Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm merge from Dave Airlie: "Highlights: - TI LCD controller KMS driver - TI OMAP KMS driver merged from staging - drop gma500 stub driver - the fbcon locking fixes - the vgacon dirty like zebra fix. - open firmware videomode and hdmi common code helpers - major locking rework for kms object handling - pageflip/cursor won't block on polling anymore! - fbcon helper and prime helper cleanups - i915: all over the map, haswell power well enhancements, valleyview macro horrors cleaned up, killing lots of legacy GTT code, - radeon: CS ioctl unification, deprecated UMS support, gpu reset rework, VM fixes - nouveau: reworked thermal code, external dp/tmds encoder support (anx9805), fences sleep instead of polling, - exynos: all over the driver fixes." Lovely conflict in radeon/evergreen_cs.c between commit de0babd60d8d ("drm/radeon: enforce use of radeon_get_ib_value when reading user cmd") and the new changes that modified that evergreen_dma_cs_parse() function. * 'drm-next' of git://people.freedesktop.org/~airlied/linux: (508 commits) drm/tilcdc: only build on arm drm/i915: Revert hdmi HDP pin checks drm/tegra: Add list of framebuffers to debugfs drm/tegra: Fix color expansion drm/tegra: Split DC_CMD_STATE_CONTROL register write drm/tegra: Implement page-flipping support drm/tegra: Implement VBLANK support drm/tegra: Implement .mode_set_base() drm/tegra: Add plane support drm/tegra: Remove bogus tegra_framebuffer structure drm: Add consistency check for page-flipping drm/radeon: Use generic HDMI infoframe helpers drm/tegra: Use generic HDMI infoframe helpers drm: Add EDID helper documentation drm: Add HDMI infoframe helpers video: Add generic HDMI infoframe helpers drm: Add some missing forward declarations drm: Move mode tables to drm_edid.c drm: Remove duplicate drm_mode_cea_vic() gma500: Fix n, m1 and m2 clock limits for sdvo and lvds ...
author: Linus Torvalds <torvalds@linux-foundation.org> 2013-02-25 16:46:44 -0800
committer: Linus Torvalds <torvalds@linux-foundation.org> 2013-02-25 16:46:44 -0800
commit: fffddfd6c8e0c10c42c6e2cc54ba880fcc36ebbb (patch)
tree: 71bc5e597124dbaf7550f1e089d675718b3ed5c0 /drivers
parent: 69086a78bdc973ec0b722be790b146e84ba8a8c4 (diff)
parent: be88298b0a3f771a4802f20c5e66af74bfd1dff1 (diff)
download: linux-fffddfd6c8e0c10c42c6e2cc54ba880fcc36ebbb.tar.gz
369 files changed, 22394 insertions, 11557 deletions
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index dbd901e94ea6..b8e2014cb9cb 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -60,7 +60,6 @@ struct intel_gtt_driver {
 };
 
 static struct _intel_private {
-	struct intel_gtt base;
 	const struct intel_gtt_driver *driver;
 	struct pci_dev *pcidev;	/* device one */
 	struct pci_dev *bridge_dev;
@@ -75,7 +74,18 @@ static struct _intel_private {
 	struct resource ifp_resource;
 	int resource_valid;
 	struct page *scratch_page;
+	phys_addr_t scratch_page_dma;
 	int refcount;
+	/* Whether i915 needs to use the dmar apis or not. */
+	unsigned int needs_dmar : 1;
+	phys_addr_t gma_bus_addr;
+	/*  Size of memory reserved for graphics by the BIOS */
+	unsigned int stolen_size;
+	/* Total number of gtt entries. */
+	unsigned int gtt_total_entries;
+	/* Part of the gtt that is mappable by the cpu, for those chips where
+	 * this is not the full gtt. */
+	unsigned int gtt_mappable_entries;
 } intel_private;
 
 #define INTEL_GTT_GEN	intel_private.driver->gen
@@ -291,15 +301,15 @@ static int intel_gtt_setup_scratch_page(void)
 	get_page(page);
 	set_pages_uc(page, 1);
 
-	if (intel_private.base.needs_dmar) {
+	if (intel_private.needs_dmar) {
 		dma_addr = pci_map_page(intel_private.pcidev, page, 0,
 				    PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 		if (pci_dma_mapping_error(intel_private.pcidev, dma_addr))
 			return -EINVAL;
 
-		intel_private.base.scratch_page_dma = dma_addr;
+		intel_private.scratch_page_dma = dma_addr;
 	} else
-		intel_private.base.scratch_page_dma = page_to_phys(page);
+		intel_private.scratch_page_dma = page_to_phys(page);
 
 	intel_private.scratch_page = page;
 
@@ -506,7 +516,7 @@ static unsigned int intel_gtt_total_entries(void)
 		/* On previous hardware, the GTT size was just what was
 		 * required to map the aperture.
 		 */
-		return intel_private.base.gtt_mappable_entries;
+		return intel_private.gtt_mappable_entries;
 	}
 }
 
@@ -546,7 +556,7 @@ static unsigned int intel_gtt_mappable_entries(void)
 static void intel_gtt_teardown_scratch_page(void)
 {
 	set_pages_wb(intel_private.scratch_page, 1);
-	pci_unmap_page(intel_private.pcidev, intel_private.base.scratch_page_dma,
+	pci_unmap_page(intel_private.pcidev, intel_private.scratch_page_dma,
 		       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 	put_page(intel_private.scratch_page);
 	__free_page(intel_private.scratch_page);
@@ -562,6 +572,40 @@ static void intel_gtt_cleanup(void)
 	intel_gtt_teardown_scratch_page();
 }
 
+/* Certain Gen5 chipsets require require idling the GPU before
+ * unmapping anything from the GTT when VT-d is enabled.
+ */
+static inline int needs_ilk_vtd_wa(void)
+{
+#ifdef CONFIG_INTEL_IOMMU
+	const unsigned short gpu_devid = intel_private.pcidev->device;
+
+	/* Query intel_iommu to see if we need the workaround. Presumably that
+	 * was loaded first.
+	 */
+	if ((gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_HB ||
+	     gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
+	     intel_iommu_gfx_mapped)
+		return 1;
+#endif
+	return 0;
+}
+
+static bool intel_gtt_can_wc(void)
+{
+	if (INTEL_GTT_GEN <= 2)
+		return false;
+
+	if (INTEL_GTT_GEN >= 6)
+		return false;
+
+	/* Reports of major corruption with ILK vt'd enabled */
+	if (needs_ilk_vtd_wa())
+		return false;
+
+	return true;
+}
+
 static int intel_gtt_init(void)
 {
 	u32 gma_addr;
@@ -572,8 +616,8 @@ static int intel_gtt_init(void)
 	if (ret != 0)
 		return ret;
 
-	intel_private.base.gtt_mappable_entries = intel_gtt_mappable_entries();
-	intel_private.base.gtt_total_entries = intel_gtt_total_entries();
+	intel_private.gtt_mappable_entries = intel_gtt_mappable_entries();
+	intel_private.gtt_total_entries = intel_gtt_total_entries();
 
 	/* save the PGETBL reg for resume */
 	intel_private.PGETBL_save =
@@ -585,13 +629,13 @@ static int intel_gtt_init(void)
 
 	dev_info(&intel_private.bridge_dev->dev,
 			"detected gtt size: %dK total, %dK mappable\n",
-			intel_private.base.gtt_total_entries * 4,
-			intel_private.base.gtt_mappable_entries * 4);
+			intel_private.gtt_total_entries * 4,
+			intel_private.gtt_mappable_entries * 4);
 
-	gtt_map_size = intel_private.base.gtt_total_entries * 4;
+	gtt_map_size = intel_private.gtt_total_entries * 4;
 
 	intel_private.gtt = NULL;
-	if (INTEL_GTT_GEN < 6 && INTEL_GTT_GEN > 2)
+	if (intel_gtt_can_wc())
 		intel_private.gtt = ioremap_wc(intel_private.gtt_bus_addr,
 					       gtt_map_size);
 	if (intel_private.gtt == NULL)
@@ -602,13 +646,12 @@ static int intel_gtt_init(void)
 		iounmap(intel_private.registers);
 		return -ENOMEM;
 	}
-	intel_private.base.gtt = intel_private.gtt;
 
 	global_cache_flush();   /* FIXME: ? */
 
-	intel_private.base.stolen_size = intel_gtt_stolen_size();
+	intel_private.stolen_size = intel_gtt_stolen_size();
 
-	intel_private.base.needs_dmar = USE_PCI_DMA_API && INTEL_GTT_GEN > 2;
+	intel_private.needs_dmar = USE_PCI_DMA_API && INTEL_GTT_GEN > 2;
 
 	ret = intel_gtt_setup_scratch_page();
 	if (ret != 0) {
@@ -623,7 +666,7 @@ static int intel_gtt_init(void)
 		pci_read_config_dword(intel_private.pcidev, I915_GMADDR,
 				      &gma_addr);
 
-	intel_private.base.gma_bus_addr = (gma_addr & PCI_BASE_ADDRESS_MEM_MASK);
+	intel_private.gma_bus_addr = (gma_addr & PCI_BASE_ADDRESS_MEM_MASK);
 
 	return 0;
 }
@@ -634,8 +677,7 @@ static int intel_fake_agp_fetch_size(void)
 	unsigned int aper_size;
 	int i;
 
-	aper_size = (intel_private.base.gtt_mappable_entries << PAGE_SHIFT)
-		    / MB(1);
+	aper_size = (intel_private.gtt_mappable_entries << PAGE_SHIFT) / MB(1);
 
 	for (i = 0; i < num_sizes; i++) {
 		if (aper_size == intel_fake_agp_sizes[i].size) {
@@ -779,7 +821,7 @@ static int intel_fake_agp_configure(void)
 	    return -EIO;
 
 	intel_private.clear_fake_agp = true;
-	agp_bridge->gart_bus_addr = intel_private.base.gma_bus_addr;
+	agp_bridge->gart_bus_addr = intel_private.gma_bus_addr;
 
 	return 0;
 }
@@ -841,12 +883,9 @@ static int intel_fake_agp_insert_entries(struct agp_memory *mem,
 {
 	int ret = -EINVAL;
 
-	if (intel_private.base.do_idle_maps)
-		return -ENODEV;
-
 	if (intel_private.clear_fake_agp) {
-		int start = intel_private.base.stolen_size / PAGE_SIZE;
-		int end = intel_private.base.gtt_mappable_entries;
+		int start = intel_private.stolen_size / PAGE_SIZE;
+		int end = intel_private.gtt_mappable_entries;
 		intel_gtt_clear_range(start, end - start);
 		intel_private.clear_fake_agp = false;
 	}
@@ -857,7 +896,7 @@ static int intel_fake_agp_insert_entries(struct agp_memory *mem,
 	if (mem->page_count == 0)
 		goto out;
 
-	if (pg_start + mem->page_count > intel_private.base.gtt_total_entries)
+	if (pg_start + mem->page_count > intel_private.gtt_total_entries)
 		goto out_err;
 
 	if (type != mem->type)
@@ -869,7 +908,7 @@ static int intel_fake_agp_insert_entries(struct agp_memory *mem,
 	if (!mem->is_flushed)
 		global_cache_flush();
 
-	if (intel_private.base.needs_dmar) {
+	if (intel_private.needs_dmar) {
 		struct sg_table st;
 
 		ret = intel_gtt_map_memory(mem->pages, mem->page_count, &st);
@@ -895,7 +934,7 @@ void intel_gtt_clear_range(unsigned int first_entry, unsigned int num_entries)
 	unsigned int i;
 
 	for (i = first_entry; i < (first_entry + num_entries); i++) {
-		intel_private.driver->write_entry(intel_private.base.scratch_page_dma,
+		intel_private.driver->write_entry(intel_private.scratch_page_dma,
 						  i, 0);
 	}
 	readl(intel_private.gtt+i-1);
@@ -908,12 +947,9 @@ static int intel_fake_agp_remove_entries(struct agp_memory *mem,
 	if (mem->page_count == 0)
 		return 0;
 
-	if (intel_private.base.do_idle_maps)
-		return -ENODEV;
-
 	intel_gtt_clear_range(pg_start, mem->page_count);
 
-	if (intel_private.base.needs_dmar) {
+	if (intel_private.needs_dmar) {
 		intel_gtt_unmap_memory(mem->sg_list, mem->num_sg);
 		mem->sg_list = NULL;
 		mem->num_sg = 0;
@@ -1070,25 +1106,6 @@ static void i965_write_entry(dma_addr_t addr,
 	writel(addr | pte_flags, intel_private.gtt + entry);
 }
 
-/* Certain Gen5 chipsets require require idling the GPU before
- * unmapping anything from the GTT when VT-d is enabled.
- */
-static inline int needs_idle_maps(void)
-{
-#ifdef CONFIG_INTEL_IOMMU
-	const unsigned short gpu_devid = intel_private.pcidev->device;
-
-	/* Query intel_iommu to see if we need the workaround. Presumably that
-	 * was loaded first.
-	 */
-	if ((gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_HB ||
-	     gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
-	     intel_iommu_gfx_mapped)
-		return 1;
-#endif
-	return 0;
-}
-
 static int i9xx_setup(void)
 {
 	u32 reg_addr, gtt_addr;
@@ -1116,9 +1133,6 @@ static int i9xx_setup(void)
 		break;
 	}
 
-	if (needs_idle_maps())
-		intel_private.base.do_idle_maps = 1;
-
 	intel_i9xx_setup_flush();
 
 	return 0;
@@ -1390,9 +1404,13 @@ int intel_gmch_probe(struct pci_dev *bridge_pdev, struct pci_dev *gpu_pdev,
 }
 EXPORT_SYMBOL(intel_gmch_probe);
 
-struct intel_gtt *intel_gtt_get(void)
+void intel_gtt_get(size_t *gtt_total, size_t *stolen_size,
+		   phys_addr_t *mappable_base, unsigned long *mappable_end)
 {
-	return &intel_private.base;
+	*gtt_total = intel_private.gtt_total_entries << PAGE_SHIFT;
+	*stolen_size = intel_private.stolen_size;
+	*mappable_base = intel_private.gma_bus_addr;
+	*mappable_end = intel_private.gtt_mappable_entries << PAGE_SHIFT;
 }
 EXPORT_SYMBOL(intel_gtt_get);
 
diff --git a/drivers/gpu/Makefile b/drivers/gpu/Makefile
index cc9277885dd0..30879df3daea 100644
--- a/drivers/gpu/Makefile
+++ b/drivers/gpu/Makefile
@@ -1 +1 @@
-obj-y			+= drm/ vga/ stub/
+obj-y			+= drm/ vga/
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 983201b450f1..1e82882da9de 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -7,6 +7,7 @@
 menuconfig DRM
 	tristate "Direct Rendering Manager (XFree86 4.1.0 and higher DRI support)"
 	depends on (AGP || AGP=n) && !EMULATED_CMPXCHG && MMU
+	select HDMI
 	select I2C
 	select I2C_ALGOBIT
 	select DMA_SHARED_BUFFER
@@ -69,6 +70,8 @@ config DRM_KMS_CMA_HELPER
 	help
 	  Choose this if you need the KMS CMA helper functions
 
+source "drivers/gpu/drm/i2c/Kconfig"
+
 config DRM_TDFX
 	tristate "3dfx Banshee/Voodoo3+"
 	depends on DRM && PCI
@@ -96,6 +99,7 @@ config DRM_RADEON
         select DRM_TTM
 	select POWER_SUPPLY
 	select HWMON
+	select BACKLIGHT_CLASS_DEVICE
 	help
 	  Choose this option if you have an ATI Radeon graphics card.  There
 	  are both PCI and AGP versions.  You don't need to choose this to
@@ -212,3 +216,7 @@ source "drivers/gpu/drm/cirrus/Kconfig"
 source "drivers/gpu/drm/shmobile/Kconfig"
 
 source "drivers/gpu/drm/tegra/Kconfig"
+
+source "drivers/gpu/drm/omapdrm/Kconfig"
+
+source "drivers/gpu/drm/tilcdc/Kconfig"
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 6f58c81cfcbc..0d59b24f8d23 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -50,4 +50,6 @@ obj-$(CONFIG_DRM_UDL) += udl/
 obj-$(CONFIG_DRM_AST) += ast/
 obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/
 obj-$(CONFIG_DRM_TEGRA) += tegra/
+obj-$(CONFIG_DRM_OMAP)	+= omapdrm/
+obj-$(CONFIG_DRM_TILCDC)	+= tilcdc/
 obj-y			+= i2c/
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 2d2c2f8d6dc6..df0d0a08097a 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -94,9 +94,9 @@ static int ast_drm_thaw(struct drm_device *dev)
 	ast_post_gpu(dev);
 
 	drm_mode_config_reset(dev);
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	drm_helper_resume_force_mode(dev);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	console_lock();
 	ast_fbdev_set_suspend(dev, 0);
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 5ccf984f063a..528429252f0f 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -98,6 +98,8 @@ struct ast_private {
 
 	struct drm_gem_object *cursor_cache;
 	uint64_t cursor_cache_gpu_addr;
+	/* Acces to this cache is protected by the crtc->mutex of the only crtc
+	 * we have. */
 	struct ttm_bo_kmap_obj cache_kmap;
 	int next_cursor;
 };
diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c
index d9ec77959dff..34931fe7d2c5 100644
--- a/drivers/gpu/drm/ast/ast_fb.c
+++ b/drivers/gpu/drm/ast/ast_fb.c
@@ -40,6 +40,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_crtc_helper.h>
 #include "ast_drv.h"
 
 static void ast_dirty_update(struct ast_fbdev *afbdev,
@@ -145,9 +146,10 @@ static int astfb_create_object(struct ast_fbdev *afbdev,
 	return ret;
 }
 
-static int astfb_create(struct ast_fbdev *afbdev,
+static int astfb_create(struct drm_fb_helper *helper,
 			struct drm_fb_helper_surface_size *sizes)
 {
+	struct ast_fbdev *afbdev = (struct ast_fbdev *)helper;
 	struct drm_device *dev = afbdev->helper.dev;
 	struct drm_mode_fb_cmd2 mode_cmd;
 	struct drm_framebuffer *fb;
@@ -248,26 +250,10 @@ static void ast_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
 	*blue = ast_crtc->lut_b[regno] << 8;
 }
 
-static int ast_find_or_create_single(struct drm_fb_helper *helper,
-					  struct drm_fb_helper_surface_size *sizes)
-{
-	struct ast_fbdev *afbdev = (struct ast_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = astfb_create(afbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static struct drm_fb_helper_funcs ast_fb_helper_funcs = {
 	.gamma_set = ast_fb_gamma_set,
 	.gamma_get = ast_fb_gamma_get,
-	.fb_probe = ast_find_or_create_single,
+	.fb_probe = astfb_create,
 };
 
 static void ast_fbdev_destroy(struct drm_device *dev,
@@ -290,6 +276,7 @@ static void ast_fbdev_destroy(struct drm_device *dev,
 	drm_fb_helper_fini(&afbdev->helper);
 
 	vfree(afbdev->sysram);
+	drm_framebuffer_unregister_private(&afb->base);
 	drm_framebuffer_cleanup(&afb->base);
 }
 
@@ -313,6 +300,10 @@ int ast_fbdev_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&afbdev->helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	drm_fb_helper_initial_config(&afbdev->helper, 32);
 	return 0;
 }
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index f668e6cc0f7a..f60fd7bd1183 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -246,16 +246,8 @@ static void ast_user_framebuffer_destroy(struct drm_framebuffer *fb)
 	kfree(fb);
 }
 
-static int ast_user_framebuffer_create_handle(struct drm_framebuffer *fb,
-					      struct drm_file *file,
-					      unsigned int *handle)
-{
-	return -EINVAL;
-}
-
 static const struct drm_framebuffer_funcs ast_fb_funcs = {
 	.destroy = ast_user_framebuffer_destroy,
-	.create_handle = ast_user_framebuffer_create_handle,
 };
 
 
@@ -266,13 +258,13 @@ int ast_framebuffer_init(struct drm_device *dev,
 {
 	int ret;
 
+	drm_helper_mode_fill_fb_struct(&ast_fb->base, mode_cmd);
+	ast_fb->obj = obj;
 	ret = drm_framebuffer_init(dev, &ast_fb->base, &ast_fb_funcs);
 	if (ret) {
 		DRM_ERROR("framebuffer init failed %d\n", ret);
 		return ret;
 	}
-	drm_helper_mode_fill_fb_struct(&ast_fb->base, mode_cmd);
-	ast_fb->obj = obj;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_fbdev.c b/drivers/gpu/drm/cirrus/cirrus_fbdev.c
index 6c6b4c87d309..e25afccaf85b 100644
--- a/drivers/gpu/drm/cirrus/cirrus_fbdev.c
+++ b/drivers/gpu/drm/cirrus/cirrus_fbdev.c
@@ -11,6 +11,7 @@
 #include <linux/module.h>
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_crtc_helper.h>
 
 #include <linux/fb.h>
 
@@ -120,9 +121,10 @@ static int cirrusfb_create_object(struct cirrus_fbdev *afbdev,
 	return ret;
 }
 
-static int cirrusfb_create(struct cirrus_fbdev *gfbdev,
+static int cirrusfb_create(struct drm_fb_helper *helper,
 			   struct drm_fb_helper_surface_size *sizes)
 {
+	struct cirrus_fbdev *gfbdev = (struct cirrus_fbdev *)helper;
 	struct drm_device *dev = gfbdev->helper.dev;
 	struct cirrus_device *cdev = gfbdev->helper.dev->dev_private;
 	struct fb_info *info;
@@ -219,23 +221,6 @@ out_iounmap:
 	return ret;
 }
 
-static int cirrus_fb_find_or_create_single(struct drm_fb_helper *helper,
-					   struct drm_fb_helper_surface_size
-					   *sizes)
-{
-	struct cirrus_fbdev *gfbdev = (struct cirrus_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = cirrusfb_create(gfbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static int cirrus_fbdev_destroy(struct drm_device *dev,
 				struct cirrus_fbdev *gfbdev)
 {
@@ -258,6 +243,7 @@ static int cirrus_fbdev_destroy(struct drm_device *dev,
 
 	vfree(gfbdev->sysram);
 	drm_fb_helper_fini(&gfbdev->helper);
+	drm_framebuffer_unregister_private(&gfb->base);
 	drm_framebuffer_cleanup(&gfb->base);
 
 	return 0;
@@ -266,7 +252,7 @@ static int cirrus_fbdev_destroy(struct drm_device *dev,
 static struct drm_fb_helper_funcs cirrus_fb_helper_funcs = {
 	.gamma_set = cirrus_crtc_fb_gamma_set,
 	.gamma_get = cirrus_crtc_fb_gamma_get,
-	.fb_probe = cirrus_fb_find_or_create_single,
+	.fb_probe = cirrusfb_create,
 };
 
 int cirrus_fbdev_init(struct cirrus_device *cdev)
@@ -290,6 +276,9 @@ int cirrus_fbdev_init(struct cirrus_device *cdev)
 		return ret;
 	}
 	drm_fb_helper_single_add_all_connectors(&gfbdev->helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(cdev->dev);
 	drm_fb_helper_initial_config(&gfbdev->helper, bpp_sel);
 
 	return 0;
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
index 6a9b12e88d46..35cbae827771 100644
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ b/drivers/gpu/drm/cirrus/cirrus_main.c
@@ -23,16 +23,8 @@ static void cirrus_user_framebuffer_destroy(struct drm_framebuffer *fb)
 	kfree(fb);
 }
 
-static int cirrus_user_framebuffer_create_handle(struct drm_framebuffer *fb,
-						 struct drm_file *file_priv,
-						 unsigned int *handle)
-{
-	return 0;
-}
-
 static const struct drm_framebuffer_funcs cirrus_fb_funcs = {
 	.destroy = cirrus_user_framebuffer_destroy,
-	.create_handle = cirrus_user_framebuffer_create_handle,
 };
 
 int cirrus_framebuffer_init(struct drm_device *dev,
@@ -42,13 +34,13 @@ int cirrus_framebuffer_init(struct drm_device *dev,
 {
 	int ret;
 
+	drm_helper_mode_fill_fb_struct(&gfb->base, mode_cmd);
+	gfb->obj = obj;
 	ret = drm_framebuffer_init(dev, &gfb->base, &cirrus_fb_funcs);
 	if (ret) {
 		DRM_ERROR("drm_framebuffer_init failed: %d\n", ret);
 		return ret;
 	}
-	drm_helper_mode_fill_fb_struct(&gfb->base, mode_cmd);
-	gfb->obj = obj;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index f2d667b8bee2..3bdf2a650d9c 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -37,6 +37,54 @@
 #include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 
+/**
+ * drm_modeset_lock_all - take all modeset locks
+ * @dev: drm device
+ *
+ * This function takes all modeset locks, suitable where a more fine-grained
+ * scheme isn't (yet) implemented.
+ */
+void drm_modeset_lock_all(struct drm_device *dev)
+{
+	struct drm_crtc *crtc;
+
+	mutex_lock(&dev->mode_config.mutex);
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
+}
+EXPORT_SYMBOL(drm_modeset_lock_all);
+
+/**
+ * drm_modeset_unlock_all - drop all modeset locks
+ * @dev: device
+ */
+void drm_modeset_unlock_all(struct drm_device *dev)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		mutex_unlock(&crtc->mutex);
+
+	mutex_unlock(&dev->mode_config.mutex);
+}
+EXPORT_SYMBOL(drm_modeset_unlock_all);
+
+/**
+ * drm_warn_on_modeset_not_all_locked - check that all modeset locks are locked
+ * @dev: device
+ */
+void drm_warn_on_modeset_not_all_locked(struct drm_device *dev)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		WARN_ON(!mutex_is_locked(&crtc->mutex));
+
+	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+}
+EXPORT_SYMBOL(drm_warn_on_modeset_not_all_locked);
+
 /* Avoid boilerplate.  I'm tired of typing. */
 #define DRM_ENUM_NAME_FN(fnname, list)				\
 	char *fnname(int val)					\
@@ -203,12 +251,10 @@ char *drm_get_connector_status_name(enum drm_connector_status status)
 }
 
 /**
- * drm_mode_object_get - allocate a new identifier
+ * drm_mode_object_get - allocate a new modeset identifier
  * @dev: DRM device
- * @ptr: object pointer, used to generate unique ID
- * @type: object type
- *
- * LOCKING:
+ * @obj: object pointer, used to generate unique ID
+ * @obj_type: object type
  *
  * Create a unique identifier based on @ptr in @dev's identifier space.  Used
  * for tracking modes, CRTCs and connectors.
@@ -231,24 +277,27 @@ again:
 
 	mutex_lock(&dev->mode_config.idr_mutex);
 	ret = idr_get_new_above(&dev->mode_config.crtc_idr, obj, 1, &new_id);
+
+	if (!ret) {
+		/*
+		 * Set up the object linking under the protection of the idr
+		 * lock so that other users can't see inconsistent state.
+		 */
+		obj->id = new_id;
+		obj->type = obj_type;
+	}
 	mutex_unlock(&dev->mode_config.idr_mutex);
+
 	if (ret == -EAGAIN)
 		goto again;
-	else if (ret)
-		return ret;
 
-	obj->id = new_id;
-	obj->type = obj_type;
-	return 0;
+	return ret;
 }
 
 /**
- * drm_mode_object_put - free an identifer
+ * drm_mode_object_put - free a modeset identifer
  * @dev: DRM device
- * @id: ID to free
- *
- * LOCKING:
- * Caller must hold DRM mode_config lock.
+ * @object: object to free
  *
  * Free @id from @dev's unique identifier pool.
  */
@@ -260,11 +309,24 @@ static void drm_mode_object_put(struct drm_device *dev,
 	mutex_unlock(&dev->mode_config.idr_mutex);
 }
 
+/**
+ * drm_mode_object_find - look up a drm object with static lifetime
+ * @dev: drm device
+ * @id: id of the mode object
+ * @type: type of the mode object
+ *
+ * Note that framebuffers cannot be looked up with this functions - since those
+ * are reference counted, they need special treatment.
+ */
 struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 		uint32_t id, uint32_t type)
 {
 	struct drm_mode_object *obj = NULL;
 
+	/* Framebuffers are reference counted and need their own lookup
+	 * function.*/
+	WARN_ON(type == DRM_MODE_OBJECT_FB);
+
 	mutex_lock(&dev->mode_config.idr_mutex);
 	obj = idr_find(&dev->mode_config.crtc_idr, id);
 	if (!obj || (obj->type != type) || (obj->id != id))
@@ -278,13 +340,18 @@ EXPORT_SYMBOL(drm_mode_object_find);
 /**
  * drm_framebuffer_init - initialize a framebuffer
  * @dev: DRM device
- *
- * LOCKING:
- * Caller must hold mode config lock.
+ * @fb: framebuffer to be initialized
+ * @funcs: ... with these functions
  *
  * Allocates an ID for the framebuffer's parent mode object, sets its mode
  * functions & device file and adds it to the master fd list.
  *
+ * IMPORTANT:
+ * This functions publishes the fb and makes it available for concurrent access
+ * by other users. Which means by this point the fb _must_ be fully set up -
+ * since all the fb attributes are invariant over its lifetime, no further
+ * locking but only correct reference counting is required.
+ *
  * RETURNS:
  * Zero on success, error code on failure.
  */
@@ -293,16 +360,23 @@ int drm_framebuffer_init(struct drm_device *dev, struct drm_framebuffer *fb,
 {
 	int ret;
 
+	mutex_lock(&dev->mode_config.fb_lock);
 	kref_init(&fb->refcount);
+	INIT_LIST_HEAD(&fb->filp_head);
+	fb->dev = dev;
+	fb->funcs = funcs;
 
 	ret = drm_mode_object_get(dev, &fb->base, DRM_MODE_OBJECT_FB);
 	if (ret)
-		return ret;
+		goto out;
+
+	/* Grab the idr reference. */
+	drm_framebuffer_reference(fb);
 
-	fb->dev = dev;
-	fb->funcs = funcs;
 	dev->mode_config.num_fb++;
 	list_add(&fb->head, &dev->mode_config.fb_list);
+out:
+	mutex_unlock(&dev->mode_config.fb_lock);
 
 	return 0;
 }
@@ -315,23 +389,63 @@ static void drm_framebuffer_free(struct kref *kref)
 	fb->funcs->destroy(fb);
 }
 
+static struct drm_framebuffer *__drm_framebuffer_lookup(struct drm_device *dev,
+							uint32_t id)
+{
+	struct drm_mode_object *obj = NULL;
+	struct drm_framebuffer *fb;
+
+	mutex_lock(&dev->mode_config.idr_mutex);
+	obj = idr_find(&dev->mode_config.crtc_idr, id);
+	if (!obj || (obj->type != DRM_MODE_OBJECT_FB) || (obj->id != id))
+		fb = NULL;
+	else
+		fb = obj_to_fb(obj);
+	mutex_unlock(&dev->mode_config.idr_mutex);
+
+	return fb;
+}
+
+/**
+ * drm_framebuffer_lookup - look up a drm framebuffer and grab a reference
+ * @dev: drm device
+ * @id: id of the fb object
+ *
+ * If successful, this grabs an additional reference to the framebuffer -
+ * callers need to make sure to eventually unreference the returned framebuffer
+ * again.
+ */
+struct drm_framebuffer *drm_framebuffer_lookup(struct drm_device *dev,
+					       uint32_t id)
+{
+	struct drm_framebuffer *fb;
+
+	mutex_lock(&dev->mode_config.fb_lock);
+	fb = __drm_framebuffer_lookup(dev, id);
+	if (fb)
+		kref_get(&fb->refcount);
+	mutex_unlock(&dev->mode_config.fb_lock);
+
+	return fb;
+}
+EXPORT_SYMBOL(drm_framebuffer_lookup);
+
 /**
  * drm_framebuffer_unreference - unref a framebuffer
+ * @fb: framebuffer to unref
  *
- * LOCKING:
- * Caller must hold mode config lock.
+ * This functions decrements the fb's refcount and frees it if it drops to zero.
  */
 void drm_framebuffer_unreference(struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = fb->dev;
 	DRM_DEBUG("FB ID: %d\n", fb->base.id);
-	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 	kref_put(&fb->refcount, drm_framebuffer_free);
 }
 EXPORT_SYMBOL(drm_framebuffer_unreference);
 
 /**
  * drm_framebuffer_reference - incr the fb refcnt
+ * @fb: framebuffer
  */
 void drm_framebuffer_reference(struct drm_framebuffer *fb)
 {
@@ -340,29 +454,74 @@ void drm_framebuffer_reference(struct drm_framebuffer *fb)
 }
 EXPORT_SYMBOL(drm_framebuffer_reference);
 
+static void drm_framebuffer_free_bug(struct kref *kref)
+{
+	BUG();
+}
+
+static void __drm_framebuffer_unreference(struct drm_framebuffer *fb)
+{
+	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	kref_put(&fb->refcount, drm_framebuffer_free_bug);
+}
+
+/* dev->mode_config.fb_lock must be held! */
+static void __drm_framebuffer_unregister(struct drm_device *dev,
+					 struct drm_framebuffer *fb)
+{
+	mutex_lock(&dev->mode_config.idr_mutex);
+	idr_remove(&dev->mode_config.crtc_idr, fb->base.id);
+	mutex_unlock(&dev->mode_config.idr_mutex);
+
+	fb->base.id = 0;
+
+	__drm_framebuffer_unreference(fb);
+}
+
+/**
+ * drm_framebuffer_unregister_private - unregister a private fb from the lookup idr
+ * @fb: fb to unregister
+ *
+ * Drivers need to call this when cleaning up driver-private framebuffers, e.g.
+ * those used for fbdev. Note that the caller must hold a reference of it's own,
+ * i.e. the object may not be destroyed through this call (since it'll lead to a
+ * locking inversion).
+ */
+void drm_framebuffer_unregister_private(struct drm_framebuffer *fb)
+{
+	struct drm_device *dev = fb->dev;
+
+	mutex_lock(&dev->mode_config.fb_lock);
+	/* Mark fb as reaped and drop idr ref. */
+	__drm_framebuffer_unregister(dev, fb);
+	mutex_unlock(&dev->mode_config.fb_lock);
+}
+EXPORT_SYMBOL(drm_framebuffer_unregister_private);
+
 /**
  * drm_framebuffer_cleanup - remove a framebuffer object
  * @fb: framebuffer to remove
  *
- * LOCKING:
- * Caller must hold mode config lock.
+ * Cleanup references to a user-created framebuffer. This function is intended
+ * to be used from the drivers ->destroy callback.
+ *
+ * Note that this function does not remove the fb from active usuage - if it is
+ * still used anywhere, hilarity can ensue since userspace could call getfb on
+ * the id and get back -EINVAL. Obviously no concern at driver unload time.
  *
- * Scans all the CRTCs in @dev's mode_config.  If they're using @fb, removes
- * it, setting it to NULL.
+ * Also, the framebuffer will not be removed from the lookup idr - for
+ * user-created framebuffers this will happen in in the rmfb ioctl. For
+ * driver-private objects (e.g. for fbdev) drivers need to explicitly call
+ * drm_framebuffer_unregister_private.
  */
 void drm_framebuffer_cleanup(struct drm_framebuffer *fb)
 {
 	struct drm_device *dev = fb->dev;
-	/*
-	 * This could be moved to drm_framebuffer_remove(), but for
-	 * debugging is nice to keep around the list of fb's that are
-	 * no longer associated w/ a drm_file but are not unreferenced
-	 * yet.  (i915 and omapdrm have debugfs files which will show
-	 * this.)
-	 */
-	drm_mode_object_put(dev, &fb->base);
+
+	mutex_lock(&dev->mode_config.fb_lock);
 	list_del(&fb->head);
 	dev->mode_config.num_fb--;
+	mutex_unlock(&dev->mode_config.fb_lock);
 }
 EXPORT_SYMBOL(drm_framebuffer_cleanup);
 
@@ -370,11 +529,13 @@ EXPORT_SYMBOL(drm_framebuffer_cleanup);
  * drm_framebuffer_remove - remove and unreference a framebuffer object
  * @fb: framebuffer to remove
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Scans all the CRTCs and planes in @dev's mode_config.  If they're
- * using @fb, removes it, setting it to NULL.
+ * using @fb, removes it, setting it to NULL. Then drops the reference to the
+ * passed-in framebuffer. Might take the modeset locks.
+ *
+ * Note that this function optimizes the cleanup away if the caller holds the
+ * last reference to the framebuffer. It is also guaranteed to not take the
+ * modeset locks in this case.
  */
 void drm_framebuffer_remove(struct drm_framebuffer *fb)
 {
@@ -384,33 +545,53 @@ void drm_framebuffer_remove(struct drm_framebuffer *fb)
 	struct drm_mode_set set;
 	int ret;
 
-	/* remove from any CRTC */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (crtc->fb == fb) {
-			/* should turn off the crtc */
-			memset(&set, 0, sizeof(struct drm_mode_set));
-			set.crtc = crtc;
-			set.fb = NULL;
-			ret = crtc->funcs->set_config(&set);
-			if (ret)
-				DRM_ERROR("failed to reset crtc %p when fb was deleted\n", crtc);
+	WARN_ON(!list_empty(&fb->filp_head));
+
+	/*
+	 * drm ABI mandates that we remove any deleted framebuffers from active
+	 * useage. But since most sane clients only remove framebuffers they no
+	 * longer need, try to optimize this away.
+	 *
+	 * Since we're holding a reference ourselves, observing a refcount of 1
+	 * means that we're the last holder and can skip it. Also, the refcount
+	 * can never increase from 1 again, so we don't need any barriers or
+	 * locks.
+	 *
+	 * Note that userspace could try to race with use and instate a new
+	 * usage _after_ we've cleared all current ones. End result will be an
+	 * in-use fb with fb-id == 0. Userspace is allowed to shoot its own foot
+	 * in this manner.
+	 */
+	if (atomic_read(&fb->refcount.refcount) > 1) {
+		drm_modeset_lock_all(dev);
+		/* remove from any CRTC */
+		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+			if (crtc->fb == fb) {
+				/* should turn off the crtc */
+				memset(&set, 0, sizeof(struct drm_mode_set));
+				set.crtc = crtc;
+				set.fb = NULL;
+				ret = drm_mode_set_config_internal(&set);
+				if (ret)
+					DRM_ERROR("failed to reset crtc %p when fb was deleted\n", crtc);
+			}
 		}
-	}
 
-	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
-		if (plane->fb == fb) {
-			/* should turn off the crtc */
-			ret = plane->funcs->disable_plane(plane);
-			if (ret)
-				DRM_ERROR("failed to disable plane with busy fb\n");
-			/* disconnect the plane from the fb and crtc: */
-			plane->fb = NULL;
-			plane->crtc = NULL;
+		list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
+			if (plane->fb == fb) {
+				/* should turn off the crtc */
+				ret = plane->funcs->disable_plane(plane);
+				if (ret)
+					DRM_ERROR("failed to disable plane with busy fb\n");
+				/* disconnect the plane from the fb and crtc: */
+				__drm_framebuffer_unreference(plane->fb);
+				plane->fb = NULL;
+				plane->crtc = NULL;
+			}
 		}
+		drm_modeset_unlock_all(dev);
 	}
 
-	list_del(&fb->filp_head);
-
 	drm_framebuffer_unreference(fb);
 }
 EXPORT_SYMBOL(drm_framebuffer_remove);
@@ -421,9 +602,6 @@ EXPORT_SYMBOL(drm_framebuffer_remove);
  * @crtc: CRTC object to init
  * @funcs: callbacks for the new CRTC
  *
- * LOCKING:
- * Takes mode_config lock.
- *
  * Inits a new object created as base part of an driver crtc object.
  *
  * RETURNS:
@@ -438,7 +616,9 @@ int drm_crtc_init(struct drm_device *dev, struct drm_crtc *crtc,
 	crtc->funcs = funcs;
 	crtc->invert_dimensions = false;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
+	mutex_init(&crtc->mutex);
+	mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
 
 	ret = drm_mode_object_get(dev, &crtc->base, DRM_MODE_OBJECT_CRTC);
 	if (ret)
@@ -450,7 +630,7 @@ int drm_crtc_init(struct drm_device *dev, struct drm_crtc *crtc,
 	dev->mode_config.num_crtc++;
 
  out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -460,9 +640,6 @@ EXPORT_SYMBOL(drm_crtc_init);
  * drm_crtc_cleanup - Cleans up the core crtc usage.
  * @crtc: CRTC to cleanup
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Cleanup @crtc. Removes from drm modesetting space
  * does NOT free object, caller does that.
  */
@@ -484,9 +661,6 @@ EXPORT_SYMBOL(drm_crtc_cleanup);
  * @connector: connector the new mode
  * @mode: mode data
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Add @mode to @connector's mode list for later use.
  */
 void drm_mode_probed_add(struct drm_connector *connector,
@@ -501,9 +675,6 @@ EXPORT_SYMBOL(drm_mode_probed_add);
  * @connector: connector list to modify
  * @mode: mode to remove
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Remove @mode from @connector's mode list, then free it.
  */
 void drm_mode_remove(struct drm_connector *connector,
@@ -519,10 +690,7 @@ EXPORT_SYMBOL(drm_mode_remove);
  * @dev: DRM device
  * @connector: the connector to init
  * @funcs: callbacks for this connector
- * @name: user visible name of the connector
- *
- * LOCKING:
- * Takes mode config lock.
+ * @connector_type: user visible type of the connector
  *
  * Initialises a preallocated connector. Connectors should be
  * subclassed as part of driver connector objects.
@@ -537,7 +705,7 @@ int drm_connector_init(struct drm_device *dev,
 {
 	int ret;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	ret = drm_mode_object_get(dev, &connector->base, DRM_MODE_OBJECT_CONNECTOR);
 	if (ret)
@@ -567,7 +735,7 @@ int drm_connector_init(struct drm_device *dev,
 				      dev->mode_config.dpms_property, 0);
 
  out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -577,9 +745,6 @@ EXPORT_SYMBOL(drm_connector_init);
  * drm_connector_cleanup - cleans up an initialised connector
  * @connector: connector to cleanup
  *
- * LOCKING:
- * Takes mode config lock.
- *
  * Cleans up the connector but doesn't free the object.
  */
 void drm_connector_cleanup(struct drm_connector *connector)
@@ -596,11 +761,9 @@ void drm_connector_cleanup(struct drm_connector *connector)
 	list_for_each_entry_safe(mode, t, &connector->user_modes, head)
 		drm_mode_remove(connector, mode);
 
-	mutex_lock(&dev->mode_config.mutex);
 	drm_mode_object_put(dev, &connector->base);
 	list_del(&connector->head);
 	dev->mode_config.num_connector--;
-	mutex_unlock(&dev->mode_config.mutex);
 }
 EXPORT_SYMBOL(drm_connector_cleanup);
 
@@ -622,7 +785,7 @@ int drm_encoder_init(struct drm_device *dev,
 {
 	int ret;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	ret = drm_mode_object_get(dev, &encoder->base, DRM_MODE_OBJECT_ENCODER);
 	if (ret)
@@ -636,7 +799,7 @@ int drm_encoder_init(struct drm_device *dev,
 	dev->mode_config.num_encoder++;
 
  out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -645,11 +808,11 @@ EXPORT_SYMBOL(drm_encoder_init);
 void drm_encoder_cleanup(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	drm_mode_object_put(dev, &encoder->base);
 	list_del(&encoder->head);
 	dev->mode_config.num_encoder--;
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 }
 EXPORT_SYMBOL(drm_encoder_cleanup);
 
@@ -661,7 +824,7 @@ int drm_plane_init(struct drm_device *dev, struct drm_plane *plane,
 {
 	int ret;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	ret = drm_mode_object_get(dev, &plane->base, DRM_MODE_OBJECT_PLANE);
 	if (ret)
@@ -695,7 +858,7 @@ int drm_plane_init(struct drm_device *dev, struct drm_plane *plane,
 	}
 
  out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -705,7 +868,7 @@ void drm_plane_cleanup(struct drm_plane *plane)
 {
 	struct drm_device *dev = plane->dev;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	kfree(plane->format_types);
 	drm_mode_object_put(dev, &plane->base);
 	/* if not added to a list, it must be a private plane */
@@ -713,7 +876,7 @@ void drm_plane_cleanup(struct drm_plane *plane)
 		list_del(&plane->head);
 		dev->mode_config.num_plane--;
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 }
 EXPORT_SYMBOL(drm_plane_cleanup);
 
@@ -721,9 +884,6 @@ EXPORT_SYMBOL(drm_plane_cleanup);
  * drm_mode_create - create a new display mode
  * @dev: DRM device
  *
- * LOCKING:
- * Caller must hold DRM mode_config lock.
- *
  * Create a new drm_display_mode, give it an ID, and return it.
  *
  * RETURNS:
@@ -751,9 +911,6 @@ EXPORT_SYMBOL(drm_mode_create);
  * @dev: DRM device
  * @mode: mode to remove
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Free @mode's unique identifier, then free it.
  */
 void drm_mode_destroy(struct drm_device *dev, struct drm_display_mode *mode)
@@ -978,16 +1135,19 @@ EXPORT_SYMBOL(drm_mode_create_dirty_info_property);
  * drm_mode_config_init - initialize DRM mode_configuration structure
  * @dev: DRM device
  *
- * LOCKING:
- * None, should happen single threaded at init time.
- *
  * Initialize @dev's mode_config structure, used for tracking the graphics
  * configuration of @dev.
+ *
+ * Since this initializes the modeset locks, no locking is possible. Which is no
+ * problem, since this should happen single threaded at init time. It is the
+ * driver's problem to ensure this guarantee.
+ *
  */
 void drm_mode_config_init(struct drm_device *dev)
 {
 	mutex_init(&dev->mode_config.mutex);
 	mutex_init(&dev->mode_config.idr_mutex);
+	mutex_init(&dev->mode_config.fb_lock);
 	INIT_LIST_HEAD(&dev->mode_config.fb_list);
 	INIT_LIST_HEAD(&dev->mode_config.crtc_list);
 	INIT_LIST_HEAD(&dev->mode_config.connector_list);
@@ -997,9 +1157,9 @@ void drm_mode_config_init(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev->mode_config.plane_list);
 	idr_init(&dev->mode_config.crtc_idr);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	drm_mode_create_standard_connector_properties(dev);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	/* Just to be sure */
 	dev->mode_config.num_fb = 0;
@@ -1057,12 +1217,13 @@ EXPORT_SYMBOL(drm_mode_group_init_legacy_group);
  * drm_mode_config_cleanup - free up DRM mode_config info
  * @dev: DRM device
  *
- * LOCKING:
- * Caller must hold mode config lock.
- *
  * Free up all the connectors and CRTCs associated with this DRM device, then
  * free up the framebuffers and associated buffer objects.
  *
+ * Note that since this /should/ happen single-threaded at driver/device
+ * teardown time, no locking is required. It's the driver's job to ensure that
+ * this guarantee actually holds true.
+ *
  * FIXME: cleanup any dangling user buffer objects too
  */
 void drm_mode_config_cleanup(struct drm_device *dev)
@@ -1089,6 +1250,15 @@ void drm_mode_config_cleanup(struct drm_device *dev)
 		drm_property_destroy(dev, property);
 	}
 
+	/*
+	 * Single-threaded teardown context, so it's not required to grab the
+	 * fb_lock to protect against concurrent fb_list access. Contrary, it
+	 * would actually deadlock with the drm_framebuffer_cleanup function.
+	 *
+	 * Also, if there are any framebuffers left, that's a driver leak now,
+	 * so politely WARN about this.
+	 */
+	WARN_ON(!list_empty(&dev->mode_config.fb_list));
 	list_for_each_entry_safe(fb, fbt, &dev->mode_config.fb_list, head) {
 		drm_framebuffer_remove(fb);
 	}
@@ -1112,9 +1282,6 @@ EXPORT_SYMBOL(drm_mode_config_cleanup);
  * @out: drm_mode_modeinfo struct to return to the user
  * @in: drm_display_mode to use
  *
- * LOCKING:
- * None.
- *
  * Convert a drm_display_mode into a drm_mode_modeinfo structure to return to
  * the user.
  */
@@ -1151,9 +1318,6 @@ static void drm_crtc_convert_to_umode(struct drm_mode_modeinfo *out,
  * @out: drm_display_mode to return to the user
  * @in: drm_mode_modeinfo to use
  *
- * LOCKING:
- * None.
- *
  * Convert a drm_mode_modeinfo into a drm_display_mode structure to return to
  * the caller.
  *
@@ -1188,13 +1352,9 @@ static int drm_crtc_convert_umode(struct drm_display_mode *out,
 
 /**
  * drm_mode_getresources - get graphics configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Construct a set of configuration description structures and return
  * them to the user, including CRTC, connector and framebuffer configuration.
@@ -1228,8 +1388,8 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
 
+	mutex_lock(&file_priv->fbs_lock);
 	/*
 	 * For the non-control nodes we need to limit the list of resources
 	 * by IDs in the group list for this node
@@ -1237,6 +1397,23 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 	list_for_each(lh, &file_priv->fbs)
 		fb_count++;
 
+	/* handle this in 4 parts */
+	/* FBs */
+	if (card_res->count_fbs >= fb_count) {
+		copied = 0;
+		fb_id = (uint32_t __user *)(unsigned long)card_res->fb_id_ptr;
+		list_for_each_entry(fb, &file_priv->fbs, filp_head) {
+			if (put_user(fb->base.id, fb_id + copied)) {
+				mutex_unlock(&file_priv->fbs_lock);
+				return -EFAULT;
+			}
+			copied++;
+		}
+	}
+	card_res->count_fbs = fb_count;
+	mutex_unlock(&file_priv->fbs_lock);
+
+	drm_modeset_lock_all(dev);
 	mode_group = &file_priv->master->minor->mode_group;
 	if (file_priv->master->minor->type == DRM_MINOR_CONTROL) {
 
@@ -1260,21 +1437,6 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 	card_res->max_width = dev->mode_config.max_width;
 	card_res->min_width = dev->mode_config.min_width;
 
-	/* handle this in 4 parts */
-	/* FBs */
-	if (card_res->count_fbs >= fb_count) {
-		copied = 0;
-		fb_id = (uint32_t __user *)(unsigned long)card_res->fb_id_ptr;
-		list_for_each_entry(fb, &file_priv->fbs, filp_head) {
-			if (put_user(fb->base.id, fb_id + copied)) {
-				ret = -EFAULT;
-				goto out;
-			}
-			copied++;
-		}
-	}
-	card_res->count_fbs = fb_count;
-
 	/* CRTCs */
 	if (card_res->count_crtcs >= crtc_count) {
 		copied = 0;
@@ -1370,19 +1532,15 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 		  card_res->count_connectors, card_res->count_encoders);
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
 /**
  * drm_mode_getcrtc - get CRTC configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Construct a CRTC configuration structure to return to the user.
  *
@@ -1402,7 +1560,7 @@ int drm_mode_getcrtc(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, crtc_resp->crtc_id,
 				   DRM_MODE_OBJECT_CRTC);
@@ -1430,19 +1588,15 @@ int drm_mode_getcrtc(struct drm_device *dev,
 	}
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
 /**
  * drm_mode_getconnector - get connector configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Construct a connector configuration structure to return to the user.
  *
@@ -1575,6 +1729,7 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 
 out:
 	mutex_unlock(&dev->mode_config.mutex);
+
 	return ret;
 }
 
@@ -1589,7 +1744,7 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, enc_resp->encoder_id,
 				   DRM_MODE_OBJECT_ENCODER);
 	if (!obj) {
@@ -1608,7 +1763,7 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 	enc_resp->possible_clones = encoder->possible_clones;
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -1618,9 +1773,6 @@ out:
  * @data: ioctl data
  * @file_priv: DRM file info
  *
- * LOCKING:
- * Takes mode config lock.
- *
  * Return an plane count and set of IDs.
  */
 int drm_mode_getplane_res(struct drm_device *dev, void *data,
@@ -1635,7 +1787,7 @@ int drm_mode_getplane_res(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	config = &dev->mode_config;
 
 	/*
@@ -1657,7 +1809,7 @@ int drm_mode_getplane_res(struct drm_device *dev, void *data,
 	plane_resp->count_planes = config->num_plane;
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -1667,9 +1819,6 @@ out:
  * @data: ioctl data
  * @file_priv: DRM file info
  *
- * LOCKING:
- * Takes mode config lock.
- *
  * Return plane info, including formats supported, gamma size, any
  * current fb, etc.
  */
@@ -1685,7 +1834,7 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, plane_resp->plane_id,
 				   DRM_MODE_OBJECT_PLANE);
 	if (!obj) {
@@ -1725,7 +1874,7 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 	plane_resp->count_format_types = plane->format_count;
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -1733,10 +1882,7 @@ out:
  * drm_mode_setplane - set up or tear down an plane
  * @dev: DRM device
  * @data: ioctl data*
- * @file_prive: DRM file info
- *
- * LOCKING:
- * Takes mode config lock.
+ * @file_priv: DRM file info
  *
  * Set plane info, including placement, fb, scaling, and other factors.
  * Or pass a NULL fb to disable.
@@ -1748,7 +1894,7 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	struct drm_mode_object *obj;
 	struct drm_plane *plane;
 	struct drm_crtc *crtc;
-	struct drm_framebuffer *fb;
+	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
 	int ret = 0;
 	unsigned int fb_width, fb_height;
 	int i;
@@ -1756,8 +1902,6 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
-
 	/*
 	 * First, find the plane, crtc, and fb objects.  If not available,
 	 * we don't bother to call the driver.
@@ -1767,16 +1911,18 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	if (!obj) {
 		DRM_DEBUG_KMS("Unknown plane ID %d\n",
 			      plane_req->plane_id);
-		ret = -ENOENT;
-		goto out;
+		return -ENOENT;
 	}
 	plane = obj_to_plane(obj);
 
 	/* No fb means shut it down */
 	if (!plane_req->fb_id) {
+		drm_modeset_lock_all(dev);
+		old_fb = plane->fb;
 		plane->funcs->disable_plane(plane);
 		plane->crtc = NULL;
 		plane->fb = NULL;
+		drm_modeset_unlock_all(dev);
 		goto out;
 	}
 
@@ -1790,15 +1936,13 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	}
 	crtc = obj_to_crtc(obj);
 
-	obj = drm_mode_object_find(dev, plane_req->fb_id,
-				   DRM_MODE_OBJECT_FB);
-	if (!obj) {
+	fb = drm_framebuffer_lookup(dev, plane_req->fb_id);
+	if (!fb) {
 		DRM_DEBUG_KMS("Unknown framebuffer ID %d\n",
 			      plane_req->fb_id);
 		ret = -ENOENT;
 		goto out;
 	}
-	fb = obj_to_fb(obj);
 
 	/* Check whether this plane supports the fb pixel format. */
 	for (i = 0; i < plane->format_count; i++)
@@ -1844,31 +1988,62 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 		goto out;
 	}
 
+	drm_modeset_lock_all(dev);
 	ret = plane->funcs->update_plane(plane, crtc, fb,
 					 plane_req->crtc_x, plane_req->crtc_y,
 					 plane_req->crtc_w, plane_req->crtc_h,
 					 plane_req->src_x, plane_req->src_y,
 					 plane_req->src_w, plane_req->src_h);
 	if (!ret) {
+		old_fb = plane->fb;
 		plane->crtc = crtc;
 		plane->fb = fb;
+		fb = NULL;
 	}
+	drm_modeset_unlock_all(dev);
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	if (fb)
+		drm_framebuffer_unreference(fb);
+	if (old_fb)
+		drm_framebuffer_unreference(old_fb);
 
 	return ret;
 }
 
 /**
- * drm_mode_setcrtc - set CRTC configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
+ * drm_mode_set_config_internal - helper to call ->set_config
+ * @set: modeset config to set
  *
- * LOCKING:
- * Takes mode config lock.
+ * This is a little helper to wrap internal calls to the ->set_config driver
+ * interface. The only thing it adds is correct refcounting dance.
+ */
+int drm_mode_set_config_internal(struct drm_mode_set *set)
+{
+	struct drm_crtc *crtc = set->crtc;
+	struct drm_framebuffer *fb, *old_fb;
+	int ret;
+
+	old_fb = crtc->fb;
+	fb = set->fb;
+
+	ret = crtc->funcs->set_config(set);
+	if (ret == 0) {
+		if (old_fb)
+			drm_framebuffer_unreference(old_fb);
+		if (fb)
+			drm_framebuffer_reference(fb);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(drm_mode_set_config_internal);
+
+/**
+ * drm_mode_setcrtc - set CRTC configuration
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Build a new CRTC configuration based on user request.
  *
@@ -1899,7 +2074,7 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 	if (crtc_req->x > INT_MAX || crtc_req->y > INT_MAX)
 		return -ERANGE;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, crtc_req->crtc_id,
 				   DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
@@ -1921,16 +2096,16 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 				goto out;
 			}
 			fb = crtc->fb;
+			/* Make refcounting symmetric with the lookup path. */
+			drm_framebuffer_reference(fb);
 		} else {
-			obj = drm_mode_object_find(dev, crtc_req->fb_id,
-						   DRM_MODE_OBJECT_FB);
-			if (!obj) {
+			fb = drm_framebuffer_lookup(dev, crtc_req->fb_id);
+			if (!fb) {
 				DRM_DEBUG_KMS("Unknown FB ID%d\n",
 						crtc_req->fb_id);
 				ret = -EINVAL;
 				goto out;
 			}
-			fb = obj_to_fb(obj);
 		}
 
 		mode = drm_mode_create(dev);
@@ -2027,12 +2202,15 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 	set.connectors = connector_set;
 	set.num_connectors = crtc_req->count_connectors;
 	set.fb = fb;
-	ret = crtc->funcs->set_config(&set);
+	ret = drm_mode_set_config_internal(&set);
 
 out:
+	if (fb)
+		drm_framebuffer_unreference(fb);
+
 	kfree(connector_set);
 	drm_mode_destroy(dev, mode);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -2050,15 +2228,14 @@ int drm_mode_cursor_ioctl(struct drm_device *dev,
 	if (!req->flags || (~DRM_MODE_CURSOR_FLAGS & req->flags))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
 	obj = drm_mode_object_find(dev, req->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", req->crtc_id);
-		ret = -EINVAL;
-		goto out;
+		return -EINVAL;
 	}
 	crtc = obj_to_crtc(obj);
 
+	mutex_lock(&crtc->mutex);
 	if (req->flags & DRM_MODE_CURSOR_BO) {
 		if (!crtc->funcs->cursor_set) {
 			ret = -ENXIO;
@@ -2078,7 +2255,8 @@ int drm_mode_cursor_ioctl(struct drm_device *dev,
 		}
 	}
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&crtc->mutex);
+
 	return ret;
 }
 
@@ -2089,7 +2267,7 @@ uint32_t drm_mode_legacy_fb_format(uint32_t bpp, uint32_t depth)
 
 	switch (bpp) {
 	case 8:
-		fmt = DRM_FORMAT_RGB332;
+		fmt = DRM_FORMAT_C8;
 		break;
 	case 16:
 		if (depth == 15)
@@ -2120,13 +2298,9 @@ EXPORT_SYMBOL(drm_mode_legacy_fb_format);
 
 /**
  * drm_mode_addfb - add an FB to the graphics configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Add a new FB to the specified CRTC, given a user request.
  *
@@ -2161,24 +2335,19 @@ int drm_mode_addfb(struct drm_device *dev,
 	if ((config->min_height > r.height) || (r.height > config->max_height))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
-
-	/* TODO check buffer is sufficiently large */
-	/* TODO setup destructor callback */
-
 	fb = dev->mode_config.funcs->fb_create(dev, file_priv, &r);
 	if (IS_ERR(fb)) {
 		DRM_DEBUG_KMS("could not create framebuffer\n");
-		ret = PTR_ERR(fb);
-		goto out;
+		drm_modeset_unlock_all(dev);
+		return PTR_ERR(fb);
 	}
 
+	mutex_lock(&file_priv->fbs_lock);
 	or->fb_id = fb->base.id;
 	list_add(&fb->filp_head, &file_priv->fbs);
 	DRM_DEBUG_KMS("[FB:%d]\n", fb->base.id);
+	mutex_unlock(&file_priv->fbs_lock);
 
-out:
-	mutex_unlock(&dev->mode_config.mutex);
 	return ret;
 }
 
@@ -2304,13 +2473,9 @@ static int framebuffer_check(const struct drm_mode_fb_cmd2 *r)
 
 /**
  * drm_mode_addfb2 - add an FB to the graphics configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Add a new FB to the specified CRTC, given a user request with format.
  *
@@ -2350,33 +2515,28 @@ int drm_mode_addfb2(struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	mutex_lock(&dev->mode_config.mutex);
-
 	fb = dev->mode_config.funcs->fb_create(dev, file_priv, r);
 	if (IS_ERR(fb)) {
 		DRM_DEBUG_KMS("could not create framebuffer\n");
-		ret = PTR_ERR(fb);
-		goto out;
+		drm_modeset_unlock_all(dev);
+		return PTR_ERR(fb);
 	}
 
+	mutex_lock(&file_priv->fbs_lock);
 	r->fb_id = fb->base.id;
 	list_add(&fb->filp_head, &file_priv->fbs);
 	DRM_DEBUG_KMS("[FB:%d]\n", fb->base.id);
+	mutex_unlock(&file_priv->fbs_lock);
+
 
-out:
-	mutex_unlock(&dev->mode_config.mutex);
 	return ret;
 }
 
 /**
  * drm_mode_rmfb - remove an FB from the configuration
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Remove the FB specified by the user.
  *
@@ -2388,50 +2548,49 @@ out:
 int drm_mode_rmfb(struct drm_device *dev,
 		   void *data, struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_framebuffer *fb = NULL;
 	struct drm_framebuffer *fbl = NULL;
 	uint32_t *id = data;
-	int ret = 0;
 	int found = 0;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
-	obj = drm_mode_object_find(dev, *id, DRM_MODE_OBJECT_FB);
-	/* TODO check that we really get a framebuffer back. */
-	if (!obj) {
-		ret = -EINVAL;
-		goto out;
-	}
-	fb = obj_to_fb(obj);
+	mutex_lock(&file_priv->fbs_lock);
+	mutex_lock(&dev->mode_config.fb_lock);
+	fb = __drm_framebuffer_lookup(dev, *id);
+	if (!fb)
+		goto fail_lookup;
 
 	list_for_each_entry(fbl, &file_priv->fbs, filp_head)
 		if (fb == fbl)
 			found = 1;
+	if (!found)
+		goto fail_lookup;
 
-	if (!found) {
-		ret = -EINVAL;
-		goto out;
-	}
+	/* Mark fb as reaped, we still have a ref from fpriv->fbs. */
+	__drm_framebuffer_unregister(dev, fb);
+
+	list_del_init(&fb->filp_head);
+	mutex_unlock(&dev->mode_config.fb_lock);
+	mutex_unlock(&file_priv->fbs_lock);
 
 	drm_framebuffer_remove(fb);
 
-out:
-	mutex_unlock(&dev->mode_config.mutex);
-	return ret;
+	return 0;
+
+fail_lookup:
+	mutex_unlock(&dev->mode_config.fb_lock);
+	mutex_unlock(&file_priv->fbs_lock);
+
+	return -EINVAL;
 }
 
 /**
  * drm_mode_getfb - get FB info
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Lookup the FB given its ID and return info about it.
  *
@@ -2444,30 +2603,28 @@ int drm_mode_getfb(struct drm_device *dev,
 		   void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_fb_cmd *r = data;
-	struct drm_mode_object *obj;
 	struct drm_framebuffer *fb;
-	int ret = 0;
+	int ret;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
-	obj = drm_mode_object_find(dev, r->fb_id, DRM_MODE_OBJECT_FB);
-	if (!obj) {
-		ret = -EINVAL;
-		goto out;
-	}
-	fb = obj_to_fb(obj);
+	fb = drm_framebuffer_lookup(dev, r->fb_id);
+	if (!fb)
+		return -EINVAL;
 
 	r->height = fb->height;
 	r->width = fb->width;
 	r->depth = fb->depth;
 	r->bpp = fb->bits_per_pixel;
 	r->pitch = fb->pitches[0];
-	fb->funcs->create_handle(fb, file_priv, &r->handle);
+	if (fb->funcs->create_handle)
+		ret = fb->funcs->create_handle(fb, file_priv, &r->handle);
+	else
+		ret = -ENODEV;
+
+	drm_framebuffer_unreference(fb);
 
-out:
-	mutex_unlock(&dev->mode_config.mutex);
 	return ret;
 }
 
@@ -2477,7 +2634,6 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev,
 	struct drm_clip_rect __user *clips_ptr;
 	struct drm_clip_rect *clips = NULL;
 	struct drm_mode_fb_dirty_cmd *r = data;
-	struct drm_mode_object *obj;
 	struct drm_framebuffer *fb;
 	unsigned flags;
 	int num_clips;
@@ -2486,13 +2642,9 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
-	obj = drm_mode_object_find(dev, r->fb_id, DRM_MODE_OBJECT_FB);
-	if (!obj) {
-		ret = -EINVAL;
-		goto out_err1;
-	}
-	fb = obj_to_fb(obj);
+	fb = drm_framebuffer_lookup(dev, r->fb_id);
+	if (!fb)
+		return -EINVAL;
 
 	num_clips = r->num_clips;
 	clips_ptr = (struct drm_clip_rect __user *)(unsigned long)r->clips_ptr;
@@ -2530,27 +2682,26 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev,
 	}
 
 	if (fb->funcs->dirty) {
+		drm_modeset_lock_all(dev);
 		ret = fb->funcs->dirty(fb, file_priv, flags, r->color,
 				       clips, num_clips);
+		drm_modeset_unlock_all(dev);
 	} else {
 		ret = -ENOSYS;
-		goto out_err2;
 	}
 
 out_err2:
 	kfree(clips);
 out_err1:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_framebuffer_unreference(fb);
+
 	return ret;
 }
 
 
 /**
  * drm_fb_release - remove and free the FBs on this file
- * @filp: file * from the ioctl
- *
- * LOCKING:
- * Takes mode config lock.
+ * @priv: drm file for the ioctl
  *
  * Destroy all the FBs associated with @filp.
  *
@@ -2564,11 +2715,20 @@ void drm_fb_release(struct drm_file *priv)
 	struct drm_device *dev = priv->minor->dev;
 	struct drm_framebuffer *fb, *tfb;
 
-	mutex_lock(&dev->mode_config.mutex);
+	mutex_lock(&priv->fbs_lock);
 	list_for_each_entry_safe(fb, tfb, &priv->fbs, filp_head) {
+
+		mutex_lock(&dev->mode_config.fb_lock);
+		/* Mark fb as reaped, we still have a ref from fpriv->fbs. */
+		__drm_framebuffer_unregister(dev, fb);
+		mutex_unlock(&dev->mode_config.fb_lock);
+
+		list_del_init(&fb->filp_head);
+
+		/* This will also drop the fpriv->fbs reference. */
 		drm_framebuffer_remove(fb);
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&priv->fbs_lock);
 }
 
 /**
@@ -2660,10 +2820,9 @@ EXPORT_SYMBOL(drm_mode_detachmode_crtc);
 
 /**
  * drm_fb_attachmode - Attach a user mode to an connector
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * This attaches a user specified mode to an connector.
  * Called by the user via ioctl.
@@ -2684,7 +2843,7 @@ int drm_mode_attachmode_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, mode_cmd->connector_id, DRM_MODE_OBJECT_CONNECTOR);
 	if (!obj) {
@@ -2708,17 +2867,16 @@ int drm_mode_attachmode_ioctl(struct drm_device *dev,
 
 	drm_mode_attachmode(dev, connector, mode);
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
 
 /**
  * drm_fb_detachmode - Detach a user specified mode from an connector
- * @inode: inode from the ioctl
- * @filp: file * from the ioctl
- * @cmd: cmd from ioctl
- * @arg: arg from ioctl
+ * @dev: drm device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: drm file for the ioctl call
  *
  * Called by the user via ioctl.
  *
@@ -2738,7 +2896,7 @@ int drm_mode_detachmode_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, mode_cmd->connector_id, DRM_MODE_OBJECT_CONNECTOR);
 	if (!obj) {
@@ -2755,7 +2913,7 @@ int drm_mode_detachmode_ioctl(struct drm_device *dev,
 
 	ret = drm_mode_detachmode(dev, connector, &mode);
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3001,7 +3159,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, out_resp->prop_id, DRM_MODE_OBJECT_PROPERTY);
 	if (!obj) {
 		ret = -EINVAL;
@@ -3079,7 +3237,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 		out_resp->count_enum_blobs = blob_count;
 	}
 done:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3130,7 +3288,7 @@ int drm_mode_getblob_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, out_resp->blob_id, DRM_MODE_OBJECT_BLOB);
 	if (!obj) {
 		ret = -EINVAL;
@@ -3148,7 +3306,7 @@ int drm_mode_getblob_ioctl(struct drm_device *dev,
 	out_resp->length = blob->length;
 
 done:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3290,7 +3448,7 @@ int drm_mode_obj_get_properties_ioctl(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, arg->obj_id, arg->obj_type);
 	if (!obj) {
@@ -3327,7 +3485,7 @@ int drm_mode_obj_get_properties_ioctl(struct drm_device *dev, void *data,
 	}
 	arg->count_props = props_count;
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3344,7 +3502,7 @@ int drm_mode_obj_set_property_ioctl(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	arg_obj = drm_mode_object_find(dev, arg->obj_id, arg->obj_type);
 	if (!arg_obj)
@@ -3382,7 +3540,7 @@ int drm_mode_obj_set_property_ioctl(struct drm_device *dev, void *data,
 	}
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3444,7 +3602,7 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		ret = -EINVAL;
@@ -3485,7 +3643,7 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 	crtc->funcs->gamma_set(crtc, r_base, g_base, b_base, 0, crtc->gamma_size);
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 
 }
@@ -3503,7 +3661,7 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		ret = -EINVAL;
@@ -3536,7 +3694,7 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 		goto out;
 	}
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -3546,7 +3704,7 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	struct drm_mode_crtc_page_flip *page_flip = data;
 	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
-	struct drm_framebuffer *fb;
+	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
 	struct drm_pending_vblank_event *e = NULL;
 	unsigned long flags;
 	int hdisplay, vdisplay;
@@ -3556,12 +3714,12 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	    page_flip->reserved != 0)
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
 	obj = drm_mode_object_find(dev, page_flip->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj)
-		goto out;
+		return -EINVAL;
 	crtc = obj_to_crtc(obj);
 
+	mutex_lock(&crtc->mutex);
 	if (crtc->fb == NULL) {
 		/* The framebuffer is currently unbound, presumably
 		 * due to a hotplug event, that userspace has not
@@ -3574,10 +3732,9 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	if (crtc->funcs->page_flip == NULL)
 		goto out;
 
-	obj = drm_mode_object_find(dev, page_flip->fb_id, DRM_MODE_OBJECT_FB);
-	if (!obj)
+	fb = drm_framebuffer_lookup(dev, page_flip->fb_id);
+	if (!fb)
 		goto out;
-	fb = obj_to_fb(obj);
 
 	hdisplay = crtc->mode.hdisplay;
 	vdisplay = crtc->mode.vdisplay;
@@ -3623,6 +3780,7 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 			(void (*) (struct drm_pending_event *)) kfree;
 	}
 
+	old_fb = crtc->fb;
 	ret = crtc->funcs->page_flip(crtc, fb, e);
 	if (ret) {
 		if (page_flip->flags & DRM_MODE_PAGE_FLIP_EVENT) {
@@ -3631,10 +3789,27 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 			spin_unlock_irqrestore(&dev->event_lock, flags);
 			kfree(e);
 		}
+		/* Keep the old fb, don't unref it. */
+		old_fb = NULL;
+	} else {
+		/*
+		 * Warn if the driver hasn't properly updated the crtc->fb
+		 * field to reflect that the new framebuffer is now used.
+		 * Failing to do so will screw with the reference counting
+		 * on framebuffers.
+		 */
+		WARN_ON(crtc->fb != fb);
+		/* Unref only the old framebuffer. */
+		fb = NULL;
 	}
 
 out:
-	mutex_unlock(&dev->mode_config.mutex);
+	if (fb)
+		drm_framebuffer_unreference(fb);
+	if (old_fb)
+		drm_framebuffer_unreference(old_fb);
+	mutex_unlock(&crtc->mutex);
+
 	return ret;
 }
 
@@ -3702,6 +3877,7 @@ void drm_fb_get_bpp_depth(uint32_t format, unsigned int *depth,
 			  int *bpp)
 {
 	switch (format) {
+	case DRM_FORMAT_C8:
 	case DRM_FORMAT_RGB332:
 	case DRM_FORMAT_BGR233:
 		*depth = 8;
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 5a3770fbd770..c194f4e680ad 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -29,11 +29,11 @@
  */
 #include <linux/kernel.h>
 #include <linux/slab.h>
+#include <linux/hdmi.h>
 #include <linux/i2c.h>
 #include <linux/module.h>
 #include <drm/drmP.h>
 #include <drm/drm_edid.h>
-#include "drm_edid_modes.h"
 
 #define version_greater(edid, maj, min) \
 	(((edid)->version > (maj)) || \
@@ -87,9 +87,6 @@ static struct edid_quirk {
 	int product_id;
 	u32 quirks;
 } edid_quirk_list[] = {
-	/* ASUS VW222S */
-	{ "ACI", 0x22a2, EDID_QUIRK_FORCE_REDUCED_BLANKING },
-
 	/* Acer AL1706 */
 	{ "ACR", 44358, EDID_QUIRK_PREFER_LARGE_60 },
 	/* Acer F51 */
@@ -130,6 +127,746 @@ static struct edid_quirk {
 	{ "VSC", 5020, EDID_QUIRK_FORCE_REDUCED_BLANKING },
 };
 
+/*
+ * Autogenerated from the DMT spec.
+ * This table is copied from xfree86/modes/xf86EdidModes.c.
+ */
+static const struct drm_display_mode drm_dmt_modes[] = {
+	/* 640x350@85Hz */
+	{ DRM_MODE("640x350", DRM_MODE_TYPE_DRIVER, 31500, 640, 672,
+		   736, 832, 0, 350, 382, 385, 445, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 640x400@85Hz */
+	{ DRM_MODE("640x400", DRM_MODE_TYPE_DRIVER, 31500, 640, 672,
+		   736, 832, 0, 400, 401, 404, 445, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 720x400@85Hz */
+	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 35500, 720, 756,
+		   828, 936, 0, 400, 401, 404, 446, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 640x480@60Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25175, 640, 656,
+		   752, 800, 0, 480, 489, 492, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 640x480@72Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 664,
+		   704, 832, 0, 480, 489, 492, 520, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 640x480@75Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 656,
+		   720, 840, 0, 480, 481, 484, 500, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 640x480@85Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 36000, 640, 696,
+		   752, 832, 0, 480, 481, 484, 509, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 800x600@56Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 36000, 800, 824,
+		   896, 1024, 0, 600, 601, 603, 625, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 800x600@60Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 40000, 800, 840,
+		   968, 1056, 0, 600, 601, 605, 628, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 800x600@72Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 50000, 800, 856,
+		   976, 1040, 0, 600, 637, 643, 666, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 800x600@75Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 49500, 800, 816,
+		   896, 1056, 0, 600, 601, 604, 625, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 800x600@85Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 56250, 800, 832,
+		   896, 1048, 0, 600, 601, 604, 631, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 800x600@120Hz RB */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 73250, 800, 848,
+		   880, 960, 0, 600, 603, 607, 636, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 848x480@60Hz */
+	{ DRM_MODE("848x480", DRM_MODE_TYPE_DRIVER, 33750, 848, 864,
+		   976, 1088, 0, 480, 486, 494, 517, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1024x768@43Hz, interlace */
+	{ DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER, 44900, 1024, 1032,
+		   1208, 1264, 0, 768, 768, 772, 817, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 1024x768@60Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048,
+		   1184, 1344, 0, 768, 771, 777, 806, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1024x768@70Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 75000, 1024, 1048,
+		   1184, 1328, 0, 768, 771, 777, 806, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1024x768@75Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 78750, 1024, 1040,
+		   1136, 1312, 0, 768, 769, 772, 800, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1024x768@85Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 94500, 1024, 1072,
+		   1168, 1376, 0, 768, 769, 772, 808, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1024x768@120Hz RB */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 115500, 1024, 1072,
+		   1104, 1184, 0, 768, 771, 775, 813, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1152x864@75Hz */
+	{ DRM_MODE("1152x864", DRM_MODE_TYPE_DRIVER, 108000, 1152, 1216,
+		   1344, 1600, 0, 864, 865, 868, 900, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x768@60Hz RB */
+	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 68250, 1280, 1328,
+		   1360, 1440, 0, 768, 771, 778, 790, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x768@60Hz */
+	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 79500, 1280, 1344,
+		   1472, 1664, 0, 768, 771, 778, 798, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x768@75Hz */
+	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 102250, 1280, 1360,
+		   1488, 1696, 0, 768, 771, 778, 805, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x768@85Hz */
+	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 117500, 1280, 1360,
+		   1496, 1712, 0, 768, 771, 778, 809, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x768@120Hz RB */
+	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 140250, 1280, 1328,
+		   1360, 1440, 0, 768, 771, 778, 813, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x800@60Hz RB */
+	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 71000, 1280, 1328,
+		   1360, 1440, 0, 800, 803, 809, 823, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x800@60Hz */
+	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 83500, 1280, 1352,
+		   1480, 1680, 0, 800, 803, 809, 831, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x800@75Hz */
+	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 106500, 1280, 1360,
+		   1488, 1696, 0, 800, 803, 809, 838, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x800@85Hz */
+	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 122500, 1280, 1360,
+		   1496, 1712, 0, 800, 803, 809, 843, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x800@120Hz RB */
+	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 146250, 1280, 1328,
+		   1360, 1440, 0, 800, 803, 809, 847, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x960@60Hz */
+	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 108000, 1280, 1376,
+		   1488, 1800, 0, 960, 961, 964, 1000, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x960@85Hz */
+	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1344,
+		   1504, 1728, 0, 960, 961, 964, 1011, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x960@120Hz RB */
+	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 175500, 1280, 1328,
+		   1360, 1440, 0, 960, 963, 967, 1017, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1280x1024@60Hz */
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 108000, 1280, 1328,
+		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x1024@75Hz */
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 135000, 1280, 1296,
+		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x1024@85Hz */
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 157500, 1280, 1344,
+		   1504, 1728, 0, 1024, 1025, 1028, 1072, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1280x1024@120Hz RB */
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 187250, 1280, 1328,
+		   1360, 1440, 0, 1024, 1027, 1034, 1084, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1360x768@60Hz */
+	{ DRM_MODE("1360x768", DRM_MODE_TYPE_DRIVER, 85500, 1360, 1424,
+		   1536, 1792, 0, 768, 771, 777, 795, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1360x768@120Hz RB */
+	{ DRM_MODE("1360x768", DRM_MODE_TYPE_DRIVER, 148250, 1360, 1408,
+		   1440, 1520, 0, 768, 771, 776, 813, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1400x1050@60Hz RB */
+	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 101000, 1400, 1448,
+		   1480, 1560, 0, 1050, 1053, 1057, 1080, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1400x1050@60Hz */
+	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 121750, 1400, 1488,
+		   1632, 1864, 0, 1050, 1053, 1057, 1089, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1400x1050@75Hz */
+	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 156000, 1400, 1504,
+		   1648, 1896, 0, 1050, 1053, 1057, 1099, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1400x1050@85Hz */
+	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 179500, 1400, 1504,
+		   1656, 1912, 0, 1050, 1053, 1057, 1105, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1400x1050@120Hz RB */
+	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 208000, 1400, 1448,
+		   1480, 1560, 0, 1050, 1053, 1057, 1112, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1440x900@60Hz RB */
+	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 88750, 1440, 1488,
+		   1520, 1600, 0, 900, 903, 909, 926, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1440x900@60Hz */
+	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 106500, 1440, 1520,
+		   1672, 1904, 0, 900, 903, 909, 934, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1440x900@75Hz */
+	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 136750, 1440, 1536,
+		   1688, 1936, 0, 900, 903, 909, 942, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1440x900@85Hz */
+	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 157000, 1440, 1544,
+		   1696, 1952, 0, 900, 903, 909, 948, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1440x900@120Hz RB */
+	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 182750, 1440, 1488,
+		   1520, 1600, 0, 900, 903, 909, 953, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1600x1200@60Hz */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 162000, 1600, 1664,
+		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1600x1200@65Hz */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 175500, 1600, 1664,
+		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1600x1200@70Hz */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 189000, 1600, 1664,
+		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1600x1200@75Hz */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 202500, 1600, 1664,
+		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1600x1200@85Hz */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 229500, 1600, 1664,
+		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1600x1200@120Hz RB */
+	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 268250, 1600, 1648,
+		   1680, 1760, 0, 1200, 1203, 1207, 1271, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1680x1050@60Hz RB */
+	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 119000, 1680, 1728,
+		   1760, 1840, 0, 1050, 1053, 1059, 1080, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1680x1050@60Hz */
+	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 146250, 1680, 1784,
+		   1960, 2240, 0, 1050, 1053, 1059, 1089, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1680x1050@75Hz */
+	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 187000, 1680, 1800,
+		   1976, 2272, 0, 1050, 1053, 1059, 1099, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1680x1050@85Hz */
+	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 214750, 1680, 1808,
+		   1984, 2288, 0, 1050, 1053, 1059, 1105, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1680x1050@120Hz RB */
+	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 245500, 1680, 1728,
+		   1760, 1840, 0, 1050, 1053, 1059, 1112, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1792x1344@60Hz */
+	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 204750, 1792, 1920,
+		   2120, 2448, 0, 1344, 1345, 1348, 1394, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1792x1344@75Hz */
+	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 261000, 1792, 1888,
+		   2104, 2456, 0, 1344, 1345, 1348, 1417, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1792x1344@120Hz RB */
+	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 333250, 1792, 1840,
+		   1872, 1952, 0, 1344, 1347, 1351, 1423, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1856x1392@60Hz */
+	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 218250, 1856, 1952,
+		   2176, 2528, 0, 1392, 1393, 1396, 1439, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1856x1392@75Hz */
+	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 288000, 1856, 1984,
+		   2208, 2560, 0, 1392, 1395, 1399, 1500, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1856x1392@120Hz RB */
+	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 356500, 1856, 1904,
+		   1936, 2016, 0, 1392, 1395, 1399, 1474, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1920x1200@60Hz RB */
+	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 154000, 1920, 1968,
+		   2000, 2080, 0, 1200, 1203, 1209, 1235, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1920x1200@60Hz */
+	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 193250, 1920, 2056,
+		   2256, 2592, 0, 1200, 1203, 1209, 1245, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1920x1200@75Hz */
+	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 245250, 1920, 2056,
+		   2264, 2608, 0, 1200, 1203, 1209, 1255, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1920x1200@85Hz */
+	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 281250, 1920, 2064,
+		   2272, 2624, 0, 1200, 1203, 1209, 1262, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1920x1200@120Hz RB */
+	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 317000, 1920, 1968,
+		   2000, 2080, 0, 1200, 1203, 1209, 1271, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 1920x1440@60Hz */
+	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 234000, 1920, 2048,
+		   2256, 2600, 0, 1440, 1441, 1444, 1500, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1920x1440@75Hz */
+	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2064,
+		   2288, 2640, 0, 1440, 1441, 1444, 1500, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 1920x1440@120Hz RB */
+	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 380500, 1920, 1968,
+		   2000, 2080, 0, 1440, 1443, 1447, 1525, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 2560x1600@60Hz RB */
+	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 268500, 2560, 2608,
+		   2640, 2720, 0, 1600, 1603, 1609, 1646, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 2560x1600@60Hz */
+	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 348500, 2560, 2752,
+		   3032, 3504, 0, 1600, 1603, 1609, 1658, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 2560x1600@75HZ */
+	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 443250, 2560, 2768,
+		   3048, 3536, 0, 1600, 1603, 1609, 1672, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 2560x1600@85HZ */
+	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 505250, 2560, 2768,
+		   3048, 3536, 0, 1600, 1603, 1609, 1682, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 2560x1600@120Hz RB */
+	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 552750, 2560, 2608,
+		   2640, 2720, 0, 1600, 1603, 1609, 1694, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
+};
+
+static const struct drm_display_mode edid_est_modes[] = {
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 40000, 800, 840,
+		   968, 1056, 0, 600, 601, 605, 628, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@60Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 36000, 800, 824,
+		   896, 1024, 0, 600, 601, 603,  625, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@56Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 656,
+		   720, 840, 0, 480, 481, 484, 500, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@75Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 664,
+		   704,  832, 0, 480, 489, 491, 520, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@72Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 30240, 640, 704,
+		   768,  864, 0, 480, 483, 486, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@67Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25200, 640, 656,
+		   752, 800, 0, 480, 490, 492, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@60Hz */
+	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 35500, 720, 738,
+		   846, 900, 0, 400, 421, 423,  449, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 720x400@88Hz */
+	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 28320, 720, 738,
+		   846,  900, 0, 400, 412, 414, 449, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 720x400@70Hz */
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 135000, 1280, 1296,
+		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1280x1024@75Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 78800, 1024, 1040,
+		   1136, 1312, 0,  768, 769, 772, 800, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1024x768@75Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 75000, 1024, 1048,
+		   1184, 1328, 0,  768, 771, 777, 806, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 1024x768@70Hz */
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048,
+		   1184, 1344, 0,  768, 771, 777, 806, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 1024x768@60Hz */
+	{ DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER,44900, 1024, 1032,
+		   1208, 1264, 0, 768, 768, 776, 817, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 1024x768@43Hz */
+	{ DRM_MODE("832x624", DRM_MODE_TYPE_DRIVER, 57284, 832, 864,
+		   928, 1152, 0, 624, 625, 628, 667, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 832x624@75Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 49500, 800, 816,
+		   896, 1056, 0, 600, 601, 604,  625, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@75Hz */
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 50000, 800, 856,
+		   976, 1040, 0, 600, 637, 643, 666, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@72Hz */
+	{ DRM_MODE("1152x864", DRM_MODE_TYPE_DRIVER, 108000, 1152, 1216,
+		   1344, 1600, 0,  864, 865, 868, 900, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1152x864@75Hz */
+};
+
+struct minimode {
+	short w;
+	short h;
+	short r;
+	short rb;
+};
+
+static const struct minimode est3_modes[] = {
+	/* byte 6 */
+	{ 640, 350, 85, 0 },
+	{ 640, 400, 85, 0 },
+	{ 720, 400, 85, 0 },
+	{ 640, 480, 85, 0 },
+	{ 848, 480, 60, 0 },
+	{ 800, 600, 85, 0 },
+	{ 1024, 768, 85, 0 },
+	{ 1152, 864, 75, 0 },
+	/* byte 7 */
+	{ 1280, 768, 60, 1 },
+	{ 1280, 768, 60, 0 },
+	{ 1280, 768, 75, 0 },
+	{ 1280, 768, 85, 0 },
+	{ 1280, 960, 60, 0 },
+	{ 1280, 960, 85, 0 },
+	{ 1280, 1024, 60, 0 },
+	{ 1280, 1024, 85, 0 },
+	/* byte 8 */
+	{ 1360, 768, 60, 0 },
+	{ 1440, 900, 60, 1 },
+	{ 1440, 900, 60, 0 },
+	{ 1440, 900, 75, 0 },
+	{ 1440, 900, 85, 0 },
+	{ 1400, 1050, 60, 1 },
+	{ 1400, 1050, 60, 0 },
+	{ 1400, 1050, 75, 0 },
+	/* byte 9 */
+	{ 1400, 1050, 85, 0 },
+	{ 1680, 1050, 60, 1 },
+	{ 1680, 1050, 60, 0 },
+	{ 1680, 1050, 75, 0 },
+	{ 1680, 1050, 85, 0 },
+	{ 1600, 1200, 60, 0 },
+	{ 1600, 1200, 65, 0 },
+	{ 1600, 1200, 70, 0 },
+	/* byte 10 */
+	{ 1600, 1200, 75, 0 },
+	{ 1600, 1200, 85, 0 },
+	{ 1792, 1344, 60, 0 },
+	{ 1792, 1344, 85, 0 },
+	{ 1856, 1392, 60, 0 },
+	{ 1856, 1392, 75, 0 },
+	{ 1920, 1200, 60, 1 },
+	{ 1920, 1200, 60, 0 },
+	/* byte 11 */
+	{ 1920, 1200, 75, 0 },
+	{ 1920, 1200, 85, 0 },
+	{ 1920, 1440, 60, 0 },
+	{ 1920, 1440, 75, 0 },
+};
+
+static const struct minimode extra_modes[] = {
+	{ 1024, 576,  60, 0 },
+	{ 1366, 768,  60, 0 },
+	{ 1600, 900,  60, 0 },
+	{ 1680, 945,  60, 0 },
+	{ 1920, 1080, 60, 0 },
+	{ 2048, 1152, 60, 0 },
+	{ 2048, 1536, 60, 0 },
+};
+
+/*
+ * Probably taken from CEA-861 spec.
+ * This table is converted from xorg's hw/xfree86/modes/xf86EdidModes.c.
+ */
+static const struct drm_display_mode edid_cea_modes[] = {
+	/* 1 - 640x480@60Hz */
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25175, 640, 656,
+		   752, 800, 0, 480, 490, 492, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 2 - 720x480@60Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 27000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 3 - 720x480@60Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 27000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 4 - 1280x720@60Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 1390,
+		   1430, 1650, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 5 - 1920x1080i@60Hz */
+	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008,
+		   2052, 2200, 0, 1080, 1084, 1094, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 6 - 1440x480i@60Hz */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 7 - 1440x480i@60Hz */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 8 - 1440x240@60Hz */
+	{ DRM_MODE("1440x240", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
+		   1602, 1716, 0, 240, 244, 247, 262, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 9 - 1440x240@60Hz */
+	{ DRM_MODE("1440x240", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
+		   1602, 1716, 0, 240, 244, 247, 262, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 10 - 2880x480i@60Hz */
+	{ DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
+		   3204, 3432, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 11 - 2880x480i@60Hz */
+	{ DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
+		   3204, 3432, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 12 - 2880x240@60Hz */
+	{ DRM_MODE("2880x240", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
+		   3204, 3432, 0, 240, 244, 247, 262, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 13 - 2880x240@60Hz */
+	{ DRM_MODE("2880x240", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
+		   3204, 3432, 0, 240, 244, 247, 262, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 14 - 1440x480@60Hz */
+	{ DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1472,
+		   1596, 1716, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 15 - 1440x480@60Hz */
+	{ DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1472,
+		   1596, 1716, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 16 - 1920x1080@60Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008,
+		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 17 - 720x576@50Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 27000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 18 - 720x576@50Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 27000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 19 - 1280x720@50Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 1720,
+		   1760, 1980, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 20 - 1920x1080i@50Hz */
+	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448,
+		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 21 - 1440x576i@50Hz */
+	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 22 - 1440x576i@50Hz */
+	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 23 - 1440x288@50Hz */
+	{ DRM_MODE("1440x288", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
+		   1590, 1728, 0, 288, 290, 293, 312, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 24 - 1440x288@50Hz */
+	{ DRM_MODE("1440x288", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
+		   1590, 1728, 0, 288, 290, 293, 312, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 25 - 2880x576i@50Hz */
+	{ DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
+		   3180, 3456, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 26 - 2880x576i@50Hz */
+	{ DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
+		   3180, 3456, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 27 - 2880x288@50Hz */
+	{ DRM_MODE("2880x288", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
+		   3180, 3456, 0, 288, 290, 293, 312, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 28 - 2880x288@50Hz */
+	{ DRM_MODE("2880x288", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
+		   3180, 3456, 0, 288, 290, 293, 312, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 29 - 1440x576@50Hz */
+	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
+		   1592, 1728, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 30 - 1440x576@50Hz */
+	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
+		   1592, 1728, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 31 - 1920x1080@50Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448,
+		   2492, 2640, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 32 - 1920x1080@24Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2558,
+		   2602, 2750, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 33 - 1920x1080@25Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448,
+		   2492, 2640, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 34 - 1920x1080@30Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008,
+		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 35 - 2880x480@60Hz */
+	{ DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2944,
+		   3192, 3432, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 36 - 2880x480@60Hz */
+	{ DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2944,
+		   3192, 3432, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 37 - 2880x576@50Hz */
+	{ DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2928,
+		   3184, 3456, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 38 - 2880x576@50Hz */
+	{ DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2928,
+		   3184, 3456, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 39 - 1920x1080i@50Hz */
+	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 72000, 1920, 1952,
+		   2120, 2304, 0, 1080, 1126, 1136, 1250, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 40 - 1920x1080i@100Hz */
+	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448,
+		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 41 - 1280x720@100Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1720,
+		   1760, 1980, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 42 - 720x576@100Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 54000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 43 - 720x576@100Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 54000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 44 - 1440x576i@100Hz */
+	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 45 - 1440x576i@100Hz */
+	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_DBLCLK) },
+	/* 46 - 1920x1080i@120Hz */
+	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008,
+		   2052, 2200, 0, 1080, 1084, 1094, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
+			DRM_MODE_FLAG_INTERLACE) },
+	/* 47 - 1280x720@120Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1390,
+		   1430, 1650, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 48 - 720x480@120Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 54000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 49 - 720x480@120Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 54000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 50 - 1440x480i@120Hz */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 51 - 1440x480i@120Hz */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 52 - 720x576@200Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 108000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 53 - 720x576@200Hz */
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 108000, 720, 732,
+		   796, 864, 0, 576, 581, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 54 - 1440x576i@200Hz */
+	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 55 - 1440x576i@200Hz */
+	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464,
+		   1590, 1728, 0, 576, 580, 586, 625, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 56 - 720x480@240Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 108000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 57 - 720x480@240Hz */
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 108000, 720, 736,
+		   798, 858, 0, 480, 489, 495, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
+	/* 58 - 1440x480i@240 */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 59 - 1440x480i@240 */
+	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478,
+		   1602, 1716, 0, 480, 488, 494, 525, 0,
+		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
+			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
+	/* 60 - 1280x720@24Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 59400, 1280, 3040,
+		   3080, 3300, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 61 - 1280x720@25Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 3700,
+		   3740, 3960, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 62 - 1280x720@30Hz */
+	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 3040,
+		   3080, 3300, 0, 720, 725, 730, 750, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 63 - 1920x1080@120Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2008,
+		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	/* 64 - 1920x1080@100Hz */
+	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2448,
+		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+};
+
 /*** DDC fetch and block validation ***/
 
 static const u8 edid_header[] = {
@@ -357,10 +1094,14 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 				break;
 			}
 		}
-		if (i == 4)
+
+		if (i == 4 && print_bad_edid) {
 			dev_warn(connector->dev->dev,
 			 "%s: Ignoring invalid EDID block %d.\n",
 			 drm_get_connector_name(connector), j);
+
+			connector->bad_edid_counter++;
+		}
 	}
 
 	if (valid_extensions != block[0x7e]) {
@@ -541,7 +1282,7 @@ struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev,
 {
 	int i;
 
-	for (i = 0; i < drm_num_dmt_modes; i++) {
+	for (i = 0; i < ARRAY_SIZE(drm_dmt_modes); i++) {
 		const struct drm_display_mode *ptr = &drm_dmt_modes[i];
 		if (hsize != ptr->hdisplay)
 			continue;
@@ -1082,7 +1823,7 @@ drm_dmt_modes_for_range(struct drm_connector *connector, struct edid *edid,
 	struct drm_display_mode *newmode;
 	struct drm_device *dev = connector->dev;
 
-	for (i = 0; i < drm_num_dmt_modes; i++) {
+	for (i = 0; i < ARRAY_SIZE(drm_dmt_modes); i++) {
 		if (mode_in_range(drm_dmt_modes + i, edid, timing) &&
 		    valid_inferred_mode(connector, drm_dmt_modes + i)) {
 			newmode = drm_mode_duplicate(dev, &drm_dmt_modes[i]);
@@ -1117,7 +1858,7 @@ drm_gtf_modes_for_range(struct drm_connector *connector, struct edid *edid,
 	struct drm_display_mode *newmode;
 	struct drm_device *dev = connector->dev;
 
-	for (i = 0; i < num_extra_modes; i++) {
+	for (i = 0; i < ARRAY_SIZE(extra_modes); i++) {
 		const struct minimode *m = &extra_modes[i];
 		newmode = drm_gtf_mode(dev, m->w, m->h, m->r, 0, 0);
 		if (!newmode)
@@ -1146,7 +1887,7 @@ drm_cvt_modes_for_range(struct drm_connector *connector, struct edid *edid,
 	struct drm_device *dev = connector->dev;
 	bool rb = drm_monitor_supports_rb(edid);
 
-	for (i = 0; i < num_extra_modes; i++) {
+	for (i = 0; i < ARRAY_SIZE(extra_modes); i++) {
 		const struct minimode *m = &extra_modes[i];
 		newmode = drm_cvt_mode(dev, m->w, m->h, m->r, rb, 0, 0);
 		if (!newmode)
@@ -1483,9 +2224,11 @@ add_detailed_modes(struct drm_connector *connector, struct edid *edid,
 #define VIDEO_BLOCK     0x02
 #define VENDOR_BLOCK    0x03
 #define SPEAKER_BLOCK	0x04
+#define VIDEO_CAPABILITY_BLOCK	0x07
 #define EDID_BASIC_AUDIO	(1 << 6)
 #define EDID_CEA_YCRCB444	(1 << 5)
 #define EDID_CEA_YCRCB422	(1 << 4)
+#define EDID_CEA_VCDB_QS	(1 << 6)
 
 /**
  * Search EDID for CEA extension block.
@@ -1513,16 +2256,19 @@ u8 *drm_find_cea_extension(struct edid *edid)
 }
 EXPORT_SYMBOL(drm_find_cea_extension);
 
-/*
- * Looks for a CEA mode matching given drm_display_mode.
- * Returns its CEA Video ID code, or 0 if not found.
+/**
+ * drm_match_cea_mode - look for a CEA mode matching given mode
+ * @to_match: display mode
+ *
+ * Returns the CEA Video ID (VIC) of the mode or 0 if it isn't a CEA-861
+ * mode.
  */
-u8 drm_match_cea_mode(struct drm_display_mode *to_match)
+u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
 {
 	struct drm_display_mode *cea_mode;
 	u8 mode;
 
-	for (mode = 0; mode < drm_num_cea_modes; mode++) {
+	for (mode = 0; mode < ARRAY_SIZE(edid_cea_modes); mode++) {
 		cea_mode = (struct drm_display_mode *)&edid_cea_modes[mode];
 
 		if (drm_mode_equal(to_match, cea_mode))
@@ -1542,7 +2288,7 @@ do_cea_modes (struct drm_connector *connector, u8 *db, u8 len)
 
 	for (mode = db; mode < db + len; mode++) {
 		cea_mode = (*mode & 127) - 1; /* CEA modes are numbered 1..127 */
-		if (cea_mode < drm_num_cea_modes) {
+		if (cea_mode < ARRAY_SIZE(edid_cea_modes)) {
 			struct drm_display_mode *newmode;
 			newmode = drm_mode_duplicate(dev,
 						     &edid_cea_modes[cea_mode]);
@@ -1902,6 +2648,37 @@ end:
 EXPORT_SYMBOL(drm_detect_monitor_audio);
 
 /**
+ * drm_rgb_quant_range_selectable - is RGB quantization range selectable?
+ *
+ * Check whether the monitor reports the RGB quantization range selection
+ * as supported. The AVI infoframe can then be used to inform the monitor
+ * which quantization range (full or limited) is used.
+ */
+bool drm_rgb_quant_range_selectable(struct edid *edid)
+{
+	u8 *edid_ext;
+	int i, start, end;
+
+	edid_ext = drm_find_cea_extension(edid);
+	if (!edid_ext)
+		return false;
+
+	if (cea_db_offsets(edid_ext, &start, &end))
+		return false;
+
+	for_each_cea_db(edid_ext, i, start, end) {
+		if (cea_db_tag(&edid_ext[i]) == VIDEO_CAPABILITY_BLOCK &&
+		    cea_db_payload_len(&edid_ext[i]) == 2) {
+			DRM_DEBUG_KMS("CEA VCDB 0x%02x\n", edid_ext[i + 2]);
+			return edid_ext[i + 2] & EDID_CEA_VCDB_QS;
+		}
+	}
+
+	return false;
+}
+EXPORT_SYMBOL(drm_rgb_quant_range_selectable);
+
+/**
  * drm_add_display_info - pull display info out if present
  * @edid: EDID data
  * @info: display info (attached to connector)
@@ -2020,7 +2797,8 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 	num_modes += add_cvt_modes(connector, edid);
 	num_modes += add_standard_modes(connector, edid);
 	num_modes += add_established_modes(connector, edid);
-	num_modes += add_inferred_modes(connector, edid);
+	if (edid->features & DRM_EDID_FEATURE_DEFAULT_GTF)
+		num_modes += add_inferred_modes(connector, edid);
 	num_modes += add_cea_modes(connector, edid);
 
 	if (quirks & (EDID_QUIRK_PREFER_LARGE_60 | EDID_QUIRK_PREFER_LARGE_75))
@@ -2081,20 +2859,33 @@ int drm_add_modes_noedid(struct drm_connector *connector,
 EXPORT_SYMBOL(drm_add_modes_noedid);
 
 /**
- * drm_mode_cea_vic - return the CEA-861 VIC of a given mode
- * @mode: mode
+ * drm_hdmi_avi_infoframe_from_display_mode() - fill an HDMI AVI infoframe with
+ *                                              data from a DRM display mode
+ * @frame: HDMI AVI infoframe
+ * @mode: DRM display mode
  *
- * RETURNS:
- * The VIC number, 0 in case it's not a CEA-861 mode.
+ * Returns 0 on success or a negative error code on failure.
  */
-uint8_t drm_mode_cea_vic(const struct drm_display_mode *mode)
+int
+drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
+					 const struct drm_display_mode *mode)
 {
-	uint8_t i;
+	int err;
+
+	if (!frame || !mode)
+		return -EINVAL;
+
+	err = hdmi_avi_infoframe_init(frame);
+	if (err < 0)
+		return err;
+
+	frame->video_code = drm_match_cea_mode(mode);
+	if (!frame->video_code)
+		return 0;
 
-	for (i = 0; i < drm_num_cea_modes; i++)
-		if (drm_mode_equal(mode, &edid_cea_modes[i]))
-			return i + 1;
+	frame->picture_aspect = HDMI_PICTURE_ASPECT_NONE;
+	frame->active_aspect = HDMI_ACTIVE_ASPECT_PICTURE;
 
 	return 0;
 }
-EXPORT_SYMBOL(drm_mode_cea_vic);
+EXPORT_SYMBOL(drm_hdmi_avi_infoframe_from_display_mode);
diff --git a/drivers/gpu/drm/drm_edid_modes.h b/drivers/gpu/drm/drm_edid_modes.h
deleted file mode 100644
index 5dbf7d2557b4..000000000000
--- a/drivers/gpu/drm/drm_edid_modes.h
+++ /dev/null
@@ -1,774 +0,0 @@
-/*
- * Copyright (c) 2007-2008 Intel Corporation
- *   Jesse Barnes <jesse.barnes@intel.com>
- * Copyright 2010 Red Hat, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sub license,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#include <linux/kernel.h>
-#include <drm/drmP.h>
-#include <drm/drm_edid.h>
-
-/*
- * Autogenerated from the DMT spec.
- * This table is copied from xfree86/modes/xf86EdidModes.c.
- */
-static const struct drm_display_mode drm_dmt_modes[] = {
-	/* 640x350@85Hz */
-	{ DRM_MODE("640x350", DRM_MODE_TYPE_DRIVER, 31500, 640, 672,
-		   736, 832, 0, 350, 382, 385, 445, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 640x400@85Hz */
-	{ DRM_MODE("640x400", DRM_MODE_TYPE_DRIVER, 31500, 640, 672,
-		   736, 832, 0, 400, 401, 404, 445, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 720x400@85Hz */
-	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 35500, 720, 756,
-		   828, 936, 0, 400, 401, 404, 446, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 640x480@60Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25175, 640, 656,
-		   752, 800, 0, 480, 489, 492, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 640x480@72Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 664,
-		   704, 832, 0, 480, 489, 492, 520, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 640x480@75Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 656,
-		   720, 840, 0, 480, 481, 484, 500, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 640x480@85Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 36000, 640, 696,
-		   752, 832, 0, 480, 481, 484, 509, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 800x600@56Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 36000, 800, 824,
-		   896, 1024, 0, 600, 601, 603, 625, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 800x600@60Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 40000, 800, 840,
-		   968, 1056, 0, 600, 601, 605, 628, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 800x600@72Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 50000, 800, 856,
-		   976, 1040, 0, 600, 637, 643, 666, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 800x600@75Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 49500, 800, 816,
-		   896, 1056, 0, 600, 601, 604, 625, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 800x600@85Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 56250, 800, 832,
-		   896, 1048, 0, 600, 601, 604, 631, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 800x600@120Hz RB */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 73250, 800, 848,
-		   880, 960, 0, 600, 603, 607, 636, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 848x480@60Hz */
-	{ DRM_MODE("848x480", DRM_MODE_TYPE_DRIVER, 33750, 848, 864,
-		   976, 1088, 0, 480, 486, 494, 517, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1024x768@43Hz, interlace */
-	{ DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER, 44900, 1024, 1032,
-		   1208, 1264, 0, 768, 768, 772, 817, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 1024x768@60Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048,
-		   1184, 1344, 0, 768, 771, 777, 806, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1024x768@70Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 75000, 1024, 1048,
-		   1184, 1328, 0, 768, 771, 777, 806, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1024x768@75Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 78750, 1024, 1040,
-		   1136, 1312, 0, 768, 769, 772, 800, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1024x768@85Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 94500, 1024, 1072,
-		   1168, 1376, 0, 768, 769, 772, 808, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1024x768@120Hz RB */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 115500, 1024, 1072,
-		   1104, 1184, 0, 768, 771, 775, 813, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1152x864@75Hz */
-	{ DRM_MODE("1152x864", DRM_MODE_TYPE_DRIVER, 108000, 1152, 1216,
-		   1344, 1600, 0, 864, 865, 868, 900, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x768@60Hz RB */
-	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 68250, 1280, 1328,
-		   1360, 1440, 0, 768, 771, 778, 790, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x768@60Hz */
-	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 79500, 1280, 1344,
-		   1472, 1664, 0, 768, 771, 778, 798, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x768@75Hz */
-	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 102250, 1280, 1360,
-		   1488, 1696, 0, 768, 771, 778, 805, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x768@85Hz */
-	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 117500, 1280, 1360,
-		   1496, 1712, 0, 768, 771, 778, 809, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x768@120Hz RB */
-	{ DRM_MODE("1280x768", DRM_MODE_TYPE_DRIVER, 140250, 1280, 1328,
-		   1360, 1440, 0, 768, 771, 778, 813, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x800@60Hz RB */
-	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 71000, 1280, 1328,
-		   1360, 1440, 0, 800, 803, 809, 823, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x800@60Hz */
-	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 83500, 1280, 1352,
-		   1480, 1680, 0, 800, 803, 809, 831, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x800@75Hz */
-	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 106500, 1280, 1360,
-		   1488, 1696, 0, 800, 803, 809, 838, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x800@85Hz */
-	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 122500, 1280, 1360,
-		   1496, 1712, 0, 800, 803, 809, 843, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x800@120Hz RB */
-	{ DRM_MODE("1280x800", DRM_MODE_TYPE_DRIVER, 146250, 1280, 1328,
-		   1360, 1440, 0, 800, 803, 809, 847, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x960@60Hz */
-	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 108000, 1280, 1376,
-		   1488, 1800, 0, 960, 961, 964, 1000, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x960@85Hz */
-	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1344,
-		   1504, 1728, 0, 960, 961, 964, 1011, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x960@120Hz RB */
-	{ DRM_MODE("1280x960", DRM_MODE_TYPE_DRIVER, 175500, 1280, 1328,
-		   1360, 1440, 0, 960, 963, 967, 1017, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1280x1024@60Hz */
-	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 108000, 1280, 1328,
-		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x1024@75Hz */
-	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 135000, 1280, 1296,
-		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x1024@85Hz */
-	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 157500, 1280, 1344,
-		   1504, 1728, 0, 1024, 1025, 1028, 1072, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1280x1024@120Hz RB */
-	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 187250, 1280, 1328,
-		   1360, 1440, 0, 1024, 1027, 1034, 1084, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1360x768@60Hz */
-	{ DRM_MODE("1360x768", DRM_MODE_TYPE_DRIVER, 85500, 1360, 1424,
-		   1536, 1792, 0, 768, 771, 777, 795, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1360x768@120Hz RB */
-	{ DRM_MODE("1360x768", DRM_MODE_TYPE_DRIVER, 148250, 1360, 1408,
-		   1440, 1520, 0, 768, 771, 776, 813, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1400x1050@60Hz RB */
-	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 101000, 1400, 1448,
-		   1480, 1560, 0, 1050, 1053, 1057, 1080, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1400x1050@60Hz */
-	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 121750, 1400, 1488,
-		   1632, 1864, 0, 1050, 1053, 1057, 1089, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1400x1050@75Hz */
-	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 156000, 1400, 1504,
-		   1648, 1896, 0, 1050, 1053, 1057, 1099, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1400x1050@85Hz */
-	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 179500, 1400, 1504,
-		   1656, 1912, 0, 1050, 1053, 1057, 1105, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1400x1050@120Hz RB */
-	{ DRM_MODE("1400x1050", DRM_MODE_TYPE_DRIVER, 208000, 1400, 1448,
-		   1480, 1560, 0, 1050, 1053, 1057, 1112, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1440x900@60Hz RB */
-	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 88750, 1440, 1488,
-		   1520, 1600, 0, 900, 903, 909, 926, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1440x900@60Hz */
-	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 106500, 1440, 1520,
-		   1672, 1904, 0, 900, 903, 909, 934, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1440x900@75Hz */
-	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 136750, 1440, 1536,
-		   1688, 1936, 0, 900, 903, 909, 942, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1440x900@85Hz */
-	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 157000, 1440, 1544,
-		   1696, 1952, 0, 900, 903, 909, 948, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1440x900@120Hz RB */
-	{ DRM_MODE("1440x900", DRM_MODE_TYPE_DRIVER, 182750, 1440, 1488,
-		   1520, 1600, 0, 900, 903, 909, 953, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1600x1200@60Hz */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 162000, 1600, 1664,
-		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1600x1200@65Hz */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 175500, 1600, 1664,
-		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1600x1200@70Hz */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 189000, 1600, 1664,
-		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1600x1200@75Hz */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 202500, 1600, 1664,
-		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1600x1200@85Hz */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 229500, 1600, 1664,
-		   1856, 2160, 0, 1200, 1201, 1204, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1600x1200@120Hz RB */
-	{ DRM_MODE("1600x1200", DRM_MODE_TYPE_DRIVER, 268250, 1600, 1648,
-		   1680, 1760, 0, 1200, 1203, 1207, 1271, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1680x1050@60Hz RB */
-	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 119000, 1680, 1728,
-		   1760, 1840, 0, 1050, 1053, 1059, 1080, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1680x1050@60Hz */
-	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 146250, 1680, 1784,
-		   1960, 2240, 0, 1050, 1053, 1059, 1089, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1680x1050@75Hz */
-	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 187000, 1680, 1800,
-		   1976, 2272, 0, 1050, 1053, 1059, 1099, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1680x1050@85Hz */
-	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 214750, 1680, 1808,
-		   1984, 2288, 0, 1050, 1053, 1059, 1105, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1680x1050@120Hz RB */
-	{ DRM_MODE("1680x1050", DRM_MODE_TYPE_DRIVER, 245500, 1680, 1728,
-		   1760, 1840, 0, 1050, 1053, 1059, 1112, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1792x1344@60Hz */
-	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 204750, 1792, 1920,
-		   2120, 2448, 0, 1344, 1345, 1348, 1394, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1792x1344@75Hz */
-	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 261000, 1792, 1888,
-		   2104, 2456, 0, 1344, 1345, 1348, 1417, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1792x1344@120Hz RB */
-	{ DRM_MODE("1792x1344", DRM_MODE_TYPE_DRIVER, 333250, 1792, 1840,
-		   1872, 1952, 0, 1344, 1347, 1351, 1423, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1856x1392@60Hz */
-	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 218250, 1856, 1952,
-		   2176, 2528, 0, 1392, 1393, 1396, 1439, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1856x1392@75Hz */
-	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 288000, 1856, 1984,
-		   2208, 2560, 0, 1392, 1395, 1399, 1500, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1856x1392@120Hz RB */
-	{ DRM_MODE("1856x1392", DRM_MODE_TYPE_DRIVER, 356500, 1856, 1904,
-		   1936, 2016, 0, 1392, 1395, 1399, 1474, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1920x1200@60Hz RB */
-	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 154000, 1920, 1968,
-		   2000, 2080, 0, 1200, 1203, 1209, 1235, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1920x1200@60Hz */
-	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 193250, 1920, 2056,
-		   2256, 2592, 0, 1200, 1203, 1209, 1245, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1920x1200@75Hz */
-	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 245250, 1920, 2056,
-		   2264, 2608, 0, 1200, 1203, 1209, 1255, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1920x1200@85Hz */
-	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 281250, 1920, 2064,
-		   2272, 2624, 0, 1200, 1203, 1209, 1262, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1920x1200@120Hz RB */
-	{ DRM_MODE("1920x1200", DRM_MODE_TYPE_DRIVER, 317000, 1920, 1968,
-		   2000, 2080, 0, 1200, 1203, 1209, 1271, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 1920x1440@60Hz */
-	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 234000, 1920, 2048,
-		   2256, 2600, 0, 1440, 1441, 1444, 1500, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1920x1440@75Hz */
-	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2064,
-		   2288, 2640, 0, 1440, 1441, 1444, 1500, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 1920x1440@120Hz RB */
-	{ DRM_MODE("1920x1440", DRM_MODE_TYPE_DRIVER, 380500, 1920, 1968,
-		   2000, 2080, 0, 1440, 1443, 1447, 1525, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 2560x1600@60Hz RB */
-	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 268500, 2560, 2608,
-		   2640, 2720, 0, 1600, 1603, 1609, 1646, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 2560x1600@60Hz */
-	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 348500, 2560, 2752,
-		   3032, 3504, 0, 1600, 1603, 1609, 1658, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 2560x1600@75HZ */
-	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 443250, 2560, 2768,
-		   3048, 3536, 0, 1600, 1603, 1609, 1672, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 2560x1600@85HZ */
-	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 505250, 2560, 2768,
-		   3048, 3536, 0, 1600, 1603, 1609, 1682, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 2560x1600@120Hz RB */
-	{ DRM_MODE("2560x1600", DRM_MODE_TYPE_DRIVER, 552750, 2560, 2608,
-		   2640, 2720, 0, 1600, 1603, 1609, 1694, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
-
-};
-static const int drm_num_dmt_modes =
-	sizeof(drm_dmt_modes) / sizeof(struct drm_display_mode);
-
-static const struct drm_display_mode edid_est_modes[] = {
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 40000, 800, 840,
-		   968, 1056, 0, 600, 601, 605, 628, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@60Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 36000, 800, 824,
-		   896, 1024, 0, 600, 601, 603,  625, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@56Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 656,
-		   720, 840, 0, 480, 481, 484, 500, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@75Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 31500, 640, 664,
-		   704,  832, 0, 480, 489, 491, 520, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@72Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 30240, 640, 704,
-		   768,  864, 0, 480, 483, 486, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@67Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25200, 640, 656,
-		   752, 800, 0, 480, 490, 492, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 640x480@60Hz */
-	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 35500, 720, 738,
-		   846, 900, 0, 400, 421, 423,  449, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 720x400@88Hz */
-	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 28320, 720, 738,
-		   846,  900, 0, 400, 412, 414, 449, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 720x400@70Hz */
-	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 135000, 1280, 1296,
-		   1440, 1688, 0, 1024, 1025, 1028, 1066, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1280x1024@75Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 78800, 1024, 1040,
-		   1136, 1312, 0,  768, 769, 772, 800, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1024x768@75Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 75000, 1024, 1048,
-		   1184, 1328, 0,  768, 771, 777, 806, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 1024x768@70Hz */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048,
-		   1184, 1344, 0,  768, 771, 777, 806, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 1024x768@60Hz */
-	{ DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER,44900, 1024, 1032,
-		   1208, 1264, 0, 768, 768, 776, 817, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 1024x768@43Hz */
-	{ DRM_MODE("832x624", DRM_MODE_TYPE_DRIVER, 57284, 832, 864,
-		   928, 1152, 0, 624, 625, 628, 667, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 832x624@75Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 49500, 800, 816,
-		   896, 1056, 0, 600, 601, 604,  625, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@75Hz */
-	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 50000, 800, 856,
-		   976, 1040, 0, 600, 637, 643, 666, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 800x600@72Hz */
-	{ DRM_MODE("1152x864", DRM_MODE_TYPE_DRIVER, 108000, 1152, 1216,
-		   1344, 1600, 0,  864, 865, 868, 900, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1152x864@75Hz */
-};
-
-struct minimode {
-	short w;
-	short h;
-	short r;
-	short rb;
-};
-
-static const struct minimode est3_modes[] = {
-	/* byte 6 */
-	{ 640, 350, 85, 0 },
-	{ 640, 400, 85, 0 },
-	{ 720, 400, 85, 0 },
-	{ 640, 480, 85, 0 },
-	{ 848, 480, 60, 0 },
-	{ 800, 600, 85, 0 },
-	{ 1024, 768, 85, 0 },
-	{ 1152, 864, 75, 0 },
-	/* byte 7 */
-	{ 1280, 768, 60, 1 },
-	{ 1280, 768, 60, 0 },
-	{ 1280, 768, 75, 0 },
-	{ 1280, 768, 85, 0 },
-	{ 1280, 960, 60, 0 },
-	{ 1280, 960, 85, 0 },
-	{ 1280, 1024, 60, 0 },
-	{ 1280, 1024, 85, 0 },
-	/* byte 8 */
-	{ 1360, 768, 60, 0 },
-	{ 1440, 900, 60, 1 },
-	{ 1440, 900, 60, 0 },
-	{ 1440, 900, 75, 0 },
-	{ 1440, 900, 85, 0 },
-	{ 1400, 1050, 60, 1 },
-	{ 1400, 1050, 60, 0 },
-	{ 1400, 1050, 75, 0 },
-	/* byte 9 */
-	{ 1400, 1050, 85, 0 },
-	{ 1680, 1050, 60, 1 },
-	{ 1680, 1050, 60, 0 },
-	{ 1680, 1050, 75, 0 },
-	{ 1680, 1050, 85, 0 },
-	{ 1600, 1200, 60, 0 },
-	{ 1600, 1200, 65, 0 },
-	{ 1600, 1200, 70, 0 },
-	/* byte 10 */
-	{ 1600, 1200, 75, 0 },
-	{ 1600, 1200, 85, 0 },
-	{ 1792, 1344, 60, 0 },
-	{ 1792, 1344, 85, 0 },
-	{ 1856, 1392, 60, 0 },
-	{ 1856, 1392, 75, 0 },
-	{ 1920, 1200, 60, 1 },
-	{ 1920, 1200, 60, 0 },
-	/* byte 11 */
-	{ 1920, 1200, 75, 0 },
-	{ 1920, 1200, 85, 0 },
-	{ 1920, 1440, 60, 0 },
-	{ 1920, 1440, 75, 0 },
-};
-static const int num_est3_modes = ARRAY_SIZE(est3_modes);
-
-static const struct minimode extra_modes[] = {
-	{ 1024, 576,  60, 0 },
-	{ 1366, 768,  60, 0 },
-	{ 1600, 900,  60, 0 },
-	{ 1680, 945,  60, 0 },
-	{ 1920, 1080, 60, 0 },
-	{ 2048, 1152, 60, 0 },
-	{ 2048, 1536, 60, 0 },
-};
-static const int num_extra_modes = ARRAY_SIZE(extra_modes);
-
-/*
- * Probably taken from CEA-861 spec.
- * This table is converted from xorg's hw/xfree86/modes/xf86EdidModes.c.
- */
-static const struct drm_display_mode edid_cea_modes[] = {
-	/* 1 - 640x480@60Hz */
-	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25175, 640, 656,
-		   752, 800, 0, 480, 490, 492, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 2 - 720x480@60Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 27000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 3 - 720x480@60Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 27000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 4 - 1280x720@60Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 1390,
-		   1430, 1650, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 5 - 1920x1080i@60Hz */
-	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008,
-		   2052, 2200, 0, 1080, 1084, 1094, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 6 - 1440x480i@60Hz */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 7 - 1440x480i@60Hz */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 8 - 1440x240@60Hz */
-	{ DRM_MODE("1440x240", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
-		   1602, 1716, 0, 240, 244, 247, 262, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 9 - 1440x240@60Hz */
-	{ DRM_MODE("1440x240", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478,
-		   1602, 1716, 0, 240, 244, 247, 262, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 10 - 2880x480i@60Hz */
-	{ DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
-		   3204, 3432, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 11 - 2880x480i@60Hz */
-	{ DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
-		   3204, 3432, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 12 - 2880x240@60Hz */
-	{ DRM_MODE("2880x240", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
-		   3204, 3432, 0, 240, 244, 247, 262, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 13 - 2880x240@60Hz */
-	{ DRM_MODE("2880x240", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956,
-		   3204, 3432, 0, 240, 244, 247, 262, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 14 - 1440x480@60Hz */
-	{ DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1472,
-		   1596, 1716, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 15 - 1440x480@60Hz */
-	{ DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1472,
-		   1596, 1716, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 16 - 1920x1080@60Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008,
-		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 17 - 720x576@50Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 27000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 18 - 720x576@50Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 27000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 19 - 1280x720@50Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 1720,
-		   1760, 1980, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 20 - 1920x1080i@50Hz */
-	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448,
-		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 21 - 1440x576i@50Hz */
-	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 22 - 1440x576i@50Hz */
-	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 23 - 1440x288@50Hz */
-	{ DRM_MODE("1440x288", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
-		   1590, 1728, 0, 288, 290, 293, 312, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 24 - 1440x288@50Hz */
-	{ DRM_MODE("1440x288", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464,
-		   1590, 1728, 0, 288, 290, 293, 312, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 25 - 2880x576i@50Hz */
-	{ DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
-		   3180, 3456, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 26 - 2880x576i@50Hz */
-	{ DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
-		   3180, 3456, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 27 - 2880x288@50Hz */
-	{ DRM_MODE("2880x288", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
-		   3180, 3456, 0, 288, 290, 293, 312, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 28 - 2880x288@50Hz */
-	{ DRM_MODE("2880x288", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928,
-		   3180, 3456, 0, 288, 290, 293, 312, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 29 - 1440x576@50Hz */
-	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
-		   1592, 1728, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 30 - 1440x576@50Hz */
-	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
-		   1592, 1728, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 31 - 1920x1080@50Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448,
-		   2492, 2640, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 32 - 1920x1080@24Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2558,
-		   2602, 2750, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 33 - 1920x1080@25Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448,
-		   2492, 2640, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 34 - 1920x1080@30Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008,
-		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 35 - 2880x480@60Hz */
-	{ DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2944,
-		   3192, 3432, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 36 - 2880x480@60Hz */
-	{ DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2944,
-		   3192, 3432, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 37 - 2880x576@50Hz */
-	{ DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2928,
-		   3184, 3456, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 38 - 2880x576@50Hz */
-	{ DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 108000, 2880, 2928,
-		   3184, 3456, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 39 - 1920x1080i@50Hz */
-	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 72000, 1920, 1952,
-		   2120, 2304, 0, 1080, 1126, 1136, 1250, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 40 - 1920x1080i@100Hz */
-	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448,
-		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 41 - 1280x720@100Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1720,
-		   1760, 1980, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 42 - 720x576@100Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 54000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 43 - 720x576@100Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 54000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 44 - 1440x576i@100Hz */
-	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 45 - 1440x576i@100Hz */
-	{ DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_DBLCLK) },
-	/* 46 - 1920x1080i@120Hz */
-	{ DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008,
-		   2052, 2200, 0, 1080, 1084, 1094, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC |
-			DRM_MODE_FLAG_INTERLACE) },
-	/* 47 - 1280x720@120Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 148500, 1280, 1390,
-		   1430, 1650, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 48 - 720x480@120Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 54000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 49 - 720x480@120Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 54000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 50 - 1440x480i@120Hz */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 51 - 1440x480i@120Hz */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 52 - 720x576@200Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 108000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 53 - 720x576@200Hz */
-	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 108000, 720, 732,
-		   796, 864, 0, 576, 581, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 54 - 1440x576i@200Hz */
-	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 55 - 1440x576i@200Hz */
-	{ DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464,
-		   1590, 1728, 0, 576, 580, 586, 625, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 56 - 720x480@240Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 108000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 57 - 720x480@240Hz */
-	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 108000, 720, 736,
-		   798, 858, 0, 480, 489, 495, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) },
-	/* 58 - 1440x480i@240 */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 59 - 1440x480i@240 */
-	{ DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478,
-		   1602, 1716, 0, 480, 488, 494, 525, 0,
-		   DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC |
-			DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) },
-	/* 60 - 1280x720@24Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 59400, 1280, 3040,
-		   3080, 3300, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 61 - 1280x720@25Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 3700,
-		   3740, 3960, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 62 - 1280x720@30Hz */
-	{ DRM_MODE("1280x720", DRM_MODE_TYPE_DRIVER, 74250, 1280, 3040,
-		   3080, 3300, 0, 720, 725, 730, 750, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 63 - 1920x1080@120Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2008,
-		   2052, 2200, 0, 1080, 1084, 1089, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-	/* 64 - 1920x1080@100Hz */
-	{ DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 297000, 1920, 2448,
-		   2492, 2640, 0, 1080, 1084, 1094, 1125, 0,
-		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
-};
-static const int drm_num_cea_modes = ARRAY_SIZE(edid_cea_modes);
diff --git a/drivers/gpu/drm/drm_encoder_slave.c b/drivers/gpu/drm/drm_encoder_slave.c
index 63e733408b6d..48c52f7df4e6 100644
--- a/drivers/gpu/drm/drm_encoder_slave.c
+++ b/drivers/gpu/drm/drm_encoder_slave.c
@@ -123,3 +123,66 @@ void drm_i2c_encoder_destroy(struct drm_encoder *drm_encoder)
 	module_put(module);
 }
 EXPORT_SYMBOL(drm_i2c_encoder_destroy);
+
+/*
+ * Wrapper fxns which can be plugged in to drm_encoder_helper_funcs:
+ */
+
+static inline struct drm_encoder_slave_funcs *
+get_slave_funcs(struct drm_encoder *enc)
+{
+	return to_encoder_slave(enc)->slave_funcs;
+}
+
+void drm_i2c_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+	get_slave_funcs(encoder)->dpms(encoder, mode);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_dpms);
+
+bool drm_i2c_encoder_mode_fixup(struct drm_encoder *encoder,
+		const struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	return get_slave_funcs(encoder)->mode_fixup(encoder, mode, adjusted_mode);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_mode_fixup);
+
+void drm_i2c_encoder_prepare(struct drm_encoder *encoder)
+{
+	drm_i2c_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_prepare);
+
+void drm_i2c_encoder_commit(struct drm_encoder *encoder)
+{
+	drm_i2c_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_commit);
+
+void drm_i2c_encoder_mode_set(struct drm_encoder *encoder,
+		struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	get_slave_funcs(encoder)->mode_set(encoder, mode, adjusted_mode);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_mode_set);
+
+enum drm_connector_status drm_i2c_encoder_detect(struct drm_encoder *encoder,
+	    struct drm_connector *connector)
+{
+	return get_slave_funcs(encoder)->detect(encoder, connector);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_detect);
+
+void drm_i2c_encoder_save(struct drm_encoder *encoder)
+{
+	get_slave_funcs(encoder)->save(encoder);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_save);
+
+void drm_i2c_encoder_restore(struct drm_encoder *encoder)
+{
+	get_slave_funcs(encoder)->restore(encoder);
+}
+EXPORT_SYMBOL(drm_i2c_encoder_restore);
diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
index fd9d0af4d536..0b5af7d0edb1 100644
--- a/drivers/gpu/drm/drm_fb_cma_helper.c
+++ b/drivers/gpu/drm/drm_fb_cma_helper.c
@@ -85,6 +85,11 @@ static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev,
 	if (!fb_cma)
 		return ERR_PTR(-ENOMEM);
 
+	drm_helper_mode_fill_fb_struct(&fb_cma->fb, mode_cmd);
+
+	for (i = 0; i < num_planes; i++)
+		fb_cma->obj[i] = obj[i];
+
 	ret = drm_framebuffer_init(dev, &fb_cma->fb, &drm_fb_cma_funcs);
 	if (ret) {
 		dev_err(dev->dev, "Failed to initalize framebuffer: %d\n", ret);
@@ -92,11 +97,6 @@ static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev,
 		return ERR_PTR(ret);
 	}
 
-	drm_helper_mode_fill_fb_struct(&fb_cma->fb, mode_cmd);
-
-	for (i = 0; i < num_planes; i++)
-		fb_cma->obj[i] = obj[i];
-
 	return fb_cma;
 }
 
@@ -180,6 +180,59 @@ struct drm_gem_cma_object *drm_fb_cma_get_gem_obj(struct drm_framebuffer *fb,
 }
 EXPORT_SYMBOL_GPL(drm_fb_cma_get_gem_obj);
 
+#ifdef CONFIG_DEBUG_FS
+/**
+ * drm_fb_cma_describe() - Helper to dump information about a single
+ * CMA framebuffer object
+ */
+void drm_fb_cma_describe(struct drm_framebuffer *fb, struct seq_file *m)
+{
+	struct drm_fb_cma *fb_cma = to_fb_cma(fb);
+	int i, n = drm_format_num_planes(fb->pixel_format);
+
+	seq_printf(m, "fb: %dx%d@%4.4s\n", fb->width, fb->height,
+			(char *)&fb->pixel_format);
+
+	for (i = 0; i < n; i++) {
+		seq_printf(m, "   %d: offset=%d pitch=%d, obj: ",
+				i, fb->offsets[i], fb->pitches[i]);
+		drm_gem_cma_describe(fb_cma->obj[i], m);
+	}
+}
+EXPORT_SYMBOL_GPL(drm_fb_cma_describe);
+
+/**
+ * drm_fb_cma_debugfs_show() - Helper to list CMA framebuffer objects
+ * in debugfs.
+ */
+int drm_fb_cma_debugfs_show(struct seq_file *m, void *arg)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_framebuffer *fb;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
+	if (ret)
+		return ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret) {
+		mutex_unlock(&dev->mode_config.mutex);
+		return ret;
+	}
+
+	list_for_each_entry(fb, &dev->mode_config.fb_list, head)
+		drm_fb_cma_describe(fb, m);
+
+	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&dev->mode_config.mutex);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(drm_fb_cma_debugfs_show);
+#endif
+
 static struct fb_ops drm_fbdev_cma_ops = {
 	.owner		= THIS_MODULE,
 	.fb_fillrect	= sys_fillrect,
@@ -266,6 +319,7 @@ static int drm_fbdev_cma_create(struct drm_fb_helper *helper,
 	return 0;
 
 err_drm_fb_cma_destroy:
+	drm_framebuffer_unregister_private(fb);
 	drm_fb_cma_destroy(fb);
 err_framebuffer_release:
 	framebuffer_release(fbi);
@@ -274,23 +328,8 @@ err_drm_gem_cma_free_object:
 	return ret;
 }
 
-static int drm_fbdev_cma_probe(struct drm_fb_helper *helper,
-	struct drm_fb_helper_surface_size *sizes)
-{
-	int ret = 0;
-
-	if (!helper->fb) {
-		ret = drm_fbdev_cma_create(helper, sizes);
-		if (ret < 0)
-			return ret;
-		ret = 1;
-	}
-
-	return ret;
-}
-
 static struct drm_fb_helper_funcs drm_fb_cma_helper_funcs = {
-	.fb_probe = drm_fbdev_cma_probe,
+	.fb_probe = drm_fbdev_cma_create,
 };
 
 /**
@@ -332,6 +371,9 @@ struct drm_fbdev_cma *drm_fbdev_cma_init(struct drm_device *dev,
 
 	}
 
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	ret = drm_fb_helper_initial_config(helper, preferred_bpp);
 	if (ret < 0) {
 		dev_err(dev->dev, "Failed to set inital hw configuration.\n");
@@ -370,8 +412,10 @@ void drm_fbdev_cma_fini(struct drm_fbdev_cma *fbdev_cma)
 		framebuffer_release(info);
 	}
 
-	if (fbdev_cma->fb)
+	if (fbdev_cma->fb) {
+		drm_framebuffer_unregister_private(&fbdev_cma->fb->fb);
 		drm_fb_cma_destroy(&fbdev_cma->fb->fb);
+	}
 
 	drm_fb_helper_fini(&fbdev_cma->fb_helper);
 	kfree(fbdev_cma);
@@ -386,8 +430,13 @@ EXPORT_SYMBOL_GPL(drm_fbdev_cma_fini);
  */
 void drm_fbdev_cma_restore_mode(struct drm_fbdev_cma *fbdev_cma)
 {
-	if (fbdev_cma)
+	if (fbdev_cma) {
+		struct drm_device *dev = fbdev_cma->fb_helper.dev;
+
+		drm_modeset_lock_all(dev);
 		drm_fb_helper_restore_fbdev_mode(&fbdev_cma->fb_helper);
+		drm_modeset_unlock_all(dev);
+	}
 }
 EXPORT_SYMBOL_GPL(drm_fbdev_cma_restore_mode);
 
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 954d175bd7fa..59d6b9bf204b 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -52,9 +52,36 @@ static LIST_HEAD(kernel_fb_helper_list);
  * mode setting driver. They can be used mostly independantely from the crtc
  * helper functions used by many drivers to implement the kernel mode setting
  * interfaces.
+ *
+ * Initialization is done as a three-step process with drm_fb_helper_init(),
+ * drm_fb_helper_single_add_all_connectors() and drm_fb_helper_initial_config().
+ * Drivers with fancier requirements than the default beheviour can override the
+ * second step with their own code.  Teardown is done with drm_fb_helper_fini().
+ *
+ * At runtime drivers should restore the fbdev console by calling
+ * drm_fb_helper_restore_fbdev_mode() from their ->lastclose callback. They
+ * should also notify the fb helper code from updates to the output
+ * configuration by calling drm_fb_helper_hotplug_event(). For easier
+ * integration with the output polling code in drm_crtc_helper.c the modeset
+ * code proves a ->output_poll_changed callback.
+ *
+ * All other functions exported by the fb helper library can be used to
+ * implement the fbdev driver interface by the driver.
  */
 
-/* simple single crtc case helper function */
+/**
+ * drm_fb_helper_single_add_all_connectors() - add all connectors to fbdev
+ * 					       emulation helper
+ * @fb_helper: fbdev initialized with drm_fb_helper_init
+ *
+ * This functions adds all the available connectors for use with the given
+ * fb_helper. This is a separate step to allow drivers to freely assign
+ * connectors to the fbdev, e.g. if some are reserved for special purposes or
+ * not adequate to be used for the fbcon.
+ *
+ * Since this is part of the initial setup before the fbdev is published, no
+ * locking is required.
+ */
 int drm_fb_helper_single_add_all_connectors(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
@@ -163,6 +190,10 @@ static void drm_fb_helper_restore_lut_atomic(struct drm_crtc *crtc)
 	crtc->funcs->gamma_set(crtc, r_base, g_base, b_base, 0, crtc->gamma_size);
 }
 
+/**
+ * drm_fb_helper_debug_enter - implementation for ->fb_debug_enter
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_debug_enter(struct fb_info *info)
 {
 	struct drm_fb_helper *helper = info->par;
@@ -208,6 +239,10 @@ static struct drm_framebuffer *drm_mode_config_fb(struct drm_crtc *crtc)
 	return NULL;
 }
 
+/**
+ * drm_fb_helper_debug_leave - implementation for ->fb_debug_leave
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_debug_leave(struct fb_info *info)
 {
 	struct drm_fb_helper *helper = info->par;
@@ -239,13 +274,24 @@ int drm_fb_helper_debug_leave(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_debug_leave);
 
+/**
+ * drm_fb_helper_restore_fbdev_mode - restore fbdev configuration
+ * @fb_helper: fbcon to restore
+ *
+ * This should be called from driver's drm ->lastclose callback
+ * when implementing an fbcon on top of kms using this helper. This ensures that
+ * the user isn't greeted with a black screen when e.g. X dies.
+ */
 bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 {
 	bool error = false;
 	int i, ret;
+
+	drm_warn_on_modeset_not_all_locked(fb_helper->dev);
+
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		struct drm_mode_set *mode_set = &fb_helper->crtc_info[i].mode_set;
-		ret = mode_set->crtc->funcs->set_config(mode_set);
+		ret = drm_mode_set_config_internal(mode_set);
 		if (ret)
 			error = true;
 	}
@@ -253,6 +299,10 @@ bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 }
 EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode);
 
+/*
+ * restore fbcon display for all kms driver's using this helper, used for sysrq
+ * and panic handling.
+ */
 static bool drm_fb_helper_force_kernel_mode(void)
 {
 	bool ret, error = false;
@@ -272,7 +322,7 @@ static bool drm_fb_helper_force_kernel_mode(void)
 	return error;
 }
 
-int drm_fb_helper_panic(struct notifier_block *n, unsigned long ununsed,
+static int drm_fb_helper_panic(struct notifier_block *n, unsigned long ununsed,
 			void *panic_str)
 {
 	/*
@@ -285,30 +335,36 @@ int drm_fb_helper_panic(struct notifier_block *n, unsigned long ununsed,
 	pr_err("panic occurred, switching back to text console\n");
 	return drm_fb_helper_force_kernel_mode();
 }
-EXPORT_SYMBOL(drm_fb_helper_panic);
 
 static struct notifier_block paniced = {
 	.notifier_call = drm_fb_helper_panic,
 };
 
-/**
- * drm_fb_helper_restore - restore the framebuffer console (kernel) config
- *
- * Restore's the kernel's fbcon mode, used for lastclose & panic paths.
- */
-void drm_fb_helper_restore(void)
+static bool drm_fb_helper_is_bound(struct drm_fb_helper *fb_helper)
 {
-	bool ret;
-	ret = drm_fb_helper_force_kernel_mode();
-	if (ret == true)
-		DRM_ERROR("Failed to restore crtc configuration\n");
+	struct drm_device *dev = fb_helper->dev;
+	struct drm_crtc *crtc;
+	int bound = 0, crtcs_bound = 0;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		if (crtc->fb)
+			crtcs_bound++;
+		if (crtc->fb == fb_helper->fb)
+			bound++;
+	}
+
+	if (bound < crtcs_bound)
+		return false;
+	return true;
 }
-EXPORT_SYMBOL(drm_fb_helper_restore);
 
 #ifdef CONFIG_MAGIC_SYSRQ
 static void drm_fb_helper_restore_work_fn(struct work_struct *ignored)
 {
-	drm_fb_helper_restore();
+	bool ret;
+	ret = drm_fb_helper_force_kernel_mode();
+	if (ret == true)
+		DRM_ERROR("Failed to restore crtc configuration\n");
 }
 static DECLARE_WORK(drm_fb_helper_restore_work, drm_fb_helper_restore_work_fn);
 
@@ -335,9 +391,22 @@ static void drm_fb_helper_dpms(struct fb_info *info, int dpms_mode)
 	int i, j;
 
 	/*
+	 * fbdev->blank can be called from irq context in case of a panic.
+	 * Since we already have our own special panic handler which will
+	 * restore the fbdev console mode completely, just bail out early.
+	 */
+	if (oops_in_progress)
+		return;
+
+	/*
 	 * For each CRTC in this fb, turn the connectors on/off.
 	 */
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
+	if (!drm_fb_helper_is_bound(fb_helper)) {
+		drm_modeset_unlock_all(dev);
+		return;
+	}
+
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		crtc = fb_helper->crtc_info[i].mode_set.crtc;
 
@@ -352,9 +421,14 @@ static void drm_fb_helper_dpms(struct fb_info *info, int dpms_mode)
 				dev->mode_config.dpms_property, dpms_mode);
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 }
 
+/**
+ * drm_fb_helper_blank - implementation for ->fb_blank
+ * @blank: desired blanking state
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_blank(int blank, struct fb_info *info)
 {
 	switch (blank) {
@@ -398,6 +472,24 @@ static void drm_fb_helper_crtc_free(struct drm_fb_helper *helper)
 	kfree(helper->crtc_info);
 }
 
+/**
+ * drm_fb_helper_init - initialize a drm_fb_helper structure
+ * @dev: drm device
+ * @fb_helper: driver-allocated fbdev helper structure to initialize
+ * @crtc_count: maximum number of crtcs to support in this fbdev emulation
+ * @max_conn_count: max connector count
+ *
+ * This allocates the structures for the fbdev helper with the given limits.
+ * Note that this won't yet touch the hardware (through the driver interfaces)
+ * nor register the fbdev. This is only done in drm_fb_helper_initial_config()
+ * to allow driver writes more control over the exact init sequence.
+ *
+ * Drivers must set fb_helper->funcs before calling
+ * drm_fb_helper_initial_config().
+ *
+ * RETURNS:
+ * Zero if everything went ok, nonzero otherwise.
+ */
 int drm_fb_helper_init(struct drm_device *dev,
 		       struct drm_fb_helper *fb_helper,
 		       int crtc_count, int max_conn_count)
@@ -526,6 +618,11 @@ static int setcolreg(struct drm_crtc *crtc, u16 red, u16 green,
 	return 0;
 }
 
+/**
+ * drm_fb_helper_setcmap - implementation for ->fb_setcmap
+ * @cmap: cmap to set
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
@@ -565,6 +662,11 @@ int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_setcmap);
 
+/**
+ * drm_fb_helper_check_var - implementation for ->fb_check_var
+ * @var: screeninfo to check
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 			    struct fb_info *info)
 {
@@ -657,13 +759,19 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 }
 EXPORT_SYMBOL(drm_fb_helper_check_var);
 
-/* this will let fbcon do the mode init */
+/**
+ * drm_fb_helper_set_par - implementation for ->fb_set_par
+ * @info: fbdev registered by the helper
+ *
+ * This will let fbcon do the mode init and is called at initialization time by
+ * the fbdev core when registering the driver, and later on through the hotplug
+ * callback.
+ */
 int drm_fb_helper_set_par(struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
 	struct drm_device *dev = fb_helper->dev;
 	struct fb_var_screeninfo *var = &info->var;
-	struct drm_crtc *crtc;
 	int ret;
 	int i;
 
@@ -672,16 +780,15 @@ int drm_fb_helper_set_par(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	for (i = 0; i < fb_helper->crtc_count; i++) {
-		crtc = fb_helper->crtc_info[i].mode_set.crtc;
-		ret = crtc->funcs->set_config(&fb_helper->crtc_info[i].mode_set);
+		ret = drm_mode_set_config_internal(&fb_helper->crtc_info[i].mode_set);
 		if (ret) {
-			mutex_unlock(&dev->mode_config.mutex);
+			drm_modeset_unlock_all(dev);
 			return ret;
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	if (fb_helper->delayed_hotplug) {
 		fb_helper->delayed_hotplug = false;
@@ -691,6 +798,11 @@ int drm_fb_helper_set_par(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_set_par);
 
+/**
+ * drm_fb_helper_pan_display - implementation for ->fb_pan_display
+ * @var: updated screen information
+ * @info: fbdev registered by the helper
+ */
 int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 			      struct fb_info *info)
 {
@@ -701,7 +813,12 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 	int ret = 0;
 	int i;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
+	if (!drm_fb_helper_is_bound(fb_helper)) {
+		drm_modeset_unlock_all(dev);
+		return -EBUSY;
+	}
+
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		crtc = fb_helper->crtc_info[i].mode_set.crtc;
 
@@ -711,22 +828,27 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 		modeset->y = var->yoffset;
 
 		if (modeset->num_connectors) {
-			ret = crtc->funcs->set_config(modeset);
+			ret = drm_mode_set_config_internal(modeset);
 			if (!ret) {
 				info->var.xoffset = var->xoffset;
 				info->var.yoffset = var->yoffset;
 			}
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 EXPORT_SYMBOL(drm_fb_helper_pan_display);
 
-int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
-				  int preferred_bpp)
+/*
+ * Allocates the backing storage and sets up the fbdev info structure through
+ * the ->fb_probe callback and then registers the fbdev and sets up the panic
+ * notifier.
+ */
+static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
+					 int preferred_bpp)
 {
-	int new_fb = 0;
+	int ret = 0;
 	int crtc_count = 0;
 	int i;
 	struct fb_info *info;
@@ -804,27 +926,30 @@ int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 	}
 
 	/* push down into drivers */
-	new_fb = (*fb_helper->funcs->fb_probe)(fb_helper, &sizes);
-	if (new_fb < 0)
-		return new_fb;
+	ret = (*fb_helper->funcs->fb_probe)(fb_helper, &sizes);
+	if (ret < 0)
+		return ret;
 
 	info = fb_helper->fbdev;
 
-	/* set the fb pointer */
+	/*
+	 * Set the fb pointer - usually drm_setup_crtcs does this for hotplug
+	 * events, but at init time drm_setup_crtcs needs to be called before
+	 * the fb is allocated (since we need to figure out the desired size of
+	 * the fb before we can allocate it ...). Hence we need to fix things up
+	 * here again.
+	 */
 	for (i = 0; i < fb_helper->crtc_count; i++)
-		fb_helper->crtc_info[i].mode_set.fb = fb_helper->fb;
+		if (fb_helper->crtc_info[i].mode_set.num_connectors)
+			fb_helper->crtc_info[i].mode_set.fb = fb_helper->fb;
 
-	if (new_fb) {
-		info->var.pixclock = 0;
-		if (register_framebuffer(info) < 0)
-			return -EINVAL;
 
-		dev_info(fb_helper->dev->dev, "fb%d: %s frame buffer device\n",
-				info->node, info->fix.id);
+	info->var.pixclock = 0;
+	if (register_framebuffer(info) < 0)
+		return -EINVAL;
 
-	} else {
-		drm_fb_helper_set_par(info);
-	}
+	dev_info(fb_helper->dev->dev, "fb%d: %s frame buffer device\n",
+			info->node, info->fix.id);
 
 	/* Switch back to kernel console on panic */
 	/* multi card linked list maybe */
@@ -834,13 +959,25 @@ int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 					       &paniced);
 		register_sysrq_key('v', &sysrq_drm_fb_helper_restore_op);
 	}
-	if (new_fb)
-		list_add(&fb_helper->kernel_fb_list, &kernel_fb_helper_list);
+
+	list_add(&fb_helper->kernel_fb_list, &kernel_fb_helper_list);
 
 	return 0;
 }
-EXPORT_SYMBOL(drm_fb_helper_single_fb_probe);
 
+/**
+ * drm_fb_helper_fill_fix - initializes fixed fbdev information
+ * @info: fbdev registered by the helper
+ * @pitch: desired pitch
+ * @depth: desired depth
+ *
+ * Helper to fill in the fixed fbdev information useful for a non-accelerated
+ * fbdev emulations. Drivers which support acceleration methods which impose
+ * additional constraints need to set up their own limits.
+ *
+ * Drivers should call this (or their equivalent setup code) from their
+ * ->fb_probe callback.
+ */
 void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
 			    uint32_t depth)
 {
@@ -861,6 +998,20 @@ void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
 }
 EXPORT_SYMBOL(drm_fb_helper_fill_fix);
 
+/**
+ * drm_fb_helper_fill_var - initalizes variable fbdev information
+ * @info: fbdev instance to set up
+ * @fb_helper: fb helper instance to use as template
+ * @fb_width: desired fb width
+ * @fb_height: desired fb height
+ *
+ * Sets up the variable fbdev metainformation from the given fb helper instance
+ * and the drm framebuffer allocated in fb_helper->fb.
+ *
+ * Drivers should call this (or their equivalent setup code) from their
+ * ->fb_probe callback after having allocated the fbdev backing
+ * storage framebuffer.
+ */
 void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helper,
 			    uint32_t fb_width, uint32_t fb_height)
 {
@@ -1284,6 +1435,7 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper)
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		modeset = &fb_helper->crtc_info[i].mode_set;
 		modeset->num_connectors = 0;
+		modeset->fb = NULL;
 	}
 
 	for (i = 0; i < fb_helper->connector_count; i++) {
@@ -1300,9 +1452,21 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper)
 			modeset->mode = drm_mode_duplicate(dev,
 							   fb_crtc->desired_mode);
 			modeset->connectors[modeset->num_connectors++] = fb_helper->connector_info[i]->connector;
+			modeset->fb = fb_helper->fb;
 		}
 	}
 
+	/* Clear out any old modes if there are no more connected outputs. */
+	for (i = 0; i < fb_helper->crtc_count; i++) {
+		modeset = &fb_helper->crtc_info[i].mode_set;
+		if (modeset->num_connectors == 0) {
+			BUG_ON(modeset->fb);
+			BUG_ON(modeset->num_connectors);
+			if (modeset->mode)
+				drm_mode_destroy(dev, modeset->mode);
+			modeset->mode = NULL;
+		}
+	}
 out:
 	kfree(crtcs);
 	kfree(modes);
@@ -1310,18 +1474,23 @@ out:
 }
 
 /**
- * drm_helper_initial_config - setup a sane initial connector configuration
+ * drm_fb_helper_initial_config - setup a sane initial connector configuration
  * @fb_helper: fb_helper device struct
  * @bpp_sel: bpp value to use for the framebuffer configuration
  *
- * LOCKING:
- * Called at init time by the driver to set up the @fb_helper initial
- * configuration, must take the mode config lock.
- *
  * Scans the CRTCs and connectors and tries to put together an initial setup.
  * At the moment, this is a cloned configuration across all heads with
  * a new framebuffer object as the backing store.
  *
+ * Note that this also registers the fbdev and so allows userspace to call into
+ * the driver through the fbdev interfaces.
+ *
+ * This function will call down into the ->fb_probe callback to let
+ * the driver allocate and initialize the fbdev info structure and the drm
+ * framebuffer used to back the fbdev. drm_fb_helper_fill_var() and
+ * drm_fb_helper_fill_fix() are provided as helpers to setup simple default
+ * values for the fbdev info structure.
+ *
  * RETURNS:
  * Zero if everything went ok, nonzero otherwise.
  */
@@ -1330,9 +1499,6 @@ bool drm_fb_helper_initial_config(struct drm_fb_helper *fb_helper, int bpp_sel)
 	struct drm_device *dev = fb_helper->dev;
 	int count = 0;
 
-	/* disable all the possible outputs/crtcs before entering KMS mode */
-	drm_helper_disable_unused_functions(fb_helper->dev);
-
 	drm_fb_helper_parse_command_line(fb_helper);
 
 	count = drm_fb_helper_probe_connector_modes(fb_helper,
@@ -1355,12 +1521,17 @@ EXPORT_SYMBOL(drm_fb_helper_initial_config);
  *                               probing all the outputs attached to the fb
  * @fb_helper: the drm_fb_helper
  *
- * LOCKING:
- * Called at runtime, must take mode config lock.
- *
  * Scan the connectors attached to the fb_helper and try to put together a
  * setup after *notification of a change in output configuration.
  *
+ * Called at runtime, takes the mode config locks to be able to check/change the
+ * modeset configuration. Must be run from process context (which usually means
+ * either the output polling work or a work item launched from the driver's
+ * hotplug interrupt).
+ *
+ * Note that the driver must ensure that this is only called _after_ the fb has
+ * been fully set up, i.e. after the call to drm_fb_helper_initial_config.
+ *
  * RETURNS:
  * 0 on success and a non-zero error code otherwise.
  */
@@ -1369,23 +1540,14 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
 	struct drm_device *dev = fb_helper->dev;
 	int count = 0;
 	u32 max_width, max_height, bpp_sel;
-	int bound = 0, crtcs_bound = 0;
-	struct drm_crtc *crtc;
 
 	if (!fb_helper->fb)
 		return 0;
 
-	mutex_lock(&dev->mode_config.mutex);
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (crtc->fb)
-			crtcs_bound++;
-		if (crtc->fb == fb_helper->fb)
-			bound++;
-	}
-
-	if (bound < crtcs_bound) {
+	drm_modeset_lock_all(dev);
+	if (!drm_fb_helper_is_bound(fb_helper)) {
 		fb_helper->delayed_hotplug = true;
-		mutex_unlock(&dev->mode_config.mutex);
+		drm_modeset_unlock_all(dev);
 		return 0;
 	}
 	DRM_DEBUG_KMS("\n");
@@ -1397,9 +1559,11 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
 	count = drm_fb_helper_probe_connector_modes(fb_helper, max_width,
 						    max_height);
 	drm_setup_crtcs(fb_helper);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
-	return drm_fb_helper_single_fb_probe(fb_helper, bpp_sel);
+	drm_fb_helper_set_par(fb_helper->fbdev);
+
+	return 0;
 }
 EXPORT_SYMBOL(drm_fb_helper_hotplug_event);
 
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 133b4132983e..13fdcd10a605 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -276,6 +276,7 @@ static int drm_open_helper(struct inode *inode, struct file *filp,
 
 	INIT_LIST_HEAD(&priv->lhead);
 	INIT_LIST_HEAD(&priv->fbs);
+	mutex_init(&priv->fbs_lock);
 	INIT_LIST_HEAD(&priv->event_list);
 	init_waitqueue_head(&priv->event_wait);
 	priv->event_space = 4096; /* set aside 4k for event buffer */
diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c
index 1aa8fee1e865..0a7e011509bd 100644
--- a/drivers/gpu/drm/drm_gem_cma_helper.c
+++ b/drivers/gpu/drm/drm_gem_cma_helper.c
@@ -249,3 +249,24 @@ int drm_gem_cma_dumb_destroy(struct drm_file *file_priv,
 	return drm_gem_handle_delete(file_priv, handle);
 }
 EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_destroy);
+
+#ifdef CONFIG_DEBUG_FS
+void drm_gem_cma_describe(struct drm_gem_cma_object *cma_obj, struct seq_file *m)
+{
+	struct drm_gem_object *obj = &cma_obj->base;
+	struct drm_device *dev = obj->dev;
+	uint64_t off = 0;
+
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	if (obj->map_list.map)
+		off = (uint64_t)obj->map_list.hash.key;
+
+	seq_printf(m, "%2d (%2d) %08llx %08Zx %p %d",
+			obj->name, obj->refcount.refcount.counter,
+			off, cma_obj->paddr, cma_obj->vaddr, obj->size);
+
+	seq_printf(m, "\n");
+}
+EXPORT_SYMBOL_GPL(drm_gem_cma_describe);
+#endif
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 19c01ca3cc76..a6a8643a6a77 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -505,6 +505,7 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc)
 
 	/* Valid dotclock? */
 	if (dotclock > 0) {
+		int frame_size;
 		/* Convert scanline length in pixels and video dot clock to
 		 * line duration, frame duration and pixel duration in
 		 * nanoseconds:
@@ -512,7 +513,10 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc)
 		pixeldur_ns = (s64) div64_u64(1000000000, dotclock);
 		linedur_ns  = (s64) div64_u64(((u64) crtc->hwmode.crtc_htotal *
 					      1000000000), dotclock);
-		framedur_ns = (s64) crtc->hwmode.crtc_vtotal * linedur_ns;
+		frame_size = crtc->hwmode.crtc_htotal *
+				crtc->hwmode.crtc_vtotal;
+		framedur_ns = (s64) div64_u64((u64) frame_size * 1000000000,
+					      dotclock);
 	} else
 		DRM_ERROR("crtc %d: Can't calculate constants, dotclock = 0!\n",
 			  crtc->base.id);
@@ -863,6 +867,7 @@ void drm_send_vblank_event(struct drm_device *dev, int crtc,
 
 		now = get_drm_timestamp();
 	}
+	e->pipe = crtc;
 	send_vblank_event(dev, e, seq, &now);
 }
 EXPORT_SYMBOL(drm_send_vblank_event);
@@ -1218,8 +1223,9 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	int ret;
 	unsigned int flags, seq, crtc, high_crtc;
 
-	if ((!drm_dev_to_irq(dev)) || (!dev->irq_enabled))
-		return -EINVAL;
+	if (drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
+		if ((!drm_dev_to_irq(dev)) || (!dev->irq_enabled))
+			return -EINVAL;
 
 	if (vblwait->request.type & _DRM_VBLANK_SIGNAL)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c
index 2aa331499f81..db1e2d6f90d7 100644
--- a/drivers/gpu/drm/drm_mm.c
+++ b/drivers/gpu/drm/drm_mm.c
@@ -102,20 +102,6 @@ int drm_mm_pre_get(struct drm_mm *mm)
 }
 EXPORT_SYMBOL(drm_mm_pre_get);
 
-static inline unsigned long drm_mm_hole_node_start(struct drm_mm_node *hole_node)
-{
-	return hole_node->start + hole_node->size;
-}
-
-static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)
-{
-	struct drm_mm_node *next_node =
-		list_entry(hole_node->node_list.next, struct drm_mm_node,
-			   node_list);
-
-	return next_node->start;
-}
-
 static void drm_mm_insert_helper(struct drm_mm_node *hole_node,
 				 struct drm_mm_node *node,
 				 unsigned long size, unsigned alignment,
@@ -127,7 +113,7 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node,
 	unsigned long adj_start = hole_start;
 	unsigned long adj_end = hole_end;
 
-	BUG_ON(!hole_node->hole_follows || node->allocated);
+	BUG_ON(node->allocated);
 
 	if (mm->color_adjust)
 		mm->color_adjust(hole_node, color, &adj_start, &adj_end);
@@ -155,12 +141,57 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node,
 	BUG_ON(node->start + node->size > adj_end);
 
 	node->hole_follows = 0;
-	if (node->start + node->size < hole_end) {
+	if (__drm_mm_hole_node_start(node) < hole_end) {
 		list_add(&node->hole_stack, &mm->hole_stack);
 		node->hole_follows = 1;
 	}
 }
 
+struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,
+					unsigned long start,
+					unsigned long size,
+					bool atomic)
+{
+	struct drm_mm_node *hole, *node;
+	unsigned long end = start + size;
+	unsigned long hole_start;
+	unsigned long hole_end;
+
+	drm_mm_for_each_hole(hole, mm, hole_start, hole_end) {
+		if (hole_start > start || hole_end < end)
+			continue;
+
+		node = drm_mm_kmalloc(mm, atomic);
+		if (unlikely(node == NULL))
+			return NULL;
+
+		node->start = start;
+		node->size = size;
+		node->mm = mm;
+		node->allocated = 1;
+
+		INIT_LIST_HEAD(&node->hole_stack);
+		list_add(&node->node_list, &hole->node_list);
+
+		if (start == hole_start) {
+			hole->hole_follows = 0;
+			list_del_init(&hole->hole_stack);
+		}
+
+		node->hole_follows = 0;
+		if (end != hole_end) {
+			list_add(&node->hole_stack, &mm->hole_stack);
+			node->hole_follows = 1;
+		}
+
+		return node;
+	}
+
+	WARN(1, "no hole found for block 0x%lx + 0x%lx\n", start, size);
+	return NULL;
+}
+EXPORT_SYMBOL(drm_mm_create_block);
+
 struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *hole_node,
 					     unsigned long size,
 					     unsigned alignment,
@@ -253,7 +284,7 @@ static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node,
 	BUG_ON(node->start + node->size > end);
 
 	node->hole_follows = 0;
-	if (node->start + node->size < hole_end) {
+	if (__drm_mm_hole_node_start(node) < hole_end) {
 		list_add(&node->hole_stack, &mm->hole_stack);
 		node->hole_follows = 1;
 	}
@@ -327,12 +358,13 @@ void drm_mm_remove_node(struct drm_mm_node *node)
 	    list_entry(node->node_list.prev, struct drm_mm_node, node_list);
 
 	if (node->hole_follows) {
-		BUG_ON(drm_mm_hole_node_start(node)
-				== drm_mm_hole_node_end(node));
+		BUG_ON(__drm_mm_hole_node_start(node) ==
+		       __drm_mm_hole_node_end(node));
 		list_del(&node->hole_stack);
 	} else
-		BUG_ON(drm_mm_hole_node_start(node)
-				!= drm_mm_hole_node_end(node));
+		BUG_ON(__drm_mm_hole_node_start(node) !=
+		       __drm_mm_hole_node_end(node));
+
 
 	if (!prev_node->hole_follows) {
 		prev_node->hole_follows = 1;
@@ -390,6 +422,8 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,
 {
 	struct drm_mm_node *entry;
 	struct drm_mm_node *best;
+	unsigned long adj_start;
+	unsigned long adj_end;
 	unsigned long best_size;
 
 	BUG_ON(mm->scanned_blocks);
@@ -397,17 +431,13 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,
 	best = NULL;
 	best_size = ~0UL;
 
-	list_for_each_entry(entry, &mm->hole_stack, hole_stack) {
-		unsigned long adj_start = drm_mm_hole_node_start(entry);
-		unsigned long adj_end = drm_mm_hole_node_end(entry);
-
+	drm_mm_for_each_hole(entry, mm, adj_start, adj_end) {
 		if (mm->color_adjust) {
 			mm->color_adjust(entry, color, &adj_start, &adj_end);
 			if (adj_end <= adj_start)
 				continue;
 		}
 
-		BUG_ON(!entry->hole_follows);
 		if (!check_free_hole(adj_start, adj_end, size, alignment))
 			continue;
 
@@ -434,6 +464,8 @@ struct drm_mm_node *drm_mm_search_free_in_range_generic(const struct drm_mm *mm,
 {
 	struct drm_mm_node *entry;
 	struct drm_mm_node *best;
+	unsigned long adj_start;
+	unsigned long adj_end;
 	unsigned long best_size;
 
 	BUG_ON(mm->scanned_blocks);
@@ -441,13 +473,11 @@ struct drm_mm_node *drm_mm_search_free_in_range_generic(const struct drm_mm *mm,
 	best = NULL;
 	best_size = ~0UL;
 
-	list_for_each_entry(entry, &mm->hole_stack, hole_stack) {
-		unsigned long adj_start = drm_mm_hole_node_start(entry) < start ?
-			start : drm_mm_hole_node_start(entry);
-		unsigned long adj_end = drm_mm_hole_node_end(entry) > end ?
-			end : drm_mm_hole_node_end(entry);
-
-		BUG_ON(!entry->hole_follows);
+	drm_mm_for_each_hole(entry, mm, adj_start, adj_end) {
+		if (adj_start < start)
+			adj_start = start;
+		if (adj_end > end)
+			adj_end = end;
 
 		if (mm->color_adjust) {
 			mm->color_adjust(entry, color, &adj_start, &adj_end);
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index d8da30e90db5..04fa6f1808d1 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -35,6 +35,8 @@
 #include <linux/export.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc.h>
+#include <video/of_videomode.h>
+#include <video/videomode.h>
 
 /**
  * drm_mode_debug_printmodeline - debug print a mode
@@ -504,6 +506,74 @@ drm_gtf_mode(struct drm_device *dev, int hdisplay, int vdisplay, int vrefresh,
 }
 EXPORT_SYMBOL(drm_gtf_mode);
 
+#if IS_ENABLED(CONFIG_VIDEOMODE)
+int drm_display_mode_from_videomode(const struct videomode *vm,
+				    struct drm_display_mode *dmode)
+{
+	dmode->hdisplay = vm->hactive;
+	dmode->hsync_start = dmode->hdisplay + vm->hfront_porch;
+	dmode->hsync_end = dmode->hsync_start + vm->hsync_len;
+	dmode->htotal = dmode->hsync_end + vm->hback_porch;
+
+	dmode->vdisplay = vm->vactive;
+	dmode->vsync_start = dmode->vdisplay + vm->vfront_porch;
+	dmode->vsync_end = dmode->vsync_start + vm->vsync_len;
+	dmode->vtotal = dmode->vsync_end + vm->vback_porch;
+
+	dmode->clock = vm->pixelclock / 1000;
+
+	dmode->flags = 0;
+	if (vm->dmt_flags & VESA_DMT_HSYNC_HIGH)
+		dmode->flags |= DRM_MODE_FLAG_PHSYNC;
+	else if (vm->dmt_flags & VESA_DMT_HSYNC_LOW)
+		dmode->flags |= DRM_MODE_FLAG_NHSYNC;
+	if (vm->dmt_flags & VESA_DMT_VSYNC_HIGH)
+		dmode->flags |= DRM_MODE_FLAG_PVSYNC;
+	else if (vm->dmt_flags & VESA_DMT_VSYNC_LOW)
+		dmode->flags |= DRM_MODE_FLAG_NVSYNC;
+	if (vm->data_flags & DISPLAY_FLAGS_INTERLACED)
+		dmode->flags |= DRM_MODE_FLAG_INTERLACE;
+	if (vm->data_flags & DISPLAY_FLAGS_DOUBLESCAN)
+		dmode->flags |= DRM_MODE_FLAG_DBLSCAN;
+	drm_mode_set_name(dmode);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(drm_display_mode_from_videomode);
+#endif
+
+#if IS_ENABLED(CONFIG_OF_VIDEOMODE)
+/**
+ * of_get_drm_display_mode - get a drm_display_mode from devicetree
+ * @np: device_node with the timing specification
+ * @dmode: will be set to the return value
+ * @index: index into the list of display timings in devicetree
+ *
+ * This function is expensive and should only be used, if only one mode is to be
+ * read from DT. To get multiple modes start with of_get_display_timings and
+ * work with that instead.
+ */
+int of_get_drm_display_mode(struct device_node *np,
+			    struct drm_display_mode *dmode, int index)
+{
+	struct videomode vm;
+	int ret;
+
+	ret = of_get_videomode(np, &vm, index);
+	if (ret)
+		return ret;
+
+	drm_display_mode_from_videomode(&vm, dmode);
+
+	pr_debug("%s: got %dx%d display mode from %s\n",
+		of_node_full_name(np), vm.hactive, vm.vactive, np->name);
+	drm_mode_debug_printmodeline(dmode);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(of_get_drm_display_mode);
+#endif
+
 /**
  * drm_mode_set_name - set the name on a mode
  * @mode: name will be set in this mode
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 754bc96e10c7..bd719e936e13 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -439,78 +439,67 @@ int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 	return 0;
 }
 
-#else
-
-int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
-{
-	return -1;
-}
-
-#endif
-
-EXPORT_SYMBOL(drm_pci_init);
-
-/*@}*/
-void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
-{
-	struct drm_device *dev, *tmp;
-	DRM_DEBUG("\n");
-
-	if (driver->driver_features & DRIVER_MODESET) {
-		pci_unregister_driver(pdriver);
-	} else {
-		list_for_each_entry_safe(dev, tmp, &driver->device_list, driver_item)
-			drm_put_dev(dev);
-	}
-	DRM_INFO("Module unloaded\n");
-}
-EXPORT_SYMBOL(drm_pci_exit);
-
 int drm_pcie_get_speed_cap_mask(struct drm_device *dev, u32 *mask)
 {
 	struct pci_dev *root;
-	int pos;
-	u32 lnkcap = 0, lnkcap2 = 0;
+	u32 lnkcap, lnkcap2;
 
 	*mask = 0;
 	if (!dev->pdev)
 		return -EINVAL;
 
-	if (!pci_is_pcie(dev->pdev))
-		return -EINVAL;
-
 	root = dev->pdev->bus->self;
 
-	pos = pci_pcie_cap(root);
-	if (!pos)
-		return -EINVAL;
-
 	/* we've been informed via and serverworks don't make the cut */
 	if (root->vendor == PCI_VENDOR_ID_VIA ||
 	    root->vendor == PCI_VENDOR_ID_SERVERWORKS)
 		return -EINVAL;
 
-	pci_read_config_dword(root, pos + PCI_EXP_LNKCAP, &lnkcap);
-	pci_read_config_dword(root, pos + PCI_EXP_LNKCAP2, &lnkcap2);
+	pcie_capability_read_dword(root, PCI_EXP_LNKCAP, &lnkcap);
+	pcie_capability_read_dword(root, PCI_EXP_LNKCAP2, &lnkcap2);
 
-	lnkcap &= PCI_EXP_LNKCAP_SLS;
-	lnkcap2 &= 0xfe;
-
-	if (lnkcap2) { /* PCIE GEN 3.0 */
+	if (lnkcap2) {	/* PCIe r3.0-compliant */
 		if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_2_5GB)
 			*mask |= DRM_PCIE_SPEED_25;
 		if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_5_0GB)
 			*mask |= DRM_PCIE_SPEED_50;
 		if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_8_0GB)
 			*mask |= DRM_PCIE_SPEED_80;
-	} else {
-		if (lnkcap & 1)
+	} else {	/* pre-r3.0 */
+		if (lnkcap & PCI_EXP_LNKCAP_SLS_2_5GB)
 			*mask |= DRM_PCIE_SPEED_25;
-		if (lnkcap & 2)
-			*mask |= DRM_PCIE_SPEED_50;
+		if (lnkcap & PCI_EXP_LNKCAP_SLS_5_0GB)
+			*mask |= (DRM_PCIE_SPEED_25 | DRM_PCIE_SPEED_50);
 	}
 
 	DRM_INFO("probing gen 2 caps for device %x:%x = %x/%x\n", root->vendor, root->device, lnkcap, lnkcap2);
 	return 0;
 }
 EXPORT_SYMBOL(drm_pcie_get_speed_cap_mask);
+
+#else
+
+int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
+{
+	return -1;
+}
+
+#endif
+
+EXPORT_SYMBOL(drm_pci_init);
+
+/*@}*/
+void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
+{
+	struct drm_device *dev, *tmp;
+	DRM_DEBUG("\n");
+
+	if (driver->driver_features & DRIVER_MODESET) {
+		pci_unregister_driver(pdriver);
+	} else {
+		list_for_each_entry_safe(dev, tmp, &driver->device_list, driver_item)
+			drm_put_dev(dev);
+	}
+	DRM_INFO("Module unloaded\n");
+}
+EXPORT_SYMBOL(drm_pci_exit);
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 7f125738f44e..366910ddcfcb 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -53,7 +53,8 @@
  * Self-importing: if userspace is using PRIME as a replacement for flink
  * then it will get a fd->handle request for a GEM object that it created.
  * Drivers should detect this situation and return back the gem object
- * from the dma-buf private.
+ * from the dma-buf private.  Prime will do this automatically for drivers that
+ * use the drm_gem_prime_{import,export} helpers.
  */
 
 struct drm_prime_member {
@@ -62,6 +63,137 @@ struct drm_prime_member {
 	uint32_t handle;
 };
 
+static struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach,
+		enum dma_data_direction dir)
+{
+	struct drm_gem_object *obj = attach->dmabuf->priv;
+	struct sg_table *sgt;
+
+	mutex_lock(&obj->dev->struct_mutex);
+
+	sgt = obj->dev->driver->gem_prime_get_sg_table(obj);
+
+	if (!IS_ERR_OR_NULL(sgt))
+		dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir);
+
+	mutex_unlock(&obj->dev->struct_mutex);
+	return sgt;
+}
+
+static void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach,
+		struct sg_table *sgt, enum dma_data_direction dir)
+{
+	dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir);
+	sg_free_table(sgt);
+	kfree(sgt);
+}
+
+static void drm_gem_dmabuf_release(struct dma_buf *dma_buf)
+{
+	struct drm_gem_object *obj = dma_buf->priv;
+
+	if (obj->export_dma_buf == dma_buf) {
+		/* drop the reference on the export fd holds */
+		obj->export_dma_buf = NULL;
+		drm_gem_object_unreference_unlocked(obj);
+	}
+}
+
+static void *drm_gem_dmabuf_vmap(struct dma_buf *dma_buf)
+{
+	struct drm_gem_object *obj = dma_buf->priv;
+	struct drm_device *dev = obj->dev;
+
+	return dev->driver->gem_prime_vmap(obj);
+}
+
+static void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
+{
+	struct drm_gem_object *obj = dma_buf->priv;
+	struct drm_device *dev = obj->dev;
+
+	dev->driver->gem_prime_vunmap(obj, vaddr);
+}
+
+static void *drm_gem_dmabuf_kmap_atomic(struct dma_buf *dma_buf,
+		unsigned long page_num)
+{
+	return NULL;
+}
+
+static void drm_gem_dmabuf_kunmap_atomic(struct dma_buf *dma_buf,
+		unsigned long page_num, void *addr)
+{
+
+}
+static void *drm_gem_dmabuf_kmap(struct dma_buf *dma_buf,
+		unsigned long page_num)
+{
+	return NULL;
+}
+
+static void drm_gem_dmabuf_kunmap(struct dma_buf *dma_buf,
+		unsigned long page_num, void *addr)
+{
+
+}
+
+static int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf,
+		struct vm_area_struct *vma)
+{
+	return -EINVAL;
+}
+
+static const struct dma_buf_ops drm_gem_prime_dmabuf_ops =  {
+	.map_dma_buf = drm_gem_map_dma_buf,
+	.unmap_dma_buf = drm_gem_unmap_dma_buf,
+	.release = drm_gem_dmabuf_release,
+	.kmap = drm_gem_dmabuf_kmap,
+	.kmap_atomic = drm_gem_dmabuf_kmap_atomic,
+	.kunmap = drm_gem_dmabuf_kunmap,
+	.kunmap_atomic = drm_gem_dmabuf_kunmap_atomic,
+	.mmap = drm_gem_dmabuf_mmap,
+	.vmap = drm_gem_dmabuf_vmap,
+	.vunmap = drm_gem_dmabuf_vunmap,
+};
+
+/**
+ * DOC: PRIME Helpers
+ *
+ * Drivers can implement @gem_prime_export and @gem_prime_import in terms of
+ * simpler APIs by using the helper functions @drm_gem_prime_export and
+ * @drm_gem_prime_import.  These functions implement dma-buf support in terms of
+ * five lower-level driver callbacks:
+ *
+ * Export callbacks:
+ *
+ *  - @gem_prime_pin (optional): prepare a GEM object for exporting
+ *
+ *  - @gem_prime_get_sg_table: provide a scatter/gather table of pinned pages
+ *
+ *  - @gem_prime_vmap: vmap a buffer exported by your driver
+ *
+ *  - @gem_prime_vunmap: vunmap a buffer exported by your driver
+ *
+ * Import callback:
+ *
+ *  - @gem_prime_import_sg_table (import): produce a GEM object from another
+ *    driver's scatter/gather table
+ */
+
+struct dma_buf *drm_gem_prime_export(struct drm_device *dev,
+				     struct drm_gem_object *obj, int flags)
+{
+	if (dev->driver->gem_prime_pin) {
+		int ret = dev->driver->gem_prime_pin(obj);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+	return dma_buf_export(obj, &drm_gem_prime_dmabuf_ops, obj->size,
+			      0600);
+}
+EXPORT_SYMBOL(drm_gem_prime_export);
+
 int drm_gem_prime_handle_to_fd(struct drm_device *dev,
 		struct drm_file *file_priv, uint32_t handle, uint32_t flags,
 		int *prime_fd)
@@ -117,6 +249,58 @@ int drm_gem_prime_handle_to_fd(struct drm_device *dev,
 }
 EXPORT_SYMBOL(drm_gem_prime_handle_to_fd);
 
+struct drm_gem_object *drm_gem_prime_import(struct drm_device *dev,
+					    struct dma_buf *dma_buf)
+{
+	struct dma_buf_attachment *attach;
+	struct sg_table *sgt;
+	struct drm_gem_object *obj;
+	int ret;
+
+	if (!dev->driver->gem_prime_import_sg_table)
+		return ERR_PTR(-EINVAL);
+
+	if (dma_buf->ops == &drm_gem_prime_dmabuf_ops) {
+		obj = dma_buf->priv;
+		if (obj->dev == dev) {
+			/*
+			 * Importing dmabuf exported from out own gem increases
+			 * refcount on gem itself instead of f_count of dmabuf.
+			 */
+			drm_gem_object_reference(obj);
+			dma_buf_put(dma_buf);
+			return obj;
+		}
+	}
+
+	attach = dma_buf_attach(dma_buf, dev->dev);
+	if (IS_ERR(attach))
+		return ERR_PTR(PTR_ERR(attach));
+
+	sgt = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL);
+	if (IS_ERR_OR_NULL(sgt)) {
+		ret = PTR_ERR(sgt);
+		goto fail_detach;
+	}
+
+	obj = dev->driver->gem_prime_import_sg_table(dev, dma_buf->size, sgt);
+	if (IS_ERR(obj)) {
+		ret = PTR_ERR(obj);
+		goto fail_unmap;
+	}
+
+	obj->import_attach = attach;
+
+	return obj;
+
+fail_unmap:
+	dma_buf_unmap_attachment(attach, sgt, DMA_BIDIRECTIONAL);
+fail_detach:
+	dma_buf_detach(dma_buf, attach);
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL(drm_gem_prime_import);
+
 int drm_gem_prime_fd_to_handle(struct drm_device *dev,
 		struct drm_file *file_priv, int prime_fd, uint32_t *handle)
 {
diff --git a/drivers/gpu/drm/drm_usb.c b/drivers/gpu/drm/drm_usb.c
index 3cec30611417..34a156f0c336 100644
--- a/drivers/gpu/drm/drm_usb.c
+++ b/drivers/gpu/drm/drm_usb.c
@@ -18,7 +18,7 @@ int drm_get_usb_dev(struct usb_interface *interface,
 
 	usbdev = interface_to_usbdev(interface);
 	dev->usbdev = usbdev;
-	dev->dev = &usbdev->dev;
+	dev->dev = &interface->dev;
 
 	mutex_lock(&drm_global_mutex);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 294c0513f587..0e04f4ea441f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -99,6 +99,10 @@ static int exynos_drm_fb_create_handle(struct drm_framebuffer *fb,
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	/* This fb should have only one gem object. */
+	if (WARN_ON(exynos_fb->buf_cnt != 1))
+		return -EINVAL;
+
 	return drm_gem_handle_create(file_priv,
 			&exynos_fb->exynos_gem_obj[0]->base, handle);
 }
@@ -217,23 +221,25 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		      struct drm_mode_fb_cmd2 *mode_cmd)
 {
 	struct drm_gem_object *obj;
+	struct exynos_drm_gem_obj *exynos_gem_obj;
 	struct exynos_drm_fb *exynos_fb;
 	int i, ret;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handles[0]);
-	if (!obj) {
-		DRM_ERROR("failed to lookup gem object\n");
-		return ERR_PTR(-ENOENT);
-	}
-
 	exynos_fb = kzalloc(sizeof(*exynos_fb), GFP_KERNEL);
 	if (!exynos_fb) {
 		DRM_ERROR("failed to allocate exynos drm framebuffer\n");
 		return ERR_PTR(-ENOMEM);
 	}
 
+	obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handles[0]);
+	if (!obj) {
+		DRM_ERROR("failed to lookup gem object\n");
+		ret = -ENOENT;
+		goto err_free;
+	}
+
 	drm_helper_mode_fill_fb_struct(&exynos_fb->fb, mode_cmd);
 	exynos_fb->exynos_gem_obj[0] = to_exynos_gem_obj(obj);
 	exynos_fb->buf_cnt = exynos_drm_format_num_buffers(mode_cmd);
@@ -241,43 +247,44 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 	DRM_DEBUG_KMS("buf_cnt = %d\n", exynos_fb->buf_cnt);
 
 	for (i = 1; i < exynos_fb->buf_cnt; i++) {
-		struct exynos_drm_gem_obj *exynos_gem_obj;
-		int ret;
-
 		obj = drm_gem_object_lookup(dev, file_priv,
 				mode_cmd->handles[i]);
 		if (!obj) {
 			DRM_ERROR("failed to lookup gem object\n");
-			kfree(exynos_fb);
-			return ERR_PTR(-ENOENT);
+			ret = -ENOENT;
+			exynos_fb->buf_cnt = i;
+			goto err_unreference;
 		}
 
 		exynos_gem_obj = to_exynos_gem_obj(obj);
+		exynos_fb->exynos_gem_obj[i] = exynos_gem_obj;
 
 		ret = check_fb_gem_memory_type(dev, exynos_gem_obj);
 		if (ret < 0) {
 			DRM_ERROR("cannot use this gem memory type for fb.\n");
-			kfree(exynos_fb);
-			return ERR_PTR(ret);
+			goto err_unreference;
 		}
-
-		exynos_fb->exynos_gem_obj[i] = to_exynos_gem_obj(obj);
 	}
 
 	ret = drm_framebuffer_init(dev, &exynos_fb->fb, &exynos_drm_fb_funcs);
 	if (ret) {
-		for (i = 0; i < exynos_fb->buf_cnt; i++) {
-			struct exynos_drm_gem_obj *gem_obj;
-
-			gem_obj = exynos_fb->exynos_gem_obj[i];
-			drm_gem_object_unreference_unlocked(&gem_obj->base);
-		}
-
-		kfree(exynos_fb);
-		return ERR_PTR(ret);
+		DRM_ERROR("failed to init framebuffer.\n");
+		goto err_unreference;
 	}
 
 	return &exynos_fb->fb;
+
+err_unreference:
+	for (i = 0; i < exynos_fb->buf_cnt; i++) {
+		struct drm_gem_object *obj;
+
+		obj = &exynos_fb->exynos_gem_obj[i]->base;
+		if (obj)
+			drm_gem_object_unreference_unlocked(obj);
+	}
+err_free:
+	kfree(exynos_fb);
+	return ERR_PTR(ret);
 }
 
 struct exynos_drm_gem_buf *exynos_drm_fb_buffer(struct drm_framebuffer *fb,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index 71f867340a88..68f0045f86b8 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -226,36 +226,8 @@ out:
 	return ret;
 }
 
-static int exynos_drm_fbdev_probe(struct drm_fb_helper *helper,
-				   struct drm_fb_helper_surface_size *sizes)
-{
-	int ret = 0;
-
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	/*
-	 * with !helper->fb, it means that this funcion is called first time
-	 * and after that, the helper->fb would be used as clone mode.
-	 */
-	if (!helper->fb) {
-		ret = exynos_drm_fbdev_create(helper, sizes);
-		if (ret < 0) {
-			DRM_ERROR("failed to create fbdev.\n");
-			return ret;
-		}
-
-		/*
-		 * fb_helper expects a value more than 1 if succeed
-		 * because register_framebuffer() should be called.
-		 */
-		ret = 1;
-	}
-
-	return ret;
-}
-
 static struct drm_fb_helper_funcs exynos_drm_fb_helper_funcs = {
-	.fb_probe =	exynos_drm_fbdev_probe,
+	.fb_probe =	exynos_drm_fbdev_create,
 };
 
 int exynos_drm_fbdev_init(struct drm_device *dev)
@@ -295,6 +267,9 @@ int exynos_drm_fbdev_init(struct drm_device *dev)
 
 	}
 
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	ret = drm_fb_helper_initial_config(helper, PREFERRED_BPP);
 	if (ret < 0) {
 		DRM_ERROR("failed to set up hw configuration.\n");
@@ -326,8 +301,10 @@ static void exynos_drm_fbdev_destroy(struct drm_device *dev,
 	/* release drm framebuffer and real buffer */
 	if (fb_helper->fb && fb_helper->fb->funcs) {
 		fb = fb_helper->fb;
-		if (fb)
+		if (fb) {
+			drm_framebuffer_unregister_private(fb);
 			drm_framebuffer_remove(fb);
+		}
 	}
 
 	/* release linux framebuffer */
@@ -374,5 +351,7 @@ void exynos_drm_fbdev_restore_mode(struct drm_device *dev)
 	if (!private || !private->fb_helper)
 		return;
 
+	drm_modeset_lock_all(dev);
 	drm_fb_helper_restore_fbdev_mode(private->fb_helper);
+	drm_modeset_unlock_all(dev);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index fb2f81b8063d..3b0da0378acf 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -19,6 +19,7 @@
 #include <linux/workqueue.h>
 #include <linux/dma-mapping.h>
 #include <linux/dma-attrs.h>
+#include <linux/of.h>
 
 #include <drm/drmP.h>
 #include <drm/exynos_drm.h>
@@ -429,7 +430,7 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct drm_device *drm_dev,
 
 	g2d_userptr->pages = pages;
 
-	sgt = kzalloc(sizeof *sgt, GFP_KERNEL);
+	sgt = kzalloc(sizeof(*sgt), GFP_KERNEL);
 	if (!sgt) {
 		DRM_ERROR("failed to allocate sg table.\n");
 		ret = -ENOMEM;
@@ -1239,6 +1240,14 @@ static int g2d_resume(struct device *dev)
 
 static SIMPLE_DEV_PM_OPS(g2d_pm_ops, g2d_suspend, g2d_resume);
 
+#ifdef CONFIG_OF
+static const struct of_device_id exynos_g2d_match[] = {
+	{ .compatible = "samsung,exynos5250-g2d" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, exynos_g2d_match);
+#endif
+
 struct platform_driver g2d_driver = {
 	.probe		= g2d_probe,
 	.remove		= g2d_remove,
@@ -1246,5 +1255,6 @@ struct platform_driver g2d_driver = {
 		.name	= "s5p-g2d",
 		.owner	= THIS_MODULE,
 		.pm	= &g2d_pm_ops,
+		.of_match_table = of_match_ptr(exynos_g2d_match),
 	},
 };
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 473180776528..67e17ce112b6 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -329,17 +329,11 @@ static struct drm_file *exynos_drm_find_drm_file(struct drm_device *drm_dev,
 {
 	struct drm_file *file_priv;
 
-	mutex_lock(&drm_dev->struct_mutex);
-
 	/* find current process's drm_file from filelist. */
-	list_for_each_entry(file_priv, &drm_dev->filelist, lhead) {
-		if (file_priv->filp == filp) {
-			mutex_unlock(&drm_dev->struct_mutex);
+	list_for_each_entry(file_priv, &drm_dev->filelist, lhead)
+		if (file_priv->filp == filp)
 			return file_priv;
-		}
-	}
 
-	mutex_unlock(&drm_dev->struct_mutex);
 	WARN_ON(1);
 
 	return ERR_PTR(-EFAULT);
@@ -400,9 +394,7 @@ static int exynos_drm_gem_mmap_buffer(struct file *filp,
 	 */
 	drm_gem_object_reference(obj);
 
-	mutex_lock(&drm_dev->struct_mutex);
 	drm_vm_open_locked(drm_dev, vma);
-	mutex_unlock(&drm_dev->struct_mutex);
 
 	return 0;
 }
@@ -432,6 +424,16 @@ int exynos_drm_gem_mmap_ioctl(struct drm_device *dev, void *data,
 	}
 
 	/*
+	 * We have to use gem object and its fops for specific mmaper,
+	 * but vm_mmap() can deliver only filp. So we have to change
+	 * filp->f_op and filp->private_data temporarily, then restore
+	 * again. So it is important to keep lock until restoration the
+	 * settings to prevent others from misuse of filp->f_op or
+	 * filp->private_data.
+	 */
+	mutex_lock(&dev->struct_mutex);
+
+	/*
 	 * Set specific mmper's fops. And it will be restored by
 	 * exynos_drm_gem_mmap_buffer to dev->driver->fops.
 	 * This is used to call specific mapper temporarily.
@@ -448,13 +450,20 @@ int exynos_drm_gem_mmap_ioctl(struct drm_device *dev, void *data,
 	addr = vm_mmap(file_priv->filp, 0, args->size,
 			PROT_READ | PROT_WRITE, MAP_SHARED, 0);
 
-	drm_gem_object_unreference_unlocked(obj);
+	drm_gem_object_unreference(obj);
 
 	if (IS_ERR((void *)addr)) {
-		file_priv->filp->private_data = file_priv;
+		/* check filp->f_op, filp->private_data are restored */
+		if (file_priv->filp->f_op == &exynos_drm_gem_fops) {
+			file_priv->filp->f_op = fops_get(dev->driver->fops);
+			file_priv->filp->private_data = file_priv;
+		}
+		mutex_unlock(&dev->struct_mutex);
 		return PTR_ERR((void *)addr);
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	args->mapped = addr;
 
 	DRM_DEBUG_KMS("mapped = 0x%lx\n", (unsigned long)args->mapped);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
index 28644539b305..7c27df03c9ff 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
@@ -124,9 +124,21 @@ static struct edid *drm_hdmi_get_edid(struct device *dev,
 static int drm_hdmi_check_timing(struct device *dev, void *timing)
 {
 	struct drm_hdmi_context *ctx = to_context(dev);
+	int ret = 0;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	/*
+	* Both, mixer and hdmi should be able to handle the requested mode.
+	* If any of the two fails, return mode as BAD.
+	*/
+
+	if (mixer_ops && mixer_ops->check_timing)
+		ret = mixer_ops->check_timing(ctx->mixer_ctx->ctx, timing);
+
+	if (ret)
+		return ret;
+
 	if (hdmi_ops && hdmi_ops->check_timing)
 		return hdmi_ops->check_timing(ctx->hdmi_ctx->ctx, timing);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
index d80516fc9ed7..b7faa3662307 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
@@ -32,7 +32,7 @@ struct exynos_hdmi_ops {
 	bool (*is_connected)(void *ctx);
 	struct edid *(*get_edid)(void *ctx,
 			struct drm_connector *connector);
-	int (*check_timing)(void *ctx, void *timing);
+	int (*check_timing)(void *ctx, struct fb_videomode *timing);
 	int (*power_on)(void *ctx, int mode);
 
 	/* manager */
@@ -58,6 +58,9 @@ struct exynos_mixer_ops {
 	void (*win_mode_set)(void *ctx, struct exynos_drm_overlay *overlay);
 	void (*win_commit)(void *ctx, int zpos);
 	void (*win_disable)(void *ctx, int zpos);
+
+	/* display */
+	int (*check_timing)(void *ctx, struct fb_videomode *timing);
 };
 
 void exynos_hdmi_drv_attach(struct exynos_drm_hdmi_context *ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.h b/drivers/gpu/drm/exynos/exynos_drm_iommu.h
index 53b7deea8ab7..598e60f57d4b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_iommu.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.h
@@ -14,7 +14,7 @@
 
 #define EXYNOS_DEV_ADDR_START	0x20000000
 #define EXYNOS_DEV_ADDR_SIZE	0x40000000
-#define EXYNOS_DEV_ADDR_ORDER	0x4
+#define EXYNOS_DEV_ADDR_ORDER	0x0
 
 #ifdef CONFIG_DRM_EXYNOS_IOMMU
 
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 233247505ff8..2c5f266154ad 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -87,6 +87,73 @@ struct hdmi_resources {
 	int				regul_count;
 };
 
+struct hdmi_tg_regs {
+	u8 cmd[1];
+	u8 h_fsz[2];
+	u8 hact_st[2];
+	u8 hact_sz[2];
+	u8 v_fsz[2];
+	u8 vsync[2];
+	u8 vsync2[2];
+	u8 vact_st[2];
+	u8 vact_sz[2];
+	u8 field_chg[2];
+	u8 vact_st2[2];
+	u8 vact_st3[2];
+	u8 vact_st4[2];
+	u8 vsync_top_hdmi[2];
+	u8 vsync_bot_hdmi[2];
+	u8 field_top_hdmi[2];
+	u8 field_bot_hdmi[2];
+	u8 tg_3d[1];
+};
+
+struct hdmi_core_regs {
+	u8 h_blank[2];
+	u8 v2_blank[2];
+	u8 v1_blank[2];
+	u8 v_line[2];
+	u8 h_line[2];
+	u8 hsync_pol[1];
+	u8 vsync_pol[1];
+	u8 int_pro_mode[1];
+	u8 v_blank_f0[2];
+	u8 v_blank_f1[2];
+	u8 h_sync_start[2];
+	u8 h_sync_end[2];
+	u8 v_sync_line_bef_2[2];
+	u8 v_sync_line_bef_1[2];
+	u8 v_sync_line_aft_2[2];
+	u8 v_sync_line_aft_1[2];
+	u8 v_sync_line_aft_pxl_2[2];
+	u8 v_sync_line_aft_pxl_1[2];
+	u8 v_blank_f2[2]; /* for 3D mode */
+	u8 v_blank_f3[2]; /* for 3D mode */
+	u8 v_blank_f4[2]; /* for 3D mode */
+	u8 v_blank_f5[2]; /* for 3D mode */
+	u8 v_sync_line_aft_3[2];
+	u8 v_sync_line_aft_4[2];
+	u8 v_sync_line_aft_5[2];
+	u8 v_sync_line_aft_6[2];
+	u8 v_sync_line_aft_pxl_3[2];
+	u8 v_sync_line_aft_pxl_4[2];
+	u8 v_sync_line_aft_pxl_5[2];
+	u8 v_sync_line_aft_pxl_6[2];
+	u8 vact_space_1[2];
+	u8 vact_space_2[2];
+	u8 vact_space_3[2];
+	u8 vact_space_4[2];
+	u8 vact_space_5[2];
+	u8 vact_space_6[2];
+};
+
+struct hdmi_v14_conf {
+	int pixel_clock;
+	struct hdmi_core_regs core;
+	struct hdmi_tg_regs tg;
+	int cea_video_id;
+};
+
 struct hdmi_context {
 	struct device			*dev;
 	struct drm_device		*drm_dev;
@@ -104,6 +171,7 @@ struct hdmi_context {
 
 	/* current hdmiphy conf index */
 	int cur_conf;
+	struct hdmi_v14_conf		mode_conf;
 
 	struct hdmi_resources		res;
 
@@ -392,586 +460,132 @@ static const struct hdmi_v13_conf hdmi_v13_confs[] = {
 };
 
 /* HDMI Version 1.4 */
-static const u8 hdmiphy_conf27_027[32] = {
-	0x01, 0xd1, 0x2d, 0x72, 0x40, 0x64, 0x12, 0x08,
-	0x43, 0xa0, 0x0e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
-	0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-	0x54, 0xe3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x00,
-};
-
-static const u8 hdmiphy_conf74_176[32] = {
-	0x01, 0xd1, 0x1f, 0x10, 0x40, 0x5b, 0xef, 0x08,
-	0x81, 0xa0, 0xb9, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-	0x5a, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-	0x54, 0xa6, 0x24, 0x01, 0x00, 0x00, 0x01, 0x00,
-};
-
-static const u8 hdmiphy_conf74_25[32] = {
-	0x01, 0xd1, 0x1f, 0x10, 0x40, 0x40, 0xf8, 0x08,
-	0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-	0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-	0x54, 0xa5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x00,
-};
-
-static const u8 hdmiphy_conf148_5[32] = {
-	0x01, 0xd1, 0x1f, 0x00, 0x40, 0x40, 0xf8, 0x08,
-	0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-	0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-	0x54, 0x4b, 0x25, 0x03, 0x00, 0x00, 0x01, 0x00,
-};
-
-struct hdmi_tg_regs {
-	u8 cmd;
-	u8 h_fsz_l;
-	u8 h_fsz_h;
-	u8 hact_st_l;
-	u8 hact_st_h;
-	u8 hact_sz_l;
-	u8 hact_sz_h;
-	u8 v_fsz_l;
-	u8 v_fsz_h;
-	u8 vsync_l;
-	u8 vsync_h;
-	u8 vsync2_l;
-	u8 vsync2_h;
-	u8 vact_st_l;
-	u8 vact_st_h;
-	u8 vact_sz_l;
-	u8 vact_sz_h;
-	u8 field_chg_l;
-	u8 field_chg_h;
-	u8 vact_st2_l;
-	u8 vact_st2_h;
-	u8 vact_st3_l;
-	u8 vact_st3_h;
-	u8 vact_st4_l;
-	u8 vact_st4_h;
-	u8 vsync_top_hdmi_l;
-	u8 vsync_top_hdmi_h;
-	u8 vsync_bot_hdmi_l;
-	u8 vsync_bot_hdmi_h;
-	u8 field_top_hdmi_l;
-	u8 field_top_hdmi_h;
-	u8 field_bot_hdmi_l;
-	u8 field_bot_hdmi_h;
-	u8 tg_3d;
-};
-
-struct hdmi_core_regs {
-	u8 h_blank[2];
-	u8 v2_blank[2];
-	u8 v1_blank[2];
-	u8 v_line[2];
-	u8 h_line[2];
-	u8 hsync_pol[1];
-	u8 vsync_pol[1];
-	u8 int_pro_mode[1];
-	u8 v_blank_f0[2];
-	u8 v_blank_f1[2];
-	u8 h_sync_start[2];
-	u8 h_sync_end[2];
-	u8 v_sync_line_bef_2[2];
-	u8 v_sync_line_bef_1[2];
-	u8 v_sync_line_aft_2[2];
-	u8 v_sync_line_aft_1[2];
-	u8 v_sync_line_aft_pxl_2[2];
-	u8 v_sync_line_aft_pxl_1[2];
-	u8 v_blank_f2[2]; /* for 3D mode */
-	u8 v_blank_f3[2]; /* for 3D mode */
-	u8 v_blank_f4[2]; /* for 3D mode */
-	u8 v_blank_f5[2]; /* for 3D mode */
-	u8 v_sync_line_aft_3[2];
-	u8 v_sync_line_aft_4[2];
-	u8 v_sync_line_aft_5[2];
-	u8 v_sync_line_aft_6[2];
-	u8 v_sync_line_aft_pxl_3[2];
-	u8 v_sync_line_aft_pxl_4[2];
-	u8 v_sync_line_aft_pxl_5[2];
-	u8 v_sync_line_aft_pxl_6[2];
-	u8 vact_space_1[2];
-	u8 vact_space_2[2];
-	u8 vact_space_3[2];
-	u8 vact_space_4[2];
-	u8 vact_space_5[2];
-	u8 vact_space_6[2];
-};
-
-struct hdmi_preset_conf {
-	struct hdmi_core_regs core;
-	struct hdmi_tg_regs tg;
-};
-
-struct hdmi_conf {
-	int width;
-	int height;
-	int vrefresh;
-	bool interlace;
-	int cea_video_id;
-	const u8 *hdmiphy_data;
-	const struct hdmi_preset_conf *conf;
-};
-
-static const struct hdmi_preset_conf hdmi_conf_480p60 = {
-	.core = {
-		.h_blank = {0x8a, 0x00},
-		.v2_blank = {0x0d, 0x02},
-		.v1_blank = {0x2d, 0x00},
-		.v_line = {0x0d, 0x02},
-		.h_line = {0x5a, 0x03},
-		.hsync_pol = {0x01},
-		.vsync_pol = {0x01},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0x0e, 0x00},
-		.h_sync_end = {0x4c, 0x00},
-		.v_sync_line_bef_2 = {0x0f, 0x00},
-		.v_sync_line_bef_1 = {0x09, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
-	},
-	.tg = {
-		0x00, /* cmd */
-		0x5a, 0x03, /* h_fsz */
-		0x8a, 0x00, 0xd0, 0x02, /* hact */
-		0x0d, 0x02, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x2d, 0x00, 0xe0, 0x01, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
-	},
+struct hdmiphy_config {
+	int pixel_clock;
+	u8 conf[32];
 };
 
-static const struct hdmi_preset_conf hdmi_conf_720p50 = {
-	.core = {
-		.h_blank = {0xbc, 0x02},
-		.v2_blank = {0xee, 0x02},
-		.v1_blank = {0x1e, 0x00},
-		.v_line = {0xee, 0x02},
-		.h_line = {0xbc, 0x07},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0xb6, 0x01},
-		.h_sync_end = {0xde, 0x01},
-		.v_sync_line_bef_2 = {0x0a, 0x00},
-		.v_sync_line_bef_1 = {0x05, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
-	},
-	.tg = {
-		0x00, /* cmd */
-		0xbc, 0x07, /* h_fsz */
-		0xbc, 0x02, 0x00, 0x05, /* hact */
-		0xee, 0x02, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x1e, 0x00, 0xd0, 0x02, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+/* list of all required phy config settings */
+static const struct hdmiphy_config hdmiphy_v14_configs[] = {
+	{
+		.pixel_clock = 25200000,
+		.conf = {
+			0x01, 0x51, 0x2A, 0x75, 0x40, 0x01, 0x00, 0x08,
+			0x82, 0x80, 0xfc, 0xd8, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xf4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_720p60 = {
-	.core = {
-		.h_blank = {0x72, 0x01},
-		.v2_blank = {0xee, 0x02},
-		.v1_blank = {0x1e, 0x00},
-		.v_line = {0xee, 0x02},
-		.h_line = {0x72, 0x06},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0x6c, 0x00},
-		.h_sync_end = {0x94, 0x00},
-		.v_sync_line_bef_2 = {0x0a, 0x00},
-		.v_sync_line_bef_1 = {0x05, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 27000000,
+		.conf = {
+			0x01, 0xd1, 0x22, 0x51, 0x40, 0x08, 0xfc, 0x20,
+			0x98, 0xa0, 0xcb, 0xd8, 0x45, 0xa0, 0xac, 0x80,
+			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xe4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x72, 0x06, /* h_fsz */
-		0x72, 0x01, 0x00, 0x05, /* hact */
-		0xee, 0x02, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x1e, 0x00, 0xd0, 0x02, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 27027000,
+		.conf = {
+			0x01, 0xd1, 0x2d, 0x72, 0x40, 0x64, 0x12, 0x08,
+			0x43, 0xa0, 0x0e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xe3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x00,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_1080i50 = {
-	.core = {
-		.h_blank = {0xd0, 0x02},
-		.v2_blank = {0x32, 0x02},
-		.v1_blank = {0x16, 0x00},
-		.v_line = {0x65, 0x04},
-		.h_line = {0x50, 0x0a},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x01},
-		.v_blank_f0 = {0x49, 0x02},
-		.v_blank_f1 = {0x65, 0x04},
-		.h_sync_start = {0x0e, 0x02},
-		.h_sync_end = {0x3a, 0x02},
-		.v_sync_line_bef_2 = {0x07, 0x00},
-		.v_sync_line_bef_1 = {0x02, 0x00},
-		.v_sync_line_aft_2 = {0x39, 0x02},
-		.v_sync_line_aft_1 = {0x34, 0x02},
-		.v_sync_line_aft_pxl_2 = {0x38, 0x07},
-		.v_sync_line_aft_pxl_1 = {0x38, 0x07},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 36000000,
+		.conf = {
+			0x01, 0x51, 0x2d, 0x55, 0x40, 0x01, 0x00, 0x08,
+			0x82, 0x80, 0x0e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xab, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x50, 0x0a, /* h_fsz */
-		0xd0, 0x02, 0x80, 0x07, /* hact */
-		0x65, 0x04, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x16, 0x00, 0x1c, 0x02, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x49, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x33, 0x02, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 40000000,
+		.conf = {
+			0x01, 0x51, 0x32, 0x55, 0x40, 0x01, 0x00, 0x08,
+			0x82, 0x80, 0x2c, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0x9a, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_1080i60 = {
-	.core = {
-		.h_blank = {0x18, 0x01},
-		.v2_blank = {0x32, 0x02},
-		.v1_blank = {0x16, 0x00},
-		.v_line = {0x65, 0x04},
-		.h_line = {0x98, 0x08},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x01},
-		.v_blank_f0 = {0x49, 0x02},
-		.v_blank_f1 = {0x65, 0x04},
-		.h_sync_start = {0x56, 0x00},
-		.h_sync_end = {0x82, 0x00},
-		.v_sync_line_bef_2 = {0x07, 0x00},
-		.v_sync_line_bef_1 = {0x02, 0x00},
-		.v_sync_line_aft_2 = {0x39, 0x02},
-		.v_sync_line_aft_1 = {0x34, 0x02},
-		.v_sync_line_aft_pxl_2 = {0xa4, 0x04},
-		.v_sync_line_aft_pxl_1 = {0xa4, 0x04},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 65000000,
+		.conf = {
+			0x01, 0xd1, 0x36, 0x34, 0x40, 0x1e, 0x0a, 0x08,
+			0x82, 0xa0, 0x45, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xbd, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x98, 0x08, /* h_fsz */
-		0x18, 0x01, 0x80, 0x07, /* hact */
-		0x65, 0x04, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x16, 0x00, 0x1c, 0x02, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x49, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x33, 0x02, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 74176000,
+		.conf = {
+			0x01, 0xd1, 0x3e, 0x35, 0x40, 0x5b, 0xde, 0x08,
+			0x82, 0xa0, 0x73, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x56, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xa6, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_1080p30 = {
-	.core = {
-		.h_blank = {0x18, 0x01},
-		.v2_blank = {0x65, 0x04},
-		.v1_blank = {0x2d, 0x00},
-		.v_line = {0x65, 0x04},
-		.h_line = {0x98, 0x08},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0x56, 0x00},
-		.h_sync_end = {0x82, 0x00},
-		.v_sync_line_bef_2 = {0x09, 0x00},
-		.v_sync_line_bef_1 = {0x04, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 74250000,
+		.conf = {
+			0x01, 0xd1, 0x1f, 0x10, 0x40, 0x40, 0xf8, 0x08,
+			0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
+			0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xa5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x00,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x98, 0x08, /* h_fsz */
-		0x18, 0x01, 0x80, 0x07, /* hact */
-		0x65, 0x04, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x2d, 0x00, 0x38, 0x04, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 83500000,
+		.conf = {
+			0x01, 0xd1, 0x23, 0x11, 0x40, 0x0c, 0xfb, 0x08,
+			0x85, 0xa0, 0xd1, 0xd8, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0x93, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_1080p50 = {
-	.core = {
-		.h_blank = {0xd0, 0x02},
-		.v2_blank = {0x65, 0x04},
-		.v1_blank = {0x2d, 0x00},
-		.v_line = {0x65, 0x04},
-		.h_line = {0x50, 0x0a},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0x0e, 0x02},
-		.h_sync_end = {0x3a, 0x02},
-		.v_sync_line_bef_2 = {0x09, 0x00},
-		.v_sync_line_bef_1 = {0x04, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		.vact_space_1 = {0xff, 0xff},
-		.vact_space_2 = {0xff, 0xff},
-		.vact_space_3 = {0xff, 0xff},
-		.vact_space_4 = {0xff, 0xff},
-		.vact_space_5 = {0xff, 0xff},
-		.vact_space_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 106500000,
+		.conf = {
+			0x01, 0xd1, 0x2c, 0x12, 0x40, 0x0c, 0x09, 0x08,
+			0x84, 0xa0, 0x0a, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0x73, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x50, 0x0a, /* h_fsz */
-		0xd0, 0x02, 0x80, 0x07, /* hact */
-		0x65, 0x04, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x2d, 0x00, 0x38, 0x04, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 108000000,
+		.conf = {
+			0x01, 0x51, 0x2d, 0x15, 0x40, 0x01, 0x00, 0x08,
+			0x82, 0x80, 0x0e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0xc7, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-};
-
-static const struct hdmi_preset_conf hdmi_conf_1080p60 = {
-	.core = {
-		.h_blank = {0x18, 0x01},
-		.v2_blank = {0x65, 0x04},
-		.v1_blank = {0x2d, 0x00},
-		.v_line = {0x65, 0x04},
-		.h_line = {0x98, 0x08},
-		.hsync_pol = {0x00},
-		.vsync_pol = {0x00},
-		.int_pro_mode = {0x00},
-		.v_blank_f0 = {0xff, 0xff},
-		.v_blank_f1 = {0xff, 0xff},
-		.h_sync_start = {0x56, 0x00},
-		.h_sync_end = {0x82, 0x00},
-		.v_sync_line_bef_2 = {0x09, 0x00},
-		.v_sync_line_bef_1 = {0x04, 0x00},
-		.v_sync_line_aft_2 = {0xff, 0xff},
-		.v_sync_line_aft_1 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_2 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_1 = {0xff, 0xff},
-		.v_blank_f2 = {0xff, 0xff},
-		.v_blank_f3 = {0xff, 0xff},
-		.v_blank_f4 = {0xff, 0xff},
-		.v_blank_f5 = {0xff, 0xff},
-		.v_sync_line_aft_3 = {0xff, 0xff},
-		.v_sync_line_aft_4 = {0xff, 0xff},
-		.v_sync_line_aft_5 = {0xff, 0xff},
-		.v_sync_line_aft_6 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_3 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_4 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_5 = {0xff, 0xff},
-		.v_sync_line_aft_pxl_6 = {0xff, 0xff},
-		/* other don't care */
+	{
+		.pixel_clock = 146250000,
+		.conf = {
+			0x01, 0xd1, 0x3d, 0x15, 0x40, 0x18, 0xfd, 0x08,
+			0x83, 0xa0, 0x6e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0x50, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
+		},
 	},
-	.tg = {
-		0x00, /* cmd */
-		0x98, 0x08, /* h_fsz */
-		0x18, 0x01, 0x80, 0x07, /* hact */
-		0x65, 0x04, /* v_fsz */
-		0x01, 0x00, 0x33, 0x02, /* vsync */
-		0x2d, 0x00, 0x38, 0x04, /* vact */
-		0x33, 0x02, /* field_chg */
-		0x48, 0x02, /* vact_st2 */
-		0x00, 0x00, /* vact_st3 */
-		0x00, 0x00, /* vact_st4 */
-		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
-		0x01, 0x00, 0x33, 0x02, /* field top/bot */
-		0x00, /* 3d FP */
+	{
+		.pixel_clock = 148500000,
+		.conf = {
+			0x01, 0xd1, 0x1f, 0x00, 0x40, 0x40, 0xf8, 0x08,
+			0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
+			0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x54, 0x4b, 0x25, 0x03, 0x00, 0x00, 0x01, 0x00,
+		},
 	},
 };
 
-static const struct hdmi_conf hdmi_confs[] = {
-	{ 720, 480, 60, false, 3, hdmiphy_conf27_027, &hdmi_conf_480p60 },
-	{ 1280, 720, 50, false, 19, hdmiphy_conf74_25, &hdmi_conf_720p50 },
-	{ 1280, 720, 60, false, 4, hdmiphy_conf74_25, &hdmi_conf_720p60 },
-	{ 1920, 1080, 50, true, 20, hdmiphy_conf74_25, &hdmi_conf_1080i50 },
-	{ 1920, 1080, 60, true, 5, hdmiphy_conf74_25, &hdmi_conf_1080i60 },
-	{ 1920, 1080, 30, false, 34, hdmiphy_conf74_176, &hdmi_conf_1080p30 },
-	{ 1920, 1080, 50, false, 31, hdmiphy_conf148_5, &hdmi_conf_1080p50 },
-	{ 1920, 1080, 60, false, 16, hdmiphy_conf148_5, &hdmi_conf_1080p60 },
-};
-
 struct hdmi_infoframe {
 	enum HDMI_PACKET_TYPE type;
 	u8 ver;
@@ -1275,31 +889,6 @@ static int hdmi_v13_conf_index(struct drm_display_mode *mode)
 	return -EINVAL;
 }
 
-static int hdmi_v14_conf_index(struct drm_display_mode *mode)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdmi_confs); ++i)
-		if (hdmi_confs[i].width == mode->hdisplay &&
-				hdmi_confs[i].height == mode->vdisplay &&
-				hdmi_confs[i].vrefresh == mode->vrefresh &&
-				hdmi_confs[i].interlace ==
-				((mode->flags & DRM_MODE_FLAG_INTERLACE) ?
-				 true : false))
-			return i;
-
-	return -EINVAL;
-}
-
-static int hdmi_conf_index(struct hdmi_context *hdata,
-			   struct drm_display_mode *mode)
-{
-	if (hdata->type == HDMI_TYPE13)
-		return hdmi_v13_conf_index(mode);
-
-	return hdmi_v14_conf_index(mode);
-}
-
 static u8 hdmi_chksum(struct hdmi_context *hdata,
 			u32 start, u8 len, u32 hdr_sum)
 {
@@ -1357,7 +946,7 @@ static void hdmi_reg_infoframe(struct hdmi_context *hdata,
 		if (hdata->type == HDMI_TYPE13)
 			vic = hdmi_v13_confs[hdata->cur_conf].cea_video_id;
 		else
-			vic = hdmi_confs[hdata->cur_conf].cea_video_id;
+			vic = hdata->mode_conf.cea_video_id;
 
 		hdmi_reg_writeb(hdata, HDMI_AVI_BYTE(4), vic);
 
@@ -1434,44 +1023,51 @@ static int hdmi_v13_check_timing(struct fb_videomode *check_timing)
 	return -EINVAL;
 }
 
+static int hdmi_v14_find_phy_conf(int pixel_clock)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmiphy_v14_configs); i++) {
+		if (hdmiphy_v14_configs[i].pixel_clock == pixel_clock)
+			return i;
+	}
+
+	DRM_DEBUG_KMS("Could not find phy config for %d\n", pixel_clock);
+	return -EINVAL;
+}
+
 static int hdmi_v14_check_timing(struct fb_videomode *check_timing)
 {
 	int i;
 
-	DRM_DEBUG_KMS("valid mode : xres=%d, yres=%d, refresh=%d, intl=%d\n",
+	DRM_DEBUG_KMS("mode: xres=%d, yres=%d, refresh=%d, clock=%d, intl=%d\n",
 			check_timing->xres, check_timing->yres,
-			check_timing->refresh, (check_timing->vmode &
-			FB_VMODE_INTERLACED) ? true : false);
+			check_timing->refresh, check_timing->pixclock,
+			(check_timing->vmode & FB_VMODE_INTERLACED) ?
+			true : false);
 
-	for (i = 0; i < ARRAY_SIZE(hdmi_confs); i++)
-		if (hdmi_confs[i].width == check_timing->xres &&
-			hdmi_confs[i].height == check_timing->yres &&
-			hdmi_confs[i].vrefresh == check_timing->refresh &&
-			hdmi_confs[i].interlace ==
-			((check_timing->vmode & FB_VMODE_INTERLACED) ?
-			 true : false))
-				return 0;
-
-	/* TODO */
+	for (i = 0; i < ARRAY_SIZE(hdmiphy_v14_configs); i++)
+		if (hdmiphy_v14_configs[i].pixel_clock ==
+			check_timing->pixclock)
+			return 0;
 
 	return -EINVAL;
 }
 
-static int hdmi_check_timing(void *ctx, void *timing)
+static int hdmi_check_timing(void *ctx, struct fb_videomode *timing)
 {
 	struct hdmi_context *hdata = ctx;
-	struct fb_videomode *check_timing = timing;
 
 	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
 
-	DRM_DEBUG_KMS("[%d]x[%d] [%d]Hz [%x]\n", check_timing->xres,
-			check_timing->yres, check_timing->refresh,
-			check_timing->vmode);
+	DRM_DEBUG_KMS("[%d]x[%d] [%d]Hz [%x]\n", timing->xres,
+			timing->yres, timing->refresh,
+			timing->vmode);
 
 	if (hdata->type == HDMI_TYPE13)
-		return hdmi_v13_check_timing(check_timing);
+		return hdmi_v13_check_timing(timing);
 	else
-		return hdmi_v14_check_timing(check_timing);
+		return hdmi_v14_check_timing(timing);
 }
 
 static void hdmi_set_acr(u32 freq, u8 *acr)
@@ -1795,9 +1391,8 @@ static void hdmi_v13_timing_apply(struct hdmi_context *hdata)
 
 static void hdmi_v14_timing_apply(struct hdmi_context *hdata)
 {
-	const struct hdmi_preset_conf *conf = hdmi_confs[hdata->cur_conf].conf;
-	const struct hdmi_core_regs *core = &conf->core;
-	const struct hdmi_tg_regs *tg = &conf->tg;
+	struct hdmi_core_regs *core = &hdata->mode_conf.core;
+	struct hdmi_tg_regs *tg = &hdata->mode_conf.tg;
 	int tries;
 
 	/* setting core registers */
@@ -1900,39 +1495,39 @@ static void hdmi_v14_timing_apply(struct hdmi_context *hdata)
 	hdmi_reg_writeb(hdata, HDMI_VACT_SPACE_6_1, core->vact_space_6[1]);
 
 	/* Timing generator registers */
-	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_L, tg->h_fsz_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_H, tg->h_fsz_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_L, tg->hact_st_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_H, tg->hact_st_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_L, tg->hact_sz_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_H, tg->hact_sz_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_L, tg->v_fsz_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_H, tg->v_fsz_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_L, tg->vsync_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_H, tg->vsync_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_L, tg->vsync2_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_H, tg->vsync2_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_L, tg->vact_st_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_H, tg->vact_st_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_L, tg->vact_sz_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_H, tg->vact_sz_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_L, tg->field_chg_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_H, tg->field_chg_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_L, tg->vact_st2_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_H, tg->vact_st2_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST3_L, tg->vact_st3_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST3_H, tg->vact_st3_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST4_L, tg->vact_st4_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST4_H, tg->vact_st4_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_L, tg->vsync_top_hdmi_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_H, tg->vsync_top_hdmi_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_L, tg->vsync_bot_hdmi_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_H, tg->vsync_bot_hdmi_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_L, tg->field_top_hdmi_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_H, tg->field_top_hdmi_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_L, tg->field_bot_hdmi_l);
-	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_H, tg->field_bot_hdmi_h);
-	hdmi_reg_writeb(hdata, HDMI_TG_3D, tg->tg_3d);
+	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_L, tg->h_fsz[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_H, tg->h_fsz[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_L, tg->hact_st[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_H, tg->hact_st[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_L, tg->hact_sz[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_H, tg->hact_sz[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_L, tg->v_fsz[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_H, tg->v_fsz[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_L, tg->vsync[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_H, tg->vsync[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_L, tg->vsync2[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_H, tg->vsync2[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_L, tg->vact_st[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_H, tg->vact_st[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_L, tg->vact_sz[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_H, tg->vact_sz[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_L, tg->field_chg[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_H, tg->field_chg[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_L, tg->vact_st2[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_H, tg->vact_st2[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST3_L, tg->vact_st3[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST3_H, tg->vact_st3[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST4_L, tg->vact_st4[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST4_H, tg->vact_st4[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_L, tg->vsync_top_hdmi[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_H, tg->vsync_top_hdmi[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_L, tg->vsync_bot_hdmi[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_H, tg->vsync_bot_hdmi[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_L, tg->field_top_hdmi[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_H, tg->field_top_hdmi[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_L, tg->field_bot_hdmi[0]);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_H, tg->field_bot_hdmi[1]);
+	hdmi_reg_writeb(hdata, HDMI_TG_3D, tg->tg_3d[0]);
 
 	/* waiting for HDMIPHY's PLL to get to steady state */
 	for (tries = 100; tries; --tries) {
@@ -2029,10 +1624,17 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 	}
 
 	/* pixel clock */
-	if (hdata->type == HDMI_TYPE13)
+	if (hdata->type == HDMI_TYPE13) {
 		hdmiphy_data = hdmi_v13_confs[hdata->cur_conf].hdmiphy_data;
-	else
-		hdmiphy_data = hdmi_confs[hdata->cur_conf].hdmiphy_data;
+	} else {
+		i = hdmi_v14_find_phy_conf(hdata->mode_conf.pixel_clock);
+		if (i < 0) {
+			DRM_ERROR("failed to find hdmiphy conf\n");
+			return;
+		}
+
+		hdmiphy_data = hdmiphy_v14_configs[i].conf;
+	}
 
 	memcpy(buffer, hdmiphy_data, 32);
 	ret = i2c_master_send(hdata->hdmiphy_port, buffer, 32);
@@ -2100,7 +1702,7 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector,
 	if (hdata->type == HDMI_TYPE13)
 		index = hdmi_v13_conf_index(adjusted_mode);
 	else
-		index = hdmi_v14_conf_index(adjusted_mode);
+		index = hdmi_v14_find_phy_conf(adjusted_mode->clock * 1000);
 
 	/* just return if user desired mode exists. */
 	if (index >= 0)
@@ -2114,7 +1716,7 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector,
 		if (hdata->type == HDMI_TYPE13)
 			index = hdmi_v13_conf_index(m);
 		else
-			index = hdmi_v14_conf_index(m);
+			index = hdmi_v14_find_phy_conf(m->clock * 1000);
 
 		if (index >= 0) {
 			struct drm_mode_object base;
@@ -2123,6 +1725,9 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector,
 			DRM_INFO("desired mode doesn't exist so\n");
 			DRM_INFO("use the most suitable mode among modes.\n");
 
+			DRM_DEBUG_KMS("Adjusted Mode: [%d]x[%d] [%d]Hz\n",
+				m->hdisplay, m->vdisplay, m->vrefresh);
+
 			/* preserve display mode header while copying. */
 			head = adjusted_mode->head;
 			base = adjusted_mode->base;
@@ -2134,6 +1739,122 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector,
 	}
 }
 
+static void hdmi_set_reg(u8 *reg_pair, int num_bytes, u32 value)
+{
+	int i;
+	BUG_ON(num_bytes > 4);
+	for (i = 0; i < num_bytes; i++)
+		reg_pair[i] = (value >> (8 * i)) & 0xff;
+}
+
+static void hdmi_v14_mode_set(struct hdmi_context *hdata,
+			struct drm_display_mode *m)
+{
+	struct hdmi_core_regs *core = &hdata->mode_conf.core;
+	struct hdmi_tg_regs *tg = &hdata->mode_conf.tg;
+
+	hdata->mode_conf.cea_video_id = drm_match_cea_mode(m);
+
+	hdata->mode_conf.pixel_clock = m->clock * 1000;
+	hdmi_set_reg(core->h_blank, 2, m->htotal - m->hdisplay);
+	hdmi_set_reg(core->v_line, 2, m->vtotal);
+	hdmi_set_reg(core->h_line, 2, m->htotal);
+	hdmi_set_reg(core->hsync_pol, 1,
+			(m->flags & DRM_MODE_FLAG_NHSYNC)  ? 1 : 0);
+	hdmi_set_reg(core->vsync_pol, 1,
+			(m->flags & DRM_MODE_FLAG_NVSYNC) ? 1 : 0);
+	hdmi_set_reg(core->int_pro_mode, 1,
+			(m->flags & DRM_MODE_FLAG_INTERLACE) ? 1 : 0);
+
+	/*
+	 * Quirk requirement for exynos 5 HDMI IP design,
+	 * 2 pixels less than the actual calculation for hsync_start
+	 * and end.
+	 */
+
+	/* Following values & calculations differ for different type of modes */
+	if (m->flags & DRM_MODE_FLAG_INTERLACE) {
+		/* Interlaced Mode */
+		hdmi_set_reg(core->v_sync_line_bef_2, 2,
+			(m->vsync_end - m->vdisplay) / 2);
+		hdmi_set_reg(core->v_sync_line_bef_1, 2,
+			(m->vsync_start - m->vdisplay) / 2);
+		hdmi_set_reg(core->v2_blank, 2, m->vtotal / 2);
+		hdmi_set_reg(core->v1_blank, 2, (m->vtotal - m->vdisplay) / 2);
+		hdmi_set_reg(core->v_blank_f0, 2, (m->vtotal +
+			((m->vsync_end - m->vsync_start) * 4) + 5) / 2);
+		hdmi_set_reg(core->v_blank_f1, 2, m->vtotal);
+		hdmi_set_reg(core->v_sync_line_aft_2, 2, (m->vtotal / 2) + 7);
+		hdmi_set_reg(core->v_sync_line_aft_1, 2, (m->vtotal / 2) + 2);
+		hdmi_set_reg(core->v_sync_line_aft_pxl_2, 2,
+			(m->htotal / 2) + (m->hsync_start - m->hdisplay));
+		hdmi_set_reg(core->v_sync_line_aft_pxl_1, 2,
+			(m->htotal / 2) + (m->hsync_start - m->hdisplay));
+		hdmi_set_reg(tg->vact_st, 2, (m->vtotal - m->vdisplay) / 2);
+		hdmi_set_reg(tg->vact_sz, 2, m->vdisplay / 2);
+		hdmi_set_reg(tg->vact_st2, 2, 0x249);/* Reset value + 1*/
+		hdmi_set_reg(tg->vact_st3, 2, 0x0);
+		hdmi_set_reg(tg->vact_st4, 2, 0x0);
+	} else {
+		/* Progressive Mode */
+		hdmi_set_reg(core->v_sync_line_bef_2, 2,
+			m->vsync_end - m->vdisplay);
+		hdmi_set_reg(core->v_sync_line_bef_1, 2,
+			m->vsync_start - m->vdisplay);
+		hdmi_set_reg(core->v2_blank, 2, m->vtotal);
+		hdmi_set_reg(core->v1_blank, 2, m->vtotal - m->vdisplay);
+		hdmi_set_reg(core->v_blank_f0, 2, 0xffff);
+		hdmi_set_reg(core->v_blank_f1, 2, 0xffff);
+		hdmi_set_reg(core->v_sync_line_aft_2, 2, 0xffff);
+		hdmi_set_reg(core->v_sync_line_aft_1, 2, 0xffff);
+		hdmi_set_reg(core->v_sync_line_aft_pxl_2, 2, 0xffff);
+		hdmi_set_reg(core->v_sync_line_aft_pxl_1, 2, 0xffff);
+		hdmi_set_reg(tg->vact_st, 2, m->vtotal - m->vdisplay);
+		hdmi_set_reg(tg->vact_sz, 2, m->vdisplay);
+		hdmi_set_reg(tg->vact_st2, 2, 0x248); /* Reset value */
+		hdmi_set_reg(tg->vact_st3, 2, 0x47b); /* Reset value */
+		hdmi_set_reg(tg->vact_st4, 2, 0x6ae); /* Reset value */
+	}
+
+	/* Following values & calculations are same irrespective of mode type */
+	hdmi_set_reg(core->h_sync_start, 2, m->hsync_start - m->hdisplay - 2);
+	hdmi_set_reg(core->h_sync_end, 2, m->hsync_end - m->hdisplay - 2);
+	hdmi_set_reg(core->vact_space_1, 2, 0xffff);
+	hdmi_set_reg(core->vact_space_2, 2, 0xffff);
+	hdmi_set_reg(core->vact_space_3, 2, 0xffff);
+	hdmi_set_reg(core->vact_space_4, 2, 0xffff);
+	hdmi_set_reg(core->vact_space_5, 2, 0xffff);
+	hdmi_set_reg(core->vact_space_6, 2, 0xffff);
+	hdmi_set_reg(core->v_blank_f2, 2, 0xffff);
+	hdmi_set_reg(core->v_blank_f3, 2, 0xffff);
+	hdmi_set_reg(core->v_blank_f4, 2, 0xffff);
+	hdmi_set_reg(core->v_blank_f5, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_3, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_4, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_5, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_6, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_pxl_3, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_pxl_4, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_pxl_5, 2, 0xffff);
+	hdmi_set_reg(core->v_sync_line_aft_pxl_6, 2, 0xffff);
+
+	/* Timing generator registers */
+	hdmi_set_reg(tg->cmd, 1, 0x0);
+	hdmi_set_reg(tg->h_fsz, 2, m->htotal);
+	hdmi_set_reg(tg->hact_st, 2, m->htotal - m->hdisplay);
+	hdmi_set_reg(tg->hact_sz, 2, m->hdisplay);
+	hdmi_set_reg(tg->v_fsz, 2, m->vtotal);
+	hdmi_set_reg(tg->vsync, 2, 0x1);
+	hdmi_set_reg(tg->vsync2, 2, 0x233); /* Reset value */
+	hdmi_set_reg(tg->field_chg, 2, 0x233); /* Reset value */
+	hdmi_set_reg(tg->vsync_top_hdmi, 2, 0x1); /* Reset value */
+	hdmi_set_reg(tg->vsync_bot_hdmi, 2, 0x233); /* Reset value */
+	hdmi_set_reg(tg->field_top_hdmi, 2, 0x1); /* Reset value */
+	hdmi_set_reg(tg->field_bot_hdmi, 2, 0x233); /* Reset value */
+	hdmi_set_reg(tg->tg_3d, 1, 0x0);
+
+}
+
 static void hdmi_mode_set(void *ctx, void *mode)
 {
 	struct hdmi_context *hdata = ctx;
@@ -2141,11 +1862,15 @@ static void hdmi_mode_set(void *ctx, void *mode)
 
 	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
 
-	conf_idx = hdmi_conf_index(hdata, mode);
-	if (conf_idx >= 0)
-		hdata->cur_conf = conf_idx;
-	else
-		DRM_DEBUG_KMS("not supported mode\n");
+	if (hdata->type == HDMI_TYPE13) {
+		conf_idx = hdmi_v13_conf_index(mode);
+		if (conf_idx >= 0)
+			hdata->cur_conf = conf_idx;
+		else
+			DRM_DEBUG_KMS("not supported mode\n");
+	} else {
+		hdmi_v14_mode_set(hdata, mode);
+	}
 }
 
 static void hdmi_get_max_resol(void *ctx, unsigned int *width,
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index c414584bfbae..e919aba29b3d 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -284,13 +284,13 @@ static void mixer_cfg_scan(struct mixer_context *ctx, unsigned int height)
 				MXR_CFG_SCAN_PROGRASSIVE);
 
 	/* choosing between porper HD and SD mode */
-	if (height == 480)
+	if (height <= 480)
 		val |= MXR_CFG_SCAN_NTSC | MXR_CFG_SCAN_SD;
-	else if (height == 576)
+	else if (height <= 576)
 		val |= MXR_CFG_SCAN_PAL | MXR_CFG_SCAN_SD;
-	else if (height == 720)
+	else if (height <= 720)
 		val |= MXR_CFG_SCAN_HD_720 | MXR_CFG_SCAN_HD;
-	else if (height == 1080)
+	else if (height <= 1080)
 		val |= MXR_CFG_SCAN_HD_1080 | MXR_CFG_SCAN_HD;
 	else
 		val |= MXR_CFG_SCAN_HD_720 | MXR_CFG_SCAN_HD;
@@ -818,6 +818,29 @@ static void mixer_win_disable(void *ctx, int win)
 	mixer_ctx->win_data[win].enabled = false;
 }
 
+int mixer_check_timing(void *ctx, struct fb_videomode *timing)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	u32 w, h;
+
+	w = timing->xres;
+	h = timing->yres;
+
+	DRM_DEBUG_KMS("%s : xres=%d, yres=%d, refresh=%d, intl=%d\n",
+		__func__, timing->xres, timing->yres,
+		timing->refresh, (timing->vmode &
+		FB_VMODE_INTERLACED) ? true : false);
+
+	if (mixer_ctx->mxr_ver == MXR_VER_0_0_0_16)
+		return 0;
+
+	if ((w >= 464 && w <= 720 && h >= 261 && h <= 576) ||
+		(w >= 1024 && w <= 1280 && h >= 576 && h <= 720) ||
+		(w >= 1664 && w <= 1920 && h >= 936 && h <= 1080))
+		return 0;
+
+	return -EINVAL;
+}
 static void mixer_wait_for_vblank(void *ctx)
 {
 	struct mixer_context *mixer_ctx = ctx;
@@ -955,6 +978,9 @@ static struct exynos_mixer_ops mixer_ops = {
 	.win_mode_set		= mixer_win_mode_set,
 	.win_commit		= mixer_win_commit,
 	.win_disable		= mixer_win_disable,
+
+	/* display */
+	.check_timing		= mixer_check_timing,
 };
 
 static irqreturn_t mixer_irq_handler(int irq, void *arg)
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index afded54dbb10..2590cac84257 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -260,13 +260,13 @@ static int psb_framebuffer_init(struct drm_device *dev,
 	default:
 		return -EINVAL;
 	}
+	drm_helper_mode_fill_fb_struct(&fb->base, mode_cmd);
+	fb->gtt = gt;
 	ret = drm_framebuffer_init(dev, &fb->base, &psb_fb_funcs);
 	if (ret) {
 		dev_err(dev->dev, "framebuffer init failed: %d\n", ret);
 		return ret;
 	}
-	drm_helper_mode_fill_fb_struct(&fb->base, mode_cmd);
-	fb->gtt = gt;
 	return 0;
 }
 
@@ -545,9 +545,7 @@ static int psbfb_probe(struct drm_fb_helper *helper,
 	struct psb_fbdev *psb_fbdev = (struct psb_fbdev *)helper;
 	struct drm_device *dev = psb_fbdev->psb_fb_helper.dev;
 	struct drm_psb_private *dev_priv = dev->dev_private;
-	int new_fb = 0;
 	int bytespp;
-	int ret;
 
 	bytespp = sizes->surface_bpp / 8;
 	if (bytespp == 3)	/* no 24bit packed */
@@ -562,13 +560,7 @@ static int psbfb_probe(struct drm_fb_helper *helper,
                 sizes->surface_depth = 16;
         }
 
-	if (!helper->fb) {
-		ret = psbfb_create(psb_fbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
+	return psbfb_create(psb_fbdev, sizes);
 }
 
 static struct drm_fb_helper_funcs psb_fb_helper_funcs = {
@@ -590,6 +582,7 @@ static int psb_fbdev_destroy(struct drm_device *dev, struct psb_fbdev *fbdev)
 		framebuffer_release(info);
 	}
 	drm_fb_helper_fini(&fbdev->psb_fb_helper);
+	drm_framebuffer_unregister_private(&psbfb->base);
 	drm_framebuffer_cleanup(&psbfb->base);
 
 	if (psbfb->gtt)
@@ -615,6 +608,10 @@ int psb_fbdev_init(struct drm_device *dev)
 							INTELFB_CONN_LIMIT);
 
 	drm_fb_helper_single_add_all_connectors(&fbdev->psb_fb_helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	drm_fb_helper_initial_config(&fbdev->psb_fb_helper, 32);
 	return 0;
 }
@@ -668,30 +665,6 @@ static void psb_user_framebuffer_destroy(struct drm_framebuffer *fb)
 {
 	struct psb_framebuffer *psbfb = to_psb_fb(fb);
 	struct gtt_range *r = psbfb->gtt;
-	struct drm_device *dev = fb->dev;
-	struct drm_psb_private *dev_priv = dev->dev_private;
-	struct psb_fbdev *fbdev = dev_priv->fbdev;
-	struct drm_crtc *crtc;
-	int reset = 0;
-
-	/* Should never get stolen memory for a user fb */
-	WARN_ON(r->stolen);
-
-	/* Check if we are erroneously live */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		if (crtc->fb == fb)
-			reset = 1;
-
-	if (reset)
-		/*
-		 * Now force a sane response before we permit the DRM CRTC
-		 * layer to do stupid things like blank the display. Instead
-		 * we reset this framebuffer as if the user had forced a reset.
-		 * We must do this before the cleanup so that the DRM layer
-		 * doesn't get a chance to stick its oar in where it isn't
-		 * wanted.
-		 */
-		drm_fb_helper_restore_fbdev_mode(&fbdev->psb_fb_helper);
 
 	/* Let DRM do its clean up */
 	drm_framebuffer_cleanup(fb);
diff --git a/drivers/gpu/drm/gma500/psb_device.c b/drivers/gpu/drm/gma500/psb_device.c
index b58c4701c4e8..f6f534b4197e 100644
--- a/drivers/gpu/drm/gma500/psb_device.c
+++ b/drivers/gpu/drm/gma500/psb_device.c
@@ -194,7 +194,7 @@ static int psb_save_display_registers(struct drm_device *dev)
 	regs->saveCHICKENBIT = PSB_RVDC32(DSPCHICKENBIT);
 
 	/* Save crtc and output state */
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		if (drm_helper_crtc_in_use(crtc))
 			crtc->funcs->save(crtc);
@@ -204,7 +204,7 @@ static int psb_save_display_registers(struct drm_device *dev)
 		if (connector->funcs->save)
 			connector->funcs->save(connector);
 
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return 0;
 }
 
@@ -234,7 +234,7 @@ static int psb_restore_display_registers(struct drm_device *dev)
 	/*make sure VGA plane is off. it initializes to on after reset!*/
 	PSB_WVDC32(0x80000000, VGACNTRL);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 		if (drm_helper_crtc_in_use(crtc))
 			crtc->funcs->restore(crtc);
@@ -243,7 +243,7 @@ static int psb_restore_display_registers(struct drm_device *dev)
 		if (connector->funcs->restore)
 			connector->funcs->restore(connector);
 
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index dd1fbfa7e467..111e3df9c5de 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -149,6 +149,16 @@ static struct drm_ioctl_desc psb_ioctls[] = {
 
 static void psb_lastclose(struct drm_device *dev)
 {
+	int ret;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_fbdev *fbdev = dev_priv->fbdev;
+
+	drm_modeset_lock_all(dev);
+	ret = drm_fb_helper_restore_fbdev_mode(&fbdev->psb_fb_helper);
+	if (ret)
+		DRM_DEBUG("failed to restore crtc mode\n");
+	drm_modeset_unlock_all(dev);
+
 	return;
 }
 
@@ -476,7 +486,7 @@ static int psb_mode_operation_ioctl(struct drm_device *dev, void *data,
 	case PSB_MODE_OPERATION_MODE_VALID:
 		umode = &arg->mode;
 
-		mutex_lock(&dev->mode_config.mutex);
+		drm_modeset_lock_all(dev);
 
 		obj = drm_mode_object_find(dev, obj_id,
 					DRM_MODE_OBJECT_CONNECTOR);
@@ -525,7 +535,7 @@ static int psb_mode_operation_ioctl(struct drm_device *dev, void *data,
 		if (mode)
 			drm_mode_destroy(dev, mode);
 mode_op_out:
-		mutex_unlock(&dev->mode_config.mutex);
+		drm_modeset_unlock_all(dev);
 		return ret;
 
 	default:
diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c
index 8033526bb53b..9edb1902a096 100644
--- a/drivers/gpu/drm/gma500/psb_intel_display.c
+++ b/drivers/gpu/drm/gma500/psb_intel_display.c
@@ -85,14 +85,14 @@ struct psb_intel_limit_t {
 #define I9XX_DOT_MAX		 400000
 #define I9XX_VCO_MIN		1400000
 #define I9XX_VCO_MAX		2800000
-#define I9XX_N_MIN		      3
-#define I9XX_N_MAX		      8
+#define I9XX_N_MIN		      1
+#define I9XX_N_MAX		      6
 #define I9XX_M_MIN		     70
 #define I9XX_M_MAX		    120
-#define I9XX_M1_MIN		     10
-#define I9XX_M1_MAX		     20
-#define I9XX_M2_MIN		      5
-#define I9XX_M2_MAX		      9
+#define I9XX_M1_MIN		      8
+#define I9XX_M1_MAX		     18
+#define I9XX_M2_MIN		      3
+#define I9XX_M2_MAX		      7
 #define I9XX_P_SDVO_DAC_MIN	      5
 #define I9XX_P_SDVO_DAC_MAX	     80
 #define I9XX_P_LVDS_MIN		      7
diff --git a/drivers/gpu/drm/i2c/Kconfig b/drivers/gpu/drm/i2c/Kconfig
new file mode 100644
index 000000000000..4d341db462a2
--- /dev/null
+++ b/drivers/gpu/drm/i2c/Kconfig
@@ -0,0 +1,28 @@
+menu "I2C encoder or helper chips"
+     depends on DRM && DRM_KMS_HELPER && I2C
+
+config DRM_I2C_CH7006
+	tristate "Chrontel ch7006 TV encoder"
+	default m if DRM_NOUVEAU
+	help
+	  Support for Chrontel ch7006 and similar TV encoders, found
+	  on some nVidia video cards.
+
+	  This driver is currently only useful if you're also using
+	  the nouveau driver.
+
+config DRM_I2C_SIL164
+	tristate "Silicon Image sil164 TMDS transmitter"
+	default m if DRM_NOUVEAU
+	help
+	  Support for sil164 and similar single-link (or dual-link
+	  when used in pairs) TMDS transmitters, used in some nVidia
+	  video cards.
+
+config DRM_I2C_NXP_TDA998X
+	tristate "NXP Semiconductors TDA998X HDMI encoder"
+	default m if DRM_TILCDC
+	help
+	  Support for NXP Semiconductors TDA998X HDMI encoders.
+
+endmenu
diff --git a/drivers/gpu/drm/i2c/Makefile b/drivers/gpu/drm/i2c/Makefile
index 92862563e7ee..43aa33baebed 100644
--- a/drivers/gpu/drm/i2c/Makefile
+++ b/drivers/gpu/drm/i2c/Makefile
@@ -5,3 +5,6 @@ obj-$(CONFIG_DRM_I2C_CH7006) += ch7006.o
 
 sil164-y := sil164_drv.o
 obj-$(CONFIG_DRM_I2C_SIL164) += sil164.o
+
+tda998x-y := tda998x_drv.o
+obj-$(CONFIG_DRM_I2C_NXP_TDA998X) += tda998x.o
diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c
index b865d0728e28..51fa32392029 100644
--- a/drivers/gpu/drm/i2c/ch7006_drv.c
+++ b/drivers/gpu/drm/i2c/ch7006_drv.c
@@ -364,7 +364,7 @@ static int ch7006_encoder_set_property(struct drm_encoder *encoder,
 				.crtc = crtc,
 			};
 
-			crtc->funcs->set_config(&modeset);
+			drm_mode_set_config_internal(&modeset);
 		}
 	}
 
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
new file mode 100644
index 000000000000..e68b58a1aaf9
--- /dev/null
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -0,0 +1,906 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+
+
+#include <linux/module.h>
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_encoder_slave.h>
+#include <drm/drm_edid.h>
+
+
+#define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
+
+struct tda998x_priv {
+	struct i2c_client *cec;
+	uint16_t rev;
+	uint8_t current_page;
+	int dpms;
+};
+
+#define to_tda998x_priv(x)  ((struct tda998x_priv *)to_encoder_slave(x)->slave_priv)
+
+/* The TDA9988 series of devices use a paged register scheme.. to simplify
+ * things we encode the page # in upper bits of the register #.  To read/
+ * write a given register, we need to make sure CURPAGE register is set
+ * appropriately.  Which implies reads/writes are not atomic.  Fun!
+ */
+
+#define REG(page, addr) (((page) << 8) | (addr))
+#define REG2ADDR(reg)   ((reg) & 0xff)
+#define REG2PAGE(reg)   (((reg) >> 8) & 0xff)
+
+#define REG_CURPAGE               0xff                /* write */
+
+
+/* Page 00h: General Control */
+#define REG_VERSION_LSB           REG(0x00, 0x00)     /* read */
+#define REG_MAIN_CNTRL0           REG(0x00, 0x01)     /* read/write */
+# define MAIN_CNTRL0_SR           (1 << 0)
+# define MAIN_CNTRL0_DECS         (1 << 1)
+# define MAIN_CNTRL0_DEHS         (1 << 2)
+# define MAIN_CNTRL0_CECS         (1 << 3)
+# define MAIN_CNTRL0_CEHS         (1 << 4)
+# define MAIN_CNTRL0_SCALER       (1 << 7)
+#define REG_VERSION_MSB           REG(0x00, 0x02)     /* read */
+#define REG_SOFTRESET             REG(0x00, 0x0a)     /* write */
+# define SOFTRESET_AUDIO          (1 << 0)
+# define SOFTRESET_I2C_MASTER     (1 << 1)
+#define REG_DDC_DISABLE           REG(0x00, 0x0b)     /* read/write */
+#define REG_CCLK_ON               REG(0x00, 0x0c)     /* read/write */
+#define REG_I2C_MASTER            REG(0x00, 0x0d)     /* read/write */
+# define I2C_MASTER_DIS_MM        (1 << 0)
+# define I2C_MASTER_DIS_FILT      (1 << 1)
+# define I2C_MASTER_APP_STRT_LAT  (1 << 2)
+#define REG_INT_FLAGS_0           REG(0x00, 0x0f)     /* read/write */
+#define REG_INT_FLAGS_1           REG(0x00, 0x10)     /* read/write */
+#define REG_INT_FLAGS_2           REG(0x00, 0x11)     /* read/write */
+# define INT_FLAGS_2_EDID_BLK_RD  (1 << 1)
+#define REG_ENA_VP_0              REG(0x00, 0x18)     /* read/write */
+#define REG_ENA_VP_1              REG(0x00, 0x19)     /* read/write */
+#define REG_ENA_VP_2              REG(0x00, 0x1a)     /* read/write */
+#define REG_ENA_AP                REG(0x00, 0x1e)     /* read/write */
+#define REG_VIP_CNTRL_0           REG(0x00, 0x20)     /* write */
+# define VIP_CNTRL_0_MIRR_A       (1 << 7)
+# define VIP_CNTRL_0_SWAP_A(x)    (((x) & 7) << 4)
+# define VIP_CNTRL_0_MIRR_B       (1 << 3)
+# define VIP_CNTRL_0_SWAP_B(x)    (((x) & 7) << 0)
+#define REG_VIP_CNTRL_1           REG(0x00, 0x21)     /* write */
+# define VIP_CNTRL_1_MIRR_C       (1 << 7)
+# define VIP_CNTRL_1_SWAP_C(x)    (((x) & 7) << 4)
+# define VIP_CNTRL_1_MIRR_D       (1 << 3)
+# define VIP_CNTRL_1_SWAP_D(x)    (((x) & 7) << 0)
+#define REG_VIP_CNTRL_2           REG(0x00, 0x22)     /* write */
+# define VIP_CNTRL_2_MIRR_E       (1 << 7)
+# define VIP_CNTRL_2_SWAP_E(x)    (((x) & 7) << 4)
+# define VIP_CNTRL_2_MIRR_F       (1 << 3)
+# define VIP_CNTRL_2_SWAP_F(x)    (((x) & 7) << 0)
+#define REG_VIP_CNTRL_3           REG(0x00, 0x23)     /* write */
+# define VIP_CNTRL_3_X_TGL        (1 << 0)
+# define VIP_CNTRL_3_H_TGL        (1 << 1)
+# define VIP_CNTRL_3_V_TGL        (1 << 2)
+# define VIP_CNTRL_3_EMB          (1 << 3)
+# define VIP_CNTRL_3_SYNC_DE      (1 << 4)
+# define VIP_CNTRL_3_SYNC_HS      (1 << 5)
+# define VIP_CNTRL_3_DE_INT       (1 << 6)
+# define VIP_CNTRL_3_EDGE         (1 << 7)
+#define REG_VIP_CNTRL_4           REG(0x00, 0x24)     /* write */
+# define VIP_CNTRL_4_BLC(x)       (((x) & 3) << 0)
+# define VIP_CNTRL_4_BLANKIT(x)   (((x) & 3) << 2)
+# define VIP_CNTRL_4_CCIR656      (1 << 4)
+# define VIP_CNTRL_4_656_ALT      (1 << 5)
+# define VIP_CNTRL_4_TST_656      (1 << 6)
+# define VIP_CNTRL_4_TST_PAT      (1 << 7)
+#define REG_VIP_CNTRL_5           REG(0x00, 0x25)     /* write */
+# define VIP_CNTRL_5_CKCASE       (1 << 0)
+# define VIP_CNTRL_5_SP_CNT(x)    (((x) & 3) << 1)
+#define REG_MAT_CONTRL            REG(0x00, 0x80)     /* write */
+# define MAT_CONTRL_MAT_SC(x)     (((x) & 3) << 0)
+# define MAT_CONTRL_MAT_BP        (1 << 2)
+#define REG_VIDFORMAT             REG(0x00, 0xa0)     /* write */
+#define REG_REFPIX_MSB            REG(0x00, 0xa1)     /* write */
+#define REG_REFPIX_LSB            REG(0x00, 0xa2)     /* write */
+#define REG_REFLINE_MSB           REG(0x00, 0xa3)     /* write */
+#define REG_REFLINE_LSB           REG(0x00, 0xa4)     /* write */
+#define REG_NPIX_MSB              REG(0x00, 0xa5)     /* write */
+#define REG_NPIX_LSB              REG(0x00, 0xa6)     /* write */
+#define REG_NLINE_MSB             REG(0x00, 0xa7)     /* write */
+#define REG_NLINE_LSB             REG(0x00, 0xa8)     /* write */
+#define REG_VS_LINE_STRT_1_MSB    REG(0x00, 0xa9)     /* write */
+#define REG_VS_LINE_STRT_1_LSB    REG(0x00, 0xaa)     /* write */
+#define REG_VS_PIX_STRT_1_MSB     REG(0x00, 0xab)     /* write */
+#define REG_VS_PIX_STRT_1_LSB     REG(0x00, 0xac)     /* write */
+#define REG_VS_LINE_END_1_MSB     REG(0x00, 0xad)     /* write */
+#define REG_VS_LINE_END_1_LSB     REG(0x00, 0xae)     /* write */
+#define REG_VS_PIX_END_1_MSB      REG(0x00, 0xaf)     /* write */
+#define REG_VS_PIX_END_1_LSB      REG(0x00, 0xb0)     /* write */
+#define REG_VS_PIX_STRT_2_MSB     REG(0x00, 0xb3)     /* write */
+#define REG_VS_PIX_STRT_2_LSB     REG(0x00, 0xb4)     /* write */
+#define REG_VS_PIX_END_2_MSB      REG(0x00, 0xb7)     /* write */
+#define REG_VS_PIX_END_2_LSB      REG(0x00, 0xb8)     /* write */
+#define REG_HS_PIX_START_MSB      REG(0x00, 0xb9)     /* write */
+#define REG_HS_PIX_START_LSB      REG(0x00, 0xba)     /* write */
+#define REG_HS_PIX_STOP_MSB       REG(0x00, 0xbb)     /* write */
+#define REG_HS_PIX_STOP_LSB       REG(0x00, 0xbc)     /* write */
+#define REG_VWIN_START_1_MSB      REG(0x00, 0xbd)     /* write */
+#define REG_VWIN_START_1_LSB      REG(0x00, 0xbe)     /* write */
+#define REG_VWIN_END_1_MSB        REG(0x00, 0xbf)     /* write */
+#define REG_VWIN_END_1_LSB        REG(0x00, 0xc0)     /* write */
+#define REG_DE_START_MSB          REG(0x00, 0xc5)     /* write */
+#define REG_DE_START_LSB          REG(0x00, 0xc6)     /* write */
+#define REG_DE_STOP_MSB           REG(0x00, 0xc7)     /* write */
+#define REG_DE_STOP_LSB           REG(0x00, 0xc8)     /* write */
+#define REG_TBG_CNTRL_0           REG(0x00, 0xca)     /* write */
+# define TBG_CNTRL_0_FRAME_DIS    (1 << 5)
+# define TBG_CNTRL_0_SYNC_MTHD    (1 << 6)
+# define TBG_CNTRL_0_SYNC_ONCE    (1 << 7)
+#define REG_TBG_CNTRL_1           REG(0x00, 0xcb)     /* write */
+# define TBG_CNTRL_1_VH_TGL_0     (1 << 0)
+# define TBG_CNTRL_1_VH_TGL_1     (1 << 1)
+# define TBG_CNTRL_1_VH_TGL_2     (1 << 2)
+# define TBG_CNTRL_1_VHX_EXT_DE   (1 << 3)
+# define TBG_CNTRL_1_VHX_EXT_HS   (1 << 4)
+# define TBG_CNTRL_1_VHX_EXT_VS   (1 << 5)
+# define TBG_CNTRL_1_DWIN_DIS     (1 << 6)
+#define REG_ENABLE_SPACE          REG(0x00, 0xd6)     /* write */
+#define REG_HVF_CNTRL_0           REG(0x00, 0xe4)     /* write */
+# define HVF_CNTRL_0_SM           (1 << 7)
+# define HVF_CNTRL_0_RWB          (1 << 6)
+# define HVF_CNTRL_0_PREFIL(x)    (((x) & 3) << 2)
+# define HVF_CNTRL_0_INTPOL(x)    (((x) & 3) << 0)
+#define REG_HVF_CNTRL_1           REG(0x00, 0xe5)     /* write */
+# define HVF_CNTRL_1_FOR          (1 << 0)
+# define HVF_CNTRL_1_YUVBLK       (1 << 1)
+# define HVF_CNTRL_1_VQR(x)       (((x) & 3) << 2)
+# define HVF_CNTRL_1_PAD(x)       (((x) & 3) << 4)
+# define HVF_CNTRL_1_SEMI_PLANAR  (1 << 6)
+#define REG_RPT_CNTRL             REG(0x00, 0xf0)     /* write */
+
+
+/* Page 02h: PLL settings */
+#define REG_PLL_SERIAL_1          REG(0x02, 0x00)     /* read/write */
+# define PLL_SERIAL_1_SRL_FDN     (1 << 0)
+# define PLL_SERIAL_1_SRL_IZ(x)   (((x) & 3) << 1)
+# define PLL_SERIAL_1_SRL_MAN_IZ  (1 << 6)
+#define REG_PLL_SERIAL_2          REG(0x02, 0x01)     /* read/write */
+# define PLL_SERIAL_2_SRL_NOSC(x) (((x) & 3) << 0)
+# define PLL_SERIAL_2_SRL_PR(x)   (((x) & 0xf) << 4)
+#define REG_PLL_SERIAL_3          REG(0x02, 0x02)     /* read/write */
+# define PLL_SERIAL_3_SRL_CCIR    (1 << 0)
+# define PLL_SERIAL_3_SRL_DE      (1 << 2)
+# define PLL_SERIAL_3_SRL_PXIN_SEL (1 << 4)
+#define REG_SERIALIZER            REG(0x02, 0x03)     /* read/write */
+#define REG_BUFFER_OUT            REG(0x02, 0x04)     /* read/write */
+#define REG_PLL_SCG1              REG(0x02, 0x05)     /* read/write */
+#define REG_PLL_SCG2              REG(0x02, 0x06)     /* read/write */
+#define REG_PLL_SCGN1             REG(0x02, 0x07)     /* read/write */
+#define REG_PLL_SCGN2             REG(0x02, 0x08)     /* read/write */
+#define REG_PLL_SCGR1             REG(0x02, 0x09)     /* read/write */
+#define REG_PLL_SCGR2             REG(0x02, 0x0a)     /* read/write */
+#define REG_AUDIO_DIV             REG(0x02, 0x0e)     /* read/write */
+#define REG_SEL_CLK               REG(0x02, 0x11)     /* read/write */
+# define SEL_CLK_SEL_CLK1         (1 << 0)
+# define SEL_CLK_SEL_VRF_CLK(x)   (((x) & 3) << 1)
+# define SEL_CLK_ENA_SC_CLK       (1 << 3)
+#define REG_ANA_GENERAL           REG(0x02, 0x12)     /* read/write */
+
+
+/* Page 09h: EDID Control */
+#define REG_EDID_DATA_0           REG(0x09, 0x00)     /* read */
+/* next 127 successive registers are the EDID block */
+#define REG_EDID_CTRL             REG(0x09, 0xfa)     /* read/write */
+#define REG_DDC_ADDR              REG(0x09, 0xfb)     /* read/write */
+#define REG_DDC_OFFS              REG(0x09, 0xfc)     /* read/write */
+#define REG_DDC_SEGM_ADDR         REG(0x09, 0xfd)     /* read/write */
+#define REG_DDC_SEGM              REG(0x09, 0xfe)     /* read/write */
+
+
+/* Page 10h: information frames and packets */
+
+
+/* Page 11h: audio settings and content info packets */
+#define REG_AIP_CNTRL_0           REG(0x11, 0x00)     /* read/write */
+# define AIP_CNTRL_0_RST_FIFO     (1 << 0)
+# define AIP_CNTRL_0_SWAP         (1 << 1)
+# define AIP_CNTRL_0_LAYOUT       (1 << 2)
+# define AIP_CNTRL_0_ACR_MAN      (1 << 5)
+# define AIP_CNTRL_0_RST_CTS      (1 << 6)
+#define REG_ENC_CNTRL             REG(0x11, 0x0d)     /* read/write */
+# define ENC_CNTRL_RST_ENC        (1 << 0)
+# define ENC_CNTRL_RST_SEL        (1 << 1)
+# define ENC_CNTRL_CTL_CODE(x)    (((x) & 3) << 2)
+
+
+/* Page 12h: HDCP and OTP */
+#define REG_TX3                   REG(0x12, 0x9a)     /* read/write */
+#define REG_TX33                  REG(0x12, 0xb8)     /* read/write */
+# define TX33_HDMI                (1 << 1)
+
+
+/* Page 13h: Gamut related metadata packets */
+
+
+
+/* CEC registers: (not paged)
+ */
+#define REG_CEC_FRO_IM_CLK_CTRL   0xfb                /* read/write */
+# define CEC_FRO_IM_CLK_CTRL_GHOST_DIS (1 << 7)
+# define CEC_FRO_IM_CLK_CTRL_ENA_OTP   (1 << 6)
+# define CEC_FRO_IM_CLK_CTRL_IMCLK_SEL (1 << 1)
+# define CEC_FRO_IM_CLK_CTRL_FRO_DIV   (1 << 0)
+#define REG_CEC_RXSHPDLEV         0xfe                /* read */
+# define CEC_RXSHPDLEV_RXSENS     (1 << 0)
+# define CEC_RXSHPDLEV_HPD        (1 << 1)
+
+#define REG_CEC_ENAMODS           0xff                /* read/write */
+# define CEC_ENAMODS_DIS_FRO      (1 << 6)
+# define CEC_ENAMODS_DIS_CCLK     (1 << 5)
+# define CEC_ENAMODS_EN_RXSENS    (1 << 2)
+# define CEC_ENAMODS_EN_HDMI      (1 << 1)
+# define CEC_ENAMODS_EN_CEC       (1 << 0)
+
+
+/* Device versions: */
+#define TDA9989N2                 0x0101
+#define TDA19989                  0x0201
+#define TDA19989N2                0x0202
+#define TDA19988                  0x0301
+
+static void
+cec_write(struct drm_encoder *encoder, uint16_t addr, uint8_t val)
+{
+	struct i2c_client *client = to_tda998x_priv(encoder)->cec;
+	uint8_t buf[] = {addr, val};
+	int ret;
+
+	ret = i2c_master_send(client, buf, ARRAY_SIZE(buf));
+	if (ret < 0)
+		dev_err(&client->dev, "Error %d writing to cec:0x%x\n", ret, addr);
+}
+
+static uint8_t
+cec_read(struct drm_encoder *encoder, uint8_t addr)
+{
+	struct i2c_client *client = to_tda998x_priv(encoder)->cec;
+	uint8_t val;
+	int ret;
+
+	ret = i2c_master_send(client, &addr, sizeof(addr));
+	if (ret < 0)
+		goto fail;
+
+	ret = i2c_master_recv(client, &val, sizeof(val));
+	if (ret < 0)
+		goto fail;
+
+	return val;
+
+fail:
+	dev_err(&client->dev, "Error %d reading from cec:0x%x\n", ret, addr);
+	return 0;
+}
+
+static void
+set_page(struct drm_encoder *encoder, uint16_t reg)
+{
+	struct tda998x_priv *priv = to_tda998x_priv(encoder);
+
+	if (REG2PAGE(reg) != priv->current_page) {
+		struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
+		uint8_t buf[] = {
+				REG_CURPAGE, REG2PAGE(reg)
+		};
+		int ret = i2c_master_send(client, buf, sizeof(buf));
+		if (ret < 0)
+			dev_err(&client->dev, "Error %d writing to REG_CURPAGE\n", ret);
+
+		priv->current_page = REG2PAGE(reg);
+	}
+}
+
+static int
+reg_read_range(struct drm_encoder *encoder, uint16_t reg, char *buf, int cnt)
+{
+	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
+	uint8_t addr = REG2ADDR(reg);
+	int ret;
+
+	set_page(encoder, reg);
+
+	ret = i2c_master_send(client, &addr, sizeof(addr));
+	if (ret < 0)
+		goto fail;
+
+	ret = i2c_master_recv(client, buf, cnt);
+	if (ret < 0)
+		goto fail;
+
+	return ret;
+
+fail:
+	dev_err(&client->dev, "Error %d reading from 0x%x\n", ret, reg);
+	return ret;
+}
+
+static uint8_t
+reg_read(struct drm_encoder *encoder, uint16_t reg)
+{
+	uint8_t val = 0;
+	reg_read_range(encoder, reg, &val, sizeof(val));
+	return val;
+}
+
+static void
+reg_write(struct drm_encoder *encoder, uint16_t reg, uint8_t val)
+{
+	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
+	uint8_t buf[] = {REG2ADDR(reg), val};
+	int ret;
+
+	set_page(encoder, reg);
+
+	ret = i2c_master_send(client, buf, ARRAY_SIZE(buf));
+	if (ret < 0)
+		dev_err(&client->dev, "Error %d writing to 0x%x\n", ret, reg);
+}
+
+static void
+reg_write16(struct drm_encoder *encoder, uint16_t reg, uint16_t val)
+{
+	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
+	uint8_t buf[] = {REG2ADDR(reg), val >> 8, val};
+	int ret;
+
+	set_page(encoder, reg);
+
+	ret = i2c_master_send(client, buf, ARRAY_SIZE(buf));
+	if (ret < 0)
+		dev_err(&client->dev, "Error %d writing to 0x%x\n", ret, reg);
+}
+
+static void
+reg_set(struct drm_encoder *encoder, uint16_t reg, uint8_t val)
+{
+	reg_write(encoder, reg, reg_read(encoder, reg) | val);
+}
+
+static void
+reg_clear(struct drm_encoder *encoder, uint16_t reg, uint8_t val)
+{
+	reg_write(encoder, reg, reg_read(encoder, reg) & ~val);
+}
+
+static void
+tda998x_reset(struct drm_encoder *encoder)
+{
+	/* reset audio and i2c master: */
+	reg_set(encoder, REG_SOFTRESET, SOFTRESET_AUDIO | SOFTRESET_I2C_MASTER);
+	msleep(50);
+	reg_clear(encoder, REG_SOFTRESET, SOFTRESET_AUDIO | SOFTRESET_I2C_MASTER);
+	msleep(50);
+
+	/* reset transmitter: */
+	reg_set(encoder, REG_MAIN_CNTRL0, MAIN_CNTRL0_SR);
+	reg_clear(encoder, REG_MAIN_CNTRL0, MAIN_CNTRL0_SR);
+
+	/* PLL registers common configuration */
+	reg_write(encoder, REG_PLL_SERIAL_1, 0x00);
+	reg_write(encoder, REG_PLL_SERIAL_2, PLL_SERIAL_2_SRL_NOSC(1));
+	reg_write(encoder, REG_PLL_SERIAL_3, 0x00);
+	reg_write(encoder, REG_SERIALIZER,   0x00);
+	reg_write(encoder, REG_BUFFER_OUT,   0x00);
+	reg_write(encoder, REG_PLL_SCG1,     0x00);
+	reg_write(encoder, REG_AUDIO_DIV,    0x03);
+	reg_write(encoder, REG_SEL_CLK,      SEL_CLK_SEL_CLK1 | SEL_CLK_ENA_SC_CLK);
+	reg_write(encoder, REG_PLL_SCGN1,    0xfa);
+	reg_write(encoder, REG_PLL_SCGN2,    0x00);
+	reg_write(encoder, REG_PLL_SCGR1,    0x5b);
+	reg_write(encoder, REG_PLL_SCGR2,    0x00);
+	reg_write(encoder, REG_PLL_SCG2,     0x10);
+}
+
+/* DRM encoder functions */
+
+static void
+tda998x_encoder_set_config(struct drm_encoder *encoder, void *params)
+{
+}
+
+static void
+tda998x_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct tda998x_priv *priv = to_tda998x_priv(encoder);
+
+	/* we only care about on or off: */
+	if (mode != DRM_MODE_DPMS_ON)
+		mode = DRM_MODE_DPMS_OFF;
+
+	if (mode == priv->dpms)
+		return;
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		/* enable audio and video ports */
+		reg_write(encoder, REG_ENA_AP, 0xff);
+		reg_write(encoder, REG_ENA_VP_0, 0xff);
+		reg_write(encoder, REG_ENA_VP_1, 0xff);
+		reg_write(encoder, REG_ENA_VP_2, 0xff);
+		/* set muxing after enabling ports: */
+		reg_write(encoder, REG_VIP_CNTRL_0,
+				VIP_CNTRL_0_SWAP_A(2) | VIP_CNTRL_0_SWAP_B(3));
+		reg_write(encoder, REG_VIP_CNTRL_1,
+				VIP_CNTRL_1_SWAP_C(0) | VIP_CNTRL_1_SWAP_D(1));
+		reg_write(encoder, REG_VIP_CNTRL_2,
+				VIP_CNTRL_2_SWAP_E(4) | VIP_CNTRL_2_SWAP_F(5));
+		break;
+	case DRM_MODE_DPMS_OFF:
+		/* disable audio and video ports */
+		reg_write(encoder, REG_ENA_AP, 0x00);
+		reg_write(encoder, REG_ENA_VP_0, 0x00);
+		reg_write(encoder, REG_ENA_VP_1, 0x00);
+		reg_write(encoder, REG_ENA_VP_2, 0x00);
+		break;
+	}
+
+	priv->dpms = mode;
+}
+
+static void
+tda998x_encoder_save(struct drm_encoder *encoder)
+{
+	DBG("");
+}
+
+static void
+tda998x_encoder_restore(struct drm_encoder *encoder)
+{
+	DBG("");
+}
+
+static bool
+tda998x_encoder_mode_fixup(struct drm_encoder *encoder,
+			  const struct drm_display_mode *mode,
+			  struct drm_display_mode *adjusted_mode)
+{
+	return true;
+}
+
+static int
+tda998x_encoder_mode_valid(struct drm_encoder *encoder,
+			  struct drm_display_mode *mode)
+{
+	return MODE_OK;
+}
+
+static void
+tda998x_encoder_mode_set(struct drm_encoder *encoder,
+			struct drm_display_mode *mode,
+			struct drm_display_mode *adjusted_mode)
+{
+	struct tda998x_priv *priv = to_tda998x_priv(encoder);
+	uint16_t hs_start, hs_end, line_start, line_end;
+	uint16_t vwin_start, vwin_end, de_start, de_end;
+	uint16_t ref_pix, ref_line, pix_start2;
+	uint8_t reg, div, rep;
+
+	hs_start   = mode->hsync_start - mode->hdisplay;
+	hs_end     = mode->hsync_end - mode->hdisplay;
+	line_start = 1;
+	line_end   = 1 + mode->vsync_end - mode->vsync_start;
+	vwin_start = mode->vtotal - mode->vsync_start;
+	vwin_end   = vwin_start + mode->vdisplay;
+	de_start   = mode->htotal - mode->hdisplay;
+	de_end     = mode->htotal;
+
+	pix_start2 = 0;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		pix_start2 = (mode->htotal / 2) + hs_start;
+
+	/* TODO how is this value calculated?  It is 2 for all common
+	 * formats in the tables in out of tree nxp driver (assuming
+	 * I've properly deciphered their byzantine table system)
+	 */
+	ref_line = 2;
+
+	/* this might changes for other color formats from the CRTC: */
+	ref_pix = 3 + hs_start;
+
+	div = 148500 / mode->clock;
+
+	DBG("clock=%d, div=%u", mode->clock, div);
+	DBG("hs_start=%u, hs_end=%u, line_start=%u, line_end=%u",
+			hs_start, hs_end, line_start, line_end);
+	DBG("vwin_start=%u, vwin_end=%u, de_start=%u, de_end=%u",
+			vwin_start, vwin_end, de_start, de_end);
+	DBG("ref_line=%u, ref_pix=%u, pix_start2=%u",
+			ref_line, ref_pix, pix_start2);
+
+	/* mute the audio FIFO: */
+	reg_set(encoder, REG_AIP_CNTRL_0, AIP_CNTRL_0_RST_FIFO);
+
+	/* set HDMI HDCP mode off: */
+	reg_set(encoder, REG_TBG_CNTRL_1, TBG_CNTRL_1_DWIN_DIS);
+	reg_clear(encoder, REG_TX33, TX33_HDMI);
+
+	reg_write(encoder, REG_ENC_CNTRL, ENC_CNTRL_CTL_CODE(0));
+	/* no pre-filter or interpolator: */
+	reg_write(encoder, REG_HVF_CNTRL_0, HVF_CNTRL_0_PREFIL(0) |
+			HVF_CNTRL_0_INTPOL(0));
+	reg_write(encoder, REG_VIP_CNTRL_5, VIP_CNTRL_5_SP_CNT(0));
+	reg_write(encoder, REG_VIP_CNTRL_4, VIP_CNTRL_4_BLANKIT(0) |
+			VIP_CNTRL_4_BLC(0));
+	reg_clear(encoder, REG_PLL_SERIAL_3, PLL_SERIAL_3_SRL_CCIR);
+
+	reg_clear(encoder, REG_PLL_SERIAL_1, PLL_SERIAL_1_SRL_MAN_IZ);
+	reg_clear(encoder, REG_PLL_SERIAL_3, PLL_SERIAL_3_SRL_DE);
+	reg_write(encoder, REG_SERIALIZER, 0);
+	reg_write(encoder, REG_HVF_CNTRL_1, HVF_CNTRL_1_VQR(0));
+
+	/* TODO enable pixel repeat for pixel rates less than 25Msamp/s */
+	rep = 0;
+	reg_write(encoder, REG_RPT_CNTRL, 0);
+	reg_write(encoder, REG_SEL_CLK, SEL_CLK_SEL_VRF_CLK(0) |
+			SEL_CLK_SEL_CLK1 | SEL_CLK_ENA_SC_CLK);
+
+	reg_write(encoder, REG_PLL_SERIAL_2, PLL_SERIAL_2_SRL_NOSC(div) |
+			PLL_SERIAL_2_SRL_PR(rep));
+
+	reg_write16(encoder, REG_VS_PIX_STRT_2_MSB, pix_start2);
+	reg_write16(encoder, REG_VS_PIX_END_2_MSB, pix_start2);
+
+	/* set color matrix bypass flag: */
+	reg_set(encoder, REG_MAT_CONTRL, MAT_CONTRL_MAT_BP);
+
+	/* set BIAS tmds value: */
+	reg_write(encoder, REG_ANA_GENERAL, 0x09);
+
+	reg_clear(encoder, REG_TBG_CNTRL_0, TBG_CNTRL_0_SYNC_MTHD);
+
+	reg_write(encoder, REG_VIP_CNTRL_3, 0);
+	reg_set(encoder, REG_VIP_CNTRL_3, VIP_CNTRL_3_SYNC_HS);
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		reg_set(encoder, REG_VIP_CNTRL_3, VIP_CNTRL_3_V_TGL);
+
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		reg_set(encoder, REG_VIP_CNTRL_3, VIP_CNTRL_3_H_TGL);
+
+	reg_write(encoder, REG_VIDFORMAT, 0x00);
+	reg_write16(encoder, REG_NPIX_MSB, mode->hdisplay - 1);
+	reg_write16(encoder, REG_NLINE_MSB, mode->vdisplay - 1);
+	reg_write16(encoder, REG_VS_LINE_STRT_1_MSB, line_start);
+	reg_write16(encoder, REG_VS_LINE_END_1_MSB, line_end);
+	reg_write16(encoder, REG_VS_PIX_STRT_1_MSB, hs_start);
+	reg_write16(encoder, REG_VS_PIX_END_1_MSB, hs_start);
+	reg_write16(encoder, REG_HS_PIX_START_MSB, hs_start);
+	reg_write16(encoder, REG_HS_PIX_STOP_MSB, hs_end);
+	reg_write16(encoder, REG_VWIN_START_1_MSB, vwin_start);
+	reg_write16(encoder, REG_VWIN_END_1_MSB, vwin_end);
+	reg_write16(encoder, REG_DE_START_MSB, de_start);
+	reg_write16(encoder, REG_DE_STOP_MSB, de_end);
+
+	if (priv->rev == TDA19988) {
+		/* let incoming pixels fill the active space (if any) */
+		reg_write(encoder, REG_ENABLE_SPACE, 0x01);
+	}
+
+	reg_write16(encoder, REG_REFPIX_MSB, ref_pix);
+	reg_write16(encoder, REG_REFLINE_MSB, ref_line);
+
+	reg = TBG_CNTRL_1_VHX_EXT_DE |
+			TBG_CNTRL_1_VHX_EXT_HS |
+			TBG_CNTRL_1_VHX_EXT_VS |
+			TBG_CNTRL_1_DWIN_DIS | /* HDCP off */
+			TBG_CNTRL_1_VH_TGL_2;
+	if (mode->flags & (DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC))
+		reg |= TBG_CNTRL_1_VH_TGL_0;
+	reg_set(encoder, REG_TBG_CNTRL_1, reg);
+
+	/* must be last register set: */
+	reg_clear(encoder, REG_TBG_CNTRL_0, TBG_CNTRL_0_SYNC_ONCE);
+}
+
+static enum drm_connector_status
+tda998x_encoder_detect(struct drm_encoder *encoder,
+		      struct drm_connector *connector)
+{
+	uint8_t val = cec_read(encoder, REG_CEC_RXSHPDLEV);
+	return (val & CEC_RXSHPDLEV_HPD) ? connector_status_connected :
+			connector_status_disconnected;
+}
+
+static int
+read_edid_block(struct drm_encoder *encoder, uint8_t *buf, int blk)
+{
+	uint8_t offset, segptr;
+	int ret, i;
+
+	/* enable EDID read irq: */
+	reg_set(encoder, REG_INT_FLAGS_2, INT_FLAGS_2_EDID_BLK_RD);
+
+	offset = (blk & 1) ? 128 : 0;
+	segptr = blk / 2;
+
+	reg_write(encoder, REG_DDC_ADDR, 0xa0);
+	reg_write(encoder, REG_DDC_OFFS, offset);
+	reg_write(encoder, REG_DDC_SEGM_ADDR, 0x60);
+	reg_write(encoder, REG_DDC_SEGM, segptr);
+
+	/* enable reading EDID: */
+	reg_write(encoder, REG_EDID_CTRL, 0x1);
+
+	/* flag must be cleared by sw: */
+	reg_write(encoder, REG_EDID_CTRL, 0x0);
+
+	/* wait for block read to complete: */
+	for (i = 100; i > 0; i--) {
+		uint8_t val = reg_read(encoder, REG_INT_FLAGS_2);
+		if (val & INT_FLAGS_2_EDID_BLK_RD)
+			break;
+		msleep(1);
+	}
+
+	if (i == 0)
+		return -ETIMEDOUT;
+
+	ret = reg_read_range(encoder, REG_EDID_DATA_0, buf, EDID_LENGTH);
+	if (ret != EDID_LENGTH) {
+		dev_err(encoder->dev->dev, "failed to read edid block %d: %d",
+				blk, ret);
+		return ret;
+	}
+
+	reg_clear(encoder, REG_INT_FLAGS_2, INT_FLAGS_2_EDID_BLK_RD);
+
+	return 0;
+}
+
+static uint8_t *
+do_get_edid(struct drm_encoder *encoder)
+{
+	int j = 0, valid_extensions = 0;
+	uint8_t *block, *new;
+	bool print_bad_edid = drm_debug & DRM_UT_KMS;
+
+	if ((block = kmalloc(EDID_LENGTH, GFP_KERNEL)) == NULL)
+		return NULL;
+
+	/* base block fetch */
+	if (read_edid_block(encoder, block, 0))
+		goto fail;
+
+	if (!drm_edid_block_valid(block, 0, print_bad_edid))
+		goto fail;
+
+	/* if there's no extensions, we're done */
+	if (block[0x7e] == 0)
+		return block;
+
+	new = krealloc(block, (block[0x7e] + 1) * EDID_LENGTH, GFP_KERNEL);
+	if (!new)
+		goto fail;
+	block = new;
+
+	for (j = 1; j <= block[0x7e]; j++) {
+		uint8_t *ext_block = block + (valid_extensions + 1) * EDID_LENGTH;
+		if (read_edid_block(encoder, ext_block, j))
+			goto fail;
+
+		if (!drm_edid_block_valid(ext_block, j, print_bad_edid))
+			goto fail;
+
+		valid_extensions++;
+	}
+
+	if (valid_extensions != block[0x7e]) {
+		block[EDID_LENGTH-1] += block[0x7e] - valid_extensions;
+		block[0x7e] = valid_extensions;
+		new = krealloc(block, (valid_extensions + 1) * EDID_LENGTH, GFP_KERNEL);
+		if (!new)
+			goto fail;
+		block = new;
+	}
+
+	return block;
+
+fail:
+	dev_warn(encoder->dev->dev, "failed to read EDID\n");
+	kfree(block);
+	return NULL;
+}
+
+static int
+tda998x_encoder_get_modes(struct drm_encoder *encoder,
+			 struct drm_connector *connector)
+{
+	struct edid *edid = (struct edid *)do_get_edid(encoder);
+	int n = 0;
+
+	if (edid) {
+		drm_mode_connector_update_edid_property(connector, edid);
+		n = drm_add_edid_modes(connector, edid);
+		kfree(edid);
+	}
+
+	return n;
+}
+
+static int
+tda998x_encoder_create_resources(struct drm_encoder *encoder,
+				struct drm_connector *connector)
+{
+	DBG("");
+	return 0;
+}
+
+static int
+tda998x_encoder_set_property(struct drm_encoder *encoder,
+			    struct drm_connector *connector,
+			    struct drm_property *property,
+			    uint64_t val)
+{
+	DBG("");
+	return 0;
+}
+
+static void
+tda998x_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct tda998x_priv *priv = to_tda998x_priv(encoder);
+	drm_i2c_encoder_destroy(encoder);
+	kfree(priv);
+}
+
+static struct drm_encoder_slave_funcs tda998x_encoder_funcs = {
+	.set_config = tda998x_encoder_set_config,
+	.destroy = tda998x_encoder_destroy,
+	.dpms = tda998x_encoder_dpms,
+	.save = tda998x_encoder_save,
+	.restore = tda998x_encoder_restore,
+	.mode_fixup = tda998x_encoder_mode_fixup,
+	.mode_valid = tda998x_encoder_mode_valid,
+	.mode_set = tda998x_encoder_mode_set,
+	.detect = tda998x_encoder_detect,
+	.get_modes = tda998x_encoder_get_modes,
+	.create_resources = tda998x_encoder_create_resources,
+	.set_property = tda998x_encoder_set_property,
+};
+
+/* I2C driver functions */
+
+static int
+tda998x_probe(struct i2c_client *client, const struct i2c_device_id *id)
+{
+	return 0;
+}
+
+static int
+tda998x_remove(struct i2c_client *client)
+{
+	return 0;
+}
+
+static int
+tda998x_encoder_init(struct i2c_client *client,
+		    struct drm_device *dev,
+		    struct drm_encoder_slave *encoder_slave)
+{
+	struct drm_encoder *encoder = &encoder_slave->base;
+	struct tda998x_priv *priv;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->current_page = 0;
+	priv->cec = i2c_new_dummy(client->adapter, 0x34);
+	priv->dpms = DRM_MODE_DPMS_OFF;
+
+	encoder_slave->slave_priv = priv;
+	encoder_slave->slave_funcs = &tda998x_encoder_funcs;
+
+	/* wake up the device: */
+	cec_write(encoder, REG_CEC_ENAMODS,
+			CEC_ENAMODS_EN_RXSENS | CEC_ENAMODS_EN_HDMI);
+
+	tda998x_reset(encoder);
+
+	/* read version: */
+	priv->rev = reg_read(encoder, REG_VERSION_LSB) |
+			reg_read(encoder, REG_VERSION_MSB) << 8;
+
+	/* mask off feature bits: */
+	priv->rev &= ~0x30; /* not-hdcp and not-scalar bit */
+
+	switch (priv->rev) {
+	case TDA9989N2:  dev_info(dev->dev, "found TDA9989 n2");  break;
+	case TDA19989:   dev_info(dev->dev, "found TDA19989");    break;
+	case TDA19989N2: dev_info(dev->dev, "found TDA19989 n2"); break;
+	case TDA19988:   dev_info(dev->dev, "found TDA19988");    break;
+	default:
+		DBG("found unsupported device: %04x", priv->rev);
+		goto fail;
+	}
+
+	/* after reset, enable DDC: */
+	reg_write(encoder, REG_DDC_DISABLE, 0x00);
+
+	/* set clock on DDC channel: */
+	reg_write(encoder, REG_TX3, 39);
+
+	/* if necessary, disable multi-master: */
+	if (priv->rev == TDA19989)
+		reg_set(encoder, REG_I2C_MASTER, I2C_MASTER_DIS_MM);
+
+	cec_write(encoder, REG_CEC_FRO_IM_CLK_CTRL,
+			CEC_FRO_IM_CLK_CTRL_GHOST_DIS | CEC_FRO_IM_CLK_CTRL_IMCLK_SEL);
+
+	return 0;
+
+fail:
+	/* if encoder_init fails, the encoder slave is never registered,
+	 * so cleanup here:
+	 */
+	if (priv->cec)
+		i2c_unregister_device(priv->cec);
+	kfree(priv);
+	encoder_slave->slave_priv = NULL;
+	encoder_slave->slave_funcs = NULL;
+	return -ENXIO;
+}
+
+static struct i2c_device_id tda998x_ids[] = {
+	{ "tda998x", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, tda998x_ids);
+
+static struct drm_i2c_encoder_driver tda998x_driver = {
+	.i2c_driver = {
+		.probe = tda998x_probe,
+		.remove = tda998x_remove,
+		.driver = {
+			.name = "tda998x",
+		},
+		.id_table = tda998x_ids,
+	},
+	.encoder_init = tda998x_encoder_init,
+};
+
+/* Module initialization */
+
+static int __init
+tda998x_init(void)
+{
+	DBG("");
+	return drm_i2c_encoder_register(THIS_MODULE, &tda998x_driver);
+}
+
+static void __exit
+tda998x_exit(void)
+{
+	DBG("");
+	drm_i2c_encoder_unregister(&tda998x_driver);
+}
+
+MODULE_AUTHOR("Rob Clark <robdclark@gmail.com");
+MODULE_DESCRIPTION("NXP Semiconductors TDA998X HDMI Encoder");
+MODULE_LICENSE("GPL");
+
+module_init(tda998x_init);
+module_exit(tda998x_exit);
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 0f2c5493242b..91f3ac6cef35 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -16,6 +16,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \
 	  i915_gem_tiling.o \
 	  i915_sysfs.o \
 	  i915_trace_points.o \
+	  i915_ums.o \
 	  intel_display.o \
 	  intel_crt.o \
 	  intel_lvds.o \
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 32158d21c632..aae31489c893 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -103,7 +103,7 @@ static const char *cache_level_str(int type)
 static void
 describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 {
-	seq_printf(m, "%p: %s%s %8zdKiB %04x %04x %d %d %d%s%s%s",
+	seq_printf(m, "%p: %s%s %8zdKiB %02x %02x %d %d %d%s%s%s",
 		   &obj->base,
 		   get_pin_flag(obj),
 		   get_tiling_flag(obj),
@@ -125,6 +125,8 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 	if (obj->gtt_space != NULL)
 		seq_printf(m, " (gtt offset: %08x, size: %08x)",
 			   obj->gtt_offset, (unsigned int)obj->gtt_space->size);
+	if (obj->stolen)
+		seq_printf(m, " (stolen: %08lx)", obj->stolen->start);
 	if (obj->pin_mappable || obj->fault_mappable) {
 		char s[3], *t = s;
 		if (obj->pin_mappable)
@@ -257,8 +259,9 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 	seq_printf(m, "%u fault mappable objects, %zu bytes\n",
 		   count, size);
 
-	seq_printf(m, "%zu [%zu] gtt total\n",
-		   dev_priv->mm.gtt_total, dev_priv->mm.mappable_gtt_total);
+	seq_printf(m, "%zu [%lu] gtt total\n",
+		   dev_priv->gtt.total,
+		   dev_priv->gtt.mappable_end - dev_priv->gtt.start);
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -388,7 +391,7 @@ static void i915_ring_seqno_info(struct seq_file *m,
 				 struct intel_ring_buffer *ring)
 {
 	if (ring->get_seqno) {
-		seq_printf(m, "Current sequence (%s): %d\n",
+		seq_printf(m, "Current sequence (%s): %u\n",
 			   ring->name, ring->get_seqno(ring, false));
 	}
 }
@@ -545,11 +548,11 @@ static int i915_hws_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
-	const volatile u32 __iomem *hws;
+	const u32 *hws;
 	int i;
 
 	ring = &dev_priv->ring[(uintptr_t)node->info_ent->data];
-	hws = (volatile u32 __iomem *)ring->status_page.page_addr;
+	hws = ring->status_page.page_addr;
 	if (hws == NULL)
 		return 0;
 
@@ -609,7 +612,7 @@ static void print_error_buffers(struct seq_file *m,
 	seq_printf(m, "%s [%d]:\n", name, count);
 
 	while (count--) {
-		seq_printf(m, "  %08x %8u %04x %04x %x %x%s%s%s%s%s%s%s",
+		seq_printf(m, "  %08x %8u %02x %02x %x %x%s%s%s%s%s%s%s",
 			   err->gtt_offset,
 			   err->size,
 			   err->read_domains,
@@ -691,7 +694,7 @@ static int i915_error_state(struct seq_file *m, void *unused)
 
 	seq_printf(m, "Time: %ld s %ld us\n", error->time.tv_sec,
 		   error->time.tv_usec);
-	seq_printf(m, "Kernel: " UTS_RELEASE);
+	seq_printf(m, "Kernel: " UTS_RELEASE "\n");
 	seq_printf(m, "PCI ID: 0x%04x\n", dev->pci_device);
 	seq_printf(m, "EIR: 0x%08x\n", error->eir);
 	seq_printf(m, "IER: 0x%08x\n", error->ier);
@@ -816,11 +819,11 @@ static int i915_error_state_open(struct inode *inode, struct file *file)
 
 	error_priv->dev = dev;
 
-	spin_lock_irqsave(&dev_priv->error_lock, flags);
-	error_priv->error = dev_priv->first_error;
+	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
+	error_priv->error = dev_priv->gpu_error.first_error;
 	if (error_priv->error)
 		kref_get(&error_priv->error->ref);
-	spin_unlock_irqrestore(&dev_priv->error_lock, flags);
+	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
 
 	return single_open(file, i915_error_state, error_priv);
 }
@@ -846,6 +849,77 @@ static const struct file_operations i915_error_state_fops = {
 	.release = i915_error_state_release,
 };
 
+static ssize_t
+i915_next_seqno_read(struct file *filp,
+		 char __user *ubuf,
+		 size_t max,
+		 loff_t *ppos)
+{
+	struct drm_device *dev = filp->private_data;
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	char buf[80];
+	int len;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	len = snprintf(buf, sizeof(buf),
+		       "next_seqno :  0x%x\n",
+		       dev_priv->next_seqno);
+
+	mutex_unlock(&dev->struct_mutex);
+
+	if (len > sizeof(buf))
+		len = sizeof(buf);
+
+	return simple_read_from_buffer(ubuf, max, ppos, buf, len);
+}
+
+static ssize_t
+i915_next_seqno_write(struct file *filp,
+		      const char __user *ubuf,
+		      size_t cnt,
+		      loff_t *ppos)
+{
+	struct drm_device *dev = filp->private_data;
+	char buf[20];
+	u32 val = 1;
+	int ret;
+
+	if (cnt > 0) {
+		if (cnt > sizeof(buf) - 1)
+			return -EINVAL;
+
+		if (copy_from_user(buf, ubuf, cnt))
+			return -EFAULT;
+		buf[cnt] = 0;
+
+		ret = kstrtouint(buf, 0, &val);
+		if (ret < 0)
+			return ret;
+	}
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	ret = i915_gem_set_seqno(dev, val);
+
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret ?: cnt;
+}
+
+static const struct file_operations i915_next_seqno_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read = i915_next_seqno_read,
+	.write = i915_next_seqno_write,
+	.llseek = default_llseek,
+};
+
 static int i915_rstdby_delays(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -888,7 +962,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 		u32 rp_state_limits = I915_READ(GEN6_RP_STATE_LIMITS);
 		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
-		u32 rpstat;
+		u32 rpstat, cagf;
 		u32 rpupei, rpcurup, rpprevup;
 		u32 rpdownei, rpcurdown, rpprevdown;
 		int max_freq;
@@ -907,6 +981,11 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		rpdownei = I915_READ(GEN6_RP_CUR_DOWN_EI);
 		rpcurdown = I915_READ(GEN6_RP_CUR_DOWN);
 		rpprevdown = I915_READ(GEN6_RP_PREV_DOWN);
+		if (IS_HASWELL(dev))
+			cagf = (rpstat & HSW_CAGF_MASK) >> HSW_CAGF_SHIFT;
+		else
+			cagf = (rpstat & GEN6_CAGF_MASK) >> GEN6_CAGF_SHIFT;
+		cagf *= GT_FREQUENCY_MULTIPLIER;
 
 		gen6_gt_force_wake_put(dev_priv);
 		mutex_unlock(&dev->struct_mutex);
@@ -919,8 +998,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 			   gt_perf_status & 0xff);
 		seq_printf(m, "Render p-state limit: %d\n",
 			   rp_state_limits & 0xff);
-		seq_printf(m, "CAGF: %dMHz\n", ((rpstat & GEN6_CAGF_MASK) >>
-						GEN6_CAGF_SHIFT) * GT_FREQUENCY_MULTIPLIER);
+		seq_printf(m, "CAGF: %dMHz\n", cagf);
 		seq_printf(m, "RP CUR UP EI: %dus\n", rpupei &
 			   GEN6_CURICONT_MASK);
 		seq_printf(m, "RP CUR UP: %dus\n", rpcurup &
@@ -1372,28 +1450,31 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 	ifbdev = dev_priv->fbdev;
 	fb = to_intel_framebuffer(ifbdev->helper.fb);
 
-	seq_printf(m, "fbcon size: %d x %d, depth %d, %d bpp, obj ",
+	seq_printf(m, "fbcon size: %d x %d, depth %d, %d bpp, refcount %d, obj ",
 		   fb->base.width,
 		   fb->base.height,
 		   fb->base.depth,
-		   fb->base.bits_per_pixel);
+		   fb->base.bits_per_pixel,
+		   atomic_read(&fb->base.refcount.refcount));
 	describe_obj(m, fb->obj);
 	seq_printf(m, "\n");
+	mutex_unlock(&dev->mode_config.mutex);
 
+	mutex_lock(&dev->mode_config.fb_lock);
 	list_for_each_entry(fb, &dev->mode_config.fb_list, base.head) {
 		if (&fb->base == ifbdev->helper.fb)
 			continue;
 
-		seq_printf(m, "user size: %d x %d, depth %d, %d bpp, obj ",
+		seq_printf(m, "user size: %d x %d, depth %d, %d bpp, refcount %d, obj ",
 			   fb->base.width,
 			   fb->base.height,
 			   fb->base.depth,
-			   fb->base.bits_per_pixel);
+			   fb->base.bits_per_pixel,
+			   atomic_read(&fb->base.refcount.refcount));
 		describe_obj(m, fb->obj);
 		seq_printf(m, "\n");
 	}
-
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&dev->mode_config.fb_lock);
 
 	return 0;
 }
@@ -1403,7 +1484,8 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	int ret;
+	struct intel_ring_buffer *ring;
+	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
 	if (ret)
@@ -1421,6 +1503,14 @@ static int i915_context_status(struct seq_file *m, void *unused)
 		seq_printf(m, "\n");
 	}
 
+	for_each_ring(ring, dev_priv, i) {
+		if (ring->default_context) {
+			seq_printf(m, "HW default context %s ring ", ring->name);
+			describe_obj(m, ring->default_context->obj);
+			seq_printf(m, "\n");
+		}
+	}
+
 	mutex_unlock(&dev->mode_config.mutex);
 
 	return 0;
@@ -1556,7 +1646,7 @@ static int i915_dpio_info(struct seq_file *m, void *data)
 		return 0;
 	}
 
-	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
+	ret = mutex_lock_interruptible(&dev_priv->dpio_lock);
 	if (ret)
 		return ret;
 
@@ -1585,7 +1675,7 @@ static int i915_dpio_info(struct seq_file *m, void *data)
 	seq_printf(m, "DPIO_FASTCLK_DISABLE: 0x%08x\n",
 		   intel_dpio_read(dev_priv, DPIO_FASTCLK_DISABLE));
 
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&dev_priv->dpio_lock);
 
 	return 0;
 }
@@ -1603,7 +1693,7 @@ i915_wedged_read(struct file *filp,
 
 	len = snprintf(buf, sizeof(buf),
 		       "wedged :  %d\n",
-		       atomic_read(&dev_priv->mm.wedged));
+		       atomic_read(&dev_priv->gpu_error.reset_counter));
 
 	if (len > sizeof(buf))
 		len = sizeof(buf);
@@ -1658,7 +1748,7 @@ i915_ring_stop_read(struct file *filp,
 	int len;
 
 	len = snprintf(buf, sizeof(buf),
-		       "0x%08x\n", dev_priv->stop_rings);
+		       "0x%08x\n", dev_priv->gpu_error.stop_rings);
 
 	if (len > sizeof(buf))
 		len = sizeof(buf);
@@ -1694,7 +1784,7 @@ i915_ring_stop_write(struct file *filp,
 	if (ret)
 		return ret;
 
-	dev_priv->stop_rings = val;
+	dev_priv->gpu_error.stop_rings = val;
 	mutex_unlock(&dev->struct_mutex);
 
 	return cnt;
@@ -1708,6 +1798,102 @@ static const struct file_operations i915_ring_stop_fops = {
 	.llseek = default_llseek,
 };
 
+#define DROP_UNBOUND 0x1
+#define DROP_BOUND 0x2
+#define DROP_RETIRE 0x4
+#define DROP_ACTIVE 0x8
+#define DROP_ALL (DROP_UNBOUND | \
+		  DROP_BOUND | \
+		  DROP_RETIRE | \
+		  DROP_ACTIVE)
+static ssize_t
+i915_drop_caches_read(struct file *filp,
+		      char __user *ubuf,
+		      size_t max,
+		      loff_t *ppos)
+{
+	char buf[20];
+	int len;
+
+	len = snprintf(buf, sizeof(buf), "0x%08x\n", DROP_ALL);
+	if (len > sizeof(buf))
+		len = sizeof(buf);
+
+	return simple_read_from_buffer(ubuf, max, ppos, buf, len);
+}
+
+static ssize_t
+i915_drop_caches_write(struct file *filp,
+		       const char __user *ubuf,
+		       size_t cnt,
+		       loff_t *ppos)
+{
+	struct drm_device *dev = filp->private_data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_object *obj, *next;
+	char buf[20];
+	int val = 0, ret;
+
+	if (cnt > 0) {
+		if (cnt > sizeof(buf) - 1)
+			return -EINVAL;
+
+		if (copy_from_user(buf, ubuf, cnt))
+			return -EFAULT;
+		buf[cnt] = 0;
+
+		val = simple_strtoul(buf, NULL, 0);
+	}
+
+	DRM_DEBUG_DRIVER("Dropping caches: 0x%08x\n", val);
+
+	/* No need to check and wait for gpu resets, only libdrm auto-restarts
+	 * on ioctls on -EAGAIN. */
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	if (val & DROP_ACTIVE) {
+		ret = i915_gpu_idle(dev);
+		if (ret)
+			goto unlock;
+	}
+
+	if (val & (DROP_RETIRE | DROP_ACTIVE))
+		i915_gem_retire_requests(dev);
+
+	if (val & DROP_BOUND) {
+		list_for_each_entry_safe(obj, next, &dev_priv->mm.inactive_list, mm_list)
+			if (obj->pin_count == 0) {
+				ret = i915_gem_object_unbind(obj);
+				if (ret)
+					goto unlock;
+			}
+	}
+
+	if (val & DROP_UNBOUND) {
+		list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list, gtt_list)
+			if (obj->pages_pin_count == 0) {
+				ret = i915_gem_object_put_pages(obj);
+				if (ret)
+					goto unlock;
+			}
+	}
+
+unlock:
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret ?: cnt;
+}
+
+static const struct file_operations i915_drop_caches_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read = i915_drop_caches_read,
+	.write = i915_drop_caches_write,
+	.llseek = default_llseek,
+};
+
 static ssize_t
 i915_max_freq_read(struct file *filp,
 		   char __user *ubuf,
@@ -2105,11 +2291,23 @@ int i915_debugfs_init(struct drm_minor *minor)
 		return ret;
 
 	ret = i915_debugfs_create(minor->debugfs_root, minor,
+				  "i915_gem_drop_caches",
+				  &i915_drop_caches_fops);
+	if (ret)
+		return ret;
+
+	ret = i915_debugfs_create(minor->debugfs_root, minor,
 				  "i915_error_state",
 				  &i915_error_state_fops);
 	if (ret)
 		return ret;
 
+	ret = i915_debugfs_create(minor->debugfs_root, minor,
+				 "i915_next_seqno",
+				 &i915_next_seqno_fops);
+	if (ret)
+		return ret;
+
 	return drm_debugfs_create_files(i915_debugfs_list,
 					I915_DEBUGFS_ENTRIES,
 					minor->debugfs_root, minor);
@@ -2129,10 +2327,14 @@ void i915_debugfs_cleanup(struct drm_minor *minor)
 				 1, minor);
 	drm_debugfs_remove_files((struct drm_info_list *) &i915_cache_sharing_fops,
 				 1, minor);
+	drm_debugfs_remove_files((struct drm_info_list *) &i915_drop_caches_fops,
+				 1, minor);
 	drm_debugfs_remove_files((struct drm_info_list *) &i915_ring_stop_fops,
 				 1, minor);
 	drm_debugfs_remove_files((struct drm_info_list *) &i915_error_state_fops,
 				 1, minor);
+	drm_debugfs_remove_files((struct drm_info_list *) &i915_next_seqno_fops,
+				 1, minor);
 }
 
 #endif /* CONFIG_DEBUG_FS */
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 99daa896105d..4fa6beb14c77 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -992,6 +992,12 @@ static int i915_getparam(struct drm_device *dev, void *data,
 	case I915_PARAM_HAS_PINNED_BATCHES:
 		value = 1;
 		break;
+	case I915_PARAM_HAS_EXEC_NO_RELOC:
+		value = 1;
+		break;
+	case I915_PARAM_HAS_EXEC_HANDLE_LUT:
+		value = 1;
+		break;
 	default:
 		DRM_DEBUG_DRIVER("Unknown parameter %d\n",
 				 param->param);
@@ -1070,7 +1076,7 @@ static int i915_set_status_page(struct drm_device *dev, void *data,
 	ring->status_page.gfx_addr = hws->addr & (0x1ffff<<12);
 
 	dev_priv->dri1.gfx_hws_cpu_addr =
-		ioremap_wc(dev_priv->mm.gtt_base_addr + hws->addr, 4096);
+		ioremap_wc(dev_priv->gtt.mappable_base + hws->addr, 4096);
 	if (dev_priv->dri1.gfx_hws_cpu_addr == NULL) {
 		i915_dma_cleanup(dev);
 		ring->status_page.gfx_addr = 0;
@@ -1297,19 +1303,21 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	if (ret)
 		goto cleanup_vga_switcheroo;
 
+	ret = drm_irq_install(dev);
+	if (ret)
+		goto cleanup_gem_stolen;
+
+	/* Important: The output setup functions called by modeset_init need
+	 * working irqs for e.g. gmbus and dp aux transfers. */
 	intel_modeset_init(dev);
 
 	ret = i915_gem_init(dev);
 	if (ret)
-		goto cleanup_gem_stolen;
-
-	intel_modeset_gem_init(dev);
+		goto cleanup_irq;
 
 	INIT_WORK(&dev_priv->console_resume_work, intel_console_resume);
 
-	ret = drm_irq_install(dev);
-	if (ret)
-		goto cleanup_gem;
+	intel_modeset_gem_init(dev);
 
 	/* Always safe in the mode setting case. */
 	/* FIXME: do pre/post-mode set stuff in core KMS code */
@@ -1317,7 +1325,25 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	ret = intel_fbdev_init(dev);
 	if (ret)
-		goto cleanup_irq;
+		goto cleanup_gem;
+
+	/* Only enable hotplug handling once the fbdev is fully set up. */
+	intel_hpd_init(dev);
+
+	/*
+	 * Some ports require correctly set-up hpd registers for detection to
+	 * work properly (leading to ghost connected connector status), e.g. VGA
+	 * on gm45.  Hence we can only set up the initial fbdev config after hpd
+	 * irqs are fully enabled. Now we should scan for the initial config
+	 * only once hotplug handling is enabled, but due to screwed-up locking
+	 * around kms/fbdev init we can't protect the fdbev initial config
+	 * scanning against hotplug events. Hence do this first and ignore the
+	 * tiny window where we will loose hotplug notifactions.
+	 */
+	intel_fbdev_initial_config(dev);
+
+	/* Only enable hotplug handling once the fbdev is fully set up. */
+	dev_priv->enable_hotplug_processing = true;
 
 	drm_kms_helper_poll_init(dev);
 
@@ -1326,13 +1352,13 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	return 0;
 
-cleanup_irq:
-	drm_irq_uninstall(dev);
 cleanup_gem:
 	mutex_lock(&dev->struct_mutex);
 	i915_gem_cleanup_ringbuffer(dev);
 	mutex_unlock(&dev->struct_mutex);
 	i915_gem_cleanup_aliasing_ppgtt(dev);
+cleanup_irq:
+	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
 cleanup_vga_switcheroo:
@@ -1400,9 +1426,9 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 	if (!ap)
 		return;
 
-	ap->ranges[0].base = dev_priv->mm.gtt->gma_bus_addr;
-	ap->ranges[0].size =
-		dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
+	ap->ranges[0].base = dev_priv->gtt.mappable_base;
+	ap->ranges[0].size = dev_priv->gtt.mappable_end - dev_priv->gtt.start;
+
 	primary =
 		pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
 
@@ -1516,18 +1542,17 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 		goto put_gmch;
 	}
 
-	aperture_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
-	dev_priv->mm.gtt_base_addr = dev_priv->mm.gtt->gma_bus_addr;
+	aperture_size = dev_priv->gtt.mappable_end;
 
-	dev_priv->mm.gtt_mapping =
-		io_mapping_create_wc(dev_priv->mm.gtt_base_addr,
+	dev_priv->gtt.mappable =
+		io_mapping_create_wc(dev_priv->gtt.mappable_base,
 				     aperture_size);
-	if (dev_priv->mm.gtt_mapping == NULL) {
+	if (dev_priv->gtt.mappable == NULL) {
 		ret = -EIO;
 		goto out_rmmap;
 	}
 
-	i915_mtrr_setup(dev_priv, dev_priv->mm.gtt_base_addr,
+	i915_mtrr_setup(dev_priv, dev_priv->gtt.mappable_base,
 			aperture_size);
 
 	/* The i915 workqueue is primarily used for batched retirement of
@@ -1580,11 +1605,12 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 		pci_enable_msi(dev->pdev);
 
 	spin_lock_init(&dev_priv->irq_lock);
-	spin_lock_init(&dev_priv->error_lock);
+	spin_lock_init(&dev_priv->gpu_error.lock);
 	spin_lock_init(&dev_priv->rps.lock);
-	spin_lock_init(&dev_priv->dpio_lock);
+	mutex_init(&dev_priv->dpio_lock);
 
 	mutex_init(&dev_priv->rps.hw_lock);
+	mutex_init(&dev_priv->modeset_restore_lock);
 
 	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
 		dev_priv->num_pipe = 3;
@@ -1614,9 +1640,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	intel_opregion_init(dev);
 	acpi_video_register();
 
-	setup_timer(&dev_priv->hangcheck_timer, i915_hangcheck_elapsed,
-		    (unsigned long) dev);
-
 	if (IS_GEN5(dev))
 		intel_gpu_ips_init(dev_priv);
 
@@ -1635,15 +1658,15 @@ out_gem_unload:
 out_mtrrfree:
 	if (dev_priv->mm.gtt_mtrr >= 0) {
 		mtrr_del(dev_priv->mm.gtt_mtrr,
-			 dev_priv->mm.gtt_base_addr,
+			 dev_priv->gtt.mappable_base,
 			 aperture_size);
 		dev_priv->mm.gtt_mtrr = -1;
 	}
-	io_mapping_free(dev_priv->mm.gtt_mapping);
+	io_mapping_free(dev_priv->gtt.mappable);
 out_rmmap:
 	pci_iounmap(dev->pdev, dev_priv->regs);
 put_gmch:
-	i915_gem_gtt_fini(dev);
+	dev_priv->gtt.gtt_remove(dev);
 put_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
 free_priv:
@@ -1673,11 +1696,11 @@ int i915_driver_unload(struct drm_device *dev)
 	/* Cancel the retire work handler, which should be idle now. */
 	cancel_delayed_work_sync(&dev_priv->mm.retire_work);
 
-	io_mapping_free(dev_priv->mm.gtt_mapping);
+	io_mapping_free(dev_priv->gtt.mappable);
 	if (dev_priv->mm.gtt_mtrr >= 0) {
 		mtrr_del(dev_priv->mm.gtt_mtrr,
-			 dev_priv->mm.gtt_base_addr,
-			 dev_priv->mm.gtt->gtt_mappable_entries * PAGE_SIZE);
+			 dev_priv->gtt.mappable_base,
+			 dev_priv->gtt.mappable_end);
 		dev_priv->mm.gtt_mtrr = -1;
 	}
 
@@ -1703,8 +1726,8 @@ int i915_driver_unload(struct drm_device *dev)
 	}
 
 	/* Free error state after interrupts are fully disabled. */
-	del_timer_sync(&dev_priv->hangcheck_timer);
-	cancel_work_sync(&dev_priv->error_work);
+	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
+	cancel_work_sync(&dev_priv->gpu_error.work);
 	i915_destroy_error_state(dev);
 
 	if (dev->pdev->msi_enabled)
@@ -1723,9 +1746,6 @@ int i915_driver_unload(struct drm_device *dev)
 		mutex_unlock(&dev->struct_mutex);
 		i915_gem_cleanup_aliasing_ppgtt(dev);
 		i915_gem_cleanup_stolen(dev);
-		drm_mm_takedown(&dev_priv->mm.stolen);
-
-		intel_cleanup_overlay(dev);
 
 		if (!I915_NEED_GFX_HWS(dev))
 			i915_free_hws(dev);
@@ -1738,6 +1758,10 @@ int i915_driver_unload(struct drm_device *dev)
 	intel_teardown_mchbar(dev);
 
 	destroy_workqueue(dev_priv->wq);
+	pm_qos_remove_request(&dev_priv->pm_qos);
+
+	if (dev_priv->slab)
+		kmem_cache_destroy(dev_priv->slab);
 
 	pci_dev_put(dev_priv->bridge_dev);
 	kfree(dev->dev_private);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 117265840b1f..c5b8c81b9440 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -276,6 +276,7 @@ static const struct intel_device_info intel_valleyview_m_info = {
 	.has_bsd_ring = 1,
 	.has_blt_ring = 1,
 	.is_valleyview = 1,
+	.display_mmio_offset = VLV_DISPLAY_BASE,
 };
 
 static const struct intel_device_info intel_valleyview_d_info = {
@@ -285,6 +286,7 @@ static const struct intel_device_info intel_valleyview_d_info = {
 	.has_bsd_ring = 1,
 	.has_blt_ring = 1,
 	.is_valleyview = 1,
+	.display_mmio_offset = VLV_DISPLAY_BASE,
 };
 
 static const struct intel_device_info intel_haswell_d_info = {
@@ -468,6 +470,13 @@ static int i915_drm_freeze(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	/* ignore lid events during suspend */
+	mutex_lock(&dev_priv->modeset_restore_lock);
+	dev_priv->modeset_restore = MODESET_SUSPENDED;
+	mutex_unlock(&dev_priv->modeset_restore_lock);
+
+	intel_set_power_well(dev, true);
+
 	drm_kms_helper_poll_disable(dev);
 
 	pci_save_state(dev->pdev);
@@ -492,9 +501,6 @@ static int i915_drm_freeze(struct drm_device *dev)
 
 	intel_opregion_fini(dev);
 
-	/* Modeset on resume, not lid events */
-	dev_priv->modeset_on_lid = 0;
-
 	console_lock();
 	intel_fbdev_set_suspend(dev, 1);
 	console_unlock();
@@ -565,12 +571,11 @@ static int __i915_drm_thaw(struct drm_device *dev)
 		intel_modeset_init_hw(dev);
 		intel_modeset_setup_hw_state(dev, false);
 		drm_irq_install(dev);
+		intel_hpd_init(dev);
 	}
 
 	intel_opregion_init(dev);
 
-	dev_priv->modeset_on_lid = 0;
-
 	/*
 	 * The console lock can be pretty contented on resume due
 	 * to all the printk activity.  Try to keep it out of the hot
@@ -583,6 +588,9 @@ static int __i915_drm_thaw(struct drm_device *dev)
 		schedule_work(&dev_priv->console_resume_work);
 	}
 
+	mutex_lock(&dev_priv->modeset_restore_lock);
+	dev_priv->modeset_restore = MODESET_DONE;
+	mutex_unlock(&dev_priv->modeset_restore_lock);
 	return error;
 }
 
@@ -778,9 +786,9 @@ int intel_gpu_reset(struct drm_device *dev)
 	}
 
 	/* Also reset the gpu hangman. */
-	if (dev_priv->stop_rings) {
+	if (dev_priv->gpu_error.stop_rings) {
 		DRM_DEBUG("Simulated gpu hang, resetting stop_rings\n");
-		dev_priv->stop_rings = 0;
+		dev_priv->gpu_error.stop_rings = 0;
 		if (ret == -ENODEV) {
 			DRM_ERROR("Reset not implemented, but ignoring "
 				  "error for simulated gpu hangs\n");
@@ -819,12 +827,12 @@ int i915_reset(struct drm_device *dev)
 	i915_gem_reset(dev);
 
 	ret = -ENODEV;
-	if (get_seconds() - dev_priv->last_gpu_reset < 5)
+	if (get_seconds() - dev_priv->gpu_error.last_reset < 5)
 		DRM_ERROR("GPU hanging too fast, declaring wedged!\n");
 	else
 		ret = intel_gpu_reset(dev);
 
-	dev_priv->last_gpu_reset = get_seconds();
+	dev_priv->gpu_error.last_reset = get_seconds();
 	if (ret) {
 		DRM_ERROR("Failed to reset chip.\n");
 		mutex_unlock(&dev->struct_mutex);
@@ -870,6 +878,7 @@ int i915_reset(struct drm_device *dev)
 
 		drm_irq_uninstall(dev);
 		drm_irq_install(dev);
+		intel_hpd_init(dev);
 	} else {
 		mutex_unlock(&dev->struct_mutex);
 	}
@@ -1113,102 +1122,6 @@ MODULE_LICENSE("GPL and additional rights");
 	((HAS_FORCE_WAKE((dev_priv)->dev)) && \
 	 ((reg) < 0x40000) &&            \
 	 ((reg) != FORCEWAKE))
-
-static bool IS_DISPLAYREG(u32 reg)
-{
-	/*
-	 * This should make it easier to transition modules over to the
-	 * new register block scheme, since we can do it incrementally.
-	 */
-	if (reg >= VLV_DISPLAY_BASE)
-		return false;
-
-	if (reg >= RENDER_RING_BASE &&
-	    reg < RENDER_RING_BASE + 0xff)
-		return false;
-	if (reg >= GEN6_BSD_RING_BASE &&
-	    reg < GEN6_BSD_RING_BASE + 0xff)
-		return false;
-	if (reg >= BLT_RING_BASE &&
-	    reg < BLT_RING_BASE + 0xff)
-		return false;
-
-	if (reg == PGTBL_ER)
-		return false;
-
-	if (reg >= IPEIR_I965 &&
-	    reg < HWSTAM)
-		return false;
-
-	if (reg == MI_MODE)
-		return false;
-
-	if (reg == GFX_MODE_GEN7)
-		return false;
-
-	if (reg == RENDER_HWS_PGA_GEN7 ||
-	    reg == BSD_HWS_PGA_GEN7 ||
-	    reg == BLT_HWS_PGA_GEN7)
-		return false;
-
-	if (reg == GEN6_BSD_SLEEP_PSMI_CONTROL ||
-	    reg == GEN6_BSD_RNCID)
-		return false;
-
-	if (reg == GEN6_BLITTER_ECOSKPD)
-		return false;
-
-	if (reg >= 0x4000c &&
-	    reg <= 0x4002c)
-		return false;
-
-	if (reg >= 0x4f000 &&
-	    reg <= 0x4f08f)
-		return false;
-
-	if (reg >= 0x4f100 &&
-	    reg <= 0x4f11f)
-		return false;
-
-	if (reg >= VLV_MASTER_IER &&
-	    reg <= GEN6_PMIER)
-		return false;
-
-	if (reg >= FENCE_REG_SANDYBRIDGE_0 &&
-	    reg < (FENCE_REG_SANDYBRIDGE_0 + (16*8)))
-		return false;
-
-	if (reg >= VLV_IIR_RW &&
-	    reg <= VLV_ISR)
-		return false;
-
-	if (reg == FORCEWAKE_VLV ||
-	    reg == FORCEWAKE_ACK_VLV)
-		return false;
-
-	if (reg == GEN6_GDRST)
-		return false;
-
-	switch (reg) {
-	case _3D_CHICKEN3:
-	case IVB_CHICKEN3:
-	case GEN7_COMMON_SLICE_CHICKEN1:
-	case GEN7_L3CNTLREG1:
-	case GEN7_L3_CHICKEN_MODE_REGISTER:
-	case GEN7_ROW_CHICKEN2:
-	case GEN7_L3SQCREG4:
-	case GEN7_SQ_CHICKEN_MBCUNIT_CONFIG:
-	case GEN7_HALF_SLICE_CHICKEN1:
-	case GEN6_MBCTL:
-	case GEN6_UCGCTL2:
-		return false;
-	default:
-		break;
-	}
-
-	return true;
-}
-
 static void
 ilk_dummy_write(struct drm_i915_private *dev_priv)
 {
@@ -1232,8 +1145,6 @@ u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \
 		if (dev_priv->forcewake_count == 0) \
 			dev_priv->gt.force_wake_put(dev_priv); \
 		spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); \
-	} else if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \
-		val = read##y(dev_priv->regs + reg + 0x180000);		\
 	} else { \
 		val = read##y(dev_priv->regs + reg); \
 	} \
@@ -1260,11 +1171,7 @@ void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val) { \
 		DRM_ERROR("Unknown unclaimed register before writing to %x\n", reg); \
 		I915_WRITE_NOTRACE(GEN7_ERR_INT, ERR_INT_MMIO_UNCLAIMED); \
 	} \
-	if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \
-		write##y(val, dev_priv->regs + reg + 0x180000);		\
-	} else {							\
-		write##y(val, dev_priv->regs + reg);			\
-	}								\
+	write##y(val, dev_priv->regs + reg); \
 	if (unlikely(__fifo_ret)) { \
 		gen6_gt_check_fifodbg(dev_priv); \
 	} \
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 12ab3bdea54d..e95337c97459 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -30,6 +30,8 @@
 #ifndef _I915_DRV_H_
 #define _I915_DRV_H_
 
+#include <uapi/drm/i915_drm.h>
+
 #include "i915_reg.h"
 #include "intel_bios.h"
 #include "intel_ringbuffer.h"
@@ -40,6 +42,7 @@
 #include <linux/backlight.h>
 #include <linux/intel-iommu.h>
 #include <linux/kref.h>
+#include <linux/pm_qos.h>
 
 /* General customization:
  */
@@ -83,7 +86,12 @@ enum port {
 };
 #define port_name(p) ((p) + 'A')
 
-#define I915_GEM_GPU_DOMAINS	(~(I915_GEM_DOMAIN_CPU | I915_GEM_DOMAIN_GTT))
+#define I915_GEM_GPU_DOMAINS \
+	(I915_GEM_DOMAIN_RENDER | \
+	 I915_GEM_DOMAIN_SAMPLER | \
+	 I915_GEM_DOMAIN_COMMAND | \
+	 I915_GEM_DOMAIN_INSTRUCTION | \
+	 I915_GEM_DOMAIN_VERTEX)
 
 #define for_each_pipe(p) for ((p) = 0; (p) < dev_priv->num_pipe; (p)++)
 
@@ -101,6 +109,19 @@ struct intel_pch_pll {
 };
 #define I915_NUM_PLLS 2
 
+/* Used by dp and fdi links */
+struct intel_link_m_n {
+	uint32_t	tu;
+	uint32_t	gmch_m;
+	uint32_t	gmch_n;
+	uint32_t	link_m;
+	uint32_t	link_n;
+};
+
+void intel_link_compute_m_n(int bpp, int nlanes,
+			    int pixel_clock, int link_clock,
+			    struct intel_link_m_n *m_n);
+
 struct intel_ddi_plls {
 	int spll_refcount;
 	int wrpll1_refcount;
@@ -279,6 +300,7 @@ struct drm_i915_display_funcs {
 			  struct drm_i915_gem_object *obj);
 	int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			    int x, int y);
+	void (*hpd_irq_setup)(struct drm_device *dev);
 	/* clock updates for mode set */
 	/* cursor updates */
 	/* render clock increase/decrease */
@@ -318,6 +340,7 @@ struct drm_i915_gt_funcs {
 	DEV_INFO_FLAG(has_llc)
 
 struct intel_device_info {
+	u32 display_mmio_offset;
 	u8 gen;
 	u8 is_mobile:1;
 	u8 is_i85x:1;
@@ -345,6 +368,50 @@ struct intel_device_info {
 	u8 has_llc:1;
 };
 
+enum i915_cache_level {
+	I915_CACHE_NONE = 0,
+	I915_CACHE_LLC,
+	I915_CACHE_LLC_MLC, /* gen6+, in docs at least! */
+};
+
+/* The Graphics Translation Table is the way in which GEN hardware translates a
+ * Graphics Virtual Address into a Physical Address. In addition to the normal
+ * collateral associated with any va->pa translations GEN hardware also has a
+ * portion of the GTT which can be mapped by the CPU and remain both coherent
+ * and correct (in cases like swizzling). That region is referred to as GMADR in
+ * the spec.
+ */
+struct i915_gtt {
+	unsigned long start;		/* Start offset of used GTT */
+	size_t total;			/* Total size GTT can map */
+	size_t stolen_size;		/* Total size of stolen memory */
+
+	unsigned long mappable_end;	/* End offset that we can CPU map */
+	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
+	phys_addr_t mappable_base;	/* PA of our GMADR */
+
+	/** "Graphics Stolen Memory" holds the global PTEs */
+	void __iomem *gsm;
+
+	bool do_idle_maps;
+	dma_addr_t scratch_page_dma;
+	struct page *scratch_page;
+
+	/* global gtt ops */
+	int (*gtt_probe)(struct drm_device *dev, size_t *gtt_total,
+			  size_t *stolen, phys_addr_t *mappable_base,
+			  unsigned long *mappable_end);
+	void (*gtt_remove)(struct drm_device *dev);
+	void (*gtt_clear_range)(struct drm_device *dev,
+				unsigned int first_entry,
+				unsigned int num_entries);
+	void (*gtt_insert_entries)(struct drm_device *dev,
+				   struct sg_table *st,
+				   unsigned int pg_start,
+				   enum i915_cache_level cache_level);
+};
+#define gtt_total_entries(gtt) ((gtt).total >> PAGE_SHIFT)
+
 #define I915_PPGTT_PD_ENTRIES 512
 #define I915_PPGTT_PT_ENTRIES 1024
 struct i915_hw_ppgtt {
@@ -354,6 +421,16 @@ struct i915_hw_ppgtt {
 	uint32_t pd_offset;
 	dma_addr_t *pt_dma_addr;
 	dma_addr_t scratch_page_dma_addr;
+
+	/* pte functions, mirroring the interface of the global gtt. */
+	void (*clear_range)(struct i915_hw_ppgtt *ppgtt,
+			    unsigned int first_entry,
+			    unsigned int num_entries);
+	void (*insert_entries)(struct i915_hw_ppgtt *ppgtt,
+			       struct sg_table *st,
+			       unsigned int pg_start,
+			       enum i915_cache_level cache_level);
+	void (*cleanup)(struct i915_hw_ppgtt *ppgtt);
 };
 
 
@@ -580,6 +657,9 @@ struct intel_gen6_power_mgmt {
 	struct mutex hw_lock;
 };
 
+/* defined intel_pm.c */
+extern spinlock_t mchdev_lock;
+
 struct intel_ilk_power_mgmt {
 	u8 cur_delay;
 	u8 min_delay;
@@ -620,8 +700,162 @@ struct intel_l3_parity {
 	struct work_struct error_work;
 };
 
+struct i915_gem_mm {
+	/** Memory allocator for GTT stolen memory */
+	struct drm_mm stolen;
+	/** Memory allocator for GTT */
+	struct drm_mm gtt_space;
+	/** List of all objects in gtt_space. Used to restore gtt
+	 * mappings on resume */
+	struct list_head bound_list;
+	/**
+	 * List of objects which are not bound to the GTT (thus
+	 * are idle and not used by the GPU) but still have
+	 * (presumably uncached) pages still attached.
+	 */
+	struct list_head unbound_list;
+
+	/** Usable portion of the GTT for GEM */
+	unsigned long stolen_base; /* limited to low memory (32-bit) */
+
+	int gtt_mtrr;
+
+	/** PPGTT used for aliasing the PPGTT with the GTT */
+	struct i915_hw_ppgtt *aliasing_ppgtt;
+
+	struct shrinker inactive_shrinker;
+	bool shrinker_no_lock_stealing;
+
+	/**
+	 * List of objects currently involved in rendering.
+	 *
+	 * Includes buffers having the contents of their GPU caches
+	 * flushed, not necessarily primitives.  last_rendering_seqno
+	 * represents when the rendering involved will be completed.
+	 *
+	 * A reference is held on the buffer while on this list.
+	 */
+	struct list_head active_list;
+
+	/**
+	 * LRU list of objects which are not in the ringbuffer and
+	 * are ready to unbind, but are still in the GTT.
+	 *
+	 * last_rendering_seqno is 0 while an object is in this list.
+	 *
+	 * A reference is not held on the buffer while on this list,
+	 * as merely being GTT-bound shouldn't prevent its being
+	 * freed, and we'll pull it off the list in the free path.
+	 */
+	struct list_head inactive_list;
+
+	/** LRU list of objects with fence regs on them. */
+	struct list_head fence_list;
+
+	/**
+	 * We leave the user IRQ off as much as possible,
+	 * but this means that requests will finish and never
+	 * be retired once the system goes idle. Set a timer to
+	 * fire periodically while the ring is running. When it
+	 * fires, go retire requests.
+	 */
+	struct delayed_work retire_work;
+
+	/**
+	 * Are we in a non-interruptible section of code like
+	 * modesetting?
+	 */
+	bool interruptible;
+
+	/**
+	 * Flag if the X Server, and thus DRM, is not currently in
+	 * control of the device.
+	 *
+	 * This is set between LeaveVT and EnterVT.  It needs to be
+	 * replaced with a semaphore.  It also needs to be
+	 * transitioned away from for kernel modesetting.
+	 */
+	int suspended;
+
+	/** Bit 6 swizzling required for X tiling */
+	uint32_t bit_6_swizzle_x;
+	/** Bit 6 swizzling required for Y tiling */
+	uint32_t bit_6_swizzle_y;
+
+	/* storage for physical objects */
+	struct drm_i915_gem_phys_object *phys_objs[I915_MAX_PHYS_OBJECT];
+
+	/* accounting, useful for userland debugging */
+	size_t object_memory;
+	u32 object_count;
+};
+
+struct i915_gpu_error {
+	/* For hangcheck timer */
+#define DRM_I915_HANGCHECK_PERIOD 1500 /* in ms */
+#define DRM_I915_HANGCHECK_JIFFIES msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD)
+	struct timer_list hangcheck_timer;
+	int hangcheck_count;
+	uint32_t last_acthd[I915_NUM_RINGS];
+	uint32_t prev_instdone[I915_NUM_INSTDONE_REG];
+
+	/* For reset and error_state handling. */
+	spinlock_t lock;
+	/* Protected by the above dev->gpu_error.lock. */
+	struct drm_i915_error_state *first_error;
+	struct work_struct work;
+
+	unsigned long last_reset;
+
+	/**
+	 * State variable and reset counter controlling the reset flow
+	 *
+	 * Upper bits are for the reset counter.  This counter is used by the
+	 * wait_seqno code to race-free noticed that a reset event happened and
+	 * that it needs to restart the entire ioctl (since most likely the
+	 * seqno it waited for won't ever signal anytime soon).
+	 *
+	 * This is important for lock-free wait paths, where no contended lock
+	 * naturally enforces the correct ordering between the bail-out of the
+	 * waiter and the gpu reset work code.
+	 *
+	 * Lowest bit controls the reset state machine: Set means a reset is in
+	 * progress. This state will (presuming we don't have any bugs) decay
+	 * into either unset (successful reset) or the special WEDGED value (hw
+	 * terminally sour). All waiters on the reset_queue will be woken when
+	 * that happens.
+	 */
+	atomic_t reset_counter;
+
+	/**
+	 * Special values/flags for reset_counter
+	 *
+	 * Note that the code relies on
+	 * 	I915_WEDGED & I915_RESET_IN_PROGRESS_FLAG
+	 * being true.
+	 */
+#define I915_RESET_IN_PROGRESS_FLAG	1
+#define I915_WEDGED			0xffffffff
+
+	/**
+	 * Waitqueue to signal when the reset has completed. Used by clients
+	 * that wait for dev_priv->mm.wedged to settle.
+	 */
+	wait_queue_head_t reset_queue;
+
+	/* For gpu hang simulation. */
+	unsigned int stop_rings;
+};
+
+enum modeset_restore {
+	MODESET_ON_LID_OPEN,
+	MODESET_DONE,
+	MODESET_SUSPENDED,
+};
+
 typedef struct drm_i915_private {
 	struct drm_device *dev;
+	struct kmem_cache *slab;
 
 	const struct intel_device_info *info;
 
@@ -636,10 +870,11 @@ typedef struct drm_i915_private {
 	/** forcewake_count is protected by gt_lock */
 	unsigned forcewake_count;
 	/** gt_lock is also taken in irq contexts. */
-	struct spinlock gt_lock;
+	spinlock_t gt_lock;
 
 	struct intel_gmbus gmbus[GMBUS_NUM_PORTS];
 
+
 	/** gmbus_mutex protects against concurrent usage of the single hw gmbus
 	 * controller on different i2c buses. */
 	struct mutex gmbus_mutex;
@@ -649,9 +884,11 @@ typedef struct drm_i915_private {
 	 */
 	uint32_t gpio_mmio_base;
 
+	wait_queue_head_t gmbus_wait_queue;
+
 	struct pci_dev *bridge_dev;
 	struct intel_ring_buffer ring[I915_NUM_RINGS];
-	uint32_t next_seqno;
+	uint32_t last_seqno, next_seqno;
 
 	drm_dma_handle_t *status_page_dmah;
 	struct resource mch_res;
@@ -661,31 +898,24 @@ typedef struct drm_i915_private {
 	/* protects the irq masks */
 	spinlock_t irq_lock;
 
+	/* To control wakeup latency, e.g. for irq-driven dp aux transfers. */
+	struct pm_qos_request pm_qos;
+
 	/* DPIO indirect register protection */
-	spinlock_t dpio_lock;
+	struct mutex dpio_lock;
 
 	/** Cached value of IMR to avoid reads in updating the bitfield */
 	u32 pipestat[2];
 	u32 irq_mask;
 	u32 gt_irq_mask;
-	u32 pch_irq_mask;
 
 	u32 hotplug_supported_mask;
 	struct work_struct hotplug_work;
+	bool enable_hotplug_processing;
 
 	int num_pipe;
 	int num_pch_pll;
 
-	/* For hangcheck timer */
-#define DRM_I915_HANGCHECK_PERIOD 1500 /* in ms */
-#define DRM_I915_HANGCHECK_JIFFIES msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD)
-	struct timer_list hangcheck_timer;
-	int hangcheck_count;
-	uint32_t last_acthd[I915_NUM_RINGS];
-	uint32_t prev_instdone[I915_NUM_INSTDONE_REG];
-
-	unsigned int stop_rings;
-
 	unsigned long cfb_size;
 	unsigned int cfb_fb;
 	enum plane cfb_plane;
@@ -696,7 +926,7 @@ typedef struct drm_i915_private {
 
 	/* overlay */
 	struct intel_overlay *overlay;
-	bool sprite_scaling_enabled;
+	unsigned int sprite_scaling_enabled;
 
 	/* LVDS info */
 	int backlight_level;  /* restore backlight to this value */
@@ -713,7 +943,6 @@ typedef struct drm_i915_private {
 	unsigned int display_clock_mode:1;
 	int lvds_ssc_freq;
 	unsigned int bios_lvds_val; /* initial [PCH_]LVDS reg val in VBIOS */
-	unsigned int lvds_val; /* used for checking LVDS channel mode */
 	struct {
 		int rate;
 		int lanes;
@@ -734,11 +963,6 @@ typedef struct drm_i915_private {
 
 	unsigned int fsb_freq, mem_freq, is_ddr3;
 
-	spinlock_t error_lock;
-	/* Protected by dev->error_lock. */
-	struct drm_i915_error_state *first_error;
-	struct work_struct error_work;
-	struct completion error_completion;
 	struct workqueue_struct *wq;
 
 	/* Display functions */
@@ -750,115 +974,12 @@ typedef struct drm_i915_private {
 
 	unsigned long quirks;
 
-	/* Register state */
-	bool modeset_on_lid;
+	enum modeset_restore modeset_restore;
+	struct mutex modeset_restore_lock;
 
-	struct {
-		/** Bridge to intel-gtt-ko */
-		struct intel_gtt *gtt;
-		/** Memory allocator for GTT stolen memory */
-		struct drm_mm stolen;
-		/** Memory allocator for GTT */
-		struct drm_mm gtt_space;
-		/** List of all objects in gtt_space. Used to restore gtt
-		 * mappings on resume */
-		struct list_head bound_list;
-		/**
-		 * List of objects which are not bound to the GTT (thus
-		 * are idle and not used by the GPU) but still have
-		 * (presumably uncached) pages still attached.
-		 */
-		struct list_head unbound_list;
-
-		/** Usable portion of the GTT for GEM */
-		unsigned long gtt_start;
-		unsigned long gtt_mappable_end;
-		unsigned long gtt_end;
-
-		struct io_mapping *gtt_mapping;
-		phys_addr_t gtt_base_addr;
-		int gtt_mtrr;
-
-		/** PPGTT used for aliasing the PPGTT with the GTT */
-		struct i915_hw_ppgtt *aliasing_ppgtt;
-
-		struct shrinker inactive_shrinker;
-		bool shrinker_no_lock_stealing;
-
-		/**
-		 * List of objects currently involved in rendering.
-		 *
-		 * Includes buffers having the contents of their GPU caches
-		 * flushed, not necessarily primitives.  last_rendering_seqno
-		 * represents when the rendering involved will be completed.
-		 *
-		 * A reference is held on the buffer while on this list.
-		 */
-		struct list_head active_list;
-
-		/**
-		 * LRU list of objects which are not in the ringbuffer and
-		 * are ready to unbind, but are still in the GTT.
-		 *
-		 * last_rendering_seqno is 0 while an object is in this list.
-		 *
-		 * A reference is not held on the buffer while on this list,
-		 * as merely being GTT-bound shouldn't prevent its being
-		 * freed, and we'll pull it off the list in the free path.
-		 */
-		struct list_head inactive_list;
-
-		/** LRU list of objects with fence regs on them. */
-		struct list_head fence_list;
-
-		/**
-		 * We leave the user IRQ off as much as possible,
-		 * but this means that requests will finish and never
-		 * be retired once the system goes idle. Set a timer to
-		 * fire periodically while the ring is running. When it
-		 * fires, go retire requests.
-		 */
-		struct delayed_work retire_work;
-
-		/**
-		 * Are we in a non-interruptible section of code like
-		 * modesetting?
-		 */
-		bool interruptible;
-
-		/**
-		 * Flag if the X Server, and thus DRM, is not currently in
-		 * control of the device.
-		 *
-		 * This is set between LeaveVT and EnterVT.  It needs to be
-		 * replaced with a semaphore.  It also needs to be
-		 * transitioned away from for kernel modesetting.
-		 */
-		int suspended;
-
-		/**
-		 * Flag if the hardware appears to be wedged.
-		 *
-		 * This is set when attempts to idle the device timeout.
-		 * It prevents command submission from occurring and makes
-		 * every pending request fail
-		 */
-		atomic_t wedged;
-
-		/** Bit 6 swizzling required for X tiling */
-		uint32_t bit_6_swizzle_x;
-		/** Bit 6 swizzling required for Y tiling */
-		uint32_t bit_6_swizzle_y;
-
-		/* storage for physical objects */
-		struct drm_i915_gem_phys_object *phys_objs[I915_MAX_PHYS_OBJECT];
-
-		/* accounting, useful for userland debugging */
-		size_t gtt_total;
-		size_t mappable_gtt_total;
-		size_t object_memory;
-		u32 object_count;
-	} mm;
+	struct i915_gtt gtt;
+
+	struct i915_gem_mm mm;
 
 	/* Kernel Modesetting */
 
@@ -900,7 +1021,7 @@ typedef struct drm_i915_private {
 	struct drm_mm_node *compressed_fb;
 	struct drm_mm_node *compressed_llb;
 
-	unsigned long last_gpu_reset;
+	struct i915_gpu_error gpu_error;
 
 	/* list of fbdev register on this device */
 	struct intel_fbdev *fbdev;
@@ -919,7 +1040,7 @@ typedef struct drm_i915_private {
 	bool hw_contexts_disabled;
 	uint32_t hw_context_size;
 
-	bool fdi_rx_polarity_reversed;
+	u32 fdi_rx_config;
 
 	struct i915_suspend_saved_registers regfile;
 
@@ -940,11 +1061,7 @@ enum hdmi_force_audio {
 	HDMI_AUDIO_ON,			/* force turn on HDMI audio */
 };
 
-enum i915_cache_level {
-	I915_CACHE_NONE = 0,
-	I915_CACHE_LLC,
-	I915_CACHE_LLC_MLC, /* gen6+, in docs at least! */
-};
+#define I915_GTT_RESERVED ((struct drm_mm_node *)0x1)
 
 struct drm_i915_gem_object_ops {
 	/* Interface between the GEM object and its backing storage.
@@ -971,6 +1088,8 @@ struct drm_i915_gem_object {
 
 	/** Current space allocated to this object in the GTT, if any. */
 	struct drm_mm_node *gtt_space;
+	/** Stolen memory for this object, instead of being backed by shmem. */
+	struct drm_mm_node *stolen;
 	struct list_head gtt_list;
 
 	/** This object's place on the active/inactive lists */
@@ -1096,13 +1215,6 @@ struct drm_i915_gem_object {
 
 	/** for phy allocated objects */
 	struct drm_i915_gem_phys_object *phys_obj;
-
-	/**
-	 * Number of crtcs where this object is currently the fb, but
-	 * will be page flipped away on the next vblank.  When it
-	 * reaches 0, dev_priv->pending_flip_queue will be woken up.
-	 */
-	atomic_t pending_flip;
 };
 #define to_gem_object(obj) (&((struct drm_i915_gem_object *)(obj))->base)
 
@@ -1141,7 +1253,7 @@ struct drm_i915_gem_request {
 
 struct drm_i915_file_private {
 	struct {
-		struct spinlock lock;
+		spinlock_t lock;
 		struct list_head request_list;
 	} mm;
 	struct idr context_idr;
@@ -1227,6 +1339,8 @@ struct drm_i915_file_private {
 
 #define HAS_PIPE_CONTROL(dev) (INTEL_INFO(dev)->gen >= 5)
 
+#define HAS_DDI(dev)		(IS_HASWELL(dev))
+
 #define INTEL_PCH_DEVICE_ID_MASK		0xff00
 #define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
 #define INTEL_PCH_CPT_DEVICE_ID_TYPE		0x1c00
@@ -1323,6 +1437,7 @@ void i915_hangcheck_elapsed(unsigned long data);
 void i915_handle_error(struct drm_device *dev, bool wedged);
 
 extern void intel_irq_init(struct drm_device *dev);
+extern void intel_hpd_init(struct drm_device *dev);
 extern void intel_gt_init(struct drm_device *dev);
 extern void intel_gt_reset(struct drm_device *dev);
 
@@ -1391,18 +1506,22 @@ int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 void i915_gem_load(struct drm_device *dev);
+void *i915_gem_object_alloc(struct drm_device *dev);
+void i915_gem_object_free(struct drm_i915_gem_object *obj);
 int i915_gem_init_object(struct drm_gem_object *obj);
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
 			 const struct drm_i915_gem_object_ops *ops);
 struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 						  size_t size);
 void i915_gem_free_object(struct drm_gem_object *obj);
+
 int __must_check i915_gem_object_pin(struct drm_i915_gem_object *obj,
 				     uint32_t alignment,
 				     bool map_and_fenceable,
 				     bool nonblocking);
 void i915_gem_object_unpin(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_object_unbind(struct drm_i915_gem_object *obj);
+int i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
 void i915_gem_release_mmap(struct drm_i915_gem_object *obj);
 void i915_gem_lastclose(struct drm_device *dev);
 
@@ -1454,8 +1573,8 @@ i915_seqno_passed(uint32_t seq1, uint32_t seq2)
 	return (int32_t)(seq1 - seq2) >= 0;
 }
 
-extern int i915_gem_get_seqno(struct drm_device *dev, u32 *seqno);
-
+int __must_check i915_gem_get_seqno(struct drm_device *dev, u32 *seqno);
+int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno);
 int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj);
 
@@ -1481,8 +1600,18 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
 
 void i915_gem_retire_requests(struct drm_device *dev);
 void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
-int __must_check i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
 				      bool interruptible);
+static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
+{
+	return unlikely(atomic_read(&error->reset_counter)
+			& I915_RESET_IN_PROGRESS_FLAG);
+}
+
+static inline bool i915_terminally_wedged(struct i915_gpu_error *error)
+{
+	return atomic_read(&error->reset_counter) == I915_WEDGED;
+}
 
 void i915_gem_reset(struct drm_device *dev);
 void i915_gem_clflush_object(struct drm_i915_gem_object *obj);
@@ -1523,9 +1652,10 @@ void i915_gem_free_all_phys_object(struct drm_device *dev);
 void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 
 uint32_t
-i915_gem_get_unfenced_gtt_alignment(struct drm_device *dev,
-				    uint32_t size,
-				    int tiling_mode);
+i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode);
+uint32_t
+i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
+			    int tiling_mode, bool fenced);
 
 int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				    enum i915_cache_level cache_level);
@@ -1548,7 +1678,6 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file);
 
 /* i915_gem_gtt.c */
-int __must_check i915_gem_init_aliasing_ppgtt(struct drm_device *dev);
 void i915_gem_cleanup_aliasing_ppgtt(struct drm_device *dev);
 void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt,
 			    struct drm_i915_gem_object *obj,
@@ -1562,12 +1691,10 @@ void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,
 				enum i915_cache_level cache_level);
 void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj);
 void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj);
-void i915_gem_init_global_gtt(struct drm_device *dev,
-			      unsigned long start,
-			      unsigned long mappable_end,
-			      unsigned long end);
+void i915_gem_init_global_gtt(struct drm_device *dev);
+void i915_gem_setup_global_gtt(struct drm_device *dev, unsigned long start,
+			       unsigned long mappable_end, unsigned long end);
 int i915_gem_gtt_init(struct drm_device *dev);
-void i915_gem_gtt_fini(struct drm_device *dev);
 static inline void i915_gem_chipset_flush(struct drm_device *dev)
 {
 	if (INTEL_INFO(dev)->gen < 6)
@@ -1585,9 +1712,22 @@ int i915_gem_evict_everything(struct drm_device *dev);
 
 /* i915_gem_stolen.c */
 int i915_gem_init_stolen(struct drm_device *dev);
+int i915_gem_stolen_setup_compression(struct drm_device *dev, int size);
+void i915_gem_stolen_cleanup_compression(struct drm_device *dev);
 void i915_gem_cleanup_stolen(struct drm_device *dev);
+struct drm_i915_gem_object *
+i915_gem_object_create_stolen(struct drm_device *dev, u32 size);
+void i915_gem_object_release_stolen(struct drm_i915_gem_object *obj);
 
 /* i915_gem_tiling.c */
+inline static bool i915_gem_object_needs_bit17_swizzle(struct drm_i915_gem_object *obj)
+{
+	drm_i915_private_t *dev_priv = obj->base.dev->dev_private;
+
+	return dev_priv->mm.bit_6_swizzle_x == I915_BIT_6_SWIZZLE_9_10_17 &&
+		obj->tiling_mode != I915_TILING_NONE;
+}
+
 void i915_gem_detect_bit_6_swizzle(struct drm_device *dev);
 void i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj);
 void i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj);
@@ -1613,9 +1753,9 @@ void i915_debugfs_cleanup(struct drm_minor *minor);
 extern int i915_save_state(struct drm_device *dev);
 extern int i915_restore_state(struct drm_device *dev);
 
-/* i915_suspend.c */
-extern int i915_save_state(struct drm_device *dev);
-extern int i915_restore_state(struct drm_device *dev);
+/* i915_ums.c */
+void i915_save_display_reg(struct drm_device *dev);
+void i915_restore_display_reg(struct drm_device *dev);
 
 /* i915_sysfs.c */
 void i915_setup_sysfs(struct drm_device *dev_priv);
@@ -1672,6 +1812,7 @@ extern void intel_modeset_cleanup(struct drm_device *dev);
 extern int intel_modeset_vga_set_state(struct drm_device *dev, bool state);
 extern void intel_modeset_setup_hw_state(struct drm_device *dev,
 					 bool force_restore);
+extern void i915_redisable_vga(struct drm_device *dev);
 extern bool intel_fbc_enabled(struct drm_device *dev);
 extern void intel_disable_fbc(struct drm_device *dev);
 extern bool ironlake_set_drps(struct drm_device *dev, u8 val);
@@ -1744,5 +1885,19 @@ __i915_write(64, q)
 #define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
 #define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)
 
+/* "Broadcast RGB" property */
+#define INTEL_BROADCAST_RGB_AUTO 0
+#define INTEL_BROADCAST_RGB_FULL 1
+#define INTEL_BROADCAST_RGB_LIMITED 2
+
+static inline uint32_t i915_vgacntrl_reg(struct drm_device *dev)
+{
+	if (HAS_PCH_SPLIT(dev))
+		return CPU_VGACNTRL;
+	else if (IS_VALLEYVIEW(dev))
+		return VLV_VGACNTRL;
+	else
+		return VGACNTRL;
+}
 
 #endif
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 8febea6daa08..8413ffced815 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -87,47 +87,43 @@ static void i915_gem_info_remove_obj(struct drm_i915_private *dev_priv,
 }
 
 static int
-i915_gem_wait_for_error(struct drm_device *dev)
+i915_gem_wait_for_error(struct i915_gpu_error *error)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct completion *x = &dev_priv->error_completion;
-	unsigned long flags;
 	int ret;
 
-	if (!atomic_read(&dev_priv->mm.wedged))
+#define EXIT_COND (!i915_reset_in_progress(error))
+	if (EXIT_COND)
 		return 0;
 
+	/* GPU is already declared terminally dead, give up. */
+	if (i915_terminally_wedged(error))
+		return -EIO;
+
 	/*
 	 * Only wait 10 seconds for the gpu reset to complete to avoid hanging
 	 * userspace. If it takes that long something really bad is going on and
 	 * we should simply try to bail out and fail as gracefully as possible.
 	 */
-	ret = wait_for_completion_interruptible_timeout(x, 10*HZ);
+	ret = wait_event_interruptible_timeout(error->reset_queue,
+					       EXIT_COND,
+					       10*HZ);
 	if (ret == 0) {
 		DRM_ERROR("Timed out waiting for the gpu reset to complete\n");
 		return -EIO;
 	} else if (ret < 0) {
 		return ret;
 	}
+#undef EXIT_COND
 
-	if (atomic_read(&dev_priv->mm.wedged)) {
-		/* GPU is hung, bump the completion count to account for
-		 * the token we just consumed so that we never hit zero and
-		 * end up waiting upon a subsequent completion event that
-		 * will never happen.
-		 */
-		spin_lock_irqsave(&x->wait.lock, flags);
-		x->done++;
-		spin_unlock_irqrestore(&x->wait.lock, flags);
-	}
 	return 0;
 }
 
 int i915_mutex_lock_interruptible(struct drm_device *dev)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	ret = i915_gem_wait_for_error(dev);
+	ret = i915_gem_wait_for_error(&dev_priv->gpu_error);
 	if (ret)
 		return ret;
 
@@ -149,6 +145,7 @@ int
 i915_gem_init_ioctl(struct drm_device *dev, void *data,
 		    struct drm_file *file)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_init *args = data;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
@@ -163,8 +160,9 @@ i915_gem_init_ioctl(struct drm_device *dev, void *data,
 		return -ENODEV;
 
 	mutex_lock(&dev->struct_mutex);
-	i915_gem_init_global_gtt(dev, args->gtt_start,
-				 args->gtt_end, args->gtt_end);
+	i915_gem_setup_global_gtt(dev, args->gtt_start, args->gtt_end,
+				  args->gtt_end);
+	dev_priv->gtt.mappable_end = args->gtt_end;
 	mutex_unlock(&dev->struct_mutex);
 
 	return 0;
@@ -186,12 +184,24 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 			pinned += obj->gtt_space->size;
 	mutex_unlock(&dev->struct_mutex);
 
-	args->aper_size = dev_priv->mm.gtt_total;
+	args->aper_size = dev_priv->gtt.total;
 	args->aper_available_size = args->aper_size - pinned;
 
 	return 0;
 }
 
+void *i915_gem_object_alloc(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	return kmem_cache_alloc(dev_priv->slab, GFP_KERNEL | __GFP_ZERO);
+}
+
+void i915_gem_object_free(struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+	kmem_cache_free(dev_priv->slab, obj);
+}
+
 static int
 i915_gem_create(struct drm_file *file,
 		struct drm_device *dev,
@@ -215,7 +225,7 @@ i915_gem_create(struct drm_file *file,
 	if (ret) {
 		drm_gem_object_release(&obj->base);
 		i915_gem_info_remove_obj(dev->dev_private, obj->base.size);
-		kfree(obj);
+		i915_gem_object_free(obj);
 		return ret;
 	}
 
@@ -259,14 +269,6 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data,
 			       args->size, &args->handle);
 }
 
-static int i915_gem_object_needs_bit17_swizzle(struct drm_i915_gem_object *obj)
-{
-	drm_i915_private_t *dev_priv = obj->base.dev->dev_private;
-
-	return dev_priv->mm.bit_6_swizzle_x == I915_BIT_6_SWIZZLE_9_10_17 &&
-		obj->tiling_mode != I915_TILING_NONE;
-}
-
 static inline int
 __copy_to_user_swizzled(char __user *cpu_vaddr,
 			const char *gpu_vaddr, int gpu_offset,
@@ -407,7 +409,6 @@ i915_gem_shmem_pread(struct drm_device *dev,
 	loff_t offset;
 	int shmem_page_offset, page_length, ret = 0;
 	int obj_do_bit17_swizzling, page_do_bit17_swizzling;
-	int hit_slowpath = 0;
 	int prefaulted = 0;
 	int needs_clflush = 0;
 	struct scatterlist *sg;
@@ -469,7 +470,6 @@ i915_gem_shmem_pread(struct drm_device *dev,
 		if (ret == 0)
 			goto next_page;
 
-		hit_slowpath = 1;
 		mutex_unlock(&dev->struct_mutex);
 
 		if (!prefaulted) {
@@ -502,12 +502,6 @@ next_page:
 out:
 	i915_gem_object_unpin_pages(obj);
 
-	if (hit_slowpath) {
-		/* Fixup: Kill any reinstated backing storage pages */
-		if (obj->madv == __I915_MADV_PURGED)
-			i915_gem_object_truncate(obj);
-	}
-
 	return ret;
 }
 
@@ -641,7 +635,7 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
 		 * source page isn't available.  Return the error and we'll
 		 * retry in the slow path.
 		 */
-		if (fast_user_write(dev_priv->mm.gtt_mapping, page_base,
+		if (fast_user_write(dev_priv->gtt.mappable, page_base,
 				    page_offset, user_data, page_length)) {
 			ret = -EFAULT;
 			goto out_unpin;
@@ -838,12 +832,13 @@ out:
 	i915_gem_object_unpin_pages(obj);
 
 	if (hit_slowpath) {
-		/* Fixup: Kill any reinstated backing storage pages */
-		if (obj->madv == __I915_MADV_PURGED)
-			i915_gem_object_truncate(obj);
-		/* and flush dirty cachelines in case the object isn't in the cpu write
-		 * domain anymore. */
-		if (obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
+		/*
+		 * Fixup: Flush cpu caches in case we didn't flush the dirty
+		 * cachelines in-line while writing and the object moved
+		 * out of the cpu write domain while we've dropped the lock.
+		 */
+		if (!needs_clflush_after &&
+		    obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
 			i915_gem_clflush_object(obj);
 			i915_gem_chipset_flush(dev);
 		}
@@ -940,26 +935,17 @@ unlock:
 }
 
 int
-i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+i915_gem_check_wedge(struct i915_gpu_error *error,
 		     bool interruptible)
 {
-	if (atomic_read(&dev_priv->mm.wedged)) {
-		struct completion *x = &dev_priv->error_completion;
-		bool recovery_complete;
-		unsigned long flags;
-
-		/* Give the error handler a chance to run. */
-		spin_lock_irqsave(&x->wait.lock, flags);
-		recovery_complete = x->done > 0;
-		spin_unlock_irqrestore(&x->wait.lock, flags);
-
+	if (i915_reset_in_progress(error)) {
 		/* Non-interruptible callers can't handle -EAGAIN, hence return
 		 * -EIO unconditionally for these. */
 		if (!interruptible)
 			return -EIO;
 
-		/* Recovery complete, but still wedged means reset failure. */
-		if (recovery_complete)
+		/* Recovery complete, but the reset failed ... */
+		if (i915_terminally_wedged(error))
 			return -EIO;
 
 		return -EAGAIN;
@@ -990,13 +976,22 @@ i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
  * __wait_seqno - wait until execution of seqno has finished
  * @ring: the ring expected to report seqno
  * @seqno: duh!
+ * @reset_counter: reset sequence associated with the given seqno
  * @interruptible: do an interruptible wait (normally yes)
  * @timeout: in - how long to wait (NULL forever); out - how much time remaining
  *
+ * Note: It is of utmost importance that the passed in seqno and reset_counter
+ * values have been read by the caller in an smp safe manner. Where read-side
+ * locks are involved, it is sufficient to read the reset_counter before
+ * unlocking the lock that protects the seqno. For lockless tricks, the
+ * reset_counter _must_ be read before, and an appropriate smp_rmb must be
+ * inserted.
+ *
  * Returns 0 if the seqno was found within the alloted time. Else returns the
  * errno with remaining time filled in timeout argument.
  */
 static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
+			unsigned reset_counter,
 			bool interruptible, struct timespec *timeout)
 {
 	drm_i915_private_t *dev_priv = ring->dev->dev_private;
@@ -1026,7 +1021,8 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 
 #define EXIT_COND \
 	(i915_seqno_passed(ring->get_seqno(ring, false), seqno) || \
-	atomic_read(&dev_priv->mm.wedged))
+	 i915_reset_in_progress(&dev_priv->gpu_error) || \
+	 reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
 	do {
 		if (interruptible)
 			end = wait_event_interruptible_timeout(ring->irq_queue,
@@ -1036,7 +1032,14 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 			end = wait_event_timeout(ring->irq_queue, EXIT_COND,
 						 timeout_jiffies);
 
-		ret = i915_gem_check_wedge(dev_priv, interruptible);
+		/* We need to check whether any gpu reset happened in between
+		 * the caller grabbing the seqno and now ... */
+		if (reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
+			end = -EAGAIN;
+
+		/* ... but upgrade the -EGAIN to an -EIO if the gpu is truely
+		 * gone. */
+		ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
 		if (ret)
 			end = ret;
 	} while (end == 0 && wait_forever);
@@ -1082,7 +1085,7 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
 	BUG_ON(seqno == 0);
 
-	ret = i915_gem_check_wedge(dev_priv, interruptible);
+	ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
 	if (ret)
 		return ret;
 
@@ -1090,7 +1093,9 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 	if (ret)
 		return ret;
 
-	return __wait_seqno(ring, seqno, interruptible, NULL);
+	return __wait_seqno(ring, seqno,
+			    atomic_read(&dev_priv->gpu_error.reset_counter),
+			    interruptible, NULL);
 }
 
 /**
@@ -1137,6 +1142,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring = obj->ring;
+	unsigned reset_counter;
 	u32 seqno;
 	int ret;
 
@@ -1147,7 +1153,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 	if (seqno == 0)
 		return 0;
 
-	ret = i915_gem_check_wedge(dev_priv, true);
+	ret = i915_gem_check_wedge(&dev_priv->gpu_error, true);
 	if (ret)
 		return ret;
 
@@ -1155,8 +1161,9 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 	if (ret)
 		return ret;
 
+	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	mutex_unlock(&dev->struct_mutex);
-	ret = __wait_seqno(ring, seqno, true, NULL);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL);
 	mutex_lock(&dev->struct_mutex);
 
 	i915_gem_retire_requests_ring(ring);
@@ -1344,6 +1351,12 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	trace_i915_gem_object_fault(obj, page_offset, true, write);
 
+	/* Access to snoopable pages through the GTT is incoherent. */
+	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev)) {
+		ret = -EINVAL;
+		goto unlock;
+	}
+
 	/* Now bind it into the GTT if needed */
 	ret = i915_gem_object_pin(obj, 0, true, false);
 	if (ret)
@@ -1359,7 +1372,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	obj->fault_mappable = true;
 
-	pfn = ((dev_priv->mm.gtt_base_addr + obj->gtt_offset) >> PAGE_SHIFT) +
+	pfn = ((dev_priv->gtt.mappable_base + obj->gtt_offset) >> PAGE_SHIFT) +
 		page_offset;
 
 	/* Finally, remap it using the new GTT offset */
@@ -1374,7 +1387,7 @@ out:
 		/* If this -EIO is due to a gpu hang, give the reset code a
 		 * chance to clean up the mess. Otherwise return the proper
 		 * SIGBUS. */
-		if (!atomic_read(&dev_priv->mm.wedged))
+		if (i915_terminally_wedged(&dev_priv->gpu_error))
 			return VM_FAULT_SIGBUS;
 	case -EAGAIN:
 		/* Give the error handler a chance to run and move the
@@ -1432,7 +1445,7 @@ i915_gem_release_mmap(struct drm_i915_gem_object *obj)
 	obj->fault_mappable = false;
 }
 
-static uint32_t
+uint32_t
 i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode)
 {
 	uint32_t gtt_size;
@@ -1460,16 +1473,15 @@ i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode)
  * Return the required GTT alignment for an object, taking into account
  * potential fence register mapping.
  */
-static uint32_t
-i915_gem_get_gtt_alignment(struct drm_device *dev,
-			   uint32_t size,
-			   int tiling_mode)
+uint32_t
+i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
+			   int tiling_mode, bool fenced)
 {
 	/*
 	 * Minimum alignment is 4k (GTT page size), but might be greater
 	 * if a fence register is needed for the object.
 	 */
-	if (INTEL_INFO(dev)->gen >= 4 ||
+	if (INTEL_INFO(dev)->gen >= 4 || (!fenced && IS_G33(dev)) ||
 	    tiling_mode == I915_TILING_NONE)
 		return 4096;
 
@@ -1480,35 +1492,6 @@ i915_gem_get_gtt_alignment(struct drm_device *dev,
 	return i915_gem_get_gtt_size(dev, size, tiling_mode);
 }
 
-/**
- * i915_gem_get_unfenced_gtt_alignment - return required GTT alignment for an
- *					 unfenced object
- * @dev: the device
- * @size: size of the object
- * @tiling_mode: tiling mode of the object
- *
- * Return the required GTT alignment for an object, only taking into account
- * unfenced tiled surface requirements.
- */
-uint32_t
-i915_gem_get_unfenced_gtt_alignment(struct drm_device *dev,
-				    uint32_t size,
-				    int tiling_mode)
-{
-	/*
-	 * Minimum alignment is 4k (GTT page size) for sane hw.
-	 */
-	if (INTEL_INFO(dev)->gen >= 4 || IS_G33(dev) ||
-	    tiling_mode == I915_TILING_NONE)
-		return 4096;
-
-	/* Previous hardware however needs to be aligned to a power-of-two
-	 * tile height. The simplest method for determining this is to reuse
-	 * the power-of-tile object size.
-	 */
-	return i915_gem_get_gtt_size(dev, size, tiling_mode);
-}
-
 static int i915_gem_object_create_mmap_offset(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
@@ -1571,7 +1554,7 @@ i915_gem_mmap_gtt(struct drm_file *file,
 		goto unlock;
 	}
 
-	if (obj->base.size > dev_priv->mm.gtt_mappable_end) {
+	if (obj->base.size > dev_priv->gtt.mappable_end) {
 		ret = -E2BIG;
 		goto out;
 	}
@@ -1689,7 +1672,7 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
 	kfree(obj->pages);
 }
 
-static int
+int
 i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 {
 	const struct drm_i915_gem_object_ops *ops = obj->ops;
@@ -1862,6 +1845,11 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 	if (obj->pages)
 		return 0;
 
+	if (obj->madv != I915_MADV_WILLNEED) {
+		DRM_ERROR("Attempting to obtain a purgeable object\n");
+		return -EINVAL;
+	}
+
 	BUG_ON(obj->pages_pin_count);
 
 	ret = ops->get_pages(obj);
@@ -1918,9 +1906,6 @@ i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 	BUG_ON(obj->base.write_domain & ~I915_GEM_GPU_DOMAINS);
 	BUG_ON(!obj->active);
 
-	if (obj->pin_count) /* are we a framebuffer? */
-		intel_mark_fb_idle(obj);
-
 	list_move_tail(&obj->mm_list, &dev_priv->mm.inactive_list);
 
 	list_del_init(&obj->ring_list);
@@ -1940,30 +1925,24 @@ i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 }
 
 static int
-i915_gem_handle_seqno_wrap(struct drm_device *dev)
+i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
 	int ret, i, j;
 
-	/* The hardware uses various monotonic 32-bit counters, if we
-	 * detect that they will wraparound we need to idle the GPU
-	 * and reset those counters.
-	 */
-	ret = 0;
+	/* Carefully retire all requests without writing to the rings */
 	for_each_ring(ring, dev_priv, i) {
-		for (j = 0; j < ARRAY_SIZE(ring->sync_seqno); j++)
-			ret |= ring->sync_seqno[j] != 0;
+		ret = intel_ring_idle(ring);
+		if (ret)
+			return ret;
 	}
-	if (ret == 0)
-		return ret;
-
-	ret = i915_gpu_idle(dev);
-	if (ret)
-		return ret;
-
 	i915_gem_retire_requests(dev);
+
+	/* Finally reset hw state */
 	for_each_ring(ring, dev_priv, i) {
+		intel_ring_init_seqno(ring, seqno);
+
 		for (j = 0; j < ARRAY_SIZE(ring->sync_seqno); j++)
 			ring->sync_seqno[j] = 0;
 	}
@@ -1971,6 +1950,32 @@ i915_gem_handle_seqno_wrap(struct drm_device *dev)
 	return 0;
 }
 
+int i915_gem_set_seqno(struct drm_device *dev, u32 seqno)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	if (seqno == 0)
+		return -EINVAL;
+
+	/* HWS page needs to be set less than what we
+	 * will inject to ring
+	 */
+	ret = i915_gem_init_seqno(dev, seqno - 1);
+	if (ret)
+		return ret;
+
+	/* Carefully set the last_seqno value so that wrap
+	 * detection still works
+	 */
+	dev_priv->next_seqno = seqno;
+	dev_priv->last_seqno = seqno - 1;
+	if (dev_priv->last_seqno == 0)
+		dev_priv->last_seqno--;
+
+	return 0;
+}
+
 int
 i915_gem_get_seqno(struct drm_device *dev, u32 *seqno)
 {
@@ -1978,14 +1983,14 @@ i915_gem_get_seqno(struct drm_device *dev, u32 *seqno)
 
 	/* reserve 0 for non-seqno */
 	if (dev_priv->next_seqno == 0) {
-		int ret = i915_gem_handle_seqno_wrap(dev);
+		int ret = i915_gem_init_seqno(dev, 0);
 		if (ret)
 			return ret;
 
 		dev_priv->next_seqno = 1;
 	}
 
-	*seqno = dev_priv->next_seqno++;
+	*seqno = dev_priv->last_seqno = dev_priv->next_seqno++;
 	return 0;
 }
 
@@ -2052,7 +2057,7 @@ i915_add_request(struct intel_ring_buffer *ring,
 
 	if (!dev_priv->mm.suspended) {
 		if (i915_enable_hangcheck) {
-			mod_timer(&dev_priv->hangcheck_timer,
+			mod_timer(&dev_priv->gpu_error.hangcheck_timer,
 				  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
 		}
 		if (was_empty) {
@@ -2317,10 +2322,12 @@ i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
 int
 i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 {
+	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct drm_i915_gem_wait *args = data;
 	struct drm_i915_gem_object *obj;
 	struct intel_ring_buffer *ring = NULL;
 	struct timespec timeout_stack, *timeout = NULL;
+	unsigned reset_counter;
 	u32 seqno = 0;
 	int ret = 0;
 
@@ -2361,9 +2368,10 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	}
 
 	drm_gem_object_unreference(&obj->base);
+	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	mutex_unlock(&dev->struct_mutex);
 
-	ret = __wait_seqno(ring, seqno, true, timeout);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, timeout);
 	if (timeout) {
 		WARN_ON(!timespec_valid(timeout));
 		args->timeout_ns = timespec_to_ns(timeout);
@@ -2427,15 +2435,15 @@ static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
 {
 	u32 old_write_domain, old_read_domains;
 
-	/* Act a barrier for all accesses through the GTT */
-	mb();
-
 	/* Force a pagefault for domain tracking on next user access */
 	i915_gem_release_mmap(obj);
 
 	if ((obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0)
 		return;
 
+	/* Wait for any direct GTT access to complete */
+	mb();
+
 	old_read_domains = obj->base.read_domains;
 	old_write_domain = obj->base.write_domain;
 
@@ -2454,7 +2462,7 @@ int
 i915_gem_object_unbind(struct drm_i915_gem_object *obj)
 {
 	drm_i915_private_t *dev_priv = obj->base.dev->dev_private;
-	int ret = 0;
+	int ret;
 
 	if (obj->gtt_space == NULL)
 		return 0;
@@ -2521,52 +2529,38 @@ int i915_gpu_idle(struct drm_device *dev)
 	return 0;
 }
 
-static void sandybridge_write_fence_reg(struct drm_device *dev, int reg,
-					struct drm_i915_gem_object *obj)
-{
-	drm_i915_private_t *dev_priv = dev->dev_private;
-	uint64_t val;
-
-	if (obj) {
-		u32 size = obj->gtt_space->size;
-
-		val = (uint64_t)((obj->gtt_offset + size - 4096) &
-				 0xfffff000) << 32;
-		val |= obj->gtt_offset & 0xfffff000;
-		val |= (uint64_t)((obj->stride / 128) - 1) <<
-			SANDYBRIDGE_FENCE_PITCH_SHIFT;
-
-		if (obj->tiling_mode == I915_TILING_Y)
-			val |= 1 << I965_FENCE_TILING_Y_SHIFT;
-		val |= I965_FENCE_REG_VALID;
-	} else
-		val = 0;
-
-	I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + reg * 8, val);
-	POSTING_READ(FENCE_REG_SANDYBRIDGE_0 + reg * 8);
-}
-
 static void i965_write_fence_reg(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
+	int fence_reg;
+	int fence_pitch_shift;
 	uint64_t val;
 
+	if (INTEL_INFO(dev)->gen >= 6) {
+		fence_reg = FENCE_REG_SANDYBRIDGE_0;
+		fence_pitch_shift = SANDYBRIDGE_FENCE_PITCH_SHIFT;
+	} else {
+		fence_reg = FENCE_REG_965_0;
+		fence_pitch_shift = I965_FENCE_PITCH_SHIFT;
+	}
+
 	if (obj) {
 		u32 size = obj->gtt_space->size;
 
 		val = (uint64_t)((obj->gtt_offset + size - 4096) &
 				 0xfffff000) << 32;
 		val |= obj->gtt_offset & 0xfffff000;
-		val |= ((obj->stride / 128) - 1) << I965_FENCE_PITCH_SHIFT;
+		val |= (uint64_t)((obj->stride / 128) - 1) << fence_pitch_shift;
 		if (obj->tiling_mode == I915_TILING_Y)
 			val |= 1 << I965_FENCE_TILING_Y_SHIFT;
 		val |= I965_FENCE_REG_VALID;
 	} else
 		val = 0;
 
-	I915_WRITE64(FENCE_REG_965_0 + reg * 8, val);
-	POSTING_READ(FENCE_REG_965_0 + reg * 8);
+	fence_reg += reg * 8;
+	I915_WRITE64(fence_reg, val);
+	POSTING_READ(fence_reg);
 }
 
 static void i915_write_fence_reg(struct drm_device *dev, int reg,
@@ -2645,18 +2639,37 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,
 	POSTING_READ(FENCE_REG_830_0 + reg * 4);
 }
 
+inline static bool i915_gem_object_needs_mb(struct drm_i915_gem_object *obj)
+{
+	return obj && obj->base.read_domains & I915_GEM_DOMAIN_GTT;
+}
+
 static void i915_gem_write_fence(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Ensure that all CPU reads are completed before installing a fence
+	 * and all writes before removing the fence.
+	 */
+	if (i915_gem_object_needs_mb(dev_priv->fence_regs[reg].obj))
+		mb();
+
 	switch (INTEL_INFO(dev)->gen) {
 	case 7:
-	case 6: sandybridge_write_fence_reg(dev, reg, obj); break;
+	case 6:
 	case 5:
 	case 4: i965_write_fence_reg(dev, reg, obj); break;
 	case 3: i915_write_fence_reg(dev, reg, obj); break;
 	case 2: i830_write_fence_reg(dev, reg, obj); break;
-	default: break;
+	default: BUG();
 	}
+
+	/* And similarly be paranoid that no direct access to this region
+	 * is reordered to before the fence is installed.
+	 */
+	if (i915_gem_object_needs_mb(obj))
+		mb();
 }
 
 static inline int fence_number(struct drm_i915_private *dev_priv,
@@ -2686,7 +2699,7 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
 }
 
 static int
-i915_gem_object_flush_fence(struct drm_i915_gem_object *obj)
+i915_gem_object_wait_fence(struct drm_i915_gem_object *obj)
 {
 	if (obj->last_fenced_seqno) {
 		int ret = i915_wait_seqno(obj->ring, obj->last_fenced_seqno);
@@ -2696,12 +2709,6 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj)
 		obj->last_fenced_seqno = 0;
 	}
 
-	/* Ensure that all CPU reads are completed before installing a fence
-	 * and all writes before removing the fence.
-	 */
-	if (obj->base.read_domains & I915_GEM_DOMAIN_GTT)
-		mb();
-
 	obj->fenced_gpu_access = false;
 	return 0;
 }
@@ -2712,7 +2719,7 @@ i915_gem_object_put_fence(struct drm_i915_gem_object *obj)
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	int ret;
 
-	ret = i915_gem_object_flush_fence(obj);
+	ret = i915_gem_object_wait_fence(obj);
 	if (ret)
 		return ret;
 
@@ -2786,7 +2793,7 @@ i915_gem_object_get_fence(struct drm_i915_gem_object *obj)
 	 * will need to serialise the write to the associated fence register?
 	 */
 	if (obj->fence_dirty) {
-		ret = i915_gem_object_flush_fence(obj);
+		ret = i915_gem_object_wait_fence(obj);
 		if (ret)
 			return ret;
 	}
@@ -2807,7 +2814,7 @@ i915_gem_object_get_fence(struct drm_i915_gem_object *obj)
 		if (reg->obj) {
 			struct drm_i915_gem_object *old = reg->obj;
 
-			ret = i915_gem_object_flush_fence(old);
+			ret = i915_gem_object_wait_fence(old);
 			if (ret)
 				return ret;
 
@@ -2830,7 +2837,7 @@ static bool i915_gem_valid_gtt_space(struct drm_device *dev,
 
 	/* On non-LLC machines we have to be careful when putting differing
 	 * types of snoopable memory together to avoid the prefetcher
-	 * crossing memory domains and dieing.
+	 * crossing memory domains and dying.
 	 */
 	if (HAS_LLC(dev))
 		return true;
@@ -2908,21 +2915,16 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
 	bool mappable, fenceable;
 	int ret;
 
-	if (obj->madv != I915_MADV_WILLNEED) {
-		DRM_ERROR("Attempting to bind a purgeable object\n");
-		return -EINVAL;
-	}
-
 	fence_size = i915_gem_get_gtt_size(dev,
 					   obj->base.size,
 					   obj->tiling_mode);
 	fence_alignment = i915_gem_get_gtt_alignment(dev,
 						     obj->base.size,
-						     obj->tiling_mode);
+						     obj->tiling_mode, true);
 	unfenced_alignment =
-		i915_gem_get_unfenced_gtt_alignment(dev,
+		i915_gem_get_gtt_alignment(dev,
 						    obj->base.size,
-						    obj->tiling_mode);
+						    obj->tiling_mode, false);
 
 	if (alignment == 0)
 		alignment = map_and_fenceable ? fence_alignment :
@@ -2938,7 +2940,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
 	 * before evicting everything in a vain attempt to find space.
 	 */
 	if (obj->base.size >
-	    (map_and_fenceable ? dev_priv->mm.gtt_mappable_end : dev_priv->mm.gtt_total)) {
+	    (map_and_fenceable ? dev_priv->gtt.mappable_end : dev_priv->gtt.total)) {
 		DRM_ERROR("Attempting to bind an object larger than the aperture\n");
 		return -E2BIG;
 	}
@@ -2959,7 +2961,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
 	if (map_and_fenceable)
 		ret = drm_mm_insert_node_in_range_generic(&dev_priv->mm.gtt_space, node,
 							  size, alignment, obj->cache_level,
-							  0, dev_priv->mm.gtt_mappable_end);
+							  0, dev_priv->gtt.mappable_end);
 	else
 		ret = drm_mm_insert_node_generic(&dev_priv->mm.gtt_space, node,
 						 size, alignment, obj->cache_level);
@@ -2999,7 +3001,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
 		(node->start & (fence_alignment - 1)) == 0;
 
 	mappable =
-		obj->gtt_offset + obj->base.size <= dev_priv->mm.gtt_mappable_end;
+		obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end;
 
 	obj->map_and_fenceable = mappable && fenceable;
 
@@ -3019,6 +3021,13 @@ i915_gem_clflush_object(struct drm_i915_gem_object *obj)
 	if (obj->pages == NULL)
 		return;
 
+	/*
+	 * Stolen memory is always coherent with the GPU as it is explicitly
+	 * marked as wc by the system, or the system is cache-coherent.
+	 */
+	if (obj->stolen)
+		return;
+
 	/* If the GPU is snooping the contents of the CPU cache,
 	 * we do not need to manually clear the CPU cache lines.  However,
 	 * the caches are only snooped when the render cache is
@@ -3107,6 +3116,13 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 
 	i915_gem_object_flush_cpu_write_domain(obj);
 
+	/* Serialise direct access to this object with the barriers for
+	 * coherent writes from the GPU, by effectively invalidating the
+	 * GTT domain upon first access.
+	 */
+	if ((obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0)
+		mb();
+
 	old_write_domain = obj->base.write_domain;
 	old_read_domains = obj->base.read_domains;
 
@@ -3413,11 +3429,17 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	unsigned long recent_enough = jiffies - msecs_to_jiffies(20);
 	struct drm_i915_gem_request *request;
 	struct intel_ring_buffer *ring = NULL;
+	unsigned reset_counter;
 	u32 seqno = 0;
 	int ret;
 
-	if (atomic_read(&dev_priv->mm.wedged))
-		return -EIO;
+	ret = i915_gem_wait_for_error(&dev_priv->gpu_error);
+	if (ret)
+		return ret;
+
+	ret = i915_gem_check_wedge(&dev_priv->gpu_error, false);
+	if (ret)
+		return ret;
 
 	spin_lock(&file_priv->mm.lock);
 	list_for_each_entry(request, &file_priv->mm.request_list, client_list) {
@@ -3427,12 +3449,13 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 		ring = request->ring;
 		seqno = request->seqno;
 	}
+	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	spin_unlock(&file_priv->mm.lock);
 
 	if (seqno == 0)
 		return 0;
 
-	ret = __wait_seqno(ring, seqno, true, NULL);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL);
 	if (ret == 0)
 		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
 
@@ -3706,14 +3729,14 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 {
 	struct drm_i915_gem_object *obj;
 	struct address_space *mapping;
-	u32 mask;
+	gfp_t mask;
 
-	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	obj = i915_gem_object_alloc(dev);
 	if (obj == NULL)
 		return NULL;
 
 	if (drm_gem_object_init(dev, &obj->base, size) != 0) {
-		kfree(obj);
+		i915_gem_object_free(obj);
 		return NULL;
 	}
 
@@ -3785,6 +3808,7 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	obj->pages_pin_count = 0;
 	i915_gem_object_put_pages(obj);
 	i915_gem_object_free_mmap_offset(obj);
+	i915_gem_object_release_stolen(obj);
 
 	BUG_ON(obj->pages);
 
@@ -3795,7 +3819,7 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	i915_gem_info_remove_obj(dev_priv, obj->base.size);
 
 	kfree(obj->bit_17);
-	kfree(obj);
+	i915_gem_object_free(obj);
 }
 
 int
@@ -3829,7 +3853,7 @@ i915_gem_idle(struct drm_device *dev)
 	 * And not confound mm.suspended!
 	 */
 	dev_priv->mm.suspended = 1;
-	del_timer_sync(&dev_priv->hangcheck_timer);
+	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
 
 	i915_kernel_lost_context(dev);
 	i915_gem_cleanup_ringbuffer(dev);
@@ -3848,7 +3872,7 @@ void i915_gem_l3_remap(struct drm_device *dev)
 	u32 misccpctl;
 	int i;
 
-	if (!IS_IVYBRIDGE(dev))
+	if (!HAS_L3_GPU_CACHE(dev))
 		return;
 
 	if (!dev_priv->l3_parity.remap_info)
@@ -3891,8 +3915,10 @@ void i915_gem_init_swizzling(struct drm_device *dev)
 	I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_SWZCTL);
 	if (IS_GEN6(dev))
 		I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_SNB));
-	else
+	else if (IS_GEN7(dev))
 		I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_IVB));
+	else
+		BUG();
 }
 
 static bool
@@ -3911,22 +3937,11 @@ intel_enable_blt(struct drm_device *dev)
 	return true;
 }
 
-int
-i915_gem_init_hw(struct drm_device *dev)
+static int i915_gem_init_rings(struct drm_device *dev)
 {
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
-		return -EIO;
-
-	if (IS_HASWELL(dev) && (I915_READ(0x120010) == 1))
-		I915_WRITE(0x9008, I915_READ(0x9008) | 0xf0000);
-
-	i915_gem_l3_remap(dev);
-
-	i915_gem_init_swizzling(dev);
-
 	ret = intel_init_render_ring_buffer(dev);
 	if (ret)
 		return ret;
@@ -3943,76 +3958,59 @@ i915_gem_init_hw(struct drm_device *dev)
 			goto cleanup_bsd_ring;
 	}
 
-	dev_priv->next_seqno = 1;
-
-	/*
-	 * XXX: There was some w/a described somewhere suggesting loading
-	 * contexts before PPGTT.
-	 */
-	i915_gem_context_init(dev);
-	i915_gem_init_ppgtt(dev);
+	ret = i915_gem_set_seqno(dev, ((u32)~0 - 0x1000));
+	if (ret)
+		goto cleanup_blt_ring;
 
 	return 0;
 
+cleanup_blt_ring:
+	intel_cleanup_ring_buffer(&dev_priv->ring[BCS]);
 cleanup_bsd_ring:
 	intel_cleanup_ring_buffer(&dev_priv->ring[VCS]);
 cleanup_render_ring:
 	intel_cleanup_ring_buffer(&dev_priv->ring[RCS]);
+
 	return ret;
 }
 
-static bool
-intel_enable_ppgtt(struct drm_device *dev)
+int
+i915_gem_init_hw(struct drm_device *dev)
 {
-	if (i915_enable_ppgtt >= 0)
-		return i915_enable_ppgtt;
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	int ret;
 
-#ifdef CONFIG_INTEL_IOMMU
-	/* Disable ppgtt on SNB if VT-d is on. */
-	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped)
-		return false;
-#endif
+	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
+		return -EIO;
 
-	return true;
+	if (IS_HASWELL(dev) && (I915_READ(0x120010) == 1))
+		I915_WRITE(0x9008, I915_READ(0x9008) | 0xf0000);
+
+	i915_gem_l3_remap(dev);
+
+	i915_gem_init_swizzling(dev);
+
+	ret = i915_gem_init_rings(dev);
+	if (ret)
+		return ret;
+
+	/*
+	 * XXX: There was some w/a described somewhere suggesting loading
+	 * contexts before PPGTT.
+	 */
+	i915_gem_context_init(dev);
+	i915_gem_init_ppgtt(dev);
+
+	return 0;
 }
 
 int i915_gem_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long gtt_size, mappable_size;
 	int ret;
 
-	gtt_size = dev_priv->mm.gtt->gtt_total_entries << PAGE_SHIFT;
-	mappable_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
-
 	mutex_lock(&dev->struct_mutex);
-	if (intel_enable_ppgtt(dev) && HAS_ALIASING_PPGTT(dev)) {
-		/* PPGTT pdes are stolen from global gtt ptes, so shrink the
-		 * aperture accordingly when using aliasing ppgtt. */
-		gtt_size -= I915_PPGTT_PD_ENTRIES*PAGE_SIZE;
-
-		i915_gem_init_global_gtt(dev, 0, mappable_size, gtt_size);
-
-		ret = i915_gem_init_aliasing_ppgtt(dev);
-		if (ret) {
-			mutex_unlock(&dev->struct_mutex);
-			return ret;
-		}
-	} else {
-		/* Let GEM Manage all of the aperture.
-		 *
-		 * However, leave one page at the end still bound to the scratch
-		 * page.  There are a number of places where the hardware
-		 * apparently prefetches past the end of the object, and we've
-		 * seen multiple hangs with the GPU head pointer stuck in a
-		 * batchbuffer bound at the last page of the aperture.  One page
-		 * should be enough to keep any prefetching inside of the
-		 * aperture.
-		 */
-		i915_gem_init_global_gtt(dev, 0, mappable_size,
-					 gtt_size);
-	}
-
+	i915_gem_init_global_gtt(dev);
 	ret = i915_gem_init_hw(dev);
 	mutex_unlock(&dev->struct_mutex);
 	if (ret) {
@@ -4047,9 +4045,9 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data,
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return 0;
 
-	if (atomic_read(&dev_priv->mm.wedged)) {
+	if (i915_reset_in_progress(&dev_priv->gpu_error)) {
 		DRM_ERROR("Reenabling wedged hardware, good luck\n");
-		atomic_set(&dev_priv->mm.wedged, 0);
+		atomic_set(&dev_priv->gpu_error.reset_counter, 0);
 	}
 
 	mutex_lock(&dev->struct_mutex);
@@ -4113,8 +4111,14 @@ init_ring_lists(struct intel_ring_buffer *ring)
 void
 i915_gem_load(struct drm_device *dev)
 {
-	int i;
 	drm_i915_private_t *dev_priv = dev->dev_private;
+	int i;
+
+	dev_priv->slab =
+		kmem_cache_create("i915_gem_object",
+				  sizeof(struct drm_i915_gem_object), 0,
+				  SLAB_HWCACHE_ALIGN,
+				  NULL);
 
 	INIT_LIST_HEAD(&dev_priv->mm.active_list);
 	INIT_LIST_HEAD(&dev_priv->mm.inactive_list);
@@ -4127,7 +4131,7 @@ i915_gem_load(struct drm_device *dev)
 		INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
 	INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
 			  i915_gem_retire_work_handler);
-	init_completion(&dev_priv->error_completion);
+	init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
 
 	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
 	if (IS_GEN3(dev)) {
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index a3f06bcad551..21177d9df423 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -126,13 +126,8 @@ static int get_context_size(struct drm_device *dev)
 
 static void do_destroy(struct i915_hw_context *ctx)
 {
-	struct drm_device *dev = ctx->obj->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
 	if (ctx->file_priv)
 		idr_remove(&ctx->file_priv->context_idr, ctx->id);
-	else
-		BUG_ON(ctx != dev_priv->ring[RCS].default_context);
 
 	drm_gem_object_unreference(&ctx->obj->base);
 	kfree(ctx);
@@ -242,7 +237,6 @@ err_destroy:
 void i915_gem_context_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t ctx_size;
 
 	if (!HAS_HW_CONTEXTS(dev)) {
 		dev_priv->hw_contexts_disabled = true;
@@ -254,11 +248,9 @@ void i915_gem_context_init(struct drm_device *dev)
 	    dev_priv->ring[RCS].default_context)
 		return;
 
-	ctx_size = get_context_size(dev);
-	dev_priv->hw_context_size = get_context_size(dev);
-	dev_priv->hw_context_size = round_up(dev_priv->hw_context_size, 4096);
+	dev_priv->hw_context_size = round_up(get_context_size(dev), 4096);
 
-	if (ctx_size <= 0 || ctx_size > (1<<20)) {
+	if (dev_priv->hw_context_size > (1<<20)) {
 		dev_priv->hw_contexts_disabled = true;
 		return;
 	}
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index abeaafef6d7e..6a5af6828624 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -281,8 +281,7 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 	if (IS_ERR(attach))
 		return ERR_CAST(attach);
 
-
-	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	obj = i915_gem_object_alloc(dev);
 	if (obj == NULL) {
 		ret = -ENOMEM;
 		goto fail_detach;
@@ -290,7 +289,7 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 
 	ret = drm_gem_private_object_init(dev, &obj->base, dma_buf->size);
 	if (ret) {
-		kfree(obj);
+		i915_gem_object_free(obj);
 		goto fail_detach;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index 776a3225184c..c86d5d9356fd 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -80,7 +80,7 @@ i915_gem_evict_something(struct drm_device *dev, int min_size,
 	if (mappable)
 		drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space,
 					    min_size, alignment, cache_level,
-					    0, dev_priv->mm.gtt_mappable_end);
+					    0, dev_priv->gtt.mappable_end);
 	else
 		drm_mm_init_scan(&dev_priv->mm.gtt_space,
 				 min_size, alignment, cache_level);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 26d08bb58218..2f2daebd0eef 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -34,61 +34,133 @@
 #include <linux/dma_remapping.h>
 
 struct eb_objects {
+	struct list_head objects;
 	int and;
-	struct hlist_head buckets[0];
+	union {
+		struct drm_i915_gem_object *lut[0];
+		struct hlist_head buckets[0];
+	};
 };
 
 static struct eb_objects *
-eb_create(int size)
+eb_create(struct drm_i915_gem_execbuffer2 *args)
 {
-	struct eb_objects *eb;
-	int count = PAGE_SIZE / sizeof(struct hlist_head) / 2;
-	BUILD_BUG_ON(!is_power_of_2(PAGE_SIZE / sizeof(struct hlist_head)));
-	while (count > size)
-		count >>= 1;
-	eb = kzalloc(count*sizeof(struct hlist_head) +
-		     sizeof(struct eb_objects),
-		     GFP_KERNEL);
-	if (eb == NULL)
-		return eb;
-
-	eb->and = count - 1;
+	struct eb_objects *eb = NULL;
+
+	if (args->flags & I915_EXEC_HANDLE_LUT) {
+		int size = args->buffer_count;
+		size *= sizeof(struct drm_i915_gem_object *);
+		size += sizeof(struct eb_objects);
+		eb = kmalloc(size, GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+	}
+
+	if (eb == NULL) {
+		int size = args->buffer_count;
+		int count = PAGE_SIZE / sizeof(struct hlist_head) / 2;
+		BUILD_BUG_ON(!is_power_of_2(PAGE_SIZE / sizeof(struct hlist_head)));
+		while (count > 2*size)
+			count >>= 1;
+		eb = kzalloc(count*sizeof(struct hlist_head) +
+			     sizeof(struct eb_objects),
+			     GFP_TEMPORARY);
+		if (eb == NULL)
+			return eb;
+
+		eb->and = count - 1;
+	} else
+		eb->and = -args->buffer_count;
+
+	INIT_LIST_HEAD(&eb->objects);
 	return eb;
 }
 
 static void
 eb_reset(struct eb_objects *eb)
 {
-	memset(eb->buckets, 0, (eb->and+1)*sizeof(struct hlist_head));
+	if (eb->and >= 0)
+		memset(eb->buckets, 0, (eb->and+1)*sizeof(struct hlist_head));
 }
 
-static void
-eb_add_object(struct eb_objects *eb, struct drm_i915_gem_object *obj)
+static int
+eb_lookup_objects(struct eb_objects *eb,
+		  struct drm_i915_gem_exec_object2 *exec,
+		  const struct drm_i915_gem_execbuffer2 *args,
+		  struct drm_file *file)
 {
-	hlist_add_head(&obj->exec_node,
-		       &eb->buckets[obj->exec_handle & eb->and]);
+	int i;
+
+	spin_lock(&file->table_lock);
+	for (i = 0; i < args->buffer_count; i++) {
+		struct drm_i915_gem_object *obj;
+
+		obj = to_intel_bo(idr_find(&file->object_idr, exec[i].handle));
+		if (obj == NULL) {
+			spin_unlock(&file->table_lock);
+			DRM_DEBUG("Invalid object handle %d at index %d\n",
+				   exec[i].handle, i);
+			return -ENOENT;
+		}
+
+		if (!list_empty(&obj->exec_list)) {
+			spin_unlock(&file->table_lock);
+			DRM_DEBUG("Object %p [handle %d, index %d] appears more than once in object list\n",
+				   obj, exec[i].handle, i);
+			return -EINVAL;
+		}
+
+		drm_gem_object_reference(&obj->base);
+		list_add_tail(&obj->exec_list, &eb->objects);
+
+		obj->exec_entry = &exec[i];
+		if (eb->and < 0) {
+			eb->lut[i] = obj;
+		} else {
+			uint32_t handle = args->flags & I915_EXEC_HANDLE_LUT ? i : exec[i].handle;
+			obj->exec_handle = handle;
+			hlist_add_head(&obj->exec_node,
+				       &eb->buckets[handle & eb->and]);
+		}
+	}
+	spin_unlock(&file->table_lock);
+
+	return 0;
 }
 
 static struct drm_i915_gem_object *
 eb_get_object(struct eb_objects *eb, unsigned long handle)
 {
-	struct hlist_head *head;
-	struct hlist_node *node;
-	struct drm_i915_gem_object *obj;
+	if (eb->and < 0) {
+		if (handle >= -eb->and)
+			return NULL;
+		return eb->lut[handle];
+	} else {
+		struct hlist_head *head;
+		struct hlist_node *node;
 
-	head = &eb->buckets[handle & eb->and];
-	hlist_for_each(node, head) {
-		obj = hlist_entry(node, struct drm_i915_gem_object, exec_node);
-		if (obj->exec_handle == handle)
-			return obj;
-	}
+		head = &eb->buckets[handle & eb->and];
+		hlist_for_each(node, head) {
+			struct drm_i915_gem_object *obj;
 
-	return NULL;
+			obj = hlist_entry(node, struct drm_i915_gem_object, exec_node);
+			if (obj->exec_handle == handle)
+				return obj;
+		}
+		return NULL;
+	}
 }
 
 static void
 eb_destroy(struct eb_objects *eb)
 {
+	while (!list_empty(&eb->objects)) {
+		struct drm_i915_gem_object *obj;
+
+		obj = list_first_entry(&eb->objects,
+				       struct drm_i915_gem_object,
+				       exec_list);
+		list_del_init(&obj->exec_list);
+		drm_gem_object_unreference(&obj->base);
+	}
 	kfree(eb);
 }
 
@@ -150,17 +222,6 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 			  reloc->write_domain);
 		return ret;
 	}
-	if (unlikely(reloc->write_domain && target_obj->pending_write_domain &&
-		     reloc->write_domain != target_obj->pending_write_domain)) {
-		DRM_DEBUG("Write domain conflict: "
-			  "obj %p target %d offset %d "
-			  "new %08x old %08x\n",
-			  obj, reloc->target_handle,
-			  (int) reloc->offset,
-			  reloc->write_domain,
-			  target_obj->pending_write_domain);
-		return ret;
-	}
 
 	target_obj->pending_read_domains |= reloc->read_domains;
 	target_obj->pending_write_domain |= reloc->write_domain;
@@ -220,7 +281,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 
 		/* Map the page containing the relocation we're going to perform.  */
 		reloc->offset += obj->gtt_offset;
-		reloc_page = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping,
+		reloc_page = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
 						      reloc->offset & PAGE_MASK);
 		reloc_entry = (uint32_t __iomem *)
 			(reloc_page + (reloc->offset & ~PAGE_MASK));
@@ -299,8 +360,7 @@ i915_gem_execbuffer_relocate_object_slow(struct drm_i915_gem_object *obj,
 
 static int
 i915_gem_execbuffer_relocate(struct drm_device *dev,
-			     struct eb_objects *eb,
-			     struct list_head *objects)
+			     struct eb_objects *eb)
 {
 	struct drm_i915_gem_object *obj;
 	int ret = 0;
@@ -313,7 +373,7 @@ i915_gem_execbuffer_relocate(struct drm_device *dev,
 	 * lockdep complains vehemently.
 	 */
 	pagefault_disable();
-	list_for_each_entry(obj, objects, exec_list) {
+	list_for_each_entry(obj, &eb->objects, exec_list) {
 		ret = i915_gem_execbuffer_relocate_object(obj, eb);
 		if (ret)
 			break;
@@ -335,7 +395,8 @@ need_reloc_mappable(struct drm_i915_gem_object *obj)
 
 static int
 i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj,
-				   struct intel_ring_buffer *ring)
+				   struct intel_ring_buffer *ring,
+				   bool *need_reloc)
 {
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
@@ -376,7 +437,20 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj,
 		obj->has_aliasing_ppgtt_mapping = 1;
 	}
 
-	entry->offset = obj->gtt_offset;
+	if (entry->offset != obj->gtt_offset) {
+		entry->offset = obj->gtt_offset;
+		*need_reloc = true;
+	}
+
+	if (entry->flags & EXEC_OBJECT_WRITE) {
+		obj->base.pending_read_domains = I915_GEM_DOMAIN_RENDER;
+		obj->base.pending_write_domain = I915_GEM_DOMAIN_RENDER;
+	}
+
+	if (entry->flags & EXEC_OBJECT_NEEDS_GTT &&
+	    !obj->has_global_gtt_mapping)
+		i915_gem_gtt_bind_object(obj, obj->cache_level);
+
 	return 0;
 }
 
@@ -402,7 +476,8 @@ i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj)
 static int
 i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 			    struct drm_file *file,
-			    struct list_head *objects)
+			    struct list_head *objects,
+			    bool *need_relocs)
 {
 	struct drm_i915_gem_object *obj;
 	struct list_head ordered_objects;
@@ -430,7 +505,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 		else
 			list_move_tail(&obj->exec_list, &ordered_objects);
 
-		obj->base.pending_read_domains = 0;
+		obj->base.pending_read_domains = I915_GEM_GPU_DOMAINS & ~I915_GEM_DOMAIN_COMMAND;
 		obj->base.pending_write_domain = 0;
 		obj->pending_fenced_gpu_access = false;
 	}
@@ -470,7 +545,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 			    (need_mappable && !obj->map_and_fenceable))
 				ret = i915_gem_object_unbind(obj);
 			else
-				ret = i915_gem_execbuffer_reserve_object(obj, ring);
+				ret = i915_gem_execbuffer_reserve_object(obj, ring, need_relocs);
 			if (ret)
 				goto err;
 		}
@@ -480,7 +555,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 			if (obj->gtt_space)
 				continue;
 
-			ret = i915_gem_execbuffer_reserve_object(obj, ring);
+			ret = i915_gem_execbuffer_reserve_object(obj, ring, need_relocs);
 			if (ret)
 				goto err;
 		}
@@ -500,21 +575,22 @@ err:		/* Decrement pin count for bound objects */
 
 static int
 i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
+				  struct drm_i915_gem_execbuffer2 *args,
 				  struct drm_file *file,
 				  struct intel_ring_buffer *ring,
-				  struct list_head *objects,
 				  struct eb_objects *eb,
-				  struct drm_i915_gem_exec_object2 *exec,
-				  int count)
+				  struct drm_i915_gem_exec_object2 *exec)
 {
 	struct drm_i915_gem_relocation_entry *reloc;
 	struct drm_i915_gem_object *obj;
+	bool need_relocs;
 	int *reloc_offset;
 	int i, total, ret;
+	int count = args->buffer_count;
 
 	/* We may process another execbuffer during the unlock... */
-	while (!list_empty(objects)) {
-		obj = list_first_entry(objects,
+	while (!list_empty(&eb->objects)) {
+		obj = list_first_entry(&eb->objects,
 				       struct drm_i915_gem_object,
 				       exec_list);
 		list_del_init(&obj->exec_list);
@@ -582,27 +658,16 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 
 	/* reacquire the objects */
 	eb_reset(eb);
-	for (i = 0; i < count; i++) {
-		obj = to_intel_bo(drm_gem_object_lookup(dev, file,
-							exec[i].handle));
-		if (&obj->base == NULL) {
-			DRM_DEBUG("Invalid object handle %d at index %d\n",
-				   exec[i].handle, i);
-			ret = -ENOENT;
-			goto err;
-		}
-
-		list_add_tail(&obj->exec_list, objects);
-		obj->exec_handle = exec[i].handle;
-		obj->exec_entry = &exec[i];
-		eb_add_object(eb, obj);
-	}
+	ret = eb_lookup_objects(eb, exec, args, file);
+	if (ret)
+		goto err;
 
-	ret = i915_gem_execbuffer_reserve(ring, file, objects);
+	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
+	ret = i915_gem_execbuffer_reserve(ring, file, &eb->objects, &need_relocs);
 	if (ret)
 		goto err;
 
-	list_for_each_entry(obj, objects, exec_list) {
+	list_for_each_entry(obj, &eb->objects, exec_list) {
 		int offset = obj->exec_entry - exec;
 		ret = i915_gem_execbuffer_relocate_object_slow(obj, eb,
 							       reloc + reloc_offset[offset]);
@@ -623,44 +688,11 @@ err:
 }
 
 static int
-i915_gem_execbuffer_wait_for_flips(struct intel_ring_buffer *ring, u32 flips)
-{
-	u32 plane, flip_mask;
-	int ret;
-
-	/* Check for any pending flips. As we only maintain a flip queue depth
-	 * of 1, we can simply insert a WAIT for the next display flip prior
-	 * to executing the batch and avoid stalling the CPU.
-	 */
-
-	for (plane = 0; flips >> plane; plane++) {
-		if (((flips >> plane) & 1) == 0)
-			continue;
-
-		if (plane)
-			flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
-		else
-			flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-
-		ret = intel_ring_begin(ring, 2);
-		if (ret)
-			return ret;
-
-		intel_ring_emit(ring, MI_WAIT_FOR_EVENT | flip_mask);
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_advance(ring);
-	}
-
-	return 0;
-}
-
-static int
 i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
 				struct list_head *objects)
 {
 	struct drm_i915_gem_object *obj;
 	uint32_t flush_domains = 0;
-	uint32_t flips = 0;
 	int ret;
 
 	list_for_each_entry(obj, objects, exec_list) {
@@ -671,18 +703,9 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
 		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
 			i915_gem_clflush_object(obj);
 
-		if (obj->base.pending_write_domain)
-			flips |= atomic_read(&obj->pending_flip);
-
 		flush_domains |= obj->base.write_domain;
 	}
 
-	if (flips) {
-		ret = i915_gem_execbuffer_wait_for_flips(ring, flips);
-		if (ret)
-			return ret;
-	}
-
 	if (flush_domains & I915_GEM_DOMAIN_CPU)
 		i915_gem_chipset_flush(ring->dev);
 
@@ -698,6 +721,9 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
 static bool
 i915_gem_check_execbuffer(struct drm_i915_gem_execbuffer2 *exec)
 {
+	if (exec->flags & __I915_EXEC_UNKNOWN_FLAGS)
+		return false;
+
 	return ((exec->batch_start_offset | exec->batch_len) & 0x7) == 0;
 }
 
@@ -711,6 +737,9 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 		char __user *ptr = (char __user *)(uintptr_t)exec[i].relocs_ptr;
 		int length; /* limited by fault_in_pages_readable() */
 
+		if (exec[i].flags & __EXEC_OBJECT_UNKNOWN_FLAGS)
+			return -EINVAL;
+
 		/* First check for malicious input causing overflow */
 		if (exec[i].relocation_count >
 		    INT_MAX / sizeof(struct drm_i915_gem_relocation_entry))
@@ -718,9 +747,6 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 
 		length = exec[i].relocation_count *
 			sizeof(struct drm_i915_gem_relocation_entry);
-		if (!access_ok(VERIFY_READ, ptr, length))
-			return -EFAULT;
-
 		/* we may also need to update the presumed offsets */
 		if (!access_ok(VERIFY_WRITE, ptr, length))
 			return -EFAULT;
@@ -742,8 +768,10 @@ i915_gem_execbuffer_move_to_active(struct list_head *objects,
 		u32 old_read = obj->base.read_domains;
 		u32 old_write = obj->base.write_domain;
 
-		obj->base.read_domains = obj->base.pending_read_domains;
 		obj->base.write_domain = obj->base.pending_write_domain;
+		if (obj->base.write_domain == 0)
+			obj->base.pending_read_domains |= obj->base.read_domains;
+		obj->base.read_domains = obj->base.pending_read_domains;
 		obj->fenced_gpu_access = obj->pending_fenced_gpu_access;
 
 		i915_gem_object_move_to_active(obj, ring);
@@ -802,21 +830,18 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		       struct drm_i915_gem_exec_object2 *exec)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct list_head objects;
 	struct eb_objects *eb;
 	struct drm_i915_gem_object *batch_obj;
 	struct drm_clip_rect *cliprects = NULL;
 	struct intel_ring_buffer *ring;
 	u32 ctx_id = i915_execbuffer2_get_context_id(*args);
 	u32 exec_start, exec_len;
-	u32 mask;
-	u32 flags;
+	u32 mask, flags;
 	int ret, mode, i;
+	bool need_relocs;
 
-	if (!i915_gem_check_execbuffer(args)) {
-		DRM_DEBUG("execbuf with invalid offset/length\n");
+	if (!i915_gem_check_execbuffer(args))
 		return -EINVAL;
-	}
 
 	ret = validate_exec_list(exec, args->buffer_count);
 	if (ret)
@@ -937,7 +962,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		goto pre_mutex_err;
 	}
 
-	eb = eb_create(args->buffer_count);
+	eb = eb_create(args);
 	if (eb == NULL) {
 		mutex_unlock(&dev->struct_mutex);
 		ret = -ENOMEM;
@@ -945,51 +970,28 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 
 	/* Look up object handles */
-	INIT_LIST_HEAD(&objects);
-	for (i = 0; i < args->buffer_count; i++) {
-		struct drm_i915_gem_object *obj;
-
-		obj = to_intel_bo(drm_gem_object_lookup(dev, file,
-							exec[i].handle));
-		if (&obj->base == NULL) {
-			DRM_DEBUG("Invalid object handle %d at index %d\n",
-				   exec[i].handle, i);
-			/* prevent error path from reading uninitialized data */
-			ret = -ENOENT;
-			goto err;
-		}
-
-		if (!list_empty(&obj->exec_list)) {
-			DRM_DEBUG("Object %p [handle %d, index %d] appears more than once in object list\n",
-				   obj, exec[i].handle, i);
-			ret = -EINVAL;
-			goto err;
-		}
-
-		list_add_tail(&obj->exec_list, &objects);
-		obj->exec_handle = exec[i].handle;
-		obj->exec_entry = &exec[i];
-		eb_add_object(eb, obj);
-	}
+	ret = eb_lookup_objects(eb, exec, args, file);
+	if (ret)
+		goto err;
 
 	/* take note of the batch buffer before we might reorder the lists */
-	batch_obj = list_entry(objects.prev,
+	batch_obj = list_entry(eb->objects.prev,
 			       struct drm_i915_gem_object,
 			       exec_list);
 
 	/* Move the objects en-masse into the GTT, evicting if necessary. */
-	ret = i915_gem_execbuffer_reserve(ring, file, &objects);
+	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
+	ret = i915_gem_execbuffer_reserve(ring, file, &eb->objects, &need_relocs);
 	if (ret)
 		goto err;
 
 	/* The objects are in their final locations, apply the relocations. */
-	ret = i915_gem_execbuffer_relocate(dev, eb, &objects);
+	if (need_relocs)
+		ret = i915_gem_execbuffer_relocate(dev, eb);
 	if (ret) {
 		if (ret == -EFAULT) {
-			ret = i915_gem_execbuffer_relocate_slow(dev, file, ring,
-								&objects, eb,
-								exec,
-								args->buffer_count);
+			ret = i915_gem_execbuffer_relocate_slow(dev, args, file, ring,
+								eb, exec);
 			BUG_ON(!mutex_is_locked(&dev->struct_mutex));
 		}
 		if (ret)
@@ -1011,7 +1013,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (flags & I915_DISPATCH_SECURE && !batch_obj->has_global_gtt_mapping)
 		i915_gem_gtt_bind_object(batch_obj, batch_obj->cache_level);
 
-	ret = i915_gem_execbuffer_move_to_gpu(ring, &objects);
+	ret = i915_gem_execbuffer_move_to_gpu(ring, &eb->objects);
 	if (ret)
 		goto err;
 
@@ -1065,20 +1067,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
 
-	i915_gem_execbuffer_move_to_active(&objects, ring);
+	i915_gem_execbuffer_move_to_active(&eb->objects, ring);
 	i915_gem_execbuffer_retire_commands(dev, file, ring);
 
 err:
 	eb_destroy(eb);
-	while (!list_empty(&objects)) {
-		struct drm_i915_gem_object *obj;
-
-		obj = list_first_entry(&objects,
-				       struct drm_i915_gem_object,
-				       exec_list);
-		list_del_init(&obj->exec_list);
-		drm_gem_object_unreference(&obj->base);
-	}
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -1187,7 +1180,7 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data,
 	}
 
 	exec2_list = kmalloc(sizeof(*exec2_list)*args->buffer_count,
-			     GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY);
+			     GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 	if (exec2_list == NULL)
 		exec2_list = drm_malloc_ab(sizeof(*exec2_list),
 					   args->buffer_count);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 2c150dee78a7..926a1e2dd234 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -44,9 +44,9 @@ typedef uint32_t gtt_pte_t;
 #define GEN6_PTE_CACHE_LLC_MLC		(3 << 1)
 #define GEN6_PTE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
 
-static inline gtt_pte_t pte_encode(struct drm_device *dev,
-				   dma_addr_t addr,
-				   enum i915_cache_level level)
+static inline gtt_pte_t gen6_pte_encode(struct drm_device *dev,
+					dma_addr_t addr,
+					enum i915_cache_level level)
 {
 	gtt_pte_t pte = GEN6_PTE_VALID;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
@@ -77,7 +77,7 @@ static inline gtt_pte_t pte_encode(struct drm_device *dev,
 }
 
 /* PPGTT support for Sandybdrige/Gen6 and later */
-static void i915_ppgtt_clear_range(struct i915_hw_ppgtt *ppgtt,
+static void gen6_ppgtt_clear_range(struct i915_hw_ppgtt *ppgtt,
 				   unsigned first_entry,
 				   unsigned num_entries)
 {
@@ -87,8 +87,9 @@ static void i915_ppgtt_clear_range(struct i915_hw_ppgtt *ppgtt,
 	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
 	unsigned last_pte, i;
 
-	scratch_pte = pte_encode(ppgtt->dev, ppgtt->scratch_page_dma_addr,
-				 I915_CACHE_LLC);
+	scratch_pte = gen6_pte_encode(ppgtt->dev,
+				      ppgtt->scratch_page_dma_addr,
+				      I915_CACHE_LLC);
 
 	while (num_entries) {
 		last_pte = first_pte + num_entries;
@@ -108,10 +109,72 @@ static void i915_ppgtt_clear_range(struct i915_hw_ppgtt *ppgtt,
 	}
 }
 
-int i915_gem_init_aliasing_ppgtt(struct drm_device *dev)
+static void gen6_ppgtt_insert_entries(struct i915_hw_ppgtt *ppgtt,
+				      struct sg_table *pages,
+				      unsigned first_entry,
+				      enum i915_cache_level cache_level)
 {
+	gtt_pte_t *pt_vaddr;
+	unsigned act_pd = first_entry / I915_PPGTT_PT_ENTRIES;
+	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
+	unsigned i, j, m, segment_len;
+	dma_addr_t page_addr;
+	struct scatterlist *sg;
+
+	/* init sg walking */
+	sg = pages->sgl;
+	i = 0;
+	segment_len = sg_dma_len(sg) >> PAGE_SHIFT;
+	m = 0;
+
+	while (i < pages->nents) {
+		pt_vaddr = kmap_atomic(ppgtt->pt_pages[act_pd]);
+
+		for (j = first_pte; j < I915_PPGTT_PT_ENTRIES; j++) {
+			page_addr = sg_dma_address(sg) + (m << PAGE_SHIFT);
+			pt_vaddr[j] = gen6_pte_encode(ppgtt->dev, page_addr,
+						      cache_level);
+
+			/* grab the next page */
+			if (++m == segment_len) {
+				if (++i == pages->nents)
+					break;
+
+				sg = sg_next(sg);
+				segment_len = sg_dma_len(sg) >> PAGE_SHIFT;
+				m = 0;
+			}
+		}
+
+		kunmap_atomic(pt_vaddr);
+
+		first_pte = 0;
+		act_pd++;
+	}
+}
+
+static void gen6_ppgtt_cleanup(struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
+
+	if (ppgtt->pt_dma_addr) {
+		for (i = 0; i < ppgtt->num_pd_entries; i++)
+			pci_unmap_page(ppgtt->dev->pdev,
+				       ppgtt->pt_dma_addr[i],
+				       4096, PCI_DMA_BIDIRECTIONAL);
+	}
+
+	kfree(ppgtt->pt_dma_addr);
+	for (i = 0; i < ppgtt->num_pd_entries; i++)
+		__free_page(ppgtt->pt_pages[i]);
+	kfree(ppgtt->pt_pages);
+	kfree(ppgtt);
+}
+
+static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
+{
+	struct drm_device *dev = ppgtt->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_ppgtt *ppgtt;
 	unsigned first_pd_entry_in_global_pt;
 	int i;
 	int ret = -ENOMEM;
@@ -119,18 +182,17 @@ int i915_gem_init_aliasing_ppgtt(struct drm_device *dev)
 	/* ppgtt PDEs reside in the global gtt pagetable, which has 512*1024
 	 * entries. For aliasing ppgtt support we just steal them at the end for
 	 * now. */
-	first_pd_entry_in_global_pt = dev_priv->mm.gtt->gtt_total_entries - I915_PPGTT_PD_ENTRIES;
-
-	ppgtt = kzalloc(sizeof(*ppgtt), GFP_KERNEL);
-	if (!ppgtt)
-		return ret;
+	first_pd_entry_in_global_pt =
+		gtt_total_entries(dev_priv->gtt) - I915_PPGTT_PD_ENTRIES;
 
-	ppgtt->dev = dev;
 	ppgtt->num_pd_entries = I915_PPGTT_PD_ENTRIES;
+	ppgtt->clear_range = gen6_ppgtt_clear_range;
+	ppgtt->insert_entries = gen6_ppgtt_insert_entries;
+	ppgtt->cleanup = gen6_ppgtt_cleanup;
 	ppgtt->pt_pages = kzalloc(sizeof(struct page *)*ppgtt->num_pd_entries,
 				  GFP_KERNEL);
 	if (!ppgtt->pt_pages)
-		goto err_ppgtt;
+		return -ENOMEM;
 
 	for (i = 0; i < ppgtt->num_pd_entries; i++) {
 		ppgtt->pt_pages[i] = alloc_page(GFP_KERNEL);
@@ -138,39 +200,32 @@ int i915_gem_init_aliasing_ppgtt(struct drm_device *dev)
 			goto err_pt_alloc;
 	}
 
-	if (dev_priv->mm.gtt->needs_dmar) {
-		ppgtt->pt_dma_addr = kzalloc(sizeof(dma_addr_t)
-						*ppgtt->num_pd_entries,
-					     GFP_KERNEL);
-		if (!ppgtt->pt_dma_addr)
-			goto err_pt_alloc;
+	ppgtt->pt_dma_addr = kzalloc(sizeof(dma_addr_t) *ppgtt->num_pd_entries,
+				     GFP_KERNEL);
+	if (!ppgtt->pt_dma_addr)
+		goto err_pt_alloc;
 
-		for (i = 0; i < ppgtt->num_pd_entries; i++) {
-			dma_addr_t pt_addr;
+	for (i = 0; i < ppgtt->num_pd_entries; i++) {
+		dma_addr_t pt_addr;
 
-			pt_addr = pci_map_page(dev->pdev, ppgtt->pt_pages[i],
-					       0, 4096,
-					       PCI_DMA_BIDIRECTIONAL);
+		pt_addr = pci_map_page(dev->pdev, ppgtt->pt_pages[i], 0, 4096,
+				       PCI_DMA_BIDIRECTIONAL);
 
-			if (pci_dma_mapping_error(dev->pdev,
-						  pt_addr)) {
-				ret = -EIO;
-				goto err_pd_pin;
+		if (pci_dma_mapping_error(dev->pdev, pt_addr)) {
+			ret = -EIO;
+			goto err_pd_pin;
 
-			}
-			ppgtt->pt_dma_addr[i] = pt_addr;
 		}
+		ppgtt->pt_dma_addr[i] = pt_addr;
 	}
 
-	ppgtt->scratch_page_dma_addr = dev_priv->mm.gtt->scratch_page_dma;
+	ppgtt->scratch_page_dma_addr = dev_priv->gtt.scratch_page_dma;
 
-	i915_ppgtt_clear_range(ppgtt, 0,
-			       ppgtt->num_pd_entries*I915_PPGTT_PT_ENTRIES);
+	ppgtt->clear_range(ppgtt, 0,
+			   ppgtt->num_pd_entries*I915_PPGTT_PT_ENTRIES);
 
 	ppgtt->pd_offset = (first_pd_entry_in_global_pt)*sizeof(gtt_pte_t);
 
-	dev_priv->mm.aliasing_ppgtt = ppgtt;
-
 	return 0;
 
 err_pd_pin:
@@ -186,94 +241,57 @@ err_pt_alloc:
 			__free_page(ppgtt->pt_pages[i]);
 	}
 	kfree(ppgtt->pt_pages);
-err_ppgtt:
-	kfree(ppgtt);
 
 	return ret;
 }
 
-void i915_gem_cleanup_aliasing_ppgtt(struct drm_device *dev)
+static int i915_gem_init_aliasing_ppgtt(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
-	int i;
+	struct i915_hw_ppgtt *ppgtt;
+	int ret;
 
+	ppgtt = kzalloc(sizeof(*ppgtt), GFP_KERNEL);
 	if (!ppgtt)
-		return;
+		return -ENOMEM;
 
-	if (ppgtt->pt_dma_addr) {
-		for (i = 0; i < ppgtt->num_pd_entries; i++)
-			pci_unmap_page(dev->pdev, ppgtt->pt_dma_addr[i],
-				       4096, PCI_DMA_BIDIRECTIONAL);
-	}
+	ppgtt->dev = dev;
 
-	kfree(ppgtt->pt_dma_addr);
-	for (i = 0; i < ppgtt->num_pd_entries; i++)
-		__free_page(ppgtt->pt_pages[i]);
-	kfree(ppgtt->pt_pages);
-	kfree(ppgtt);
+	ret = gen6_ppgtt_init(ppgtt);
+	if (ret)
+		kfree(ppgtt);
+	else
+		dev_priv->mm.aliasing_ppgtt = ppgtt;
+
+	return ret;
 }
 
-static void i915_ppgtt_insert_sg_entries(struct i915_hw_ppgtt *ppgtt,
-					 const struct sg_table *pages,
-					 unsigned first_entry,
-					 enum i915_cache_level cache_level)
+void i915_gem_cleanup_aliasing_ppgtt(struct drm_device *dev)
 {
-	gtt_pte_t *pt_vaddr;
-	unsigned act_pd = first_entry / I915_PPGTT_PT_ENTRIES;
-	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
-	unsigned i, j, m, segment_len;
-	dma_addr_t page_addr;
-	struct scatterlist *sg;
-
-	/* init sg walking */
-	sg = pages->sgl;
-	i = 0;
-	segment_len = sg_dma_len(sg) >> PAGE_SHIFT;
-	m = 0;
-
-	while (i < pages->nents) {
-		pt_vaddr = kmap_atomic(ppgtt->pt_pages[act_pd]);
-
-		for (j = first_pte; j < I915_PPGTT_PT_ENTRIES; j++) {
-			page_addr = sg_dma_address(sg) + (m << PAGE_SHIFT);
-			pt_vaddr[j] = pte_encode(ppgtt->dev, page_addr,
-						 cache_level);
-
-			/* grab the next page */
-			if (++m == segment_len) {
-				if (++i == pages->nents)
-					break;
-
-				sg = sg_next(sg);
-				segment_len = sg_dma_len(sg) >> PAGE_SHIFT;
-				m = 0;
-			}
-		}
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 
-		kunmap_atomic(pt_vaddr);
+	if (!ppgtt)
+		return;
 
-		first_pte = 0;
-		act_pd++;
-	}
+	ppgtt->cleanup(ppgtt);
 }
 
 void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt,
 			    struct drm_i915_gem_object *obj,
 			    enum i915_cache_level cache_level)
 {
-	i915_ppgtt_insert_sg_entries(ppgtt,
-				     obj->pages,
-				     obj->gtt_space->start >> PAGE_SHIFT,
-				     cache_level);
+	ppgtt->insert_entries(ppgtt, obj->pages,
+			      obj->gtt_space->start >> PAGE_SHIFT,
+			      cache_level);
 }
 
 void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt,
 			      struct drm_i915_gem_object *obj)
 {
-	i915_ppgtt_clear_range(ppgtt,
-			       obj->gtt_space->start >> PAGE_SHIFT,
-			       obj->base.size >> PAGE_SHIFT);
+	ppgtt->clear_range(ppgtt,
+			   obj->gtt_space->start >> PAGE_SHIFT,
+			   obj->base.size >> PAGE_SHIFT);
 }
 
 void i915_gem_init_ppgtt(struct drm_device *dev)
@@ -282,7 +300,7 @@ void i915_gem_init_ppgtt(struct drm_device *dev)
 	uint32_t pd_offset;
 	struct intel_ring_buffer *ring;
 	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
-	uint32_t __iomem *pd_addr;
+	gtt_pte_t __iomem *pd_addr;
 	uint32_t pd_entry;
 	int i;
 
@@ -290,15 +308,11 @@ void i915_gem_init_ppgtt(struct drm_device *dev)
 		return;
 
 
-	pd_addr = dev_priv->mm.gtt->gtt + ppgtt->pd_offset/sizeof(uint32_t);
+	pd_addr = (gtt_pte_t __iomem*)dev_priv->gtt.gsm + ppgtt->pd_offset/sizeof(gtt_pte_t);
 	for (i = 0; i < ppgtt->num_pd_entries; i++) {
 		dma_addr_t pt_addr;
 
-		if (dev_priv->mm.gtt->needs_dmar)
-			pt_addr = ppgtt->pt_dma_addr[i];
-		else
-			pt_addr = page_to_phys(ppgtt->pt_pages[i]);
-
+		pt_addr = ppgtt->pt_dma_addr[i];
 		pd_entry = GEN6_PDE_ADDR_ENCODE(pt_addr);
 		pd_entry |= GEN6_PDE_VALID;
 
@@ -338,11 +352,27 @@ void i915_gem_init_ppgtt(struct drm_device *dev)
 	}
 }
 
+extern int intel_iommu_gfx_mapped;
+/* Certain Gen5 chipsets require require idling the GPU before
+ * unmapping anything from the GTT when VT-d is enabled.
+ */
+static inline bool needs_idle_maps(struct drm_device *dev)
+{
+#ifdef CONFIG_INTEL_IOMMU
+	/* Query intel_iommu to see if we need the workaround. Presumably that
+	 * was loaded first.
+	 */
+	if (IS_GEN5(dev) && IS_MOBILE(dev) && intel_iommu_gfx_mapped)
+		return true;
+#endif
+	return false;
+}
+
 static bool do_idling(struct drm_i915_private *dev_priv)
 {
 	bool ret = dev_priv->mm.interruptible;
 
-	if (unlikely(dev_priv->mm.gtt->do_idle_maps)) {
+	if (unlikely(dev_priv->gtt.do_idle_maps)) {
 		dev_priv->mm.interruptible = false;
 		if (i915_gpu_idle(dev_priv->dev)) {
 			DRM_ERROR("Couldn't idle GPU\n");
@@ -356,45 +386,18 @@ static bool do_idling(struct drm_i915_private *dev_priv)
 
 static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
 {
-	if (unlikely(dev_priv->mm.gtt->do_idle_maps))
+	if (unlikely(dev_priv->gtt.do_idle_maps))
 		dev_priv->mm.interruptible = interruptible;
 }
 
-
-static void i915_ggtt_clear_range(struct drm_device *dev,
-				 unsigned first_entry,
-				 unsigned num_entries)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	gtt_pte_t scratch_pte;
-	gtt_pte_t __iomem *gtt_base = dev_priv->mm.gtt->gtt + first_entry;
-	const int max_entries = dev_priv->mm.gtt->gtt_total_entries - first_entry;
-	int i;
-
-	if (INTEL_INFO(dev)->gen < 6) {
-		intel_gtt_clear_range(first_entry, num_entries);
-		return;
-	}
-
-	if (WARN(num_entries > max_entries,
-		 "First entry = %d; Num entries = %d (max=%d)\n",
-		 first_entry, num_entries, max_entries))
-		num_entries = max_entries;
-
-	scratch_pte = pte_encode(dev, dev_priv->mm.gtt->scratch_page_dma, I915_CACHE_LLC);
-	for (i = 0; i < num_entries; i++)
-		iowrite32(scratch_pte, &gtt_base[i]);
-	readl(gtt_base);
-}
-
 void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 
 	/* First fill our portion of the GTT with scratch pages */
-	i915_ggtt_clear_range(dev, dev_priv->mm.gtt_start / PAGE_SIZE,
-			      (dev_priv->mm.gtt_end - dev_priv->mm.gtt_start) / PAGE_SIZE);
+	dev_priv->gtt.gtt_clear_range(dev, dev_priv->gtt.start / PAGE_SIZE,
+				      dev_priv->gtt.total / PAGE_SIZE);
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) {
 		i915_gem_clflush_object(obj);
@@ -423,16 +426,15 @@ int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj)
  * within the global GTT as well as accessible by the GPU through the GMADR
  * mapped BAR (dev_priv->mm.gtt->gtt).
  */
-static void gen6_ggtt_bind_object(struct drm_i915_gem_object *obj,
-				  enum i915_cache_level level)
+static void gen6_ggtt_insert_entries(struct drm_device *dev,
+				     struct sg_table *st,
+				     unsigned int first_entry,
+				     enum i915_cache_level level)
 {
-	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct sg_table *st = obj->pages;
 	struct scatterlist *sg = st->sgl;
-	const int first_entry = obj->gtt_space->start >> PAGE_SHIFT;
-	const int max_entries = dev_priv->mm.gtt->gtt_total_entries - first_entry;
-	gtt_pte_t __iomem *gtt_entries = dev_priv->mm.gtt->gtt + first_entry;
+	gtt_pte_t __iomem *gtt_entries =
+		(gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
 	int unused, i = 0;
 	unsigned int len, m = 0;
 	dma_addr_t addr;
@@ -441,14 +443,12 @@ static void gen6_ggtt_bind_object(struct drm_i915_gem_object *obj,
 		len = sg_dma_len(sg) >> PAGE_SHIFT;
 		for (m = 0; m < len; m++) {
 			addr = sg_dma_address(sg) + (m << PAGE_SHIFT);
-			iowrite32(pte_encode(dev, addr, level), &gtt_entries[i]);
+			iowrite32(gen6_pte_encode(dev, addr, level),
+				  &gtt_entries[i]);
 			i++;
 		}
 	}
 
-	BUG_ON(i > max_entries);
-	BUG_ON(i != obj->base.size / PAGE_SIZE);
-
 	/* XXX: This serves as a posting read to make sure that the PTE has
 	 * actually been updated. There is some concern that even though
 	 * registers and PTEs are within the same BAR that they are potentially
@@ -456,7 +456,8 @@ static void gen6_ggtt_bind_object(struct drm_i915_gem_object *obj,
 	 * hardware should work, we must keep this posting read for paranoia.
 	 */
 	if (i != 0)
-		WARN_ON(readl(&gtt_entries[i-1]) != pte_encode(dev, addr, level));
+		WARN_ON(readl(&gtt_entries[i-1])
+			!= gen6_pte_encode(dev, addr, level));
 
 	/* This next bit makes the above posting read even more important. We
 	 * want to flush the TLBs only after we're certain all the PTE updates
@@ -466,28 +467,70 @@ static void gen6_ggtt_bind_object(struct drm_i915_gem_object *obj,
 	POSTING_READ(GFX_FLSH_CNTL_GEN6);
 }
 
+static void gen6_ggtt_clear_range(struct drm_device *dev,
+				  unsigned int first_entry,
+				  unsigned int num_entries)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	gtt_pte_t scratch_pte;
+	gtt_pte_t __iomem *gtt_base = (gtt_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
+	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
+	int i;
+
+	if (WARN(num_entries > max_entries,
+		 "First entry = %d; Num entries = %d (max=%d)\n",
+		 first_entry, num_entries, max_entries))
+		num_entries = max_entries;
+
+	scratch_pte = gen6_pte_encode(dev, dev_priv->gtt.scratch_page_dma,
+				      I915_CACHE_LLC);
+	for (i = 0; i < num_entries; i++)
+		iowrite32(scratch_pte, &gtt_base[i]);
+	readl(gtt_base);
+}
+
+
+static void i915_ggtt_insert_entries(struct drm_device *dev,
+				     struct sg_table *st,
+				     unsigned int pg_start,
+				     enum i915_cache_level cache_level)
+{
+	unsigned int flags = (cache_level == I915_CACHE_NONE) ?
+		AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
+
+	intel_gtt_insert_sg_entries(st, pg_start, flags);
+
+}
+
+static void i915_ggtt_clear_range(struct drm_device *dev,
+				  unsigned int first_entry,
+				  unsigned int num_entries)
+{
+	intel_gtt_clear_range(first_entry, num_entries);
+}
+
+
 void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,
 			      enum i915_cache_level cache_level)
 {
 	struct drm_device *dev = obj->base.dev;
-	if (INTEL_INFO(dev)->gen < 6) {
-		unsigned int flags = (cache_level == I915_CACHE_NONE) ?
-			AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
-		intel_gtt_insert_sg_entries(obj->pages,
-					    obj->gtt_space->start >> PAGE_SHIFT,
-					    flags);
-	} else {
-		gen6_ggtt_bind_object(obj, cache_level);
-	}
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->gtt.gtt_insert_entries(dev, obj->pages,
+					 obj->gtt_space->start >> PAGE_SHIFT,
+					 cache_level);
 
 	obj->has_global_gtt_mapping = 1;
 }
 
 void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)
 {
-	i915_ggtt_clear_range(obj->base.dev,
-			      obj->gtt_space->start >> PAGE_SHIFT,
-			      obj->base.size >> PAGE_SHIFT);
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->gtt.gtt_clear_range(obj->base.dev,
+				      obj->gtt_space->start >> PAGE_SHIFT,
+				      obj->base.size >> PAGE_SHIFT);
 
 	obj->has_global_gtt_mapping = 0;
 }
@@ -524,27 +567,101 @@ static void i915_gtt_color_adjust(struct drm_mm_node *node,
 			*end -= 4096;
 	}
 }
-
-void i915_gem_init_global_gtt(struct drm_device *dev,
-			      unsigned long start,
-			      unsigned long mappable_end,
-			      unsigned long end)
+void i915_gem_setup_global_gtt(struct drm_device *dev,
+			       unsigned long start,
+			       unsigned long mappable_end,
+			       unsigned long end)
 {
+	/* Let GEM Manage all of the aperture.
+	 *
+	 * However, leave one page at the end still bound to the scratch page.
+	 * There are a number of places where the hardware apparently prefetches
+	 * past the end of the object, and we've seen multiple hangs with the
+	 * GPU head pointer stuck in a batchbuffer bound at the last page of the
+	 * aperture.  One page should be enough to keep any prefetching inside
+	 * of the aperture.
+	 */
 	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_mm_node *entry;
+	struct drm_i915_gem_object *obj;
+	unsigned long hole_start, hole_end;
 
-	/* Substract the guard page ... */
+	BUG_ON(mappable_end > end);
+
+	/* Subtract the guard page ... */
 	drm_mm_init(&dev_priv->mm.gtt_space, start, end - start - PAGE_SIZE);
 	if (!HAS_LLC(dev))
 		dev_priv->mm.gtt_space.color_adjust = i915_gtt_color_adjust;
 
-	dev_priv->mm.gtt_start = start;
-	dev_priv->mm.gtt_mappable_end = mappable_end;
-	dev_priv->mm.gtt_end = end;
-	dev_priv->mm.gtt_total = end - start;
-	dev_priv->mm.mappable_gtt_total = min(end, mappable_end) - start;
+	/* Mark any preallocated objects as occupied */
+	list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) {
+		DRM_DEBUG_KMS("reserving preallocated space: %x + %zx\n",
+			      obj->gtt_offset, obj->base.size);
+
+		BUG_ON(obj->gtt_space != I915_GTT_RESERVED);
+		obj->gtt_space = drm_mm_create_block(&dev_priv->mm.gtt_space,
+						     obj->gtt_offset,
+						     obj->base.size,
+						     false);
+		obj->has_global_gtt_mapping = 1;
+	}
+
+	dev_priv->gtt.start = start;
+	dev_priv->gtt.total = end - start;
+
+	/* Clear any non-preallocated blocks */
+	drm_mm_for_each_hole(entry, &dev_priv->mm.gtt_space,
+			     hole_start, hole_end) {
+		DRM_DEBUG_KMS("clearing unused GTT space: [%lx, %lx]\n",
+			      hole_start, hole_end);
+		dev_priv->gtt.gtt_clear_range(dev, hole_start / PAGE_SIZE,
+					      (hole_end-hole_start) / PAGE_SIZE);
+	}
 
-	/* ... but ensure that we clear the entire range. */
-	i915_ggtt_clear_range(dev, start / PAGE_SIZE, (end-start) / PAGE_SIZE);
+	/* And finally clear the reserved guard page */
+	dev_priv->gtt.gtt_clear_range(dev, end / PAGE_SIZE - 1, 1);
+}
+
+static bool
+intel_enable_ppgtt(struct drm_device *dev)
+{
+	if (i915_enable_ppgtt >= 0)
+		return i915_enable_ppgtt;
+
+#ifdef CONFIG_INTEL_IOMMU
+	/* Disable ppgtt on SNB if VT-d is on. */
+	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped)
+		return false;
+#endif
+
+	return true;
+}
+
+void i915_gem_init_global_gtt(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long gtt_size, mappable_size;
+
+	gtt_size = dev_priv->gtt.total;
+	mappable_size = dev_priv->gtt.mappable_end;
+
+	if (intel_enable_ppgtt(dev) && HAS_ALIASING_PPGTT(dev)) {
+		int ret;
+		/* PPGTT pdes are stolen from global gtt ptes, so shrink the
+		 * aperture accordingly when using aliasing ppgtt. */
+		gtt_size -= I915_PPGTT_PD_ENTRIES*PAGE_SIZE;
+
+		i915_gem_setup_global_gtt(dev, 0, mappable_size, gtt_size);
+
+		ret = i915_gem_init_aliasing_ppgtt(dev);
+		if (!ret)
+			return;
+
+		DRM_ERROR("Aliased PPGTT setup failed %d\n", ret);
+		drm_mm_takedown(&dev_priv->mm.gtt_space);
+		gtt_size += I915_PPGTT_PD_ENTRIES*PAGE_SIZE;
+	}
+	i915_gem_setup_global_gtt(dev, 0, mappable_size, gtt_size);
 }
 
 static int setup_scratch_page(struct drm_device *dev)
@@ -567,8 +684,8 @@ static int setup_scratch_page(struct drm_device *dev)
 #else
 	dma_addr = page_to_phys(page);
 #endif
-	dev_priv->mm.gtt->scratch_page = page;
-	dev_priv->mm.gtt->scratch_page_dma = dma_addr;
+	dev_priv->gtt.scratch_page = page;
+	dev_priv->gtt.scratch_page_dma = dma_addr;
 
 	return 0;
 }
@@ -576,11 +693,11 @@ static int setup_scratch_page(struct drm_device *dev)
 static void teardown_scratch_page(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	set_pages_wb(dev_priv->mm.gtt->scratch_page, 1);
-	pci_unmap_page(dev->pdev, dev_priv->mm.gtt->scratch_page_dma,
+	set_pages_wb(dev_priv->gtt.scratch_page, 1);
+	pci_unmap_page(dev->pdev, dev_priv->gtt.scratch_page_dma,
 		       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
-	put_page(dev_priv->mm.gtt->scratch_page);
-	__free_page(dev_priv->mm.gtt->scratch_page);
+	put_page(dev_priv->gtt.scratch_page);
+	__free_page(dev_priv->gtt.scratch_page);
 }
 
 static inline unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
@@ -590,14 +707,14 @@ static inline unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
 	return snb_gmch_ctl << 20;
 }
 
-static inline unsigned int gen6_get_stolen_size(u16 snb_gmch_ctl)
+static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GMS_SHIFT;
 	snb_gmch_ctl &= SNB_GMCH_GMS_MASK;
 	return snb_gmch_ctl << 25; /* 32 MB units */
 }
 
-static inline unsigned int gen7_get_stolen_size(u16 snb_gmch_ctl)
+static inline size_t gen7_get_stolen_size(u16 snb_gmch_ctl)
 {
 	static const int stolen_decoder[] = {
 		0, 0, 0, 0, 0, 32, 48, 64, 128, 256, 96, 160, 224, 352};
@@ -606,103 +723,127 @@ static inline unsigned int gen7_get_stolen_size(u16 snb_gmch_ctl)
 	return stolen_decoder[snb_gmch_ctl] << 20;
 }
 
-int i915_gem_gtt_init(struct drm_device *dev)
+static int gen6_gmch_probe(struct drm_device *dev,
+			   size_t *gtt_total,
+			   size_t *stolen,
+			   phys_addr_t *mappable_base,
+			   unsigned long *mappable_end)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	phys_addr_t gtt_bus_addr;
+	unsigned int gtt_size;
 	u16 snb_gmch_ctl;
 	int ret;
 
-	/* On modern platforms we need not worry ourself with the legacy
-	 * hostbridge query stuff. Skip it entirely
-	 */
-	if (INTEL_INFO(dev)->gen < 6) {
-		ret = intel_gmch_probe(dev_priv->bridge_dev, dev->pdev, NULL);
-		if (!ret) {
-			DRM_ERROR("failed to set up gmch\n");
-			return -EIO;
-		}
+	*mappable_base = pci_resource_start(dev->pdev, 2);
+	*mappable_end = pci_resource_len(dev->pdev, 2);
 
-		dev_priv->mm.gtt = intel_gtt_get();
-		if (!dev_priv->mm.gtt) {
-			DRM_ERROR("Failed to initialize GTT\n");
-			intel_gmch_remove();
-			return -ENODEV;
-		}
-		return 0;
+	/* 64/512MB is the current min/max we actually know of, but this is just
+	 * a coarse sanity check.
+	 */
+	if ((*mappable_end < (64<<20) || (*mappable_end > (512<<20)))) {
+		DRM_ERROR("Unknown GMADR size (%lx)\n",
+			  dev_priv->gtt.mappable_end);
+		return -ENXIO;
 	}
 
-	dev_priv->mm.gtt = kzalloc(sizeof(*dev_priv->mm.gtt), GFP_KERNEL);
-	if (!dev_priv->mm.gtt)
-		return -ENOMEM;
-
 	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(40)))
 		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(40));
+	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
+	gtt_size = gen6_get_total_gtt_size(snb_gmch_ctl);
 
-#ifdef CONFIG_INTEL_IOMMU
-	dev_priv->mm.gtt->needs_dmar = 1;
-#endif
+	if (IS_GEN7(dev))
+		*stolen = gen7_get_stolen_size(snb_gmch_ctl);
+	else
+		*stolen = gen6_get_stolen_size(snb_gmch_ctl);
+
+	*gtt_total = (gtt_size / sizeof(gtt_pte_t)) << PAGE_SHIFT;
 
 	/* For GEN6+ the PTEs for the ggtt live at 2MB + BAR0 */
 	gtt_bus_addr = pci_resource_start(dev->pdev, 0) + (2<<20);
-	dev_priv->mm.gtt->gma_bus_addr = pci_resource_start(dev->pdev, 2);
-
-	/* i9xx_setup */
-	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
-	dev_priv->mm.gtt->gtt_total_entries =
-		gen6_get_total_gtt_size(snb_gmch_ctl) / sizeof(gtt_pte_t);
-	if (INTEL_INFO(dev)->gen < 7)
-		dev_priv->mm.gtt->stolen_size = gen6_get_stolen_size(snb_gmch_ctl);
-	else
-		dev_priv->mm.gtt->stolen_size = gen7_get_stolen_size(snb_gmch_ctl);
-
-	dev_priv->mm.gtt->gtt_mappable_entries = pci_resource_len(dev->pdev, 2) >> PAGE_SHIFT;
-	/* 64/512MB is the current min/max we actually know of, but this is just a
-	 * coarse sanity check.
-	 */
-	if ((dev_priv->mm.gtt->gtt_mappable_entries >> 8) < 64 ||
-	    dev_priv->mm.gtt->gtt_mappable_entries > dev_priv->mm.gtt->gtt_total_entries) {
-		DRM_ERROR("Unknown GMADR entries (%d)\n",
-			  dev_priv->mm.gtt->gtt_mappable_entries);
-		ret = -ENXIO;
-		goto err_out;
+	dev_priv->gtt.gsm = ioremap_wc(gtt_bus_addr, gtt_size);
+	if (!dev_priv->gtt.gsm) {
+		DRM_ERROR("Failed to map the gtt page table\n");
+		return -ENOMEM;
 	}
 
 	ret = setup_scratch_page(dev);
-	if (ret) {
+	if (ret)
 		DRM_ERROR("Scratch setup failed\n");
-		goto err_out;
-	}
 
-	dev_priv->mm.gtt->gtt = ioremap_wc(gtt_bus_addr,
-					   dev_priv->mm.gtt->gtt_total_entries * sizeof(gtt_pte_t));
-	if (!dev_priv->mm.gtt->gtt) {
-		DRM_ERROR("Failed to map the gtt page table\n");
-		teardown_scratch_page(dev);
-		ret = -ENOMEM;
-		goto err_out;
+	dev_priv->gtt.gtt_clear_range = gen6_ggtt_clear_range;
+	dev_priv->gtt.gtt_insert_entries = gen6_ggtt_insert_entries;
+
+	return ret;
+}
+
+static void gen6_gmch_remove(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	iounmap(dev_priv->gtt.gsm);
+	teardown_scratch_page(dev_priv->dev);
+}
+
+static int i915_gmch_probe(struct drm_device *dev,
+			   size_t *gtt_total,
+			   size_t *stolen,
+			   phys_addr_t *mappable_base,
+			   unsigned long *mappable_end)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	ret = intel_gmch_probe(dev_priv->bridge_dev, dev_priv->dev->pdev, NULL);
+	if (!ret) {
+		DRM_ERROR("failed to set up gmch\n");
+		return -EIO;
 	}
 
-	/* GMADR is the PCI aperture used by SW to access tiled GFX surfaces in a linear fashion. */
-	DRM_INFO("Memory usable by graphics device = %dM\n", dev_priv->mm.gtt->gtt_total_entries >> 8);
-	DRM_DEBUG_DRIVER("GMADR size = %dM\n", dev_priv->mm.gtt->gtt_mappable_entries >> 8);
-	DRM_DEBUG_DRIVER("GTT stolen size = %dM\n", dev_priv->mm.gtt->stolen_size >> 20);
+	intel_gtt_get(gtt_total, stolen, mappable_base, mappable_end);
+
+	dev_priv->gtt.do_idle_maps = needs_idle_maps(dev_priv->dev);
+	dev_priv->gtt.gtt_clear_range = i915_ggtt_clear_range;
+	dev_priv->gtt.gtt_insert_entries = i915_ggtt_insert_entries;
 
 	return 0;
+}
 
-err_out:
-	kfree(dev_priv->mm.gtt);
-	if (INTEL_INFO(dev)->gen < 6)
-		intel_gmch_remove();
-	return ret;
+static void i915_gmch_remove(struct drm_device *dev)
+{
+	intel_gmch_remove();
 }
 
-void i915_gem_gtt_fini(struct drm_device *dev)
+int i915_gem_gtt_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	iounmap(dev_priv->mm.gtt->gtt);
-	teardown_scratch_page(dev);
-	if (INTEL_INFO(dev)->gen < 6)
-		intel_gmch_remove();
-	kfree(dev_priv->mm.gtt);
+	struct i915_gtt *gtt = &dev_priv->gtt;
+	unsigned long gtt_size;
+	int ret;
+
+	if (INTEL_INFO(dev)->gen <= 5) {
+		dev_priv->gtt.gtt_probe = i915_gmch_probe;
+		dev_priv->gtt.gtt_remove = i915_gmch_remove;
+	} else {
+		dev_priv->gtt.gtt_probe = gen6_gmch_probe;
+		dev_priv->gtt.gtt_remove = gen6_gmch_remove;
+	}
+
+	ret = dev_priv->gtt.gtt_probe(dev, &dev_priv->gtt.total,
+				     &dev_priv->gtt.stolen_size,
+				     &gtt->mappable_base,
+				     &gtt->mappable_end);
+	if (ret)
+		return ret;
+
+	gtt_size = (dev_priv->gtt.total >> PAGE_SHIFT) * sizeof(gtt_pte_t);
+
+	/* GMADR is the PCI mmio aperture into the global GTT. */
+	DRM_INFO("Memory usable by graphics device = %zdM\n",
+		 dev_priv->gtt.total >> 20);
+	DRM_DEBUG_DRIVER("GMADR size = %ldM\n",
+			 dev_priv->gtt.mappable_end >> 20);
+	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n",
+			 dev_priv->gtt.stolen_size >> 20);
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 8e91083b126f..69d97cbac13c 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -42,85 +42,73 @@
  * for is a boon.
  */
 
-#define PTE_ADDRESS_MASK		0xfffff000
-#define PTE_ADDRESS_MASK_HIGH		0x000000f0 /* i915+ */
-#define PTE_MAPPING_TYPE_UNCACHED	(0 << 1)
-#define PTE_MAPPING_TYPE_DCACHE		(1 << 1) /* i830 only */
-#define PTE_MAPPING_TYPE_CACHED		(3 << 1)
-#define PTE_MAPPING_TYPE_MASK		(3 << 1)
-#define PTE_VALID			(1 << 0)
-
-/**
- * i915_stolen_to_phys - take an offset into stolen memory and turn it into
- *                       a physical one
- * @dev: drm device
- * @offset: address to translate
- *
- * Some chip functions require allocations from stolen space and need the
- * physical address of the memory in question.
- */
-static unsigned long i915_stolen_to_phys(struct drm_device *dev, u32 offset)
+static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct pci_dev *pdev = dev_priv->bridge_dev;
 	u32 base;
 
-#if 0
 	/* On the machines I have tested the Graphics Base of Stolen Memory
-	 * is unreliable, so compute the base by subtracting the stolen memory
-	 * from the Top of Low Usable DRAM which is where the BIOS places
-	 * the graphics stolen memory.
+	 * is unreliable, so on those compute the base by subtracting the
+	 * stolen memory from the Top of Low Usable DRAM which is where the
+	 * BIOS places the graphics stolen memory.
+	 *
+	 * On gen2, the layout is slightly different with the Graphics Segment
+	 * immediately following Top of Memory (or Top of Usable DRAM). Note
+	 * it appears that TOUD is only reported by 865g, so we just use the
+	 * top of memory as determined by the e820 probe.
+	 *
+	 * XXX gen2 requires an unavailable symbol and 945gm fails with
+	 * its value of TOLUD.
 	 */
-	if (INTEL_INFO(dev)->gen > 3 || IS_G33(dev)) {
-		/* top 32bits are reserved = 0 */
+	base = 0;
+	if (INTEL_INFO(dev)->gen >= 6) {
+		/* Read Base Data of Stolen Memory Register (BDSM) directly.
+		 * Note that there is also a MCHBAR miror at 0x1080c0 or
+		 * we could use device 2:0x5c instead.
+		*/
+		pci_read_config_dword(pdev, 0xB0, &base);
+		base &= ~4095; /* lower bits used for locking register */
+	} else if (INTEL_INFO(dev)->gen > 3 || IS_G33(dev)) {
+		/* Read Graphics Base of Stolen Memory directly */
 		pci_read_config_dword(pdev, 0xA4, &base);
-	} else {
-		/* XXX presume 8xx is the same as i915 */
-		pci_bus_read_config_dword(pdev->bus, 2, 0x5C, &base);
-	}
-#else
-	if (INTEL_INFO(dev)->gen > 3 || IS_G33(dev)) {
-		u16 val;
-		pci_read_config_word(pdev, 0xb0, &val);
-		base = val >> 4 << 20;
-	} else {
+#if 0
+	} else if (IS_GEN3(dev)) {
 		u8 val;
+		/* Stolen is immediately below Top of Low Usable DRAM */
 		pci_read_config_byte(pdev, 0x9c, &val);
 		base = val >> 3 << 27;
-	}
-	base -= dev_priv->mm.gtt->stolen_size;
+		base -= dev_priv->mm.gtt->stolen_size;
+	} else {
+		/* Stolen is immediately above Top of Memory */
+		base = max_low_pfn_mapped << PAGE_SHIFT;
 #endif
+	}
 
-	return base + offset;
+	return base;
 }
 
-static void i915_warn_stolen(struct drm_device *dev)
-{
-	DRM_INFO("not enough stolen space for compressed buffer, disabling\n");
-	DRM_INFO("hint: you may be able to increase stolen memory size in the BIOS to avoid this\n");
-}
-
-static void i915_setup_compression(struct drm_device *dev, int size)
+static int i915_setup_compression(struct drm_device *dev, int size)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_mm_node *compressed_fb, *uninitialized_var(compressed_llb);
-	unsigned long cfb_base;
-	unsigned long ll_base = 0;
-
-	/* Just in case the BIOS is doing something questionable. */
-	intel_disable_fbc(dev);
 
-	compressed_fb = drm_mm_search_free(&dev_priv->mm.stolen, size, 4096, 0);
+	/* Try to over-allocate to reduce reallocations and fragmentation */
+	compressed_fb = drm_mm_search_free(&dev_priv->mm.stolen,
+					   size <<= 1, 4096, 0);
+	if (!compressed_fb)
+		compressed_fb = drm_mm_search_free(&dev_priv->mm.stolen,
+						   size >>= 1, 4096, 0);
 	if (compressed_fb)
 		compressed_fb = drm_mm_get_block(compressed_fb, size, 4096);
 	if (!compressed_fb)
 		goto err;
 
-	cfb_base = i915_stolen_to_phys(dev, compressed_fb->start);
-	if (!cfb_base)
-		goto err_fb;
-
-	if (!(IS_GM45(dev) || HAS_PCH_SPLIT(dev))) {
+	if (HAS_PCH_SPLIT(dev))
+		I915_WRITE(ILK_DPFC_CB_BASE, compressed_fb->start);
+	else if (IS_GM45(dev)) {
+		I915_WRITE(DPFC_CB_BASE, compressed_fb->start);
+	} else {
 		compressed_llb = drm_mm_search_free(&dev_priv->mm.stolen,
 						    4096, 4096, 0);
 		if (compressed_llb)
@@ -129,73 +117,206 @@ static void i915_setup_compression(struct drm_device *dev, int size)
 		if (!compressed_llb)
 			goto err_fb;
 
-		ll_base = i915_stolen_to_phys(dev, compressed_llb->start);
-		if (!ll_base)
-			goto err_llb;
+		dev_priv->compressed_llb = compressed_llb;
+
+		I915_WRITE(FBC_CFB_BASE,
+			   dev_priv->mm.stolen_base + compressed_fb->start);
+		I915_WRITE(FBC_LL_BASE,
+			   dev_priv->mm.stolen_base + compressed_llb->start);
 	}
 
+	dev_priv->compressed_fb = compressed_fb;
 	dev_priv->cfb_size = size;
 
-	dev_priv->compressed_fb = compressed_fb;
-	if (HAS_PCH_SPLIT(dev))
-		I915_WRITE(ILK_DPFC_CB_BASE, compressed_fb->start);
-	else if (IS_GM45(dev)) {
-		I915_WRITE(DPFC_CB_BASE, compressed_fb->start);
-	} else {
-		I915_WRITE(FBC_CFB_BASE, cfb_base);
-		I915_WRITE(FBC_LL_BASE, ll_base);
-		dev_priv->compressed_llb = compressed_llb;
-	}
+	DRM_DEBUG_KMS("reserved %d bytes of contiguous stolen space for FBC\n",
+		      size);
 
-	DRM_DEBUG_KMS("FBC base 0x%08lx, ll base 0x%08lx, size %dM\n",
-		      cfb_base, ll_base, size >> 20);
-	return;
+	return 0;
 
-err_llb:
-	drm_mm_put_block(compressed_llb);
 err_fb:
 	drm_mm_put_block(compressed_fb);
 err:
-	dev_priv->no_fbc_reason = FBC_STOLEN_TOO_SMALL;
-	i915_warn_stolen(dev);
+	return -ENOSPC;
+}
+
+int i915_gem_stolen_setup_compression(struct drm_device *dev, int size)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->mm.stolen_base == 0)
+		return -ENODEV;
+
+	if (size < dev_priv->cfb_size)
+		return 0;
+
+	/* Release any current block */
+	i915_gem_stolen_cleanup_compression(dev);
+
+	return i915_setup_compression(dev, size);
 }
 
-static void i915_cleanup_compression(struct drm_device *dev)
+void i915_gem_stolen_cleanup_compression(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	drm_mm_put_block(dev_priv->compressed_fb);
+	if (dev_priv->cfb_size == 0)
+		return;
+
+	if (dev_priv->compressed_fb)
+		drm_mm_put_block(dev_priv->compressed_fb);
+
 	if (dev_priv->compressed_llb)
 		drm_mm_put_block(dev_priv->compressed_llb);
+
+	dev_priv->cfb_size = 0;
 }
 
 void i915_gem_cleanup_stolen(struct drm_device *dev)
 {
-	if (I915_HAS_FBC(dev) && i915_powersave)
-		i915_cleanup_compression(dev);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	i915_gem_stolen_cleanup_compression(dev);
+	drm_mm_takedown(&dev_priv->mm.stolen);
 }
 
 int i915_gem_init_stolen(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long prealloc_size = dev_priv->mm.gtt->stolen_size;
+
+	dev_priv->mm.stolen_base = i915_stolen_to_physical(dev);
+	if (dev_priv->mm.stolen_base == 0)
+		return 0;
+
+	DRM_DEBUG_KMS("found %zd bytes of stolen memory at %08lx\n",
+		      dev_priv->gtt.stolen_size, dev_priv->mm.stolen_base);
 
 	/* Basic memrange allocator for stolen space */
-	drm_mm_init(&dev_priv->mm.stolen, 0, prealloc_size);
+	drm_mm_init(&dev_priv->mm.stolen, 0, dev_priv->gtt.stolen_size);
+
+	return 0;
+}
 
-	/* Try to set up FBC with a reasonable compressed buffer size */
-	if (I915_HAS_FBC(dev) && i915_powersave) {
-		int cfb_size;
+static struct sg_table *
+i915_pages_create_for_stolen(struct drm_device *dev,
+			     u32 offset, u32 size)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct sg_table *st;
+	struct scatterlist *sg;
+
+	DRM_DEBUG_DRIVER("offset=0x%x, size=%d\n", offset, size);
+	BUG_ON(offset > dev_priv->gtt.stolen_size - size);
 
-		/* Leave 1M for line length buffer & misc. */
+	/* We hide that we have no struct page backing our stolen object
+	 * by wrapping the contiguous physical allocation with a fake
+	 * dma mapping in a single scatterlist.
+	 */
+
+	st = kmalloc(sizeof(*st), GFP_KERNEL);
+	if (st == NULL)
+		return NULL;
 
-		/* Try to get a 32M buffer... */
-		if (prealloc_size > (36*1024*1024))
-			cfb_size = 32*1024*1024;
-		else /* fall back to 7/8 of the stolen space */
-			cfb_size = prealloc_size * 7 / 8;
-		i915_setup_compression(dev, cfb_size);
+	if (sg_alloc_table(st, 1, GFP_KERNEL)) {
+		kfree(st);
+		return NULL;
 	}
 
-	return 0;
+	sg = st->sgl;
+	sg->offset = offset;
+	sg->length = size;
+
+	sg_dma_address(sg) = (dma_addr_t)dev_priv->mm.stolen_base + offset;
+	sg_dma_len(sg) = size;
+
+	return st;
+}
+
+static int i915_gem_object_get_pages_stolen(struct drm_i915_gem_object *obj)
+{
+	BUG();
+	return -EINVAL;
+}
+
+static void i915_gem_object_put_pages_stolen(struct drm_i915_gem_object *obj)
+{
+	/* Should only be called during free */
+	sg_free_table(obj->pages);
+	kfree(obj->pages);
+}
+
+static const struct drm_i915_gem_object_ops i915_gem_object_stolen_ops = {
+	.get_pages = i915_gem_object_get_pages_stolen,
+	.put_pages = i915_gem_object_put_pages_stolen,
+};
+
+static struct drm_i915_gem_object *
+_i915_gem_object_create_stolen(struct drm_device *dev,
+			       struct drm_mm_node *stolen)
+{
+	struct drm_i915_gem_object *obj;
+
+	obj = i915_gem_object_alloc(dev);
+	if (obj == NULL)
+		return NULL;
+
+	if (drm_gem_private_object_init(dev, &obj->base, stolen->size))
+		goto cleanup;
+
+	i915_gem_object_init(obj, &i915_gem_object_stolen_ops);
+
+	obj->pages = i915_pages_create_for_stolen(dev,
+						  stolen->start, stolen->size);
+	if (obj->pages == NULL)
+		goto cleanup;
+
+	obj->has_dma_mapping = true;
+	obj->pages_pin_count = 1;
+	obj->stolen = stolen;
+
+	obj->base.write_domain = I915_GEM_DOMAIN_GTT;
+	obj->base.read_domains = I915_GEM_DOMAIN_GTT;
+	obj->cache_level = I915_CACHE_NONE;
+
+	return obj;
+
+cleanup:
+	i915_gem_object_free(obj);
+	return NULL;
+}
+
+struct drm_i915_gem_object *
+i915_gem_object_create_stolen(struct drm_device *dev, u32 size)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_object *obj;
+	struct drm_mm_node *stolen;
+
+	if (dev_priv->mm.stolen_base == 0)
+		return NULL;
+
+	DRM_DEBUG_KMS("creating stolen object: size=%x\n", size);
+	if (size == 0)
+		return NULL;
+
+	stolen = drm_mm_search_free(&dev_priv->mm.stolen, size, 4096, 0);
+	if (stolen)
+		stolen = drm_mm_get_block(stolen, size, 4096);
+	if (stolen == NULL)
+		return NULL;
+
+	obj = _i915_gem_object_create_stolen(dev, stolen);
+	if (obj)
+		return obj;
+
+	drm_mm_put_block(stolen);
+	return NULL;
+}
+
+void
+i915_gem_object_release_stolen(struct drm_i915_gem_object *obj)
+{
+	if (obj->stolen) {
+		drm_mm_put_block(obj->stolen);
+		obj->stolen = NULL;
+	}
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index cedbfd7b3dfa..abcba2f5a788 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -272,18 +272,7 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)
 			return false;
 	}
 
-	/*
-	 * Previous chips need to be aligned to the size of the smallest
-	 * fence register that can contain the object.
-	 */
-	if (INTEL_INFO(obj->base.dev)->gen == 3)
-		size = 1024*1024;
-	else
-		size = 512*1024;
-
-	while (size < obj->base.size)
-		size <<= 1;
-
+	size = i915_gem_get_gtt_size(obj->base.dev, obj->base.size, tiling_mode);
 	if (obj->gtt_space->size != size)
 		return false;
 
@@ -368,15 +357,15 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 
 		obj->map_and_fenceable =
 			obj->gtt_space == NULL ||
-			(obj->gtt_offset + obj->base.size <= dev_priv->mm.gtt_mappable_end &&
+			(obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end &&
 			 i915_gem_object_fence_ok(obj, args->tiling_mode));
 
 		/* Rebind if we need a change of alignment */
 		if (!obj->map_and_fenceable) {
 			u32 unfenced_alignment =
-				i915_gem_get_unfenced_gtt_alignment(dev,
-								    obj->base.size,
-								    args->tiling_mode);
+				i915_gem_get_gtt_alignment(dev, obj->base.size,
+							    args->tiling_mode,
+							    false);
 			if (obj->gtt_offset & (unfenced_alignment - 1))
 				ret = i915_gem_object_unbind(obj);
 		}
@@ -396,6 +385,18 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	/* we have to maintain this existing ABI... */
 	args->stride = obj->stride;
 	args->tiling_mode = obj->tiling_mode;
+
+	/* Try to preallocate memory required to save swizzling on put-pages */
+	if (i915_gem_object_needs_bit17_swizzle(obj)) {
+		if (obj->bit_17 == NULL) {
+			obj->bit_17 = kmalloc(BITS_TO_LONGS(obj->base.size >> PAGE_SHIFT) *
+					      sizeof(long), GFP_KERNEL);
+		}
+	} else {
+		kfree(obj->bit_17);
+		obj->bit_17 = NULL;
+	}
+
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index fe843389c7b4..2cd97d1cc920 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -287,6 +287,10 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	struct intel_encoder *encoder;
 
+	/* HPD irq before everything is fully set up. */
+	if (!dev_priv->enable_hotplug_processing)
+		return;
+
 	mutex_lock(&mode_config->mutex);
 	DRM_DEBUG_KMS("running encoder hotplug functions\n");
 
@@ -300,9 +304,6 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	drm_helper_hpd_irq_event(dev);
 }
 
-/* defined intel_pm.c */
-extern spinlock_t mchdev_lock;
-
 static void ironlake_handle_rps_change(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
@@ -355,8 +356,8 @@ static void notify_ring(struct drm_device *dev,
 
 	wake_up_all(&ring->irq_queue);
 	if (i915_enable_hangcheck) {
-		dev_priv->hangcheck_count = 0;
-		mod_timer(&dev_priv->hangcheck_timer,
+		dev_priv->gpu_error.hangcheck_count = 0;
+		mod_timer(&dev_priv->gpu_error.hangcheck_timer,
 			  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
 	}
 }
@@ -524,6 +525,20 @@ static void gen6_queue_rps_work(struct drm_i915_private *dev_priv,
 	queue_work(dev_priv->wq, &dev_priv->rps.work);
 }
 
+static void gmbus_irq_handler(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = (drm_i915_private_t *) dev->dev_private;
+
+	wake_up_all(&dev_priv->gmbus_wait_queue);
+}
+
+static void dp_aux_irq_handler(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = (drm_i915_private_t *) dev->dev_private;
+
+	wake_up_all(&dev_priv->gmbus_wait_queue);
+}
+
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
@@ -533,7 +548,6 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 	unsigned long irqflags;
 	int pipe;
 	u32 pipe_stats[I915_MAX_PIPES];
-	bool blc_event;
 
 	atomic_inc(&dev_priv->irq_received);
 
@@ -590,8 +604,8 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 			I915_READ(PORT_HOTPLUG_STAT);
 		}
 
-		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
-			blc_event = true;
+		if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
+			gmbus_irq_handler(dev);
 
 		if (pm_iir & GEN6_PM_DEFERRED_EVENTS)
 			gen6_queue_rps_work(dev_priv, pm_iir);
@@ -618,8 +632,11 @@ static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 				 (pch_iir & SDE_AUDIO_POWER_MASK) >>
 				 SDE_AUDIO_POWER_SHIFT);
 
+	if (pch_iir & SDE_AUX_MASK)
+		dp_aux_irq_handler(dev);
+
 	if (pch_iir & SDE_GMBUS)
-		DRM_DEBUG_DRIVER("PCH GMBUS interrupt\n");
+		gmbus_irq_handler(dev);
 
 	if (pch_iir & SDE_AUDIO_HDCP_MASK)
 		DRM_DEBUG_DRIVER("PCH HDCP audio interrupt\n");
@@ -662,10 +679,10 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 				 SDE_AUDIO_POWER_SHIFT_CPT);
 
 	if (pch_iir & SDE_AUX_MASK_CPT)
-		DRM_DEBUG_DRIVER("AUX channel interrupt\n");
+		dp_aux_irq_handler(dev);
 
 	if (pch_iir & SDE_GMBUS_CPT)
-		DRM_DEBUG_DRIVER("PCH GMBUS interrupt\n");
+		gmbus_irq_handler(dev);
 
 	if (pch_iir & SDE_AUDIO_CP_REQ_CPT)
 		DRM_DEBUG_DRIVER("Audio CP request interrupt\n");
@@ -703,6 +720,9 @@ static irqreturn_t ivybridge_irq_handler(int irq, void *arg)
 
 	de_iir = I915_READ(DEIIR);
 	if (de_iir) {
+		if (de_iir & DE_AUX_CHANNEL_A_IVB)
+			dp_aux_irq_handler(dev);
+
 		if (de_iir & DE_GSE_IVB)
 			intel_opregion_gse_intr(dev);
 
@@ -758,7 +778,7 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	struct drm_device *dev = (struct drm_device *) arg;
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	int ret = IRQ_NONE;
-	u32 de_iir, gt_iir, de_ier, pch_iir, pm_iir;
+	u32 de_iir, gt_iir, de_ier, pm_iir;
 
 	atomic_inc(&dev_priv->irq_received);
 
@@ -769,11 +789,9 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 
 	de_iir = I915_READ(DEIIR);
 	gt_iir = I915_READ(GTIIR);
-	pch_iir = I915_READ(SDEIIR);
 	pm_iir = I915_READ(GEN6_PMIIR);
 
-	if (de_iir == 0 && gt_iir == 0 && pch_iir == 0 &&
-	    (!IS_GEN6(dev) || pm_iir == 0))
+	if (de_iir == 0 && gt_iir == 0 && (!IS_GEN6(dev) || pm_iir == 0))
 		goto done;
 
 	ret = IRQ_HANDLED;
@@ -783,6 +801,9 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	else
 		snb_gt_irq_handler(dev, dev_priv, gt_iir);
 
+	if (de_iir & DE_AUX_CHANNEL_A)
+		dp_aux_irq_handler(dev);
+
 	if (de_iir & DE_GSE)
 		intel_opregion_gse_intr(dev);
 
@@ -804,10 +825,15 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 
 	/* check event from PCH */
 	if (de_iir & DE_PCH_EVENT) {
+		u32 pch_iir = I915_READ(SDEIIR);
+
 		if (HAS_PCH_CPT(dev))
 			cpt_irq_handler(dev, pch_iir);
 		else
 			ibx_irq_handler(dev, pch_iir);
+
+		/* should clear PCH hotplug event before clear CPU irq */
+		I915_WRITE(SDEIIR, pch_iir);
 	}
 
 	if (IS_GEN5(dev) &&  de_iir & DE_PCU_EVENT)
@@ -816,8 +842,6 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	if (IS_GEN6(dev) && pm_iir & GEN6_PM_DEFERRED_EVENTS)
 		gen6_queue_rps_work(dev_priv, pm_iir);
 
-	/* should clear PCH hotplug event before clear CPU irq */
-	I915_WRITE(SDEIIR, pch_iir);
 	I915_WRITE(GTIIR, gt_iir);
 	I915_WRITE(DEIIR, de_iir);
 	I915_WRITE(GEN6_PMIIR, pm_iir);
@@ -838,23 +862,60 @@ done:
  */
 static void i915_error_work_func(struct work_struct *work)
 {
-	drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
-						    error_work);
+	struct i915_gpu_error *error = container_of(work, struct i915_gpu_error,
+						    work);
+	drm_i915_private_t *dev_priv = container_of(error, drm_i915_private_t,
+						    gpu_error);
 	struct drm_device *dev = dev_priv->dev;
+	struct intel_ring_buffer *ring;
 	char *error_event[] = { "ERROR=1", NULL };
 	char *reset_event[] = { "RESET=1", NULL };
 	char *reset_done_event[] = { "ERROR=0", NULL };
+	int i, ret;
 
 	kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, error_event);
 
-	if (atomic_read(&dev_priv->mm.wedged)) {
+	/*
+	 * Note that there's only one work item which does gpu resets, so we
+	 * need not worry about concurrent gpu resets potentially incrementing
+	 * error->reset_counter twice. We only need to take care of another
+	 * racing irq/hangcheck declaring the gpu dead for a second time. A
+	 * quick check for that is good enough: schedule_work ensures the
+	 * correct ordering between hang detection and this work item, and since
+	 * the reset in-progress bit is only ever set by code outside of this
+	 * work we don't need to worry about any other races.
+	 */
+	if (i915_reset_in_progress(error) && !i915_terminally_wedged(error)) {
 		DRM_DEBUG_DRIVER("resetting chip\n");
-		kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, reset_event);
-		if (!i915_reset(dev)) {
-			atomic_set(&dev_priv->mm.wedged, 0);
-			kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, reset_done_event);
+		kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE,
+				   reset_event);
+
+		ret = i915_reset(dev);
+
+		if (ret == 0) {
+			/*
+			 * After all the gem state is reset, increment the reset
+			 * counter and wake up everyone waiting for the reset to
+			 * complete.
+			 *
+			 * Since unlock operations are a one-sided barrier only,
+			 * we need to insert a barrier here to order any seqno
+			 * updates before
+			 * the counter increment.
+			 */
+			smp_mb__before_atomic_inc();
+			atomic_inc(&dev_priv->gpu_error.reset_counter);
+
+			kobject_uevent_env(&dev->primary->kdev.kobj,
+					   KOBJ_CHANGE, reset_done_event);
+		} else {
+			atomic_set(&error->reset_counter, I915_WEDGED);
 		}
-		complete_all(&dev_priv->error_completion);
+
+		for_each_ring(ring, dev_priv, i)
+			wake_up_all(&ring->irq_queue);
+
+		wake_up_all(&dev_priv->gpu_error.reset_queue);
 	}
 }
 
@@ -915,7 +976,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			goto unwind;
 
 		local_irq_save(flags);
-		if (reloc_offset < dev_priv->mm.gtt_mappable_end &&
+		if (reloc_offset < dev_priv->gtt.mappable_end &&
 		    src->has_global_gtt_mapping) {
 			void __iomem *s;
 
@@ -924,10 +985,18 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			 * captures what the GPU read.
 			 */
 
-			s = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping,
+			s = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
 						     reloc_offset);
 			memcpy_fromio(d, s, PAGE_SIZE);
 			io_mapping_unmap_atomic(s);
+		} else if (src->stolen) {
+			unsigned long offset;
+
+			offset = dev_priv->mm.stolen_base;
+			offset += src->stolen->start;
+			offset += i << PAGE_SHIFT;
+
+			memcpy_fromio(d, (void __iomem *) offset, PAGE_SIZE);
 		} else {
 			struct page *page;
 			void *s;
@@ -1074,6 +1143,8 @@ static void i915_gem_record_fences(struct drm_device *dev,
 			error->fence[i] = I915_READ(FENCE_REG_830_0 + (i * 4));
 		break;
 
+	default:
+		BUG();
 	}
 }
 
@@ -1222,9 +1293,9 @@ static void i915_capture_error_state(struct drm_device *dev)
 	unsigned long flags;
 	int i, pipe;
 
-	spin_lock_irqsave(&dev_priv->error_lock, flags);
-	error = dev_priv->first_error;
-	spin_unlock_irqrestore(&dev_priv->error_lock, flags);
+	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
+	error = dev_priv->gpu_error.first_error;
+	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
 	if (error)
 		return;
 
@@ -1235,7 +1306,8 @@ static void i915_capture_error_state(struct drm_device *dev)
 		return;
 	}
 
-	DRM_INFO("capturing error event; look for more information in /debug/dri/%d/i915_error_state\n",
+	DRM_INFO("capturing error event; look for more information in"
+		 "/sys/kernel/debug/dri/%d/i915_error_state\n",
 		 dev->primary->index);
 
 	kref_init(&error->ref);
@@ -1318,12 +1390,12 @@ static void i915_capture_error_state(struct drm_device *dev)
 	error->overlay = intel_overlay_capture_error_state(dev);
 	error->display = intel_display_capture_error_state(dev);
 
-	spin_lock_irqsave(&dev_priv->error_lock, flags);
-	if (dev_priv->first_error == NULL) {
-		dev_priv->first_error = error;
+	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
+	if (dev_priv->gpu_error.first_error == NULL) {
+		dev_priv->gpu_error.first_error = error;
 		error = NULL;
 	}
-	spin_unlock_irqrestore(&dev_priv->error_lock, flags);
+	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
 
 	if (error)
 		i915_error_state_free(&error->ref);
@@ -1335,10 +1407,10 @@ void i915_destroy_error_state(struct drm_device *dev)
 	struct drm_i915_error_state *error;
 	unsigned long flags;
 
-	spin_lock_irqsave(&dev_priv->error_lock, flags);
-	error = dev_priv->first_error;
-	dev_priv->first_error = NULL;
-	spin_unlock_irqrestore(&dev_priv->error_lock, flags);
+	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
+	error = dev_priv->gpu_error.first_error;
+	dev_priv->gpu_error.first_error = NULL;
+	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
 
 	if (error)
 		kref_put(&error->ref, i915_error_state_free);
@@ -1459,17 +1531,18 @@ void i915_handle_error(struct drm_device *dev, bool wedged)
 	i915_report_and_clear_eir(dev);
 
 	if (wedged) {
-		INIT_COMPLETION(dev_priv->error_completion);
-		atomic_set(&dev_priv->mm.wedged, 1);
+		atomic_set_mask(I915_RESET_IN_PROGRESS_FLAG,
+				&dev_priv->gpu_error.reset_counter);
 
 		/*
-		 * Wakeup waiting processes so they don't hang
+		 * Wakeup waiting processes so that the reset work item
+		 * doesn't deadlock trying to grab various locks.
 		 */
 		for_each_ring(ring, dev_priv, i)
 			wake_up_all(&ring->irq_queue);
 	}
 
-	queue_work(dev_priv->wq, &dev_priv->error_work);
+	queue_work(dev_priv->wq, &dev_priv->gpu_error.work);
 }
 
 static void i915_pageflip_stall_check(struct drm_device *dev, int pipe)
@@ -1700,7 +1773,7 @@ static bool i915_hangcheck_hung(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
 
-	if (dev_priv->hangcheck_count++ > 1) {
+	if (dev_priv->gpu_error.hangcheck_count++ > 1) {
 		bool hung = true;
 
 		DRM_ERROR("Hangcheck timer elapsed... GPU hung\n");
@@ -1759,25 +1832,29 @@ void i915_hangcheck_elapsed(unsigned long data)
 			goto repeat;
 		}
 
-		dev_priv->hangcheck_count = 0;
+		dev_priv->gpu_error.hangcheck_count = 0;
 		return;
 	}
 
 	i915_get_extra_instdone(dev, instdone);
-	if (memcmp(dev_priv->last_acthd, acthd, sizeof(acthd)) == 0 &&
-	    memcmp(dev_priv->prev_instdone, instdone, sizeof(instdone)) == 0) {
+	if (memcmp(dev_priv->gpu_error.last_acthd, acthd,
+		   sizeof(acthd)) == 0 &&
+	    memcmp(dev_priv->gpu_error.prev_instdone, instdone,
+		   sizeof(instdone)) == 0) {
 		if (i915_hangcheck_hung(dev))
 			return;
 	} else {
-		dev_priv->hangcheck_count = 0;
+		dev_priv->gpu_error.hangcheck_count = 0;
 
-		memcpy(dev_priv->last_acthd, acthd, sizeof(acthd));
-		memcpy(dev_priv->prev_instdone, instdone, sizeof(instdone));
+		memcpy(dev_priv->gpu_error.last_acthd, acthd,
+		       sizeof(acthd));
+		memcpy(dev_priv->gpu_error.prev_instdone, instdone,
+		       sizeof(instdone));
 	}
 
 repeat:
 	/* Reset timer case chip hangs without another request being added */
-	mod_timer(&dev_priv->hangcheck_timer,
+	mod_timer(&dev_priv->gpu_error.hangcheck_timer,
 		  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
 }
 
@@ -1847,7 +1924,7 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
  * This register is the same on all known PCH chips.
  */
 
-static void ironlake_enable_pch_hotplug(struct drm_device *dev)
+static void ibx_enable_hotplug(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32	hotplug;
@@ -1860,14 +1937,36 @@ static void ironlake_enable_pch_hotplug(struct drm_device *dev)
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 }
 
+static void ibx_irq_postinstall(struct drm_device *dev)
+{
+	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	u32 mask;
+
+	if (HAS_PCH_IBX(dev))
+		mask = SDE_HOTPLUG_MASK |
+		       SDE_GMBUS |
+		       SDE_AUX_MASK;
+	else
+		mask = SDE_HOTPLUG_MASK_CPT |
+		       SDE_GMBUS_CPT |
+		       SDE_AUX_MASK_CPT;
+
+	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
+	I915_WRITE(SDEIMR, ~mask);
+	I915_WRITE(SDEIER, mask);
+	POSTING_READ(SDEIER);
+
+	ibx_enable_hotplug(dev);
+}
+
 static int ironlake_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	/* enable kind of interrupts always enabled */
 	u32 display_mask = DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
-			   DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE;
+			   DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
+			   DE_AUX_CHANNEL_A;
 	u32 render_irqs;
-	u32 hotplug_mask;
 
 	dev_priv->irq_mask = ~display_mask;
 
@@ -1895,27 +1994,7 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(GTIER, render_irqs);
 	POSTING_READ(GTIER);
 
-	if (HAS_PCH_CPT(dev)) {
-		hotplug_mask = (SDE_CRT_HOTPLUG_CPT |
-				SDE_PORTB_HOTPLUG_CPT |
-				SDE_PORTC_HOTPLUG_CPT |
-				SDE_PORTD_HOTPLUG_CPT);
-	} else {
-		hotplug_mask = (SDE_CRT_HOTPLUG |
-				SDE_PORTB_HOTPLUG |
-				SDE_PORTC_HOTPLUG |
-				SDE_PORTD_HOTPLUG |
-				SDE_AUX_MASK);
-	}
-
-	dev_priv->pch_irq_mask = ~hotplug_mask;
-
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
-	I915_WRITE(SDEIMR, dev_priv->pch_irq_mask);
-	I915_WRITE(SDEIER, hotplug_mask);
-	POSTING_READ(SDEIER);
-
-	ironlake_enable_pch_hotplug(dev);
+	ibx_irq_postinstall(dev);
 
 	if (IS_IRONLAKE_M(dev)) {
 		/* Clear & enable PCU event interrupts */
@@ -1935,9 +2014,9 @@ static int ivybridge_irq_postinstall(struct drm_device *dev)
 		DE_MASTER_IRQ_CONTROL | DE_GSE_IVB | DE_PCH_EVENT_IVB |
 		DE_PLANEC_FLIP_DONE_IVB |
 		DE_PLANEB_FLIP_DONE_IVB |
-		DE_PLANEA_FLIP_DONE_IVB;
+		DE_PLANEA_FLIP_DONE_IVB |
+		DE_AUX_CHANNEL_A_IVB;
 	u32 render_irqs;
-	u32 hotplug_mask;
 
 	dev_priv->irq_mask = ~display_mask;
 
@@ -1961,18 +2040,7 @@ static int ivybridge_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(GTIER, render_irqs);
 	POSTING_READ(GTIER);
 
-	hotplug_mask = (SDE_CRT_HOTPLUG_CPT |
-			SDE_PORTB_HOTPLUG_CPT |
-			SDE_PORTC_HOTPLUG_CPT |
-			SDE_PORTD_HOTPLUG_CPT);
-	dev_priv->pch_irq_mask = ~hotplug_mask;
-
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
-	I915_WRITE(SDEIMR, dev_priv->pch_irq_mask);
-	I915_WRITE(SDEIER, hotplug_mask);
-	POSTING_READ(SDEIER);
-
-	ironlake_enable_pch_hotplug(dev);
+	ibx_irq_postinstall(dev);
 
 	return 0;
 }
@@ -1981,7 +2049,6 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32 enable_mask;
-	u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
 	u32 pipestat_enable = PLANE_FLIP_DONE_INT_EN_VLV;
 	u32 render_irqs;
 	u16 msid;
@@ -2010,6 +2077,9 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 	msid |= (1<<14);
 	pci_write_config_word(dev_priv->dev->pdev, 0x98, msid);
 
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	POSTING_READ(PORT_HOTPLUG_EN);
+
 	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
 	I915_WRITE(VLV_IER, enable_mask);
 	I915_WRITE(VLV_IIR, 0xffffffff);
@@ -2018,6 +2088,7 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 
 	i915_enable_pipestat(dev_priv, 0, pipestat_enable);
+	i915_enable_pipestat(dev_priv, 0, PIPE_GMBUS_EVENT_ENABLE);
 	i915_enable_pipestat(dev_priv, 1, pipestat_enable);
 
 	I915_WRITE(VLV_IIR, 0xffffffff);
@@ -2038,13 +2109,22 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 #endif
 
 	I915_WRITE(VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE);
+
+	return 0;
+}
+
+static void valleyview_hpd_irq_setup(struct drm_device *dev)
+{
+	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
+
 	/* Note HDMI and DP share bits */
-	if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMIB_HOTPLUG_INT_EN;
-	if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMIC_HOTPLUG_INT_EN;
-	if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMID_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTB_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTB_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTC_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTC_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTD_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTD_HOTPLUG_INT_EN;
 	if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I915)
 		hotplug_en |= SDVOC_HOTPLUG_INT_EN;
 	if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I915)
@@ -2055,8 +2135,6 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 	}
 
 	I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
-
-	return 0;
 }
 
 static void valleyview_irq_uninstall(struct drm_device *dev)
@@ -2286,6 +2364,9 @@ static int i915_irq_postinstall(struct drm_device *dev)
 		I915_USER_INTERRUPT;
 
 	if (I915_HAS_HOTPLUG(dev)) {
+		I915_WRITE(PORT_HOTPLUG_EN, 0);
+		POSTING_READ(PORT_HOTPLUG_EN);
+
 		/* Enable in IER... */
 		enable_mask |= I915_DISPLAY_PORT_INTERRUPT;
 		/* and unmask in IMR */
@@ -2296,15 +2377,25 @@ static int i915_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(IER, enable_mask);
 	POSTING_READ(IER);
 
+	intel_opregion_enable_asle(dev);
+
+	return 0;
+}
+
+static void i915_hpd_irq_setup(struct drm_device *dev)
+{
+	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	u32 hotplug_en;
+
 	if (I915_HAS_HOTPLUG(dev)) {
-		u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
-
-		if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMIB_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMIC_HOTPLUG_INT_EN;
-		if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
-			hotplug_en |= HDMID_HOTPLUG_INT_EN;
+		hotplug_en = I915_READ(PORT_HOTPLUG_EN);
+
+		if (dev_priv->hotplug_supported_mask & PORTB_HOTPLUG_INT_STATUS)
+			hotplug_en |= PORTB_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & PORTC_HOTPLUG_INT_STATUS)
+			hotplug_en |= PORTC_HOTPLUG_INT_EN;
+		if (dev_priv->hotplug_supported_mask & PORTD_HOTPLUG_INT_STATUS)
+			hotplug_en |= PORTD_HOTPLUG_INT_EN;
 		if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I915)
 			hotplug_en |= SDVOC_HOTPLUG_INT_EN;
 		if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I915)
@@ -2318,10 +2409,6 @@ static int i915_irq_postinstall(struct drm_device *dev)
 
 		I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
 	}
-
-	intel_opregion_enable_asle(dev);
-
-	return 0;
 }
 
 static irqreturn_t i915_irq_handler(int irq, void *arg)
@@ -2481,7 +2568,6 @@ static void i965_irq_preinstall(struct drm_device * dev)
 static int i965_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
-	u32 hotplug_en;
 	u32 enable_mask;
 	u32 error_mask;
 
@@ -2502,6 +2588,7 @@ static int i965_irq_postinstall(struct drm_device *dev)
 
 	dev_priv->pipestat[0] = 0;
 	dev_priv->pipestat[1] = 0;
+	i915_enable_pipestat(dev_priv, 0, PIPE_GMBUS_EVENT_ENABLE);
 
 	/*
 	 * Enable some error detection, note the instruction error mask
@@ -2522,14 +2609,27 @@ static int i965_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(IER, enable_mask);
 	POSTING_READ(IER);
 
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	POSTING_READ(PORT_HOTPLUG_EN);
+
+	intel_opregion_enable_asle(dev);
+
+	return 0;
+}
+
+static void i965_hpd_irq_setup(struct drm_device *dev)
+{
+	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	u32 hotplug_en;
+
 	/* Note HDMI and DP share hotplug bits */
 	hotplug_en = 0;
-	if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMIB_HOTPLUG_INT_EN;
-	if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMIC_HOTPLUG_INT_EN;
-	if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
-		hotplug_en |= HDMID_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTB_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTB_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTC_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTC_HOTPLUG_INT_EN;
+	if (dev_priv->hotplug_supported_mask & PORTD_HOTPLUG_INT_STATUS)
+		hotplug_en |= PORTD_HOTPLUG_INT_EN;
 	if (IS_G4X(dev)) {
 		if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_G4X)
 			hotplug_en |= SDVOC_HOTPLUG_INT_EN;
@@ -2556,10 +2656,6 @@ static int i965_irq_postinstall(struct drm_device *dev)
 	/* Ignore TV since it's buggy */
 
 	I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
-
-	intel_opregion_enable_asle(dev);
-
-	return 0;
 }
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
@@ -2655,6 +2751,9 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		if (blc_event || (iir & I915_ASLE_INTERRUPT))
 			intel_opregion_asle_intr(dev);
 
+		if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
+			gmbus_irq_handler(dev);
+
 		/* With MSI, interrupts are only generated when iir
 		 * transitions from zero to nonzero.  If another bit got
 		 * set while we were handling the existing iir bits, then
@@ -2706,10 +2805,16 @@ void intel_irq_init(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func);
-	INIT_WORK(&dev_priv->error_work, i915_error_work_func);
+	INIT_WORK(&dev_priv->gpu_error.work, i915_error_work_func);
 	INIT_WORK(&dev_priv->rps.work, gen6_pm_rps_work);
 	INIT_WORK(&dev_priv->l3_parity.error_work, ivybridge_parity_work);
 
+	setup_timer(&dev_priv->gpu_error.hangcheck_timer,
+		    i915_hangcheck_elapsed,
+		    (unsigned long) dev);
+
+	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY, PM_QOS_DEFAULT_VALUE);
+
 	dev->driver->get_vblank_counter = i915_get_vblank_counter;
 	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
@@ -2730,7 +2835,8 @@ void intel_irq_init(struct drm_device *dev)
 		dev->driver->irq_uninstall = valleyview_irq_uninstall;
 		dev->driver->enable_vblank = valleyview_enable_vblank;
 		dev->driver->disable_vblank = valleyview_disable_vblank;
-	} else if (IS_IVYBRIDGE(dev)) {
+		dev_priv->display.hpd_irq_setup = valleyview_hpd_irq_setup;
+	} else if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
 		/* Share pre & uninstall handlers with ILK/SNB */
 		dev->driver->irq_handler = ivybridge_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_preinstall;
@@ -2738,14 +2844,6 @@ void intel_irq_init(struct drm_device *dev)
 		dev->driver->irq_uninstall = ironlake_irq_uninstall;
 		dev->driver->enable_vblank = ivybridge_enable_vblank;
 		dev->driver->disable_vblank = ivybridge_disable_vblank;
-	} else if (IS_HASWELL(dev)) {
-		/* Share interrupts handling with IVB */
-		dev->driver->irq_handler = ivybridge_irq_handler;
-		dev->driver->irq_preinstall = ironlake_irq_preinstall;
-		dev->driver->irq_postinstall = ivybridge_irq_postinstall;
-		dev->driver->irq_uninstall = ironlake_irq_uninstall;
-		dev->driver->enable_vblank = ivybridge_enable_vblank;
-		dev->driver->disable_vblank = ivybridge_disable_vblank;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_preinstall;
@@ -2764,13 +2862,23 @@ void intel_irq_init(struct drm_device *dev)
 			dev->driver->irq_postinstall = i915_irq_postinstall;
 			dev->driver->irq_uninstall = i915_irq_uninstall;
 			dev->driver->irq_handler = i915_irq_handler;
+			dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
 		} else {
 			dev->driver->irq_preinstall = i965_irq_preinstall;
 			dev->driver->irq_postinstall = i965_irq_postinstall;
 			dev->driver->irq_uninstall = i965_irq_uninstall;
 			dev->driver->irq_handler = i965_irq_handler;
+			dev_priv->display.hpd_irq_setup = i965_hpd_irq_setup;
 		}
 		dev->driver->enable_vblank = i915_enable_vblank;
 		dev->driver->disable_vblank = i915_disable_vblank;
 	}
 }
+
+void intel_hpd_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->display.hpd_irq_setup)
+		dev_priv->display.hpd_irq_setup(dev);
+}
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 59afb7eb6db6..527b664d3434 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -141,8 +141,15 @@
 #define   VGA_MSR_MEM_EN (1<<1)
 #define   VGA_MSR_CGA_MODE (1<<0)
 
-#define VGA_SR_INDEX 0x3c4
-#define VGA_SR_DATA 0x3c5
+/*
+ * SR01 is the only VGA register touched on non-UMS setups.
+ * VLV doesn't do UMS, so the sequencer index/data registers
+ * are the only VGA registers which need to include
+ * display_mmio_offset.
+ */
+#define VGA_SR_INDEX (dev_priv->info->display_mmio_offset + 0x3c4)
+#define SR01			1
+#define VGA_SR_DATA (dev_priv->info->display_mmio_offset + 0x3c5)
 
 #define VGA_AR_INDEX 0x3c0
 #define   VGA_AR_VID_EN (1<<5)
@@ -301,6 +308,7 @@
 #define   DISPLAY_PLANE_A           (0<<20)
 #define   DISPLAY_PLANE_B           (1<<20)
 #define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2))
+#define   PIPE_CONTROL_GLOBAL_GTT_IVB			(1<<24) /* gen7+ */
 #define   PIPE_CONTROL_CS_STALL				(1<<20)
 #define   PIPE_CONTROL_TLB_INVALIDATE			(1<<18)
 #define   PIPE_CONTROL_QW_WRITE				(1<<14)
@@ -335,17 +343,19 @@
  *  0x801c/3c: core clock bits
  *  0x8048/68: low pass filter coefficients
  *  0x8100: fast clock controls
+ *
+ * DPIO is VLV only.
  */
-#define DPIO_PKT			0x2100
+#define DPIO_PKT			(VLV_DISPLAY_BASE + 0x2100)
 #define  DPIO_RID			(0<<24)
 #define  DPIO_OP_WRITE			(1<<16)
 #define  DPIO_OP_READ			(0<<16)
 #define  DPIO_PORTID			(0x12<<8)
 #define  DPIO_BYTE			(0xf<<4)
 #define  DPIO_BUSY			(1<<0) /* status only */
-#define DPIO_DATA			0x2104
-#define DPIO_REG			0x2108
-#define DPIO_CTL			0x2110
+#define DPIO_DATA			(VLV_DISPLAY_BASE + 0x2104)
+#define DPIO_REG			(VLV_DISPLAY_BASE + 0x2108)
+#define DPIO_CTL			(VLV_DISPLAY_BASE + 0x2110)
 #define  DPIO_MODSEL1			(1<<3) /* if ref clk b == 27 */
 #define  DPIO_MODSEL0			(1<<2) /* if ref clk a == 27 */
 #define  DPIO_SFR_BYPASS		(1<<1)
@@ -556,13 +566,13 @@
 #define IIR		0x020a4
 #define IMR		0x020a8
 #define ISR		0x020ac
-#define VLV_GUNIT_CLOCK_GATE	0x182060
+#define VLV_GUNIT_CLOCK_GATE	(VLV_DISPLAY_BASE + 0x2060)
 #define   GCFG_DIS		(1<<8)
-#define VLV_IIR_RW	0x182084
-#define VLV_IER		0x1820a0
-#define VLV_IIR		0x1820a4
-#define VLV_IMR		0x1820a8
-#define VLV_ISR		0x1820ac
+#define VLV_IIR_RW	(VLV_DISPLAY_BASE + 0x2084)
+#define VLV_IER		(VLV_DISPLAY_BASE + 0x20a0)
+#define VLV_IIR		(VLV_DISPLAY_BASE + 0x20a4)
+#define VLV_IMR		(VLV_DISPLAY_BASE + 0x20a8)
+#define VLV_ISR		(VLV_DISPLAY_BASE + 0x20ac)
 #define   I915_PIPE_CONTROL_NOTIFY_INTERRUPT		(1<<18)
 #define   I915_DISPLAY_PORT_INTERRUPT			(1<<17)
 #define   I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT	(1<<15)
@@ -735,6 +745,7 @@
 #define   GEN7_FF_TS_SCHED_HS0		(0x3<<16)
 #define   GEN7_FF_TS_SCHED_LOAD_BALANCE	(0x1<<16)
 #define   GEN7_FF_TS_SCHED_HW		(0x0<<16) /* Default */
+#define   GEN7_FF_VS_REF_CNT_FFME	(1 << 15)
 #define   GEN7_FF_VS_SCHED_HS1		(0x5<<12)
 #define   GEN7_FF_VS_SCHED_HS0		(0x3<<12)
 #define   GEN7_FF_VS_SCHED_LOAD_BALANCE	(0x1<<12) /* Default */
@@ -921,8 +932,8 @@
 #define   VGA1_PD_P1_DIV_2	(1 << 13)
 #define   VGA1_PD_P1_SHIFT	8
 #define   VGA1_PD_P1_MASK	(0x1f << 8)
-#define _DPLL_A	0x06014
-#define _DPLL_B	0x06018
+#define _DPLL_A	(dev_priv->info->display_mmio_offset + 0x6014)
+#define _DPLL_B	(dev_priv->info->display_mmio_offset + 0x6018)
 #define DPLL(pipe) _PIPE(pipe, _DPLL_A, _DPLL_B)
 #define   DPLL_VCO_ENABLE		(1 << 31)
 #define   DPLL_DVO_HIGH_SPEED		(1 << 30)
@@ -943,23 +954,6 @@
 #define   DPLL_LOCK_VLV			(1<<15)
 #define   DPLL_INTEGRATED_CLOCK_VLV	(1<<13)
 
-#define SRX_INDEX		0x3c4
-#define SRX_DATA		0x3c5
-#define SR01			1
-#define SR01_SCREEN_OFF		(1<<5)
-
-#define PPCR			0x61204
-#define PPCR_ON			(1<<0)
-
-#define DVOB			0x61140
-#define DVOB_ON			(1<<31)
-#define DVOC			0x61160
-#define DVOC_ON			(1<<31)
-#define LVDS			0x61180
-#define LVDS_ON			(1<<31)
-
-/* Scratch pad debug 0 reg:
- */
 #define   DPLL_FPA01_P1_POST_DIV_MASK_I830	0x001f0000
 /*
  * The i830 generation, in LVDS mode, defines P1 as the bit number set within
@@ -998,7 +992,7 @@
 #define   SDVO_MULTIPLIER_MASK			0x000000ff
 #define   SDVO_MULTIPLIER_SHIFT_HIRES		4
 #define   SDVO_MULTIPLIER_SHIFT_VGA		0
-#define _DPLL_A_MD 0x0601c /* 965+ only */
+#define _DPLL_A_MD (dev_priv->info->display_mmio_offset + 0x601c) /* 965+ only */
 /*
  * UDI pixel divider, controlling how many pixels are stuffed into a packet.
  *
@@ -1035,7 +1029,7 @@
  */
 #define   DPLL_MD_VGA_UDI_MULTIPLIER_MASK	0x0000003f
 #define   DPLL_MD_VGA_UDI_MULTIPLIER_SHIFT	0
-#define _DPLL_B_MD 0x06020 /* 965+ only */
+#define _DPLL_B_MD (dev_priv->info->display_mmio_offset + 0x6020) /* 965+ only */
 #define DPLL_MD(pipe) _PIPE(pipe, _DPLL_A_MD, _DPLL_B_MD)
 
 #define _FPA0	0x06040
@@ -1178,15 +1172,15 @@
 #define RAMCLK_GATE_D		0x6210		/* CRL only */
 #define DEUC			0x6214          /* CRL only */
 
-#define FW_BLC_SELF_VLV		0x6500
+#define FW_BLC_SELF_VLV		(VLV_DISPLAY_BASE + 0x6500)
 #define  FW_CSPWRDWNEN		(1<<15)
 
 /*
  * Palette regs
  */
 
-#define _PALETTE_A		0x0a000
-#define _PALETTE_B		0x0a800
+#define _PALETTE_A		(dev_priv->info->display_mmio_offset + 0xa000)
+#define _PALETTE_B		(dev_priv->info->display_mmio_offset + 0xa800)
 #define PALETTE(pipe) _PIPE(pipe, _PALETTE_A, _PALETTE_B)
 
 /* MCH MMIO space */
@@ -1242,6 +1236,10 @@
 #define   MAD_DIMM_A_SIZE_SHIFT		0
 #define   MAD_DIMM_A_SIZE_MASK		(0xff << MAD_DIMM_A_SIZE_SHIFT)
 
+/** snb MCH registers for priority tuning */
+#define MCH_SSKPD			(MCHBAR_MIRROR_BASE_SNB + 0x5d10)
+#define   MCH_SSKPD_WM0_MASK		0x3f
+#define   MCH_SSKPD_WM0_VAL		0xc
 
 /* Clocking configuration register */
 #define CLKCFG			0x10c00
@@ -1551,26 +1549,26 @@
  */
 
 /* Pipe A timing regs */
-#define _HTOTAL_A	0x60000
-#define _HBLANK_A	0x60004
-#define _HSYNC_A		0x60008
-#define _VTOTAL_A	0x6000c
-#define _VBLANK_A	0x60010
-#define _VSYNC_A		0x60014
-#define _PIPEASRC	0x6001c
-#define _BCLRPAT_A	0x60020
-#define _VSYNCSHIFT_A	0x60028
+#define _HTOTAL_A	(dev_priv->info->display_mmio_offset + 0x60000)
+#define _HBLANK_A	(dev_priv->info->display_mmio_offset + 0x60004)
+#define _HSYNC_A	(dev_priv->info->display_mmio_offset + 0x60008)
+#define _VTOTAL_A	(dev_priv->info->display_mmio_offset + 0x6000c)
+#define _VBLANK_A	(dev_priv->info->display_mmio_offset + 0x60010)
+#define _VSYNC_A	(dev_priv->info->display_mmio_offset + 0x60014)
+#define _PIPEASRC	(dev_priv->info->display_mmio_offset + 0x6001c)
+#define _BCLRPAT_A	(dev_priv->info->display_mmio_offset + 0x60020)
+#define _VSYNCSHIFT_A	(dev_priv->info->display_mmio_offset + 0x60028)
 
 /* Pipe B timing regs */
-#define _HTOTAL_B	0x61000
-#define _HBLANK_B	0x61004
-#define _HSYNC_B		0x61008
-#define _VTOTAL_B	0x6100c
-#define _VBLANK_B	0x61010
-#define _VSYNC_B		0x61014
-#define _PIPEBSRC	0x6101c
-#define _BCLRPAT_B	0x61020
-#define _VSYNCSHIFT_B	0x61028
+#define _HTOTAL_B	(dev_priv->info->display_mmio_offset + 0x61000)
+#define _HBLANK_B	(dev_priv->info->display_mmio_offset + 0x61004)
+#define _HSYNC_B	(dev_priv->info->display_mmio_offset + 0x61008)
+#define _VTOTAL_B	(dev_priv->info->display_mmio_offset + 0x6100c)
+#define _VBLANK_B	(dev_priv->info->display_mmio_offset + 0x61010)
+#define _VSYNC_B	(dev_priv->info->display_mmio_offset + 0x61014)
+#define _PIPEBSRC	(dev_priv->info->display_mmio_offset + 0x6101c)
+#define _BCLRPAT_B	(dev_priv->info->display_mmio_offset + 0x61020)
+#define _VSYNCSHIFT_B	(dev_priv->info->display_mmio_offset + 0x61028)
 
 
 #define HTOTAL(trans) _TRANSCODER(trans, _HTOTAL_A, _HTOTAL_B)
@@ -1631,13 +1629,10 @@
 
 
 /* Hotplug control (945+ only) */
-#define PORT_HOTPLUG_EN		0x61110
-#define   HDMIB_HOTPLUG_INT_EN			(1 << 29)
-#define   DPB_HOTPLUG_INT_EN			(1 << 29)
-#define   HDMIC_HOTPLUG_INT_EN			(1 << 28)
-#define   DPC_HOTPLUG_INT_EN			(1 << 28)
-#define   HDMID_HOTPLUG_INT_EN			(1 << 27)
-#define   DPD_HOTPLUG_INT_EN			(1 << 27)
+#define PORT_HOTPLUG_EN		(dev_priv->info->display_mmio_offset + 0x61110)
+#define   PORTB_HOTPLUG_INT_EN			(1 << 29)
+#define   PORTC_HOTPLUG_INT_EN			(1 << 28)
+#define   PORTD_HOTPLUG_INT_EN			(1 << 27)
 #define   SDVOB_HOTPLUG_INT_EN			(1 << 26)
 #define   SDVOC_HOTPLUG_INT_EN			(1 << 25)
 #define   TV_HOTPLUG_INT_EN			(1 << 18)
@@ -1658,21 +1653,14 @@
 #define CRT_HOTPLUG_DETECT_VOLTAGE_325MV	(0 << 2)
 #define CRT_HOTPLUG_DETECT_VOLTAGE_475MV	(1 << 2)
 
-#define PORT_HOTPLUG_STAT	0x61114
+#define PORT_HOTPLUG_STAT	(dev_priv->info->display_mmio_offset + 0x61114)
 /* HDMI/DP bits are gen4+ */
-#define   DPB_HOTPLUG_LIVE_STATUS               (1 << 29)
-#define   DPC_HOTPLUG_LIVE_STATUS               (1 << 28)
-#define   DPD_HOTPLUG_LIVE_STATUS               (1 << 27)
-#define   DPD_HOTPLUG_INT_STATUS		(3 << 21)
-#define   DPC_HOTPLUG_INT_STATUS		(3 << 19)
-#define   DPB_HOTPLUG_INT_STATUS		(3 << 17)
-/* HDMI bits are shared with the DP bits */
-#define   HDMIB_HOTPLUG_LIVE_STATUS             (1 << 29)
-#define   HDMIC_HOTPLUG_LIVE_STATUS             (1 << 28)
-#define   HDMID_HOTPLUG_LIVE_STATUS             (1 << 27)
-#define   HDMID_HOTPLUG_INT_STATUS		(3 << 21)
-#define   HDMIC_HOTPLUG_INT_STATUS		(3 << 19)
-#define   HDMIB_HOTPLUG_INT_STATUS		(3 << 17)
+#define   PORTB_HOTPLUG_LIVE_STATUS               (1 << 29)
+#define   PORTC_HOTPLUG_LIVE_STATUS               (1 << 28)
+#define   PORTD_HOTPLUG_LIVE_STATUS               (1 << 27)
+#define   PORTD_HOTPLUG_INT_STATUS		(3 << 21)
+#define   PORTC_HOTPLUG_INT_STATUS		(3 << 19)
+#define   PORTB_HOTPLUG_INT_STATUS		(3 << 17)
 /* CRT/TV common between gen3+ */
 #define   CRT_HOTPLUG_INT_STATUS		(1 << 11)
 #define   TV_HOTPLUG_INT_STATUS			(1 << 10)
@@ -1877,7 +1865,7 @@
 #define PP_DIVISOR	0x61210
 
 /* Panel fitting */
-#define PFIT_CONTROL	0x61230
+#define PFIT_CONTROL	(dev_priv->info->display_mmio_offset + 0x61230)
 #define   PFIT_ENABLE		(1 << 31)
 #define   PFIT_PIPE_MASK	(3 << 29)
 #define   PFIT_PIPE_SHIFT	29
@@ -1895,9 +1883,7 @@
 #define   PFIT_SCALING_PROGRAMMED (1 << 26)
 #define   PFIT_SCALING_PILLAR	(2 << 26)
 #define   PFIT_SCALING_LETTER	(3 << 26)
-#define PFIT_PGM_RATIOS	0x61234
-#define   PFIT_VERT_SCALE_MASK			0xfff00000
-#define   PFIT_HORIZ_SCALE_MASK			0x0000fff0
+#define PFIT_PGM_RATIOS	(dev_priv->info->display_mmio_offset + 0x61234)
 /* Pre-965 */
 #define		PFIT_VERT_SCALE_SHIFT		20
 #define		PFIT_VERT_SCALE_MASK		0xfff00000
@@ -1909,7 +1895,7 @@
 #define		PFIT_HORIZ_SCALE_SHIFT_965	0
 #define		PFIT_HORIZ_SCALE_MASK_965	0x00001fff
 
-#define PFIT_AUTO_RATIOS 0x61238
+#define PFIT_AUTO_RATIOS (dev_priv->info->display_mmio_offset + 0x61238)
 
 /* Backlight control */
 #define BLC_PWM_CTL2		0x61250 /* 965+ only */
@@ -2639,10 +2625,10 @@
 /* Display & cursor control */
 
 /* Pipe A */
-#define _PIPEADSL		0x70000
+#define _PIPEADSL		(dev_priv->info->display_mmio_offset + 0x70000)
 #define   DSL_LINEMASK_GEN2	0x00000fff
 #define   DSL_LINEMASK_GEN3	0x00001fff
-#define _PIPEACONF		0x70008
+#define _PIPEACONF		(dev_priv->info->display_mmio_offset + 0x70008)
 #define   PIPECONF_ENABLE	(1<<31)
 #define   PIPECONF_DISABLE	0
 #define   PIPECONF_DOUBLE_WIDE	(1<<30)
@@ -2671,18 +2657,19 @@
 #define   PIPECONF_INTERLACED_DBL_ILK		(4 << 21) /* ilk/snb only */
 #define   PIPECONF_PFIT_PF_INTERLACED_DBL_ILK	(5 << 21) /* ilk/snb only */
 #define   PIPECONF_CXSR_DOWNCLOCK	(1<<16)
-#define   PIPECONF_BPP_MASK	(0x000000e0)
-#define   PIPECONF_BPP_8	(0<<5)
-#define   PIPECONF_BPP_10	(1<<5)
-#define   PIPECONF_BPP_6	(2<<5)
-#define   PIPECONF_BPP_12	(3<<5)
+#define   PIPECONF_COLOR_RANGE_SELECT	(1 << 13)
+#define   PIPECONF_BPC_MASK	(0x7 << 5)
+#define   PIPECONF_8BPC		(0<<5)
+#define   PIPECONF_10BPC	(1<<5)
+#define   PIPECONF_6BPC		(2<<5)
+#define   PIPECONF_12BPC	(3<<5)
 #define   PIPECONF_DITHER_EN	(1<<4)
 #define   PIPECONF_DITHER_TYPE_MASK (0x0000000c)
 #define   PIPECONF_DITHER_TYPE_SP (0<<2)
 #define   PIPECONF_DITHER_TYPE_ST1 (1<<2)
 #define   PIPECONF_DITHER_TYPE_ST2 (2<<2)
 #define   PIPECONF_DITHER_TYPE_TEMP (3<<2)
-#define _PIPEASTAT		0x70024
+#define _PIPEASTAT		(dev_priv->info->display_mmio_offset + 0x70024)
 #define   PIPE_FIFO_UNDERRUN_STATUS		(1UL<<31)
 #define   SPRITE1_FLIPDONE_INT_EN_VLV		(1UL<<30)
 #define   PIPE_CRC_ERROR_ENABLE			(1UL<<29)
@@ -2693,7 +2680,7 @@
 #define   PIPE_VSYNC_INTERRUPT_ENABLE		(1UL<<25)
 #define   PIPE_DISPLAY_LINE_COMPARE_ENABLE	(1UL<<24)
 #define   PIPE_DPST_EVENT_ENABLE		(1UL<<23)
-#define   SPRITE0_FLIP_DONE_INT_EN_VLV		(1UL<<26)
+#define   SPRITE0_FLIP_DONE_INT_EN_VLV		(1UL<<22)
 #define   PIPE_LEGACY_BLC_EVENT_ENABLE		(1UL<<22)
 #define   PIPE_ODD_FIELD_INTERRUPT_ENABLE	(1UL<<21)
 #define   PIPE_EVEN_FIELD_INTERRUPT_ENABLE	(1UL<<20)
@@ -2703,7 +2690,7 @@
 #define   PIPEA_HBLANK_INT_EN_VLV		(1UL<<16)
 #define   PIPE_OVERLAY_UPDATED_ENABLE		(1UL<<16)
 #define   SPRITE1_FLIPDONE_INT_STATUS_VLV	(1UL<<15)
-#define   SPRITE0_FLIPDONE_INT_STATUS_VLV	(1UL<<15)
+#define   SPRITE0_FLIPDONE_INT_STATUS_VLV	(1UL<<14)
 #define   PIPE_CRC_ERROR_INTERRUPT_STATUS	(1UL<<13)
 #define   PIPE_CRC_DONE_INTERRUPT_STATUS	(1UL<<12)
 #define   PIPE_GMBUS_INTERRUPT_STATUS		(1UL<<11)
@@ -2719,11 +2706,6 @@
 #define   PIPE_START_VBLANK_INTERRUPT_STATUS	(1UL<<2) /* 965 or later */
 #define   PIPE_VBLANK_INTERRUPT_STATUS		(1UL<<1)
 #define   PIPE_OVERLAY_UPDATED_STATUS		(1UL<<0)
-#define   PIPE_BPC_MASK				(7 << 5) /* Ironlake */
-#define   PIPE_8BPC				(0 << 5)
-#define   PIPE_10BPC				(1 << 5)
-#define   PIPE_6BPC				(2 << 5)
-#define   PIPE_12BPC				(3 << 5)
 
 #define PIPESRC(pipe) _PIPE(pipe, _PIPEASRC, _PIPEBSRC)
 #define PIPECONF(tran) _TRANSCODER(tran, _PIPEACONF, _PIPEBCONF)
@@ -2732,7 +2714,7 @@
 #define PIPEFRAMEPIXEL(pipe)  _PIPE(pipe, _PIPEAFRAMEPIXEL, _PIPEBFRAMEPIXEL)
 #define PIPESTAT(pipe) _PIPE(pipe, _PIPEASTAT, _PIPEBSTAT)
 
-#define VLV_DPFLIPSTAT				0x70028
+#define VLV_DPFLIPSTAT				(VLV_DISPLAY_BASE + 0x70028)
 #define   PIPEB_LINE_COMPARE_INT_EN		(1<<29)
 #define   PIPEB_HLINE_INT_EN			(1<<28)
 #define   PIPEB_VBLANK_INT_EN			(1<<27)
@@ -2746,7 +2728,7 @@
 #define   SPRITEA_FLIPDONE_INT_EN		(1<<17)
 #define   PLANEA_FLIPDONE_INT_EN		(1<<16)
 
-#define DPINVGTT				0x7002c /* VLV only */
+#define DPINVGTT				(VLV_DISPLAY_BASE + 0x7002c) /* VLV only */
 #define   CURSORB_INVALID_GTT_INT_EN		(1<<23)
 #define   CURSORA_INVALID_GTT_INT_EN		(1<<22)
 #define   SPRITED_INVALID_GTT_INT_EN		(1<<21)
@@ -2774,7 +2756,7 @@
 #define   DSPARB_BEND_SHIFT	9 /* on 855 */
 #define   DSPARB_AEND_SHIFT	0
 
-#define DSPFW1			0x70034
+#define DSPFW1			(dev_priv->info->display_mmio_offset + 0x70034)
 #define   DSPFW_SR_SHIFT	23
 #define   DSPFW_SR_MASK		(0x1ff<<23)
 #define   DSPFW_CURSORB_SHIFT	16
@@ -2782,11 +2764,11 @@
 #define   DSPFW_PLANEB_SHIFT	8
 #define   DSPFW_PLANEB_MASK	(0x7f<<8)
 #define   DSPFW_PLANEA_MASK	(0x7f)
-#define DSPFW2			0x70038
+#define DSPFW2			(dev_priv->info->display_mmio_offset + 0x70038)
 #define   DSPFW_CURSORA_MASK	0x00003f00
 #define   DSPFW_CURSORA_SHIFT	8
 #define   DSPFW_PLANEC_MASK	(0x7f)
-#define DSPFW3			0x7003c
+#define DSPFW3			(dev_priv->info->display_mmio_offset + 0x7003c)
 #define   DSPFW_HPLL_SR_EN	(1<<31)
 #define   DSPFW_CURSOR_SR_SHIFT	24
 #define   PINEVIEW_SELF_REFRESH_EN	(1<<30)
@@ -2798,13 +2780,13 @@
 /* drain latency register values*/
 #define DRAIN_LATENCY_PRECISION_32	32
 #define DRAIN_LATENCY_PRECISION_16	16
-#define VLV_DDL1			0x70050
+#define VLV_DDL1			(VLV_DISPLAY_BASE + 0x70050)
 #define DDL_CURSORA_PRECISION_32	(1<<31)
 #define DDL_CURSORA_PRECISION_16	(0<<31)
 #define DDL_CURSORA_SHIFT		24
 #define DDL_PLANEA_PRECISION_32		(1<<7)
 #define DDL_PLANEA_PRECISION_16		(0<<7)
-#define VLV_DDL2			0x70054
+#define VLV_DDL2			(VLV_DISPLAY_BASE + 0x70054)
 #define DDL_CURSORB_PRECISION_32	(1<<31)
 #define DDL_CURSORB_PRECISION_16	(0<<31)
 #define DDL_CURSORB_SHIFT		24
@@ -2948,10 +2930,10 @@
  *  } while (high1 != high2);
  *  frame = (high1 << 8) | low1;
  */
-#define _PIPEAFRAMEHIGH          0x70040
+#define _PIPEAFRAMEHIGH          (dev_priv->info->display_mmio_offset + 0x70040)
 #define   PIPE_FRAME_HIGH_MASK    0x0000ffff
 #define   PIPE_FRAME_HIGH_SHIFT   0
-#define _PIPEAFRAMEPIXEL         0x70044
+#define _PIPEAFRAMEPIXEL         (dev_priv->info->display_mmio_offset + 0x70044)
 #define   PIPE_FRAME_LOW_MASK     0xff000000
 #define   PIPE_FRAME_LOW_SHIFT    24
 #define   PIPE_PIXEL_MASK         0x00ffffff
@@ -2962,11 +2944,12 @@
 #define PIPE_FRMCOUNT_GM45(pipe) _PIPE(pipe, _PIPEA_FRMCOUNT_GM45, _PIPEB_FRMCOUNT_GM45)
 
 /* Cursor A & B regs */
-#define _CURACNTR		0x70080
+#define _CURACNTR		(dev_priv->info->display_mmio_offset + 0x70080)
 /* Old style CUR*CNTR flags (desktop 8xx) */
 #define   CURSOR_ENABLE		0x80000000
 #define   CURSOR_GAMMA_ENABLE	0x40000000
 #define   CURSOR_STRIDE_MASK	0x30000000
+#define   CURSOR_PIPE_CSC_ENABLE (1<<24)
 #define   CURSOR_FORMAT_SHIFT	24
 #define   CURSOR_FORMAT_MASK	(0x07 << CURSOR_FORMAT_SHIFT)
 #define   CURSOR_FORMAT_2C	(0x00 << CURSOR_FORMAT_SHIFT)
@@ -2983,16 +2966,16 @@
 #define   MCURSOR_PIPE_A	0x00
 #define   MCURSOR_PIPE_B	(1 << 28)
 #define   MCURSOR_GAMMA_ENABLE  (1 << 26)
-#define _CURABASE		0x70084
-#define _CURAPOS			0x70088
+#define _CURABASE		(dev_priv->info->display_mmio_offset + 0x70084)
+#define _CURAPOS		(dev_priv->info->display_mmio_offset + 0x70088)
 #define   CURSOR_POS_MASK       0x007FF
 #define   CURSOR_POS_SIGN       0x8000
 #define   CURSOR_X_SHIFT        0
 #define   CURSOR_Y_SHIFT        16
 #define CURSIZE			0x700a0
-#define _CURBCNTR		0x700c0
-#define _CURBBASE		0x700c4
-#define _CURBPOS			0x700c8
+#define _CURBCNTR		(dev_priv->info->display_mmio_offset + 0x700c0)
+#define _CURBBASE		(dev_priv->info->display_mmio_offset + 0x700c4)
+#define _CURBPOS		(dev_priv->info->display_mmio_offset + 0x700c8)
 
 #define _CURBCNTR_IVB		0x71080
 #define _CURBBASE_IVB		0x71084
@@ -3007,7 +2990,7 @@
 #define CURPOS_IVB(pipe) _PIPE(pipe, _CURAPOS, _CURBPOS_IVB)
 
 /* Display A control */
-#define _DSPACNTR                0x70180
+#define _DSPACNTR                (dev_priv->info->display_mmio_offset + 0x70180)
 #define   DISPLAY_PLANE_ENABLE			(1<<31)
 #define   DISPLAY_PLANE_DISABLE			0
 #define   DISPPLANE_GAMMA_ENABLE		(1<<30)
@@ -3028,6 +3011,7 @@
 #define   DISPPLANE_RGBA888			(0xf<<26)
 #define   DISPPLANE_STEREO_ENABLE		(1<<25)
 #define   DISPPLANE_STEREO_DISABLE		0
+#define   DISPPLANE_PIPE_CSC_ENABLE		(1<<24)
 #define   DISPPLANE_SEL_PIPE_SHIFT		24
 #define   DISPPLANE_SEL_PIPE_MASK		(3<<DISPPLANE_SEL_PIPE_SHIFT)
 #define   DISPPLANE_SEL_PIPE_A			0
@@ -3040,14 +3024,14 @@
 #define   DISPPLANE_STEREO_POLARITY_SECOND	(1<<18)
 #define   DISPPLANE_TRICKLE_FEED_DISABLE	(1<<14) /* Ironlake */
 #define   DISPPLANE_TILED			(1<<10)
-#define _DSPAADDR		0x70184
-#define _DSPASTRIDE		0x70188
-#define _DSPAPOS			0x7018C /* reserved */
-#define _DSPASIZE		0x70190
-#define _DSPASURF		0x7019C /* 965+ only */
-#define _DSPATILEOFF		0x701A4 /* 965+ only */
-#define _DSPAOFFSET		0x701A4 /* HSW */
-#define _DSPASURFLIVE		0x701AC
+#define _DSPAADDR		(dev_priv->info->display_mmio_offset + 0x70184)
+#define _DSPASTRIDE		(dev_priv->info->display_mmio_offset + 0x70188)
+#define _DSPAPOS		(dev_priv->info->display_mmio_offset + 0x7018C) /* reserved */
+#define _DSPASIZE		(dev_priv->info->display_mmio_offset + 0x70190)
+#define _DSPASURF		(dev_priv->info->display_mmio_offset + 0x7019C) /* 965+ only */
+#define _DSPATILEOFF		(dev_priv->info->display_mmio_offset + 0x701A4) /* 965+ only */
+#define _DSPAOFFSET		(dev_priv->info->display_mmio_offset + 0x701A4) /* HSW */
+#define _DSPASURFLIVE		(dev_priv->info->display_mmio_offset + 0x701AC)
 
 #define DSPCNTR(plane) _PIPE(plane, _DSPACNTR, _DSPBCNTR)
 #define DSPADDR(plane) _PIPE(plane, _DSPAADDR, _DSPBADDR)
@@ -3068,44 +3052,44 @@
 		(I915_WRITE((reg), (gfx_addr) | I915_LO_DISPBASE(I915_READ(reg))))
 
 /* VBIOS flags */
-#define SWF00			0x71410
-#define SWF01			0x71414
-#define SWF02			0x71418
-#define SWF03			0x7141c
-#define SWF04			0x71420
-#define SWF05			0x71424
-#define SWF06			0x71428
-#define SWF10			0x70410
-#define SWF11			0x70414
-#define SWF14			0x71420
-#define SWF30			0x72414
-#define SWF31			0x72418
-#define SWF32			0x7241c
+#define SWF00			(dev_priv->info->display_mmio_offset + 0x71410)
+#define SWF01			(dev_priv->info->display_mmio_offset + 0x71414)
+#define SWF02			(dev_priv->info->display_mmio_offset + 0x71418)
+#define SWF03			(dev_priv->info->display_mmio_offset + 0x7141c)
+#define SWF04			(dev_priv->info->display_mmio_offset + 0x71420)
+#define SWF05			(dev_priv->info->display_mmio_offset + 0x71424)
+#define SWF06			(dev_priv->info->display_mmio_offset + 0x71428)
+#define SWF10			(dev_priv->info->display_mmio_offset + 0x70410)
+#define SWF11			(dev_priv->info->display_mmio_offset + 0x70414)
+#define SWF14			(dev_priv->info->display_mmio_offset + 0x71420)
+#define SWF30			(dev_priv->info->display_mmio_offset + 0x72414)
+#define SWF31			(dev_priv->info->display_mmio_offset + 0x72418)
+#define SWF32			(dev_priv->info->display_mmio_offset + 0x7241c)
 
 /* Pipe B */
-#define _PIPEBDSL		0x71000
-#define _PIPEBCONF		0x71008
-#define _PIPEBSTAT		0x71024
-#define _PIPEBFRAMEHIGH		0x71040
-#define _PIPEBFRAMEPIXEL		0x71044
+#define _PIPEBDSL		(dev_priv->info->display_mmio_offset + 0x71000)
+#define _PIPEBCONF		(dev_priv->info->display_mmio_offset + 0x71008)
+#define _PIPEBSTAT		(dev_priv->info->display_mmio_offset + 0x71024)
+#define _PIPEBFRAMEHIGH		(dev_priv->info->display_mmio_offset + 0x71040)
+#define _PIPEBFRAMEPIXEL	(dev_priv->info->display_mmio_offset + 0x71044)
 #define _PIPEB_FRMCOUNT_GM45	0x71040
 #define _PIPEB_FLIPCOUNT_GM45	0x71044
 
 
 /* Display B control */
-#define _DSPBCNTR		0x71180
+#define _DSPBCNTR		(dev_priv->info->display_mmio_offset + 0x71180)
 #define   DISPPLANE_ALPHA_TRANS_ENABLE		(1<<15)
 #define   DISPPLANE_ALPHA_TRANS_DISABLE		0
 #define   DISPPLANE_SPRITE_ABOVE_DISPLAY	0
 #define   DISPPLANE_SPRITE_ABOVE_OVERLAY	(1)
-#define _DSPBADDR		0x71184
-#define _DSPBSTRIDE		0x71188
-#define _DSPBPOS			0x7118C
-#define _DSPBSIZE		0x71190
-#define _DSPBSURF		0x7119C
-#define _DSPBTILEOFF		0x711A4
-#define _DSPBOFFSET		0x711A4
-#define _DSPBSURFLIVE		0x711AC
+#define _DSPBADDR		(dev_priv->info->display_mmio_offset + 0x71184)
+#define _DSPBSTRIDE		(dev_priv->info->display_mmio_offset + 0x71188)
+#define _DSPBPOS		(dev_priv->info->display_mmio_offset + 0x7118C)
+#define _DSPBSIZE		(dev_priv->info->display_mmio_offset + 0x71190)
+#define _DSPBSURF		(dev_priv->info->display_mmio_offset + 0x7119C)
+#define _DSPBTILEOFF		(dev_priv->info->display_mmio_offset + 0x711A4)
+#define _DSPBOFFSET		(dev_priv->info->display_mmio_offset + 0x711A4)
+#define _DSPBSURFLIVE		(dev_priv->info->display_mmio_offset + 0x711AC)
 
 /* Sprite A control */
 #define _DVSACNTR		0x72180
@@ -3116,6 +3100,7 @@
 #define   DVS_FORMAT_RGBX101010	(1<<25)
 #define   DVS_FORMAT_RGBX888	(2<<25)
 #define   DVS_FORMAT_RGBX161616	(3<<25)
+#define   DVS_PIPE_CSC_ENABLE   (1<<24)
 #define   DVS_SOURCE_KEY	(1<<22)
 #define   DVS_RGB_ORDER_XBGR	(1<<20)
 #define   DVS_YUV_BYTE_ORDER_MASK (3<<16)
@@ -3183,7 +3168,7 @@
 #define   SPRITE_FORMAT_RGBX161616	(3<<25)
 #define   SPRITE_FORMAT_YUV444		(4<<25)
 #define   SPRITE_FORMAT_XR_BGR101010	(5<<25) /* Extended range */
-#define   SPRITE_CSC_ENABLE		(1<<24)
+#define   SPRITE_PIPE_CSC_ENABLE	(1<<24)
 #define   SPRITE_SOURCE_KEY		(1<<22)
 #define   SPRITE_RGB_ORDER_RGBX		(1<<20) /* only for 888 and 161616 */
 #define   SPRITE_YUV_TO_RGB_CSC_DISABLE	(1<<19)
@@ -3254,6 +3239,8 @@
 # define VGA_2X_MODE				(1 << 30)
 # define VGA_PIPE_B_SELECT			(1 << 29)
 
+#define VLV_VGACNTRL		(VLV_DISPLAY_BASE + 0x71400)
+
 /* Ironlake */
 
 #define CPU_VGACNTRL	0x41000
@@ -3294,41 +3281,41 @@
 #define  FDI_PLL_FREQ_DISABLE_COUNT_LIMIT_MASK  0xff
 
 
-#define _PIPEA_DATA_M1           0x60030
+#define _PIPEA_DATA_M1           (dev_priv->info->display_mmio_offset + 0x60030)
 #define  TU_SIZE(x)             (((x)-1) << 25) /* default size 64 */
 #define  TU_SIZE_MASK           0x7e000000
 #define  PIPE_DATA_M1_OFFSET    0
-#define _PIPEA_DATA_N1           0x60034
+#define _PIPEA_DATA_N1           (dev_priv->info->display_mmio_offset + 0x60034)
 #define  PIPE_DATA_N1_OFFSET    0
 
-#define _PIPEA_DATA_M2           0x60038
+#define _PIPEA_DATA_M2           (dev_priv->info->display_mmio_offset + 0x60038)
 #define  PIPE_DATA_M2_OFFSET    0
-#define _PIPEA_DATA_N2           0x6003c
+#define _PIPEA_DATA_N2           (dev_priv->info->display_mmio_offset + 0x6003c)
 #define  PIPE_DATA_N2_OFFSET    0
 
-#define _PIPEA_LINK_M1           0x60040
+#define _PIPEA_LINK_M1           (dev_priv->info->display_mmio_offset + 0x60040)
 #define  PIPE_LINK_M1_OFFSET    0
-#define _PIPEA_LINK_N1           0x60044
+#define _PIPEA_LINK_N1           (dev_priv->info->display_mmio_offset + 0x60044)
 #define  PIPE_LINK_N1_OFFSET    0
 
-#define _PIPEA_LINK_M2           0x60048
+#define _PIPEA_LINK_M2           (dev_priv->info->display_mmio_offset + 0x60048)
 #define  PIPE_LINK_M2_OFFSET    0
-#define _PIPEA_LINK_N2           0x6004c
+#define _PIPEA_LINK_N2           (dev_priv->info->display_mmio_offset + 0x6004c)
 #define  PIPE_LINK_N2_OFFSET    0
 
 /* PIPEB timing regs are same start from 0x61000 */
 
-#define _PIPEB_DATA_M1           0x61030
-#define _PIPEB_DATA_N1           0x61034
+#define _PIPEB_DATA_M1           (dev_priv->info->display_mmio_offset + 0x61030)
+#define _PIPEB_DATA_N1           (dev_priv->info->display_mmio_offset + 0x61034)
 
-#define _PIPEB_DATA_M2           0x61038
-#define _PIPEB_DATA_N2           0x6103c
+#define _PIPEB_DATA_M2           (dev_priv->info->display_mmio_offset + 0x61038)
+#define _PIPEB_DATA_N2           (dev_priv->info->display_mmio_offset + 0x6103c)
 
-#define _PIPEB_LINK_M1           0x61040
-#define _PIPEB_LINK_N1           0x61044
+#define _PIPEB_LINK_M1           (dev_priv->info->display_mmio_offset + 0x61040)
+#define _PIPEB_LINK_N1           (dev_priv->info->display_mmio_offset + 0x61044)
 
-#define _PIPEB_LINK_M2           0x61048
-#define _PIPEB_LINK_N2           0x6104c
+#define _PIPEB_LINK_M2           (dev_priv->info->display_mmio_offset + 0x61048)
+#define _PIPEB_LINK_N2           (dev_priv->info->display_mmio_offset + 0x6104c)
 
 #define PIPE_DATA_M1(tran) _TRANSCODER(tran, _PIPEA_DATA_M1, _PIPEB_DATA_M1)
 #define PIPE_DATA_N1(tran) _TRANSCODER(tran, _PIPEA_DATA_N1, _PIPEB_DATA_N1)
@@ -3581,27 +3568,30 @@
 #define PORTD_PULSE_DURATION_6ms        (2 << 18)
 #define PORTD_PULSE_DURATION_100ms      (3 << 18)
 #define PORTD_PULSE_DURATION_MASK	(3 << 18)
-#define PORTD_HOTPLUG_NO_DETECT         (0)
-#define PORTD_HOTPLUG_SHORT_DETECT      (1 << 16)
-#define PORTD_HOTPLUG_LONG_DETECT       (1 << 17)
+#define PORTD_HOTPLUG_STATUS_MASK	(0x3 << 16)
+#define  PORTD_HOTPLUG_NO_DETECT	(0 << 16)
+#define  PORTD_HOTPLUG_SHORT_DETECT	(1 << 16)
+#define  PORTD_HOTPLUG_LONG_DETECT	(2 << 16)
 #define PORTC_HOTPLUG_ENABLE            (1 << 12)
 #define PORTC_PULSE_DURATION_2ms        (0)
 #define PORTC_PULSE_DURATION_4_5ms      (1 << 10)
 #define PORTC_PULSE_DURATION_6ms        (2 << 10)
 #define PORTC_PULSE_DURATION_100ms      (3 << 10)
 #define PORTC_PULSE_DURATION_MASK	(3 << 10)
-#define PORTC_HOTPLUG_NO_DETECT         (0)
-#define PORTC_HOTPLUG_SHORT_DETECT      (1 << 8)
-#define PORTC_HOTPLUG_LONG_DETECT       (1 << 9)
+#define PORTC_HOTPLUG_STATUS_MASK	(0x3 << 8)
+#define  PORTC_HOTPLUG_NO_DETECT	(0 << 8)
+#define  PORTC_HOTPLUG_SHORT_DETECT	(1 << 8)
+#define  PORTC_HOTPLUG_LONG_DETECT	(2 << 8)
 #define PORTB_HOTPLUG_ENABLE            (1 << 4)
 #define PORTB_PULSE_DURATION_2ms        (0)
 #define PORTB_PULSE_DURATION_4_5ms      (1 << 2)
 #define PORTB_PULSE_DURATION_6ms        (2 << 2)
 #define PORTB_PULSE_DURATION_100ms      (3 << 2)
 #define PORTB_PULSE_DURATION_MASK	(3 << 2)
-#define PORTB_HOTPLUG_NO_DETECT         (0)
-#define PORTB_HOTPLUG_SHORT_DETECT      (1 << 0)
-#define PORTB_HOTPLUG_LONG_DETECT       (1 << 1)
+#define PORTB_HOTPLUG_STATUS_MASK	(0x3 << 0)
+#define  PORTB_HOTPLUG_NO_DETECT	(0 << 0)
+#define  PORTB_HOTPLUG_SHORT_DETECT	(1 << 0)
+#define  PORTB_HOTPLUG_LONG_DETECT	(2 << 0)
 
 #define PCH_GPIOA               0xc5010
 #define PCH_GPIOB               0xc5014
@@ -3722,13 +3712,13 @@
 #define TVIDEO_DIP_DATA(pipe) _PIPE(pipe, _VIDEO_DIP_DATA_A, _VIDEO_DIP_DATA_B)
 #define TVIDEO_DIP_GCP(pipe) _PIPE(pipe, _VIDEO_DIP_GCP_A, _VIDEO_DIP_GCP_B)
 
-#define VLV_VIDEO_DIP_CTL_A		0x60200
-#define VLV_VIDEO_DIP_DATA_A		0x60208
-#define VLV_VIDEO_DIP_GDCP_PAYLOAD_A	0x60210
+#define VLV_VIDEO_DIP_CTL_A		(VLV_DISPLAY_BASE + 0x60200)
+#define VLV_VIDEO_DIP_DATA_A		(VLV_DISPLAY_BASE + 0x60208)
+#define VLV_VIDEO_DIP_GDCP_PAYLOAD_A	(VLV_DISPLAY_BASE + 0x60210)
 
-#define VLV_VIDEO_DIP_CTL_B		0x61170
-#define VLV_VIDEO_DIP_DATA_B		0x61174
-#define VLV_VIDEO_DIP_GDCP_PAYLOAD_B	0x61178
+#define VLV_VIDEO_DIP_CTL_B		(VLV_DISPLAY_BASE + 0x61170)
+#define VLV_VIDEO_DIP_DATA_B		(VLV_DISPLAY_BASE + 0x61174)
+#define VLV_VIDEO_DIP_GDCP_PAYLOAD_B	(VLV_DISPLAY_BASE + 0x61178)
 
 #define VLV_TVIDEO_DIP_CTL(pipe) \
 	 _PIPE(pipe, VLV_VIDEO_DIP_CTL_A, VLV_VIDEO_DIP_CTL_B)
@@ -3820,8 +3810,6 @@
 #define  TRANS_FSYNC_DELAY_HB2  (1<<27)
 #define  TRANS_FSYNC_DELAY_HB3  (2<<27)
 #define  TRANS_FSYNC_DELAY_HB4  (3<<27)
-#define  TRANS_DP_AUDIO_ONLY    (1<<26)
-#define  TRANS_DP_VIDEO_AUDIO   (0<<26)
 #define  TRANS_INTERLACE_MASK   (7<<21)
 #define  TRANS_PROGRESSIVE      (0<<21)
 #define  TRANS_INTERLACED       (3<<21)
@@ -3927,7 +3915,7 @@
 #define  FDI_10BPC                      (1<<16)
 #define  FDI_6BPC                       (2<<16)
 #define  FDI_12BPC                      (3<<16)
-#define  FDI_LINK_REVERSE_OVERWRITE     (1<<15)
+#define  FDI_RX_LINK_REVERSAL_OVERRIDE  (1<<15)
 #define  FDI_DMI_LINK_REVERSE_MASK      (1<<14)
 #define  FDI_RX_PLL_ENABLE              (1<<13)
 #define  FDI_FS_ERR_CORRECT_ENABLE      (1<<11)
@@ -4020,17 +4008,17 @@
 #define  LVDS_DETECTED	(1 << 1)
 
 /* vlv has 2 sets of panel control regs. */
-#define PIPEA_PP_STATUS         0x61200
-#define PIPEA_PP_CONTROL        0x61204
-#define PIPEA_PP_ON_DELAYS      0x61208
-#define PIPEA_PP_OFF_DELAYS     0x6120c
-#define PIPEA_PP_DIVISOR        0x61210
-
-#define PIPEB_PP_STATUS         0x61300
-#define PIPEB_PP_CONTROL        0x61304
-#define PIPEB_PP_ON_DELAYS      0x61308
-#define PIPEB_PP_OFF_DELAYS     0x6130c
-#define PIPEB_PP_DIVISOR        0x61310
+#define PIPEA_PP_STATUS         (VLV_DISPLAY_BASE + 0x61200)
+#define PIPEA_PP_CONTROL        (VLV_DISPLAY_BASE + 0x61204)
+#define PIPEA_PP_ON_DELAYS      (VLV_DISPLAY_BASE + 0x61208)
+#define PIPEA_PP_OFF_DELAYS     (VLV_DISPLAY_BASE + 0x6120c)
+#define PIPEA_PP_DIVISOR        (VLV_DISPLAY_BASE + 0x61210)
+
+#define PIPEB_PP_STATUS         (VLV_DISPLAY_BASE + 0x61300)
+#define PIPEB_PP_CONTROL        (VLV_DISPLAY_BASE + 0x61304)
+#define PIPEB_PP_ON_DELAYS      (VLV_DISPLAY_BASE + 0x61308)
+#define PIPEB_PP_OFF_DELAYS     (VLV_DISPLAY_BASE + 0x6130c)
+#define PIPEB_PP_DIVISOR        (VLV_DISPLAY_BASE + 0x61310)
 
 #define PCH_PP_STATUS		0xc7200
 #define PCH_PP_CONTROL		0xc7204
@@ -4211,7 +4199,9 @@
 #define GEN6_RP_INTERRUPT_LIMITS		0xA014
 #define GEN6_RPSTAT1				0xA01C
 #define   GEN6_CAGF_SHIFT			8
+#define   HSW_CAGF_SHIFT			7
 #define   GEN6_CAGF_MASK			(0x7f << GEN6_CAGF_SHIFT)
+#define   HSW_CAGF_MASK				(0x7f << HSW_CAGF_SHIFT)
 #define GEN6_RP_CONTROL				0xA024
 #define   GEN6_RP_MEDIA_TURBO			(1<<11)
 #define   GEN6_RP_MEDIA_MODE_MASK		(3<<9)
@@ -4280,8 +4270,8 @@
 #define   GEN6_PCODE_READ_MIN_FREQ_TABLE	0x9
 #define	  GEN6_PCODE_WRITE_RC6VIDS		0x4
 #define	  GEN6_PCODE_READ_RC6VIDS		0x5
-#define   GEN6_ENCODE_RC6_VID(mv)		(((mv) / 5) - 245) < 0 ?: 0
-#define   GEN6_DECODE_RC6_VID(vids)		(((vids) * 5) > 0 ? ((vids) * 5) + 245 : 0)
+#define   GEN6_ENCODE_RC6_VID(mv)		(((mv) - 245) / 5)
+#define   GEN6_DECODE_RC6_VID(vids)		(((vids) * 5) + 245)
 #define GEN6_PCODE_DATA				0x138128
 #define   GEN6_PCODE_FREQ_IA_RATIO_SHIFT	8
 
@@ -4322,7 +4312,7 @@
 #define GEN7_ROW_CHICKEN2_GT2		0xf4f4
 #define   DOP_CLOCK_GATING_DISABLE	(1<<0)
 
-#define G4X_AUD_VID_DID			0x62020
+#define G4X_AUD_VID_DID			(dev_priv->info->display_mmio_offset + 0x62020)
 #define INTEL_AUDIO_DEVCL		0x808629FB
 #define INTEL_AUDIO_DEVBLC		0x80862801
 #define INTEL_AUDIO_DEVCTG		0x80862802
@@ -4438,10 +4428,10 @@
 #define   AUDIO_CP_READY_C		(1<<9)
 
 /* HSW Power Wells */
-#define HSW_PWR_WELL_CTL1			0x45400 /* BIOS */
-#define HSW_PWR_WELL_CTL2			0x45404 /* Driver */
-#define HSW_PWR_WELL_CTL3			0x45408 /* KVMR */
-#define HSW_PWR_WELL_CTL4			0x4540C /* Debug */
+#define HSW_PWR_WELL_BIOS			0x45400 /* CTL1 */
+#define HSW_PWR_WELL_DRIVER			0x45404 /* CTL2 */
+#define HSW_PWR_WELL_KVMR			0x45408 /* CTL3 */
+#define HSW_PWR_WELL_DEBUG			0x4540C /* CTL4 */
 #define   HSW_PWR_WELL_ENABLE			(1<<31)
 #define   HSW_PWR_WELL_STATE			(1<<30)
 #define HSW_PWR_WELL_CTL5			0x45410
@@ -4524,6 +4514,7 @@
 #define  DDI_BUF_EMP_800MV_0DB_HSW		(7<<24)   /* Sel7 */
 #define  DDI_BUF_EMP_800MV_3_5DB_HSW		(8<<24)   /* Sel8 */
 #define  DDI_BUF_EMP_MASK			(0xf<<24)
+#define  DDI_BUF_PORT_REVERSAL			(1<<16)
 #define  DDI_BUF_IS_IDLE			(1<<7)
 #define  DDI_A_4_LANES				(1<<4)
 #define  DDI_PORT_WIDTH_X1			(0<<1)
@@ -4657,4 +4648,51 @@
 #define  WM_DBG_DISALLOW_MAXFIFO	(1<<1)
 #define  WM_DBG_DISALLOW_SPRITE		(1<<2)
 
+/* pipe CSC */
+#define _PIPE_A_CSC_COEFF_RY_GY	0x49010
+#define _PIPE_A_CSC_COEFF_BY	0x49014
+#define _PIPE_A_CSC_COEFF_RU_GU	0x49018
+#define _PIPE_A_CSC_COEFF_BU	0x4901c
+#define _PIPE_A_CSC_COEFF_RV_GV	0x49020
+#define _PIPE_A_CSC_COEFF_BV	0x49024
+#define _PIPE_A_CSC_MODE	0x49028
+#define _PIPE_A_CSC_PREOFF_HI	0x49030
+#define _PIPE_A_CSC_PREOFF_ME	0x49034
+#define _PIPE_A_CSC_PREOFF_LO	0x49038
+#define _PIPE_A_CSC_POSTOFF_HI	0x49040
+#define _PIPE_A_CSC_POSTOFF_ME	0x49044
+#define _PIPE_A_CSC_POSTOFF_LO	0x49048
+
+#define _PIPE_B_CSC_COEFF_RY_GY	0x49110
+#define _PIPE_B_CSC_COEFF_BY	0x49114
+#define _PIPE_B_CSC_COEFF_RU_GU	0x49118
+#define _PIPE_B_CSC_COEFF_BU	0x4911c
+#define _PIPE_B_CSC_COEFF_RV_GV	0x49120
+#define _PIPE_B_CSC_COEFF_BV	0x49124
+#define _PIPE_B_CSC_MODE	0x49128
+#define _PIPE_B_CSC_PREOFF_HI	0x49130
+#define _PIPE_B_CSC_PREOFF_ME	0x49134
+#define _PIPE_B_CSC_PREOFF_LO	0x49138
+#define _PIPE_B_CSC_POSTOFF_HI	0x49140
+#define _PIPE_B_CSC_POSTOFF_ME	0x49144
+#define _PIPE_B_CSC_POSTOFF_LO	0x49148
+
+#define CSC_BLACK_SCREEN_OFFSET (1 << 2)
+#define CSC_POSITION_BEFORE_GAMMA (1 << 1)
+#define CSC_MODE_YUV_TO_RGB (1 << 0)
+
+#define PIPE_CSC_COEFF_RY_GY(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RY_GY, _PIPE_B_CSC_COEFF_RY_GY)
+#define PIPE_CSC_COEFF_BY(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BY, _PIPE_B_CSC_COEFF_BY)
+#define PIPE_CSC_COEFF_RU_GU(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RU_GU, _PIPE_B_CSC_COEFF_RU_GU)
+#define PIPE_CSC_COEFF_BU(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BU, _PIPE_B_CSC_COEFF_BU)
+#define PIPE_CSC_COEFF_RV_GV(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RV_GV, _PIPE_B_CSC_COEFF_RV_GV)
+#define PIPE_CSC_COEFF_BV(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BV, _PIPE_B_CSC_COEFF_BV)
+#define PIPE_CSC_MODE(pipe) _PIPE(pipe, _PIPE_A_CSC_MODE, _PIPE_B_CSC_MODE)
+#define PIPE_CSC_PREOFF_HI(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_HI, _PIPE_B_CSC_PREOFF_HI)
+#define PIPE_CSC_PREOFF_ME(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_ME, _PIPE_B_CSC_PREOFF_ME)
+#define PIPE_CSC_PREOFF_LO(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_LO, _PIPE_B_CSC_PREOFF_LO)
+#define PIPE_CSC_POSTOFF_HI(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_HI, _PIPE_B_CSC_POSTOFF_HI)
+#define PIPE_CSC_POSTOFF_ME(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_ME, _PIPE_B_CSC_POSTOFF_ME)
+#define PIPE_CSC_POSTOFF_LO(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_LO, _PIPE_B_CSC_POSTOFF_LO)
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 63d4d30c39de..2135f21ea458 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -29,67 +29,6 @@
 #include "intel_drv.h"
 #include "i915_reg.h"
 
-static bool i915_pipe_enabled(struct drm_device *dev, enum pipe pipe)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32	dpll_reg;
-
-	/* On IVB, 3rd pipe shares PLL with another one */
-	if (pipe > 1)
-		return false;
-
-	if (HAS_PCH_SPLIT(dev))
-		dpll_reg = _PCH_DPLL(pipe);
-	else
-		dpll_reg = (pipe == PIPE_A) ? _DPLL_A : _DPLL_B;
-
-	return (I915_READ(dpll_reg) & DPLL_VCO_ENABLE);
-}
-
-static void i915_save_palette(struct drm_device *dev, enum pipe pipe)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long reg = (pipe == PIPE_A ? _PALETTE_A : _PALETTE_B);
-	u32 *array;
-	int i;
-
-	if (!i915_pipe_enabled(dev, pipe))
-		return;
-
-	if (HAS_PCH_SPLIT(dev))
-		reg = (pipe == PIPE_A) ? _LGC_PALETTE_A : _LGC_PALETTE_B;
-
-	if (pipe == PIPE_A)
-		array = dev_priv->regfile.save_palette_a;
-	else
-		array = dev_priv->regfile.save_palette_b;
-
-	for (i = 0; i < 256; i++)
-		array[i] = I915_READ(reg + (i << 2));
-}
-
-static void i915_restore_palette(struct drm_device *dev, enum pipe pipe)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long reg = (pipe == PIPE_A ? _PALETTE_A : _PALETTE_B);
-	u32 *array;
-	int i;
-
-	if (!i915_pipe_enabled(dev, pipe))
-		return;
-
-	if (HAS_PCH_SPLIT(dev))
-		reg = (pipe == PIPE_A) ? _LGC_PALETTE_A : _LGC_PALETTE_B;
-
-	if (pipe == PIPE_A)
-		array = dev_priv->regfile.save_palette_a;
-	else
-		array = dev_priv->regfile.save_palette_b;
-
-	for (i = 0; i < 256; i++)
-		I915_WRITE(reg + (i << 2), array[i]);
-}
-
 static u8 i915_read_indexed(struct drm_device *dev, u16 index_port, u16 data_port, u8 reg)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -130,6 +69,12 @@ static void i915_save_vga(struct drm_device *dev)
 	int i;
 	u16 cr_index, cr_data, st01;
 
+	/* VGA state */
+	dev_priv->regfile.saveVGA0 = I915_READ(VGA0);
+	dev_priv->regfile.saveVGA1 = I915_READ(VGA1);
+	dev_priv->regfile.saveVGA_PD = I915_READ(VGA_PD);
+	dev_priv->regfile.saveVGACNTRL = I915_READ(i915_vgacntrl_reg(dev));
+
 	/* VGA color palette registers */
 	dev_priv->regfile.saveDACMASK = I915_READ8(VGA_DACMASK);
 
@@ -188,6 +133,15 @@ static void i915_restore_vga(struct drm_device *dev)
 	int i;
 	u16 cr_index, cr_data, st01;
 
+	/* VGA state */
+	I915_WRITE(i915_vgacntrl_reg(dev), dev_priv->regfile.saveVGACNTRL);
+
+	I915_WRITE(VGA0, dev_priv->regfile.saveVGA0);
+	I915_WRITE(VGA1, dev_priv->regfile.saveVGA1);
+	I915_WRITE(VGA_PD, dev_priv->regfile.saveVGA_PD);
+	POSTING_READ(VGA_PD);
+	udelay(150);
+
 	/* MSR bits */
 	I915_WRITE8(VGA_MSR_WRITE, dev_priv->regfile.saveMSR);
 	if (dev_priv->regfile.saveMSR & VGA_MSR_CGA_MODE) {
@@ -235,396 +189,18 @@ static void i915_restore_vga(struct drm_device *dev)
 	I915_WRITE8(VGA_DACMASK, dev_priv->regfile.saveDACMASK);
 }
 
-static void i915_save_modeset_reg(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i;
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
-	/* Cursor state */
-	dev_priv->regfile.saveCURACNTR = I915_READ(_CURACNTR);
-	dev_priv->regfile.saveCURAPOS = I915_READ(_CURAPOS);
-	dev_priv->regfile.saveCURABASE = I915_READ(_CURABASE);
-	dev_priv->regfile.saveCURBCNTR = I915_READ(_CURBCNTR);
-	dev_priv->regfile.saveCURBPOS = I915_READ(_CURBPOS);
-	dev_priv->regfile.saveCURBBASE = I915_READ(_CURBBASE);
-	if (IS_GEN2(dev))
-		dev_priv->regfile.saveCURSIZE = I915_READ(CURSIZE);
-
-	if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->regfile.savePCH_DREF_CONTROL = I915_READ(PCH_DREF_CONTROL);
-		dev_priv->regfile.saveDISP_ARB_CTL = I915_READ(DISP_ARB_CTL);
-	}
-
-	/* Pipe & plane A info */
-	dev_priv->regfile.savePIPEACONF = I915_READ(_PIPEACONF);
-	dev_priv->regfile.savePIPEASRC = I915_READ(_PIPEASRC);
-	if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->regfile.saveFPA0 = I915_READ(_PCH_FPA0);
-		dev_priv->regfile.saveFPA1 = I915_READ(_PCH_FPA1);
-		dev_priv->regfile.saveDPLL_A = I915_READ(_PCH_DPLL_A);
-	} else {
-		dev_priv->regfile.saveFPA0 = I915_READ(_FPA0);
-		dev_priv->regfile.saveFPA1 = I915_READ(_FPA1);
-		dev_priv->regfile.saveDPLL_A = I915_READ(_DPLL_A);
-	}
-	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveDPLL_A_MD = I915_READ(_DPLL_A_MD);
-	dev_priv->regfile.saveHTOTAL_A = I915_READ(_HTOTAL_A);
-	dev_priv->regfile.saveHBLANK_A = I915_READ(_HBLANK_A);
-	dev_priv->regfile.saveHSYNC_A = I915_READ(_HSYNC_A);
-	dev_priv->regfile.saveVTOTAL_A = I915_READ(_VTOTAL_A);
-	dev_priv->regfile.saveVBLANK_A = I915_READ(_VBLANK_A);
-	dev_priv->regfile.saveVSYNC_A = I915_READ(_VSYNC_A);
-	if (!HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveBCLRPAT_A = I915_READ(_BCLRPAT_A);
-
-	if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->regfile.savePIPEA_DATA_M1 = I915_READ(_PIPEA_DATA_M1);
-		dev_priv->regfile.savePIPEA_DATA_N1 = I915_READ(_PIPEA_DATA_N1);
-		dev_priv->regfile.savePIPEA_LINK_M1 = I915_READ(_PIPEA_LINK_M1);
-		dev_priv->regfile.savePIPEA_LINK_N1 = I915_READ(_PIPEA_LINK_N1);
-
-		dev_priv->regfile.saveFDI_TXA_CTL = I915_READ(_FDI_TXA_CTL);
-		dev_priv->regfile.saveFDI_RXA_CTL = I915_READ(_FDI_RXA_CTL);
-
-		dev_priv->regfile.savePFA_CTL_1 = I915_READ(_PFA_CTL_1);
-		dev_priv->regfile.savePFA_WIN_SZ = I915_READ(_PFA_WIN_SZ);
-		dev_priv->regfile.savePFA_WIN_POS = I915_READ(_PFA_WIN_POS);
-
-		dev_priv->regfile.saveTRANSACONF = I915_READ(_TRANSACONF);
-		dev_priv->regfile.saveTRANS_HTOTAL_A = I915_READ(_TRANS_HTOTAL_A);
-		dev_priv->regfile.saveTRANS_HBLANK_A = I915_READ(_TRANS_HBLANK_A);
-		dev_priv->regfile.saveTRANS_HSYNC_A = I915_READ(_TRANS_HSYNC_A);
-		dev_priv->regfile.saveTRANS_VTOTAL_A = I915_READ(_TRANS_VTOTAL_A);
-		dev_priv->regfile.saveTRANS_VBLANK_A = I915_READ(_TRANS_VBLANK_A);
-		dev_priv->regfile.saveTRANS_VSYNC_A = I915_READ(_TRANS_VSYNC_A);
-	}
-
-	dev_priv->regfile.saveDSPACNTR = I915_READ(_DSPACNTR);
-	dev_priv->regfile.saveDSPASTRIDE = I915_READ(_DSPASTRIDE);
-	dev_priv->regfile.saveDSPASIZE = I915_READ(_DSPASIZE);
-	dev_priv->regfile.saveDSPAPOS = I915_READ(_DSPAPOS);
-	dev_priv->regfile.saveDSPAADDR = I915_READ(_DSPAADDR);
-	if (INTEL_INFO(dev)->gen >= 4) {
-		dev_priv->regfile.saveDSPASURF = I915_READ(_DSPASURF);
-		dev_priv->regfile.saveDSPATILEOFF = I915_READ(_DSPATILEOFF);
-	}
-	i915_save_palette(dev, PIPE_A);
-	dev_priv->regfile.savePIPEASTAT = I915_READ(_PIPEASTAT);
-
-	/* Pipe & plane B info */
-	dev_priv->regfile.savePIPEBCONF = I915_READ(_PIPEBCONF);
-	dev_priv->regfile.savePIPEBSRC = I915_READ(_PIPEBSRC);
-	if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->regfile.saveFPB0 = I915_READ(_PCH_FPB0);
-		dev_priv->regfile.saveFPB1 = I915_READ(_PCH_FPB1);
-		dev_priv->regfile.saveDPLL_B = I915_READ(_PCH_DPLL_B);
-	} else {
-		dev_priv->regfile.saveFPB0 = I915_READ(_FPB0);
-		dev_priv->regfile.saveFPB1 = I915_READ(_FPB1);
-		dev_priv->regfile.saveDPLL_B = I915_READ(_DPLL_B);
-	}
-	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveDPLL_B_MD = I915_READ(_DPLL_B_MD);
-	dev_priv->regfile.saveHTOTAL_B = I915_READ(_HTOTAL_B);
-	dev_priv->regfile.saveHBLANK_B = I915_READ(_HBLANK_B);
-	dev_priv->regfile.saveHSYNC_B = I915_READ(_HSYNC_B);
-	dev_priv->regfile.saveVTOTAL_B = I915_READ(_VTOTAL_B);
-	dev_priv->regfile.saveVBLANK_B = I915_READ(_VBLANK_B);
-	dev_priv->regfile.saveVSYNC_B = I915_READ(_VSYNC_B);
-	if (!HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveBCLRPAT_B = I915_READ(_BCLRPAT_B);
-
-	if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->regfile.savePIPEB_DATA_M1 = I915_READ(_PIPEB_DATA_M1);
-		dev_priv->regfile.savePIPEB_DATA_N1 = I915_READ(_PIPEB_DATA_N1);
-		dev_priv->regfile.savePIPEB_LINK_M1 = I915_READ(_PIPEB_LINK_M1);
-		dev_priv->regfile.savePIPEB_LINK_N1 = I915_READ(_PIPEB_LINK_N1);
-
-		dev_priv->regfile.saveFDI_TXB_CTL = I915_READ(_FDI_TXB_CTL);
-		dev_priv->regfile.saveFDI_RXB_CTL = I915_READ(_FDI_RXB_CTL);
-
-		dev_priv->regfile.savePFB_CTL_1 = I915_READ(_PFB_CTL_1);
-		dev_priv->regfile.savePFB_WIN_SZ = I915_READ(_PFB_WIN_SZ);
-		dev_priv->regfile.savePFB_WIN_POS = I915_READ(_PFB_WIN_POS);
-
-		dev_priv->regfile.saveTRANSBCONF = I915_READ(_TRANSBCONF);
-		dev_priv->regfile.saveTRANS_HTOTAL_B = I915_READ(_TRANS_HTOTAL_B);
-		dev_priv->regfile.saveTRANS_HBLANK_B = I915_READ(_TRANS_HBLANK_B);
-		dev_priv->regfile.saveTRANS_HSYNC_B = I915_READ(_TRANS_HSYNC_B);
-		dev_priv->regfile.saveTRANS_VTOTAL_B = I915_READ(_TRANS_VTOTAL_B);
-		dev_priv->regfile.saveTRANS_VBLANK_B = I915_READ(_TRANS_VBLANK_B);
-		dev_priv->regfile.saveTRANS_VSYNC_B = I915_READ(_TRANS_VSYNC_B);
-	}
-
-	dev_priv->regfile.saveDSPBCNTR = I915_READ(_DSPBCNTR);
-	dev_priv->regfile.saveDSPBSTRIDE = I915_READ(_DSPBSTRIDE);
-	dev_priv->regfile.saveDSPBSIZE = I915_READ(_DSPBSIZE);
-	dev_priv->regfile.saveDSPBPOS = I915_READ(_DSPBPOS);
-	dev_priv->regfile.saveDSPBADDR = I915_READ(_DSPBADDR);
-	if (INTEL_INFO(dev)->gen >= 4) {
-		dev_priv->regfile.saveDSPBSURF = I915_READ(_DSPBSURF);
-		dev_priv->regfile.saveDSPBTILEOFF = I915_READ(_DSPBTILEOFF);
-	}
-	i915_save_palette(dev, PIPE_B);
-	dev_priv->regfile.savePIPEBSTAT = I915_READ(_PIPEBSTAT);
-
-	/* Fences */
-	switch (INTEL_INFO(dev)->gen) {
-	case 7:
-	case 6:
-		for (i = 0; i < 16; i++)
-			dev_priv->regfile.saveFENCE[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
-		break;
-	case 5:
-	case 4:
-		for (i = 0; i < 16; i++)
-			dev_priv->regfile.saveFENCE[i] = I915_READ64(FENCE_REG_965_0 + (i * 8));
-		break;
-	case 3:
-		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
-			for (i = 0; i < 8; i++)
-				dev_priv->regfile.saveFENCE[i+8] = I915_READ(FENCE_REG_945_8 + (i * 4));
-	case 2:
-		for (i = 0; i < 8; i++)
-			dev_priv->regfile.saveFENCE[i] = I915_READ(FENCE_REG_830_0 + (i * 4));
-		break;
-	}
-
-	/* CRT state */
-	if (HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveADPA = I915_READ(PCH_ADPA);
-	else
-		dev_priv->regfile.saveADPA = I915_READ(ADPA);
-
-	return;
-}
-
-static void i915_restore_modeset_reg(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int dpll_a_reg, fpa0_reg, fpa1_reg;
-	int dpll_b_reg, fpb0_reg, fpb1_reg;
-	int i;
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
-	/* Fences */
-	switch (INTEL_INFO(dev)->gen) {
-	case 7:
-	case 6:
-		for (i = 0; i < 16; i++)
-			I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + (i * 8), dev_priv->regfile.saveFENCE[i]);
-		break;
-	case 5:
-	case 4:
-		for (i = 0; i < 16; i++)
-			I915_WRITE64(FENCE_REG_965_0 + (i * 8), dev_priv->regfile.saveFENCE[i]);
-		break;
-	case 3:
-	case 2:
-		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
-			for (i = 0; i < 8; i++)
-				I915_WRITE(FENCE_REG_945_8 + (i * 4), dev_priv->regfile.saveFENCE[i+8]);
-		for (i = 0; i < 8; i++)
-			I915_WRITE(FENCE_REG_830_0 + (i * 4), dev_priv->regfile.saveFENCE[i]);
-		break;
-	}
-
-
-	if (HAS_PCH_SPLIT(dev)) {
-		dpll_a_reg = _PCH_DPLL_A;
-		dpll_b_reg = _PCH_DPLL_B;
-		fpa0_reg = _PCH_FPA0;
-		fpb0_reg = _PCH_FPB0;
-		fpa1_reg = _PCH_FPA1;
-		fpb1_reg = _PCH_FPB1;
-	} else {
-		dpll_a_reg = _DPLL_A;
-		dpll_b_reg = _DPLL_B;
-		fpa0_reg = _FPA0;
-		fpb0_reg = _FPB0;
-		fpa1_reg = _FPA1;
-		fpb1_reg = _FPB1;
-	}
-
-	if (HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(PCH_DREF_CONTROL, dev_priv->regfile.savePCH_DREF_CONTROL);
-		I915_WRITE(DISP_ARB_CTL, dev_priv->regfile.saveDISP_ARB_CTL);
-	}
-
-	/* Pipe & plane A info */
-	/* Prime the clock */
-	if (dev_priv->regfile.saveDPLL_A & DPLL_VCO_ENABLE) {
-		I915_WRITE(dpll_a_reg, dev_priv->regfile.saveDPLL_A &
-			   ~DPLL_VCO_ENABLE);
-		POSTING_READ(dpll_a_reg);
-		udelay(150);
-	}
-	I915_WRITE(fpa0_reg, dev_priv->regfile.saveFPA0);
-	I915_WRITE(fpa1_reg, dev_priv->regfile.saveFPA1);
-	/* Actually enable it */
-	I915_WRITE(dpll_a_reg, dev_priv->regfile.saveDPLL_A);
-	POSTING_READ(dpll_a_reg);
-	udelay(150);
-	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(_DPLL_A_MD, dev_priv->regfile.saveDPLL_A_MD);
-		POSTING_READ(_DPLL_A_MD);
-	}
-	udelay(150);
-
-	/* Restore mode */
-	I915_WRITE(_HTOTAL_A, dev_priv->regfile.saveHTOTAL_A);
-	I915_WRITE(_HBLANK_A, dev_priv->regfile.saveHBLANK_A);
-	I915_WRITE(_HSYNC_A, dev_priv->regfile.saveHSYNC_A);
-	I915_WRITE(_VTOTAL_A, dev_priv->regfile.saveVTOTAL_A);
-	I915_WRITE(_VBLANK_A, dev_priv->regfile.saveVBLANK_A);
-	I915_WRITE(_VSYNC_A, dev_priv->regfile.saveVSYNC_A);
-	if (!HAS_PCH_SPLIT(dev))
-		I915_WRITE(_BCLRPAT_A, dev_priv->regfile.saveBCLRPAT_A);
-
-	if (HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(_PIPEA_DATA_M1, dev_priv->regfile.savePIPEA_DATA_M1);
-		I915_WRITE(_PIPEA_DATA_N1, dev_priv->regfile.savePIPEA_DATA_N1);
-		I915_WRITE(_PIPEA_LINK_M1, dev_priv->regfile.savePIPEA_LINK_M1);
-		I915_WRITE(_PIPEA_LINK_N1, dev_priv->regfile.savePIPEA_LINK_N1);
-
-		I915_WRITE(_FDI_RXA_CTL, dev_priv->regfile.saveFDI_RXA_CTL);
-		I915_WRITE(_FDI_TXA_CTL, dev_priv->regfile.saveFDI_TXA_CTL);
-
-		I915_WRITE(_PFA_CTL_1, dev_priv->regfile.savePFA_CTL_1);
-		I915_WRITE(_PFA_WIN_SZ, dev_priv->regfile.savePFA_WIN_SZ);
-		I915_WRITE(_PFA_WIN_POS, dev_priv->regfile.savePFA_WIN_POS);
-
-		I915_WRITE(_TRANSACONF, dev_priv->regfile.saveTRANSACONF);
-		I915_WRITE(_TRANS_HTOTAL_A, dev_priv->regfile.saveTRANS_HTOTAL_A);
-		I915_WRITE(_TRANS_HBLANK_A, dev_priv->regfile.saveTRANS_HBLANK_A);
-		I915_WRITE(_TRANS_HSYNC_A, dev_priv->regfile.saveTRANS_HSYNC_A);
-		I915_WRITE(_TRANS_VTOTAL_A, dev_priv->regfile.saveTRANS_VTOTAL_A);
-		I915_WRITE(_TRANS_VBLANK_A, dev_priv->regfile.saveTRANS_VBLANK_A);
-		I915_WRITE(_TRANS_VSYNC_A, dev_priv->regfile.saveTRANS_VSYNC_A);
-	}
-
-	/* Restore plane info */
-	I915_WRITE(_DSPASIZE, dev_priv->regfile.saveDSPASIZE);
-	I915_WRITE(_DSPAPOS, dev_priv->regfile.saveDSPAPOS);
-	I915_WRITE(_PIPEASRC, dev_priv->regfile.savePIPEASRC);
-	I915_WRITE(_DSPAADDR, dev_priv->regfile.saveDSPAADDR);
-	I915_WRITE(_DSPASTRIDE, dev_priv->regfile.saveDSPASTRIDE);
-	if (INTEL_INFO(dev)->gen >= 4) {
-		I915_WRITE(_DSPASURF, dev_priv->regfile.saveDSPASURF);
-		I915_WRITE(_DSPATILEOFF, dev_priv->regfile.saveDSPATILEOFF);
-	}
-
-	I915_WRITE(_PIPEACONF, dev_priv->regfile.savePIPEACONF);
-
-	i915_restore_palette(dev, PIPE_A);
-	/* Enable the plane */
-	I915_WRITE(_DSPACNTR, dev_priv->regfile.saveDSPACNTR);
-	I915_WRITE(_DSPAADDR, I915_READ(_DSPAADDR));
-
-	/* Pipe & plane B info */
-	if (dev_priv->regfile.saveDPLL_B & DPLL_VCO_ENABLE) {
-		I915_WRITE(dpll_b_reg, dev_priv->regfile.saveDPLL_B &
-			   ~DPLL_VCO_ENABLE);
-		POSTING_READ(dpll_b_reg);
-		udelay(150);
-	}
-	I915_WRITE(fpb0_reg, dev_priv->regfile.saveFPB0);
-	I915_WRITE(fpb1_reg, dev_priv->regfile.saveFPB1);
-	/* Actually enable it */
-	I915_WRITE(dpll_b_reg, dev_priv->regfile.saveDPLL_B);
-	POSTING_READ(dpll_b_reg);
-	udelay(150);
-	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(_DPLL_B_MD, dev_priv->regfile.saveDPLL_B_MD);
-		POSTING_READ(_DPLL_B_MD);
-	}
-	udelay(150);
-
-	/* Restore mode */
-	I915_WRITE(_HTOTAL_B, dev_priv->regfile.saveHTOTAL_B);
-	I915_WRITE(_HBLANK_B, dev_priv->regfile.saveHBLANK_B);
-	I915_WRITE(_HSYNC_B, dev_priv->regfile.saveHSYNC_B);
-	I915_WRITE(_VTOTAL_B, dev_priv->regfile.saveVTOTAL_B);
-	I915_WRITE(_VBLANK_B, dev_priv->regfile.saveVBLANK_B);
-	I915_WRITE(_VSYNC_B, dev_priv->regfile.saveVSYNC_B);
-	if (!HAS_PCH_SPLIT(dev))
-		I915_WRITE(_BCLRPAT_B, dev_priv->regfile.saveBCLRPAT_B);
-
-	if (HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(_PIPEB_DATA_M1, dev_priv->regfile.savePIPEB_DATA_M1);
-		I915_WRITE(_PIPEB_DATA_N1, dev_priv->regfile.savePIPEB_DATA_N1);
-		I915_WRITE(_PIPEB_LINK_M1, dev_priv->regfile.savePIPEB_LINK_M1);
-		I915_WRITE(_PIPEB_LINK_N1, dev_priv->regfile.savePIPEB_LINK_N1);
-
-		I915_WRITE(_FDI_RXB_CTL, dev_priv->regfile.saveFDI_RXB_CTL);
-		I915_WRITE(_FDI_TXB_CTL, dev_priv->regfile.saveFDI_TXB_CTL);
-
-		I915_WRITE(_PFB_CTL_1, dev_priv->regfile.savePFB_CTL_1);
-		I915_WRITE(_PFB_WIN_SZ, dev_priv->regfile.savePFB_WIN_SZ);
-		I915_WRITE(_PFB_WIN_POS, dev_priv->regfile.savePFB_WIN_POS);
-
-		I915_WRITE(_TRANSBCONF, dev_priv->regfile.saveTRANSBCONF);
-		I915_WRITE(_TRANS_HTOTAL_B, dev_priv->regfile.saveTRANS_HTOTAL_B);
-		I915_WRITE(_TRANS_HBLANK_B, dev_priv->regfile.saveTRANS_HBLANK_B);
-		I915_WRITE(_TRANS_HSYNC_B, dev_priv->regfile.saveTRANS_HSYNC_B);
-		I915_WRITE(_TRANS_VTOTAL_B, dev_priv->regfile.saveTRANS_VTOTAL_B);
-		I915_WRITE(_TRANS_VBLANK_B, dev_priv->regfile.saveTRANS_VBLANK_B);
-		I915_WRITE(_TRANS_VSYNC_B, dev_priv->regfile.saveTRANS_VSYNC_B);
-	}
-
-	/* Restore plane info */
-	I915_WRITE(_DSPBSIZE, dev_priv->regfile.saveDSPBSIZE);
-	I915_WRITE(_DSPBPOS, dev_priv->regfile.saveDSPBPOS);
-	I915_WRITE(_PIPEBSRC, dev_priv->regfile.savePIPEBSRC);
-	I915_WRITE(_DSPBADDR, dev_priv->regfile.saveDSPBADDR);
-	I915_WRITE(_DSPBSTRIDE, dev_priv->regfile.saveDSPBSTRIDE);
-	if (INTEL_INFO(dev)->gen >= 4) {
-		I915_WRITE(_DSPBSURF, dev_priv->regfile.saveDSPBSURF);
-		I915_WRITE(_DSPBTILEOFF, dev_priv->regfile.saveDSPBTILEOFF);
-	}
-
-	I915_WRITE(_PIPEBCONF, dev_priv->regfile.savePIPEBCONF);
-
-	i915_restore_palette(dev, PIPE_B);
-	/* Enable the plane */
-	I915_WRITE(_DSPBCNTR, dev_priv->regfile.saveDSPBCNTR);
-	I915_WRITE(_DSPBADDR, I915_READ(_DSPBADDR));
-
-	/* Cursor state */
-	I915_WRITE(_CURAPOS, dev_priv->regfile.saveCURAPOS);
-	I915_WRITE(_CURACNTR, dev_priv->regfile.saveCURACNTR);
-	I915_WRITE(_CURABASE, dev_priv->regfile.saveCURABASE);
-	I915_WRITE(_CURBPOS, dev_priv->regfile.saveCURBPOS);
-	I915_WRITE(_CURBCNTR, dev_priv->regfile.saveCURBCNTR);
-	I915_WRITE(_CURBBASE, dev_priv->regfile.saveCURBBASE);
-	if (IS_GEN2(dev))
-		I915_WRITE(CURSIZE, dev_priv->regfile.saveCURSIZE);
-
-	/* CRT state */
-	if (HAS_PCH_SPLIT(dev))
-		I915_WRITE(PCH_ADPA, dev_priv->regfile.saveADPA);
-	else
-		I915_WRITE(ADPA, dev_priv->regfile.saveADPA);
-
-	return;
-}
-
 static void i915_save_display(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	/* Display arbitration control */
-	dev_priv->regfile.saveDSPARB = I915_READ(DSPARB);
+	if (INTEL_INFO(dev)->gen <= 4)
+		dev_priv->regfile.saveDSPARB = I915_READ(DSPARB);
 
 	/* This is only meaningful in non-KMS mode */
 	/* Don't regfile.save them in KMS mode */
-	i915_save_modeset_reg(dev);
+	if (!drm_core_check_feature(dev, DRIVER_MODESET))
+		i915_save_display_reg(dev);
 
 	/* LVDS state */
 	if (HAS_PCH_SPLIT(dev)) {
@@ -658,24 +234,6 @@ static void i915_save_display(struct drm_device *dev)
 		dev_priv->regfile.savePP_DIVISOR = I915_READ(PP_DIVISOR);
 	}
 
-	if (!drm_core_check_feature(dev, DRIVER_MODESET)) {
-		/* Display Port state */
-		if (SUPPORTS_INTEGRATED_DP(dev)) {
-			dev_priv->regfile.saveDP_B = I915_READ(DP_B);
-			dev_priv->regfile.saveDP_C = I915_READ(DP_C);
-			dev_priv->regfile.saveDP_D = I915_READ(DP_D);
-			dev_priv->regfile.savePIPEA_GMCH_DATA_M = I915_READ(_PIPEA_GMCH_DATA_M);
-			dev_priv->regfile.savePIPEB_GMCH_DATA_M = I915_READ(_PIPEB_GMCH_DATA_M);
-			dev_priv->regfile.savePIPEA_GMCH_DATA_N = I915_READ(_PIPEA_GMCH_DATA_N);
-			dev_priv->regfile.savePIPEB_GMCH_DATA_N = I915_READ(_PIPEB_GMCH_DATA_N);
-			dev_priv->regfile.savePIPEA_DP_LINK_M = I915_READ(_PIPEA_DP_LINK_M);
-			dev_priv->regfile.savePIPEB_DP_LINK_M = I915_READ(_PIPEB_DP_LINK_M);
-			dev_priv->regfile.savePIPEA_DP_LINK_N = I915_READ(_PIPEA_DP_LINK_N);
-			dev_priv->regfile.savePIPEB_DP_LINK_N = I915_READ(_PIPEB_DP_LINK_N);
-		}
-		/* FIXME: regfile.save TV & SDVO state */
-	}
-
 	/* Only regfile.save FBC state on the platform that supports FBC */
 	if (I915_HAS_FBC(dev)) {
 		if (HAS_PCH_SPLIT(dev)) {
@@ -690,16 +248,8 @@ static void i915_save_display(struct drm_device *dev)
 		}
 	}
 
-	/* VGA state */
-	dev_priv->regfile.saveVGA0 = I915_READ(VGA0);
-	dev_priv->regfile.saveVGA1 = I915_READ(VGA1);
-	dev_priv->regfile.saveVGA_PD = I915_READ(VGA_PD);
-	if (HAS_PCH_SPLIT(dev))
-		dev_priv->regfile.saveVGACNTRL = I915_READ(CPU_VGACNTRL);
-	else
-		dev_priv->regfile.saveVGACNTRL = I915_READ(VGACNTRL);
-
-	i915_save_vga(dev);
+	if (!drm_core_check_feature(dev, DRIVER_MODESET))
+		i915_save_vga(dev);
 }
 
 static void i915_restore_display(struct drm_device *dev)
@@ -707,25 +257,11 @@ static void i915_restore_display(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	/* Display arbitration */
-	I915_WRITE(DSPARB, dev_priv->regfile.saveDSPARB);
+	if (INTEL_INFO(dev)->gen <= 4)
+		I915_WRITE(DSPARB, dev_priv->regfile.saveDSPARB);
 
-	if (!drm_core_check_feature(dev, DRIVER_MODESET)) {
-		/* Display port ratios (must be done before clock is set) */
-		if (SUPPORTS_INTEGRATED_DP(dev)) {
-			I915_WRITE(_PIPEA_GMCH_DATA_M, dev_priv->regfile.savePIPEA_GMCH_DATA_M);
-			I915_WRITE(_PIPEB_GMCH_DATA_M, dev_priv->regfile.savePIPEB_GMCH_DATA_M);
-			I915_WRITE(_PIPEA_GMCH_DATA_N, dev_priv->regfile.savePIPEA_GMCH_DATA_N);
-			I915_WRITE(_PIPEB_GMCH_DATA_N, dev_priv->regfile.savePIPEB_GMCH_DATA_N);
-			I915_WRITE(_PIPEA_DP_LINK_M, dev_priv->regfile.savePIPEA_DP_LINK_M);
-			I915_WRITE(_PIPEB_DP_LINK_M, dev_priv->regfile.savePIPEB_DP_LINK_M);
-			I915_WRITE(_PIPEA_DP_LINK_N, dev_priv->regfile.savePIPEA_DP_LINK_N);
-			I915_WRITE(_PIPEB_DP_LINK_N, dev_priv->regfile.savePIPEB_DP_LINK_N);
-		}
-	}
-
-	/* This is only meaningful in non-KMS mode */
-	/* Don't restore them in KMS mode */
-	i915_restore_modeset_reg(dev);
+	if (!drm_core_check_feature(dev, DRIVER_MODESET))
+		i915_restore_display_reg(dev);
 
 	/* LVDS state */
 	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev))
@@ -763,16 +299,6 @@ static void i915_restore_display(struct drm_device *dev)
 		I915_WRITE(PP_CONTROL, dev_priv->regfile.savePP_CONTROL);
 	}
 
-	if (!drm_core_check_feature(dev, DRIVER_MODESET)) {
-		/* Display Port state */
-		if (SUPPORTS_INTEGRATED_DP(dev)) {
-			I915_WRITE(DP_B, dev_priv->regfile.saveDP_B);
-			I915_WRITE(DP_C, dev_priv->regfile.saveDP_C);
-			I915_WRITE(DP_D, dev_priv->regfile.saveDP_D);
-		}
-		/* FIXME: restore TV & SDVO state */
-	}
-
 	/* only restore FBC info on the platform that supports FBC*/
 	intel_disable_fbc(dev);
 	if (I915_HAS_FBC(dev)) {
@@ -787,19 +313,11 @@ static void i915_restore_display(struct drm_device *dev)
 			I915_WRITE(FBC_CONTROL, dev_priv->regfile.saveFBC_CONTROL);
 		}
 	}
-	/* VGA state */
-	if (HAS_PCH_SPLIT(dev))
-		I915_WRITE(CPU_VGACNTRL, dev_priv->regfile.saveVGACNTRL);
-	else
-		I915_WRITE(VGACNTRL, dev_priv->regfile.saveVGACNTRL);
 
-	I915_WRITE(VGA0, dev_priv->regfile.saveVGA0);
-	I915_WRITE(VGA1, dev_priv->regfile.saveVGA1);
-	I915_WRITE(VGA_PD, dev_priv->regfile.saveVGA_PD);
-	POSTING_READ(VGA_PD);
-	udelay(150);
-
-	i915_restore_vga(dev);
+	if (!drm_core_check_feature(dev, DRIVER_MODESET))
+		i915_restore_vga(dev);
+	else
+		i915_redisable_vga(dev);
 }
 
 int i915_save_state(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_ums.c b/drivers/gpu/drm/i915/i915_ums.c
new file mode 100644
index 000000000000..985a09716237
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_ums.c
@@ -0,0 +1,503 @@
+/*
+ *
+ * Copyright 2008 (c) Intel Corporation
+ *   Jesse Barnes <jbarnes@virtuousgeek.org>
+ * Copyright 2013 (c) Intel Corporation
+ *   Daniel Vetter <daniel.vetter@ffwll.ch>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sub license, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial portions
+ * of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
+ * IN NO EVENT SHALL TUNGSTEN GRAPHICS AND/OR ITS SUPPLIERS BE LIABLE FOR
+ * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+ * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+ * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <drm/drmP.h>
+#include <drm/i915_drm.h>
+#include "intel_drv.h"
+#include "i915_reg.h"
+
+static bool i915_pipe_enabled(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32	dpll_reg;
+
+	/* On IVB, 3rd pipe shares PLL with another one */
+	if (pipe > 1)
+		return false;
+
+	if (HAS_PCH_SPLIT(dev))
+		dpll_reg = _PCH_DPLL(pipe);
+	else
+		dpll_reg = (pipe == PIPE_A) ? _DPLL_A : _DPLL_B;
+
+	return (I915_READ(dpll_reg) & DPLL_VCO_ENABLE);
+}
+
+static void i915_save_palette(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long reg = (pipe == PIPE_A ? _PALETTE_A : _PALETTE_B);
+	u32 *array;
+	int i;
+
+	if (!i915_pipe_enabled(dev, pipe))
+		return;
+
+	if (HAS_PCH_SPLIT(dev))
+		reg = (pipe == PIPE_A) ? _LGC_PALETTE_A : _LGC_PALETTE_B;
+
+	if (pipe == PIPE_A)
+		array = dev_priv->regfile.save_palette_a;
+	else
+		array = dev_priv->regfile.save_palette_b;
+
+	for (i = 0; i < 256; i++)
+		array[i] = I915_READ(reg + (i << 2));
+}
+
+static void i915_restore_palette(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long reg = (pipe == PIPE_A ? _PALETTE_A : _PALETTE_B);
+	u32 *array;
+	int i;
+
+	if (!i915_pipe_enabled(dev, pipe))
+		return;
+
+	if (HAS_PCH_SPLIT(dev))
+		reg = (pipe == PIPE_A) ? _LGC_PALETTE_A : _LGC_PALETTE_B;
+
+	if (pipe == PIPE_A)
+		array = dev_priv->regfile.save_palette_a;
+	else
+		array = dev_priv->regfile.save_palette_b;
+
+	for (i = 0; i < 256; i++)
+		I915_WRITE(reg + (i << 2), array[i]);
+}
+
+void i915_save_display_reg(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int i;
+
+	/* Cursor state */
+	dev_priv->regfile.saveCURACNTR = I915_READ(_CURACNTR);
+	dev_priv->regfile.saveCURAPOS = I915_READ(_CURAPOS);
+	dev_priv->regfile.saveCURABASE = I915_READ(_CURABASE);
+	dev_priv->regfile.saveCURBCNTR = I915_READ(_CURBCNTR);
+	dev_priv->regfile.saveCURBPOS = I915_READ(_CURBPOS);
+	dev_priv->regfile.saveCURBBASE = I915_READ(_CURBBASE);
+	if (IS_GEN2(dev))
+		dev_priv->regfile.saveCURSIZE = I915_READ(CURSIZE);
+
+	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->regfile.savePCH_DREF_CONTROL = I915_READ(PCH_DREF_CONTROL);
+		dev_priv->regfile.saveDISP_ARB_CTL = I915_READ(DISP_ARB_CTL);
+	}
+
+	/* Pipe & plane A info */
+	dev_priv->regfile.savePIPEACONF = I915_READ(_PIPEACONF);
+	dev_priv->regfile.savePIPEASRC = I915_READ(_PIPEASRC);
+	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->regfile.saveFPA0 = I915_READ(_PCH_FPA0);
+		dev_priv->regfile.saveFPA1 = I915_READ(_PCH_FPA1);
+		dev_priv->regfile.saveDPLL_A = I915_READ(_PCH_DPLL_A);
+	} else {
+		dev_priv->regfile.saveFPA0 = I915_READ(_FPA0);
+		dev_priv->regfile.saveFPA1 = I915_READ(_FPA1);
+		dev_priv->regfile.saveDPLL_A = I915_READ(_DPLL_A);
+	}
+	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev))
+		dev_priv->regfile.saveDPLL_A_MD = I915_READ(_DPLL_A_MD);
+	dev_priv->regfile.saveHTOTAL_A = I915_READ(_HTOTAL_A);
+	dev_priv->regfile.saveHBLANK_A = I915_READ(_HBLANK_A);
+	dev_priv->regfile.saveHSYNC_A = I915_READ(_HSYNC_A);
+	dev_priv->regfile.saveVTOTAL_A = I915_READ(_VTOTAL_A);
+	dev_priv->regfile.saveVBLANK_A = I915_READ(_VBLANK_A);
+	dev_priv->regfile.saveVSYNC_A = I915_READ(_VSYNC_A);
+	if (!HAS_PCH_SPLIT(dev))
+		dev_priv->regfile.saveBCLRPAT_A = I915_READ(_BCLRPAT_A);
+
+	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->regfile.savePIPEA_DATA_M1 = I915_READ(_PIPEA_DATA_M1);
+		dev_priv->regfile.savePIPEA_DATA_N1 = I915_READ(_PIPEA_DATA_N1);
+		dev_priv->regfile.savePIPEA_LINK_M1 = I915_READ(_PIPEA_LINK_M1);
+		dev_priv->regfile.savePIPEA_LINK_N1 = I915_READ(_PIPEA_LINK_N1);
+
+		dev_priv->regfile.saveFDI_TXA_CTL = I915_READ(_FDI_TXA_CTL);
+		dev_priv->regfile.saveFDI_RXA_CTL = I915_READ(_FDI_RXA_CTL);
+
+		dev_priv->regfile.savePFA_CTL_1 = I915_READ(_PFA_CTL_1);
+		dev_priv->regfile.savePFA_WIN_SZ = I915_READ(_PFA_WIN_SZ);
+		dev_priv->regfile.savePFA_WIN_POS = I915_READ(_PFA_WIN_POS);
+
+		dev_priv->regfile.saveTRANSACONF = I915_READ(_TRANSACONF);
+		dev_priv->regfile.saveTRANS_HTOTAL_A = I915_READ(_TRANS_HTOTAL_A);
+		dev_priv->regfile.saveTRANS_HBLANK_A = I915_READ(_TRANS_HBLANK_A);
+		dev_priv->regfile.saveTRANS_HSYNC_A = I915_READ(_TRANS_HSYNC_A);
+		dev_priv->regfile.saveTRANS_VTOTAL_A = I915_READ(_TRANS_VTOTAL_A);
+		dev_priv->regfile.saveTRANS_VBLANK_A = I915_READ(_TRANS_VBLANK_A);
+		dev_priv->regfile.saveTRANS_VSYNC_A = I915_READ(_TRANS_VSYNC_A);
+	}
+
+	dev_priv->regfile.saveDSPACNTR = I915_READ(_DSPACNTR);
+	dev_priv->regfile.saveDSPASTRIDE = I915_READ(_DSPASTRIDE);
+	dev_priv->regfile.saveDSPASIZE = I915_READ(_DSPASIZE);
+	dev_priv->regfile.saveDSPAPOS = I915_READ(_DSPAPOS);
+	dev_priv->regfile.saveDSPAADDR = I915_READ(_DSPAADDR);
+	if (INTEL_INFO(dev)->gen >= 4) {
+		dev_priv->regfile.saveDSPASURF = I915_READ(_DSPASURF);
+		dev_priv->regfile.saveDSPATILEOFF = I915_READ(_DSPATILEOFF);
+	}
+	i915_save_palette(dev, PIPE_A);
+	dev_priv->regfile.savePIPEASTAT = I915_READ(_PIPEASTAT);
+
+	/* Pipe & plane B info */
+	dev_priv->regfile.savePIPEBCONF = I915_READ(_PIPEBCONF);
+	dev_priv->regfile.savePIPEBSRC = I915_READ(_PIPEBSRC);
+	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->regfile.saveFPB0 = I915_READ(_PCH_FPB0);
+		dev_priv->regfile.saveFPB1 = I915_READ(_PCH_FPB1);
+		dev_priv->regfile.saveDPLL_B = I915_READ(_PCH_DPLL_B);
+	} else {
+		dev_priv->regfile.saveFPB0 = I915_READ(_FPB0);
+		dev_priv->regfile.saveFPB1 = I915_READ(_FPB1);
+		dev_priv->regfile.saveDPLL_B = I915_READ(_DPLL_B);
+	}
+	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev))
+		dev_priv->regfile.saveDPLL_B_MD = I915_READ(_DPLL_B_MD);
+	dev_priv->regfile.saveHTOTAL_B = I915_READ(_HTOTAL_B);
+	dev_priv->regfile.saveHBLANK_B = I915_READ(_HBLANK_B);
+	dev_priv->regfile.saveHSYNC_B = I915_READ(_HSYNC_B);
+	dev_priv->regfile.saveVTOTAL_B = I915_READ(_VTOTAL_B);
+	dev_priv->regfile.saveVBLANK_B = I915_READ(_VBLANK_B);
+	dev_priv->regfile.saveVSYNC_B = I915_READ(_VSYNC_B);
+	if (!HAS_PCH_SPLIT(dev))
+		dev_priv->regfile.saveBCLRPAT_B = I915_READ(_BCLRPAT_B);
+
+	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->regfile.savePIPEB_DATA_M1 = I915_READ(_PIPEB_DATA_M1);
+		dev_priv->regfile.savePIPEB_DATA_N1 = I915_READ(_PIPEB_DATA_N1);
+		dev_priv->regfile.savePIPEB_LINK_M1 = I915_READ(_PIPEB_LINK_M1);
+		dev_priv->regfile.savePIPEB_LINK_N1 = I915_READ(_PIPEB_LINK_N1);
+
+		dev_priv->regfile.saveFDI_TXB_CTL = I915_READ(_FDI_TXB_CTL);
+		dev_priv->regfile.saveFDI_RXB_CTL = I915_READ(_FDI_RXB_CTL);
+
+		dev_priv->regfile.savePFB_CTL_1 = I915_READ(_PFB_CTL_1);
+		dev_priv->regfile.savePFB_WIN_SZ = I915_READ(_PFB_WIN_SZ);
+		dev_priv->regfile.savePFB_WIN_POS = I915_READ(_PFB_WIN_POS);
+
+		dev_priv->regfile.saveTRANSBCONF = I915_READ(_TRANSBCONF);
+		dev_priv->regfile.saveTRANS_HTOTAL_B = I915_READ(_TRANS_HTOTAL_B);
+		dev_priv->regfile.saveTRANS_HBLANK_B = I915_READ(_TRANS_HBLANK_B);
+		dev_priv->regfile.saveTRANS_HSYNC_B = I915_READ(_TRANS_HSYNC_B);
+		dev_priv->regfile.saveTRANS_VTOTAL_B = I915_READ(_TRANS_VTOTAL_B);
+		dev_priv->regfile.saveTRANS_VBLANK_B = I915_READ(_TRANS_VBLANK_B);
+		dev_priv->regfile.saveTRANS_VSYNC_B = I915_READ(_TRANS_VSYNC_B);
+	}
+
+	dev_priv->regfile.saveDSPBCNTR = I915_READ(_DSPBCNTR);
+	dev_priv->regfile.saveDSPBSTRIDE = I915_READ(_DSPBSTRIDE);
+	dev_priv->regfile.saveDSPBSIZE = I915_READ(_DSPBSIZE);
+	dev_priv->regfile.saveDSPBPOS = I915_READ(_DSPBPOS);
+	dev_priv->regfile.saveDSPBADDR = I915_READ(_DSPBADDR);
+	if (INTEL_INFO(dev)->gen >= 4) {
+		dev_priv->regfile.saveDSPBSURF = I915_READ(_DSPBSURF);
+		dev_priv->regfile.saveDSPBTILEOFF = I915_READ(_DSPBTILEOFF);
+	}
+	i915_save_palette(dev, PIPE_B);
+	dev_priv->regfile.savePIPEBSTAT = I915_READ(_PIPEBSTAT);
+
+	/* Fences */
+	switch (INTEL_INFO(dev)->gen) {
+	case 7:
+	case 6:
+		for (i = 0; i < 16; i++)
+			dev_priv->regfile.saveFENCE[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
+		break;
+	case 5:
+	case 4:
+		for (i = 0; i < 16; i++)
+			dev_priv->regfile.saveFENCE[i] = I915_READ64(FENCE_REG_965_0 + (i * 8));
+		break;
+	case 3:
+		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
+			for (i = 0; i < 8; i++)
+				dev_priv->regfile.saveFENCE[i+8] = I915_READ(FENCE_REG_945_8 + (i * 4));
+	case 2:
+		for (i = 0; i < 8; i++)
+			dev_priv->regfile.saveFENCE[i] = I915_READ(FENCE_REG_830_0 + (i * 4));
+		break;
+	}
+
+	/* CRT state */
+	if (HAS_PCH_SPLIT(dev))
+		dev_priv->regfile.saveADPA = I915_READ(PCH_ADPA);
+	else
+		dev_priv->regfile.saveADPA = I915_READ(ADPA);
+
+	/* Display Port state */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		dev_priv->regfile.saveDP_B = I915_READ(DP_B);
+		dev_priv->regfile.saveDP_C = I915_READ(DP_C);
+		dev_priv->regfile.saveDP_D = I915_READ(DP_D);
+		dev_priv->regfile.savePIPEA_GMCH_DATA_M = I915_READ(_PIPEA_GMCH_DATA_M);
+		dev_priv->regfile.savePIPEB_GMCH_DATA_M = I915_READ(_PIPEB_GMCH_DATA_M);
+		dev_priv->regfile.savePIPEA_GMCH_DATA_N = I915_READ(_PIPEA_GMCH_DATA_N);
+		dev_priv->regfile.savePIPEB_GMCH_DATA_N = I915_READ(_PIPEB_GMCH_DATA_N);
+		dev_priv->regfile.savePIPEA_DP_LINK_M = I915_READ(_PIPEA_DP_LINK_M);
+		dev_priv->regfile.savePIPEB_DP_LINK_M = I915_READ(_PIPEB_DP_LINK_M);
+		dev_priv->regfile.savePIPEA_DP_LINK_N = I915_READ(_PIPEA_DP_LINK_N);
+		dev_priv->regfile.savePIPEB_DP_LINK_N = I915_READ(_PIPEB_DP_LINK_N);
+	}
+	/* FIXME: regfile.save TV & SDVO state */
+
+	return;
+}
+
+void i915_restore_display_reg(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int dpll_a_reg, fpa0_reg, fpa1_reg;
+	int dpll_b_reg, fpb0_reg, fpb1_reg;
+	int i;
+
+	/* Display port ratios (must be done before clock is set) */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		I915_WRITE(_PIPEA_GMCH_DATA_M, dev_priv->regfile.savePIPEA_GMCH_DATA_M);
+		I915_WRITE(_PIPEB_GMCH_DATA_M, dev_priv->regfile.savePIPEB_GMCH_DATA_M);
+		I915_WRITE(_PIPEA_GMCH_DATA_N, dev_priv->regfile.savePIPEA_GMCH_DATA_N);
+		I915_WRITE(_PIPEB_GMCH_DATA_N, dev_priv->regfile.savePIPEB_GMCH_DATA_N);
+		I915_WRITE(_PIPEA_DP_LINK_M, dev_priv->regfile.savePIPEA_DP_LINK_M);
+		I915_WRITE(_PIPEB_DP_LINK_M, dev_priv->regfile.savePIPEB_DP_LINK_M);
+		I915_WRITE(_PIPEA_DP_LINK_N, dev_priv->regfile.savePIPEA_DP_LINK_N);
+		I915_WRITE(_PIPEB_DP_LINK_N, dev_priv->regfile.savePIPEB_DP_LINK_N);
+	}
+
+	/* Fences */
+	switch (INTEL_INFO(dev)->gen) {
+	case 7:
+	case 6:
+		for (i = 0; i < 16; i++)
+			I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + (i * 8), dev_priv->regfile.saveFENCE[i]);
+		break;
+	case 5:
+	case 4:
+		for (i = 0; i < 16; i++)
+			I915_WRITE64(FENCE_REG_965_0 + (i * 8), dev_priv->regfile.saveFENCE[i]);
+		break;
+	case 3:
+	case 2:
+		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
+			for (i = 0; i < 8; i++)
+				I915_WRITE(FENCE_REG_945_8 + (i * 4), dev_priv->regfile.saveFENCE[i+8]);
+		for (i = 0; i < 8; i++)
+			I915_WRITE(FENCE_REG_830_0 + (i * 4), dev_priv->regfile.saveFENCE[i]);
+		break;
+	}
+
+
+	if (HAS_PCH_SPLIT(dev)) {
+		dpll_a_reg = _PCH_DPLL_A;
+		dpll_b_reg = _PCH_DPLL_B;
+		fpa0_reg = _PCH_FPA0;
+		fpb0_reg = _PCH_FPB0;
+		fpa1_reg = _PCH_FPA1;
+		fpb1_reg = _PCH_FPB1;
+	} else {
+		dpll_a_reg = _DPLL_A;
+		dpll_b_reg = _DPLL_B;
+		fpa0_reg = _FPA0;
+		fpb0_reg = _FPB0;
+		fpa1_reg = _FPA1;
+		fpb1_reg = _FPB1;
+	}
+
+	if (HAS_PCH_SPLIT(dev)) {
+		I915_WRITE(PCH_DREF_CONTROL, dev_priv->regfile.savePCH_DREF_CONTROL);
+		I915_WRITE(DISP_ARB_CTL, dev_priv->regfile.saveDISP_ARB_CTL);
+	}
+
+	/* Pipe & plane A info */
+	/* Prime the clock */
+	if (dev_priv->regfile.saveDPLL_A & DPLL_VCO_ENABLE) {
+		I915_WRITE(dpll_a_reg, dev_priv->regfile.saveDPLL_A &
+			   ~DPLL_VCO_ENABLE);
+		POSTING_READ(dpll_a_reg);
+		udelay(150);
+	}
+	I915_WRITE(fpa0_reg, dev_priv->regfile.saveFPA0);
+	I915_WRITE(fpa1_reg, dev_priv->regfile.saveFPA1);
+	/* Actually enable it */
+	I915_WRITE(dpll_a_reg, dev_priv->regfile.saveDPLL_A);
+	POSTING_READ(dpll_a_reg);
+	udelay(150);
+	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev)) {
+		I915_WRITE(_DPLL_A_MD, dev_priv->regfile.saveDPLL_A_MD);
+		POSTING_READ(_DPLL_A_MD);
+	}
+	udelay(150);
+
+	/* Restore mode */
+	I915_WRITE(_HTOTAL_A, dev_priv->regfile.saveHTOTAL_A);
+	I915_WRITE(_HBLANK_A, dev_priv->regfile.saveHBLANK_A);
+	I915_WRITE(_HSYNC_A, dev_priv->regfile.saveHSYNC_A);
+	I915_WRITE(_VTOTAL_A, dev_priv->regfile.saveVTOTAL_A);
+	I915_WRITE(_VBLANK_A, dev_priv->regfile.saveVBLANK_A);
+	I915_WRITE(_VSYNC_A, dev_priv->regfile.saveVSYNC_A);
+	if (!HAS_PCH_SPLIT(dev))
+		I915_WRITE(_BCLRPAT_A, dev_priv->regfile.saveBCLRPAT_A);
+
+	if (HAS_PCH_SPLIT(dev)) {
+		I915_WRITE(_PIPEA_DATA_M1, dev_priv->regfile.savePIPEA_DATA_M1);
+		I915_WRITE(_PIPEA_DATA_N1, dev_priv->regfile.savePIPEA_DATA_N1);
+		I915_WRITE(_PIPEA_LINK_M1, dev_priv->regfile.savePIPEA_LINK_M1);
+		I915_WRITE(_PIPEA_LINK_N1, dev_priv->regfile.savePIPEA_LINK_N1);
+
+		I915_WRITE(_FDI_RXA_CTL, dev_priv->regfile.saveFDI_RXA_CTL);
+		I915_WRITE(_FDI_TXA_CTL, dev_priv->regfile.saveFDI_TXA_CTL);
+
+		I915_WRITE(_PFA_CTL_1, dev_priv->regfile.savePFA_CTL_1);
+		I915_WRITE(_PFA_WIN_SZ, dev_priv->regfile.savePFA_WIN_SZ);
+		I915_WRITE(_PFA_WIN_POS, dev_priv->regfile.savePFA_WIN_POS);
+
+		I915_WRITE(_TRANSACONF, dev_priv->regfile.saveTRANSACONF);
+		I915_WRITE(_TRANS_HTOTAL_A, dev_priv->regfile.saveTRANS_HTOTAL_A);
+		I915_WRITE(_TRANS_HBLANK_A, dev_priv->regfile.saveTRANS_HBLANK_A);
+		I915_WRITE(_TRANS_HSYNC_A, dev_priv->regfile.saveTRANS_HSYNC_A);
+		I915_WRITE(_TRANS_VTOTAL_A, dev_priv->regfile.saveTRANS_VTOTAL_A);
+		I915_WRITE(_TRANS_VBLANK_A, dev_priv->regfile.saveTRANS_VBLANK_A);
+		I915_WRITE(_TRANS_VSYNC_A, dev_priv->regfile.saveTRANS_VSYNC_A);
+	}
+
+	/* Restore plane info */
+	I915_WRITE(_DSPASIZE, dev_priv->regfile.saveDSPASIZE);
+	I915_WRITE(_DSPAPOS, dev_priv->regfile.saveDSPAPOS);
+	I915_WRITE(_PIPEASRC, dev_priv->regfile.savePIPEASRC);
+	I915_WRITE(_DSPAADDR, dev_priv->regfile.saveDSPAADDR);
+	I915_WRITE(_DSPASTRIDE, dev_priv->regfile.saveDSPASTRIDE);
+	if (INTEL_INFO(dev)->gen >= 4) {
+		I915_WRITE(_DSPASURF, dev_priv->regfile.saveDSPASURF);
+		I915_WRITE(_DSPATILEOFF, dev_priv->regfile.saveDSPATILEOFF);
+	}
+
+	I915_WRITE(_PIPEACONF, dev_priv->regfile.savePIPEACONF);
+
+	i915_restore_palette(dev, PIPE_A);
+	/* Enable the plane */
+	I915_WRITE(_DSPACNTR, dev_priv->regfile.saveDSPACNTR);
+	I915_WRITE(_DSPAADDR, I915_READ(_DSPAADDR));
+
+	/* Pipe & plane B info */
+	if (dev_priv->regfile.saveDPLL_B & DPLL_VCO_ENABLE) {
+		I915_WRITE(dpll_b_reg, dev_priv->regfile.saveDPLL_B &
+			   ~DPLL_VCO_ENABLE);
+		POSTING_READ(dpll_b_reg);
+		udelay(150);
+	}
+	I915_WRITE(fpb0_reg, dev_priv->regfile.saveFPB0);
+	I915_WRITE(fpb1_reg, dev_priv->regfile.saveFPB1);
+	/* Actually enable it */
+	I915_WRITE(dpll_b_reg, dev_priv->regfile.saveDPLL_B);
+	POSTING_READ(dpll_b_reg);
+	udelay(150);
+	if (INTEL_INFO(dev)->gen >= 4 && !HAS_PCH_SPLIT(dev)) {
+		I915_WRITE(_DPLL_B_MD, dev_priv->regfile.saveDPLL_B_MD);
+		POSTING_READ(_DPLL_B_MD);
+	}
+	udelay(150);
+
+	/* Restore mode */
+	I915_WRITE(_HTOTAL_B, dev_priv->regfile.saveHTOTAL_B);
+	I915_WRITE(_HBLANK_B, dev_priv->regfile.saveHBLANK_B);
+	I915_WRITE(_HSYNC_B, dev_priv->regfile.saveHSYNC_B);
+	I915_WRITE(_VTOTAL_B, dev_priv->regfile.saveVTOTAL_B);
+	I915_WRITE(_VBLANK_B, dev_priv->regfile.saveVBLANK_B);
+	I915_WRITE(_VSYNC_B, dev_priv->regfile.saveVSYNC_B);
+	if (!HAS_PCH_SPLIT(dev))
+		I915_WRITE(_BCLRPAT_B, dev_priv->regfile.saveBCLRPAT_B);
+
+	if (HAS_PCH_SPLIT(dev)) {
+		I915_WRITE(_PIPEB_DATA_M1, dev_priv->regfile.savePIPEB_DATA_M1);
+		I915_WRITE(_PIPEB_DATA_N1, dev_priv->regfile.savePIPEB_DATA_N1);
+		I915_WRITE(_PIPEB_LINK_M1, dev_priv->regfile.savePIPEB_LINK_M1);
+		I915_WRITE(_PIPEB_LINK_N1, dev_priv->regfile.savePIPEB_LINK_N1);
+
+		I915_WRITE(_FDI_RXB_CTL, dev_priv->regfile.saveFDI_RXB_CTL);
+		I915_WRITE(_FDI_TXB_CTL, dev_priv->regfile.saveFDI_TXB_CTL);
+
+		I915_WRITE(_PFB_CTL_1, dev_priv->regfile.savePFB_CTL_1);
+		I915_WRITE(_PFB_WIN_SZ, dev_priv->regfile.savePFB_WIN_SZ);
+		I915_WRITE(_PFB_WIN_POS, dev_priv->regfile.savePFB_WIN_POS);
+
+		I915_WRITE(_TRANSBCONF, dev_priv->regfile.saveTRANSBCONF);
+		I915_WRITE(_TRANS_HTOTAL_B, dev_priv->regfile.saveTRANS_HTOTAL_B);
+		I915_WRITE(_TRANS_HBLANK_B, dev_priv->regfile.saveTRANS_HBLANK_B);
+		I915_WRITE(_TRANS_HSYNC_B, dev_priv->regfile.saveTRANS_HSYNC_B);
+		I915_WRITE(_TRANS_VTOTAL_B, dev_priv->regfile.saveTRANS_VTOTAL_B);
+		I915_WRITE(_TRANS_VBLANK_B, dev_priv->regfile.saveTRANS_VBLANK_B);
+		I915_WRITE(_TRANS_VSYNC_B, dev_priv->regfile.saveTRANS_VSYNC_B);
+	}
+
+	/* Restore plane info */
+	I915_WRITE(_DSPBSIZE, dev_priv->regfile.saveDSPBSIZE);
+	I915_WRITE(_DSPBPOS, dev_priv->regfile.saveDSPBPOS);
+	I915_WRITE(_PIPEBSRC, dev_priv->regfile.savePIPEBSRC);
+	I915_WRITE(_DSPBADDR, dev_priv->regfile.saveDSPBADDR);
+	I915_WRITE(_DSPBSTRIDE, dev_priv->regfile.saveDSPBSTRIDE);
+	if (INTEL_INFO(dev)->gen >= 4) {
+		I915_WRITE(_DSPBSURF, dev_priv->regfile.saveDSPBSURF);
+		I915_WRITE(_DSPBTILEOFF, dev_priv->regfile.saveDSPBTILEOFF);
+	}
+
+	I915_WRITE(_PIPEBCONF, dev_priv->regfile.savePIPEBCONF);
+
+	i915_restore_palette(dev, PIPE_B);
+	/* Enable the plane */
+	I915_WRITE(_DSPBCNTR, dev_priv->regfile.saveDSPBCNTR);
+	I915_WRITE(_DSPBADDR, I915_READ(_DSPBADDR));
+
+	/* Cursor state */
+	I915_WRITE(_CURAPOS, dev_priv->regfile.saveCURAPOS);
+	I915_WRITE(_CURACNTR, dev_priv->regfile.saveCURACNTR);
+	I915_WRITE(_CURABASE, dev_priv->regfile.saveCURABASE);
+	I915_WRITE(_CURBPOS, dev_priv->regfile.saveCURBPOS);
+	I915_WRITE(_CURBCNTR, dev_priv->regfile.saveCURBCNTR);
+	I915_WRITE(_CURBBASE, dev_priv->regfile.saveCURBBASE);
+	if (IS_GEN2(dev))
+		I915_WRITE(CURSIZE, dev_priv->regfile.saveCURSIZE);
+
+	/* CRT state */
+	if (HAS_PCH_SPLIT(dev))
+		I915_WRITE(PCH_ADPA, dev_priv->regfile.saveADPA);
+	else
+		I915_WRITE(ADPA, dev_priv->regfile.saveADPA);
+
+	/* Display Port state */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		I915_WRITE(DP_B, dev_priv->regfile.saveDP_B);
+		I915_WRITE(DP_C, dev_priv->regfile.saveDP_C);
+		I915_WRITE(DP_D, dev_priv->regfile.saveDP_D);
+	}
+	/* FIXME: restore TV & SDVO state */
+
+	return;
+}
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 9293878ec7eb..969d08c72d10 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -267,27 +267,27 @@ static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
 
 		crt->force_hotplug_required = 0;
 
-		save_adpa = adpa = I915_READ(PCH_ADPA);
+		save_adpa = adpa = I915_READ(crt->adpa_reg);
 		DRM_DEBUG_KMS("trigger hotplug detect cycle: adpa=0x%x\n", adpa);
 
 		adpa |= ADPA_CRT_HOTPLUG_FORCE_TRIGGER;
 		if (turn_off_dac)
 			adpa &= ~ADPA_DAC_ENABLE;
 
-		I915_WRITE(PCH_ADPA, adpa);
+		I915_WRITE(crt->adpa_reg, adpa);
 
-		if (wait_for((I915_READ(PCH_ADPA) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0,
+		if (wait_for((I915_READ(crt->adpa_reg) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0,
 			     1000))
 			DRM_DEBUG_KMS("timed out waiting for FORCE_TRIGGER");
 
 		if (turn_off_dac) {
-			I915_WRITE(PCH_ADPA, save_adpa);
-			POSTING_READ(PCH_ADPA);
+			I915_WRITE(crt->adpa_reg, save_adpa);
+			POSTING_READ(crt->adpa_reg);
 		}
 	}
 
 	/* Check the status to see if both blue and green are on now */
-	adpa = I915_READ(PCH_ADPA);
+	adpa = I915_READ(crt->adpa_reg);
 	if ((adpa & ADPA_CRT_HOTPLUG_MONITOR_MASK) != 0)
 		ret = true;
 	else
@@ -300,26 +300,27 @@ static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
 static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
+	struct intel_crt *crt = intel_attached_crt(connector);
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 adpa;
 	bool ret;
 	u32 save_adpa;
 
-	save_adpa = adpa = I915_READ(ADPA);
+	save_adpa = adpa = I915_READ(crt->adpa_reg);
 	DRM_DEBUG_KMS("trigger hotplug detect cycle: adpa=0x%x\n", adpa);
 
 	adpa |= ADPA_CRT_HOTPLUG_FORCE_TRIGGER;
 
-	I915_WRITE(ADPA, adpa);
+	I915_WRITE(crt->adpa_reg, adpa);
 
-	if (wait_for((I915_READ(ADPA) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0,
+	if (wait_for((I915_READ(crt->adpa_reg) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0,
 		     1000)) {
 		DRM_DEBUG_KMS("timed out waiting for FORCE_TRIGGER");
-		I915_WRITE(ADPA, save_adpa);
+		I915_WRITE(crt->adpa_reg, save_adpa);
 	}
 
 	/* Check the status to see if both blue and green are on now */
-	adpa = I915_READ(ADPA);
+	adpa = I915_READ(crt->adpa_reg);
 	if ((adpa & ADPA_CRT_HOTPLUG_MONITOR_MASK) != 0)
 		ret = true;
 	else
@@ -665,11 +666,11 @@ static void intel_crt_reset(struct drm_connector *connector)
 	if (HAS_PCH_SPLIT(dev)) {
 		u32 adpa;
 
-		adpa = I915_READ(PCH_ADPA);
+		adpa = I915_READ(crt->adpa_reg);
 		adpa &= ~ADPA_CRT_HOTPLUG_MASK;
 		adpa |= ADPA_HOTPLUG_BITS;
-		I915_WRITE(PCH_ADPA, adpa);
-		POSTING_READ(PCH_ADPA);
+		I915_WRITE(crt->adpa_reg, adpa);
+		POSTING_READ(crt->adpa_reg);
 
 		DRM_DEBUG_KMS("pch crt adpa set to 0x%x\n", adpa);
 		crt->force_hotplug_required = 1;
@@ -684,7 +685,6 @@ static void intel_crt_reset(struct drm_connector *connector)
 static const struct drm_encoder_helper_funcs crt_encoder_funcs = {
 	.mode_fixup = intel_crt_mode_fixup,
 	.mode_set = intel_crt_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_crt_connector_funcs = {
@@ -776,7 +776,7 @@ void intel_crt_init(struct drm_device *dev)
 
 	crt->base.disable = intel_disable_crt;
 	crt->base.enable = intel_enable_crt;
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		crt->base.get_hw_state = intel_ddi_get_hw_state;
 	else
 		crt->base.get_hw_state = intel_crt_get_hw_state;
@@ -800,10 +800,14 @@ void intel_crt_init(struct drm_device *dev)
 	dev_priv->hotplug_supported_mask |= CRT_HOTPLUG_INT_STATUS;
 
 	/*
-	 * TODO: find a proper way to discover whether we need to set the
-	 * polarity reversal bit or not, instead of relying on the BIOS.
+	 * TODO: find a proper way to discover whether we need to set the the
+	 * polarity and link reversal bits or not, instead of relying on the
+	 * BIOS.
 	 */
-	if (HAS_PCH_LPT(dev))
-		dev_priv->fdi_rx_polarity_reversed =
-		     !!(I915_READ(_FDI_RXA_CTL) & FDI_RX_POLARITY_REVERSED_LPT);
+	if (HAS_PCH_LPT(dev)) {
+		u32 fdi_config = FDI_RX_POLARITY_REVERSED_LPT |
+				 FDI_RX_LINK_REVERSAL_OVERRIDE;
+
+		dev_priv->fdi_rx_config = I915_READ(_FDI_RXA_CTL) & fdi_config;
+	}
 }
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 4bad0f724019..d64af5aa4a1c 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -84,7 +84,8 @@ static enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
  * in either FDI or DP modes only, as HDMI connections will work with both
  * of those
  */
-void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port, bool use_fdi_mode)
+static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port,
+				      bool use_fdi_mode)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 reg;
@@ -114,16 +115,17 @@ void intel_prepare_ddi(struct drm_device *dev)
 {
 	int port;
 
-	if (IS_HASWELL(dev)) {
-		for (port = PORT_A; port < PORT_E; port++)
-			intel_prepare_ddi_buffers(dev, port, false);
+	if (!HAS_DDI(dev))
+		return;
 
-		/* DDI E is the suggested one to work in FDI mode, so program is as such by
-		 * default. It will have to be re-programmed in case a digital DP output
-		 * will be detected on it
-		 */
-		intel_prepare_ddi_buffers(dev, PORT_E, true);
-	}
+	for (port = PORT_A; port < PORT_E; port++)
+		intel_prepare_ddi_buffers(dev, port, false);
+
+	/* DDI E is the suggested one to work in FDI mode, so program is as such
+	 * by default. It will have to be re-programmed in case a digital DP
+	 * output will be detected on it
+	 */
+	intel_prepare_ddi_buffers(dev, PORT_E, true);
 }
 
 static const long hsw_ddi_buf_ctl_values[] = {
@@ -178,10 +180,8 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 				  FDI_RX_TP1_TO_TP2_48 | FDI_RX_FDI_DELAY_90);
 
 	/* Enable the PCH Receiver FDI PLL */
-	rx_ctl_val = FDI_RX_PLL_ENABLE | FDI_RX_ENHANCE_FRAME_ENABLE |
-		     ((intel_crtc->fdi_lanes - 1) << 19);
-	if (dev_priv->fdi_rx_polarity_reversed)
-		rx_ctl_val |= FDI_RX_POLARITY_REVERSED_LPT;
+	rx_ctl_val = dev_priv->fdi_rx_config | FDI_RX_ENHANCE_FRAME_ENABLE |
+		     FDI_RX_PLL_ENABLE | ((intel_crtc->fdi_lanes - 1) << 19);
 	I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
 	POSTING_READ(_FDI_RXA_CTL);
 	udelay(220);
@@ -203,7 +203,10 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 					DP_TP_CTL_LINK_TRAIN_PAT1 |
 					DP_TP_CTL_ENABLE);
 
-		/* Configure and enable DDI_BUF_CTL for DDI E with next voltage */
+		/* Configure and enable DDI_BUF_CTL for DDI E with next voltage.
+		 * DDI E does not support port reversal, the functionality is
+		 * achieved on the PCH side in FDI_RX_CTL, so no need to set the
+		 * port reversal bit */
 		I915_WRITE(DDI_BUF_CTL(PORT_E),
 			   DDI_BUF_CTL_ENABLE |
 			   ((intel_crtc->fdi_lanes - 1) << 1) |
@@ -675,10 +678,14 @@ static void intel_ddi_mode_set(struct drm_encoder *encoder,
 	DRM_DEBUG_KMS("Preparing DDI mode for Haswell on port %c, pipe %c\n",
 		      port_name(port), pipe_name(pipe));
 
+	intel_crtc->eld_vld = false;
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+		struct intel_digital_port *intel_dig_port =
+			enc_to_dig_port(encoder);
 
-		intel_dp->DP = DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
+		intel_dp->DP = intel_dig_port->port_reversal |
+			       DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
 		switch (intel_dp->lane_count) {
 		case 1:
 			intel_dp->DP |= DDI_PORT_WIDTH_X1;
@@ -985,7 +992,13 @@ void intel_ddi_enable_pipe_func(struct drm_crtc *crtc)
 	if (cpu_transcoder == TRANSCODER_EDP) {
 		switch (pipe) {
 		case PIPE_A:
-			temp |= TRANS_DDI_EDP_INPUT_A_ONOFF;
+			/* Can only use the always-on power well for eDP when
+			 * not using the panel fitter, and when not using motion
+			  * blur mitigation (which we don't support). */
+			if (dev_priv->pch_pf_size)
+				temp |= TRANS_DDI_EDP_INPUT_A_ONOFF;
+			else
+				temp |= TRANS_DDI_EDP_INPUT_A_ON;
 			break;
 		case PIPE_B:
 			temp |= TRANS_DDI_EDP_INPUT_B_ONOFF;
@@ -1069,7 +1082,7 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 	if (port == PORT_A)
 		cpu_transcoder = TRANSCODER_EDP;
 	else
-		cpu_transcoder = pipe;
+		cpu_transcoder = (enum transcoder) pipe;
 
 	tmp = I915_READ(TRANS_DDI_FUNC_CTL(cpu_transcoder));
 
@@ -1285,34 +1298,58 @@ static void intel_ddi_post_disable(struct intel_encoder *intel_encoder)
 static void intel_enable_ddi(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
 	int type = intel_encoder->type;
+	uint32_t tmp;
 
 	if (type == INTEL_OUTPUT_HDMI) {
+		struct intel_digital_port *intel_dig_port =
+			enc_to_dig_port(encoder);
+
 		/* In HDMI/DVI mode, the port width, and swing/emphasis values
 		 * are ignored so nothing special needs to be done besides
 		 * enabling the port.
 		 */
-		I915_WRITE(DDI_BUF_CTL(port), DDI_BUF_CTL_ENABLE);
+		I915_WRITE(DDI_BUF_CTL(port),
+			   intel_dig_port->port_reversal | DDI_BUF_CTL_ENABLE);
 	} else if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
 		ironlake_edp_backlight_on(intel_dp);
 	}
+
+	if (intel_crtc->eld_vld) {
+		tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
+		tmp |= ((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) << (pipe * 4));
+		I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
+	}
 }
 
 static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
 	int type = intel_encoder->type;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t tmp;
 
 	if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
 		ironlake_edp_backlight_off(intel_dp);
 	}
+
+	tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
+	tmp &= ~((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) << (pipe * 4));
+	I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
 }
 
 int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv)
@@ -1452,11 +1489,11 @@ static const struct drm_encoder_funcs intel_ddi_funcs = {
 static const struct drm_encoder_helper_funcs intel_ddi_helper_funcs = {
 	.mode_fixup = intel_ddi_mode_fixup,
 	.mode_set = intel_ddi_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 void intel_ddi_init(struct drm_device *dev, enum port port)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_digital_port *intel_dig_port;
 	struct intel_encoder *intel_encoder;
 	struct drm_encoder *encoder;
@@ -1497,6 +1534,8 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 	intel_encoder->get_hw_state = intel_ddi_get_hw_state;
 
 	intel_dig_port->port = port;
+	intel_dig_port->port_reversal = I915_READ(DDI_BUF_CTL(port)) &
+					DDI_BUF_PORT_REVERSAL;
 	if (hdmi_connector)
 		intel_dig_port->hdmi.sdvox_reg = DDI_BUF_CTL(port);
 	else
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index da1ad9c80bb5..a05ac2c91ba2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -154,8 +154,8 @@ static const intel_limit_t intel_limits_i9xx_sdvo = {
 	.vco = { .min = 1400000, .max = 2800000 },
 	.n = { .min = 1, .max = 6 },
 	.m = { .min = 70, .max = 120 },
-	.m1 = { .min = 10, .max = 22 },
-	.m2 = { .min = 5, .max = 9 },
+	.m1 = { .min = 8, .max = 18 },
+	.m2 = { .min = 3, .max = 7 },
 	.p = { .min = 5, .max = 80 },
 	.p1 = { .min = 1, .max = 8 },
 	.p2 = { .dot_limit = 200000,
@@ -168,8 +168,8 @@ static const intel_limit_t intel_limits_i9xx_lvds = {
 	.vco = { .min = 1400000, .max = 2800000 },
 	.n = { .min = 1, .max = 6 },
 	.m = { .min = 70, .max = 120 },
-	.m1 = { .min = 10, .max = 22 },
-	.m2 = { .min = 5, .max = 9 },
+	.m1 = { .min = 8, .max = 18 },
+	.m2 = { .min = 3, .max = 7 },
 	.p = { .min = 7, .max = 98 },
 	.p1 = { .min = 1, .max = 8 },
 	.p2 = { .dot_limit = 112000,
@@ -416,13 +416,11 @@ static const intel_limit_t intel_limits_vlv_dp = {
 
 u32 intel_dpio_read(struct drm_i915_private *dev_priv, int reg)
 {
-	unsigned long flags;
-	u32 val = 0;
+	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
 
-	spin_lock_irqsave(&dev_priv->dpio_lock, flags);
 	if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) {
 		DRM_ERROR("DPIO idle wait timed out\n");
-		goto out_unlock;
+		return 0;
 	}
 
 	I915_WRITE(DPIO_REG, reg);
@@ -430,24 +428,20 @@ u32 intel_dpio_read(struct drm_i915_private *dev_priv, int reg)
 		   DPIO_BYTE);
 	if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) {
 		DRM_ERROR("DPIO read wait timed out\n");
-		goto out_unlock;
+		return 0;
 	}
-	val = I915_READ(DPIO_DATA);
 
-out_unlock:
-	spin_unlock_irqrestore(&dev_priv->dpio_lock, flags);
-	return val;
+	return I915_READ(DPIO_DATA);
 }
 
 static void intel_dpio_write(struct drm_i915_private *dev_priv, int reg,
 			     u32 val)
 {
-	unsigned long flags;
+	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
 
-	spin_lock_irqsave(&dev_priv->dpio_lock, flags);
 	if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) {
 		DRM_ERROR("DPIO idle wait timed out\n");
-		goto out_unlock;
+		return;
 	}
 
 	I915_WRITE(DPIO_DATA, val);
@@ -456,9 +450,6 @@ static void intel_dpio_write(struct drm_i915_private *dev_priv, int reg,
 		   DPIO_BYTE);
 	if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100))
 		DRM_ERROR("DPIO write wait timed out\n");
-
-out_unlock:
-       spin_unlock_irqrestore(&dev_priv->dpio_lock, flags);
 }
 
 static void vlv_init_dpio(struct drm_device *dev)
@@ -472,61 +463,14 @@ static void vlv_init_dpio(struct drm_device *dev)
 	POSTING_READ(DPIO_CTL);
 }
 
-static int intel_dual_link_lvds_callback(const struct dmi_system_id *id)
-{
-	DRM_INFO("Forcing lvds to dual link mode on %s\n", id->ident);
-	return 1;
-}
-
-static const struct dmi_system_id intel_dual_link_lvds[] = {
-	{
-		.callback = intel_dual_link_lvds_callback,
-		.ident = "Apple MacBook Pro (Core i5/i7 Series)",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro8,2"),
-		},
-	},
-	{ }	/* terminating entry */
-};
-
-static bool is_dual_link_lvds(struct drm_i915_private *dev_priv,
-			      unsigned int reg)
-{
-	unsigned int val;
-
-	/* use the module option value if specified */
-	if (i915_lvds_channel_mode > 0)
-		return i915_lvds_channel_mode == 2;
-
-	if (dmi_check_system(intel_dual_link_lvds))
-		return true;
-
-	if (dev_priv->lvds_val)
-		val = dev_priv->lvds_val;
-	else {
-		/* BIOS should set the proper LVDS register value at boot, but
-		 * in reality, it doesn't set the value when the lid is closed;
-		 * we need to check "the value to be set" in VBT when LVDS
-		 * register is uninitialized.
-		 */
-		val = I915_READ(reg);
-		if (!(val & ~(LVDS_PIPE_MASK | LVDS_DETECTED)))
-			val = dev_priv->bios_lvds_val;
-		dev_priv->lvds_val = val;
-	}
-	return (val & LVDS_CLKB_POWER_MASK) == LVDS_CLKB_POWER_UP;
-}
-
 static const intel_limit_t *intel_ironlake_limit(struct drm_crtc *crtc,
 						int refclk)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	const intel_limit_t *limit;
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) {
-		if (is_dual_link_lvds(dev_priv, PCH_LVDS)) {
+		if (intel_is_dual_link_lvds(dev)) {
 			/* LVDS dual channel */
 			if (refclk == 100000)
 				limit = &intel_limits_ironlake_dual_lvds_100m;
@@ -550,11 +494,10 @@ static const intel_limit_t *intel_ironlake_limit(struct drm_crtc *crtc,
 static const intel_limit_t *intel_g4x_limit(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	const intel_limit_t *limit;
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) {
-		if (is_dual_link_lvds(dev_priv, LVDS))
+		if (intel_is_dual_link_lvds(dev))
 			/* LVDS with dual channel */
 			limit = &intel_limits_g4x_dual_channel_lvds;
 		else
@@ -686,19 +629,16 @@ intel_find_best_PLL(const intel_limit_t *limit, struct drm_crtc *crtc,
 
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	intel_clock_t clock;
 	int err = target;
 
-	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) &&
-	    (I915_READ(LVDS)) != 0) {
+	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) {
 		/*
-		 * For LVDS, if the panel is on, just rely on its current
-		 * settings for dual-channel.  We haven't figured out how to
-		 * reliably set up different single/dual channel state, if we
-		 * even can.
+		 * For LVDS just rely on its current settings for dual-channel.
+		 * We haven't figured out how to reliably set up different
+		 * single/dual channel state, if we even can.
 		 */
-		if (is_dual_link_lvds(dev_priv, LVDS))
+		if (intel_is_dual_link_lvds(dev))
 			clock.p2 = limit->p2.p2_fast;
 		else
 			clock.p2 = limit->p2.p2_slow;
@@ -751,7 +691,6 @@ intel_g4x_find_best_PLL(const intel_limit_t *limit, struct drm_crtc *crtc,
 			intel_clock_t *best_clock)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	intel_clock_t clock;
 	int max_n;
 	bool found;
@@ -766,8 +705,7 @@ intel_g4x_find_best_PLL(const intel_limit_t *limit, struct drm_crtc *crtc,
 			lvds_reg = PCH_LVDS;
 		else
 			lvds_reg = LVDS;
-		if ((I915_READ(lvds_reg) & LVDS_CLKB_POWER_MASK) ==
-		    LVDS_CLKB_POWER_UP)
+		if (intel_is_dual_link_lvds(dev))
 			clock.p2 = limit->p2.p2_fast;
 		else
 			clock.p2 = limit->p2.p2_slow;
@@ -1047,6 +985,51 @@ void intel_wait_for_pipe_off(struct drm_device *dev, int pipe)
 	}
 }
 
+/*
+ * ibx_digital_port_connected - is the specified port connected?
+ * @dev_priv: i915 private structure
+ * @port: the port to test
+ *
+ * Returns true if @port is connected, false otherwise.
+ */
+bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
+				struct intel_digital_port *port)
+{
+	u32 bit;
+
+	if (HAS_PCH_IBX(dev_priv->dev)) {
+		switch(port->port) {
+		case PORT_B:
+			bit = SDE_PORTB_HOTPLUG;
+			break;
+		case PORT_C:
+			bit = SDE_PORTC_HOTPLUG;
+			break;
+		case PORT_D:
+			bit = SDE_PORTD_HOTPLUG;
+			break;
+		default:
+			return true;
+		}
+	} else {
+		switch(port->port) {
+		case PORT_B:
+			bit = SDE_PORTB_HOTPLUG_CPT;
+			break;
+		case PORT_C:
+			bit = SDE_PORTC_HOTPLUG_CPT;
+			break;
+		case PORT_D:
+			bit = SDE_PORTD_HOTPLUG_CPT;
+			break;
+		default:
+			return true;
+		}
+	}
+
+	return I915_READ(SDEISR) & bit;
+}
+
 static const char *state_string(bool enabled)
 {
 	return enabled ? "on" : "off";
@@ -1125,8 +1108,8 @@ static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
 
-	if (IS_HASWELL(dev_priv->dev)) {
-		/* On Haswell, DDI is used instead of FDI_TX_CTL */
+	if (HAS_DDI(dev_priv->dev)) {
+		/* DDI does not have a specific FDI_TX register */
 		reg = TRANS_DDI_FUNC_CTL(cpu_transcoder);
 		val = I915_READ(reg);
 		cur_state = !!(val & TRANS_DDI_FUNC_ENABLE);
@@ -1170,7 +1153,7 @@ static void assert_fdi_tx_pll_enabled(struct drm_i915_private *dev_priv,
 		return;
 
 	/* On Haswell, DDI ports are responsible for the FDI PLL setup */
-	if (IS_HASWELL(dev_priv->dev))
+	if (HAS_DDI(dev_priv->dev))
 		return;
 
 	reg = FDI_TX_CTL(pipe);
@@ -1231,9 +1214,15 @@ void assert_pipe(struct drm_i915_private *dev_priv,
 	if (pipe == PIPE_A && dev_priv->quirks & QUIRK_PIPEA_FORCE)
 		state = true;
 
-	reg = PIPECONF(cpu_transcoder);
-	val = I915_READ(reg);
-	cur_state = !!(val & PIPECONF_ENABLE);
+	if (IS_HASWELL(dev_priv->dev) && cpu_transcoder != TRANSCODER_EDP &&
+	    !(I915_READ(HSW_PWR_WELL_DRIVER) & HSW_PWR_WELL_ENABLE)) {
+		cur_state = false;
+	} else {
+		reg = PIPECONF(cpu_transcoder);
+		val = I915_READ(reg);
+		cur_state = !!(val & PIPECONF_ENABLE);
+	}
+
 	WARN(cur_state != state,
 	     "pipe %c assertion failure (expected %s, current %s)\n",
 	     pipe_name(pipe), state_string(state), state_string(cur_state));
@@ -1509,13 +1498,14 @@ static void
 intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 		enum intel_sbi_destination destination)
 {
-	unsigned long flags;
 	u32 tmp;
 
-	spin_lock_irqsave(&dev_priv->dpio_lock, flags);
-	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0, 100)) {
+	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
+
+	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
+				100)) {
 		DRM_ERROR("timeout waiting for SBI to become ready\n");
-		goto out_unlock;
+		return;
 	}
 
 	I915_WRITE(SBI_ADDR, (reg << 16));
@@ -1530,24 +1520,21 @@ intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 	if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0,
 				100)) {
 		DRM_ERROR("timeout waiting for SBI to complete write transaction\n");
-		goto out_unlock;
+		return;
 	}
-
-out_unlock:
-	spin_unlock_irqrestore(&dev_priv->dpio_lock, flags);
 }
 
 static u32
 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 	       enum intel_sbi_destination destination)
 {
-	unsigned long flags;
 	u32 value = 0;
+	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
 
-	spin_lock_irqsave(&dev_priv->dpio_lock, flags);
-	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0, 100)) {
+	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
+				100)) {
 		DRM_ERROR("timeout waiting for SBI to become ready\n");
-		goto out_unlock;
+		return 0;
 	}
 
 	I915_WRITE(SBI_ADDR, (reg << 16));
@@ -1561,14 +1548,10 @@ intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 	if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0,
 				100)) {
 		DRM_ERROR("timeout waiting for SBI to complete read transaction\n");
-		goto out_unlock;
+		return 0;
 	}
 
-	value = I915_READ(SBI_DATA);
-
-out_unlock:
-	spin_unlock_irqrestore(&dev_priv->dpio_lock, flags);
-	return value;
+	return I915_READ(SBI_DATA);
 }
 
 /**
@@ -1700,8 +1683,8 @@ static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 		 * make the BPC in transcoder be consistent with
 		 * that in pipeconf reg.
 		 */
-		val &= ~PIPE_BPC_MASK;
-		val |= pipeconf_val & PIPE_BPC_MASK;
+		val &= ~PIPECONF_BPC_MASK;
+		val |= pipeconf_val & PIPECONF_BPC_MASK;
 	}
 
 	val &= ~TRANS_INTERLACE_MASK;
@@ -1728,7 +1711,7 @@ static void lpt_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 	BUG_ON(dev_priv->info->gen < 5);
 
 	/* FDI must be feeding us bits for PCH ports */
-	assert_fdi_tx_enabled(dev_priv, cpu_transcoder);
+	assert_fdi_tx_enabled(dev_priv, (enum pipe) cpu_transcoder);
 	assert_fdi_rx_enabled(dev_priv, TRANSCODER_A);
 
 	/* Workaround: set timing override bit. */
@@ -1816,11 +1799,11 @@ static void intel_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
 {
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
-	enum transcoder pch_transcoder;
+	enum pipe pch_transcoder;
 	int reg;
 	u32 val;
 
-	if (IS_HASWELL(dev_priv->dev))
+	if (HAS_PCH_LPT(dev_priv->dev))
 		pch_transcoder = TRANSCODER_A;
 	else
 		pch_transcoder = pipe;
@@ -1836,7 +1819,8 @@ static void intel_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
 		if (pch_port) {
 			/* if driving the PCH, we need FDI enabled */
 			assert_fdi_rx_pll_enabled(dev_priv, pch_transcoder);
-			assert_fdi_tx_pll_enabled(dev_priv, cpu_transcoder);
+			assert_fdi_tx_pll_enabled(dev_priv,
+						  (enum pipe) cpu_transcoder);
 		}
 		/* FIXME: assert CPU port conditions for SNB+ */
 	}
@@ -2017,18 +2001,29 @@ void intel_unpin_fb_obj(struct drm_i915_gem_object *obj)
 
 /* Computes the linear offset to the base tile and adjusts x, y. bytes per pixel
  * is assumed to be a power-of-two. */
-unsigned long intel_gen4_compute_offset_xtiled(int *x, int *y,
-					       unsigned int bpp,
-					       unsigned int pitch)
+unsigned long intel_gen4_compute_page_offset(int *x, int *y,
+					     unsigned int tiling_mode,
+					     unsigned int cpp,
+					     unsigned int pitch)
 {
-	int tile_rows, tiles;
+	if (tiling_mode != I915_TILING_NONE) {
+		unsigned int tile_rows, tiles;
+
+		tile_rows = *y / 8;
+		*y %= 8;
+
+		tiles = *x / (512/cpp);
+		*x %= 512/cpp;
 
-	tile_rows = *y / 8;
-	*y %= 8;
-	tiles = *x / (512/bpp);
-	*x %= 512/bpp;
+		return tile_rows * pitch * 8 + tiles * 4096;
+	} else {
+		unsigned int offset;
 
-	return tile_rows * pitch * 8 + tiles * 4096;
+		offset = *y * pitch + *x * cpp;
+		*y = 0;
+		*x = (offset & 4095) / cpp;
+		return offset & -4096;
+	}
 }
 
 static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
@@ -2105,9 +2100,9 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 
 	if (INTEL_INFO(dev)->gen >= 4) {
 		intel_crtc->dspaddr_offset =
-			intel_gen4_compute_offset_xtiled(&x, &y,
-							 fb->bits_per_pixel / 8,
-							 fb->pitches[0]);
+			intel_gen4_compute_page_offset(&x, &y, obj->tiling_mode,
+						       fb->bits_per_pixel / 8,
+						       fb->pitches[0]);
 		linear_offset -= intel_crtc->dspaddr_offset;
 	} else {
 		intel_crtc->dspaddr_offset = linear_offset;
@@ -2198,9 +2193,9 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
 
 	linear_offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
 	intel_crtc->dspaddr_offset =
-		intel_gen4_compute_offset_xtiled(&x, &y,
-						 fb->bits_per_pixel / 8,
-						 fb->pitches[0]);
+		intel_gen4_compute_page_offset(&x, &y, obj->tiling_mode,
+					       fb->bits_per_pixel / 8,
+					       fb->pitches[0]);
 	linear_offset -= intel_crtc->dspaddr_offset;
 
 	DRM_DEBUG_KMS("Writing base %08X %08lX %d %d %d\n",
@@ -2242,10 +2237,6 @@ intel_finish_fb(struct drm_framebuffer *old_fb)
 	bool was_interruptible = dev_priv->mm.interruptible;
 	int ret;
 
-	wait_event(dev_priv->pending_flip_queue,
-		   atomic_read(&dev_priv->mm.wedged) ||
-		   atomic_read(&obj->pending_flip) == 0);
-
 	/* Big Hammer, we also need to ensure that any pending
 	 * MI_WAIT_FOR_EVENT inside a user batch buffer on the
 	 * current scanout is retired before unpinning the old
@@ -2350,43 +2341,6 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 	return 0;
 }
 
-static void ironlake_set_pll_edp(struct drm_crtc *crtc, int clock)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 dpa_ctl;
-
-	DRM_DEBUG_KMS("eDP PLL enable for clock %d\n", clock);
-	dpa_ctl = I915_READ(DP_A);
-	dpa_ctl &= ~DP_PLL_FREQ_MASK;
-
-	if (clock < 200000) {
-		u32 temp;
-		dpa_ctl |= DP_PLL_FREQ_160MHZ;
-		/* workaround for 160Mhz:
-		   1) program 0x4600c bits 15:0 = 0x8124
-		   2) program 0x46010 bit 0 = 1
-		   3) program 0x46034 bit 24 = 1
-		   4) program 0x64000 bit 14 = 1
-		   */
-		temp = I915_READ(0x4600c);
-		temp &= 0xffff0000;
-		I915_WRITE(0x4600c, temp | 0x8124);
-
-		temp = I915_READ(0x46010);
-		I915_WRITE(0x46010, temp | 1);
-
-		temp = I915_READ(0x46034);
-		I915_WRITE(0x46034, temp | (1 << 24));
-	} else {
-		dpa_ctl |= DP_PLL_FREQ_270MHZ;
-	}
-	I915_WRITE(DP_A, dpa_ctl);
-
-	POSTING_READ(DP_A);
-	udelay(500);
-}
-
 static void intel_fdi_normal_train(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -2815,7 +2769,7 @@ static void ironlake_fdi_pll_enable(struct intel_crtc *intel_crtc)
 	temp = I915_READ(reg);
 	temp &= ~((0x7 << 19) | (0x7 << 16));
 	temp |= (intel_crtc->fdi_lanes - 1) << 19;
-	temp |= (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) << 11;
+	temp |= (I915_READ(PIPECONF(pipe)) & PIPECONF_BPC_MASK) << 11;
 	I915_WRITE(reg, temp | FDI_RX_PLL_ENABLE);
 
 	POSTING_READ(reg);
@@ -2828,18 +2782,14 @@ static void ironlake_fdi_pll_enable(struct intel_crtc *intel_crtc)
 	POSTING_READ(reg);
 	udelay(200);
 
-	/* On Haswell, the PLL configuration for ports and pipes is handled
-	 * separately, as part of DDI setup */
-	if (!IS_HASWELL(dev)) {
-		/* Enable CPU FDI TX PLL, always on for Ironlake */
-		reg = FDI_TX_CTL(pipe);
-		temp = I915_READ(reg);
-		if ((temp & FDI_TX_PLL_ENABLE) == 0) {
-			I915_WRITE(reg, temp | FDI_TX_PLL_ENABLE);
+	/* Enable CPU FDI TX PLL, always on for Ironlake */
+	reg = FDI_TX_CTL(pipe);
+	temp = I915_READ(reg);
+	if ((temp & FDI_TX_PLL_ENABLE) == 0) {
+		I915_WRITE(reg, temp | FDI_TX_PLL_ENABLE);
 
-			POSTING_READ(reg);
-			udelay(100);
-		}
+		POSTING_READ(reg);
+		udelay(100);
 	}
 }
 
@@ -2889,7 +2839,7 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc)
 	reg = FDI_RX_CTL(pipe);
 	temp = I915_READ(reg);
 	temp &= ~(0x7 << 16);
-	temp |= (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) << 11;
+	temp |= (I915_READ(PIPECONF(pipe)) & PIPECONF_BPC_MASK) << 11;
 	I915_WRITE(reg, temp & ~FDI_RX_ENABLE);
 
 	POSTING_READ(reg);
@@ -2918,7 +2868,7 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc)
 	}
 	/* BPC in FDI rx is consistent with that in PIPECONF */
 	temp &= ~(0x07 << 16);
-	temp |= (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) << 11;
+	temp |= (I915_READ(PIPECONF(pipe)) & PIPECONF_BPC_MASK) << 11;
 	I915_WRITE(reg, temp);
 
 	POSTING_READ(reg);
@@ -2929,10 +2879,12 @@ static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	unsigned long flags;
 	bool pending;
 
-	if (atomic_read(&dev_priv->mm.wedged))
+	if (i915_reset_in_progress(&dev_priv->gpu_error) ||
+	    intel_crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
 		return false;
 
 	spin_lock_irqsave(&dev->event_lock, flags);
@@ -2950,6 +2902,8 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 	if (crtc->fb == NULL)
 		return;
 
+	WARN_ON(waitqueue_active(&dev_priv->pending_flip_queue));
+
 	wait_event(dev_priv->pending_flip_queue,
 		   !intel_crtc_has_pending_flip(crtc));
 
@@ -2992,6 +2946,8 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 	u32 divsel, phaseinc, auxdiv, phasedir = 0;
 	u32 temp;
 
+	mutex_lock(&dev_priv->dpio_lock);
+
 	/* It is necessary to ungate the pixclk gate prior to programming
 	 * the divisors, and gate it back when it is done.
 	 */
@@ -3066,6 +3022,8 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 	udelay(24);
 
 	I915_WRITE(PIXCLK_GATE, PIXCLK_GATE_UNGATE);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 /*
@@ -3146,7 +3104,7 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 	if (HAS_PCH_CPT(dev) &&
 	    (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT) ||
 	     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
-		u32 bpc = (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) >> 5;
+		u32 bpc = (I915_READ(PIPECONF(pipe)) & PIPECONF_BPC_MASK) >> 5;
 		reg = TRANS_DP_CTL(pipe);
 		temp = I915_READ(reg);
 		temp &= ~(TRANS_DP_PORT_SEL_MASK |
@@ -3623,7 +3581,7 @@ static void haswell_crtc_off(struct drm_crtc *crtc)
 
 	/* Stop saying we're using TRANSCODER_EDP because some other CRTC might
 	 * start using it. */
-	intel_crtc->cpu_transcoder = intel_crtc->pipe;
+	intel_crtc->cpu_transcoder = (enum transcoder) intel_crtc->pipe;
 
 	intel_ddi_put_crtc_pll(crtc);
 }
@@ -3664,6 +3622,11 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 	intel_update_watermarks(dev);
 
 	intel_enable_pll(dev_priv, pipe);
+
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->pre_enable)
+			encoder->pre_enable(encoder);
+
 	intel_enable_pipe(dev_priv, pipe, false);
 	intel_enable_plane(dev_priv, plane, pipe);
 
@@ -3686,6 +3649,7 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
+	u32 pctl;
 
 
 	if (!intel_crtc->active)
@@ -3705,6 +3669,13 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 
 	intel_disable_plane(dev_priv, plane, pipe);
 	intel_disable_pipe(dev_priv, pipe);
+
+	/* Disable pannel fitter if it is on this pipe. */
+	pctl = I915_READ(PFIT_CONTROL);
+	if ((pctl & PFIT_ENABLE) &&
+	    ((pctl & PFIT_PIPE_MASK) >> PFIT_PIPE_SHIFT) == pipe)
+		I915_WRITE(PFIT_CONTROL, 0);
+
 	intel_disable_pll(dev_priv, pipe);
 
 	intel_crtc->active = false;
@@ -3767,19 +3738,17 @@ void intel_crtc_update_dpms(struct drm_crtc *crtc)
 	intel_crtc_update_sarea(crtc, enable);
 }
 
-static void intel_crtc_noop(struct drm_crtc *crtc)
-{
-}
-
 static void intel_crtc_disable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_connector *connector;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	/* crtc should still be enabled when we disable it. */
 	WARN_ON(!crtc->enabled);
 
+	intel_crtc->eld_vld = false;
 	dev_priv->display.crtc_disable(crtc);
 	intel_crtc_update_sarea(crtc, false);
 	dev_priv->display.off(crtc);
@@ -3817,10 +3786,6 @@ void intel_modeset_disable(struct drm_device *dev)
 	}
 }
 
-void intel_encoder_noop(struct drm_encoder *encoder)
-{
-}
-
 void intel_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
@@ -4012,16 +3977,8 @@ static int i830_get_display_clock_speed(struct drm_device *dev)
 	return 133000;
 }
 
-struct fdi_m_n {
-	u32        tu;
-	u32        gmch_m;
-	u32        gmch_n;
-	u32        link_m;
-	u32        link_n;
-};
-
 static void
-fdi_reduce_ratio(u32 *num, u32 *den)
+intel_reduce_ratio(uint32_t *num, uint32_t *den)
 {
 	while (*num > 0xffffff || *den > 0xffffff) {
 		*num >>= 1;
@@ -4029,20 +3986,18 @@ fdi_reduce_ratio(u32 *num, u32 *den)
 	}
 }
 
-static void
-ironlake_compute_m_n(int bits_per_pixel, int nlanes, int pixel_clock,
-		     int link_clock, struct fdi_m_n *m_n)
+void
+intel_link_compute_m_n(int bits_per_pixel, int nlanes,
+		       int pixel_clock, int link_clock,
+		       struct intel_link_m_n *m_n)
 {
-	m_n->tu = 64; /* default size */
-
-	/* BUG_ON(pixel_clock > INT_MAX / 36); */
+	m_n->tu = 64;
 	m_n->gmch_m = bits_per_pixel * pixel_clock;
 	m_n->gmch_n = link_clock * nlanes * 8;
-	fdi_reduce_ratio(&m_n->gmch_m, &m_n->gmch_n);
-
+	intel_reduce_ratio(&m_n->gmch_m, &m_n->gmch_n);
 	m_n->link_m = pixel_clock;
 	m_n->link_n = link_clock;
-	fdi_reduce_ratio(&m_n->link_m, &m_n->link_n);
+	intel_reduce_ratio(&m_n->link_m, &m_n->link_n);
 }
 
 static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
@@ -4289,51 +4244,6 @@ static void i9xx_update_pll_dividers(struct drm_crtc *crtc,
 	}
 }
 
-static void intel_update_lvds(struct drm_crtc *crtc, intel_clock_t *clock,
-			      struct drm_display_mode *adjusted_mode)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	u32 temp;
-
-	temp = I915_READ(LVDS);
-	temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
-	if (pipe == 1) {
-		temp |= LVDS_PIPEB_SELECT;
-	} else {
-		temp &= ~LVDS_PIPEB_SELECT;
-	}
-	/* set the corresponsding LVDS_BORDER bit */
-	temp |= dev_priv->lvds_border_bits;
-	/* Set the B0-B3 data pairs corresponding to whether we're going to
-	 * set the DPLLs for dual-channel mode or not.
-	 */
-	if (clock->p2 == 7)
-		temp |= LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP;
-	else
-		temp &= ~(LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP);
-
-	/* It would be nice to set 24 vs 18-bit mode (LVDS_A3_POWER_UP)
-	 * appropriately here, but we need to look more thoroughly into how
-	 * panels behave in the two modes.
-	 */
-	/* set the dithering flag on LVDS as needed */
-	if (INTEL_INFO(dev)->gen >= 4) {
-		if (dev_priv->lvds_dither)
-			temp |= LVDS_ENABLE_DITHER;
-		else
-			temp &= ~LVDS_ENABLE_DITHER;
-	}
-	temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY);
-	if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC)
-		temp |= LVDS_HSYNC_POLARITY;
-	if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC)
-		temp |= LVDS_VSYNC_POLARITY;
-	I915_WRITE(LVDS, temp);
-}
-
 static void vlv_update_pll(struct drm_crtc *crtc,
 			   struct drm_display_mode *mode,
 			   struct drm_display_mode *adjusted_mode,
@@ -4349,6 +4259,8 @@ static void vlv_update_pll(struct drm_crtc *crtc,
 	bool is_sdvo;
 	u32 temp;
 
+	mutex_lock(&dev_priv->dpio_lock);
+
 	is_sdvo = intel_pipe_has_type(crtc, INTEL_OUTPUT_SDVO) ||
 		intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI);
 
@@ -4432,6 +4344,8 @@ static void vlv_update_pll(struct drm_crtc *crtc,
 			temp |= (1 << 21);
 		intel_dpio_write(dev_priv, DPIO_DATA_CHANNEL2, temp);
 	}
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 static void i9xx_update_pll(struct drm_crtc *crtc,
@@ -4443,6 +4357,7 @@ static void i9xx_update_pll(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	u32 dpll;
 	bool is_sdvo;
@@ -4511,12 +4426,9 @@ static void i9xx_update_pll(struct drm_crtc *crtc,
 	POSTING_READ(DPLL(pipe));
 	udelay(150);
 
-	/* The LVDS pin pair needs to be on before the DPLLs are enabled.
-	 * This is an exception to the general rule that mode_set doesn't turn
-	 * things on.
-	 */
-	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-		intel_update_lvds(crtc, clock, adjusted_mode);
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->pre_pll_enable)
+			encoder->pre_pll_enable(encoder);
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT))
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
@@ -4555,6 +4467,7 @@ static void i8xx_update_pll(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	u32 dpll;
 
@@ -4588,12 +4501,9 @@ static void i8xx_update_pll(struct drm_crtc *crtc,
 	POSTING_READ(DPLL(pipe));
 	udelay(150);
 
-	/* The LVDS pin pair needs to be on before the DPLLs are enabled.
-	 * This is an exception to the general rule that mode_set doesn't turn
-	 * things on.
-	 */
-	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-		intel_update_lvds(crtc, clock, adjusted_mode);
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->pre_pll_enable)
+			encoder->pre_pll_enable(encoder);
 
 	I915_WRITE(DPLL(pipe), dpll);
 
@@ -4783,10 +4693,10 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	}
 
 	/* default to 8bpc */
-	pipeconf &= ~(PIPECONF_BPP_MASK | PIPECONF_DITHER_EN);
+	pipeconf &= ~(PIPECONF_BPC_MASK | PIPECONF_DITHER_EN);
 	if (is_dp) {
 		if (adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
-			pipeconf |= PIPECONF_BPP_6 |
+			pipeconf |= PIPECONF_6BPC |
 				    PIPECONF_DITHER_EN |
 				    PIPECONF_DITHER_TYPE_SP;
 		}
@@ -4794,7 +4704,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 
 	if (IS_VALLEYVIEW(dev) && intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) {
 		if (adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
-			pipeconf |= PIPECONF_BPP_6 |
+			pipeconf |= PIPECONF_6BPC |
 					PIPECONF_ENABLE |
 					I965_PIPECONF_ACTIVE;
 		}
@@ -4981,6 +4891,8 @@ static void lpt_init_pch_refclk(struct drm_device *dev)
 	if (!has_vga)
 		return;
 
+	mutex_lock(&dev_priv->dpio_lock);
+
 	/* XXX: Rip out SDV support once Haswell ships for real. */
 	if (IS_HASWELL(dev) && (dev->pci_device & 0xFF00) == 0x0C00)
 		is_sdv = true;
@@ -5123,6 +5035,8 @@ static void lpt_init_pch_refclk(struct drm_device *dev)
 	tmp = intel_sbi_read(dev_priv, SBI_DBUFF0, SBI_ICLK);
 	tmp |= SBI_DBUFF0_ENABLE;
 	intel_sbi_write(dev_priv, SBI_DBUFF0, tmp, SBI_ICLK);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 /*
@@ -5177,19 +5091,19 @@ static void ironlake_set_pipeconf(struct drm_crtc *crtc,
 
 	val = I915_READ(PIPECONF(pipe));
 
-	val &= ~PIPE_BPC_MASK;
+	val &= ~PIPECONF_BPC_MASK;
 	switch (intel_crtc->bpp) {
 	case 18:
-		val |= PIPE_6BPC;
+		val |= PIPECONF_6BPC;
 		break;
 	case 24:
-		val |= PIPE_8BPC;
+		val |= PIPECONF_8BPC;
 		break;
 	case 30:
-		val |= PIPE_10BPC;
+		val |= PIPECONF_10BPC;
 		break;
 	case 36:
-		val |= PIPE_12BPC;
+		val |= PIPECONF_12BPC;
 		break;
 	default:
 		/* Case prevented by intel_choose_pipe_bpp_dither. */
@@ -5206,10 +5120,80 @@ static void ironlake_set_pipeconf(struct drm_crtc *crtc,
 	else
 		val |= PIPECONF_PROGRESSIVE;
 
+	if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+		val |= PIPECONF_COLOR_RANGE_SELECT;
+	else
+		val &= ~PIPECONF_COLOR_RANGE_SELECT;
+
 	I915_WRITE(PIPECONF(pipe), val);
 	POSTING_READ(PIPECONF(pipe));
 }
 
+/*
+ * Set up the pipe CSC unit.
+ *
+ * Currently only full range RGB to limited range RGB conversion
+ * is supported, but eventually this should handle various
+ * RGB<->YCbCr scenarios as well.
+ */
+static void intel_set_pipe_csc(struct drm_crtc *crtc,
+			       const struct drm_display_mode *adjusted_mode)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	uint16_t coeff = 0x7800; /* 1.0 */
+
+	/*
+	 * TODO: Check what kind of values actually come out of the pipe
+	 * with these coeff/postoff values and adjust to get the best
+	 * accuracy. Perhaps we even need to take the bpc value into
+	 * consideration.
+	 */
+
+	if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+		coeff = ((235 - 16) * (1 << 12) / 255) & 0xff8; /* 0.xxx... */
+
+	/*
+	 * GY/GU and RY/RU should be the other way around according
+	 * to BSpec, but reality doesn't agree. Just set them up in
+	 * a way that results in the correct picture.
+	 */
+	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeff << 16);
+	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), 0);
+
+	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeff);
+	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), 0);
+
+	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), 0);
+	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeff << 16);
+
+	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
+
+	if (INTEL_INFO(dev)->gen > 6) {
+		uint16_t postoff = 0;
+
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			postoff = (16 * (1 << 13) / 255) & 0x1fff;
+
+		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff);
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), 0);
+	} else {
+		uint32_t mode = CSC_MODE_YUV_TO_RGB;
+
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			mode |= CSC_BLACK_SCREEN_OFFSET;
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), mode);
+	}
+}
+
 static void haswell_set_pipeconf(struct drm_crtc *crtc,
 				 struct drm_display_mode *adjusted_mode,
 				 bool dither)
@@ -5400,7 +5384,7 @@ static void ironlake_set_m_n(struct drm_crtc *crtc,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum transcoder cpu_transcoder = intel_crtc->cpu_transcoder;
 	struct intel_encoder *intel_encoder, *edp_encoder = NULL;
-	struct fdi_m_n m_n = {0};
+	struct intel_link_m_n m_n = {0};
 	int target_clock, pixel_multiplier, lane, link_bw;
 	bool is_dp = false, is_cpu_edp = false;
 
@@ -5452,8 +5436,7 @@ static void ironlake_set_m_n(struct drm_crtc *crtc,
 
 	if (pixel_multiplier > 1)
 		link_bw *= pixel_multiplier;
-	ironlake_compute_m_n(intel_crtc->bpp, lane, target_clock, link_bw,
-			     &m_n);
+	intel_link_compute_m_n(intel_crtc->bpp, lane, target_clock, link_bw, &m_n);
 
 	I915_WRITE(PIPE_DATA_M1(cpu_transcoder), TU_SIZE(m_n.tu) | m_n.gmch_m);
 	I915_WRITE(PIPE_DATA_N1(cpu_transcoder), m_n.gmch_n);
@@ -5506,7 +5489,7 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 	if (is_lvds) {
 		if ((intel_panel_use_ssc(dev_priv) &&
 		     dev_priv->lvds_ssc_freq == 100) ||
-		    (I915_READ(PCH_LVDS) & LVDS_CLKB_POWER_MASK) == LVDS_CLKB_POWER_UP)
+		    intel_is_dual_link_lvds(dev))
 			factor = 25;
 	} else if (is_sdvo && is_tv)
 		factor = 20;
@@ -5581,7 +5564,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	bool ok, has_reduced_clock = false;
 	bool is_lvds = false, is_dp = false, is_cpu_edp = false;
 	struct intel_encoder *encoder;
-	u32 temp;
 	int ret;
 	bool dither, fdi_config_ok;
 
@@ -5645,54 +5627,12 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	} else
 		intel_put_pch_pll(intel_crtc);
 
-	/* The LVDS pin pair needs to be on before the DPLLs are enabled.
-	 * This is an exception to the general rule that mode_set doesn't turn
-	 * things on.
-	 */
-	if (is_lvds) {
-		temp = I915_READ(PCH_LVDS);
-		temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
-		if (HAS_PCH_CPT(dev)) {
-			temp &= ~PORT_TRANS_SEL_MASK;
-			temp |= PORT_TRANS_SEL_CPT(pipe);
-		} else {
-			if (pipe == 1)
-				temp |= LVDS_PIPEB_SELECT;
-			else
-				temp &= ~LVDS_PIPEB_SELECT;
-		}
-
-		/* set the corresponsding LVDS_BORDER bit */
-		temp |= dev_priv->lvds_border_bits;
-		/* Set the B0-B3 data pairs corresponding to whether we're going to
-		 * set the DPLLs for dual-channel mode or not.
-		 */
-		if (clock.p2 == 7)
-			temp |= LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP;
-		else
-			temp &= ~(LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP);
-
-		/* It would be nice to set 24 vs 18-bit mode (LVDS_A3_POWER_UP)
-		 * appropriately here, but we need to look more thoroughly into how
-		 * panels behave in the two modes.
-		 */
-		temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY);
-		if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC)
-			temp |= LVDS_HSYNC_POLARITY;
-		if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC)
-			temp |= LVDS_VSYNC_POLARITY;
-		I915_WRITE(PCH_LVDS, temp);
-	}
-
-	if (is_dp && !is_cpu_edp) {
+	if (is_dp && !is_cpu_edp)
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
-	} else {
-		/* For non-DP output, clear any trans DP clock recovery setting.*/
-		I915_WRITE(TRANSDATA_M1(pipe), 0);
-		I915_WRITE(TRANSDATA_N1(pipe), 0);
-		I915_WRITE(TRANSDPLINK_M1(pipe), 0);
-		I915_WRITE(TRANSDPLINK_N1(pipe), 0);
-	}
+
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->pre_pll_enable)
+			encoder->pre_pll_enable(encoder);
 
 	if (intel_crtc->pch_pll) {
 		I915_WRITE(intel_crtc->pch_pll->pll_reg, dpll);
@@ -5727,9 +5667,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 
 	fdi_config_ok = ironlake_check_fdi_lanes(intel_crtc);
 
-	if (is_cpu_edp)
-		ironlake_set_pll_edp(crtc, adjusted_mode->clock);
-
 	ironlake_set_pipeconf(crtc, adjusted_mode, dither);
 
 	intel_wait_for_vblank(dev, pipe);
@@ -5747,6 +5684,35 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	return fdi_config_ok ? ret : -EINVAL;
 }
 
+static void haswell_modeset_global_resources(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	bool enable = false;
+	struct intel_crtc *crtc;
+	struct intel_encoder *encoder;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+		if (crtc->pipe != PIPE_A && crtc->base.enabled)
+			enable = true;
+		/* XXX: Should check for edp transcoder here, but thanks to init
+		 * sequence that's not yet available. Just in case desktop eDP
+		 * on PORT D is possible on haswell, too. */
+	}
+
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
+			    base.head) {
+		if (encoder->type != INTEL_OUTPUT_EDP &&
+		    encoder->connectors_active)
+			enable = true;
+	}
+
+	/* Even the eDP panel fitter is outside the always-on well. */
+	if (dev_priv->pch_pf_size)
+		enable = true;
+
+	intel_set_power_well(dev, enable);
+}
+
 static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 				 struct drm_display_mode *mode,
 				 struct drm_display_mode *adjusted_mode,
@@ -5759,20 +5725,13 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
 	int num_connectors = 0;
-	intel_clock_t clock, reduced_clock;
-	u32 dpll = 0, fp = 0, fp2 = 0;
-	bool ok, has_reduced_clock = false;
-	bool is_lvds = false, is_dp = false, is_cpu_edp = false;
+	bool is_dp = false, is_cpu_edp = false;
 	struct intel_encoder *encoder;
-	u32 temp;
 	int ret;
 	bool dither;
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		switch (encoder->type) {
-		case INTEL_OUTPUT_LVDS:
-			is_lvds = true;
-			break;
 		case INTEL_OUTPUT_DISPLAYPORT:
 			is_dp = true;
 			break;
@@ -5786,11 +5745,6 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 		num_connectors++;
 	}
 
-	if (is_cpu_edp)
-		intel_crtc->cpu_transcoder = TRANSCODER_EDP;
-	else
-		intel_crtc->cpu_transcoder = pipe;
-
 	/* We are not sure yet this won't happen. */
 	WARN(!HAS_PCH_LPT(dev), "Unexpected PCH type %d\n",
 	     INTEL_PCH_TYPE(dev));
@@ -5806,147 +5760,32 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 	if (!intel_ddi_pll_mode_set(crtc, adjusted_mode->clock))
 		return -EINVAL;
 
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) {
-		ok = ironlake_compute_clocks(crtc, adjusted_mode, &clock,
-					     &has_reduced_clock,
-					     &reduced_clock);
-		if (!ok) {
-			DRM_ERROR("Couldn't find PLL settings for mode!\n");
-			return -EINVAL;
-		}
-	}
-
 	/* Ensure that the cursor is valid for the new mode before changing... */
 	intel_crtc_update_cursor(crtc, true);
 
 	/* determine panel color depth */
 	dither = intel_choose_pipe_bpp_dither(crtc, fb, &intel_crtc->bpp,
 					      adjusted_mode);
-	if (is_lvds && dev_priv->lvds_dither)
-		dither = true;
 
 	DRM_DEBUG_KMS("Mode for pipe %d:\n", pipe);
 	drm_mode_debug_printmodeline(mode);
 
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) {
-		fp = clock.n << 16 | clock.m1 << 8 | clock.m2;
-		if (has_reduced_clock)
-			fp2 = reduced_clock.n << 16 | reduced_clock.m1 << 8 |
-			      reduced_clock.m2;
-
-		dpll = ironlake_compute_dpll(intel_crtc, adjusted_mode, &clock,
-					     fp);
-
-		/* CPU eDP is the only output that doesn't need a PCH PLL of its
-		 * own on pre-Haswell/LPT generation */
-		if (!is_cpu_edp) {
-			struct intel_pch_pll *pll;
-
-			pll = intel_get_pch_pll(intel_crtc, dpll, fp);
-			if (pll == NULL) {
-				DRM_DEBUG_DRIVER("failed to find PLL for pipe %d\n",
-						 pipe);
-				return -EINVAL;
-			}
-		} else
-			intel_put_pch_pll(intel_crtc);
-
-		/* The LVDS pin pair needs to be on before the DPLLs are
-		 * enabled.  This is an exception to the general rule that
-		 * mode_set doesn't turn things on.
-		 */
-		if (is_lvds) {
-			temp = I915_READ(PCH_LVDS);
-			temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
-			if (HAS_PCH_CPT(dev)) {
-				temp &= ~PORT_TRANS_SEL_MASK;
-				temp |= PORT_TRANS_SEL_CPT(pipe);
-			} else {
-				if (pipe == 1)
-					temp |= LVDS_PIPEB_SELECT;
-				else
-					temp &= ~LVDS_PIPEB_SELECT;
-			}
-
-			/* set the corresponsding LVDS_BORDER bit */
-			temp |= dev_priv->lvds_border_bits;
-			/* Set the B0-B3 data pairs corresponding to whether
-			 * we're going to set the DPLLs for dual-channel mode or
-			 * not.
-			 */
-			if (clock.p2 == 7)
-				temp |= LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP;
-			else
-				temp &= ~(LVDS_B0B3_POWER_UP |
-					  LVDS_CLKB_POWER_UP);
-
-			/* It would be nice to set 24 vs 18-bit mode
-			 * (LVDS_A3_POWER_UP) appropriately here, but we need to
-			 * look more thoroughly into how panels behave in the
-			 * two modes.
-			 */
-			temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY);
-			if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC)
-				temp |= LVDS_HSYNC_POLARITY;
-			if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC)
-				temp |= LVDS_VSYNC_POLARITY;
-			I915_WRITE(PCH_LVDS, temp);
-		}
-	}
-
-	if (is_dp && !is_cpu_edp) {
+	if (is_dp && !is_cpu_edp)
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
-	} else {
-		if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) {
-			/* For non-DP output, clear any trans DP clock recovery
-			 * setting.*/
-			I915_WRITE(TRANSDATA_M1(pipe), 0);
-			I915_WRITE(TRANSDATA_N1(pipe), 0);
-			I915_WRITE(TRANSDPLINK_M1(pipe), 0);
-			I915_WRITE(TRANSDPLINK_N1(pipe), 0);
-		}
-	}
 
 	intel_crtc->lowfreq_avail = false;
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) {
-		if (intel_crtc->pch_pll) {
-			I915_WRITE(intel_crtc->pch_pll->pll_reg, dpll);
-
-			/* Wait for the clocks to stabilize. */
-			POSTING_READ(intel_crtc->pch_pll->pll_reg);
-			udelay(150);
-
-			/* The pixel multiplier can only be updated once the
-			 * DPLL is enabled and the clocks are stable.
-			 *
-			 * So write it again.
-			 */
-			I915_WRITE(intel_crtc->pch_pll->pll_reg, dpll);
-		}
-
-		if (intel_crtc->pch_pll) {
-			if (is_lvds && has_reduced_clock && i915_powersave) {
-				I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp2);
-				intel_crtc->lowfreq_avail = true;
-			} else {
-				I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp);
-			}
-		}
-	}
 
 	intel_set_pipe_timings(intel_crtc, mode, adjusted_mode);
 
 	if (!is_dp || is_cpu_edp)
 		ironlake_set_m_n(crtc, mode, adjusted_mode);
 
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
-		if (is_cpu_edp)
-			ironlake_set_pll_edp(crtc, adjusted_mode->clock);
-
 	haswell_set_pipeconf(crtc, adjusted_mode, dither);
 
+	intel_set_pipe_csc(crtc, adjusted_mode);
+
 	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
 	POSTING_READ(DSPCNTR(plane));
 
 	ret = intel_pipe_set_base(crtc, x, y, fb);
@@ -5972,6 +5811,11 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 	int pipe = intel_crtc->pipe;
 	int ret;
 
+	if (IS_HASWELL(dev) && intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))
+		intel_crtc->cpu_transcoder = TRANSCODER_EDP;
+	else
+		intel_crtc->cpu_transcoder = pipe;
+
 	drm_vblank_pre_modeset(dev, pipe);
 
 	ret = dev_priv->display.crtc_mode_set(crtc, mode, adjusted_mode,
@@ -6068,6 +5912,7 @@ static void haswell_write_eld(struct drm_connector *connector,
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
 	struct drm_device *dev = crtc->dev;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t eldv;
 	uint32_t i;
 	int len;
@@ -6109,6 +5954,7 @@ static void haswell_write_eld(struct drm_connector *connector,
 	DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe));
 
 	eldv = AUDIO_ELD_VALID_A << (pipe * 4);
+	intel_crtc->eld_vld = true;
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
 		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
@@ -6344,6 +6190,8 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
 			cntl |= CURSOR_MODE_DISABLE;
 		}
+		if (IS_HASWELL(dev))
+			cntl |= CURSOR_PIPE_CSC_ENABLE;
 		I915_WRITE(CURCNTR_IVB(pipe), cntl);
 
 		intel_crtc->cursor_visible = visible;
@@ -6700,6 +6548,8 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	if (encoder->crtc) {
 		crtc = encoder->crtc;
 
+		mutex_lock(&crtc->mutex);
+
 		old->dpms_mode = connector->dpms;
 		old->load_detect_temp = false;
 
@@ -6729,6 +6579,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		return false;
 	}
 
+	mutex_lock(&crtc->mutex);
 	intel_encoder->new_crtc = to_intel_crtc(crtc);
 	to_intel_connector(connector)->new_encoder = intel_encoder;
 
@@ -6756,13 +6607,15 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		DRM_DEBUG_KMS("reusing fbdev for load-detection framebuffer\n");
 	if (IS_ERR(fb)) {
 		DRM_DEBUG_KMS("failed to allocate framebuffer for load-detection\n");
+		mutex_unlock(&crtc->mutex);
 		return false;
 	}
 
-	if (!intel_set_mode(crtc, mode, 0, 0, fb)) {
+	if (intel_set_mode(crtc, mode, 0, 0, fb)) {
 		DRM_DEBUG_KMS("failed to set mode on load-detect pipe\n");
 		if (old->release_fb)
 			old->release_fb->funcs->destroy(old->release_fb);
+		mutex_unlock(&crtc->mutex);
 		return false;
 	}
 
@@ -6777,27 +6630,31 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(connector);
 	struct drm_encoder *encoder = &intel_encoder->base;
+	struct drm_crtc *crtc = encoder->crtc;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 		      connector->base.id, drm_get_connector_name(connector),
 		      encoder->base.id, drm_get_encoder_name(encoder));
 
 	if (old->load_detect_temp) {
-		struct drm_crtc *crtc = encoder->crtc;
-
 		to_intel_connector(connector)->new_encoder = NULL;
 		intel_encoder->new_crtc = NULL;
 		intel_set_mode(crtc, NULL, 0, 0, NULL);
 
-		if (old->release_fb)
-			old->release_fb->funcs->destroy(old->release_fb);
+		if (old->release_fb) {
+			drm_framebuffer_unregister_private(old->release_fb);
+			drm_framebuffer_unreference(old->release_fb);
+		}
 
+		mutex_unlock(&crtc->mutex);
 		return;
 	}
 
 	/* Switch crtc and encoder back off if necessary */
 	if (old->dpms_mode != DRM_MODE_DPMS_ON)
 		connector->funcs->dpms(connector, old->dpms_mode);
+
+	mutex_unlock(&crtc->mutex);
 }
 
 /* Returns the clock of the currently programmed mode of the given pipe. */
@@ -6993,11 +6850,6 @@ void intel_mark_busy(struct drm_device *dev)
 
 void intel_mark_idle(struct drm_device *dev)
 {
-}
-
-void intel_mark_fb_busy(struct drm_i915_gem_object *obj)
-{
-	struct drm_device *dev = obj->base.dev;
 	struct drm_crtc *crtc;
 
 	if (!i915_powersave)
@@ -7007,12 +6859,11 @@ void intel_mark_fb_busy(struct drm_i915_gem_object *obj)
 		if (!crtc->fb)
 			continue;
 
-		if (to_intel_framebuffer(crtc->fb)->obj == obj)
-			intel_increase_pllclock(crtc);
+		intel_decrease_pllclock(crtc);
 	}
 }
 
-void intel_mark_fb_idle(struct drm_i915_gem_object *obj)
+void intel_mark_fb_busy(struct drm_i915_gem_object *obj)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_crtc *crtc;
@@ -7025,7 +6876,7 @@ void intel_mark_fb_idle(struct drm_i915_gem_object *obj)
 			continue;
 
 		if (to_intel_framebuffer(crtc->fb)->obj == obj)
-			intel_decrease_pllclock(crtc);
+			intel_increase_pllclock(crtc);
 	}
 }
 
@@ -7109,9 +6960,7 @@ static void do_intel_finish_page_flip(struct drm_device *dev,
 
 	obj = work->old_fb_obj;
 
-	atomic_clear_mask(1 << intel_crtc->plane,
-			  &obj->pending_flip.counter);
-	wake_up(&dev_priv->pending_flip_queue);
+	wake_up_all(&dev_priv->pending_flip_queue);
 
 	queue_work(dev_priv->wq, &work->work);
 
@@ -7474,11 +7323,8 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	work->enable_stall_check = true;
 
-	/* Block clients from rendering to the new back buffer until
-	 * the flip occurs and the object is no longer visible.
-	 */
-	atomic_add(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
 	atomic_inc(&intel_crtc->unpin_work_count);
+	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
 	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj);
 	if (ret)
@@ -7494,7 +7340,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 cleanup_pending:
 	atomic_dec(&intel_crtc->unpin_work_count);
-	atomic_sub(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
 	drm_gem_object_unreference(&work->old_fb_obj->base);
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
@@ -7514,7 +7359,6 @@ free_work:
 static struct drm_crtc_helper_funcs intel_helper_funcs = {
 	.mode_set_base_atomic = intel_pipe_set_base_atomic,
 	.load_lut = intel_crtc_load_lut,
-	.disable = intel_crtc_noop,
 };
 
 bool intel_encoder_check_is_cloned(struct intel_encoder *encoder)
@@ -7904,16 +7748,21 @@ intel_modeset_check_state(struct drm_device *dev)
 	}
 }
 
-bool intel_set_mode(struct drm_crtc *crtc,
-		    struct drm_display_mode *mode,
-		    int x, int y, struct drm_framebuffer *fb)
+int intel_set_mode(struct drm_crtc *crtc,
+		   struct drm_display_mode *mode,
+		   int x, int y, struct drm_framebuffer *fb)
 {
 	struct drm_device *dev = crtc->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct drm_display_mode *adjusted_mode, saved_mode, saved_hwmode;
+	struct drm_display_mode *adjusted_mode, *saved_mode, *saved_hwmode;
 	struct intel_crtc *intel_crtc;
 	unsigned disable_pipes, prepare_pipes, modeset_pipes;
-	bool ret = true;
+	int ret = 0;
+
+	saved_mode = kmalloc(2 * sizeof(*saved_mode), GFP_KERNEL);
+	if (!saved_mode)
+		return -ENOMEM;
+	saved_hwmode = saved_mode + 1;
 
 	intel_modeset_affected_pipes(crtc, &modeset_pipes,
 				     &prepare_pipes, &disable_pipes);
@@ -7924,8 +7773,8 @@ bool intel_set_mode(struct drm_crtc *crtc,
 	for_each_intel_crtc_masked(dev, disable_pipes, intel_crtc)
 		intel_crtc_disable(&intel_crtc->base);
 
-	saved_hwmode = crtc->hwmode;
-	saved_mode = crtc->mode;
+	*saved_hwmode = crtc->hwmode;
+	*saved_mode = crtc->mode;
 
 	/* Hack: Because we don't (yet) support global modeset on multiple
 	 * crtcs, we don't keep track of the new mode for more than one crtc.
@@ -7936,7 +7785,8 @@ bool intel_set_mode(struct drm_crtc *crtc,
 	if (modeset_pipes) {
 		adjusted_mode = intel_modeset_adjusted_mode(crtc, mode);
 		if (IS_ERR(adjusted_mode)) {
-			return false;
+			ret = PTR_ERR(adjusted_mode);
+			goto out;
 		}
 	}
 
@@ -7962,11 +7812,11 @@ bool intel_set_mode(struct drm_crtc *crtc,
 	 * on the DPLL.
 	 */
 	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		ret = !intel_crtc_mode_set(&intel_crtc->base,
-					   mode, adjusted_mode,
-					   x, y, fb);
-		if (!ret)
-		    goto done;
+		ret = intel_crtc_mode_set(&intel_crtc->base,
+					  mode, adjusted_mode,
+					  x, y, fb);
+		if (ret)
+			goto done;
 	}
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
@@ -7987,16 +7837,23 @@ bool intel_set_mode(struct drm_crtc *crtc,
 	/* FIXME: add subpixel order */
 done:
 	drm_mode_destroy(dev, adjusted_mode);
-	if (!ret && crtc->enabled) {
-		crtc->hwmode = saved_hwmode;
-		crtc->mode = saved_mode;
+	if (ret && crtc->enabled) {
+		crtc->hwmode = *saved_hwmode;
+		crtc->mode = *saved_mode;
 	} else {
 		intel_modeset_check_state(dev);
 	}
 
+out:
+	kfree(saved_mode);
 	return ret;
 }
 
+void intel_crtc_restore_mode(struct drm_crtc *crtc)
+{
+	intel_set_mode(crtc, &crtc->mode, crtc->x, crtc->y, crtc->fb);
+}
+
 #undef for_each_intel_crtc_masked
 
 static void intel_set_config_free(struct intel_set_config *config)
@@ -8109,7 +7966,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 	struct intel_encoder *encoder;
 	int count, ro;
 
-	/* The upper layers ensure that we either disabl a crtc or have a list
+	/* The upper layers ensure that we either disable a crtc or have a list
 	 * of connectors. For paranoia, double-check this. */
 	WARN_ON(!set->fb && (set->num_connectors != 0));
 	WARN_ON(set->fb && (set->num_connectors == 0));
@@ -8211,14 +8068,9 @@ static int intel_crtc_set_config(struct drm_mode_set *set)
 	BUG_ON(!set->crtc);
 	BUG_ON(!set->crtc->helper_private);
 
-	if (!set->mode)
-		set->fb = NULL;
-
-	/* The fb helper likes to play gross jokes with ->mode_set_config.
-	 * Unfortunately the crtc helper doesn't do much at all for this case,
-	 * so we have to cope with this madness until the fb helper is fixed up. */
-	if (set->fb && set->num_connectors == 0)
-		return 0;
+	/* Enforce sane interface api - has been abused by the fb helper. */
+	BUG_ON(!set->mode && set->fb);
+	BUG_ON(set->fb && set->num_connectors == 0);
 
 	if (set->fb) {
 		DRM_DEBUG_KMS("[CRTC:%d] [FB:%d] #connectors=%d (x y) (%i %i)\n",
@@ -8262,11 +8114,11 @@ static int intel_crtc_set_config(struct drm_mode_set *set)
 			drm_mode_debug_printmodeline(set->mode);
 		}
 
-		if (!intel_set_mode(set->crtc, set->mode,
-				    set->x, set->y, set->fb)) {
-			DRM_ERROR("failed to set mode on [CRTC:%d]\n",
-				  set->crtc->base.id);
-			ret = -EINVAL;
+		ret = intel_set_mode(set->crtc, set->mode,
+				     set->x, set->y, set->fb);
+		if (ret) {
+			DRM_ERROR("failed to set mode on [CRTC:%d], err = %d\n",
+				  set->crtc->base.id, ret);
 			goto fail;
 		}
 	} else if (config->fb_changed) {
@@ -8283,8 +8135,8 @@ fail:
 
 	/* Try to restore the config */
 	if (config->mode_changed &&
-	    !intel_set_mode(save_set.crtc, save_set.mode,
-			    save_set.x, save_set.y, save_set.fb))
+	    intel_set_mode(save_set.crtc, save_set.mode,
+			   save_set.x, save_set.y, save_set.fb))
 		DRM_ERROR("failed to restore config after modeset failure\n");
 
 out_config:
@@ -8303,7 +8155,7 @@ static const struct drm_crtc_funcs intel_crtc_funcs = {
 
 static void intel_cpu_pll_init(struct drm_device *dev)
 {
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		intel_ddi_pll_init(dev);
 }
 
@@ -8439,11 +8291,10 @@ static void intel_setup_outputs(struct drm_device *dev)
 		I915_WRITE(PFIT_CONTROL, 0);
 	}
 
-	if (!(IS_HASWELL(dev) &&
-	      (I915_READ(DDI_BUF_CTL(PORT_A)) & DDI_A_4_LANES)))
+	if (!(HAS_DDI(dev) && (I915_READ(DDI_BUF_CTL(PORT_A)) & DDI_A_4_LANES)))
 		intel_crt_init(dev);
 
-	if (IS_HASWELL(dev)) {
+	if (HAS_DDI(dev)) {
 		int found;
 
 		/* Haswell uses DDI functions to detect digital outputs */
@@ -8490,23 +8341,18 @@ static void intel_setup_outputs(struct drm_device *dev)
 		if (I915_READ(PCH_DP_D) & DP_DETECTED)
 			intel_dp_init(dev, PCH_DP_D, PORT_D);
 	} else if (IS_VALLEYVIEW(dev)) {
-		int found;
-
 		/* Check for built-in panel first. Shares lanes with HDMI on SDVOC */
-		if (I915_READ(DP_C) & DP_DETECTED)
-			intel_dp_init(dev, DP_C, PORT_C);
+		if (I915_READ(VLV_DISPLAY_BASE + DP_C) & DP_DETECTED)
+			intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
 
-		if (I915_READ(SDVOB) & PORT_DETECTED) {
-			/* SDVOB multiplex with HDMIB */
-			found = intel_sdvo_init(dev, SDVOB, true);
-			if (!found)
-				intel_hdmi_init(dev, SDVOB, PORT_B);
-			if (!found && (I915_READ(DP_B) & DP_DETECTED))
-				intel_dp_init(dev, DP_B, PORT_B);
+		if (I915_READ(VLV_DISPLAY_BASE + SDVOB) & PORT_DETECTED) {
+			intel_hdmi_init(dev, VLV_DISPLAY_BASE + SDVOB, PORT_B);
+			if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED)
+				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
 		}
 
-		if (I915_READ(SDVOC) & PORT_DETECTED)
-			intel_hdmi_init(dev, SDVOC, PORT_C);
+		if (I915_READ(VLV_DISPLAY_BASE + SDVOC) & PORT_DETECTED)
+			intel_hdmi_init(dev, VLV_DISPLAY_BASE + SDVOC, PORT_C);
 
 	} else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
 		bool found = false;
@@ -8666,14 +8512,15 @@ int intel_framebuffer_init(struct drm_device *dev,
 	if (mode_cmd->offsets[0] != 0)
 		return -EINVAL;
 
+	drm_helper_mode_fill_fb_struct(&intel_fb->base, mode_cmd);
+	intel_fb->obj = obj;
+
 	ret = drm_framebuffer_init(dev, &intel_fb->base, &intel_fb_funcs);
 	if (ret) {
 		DRM_ERROR("framebuffer init failed %d\n", ret);
 		return ret;
 	}
 
-	drm_helper_mode_fill_fb_struct(&intel_fb->base, mode_cmd);
-	intel_fb->obj = obj;
 	return 0;
 }
 
@@ -8703,7 +8550,7 @@ static void intel_init_display(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	/* We always want a DPMS function */
-	if (IS_HASWELL(dev)) {
+	if (HAS_DDI(dev)) {
 		dev_priv->display.crtc_mode_set = haswell_crtc_mode_set;
 		dev_priv->display.crtc_enable = haswell_crtc_enable;
 		dev_priv->display.crtc_disable = haswell_crtc_disable;
@@ -8765,8 +8612,9 @@ static void intel_init_display(struct drm_device *dev)
 		} else if (IS_HASWELL(dev)) {
 			dev_priv->display.fdi_link_train = hsw_fdi_link_train;
 			dev_priv->display.write_eld = haswell_write_eld;
-		} else
-			dev_priv->display.update_wm = NULL;
+			dev_priv->display.modeset_global_resources =
+				haswell_modeset_global_resources;
+		}
 	} else if (IS_G4X(dev)) {
 		dev_priv->display.write_eld = g4x_write_eld;
 	}
@@ -8888,6 +8736,18 @@ static struct intel_quirk intel_quirks[] = {
 
 	/* Acer Aspire 5734Z must invert backlight brightness */
 	{ 0x2a42, 0x1025, 0x0459, quirk_invert_brightness },
+
+	/* Acer/eMachines G725 */
+	{ 0x2a42, 0x1025, 0x0210, quirk_invert_brightness },
+
+	/* Acer/eMachines e725 */
+	{ 0x2a42, 0x1025, 0x0212, quirk_invert_brightness },
+
+	/* Acer/Packard Bell NCL20 */
+	{ 0x2a42, 0x1025, 0x034b, quirk_invert_brightness },
+
+	/* Acer Aspire 4736Z */
+	{ 0x2a42, 0x1025, 0x0260, quirk_invert_brightness },
 };
 
 static void intel_init_quirks(struct drm_device *dev)
@@ -8916,12 +8776,7 @@ static void i915_disable_vga(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u8 sr1;
-	u32 vga_reg;
-
-	if (HAS_PCH_SPLIT(dev))
-		vga_reg = CPU_VGACNTRL;
-	else
-		vga_reg = VGACNTRL;
+	u32 vga_reg = i915_vgacntrl_reg(dev);
 
 	vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO);
 	outb(SR01, VGA_SR_INDEX);
@@ -8936,10 +8791,7 @@ static void i915_disable_vga(struct drm_device *dev)
 
 void intel_modeset_init_hw(struct drm_device *dev)
 {
-	/* We attempt to init the necessary power wells early in the initialization
-	 * time, so the subsystems that expect power to be enabled can work.
-	 */
-	intel_init_power_wells(dev);
+	intel_init_power_well(dev);
 
 	intel_prepare_ddi(dev);
 
@@ -8981,7 +8833,7 @@ void intel_modeset_init(struct drm_device *dev)
 		dev->mode_config.max_width = 8192;
 		dev->mode_config.max_height = 8192;
 	}
-	dev->mode_config.fb_base = dev_priv->mm.gtt_base_addr;
+	dev->mode_config.fb_base = dev_priv->gtt.mappable_base;
 
 	DRM_DEBUG_KMS("%d display pipe%s available.\n",
 		      dev_priv->num_pipe, dev_priv->num_pipe > 1 ? "s" : "");
@@ -8999,6 +8851,9 @@ void intel_modeset_init(struct drm_device *dev)
 	/* Just disable it once at startup */
 	i915_disable_vga(dev);
 	intel_setup_outputs(dev);
+
+	/* Just in case the BIOS is doing something questionable. */
+	intel_disable_fbc(dev);
 }
 
 static void
@@ -9180,20 +9035,14 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	 * the crtc fixup. */
 }
 
-static void i915_redisable_vga(struct drm_device *dev)
+void i915_redisable_vga(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 vga_reg;
-
-	if (HAS_PCH_SPLIT(dev))
-		vga_reg = CPU_VGACNTRL;
-	else
-		vga_reg = VGACNTRL;
+	u32 vga_reg = i915_vgacntrl_reg(dev);
 
 	if (I915_READ(vga_reg) != VGA_DISP_DISABLE) {
 		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
-		I915_WRITE(vga_reg, VGA_DISP_DISABLE);
-		POSTING_READ(vga_reg);
+		i915_disable_vga(dev);
 	}
 }
 
@@ -9209,7 +9058,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
 
-	if (IS_HASWELL(dev)) {
+	if (HAS_DDI(dev)) {
 		tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP));
 
 		if (tmp & TRANS_DDI_FUNC_ENABLE) {
@@ -9250,7 +9099,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 			      crtc->active ? "enabled" : "disabled");
 	}
 
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		intel_ddi_setup_hw_pll_state(dev);
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
@@ -9301,9 +9150,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 
 	if (force_restore) {
 		for_each_pipe(pipe) {
-			crtc = to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
-			intel_set_mode(&crtc->base, &crtc->base.mode,
-				       crtc->base.x, crtc->base.y, crtc->base.fb);
+			intel_crtc_restore_mode(dev_priv->pipe_to_crtc_mapping[pipe]);
 		}
 
 		i915_redisable_vga(dev);
@@ -9367,6 +9214,8 @@ void intel_modeset_cleanup(struct drm_device *dev)
 	flush_scheduled_work();
 
 	drm_mode_config_cleanup(dev);
+
+	intel_cleanup_overlay(dev);
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index fb3715b4b09d..f61cb7998c72 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -148,15 +148,6 @@ intel_dp_max_link_bw(struct intel_dp *intel_dp)
 	return max_link_bw;
 }
 
-static int
-intel_dp_link_clock(uint8_t link_bw)
-{
-	if (link_bw == DP_LINK_BW_2_7)
-		return 270000;
-	else
-		return 162000;
-}
-
 /*
  * The units on the numbers in the next two are... bizarre.  Examples will
  * make it clearer; this one parallels an example in the eDP spec.
@@ -191,7 +182,8 @@ intel_dp_adjust_dithering(struct intel_dp *intel_dp,
 			  struct drm_display_mode *mode,
 			  bool adjust_mode)
 {
-	int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp));
+	int max_link_clock =
+		drm_dp_bw_code_to_link_rate(intel_dp_max_link_bw(intel_dp));
 	int max_lanes = drm_dp_max_lane_count(intel_dp->dpcd);
 	int max_rate, mode_rate;
 
@@ -330,6 +322,48 @@ intel_dp_check_edp(struct intel_dp *intel_dp)
 	}
 }
 
+static uint32_t
+intel_dp_aux_wait_done(struct intel_dp *intel_dp, bool has_aux_irq)
+{
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t ch_ctl = intel_dp->output_reg + 0x10;
+	uint32_t status;
+	bool done;
+
+	if (IS_HASWELL(dev)) {
+		switch (intel_dig_port->port) {
+		case PORT_A:
+			ch_ctl = DPA_AUX_CH_CTL;
+			break;
+		case PORT_B:
+			ch_ctl = PCH_DPB_AUX_CH_CTL;
+			break;
+		case PORT_C:
+			ch_ctl = PCH_DPC_AUX_CH_CTL;
+			break;
+		case PORT_D:
+			ch_ctl = PCH_DPD_AUX_CH_CTL;
+			break;
+		default:
+			BUG();
+		}
+	}
+
+#define C (((status = I915_READ_NOTRACE(ch_ctl)) & DP_AUX_CH_CTL_SEND_BUSY) == 0)
+	if (has_aux_irq)
+		done = wait_event_timeout(dev_priv->gmbus_wait_queue, C, 10);
+	else
+		done = wait_for_atomic(C, 10) == 0;
+	if (!done)
+		DRM_ERROR("dp aux hw did not signal timeout (has irq: %i)!\n",
+			  has_aux_irq);
+#undef C
+
+	return status;
+}
+
 static int
 intel_dp_aux_ch(struct intel_dp *intel_dp,
 		uint8_t *send, int send_bytes,
@@ -341,11 +375,17 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t ch_ctl = output_reg + 0x10;
 	uint32_t ch_data = ch_ctl + 4;
-	int i;
-	int recv_bytes;
+	int i, ret, recv_bytes;
 	uint32_t status;
 	uint32_t aux_clock_divider;
 	int try, precharge;
+	bool has_aux_irq = INTEL_INFO(dev)->gen >= 5 && !IS_VALLEYVIEW(dev);
+
+	/* dp aux is extremely sensitive to irq latency, hence request the
+	 * lowest possible wakeup latency and so prevent the cpu from going into
+	 * deep sleep states.
+	 */
+	pm_qos_update_request(&dev_priv->pm_qos, 0);
 
 	if (IS_HASWELL(dev)) {
 		switch (intel_dig_port->port) {
@@ -379,7 +419,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	 * clock divider.
 	 */
 	if (is_cpu_edp(intel_dp)) {
-		if (IS_HASWELL(dev))
+		if (HAS_DDI(dev))
 			aux_clock_divider = intel_ddi_get_cdclk_freq(dev_priv) >> 1;
 		else if (IS_VALLEYVIEW(dev))
 			aux_clock_divider = 100;
@@ -399,7 +439,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 
 	/* Try to wait for any previous AUX channel activity */
 	for (try = 0; try < 3; try++) {
-		status = I915_READ(ch_ctl);
+		status = I915_READ_NOTRACE(ch_ctl);
 		if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0)
 			break;
 		msleep(1);
@@ -408,7 +448,8 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	if (try == 3) {
 		WARN(1, "dp_aux_ch not started status 0x%08x\n",
 		     I915_READ(ch_ctl));
-		return -EBUSY;
+		ret = -EBUSY;
+		goto out;
 	}
 
 	/* Must try at least 3 times according to DP spec */
@@ -421,6 +462,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 		/* Send the command and wait for it to complete */
 		I915_WRITE(ch_ctl,
 			   DP_AUX_CH_CTL_SEND_BUSY |
+			   (has_aux_irq ? DP_AUX_CH_CTL_INTERRUPT : 0) |
 			   DP_AUX_CH_CTL_TIME_OUT_400us |
 			   (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
 			   (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
@@ -428,12 +470,8 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 			   DP_AUX_CH_CTL_DONE |
 			   DP_AUX_CH_CTL_TIME_OUT_ERROR |
 			   DP_AUX_CH_CTL_RECEIVE_ERROR);
-		for (;;) {
-			status = I915_READ(ch_ctl);
-			if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0)
-				break;
-			udelay(100);
-		}
+
+		status = intel_dp_aux_wait_done(intel_dp, has_aux_irq);
 
 		/* Clear done status and any errors */
 		I915_WRITE(ch_ctl,
@@ -451,7 +489,8 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 
 	if ((status & DP_AUX_CH_CTL_DONE) == 0) {
 		DRM_ERROR("dp_aux_ch not done status 0x%08x\n", status);
-		return -EBUSY;
+		ret = -EBUSY;
+		goto out;
 	}
 
 	/* Check for timeout or receive error.
@@ -459,14 +498,16 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	 */
 	if (status & DP_AUX_CH_CTL_RECEIVE_ERROR) {
 		DRM_ERROR("dp_aux_ch receive error status 0x%08x\n", status);
-		return -EIO;
+		ret = -EIO;
+		goto out;
 	}
 
 	/* Timeouts occur when the device isn't connected, so they're
 	 * "normal" -- don't fill the kernel log with these */
 	if (status & DP_AUX_CH_CTL_TIME_OUT_ERROR) {
 		DRM_DEBUG_KMS("dp_aux_ch timeout status 0x%08x\n", status);
-		return -ETIMEDOUT;
+		ret = -ETIMEDOUT;
+		goto out;
 	}
 
 	/* Unload any bytes sent back from the other side */
@@ -479,7 +520,11 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 		unpack_aux(I915_READ(ch_data + i),
 			   recv + i, recv_bytes - i);
 
-	return recv_bytes;
+	ret = recv_bytes;
+out:
+	pm_qos_update_request(&dev_priv->pm_qos, PM_QOS_DEFAULT_VALUE);
+
+	return ret;
 }
 
 /* Write data to the aux channel in native mode */
@@ -718,16 +763,35 @@ intel_dp_mode_fixup(struct drm_encoder *encoder,
 		return false;
 
 	bpp = adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 24;
+
+	if (intel_dp->color_range_auto) {
+		/*
+		 * See:
+		 * CEA-861-E - 5.1 Default Encoding Parameters
+		 * VESA DisplayPort Ver.1.2a - 5.1.1.1 Video Colorimetry
+		 */
+		if (bpp != 18 && drm_match_cea_mode(adjusted_mode) > 1)
+			intel_dp->color_range = DP_COLOR_RANGE_16_235;
+		else
+			intel_dp->color_range = 0;
+	}
+
+	if (intel_dp->color_range)
+		adjusted_mode->private_flags |= INTEL_MODE_LIMITED_COLOR_RANGE;
+
 	mode_rate = intel_dp_link_required(adjusted_mode->clock, bpp);
 
 	for (clock = 0; clock <= max_clock; clock++) {
 		for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
-			int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count);
+			int link_bw_clock =
+				drm_dp_bw_code_to_link_rate(bws[clock]);
+			int link_avail = intel_dp_max_data_rate(link_bw_clock,
+								lane_count);
 
 			if (mode_rate <= link_avail) {
 				intel_dp->link_bw = bws[clock];
 				intel_dp->lane_count = lane_count;
-				adjusted_mode->clock = intel_dp_link_clock(intel_dp->link_bw);
+				adjusted_mode->clock = link_bw_clock;
 				DRM_DEBUG_KMS("DP link bw %02x lane "
 						"count %d clock %d bpp %d\n",
 				       intel_dp->link_bw, intel_dp->lane_count,
@@ -742,39 +806,6 @@ intel_dp_mode_fixup(struct drm_encoder *encoder,
 	return false;
 }
 
-struct intel_dp_m_n {
-	uint32_t	tu;
-	uint32_t	gmch_m;
-	uint32_t	gmch_n;
-	uint32_t	link_m;
-	uint32_t	link_n;
-};
-
-static void
-intel_reduce_ratio(uint32_t *num, uint32_t *den)
-{
-	while (*num > 0xffffff || *den > 0xffffff) {
-		*num >>= 1;
-		*den >>= 1;
-	}
-}
-
-static void
-intel_dp_compute_m_n(int bpp,
-		     int nlanes,
-		     int pixel_clock,
-		     int link_clock,
-		     struct intel_dp_m_n *m_n)
-{
-	m_n->tu = 64;
-	m_n->gmch_m = (pixel_clock * bpp) >> 3;
-	m_n->gmch_n = link_clock * nlanes;
-	intel_reduce_ratio(&m_n->gmch_m, &m_n->gmch_n);
-	m_n->link_m = pixel_clock;
-	m_n->link_n = link_clock;
-	intel_reduce_ratio(&m_n->link_m, &m_n->link_n);
-}
-
 void
 intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 		 struct drm_display_mode *adjusted_mode)
@@ -785,7 +816,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int lane_count = 4;
-	struct intel_dp_m_n m_n;
+	struct intel_link_m_n m_n;
 	int pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->cpu_transcoder;
 
@@ -808,8 +839,8 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 	 * the number of bytes_per_pixel post-LUT, which we always
 	 * set up for 8-bits of R/G/B, or 3 bytes total.
 	 */
-	intel_dp_compute_m_n(intel_crtc->bpp, lane_count,
-			     mode->clock, adjusted_mode->clock, &m_n);
+	intel_link_compute_m_n(intel_crtc->bpp, lane_count,
+			       mode->clock, adjusted_mode->clock, &m_n);
 
 	if (IS_HASWELL(dev)) {
 		I915_WRITE(PIPE_DATA_M1(cpu_transcoder),
@@ -851,6 +882,32 @@ void intel_dp_init_link_config(struct intel_dp *intel_dp)
 	}
 }
 
+static void ironlake_set_pll_edp(struct drm_crtc *crtc, int clock)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 dpa_ctl;
+
+	DRM_DEBUG_KMS("eDP PLL enable for clock %d\n", clock);
+	dpa_ctl = I915_READ(DP_A);
+	dpa_ctl &= ~DP_PLL_FREQ_MASK;
+
+	if (clock < 200000) {
+		/* For a long time we've carried around a ILK-DevA w/a for the
+		 * 160MHz clock. If we're really unlucky, it's still required.
+		 */
+		DRM_DEBUG_KMS("160MHz cpu eDP clock, might need ilk devA w/a\n");
+		dpa_ctl |= DP_PLL_FREQ_160MHZ;
+	} else {
+		dpa_ctl |= DP_PLL_FREQ_270MHZ;
+	}
+
+	I915_WRITE(DP_A, dpa_ctl);
+
+	POSTING_READ(DP_A);
+	udelay(500);
+}
+
 static void
 intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 		  struct drm_display_mode *adjusted_mode)
@@ -926,7 +983,8 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 		else
 			intel_dp->DP |= DP_PLL_FREQ_270MHZ;
 	} else if (!HAS_PCH_CPT(dev) || is_cpu_edp(intel_dp)) {
-		intel_dp->DP |= intel_dp->color_range;
+		if (!HAS_PCH_SPLIT(dev))
+			intel_dp->DP |= intel_dp->color_range;
 
 		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
 			intel_dp->DP |= DP_SYNC_HS_HIGH;
@@ -950,6 +1008,9 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 	} else {
 		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 	}
+
+	if (is_cpu_edp(intel_dp))
+		ironlake_set_pll_edp(crtc, adjusted_mode->clock);
 }
 
 #define IDLE_ON_MASK		(PP_ON | 0 	  | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
@@ -1057,6 +1118,8 @@ static void ironlake_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 pp;
 
+	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+
 	if (!intel_dp->want_panel_vdd && ironlake_edp_have_panel_vdd(intel_dp)) {
 		pp = ironlake_get_pp_control(dev_priv);
 		pp &= ~EDP_FORCE_VDD;
@@ -1543,7 +1606,7 @@ intel_get_adjust_train(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_ST
 }
 
 static uint32_t
-intel_dp_signal_levels(uint8_t train_set)
+intel_gen4_signal_levels(uint8_t train_set)
 {
 	uint32_t	signal_levels = 0;
 
@@ -1641,7 +1704,7 @@ intel_gen7_edp_signal_levels(uint8_t train_set)
 
 /* Gen7.5's (HSW) DP voltage swing and pre-emphasis control */
 static uint32_t
-intel_dp_signal_levels_hsw(uint8_t train_set)
+intel_hsw_signal_levels(uint8_t train_set)
 {
 	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
 					 DP_TRAIN_PRE_EMPHASIS_MASK);
@@ -1673,6 +1736,34 @@ intel_dp_signal_levels_hsw(uint8_t train_set)
 	}
 }
 
+/* Properly updates "DP" with the correct signal levels. */
+static void
+intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
+{
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	uint32_t signal_levels, mask;
+	uint8_t train_set = intel_dp->train_set[0];
+
+	if (IS_HASWELL(dev)) {
+		signal_levels = intel_hsw_signal_levels(train_set);
+		mask = DDI_BUF_EMP_MASK;
+	} else if (IS_GEN7(dev) && is_cpu_edp(intel_dp) && !IS_VALLEYVIEW(dev)) {
+		signal_levels = intel_gen7_edp_signal_levels(train_set);
+		mask = EDP_LINK_TRAIN_VOL_EMP_MASK_IVB;
+	} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
+		signal_levels = intel_gen6_edp_signal_levels(train_set);
+		mask = EDP_LINK_TRAIN_VOL_EMP_MASK_SNB;
+	} else {
+		signal_levels = intel_gen4_signal_levels(train_set);
+		mask = DP_VOLTAGE_MASK | DP_PRE_EMPHASIS_MASK;
+	}
+
+	DRM_DEBUG_KMS("Using signal levels %08x\n", signal_levels);
+
+	*DP = (*DP & ~mask) | signal_levels;
+}
+
 static bool
 intel_dp_set_link_train(struct intel_dp *intel_dp,
 			uint32_t dp_reg_value,
@@ -1696,14 +1787,18 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 		temp &= ~DP_TP_CTL_LINK_TRAIN_MASK;
 		switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
 		case DP_TRAINING_PATTERN_DISABLE:
-			temp |= DP_TP_CTL_LINK_TRAIN_IDLE;
-			I915_WRITE(DP_TP_CTL(port), temp);
 
-			if (wait_for((I915_READ(DP_TP_STATUS(port)) &
-				      DP_TP_STATUS_IDLE_DONE), 1))
-				DRM_ERROR("Timed out waiting for DP idle patterns\n");
+			if (port != PORT_A) {
+				temp |= DP_TP_CTL_LINK_TRAIN_IDLE;
+				I915_WRITE(DP_TP_CTL(port), temp);
+
+				if (wait_for((I915_READ(DP_TP_STATUS(port)) &
+					      DP_TP_STATUS_IDLE_DONE), 1))
+					DRM_ERROR("Timed out waiting for DP idle patterns\n");
+
+				temp &= ~DP_TP_CTL_LINK_TRAIN_MASK;
+			}
 
-			temp &= ~DP_TP_CTL_LINK_TRAIN_MASK;
 			temp |= DP_TP_CTL_LINK_TRAIN_NORMAL;
 
 			break;
@@ -1791,7 +1886,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	int voltage_tries, loop_tries;
 	uint32_t DP = intel_dp->DP;
 
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		intel_ddi_prepare_link_retrain(encoder);
 
 	/* Write the link configuration data */
@@ -1809,24 +1904,8 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	for (;;) {
 		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
 		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
-		uint32_t    signal_levels;
-
-		if (IS_HASWELL(dev)) {
-			signal_levels = intel_dp_signal_levels_hsw(
-							intel_dp->train_set[0]);
-			DP = (DP & ~DDI_BUF_EMP_MASK) | signal_levels;
-		} else if (IS_GEN7(dev) && is_cpu_edp(intel_dp) && !IS_VALLEYVIEW(dev)) {
-			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
-		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
-			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
-		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
-		}
-		DRM_DEBUG_KMS("training pattern 1 signal levels %08x\n",
-			      signal_levels);
+
+		intel_dp_set_signal_levels(intel_dp, &DP);
 
 		/* Set training pattern 1 */
 		if (!intel_dp_set_link_train(intel_dp, DP,
@@ -1882,7 +1961,6 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 void
 intel_dp_complete_link_train(struct intel_dp *intel_dp)
 {
-	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	bool channel_eq = false;
 	int tries, cr_tries;
 	uint32_t DP = intel_dp->DP;
@@ -1892,8 +1970,6 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 	cr_tries = 0;
 	channel_eq = false;
 	for (;;) {
-		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
-		uint32_t    signal_levels;
 		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 
 		if (cr_tries > 5) {
@@ -1902,19 +1978,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		if (IS_HASWELL(dev)) {
-			signal_levels = intel_dp_signal_levels_hsw(intel_dp->train_set[0]);
-			DP = (DP & ~DDI_BUF_EMP_MASK) | signal_levels;
-		} else if (IS_GEN7(dev) && is_cpu_edp(intel_dp) && !IS_VALLEYVIEW(dev)) {
-			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
-		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
-			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
-		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
-			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
-		}
+		intel_dp_set_signal_levels(intel_dp, &DP);
 
 		/* channel eq pattern */
 		if (!intel_dp_set_link_train(intel_dp, DP,
@@ -1964,6 +2028,8 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(intel_dig_port->base.base.crtc);
 	uint32_t DP = intel_dp->DP;
 
 	/*
@@ -1981,7 +2047,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	 *   intel_ddi_prepare_link_retrain will take care of redoing the link
 	 *   train.
 	 */
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		return;
 
 	if (WARN_ON((I915_READ(intel_dp->output_reg) & DP_PORT_EN) == 0))
@@ -1998,7 +2064,8 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	}
 	POSTING_READ(intel_dp->output_reg);
 
-	msleep(17);
+	/* We don't really know why we're doing this */
+	intel_wait_for_vblank(dev, intel_crtc->pipe);
 
 	if (HAS_PCH_IBX(dev) &&
 	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
@@ -2018,19 +2085,14 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 		/* Changes to enable or select take place the vblank
 		 * after being written.
 		 */
-		if (crtc == NULL) {
-			/* We can arrive here never having been attached
-			 * to a CRTC, for instance, due to inheriting
-			 * random state from the BIOS.
-			 *
-			 * If the pipe is not running, play safe and
-			 * wait for the clocks to stabilise before
-			 * continuing.
-			 */
+		if (WARN_ON(crtc == NULL)) {
+			/* We should never try to disable a port without a crtc
+			 * attached. For paranoia keep the code around for a
+			 * bit. */
 			POSTING_READ(intel_dp->output_reg);
 			msleep(50);
 		} else
-			intel_wait_for_vblank(dev, to_intel_crtc(crtc)->pipe);
+			intel_wait_for_vblank(dev, intel_crtc->pipe);
 	}
 
 	DP &= ~DP_AUDIO_OUTPUT_ENABLE;
@@ -2042,10 +2104,16 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 static bool
 intel_dp_get_dpcd(struct intel_dp *intel_dp)
 {
+	char dpcd_hex_dump[sizeof(intel_dp->dpcd) * 3];
+
 	if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd,
 					   sizeof(intel_dp->dpcd)) == 0)
 		return false; /* aux transfer failed */
 
+	hex_dump_to_buffer(intel_dp->dpcd, sizeof(intel_dp->dpcd),
+			   32, 1, dpcd_hex_dump, sizeof(dpcd_hex_dump), false);
+	DRM_DEBUG_KMS("DPCD: %s\n", dpcd_hex_dump);
+
 	if (intel_dp->dpcd[DP_DPCD_REV] == 0)
 		return false; /* DPCD not present */
 
@@ -2206,6 +2274,8 @@ static enum drm_connector_status
 ironlake_dp_detect(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	enum drm_connector_status status;
 
 	/* Can't disconnect eDP, but you can close the lid... */
@@ -2216,6 +2286,9 @@ ironlake_dp_detect(struct intel_dp *intel_dp)
 		return status;
 	}
 
+	if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
+		return connector_status_disconnected;
+
 	return intel_dp_detect_dpcd(intel_dp);
 }
 
@@ -2224,17 +2297,18 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	uint32_t bit;
 
-	switch (intel_dp->output_reg) {
-	case DP_B:
-		bit = DPB_HOTPLUG_LIVE_STATUS;
+	switch (intel_dig_port->port) {
+	case PORT_B:
+		bit = PORTB_HOTPLUG_LIVE_STATUS;
 		break;
-	case DP_C:
-		bit = DPC_HOTPLUG_LIVE_STATUS;
+	case PORT_C:
+		bit = PORTC_HOTPLUG_LIVE_STATUS;
 		break;
-	case DP_D:
-		bit = DPD_HOTPLUG_LIVE_STATUS;
+	case PORT_D:
+		bit = PORTD_HOTPLUG_LIVE_STATUS;
 		break;
 	default:
 		return connector_status_unknown;
@@ -2290,13 +2364,6 @@ intel_dp_get_edid_modes(struct drm_connector *connector, struct i2c_adapter *ada
 	return intel_ddc_get_modes(connector, adapter);
 }
 
-
-/**
- * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect DP connection.
- *
- * \return true if DP port is connected.
- * \return false if DP port is disconnected.
- */
 static enum drm_connector_status
 intel_dp_detect(struct drm_connector *connector, bool force)
 {
@@ -2306,7 +2373,6 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	struct drm_device *dev = connector->dev;
 	enum drm_connector_status status;
 	struct edid *edid = NULL;
-	char dpcd_hex_dump[sizeof(intel_dp->dpcd) * 3];
 
 	intel_dp->has_audio = false;
 
@@ -2315,10 +2381,6 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	else
 		status = g4x_dp_detect(intel_dp);
 
-	hex_dump_to_buffer(intel_dp->dpcd, sizeof(intel_dp->dpcd),
-			   32, 1, dpcd_hex_dump, sizeof(dpcd_hex_dump), false);
-	DRM_DEBUG_KMS("DPCD: %s\n", dpcd_hex_dump);
-
 	if (status != connector_status_connected)
 		return status;
 
@@ -2419,10 +2481,21 @@ intel_dp_set_property(struct drm_connector *connector,
 	}
 
 	if (property == dev_priv->broadcast_rgb_property) {
-		if (val == !!intel_dp->color_range)
-			return 0;
-
-		intel_dp->color_range = val ? DP_COLOR_RANGE_16_235 : 0;
+		switch (val) {
+		case INTEL_BROADCAST_RGB_AUTO:
+			intel_dp->color_range_auto = true;
+			break;
+		case INTEL_BROADCAST_RGB_FULL:
+			intel_dp->color_range_auto = false;
+			intel_dp->color_range = 0;
+			break;
+		case INTEL_BROADCAST_RGB_LIMITED:
+			intel_dp->color_range_auto = false;
+			intel_dp->color_range = DP_COLOR_RANGE_16_235;
+			break;
+		default:
+			return -EINVAL;
+		}
 		goto done;
 	}
 
@@ -2445,11 +2518,8 @@ intel_dp_set_property(struct drm_connector *connector,
 	return -EINVAL;
 
 done:
-	if (intel_encoder->base.crtc) {
-		struct drm_crtc *crtc = intel_encoder->base.crtc;
-		intel_set_mode(crtc, &crtc->mode,
-			       crtc->x, crtc->y, crtc->fb);
-	}
+	if (intel_encoder->base.crtc)
+		intel_crtc_restore_mode(intel_encoder->base.crtc);
 
 	return 0;
 }
@@ -2491,7 +2561,6 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs intel_dp_helper_funcs = {
 	.mode_fixup = intel_dp_mode_fixup,
 	.mode_set = intel_dp_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_dp_connector_funcs = {
@@ -2566,6 +2635,7 @@ intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connect
 
 	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
+	intel_dp->color_range_auto = true;
 
 	if (is_edp(intel_dp)) {
 		drm_mode_create_scaling_mode_property(connector->dev);
@@ -2755,7 +2825,7 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 	drm_sysfs_connector_add(connector);
 
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		intel_connector->get_hw_state = intel_ddi_connector_get_hw_state;
 	else
 		intel_connector->get_hw_state = intel_connector_get_hw_state;
@@ -2767,15 +2837,15 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 		name = "DPDDC-A";
 		break;
 	case PORT_B:
-		dev_priv->hotplug_supported_mask |= DPB_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTB_HOTPLUG_INT_STATUS;
 		name = "DPDDC-B";
 		break;
 	case PORT_C:
-		dev_priv->hotplug_supported_mask |= DPC_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTC_HOTPLUG_INT_STATUS;
 		name = "DPDDC-C";
 		break;
 	case PORT_D:
-		dev_priv->hotplug_supported_mask |= DPD_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTD_HOTPLUG_INT_STATUS;
 		name = "DPDDC-D";
 		break;
 	default:
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 8a1bd4a3ad0d..07ebac6fe8ca 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -109,6 +109,11 @@
  * timings in the mode to prevent the crtc fixup from overwriting them.
  * Currently only lvds needs that. */
 #define INTEL_MODE_CRTC_TIMINGS_SET (0x20)
+/*
+ * Set when limited 16-235 (as opposed to full 0-255) RGB color range is
+ * to be used.
+ */
+#define INTEL_MODE_LIMITED_COLOR_RANGE (0x40)
 
 static inline void
 intel_mode_set_pixel_multiplier(struct drm_display_mode *mode,
@@ -153,6 +158,7 @@ struct intel_encoder {
 	bool cloneable;
 	bool connectors_active;
 	void (*hot_plug)(struct intel_encoder *);
+	void (*pre_pll_enable)(struct intel_encoder *);
 	void (*pre_enable)(struct intel_encoder *);
 	void (*enable)(struct intel_encoder *);
 	void (*disable)(struct intel_encoder *);
@@ -205,6 +211,7 @@ struct intel_crtc {
 	 * some outputs connected to this crtc.
 	 */
 	bool active;
+	bool eld_vld;
 	bool primary_disabled; /* is the crtc obscured by a plane? */
 	bool lowfreq_avail;
 	struct intel_overlay *overlay;
@@ -228,6 +235,9 @@ struct intel_crtc {
 	/* We can share PLLs across outputs if the timings match */
 	struct intel_pch_pll *pch_pll;
 	uint32_t ddi_pll_sel;
+
+	/* reset counter value when the last flip was submitted */
+	unsigned int reset_counter;
 };
 
 struct intel_plane {
@@ -283,6 +293,9 @@ struct cxsr_latency {
 #define DIP_LEN_AVI     13
 #define DIP_AVI_PR_1    0
 #define DIP_AVI_PR_2    1
+#define DIP_AVI_RGB_QUANT_RANGE_DEFAULT	(0 << 2)
+#define DIP_AVI_RGB_QUANT_RANGE_LIMITED	(1 << 2)
+#define DIP_AVI_RGB_QUANT_RANGE_FULL	(2 << 2)
 
 #define DIP_TYPE_SPD	0x83
 #define DIP_VERSION_SPD	0x1
@@ -337,9 +350,11 @@ struct intel_hdmi {
 	u32 sdvox_reg;
 	int ddc_bus;
 	uint32_t color_range;
+	bool color_range_auto;
 	bool has_hdmi_sink;
 	bool has_audio;
 	enum hdmi_force_audio force_audio;
+	bool rgb_quant_range_selectable;
 	void (*write_infoframe)(struct drm_encoder *encoder,
 				struct dip_infoframe *frame);
 	void (*set_infoframes)(struct drm_encoder *encoder,
@@ -356,6 +371,7 @@ struct intel_dp {
 	bool has_audio;
 	enum hdmi_force_audio force_audio;
 	uint32_t color_range;
+	bool color_range_auto;
 	uint8_t link_bw;
 	uint8_t lane_count;
 	uint8_t dpcd[DP_RECEIVER_CAP_SIZE];
@@ -377,6 +393,7 @@ struct intel_dp {
 struct intel_digital_port {
 	struct intel_encoder base;
 	enum port port;
+	u32 port_reversal;
 	struct intel_dp dp;
 	struct intel_hdmi hdmi;
 };
@@ -439,10 +456,10 @@ extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg,
 extern void intel_dvo_init(struct drm_device *dev);
 extern void intel_tv_init(struct drm_device *dev);
 extern void intel_mark_busy(struct drm_device *dev);
-extern void intel_mark_idle(struct drm_device *dev);
 extern void intel_mark_fb_busy(struct drm_i915_gem_object *obj);
-extern void intel_mark_fb_idle(struct drm_i915_gem_object *obj);
+extern void intel_mark_idle(struct drm_device *dev);
 extern bool intel_lvds_init(struct drm_device *dev);
+extern bool intel_is_dual_link_lvds(struct drm_device *dev);
 extern void intel_dp_init(struct drm_device *dev, int output_reg,
 			  enum port port);
 extern void intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
@@ -502,12 +519,12 @@ struct intel_set_config {
 	bool mode_changed;
 };
 
-extern bool intel_set_mode(struct drm_crtc *crtc, struct drm_display_mode *mode,
-			   int x, int y, struct drm_framebuffer *old_fb);
+extern int intel_set_mode(struct drm_crtc *crtc, struct drm_display_mode *mode,
+			  int x, int y, struct drm_framebuffer *old_fb);
 extern void intel_modeset_disable(struct drm_device *dev);
+extern void intel_crtc_restore_mode(struct drm_crtc *crtc);
 extern void intel_crtc_load_lut(struct drm_crtc *crtc);
 extern void intel_crtc_update_dpms(struct drm_crtc *crtc);
-extern void intel_encoder_noop(struct drm_encoder *encoder);
 extern void intel_encoder_destroy(struct drm_encoder *encoder);
 extern void intel_encoder_dpms(struct intel_encoder *encoder, int mode);
 extern bool intel_encoder_check_is_cloned(struct intel_encoder *encoder);
@@ -546,6 +563,9 @@ hdmi_to_dig_port(struct intel_hdmi *intel_hdmi)
 	return container_of(intel_hdmi, struct intel_digital_port, hdmi);
 }
 
+bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
+				struct intel_digital_port *port);
+
 extern void intel_connector_attach_encoder(struct intel_connector *connector,
 					   struct intel_encoder *encoder);
 extern struct drm_encoder *intel_best_encoder(struct drm_connector *connector);
@@ -589,6 +609,7 @@ extern int intel_framebuffer_init(struct drm_device *dev,
 				  struct drm_mode_fb_cmd2 *mode_cmd,
 				  struct drm_i915_gem_object *obj);
 extern int intel_fbdev_init(struct drm_device *dev);
+extern void intel_fbdev_initial_config(struct drm_device *dev);
 extern void intel_fbdev_fini(struct drm_device *dev);
 extern void intel_fbdev_set_suspend(struct drm_device *dev, int state);
 extern void intel_prepare_page_flip(struct drm_device *dev, int plane);
@@ -627,9 +648,10 @@ extern void intel_update_sprite_watermarks(struct drm_device *dev, int pipe,
 extern void intel_update_linetime_watermarks(struct drm_device *dev, int pipe,
 			 struct drm_display_mode *mode);
 
-extern unsigned long intel_gen4_compute_offset_xtiled(int *x, int *y,
-						      unsigned int bpp,
-						      unsigned int pitch);
+extern unsigned long intel_gen4_compute_page_offset(int *x, int *y,
+						    unsigned int tiling_mode,
+						    unsigned int bpp,
+						    unsigned int pitch);
 
 extern int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 				     struct drm_file *file_priv);
@@ -648,7 +670,8 @@ extern void intel_update_fbc(struct drm_device *dev);
 extern void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
 extern void intel_gpu_ips_teardown(void);
 
-extern void intel_init_power_wells(struct drm_device *dev);
+extern void intel_init_power_well(struct drm_device *dev);
+extern void intel_set_power_well(struct drm_device *dev, bool enable);
 extern void intel_enable_gt_powersave(struct drm_device *dev);
 extern void intel_disable_gt_powersave(struct drm_device *dev);
 extern void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 15da99533e5b..00e70dbe82da 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -345,7 +345,6 @@ static void intel_dvo_destroy(struct drm_connector *connector)
 static const struct drm_encoder_helper_funcs intel_dvo_helper_funcs = {
 	.mode_fixup = intel_dvo_mode_fixup,
 	.mode_set = intel_dvo_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_dvo_connector_funcs = {
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index 7b30b5c2c4ee..981bdce3634e 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -57,9 +57,10 @@ static struct fb_ops intelfb_ops = {
 	.fb_debug_leave = drm_fb_helper_debug_leave,
 };
 
-static int intelfb_create(struct intel_fbdev *ifbdev,
+static int intelfb_create(struct drm_fb_helper *helper,
 			  struct drm_fb_helper_surface_size *sizes)
 {
+	struct intel_fbdev *ifbdev = (struct intel_fbdev *)helper;
 	struct drm_device *dev = ifbdev->helper.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct fb_info *info;
@@ -83,7 +84,9 @@ static int intelfb_create(struct intel_fbdev *ifbdev,
 
 	size = mode_cmd.pitches[0] * mode_cmd.height;
 	size = ALIGN(size, PAGE_SIZE);
-	obj = i915_gem_alloc_object(dev, size);
+	obj = i915_gem_object_create_stolen(dev, size);
+	if (obj == NULL)
+		obj = i915_gem_alloc_object(dev, size);
 	if (!obj) {
 		DRM_ERROR("failed to allocate framebuffer\n");
 		ret = -ENOMEM;
@@ -133,14 +136,13 @@ static int intelfb_create(struct intel_fbdev *ifbdev,
 		goto out_unpin;
 	}
 	info->apertures->ranges[0].base = dev->mode_config.fb_base;
-	info->apertures->ranges[0].size =
-		dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
+	info->apertures->ranges[0].size = dev_priv->gtt.mappable_end;
 
 	info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset;
 	info->fix.smem_len = size;
 
 	info->screen_base =
-		ioremap_wc(dev_priv->mm.gtt_base_addr + obj->gtt_offset,
+		ioremap_wc(dev_priv->gtt.mappable_base + obj->gtt_offset,
 			   size);
 	if (!info->screen_base) {
 		ret = -ENOSPC;
@@ -153,6 +155,13 @@ static int intelfb_create(struct intel_fbdev *ifbdev,
 	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &ifbdev->helper, sizes->fb_width, sizes->fb_height);
 
+	/* If the object is shmemfs backed, it will have given us zeroed pages.
+	 * If the object is stolen however, it will be full of whatever
+	 * garbage was left in there.
+	 */
+	if (ifbdev->ifb.obj->stolen)
+		memset_io(info->screen_base, 0, info->screen_size);
+
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
 
 	DRM_DEBUG_KMS("allocated %dx%d fb: 0x%08x, bo %p\n",
@@ -173,26 +182,10 @@ out:
 	return ret;
 }
 
-static int intel_fb_find_or_create_single(struct drm_fb_helper *helper,
-					  struct drm_fb_helper_surface_size *sizes)
-{
-	struct intel_fbdev *ifbdev = (struct intel_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = intelfb_create(ifbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static struct drm_fb_helper_funcs intel_fb_helper_funcs = {
 	.gamma_set = intel_crtc_fb_gamma_set,
 	.gamma_get = intel_crtc_fb_gamma_get,
-	.fb_probe = intel_fb_find_or_create_single,
+	.fb_probe = intelfb_create,
 };
 
 static void intel_fbdev_destroy(struct drm_device *dev,
@@ -212,6 +205,7 @@ static void intel_fbdev_destroy(struct drm_device *dev,
 
 	drm_fb_helper_fini(&ifbdev->helper);
 
+	drm_framebuffer_unregister_private(&ifb->base);
 	drm_framebuffer_cleanup(&ifb->base);
 	if (ifb->obj) {
 		drm_gem_object_unreference_unlocked(&ifb->obj->base);
@@ -241,10 +235,18 @@ int intel_fbdev_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&ifbdev->helper);
-	drm_fb_helper_initial_config(&ifbdev->helper, 32);
+
 	return 0;
 }
 
+void intel_fbdev_initial_config(struct drm_device *dev)
+{
+	drm_i915_private_t *dev_priv = dev->dev_private;
+
+	/* Due to peculiar init order wrt to hpd handling this is separate. */
+	drm_fb_helper_initial_config(&dev_priv->fbdev->helper, 32);
+}
+
 void intel_fbdev_fini(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
@@ -280,7 +282,7 @@ void intel_fb_restore_mode(struct drm_device *dev)
 	struct drm_mode_config *config = &dev->mode_config;
 	struct drm_plane *plane;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	ret = drm_fb_helper_restore_fbdev_mode(&dev_priv->fbdev->helper);
 	if (ret)
@@ -288,7 +290,8 @@ void intel_fb_restore_mode(struct drm_device *dev)
 
 	/* Be sure to shut off any planes that may be active */
 	list_for_each_entry(plane, &config->plane_list, head)
-		plane->funcs->disable_plane(plane);
+		if (plane->enabled)
+			plane->funcs->disable_plane(plane);
 
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 }
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 2ee9821b9d93..fa8ec4a26041 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -48,7 +48,7 @@ assert_hdmi_port_disabled(struct intel_hdmi *intel_hdmi)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t enabled_bits;
 
-	enabled_bits = IS_HASWELL(dev) ? DDI_BUF_CTL_ENABLE : SDVO_ENABLE;
+	enabled_bits = HAS_DDI(dev) ? DDI_BUF_CTL_ENABLE : SDVO_ENABLE;
 
 	WARN(I915_READ(intel_hdmi->sdvox_reg) & enabled_bits,
 	     "HDMI port enabled, expecting disabled\n");
@@ -331,6 +331,7 @@ static void intel_set_infoframe(struct drm_encoder *encoder,
 static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
 					 struct drm_display_mode *adjusted_mode)
 {
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	struct dip_infoframe avi_if = {
 		.type = DIP_TYPE_AVI,
 		.ver = DIP_VERSION_AVI,
@@ -340,7 +341,14 @@ static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
 		avi_if.body.avi.YQ_CN_PR |= DIP_AVI_PR_2;
 
-	avi_if.body.avi.VIC = drm_mode_cea_vic(adjusted_mode);
+	if (intel_hdmi->rgb_quant_range_selectable) {
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			avi_if.body.avi.ITC_EC_Q_SC |= DIP_AVI_RGB_QUANT_RANGE_LIMITED;
+		else
+			avi_if.body.avi.ITC_EC_Q_SC |= DIP_AVI_RGB_QUANT_RANGE_FULL;
+	}
+
+	avi_if.body.avi.VIC = drm_match_cea_mode(adjusted_mode);
 
 	intel_set_infoframe(encoder, &avi_if);
 }
@@ -364,7 +372,8 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
-	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
+	struct intel_hdmi *intel_hdmi = &intel_dig_port->hdmi;
 	u32 reg = VIDEO_DIP_CTL;
 	u32 val = I915_READ(reg);
 	u32 port;
@@ -391,11 +400,11 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 		return;
 	}
 
-	switch (intel_hdmi->sdvox_reg) {
-	case SDVOB:
+	switch (intel_dig_port->port) {
+	case PORT_B:
 		port = VIDEO_DIP_PORT_B;
 		break;
-	case SDVOC:
+	case PORT_C:
 		port = VIDEO_DIP_PORT_C;
 		break;
 	default:
@@ -428,7 +437,8 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
-	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
+	struct intel_hdmi *intel_hdmi = &intel_dig_port->hdmi;
 	u32 reg = TVIDEO_DIP_CTL(intel_crtc->pipe);
 	u32 val = I915_READ(reg);
 	u32 port;
@@ -447,14 +457,14 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 		return;
 	}
 
-	switch (intel_hdmi->sdvox_reg) {
-	case HDMIB:
+	switch (intel_dig_port->port) {
+	case PORT_B:
 		port = VIDEO_DIP_PORT_B;
 		break;
-	case HDMIC:
+	case PORT_C:
 		port = VIDEO_DIP_PORT_C;
 		break;
-	case HDMID:
+	case PORT_D:
 		port = VIDEO_DIP_PORT_D;
 		break;
 	default:
@@ -766,46 +776,38 @@ bool intel_hdmi_mode_fixup(struct drm_encoder *encoder,
 			   const struct drm_display_mode *mode,
 			   struct drm_display_mode *adjusted_mode)
 {
-	return true;
-}
-
-static bool g4x_hdmi_connected(struct intel_hdmi *intel_hdmi)
-{
-	struct drm_device *dev = intel_hdmi_to_dev(intel_hdmi);
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t bit;
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 
-	switch (intel_hdmi->sdvox_reg) {
-	case SDVOB:
-		bit = HDMIB_HOTPLUG_LIVE_STATUS;
-		break;
-	case SDVOC:
-		bit = HDMIC_HOTPLUG_LIVE_STATUS;
-		break;
-	default:
-		bit = 0;
-		break;
+	if (intel_hdmi->color_range_auto) {
+		/* See CEA-861-E - 5.1 Default Encoding Parameters */
+		if (intel_hdmi->has_hdmi_sink &&
+		    drm_match_cea_mode(adjusted_mode) > 1)
+			intel_hdmi->color_range = SDVO_COLOR_RANGE_16_235;
+		else
+			intel_hdmi->color_range = 0;
 	}
 
-	return I915_READ(PORT_HOTPLUG_STAT) & bit;
+	if (intel_hdmi->color_range)
+		adjusted_mode->private_flags |= INTEL_MODE_LIMITED_COLOR_RANGE;
+
+	return true;
 }
 
 static enum drm_connector_status
 intel_hdmi_detect(struct drm_connector *connector, bool force)
 {
+	struct drm_device *dev = connector->dev;
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct intel_digital_port *intel_dig_port =
 		hdmi_to_dig_port(intel_hdmi);
 	struct intel_encoder *intel_encoder = &intel_dig_port->base;
-	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct edid *edid;
 	enum drm_connector_status status = connector_status_disconnected;
 
-	if (IS_G4X(connector->dev) && !g4x_hdmi_connected(intel_hdmi))
-		return status;
-
 	intel_hdmi->has_hdmi_sink = false;
 	intel_hdmi->has_audio = false;
+	intel_hdmi->rgb_quant_range_selectable = false;
 	edid = drm_get_edid(connector,
 			    intel_gmbus_get_adapter(dev_priv,
 						    intel_hdmi->ddc_bus));
@@ -817,6 +819,8 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 				intel_hdmi->has_hdmi_sink =
 						drm_detect_hdmi_monitor(edid);
 			intel_hdmi->has_audio = drm_detect_monitor_audio(edid);
+			intel_hdmi->rgb_quant_range_selectable =
+				drm_rgb_quant_range_selectable(edid);
 		}
 		kfree(edid);
 	}
@@ -902,21 +906,29 @@ intel_hdmi_set_property(struct drm_connector *connector,
 	}
 
 	if (property == dev_priv->broadcast_rgb_property) {
-		if (val == !!intel_hdmi->color_range)
-			return 0;
-
-		intel_hdmi->color_range = val ? SDVO_COLOR_RANGE_16_235 : 0;
+		switch (val) {
+		case INTEL_BROADCAST_RGB_AUTO:
+			intel_hdmi->color_range_auto = true;
+			break;
+		case INTEL_BROADCAST_RGB_FULL:
+			intel_hdmi->color_range_auto = false;
+			intel_hdmi->color_range = 0;
+			break;
+		case INTEL_BROADCAST_RGB_LIMITED:
+			intel_hdmi->color_range_auto = false;
+			intel_hdmi->color_range = SDVO_COLOR_RANGE_16_235;
+			break;
+		default:
+			return -EINVAL;
+		}
 		goto done;
 	}
 
 	return -EINVAL;
 
 done:
-	if (intel_dig_port->base.base.crtc) {
-		struct drm_crtc *crtc = intel_dig_port->base.base.crtc;
-		intel_set_mode(crtc, &crtc->mode,
-			       crtc->x, crtc->y, crtc->fb);
-	}
+	if (intel_dig_port->base.base.crtc)
+		intel_crtc_restore_mode(intel_dig_port->base.base.crtc);
 
 	return 0;
 }
@@ -931,7 +943,6 @@ static void intel_hdmi_destroy(struct drm_connector *connector)
 static const struct drm_encoder_helper_funcs intel_hdmi_helper_funcs = {
 	.mode_fixup = intel_hdmi_mode_fixup,
 	.mode_set = intel_hdmi_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_hdmi_connector_funcs = {
@@ -957,6 +968,7 @@ intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *c
 {
 	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
+	intel_hdmi->color_range_auto = true;
 }
 
 void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
@@ -980,15 +992,15 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 	switch (port) {
 	case PORT_B:
 		intel_hdmi->ddc_bus = GMBUS_PORT_DPB;
-		dev_priv->hotplug_supported_mask |= HDMIB_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTB_HOTPLUG_INT_STATUS;
 		break;
 	case PORT_C:
 		intel_hdmi->ddc_bus = GMBUS_PORT_DPC;
-		dev_priv->hotplug_supported_mask |= HDMIC_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTC_HOTPLUG_INT_STATUS;
 		break;
 	case PORT_D:
 		intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
-		dev_priv->hotplug_supported_mask |= HDMID_HOTPLUG_INT_STATUS;
+		dev_priv->hotplug_supported_mask |= PORTD_HOTPLUG_INT_STATUS;
 		break;
 	case PORT_A:
 		/* Internal port only for eDP. */
@@ -1013,7 +1025,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 		intel_hdmi->set_infoframes = cpt_set_infoframes;
 	}
 
-	if (IS_HASWELL(dev))
+	if (HAS_DDI(dev))
 		intel_connector->get_hw_state = intel_ddi_connector_get_hw_state;
 	else
 		intel_connector->get_hw_state = intel_connector_get_hw_state;
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index 3ef5af15b812..acf8aec9ada7 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -63,6 +63,7 @@ intel_i2c_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS0, 0);
+	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS4, 0);
 }
 
 static void intel_i2c_quirk_set(struct drm_i915_private *dev_priv, bool enable)
@@ -202,6 +203,68 @@ intel_gpio_setup(struct intel_gmbus *bus, u32 pin)
 	algo->data = bus;
 }
 
+#define HAS_GMBUS_IRQ(dev) (INTEL_INFO(dev)->gen >= 4)
+static int
+gmbus_wait_hw_status(struct drm_i915_private *dev_priv,
+		     u32 gmbus2_status,
+		     u32 gmbus4_irq_en)
+{
+	int i;
+	int reg_offset = dev_priv->gpio_mmio_base;
+	u32 gmbus2 = 0;
+	DEFINE_WAIT(wait);
+
+	/* Important: The hw handles only the first bit, so set only one! Since
+	 * we also need to check for NAKs besides the hw ready/idle signal, we
+	 * need to wake up periodically and check that ourselves. */
+	I915_WRITE(GMBUS4 + reg_offset, gmbus4_irq_en);
+
+	for (i = 0; i < msecs_to_jiffies(50) + 1; i++) {
+		prepare_to_wait(&dev_priv->gmbus_wait_queue, &wait,
+				TASK_UNINTERRUPTIBLE);
+
+		gmbus2 = I915_READ_NOTRACE(GMBUS2 + reg_offset);
+		if (gmbus2 & (GMBUS_SATOER | gmbus2_status))
+			break;
+
+		schedule_timeout(1);
+	}
+	finish_wait(&dev_priv->gmbus_wait_queue, &wait);
+
+	I915_WRITE(GMBUS4 + reg_offset, 0);
+
+	if (gmbus2 & GMBUS_SATOER)
+		return -ENXIO;
+	if (gmbus2 & gmbus2_status)
+		return 0;
+	return -ETIMEDOUT;
+}
+
+static int
+gmbus_wait_idle(struct drm_i915_private *dev_priv)
+{
+	int ret;
+	int reg_offset = dev_priv->gpio_mmio_base;
+
+#define C ((I915_READ_NOTRACE(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0)
+
+	if (!HAS_GMBUS_IRQ(dev_priv->dev))
+		return wait_for(C, 10);
+
+	/* Important: The hw handles only the first bit, so set only one! */
+	I915_WRITE(GMBUS4 + reg_offset, GMBUS_IDLE_EN);
+
+	ret = wait_event_timeout(dev_priv->gmbus_wait_queue, C, 10);
+
+	I915_WRITE(GMBUS4 + reg_offset, 0);
+
+	if (ret)
+		return 0;
+	else
+		return -ETIMEDOUT;
+#undef C
+}
+
 static int
 gmbus_xfer_read(struct drm_i915_private *dev_priv, struct i2c_msg *msg,
 		u32 gmbus1_index)
@@ -219,15 +282,11 @@ gmbus_xfer_read(struct drm_i915_private *dev_priv, struct i2c_msg *msg,
 	while (len) {
 		int ret;
 		u32 val, loop = 0;
-		u32 gmbus2;
 
-		ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) &
-			       (GMBUS_SATOER | GMBUS_HW_RDY),
-			       50);
+		ret = gmbus_wait_hw_status(dev_priv, GMBUS_HW_RDY,
+					   GMBUS_HW_RDY_EN);
 		if (ret)
-			return -ETIMEDOUT;
-		if (gmbus2 & GMBUS_SATOER)
-			return -ENXIO;
+			return ret;
 
 		val = I915_READ(GMBUS3 + reg_offset);
 		do {
@@ -261,7 +320,6 @@ gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg)
 		   GMBUS_SLAVE_WRITE | GMBUS_SW_RDY);
 	while (len) {
 		int ret;
-		u32 gmbus2;
 
 		val = loop = 0;
 		do {
@@ -270,13 +328,10 @@ gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg)
 
 		I915_WRITE(GMBUS3 + reg_offset, val);
 
-		ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) &
-			       (GMBUS_SATOER | GMBUS_HW_RDY),
-			       50);
+		ret = gmbus_wait_hw_status(dev_priv, GMBUS_HW_RDY,
+					   GMBUS_HW_RDY_EN);
 		if (ret)
-			return -ETIMEDOUT;
-		if (gmbus2 & GMBUS_SATOER)
-			return -ENXIO;
+			return ret;
 	}
 	return 0;
 }
@@ -345,8 +400,6 @@ gmbus_xfer(struct i2c_adapter *adapter,
 	I915_WRITE(GMBUS0 + reg_offset, bus->reg0);
 
 	for (i = 0; i < num; i++) {
-		u32 gmbus2;
-
 		if (gmbus_is_index_read(msgs, i, num)) {
 			ret = gmbus_xfer_index_read(dev_priv, &msgs[i]);
 			i += 1;  /* set i to the index of the read xfer */
@@ -361,13 +414,12 @@ gmbus_xfer(struct i2c_adapter *adapter,
 		if (ret == -ENXIO)
 			goto clear_err;
 
-		ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) &
-			       (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE),
-			       50);
+		ret = gmbus_wait_hw_status(dev_priv, GMBUS_HW_WAIT_PHASE,
+					   GMBUS_HW_WAIT_EN);
+		if (ret == -ENXIO)
+			goto clear_err;
 		if (ret)
 			goto timeout;
-		if (gmbus2 & GMBUS_SATOER)
-			goto clear_err;
 	}
 
 	/* Generate a STOP condition on the bus. Note that gmbus can't generata
@@ -380,8 +432,7 @@ gmbus_xfer(struct i2c_adapter *adapter,
 	 * We will re-enable it at the start of the next xfer,
 	 * till then let it sleep.
 	 */
-	if (wait_for((I915_READ(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0,
-		     10)) {
+	if (gmbus_wait_idle(dev_priv)) {
 		DRM_DEBUG_KMS("GMBUS [%s] timed out waiting for idle\n",
 			 adapter->name);
 		ret = -ETIMEDOUT;
@@ -405,8 +456,7 @@ clear_err:
 	 * it's slow responding and only answers on the 2nd retry.
 	 */
 	ret = -ENXIO;
-	if (wait_for((I915_READ(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0,
-		     10)) {
+	if (gmbus_wait_idle(dev_priv)) {
 		DRM_DEBUG_KMS("GMBUS [%s] timed out after NAK\n",
 			      adapter->name);
 		ret = -ETIMEDOUT;
@@ -465,10 +515,13 @@ int intel_setup_gmbus(struct drm_device *dev)
 
 	if (HAS_PCH_SPLIT(dev))
 		dev_priv->gpio_mmio_base = PCH_GPIOA - GPIOA;
+	else if (IS_VALLEYVIEW(dev))
+		dev_priv->gpio_mmio_base = VLV_DISPLAY_BASE;
 	else
 		dev_priv->gpio_mmio_base = 0;
 
 	mutex_init(&dev_priv->gmbus_mutex);
+	init_waitqueue_head(&dev_priv->gmbus_wait_queue);
 
 	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
 		struct intel_gmbus *bus = &dev_priv->gmbus[i];
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 17aee74258ad..3d1d97488cc9 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -51,7 +51,8 @@ struct intel_lvds_encoder {
 
 	u32 pfit_control;
 	u32 pfit_pgm_ratios;
-	bool pfit_dirty;
+	bool is_dual_link;
+	u32 reg;
 
 	struct intel_lvds_connector *attached_connector;
 };
@@ -71,15 +72,10 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 lvds_reg, tmp;
-
-	if (HAS_PCH_SPLIT(dev)) {
-		lvds_reg = PCH_LVDS;
-	} else {
-		lvds_reg = LVDS;
-	}
+	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
+	u32 tmp;
 
-	tmp = I915_READ(lvds_reg);
+	tmp = I915_READ(lvds_encoder->reg);
 
 	if (!(tmp & LVDS_PORT_EN))
 		return false;
@@ -92,6 +88,91 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 	return true;
 }
 
+/* The LVDS pin pair needs to be on before the DPLLs are enabled.
+ * This is an exception to the general rule that mode_set doesn't turn
+ * things on.
+ */
+static void intel_pre_pll_enable_lvds(struct intel_encoder *encoder)
+{
+	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *fixed_mode =
+		lvds_encoder->attached_connector->base.panel.fixed_mode;
+	int pipe = intel_crtc->pipe;
+	u32 temp;
+
+	temp = I915_READ(lvds_encoder->reg);
+	temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
+
+	if (HAS_PCH_CPT(dev)) {
+		temp &= ~PORT_TRANS_SEL_MASK;
+		temp |= PORT_TRANS_SEL_CPT(pipe);
+	} else {
+		if (pipe == 1) {
+			temp |= LVDS_PIPEB_SELECT;
+		} else {
+			temp &= ~LVDS_PIPEB_SELECT;
+		}
+	}
+
+	/* set the corresponsding LVDS_BORDER bit */
+	temp |= dev_priv->lvds_border_bits;
+	/* Set the B0-B3 data pairs corresponding to whether we're going to
+	 * set the DPLLs for dual-channel mode or not.
+	 */
+	if (lvds_encoder->is_dual_link)
+		temp |= LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP;
+	else
+		temp &= ~(LVDS_B0B3_POWER_UP | LVDS_CLKB_POWER_UP);
+
+	/* It would be nice to set 24 vs 18-bit mode (LVDS_A3_POWER_UP)
+	 * appropriately here, but we need to look more thoroughly into how
+	 * panels behave in the two modes.
+	 */
+
+	/* Set the dithering flag on LVDS as needed, note that there is no
+	 * special lvds dither control bit on pch-split platforms, dithering is
+	 * only controlled through the PIPECONF reg. */
+	if (INTEL_INFO(dev)->gen == 4) {
+		if (dev_priv->lvds_dither)
+			temp |= LVDS_ENABLE_DITHER;
+		else
+			temp &= ~LVDS_ENABLE_DITHER;
+	}
+	temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY);
+	if (fixed_mode->flags & DRM_MODE_FLAG_NHSYNC)
+		temp |= LVDS_HSYNC_POLARITY;
+	if (fixed_mode->flags & DRM_MODE_FLAG_NVSYNC)
+		temp |= LVDS_VSYNC_POLARITY;
+
+	I915_WRITE(lvds_encoder->reg, temp);
+}
+
+static void intel_pre_enable_lvds(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct intel_lvds_encoder *enc = to_lvds_encoder(&encoder->base);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (HAS_PCH_SPLIT(dev) || !enc->pfit_control)
+		return;
+
+	/*
+	 * Enable automatic panel scaling so that non-native modes
+	 * fill the screen.  The panel fitter should only be
+	 * adjusted whilst the pipe is disabled, according to
+	 * register description and PRM.
+	 */
+	DRM_DEBUG_KMS("applying panel-fitter: %x, %x\n",
+		      enc->pfit_control,
+		      enc->pfit_pgm_ratios);
+
+	I915_WRITE(PFIT_PGM_RATIOS, enc->pfit_pgm_ratios);
+	I915_WRITE(PFIT_CONTROL, enc->pfit_control);
+}
+
 /**
  * Sets the power state for the panel.
  */
@@ -101,38 +182,20 @@ static void intel_enable_lvds(struct intel_encoder *encoder)
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 ctl_reg, lvds_reg, stat_reg;
+	u32 ctl_reg, stat_reg;
 
 	if (HAS_PCH_SPLIT(dev)) {
 		ctl_reg = PCH_PP_CONTROL;
-		lvds_reg = PCH_LVDS;
 		stat_reg = PCH_PP_STATUS;
 	} else {
 		ctl_reg = PP_CONTROL;
-		lvds_reg = LVDS;
 		stat_reg = PP_STATUS;
 	}
 
-	I915_WRITE(lvds_reg, I915_READ(lvds_reg) | LVDS_PORT_EN);
-
-	if (lvds_encoder->pfit_dirty) {
-		/*
-		 * Enable automatic panel scaling so that non-native modes
-		 * fill the screen.  The panel fitter should only be
-		 * adjusted whilst the pipe is disabled, according to
-		 * register description and PRM.
-		 */
-		DRM_DEBUG_KMS("applying panel-fitter: %x, %x\n",
-			      lvds_encoder->pfit_control,
-			      lvds_encoder->pfit_pgm_ratios);
-
-		I915_WRITE(PFIT_PGM_RATIOS, lvds_encoder->pfit_pgm_ratios);
-		I915_WRITE(PFIT_CONTROL, lvds_encoder->pfit_control);
-		lvds_encoder->pfit_dirty = false;
-	}
+	I915_WRITE(lvds_encoder->reg, I915_READ(lvds_encoder->reg) | LVDS_PORT_EN);
 
 	I915_WRITE(ctl_reg, I915_READ(ctl_reg) | POWER_TARGET_ON);
-	POSTING_READ(lvds_reg);
+	POSTING_READ(lvds_encoder->reg);
 	if (wait_for((I915_READ(stat_reg) & PP_ON) != 0, 1000))
 		DRM_ERROR("timed out waiting for panel to power on\n");
 
@@ -144,15 +207,13 @@ static void intel_disable_lvds(struct intel_encoder *encoder)
 	struct drm_device *dev = encoder->base.dev;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 ctl_reg, lvds_reg, stat_reg;
+	u32 ctl_reg, stat_reg;
 
 	if (HAS_PCH_SPLIT(dev)) {
 		ctl_reg = PCH_PP_CONTROL;
-		lvds_reg = PCH_LVDS;
 		stat_reg = PCH_PP_STATUS;
 	} else {
 		ctl_reg = PP_CONTROL;
-		lvds_reg = LVDS;
 		stat_reg = PP_STATUS;
 	}
 
@@ -162,13 +223,8 @@ static void intel_disable_lvds(struct intel_encoder *encoder)
 	if (wait_for((I915_READ(stat_reg) & PP_ON) == 0, 1000))
 		DRM_ERROR("timed out waiting for panel to power off\n");
 
-	if (lvds_encoder->pfit_control) {
-		I915_WRITE(PFIT_CONTROL, 0);
-		lvds_encoder->pfit_dirty = true;
-	}
-
-	I915_WRITE(lvds_reg, I915_READ(lvds_reg) & ~LVDS_PORT_EN);
-	POSTING_READ(lvds_reg);
+	I915_WRITE(lvds_encoder->reg, I915_READ(lvds_encoder->reg) & ~LVDS_PORT_EN);
+	POSTING_READ(lvds_encoder->reg);
 }
 
 static int intel_lvds_mode_valid(struct drm_connector *connector,
@@ -406,7 +462,6 @@ out:
 	    pfit_pgm_ratios != lvds_encoder->pfit_pgm_ratios) {
 		lvds_encoder->pfit_control = pfit_control;
 		lvds_encoder->pfit_pgm_ratios = pfit_pgm_ratios;
-		lvds_encoder->pfit_dirty = true;
 	}
 	dev_priv->lvds_border_bits = border;
 
@@ -492,13 +547,14 @@ static const struct dmi_system_id intel_no_modeset_on_lid[] = {
 };
 
 /*
- * Lid events. Note the use of 'modeset_on_lid':
- *  - we set it on lid close, and reset it on open
+ * Lid events. Note the use of 'modeset':
+ *  - we set it to MODESET_ON_LID_OPEN on lid close,
+ *    and set it to MODESET_DONE on open
  *  - we use it as a "only once" bit (ie we ignore
- *    duplicate events where it was already properly
- *    set/reset)
- *  - the suspend/resume paths will also set it to
- *    zero, since they restore the mode ("lid open").
+ *    duplicate events where it was already properly set)
+ *  - the suspend/resume paths will set it to
+ *    MODESET_SUSPENDED and ignore the lid open event,
+ *    because they restore the mode ("lid open").
  */
 static int intel_lid_notify(struct notifier_block *nb, unsigned long val,
 			    void *unused)
@@ -512,6 +568,9 @@ static int intel_lid_notify(struct notifier_block *nb, unsigned long val,
 	if (dev->switch_power_state != DRM_SWITCH_POWER_ON)
 		return NOTIFY_OK;
 
+	mutex_lock(&dev_priv->modeset_restore_lock);
+	if (dev_priv->modeset_restore == MODESET_SUSPENDED)
+		goto exit;
 	/*
 	 * check and update the status of LVDS connector after receiving
 	 * the LID nofication event.
@@ -520,21 +579,24 @@ static int intel_lid_notify(struct notifier_block *nb, unsigned long val,
 
 	/* Don't force modeset on machines where it causes a GPU lockup */
 	if (dmi_check_system(intel_no_modeset_on_lid))
-		return NOTIFY_OK;
+		goto exit;
 	if (!acpi_lid_open()) {
-		dev_priv->modeset_on_lid = 1;
-		return NOTIFY_OK;
+		/* do modeset on next lid open event */
+		dev_priv->modeset_restore = MODESET_ON_LID_OPEN;
+		goto exit;
 	}
 
-	if (!dev_priv->modeset_on_lid)
-		return NOTIFY_OK;
-
-	dev_priv->modeset_on_lid = 0;
+	if (dev_priv->modeset_restore == MODESET_DONE)
+		goto exit;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	intel_modeset_setup_hw_state(dev, true);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
+
+	dev_priv->modeset_restore = MODESET_DONE;
 
+exit:
+	mutex_unlock(&dev_priv->modeset_restore_lock);
 	return NOTIFY_OK;
 }
 
@@ -591,8 +653,7 @@ static int intel_lvds_set_property(struct drm_connector *connector,
 			 * If the CRTC is enabled, the display will be changed
 			 * according to the new panel fitting mode.
 			 */
-			intel_set_mode(crtc, &crtc->mode,
-				       crtc->x, crtc->y, crtc->fb);
+			intel_crtc_restore_mode(crtc);
 		}
 	}
 
@@ -602,7 +663,6 @@ static int intel_lvds_set_property(struct drm_connector *connector,
 static const struct drm_encoder_helper_funcs intel_lvds_helper_funcs = {
 	.mode_fixup = intel_lvds_mode_fixup,
 	.mode_set = intel_lvds_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_helper_funcs intel_lvds_connector_helper_funcs = {
@@ -895,6 +955,66 @@ static bool lvds_is_present_in_vbt(struct drm_device *dev,
 	return false;
 }
 
+static int intel_dual_link_lvds_callback(const struct dmi_system_id *id)
+{
+	DRM_INFO("Forcing lvds to dual link mode on %s\n", id->ident);
+	return 1;
+}
+
+static const struct dmi_system_id intel_dual_link_lvds[] = {
+	{
+		.callback = intel_dual_link_lvds_callback,
+		.ident = "Apple MacBook Pro (Core i5/i7 Series)",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro8,2"),
+		},
+	},
+	{ }	/* terminating entry */
+};
+
+bool intel_is_dual_link_lvds(struct drm_device *dev)
+{
+	struct intel_encoder *encoder;
+	struct intel_lvds_encoder *lvds_encoder;
+
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
+			    base.head) {
+		if (encoder->type == INTEL_OUTPUT_LVDS) {
+			lvds_encoder = to_lvds_encoder(&encoder->base);
+
+			return lvds_encoder->is_dual_link;
+		}
+	}
+
+	return false;
+}
+
+static bool compute_is_dual_link_lvds(struct intel_lvds_encoder *lvds_encoder)
+{
+	struct drm_device *dev = lvds_encoder->base.base.dev;
+	unsigned int val;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* use the module option value if specified */
+	if (i915_lvds_channel_mode > 0)
+		return i915_lvds_channel_mode == 2;
+
+	if (dmi_check_system(intel_dual_link_lvds))
+		return true;
+
+	/* BIOS should set the proper LVDS register value at boot, but
+	 * in reality, it doesn't set the value when the lid is closed;
+	 * we need to check "the value to be set" in VBT when LVDS
+	 * register is uninitialized.
+	 */
+	val = I915_READ(lvds_encoder->reg);
+	if (!(val & ~(LVDS_PIPE_MASK | LVDS_DETECTED)))
+		val = dev_priv->bios_lvds_val;
+
+	return (val & LVDS_CLKB_POWER_MASK) == LVDS_CLKB_POWER_UP;
+}
+
 static bool intel_lvds_supported(struct drm_device *dev)
 {
 	/* With the introduction of the PCH we gained a dedicated
@@ -980,6 +1100,8 @@ bool intel_lvds_init(struct drm_device *dev)
 			 DRM_MODE_ENCODER_LVDS);
 
 	intel_encoder->enable = intel_enable_lvds;
+	intel_encoder->pre_enable = intel_pre_enable_lvds;
+	intel_encoder->pre_pll_enable = intel_pre_pll_enable_lvds;
 	intel_encoder->disable = intel_disable_lvds;
 	intel_encoder->get_hw_state = intel_lvds_get_hw_state;
 	intel_connector->get_hw_state = intel_connector_get_hw_state;
@@ -1001,6 +1123,12 @@ bool intel_lvds_init(struct drm_device *dev)
 	connector->interlace_allowed = false;
 	connector->doublescan_allowed = false;
 
+	if (HAS_PCH_SPLIT(dev)) {
+		lvds_encoder->reg = PCH_LVDS;
+	} else {
+		lvds_encoder->reg = LVDS;
+	}
+
 	/* create the scaling mode property */
 	drm_mode_create_scaling_mode_property(dev);
 	drm_object_attach_property(&connector->base,
@@ -1101,6 +1229,10 @@ bool intel_lvds_init(struct drm_device *dev)
 		goto failed;
 
 out:
+	lvds_encoder->is_dual_link = compute_is_dual_link_lvds(lvds_encoder);
+	DRM_DEBUG_KMS("detected %s-link lvds configuration\n",
+		      lvds_encoder->is_dual_link ? "dual" : "single");
+
 	/*
 	 * Unlock registers and just
 	 * leave them unlocked
diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c
index b00f1c83adce..0e860f39933d 100644
--- a/drivers/gpu/drm/i915/intel_modes.c
+++ b/drivers/gpu/drm/i915/intel_modes.c
@@ -28,7 +28,6 @@
 #include <linux/fb.h>
 #include <drm/drm_edid.h>
 #include <drm/drmP.h>
-#include <drm/drm_edid.h>
 #include "intel_drv.h"
 #include "i915_drv.h"
 
@@ -101,8 +100,9 @@ intel_attach_force_audio_property(struct drm_connector *connector)
 }
 
 static const struct drm_prop_enum_list broadcast_rgb_names[] = {
-	{ 0, "Full" },
-	{ 1, "Limited 16:235" },
+	{ INTEL_BROADCAST_RGB_AUTO, "Automatic" },
+	{ INTEL_BROADCAST_RGB_FULL, "Full" },
+	{ INTEL_BROADCAST_RGB_LIMITED, "Limited 16:235" },
 };
 
 void
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 7741c22c934c..4d338740f2cb 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -347,7 +347,7 @@ static void intel_didl_outputs(struct drm_device *dev)
 	int i = 0;
 
 	handle = DEVICE_ACPI_HANDLE(&dev->pdev->dev);
-	if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev)))
+	if (!handle || acpi_bus_get_device(handle, &acpi_dev))
 		return;
 
 	if (acpi_is_video_device(acpi_dev))
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index d7bc817f51a0..67a2501d519d 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -195,7 +195,7 @@ intel_overlay_map_regs(struct intel_overlay *overlay)
 	if (OVERLAY_NEEDS_PHYSICAL(overlay->dev))
 		regs = (struct overlay_registers __iomem *)overlay->reg_bo->phys_obj->handle->vaddr;
 	else
-		regs = io_mapping_map_wc(dev_priv->mm.gtt_mapping,
+		regs = io_mapping_map_wc(dev_priv->gtt.mappable,
 					 overlay->reg_bo->gtt_offset);
 
 	return regs;
@@ -1045,13 +1045,13 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 	}
 
 	if (!(put_image_rec->flags & I915_OVERLAY_ENABLE)) {
-		mutex_lock(&dev->mode_config.mutex);
+		drm_modeset_lock_all(dev);
 		mutex_lock(&dev->struct_mutex);
 
 		ret = intel_overlay_switch_off(overlay);
 
 		mutex_unlock(&dev->struct_mutex);
-		mutex_unlock(&dev->mode_config.mutex);
+		drm_modeset_unlock_all(dev);
 
 		return ret;
 	}
@@ -1075,7 +1075,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 		goto out_free;
 	}
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	mutex_lock(&dev->struct_mutex);
 
 	if (new_bo->tiling_mode) {
@@ -1157,7 +1157,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 		goto out_unlock;
 
 	mutex_unlock(&dev->struct_mutex);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	kfree(params);
 
@@ -1165,7 +1165,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 
 out_unlock:
 	mutex_unlock(&dev->struct_mutex);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	drm_gem_object_unreference_unlocked(&new_bo->base);
 out_free:
 	kfree(params);
@@ -1241,7 +1241,7 @@ int intel_overlay_attrs(struct drm_device *dev, void *data,
 		return -ENODEV;
 	}
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	mutex_lock(&dev->struct_mutex);
 
 	ret = -EINVAL;
@@ -1307,7 +1307,7 @@ int intel_overlay_attrs(struct drm_device *dev, void *data,
 	ret = 0;
 out_unlock:
 	mutex_unlock(&dev->struct_mutex);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -1333,8 +1333,10 @@ void intel_setup_overlay(struct drm_device *dev)
 
 	overlay->dev = dev;
 
-	reg_bo = i915_gem_alloc_object(dev, PAGE_SIZE);
-	if (!reg_bo)
+	reg_bo = i915_gem_object_create_stolen(dev, PAGE_SIZE);
+	if (reg_bo == NULL)
+		reg_bo = i915_gem_alloc_object(dev, PAGE_SIZE);
+	if (reg_bo == NULL)
 		goto out_free;
 	overlay->reg_bo = reg_bo;
 
@@ -1432,7 +1434,7 @@ intel_overlay_map_regs_atomic(struct intel_overlay *overlay)
 		regs = (struct overlay_registers __iomem *)
 			overlay->reg_bo->phys_obj->handle->vaddr;
 	else
-		regs = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping,
+		regs = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
 						overlay->reg_bo->gtt_offset);
 
 	return regs;
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index bee8cb6108a7..a3730e0289e5 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -321,6 +321,9 @@ void intel_panel_enable_backlight(struct drm_device *dev,
 	if (dev_priv->backlight_level == 0)
 		dev_priv->backlight_level = intel_panel_get_max_backlight(dev);
 
+	dev_priv->backlight_enabled = true;
+	intel_panel_actually_set_backlight(dev, dev_priv->backlight_level);
+
 	if (INTEL_INFO(dev)->gen >= 4) {
 		uint32_t reg, tmp;
 
@@ -356,12 +359,12 @@ void intel_panel_enable_backlight(struct drm_device *dev,
 	}
 
 set_level:
-	/* Call below after setting BLC_PWM_CPU_CTL2 and BLC_PWM_PCH_CTL1.
-	 * BLC_PWM_CPU_CTL may be cleared to zero automatically when these
-	 * registers are set.
+	/* Check the current backlight level and try to set again if it's zero.
+	 * On some machines, BLC_PWM_CPU_CTL is cleared to zero automatically
+	 * when BLC_PWM_CPU_CTL2 and BLC_PWM_PCH_CTL1 are written.
 	 */
-	dev_priv->backlight_enabled = true;
-	intel_panel_actually_set_backlight(dev, dev_priv->backlight_level);
+	if (!intel_panel_get_backlight(dev))
+		intel_panel_actually_set_backlight(dev, dev_priv->backlight_level);
 }
 
 static void intel_panel_init_backlight(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 3280cffe50f4..61fee7fcdc2c 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -447,12 +447,6 @@ void intel_update_fbc(struct drm_device *dev)
 		dev_priv->no_fbc_reason = FBC_MODULE_PARAM;
 		goto out_disable;
 	}
-	if (intel_fb->obj->base.size > dev_priv->cfb_size) {
-		DRM_DEBUG_KMS("framebuffer too large, disabling "
-			      "compression\n");
-		dev_priv->no_fbc_reason = FBC_STOLEN_TOO_SMALL;
-		goto out_disable;
-	}
 	if ((crtc->mode.flags & DRM_MODE_FLAG_INTERLACE) ||
 	    (crtc->mode.flags & DRM_MODE_FLAG_DBLSCAN)) {
 		DRM_DEBUG_KMS("mode incompatible with compression, "
@@ -486,6 +480,14 @@ void intel_update_fbc(struct drm_device *dev)
 	if (in_dbg_master())
 		goto out_disable;
 
+	if (i915_gem_stolen_setup_compression(dev, intel_fb->obj->base.size)) {
+		DRM_INFO("not enough stolen space for compressed buffer (need %zd bytes), disabling\n", intel_fb->obj->base.size);
+		DRM_INFO("hint: you may be able to increase stolen memory size in the BIOS to avoid this\n");
+		DRM_DEBUG_KMS("framebuffer too large, disabling compression\n");
+		dev_priv->no_fbc_reason = FBC_STOLEN_TOO_SMALL;
+		goto out_disable;
+	}
+
 	/* If the scanout has not changed, don't modify the FBC settings.
 	 * Note that we make the fundamental assumption that the fb->obj
 	 * cannot be unpinned (and have its GTT offset and fence revoked)
@@ -533,6 +535,7 @@ out_disable:
 		DRM_DEBUG_KMS("unsupported config, disabling FBC\n");
 		intel_disable_fbc(dev);
 	}
+	i915_gem_stolen_cleanup_compression(dev);
 }
 
 static void i915_pineview_get_mem_freq(struct drm_device *dev)
@@ -2286,7 +2289,6 @@ err_unpin:
 	i915_gem_object_unpin(ctx);
 err_unref:
 	drm_gem_object_unreference(&ctx->base);
-	mutex_unlock(&dev->struct_mutex);
 	return NULL;
 }
 
@@ -3581,6 +3583,19 @@ static void cpt_init_clock_gating(struct drm_device *dev)
 	}
 }
 
+static void gen6_check_mch_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t tmp;
+
+	tmp = I915_READ(MCH_SSKPD);
+	if ((tmp & MCH_SSKPD_WM0_MASK) != MCH_SSKPD_WM0_VAL) {
+		DRM_INFO("Wrong MCH_SSKPD value: 0x%08x\n", tmp);
+		DRM_INFO("This can cause pipe underruns and display issues.\n");
+		DRM_INFO("Please upgrade your BIOS to fix this.\n");
+	}
+}
+
 static void gen6_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3673,6 +3688,8 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN6_GT_MODE, _MASKED_BIT_ENABLE(GEN6_GT_MODE_HI));
 
 	cpt_init_clock_gating(dev);
+
+	gen6_check_mch_setup(dev);
 }
 
 static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv)
@@ -3684,6 +3701,10 @@ static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv)
 	reg |= GEN7_FF_VS_SCHED_HW;
 	reg |= GEN7_FF_DS_SCHED_HW;
 
+	/* WaVSRefCountFullforceMissDisable */
+	if (IS_HASWELL(dev_priv->dev))
+		reg &= ~GEN7_FF_VS_REF_CNT_FFME;
+
 	I915_WRITE(GEN7_FF_THREAD_MODE, reg);
 }
 
@@ -3854,6 +3875,8 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN6_MBCUNIT_SNPCR, snpcr);
 
 	cpt_init_clock_gating(dev);
+
+	gen6_check_mch_setup(dev);
 }
 
 static void valleyview_init_clock_gating(struct drm_device *dev)
@@ -4047,35 +4070,57 @@ void intel_init_clock_gating(struct drm_device *dev)
 	dev_priv->display.init_clock_gating(dev);
 }
 
-/* Starting with Haswell, we have different power wells for
- * different parts of the GPU. This attempts to enable them all.
- */
-void intel_init_power_wells(struct drm_device *dev)
+void intel_set_power_well(struct drm_device *dev, bool enable)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long power_wells[] = {
-		HSW_PWR_WELL_CTL1,
-		HSW_PWR_WELL_CTL2,
-		HSW_PWR_WELL_CTL4
-	};
-	int i;
+	bool is_enabled, enable_requested;
+	uint32_t tmp;
 
 	if (!IS_HASWELL(dev))
 		return;
 
-	mutex_lock(&dev->struct_mutex);
+	tmp = I915_READ(HSW_PWR_WELL_DRIVER);
+	is_enabled = tmp & HSW_PWR_WELL_STATE;
+	enable_requested = tmp & HSW_PWR_WELL_ENABLE;
 
-	for (i = 0; i < ARRAY_SIZE(power_wells); i++) {
-		int well = I915_READ(power_wells[i]);
+	if (enable) {
+		if (!enable_requested)
+			I915_WRITE(HSW_PWR_WELL_DRIVER, HSW_PWR_WELL_ENABLE);
 
-		if ((well & HSW_PWR_WELL_STATE) == 0) {
-			I915_WRITE(power_wells[i], well & HSW_PWR_WELL_ENABLE);
-			if (wait_for((I915_READ(power_wells[i]) & HSW_PWR_WELL_STATE), 20))
-				DRM_ERROR("Error enabling power well %lx\n", power_wells[i]);
+		if (!is_enabled) {
+			DRM_DEBUG_KMS("Enabling power well\n");
+			if (wait_for((I915_READ(HSW_PWR_WELL_DRIVER) &
+				      HSW_PWR_WELL_STATE), 20))
+				DRM_ERROR("Timeout enabling power well\n");
+		}
+	} else {
+		if (enable_requested) {
+			I915_WRITE(HSW_PWR_WELL_DRIVER, 0);
+			DRM_DEBUG_KMS("Requesting to disable the power well\n");
 		}
 	}
+}
 
-	mutex_unlock(&dev->struct_mutex);
+/*
+ * Starting with Haswell, we have a "Power Down Well" that can be turned off
+ * when not needed anymore. We have 4 registers that can request the power well
+ * to be enabled, and it will only be disabled if none of the registers is
+ * requesting it to be enabled.
+ */
+void intel_init_power_well(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!IS_HASWELL(dev))
+		return;
+
+	/* For now, we need the power well to be always enabled. */
+	intel_set_power_well(dev, true);
+
+	/* We're taking over the BIOS, so clear any requests made by it since
+	 * the driver is in charge now. */
+	if (I915_READ(HSW_PWR_WELL_BIOS) & HSW_PWR_WELL_ENABLE)
+		I915_WRITE(HSW_PWR_WELL_BIOS, 0);
 }
 
 /* Set up chip specific power management-related functions */
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 42ff97d667d2..1d5d613eb6be 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -318,6 +318,7 @@ gen7_render_ring_flush(struct intel_ring_buffer *ring,
 		 * TLB invalidate requires a post-sync write.
 		 */
 		flags |= PIPE_CONTROL_QW_WRITE;
+		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
 
 		/* Workaround: we must issue a pipe_control with CS-stall bit
 		 * set before a pipe_control command that has the state cache
@@ -331,7 +332,7 @@ gen7_render_ring_flush(struct intel_ring_buffer *ring,
 
 	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
 	intel_ring_emit(ring, flags);
-	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(ring, scratch_addr);
 	intel_ring_emit(ring, 0);
 	intel_ring_advance(ring);
 
@@ -467,6 +468,9 @@ init_pipe_control(struct intel_ring_buffer *ring)
 	if (pc->cpu_page == NULL)
 		goto err_unpin;
 
+	DRM_DEBUG_DRIVER("%s pipe control offset: 0x%08x\n",
+			 ring->name, pc->gtt_offset);
+
 	pc->obj = obj;
 	ring->private = pc;
 	return 0;
@@ -613,6 +617,13 @@ gen6_add_request(struct intel_ring_buffer *ring)
 	return 0;
 }
 
+static inline bool i915_gem_has_seqno_wrapped(struct drm_device *dev,
+					      u32 seqno)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	return dev_priv->last_seqno < seqno;
+}
+
 /**
  * intel_ring_sync - sync the waiter to the signaller on seqno
  *
@@ -643,11 +654,20 @@ gen6_ring_sync(struct intel_ring_buffer *waiter,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(waiter,
-			dw1 | signaller->semaphore_register[waiter->id]);
-	intel_ring_emit(waiter, seqno);
-	intel_ring_emit(waiter, 0);
-	intel_ring_emit(waiter, MI_NOOP);
+	/* If seqno wrap happened, omit the wait with no-ops */
+	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
+		intel_ring_emit(waiter,
+				dw1 |
+				signaller->semaphore_register[waiter->id]);
+		intel_ring_emit(waiter, seqno);
+		intel_ring_emit(waiter, 0);
+		intel_ring_emit(waiter, MI_NOOP);
+	} else {
+		intel_ring_emit(waiter, MI_NOOP);
+		intel_ring_emit(waiter, MI_NOOP);
+		intel_ring_emit(waiter, MI_NOOP);
+		intel_ring_emit(waiter, MI_NOOP);
+	}
 	intel_ring_advance(waiter);
 
 	return 0;
@@ -728,6 +748,12 @@ ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
 	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
 }
 
+static void
+ring_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+{
+	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+}
+
 static u32
 pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
 {
@@ -735,6 +761,13 @@ pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
 	return pc->cpu_page[0];
 }
 
+static void
+pc_render_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+{
+	struct pipe_control *pc = ring->private;
+	pc->cpu_page[0] = seqno;
+}
+
 static bool
 gen5_ring_get_irq(struct intel_ring_buffer *ring)
 {
@@ -1164,7 +1197,11 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 			return ret;
 	}
 
-	obj = i915_gem_alloc_object(dev, ring->size);
+	obj = NULL;
+	if (!HAS_LLC(dev))
+		obj = i915_gem_object_create_stolen(dev, ring->size);
+	if (obj == NULL)
+		obj = i915_gem_alloc_object(dev, ring->size);
 	if (obj == NULL) {
 		DRM_ERROR("Failed to allocate ringbuffer\n");
 		ret = -ENOMEM;
@@ -1182,7 +1219,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 		goto err_unpin;
 
 	ring->virtual_start =
-		ioremap_wc(dev_priv->mm.gtt->gma_bus_addr + obj->gtt_offset,
+		ioremap_wc(dev_priv->gtt.mappable_base + obj->gtt_offset,
 			   ring->size);
 	if (ring->virtual_start == NULL) {
 		DRM_ERROR("Failed to map ringbuffer.\n");
@@ -1348,7 +1385,8 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 
 		msleep(1);
 
-		ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
+		ret = i915_gem_check_wedge(&dev_priv->gpu_error,
+					   dev_priv->mm.interruptible);
 		if (ret)
 			return ret;
 	} while (!time_after(jiffies, end));
@@ -1410,14 +1448,35 @@ intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
 	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_request);
 }
 
+static int __intel_ring_begin(struct intel_ring_buffer *ring,
+			      int bytes)
+{
+	int ret;
+
+	if (unlikely(ring->tail + bytes > ring->effective_size)) {
+		ret = intel_wrap_ring_buffer(ring);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	if (unlikely(ring->space < bytes)) {
+		ret = ring_wait_for_space(ring, bytes);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	ring->space -= bytes;
+	return 0;
+}
+
 int intel_ring_begin(struct intel_ring_buffer *ring,
 		     int num_dwords)
 {
 	drm_i915_private_t *dev_priv = ring->dev->dev_private;
-	int n = 4*num_dwords;
 	int ret;
 
-	ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
+	ret = i915_gem_check_wedge(&dev_priv->gpu_error,
+				   dev_priv->mm.interruptible);
 	if (ret)
 		return ret;
 
@@ -1426,20 +1485,21 @@ int intel_ring_begin(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	if (unlikely(ring->tail + n > ring->effective_size)) {
-		ret = intel_wrap_ring_buffer(ring);
-		if (unlikely(ret))
-			return ret;
-	}
+	return __intel_ring_begin(ring, num_dwords * sizeof(uint32_t));
+}
 
-	if (unlikely(ring->space < n)) {
-		ret = ring_wait_for_space(ring, n);
-		if (unlikely(ret))
-			return ret;
+void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+
+	BUG_ON(ring->outstanding_lazy_request);
+
+	if (INTEL_INFO(ring->dev)->gen >= 6) {
+		I915_WRITE(RING_SYNC_0(ring->mmio_base), 0);
+		I915_WRITE(RING_SYNC_1(ring->mmio_base), 0);
 	}
 
-	ring->space -= n;
-	return 0;
+	ring->set_seqno(ring, seqno);
 }
 
 void intel_ring_advance(struct intel_ring_buffer *ring)
@@ -1447,7 +1507,7 @@ void intel_ring_advance(struct intel_ring_buffer *ring)
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
 	ring->tail &= ring->size - 1;
-	if (dev_priv->stop_rings & intel_ring_flag(ring))
+	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
 		return;
 	ring->write_tail(ring, ring->tail);
 }
@@ -1604,6 +1664,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->irq_put = gen6_ring_put_irq;
 		ring->irq_enable_mask = GT_USER_INTERRUPT;
 		ring->get_seqno = gen6_ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
 		ring->sync_to = gen6_ring_sync;
 		ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID;
 		ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV;
@@ -1614,6 +1675,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->add_request = pc_render_add_request;
 		ring->flush = gen4_render_ring_flush;
 		ring->get_seqno = pc_render_get_seqno;
+		ring->set_seqno = pc_render_set_seqno;
 		ring->irq_get = gen5_ring_get_irq;
 		ring->irq_put = gen5_ring_put_irq;
 		ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY;
@@ -1624,6 +1686,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		else
 			ring->flush = gen4_render_ring_flush;
 		ring->get_seqno = ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
 		if (IS_GEN2(dev)) {
 			ring->irq_get = i8xx_ring_get_irq;
 			ring->irq_put = i8xx_ring_put_irq;
@@ -1695,6 +1758,7 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 	else
 		ring->flush = gen4_render_ring_flush;
 	ring->get_seqno = ring_get_seqno;
+	ring->set_seqno = ring_set_seqno;
 	if (IS_GEN2(dev)) {
 		ring->irq_get = i8xx_ring_get_irq;
 		ring->irq_put = i8xx_ring_put_irq;
@@ -1755,6 +1819,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 		ring->flush = gen6_ring_flush;
 		ring->add_request = gen6_add_request;
 		ring->get_seqno = gen6_ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
 		ring->irq_enable_mask = GEN6_BSD_USER_INTERRUPT;
 		ring->irq_get = gen6_ring_get_irq;
 		ring->irq_put = gen6_ring_put_irq;
@@ -1770,6 +1835,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 		ring->flush = bsd_ring_flush;
 		ring->add_request = i9xx_add_request;
 		ring->get_seqno = ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
 		if (IS_GEN5(dev)) {
 			ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
 			ring->irq_get = gen5_ring_get_irq;
@@ -1799,6 +1865,7 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 	ring->flush = blt_ring_flush;
 	ring->add_request = gen6_add_request;
 	ring->get_seqno = gen6_ring_get_seqno;
+	ring->set_seqno = ring_set_seqno;
 	ring->irq_enable_mask = GEN6_BLITTER_USER_INTERRUPT;
 	ring->irq_get = gen6_ring_get_irq;
 	ring->irq_put = gen6_ring_put_irq;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 6af87cd05725..d66208c2c48b 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -90,6 +90,8 @@ struct  intel_ring_buffer {
 	 */
 	u32		(*get_seqno)(struct intel_ring_buffer *ring,
 				     bool lazy_coherency);
+	void		(*set_seqno)(struct intel_ring_buffer *ring,
+				     u32 seqno);
 	int		(*dispatch_execbuffer)(struct intel_ring_buffer *ring,
 					       u32 offset, u32 length,
 					       unsigned flags);
@@ -178,6 +180,13 @@ intel_read_status_page(struct intel_ring_buffer *ring,
 	return ring->status_page.page_addr[reg];
 }
 
+static inline void
+intel_write_status_page(struct intel_ring_buffer *ring,
+			int reg, u32 value)
+{
+	ring->status_page.page_addr[reg] = value;
+}
+
 /**
  * Reads a dword out of the status page, which is written to from the command
  * queue by automatic updates, MI_REPORT_HEAD, MI_STORE_DATA_INDEX, or
@@ -208,7 +217,7 @@ static inline void intel_ring_emit(struct intel_ring_buffer *ring,
 }
 void intel_ring_advance(struct intel_ring_buffer *ring);
 int __must_check intel_ring_idle(struct intel_ring_buffer *ring);
-
+void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno);
 int intel_ring_flush_all_caches(struct intel_ring_buffer *ring);
 int intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring);
 
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index c275bf0fa36d..d07a8cdf998e 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -103,6 +103,7 @@ struct intel_sdvo {
 	 * It is only valid when using TMDS encoding and 8 bit per color mode.
 	 */
 	uint32_t color_range;
+	bool color_range_auto;
 
 	/**
 	 * This is set if we're going to treat the device as TV-out.
@@ -125,6 +126,7 @@ struct intel_sdvo {
 	bool is_hdmi;
 	bool has_hdmi_monitor;
 	bool has_hdmi_audio;
+	bool rgb_quant_range_selectable;
 
 	/**
 	 * This is set if we detect output of sdvo device as LVDS and
@@ -946,7 +948,8 @@ static bool intel_sdvo_write_infoframe(struct intel_sdvo *intel_sdvo,
 				    &tx_rate, 1);
 }
 
-static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo)
+static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo,
+					 const struct drm_display_mode *adjusted_mode)
 {
 	struct dip_infoframe avi_if = {
 		.type = DIP_TYPE_AVI,
@@ -955,6 +958,13 @@ static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo)
 	};
 	uint8_t sdvo_data[4 + sizeof(avi_if.body.avi)];
 
+	if (intel_sdvo->rgb_quant_range_selectable) {
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			avi_if.body.avi.ITC_EC_Q_SC |= DIP_AVI_RGB_QUANT_RANGE_LIMITED;
+		else
+			avi_if.body.avi.ITC_EC_Q_SC |= DIP_AVI_RGB_QUANT_RANGE_FULL;
+	}
+
 	intel_dip_infoframe_csum(&avi_if);
 
 	/* sdvo spec says that the ecc is handled by the hw, and it looks like
@@ -1064,6 +1074,18 @@ static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder,
 	multiplier = intel_sdvo_get_pixel_multiplier(adjusted_mode);
 	intel_mode_set_pixel_multiplier(adjusted_mode, multiplier);
 
+	if (intel_sdvo->color_range_auto) {
+		/* See CEA-861-E - 5.1 Default Encoding Parameters */
+		if (intel_sdvo->has_hdmi_monitor &&
+		    drm_match_cea_mode(adjusted_mode) > 1)
+			intel_sdvo->color_range = SDVO_COLOR_RANGE_16_235;
+		else
+			intel_sdvo->color_range = 0;
+	}
+
+	if (intel_sdvo->color_range)
+		adjusted_mode->private_flags |= INTEL_MODE_LIMITED_COLOR_RANGE;
+
 	return true;
 }
 
@@ -1121,7 +1143,7 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder,
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_HDMI);
 		intel_sdvo_set_colorimetry(intel_sdvo,
 					   SDVO_COLORIMETRY_RGB256);
-		intel_sdvo_set_avi_infoframe(intel_sdvo);
+		intel_sdvo_set_avi_infoframe(intel_sdvo, adjusted_mode);
 	} else
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_DVI);
 
@@ -1153,7 +1175,7 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder,
 		/* The real mode polarity is set by the SDVO commands, using
 		 * struct intel_sdvo_dtd. */
 		sdvox = SDVO_VSYNC_ACTIVE_HIGH | SDVO_HSYNC_ACTIVE_HIGH;
-		if (intel_sdvo->is_hdmi)
+		if (!HAS_PCH_SPLIT(dev) && intel_sdvo->is_hdmi)
 			sdvox |= intel_sdvo->color_range;
 		if (INTEL_INFO(dev)->gen < 5)
 			sdvox |= SDVO_BORDER_ENABLE;
@@ -1513,6 +1535,8 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 			if (intel_sdvo->is_hdmi) {
 				intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid);
 				intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
+				intel_sdvo->rgb_quant_range_selectable =
+					drm_rgb_quant_range_selectable(edid);
 			}
 		} else
 			status = connector_status_disconnected;
@@ -1564,6 +1588,7 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 
 	intel_sdvo->has_hdmi_monitor = false;
 	intel_sdvo->has_hdmi_audio = false;
+	intel_sdvo->rgb_quant_range_selectable = false;
 
 	if ((intel_sdvo_connector->output_flag & response) == 0)
 		ret = connector_status_disconnected;
@@ -1897,10 +1922,21 @@ intel_sdvo_set_property(struct drm_connector *connector,
 	}
 
 	if (property == dev_priv->broadcast_rgb_property) {
-		if (val == !!intel_sdvo->color_range)
-			return 0;
-
-		intel_sdvo->color_range = val ? SDVO_COLOR_RANGE_16_235 : 0;
+		switch (val) {
+		case INTEL_BROADCAST_RGB_AUTO:
+			intel_sdvo->color_range_auto = true;
+			break;
+		case INTEL_BROADCAST_RGB_FULL:
+			intel_sdvo->color_range_auto = false;
+			intel_sdvo->color_range = 0;
+			break;
+		case INTEL_BROADCAST_RGB_LIMITED:
+			intel_sdvo->color_range_auto = false;
+			intel_sdvo->color_range = SDVO_COLOR_RANGE_16_235;
+			break;
+		default:
+			return -EINVAL;
+		}
 		goto done;
 	}
 
@@ -1997,11 +2033,8 @@ set_value:
 
 
 done:
-	if (intel_sdvo->base.base.crtc) {
-		struct drm_crtc *crtc = intel_sdvo->base.base.crtc;
-		intel_set_mode(crtc, &crtc->mode,
-			       crtc->x, crtc->y, crtc->fb);
-	}
+	if (intel_sdvo->base.base.crtc)
+		intel_crtc_restore_mode(intel_sdvo->base.base.crtc);
 
 	return 0;
 #undef CHECK_PROPERTY
@@ -2010,7 +2043,6 @@ done:
 static const struct drm_encoder_helper_funcs intel_sdvo_helper_funcs = {
 	.mode_fixup = intel_sdvo_mode_fixup,
 	.mode_set = intel_sdvo_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_sdvo_connector_funcs = {
@@ -2200,13 +2232,16 @@ intel_sdvo_connector_init(struct intel_sdvo_connector *connector,
 }
 
 static void
-intel_sdvo_add_hdmi_properties(struct intel_sdvo_connector *connector)
+intel_sdvo_add_hdmi_properties(struct intel_sdvo *intel_sdvo,
+			       struct intel_sdvo_connector *connector)
 {
 	struct drm_device *dev = connector->base.base.dev;
 
 	intel_attach_force_audio_property(&connector->base.base);
-	if (INTEL_INFO(dev)->gen >= 4 && IS_MOBILE(dev))
+	if (INTEL_INFO(dev)->gen >= 4 && IS_MOBILE(dev)) {
 		intel_attach_broadcast_rgb_property(&connector->base.base);
+		intel_sdvo->color_range_auto = true;
+	}
 }
 
 static bool
@@ -2254,7 +2289,7 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device)
 
 	intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo);
 	if (intel_sdvo->is_hdmi)
-		intel_sdvo_add_hdmi_properties(intel_sdvo_connector);
+		intel_sdvo_add_hdmi_properties(intel_sdvo, intel_sdvo_connector);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index d7b060e0a231..1b6eb76beb7c 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -50,6 +50,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
 	u32 sprctl, sprscale = 0;
 	unsigned long sprsurf_offset, linear_offset;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	bool scaling_was_enabled = dev_priv->sprite_scaling_enabled;
 
 	sprctl = I915_READ(SPRCTL(pipe));
 
@@ -89,6 +90,9 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
 	sprctl |= SPRITE_TRICKLE_FEED_DISABLE;
 	sprctl |= SPRITE_ENABLE;
 
+	if (IS_HASWELL(dev))
+		sprctl |= SPRITE_PIPE_CSC_ENABLE;
+
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
@@ -103,27 +107,23 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
 	 * when scaling is disabled.
 	 */
 	if (crtc_w != src_w || crtc_h != src_h) {
-		if (!dev_priv->sprite_scaling_enabled) {
-			dev_priv->sprite_scaling_enabled = true;
+		dev_priv->sprite_scaling_enabled |= 1 << pipe;
+
+		if (!scaling_was_enabled) {
 			intel_update_watermarks(dev);
 			intel_wait_for_vblank(dev, pipe);
 		}
 		sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h;
-	} else {
-		if (dev_priv->sprite_scaling_enabled) {
-			dev_priv->sprite_scaling_enabled = false;
-			/* potentially re-enable LP watermarks */
-			intel_update_watermarks(dev);
-		}
-	}
+	} else
+		dev_priv->sprite_scaling_enabled &= ~(1 << pipe);
 
 	I915_WRITE(SPRSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(SPRPOS(pipe), (crtc_y << 16) | crtc_x);
 
 	linear_offset = y * fb->pitches[0] + x * pixel_size;
 	sprsurf_offset =
-		intel_gen4_compute_offset_xtiled(&x, &y,
-						 pixel_size, fb->pitches[0]);
+		intel_gen4_compute_page_offset(&x, &y, obj->tiling_mode,
+					       pixel_size, fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
 	/* HSW consolidates SPRTILEOFF and SPRLINOFF into a single SPROFFSET
@@ -141,6 +141,10 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
 	I915_WRITE(SPRCTL(pipe), sprctl);
 	I915_MODIFY_DISPBASE(SPRSURF(pipe), obj->gtt_offset + sprsurf_offset);
 	POSTING_READ(SPRSURF(pipe));
+
+	/* potentially re-enable LP watermarks */
+	if (scaling_was_enabled && !dev_priv->sprite_scaling_enabled)
+		intel_update_watermarks(dev);
 }
 
 static void
@@ -150,6 +154,7 @@ ivb_disable_plane(struct drm_plane *plane)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 	int pipe = intel_plane->pipe;
+	bool scaling_was_enabled = dev_priv->sprite_scaling_enabled;
 
 	I915_WRITE(SPRCTL(pipe), I915_READ(SPRCTL(pipe)) & ~SPRITE_ENABLE);
 	/* Can't leave the scaler enabled... */
@@ -159,8 +164,11 @@ ivb_disable_plane(struct drm_plane *plane)
 	I915_MODIFY_DISPBASE(SPRSURF(pipe), 0);
 	POSTING_READ(SPRSURF(pipe));
 
-	dev_priv->sprite_scaling_enabled = false;
-	intel_update_watermarks(dev);
+	dev_priv->sprite_scaling_enabled &= ~(1 << pipe);
+
+	/* potentially re-enable LP watermarks */
+	if (scaling_was_enabled && !dev_priv->sprite_scaling_enabled)
+		intel_update_watermarks(dev);
 }
 
 static int
@@ -287,8 +295,8 @@ ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
 
 	linear_offset = y * fb->pitches[0] + x * pixel_size;
 	dvssurf_offset =
-		intel_gen4_compute_offset_xtiled(&x, &y,
-						 pixel_size, fb->pitches[0]);
+		intel_gen4_compute_page_offset(&x, &y, obj->tiling_mode,
+					       pixel_size, fb->pitches[0]);
 	linear_offset -= dvssurf_offset;
 
 	if (obj->tiling_mode != I915_TILING_NONE)
@@ -593,7 +601,7 @@ int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 	if ((set->flags & (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE)) == (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE))
 		return -EINVAL;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, set->plane_id, DRM_MODE_OBJECT_PLANE);
 	if (!obj) {
@@ -606,7 +614,7 @@ int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 	ret = intel_plane->update_colorkey(plane, set);
 
 out_unlock:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
@@ -622,7 +630,7 @@ int intel_sprite_get_colorkey(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -ENODEV;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 
 	obj = drm_mode_object_find(dev, get->plane_id, DRM_MODE_OBJECT_PLANE);
 	if (!obj) {
@@ -635,7 +643,7 @@ int intel_sprite_get_colorkey(struct drm_device *dev, void *data,
 	intel_plane->get_colorkey(plane, get);
 
 out_unlock:
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index ea93520c1278..d808421c1c80 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1479,8 +1479,7 @@ intel_tv_set_property(struct drm_connector *connector, struct drm_property *prop
 	}
 
 	if (changed && crtc)
-		intel_set_mode(crtc, &crtc->mode,
-			       crtc->x, crtc->y, crtc->fb);
+		intel_crtc_restore_mode(crtc);
 out:
 	return ret;
 }
@@ -1488,7 +1487,6 @@ out:
 static const struct drm_encoder_helper_funcs intel_tv_helper_funcs = {
 	.mode_fixup = intel_tv_mode_fixup,
 	.mode_set = intel_tv_mode_set,
-	.disable = intel_encoder_noop,
 };
 
 static const struct drm_connector_funcs intel_tv_connector_funcs = {
diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c
index 2f486481d79a..d2253f639481 100644
--- a/drivers/gpu/drm/mgag200/mgag200_fb.c
+++ b/drivers/gpu/drm/mgag200/mgag200_fb.c
@@ -13,6 +13,7 @@
 #include <linux/module.h>
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_crtc_helper.h>
 
 #include <linux/fb.h>
 
@@ -120,9 +121,10 @@ static int mgag200fb_create_object(struct mga_fbdev *afbdev,
 	return ret;
 }
 
-static int mgag200fb_create(struct mga_fbdev *mfbdev,
+static int mgag200fb_create(struct drm_fb_helper *helper,
 			   struct drm_fb_helper_surface_size *sizes)
 {
+	struct mga_fbdev *mfbdev = (struct mga_fbdev *)helper;
 	struct drm_device *dev = mfbdev->helper.dev;
 	struct drm_mode_fb_cmd2 mode_cmd;
 	struct mga_device *mdev = dev->dev_private;
@@ -209,23 +211,6 @@ out:
 	return ret;
 }
 
-static int mga_fb_find_or_create_single(struct drm_fb_helper *helper,
-					   struct drm_fb_helper_surface_size
-					   *sizes)
-{
-	struct mga_fbdev *mfbdev = (struct mga_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = mgag200fb_create(mfbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static int mga_fbdev_destroy(struct drm_device *dev,
 				struct mga_fbdev *mfbdev)
 {
@@ -247,6 +232,7 @@ static int mga_fbdev_destroy(struct drm_device *dev,
 	}
 	drm_fb_helper_fini(&mfbdev->helper);
 	vfree(mfbdev->sysram);
+	drm_framebuffer_unregister_private(&mfb->base);
 	drm_framebuffer_cleanup(&mfb->base);
 
 	return 0;
@@ -255,7 +241,7 @@ static int mga_fbdev_destroy(struct drm_device *dev,
 static struct drm_fb_helper_funcs mga_fb_helper_funcs = {
 	.gamma_set = mga_crtc_fb_gamma_set,
 	.gamma_get = mga_crtc_fb_gamma_get,
-	.fb_probe = mga_fb_find_or_create_single,
+	.fb_probe = mgag200fb_create,
 };
 
 int mgag200_fbdev_init(struct mga_device *mdev)
@@ -277,6 +263,10 @@ int mgag200_fbdev_init(struct mga_device *mdev)
 		return ret;
 	}
 	drm_fb_helper_single_add_all_connectors(&mfbdev->helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(mdev->dev);
+
 	drm_fb_helper_initial_config(&mfbdev->helper, 32);
 
 	return 0;
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index 70dd3c5529d4..64297c72464f 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -23,16 +23,8 @@ static void mga_user_framebuffer_destroy(struct drm_framebuffer *fb)
 	kfree(fb);
 }
 
-static int mga_user_framebuffer_create_handle(struct drm_framebuffer *fb,
-						 struct drm_file *file_priv,
-						 unsigned int *handle)
-{
-	return 0;
-}
-
 static const struct drm_framebuffer_funcs mga_fb_funcs = {
 	.destroy = mga_user_framebuffer_destroy,
-	.create_handle = mga_user_framebuffer_create_handle,
 };
 
 int mgag200_framebuffer_init(struct drm_device *dev,
@@ -40,13 +32,15 @@ int mgag200_framebuffer_init(struct drm_device *dev,
 			     struct drm_mode_fb_cmd2 *mode_cmd,
 			     struct drm_gem_object *obj)
 {
-	int ret = drm_framebuffer_init(dev, &gfb->base, &mga_fb_funcs);
+	int ret;
+	
+	drm_helper_mode_fill_fb_struct(&gfb->base, mode_cmd);
+	gfb->obj = obj;
+	ret = drm_framebuffer_init(dev, &gfb->base, &mga_fb_funcs);
 	if (ret) {
 		DRM_ERROR("drm_framebuffer_init failed: %d\n", ret);
 		return ret;
 	}
-	drm_helper_mode_fill_fb_struct(&gfb->base, mode_cmd);
-	gfb->obj = obj;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index 8a55beeb8bdc..a7ff6d5a34b9 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -11,8 +11,9 @@ config DRM_NOUVEAU
 	select FRAMEBUFFER_CONSOLE if !EXPERT
 	select FB_BACKLIGHT if DRM_NOUVEAU_BACKLIGHT
 	select ACPI_VIDEO if ACPI && X86 && BACKLIGHT_CLASS_DEVICE && VIDEO_OUTPUT_CONTROL && INPUT
-	select ACPI_WMI if ACPI
-	select MXM_WMI if ACPI
+	select X86_PLATFORM_DEVICES if ACPI && X86
+	select ACPI_WMI if ACPI && X86
+	select MXM_WMI if ACPI && X86
 	select POWER_SUPPLY
 	help
 	  Choose this option for open-source nVidia support.
@@ -52,26 +53,3 @@ config DRM_NOUVEAU_BACKLIGHT
 	help
 	  Say Y here if you want to control the backlight of your display
 	  (e.g. a laptop panel).
-
-menu "I2C encoder or helper chips"
-     depends on DRM && DRM_KMS_HELPER && I2C
-
-config DRM_I2C_CH7006
-	tristate "Chrontel ch7006 TV encoder"
-	default m if DRM_NOUVEAU
-	help
-	  Support for Chrontel ch7006 and similar TV encoders, found
-	  on some nVidia video cards.
-
-	  This driver is currently only useful if you're also using
-	  the nouveau driver.
-
-config DRM_I2C_SIL164
-	tristate "Silicon Image sil164 TMDS transmitter"
-	default m if DRM_NOUVEAU
-	help
-	  Support for sil164 and similar single-link (or dual-link
-	  when used in pairs) TMDS transmitters, used in some nVidia
-	  video cards.
-
-endmenu
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index ab25752a0b1e..90f9140eeefd 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -11,6 +11,7 @@ nouveau-y := core/core/client.o
 nouveau-y += core/core/engctx.o
 nouveau-y += core/core/engine.o
 nouveau-y += core/core/enum.o
+nouveau-y += core/core/event.o
 nouveau-y += core/core/falcon.o
 nouveau-y += core/core/gpuobj.o
 nouveau-y += core/core/handle.o
@@ -40,6 +41,11 @@ nouveau-y += core/subdev/bios/mxm.o
 nouveau-y += core/subdev/bios/perf.o
 nouveau-y += core/subdev/bios/pll.o
 nouveau-y += core/subdev/bios/therm.o
+nouveau-y += core/subdev/bios/xpio.o
+nouveau-y += core/subdev/bus/nv04.o
+nouveau-y += core/subdev/bus/nv31.o
+nouveau-y += core/subdev/bus/nv50.o
+nouveau-y += core/subdev/bus/nvc0.o
 nouveau-y += core/subdev/clock/nv04.o
 nouveau-y += core/subdev/clock/nv40.o
 nouveau-y += core/subdev/clock/nv50.o
@@ -85,9 +91,16 @@ nouveau-y += core/subdev/gpio/base.o
 nouveau-y += core/subdev/gpio/nv10.o
 nouveau-y += core/subdev/gpio/nv50.o
 nouveau-y += core/subdev/gpio/nvd0.o
+nouveau-y += core/subdev/gpio/nve0.o
 nouveau-y += core/subdev/i2c/base.o
+nouveau-y += core/subdev/i2c/anx9805.o
 nouveau-y += core/subdev/i2c/aux.o
 nouveau-y += core/subdev/i2c/bit.o
+nouveau-y += core/subdev/i2c/nv04.o
+nouveau-y += core/subdev/i2c/nv4e.o
+nouveau-y += core/subdev/i2c/nv50.o
+nouveau-y += core/subdev/i2c/nv94.o
+nouveau-y += core/subdev/i2c/nvd0.o
 nouveau-y += core/subdev/ibus/nvc0.o
 nouveau-y += core/subdev/ibus/nve0.o
 nouveau-y += core/subdev/instmem/base.o
@@ -106,10 +119,15 @@ nouveau-y += core/subdev/mxm/mxms.o
 nouveau-y += core/subdev/mxm/nv50.o
 nouveau-y += core/subdev/therm/base.o
 nouveau-y += core/subdev/therm/fan.o
+nouveau-y += core/subdev/therm/fannil.o
+nouveau-y += core/subdev/therm/fanpwm.o
+nouveau-y += core/subdev/therm/fantog.o
 nouveau-y += core/subdev/therm/ic.o
+nouveau-y += core/subdev/therm/temp.o
 nouveau-y += core/subdev/therm/nv40.o
 nouveau-y += core/subdev/therm/nv50.o
-nouveau-y += core/subdev/therm/temp.o
+nouveau-y += core/subdev/therm/nva3.o
+nouveau-y += core/subdev/therm/nvd0.o
 nouveau-y += core/subdev/timer/base.o
 nouveau-y += core/subdev/timer/nv04.o
 nouveau-y += core/subdev/vm/base.o
@@ -132,6 +150,7 @@ nouveau-y += core/engine/copy/nvc0.o
 nouveau-y += core/engine/copy/nve0.o
 nouveau-y += core/engine/crypt/nv84.o
 nouveau-y += core/engine/crypt/nv98.o
+nouveau-y += core/engine/disp/base.o
 nouveau-y += core/engine/disp/nv04.o
 nouveau-y += core/engine/disp/nv50.o
 nouveau-y += core/engine/disp/nv84.o
@@ -141,11 +160,13 @@ nouveau-y += core/engine/disp/nva3.o
 nouveau-y += core/engine/disp/nvd0.o
 nouveau-y += core/engine/disp/nve0.o
 nouveau-y += core/engine/disp/dacnv50.o
+nouveau-y += core/engine/disp/dport.o
 nouveau-y += core/engine/disp/hdanva3.o
 nouveau-y += core/engine/disp/hdanvd0.o
 nouveau-y += core/engine/disp/hdminv84.o
 nouveau-y += core/engine/disp/hdminva3.o
 nouveau-y += core/engine/disp/hdminvd0.o
+nouveau-y += core/engine/disp/piornv50.o
 nouveau-y += core/engine/disp/sornv50.o
 nouveau-y += core/engine/disp/sornv94.o
 nouveau-y += core/engine/disp/sornvd0.o
@@ -194,7 +215,8 @@ nouveau-y += nouveau_drm.o nouveau_chan.o nouveau_dma.o nouveau_fence.o
 nouveau-y += nouveau_irq.o nouveau_vga.o nouveau_agp.o
 nouveau-y += nouveau_ttm.o nouveau_sgdma.o nouveau_bo.o nouveau_gem.o
 nouveau-y += nouveau_prime.o nouveau_abi16.o
-nouveau-y += nv04_fence.o nv10_fence.o nv50_fence.o nv84_fence.o nvc0_fence.o
+nouveau-y += nv04_fence.o nv10_fence.o nv17_fence.o
+nouveau-y += nv50_fence.o nv84_fence.o nvc0_fence.o
 
 # drm/kms
 nouveau-y += nouveau_bios.o nouveau_fbcon.o nouveau_display.o
@@ -216,7 +238,10 @@ nouveau-y += nouveau_mem.o
 
 # other random bits
 nouveau-$(CONFIG_COMPAT) += nouveau_ioc32.o
+ifdef CONFIG_X86
 nouveau-$(CONFIG_ACPI) += nouveau_acpi.o
+endif
 nouveau-$(CONFIG_DRM_NOUVEAU_BACKLIGHT) += nouveau_backlight.o
+nouveau-$(CONFIG_DEBUG_FS) += nouveau_debugfs.o
 
 obj-$(CONFIG_DRM_NOUVEAU)+= nouveau.o
diff --git a/drivers/gpu/drm/nouveau/core/core/client.c b/drivers/gpu/drm/nouveau/core/core/client.c
index 8bbb58f94a19..295c22165eac 100644
--- a/drivers/gpu/drm/nouveau/core/core/client.c
+++ b/drivers/gpu/drm/nouveau/core/core/client.c
@@ -99,3 +99,13 @@ nouveau_client_fini(struct nouveau_client *client, bool suspend)
 	nv_debug(client, "%s completed with %d\n", name[suspend], ret);
 	return ret;
 }
+
+const char *
+nouveau_client_name(void *obj)
+{
+	const char *client_name = "unknown";
+	struct nouveau_client *client = nouveau_client(obj);
+	if (client)
+		client_name = client->name;
+	return client_name;
+}
diff --git a/drivers/gpu/drm/nouveau/core/core/enum.c b/drivers/gpu/drm/nouveau/core/core/enum.c
index 7cc7133d82de..dd434790ccc4 100644
--- a/drivers/gpu/drm/nouveau/core/core/enum.c
+++ b/drivers/gpu/drm/nouveau/core/core/enum.c
@@ -40,14 +40,15 @@ nouveau_enum_find(const struct nouveau_enum *en, u32 value)
 	return NULL;
 }
 
-void
+const struct nouveau_enum *
 nouveau_enum_print(const struct nouveau_enum *en, u32 value)
 {
 	en = nouveau_enum_find(en, value);
 	if (en)
-		printk("%s", en->name);
+		pr_cont("%s", en->name);
 	else
-		printk("(unknown enum 0x%08x)", value);
+		pr_cont("(unknown enum 0x%08x)", value);
+	return en;
 }
 
 void
@@ -55,7 +56,7 @@ nouveau_bitfield_print(const struct nouveau_bitfield *bf, u32 value)
 {
 	while (bf->name) {
 		if (value & bf->mask) {
-			printk(" %s", bf->name);
+			pr_cont(" %s", bf->name);
 			value &= ~bf->mask;
 		}
 
@@ -63,5 +64,5 @@ nouveau_bitfield_print(const struct nouveau_bitfield *bf, u32 value)
 	}
 
 	if (value)
-		printk(" (unknown bits 0x%08x)", value);
+		pr_cont(" (unknown bits 0x%08x)", value);
 }
diff --git a/drivers/gpu/drm/nouveau/core/core/event.c b/drivers/gpu/drm/nouveau/core/core/event.c
new file mode 100644
index 000000000000..6d01e0f0fc8a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/core/event.c
@@ -0,0 +1,106 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <core/os.h>
+#include <core/event.h>
+
+static void
+nouveau_event_put_locked(struct nouveau_event *event, int index,
+			 struct nouveau_eventh *handler)
+{
+	if (!--event->index[index].refs)
+		event->disable(event, index);
+	list_del(&handler->head);
+}
+
+void
+nouveau_event_put(struct nouveau_event *event, int index,
+		  struct nouveau_eventh *handler)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&event->lock, flags);
+	if (index < event->index_nr)
+		nouveau_event_put_locked(event, index, handler);
+	spin_unlock_irqrestore(&event->lock, flags);
+}
+
+void
+nouveau_event_get(struct nouveau_event *event, int index,
+		  struct nouveau_eventh *handler)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&event->lock, flags);
+	if (index < event->index_nr) {
+		list_add(&handler->head, &event->index[index].list);
+		if (!event->index[index].refs++)
+			event->enable(event, index);
+	}
+	spin_unlock_irqrestore(&event->lock, flags);
+}
+
+void
+nouveau_event_trigger(struct nouveau_event *event, int index)
+{
+	struct nouveau_eventh *handler, *temp;
+	unsigned long flags;
+
+	if (index >= event->index_nr)
+		return;
+
+	spin_lock_irqsave(&event->lock, flags);
+	list_for_each_entry_safe(handler, temp, &event->index[index].list, head) {
+		if (handler->func(handler, index) == NVKM_EVENT_DROP) {
+			nouveau_event_put_locked(event, index, handler);
+		}
+	}
+	spin_unlock_irqrestore(&event->lock, flags);
+}
+
+void
+nouveau_event_destroy(struct nouveau_event **pevent)
+{
+	struct nouveau_event *event = *pevent;
+	if (event) {
+		kfree(event);
+		*pevent = NULL;
+	}
+}
+
+int
+nouveau_event_create(int index_nr, struct nouveau_event **pevent)
+{
+	struct nouveau_event *event;
+	int i;
+
+	event = *pevent = kzalloc(sizeof(*event) + index_nr *
+				  sizeof(event->index[0]), GFP_KERNEL);
+	if (!event)
+		return -ENOMEM;
+
+	spin_lock_init(&event->lock);
+	for (i = 0; i < index_nr; i++)
+		INIT_LIST_HEAD(&event->index[i].list);
+	event->index_nr = index_nr;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/copy/nva3.c b/drivers/gpu/drm/nouveau/core/engine/copy/nva3.c
index 283248c7b050..d6dc2a65ccd1 100644
--- a/drivers/gpu/drm/nouveau/core/engine/copy/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/engine/copy/nva3.c
@@ -22,6 +22,7 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/falcon.h>
 #include <core/class.h>
 #include <core/enum.h>
@@ -100,8 +101,9 @@ nva3_copy_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00000040) {
 		nv_error(falcon, "DISPATCH_ERROR [");
 		nouveau_enum_print(nva3_copy_isr_error_name, ssta);
-		printk("] ch %d [0x%010llx] subc %d mthd 0x%04x data 0x%08x\n",
-		       chid, inst << 12, subc, mthd, data);
+		pr_cont("] ch %d [0x%010llx %s] subc %d mthd 0x%04x data 0x%08x\n",
+		       chid, inst << 12, nouveau_client_name(engctx), subc,
+		       mthd, data);
 		nv_wo32(falcon, 0x004, 0x00000040);
 		stat &= ~0x00000040;
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/crypt/nv84.c b/drivers/gpu/drm/nouveau/core/engine/crypt/nv84.c
index b97490512723..5bc021f471f9 100644
--- a/drivers/gpu/drm/nouveau/core/engine/crypt/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/crypt/nv84.c
@@ -22,6 +22,7 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/enum.h>
 #include <core/class.h>
@@ -126,10 +127,11 @@ nv84_crypt_intr(struct nouveau_subdev *subdev)
 	chid   = pfifo->chid(pfifo, engctx);
 
 	if (stat) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv84_crypt_intr_mask, stat);
-		printk(" ch %d [0x%010llx] mthd 0x%04x data 0x%08x\n",
-		       chid, (u64)inst << 12, mthd, data);
+		pr_cont(" ch %d [0x%010llx %s] mthd 0x%04x data 0x%08x\n",
+		       chid, (u64)inst << 12, nouveau_client_name(engctx),
+		       mthd, data);
 	}
 
 	nv_wr32(priv, 0x102130, stat);
diff --git a/drivers/gpu/drm/nouveau/core/engine/crypt/nv98.c b/drivers/gpu/drm/nouveau/core/engine/crypt/nv98.c
index 21986f3bf0c8..8bf8955051d4 100644
--- a/drivers/gpu/drm/nouveau/core/engine/crypt/nv98.c
+++ b/drivers/gpu/drm/nouveau/core/engine/crypt/nv98.c
@@ -22,6 +22,7 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/enum.h>
 #include <core/class.h>
@@ -102,8 +103,9 @@ nv98_crypt_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00000040) {
 		nv_error(priv, "DISPATCH_ERROR [");
 		nouveau_enum_print(nv98_crypt_isr_error_name, ssta);
-		printk("] ch %d [0x%010llx] subc %d mthd 0x%04x data 0x%08x\n",
-		       chid, (u64)inst << 12, subc, mthd, data);
+		pr_cont("] ch %d [0x%010llx %s] subc %d mthd 0x%04x data 0x%08x\n",
+		       chid, (u64)inst << 12, nouveau_client_name(engctx),
+		       subc, mthd, data);
 		nv_wr32(priv, 0x087004, 0x00000040);
 		stat &= ~0x00000040;
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/base.c b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
new file mode 100644
index 000000000000..7a5cae42834f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
@@ -0,0 +1,52 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <engine/disp.h>
+
+void
+_nouveau_disp_dtor(struct nouveau_object *object)
+{
+	struct nouveau_disp *disp = (void *)object;
+	nouveau_event_destroy(&disp->vblank);
+	nouveau_engine_destroy(&disp->base);
+}
+
+int
+nouveau_disp_create_(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, int heads,
+		     const char *intname, const char *extname,
+		     int length, void **pobject)
+{
+	struct nouveau_disp *disp;
+	int ret;
+
+	ret = nouveau_engine_create_(parent, engine, oclass, true,
+				     intname, extname, length, pobject);
+	disp = *pobject;
+	if (ret)
+		return ret;
+
+	return nouveau_event_create(heads, &disp->vblank);
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
new file mode 100644
index 000000000000..fa27b02ff829
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
@@ -0,0 +1,346 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/dcb.h>
+#include <subdev/bios/dp.h>
+#include <subdev/bios/init.h>
+#include <subdev/i2c.h>
+
+#include <engine/disp.h>
+
+#include "dport.h"
+
+#define DBG(fmt, args...) nv_debug(dp->disp, "DP:%04x:%04x: " fmt,             \
+				   dp->outp->hasht, dp->outp->hashm, ##args)
+#define ERR(fmt, args...) nv_error(dp->disp, "DP:%04x:%04x: " fmt,             \
+				   dp->outp->hasht, dp->outp->hashm, ##args)
+
+/******************************************************************************
+ * link training
+ *****************************************************************************/
+struct dp_state {
+	const struct nouveau_dp_func *func;
+	struct nouveau_disp *disp;
+	struct dcb_output *outp;
+	struct nvbios_dpout info;
+	u8 version;
+	struct nouveau_i2c_port *aux;
+	int head;
+	u8  dpcd[4];
+	int link_nr;
+	u32 link_bw;
+	u8  stat[6];
+	u8  conf[4];
+};
+
+static int
+dp_set_link_config(struct dp_state *dp)
+{
+	struct nouveau_disp *disp = dp->disp;
+	struct nouveau_bios *bios = nouveau_bios(disp);
+	struct nvbios_init init = {
+		.subdev = nv_subdev(dp->disp),
+		.bios = bios,
+		.offset = 0x0000,
+		.outp = dp->outp,
+		.crtc = dp->head,
+		.execute = 1,
+	};
+	u32 lnkcmp;
+	u8 sink[2];
+
+	DBG("%d lanes at %d KB/s\n", dp->link_nr, dp->link_bw);
+
+	/* set desired link configuration on the sink */
+	sink[0] = dp->link_bw / 27000;
+	sink[1] = dp->link_nr;
+	if (dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
+		sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN;
+
+	nv_wraux(dp->aux, DPCD_LC00, sink, 2);
+
+	/* set desired link configuration on the source */
+	if ((lnkcmp = dp->info.lnkcmp)) {
+		if (dp->version < 0x30) {
+			while ((dp->link_bw / 10) < nv_ro16(bios, lnkcmp))
+				lnkcmp += 4;
+			init.offset = nv_ro16(bios, lnkcmp + 2);
+		} else {
+			while ((dp->link_bw / 27000) < nv_ro08(bios, lnkcmp))
+				lnkcmp += 3;
+			init.offset = nv_ro16(bios, lnkcmp + 1);
+		}
+
+		nvbios_exec(&init);
+	}
+
+	return dp->func->lnk_ctl(dp->disp, dp->outp, dp->head,
+				 dp->link_nr, dp->link_bw / 27000,
+				 dp->dpcd[DPCD_RC02] &
+					  DPCD_RC02_ENHANCED_FRAME_CAP);
+}
+
+static void
+dp_set_training_pattern(struct dp_state *dp, u8 pattern)
+{
+	u8 sink_tp;
+
+	DBG("training pattern %d\n", pattern);
+	dp->func->pattern(dp->disp, dp->outp, dp->head, pattern);
+
+	nv_rdaux(dp->aux, DPCD_LC02, &sink_tp, 1);
+	sink_tp &= ~DPCD_LC02_TRAINING_PATTERN_SET;
+	sink_tp |= pattern;
+	nv_wraux(dp->aux, DPCD_LC02, &sink_tp, 1);
+}
+
+static int
+dp_link_train_commit(struct dp_state *dp)
+{
+	int i;
+
+	for (i = 0; i < dp->link_nr; i++) {
+		u8 lane = (dp->stat[4 + (i >> 1)] >> ((i & 1) * 4)) & 0xf;
+		u8 lpre = (lane & 0x0c) >> 2;
+		u8 lvsw = (lane & 0x03) >> 0;
+
+		dp->conf[i] = (lpre << 3) | lvsw;
+		if (lvsw == 3)
+			dp->conf[i] |= DPCD_LC03_MAX_SWING_REACHED;
+		if (lpre == 3)
+			dp->conf[i] |= DPCD_LC03_MAX_PRE_EMPHASIS_REACHED;
+
+		DBG("config lane %d %02x\n", i, dp->conf[i]);
+		dp->func->drv_ctl(dp->disp, dp->outp, dp->head, i, lvsw, lpre);
+	}
+
+	return nv_wraux(dp->aux, DPCD_LC03(0), dp->conf, 4);
+}
+
+static int
+dp_link_train_update(struct dp_state *dp, u32 delay)
+{
+	int ret;
+
+	udelay(delay);
+
+	ret = nv_rdaux(dp->aux, DPCD_LS02, dp->stat, 6);
+	if (ret)
+		return ret;
+
+	DBG("status %*ph\n", 6, dp->stat);
+	return 0;
+}
+
+static int
+dp_link_train_cr(struct dp_state *dp)
+{
+	bool cr_done = false, abort = false;
+	int voltage = dp->conf[0] & DPCD_LC03_VOLTAGE_SWING_SET;
+	int tries = 0, i;
+
+	dp_set_training_pattern(dp, 1);
+
+	do {
+		if (dp_link_train_commit(dp) ||
+		    dp_link_train_update(dp, 100))
+			break;
+
+		cr_done = true;
+		for (i = 0; i < dp->link_nr; i++) {
+			u8 lane = (dp->stat[i >> 1] >> ((i & 1) * 4)) & 0xf;
+			if (!(lane & DPCD_LS02_LANE0_CR_DONE)) {
+				cr_done = false;
+				if (dp->conf[i] & DPCD_LC03_MAX_SWING_REACHED)
+					abort = true;
+				break;
+			}
+		}
+
+		if ((dp->conf[0] & DPCD_LC03_VOLTAGE_SWING_SET) != voltage) {
+			voltage = dp->conf[0] & DPCD_LC03_VOLTAGE_SWING_SET;
+			tries = 0;
+		}
+	} while (!cr_done && !abort && ++tries < 5);
+
+	return cr_done ? 0 : -1;
+}
+
+static int
+dp_link_train_eq(struct dp_state *dp)
+{
+	bool eq_done, cr_done = true;
+	int tries = 0, i;
+
+	dp_set_training_pattern(dp, 2);
+
+	do {
+		if (dp_link_train_update(dp, 400))
+			break;
+
+		eq_done = !!(dp->stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE);
+		for (i = 0; i < dp->link_nr && eq_done; i++) {
+			u8 lane = (dp->stat[i >> 1] >> ((i & 1) * 4)) & 0xf;
+			if (!(lane & DPCD_LS02_LANE0_CR_DONE))
+				cr_done = false;
+			if (!(lane & DPCD_LS02_LANE0_CHANNEL_EQ_DONE) ||
+			    !(lane & DPCD_LS02_LANE0_SYMBOL_LOCKED))
+				eq_done = false;
+		}
+
+		if (dp_link_train_commit(dp))
+			break;
+	} while (!eq_done && cr_done && ++tries <= 5);
+
+	return eq_done ? 0 : -1;
+}
+
+static void
+dp_link_train_init(struct dp_state *dp, bool spread)
+{
+	struct nvbios_init init = {
+		.subdev = nv_subdev(dp->disp),
+		.bios = nouveau_bios(dp->disp),
+		.outp = dp->outp,
+		.crtc = dp->head,
+		.execute = 1,
+	};
+
+	/* set desired spread */
+	if (spread)
+		init.offset = dp->info.script[2];
+	else
+		init.offset = dp->info.script[3];
+	nvbios_exec(&init);
+
+	/* pre-train script */
+	init.offset = dp->info.script[0];
+	nvbios_exec(&init);
+}
+
+static void
+dp_link_train_fini(struct dp_state *dp)
+{
+	struct nvbios_init init = {
+		.subdev = nv_subdev(dp->disp),
+		.bios = nouveau_bios(dp->disp),
+		.outp = dp->outp,
+		.crtc = dp->head,
+		.execute = 1,
+	};
+
+	/* post-train script */
+	init.offset = dp->info.script[1],
+	nvbios_exec(&init);
+}
+
+int
+nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
+		 struct dcb_output *outp, int head, u32 datarate)
+{
+	struct nouveau_bios *bios = nouveau_bios(disp);
+	struct nouveau_i2c *i2c = nouveau_i2c(disp);
+	struct dp_state _dp = {
+		.disp = disp,
+		.func = func,
+		.outp = outp,
+		.head = head,
+	}, *dp = &_dp;
+	const u32 bw_list[] = { 270000, 162000, 0 };
+	const u32 *link_bw = bw_list;
+	u8  hdr, cnt, len;
+	u32 data;
+	int ret;
+
+	/* find the bios displayport data relevant to this output */
+	data = nvbios_dpout_match(bios, outp->hasht, outp->hashm, &dp->version,
+				 &hdr, &cnt, &len, &dp->info);
+	if (!data) {
+		ERR("bios data not found\n");
+		return -EINVAL;
+	}
+
+	/* acquire the aux channel and fetch some info about the display */
+	if (outp->location)
+		dp->aux = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
+	else
+		dp->aux = i2c->find(i2c, NV_I2C_TYPE_DCBI2C(outp->i2c_index));
+	if (!dp->aux) {
+		ERR("no aux channel?!\n");
+		return -ENODEV;
+	}
+
+	ret = nv_rdaux(dp->aux, 0x00000, dp->dpcd, sizeof(dp->dpcd));
+	if (ret) {
+		ERR("failed to read DPCD\n");
+		return ret;
+	}
+
+	/* adjust required bandwidth for 8B/10B coding overhead */
+	datarate = (datarate / 8) * 10;
+
+	/* enable down-spreading and execute pre-train script from vbios */
+	dp_link_train_init(dp, dp->dpcd[3] & 0x01);
+
+	/* start off at highest link rate supported by encoder and display */
+	while (*link_bw > (dp->dpcd[1] * 27000))
+		link_bw++;
+
+	while (link_bw[0]) {
+		/* find minimum required lane count at this link rate */
+		dp->link_nr = dp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT;
+		while ((dp->link_nr >> 1) * link_bw[0] > datarate)
+			dp->link_nr >>= 1;
+
+		/* drop link rate to minimum with this lane count */
+		while ((link_bw[1] * dp->link_nr) > datarate)
+			link_bw++;
+		dp->link_bw = link_bw[0];
+
+		/* program selected link configuration */
+		ret = dp_set_link_config(dp);
+		if (ret == 0) {
+			/* attempt to train the link at this configuration */
+			memset(dp->stat, 0x00, sizeof(dp->stat));
+			if (!dp_link_train_cr(dp) &&
+			    !dp_link_train_eq(dp))
+				break;
+		} else
+		if (ret >= 1) {
+			/* dp_set_link_config() handled training */
+			break;
+		}
+
+		/* retry at lower rate */
+		link_bw++;
+	}
+
+	/* finish link training */
+	dp_set_training_pattern(dp, 0);
+
+	/* execute post-train script from vbios */
+	dp_link_train_fini(dp);
+	return true;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.h b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
new file mode 100644
index 000000000000..0e1bbd18ff6c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
@@ -0,0 +1,78 @@
+#ifndef __NVKM_DISP_DPORT_H__
+#define __NVKM_DISP_DPORT_H__
+
+/* DPCD Receiver Capabilities */
+#define DPCD_RC00                                                       0x00000
+#define DPCD_RC00_DPCD_REV                                                 0xff
+#define DPCD_RC01                                                       0x00001
+#define DPCD_RC01_MAX_LINK_RATE                                            0xff
+#define DPCD_RC02                                                       0x00002
+#define DPCD_RC02_ENHANCED_FRAME_CAP                                       0x80
+#define DPCD_RC02_MAX_LANE_COUNT                                           0x1f
+#define DPCD_RC03                                                       0x00003
+#define DPCD_RC03_MAX_DOWNSPREAD                                           0x01
+
+/* DPCD Link Configuration */
+#define DPCD_LC00                                                       0x00100
+#define DPCD_LC00_LINK_BW_SET                                              0xff
+#define DPCD_LC01                                                       0x00101
+#define DPCD_LC01_ENHANCED_FRAME_EN                                        0x80
+#define DPCD_LC01_LANE_COUNT_SET                                           0x1f
+#define DPCD_LC02                                                       0x00102
+#define DPCD_LC02_TRAINING_PATTERN_SET                                     0x03
+#define DPCD_LC03(l)                                            ((l) +  0x00103)
+#define DPCD_LC03_MAX_PRE_EMPHASIS_REACHED                                 0x20
+#define DPCD_LC03_PRE_EMPHASIS_SET                                         0x18
+#define DPCD_LC03_MAX_SWING_REACHED                                        0x04
+#define DPCD_LC03_VOLTAGE_SWING_SET                                        0x03
+
+/* DPCD Link/Sink Status */
+#define DPCD_LS02                                                       0x00202
+#define DPCD_LS02_LANE1_SYMBOL_LOCKED                                      0x40
+#define DPCD_LS02_LANE1_CHANNEL_EQ_DONE                                    0x20
+#define DPCD_LS02_LANE1_CR_DONE                                            0x10
+#define DPCD_LS02_LANE0_SYMBOL_LOCKED                                      0x04
+#define DPCD_LS02_LANE0_CHANNEL_EQ_DONE                                    0x02
+#define DPCD_LS02_LANE0_CR_DONE                                            0x01
+#define DPCD_LS03                                                       0x00203
+#define DPCD_LS03_LANE3_SYMBOL_LOCKED                                      0x40
+#define DPCD_LS03_LANE3_CHANNEL_EQ_DONE                                    0x20
+#define DPCD_LS03_LANE3_CR_DONE                                            0x10
+#define DPCD_LS03_LANE2_SYMBOL_LOCKED                                      0x04
+#define DPCD_LS03_LANE2_CHANNEL_EQ_DONE                                    0x02
+#define DPCD_LS03_LANE2_CR_DONE                                            0x01
+#define DPCD_LS04                                                       0x00204
+#define DPCD_LS04_LINK_STATUS_UPDATED                                      0x80
+#define DPCD_LS04_DOWNSTREAM_PORT_STATUS_CHANGED                           0x40
+#define DPCD_LS04_INTERLANE_ALIGN_DONE                                     0x01
+#define DPCD_LS06                                                       0x00206
+#define DPCD_LS06_LANE1_PRE_EMPHASIS                                       0xc0
+#define DPCD_LS06_LANE1_VOLTAGE_SWING                                      0x30
+#define DPCD_LS06_LANE0_PRE_EMPHASIS                                       0x0c
+#define DPCD_LS06_LANE0_VOLTAGE_SWING                                      0x03
+#define DPCD_LS07                                                       0x00207
+#define DPCD_LS07_LANE3_PRE_EMPHASIS                                       0xc0
+#define DPCD_LS07_LANE3_VOLTAGE_SWING                                      0x30
+#define DPCD_LS07_LANE2_PRE_EMPHASIS                                       0x0c
+#define DPCD_LS07_LANE2_VOLTAGE_SWING                                      0x03
+
+struct nouveau_disp;
+struct dcb_output;
+
+struct nouveau_dp_func {
+	int (*pattern)(struct nouveau_disp *, struct dcb_output *,
+		       int head, int pattern);
+	int (*lnk_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
+		       int link_nr, int link_bw, bool enh_frame);
+	int (*drv_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
+		       int lane, int swing, int preem);
+};
+
+extern const struct nouveau_dp_func nv94_sor_dp_func;
+extern const struct nouveau_dp_func nvd0_sor_dp_func;
+extern const struct nouveau_dp_func nv50_pior_dp_func;
+
+int nouveau_dp_train(struct nouveau_disp *, const struct nouveau_dp_func *,
+		     struct dcb_output *, int, u32);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
index 1c919f2af89f..05e903f08a36 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
@@ -24,21 +24,33 @@
 
 #include <engine/disp.h>
 
+#include <core/event.h>
+#include <core/class.h>
+
 struct nv04_disp_priv {
 	struct nouveau_disp base;
 };
 
 static struct nouveau_oclass
 nv04_disp_sclass[] = {
+	{ NV04_DISP_CLASS, &nouveau_object_ofuncs },
 	{},
 };
 
+/*******************************************************************************
+ * Display engine implementation
+ ******************************************************************************/
+
+static void
+nv04_disp_vblank_enable(struct nouveau_event *event, int head)
+{
+	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000001);
+}
+
 static void
-nv04_disp_intr_vblank(struct nv04_disp_priv *priv, int crtc)
+nv04_disp_vblank_disable(struct nouveau_event *event, int head)
 {
-	struct nouveau_disp *disp = &priv->base;
-	if (disp->vblank.notify)
-		disp->vblank.notify(disp->vblank.data, crtc);
+	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000000);
 }
 
 static void
@@ -49,25 +61,25 @@ nv04_disp_intr(struct nouveau_subdev *subdev)
 	u32 crtc1 = nv_rd32(priv, 0x602100);
 
 	if (crtc0 & 0x00000001) {
-		nv04_disp_intr_vblank(priv, 0);
+		nouveau_event_trigger(priv->base.vblank, 0);
 		nv_wr32(priv, 0x600100, 0x00000001);
 	}
 
 	if (crtc1 & 0x00000001) {
-		nv04_disp_intr_vblank(priv, 1);
+		nouveau_event_trigger(priv->base.vblank, 1);
 		nv_wr32(priv, 0x602100, 0x00000001);
 	}
 }
 
 static int
 nv04_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-		  struct nouveau_oclass *oclass, void *data, u32 size,
-		  struct nouveau_object **pobject)
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
 {
 	struct nv04_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "DISPLAY",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "DISPLAY",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -75,6 +87,9 @@ nv04_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 	nv_engine(priv)->sclass = nv04_disp_sclass;
 	nv_subdev(priv)->intr = nv04_disp_intr;
+	priv->base.vblank->priv = priv;
+	priv->base.vblank->enable = nv04_disp_vblank_enable;
+	priv->base.vblank->disable = nv04_disp_vblank_disable;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
index ca1a7d76a95b..5fa13267bd9f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
@@ -27,7 +27,6 @@
 #include <core/handle.h>
 #include <core/class.h>
 
-#include <engine/software.h>
 #include <engine/disp.h>
 
 #include <subdev/bios.h>
@@ -37,7 +36,6 @@
 #include <subdev/bios/pll.h>
 #include <subdev/timer.h>
 #include <subdev/fb.h>
-#include <subdev/bar.h>
 #include <subdev/clock.h>
 
 #include "nv50.h"
@@ -335,7 +333,7 @@ nv50_disp_sync_ctor(struct nouveau_object *parent,
 	struct nv50_disp_dmac *dmac;
 	int ret;
 
-	if (size < sizeof(*data) || args->head > 1)
+	if (size < sizeof(*args) || args->head > 1)
 		return -EINVAL;
 
 	ret = nv50_disp_dmac_create_(parent, engine, oclass, args->pushbuf,
@@ -374,7 +372,7 @@ nv50_disp_ovly_ctor(struct nouveau_object *parent,
 	struct nv50_disp_dmac *dmac;
 	int ret;
 
-	if (size < sizeof(*data) || args->head > 1)
+	if (size < sizeof(*args) || args->head > 1)
 		return -EINVAL;
 
 	ret = nv50_disp_dmac_create_(parent, engine, oclass, args->pushbuf,
@@ -543,6 +541,18 @@ nv50_disp_curs_ofuncs = {
  * Base display object
  ******************************************************************************/
 
+static void
+nv50_disp_base_vblank_enable(struct nouveau_event *event, int head)
+{
+	nv_mask(event->priv, 0x61002c, (1 << head), (1 << head));
+}
+
+static void
+nv50_disp_base_vblank_disable(struct nouveau_event *event, int head)
+{
+	nv_mask(event->priv, 0x61002c, (1 << head), (0 << head));
+}
+
 static int
 nv50_disp_base_ctor(struct nouveau_object *parent,
 		    struct nouveau_object *engine,
@@ -559,6 +569,9 @@ nv50_disp_base_ctor(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
+	priv->base.vblank->priv = priv;
+	priv->base.vblank->enable = nv50_disp_base_vblank_enable;
+	priv->base.vblank->disable = nv50_disp_base_vblank_disable;
 	return nouveau_ramht_new(parent, parent, 0x1000, 0, &base->ramht);
 }
 
@@ -613,7 +626,7 @@ nv50_disp_base_init(struct nouveau_object *object)
 		nv_wr32(priv, 0x6101e0 + (i * 0x04), tmp);
 	}
 
-	/* ... EXT caps */
+	/* ... PIOR caps */
 	for (i = 0; i < 3; i++) {
 		tmp = nv_rd32(priv, 0x61e000 + (i * 0x800));
 		nv_wr32(priv, 0x6101f0 + (i * 0x04), tmp);
@@ -665,6 +678,9 @@ nv50_disp_base_omthds[] = {
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_TMDS_PWR)  , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_DP_PWR)    , nv50_pior_mthd },
 	{},
 };
 
@@ -756,50 +772,6 @@ nv50_disp_intr_error(struct nv50_disp_priv *priv)
 	}
 }
 
-static void
-nv50_disp_intr_vblank(struct nv50_disp_priv *priv, int crtc)
-{
-	struct nouveau_bar *bar = nouveau_bar(priv);
-	struct nouveau_disp *disp = &priv->base;
-	struct nouveau_software_chan *chan, *temp;
-	unsigned long flags;
-
-	spin_lock_irqsave(&disp->vblank.lock, flags);
-	list_for_each_entry_safe(chan, temp, &disp->vblank.list, vblank.head) {
-		if (chan->vblank.crtc != crtc)
-			continue;
-
-		if (nv_device(priv)->chipset >= 0xc0) {
-			nv_wr32(priv, 0x001718, 0x80000000 | chan->vblank.channel);
-			bar->flush(bar);
-			nv_wr32(priv, 0x06000c,
-				upper_32_bits(chan->vblank.offset));
-			nv_wr32(priv, 0x060010,
-				lower_32_bits(chan->vblank.offset));
-			nv_wr32(priv, 0x060014, chan->vblank.value);
-		} else {
-			nv_wr32(priv, 0x001704, chan->vblank.channel);
-			nv_wr32(priv, 0x001710, 0x80000000 | chan->vblank.ctxdma);
-			bar->flush(bar);
-			if (nv_device(priv)->chipset == 0x50) {
-				nv_wr32(priv, 0x001570, chan->vblank.offset);
-				nv_wr32(priv, 0x001574, chan->vblank.value);
-			} else {
-				nv_wr32(priv, 0x060010, chan->vblank.offset);
-				nv_wr32(priv, 0x060014, chan->vblank.value);
-			}
-		}
-
-		list_del(&chan->vblank.head);
-		if (disp->vblank.put)
-			disp->vblank.put(disp->vblank.data, crtc);
-	}
-	spin_unlock_irqrestore(&disp->vblank.lock, flags);
-
-	if (disp->vblank.notify)
-		disp->vblank.notify(disp->vblank.data, crtc);
-}
-
 static u16
 exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 	    struct dcb_output *dcb, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
@@ -811,8 +783,8 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 	if (outp < 4) {
 		type = DCB_OUTPUT_ANALOG;
 		mask = 0;
-	} else {
-		outp -= 4;
+	} else
+	if (outp < 8) {
 		switch (ctrl & 0x00000f00) {
 		case 0x00000000: type = DCB_OUTPUT_LVDS; mask = 1; break;
 		case 0x00000100: type = DCB_OUTPUT_TMDS; mask = 1; break;
@@ -824,6 +796,17 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
 			return 0x0000;
 		}
+		outp -= 4;
+	} else {
+		outp = outp - 8;
+		type = 0x0010;
+		mask = 0;
+		switch (ctrl & 0x00000f00) {
+		case 0x00000000: type |= priv->pior.type[outp]; break;
+		default:
+			nv_error(priv, "unknown PIOR mc 0x%08x\n", ctrl);
+			return 0x0000;
+		}
 	}
 
 	mask  = 0x00c0 & (mask << 6);
@@ -834,6 +817,10 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 	if (!data)
 		return 0x0000;
 
+	/* off-chip encoders require matching the exact encoder type */
+	if (dcb->location != 0)
+		type |= dcb->extdev << 8;
+
 	return nvbios_outp_match(bios, type, mask, ver, hdr, cnt, len, info);
 }
 
@@ -848,9 +835,11 @@ exec_script(struct nv50_disp_priv *priv, int head, int id)
 	u32 ctrl = 0x00000000;
 	int i;
 
+	/* DAC */
 	for (i = 0; !(ctrl & (1 << head)) && i < 3; i++)
 		ctrl = nv_rd32(priv, 0x610b5c + (i * 8));
 
+	/* SOR */
 	if (!(ctrl & (1 << head))) {
 		if (nv_device(priv)->chipset  < 0x90 ||
 		    nv_device(priv)->chipset == 0x92 ||
@@ -865,6 +854,13 @@ exec_script(struct nv50_disp_priv *priv, int head, int id)
 		}
 	}
 
+	/* PIOR */
+	if (!(ctrl & (1 << head))) {
+		for (i = 0; !(ctrl & (1 << head)) && i < 3; i++)
+			ctrl = nv_rd32(priv, 0x610b84 + (i * 8));
+		i += 8;
+	}
+
 	if (!(ctrl & (1 << head)))
 		return false;
 	i--;
@@ -894,13 +890,15 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
 	u8  ver, hdr, cnt, len;
-	u16 data, conf;
 	u32 ctrl = 0x00000000;
+	u32 data, conf = ~0;
 	int i;
 
+	/* DAC */
 	for (i = 0; !(ctrl & (1 << head)) && i < 3; i++)
 		ctrl = nv_rd32(priv, 0x610b58 + (i * 8));
 
+	/* SOR */
 	if (!(ctrl & (1 << head))) {
 		if (nv_device(priv)->chipset  < 0x90 ||
 		    nv_device(priv)->chipset == 0x92 ||
@@ -915,34 +913,46 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 		}
 	}
 
+	/* PIOR */
+	if (!(ctrl & (1 << head))) {
+		for (i = 0; !(ctrl & (1 << head)) && i < 3; i++)
+			ctrl = nv_rd32(priv, 0x610b80 + (i * 8));
+		i += 8;
+	}
+
 	if (!(ctrl & (1 << head)))
-		return 0x0000;
+		return conf;
 	i--;
 
 	data = exec_lookup(priv, head, i, ctrl, outp, &ver, &hdr, &cnt, &len, &info1);
 	if (!data)
-		return 0x0000;
-
-	switch (outp->type) {
-	case DCB_OUTPUT_TMDS:
-		conf = (ctrl & 0x00000f00) >> 8;
-		if (pclk >= 165000)
-			conf |= 0x0100;
-		break;
-	case DCB_OUTPUT_LVDS:
-		conf = priv->sor.lvdsconf;
-		break;
-	case DCB_OUTPUT_DP:
+		return conf;
+
+	if (outp->location == 0) {
+		switch (outp->type) {
+		case DCB_OUTPUT_TMDS:
+			conf = (ctrl & 0x00000f00) >> 8;
+			if (pclk >= 165000)
+				conf |= 0x0100;
+			break;
+		case DCB_OUTPUT_LVDS:
+			conf = priv->sor.lvdsconf;
+			break;
+		case DCB_OUTPUT_DP:
+			conf = (ctrl & 0x00000f00) >> 8;
+			break;
+		case DCB_OUTPUT_ANALOG:
+		default:
+			conf = 0x00ff;
+			break;
+		}
+	} else {
 		conf = (ctrl & 0x00000f00) >> 8;
-		break;
-	case DCB_OUTPUT_ANALOG:
-	default:
-		conf = 0x00ff;
-		break;
+		pclk = pclk / 2;
 	}
 
 	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
-	if (data) {
+	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
 			struct nvbios_init init = {
@@ -954,32 +964,37 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 				.execute = 1,
 			};
 
-			if (nvbios_exec(&init))
-				return 0x0000;
-			return conf;
+			nvbios_exec(&init);
 		}
 	}
 
-	return 0x0000;
+	return conf;
 }
 
 static void
-nv50_disp_intr_unk10(struct nv50_disp_priv *priv, u32 super)
+nv50_disp_intr_unk10_0(struct nv50_disp_priv *priv, int head)
 {
-	int head = ffs((super & 0x00000060) >> 5) - 1;
-	if (head >= 0) {
-		head = ffs((super & 0x00000180) >> 7) - 1;
-		if (head >= 0)
-			exec_script(priv, head, 1);
-	}
+	exec_script(priv, head, 1);
+}
 
-	nv_wr32(priv, 0x610024, 0x00000010);
-	nv_wr32(priv, 0x610030, 0x80000000);
+static void
+nv50_disp_intr_unk20_0(struct nv50_disp_priv *priv, int head)
+{
+	exec_script(priv, head, 2);
+}
+
+static void
+nv50_disp_intr_unk20_1(struct nv50_disp_priv *priv, int head)
+{
+	struct nouveau_clock *clk = nouveau_clock(priv);
+	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
+	if (pclk)
+		clk->pll_set(clk, PLL_VPLL0 + head, pclk);
 }
 
 static void
-nv50_disp_intr_unk20_dp(struct nv50_disp_priv *priv,
-		        struct dcb_output *outp, u32 pclk)
+nv50_disp_intr_unk20_2_dp(struct nv50_disp_priv *priv,
+			  struct dcb_output *outp, u32 pclk)
 {
 	const int link = !(outp->sorconf.link & 1);
 	const int   or = ffs(outp->or) - 1;
@@ -1085,53 +1100,54 @@ nv50_disp_intr_unk20_dp(struct nv50_disp_priv *priv,
 }
 
 static void
-nv50_disp_intr_unk20(struct nv50_disp_priv *priv, u32 super)
+nv50_disp_intr_unk20_2(struct nv50_disp_priv *priv, int head)
 {
 	struct dcb_output outp;
-	u32 addr, mask, data;
-	int head;
+	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
+	u32 hval, hreg = 0x614200 + (head * 0x800);
+	u32 oval, oreg;
+	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
+	if (conf != ~0) {
+		if (outp.location == 0 && outp.type == DCB_OUTPUT_DP) {
+			u32 soff = (ffs(outp.or) - 1) * 0x08;
+			u32 ctrl = nv_rd32(priv, 0x610798 + soff);
+			u32 datarate;
+
+			switch ((ctrl & 0x000f0000) >> 16) {
+			case 6: datarate = pclk * 30 / 8; break;
+			case 5: datarate = pclk * 24 / 8; break;
+			case 2:
+			default:
+				datarate = pclk * 18 / 8;
+				break;
+			}
 
-	/* finish detaching encoder? */
-	head = ffs((super & 0x00000180) >> 7) - 1;
-	if (head >= 0)
-		exec_script(priv, head, 2);
-
-	/* check whether a vpll change is required */
-	head = ffs((super & 0x00000600) >> 9) - 1;
-	if (head >= 0) {
-		u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-		if (pclk) {
-			struct nouveau_clock *clk = nouveau_clock(priv);
-			clk->pll_set(clk, PLL_VPLL0 + head, pclk);
+			nouveau_dp_train(&priv->base, priv->sor.dp,
+					 &outp, head, datarate);
 		}
 
-		nv_mask(priv, 0x614200 + head * 0x800, 0x0000000f, 0x00000000);
-	}
-
-	/* (re)attach the relevant OR to the head */
-	head = ffs((super & 0x00000180) >> 7) - 1;
-	if (head >= 0) {
-		u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-		u32 conf = exec_clkcmp(priv, head, 0, pclk, &outp);
-		if (conf) {
-			if (outp.type == DCB_OUTPUT_ANALOG) {
-				addr = 0x614280 + (ffs(outp.or) - 1) * 0x800;
-				mask = 0xffffffff;
-				data = 0x00000000;
-			} else {
-				if (outp.type == DCB_OUTPUT_DP)
-					nv50_disp_intr_unk20_dp(priv, &outp, pclk);
-				addr = 0x614300 + (ffs(outp.or) - 1) * 0x800;
-				mask = 0x00000707;
-				data = (conf & 0x0100) ? 0x0101 : 0x0000;
-			}
-
-			nv_mask(priv, addr, mask, data);
+		exec_clkcmp(priv, head, 0, pclk, &outp);
+
+		if (!outp.location && outp.type == DCB_OUTPUT_ANALOG) {
+			oreg = 0x614280 + (ffs(outp.or) - 1) * 0x800;
+			oval = 0x00000000;
+			hval = 0x00000000;
+		} else
+		if (!outp.location) {
+			if (outp.type == DCB_OUTPUT_DP)
+				nv50_disp_intr_unk20_2_dp(priv, &outp, pclk);
+			oreg = 0x614300 + (ffs(outp.or) - 1) * 0x800;
+			oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
+			hval = 0x00000000;
+		} else {
+			oreg = 0x614380 + (ffs(outp.or) - 1) * 0x800;
+			oval = 0x00000001;
+			hval = 0x00000001;
 		}
-	}
 
-	nv_wr32(priv, 0x610024, 0x00000020);
-	nv_wr32(priv, 0x610030, 0x80000000);
+		nv_mask(priv, hreg, 0x0000000f, hval);
+		nv_mask(priv, oreg, 0x00000707, oval);
+	}
 }
 
 /* If programming a TMDS output on a SOR that can also be configured for
@@ -1143,7 +1159,7 @@ nv50_disp_intr_unk20(struct nv50_disp_priv *priv, u32 super)
  * programmed for DisplayPort.
  */
 static void
-nv50_disp_intr_unk40_tmds(struct nv50_disp_priv *priv, struct dcb_output *outp)
+nv50_disp_intr_unk40_0_tmds(struct nv50_disp_priv *priv, struct dcb_output *outp)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
 	const int link = !(outp->sorconf.link & 1);
@@ -1157,35 +1173,79 @@ nv50_disp_intr_unk40_tmds(struct nv50_disp_priv *priv, struct dcb_output *outp)
 }
 
 static void
-nv50_disp_intr_unk40(struct nv50_disp_priv *priv, u32 super)
+nv50_disp_intr_unk40_0(struct nv50_disp_priv *priv, int head)
 {
-	int head = ffs((super & 0x00000180) >> 7) - 1;
-	if (head >= 0) {
-		struct dcb_output outp;
-		u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-		if (pclk && exec_clkcmp(priv, head, 1, pclk, &outp)) {
-			if (outp.type == DCB_OUTPUT_TMDS)
-				nv50_disp_intr_unk40_tmds(priv, &outp);
+	struct dcb_output outp;
+	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
+	if (exec_clkcmp(priv, head, 1, pclk, &outp) != ~0) {
+		if (outp.location == 0 && outp.type == DCB_OUTPUT_TMDS)
+			nv50_disp_intr_unk40_0_tmds(priv, &outp);
+		else
+		if (outp.location == 1 && outp.type == DCB_OUTPUT_DP) {
+			u32 soff = (ffs(outp.or) - 1) * 0x08;
+			u32 ctrl = nv_rd32(priv, 0x610b84 + soff);
+			u32 datarate;
+
+			switch ((ctrl & 0x000f0000) >> 16) {
+			case 6: datarate = pclk * 30 / 8; break;
+			case 5: datarate = pclk * 24 / 8; break;
+			case 2:
+			default:
+				datarate = pclk * 18 / 8;
+				break;
+			}
+
+			nouveau_dp_train(&priv->base, priv->pior.dp,
+					 &outp, head, datarate);
 		}
 	}
-
-	nv_wr32(priv, 0x610024, 0x00000040);
-	nv_wr32(priv, 0x610030, 0x80000000);
 }
 
-static void
-nv50_disp_intr_super(struct nv50_disp_priv *priv, u32 intr1)
+void
+nv50_disp_intr_supervisor(struct work_struct *work)
 {
+	struct nv50_disp_priv *priv =
+		container_of(work, struct nv50_disp_priv, supervisor);
 	u32 super = nv_rd32(priv, 0x610030);
+	int head;
 
-	nv_debug(priv, "supervisor 0x%08x 0x%08x\n", intr1, super);
+	nv_debug(priv, "supervisor 0x%08x 0x%08x\n", priv->super, super);
 
-	if (intr1 & 0x00000010)
-		nv50_disp_intr_unk10(priv, super);
-	if (intr1 & 0x00000020)
-		nv50_disp_intr_unk20(priv, super);
-	if (intr1 & 0x00000040)
-		nv50_disp_intr_unk40(priv, super);
+	if (priv->super & 0x00000010) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(super & (0x00000020 << head)))
+				continue;
+			if (!(super & (0x00000080 << head)))
+				continue;
+			nv50_disp_intr_unk10_0(priv, head);
+		}
+	} else
+	if (priv->super & 0x00000020) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(super & (0x00000080 << head)))
+				continue;
+			nv50_disp_intr_unk20_0(priv, head);
+		}
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(super & (0x00000200 << head)))
+				continue;
+			nv50_disp_intr_unk20_1(priv, head);
+		}
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(super & (0x00000080 << head)))
+				continue;
+			nv50_disp_intr_unk20_2(priv, head);
+		}
+	} else
+	if (priv->super & 0x00000040) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(super & (0x00000080 << head)))
+				continue;
+			nv50_disp_intr_unk40_0(priv, head);
+		}
+	}
+
+	nv_wr32(priv, 0x610030, 0x80000000);
 }
 
 void
@@ -1201,19 +1261,21 @@ nv50_disp_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (intr1 & 0x00000004) {
-		nv50_disp_intr_vblank(priv, 0);
+		nouveau_event_trigger(priv->base.vblank, 0);
 		nv_wr32(priv, 0x610024, 0x00000004);
 		intr1 &= ~0x00000004;
 	}
 
 	if (intr1 & 0x00000008) {
-		nv50_disp_intr_vblank(priv, 1);
+		nouveau_event_trigger(priv->base.vblank, 1);
 		nv_wr32(priv, 0x610024, 0x00000008);
 		intr1 &= ~0x00000008;
 	}
 
 	if (intr1 & 0x00000070) {
-		nv50_disp_intr_super(priv, intr1);
+		priv->super = (intr1 & 0x00000070);
+		schedule_work(&priv->supervisor);
+		nv_wr32(priv, 0x610024, priv->super);
 		intr1 &= ~0x00000070;
 	}
 }
@@ -1226,7 +1288,7 @@ nv50_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "PDISP",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -1235,16 +1297,17 @@ nv50_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nv50_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nv50_disp_intr;
+	INIT_WORK(&priv->supervisor, nv50_disp_intr_supervisor);
 	priv->sclass = nv50_disp_sclass;
 	priv->head.nr = 2;
 	priv->dac.nr = 3;
 	priv->sor.nr = 2;
+	priv->pior.nr = 3;
 	priv->dac.power = nv50_dac_power;
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->pior.power = nv50_pior_power;
+	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
index a6bb931450f1..1ae6ceb56704 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
@@ -3,16 +3,22 @@
 
 #include <core/parent.h>
 #include <core/namedb.h>
+#include <core/engctx.h>
 #include <core/ramht.h>
+#include <core/event.h>
 
 #include <engine/dmaobj.h>
 #include <engine/disp.h>
 
-struct dcb_output;
+#include "dport.h"
 
 struct nv50_disp_priv {
 	struct nouveau_disp base;
 	struct nouveau_oclass *sclass;
+
+	struct work_struct supervisor;
+	u32 super;
+
 	struct {
 		int nr;
 	} head;
@@ -26,23 +32,15 @@ struct nv50_disp_priv {
 		int (*power)(struct nv50_disp_priv *, int sor, u32 data);
 		int (*hda_eld)(struct nv50_disp_priv *, int sor, u8 *, u32);
 		int (*hdmi)(struct nv50_disp_priv *, int head, int sor, u32);
-		int (*dp_train_init)(struct nv50_disp_priv *, int sor, int link,
-				     int head, u16 type, u16 mask, u32 data,
-				     struct dcb_output *);
-		int (*dp_train_fini)(struct nv50_disp_priv *, int sor, int link,
-				     int head, u16 type, u16 mask, u32 data,
-				     struct dcb_output *);
-		int (*dp_train)(struct nv50_disp_priv *, int sor, int link,
-				u16 type, u16 mask, u32 data,
-				struct dcb_output *);
-		int (*dp_lnkctl)(struct nv50_disp_priv *, int sor, int link,
-				 int head, u16 type, u16 mask, u32 data,
-				 struct dcb_output *);
-		int (*dp_drvctl)(struct nv50_disp_priv *, int sor, int link,
-				 int lane, u16 type, u16 mask, u32 data,
-				 struct dcb_output *);
 		u32 lvdsconf;
+		const struct nouveau_dp_func *dp;
 	} sor;
+	struct {
+		int nr;
+		int (*power)(struct nv50_disp_priv *, int ext, u32 data);
+		u8 type[3];
+		const struct nouveau_dp_func *dp;
+	} pior;
 };
 
 #define DAC_MTHD(n) (n), (n) + 0x03
@@ -81,6 +79,11 @@ int nvd0_sor_dp_lnkctl(struct nv50_disp_priv *, int, int, int, u16, u16, u32,
 int nvd0_sor_dp_drvctl(struct nv50_disp_priv *, int, int, int, u16, u16, u32,
 		       struct dcb_output *);
 
+#define PIOR_MTHD(n) (n), (n) + 0x03
+
+int nv50_pior_mthd(struct nouveau_object *, u32, void *, u32);
+int nv50_pior_power(struct nv50_disp_priv *, int, u32);
+
 struct nv50_disp_base {
 	struct nouveau_parent base;
 	struct nouveau_ramht *ramht;
@@ -124,6 +127,7 @@ extern struct nouveau_ofuncs nv50_disp_oimm_ofuncs;
 extern struct nouveau_ofuncs nv50_disp_curs_ofuncs;
 extern struct nouveau_ofuncs nv50_disp_base_ofuncs;
 extern struct nouveau_oclass nv50_disp_cclass;
+void nv50_disp_intr_supervisor(struct work_struct *);
 void nv50_disp_intr(struct nouveau_subdev *);
 
 extern struct nouveau_omthds nv84_disp_base_omthds[];
@@ -137,6 +141,7 @@ extern struct nouveau_ofuncs nvd0_disp_oimm_ofuncs;
 extern struct nouveau_ofuncs nvd0_disp_curs_ofuncs;
 extern struct nouveau_ofuncs nvd0_disp_base_ofuncs;
 extern struct nouveau_oclass nvd0_disp_cclass;
+void nvd0_disp_intr_supervisor(struct work_struct *);
 void nvd0_disp_intr(struct nouveau_subdev *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
index fc84eacdfbec..d8c74c0883a1 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
@@ -46,6 +46,9 @@ nv84_disp_base_omthds[] = {
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_TMDS_PWR)  , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_DP_PWR)    , nv50_pior_mthd },
 	{},
 };
 
@@ -63,7 +66,7 @@ nv84_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "PDISP",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -72,17 +75,18 @@ nv84_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nv84_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nv50_disp_intr;
+	INIT_WORK(&priv->supervisor, nv50_disp_intr_supervisor);
 	priv->sclass = nv84_disp_sclass;
 	priv->head.nr = 2;
 	priv->dac.nr = 3;
 	priv->sor.nr = 2;
+	priv->pior.nr = 3;
 	priv->dac.power = nv50_dac_power;
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->pior.power = nv50_pior_power;
+	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
index ba9dfd4669a2..a66f949c1f84 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
@@ -44,14 +44,11 @@ nv94_disp_base_omthds[] = {
 	{ SOR_MTHD(NV50_DISP_SOR_PWR)         , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_TRAIN)    , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_LNKCTL)   , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(0)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(1)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(2)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(3)), nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_TMDS_PWR)  , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_DP_PWR)    , nv50_pior_mthd },
 	{},
 };
 
@@ -69,7 +66,7 @@ nv94_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "PDISP",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -78,22 +75,19 @@ nv94_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nv94_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nv50_disp_intr;
+	INIT_WORK(&priv->supervisor, nv50_disp_intr_supervisor);
 	priv->sclass = nv94_disp_sclass;
 	priv->head.nr = 2;
 	priv->dac.nr = 3;
 	priv->sor.nr = 4;
+	priv->pior.nr = 3;
 	priv->dac.power = nv50_dac_power;
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
-	priv->sor.dp_train = nv94_sor_dp_train;
-	priv->sor.dp_train_init = nv94_sor_dp_train_init;
-	priv->sor.dp_train_fini = nv94_sor_dp_train_fini;
-	priv->sor.dp_lnkctl = nv94_sor_dp_lnkctl;
-	priv->sor.dp_drvctl = nv94_sor_dp_drvctl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->sor.dp = &nv94_sor_dp_func;
+	priv->pior.power = nv50_pior_power;
+	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
index 5d63902cdeda..6cf8eefac368 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
@@ -53,7 +53,7 @@ nva0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "PDISP",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -62,17 +62,18 @@ nva0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nva0_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nv50_disp_intr;
+	INIT_WORK(&priv->supervisor, nv50_disp_intr_supervisor);
 	priv->sclass = nva0_disp_sclass;
 	priv->head.nr = 2;
 	priv->dac.nr = 3;
 	priv->sor.nr = 2;
+	priv->pior.nr = 3;
 	priv->dac.power = nv50_dac_power;
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->pior.power = nv50_pior_power;
+	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
index e9192ca389fa..b75413169eae 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
@@ -45,14 +45,11 @@ nva3_disp_base_omthds[] = {
 	{ SOR_MTHD(NVA3_DISP_SOR_HDA_ELD)     , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_TRAIN)    , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_LNKCTL)   , nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(0)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(1)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(2)), nv50_sor_mthd },
-	{ SOR_MTHD(NV94_DISP_SOR_DP_DRVCTL(3)), nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_TMDS_PWR)  , nv50_pior_mthd },
+	{ PIOR_MTHD(NV50_DISP_PIOR_DP_PWR)    , nv50_pior_mthd },
 	{},
 };
 
@@ -70,7 +67,7 @@ nva3_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_disp_priv *priv;
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
+	ret = nouveau_disp_create(parent, engine, oclass, 2, "PDISP",
 				  "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
@@ -79,23 +76,20 @@ nva3_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nva3_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nv50_disp_intr;
+	INIT_WORK(&priv->supervisor, nv50_disp_intr_supervisor);
 	priv->sclass = nva3_disp_sclass;
 	priv->head.nr = 2;
 	priv->dac.nr = 3;
 	priv->sor.nr = 4;
+	priv->pior.nr = 3;
 	priv->dac.power = nv50_dac_power;
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nva3_hda_eld;
 	priv->sor.hdmi = nva3_hdmi_ctrl;
-	priv->sor.dp_train = nv94_sor_dp_train;
-	priv->sor.dp_train_init = nv94_sor_dp_train_init;
-	priv->sor.dp_train_fini = nv94_sor_dp_train_fini;
-	priv->sor.dp_lnkctl = nv94_sor_dp_lnkctl;
-	priv->sor.dp_drvctl = nv94_sor_dp_drvctl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->sor.dp = &nv94_sor_dp_func;
+	priv->pior.power = nv50_pior_power;
+	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
index 9e38ebff5fb3..788dd34ccb54 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
@@ -27,12 +27,10 @@
 #include <core/handle.h>
 #include <core/class.h>
 
-#include <engine/software.h>
 #include <engine/disp.h>
 
 #include <subdev/timer.h>
 #include <subdev/fb.h>
-#include <subdev/bar.h>
 #include <subdev/clock.h>
 
 #include <subdev/bios.h>
@@ -230,7 +228,7 @@ nvd0_disp_sync_ctor(struct nouveau_object *parent,
 	struct nv50_disp_dmac *dmac;
 	int ret;
 
-	if (size < sizeof(*data) || args->head >= priv->head.nr)
+	if (size < sizeof(*args) || args->head >= priv->head.nr)
 		return -EINVAL;
 
 	ret = nv50_disp_dmac_create_(parent, engine, oclass, args->pushbuf,
@@ -270,7 +268,7 @@ nvd0_disp_ovly_ctor(struct nouveau_object *parent,
 	struct nv50_disp_dmac *dmac;
 	int ret;
 
-	if (size < sizeof(*data) || args->head >= priv->head.nr)
+	if (size < sizeof(*args) || args->head >= priv->head.nr)
 		return -EINVAL;
 
 	ret = nv50_disp_dmac_create_(parent, engine, oclass, args->pushbuf,
@@ -443,6 +441,18 @@ nvd0_disp_curs_ofuncs = {
  * Base display object
  ******************************************************************************/
 
+static void
+nvd0_disp_base_vblank_enable(struct nouveau_event *event, int head)
+{
+	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000001);
+}
+
+static void
+nvd0_disp_base_vblank_disable(struct nouveau_event *event, int head)
+{
+	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000000);
+}
+
 static int
 nvd0_disp_base_ctor(struct nouveau_object *parent,
 		    struct nouveau_object *engine,
@@ -459,6 +469,10 @@ nvd0_disp_base_ctor(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
+	priv->base.vblank->priv = priv;
+	priv->base.vblank->enable = nvd0_disp_base_vblank_enable;
+	priv->base.vblank->disable = nvd0_disp_base_vblank_disable;
+
 	return nouveau_ramht_new(parent, parent, 0x1000, 0, &base->ramht);
 }
 
@@ -609,13 +623,24 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 }
 
 static bool
-exec_script(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl, int id)
+exec_script(struct nv50_disp_priv *priv, int head, int id)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
 	struct nvbios_outp info;
 	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
+	u32 ctrl = 0x00000000;
 	u16 data;
+	int outp;
+
+	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
+		ctrl = nv_rd32(priv, 0x640180 + (outp * 0x20));
+		if (ctrl & (1 << head))
+			break;
+	}
+
+	if (outp == 8)
+		return false;
 
 	data = exec_lookup(priv, head, outp, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info);
 	if (data) {
@@ -635,21 +660,31 @@ exec_script(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl, int id)
 }
 
 static u32
-exec_clkcmp(struct nv50_disp_priv *priv, int head, int outp,
-	    u32 ctrl, int id, u32 pclk)
+exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
+	    u32 pclk, struct dcb_output *dcb)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
-	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
-	u16 data, conf;
+	u32 ctrl = 0x00000000;
+	u32 data, conf = ~0;
+	int outp;
 
-	data = exec_lookup(priv, head, outp, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info1);
-	if (data == 0x0000)
+	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
+		ctrl = nv_rd32(priv, 0x660180 + (outp * 0x20));
+		if (ctrl & (1 << head))
+			break;
+	}
+
+	if (outp == 8)
 		return false;
 
-	switch (dcb.type) {
+	data = exec_lookup(priv, head, outp, ctrl, dcb, &ver, &hdr, &cnt, &len, &info1);
+	if (data == 0x0000)
+		return conf;
+
+	switch (dcb->type) {
 	case DCB_OUTPUT_TMDS:
 		conf = (ctrl & 0x00000f00) >> 8;
 		if (pclk >= 165000)
@@ -668,46 +703,52 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int outp,
 	}
 
 	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
-	if (data) {
+	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
 			struct nvbios_init init = {
 				.subdev = nv_subdev(priv),
 				.bios = bios,
 				.offset = data,
-				.outp = &dcb,
+				.outp = dcb,
 				.crtc = head,
 				.execute = 1,
 			};
 
-			if (nvbios_exec(&init))
-				return 0x0000;
-			return conf;
+			nvbios_exec(&init);
 		}
 	}
 
-	return 0x0000;
+	return conf;
 }
 
 static void
-nvd0_display_unk1_handler(struct nv50_disp_priv *priv, u32 head, u32 mask)
+nvd0_disp_intr_unk1_0(struct nv50_disp_priv *priv, int head)
 {
-	int i;
+	exec_script(priv, head, 1);
+}
 
-	for (i = 0; mask && i < 8; i++) {
-		u32 mcc = nv_rd32(priv, 0x640180 + (i * 0x20));
-		if (mcc & (1 << head))
-			exec_script(priv, head, i, mcc, 1);
-	}
+static void
+nvd0_disp_intr_unk2_0(struct nv50_disp_priv *priv, int head)
+{
+	exec_script(priv, head, 2);
+}
 
-	nv_wr32(priv, 0x6101d4, 0x00000000);
-	nv_wr32(priv, 0x6109d4, 0x00000000);
-	nv_wr32(priv, 0x6101d0, 0x80000000);
+static void
+nvd0_disp_intr_unk2_1(struct nv50_disp_priv *priv, int head)
+{
+	struct nouveau_clock *clk = nouveau_clock(priv);
+	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
+	if (pclk)
+		clk->pll_set(clk, PLL_VPLL0 + head, pclk);
+	nv_wr32(priv, 0x612200 + (head * 0x800), 0x00000000);
 }
 
 static void
-nvd0_display_unk2_calc_tu(struct nv50_disp_priv *priv, int head, int or)
+nvd0_disp_intr_unk2_2_tu(struct nv50_disp_priv *priv, int head,
+			 struct dcb_output *outp)
 {
+	const int or = ffs(outp->or) - 1;
 	const u32 ctrl = nv_rd32(priv, 0x660200 + (or   * 0x020));
 	const u32 conf = nv_rd32(priv, 0x660404 + (head * 0x300));
 	const u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
@@ -750,105 +791,102 @@ nvd0_display_unk2_calc_tu(struct nv50_disp_priv *priv, int head, int or)
 }
 
 static void
-nvd0_display_unk2_handler(struct nv50_disp_priv *priv, u32 head, u32 mask)
+nvd0_disp_intr_unk2_2(struct nv50_disp_priv *priv, int head)
 {
-	u32 pclk;
-	int i;
-
-	for (i = 0; mask && i < 8; i++) {
-		u32 mcc = nv_rd32(priv, 0x640180 + (i * 0x20));
-		if (mcc & (1 << head))
-			exec_script(priv, head, i, mcc, 2);
-	}
+	struct dcb_output outp;
+	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
+	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
+	if (conf != ~0) {
+		u32 addr, data;
+
+		if (outp.type == DCB_OUTPUT_DP) {
+			u32 sync = nv_rd32(priv, 0x660404 + (head * 0x300));
+			switch ((sync & 0x000003c0) >> 6) {
+			case 6: pclk = pclk * 30 / 8; break;
+			case 5: pclk = pclk * 24 / 8; break;
+			case 2:
+			default:
+				pclk = pclk * 18 / 8;
+				break;
+			}
 
-	pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
-	nv_debug(priv, "head %d pclk %d mask 0x%08x\n", head, pclk, mask);
-	if (pclk && (mask & 0x00010000)) {
-		struct nouveau_clock *clk = nouveau_clock(priv);
-		clk->pll_set(clk, PLL_VPLL0 + head, pclk);
-	}
+			nouveau_dp_train(&priv->base, priv->sor.dp,
+					 &outp, head, pclk);
+		}
 
-	nv_wr32(priv, 0x612200 + (head * 0x800), 0x00000000);
+		exec_clkcmp(priv, head, 0, pclk, &outp);
 
-	for (i = 0; mask && i < 8; i++) {
-		u32 mcp = nv_rd32(priv, 0x660180 + (i * 0x20)), cfg;
-		if (mcp & (1 << head)) {
-			if ((cfg = exec_clkcmp(priv, head, i, mcp, 0, pclk))) {
-				u32 addr, mask, data = 0x00000000;
-				if (i < 4) {
-					addr = 0x612280 + ((i - 0) * 0x800);
-					mask = 0xffffffff;
-				} else {
-					switch (mcp & 0x00000f00) {
-					case 0x00000800:
-					case 0x00000900:
-						nvd0_display_unk2_calc_tu(priv, head, i - 4);
-						break;
-					default:
-						break;
-					}
-
-					addr = 0x612300 + ((i - 4) * 0x800);
-					mask = 0x00000707;
-					if (cfg & 0x00000100)
-						data = 0x00000101;
-				}
-				nv_mask(priv, addr, mask, data);
-			}
-			break;
+		if (outp.type == DCB_OUTPUT_ANALOG) {
+			addr = 0x612280 + (ffs(outp.or) - 1) * 0x800;
+			data = 0x00000000;
+		} else {
+			if (outp.type == DCB_OUTPUT_DP)
+				nvd0_disp_intr_unk2_2_tu(priv, head, &outp);
+			addr = 0x612300 + (ffs(outp.or) - 1) * 0x800;
+			data = (conf & 0x0100) ? 0x00000101 : 0x00000000;
 		}
-	}
 
-	nv_wr32(priv, 0x6101d4, 0x00000000);
-	nv_wr32(priv, 0x6109d4, 0x00000000);
-	nv_wr32(priv, 0x6101d0, 0x80000000);
+		nv_mask(priv, addr, 0x00000707, data);
+	}
 }
 
 static void
-nvd0_display_unk4_handler(struct nv50_disp_priv *priv, u32 head, u32 mask)
+nvd0_disp_intr_unk4_0(struct nv50_disp_priv *priv, int head)
 {
-	int pclk, i;
-
-	pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
+	struct dcb_output outp;
+	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
+	exec_clkcmp(priv, head, 1, pclk, &outp);
+}
 
-	for (i = 0; mask && i < 8; i++) {
-		u32 mcp = nv_rd32(priv, 0x660180 + (i * 0x20));
-		if (mcp & (1 << head))
-			exec_clkcmp(priv, head, i, mcp, 1, pclk);
+void
+nvd0_disp_intr_supervisor(struct work_struct *work)
+{
+	struct nv50_disp_priv *priv =
+		container_of(work, struct nv50_disp_priv, supervisor);
+	u32 mask[4];
+	int head;
+
+	nv_debug(priv, "supervisor %08x\n", priv->super);
+	for (head = 0; head < priv->head.nr; head++) {
+		mask[head] = nv_rd32(priv, 0x6101d4 + (head * 0x800));
+		nv_debug(priv, "head %d: 0x%08x\n", head, mask[head]);
 	}
 
-	nv_wr32(priv, 0x6101d4, 0x00000000);
-	nv_wr32(priv, 0x6109d4, 0x00000000);
-	nv_wr32(priv, 0x6101d0, 0x80000000);
-}
-
-static void
-nvd0_disp_intr_vblank(struct nv50_disp_priv *priv, int crtc)
-{
-	struct nouveau_bar *bar = nouveau_bar(priv);
-	struct nouveau_disp *disp = &priv->base;
-	struct nouveau_software_chan *chan, *temp;
-	unsigned long flags;
-
-	spin_lock_irqsave(&disp->vblank.lock, flags);
-	list_for_each_entry_safe(chan, temp, &disp->vblank.list, vblank.head) {
-		if (chan->vblank.crtc != crtc)
-			continue;
-
-		nv_wr32(priv, 0x001718, 0x80000000 | chan->vblank.channel);
-		bar->flush(bar);
-		nv_wr32(priv, 0x06000c, upper_32_bits(chan->vblank.offset));
-		nv_wr32(priv, 0x060010, lower_32_bits(chan->vblank.offset));
-		nv_wr32(priv, 0x060014, chan->vblank.value);
-
-		list_del(&chan->vblank.head);
-		if (disp->vblank.put)
-			disp->vblank.put(disp->vblank.data, crtc);
+	if (priv->super & 0x00000001) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(mask[head] & 0x00001000))
+				continue;
+			nvd0_disp_intr_unk1_0(priv, head);
+		}
+	} else
+	if (priv->super & 0x00000002) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(mask[head] & 0x00001000))
+				continue;
+			nvd0_disp_intr_unk2_0(priv, head);
+		}
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(mask[head] & 0x00010000))
+				continue;
+			nvd0_disp_intr_unk2_1(priv, head);
+		}
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(mask[head] & 0x00001000))
+				continue;
+			nvd0_disp_intr_unk2_2(priv, head);
+		}
+	} else
+	if (priv->super & 0x00000004) {
+		for (head = 0; head < priv->head.nr; head++) {
+			if (!(mask[head] & 0x00001000))
+				continue;
+			nvd0_disp_intr_unk4_0(priv, head);
+		}
 	}
-	spin_unlock_irqrestore(&disp->vblank.lock, flags);
 
-	if (disp->vblank.notify)
-		disp->vblank.notify(disp->vblank.data, crtc);
+	for (head = 0; head < priv->head.nr; head++)
+		nv_wr32(priv, 0x6101d4 + (head * 0x800), 0x00000000);
+	nv_wr32(priv, 0x6101d0, 0x80000000);
 }
 
 void
@@ -884,27 +922,11 @@ nvd0_disp_intr(struct nouveau_subdev *subdev)
 
 	if (intr & 0x00100000) {
 		u32 stat = nv_rd32(priv, 0x6100ac);
-		u32 mask = 0, crtc = ~0;
-
-		while (!mask && ++crtc < priv->head.nr)
-			mask = nv_rd32(priv, 0x6101d4 + (crtc * 0x800));
-
-		if (stat & 0x00000001) {
-			nv_wr32(priv, 0x6100ac, 0x00000001);
-			nvd0_display_unk1_handler(priv, crtc, mask);
-			stat &= ~0x00000001;
-		}
-
-		if (stat & 0x00000002) {
-			nv_wr32(priv, 0x6100ac, 0x00000002);
-			nvd0_display_unk2_handler(priv, crtc, mask);
-			stat &= ~0x00000002;
-		}
-
-		if (stat & 0x00000004) {
-			nv_wr32(priv, 0x6100ac, 0x00000004);
-			nvd0_display_unk4_handler(priv, crtc, mask);
-			stat &= ~0x00000004;
+		if (stat & 0x00000007) {
+			priv->super = (stat & 0x00000007);
+			schedule_work(&priv->supervisor);
+			nv_wr32(priv, 0x6100ac, priv->super);
+			stat &= ~0x00000007;
 		}
 
 		if (stat) {
@@ -920,7 +942,7 @@ nvd0_disp_intr(struct nouveau_subdev *subdev)
 		if (mask & intr) {
 			u32 stat = nv_rd32(priv, 0x6100bc + (i * 0x800));
 			if (stat & 0x00000001)
-				nvd0_disp_intr_vblank(priv, i);
+				nouveau_event_trigger(priv->base.vblank, i);
 			nv_mask(priv, 0x6100bc + (i * 0x800), 0, 0);
 			nv_rd32(priv, 0x6100c0 + (i * 0x800));
 		}
@@ -933,10 +955,11 @@ nvd0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_object **pobject)
 {
 	struct nv50_disp_priv *priv;
+	int heads = nv_rd32(parent, 0x022448);
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
-				  "display", &priv);
+	ret = nouveau_disp_create(parent, engine, oclass, heads,
+				  "PDISP", "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -944,8 +967,9 @@ nvd0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nvd0_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nvd0_disp_intr;
+	INIT_WORK(&priv->supervisor, nvd0_disp_intr_supervisor);
 	priv->sclass = nvd0_disp_sclass;
-	priv->head.nr = nv_rd32(priv, 0x022448);
+	priv->head.nr = heads;
 	priv->dac.nr = 3;
 	priv->sor.nr = 4;
 	priv->dac.power = nv50_dac_power;
@@ -953,14 +977,7 @@ nvd0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp_train = nvd0_sor_dp_train;
-	priv->sor.dp_train_init = nv94_sor_dp_train_init;
-	priv->sor.dp_train_fini = nv94_sor_dp_train_fini;
-	priv->sor.dp_lnkctl = nvd0_sor_dp_lnkctl;
-	priv->sor.dp_drvctl = nvd0_sor_dp_drvctl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
index 259537c4587e..20725b363d58 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
@@ -51,10 +51,11 @@ nve0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_object **pobject)
 {
 	struct nv50_disp_priv *priv;
+	int heads = nv_rd32(parent, 0x022448);
 	int ret;
 
-	ret = nouveau_disp_create(parent, engine, oclass, "PDISP",
-				  "display", &priv);
+	ret = nouveau_disp_create(parent, engine, oclass, heads,
+				  "PDISP", "display", &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -62,8 +63,9 @@ nve0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_engine(priv)->sclass = nve0_disp_base_oclass;
 	nv_engine(priv)->cclass = &nv50_disp_cclass;
 	nv_subdev(priv)->intr = nvd0_disp_intr;
+	INIT_WORK(&priv->supervisor, nvd0_disp_intr_supervisor);
 	priv->sclass = nve0_disp_sclass;
-	priv->head.nr = nv_rd32(priv, 0x022448);
+	priv->head.nr = heads;
 	priv->dac.nr = 3;
 	priv->sor.nr = 4;
 	priv->dac.power = nv50_dac_power;
@@ -71,14 +73,7 @@ nve0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp_train = nvd0_sor_dp_train;
-	priv->sor.dp_train_init = nv94_sor_dp_train_init;
-	priv->sor.dp_train_fini = nv94_sor_dp_train_fini;
-	priv->sor.dp_lnkctl = nvd0_sor_dp_lnkctl;
-	priv->sor.dp_drvctl = nvd0_sor_dp_drvctl;
-
-	INIT_LIST_HEAD(&priv->base.vblank.list);
-	spin_lock_init(&priv->base.vblank.lock);
+	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
new file mode 100644
index 000000000000..2c8ce351b52d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
@@ -0,0 +1,140 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <core/os.h>
+#include <core/class.h>
+
+#include <subdev/bios.h>
+#include <subdev/bios/dcb.h>
+#include <subdev/timer.h>
+#include <subdev/i2c.h>
+
+#include "nv50.h"
+
+/******************************************************************************
+ * DisplayPort
+ *****************************************************************************/
+static struct nouveau_i2c_port *
+nv50_pior_dp_find(struct nouveau_disp *disp, struct dcb_output *outp)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(disp);
+	return i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
+}
+
+static int
+nv50_pior_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
+		     int head, int pattern)
+{
+	struct nouveau_i2c_port *port;
+	int ret = -EINVAL;
+
+	port = nv50_pior_dp_find(disp, outp);
+	if (port) {
+		if (port->func->pattern)
+			ret = port->func->pattern(port, pattern);
+		else
+			ret = 0;
+	}
+
+	return ret;
+}
+
+static int
+nv50_pior_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		     int head, int lane_nr, int link_bw, bool enh)
+{
+	struct nouveau_i2c_port *port;
+	int ret = -EINVAL;
+
+	port = nv50_pior_dp_find(disp, outp);
+	if (port && port->func->lnk_ctl)
+		ret = port->func->lnk_ctl(port, lane_nr, link_bw, enh);
+
+	return ret;
+}
+
+static int
+nv50_pior_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		     int head, int lane, int vsw, int pre)
+{
+	struct nouveau_i2c_port *port;
+	int ret = -EINVAL;
+
+	port = nv50_pior_dp_find(disp, outp);
+	if (port) {
+		if (port->func->drv_ctl)
+			ret = port->func->drv_ctl(port, lane, vsw, pre);
+		else
+			ret = 0;
+	}
+
+	return ret;
+}
+
+const struct nouveau_dp_func
+nv50_pior_dp_func = {
+	.pattern = nv50_pior_dp_pattern,
+	.lnk_ctl = nv50_pior_dp_lnk_ctl,
+	.drv_ctl = nv50_pior_dp_drv_ctl,
+};
+
+/******************************************************************************
+ * General PIOR handling
+ *****************************************************************************/
+int
+nv50_pior_power(struct nv50_disp_priv *priv, int or, u32 data)
+{
+	const u32 stat = data & NV50_DISP_PIOR_PWR_STATE;
+	const u32 soff = (or * 0x800);
+	nv_wait(priv, 0x61e004 + soff, 0x80000000, 0x00000000);
+	nv_mask(priv, 0x61e004 + soff, 0x80000101, 0x80000000 | stat);
+	nv_wait(priv, 0x61e004 + soff, 0x80000000, 0x00000000);
+	return 0;
+}
+
+int
+nv50_pior_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
+{
+	struct nv50_disp_priv *priv = (void *)object->engine;
+	const u8 type = (mthd & NV50_DISP_PIOR_MTHD_TYPE) >> 12;
+	const u8 or   = (mthd & NV50_DISP_PIOR_MTHD_OR);
+	u32 *data = args;
+	int ret;
+
+	if (size < sizeof(u32))
+		return -EINVAL;
+
+	mthd &= ~NV50_DISP_PIOR_MTHD_TYPE;
+	mthd &= ~NV50_DISP_PIOR_MTHD_OR;
+	switch (mthd) {
+	case NV50_DISP_PIOR_PWR:
+		ret = priv->pior.power(priv, or, data[0]);
+		priv->pior.type[or] = type;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
index 39b6b67732d0..ab1e918469a8 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
@@ -79,31 +79,6 @@ nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 		priv->sor.lvdsconf = data & NV50_DISP_SOR_LVDS_SCRIPT_ID;
 		ret = 0;
 		break;
-	case NV94_DISP_SOR_DP_TRAIN:
-		switch (data & NV94_DISP_SOR_DP_TRAIN_OP) {
-		case NV94_DISP_SOR_DP_TRAIN_OP_PATTERN:
-			ret = priv->sor.dp_train(priv, or, link, type, mask, data, &outp);
-			break;
-		case NV94_DISP_SOR_DP_TRAIN_OP_INIT:
-			ret = priv->sor.dp_train_init(priv, or, link, head, type, mask, data, &outp);
-			break;
-		case NV94_DISP_SOR_DP_TRAIN_OP_FINI:
-			ret = priv->sor.dp_train_fini(priv, or, link, head, type, mask, data, &outp);
-			break;
-		default:
-			break;
-		}
-		break;
-	case NV94_DISP_SOR_DP_LNKCTL:
-		ret = priv->sor.dp_lnkctl(priv, or, link, head, type, mask, data, &outp);
-		break;
-	case NV94_DISP_SOR_DP_DRVCTL(0):
-	case NV94_DISP_SOR_DP_DRVCTL(1):
-	case NV94_DISP_SOR_DP_DRVCTL(2):
-	case NV94_DISP_SOR_DP_DRVCTL(3):
-		ret = priv->sor.dp_drvctl(priv, or, link, (mthd & 0xc0) >> 6,
-				          type, mask, data, &outp);
-		break;
 	default:
 		BUG_ON(1);
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
index f6edd009762e..7ec4ee83fb64 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
@@ -33,124 +33,53 @@
 #include "nv50.h"
 
 static inline u32
-nv94_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
+nv94_sor_soff(struct dcb_output *outp)
 {
-	static const u8 nvaf[] = { 24, 16, 8, 0 }; /* thanks, apple.. */
-	static const u8 nv94[] = { 16, 8, 0, 24 };
-	if (nv_device(priv)->chipset == 0xaf)
-		return nvaf[lane];
-	return nv94[lane];
+	return (ffs(outp->or) - 1) * 0x800;
 }
 
-int
-nv94_sor_dp_train_init(struct nv50_disp_priv *priv, int or, int link, int head,
-		       u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static inline u32
+nv94_sor_loff(struct dcb_output *outp)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	struct nvbios_dpout info;
-	u8  ver, hdr, cnt, len;
-	u16 outp;
-
-	outp = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &info);
-	if (outp) {
-		struct nvbios_init init = {
-			.subdev = nv_subdev(priv),
-			.bios = bios,
-			.outp = dcbo,
-			.crtc = head,
-			.execute = 1,
-		};
-
-		if (data & NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD_ON)
-			init.offset = info.script[2];
-		else
-			init.offset = info.script[3];
-		nvbios_exec(&init);
-
-		init.offset = info.script[0];
-		nvbios_exec(&init);
-	}
-
-	return 0;
+	return nv94_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
 }
 
-int
-nv94_sor_dp_train_fini(struct nv50_disp_priv *priv, int or, int link, int head,
-		       u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static inline u32
+nv94_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	struct nvbios_dpout info;
-	u8  ver, hdr, cnt, len;
-	u16 outp;
-
-	outp = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &info);
-	if (outp) {
-		struct nvbios_init init = {
-			.subdev = nv_subdev(priv),
-			.bios = bios,
-			.offset = info.script[1],
-			.outp = dcbo,
-			.crtc = head,
-			.execute = 1,
-		};
-
-		nvbios_exec(&init);
-	}
-
-	return 0;
+	static const u8 nvaf[] = { 24, 16, 8, 0 }; /* thanks, apple.. */
+	static const u8 nv94[] = { 16, 8, 0, 24 };
+	if (nv_device(priv)->chipset == 0xaf)
+		return nvaf[lane];
+	return nv94[lane];
 }
 
-int
-nv94_sor_dp_train(struct nv50_disp_priv *priv, int or, int link,
-		  u16 type, u16 mask, u32 data, struct dcb_output *info)
+static int
+nv94_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int pattern)
 {
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u32 patt = (data & NV94_DISP_SOR_DP_TRAIN_PATTERN);
-	nv_mask(priv, 0x61c10c + loff, 0x0f000000, patt << 24);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 loff = nv94_sor_loff(outp);
+	nv_mask(priv, 0x61c10c + loff, 0x0f000000, pattern << 24);
 	return 0;
 }
 
-int
-nv94_sor_dp_lnkctl(struct nv50_disp_priv *priv, int or, int link, int head,
-		   u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static int
+nv94_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int link_nr, int link_bw, bool enh_frame)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u32 soff = (or * 0x800);
-	u16 link_bw = (data & NV94_DISP_SOR_DP_LNKCTL_WIDTH) >> 8;
-	u8  link_nr = (data & NV94_DISP_SOR_DP_LNKCTL_COUNT);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 soff = nv94_sor_soff(outp);
+	const u32 loff = nv94_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 outp, lane = 0;
-	u8  ver, hdr, cnt, len;
-	struct nvbios_dpout info;
+	u32 lane = 0;
 	int i;
 
-	/* -> 10Khz units */
-	link_bw *= 2700;
-
-	outp = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &info);
-	if (outp && info.lnkcmp) {
-		struct nvbios_init init = {
-			.subdev = nv_subdev(priv),
-			.bios = bios,
-			.offset = 0x0000,
-			.outp = dcbo,
-			.crtc = head,
-			.execute = 1,
-		};
-
-		while (link_bw < nv_ro16(bios, info.lnkcmp))
-			info.lnkcmp += 4;
-		init.offset = nv_ro16(bios, info.lnkcmp + 2);
-
-		nvbios_exec(&init);
-	}
-
 	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (data & NV94_DISP_SOR_DP_LNKCTL_FRAME_ENH)
+	if (enh_frame)
 		dpctrl |= 0x00004000;
-	if (link_bw > 16200)
+	if (link_bw > 0x06)
 		clksor |= 0x00040000;
 
 	for (i = 0; i < link_nr; i++)
@@ -162,24 +91,25 @@ nv94_sor_dp_lnkctl(struct nv50_disp_priv *priv, int or, int link, int head,
 	return 0;
 }
 
-int
-nv94_sor_dp_drvctl(struct nv50_disp_priv *priv, int or, int link, int lane,
-		   u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static int
+nv94_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int lane, int swing, int preem)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u8 swing = (data & NV94_DISP_SOR_DP_DRVCTL_VS) >> 8;
-	const u8 preem = (data & NV94_DISP_SOR_DP_DRVCTL_PE);
+	struct nouveau_bios *bios = nouveau_bios(disp);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 loff = nv94_sor_loff(outp);
 	u32 addr, shift = nv94_sor_dp_lane_map(priv, lane);
 	u8  ver, hdr, cnt, len;
-	struct nvbios_dpout outp;
+	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &outp);
+	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem, &ver, &hdr, &cnt, &len, &ocfg);
+	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
@@ -188,3 +118,10 @@ nv94_sor_dp_drvctl(struct nv50_disp_priv *priv, int or, int link, int lane,
 	nv_mask(priv, 0x61c130 + loff, 0x0000ff00, ocfg.unk << 8);
 	return 0;
 }
+
+const struct nouveau_dp_func
+nv94_sor_dp_func = {
+	.pattern = nv94_sor_dp_pattern,
+	.lnk_ctl = nv94_sor_dp_lnk_ctl,
+	.drv_ctl = nv94_sor_dp_drv_ctl,
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
index c37ce7e29f5d..9e1d435d7282 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
@@ -33,59 +33,49 @@
 #include "nv50.h"
 
 static inline u32
+nvd0_sor_soff(struct dcb_output *outp)
+{
+	return (ffs(outp->or) - 1) * 0x800;
+}
+
+static inline u32
+nvd0_sor_loff(struct dcb_output *outp)
+{
+	return nvd0_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
+}
+
+static inline u32
 nvd0_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 {
 	static const u8 nvd0[] = { 16, 8, 0, 24 };
 	return nvd0[lane];
 }
 
-int
-nvd0_sor_dp_train(struct nv50_disp_priv *priv, int or, int link,
-		  u16 type, u16 mask, u32 data, struct dcb_output *info)
+static int
+nvd0_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int pattern)
 {
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u32 patt = (data & NV94_DISP_SOR_DP_TRAIN_PATTERN);
-	nv_mask(priv, 0x61c110 + loff, 0x0f0f0f0f, 0x01010101 * patt);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 loff = nvd0_sor_loff(outp);
+	nv_mask(priv, 0x61c110 + loff, 0x0f0f0f0f, 0x01010101 * pattern);
 	return 0;
 }
 
-int
-nvd0_sor_dp_lnkctl(struct nv50_disp_priv *priv, int or, int link, int head,
-		   u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static int
+nvd0_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int link_nr, int link_bw, bool enh_frame)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u32 soff = (or * 0x800);
-	const u8  link_bw = (data & NV94_DISP_SOR_DP_LNKCTL_WIDTH) >> 8;
-	const u8  link_nr = (data & NV94_DISP_SOR_DP_LNKCTL_COUNT);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 soff = nvd0_sor_soff(outp);
+	const u32 loff = nvd0_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 outp, lane = 0;
-	u8  ver, hdr, cnt, len;
-	struct nvbios_dpout info;
+	u32 lane = 0;
 	int i;
 
-	outp = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &info);
-	if (outp && info.lnkcmp) {
-		struct nvbios_init init = {
-			.subdev = nv_subdev(priv),
-			.bios = bios,
-			.offset = 0x0000,
-			.outp = dcbo,
-			.crtc = head,
-			.execute = 1,
-		};
-
-		while (nv_ro08(bios, info.lnkcmp) < link_bw)
-			info.lnkcmp += 3;
-		init.offset = nv_ro16(bios, info.lnkcmp + 1);
-
-		nvbios_exec(&init);
-	}
-
 	clksor |= link_bw << 18;
 	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (data & NV94_DISP_SOR_DP_LNKCTL_FRAME_ENH)
+	if (enh_frame)
 		dpctrl |= 0x00004000;
 
 	for (i = 0; i < link_nr; i++)
@@ -97,24 +87,25 @@ nvd0_sor_dp_lnkctl(struct nv50_disp_priv *priv, int or, int link, int head,
 	return 0;
 }
 
-int
-nvd0_sor_dp_drvctl(struct nv50_disp_priv *priv, int or, int link, int lane,
-		   u16 type, u16 mask, u32 data, struct dcb_output *dcbo)
+static int
+nvd0_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
+		    int head, int lane, int swing, int preem)
 {
-	struct nouveau_bios *bios = nouveau_bios(priv);
-	const u32 loff = (or * 0x800) + (link * 0x80);
-	const u8 swing = (data & NV94_DISP_SOR_DP_DRVCTL_VS) >> 8;
-	const u8 preem = (data & NV94_DISP_SOR_DP_DRVCTL_PE);
+	struct nouveau_bios *bios = nouveau_bios(disp);
+	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 loff = nvd0_sor_loff(outp);
 	u32 addr, shift = nvd0_sor_dp_lane_map(priv, lane);
 	u8  ver, hdr, cnt, len;
-	struct nvbios_dpout outp;
+	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, type, mask, &ver, &hdr, &cnt, &len, &outp);
+	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem, &ver, &hdr, &cnt, &len, &ocfg);
+	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
@@ -124,3 +115,10 @@ nvd0_sor_dp_drvctl(struct nv50_disp_priv *priv, int or, int link, int lane,
 	nv_mask(priv, 0x61c13c + loff, 0x00000000, 0x00000000);
 	return 0;
 }
+
+const struct nouveau_dp_func
+nvd0_sor_dp_func = {
+	.pattern = nvd0_sor_dp_pattern,
+	.lnk_ctl = nvd0_sor_dp_lnk_ctl,
+	.drv_ctl = nvd0_sor_dp_drv_ctl,
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
index c2b9db335816..7341ebe131fa 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
@@ -22,8 +22,10 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/object.h>
 #include <core/handle.h>
+#include <core/event.h>
 #include <core/class.h>
 
 #include <engine/dmaobj.h>
@@ -146,10 +148,25 @@ nouveau_fifo_chid(struct nouveau_fifo *priv, struct nouveau_object *object)
 	return -1;
 }
 
+const char *
+nouveau_client_name_for_fifo_chid(struct nouveau_fifo *fifo, u32 chid)
+{
+	struct nouveau_fifo_chan *chan = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&fifo->lock, flags);
+	if (chid >= fifo->min && chid <= fifo->max)
+		chan = (void *)fifo->channel[chid];
+	spin_unlock_irqrestore(&fifo->lock, flags);
+
+	return nouveau_client_name(chan);
+}
+
 void
 nouveau_fifo_destroy(struct nouveau_fifo *priv)
 {
 	kfree(priv->channel);
+	nouveau_event_destroy(&priv->uevent);
 	nouveau_engine_destroy(&priv->base);
 }
 
@@ -174,6 +191,10 @@ nouveau_fifo_create_(struct nouveau_object *parent,
 	if (!priv->channel)
 		return -ENOMEM;
 
+	ret = nouveau_event_create(1, &priv->uevent);
+	if (ret)
+		return ret;
+
 	priv->chid = nouveau_fifo_chid;
 	spin_lock_init(&priv->lock);
 	return 0;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
index a47a8548f9e0..f877bd524a92 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
@@ -28,6 +28,7 @@
 #include <core/namedb.h>
 #include <core/handle.h>
 #include <core/ramht.h>
+#include <core/event.h>
 
 #include <subdev/instmem.h>
 #include <subdev/instmem/nv04.h>
@@ -398,6 +399,98 @@ out:
 	return handled;
 }
 
+static void
+nv04_fifo_cache_error(struct nouveau_device *device,
+		struct nv04_fifo_priv *priv, u32 chid, u32 get)
+{
+	u32 mthd, data;
+	int ptr;
+
+	/* NV_PFIFO_CACHE1_GET actually goes to 0xffc before wrapping on my
+	 * G80 chips, but CACHE1 isn't big enough for this much data.. Tests
+	 * show that it wraps around to the start at GET=0x800.. No clue as to
+	 * why..
+	 */
+	ptr = (get & 0x7ff) >> 2;
+
+	if (device->card_type < NV_40) {
+		mthd = nv_rd32(priv, NV04_PFIFO_CACHE1_METHOD(ptr));
+		data = nv_rd32(priv, NV04_PFIFO_CACHE1_DATA(ptr));
+	} else {
+		mthd = nv_rd32(priv, NV40_PFIFO_CACHE1_METHOD(ptr));
+		data = nv_rd32(priv, NV40_PFIFO_CACHE1_DATA(ptr));
+	}
+
+	if (!nv04_fifo_swmthd(priv, chid, mthd, data)) {
+		const char *client_name =
+			nouveau_client_name_for_fifo_chid(&priv->base, chid);
+		nv_error(priv,
+			 "CACHE_ERROR - ch %d [%s] subc %d mthd 0x%04x data 0x%08x\n",
+			 chid, client_name, (mthd >> 13) & 7, mthd & 0x1ffc,
+			 data);
+	}
+
+	nv_wr32(priv, NV04_PFIFO_CACHE1_DMA_PUSH, 0);
+	nv_wr32(priv, NV03_PFIFO_INTR_0, NV_PFIFO_INTR_CACHE_ERROR);
+
+	nv_wr32(priv, NV03_PFIFO_CACHE1_PUSH0,
+		nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH0) & ~1);
+	nv_wr32(priv, NV03_PFIFO_CACHE1_GET, get + 4);
+	nv_wr32(priv, NV03_PFIFO_CACHE1_PUSH0,
+		nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH0) | 1);
+	nv_wr32(priv, NV04_PFIFO_CACHE1_HASH, 0);
+
+	nv_wr32(priv, NV04_PFIFO_CACHE1_DMA_PUSH,
+		nv_rd32(priv, NV04_PFIFO_CACHE1_DMA_PUSH) | 1);
+	nv_wr32(priv, NV04_PFIFO_CACHE1_PULL0, 1);
+}
+
+static void
+nv04_fifo_dma_pusher(struct nouveau_device *device, struct nv04_fifo_priv *priv,
+		u32 chid)
+{
+	const char *client_name;
+	u32 dma_get = nv_rd32(priv, 0x003244);
+	u32 dma_put = nv_rd32(priv, 0x003240);
+	u32 push = nv_rd32(priv, 0x003220);
+	u32 state = nv_rd32(priv, 0x003228);
+
+	client_name = nouveau_client_name_for_fifo_chid(&priv->base, chid);
+
+	if (device->card_type == NV_50) {
+		u32 ho_get = nv_rd32(priv, 0x003328);
+		u32 ho_put = nv_rd32(priv, 0x003320);
+		u32 ib_get = nv_rd32(priv, 0x003334);
+		u32 ib_put = nv_rd32(priv, 0x003330);
+
+		nv_error(priv,
+			 "DMA_PUSHER - ch %d [%s] get 0x%02x%08x put 0x%02x%08x ib_get 0x%08x ib_put 0x%08x state 0x%08x (err: %s) push 0x%08x\n",
+			 chid, client_name, ho_get, dma_get, ho_put, dma_put,
+			 ib_get, ib_put, state, nv_dma_state_err(state), push);
+
+		/* METHOD_COUNT, in DMA_STATE on earlier chipsets */
+		nv_wr32(priv, 0x003364, 0x00000000);
+		if (dma_get != dma_put || ho_get != ho_put) {
+			nv_wr32(priv, 0x003244, dma_put);
+			nv_wr32(priv, 0x003328, ho_put);
+		} else
+		if (ib_get != ib_put)
+			nv_wr32(priv, 0x003334, ib_put);
+	} else {
+		nv_error(priv,
+			 "DMA_PUSHER - ch %d [%s] get 0x%08x put 0x%08x state 0x%08x (err: %s) push 0x%08x\n",
+			 chid, client_name, dma_get, dma_put, state,
+			 nv_dma_state_err(state), push);
+
+		if (dma_get != dma_put)
+			nv_wr32(priv, 0x003244, dma_put);
+	}
+
+	nv_wr32(priv, 0x003228, 0x00000000);
+	nv_wr32(priv, 0x003220, 0x00000001);
+	nv_wr32(priv, 0x002100, NV_PFIFO_INTR_DMA_PUSHER);
+}
+
 void
 nv04_fifo_intr(struct nouveau_subdev *subdev)
 {
@@ -416,96 +509,12 @@ nv04_fifo_intr(struct nouveau_subdev *subdev)
 		get  = nv_rd32(priv, NV03_PFIFO_CACHE1_GET);
 
 		if (status & NV_PFIFO_INTR_CACHE_ERROR) {
-			uint32_t mthd, data;
-			int ptr;
-
-			/* NV_PFIFO_CACHE1_GET actually goes to 0xffc before
-			 * wrapping on my G80 chips, but CACHE1 isn't big
-			 * enough for this much data.. Tests show that it
-			 * wraps around to the start at GET=0x800.. No clue
-			 * as to why..
-			 */
-			ptr = (get & 0x7ff) >> 2;
-
-			if (device->card_type < NV_40) {
-				mthd = nv_rd32(priv,
-					NV04_PFIFO_CACHE1_METHOD(ptr));
-				data = nv_rd32(priv,
-					NV04_PFIFO_CACHE1_DATA(ptr));
-			} else {
-				mthd = nv_rd32(priv,
-					NV40_PFIFO_CACHE1_METHOD(ptr));
-				data = nv_rd32(priv,
-					NV40_PFIFO_CACHE1_DATA(ptr));
-			}
-
-			if (!nv04_fifo_swmthd(priv, chid, mthd, data)) {
-				nv_error(priv, "CACHE_ERROR - Ch %d/%d "
-					      "Mthd 0x%04x Data 0x%08x\n",
-					chid, (mthd >> 13) & 7, mthd & 0x1ffc,
-					data);
-			}
-
-			nv_wr32(priv, NV04_PFIFO_CACHE1_DMA_PUSH, 0);
-			nv_wr32(priv, NV03_PFIFO_INTR_0,
-						NV_PFIFO_INTR_CACHE_ERROR);
-
-			nv_wr32(priv, NV03_PFIFO_CACHE1_PUSH0,
-				nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH0) & ~1);
-			nv_wr32(priv, NV03_PFIFO_CACHE1_GET, get + 4);
-			nv_wr32(priv, NV03_PFIFO_CACHE1_PUSH0,
-				nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH0) | 1);
-			nv_wr32(priv, NV04_PFIFO_CACHE1_HASH, 0);
-
-			nv_wr32(priv, NV04_PFIFO_CACHE1_DMA_PUSH,
-				nv_rd32(priv, NV04_PFIFO_CACHE1_DMA_PUSH) | 1);
-			nv_wr32(priv, NV04_PFIFO_CACHE1_PULL0, 1);
-
+			nv04_fifo_cache_error(device, priv, chid, get);
 			status &= ~NV_PFIFO_INTR_CACHE_ERROR;
 		}
 
 		if (status & NV_PFIFO_INTR_DMA_PUSHER) {
-			u32 dma_get = nv_rd32(priv, 0x003244);
-			u32 dma_put = nv_rd32(priv, 0x003240);
-			u32 push = nv_rd32(priv, 0x003220);
-			u32 state = nv_rd32(priv, 0x003228);
-
-			if (device->card_type == NV_50) {
-				u32 ho_get = nv_rd32(priv, 0x003328);
-				u32 ho_put = nv_rd32(priv, 0x003320);
-				u32 ib_get = nv_rd32(priv, 0x003334);
-				u32 ib_put = nv_rd32(priv, 0x003330);
-
-				nv_error(priv, "DMA_PUSHER - Ch %d Get 0x%02x%08x "
-				     "Put 0x%02x%08x IbGet 0x%08x IbPut 0x%08x "
-				     "State 0x%08x (err: %s) Push 0x%08x\n",
-					chid, ho_get, dma_get, ho_put,
-					dma_put, ib_get, ib_put, state,
-					nv_dma_state_err(state),
-					push);
-
-				/* METHOD_COUNT, in DMA_STATE on earlier chipsets */
-				nv_wr32(priv, 0x003364, 0x00000000);
-				if (dma_get != dma_put || ho_get != ho_put) {
-					nv_wr32(priv, 0x003244, dma_put);
-					nv_wr32(priv, 0x003328, ho_put);
-				} else
-				if (ib_get != ib_put) {
-					nv_wr32(priv, 0x003334, ib_put);
-				}
-			} else {
-				nv_error(priv, "DMA_PUSHER - Ch %d Get 0x%08x "
-					     "Put 0x%08x State 0x%08x (err: %s) Push 0x%08x\n",
-					chid, dma_get, dma_put, state,
-					nv_dma_state_err(state), push);
-
-				if (dma_get != dma_put)
-					nv_wr32(priv, 0x003244, dma_put);
-			}
-
-			nv_wr32(priv, 0x003228, 0x00000000);
-			nv_wr32(priv, 0x003220, 0x00000001);
-			nv_wr32(priv, 0x002100, NV_PFIFO_INTR_DMA_PUSHER);
+			nv04_fifo_dma_pusher(device, priv, chid);
 			status &= ~NV_PFIFO_INTR_DMA_PUSHER;
 		}
 
@@ -528,6 +537,12 @@ nv04_fifo_intr(struct nouveau_subdev *subdev)
 				status &= ~0x00000010;
 				nv_wr32(priv, 0x002100, 0x00000010);
 			}
+
+			if (status & 0x40000000) {
+				nouveau_event_trigger(priv->base.uevent, 0);
+				nv_wr32(priv, 0x002100, 0x40000000);
+				status &= ~0x40000000;
+			}
 		}
 
 		if (status) {
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
index bd096364f680..840af6172788 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
@@ -129,7 +129,8 @@ nv50_fifo_context_detach(struct nouveau_object *parent, bool suspend,
 	/* do the kickoff... */
 	nv_wr32(priv, 0x0032fc, nv_gpuobj(base)->addr >> 12);
 	if (!nv_wait_ne(priv, 0x0032fc, 0xffffffff, 0xffffffff)) {
-		nv_error(priv, "channel %d unload timeout\n", chan->base.chid);
+		nv_error(priv, "channel %d [%s] unload timeout\n",
+			 chan->base.chid, nouveau_client_name(chan));
 		if (suspend)
 			ret = -EBUSY;
 	}
@@ -480,7 +481,7 @@ nv50_fifo_init(struct nouveau_object *object)
 	nv_wr32(priv, 0x002044, 0x01003fff);
 
 	nv_wr32(priv, 0x002100, 0xffffffff);
-	nv_wr32(priv, 0x002140, 0xffffffff);
+	nv_wr32(priv, 0x002140, 0xbfffffff);
 
 	for (i = 0; i < 128; i++)
 		nv_wr32(priv, 0x002600 + (i * 4), 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
index 1eb1c512f503..094000e87871 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
@@ -26,6 +26,7 @@
 #include <core/client.h>
 #include <core/engctx.h>
 #include <core/ramht.h>
+#include <core/event.h>
 #include <core/class.h>
 #include <core/math.h>
 
@@ -100,7 +101,8 @@ nv84_fifo_context_detach(struct nouveau_object *parent, bool suspend,
 	done = nv_wait_ne(priv, 0x0032fc, 0xffffffff, 0xffffffff);
 	nv_wr32(priv, 0x002520, save);
 	if (!done) {
-		nv_error(priv, "channel %d unload timeout\n", chan->base.chid);
+		nv_error(priv, "channel %d [%s] unload timeout\n",
+			 chan->base.chid, nouveau_client_name(chan));
 		if (suspend)
 			return -EBUSY;
 	}
@@ -378,6 +380,20 @@ nv84_fifo_cclass = {
  * PFIFO engine
  ******************************************************************************/
 
+static void
+nv84_fifo_uevent_enable(struct nouveau_event *event, int index)
+{
+	struct nv84_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x40000000, 0x40000000);
+}
+
+static void
+nv84_fifo_uevent_disable(struct nouveau_event *event, int index)
+{
+	struct nv84_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x40000000, 0x00000000);
+}
+
 static int
 nv84_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
@@ -401,6 +417,10 @@ nv84_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	priv->base.uevent->enable = nv84_fifo_uevent_enable;
+	priv->base.uevent->disable = nv84_fifo_uevent_disable;
+	priv->base.uevent->priv = priv;
+
 	nv_subdev(priv)->unit = 0x00000100;
 	nv_subdev(priv)->intr = nv04_fifo_intr;
 	nv_engine(priv)->cclass = &nv84_fifo_cclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
index b4365dde1859..4f226afb5591 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
@@ -27,6 +27,7 @@
 #include <core/namedb.h>
 #include <core/gpuobj.h>
 #include <core/engctx.h>
+#include <core/event.h>
 #include <core/class.h>
 #include <core/math.h>
 #include <core/enum.h>
@@ -149,7 +150,8 @@ nvc0_fifo_context_detach(struct nouveau_object *parent, bool suspend,
 
 	nv_wr32(priv, 0x002634, chan->base.chid);
 	if (!nv_wait(priv, 0x002634, 0xffffffff, chan->base.chid)) {
-		nv_error(priv, "channel %d kick timeout\n", chan->base.chid);
+		nv_error(priv, "channel %d [%s] kick timeout\n",
+			 chan->base.chid, nouveau_client_name(chan));
 		if (suspend)
 			return -EBUSY;
 	}
@@ -333,17 +335,17 @@ nvc0_fifo_cclass = {
  ******************************************************************************/
 
 static const struct nouveau_enum nvc0_fifo_fault_unit[] = {
-	{ 0x00, "PGRAPH" },
+	{ 0x00, "PGRAPH", NULL, NVDEV_ENGINE_GR },
 	{ 0x03, "PEEPHOLE" },
 	{ 0x04, "BAR1" },
 	{ 0x05, "BAR3" },
-	{ 0x07, "PFIFO" },
-	{ 0x10, "PBSP" },
-	{ 0x11, "PPPP" },
+	{ 0x07, "PFIFO", NULL, NVDEV_ENGINE_FIFO },
+	{ 0x10, "PBSP", NULL, NVDEV_ENGINE_BSP },
+	{ 0x11, "PPPP", NULL, NVDEV_ENGINE_PPP },
 	{ 0x13, "PCOUNTER" },
-	{ 0x14, "PVP" },
-	{ 0x15, "PCOPY0" },
-	{ 0x16, "PCOPY1" },
+	{ 0x14, "PVP", NULL, NVDEV_ENGINE_VP },
+	{ 0x15, "PCOPY0", NULL, NVDEV_ENGINE_COPY0 },
+	{ 0x16, "PCOPY1", NULL, NVDEV_ENGINE_COPY1 },
 	{ 0x17, "PDAEMON" },
 	{}
 };
@@ -402,6 +404,9 @@ nvc0_fifo_isr_vm_fault(struct nvc0_fifo_priv *priv, int unit)
 	u32 vahi = nv_rd32(priv, 0x002808 + (unit * 0x10));
 	u32 stat = nv_rd32(priv, 0x00280c + (unit * 0x10));
 	u32 client = (stat & 0x00001f00) >> 8;
+	const struct nouveau_enum *en;
+	struct nouveau_engine *engine;
+	struct nouveau_object *engctx = NULL;
 
 	switch (unit) {
 	case 3: /* PEEPHOLE */
@@ -420,16 +425,26 @@ nvc0_fifo_isr_vm_fault(struct nvc0_fifo_priv *priv, int unit)
 	nv_error(priv, "%s fault at 0x%010llx [", (stat & 0x00000080) ?
 		 "write" : "read", (u64)vahi << 32 | valo);
 	nouveau_enum_print(nvc0_fifo_fault_reason, stat & 0x0000000f);
-	printk("] from ");
-	nouveau_enum_print(nvc0_fifo_fault_unit, unit);
+	pr_cont("] from ");
+	en = nouveau_enum_print(nvc0_fifo_fault_unit, unit);
 	if (stat & 0x00000040) {
-		printk("/");
+		pr_cont("/");
 		nouveau_enum_print(nvc0_fifo_fault_hubclient, client);
 	} else {
-		printk("/GPC%d/", (stat & 0x1f000000) >> 24);
+		pr_cont("/GPC%d/", (stat & 0x1f000000) >> 24);
 		nouveau_enum_print(nvc0_fifo_fault_gpcclient, client);
 	}
-	printk(" on channel 0x%010llx\n", (u64)inst << 12);
+
+	if (en && en->data2) {
+		engine = nouveau_engine(priv, en->data2);
+		if (engine)
+			engctx = nouveau_engctx_get(engine, inst);
+
+	}
+	pr_cont(" on channel 0x%010llx [%s]\n", (u64)inst << 12,
+			nouveau_client_name(engctx));
+
+	nouveau_engctx_put(engctx);
 }
 
 static int
@@ -484,10 +499,12 @@ nvc0_fifo_isr_subfifo_intr(struct nvc0_fifo_priv *priv, int unit)
 	if (show) {
 		nv_error(priv, "SUBFIFO%d:", unit);
 		nouveau_bitfield_print(nvc0_fifo_subfifo_intr, show);
-		printk("\n");
-		nv_error(priv, "SUBFIFO%d: ch %d subc %d mthd 0x%04x "
-			       "data 0x%08x\n",
-			 unit, chid, subc, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "SUBFIFO%d: ch %d [%s] subc %d mthd 0x%04x data 0x%08x\n",
+			 unit, chid,
+			 nouveau_client_name_for_fifo_chid(&priv->base, chid),
+			 subc, mthd, data);
 	}
 
 	nv_wr32(priv, 0x0400c0 + (unit * 0x2000), 0x80600008);
@@ -501,12 +518,34 @@ nvc0_fifo_intr(struct nouveau_subdev *subdev)
 	u32 mask = nv_rd32(priv, 0x002140);
 	u32 stat = nv_rd32(priv, 0x002100) & mask;
 
+	if (stat & 0x00000001) {
+		u32 intr = nv_rd32(priv, 0x00252c);
+		nv_warn(priv, "INTR 0x00000001: 0x%08x\n", intr);
+		nv_wr32(priv, 0x002100, 0x00000001);
+		stat &= ~0x00000001;
+	}
+
 	if (stat & 0x00000100) {
-		nv_warn(priv, "unknown status 0x00000100\n");
+		u32 intr = nv_rd32(priv, 0x00254c);
+		nv_warn(priv, "INTR 0x00000100: 0x%08x\n", intr);
 		nv_wr32(priv, 0x002100, 0x00000100);
 		stat &= ~0x00000100;
 	}
 
+	if (stat & 0x00010000) {
+		u32 intr = nv_rd32(priv, 0x00256c);
+		nv_warn(priv, "INTR 0x00010000: 0x%08x\n", intr);
+		nv_wr32(priv, 0x002100, 0x00010000);
+		stat &= ~0x00010000;
+	}
+
+	if (stat & 0x01000000) {
+		u32 intr = nv_rd32(priv, 0x00258c);
+		nv_warn(priv, "INTR 0x01000000: 0x%08x\n", intr);
+		nv_wr32(priv, 0x002100, 0x01000000);
+		stat &= ~0x01000000;
+	}
+
 	if (stat & 0x10000000) {
 		u32 units = nv_rd32(priv, 0x00259c);
 		u32 u = units;
@@ -536,11 +575,20 @@ nvc0_fifo_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (stat & 0x40000000) {
-		nv_warn(priv, "unknown status 0x40000000\n");
-		nv_mask(priv, 0x002a00, 0x00000000, 0x00000000);
+		u32 intr0 = nv_rd32(priv, 0x0025a4);
+		u32 intr1 = nv_mask(priv, 0x002a00, 0x00000000, 0x00000);
+		nv_debug(priv, "INTR 0x40000000: 0x%08x 0x%08x\n",
+			       intr0, intr1);
 		stat &= ~0x40000000;
 	}
 
+	if (stat & 0x80000000) {
+		u32 intr = nv_mask(priv, 0x0025a8, 0x00000000, 0x00000000);
+		nouveau_event_trigger(priv->base.uevent, 0);
+		nv_debug(priv, "INTR 0x80000000: 0x%08x\n", intr);
+		stat &= ~0x80000000;
+	}
+
 	if (stat) {
 		nv_fatal(priv, "unhandled status 0x%08x\n", stat);
 		nv_wr32(priv, 0x002100, stat);
@@ -548,6 +596,20 @@ nvc0_fifo_intr(struct nouveau_subdev *subdev)
 	}
 }
 
+static void
+nvc0_fifo_uevent_enable(struct nouveau_event *event, int index)
+{
+	struct nvc0_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
+}
+
+static void
+nvc0_fifo_uevent_disable(struct nouveau_event *event, int index)
+{
+	struct nvc0_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
+}
+
 static int
 nvc0_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
@@ -581,6 +643,10 @@ nvc0_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	priv->base.uevent->enable = nvc0_fifo_uevent_enable;
+	priv->base.uevent->disable = nvc0_fifo_uevent_disable;
+	priv->base.uevent->priv = priv;
+
 	nv_subdev(priv)->unit = 0x00000100;
 	nv_subdev(priv)->intr = nvc0_fifo_intr;
 	nv_engine(priv)->cclass = &nvc0_fifo_cclass;
@@ -639,7 +705,8 @@ nvc0_fifo_init(struct nouveau_object *object)
 
 	nv_wr32(priv, 0x002a00, 0xffffffff); /* clears PFIFO.INTR bit 30 */
 	nv_wr32(priv, 0x002100, 0xffffffff);
-	nv_wr32(priv, 0x002140, 0xbfffffff);
+	nv_wr32(priv, 0x002140, 0x3fffffff);
+	nv_wr32(priv, 0x002628, 0x00000001); /* makes mthd 0x20 work */
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index c930da99c2c1..4419e40d88e9 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -27,6 +27,7 @@
 #include <core/namedb.h>
 #include <core/gpuobj.h>
 #include <core/engctx.h>
+#include <core/event.h>
 #include <core/class.h>
 #include <core/math.h>
 #include <core/enum.h>
@@ -184,7 +185,8 @@ nve0_fifo_context_detach(struct nouveau_object *parent, bool suspend,
 
 	nv_wr32(priv, 0x002634, chan->base.chid);
 	if (!nv_wait(priv, 0x002634, 0xffffffff, chan->base.chid)) {
-		nv_error(priv, "channel %d kick timeout\n", chan->base.chid);
+		nv_error(priv, "channel %d [%s] kick timeout\n",
+			 chan->base.chid, nouveau_client_name(chan));
 		if (suspend)
 			return -EBUSY;
 	}
@@ -412,20 +414,34 @@ nve0_fifo_isr_vm_fault(struct nve0_fifo_priv *priv, int unit)
 	u32 vahi = nv_rd32(priv, 0x2808 + (unit * 0x10));
 	u32 stat = nv_rd32(priv, 0x280c + (unit * 0x10));
 	u32 client = (stat & 0x00001f00) >> 8;
+	const struct nouveau_enum *en;
+	struct nouveau_engine *engine;
+	struct nouveau_object *engctx = NULL;
 
 	nv_error(priv, "PFIFO: %s fault at 0x%010llx [", (stat & 0x00000080) ?
 		       "write" : "read", (u64)vahi << 32 | valo);
 	nouveau_enum_print(nve0_fifo_fault_reason, stat & 0x0000000f);
-	printk("] from ");
-	nouveau_enum_print(nve0_fifo_fault_unit, unit);
+	pr_cont("] from ");
+	en = nouveau_enum_print(nve0_fifo_fault_unit, unit);
 	if (stat & 0x00000040) {
-		printk("/");
+		pr_cont("/");
 		nouveau_enum_print(nve0_fifo_fault_hubclient, client);
 	} else {
-		printk("/GPC%d/", (stat & 0x1f000000) >> 24);
+		pr_cont("/GPC%d/", (stat & 0x1f000000) >> 24);
 		nouveau_enum_print(nve0_fifo_fault_gpcclient, client);
 	}
-	printk(" on channel 0x%010llx\n", (u64)inst << 12);
+
+	if (en && en->data2) {
+		engine = nouveau_engine(priv, en->data2);
+		if (engine)
+			engctx = nouveau_engctx_get(engine, inst);
+
+	}
+
+	pr_cont(" on channel 0x%010llx [%s]\n", (u64)inst << 12,
+			nouveau_client_name(engctx));
+
+	nouveau_engctx_put(engctx);
 }
 
 static int
@@ -480,10 +496,12 @@ nve0_fifo_isr_subfifo_intr(struct nve0_fifo_priv *priv, int unit)
 	if (show) {
 		nv_error(priv, "SUBFIFO%d:", unit);
 		nouveau_bitfield_print(nve0_fifo_subfifo_intr, show);
-		printk("\n");
-		nv_error(priv, "SUBFIFO%d: ch %d subc %d mthd 0x%04x "
-			       "data 0x%08x\n",
-			 unit, chid, subc, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "SUBFIFO%d: ch %d [%s] subc %d mthd 0x%04x data 0x%08x\n",
+			 unit, chid,
+			 nouveau_client_name_for_fifo_chid(&priv->base, chid),
+			 subc, mthd, data);
 	}
 
 	nv_wr32(priv, 0x0400c0 + (unit * 0x2000), 0x80600008);
@@ -537,6 +555,12 @@ nve0_fifo_intr(struct nouveau_subdev *subdev)
 		stat &= ~0x40000000;
 	}
 
+	if (stat & 0x80000000) {
+		nouveau_event_trigger(priv->base.uevent, 0);
+		nv_wr32(priv, 0x002100, 0x80000000);
+		stat &= ~0x80000000;
+	}
+
 	if (stat) {
 		nv_fatal(priv, "unhandled status 0x%08x\n", stat);
 		nv_wr32(priv, 0x002100, stat);
@@ -544,6 +568,20 @@ nve0_fifo_intr(struct nouveau_subdev *subdev)
 	}
 }
 
+static void
+nve0_fifo_uevent_enable(struct nouveau_event *event, int index)
+{
+	struct nve0_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
+}
+
+static void
+nve0_fifo_uevent_disable(struct nouveau_event *event, int index)
+{
+	struct nve0_fifo_priv *priv = event->priv;
+	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
+}
+
 static int
 nve0_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
@@ -567,6 +605,10 @@ nve0_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	priv->base.uevent->enable = nve0_fifo_uevent_enable;
+	priv->base.uevent->disable = nve0_fifo_uevent_disable;
+	priv->base.uevent->priv = priv;
+
 	nv_subdev(priv)->unit = 0x00000100;
 	nv_subdev(priv)->intr = nve0_fifo_intr;
 	nv_engine(priv)->cclass = &nve0_fifo_cclass;
@@ -617,7 +659,7 @@ nve0_fifo_init(struct nouveau_object *object)
 
 	nv_wr32(priv, 0x002a00, 0xffffffff);
 	nv_wr32(priv, 0x002100, 0xffffffff);
-	nv_wr32(priv, 0x002140, 0xbfffffff);
+	nv_wr32(priv, 0x002140, 0x3fffffff);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv04.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv04.c
index e30a9c5ff1fc..ad13dcdd15f9 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv04.c
@@ -22,6 +22,7 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/class.h>
 #include <core/handle.h>
@@ -1297,16 +1298,17 @@ nv04_graph_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, NV04_PGRAPH_FIFO, 0x00000001);
 
 	if (show) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv04_graph_intr_name, show);
-		printk(" nsource:");
+		pr_cont(" nsource:");
 		nouveau_bitfield_print(nv04_graph_nsource, nsource);
-		printk(" nstatus:");
+		pr_cont(" nstatus:");
 		nouveau_bitfield_print(nv04_graph_nstatus, nstatus);
-		printk("\n");
-		nv_error(priv, "ch %d/%d class 0x%04x "
-			       "mthd 0x%04x data 0x%08x\n",
-			 chid, subc, class, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "ch %d [%s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, nouveau_client_name(chan), subc, class, mthd,
+			 data);
 	}
 
 	nouveau_namedb_put(handle);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
index 5c0f843ea249..23c143aaa556 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
@@ -22,6 +22,7 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/class.h>
 #include <core/handle.h>
@@ -1193,16 +1194,17 @@ nv10_graph_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, NV04_PGRAPH_FIFO, 0x00000001);
 
 	if (show) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv10_graph_intr_name, show);
-		printk(" nsource:");
+		pr_cont(" nsource:");
 		nouveau_bitfield_print(nv04_graph_nsource, nsource);
-		printk(" nstatus:");
+		pr_cont(" nstatus:");
 		nouveau_bitfield_print(nv10_graph_nstatus, nstatus);
-		printk("\n");
-		nv_error(priv, "ch %d/%d class 0x%04x "
-			       "mthd 0x%04x data 0x%08x\n",
-			 chid, subc, class, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "ch %d [%s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, nouveau_client_name(chan), subc, class, mthd,
+			 data);
 	}
 
 	nouveau_namedb_put(handle);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv20.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv20.c
index 5b20401bf911..0607b9801748 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv20.c
@@ -1,3 +1,4 @@
+#include <core/client.h>
 #include <core/os.h>
 #include <core/class.h>
 #include <core/engctx.h>
@@ -224,15 +225,17 @@ nv20_graph_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, NV04_PGRAPH_FIFO, 0x00000001);
 
 	if (show) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv10_graph_intr_name, show);
-		printk(" nsource:");
+		pr_cont(" nsource:");
 		nouveau_bitfield_print(nv04_graph_nsource, nsource);
-		printk(" nstatus:");
+		pr_cont(" nstatus:");
 		nouveau_bitfield_print(nv10_graph_nstatus, nstatus);
-		printk("\n");
-		nv_error(priv, "ch %d/%d class 0x%04x mthd 0x%04x data 0x%08x\n",
-			chid, subc, class, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "ch %d [%s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, nouveau_client_name(engctx), subc, class, mthd,
+			 data);
 	}
 
 	nouveau_engctx_put(engctx);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv40.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv40.c
index 0b36dd3deebd..17049d5c723d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv40.c
@@ -22,6 +22,7 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/class.h>
 #include <core/handle.h>
@@ -321,16 +322,17 @@ nv40_graph_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, NV04_PGRAPH_FIFO, 0x00000001);
 
 	if (show) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv10_graph_intr_name, show);
-		printk(" nsource:");
+		pr_cont(" nsource:");
 		nouveau_bitfield_print(nv04_graph_nsource, nsource);
-		printk(" nstatus:");
+		pr_cont(" nstatus:");
 		nouveau_bitfield_print(nv10_graph_nstatus, nstatus);
-		printk("\n");
-		nv_error(priv, "ch %d [0x%08x] subc %d class 0x%04x "
-			       "mthd 0x%04x data 0x%08x\n",
-			 chid, inst << 4, subc, class, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "ch %d [0x%08x %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, inst << 4, nouveau_client_name(engctx), subc,
+			 class, mthd, data);
 	}
 
 	nouveau_engctx_put(engctx);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
index b1c3d835b4c2..f2b1a7a124f2 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
@@ -24,6 +24,7 @@
 
 #include <core/os.h>
 #include <core/class.h>
+#include <core/client.h>
 #include <core/handle.h>
 #include <core/engctx.h>
 #include <core/enum.h>
@@ -418,7 +419,7 @@ nv50_priv_mp_trap(struct nv50_graph_priv *priv, int tpid, int display)
 			nv_error(priv, "TRAP_MP_EXEC - "
 					"TP %d MP %d: ", tpid, i);
 			nouveau_enum_print(nv50_mp_exec_error_names, status);
-			printk(" at %06x warp %d, opcode %08x %08x\n",
+			pr_cont(" at %06x warp %d, opcode %08x %08x\n",
 					pc&0xffffff, pc >> 24,
 					oplow, ophigh);
 		}
@@ -532,7 +533,7 @@ nv50_priv_tp_trap(struct nv50_graph_priv *priv, int type, u32 ustatus_old,
 
 static int
 nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
-			int chid, u64 inst)
+			int chid, u64 inst, struct nouveau_object *engctx)
 {
 	u32 status = nv_rd32(priv, 0x400108);
 	u32 ustatus;
@@ -565,12 +566,11 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 
 			nv_error(priv, "TRAP DISPATCH_FAULT\n");
 			if (display && (addr & 0x80000000)) {
-				nv_error(priv, "ch %d [0x%010llx] "
-					     "subc %d class 0x%04x mthd 0x%04x "
-					     "data 0x%08x%08x "
-					     "400808 0x%08x 400848 0x%08x\n",
-					chid, inst, subc, class, mthd, datah,
-					datal, addr, r848);
+				nv_error(priv,
+					 "ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x%08x 400808 0x%08x 400848 0x%08x\n",
+					 chid, inst,
+					 nouveau_client_name(engctx), subc,
+					 class, mthd, datah, datal, addr, r848);
 			} else
 			if (display) {
 				nv_error(priv, "no stuck command?\n");
@@ -591,11 +591,11 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 
 			nv_error(priv, "TRAP DISPATCH_QUERY\n");
 			if (display && (addr & 0x80000000)) {
-				nv_error(priv, "ch %d [0x%010llx] "
-					     "subc %d class 0x%04x mthd 0x%04x "
-					     "data 0x%08x 40084c 0x%08x\n",
-					chid, inst, subc, class, mthd,
-					data, addr);
+				nv_error(priv,
+					 "ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x 40084c 0x%08x\n",
+					 chid, inst,
+					 nouveau_client_name(engctx), subc,
+					 class, mthd, data, addr);
 			} else
 			if (display) {
 				nv_error(priv, "no stuck command?\n");
@@ -623,7 +623,7 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 		if (display) {
 			nv_error(priv, "TRAP_M2MF");
 			nouveau_bitfield_print(nv50_graph_trap_m2mf, ustatus);
-			printk("\n");
+			pr_cont("\n");
 			nv_error(priv, "TRAP_M2MF %08x %08x %08x %08x\n",
 				nv_rd32(priv, 0x406804), nv_rd32(priv, 0x406808),
 				nv_rd32(priv, 0x40680c), nv_rd32(priv, 0x406810));
@@ -644,7 +644,7 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 		if (display) {
 			nv_error(priv, "TRAP_VFETCH");
 			nouveau_bitfield_print(nv50_graph_trap_vfetch, ustatus);
-			printk("\n");
+			pr_cont("\n");
 			nv_error(priv, "TRAP_VFETCH %08x %08x %08x %08x\n",
 				nv_rd32(priv, 0x400c00), nv_rd32(priv, 0x400c08),
 				nv_rd32(priv, 0x400c0c), nv_rd32(priv, 0x400c10));
@@ -661,7 +661,7 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 		if (display) {
 			nv_error(priv, "TRAP_STRMOUT");
 			nouveau_bitfield_print(nv50_graph_trap_strmout, ustatus);
-			printk("\n");
+			pr_cont("\n");
 			nv_error(priv, "TRAP_STRMOUT %08x %08x %08x %08x\n",
 				nv_rd32(priv, 0x401804), nv_rd32(priv, 0x401808),
 				nv_rd32(priv, 0x40180c), nv_rd32(priv, 0x401810));
@@ -682,7 +682,7 @@ nv50_graph_trap_handler(struct nv50_graph_priv *priv, u32 display,
 		if (display) {
 			nv_error(priv, "TRAP_CCACHE");
 			nouveau_bitfield_print(nv50_graph_trap_ccache, ustatus);
-			printk("\n");
+			pr_cont("\n");
 			nv_error(priv, "TRAP_CCACHE %08x %08x %08x %08x"
 				     " %08x %08x %08x\n",
 				nv_rd32(priv, 0x405000), nv_rd32(priv, 0x405004),
@@ -774,11 +774,12 @@ nv50_graph_intr(struct nouveau_subdev *subdev)
 		u32 ecode = nv_rd32(priv, 0x400110);
 		nv_error(priv, "DATA_ERROR ");
 		nouveau_enum_print(nv50_data_error_names, ecode);
-		printk("\n");
+		pr_cont("\n");
 	}
 
 	if (stat & 0x00200000) {
-		if (!nv50_graph_trap_handler(priv, show, chid, (u64)inst << 12))
+		if (!nv50_graph_trap_handler(priv, show, chid, (u64)inst << 12,
+				engctx))
 			show &= ~0x00200000;
 	}
 
@@ -786,12 +787,13 @@ nv50_graph_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, 0x400500, 0x00010001);
 
 	if (show) {
-		nv_error(priv, "");
+		nv_error(priv, "%s", "");
 		nouveau_bitfield_print(nv50_graph_intr_name, show);
-		printk("\n");
-		nv_error(priv, "ch %d [0x%010llx] subc %d class 0x%04x "
-			       "mthd 0x%04x data 0x%08x\n",
-			 chid, (u64)inst << 12, subc, class, mthd, data);
+		pr_cont("\n");
+		nv_error(priv,
+			 "ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, (u64)inst << 12, nouveau_client_name(engctx),
+			 subc, class, mthd, data);
 	}
 
 	if (nv_rd32(priv, 0x400824) & (1 << 31))
@@ -907,9 +909,8 @@ nv50_graph_init(struct nouveau_object *object)
 	nv_wr32(priv, 0x400828, 0x00000000);
 	nv_wr32(priv, 0x40082c, 0x00000000);
 	nv_wr32(priv, 0x400830, 0x00000000);
-	nv_wr32(priv, 0x400724, 0x00000000);
 	nv_wr32(priv, 0x40032c, 0x00000000);
-	nv_wr32(priv, 0x400320, 4);	/* CTXCTL_CMD = NEWCTXDMA */
+	nv_wr32(priv, 0x400330, 0x00000000);
 
 	/* some unknown zcull magic */
 	switch (nv_device(priv)->chipset & 0xf0) {
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
index 45aff5f5085a..0de0dd724aff 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
@@ -433,10 +433,10 @@ nvc0_graph_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00000010) {
 		handle = nouveau_handle_get_class(engctx, class);
 		if (!handle || nv_call(handle->object, mthd, data)) {
-			nv_error(priv, "ILLEGAL_MTHD ch %d [0x%010llx] "
-				     "subc %d class 0x%04x mthd 0x%04x "
-				     "data 0x%08x\n",
-				 chid, inst << 12, subc, class, mthd, data);
+			nv_error(priv,
+				 "ILLEGAL_MTHD ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+				 chid, inst << 12, nouveau_client_name(engctx),
+				 subc, class, mthd, data);
 		}
 		nouveau_handle_put(handle);
 		nv_wr32(priv, 0x400100, 0x00000010);
@@ -444,9 +444,10 @@ nvc0_graph_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (stat & 0x00000020) {
-		nv_error(priv, "ILLEGAL_CLASS ch %d [0x%010llx] subc %d "
-			     "class 0x%04x mthd 0x%04x data 0x%08x\n",
-			chid, inst << 12, subc, class, mthd, data);
+		nv_error(priv,
+			 "ILLEGAL_CLASS ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, inst << 12, nouveau_client_name(engctx), subc,
+			 class, mthd, data);
 		nv_wr32(priv, 0x400100, 0x00000020);
 		stat &= ~0x00000020;
 	}
@@ -454,15 +455,16 @@ nvc0_graph_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00100000) {
 		nv_error(priv, "DATA_ERROR [");
 		nouveau_enum_print(nv50_data_error_names, code);
-		printk("] ch %d [0x%010llx] subc %d class 0x%04x "
-		       "mthd 0x%04x data 0x%08x\n",
-		       chid, inst << 12, subc, class, mthd, data);
+		pr_cont("] ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			chid, inst << 12, nouveau_client_name(engctx), subc,
+			class, mthd, data);
 		nv_wr32(priv, 0x400100, 0x00100000);
 		stat &= ~0x00100000;
 	}
 
 	if (stat & 0x00200000) {
-		nv_error(priv, "TRAP ch %d [0x%010llx]\n", chid, inst << 12);
+		nv_error(priv, "TRAP ch %d [0x%010llx %s]\n", chid, inst << 12,
+			 nouveau_client_name(engctx));
 		nvc0_graph_trap_intr(priv);
 		nv_wr32(priv, 0x400100, 0x00200000);
 		stat &= ~0x00200000;
@@ -611,10 +613,8 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 static void
 nvc0_graph_dtor_fw(struct nvc0_graph_fuc *fuc)
 {
-	if (fuc->data) {
-		kfree(fuc->data);
-		fuc->data = NULL;
-	}
+	kfree(fuc->data);
+	fuc->data = NULL;
 }
 
 void
@@ -622,8 +622,7 @@ nvc0_graph_dtor(struct nouveau_object *object)
 {
 	struct nvc0_graph_priv *priv = (void *)object;
 
-	if (priv->data)
-		kfree(priv->data);
+	kfree(priv->data);
 
 	nvc0_graph_dtor_fw(&priv->fuc409c);
 	nvc0_graph_dtor_fw(&priv->fuc409d);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nve0.c b/drivers/gpu/drm/nouveau/core/engine/graph/nve0.c
index 9f82e9702b46..61cec0f6ff1c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nve0.c
@@ -78,15 +78,16 @@ nve0_graph_ctxctl_isr(struct nvc0_graph_priv *priv)
 }
 
 static void
-nve0_graph_trap_isr(struct nvc0_graph_priv *priv, int chid, u64 inst)
+nve0_graph_trap_isr(struct nvc0_graph_priv *priv, int chid, u64 inst,
+		struct nouveau_object *engctx)
 {
 	u32 trap = nv_rd32(priv, 0x400108);
 	int rop;
 
 	if (trap & 0x00000001) {
 		u32 stat = nv_rd32(priv, 0x404000);
-		nv_error(priv, "DISPATCH ch %d [0x%010llx] 0x%08x\n",
-			 chid, inst, stat);
+		nv_error(priv, "DISPATCH ch %d [0x%010llx %s] 0x%08x\n",
+			 chid, inst, nouveau_client_name(engctx), stat);
 		nv_wr32(priv, 0x404000, 0xc0000000);
 		nv_wr32(priv, 0x400108, 0x00000001);
 		trap &= ~0x00000001;
@@ -94,8 +95,8 @@ nve0_graph_trap_isr(struct nvc0_graph_priv *priv, int chid, u64 inst)
 
 	if (trap & 0x00000010) {
 		u32 stat = nv_rd32(priv, 0x405840);
-		nv_error(priv, "SHADER ch %d [0x%010llx] 0x%08x\n",
-			 chid, inst, stat);
+		nv_error(priv, "SHADER ch %d [0x%010llx %s] 0x%08x\n",
+			 chid, inst, nouveau_client_name(engctx), stat);
 		nv_wr32(priv, 0x405840, 0xc0000000);
 		nv_wr32(priv, 0x400108, 0x00000010);
 		trap &= ~0x00000010;
@@ -105,8 +106,10 @@ nve0_graph_trap_isr(struct nvc0_graph_priv *priv, int chid, u64 inst)
 		for (rop = 0; rop < priv->rop_nr; rop++) {
 			u32 statz = nv_rd32(priv, ROP_UNIT(rop, 0x070));
 			u32 statc = nv_rd32(priv, ROP_UNIT(rop, 0x144));
-			nv_error(priv, "ROP%d ch %d [0x%010llx] 0x%08x 0x%08x\n",
-				 rop, chid, inst, statz, statc);
+			nv_error(priv,
+				 "ROP%d ch %d [0x%010llx %s] 0x%08x 0x%08x\n",
+				 rop, chid, inst, nouveau_client_name(engctx),
+				 statz, statc);
 			nv_wr32(priv, ROP_UNIT(rop, 0x070), 0xc0000000);
 			nv_wr32(priv, ROP_UNIT(rop, 0x144), 0xc0000000);
 		}
@@ -115,8 +118,8 @@ nve0_graph_trap_isr(struct nvc0_graph_priv *priv, int chid, u64 inst)
 	}
 
 	if (trap) {
-		nv_error(priv, "TRAP ch %d [0x%010llx] 0x%08x\n",
-			 chid, inst, trap);
+		nv_error(priv, "TRAP ch %d [0x%010llx %s] 0x%08x\n",
+			 chid, inst, nouveau_client_name(engctx), trap);
 		nv_wr32(priv, 0x400108, trap);
 	}
 }
@@ -145,10 +148,10 @@ nve0_graph_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00000010) {
 		handle = nouveau_handle_get_class(engctx, class);
 		if (!handle || nv_call(handle->object, mthd, data)) {
-			nv_error(priv, "ILLEGAL_MTHD ch %d [0x%010llx] "
-				     "subc %d class 0x%04x mthd 0x%04x "
-				     "data 0x%08x\n",
-				 chid, inst, subc, class, mthd, data);
+			nv_error(priv,
+				 "ILLEGAL_MTHD ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+				 chid, inst, nouveau_client_name(engctx), subc,
+				 class, mthd, data);
 		}
 		nouveau_handle_put(handle);
 		nv_wr32(priv, 0x400100, 0x00000010);
@@ -156,9 +159,10 @@ nve0_graph_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (stat & 0x00000020) {
-		nv_error(priv, "ILLEGAL_CLASS ch %d [0x%010llx] subc %d "
-			     "class 0x%04x mthd 0x%04x data 0x%08x\n",
-			 chid, inst, subc, class, mthd, data);
+		nv_error(priv,
+			 "ILLEGAL_CLASS ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			 chid, inst, nouveau_client_name(engctx), subc, class,
+			 mthd, data);
 		nv_wr32(priv, 0x400100, 0x00000020);
 		stat &= ~0x00000020;
 	}
@@ -166,15 +170,15 @@ nve0_graph_intr(struct nouveau_subdev *subdev)
 	if (stat & 0x00100000) {
 		nv_error(priv, "DATA_ERROR [");
 		nouveau_enum_print(nv50_data_error_names, code);
-		printk("] ch %d [0x%010llx] subc %d class 0x%04x "
-		       "mthd 0x%04x data 0x%08x\n",
-		       chid, inst, subc, class, mthd, data);
+		pr_cont("] ch %d [0x%010llx %s] subc %d class 0x%04x mthd 0x%04x data 0x%08x\n",
+			chid, inst, nouveau_client_name(engctx), subc, class,
+			mthd, data);
 		nv_wr32(priv, 0x400100, 0x00100000);
 		stat &= ~0x00100000;
 	}
 
 	if (stat & 0x00200000) {
-		nve0_graph_trap_isr(priv, chid, inst);
+		nve0_graph_trap_isr(priv, chid, inst, engctx);
 		nv_wr32(priv, 0x400100, 0x00200000);
 		stat &= ~0x00200000;
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
index 9fd86375f4c4..49ecbb859b25 100644
--- a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
+++ b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
@@ -22,6 +22,7 @@
  * Authors: Ben Skeggs
  */
 
+#include <core/client.h>
 #include <core/os.h>
 #include <core/class.h>
 #include <core/engctx.h>
@@ -231,8 +232,10 @@ nv31_mpeg_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, 0x00b230, 0x00000001);
 
 	if (show) {
-		nv_error(priv, "ch %d [0x%08x] 0x%08x 0x%08x 0x%08x 0x%08x\n",
-			 chid, inst << 4, stat, type, mthd, data);
+		nv_error(priv,
+			 "ch %d [0x%08x %s] 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			 chid, inst << 4, nouveau_client_name(engctx), stat,
+			 type, mthd, data);
 	}
 
 	nouveau_engctx_put(engctx);
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
index b0e7e1c01ce6..c48e74953771 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
@@ -28,6 +28,9 @@
 #include <core/namedb.h>
 #include <core/handle.h>
 #include <core/gpuobj.h>
+#include <core/event.h>
+
+#include <subdev/bar.h>
 
 #include <engine/software.h>
 #include <engine/disp.h>
@@ -90,18 +93,11 @@ nv50_software_mthd_vblsem_release(struct nouveau_object *object, u32 mthd,
 {
 	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
 	struct nouveau_disp *disp = nouveau_disp(object);
-	unsigned long flags;
 	u32 crtc = *(u32 *)args;
-
 	if (crtc > 1)
 		return -EINVAL;
 
-	disp->vblank.get(disp->vblank.data, crtc);
-
-	spin_lock_irqsave(&disp->vblank.lock, flags);
-	list_add(&chan->base.vblank.head, &disp->vblank.list);
-	chan->base.vblank.crtc = crtc;
-	spin_unlock_irqrestore(&disp->vblank.lock, flags);
+	nouveau_event_get(disp->vblank, crtc, &chan->base.vblank.event);
 	return 0;
 }
 
@@ -136,6 +132,29 @@ nv50_software_sclass[] = {
  ******************************************************************************/
 
 static int
+nv50_software_vblsem_release(struct nouveau_eventh *event, int head)
+{
+	struct nouveau_software_chan *chan =
+		container_of(event, struct nouveau_software_chan, vblank.event);
+	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
+	struct nouveau_bar *bar = nouveau_bar(priv);
+
+	nv_wr32(priv, 0x001704, chan->vblank.channel);
+	nv_wr32(priv, 0x001710, 0x80000000 | chan->vblank.ctxdma);
+	bar->flush(bar);
+
+	if (nv_device(priv)->chipset == 0x50) {
+		nv_wr32(priv, 0x001570, chan->vblank.offset);
+		nv_wr32(priv, 0x001574, chan->vblank.value);
+	} else {
+		nv_wr32(priv, 0x060010, chan->vblank.offset);
+		nv_wr32(priv, 0x060014, chan->vblank.value);
+	}
+
+	return NVKM_EVENT_DROP;
+}
+
+static int
 nv50_software_context_ctor(struct nouveau_object *parent,
 			   struct nouveau_object *engine,
 			   struct nouveau_oclass *oclass, void *data, u32 size,
@@ -150,6 +169,7 @@ nv50_software_context_ctor(struct nouveau_object *parent,
 		return ret;
 
 	chan->base.vblank.channel = nv_gpuobj(parent->parent)->addr >> 12;
+	chan->base.vblank.event.func = nv50_software_vblsem_release;
 	return 0;
 }
 
@@ -170,8 +190,8 @@ nv50_software_cclass = {
 
 static int
 nv50_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
+		   struct nouveau_oclass *oclass, void *data, u32 size,
+		   struct nouveau_object **pobject)
 {
 	struct nv50_software_priv *priv;
 	int ret;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
index 282a1cd1bc2f..a523eaad47e3 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
@@ -25,6 +25,9 @@
 #include <core/os.h>
 #include <core/class.h>
 #include <core/engctx.h>
+#include <core/event.h>
+
+#include <subdev/bar.h>
 
 #include <engine/software.h>
 #include <engine/disp.h>
@@ -72,18 +75,12 @@ nvc0_software_mthd_vblsem_release(struct nouveau_object *object, u32 mthd,
 {
 	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
 	struct nouveau_disp *disp = nouveau_disp(object);
-	unsigned long flags;
 	u32 crtc = *(u32 *)args;
 
 	if ((nv_device(object)->card_type < NV_E0 && crtc > 1) || crtc > 3)
 		return -EINVAL;
 
-	disp->vblank.get(disp->vblank.data, crtc);
-
-	spin_lock_irqsave(&disp->vblank.lock, flags);
-	list_add(&chan->base.vblank.head, &disp->vblank.list);
-	chan->base.vblank.crtc = crtc;
-	spin_unlock_irqrestore(&disp->vblank.lock, flags);
+	nouveau_event_get(disp->vblank, crtc, &chan->base.vblank.event);
 	return 0;
 }
 
@@ -118,6 +115,23 @@ nvc0_software_sclass[] = {
  ******************************************************************************/
 
 static int
+nvc0_software_vblsem_release(struct nouveau_eventh *event, int head)
+{
+	struct nouveau_software_chan *chan =
+		container_of(event, struct nouveau_software_chan, vblank.event);
+	struct nvc0_software_priv *priv = (void *)nv_object(chan)->engine;
+	struct nouveau_bar *bar = nouveau_bar(priv);
+
+	nv_wr32(priv, 0x001718, 0x80000000 | chan->vblank.channel);
+	bar->flush(bar);
+	nv_wr32(priv, 0x06000c, upper_32_bits(chan->vblank.offset));
+	nv_wr32(priv, 0x060010, lower_32_bits(chan->vblank.offset));
+	nv_wr32(priv, 0x060014, chan->vblank.value);
+
+	return NVKM_EVENT_DROP;
+}
+
+static int
 nvc0_software_context_ctor(struct nouveau_object *parent,
 			   struct nouveau_object *engine,
 			   struct nouveau_oclass *oclass, void *data, u32 size,
@@ -132,6 +146,7 @@ nvc0_software_context_ctor(struct nouveau_object *parent,
 		return ret;
 
 	chan->base.vblank.channel = nv_gpuobj(parent->parent)->addr >> 12;
+	chan->base.vblank.event.func = nvc0_software_vblsem_release;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/include/core/class.h b/drivers/gpu/drm/nouveau/core/include/core/class.h
index 47c4b3a5bd3a..92d3ab11d962 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/class.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/class.h
@@ -154,6 +154,14 @@ struct nve0_channel_ind_class {
 	u32 engine;
 };
 
+/* 0046: NV04_DISP
+ */
+
+#define NV04_DISP_CLASS                                              0x00000046
+
+struct nv04_display_class {
+};
+
 /* 5070: NV50_DISP
  * 8270: NV84_DISP
  * 8370: NVA0_DISP
@@ -190,25 +198,6 @@ struct nve0_channel_ind_class {
 #define NV84_DISP_SOR_HDMI_PWR_REKEY                                 0x0000007f
 #define NV50_DISP_SOR_LVDS_SCRIPT                                    0x00013000
 #define NV50_DISP_SOR_LVDS_SCRIPT_ID                                 0x0000ffff
-#define NV94_DISP_SOR_DP_TRAIN                                       0x00016000
-#define NV94_DISP_SOR_DP_TRAIN_OP                                    0xf0000000
-#define NV94_DISP_SOR_DP_TRAIN_OP_PATTERN                            0x00000000
-#define NV94_DISP_SOR_DP_TRAIN_OP_INIT                               0x10000000
-#define NV94_DISP_SOR_DP_TRAIN_OP_FINI                               0x20000000
-#define NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD                           0x00000001
-#define NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD_OFF                       0x00000000
-#define NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD_ON                        0x00000001
-#define NV94_DISP_SOR_DP_TRAIN_PATTERN                               0x00000003
-#define NV94_DISP_SOR_DP_TRAIN_PATTERN_DISABLED                      0x00000000
-#define NV94_DISP_SOR_DP_LNKCTL                                      0x00016040
-#define NV94_DISP_SOR_DP_LNKCTL_FRAME                                0x80000000
-#define NV94_DISP_SOR_DP_LNKCTL_FRAME_STD                            0x00000000
-#define NV94_DISP_SOR_DP_LNKCTL_FRAME_ENH                            0x80000000
-#define NV94_DISP_SOR_DP_LNKCTL_WIDTH                                0x00001f00
-#define NV94_DISP_SOR_DP_LNKCTL_COUNT                                0x00000007
-#define NV94_DISP_SOR_DP_DRVCTL(l)                     ((l) * 0x40 + 0x00016100)
-#define NV94_DISP_SOR_DP_DRVCTL_VS                                   0x00000300
-#define NV94_DISP_SOR_DP_DRVCTL_PE                                   0x00000003
 
 #define NV50_DISP_DAC_MTHD                                           0x00020000
 #define NV50_DISP_DAC_MTHD_TYPE                                      0x0000f000
@@ -230,6 +219,23 @@ struct nve0_channel_ind_class {
 #define NV50_DISP_DAC_LOAD                                           0x0002000c
 #define NV50_DISP_DAC_LOAD_VALUE                                     0x00000007
 
+#define NV50_DISP_PIOR_MTHD                                          0x00030000
+#define NV50_DISP_PIOR_MTHD_TYPE                                     0x0000f000
+#define NV50_DISP_PIOR_MTHD_OR                                       0x00000003
+
+#define NV50_DISP_PIOR_PWR                                           0x00030000
+#define NV50_DISP_PIOR_PWR_STATE                                     0x00000001
+#define NV50_DISP_PIOR_PWR_STATE_ON                                  0x00000001
+#define NV50_DISP_PIOR_PWR_STATE_OFF                                 0x00000000
+#define NV50_DISP_PIOR_TMDS_PWR                                      0x00032000
+#define NV50_DISP_PIOR_TMDS_PWR_STATE                                0x00000001
+#define NV50_DISP_PIOR_TMDS_PWR_STATE_ON                             0x00000001
+#define NV50_DISP_PIOR_TMDS_PWR_STATE_OFF                            0x00000000
+#define NV50_DISP_PIOR_DP_PWR                                        0x00036000
+#define NV50_DISP_PIOR_DP_PWR_STATE                                  0x00000001
+#define NV50_DISP_PIOR_DP_PWR_STATE_ON                               0x00000001
+#define NV50_DISP_PIOR_DP_PWR_STATE_OFF                              0x00000000
+
 struct nv50_display_class {
 };
 
diff --git a/drivers/gpu/drm/nouveau/core/include/core/client.h b/drivers/gpu/drm/nouveau/core/include/core/client.h
index 63acc0346ff2..c66eac513803 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/client.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/client.h
@@ -7,7 +7,7 @@ struct nouveau_client {
 	struct nouveau_namedb base;
 	struct nouveau_handle *root;
 	struct nouveau_object *device;
-	char name[16];
+	char name[32];
 	u32 debug;
 	struct nouveau_vm *vm;
 };
@@ -41,5 +41,6 @@ int  nouveau_client_create_(const char *name, u64 device, const char *cfg,
 
 int  nouveau_client_init(struct nouveau_client *);
 int  nouveau_client_fini(struct nouveau_client *, bool suspend);
+const char *nouveau_client_name(void *obj);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/core/device.h b/drivers/gpu/drm/nouveau/core/include/core/device.h
index e58b6f0984c1..d351a4e5819c 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/device.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/device.h
@@ -26,6 +26,7 @@ enum nv_subdev_type {
 	 */
 	NVDEV_SUBDEV_MXM,
 	NVDEV_SUBDEV_MC,
+	NVDEV_SUBDEV_BUS,
 	NVDEV_SUBDEV_TIMER,
 	NVDEV_SUBDEV_FB,
 	NVDEV_SUBDEV_LTCG,
diff --git a/drivers/gpu/drm/nouveau/core/include/core/enum.h b/drivers/gpu/drm/nouveau/core/include/core/enum.h
index e7b1e181943b..4fc62bb8c1f0 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/enum.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/enum.h
@@ -5,12 +5,13 @@ struct nouveau_enum {
 	u32 value;
 	const char *name;
 	const void *data;
+	u32 data2;
 };
 
 const struct nouveau_enum *
 nouveau_enum_find(const struct nouveau_enum *, u32 value);
 
-void
+const struct nouveau_enum *
 nouveau_enum_print(const struct nouveau_enum *en, u32 value);
 
 struct nouveau_bitfield {
diff --git a/drivers/gpu/drm/nouveau/core/include/core/event.h b/drivers/gpu/drm/nouveau/core/include/core/event.h
new file mode 100644
index 000000000000..9e094408f14e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/core/event.h
@@ -0,0 +1,36 @@
+#ifndef __NVKM_EVENT_H__
+#define __NVKM_EVENT_H__
+
+/* return codes from event handlers */
+#define NVKM_EVENT_DROP 0
+#define NVKM_EVENT_KEEP 1
+
+struct nouveau_eventh {
+	struct list_head head;
+	int (*func)(struct nouveau_eventh *, int index);
+};
+
+struct nouveau_event {
+	spinlock_t lock;
+
+	void *priv;
+	void (*enable)(struct nouveau_event *, int index);
+	void (*disable)(struct nouveau_event *, int index);
+
+	int index_nr;
+	struct {
+		struct list_head list;
+		int refs;
+	} index[];
+};
+
+int  nouveau_event_create(int index_nr, struct nouveau_event **);
+void nouveau_event_destroy(struct nouveau_event **);
+void nouveau_event_trigger(struct nouveau_event *, int index);
+
+void nouveau_event_get(struct nouveau_event *, int index,
+		       struct nouveau_eventh *);
+void nouveau_event_put(struct nouveau_event *, int index,
+		       struct nouveau_eventh *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/core/object.h b/drivers/gpu/drm/nouveau/core/include/core/object.h
index 106bb19fdd9a..62e68baef087 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/object.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/object.h
@@ -136,7 +136,7 @@ static inline u8
 nv_ro08(void *obj, u64 addr)
 {
 	u8 data = nv_ofuncs(obj)->rd08(obj, addr);
-	nv_spam(obj, "nv_ro08 0x%08x 0x%02x\n", addr, data);
+	nv_spam(obj, "nv_ro08 0x%08llx 0x%02x\n", addr, data);
 	return data;
 }
 
@@ -144,7 +144,7 @@ static inline u16
 nv_ro16(void *obj, u64 addr)
 {
 	u16 data = nv_ofuncs(obj)->rd16(obj, addr);
-	nv_spam(obj, "nv_ro16 0x%08x 0x%04x\n", addr, data);
+	nv_spam(obj, "nv_ro16 0x%08llx 0x%04x\n", addr, data);
 	return data;
 }
 
@@ -152,28 +152,28 @@ static inline u32
 nv_ro32(void *obj, u64 addr)
 {
 	u32 data = nv_ofuncs(obj)->rd32(obj, addr);
-	nv_spam(obj, "nv_ro32 0x%08x 0x%08x\n", addr, data);
+	nv_spam(obj, "nv_ro32 0x%08llx 0x%08x\n", addr, data);
 	return data;
 }
 
 static inline void
 nv_wo08(void *obj, u64 addr, u8 data)
 {
-	nv_spam(obj, "nv_wo08 0x%08x 0x%02x\n", addr, data);
+	nv_spam(obj, "nv_wo08 0x%08llx 0x%02x\n", addr, data);
 	nv_ofuncs(obj)->wr08(obj, addr, data);
 }
 
 static inline void
 nv_wo16(void *obj, u64 addr, u16 data)
 {
-	nv_spam(obj, "nv_wo16 0x%08x 0x%04x\n", addr, data);
+	nv_spam(obj, "nv_wo16 0x%08llx 0x%04x\n", addr, data);
 	nv_ofuncs(obj)->wr16(obj, addr, data);
 }
 
 static inline void
 nv_wo32(void *obj, u64 addr, u32 data)
 {
-	nv_spam(obj, "nv_wo32 0x%08x 0x%08x\n", addr, data);
+	nv_spam(obj, "nv_wo32 0x%08llx 0x%08x\n", addr, data);
 	nv_ofuncs(obj)->wr32(obj, addr, data);
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/include/core/printk.h b/drivers/gpu/drm/nouveau/core/include/core/printk.h
index 1d629664f32d..febed2ea5c80 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/printk.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/printk.h
@@ -15,7 +15,8 @@ struct nouveau_object;
 #define NV_PRINTK_TRACE    KERN_DEBUG
 #define NV_PRINTK_SPAM     KERN_DEBUG
 
-void nv_printk_(struct nouveau_object *, const char *, int, const char *, ...);
+void __printf(4, 5)
+nv_printk_(struct nouveau_object *, const char *, int, const char *, ...);
 
 #define nv_printk(o,l,f,a...) do {                                             \
 	if (NV_DBG_##l <= CONFIG_NOUVEAU_DEBUG)                                \
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/disp.h b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
index 46948285f3e7..28da6772c095 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
@@ -4,18 +4,11 @@
 #include <core/object.h>
 #include <core/engine.h>
 #include <core/device.h>
+#include <core/event.h>
 
 struct nouveau_disp {
 	struct nouveau_engine base;
-
-	struct {
-		struct list_head list;
-		spinlock_t lock;
-		void (*notify)(void *, int);
-		void (*get)(void *, int);
-		void (*put)(void *, int);
-		void *data;
-	} vblank;
+	struct nouveau_event *vblank;
 };
 
 static inline struct nouveau_disp *
@@ -24,16 +17,22 @@ nouveau_disp(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_ENGINE_DISP];
 }
 
-#define nouveau_disp_create(p,e,c,i,x,d)                                       \
-	nouveau_engine_create((p), (e), (c), true, (i), (x), (d))
-#define nouveau_disp_destroy(d)                                                \
-	nouveau_engine_destroy(&(d)->base)
+#define nouveau_disp_create(p,e,c,h,i,x,d)                                     \
+	nouveau_disp_create_((p), (e), (c), (h), (i), (x),                     \
+			     sizeof(**d), (void **)d)
+#define nouveau_disp_destroy(d) ({                                             \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_dtor(nv_object(disp));                                   \
+})
 #define nouveau_disp_init(d)                                                   \
 	nouveau_engine_init(&(d)->base)
 #define nouveau_disp_fini(d,s)                                                 \
 	nouveau_engine_fini(&(d)->base, (s))
 
-#define _nouveau_disp_dtor _nouveau_engine_dtor
+int  nouveau_disp_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int heads,
+			  const char *, const char *, int, void **);
+void _nouveau_disp_dtor(struct nouveau_object *);
 #define _nouveau_disp_init _nouveau_engine_init
 #define _nouveau_disp_fini _nouveau_engine_fini
 
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
index f18846c8c6fe..b46c197709f3 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
@@ -65,6 +65,8 @@ struct nouveau_fifo_base {
 struct nouveau_fifo {
 	struct nouveau_engine base;
 
+	struct nouveau_event *uevent;
+
 	struct nouveau_object **channel;
 	spinlock_t lock;
 	u16 min;
@@ -92,6 +94,8 @@ int nouveau_fifo_create_(struct nouveau_object *, struct nouveau_object *,
 			 struct nouveau_oclass *, int min, int max,
 			 int size, void **);
 void nouveau_fifo_destroy(struct nouveau_fifo *);
+const char *
+nouveau_client_name_for_fifo_chid(struct nouveau_fifo *fifo, u32 chid);
 
 #define _nouveau_fifo_init _nouveau_engine_init
 #define _nouveau_fifo_fini _nouveau_engine_fini
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/software.h b/drivers/gpu/drm/nouveau/core/include/engine/software.h
index c945691c8564..45799487e573 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/software.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/software.h
@@ -3,17 +3,17 @@
 
 #include <core/engine.h>
 #include <core/engctx.h>
+#include <core/event.h>
 
 struct nouveau_software_chan {
 	struct nouveau_engctx base;
 
 	struct {
-		struct list_head head;
+		struct nouveau_eventh event;
 		u32 channel;
 		u32 ctxdma;
 		u64 offset;
 		u32 value;
-		u32 crtc;
 	} vblank;
 
 	int (*flip)(void *);
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dcb.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dcb.h
index b79025da581e..123270e9813a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dcb.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dcb.h
@@ -16,6 +16,8 @@ enum dcb_output_type {
 
 struct dcb_output {
 	int index;	/* may not be raw dcb index if merging has happened */
+	u16 hasht;
+	u16 hashm;
 	enum dcb_output_type type;
 	uint8_t i2c_index;
 	uint8_t heads;
@@ -25,6 +27,7 @@ struct dcb_output {
 	uint8_t or;
 	uint8_t link;
 	bool duallink_possible;
+	uint8_t extdev;
 	union {
 		struct sor_conf {
 			int link;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
index e6563b5cb08e..96d3364f6db3 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
@@ -1,17 +1,22 @@
 #ifndef __NVBIOS_GPIO_H__
 #define __NVBIOS_GPIO_H__
 
-struct nouveau_bios;
-
 enum dcb_gpio_func_name {
 	DCB_GPIO_PANEL_POWER = 0x01,
 	DCB_GPIO_TVDAC0 = 0x0c,
 	DCB_GPIO_TVDAC1 = 0x2d,
-	DCB_GPIO_PWM_FAN = 0x09,
+	DCB_GPIO_FAN = 0x09,
 	DCB_GPIO_FAN_SENSE = 0x3d,
 	DCB_GPIO_UNUSED = 0xff
 };
 
+#define DCB_GPIO_LOG_DIR     0x02
+#define DCB_GPIO_LOG_DIR_OUT 0x00
+#define DCB_GPIO_LOG_DIR_IN  0x02
+#define DCB_GPIO_LOG_VAL     0x01
+#define DCB_GPIO_LOG_VAL_LO  0x00
+#define DCB_GPIO_LOG_VAL_HI  0x01
+
 struct dcb_gpio_func {
 	u8 func;
 	u8 line;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/i2c.h
index 5079bedfd985..10b57a19a7de 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/i2c.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/i2c.h
@@ -15,7 +15,7 @@ struct dcb_i2c_entry {
 	enum dcb_i2c_type type;
 	u8 drive;
 	u8 sense;
-	u32 data;
+	u8 share;
 };
 
 u16 dcb_i2c_table(struct nouveau_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/therm.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/therm.h
index a2c4296fc5f6..083541dbe9c8 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/therm.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/therm.h
@@ -23,11 +23,27 @@ struct nvbios_therm_sensor {
 	struct nvbios_therm_threshold thrs_shutdown;
 };
 
+/* no vbios have more than 6 */
+#define NOUVEAU_TEMP_FAN_TRIP_MAX 10
+struct nouveau_therm_trip_point {
+	int fan_duty;
+	int temp;
+	int hysteresis;
+};
+
 struct nvbios_therm_fan {
 	u16 pwm_freq;
 
 	u8 min_duty;
 	u8 max_duty;
+
+	u16 bump_period;
+	u16 slow_down_period;
+
+	struct nouveau_therm_trip_point trip[NOUVEAU_TEMP_FAN_TRIP_MAX];
+	u8 nr_fan_trip;
+	u8 linear_min_temp;
+	u8 linear_max_temp;
 };
 
 enum nvbios_therm_domain {
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/xpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/xpio.h
new file mode 100644
index 000000000000..360baab52e4c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/xpio.h
@@ -0,0 +1,19 @@
+#ifndef __NVBIOS_XPIO_H__
+#define __NVBIOS_XPIO_H__
+
+#define NVBIOS_XPIO_FLAG_AUX  0x10
+#define NVBIOS_XPIO_FLAG_AUX0 0x00
+#define NVBIOS_XPIO_FLAG_AUX1 0x10
+
+struct nvbios_xpio {
+	u8 type;
+	u8 addr;
+	u8 flags;
+};
+
+u16 dcb_xpio_table(struct nouveau_bios *, u8 idx,
+		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 dcb_xpio_parse(struct nouveau_bios *, u8 idx,
+		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_xpio *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bus.h b/drivers/gpu/drm/nouveau/core/include/subdev/bus.h
new file mode 100644
index 000000000000..7d88ec4a6d06
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bus.h
@@ -0,0 +1,41 @@
+#ifndef __NOUVEAU_BUS_H__
+#define __NOUVEAU_BUS_H__
+
+#include <core/subdev.h>
+#include <core/device.h>
+
+struct nouveau_bus_intr {
+	u32 stat;
+	u32 unit;
+};
+
+struct nouveau_bus {
+	struct nouveau_subdev base;
+};
+
+static inline struct nouveau_bus *
+nouveau_bus(void *obj)
+{
+	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_BUS];
+}
+
+#define nouveau_bus_create(p, e, o, d)                                         \
+	nouveau_subdev_create_((p), (e), (o), 0, "PBUS", "master",             \
+			       sizeof(**d), (void **)d)
+#define nouveau_bus_destroy(p)                                                 \
+	nouveau_subdev_destroy(&(p)->base)
+#define nouveau_bus_init(p)                                                    \
+	nouveau_subdev_init(&(p)->base)
+#define nouveau_bus_fini(p, s)                                                 \
+	nouveau_subdev_fini(&(p)->base, (s))
+
+#define _nouveau_bus_dtor _nouveau_subdev_dtor
+#define _nouveau_bus_init _nouveau_subdev_init
+#define _nouveau_bus_fini _nouveau_subdev_fini
+
+extern struct nouveau_oclass nv04_bus_oclass;
+extern struct nouveau_oclass nv31_bus_oclass;
+extern struct nouveau_oclass nv50_bus_oclass;
+extern struct nouveau_oclass nvc0_bus_oclass;
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
index b75e8f18e52c..c85b9f1579ad 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
@@ -3,6 +3,7 @@
 
 #include <core/subdev.h>
 #include <core/device.h>
+#include <core/event.h>
 
 #include <subdev/bios.h>
 #include <subdev/bios/gpio.h>
@@ -10,28 +11,18 @@
 struct nouveau_gpio {
 	struct nouveau_subdev base;
 
+	struct nouveau_event *events;
+
 	/* hardware interfaces */
 	void (*reset)(struct nouveau_gpio *, u8 func);
 	int  (*drive)(struct nouveau_gpio *, int line, int dir, int out);
 	int  (*sense)(struct nouveau_gpio *, int line);
-	void (*irq_enable)(struct nouveau_gpio *, int line, bool);
 
 	/* software interfaces */
 	int  (*find)(struct nouveau_gpio *, int idx, u8 tag, u8 line,
 		     struct dcb_gpio_func *);
 	int  (*set)(struct nouveau_gpio *, int idx, u8 tag, u8 line, int state);
 	int  (*get)(struct nouveau_gpio *, int idx, u8 tag, u8 line);
-	int  (*irq)(struct nouveau_gpio *, int idx, u8 tag, u8 line, bool on);
-
-	/* interrupt handling */
-	struct list_head isr;
-	spinlock_t lock;
-
-	void (*isr_run)(struct nouveau_gpio *, int idx, u32 mask);
-	int  (*isr_add)(struct nouveau_gpio *, int idx, u8 tag, u8 line,
-			void (*)(void *, int state), void *data);
-	void (*isr_del)(struct nouveau_gpio *, int idx, u8 tag, u8 line,
-			void (*)(void *, int state), void *data);
 };
 
 static inline struct nouveau_gpio *
@@ -40,25 +31,23 @@ nouveau_gpio(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_GPIO];
 }
 
-#define nouveau_gpio_create(p,e,o,d)                                           \
-	nouveau_gpio_create_((p), (e), (o), sizeof(**d), (void **)d)
-#define nouveau_gpio_destroy(p)                                                \
-	nouveau_subdev_destroy(&(p)->base)
+#define nouveau_gpio_create(p,e,o,l,d)                                         \
+	nouveau_gpio_create_((p), (e), (o), (l), sizeof(**d), (void **)d)
+#define nouveau_gpio_destroy(p) ({                                             \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_dtor(nv_object(gpio));                                   \
+})
 #define nouveau_gpio_fini(p,s)                                                 \
 	nouveau_subdev_fini(&(p)->base, (s))
 
-int nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
-			 struct nouveau_oclass *, int, void **);
-int nouveau_gpio_init(struct nouveau_gpio *);
+int  nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int, int, void **);
+void _nouveau_gpio_dtor(struct nouveau_object *);
+int  nouveau_gpio_init(struct nouveau_gpio *);
 
 extern struct nouveau_oclass nv10_gpio_oclass;
 extern struct nouveau_oclass nv50_gpio_oclass;
 extern struct nouveau_oclass nvd0_gpio_oclass;
-
-void nv50_gpio_dtor(struct nouveau_object *);
-int  nv50_gpio_init(struct nouveau_object *);
-int  nv50_gpio_fini(struct nouveau_object *, bool);
-void nv50_gpio_intr(struct nouveau_subdev *);
-void nv50_gpio_irq_enable(struct nouveau_gpio *, int line, bool);
+extern struct nouveau_oclass nve0_gpio_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
index b93ab01e3785..888384c0bed8 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
@@ -10,23 +10,59 @@
 #define NV_I2C_PORT(n)    (0x00 + (n))
 #define NV_I2C_DEFAULT(n) (0x80 + (n))
 
+#define NV_I2C_TYPE_DCBI2C(n) (0x0000 | (n))
+#define NV_I2C_TYPE_EXTDDC(e) (0x0005 | (e) << 8)
+#define NV_I2C_TYPE_EXTAUX(e) (0x0006 | (e) << 8)
+
 struct nouveau_i2c_port {
+	struct nouveau_object base;
 	struct i2c_adapter adapter;
-	struct nouveau_i2c *i2c;
-	struct i2c_algo_bit_data bit;
+
 	struct list_head head;
 	u8  index;
-	u8  type;
-	u32 dcb;
-	u32 drive;
-	u32 sense;
-	u32 state;
+
+	const struct nouveau_i2c_func *func;
+};
+
+struct nouveau_i2c_func {
+	void (*acquire)(struct nouveau_i2c_port *);
+	void (*release)(struct nouveau_i2c_port *);
+
+	void (*drive_scl)(struct nouveau_i2c_port *, int);
+	void (*drive_sda)(struct nouveau_i2c_port *, int);
+	int  (*sense_scl)(struct nouveau_i2c_port *);
+	int  (*sense_sda)(struct nouveau_i2c_port *);
+
+	int  (*aux)(struct nouveau_i2c_port *, u8, u32, u8 *, u8);
+	int  (*pattern)(struct nouveau_i2c_port *, int pattern);
+	int  (*lnk_ctl)(struct nouveau_i2c_port *, int nr, int bw, bool enh);
+	int  (*drv_ctl)(struct nouveau_i2c_port *, int lane, int sw, int pe);
 };
 
+#define nouveau_i2c_port_create(p,e,o,i,a,d)                                   \
+	nouveau_i2c_port_create_((p), (e), (o), (i), (a),                      \
+				 sizeof(**d), (void **)d)
+#define nouveau_i2c_port_destroy(p) ({                                         \
+	struct nouveau_i2c_port *port = (p);                                   \
+	_nouveau_i2c_port_dtor(nv_object(i2c));                                \
+})
+#define nouveau_i2c_port_init(p)                                               \
+	nouveau_object_init(&(p)->base)
+#define nouveau_i2c_port_fini(p,s)                                             \
+	nouveau_object_fini(&(p)->base, (s))
+
+int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *,
+			     struct nouveau_oclass *, u8,
+			     const struct i2c_algorithm *, int, void **);
+void _nouveau_i2c_port_dtor(struct nouveau_object *);
+#define _nouveau_i2c_port_init nouveau_object_init
+#define _nouveau_i2c_port_fini nouveau_object_fini
+
 struct nouveau_i2c {
 	struct nouveau_subdev base;
 
 	struct nouveau_i2c_port *(*find)(struct nouveau_i2c *, u8 index);
+	struct nouveau_i2c_port *(*find_type)(struct nouveau_i2c *, u16 type);
 	int (*identify)(struct nouveau_i2c *, int index,
 			const char *what, struct i2c_board_info *,
 			bool (*match)(struct nouveau_i2c_port *,
@@ -40,21 +76,76 @@ nouveau_i2c(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_I2C];
 }
 
-extern struct nouveau_oclass nouveau_i2c_oclass;
+#define nouveau_i2c_create(p,e,o,s,d)                                          \
+	nouveau_i2c_create_((p), (e), (o), (s), sizeof(**d), (void **)d)
+#define nouveau_i2c_destroy(p) ({                                              \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_dtor(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_init(p) ({                                                 \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_init(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_fini(p,s) ({                                               \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_fini(nv_object(i2c), (s));                                \
+})
 
-void nouveau_i2c_drive_scl(void *, int);
-void nouveau_i2c_drive_sda(void *, int);
-int  nouveau_i2c_sense_scl(void *);
-int  nouveau_i2c_sense_sda(void *);
+int nouveau_i2c_create_(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, struct nouveau_oclass *,
+			int, void **);
+void _nouveau_i2c_dtor(struct nouveau_object *);
+int  _nouveau_i2c_init(struct nouveau_object *);
+int  _nouveau_i2c_fini(struct nouveau_object *, bool);
 
-int  nv_rdi2cr(struct nouveau_i2c_port *, u8 addr, u8 reg);
-int  nv_wri2cr(struct nouveau_i2c_port *, u8 addr, u8 reg, u8 val);
-bool nv_probe_i2c(struct nouveau_i2c_port *, u8 addr);
-
-int nv_rdaux(struct nouveau_i2c_port *, u32 addr, u8 *data, u8 size);
-int nv_wraux(struct nouveau_i2c_port *, u32 addr, u8 *data, u8 size);
+extern struct nouveau_oclass nv04_i2c_oclass;
+extern struct nouveau_oclass nv4e_i2c_oclass;
+extern struct nouveau_oclass nv50_i2c_oclass;
+extern struct nouveau_oclass nv94_i2c_oclass;
+extern struct nouveau_oclass nvd0_i2c_oclass;
+extern struct nouveau_oclass nouveau_anx9805_sclass[];
 
 extern const struct i2c_algorithm nouveau_i2c_bit_algo;
 extern const struct i2c_algorithm nouveau_i2c_aux_algo;
 
+static inline int
+nv_rdi2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg)
+{
+	u8 val;
+	struct i2c_msg msgs[] = {
+		{ .addr = addr, .flags = 0, .len = 1, .buf = &reg },
+		{ .addr = addr, .flags = I2C_M_RD, .len = 1, .buf = &val },
+	};
+
+	int ret = i2c_transfer(&port->adapter, msgs, 2);
+	if (ret != 2)
+		return -EIO;
+
+	return val;
+}
+
+static inline int
+nv_wri2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg, u8 val)
+{
+	u8 buf[2] = { reg, val };
+	struct i2c_msg msgs[] = {
+		{ .addr = addr, .flags = 0, .len = 2, .buf = buf },
+	};
+
+	int ret = i2c_transfer(&port->adapter, msgs, 1);
+	if (ret != 1)
+		return -EIO;
+
+	return 0;
+}
+
+static inline bool
+nv_probe_i2c(struct nouveau_i2c_port *port, u8 addr)
+{
+	return nv_rdi2cr(port, addr, 0) >= 0;
+}
+
+int nv_rdaux(struct nouveau_i2c_port *, u32 addr, u8 *data, u8 size);
+int nv_wraux(struct nouveau_i2c_port *, u32 addr, u8 *data, u8 size);
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/therm.h b/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
index faee569fd458..6b17b614629f 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
@@ -4,10 +4,10 @@
 #include <core/device.h>
 #include <core/subdev.h>
 
-enum nouveau_therm_fan_mode {
-	FAN_CONTROL_NONE = 0,
-	FAN_CONTROL_MANUAL = 1,
-	FAN_CONTROL_NR,
+enum nouveau_therm_mode {
+	NOUVEAU_THERM_CTRL_NONE = 0,
+	NOUVEAU_THERM_CTRL_MANUAL = 1,
+	NOUVEAU_THERM_CTRL_AUTO = 2,
 };
 
 enum nouveau_therm_attr_type {
@@ -28,6 +28,11 @@ enum nouveau_therm_attr_type {
 struct nouveau_therm {
 	struct nouveau_subdev base;
 
+	int (*pwm_ctrl)(struct nouveau_therm *, int line, bool);
+	int (*pwm_get)(struct nouveau_therm *, int line, u32 *, u32 *);
+	int (*pwm_set)(struct nouveau_therm *, int line, u32, u32);
+	int (*pwm_clock)(struct nouveau_therm *);
+
 	int (*fan_get)(struct nouveau_therm *);
 	int (*fan_set)(struct nouveau_therm *, int);
 	int (*fan_sense)(struct nouveau_therm *);
@@ -46,13 +51,29 @@ nouveau_therm(void *obj)
 }
 
 #define nouveau_therm_create(p,e,o,d)                                          \
-	nouveau_subdev_create((p), (e), (o), 0, "THERM", "therm", d)
-#define nouveau_therm_destroy(p)                                               \
-	nouveau_subdev_destroy(&(p)->base)
+	nouveau_therm_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_therm_destroy(p) ({                                            \
+	struct nouveau_therm *therm = (p);                                     \
+        _nouveau_therm_dtor(nv_object(therm));                                 \
+})
+#define nouveau_therm_init(p) ({                                               \
+	struct nouveau_therm *therm = (p);                                     \
+        _nouveau_therm_init(nv_object(therm));                                 \
+})
+#define nouveau_therm_fini(p,s) ({                                             \
+	struct nouveau_therm *therm = (p);                                     \
+        _nouveau_therm_init(nv_object(therm), (s));                            \
+})
 
-#define _nouveau_therm_dtor _nouveau_subdev_dtor
+int  nouveau_therm_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, int, void **);
+void _nouveau_therm_dtor(struct nouveau_object *);
+int  _nouveau_therm_init(struct nouveau_object *);
+int  _nouveau_therm_fini(struct nouveau_object *, bool);
 
 extern struct nouveau_oclass nv40_therm_oclass;
 extern struct nouveau_oclass nv50_therm_oclass;
+extern struct nouveau_oclass nva3_therm_oclass;
+extern struct nouveau_oclass nvd0_therm_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/timer.h b/drivers/gpu/drm/nouveau/core/include/subdev/timer.h
index c24ec8ab3db4..e465d158d352 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/timer.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/timer.h
@@ -10,6 +10,14 @@ struct nouveau_alarm {
 	void (*func)(struct nouveau_alarm *);
 };
 
+static inline void
+nouveau_alarm_init(struct nouveau_alarm *alarm,
+		   void (*func)(struct nouveau_alarm *))
+{
+	INIT_LIST_HEAD(&alarm->head);
+	alarm->func = func;
+}
+
 bool nouveau_timer_wait_eq(void *, u64 nsec, u32 addr, u32 mask, u32 data);
 bool nouveau_timer_wait_ne(void *, u64 nsec, u32 addr, u32 mask, u32 data);
 bool nouveau_timer_wait_cb(void *, u64 nsec, bool (*func)(void *), void *data);
diff --git a/drivers/gpu/drm/nouveau/core/os.h b/drivers/gpu/drm/nouveau/core/os.h
index cfe3b9cad156..eb496033b55c 100644
--- a/drivers/gpu/drm/nouveau/core/os.h
+++ b/drivers/gpu/drm/nouveau/core/os.h
@@ -16,6 +16,7 @@
 #include <linux/vmalloc.h>
 #include <linux/acpi.h>
 #include <linux/dmi.h>
+#include <linux/reboot.h>
 
 #include <asm/unaligned.h>
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
index f621f69fa1a2..e816f06637a7 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
@@ -172,7 +172,7 @@ out:
 	nv_wr32(bios, pcireg, access);
 }
 
-#if defined(CONFIG_ACPI)
+#if defined(CONFIG_ACPI) && defined(CONFIG_X86)
 int nouveau_acpi_get_bios_chunk(uint8_t *bios, int offset, int len);
 bool nouveau_acpi_rom_supported(struct pci_dev *pdev);
 #else
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/dcb.c b/drivers/gpu/drm/nouveau/core/subdev/bios/dcb.c
index 0fd87df99dd6..2d9b9d7a7992 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/dcb.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/dcb.c
@@ -107,6 +107,18 @@ dcb_outp(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len)
 	return 0x0000;
 }
 
+static inline u16
+dcb_outp_hasht(struct dcb_output *outp)
+{
+	return (outp->extdev << 8) | (outp->location << 4) | outp->type;
+}
+
+static inline u16
+dcb_outp_hashm(struct dcb_output *outp)
+{
+	return (outp->heads << 8) | (outp->link << 6) | outp->or;
+}
+
 u16
 dcb_outp_parse(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len,
 	       struct dcb_output *outp)
@@ -135,34 +147,28 @@ dcb_outp_parse(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len,
 			case DCB_OUTPUT_DP:
 				outp->link = (conf & 0x00000030) >> 4;
 				outp->sorconf.link = outp->link; /*XXX*/
+				outp->extdev = 0x00;
+				if (outp->location != 0)
+					outp->extdev = (conf & 0x0000ff00) >> 8;
 				break;
 			default:
 				break;
 			}
 		}
+
+		outp->hasht = dcb_outp_hasht(outp);
+		outp->hashm = dcb_outp_hashm(outp);
 	}
 	return dcb;
 }
 
-static inline u16
-dcb_outp_hasht(struct dcb_output *outp)
-{
-	return outp->type;
-}
-
-static inline u16
-dcb_outp_hashm(struct dcb_output *outp)
-{
-	return (outp->heads << 8) | (outp->link << 6) | outp->or;
-}
-
 u16
 dcb_outp_match(struct nouveau_bios *bios, u16 type, u16 mask,
 	       u8 *ver, u8 *len, struct dcb_output *outp)
 {
 	u16 dcb, idx = 0;
 	while ((dcb = dcb_outp_parse(bios, idx++, ver, len, outp))) {
-		if (dcb_outp_hasht(outp) == type) {
+		if ((dcb_outp_hasht(outp) & 0x00ff) == (type & 0x00ff)) {
 			if ((dcb_outp_hashm(outp) & mask) == mask)
 				break;
 		}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/extdev.c b/drivers/gpu/drm/nouveau/core/subdev/bios/extdev.c
index 5afb568b2d69..b2a676e53580 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/extdev.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/extdev.c
@@ -48,7 +48,7 @@ extdev_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *len, u8 *cnt)
 	return extdev + *hdr;
 }
 
-u16
+static u16
 nvbios_extdev_entry(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len)
 {
 	u8 hdr, cnt;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/gpio.c b/drivers/gpu/drm/nouveau/core/subdev/bios/gpio.c
index c84e93fa6d95..172a4f999990 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/gpio.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/gpio.c
@@ -25,6 +25,7 @@
 #include <subdev/bios.h>
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/gpio.h>
+#include <subdev/bios/xpio.h>
 
 u16
 dcb_gpio_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
@@ -60,8 +61,14 @@ dcb_gpio_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 u16
 dcb_gpio_entry(struct nouveau_bios *bios, int idx, int ent, u8 *ver, u8 *len)
 {
-	u8  hdr, cnt;
-	u16 gpio = !idx ? dcb_gpio_table(bios, ver, &hdr, &cnt, len) : 0x0000;
+	u8  hdr, cnt, xver; /* use gpio version for xpio entry parsing */
+	u16 gpio;
+
+	if (!idx--)
+		gpio = dcb_gpio_table(bios, ver, &hdr, &cnt, len);
+	else
+		gpio = dcb_xpio_table(bios, idx, &xver, &hdr, &cnt, len);
+
 	if (gpio && ent < cnt)
 		return gpio + hdr + (ent * *len);
 	return 0x0000;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/i2c.c b/drivers/gpu/drm/nouveau/core/subdev/bios/i2c.c
index ad577db83766..cfb9288c6d28 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/i2c.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/i2c.c
@@ -70,12 +70,12 @@ dcb_i2c_parse(struct nouveau_bios *bios, u8 idx, struct dcb_i2c_entry *info)
 	u8  ver, len;
 	u16 ent = dcb_i2c_entry(bios, idx, &ver, &len);
 	if (ent) {
-		info->data = nv_ro32(bios, ent + 0);
-		info->type = nv_ro08(bios, ent + 3);
+		info->type  = nv_ro08(bios, ent + 3);
+		info->share = DCB_I2C_UNUSED;
 		if (ver < 0x30) {
 			info->type &= 0x07;
 			if (info->type == 0x07)
-				info->type = 0xff;
+				info->type = DCB_I2C_UNUSED;
 		}
 
 		switch (info->type) {
@@ -88,7 +88,11 @@ dcb_i2c_parse(struct nouveau_bios *bios, u8 idx, struct dcb_i2c_entry *info)
 			return 0;
 		case DCB_I2C_NVIO_BIT:
 		case DCB_I2C_NVIO_AUX:
-			info->drive = nv_ro08(bios, ent + 0);
+			info->drive = nv_ro08(bios, ent + 0) & 0x0f;
+			if (nv_ro08(bios, ent + 1) & 0x01) {
+				info->share  = nv_ro08(bios, ent + 1) >> 1;
+				info->share &= 0x0f;
+			}
 			return 0;
 		case DCB_I2C_UNUSED:
 			return 0;
@@ -121,7 +125,8 @@ dcb_i2c_parse(struct nouveau_bios *bios, u8 idx, struct dcb_i2c_entry *info)
 			if (!info->sense) info->sense = 0x36;
 		}
 
-		info->type = DCB_I2C_NV04_BIT;
+		info->type  = DCB_I2C_NV04_BIT;
+		info->share = DCB_I2C_UNUSED;
 		return 0;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
index 690ed438b2ad..2cc1e6a5eb6a 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
@@ -231,6 +231,11 @@ init_i2c(struct nvbios_init *init, int index)
 			return NULL;
 		}
 
+		if (index == -2 && init->outp->location) {
+			index = NV_I2C_TYPE_EXTAUX(init->outp->extdev);
+			return i2c->find_type(i2c, index);
+		}
+
 		index = init->outp->i2c_index;
 	}
 
@@ -258,7 +263,7 @@ init_wri2cr(struct nvbios_init *init, u8 index, u8 addr, u8 reg, u8 val)
 static int
 init_rdauxr(struct nvbios_init *init, u32 addr)
 {
-	struct nouveau_i2c_port *port = init_i2c(init, -1);
+	struct nouveau_i2c_port *port = init_i2c(init, -2);
 	u8 data;
 
 	if (port && init_exec(init)) {
@@ -274,7 +279,7 @@ init_rdauxr(struct nvbios_init *init, u32 addr)
 static int
 init_wrauxr(struct nvbios_init *init, u32 addr, u8 data)
 {
-	struct nouveau_i2c_port *port = init_i2c(init, -1);
+	struct nouveau_i2c_port *port = init_i2c(init, -2);
 	if (port && init_exec(init))
 		return nv_wraux(port, addr, &data, 1);
 	return -ENODEV;
@@ -1816,7 +1821,7 @@ init_ram_restrict_zm_reg_group(struct nvbios_init *init)
 	u8 i, j;
 
 	trace("RAM_RESTRICT_ZM_REG_GROUP\t"
-	      "R[%08x] 0x%02x 0x%02x\n", addr, incr, num);
+	      "R[0x%08x] 0x%02x 0x%02x\n", addr, incr, num);
 	init->offset += 7;
 
 	for (i = 0; i < num; i++) {
@@ -1849,7 +1854,7 @@ init_copy_zm_reg(struct nvbios_init *init)
 	u32 sreg = nv_ro32(bios, init->offset + 1);
 	u32 dreg = nv_ro32(bios, init->offset + 5);
 
-	trace("COPY_ZM_REG\tR[0x%06x] = R[0x%06x]\n", sreg, dreg);
+	trace("COPY_ZM_REG\tR[0x%06x] = R[0x%06x]\n", dreg, sreg);
 	init->offset += 9;
 
 	init_wr32(init, dreg, init_rd32(init, sreg));
@@ -1866,7 +1871,7 @@ init_zm_reg_group(struct nvbios_init *init)
 	u32 addr = nv_ro32(bios, init->offset + 1);
 	u8 count = nv_ro08(bios, init->offset + 5);
 
-	trace("ZM_REG_GROUP\tR[0x%06x] =\n");
+	trace("ZM_REG_GROUP\tR[0x%06x] =\n", addr);
 	init->offset += 6;
 
 	while (count--) {
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c b/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c
index 862a08a2ae27..22a20573ed1b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/therm.c
@@ -55,7 +55,7 @@ therm_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *len, u8 *cnt)
 	return therm + nv_ro08(bios, therm + 1);
 }
 
-u16
+static u16
 nvbios_therm_entry(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len)
 {
 	u8 hdr, cnt;
@@ -155,10 +155,15 @@ int
 nvbios_therm_fan_parse(struct nouveau_bios *bios,
 			  struct nvbios_therm_fan *fan)
 {
+	struct nouveau_therm_trip_point *cur_trip = NULL;
 	u8 ver, len, i;
 	u16 entry;
 
+	uint8_t duty_lut[] = { 0, 0, 25, 0, 40, 0, 50, 0,
+				75, 0, 85, 0, 100, 0, 100, 0 };
+
 	i = 0;
+	fan->nr_fan_trip = 0;
 	while ((entry = nvbios_therm_entry(bios, i++, &ver, &len))) {
 		s16 value = nv_ro16(bios, entry + 1);
 
@@ -167,9 +172,30 @@ nvbios_therm_fan_parse(struct nouveau_bios *bios,
 			fan->min_duty = value & 0xff;
 			fan->max_duty = (value & 0xff00) >> 8;
 			break;
+		case 0x24:
+			fan->nr_fan_trip++;
+			cur_trip = &fan->trip[fan->nr_fan_trip - 1];
+			cur_trip->hysteresis = value & 0xf;
+			cur_trip->temp = (value & 0xff0) >> 4;
+			cur_trip->fan_duty = duty_lut[(value & 0xf000) >> 12];
+			break;
+		case 0x25:
+			cur_trip = &fan->trip[fan->nr_fan_trip - 1];
+			cur_trip->fan_duty = value;
+			break;
 		case 0x26:
 			fan->pwm_freq = value;
 			break;
+		case 0x3b:
+			fan->bump_period = value;
+			break;
+		case 0x3c:
+			fan->slow_down_period = value;
+			break;
+		case 0x46:
+			fan->linear_min_temp = nv_ro08(bios, entry + 1);
+			fan->linear_max_temp = nv_ro08(bios, entry + 2);
+			break;
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/xpio.c b/drivers/gpu/drm/nouveau/core/subdev/bios/xpio.c
new file mode 100644
index 000000000000..e9b8e5d30a7a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/xpio.c
@@ -0,0 +1,76 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/gpio.h>
+#include <subdev/bios/xpio.h>
+
+static u16
+dcb_xpiod_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	u16 data = dcb_gpio_table(bios, ver, hdr, cnt, len);
+	if (data && *ver >= 0x40 && *hdr >= 0x06) {
+		u16 xpio = nv_ro16(bios, data + 0x04);
+		if (xpio) {
+			*ver = nv_ro08(bios, data + 0x00);
+			*hdr = nv_ro08(bios, data + 0x01);
+			*cnt = nv_ro08(bios, data + 0x02);
+			*len = nv_ro08(bios, data + 0x03);
+			return xpio;
+		}
+	}
+	return 0x0000;
+}
+
+u16
+dcb_xpio_table(struct nouveau_bios *bios, u8 idx,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	u16 data = dcb_xpiod_table(bios, ver, hdr, cnt, len);
+	if (data && idx < *cnt) {
+		u16 xpio = nv_ro16(bios, data + *hdr + (idx * *len));
+		if (xpio) {
+			*ver = nv_ro08(bios, data + 0x00);
+			*hdr = nv_ro08(bios, data + 0x01);
+			*cnt = nv_ro08(bios, data + 0x02);
+			*len = nv_ro08(bios, data + 0x03);
+			return xpio;
+		}
+	}
+	return 0x0000;
+}
+
+u16
+dcb_xpio_parse(struct nouveau_bios *bios, u8 idx,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+	       struct nvbios_xpio *info)
+{
+	u16 data = dcb_xpio_table(bios, idx, ver, hdr, cnt, len);
+	if (data && *len >= 6) {
+		info->type = nv_ro08(bios, data + 0x04);
+		info->addr = nv_ro08(bios, data + 0x05);
+		info->flags = nv_ro08(bios, data + 0x06);
+	}
+	return 0x0000;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c
new file mode 100644
index 000000000000..8c7f8057a185
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c
@@ -0,0 +1,95 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres <martin.peres@labri.fr>
+ *          Ben Skeggs
+ */
+
+#include <subdev/bus.h>
+
+struct nv04_bus_priv {
+	struct nouveau_bus base;
+};
+
+static void
+nv04_bus_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_bus *pbus = nouveau_bus(subdev);
+	u32 stat = nv_rd32(pbus, 0x001100) & nv_rd32(pbus, 0x001140);
+
+	if (stat & 0x00000001) {
+		nv_error(pbus, "BUS ERROR\n");
+		stat &= ~0x00000001;
+		nv_wr32(pbus, 0x001100, 0x00000001);
+	}
+
+	if (stat & 0x00000110) {
+		subdev = nouveau_subdev(subdev, NVDEV_SUBDEV_GPIO);
+		if (subdev && subdev->intr)
+			subdev->intr(subdev);
+		stat &= ~0x00000110;
+		nv_wr32(pbus, 0x001100, 0x00000110);
+	}
+
+	if (stat) {
+		nv_error(pbus, "unknown intr 0x%08x\n", stat);
+		nv_mask(pbus, 0x001140, stat, 0x00000000);
+	}
+}
+
+static int
+nv04_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv04_bus_priv *priv;
+	int ret;
+
+	ret = nouveau_bus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = nv04_bus_intr;
+	return 0;
+}
+
+static int
+nv04_bus_init(struct nouveau_object *object)
+{
+	struct nv04_bus_priv *priv = (void *)object;
+
+	nv_wr32(priv, 0x001100, 0xffffffff);
+	nv_wr32(priv, 0x001140, 0x00000111);
+
+	return nouveau_bus_init(&priv->base);
+}
+
+struct nouveau_oclass
+nv04_bus_oclass = {
+	.handle = NV_SUBDEV(BUS, 0x04),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_bus_ctor,
+		.dtor = _nouveau_bus_dtor,
+		.init = nv04_bus_init,
+		.fini = _nouveau_bus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c
new file mode 100644
index 000000000000..34132aef34e1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c
@@ -0,0 +1,112 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres <martin.peres@labri.fr>
+ *          Ben Skeggs
+ */
+
+#include <subdev/bus.h>
+
+struct nv31_bus_priv {
+	struct nouveau_bus base;
+};
+
+static void
+nv31_bus_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_bus *pbus = nouveau_bus(subdev);
+	u32 stat = nv_rd32(pbus, 0x001100) & nv_rd32(pbus, 0x001140);
+	u32 gpio = nv_rd32(pbus, 0x001104) & nv_rd32(pbus, 0x001144);
+
+	if (gpio) {
+		subdev = nouveau_subdev(pbus, NVDEV_SUBDEV_GPIO);
+		if (subdev && subdev->intr)
+			subdev->intr(subdev);
+	}
+
+	if (stat & 0x00000008) {  /* NV41- */
+		u32 addr = nv_rd32(pbus, 0x009084);
+		u32 data = nv_rd32(pbus, 0x009088);
+
+		nv_error(pbus, "MMIO %s of 0x%08x FAULT at 0x%06x\n",
+			 (addr & 0x00000002) ? "write" : "read", data,
+			 (addr & 0x00fffffc));
+
+		stat &= ~0x00000008;
+		nv_wr32(pbus, 0x001100, 0x00000008);
+	}
+
+	if (stat & 0x00070000) {
+		subdev = nouveau_subdev(pbus, NVDEV_SUBDEV_THERM);
+		if (subdev && subdev->intr)
+			subdev->intr(subdev);
+		stat &= ~0x00070000;
+		nv_wr32(pbus, 0x001100, 0x00070000);
+	}
+
+	if (stat) {
+		nv_error(pbus, "unknown intr 0x%08x\n", stat);
+		nv_mask(pbus, 0x001140, stat, 0x00000000);
+	}
+}
+
+static int
+nv31_bus_init(struct nouveau_object *object)
+{
+	struct nv31_bus_priv *priv = (void *)object;
+	int ret;
+
+	ret = nouveau_bus_init(&priv->base);
+	if (ret)
+		return ret;
+
+	nv_wr32(priv, 0x001100, 0xffffffff);
+	nv_wr32(priv, 0x001140, 0x00070008);
+	return 0;
+}
+
+static int
+nv31_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv31_bus_priv *priv;
+	int ret;
+
+	ret = nouveau_bus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = nv31_bus_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+nv31_bus_oclass = {
+	.handle = NV_SUBDEV(BUS, 0x31),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv31_bus_ctor,
+		.dtor = _nouveau_bus_dtor,
+		.init = nv31_bus_init,
+		.fini = _nouveau_bus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c
new file mode 100644
index 000000000000..f5b2117fa8c6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c
@@ -0,0 +1,105 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres <martin.peres@labri.fr>
+ *          Ben Skeggs
+ */
+
+#include <subdev/bus.h>
+
+struct nv50_bus_priv {
+	struct nouveau_bus base;
+};
+
+static void
+nv50_bus_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_bus *pbus = nouveau_bus(subdev);
+	u32 stat = nv_rd32(pbus, 0x001100) & nv_rd32(pbus, 0x001140);
+
+	if (stat & 0x00000008) {
+		u32 addr = nv_rd32(pbus, 0x009084);
+		u32 data = nv_rd32(pbus, 0x009088);
+
+		nv_error(pbus, "MMIO %s of 0x%08x FAULT at 0x%06x\n",
+			 (addr & 0x00000002) ? "write" : "read", data,
+			 (addr & 0x00fffffc));
+
+		stat &= ~0x00000008;
+		nv_wr32(pbus, 0x001100, 0x00000008);
+	}
+
+	if (stat & 0x00010000) {
+		subdev = nouveau_subdev(pbus, NVDEV_SUBDEV_THERM);
+		if (subdev && subdev->intr)
+			subdev->intr(subdev);
+		stat &= ~0x00010000;
+		nv_wr32(pbus, 0x001100, 0x00010000);
+	}
+
+	if (stat) {
+		nv_error(pbus, "unknown intr 0x%08x\n", stat);
+		nv_mask(pbus, 0x001140, stat, 0);
+	}
+}
+
+static int
+nv50_bus_init(struct nouveau_object *object)
+{
+	struct nv50_bus_priv *priv = (void *)object;
+	int ret;
+
+	ret = nouveau_bus_init(&priv->base);
+	if (ret)
+		return ret;
+
+	nv_wr32(priv, 0x001100, 0xffffffff);
+	nv_wr32(priv, 0x001140, 0x00010008);
+	return 0;
+}
+
+static int
+nv50_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv50_bus_priv *priv;
+	int ret;
+
+	ret = nouveau_bus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = nv50_bus_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+nv50_bus_oclass = {
+	.handle = NV_SUBDEV(BUS, 0x50),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_bus_ctor,
+		.dtor = _nouveau_bus_dtor,
+		.init = nv50_bus_init,
+		.fini = _nouveau_bus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c
new file mode 100644
index 000000000000..b192d6246363
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c
@@ -0,0 +1,101 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres <martin.peres@labri.fr>
+ *          Ben Skeggs
+ */
+
+#include <subdev/bus.h>
+
+struct nvc0_bus_priv {
+	struct nouveau_bus base;
+};
+
+static void
+nvc0_bus_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_bus *pbus = nouveau_bus(subdev);
+	u32 stat = nv_rd32(pbus, 0x001100) & nv_rd32(pbus, 0x001140);
+
+	if (stat & 0x0000000e) {
+		u32 addr = nv_rd32(pbus, 0x009084);
+		u32 data = nv_rd32(pbus, 0x009088);
+
+		nv_error(pbus, "MMIO %s of 0x%08x FAULT at 0x%06x [ %s%s%s]\n",
+			 (addr & 0x00000002) ? "write" : "read", data,
+			 (addr & 0x00fffffc),
+			 (stat & 0x00000002) ? "!ENGINE " : "",
+			 (stat & 0x00000004) ? "IBUS " : "",
+			 (stat & 0x00000008) ? "TIMEOUT " : "");
+
+		nv_wr32(pbus, 0x009084, 0x00000000);
+		nv_wr32(pbus, 0x001100, (stat & 0x0000000e));
+		stat &= ~0x0000000e;
+	}
+
+	if (stat) {
+		nv_error(pbus, "unknown intr 0x%08x\n", stat);
+		nv_mask(pbus, 0x001140, stat, 0x00000000);
+	}
+}
+
+static int
+nvc0_bus_init(struct nouveau_object *object)
+{
+	struct nvc0_bus_priv *priv = (void *)object;
+	int ret;
+
+	ret = nouveau_bus_init(&priv->base);
+	if (ret)
+		return ret;
+
+	nv_wr32(priv, 0x001100, 0xffffffff);
+	nv_wr32(priv, 0x001140, 0x0000000e);
+	return 0;
+}
+
+static int
+nvc0_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nvc0_bus_priv *priv;
+	int ret;
+
+	ret = nouveau_bus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = nvc0_bus_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+nvc0_bus_oclass = {
+	.handle = NV_SUBDEV(BUS, 0xc0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_bus_ctor,
+		.dtor = _nouveau_bus_dtor,
+		.init = nvc0_bus_init,
+		.fini = _nouveau_bus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/base.c b/drivers/gpu/drm/nouveau/core/subdev/device/base.c
index f8a7ed4166cf..3937ced5c753 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/base.c
@@ -66,6 +66,7 @@ static const u64 disable_map[] = {
 	[NVDEV_SUBDEV_CLOCK]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_MXM]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_MC]	= NV_DEVICE_DISABLE_CORE,
+	[NVDEV_SUBDEV_BUS]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_TIMER]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_FB]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_LTCG]	= NV_DEVICE_DISABLE_CORE,
@@ -103,8 +104,8 @@ nouveau_devobj_ctor(struct nouveau_object *parent,
 	struct nouveau_device *device;
 	struct nouveau_devobj *devobj;
 	struct nv_device_class *args = data;
-	u64 disable, boot0, strap;
-	u64 mmio_base, mmio_size;
+	u32 boot0, strap;
+	u64 disable, mmio_base, mmio_size;
 	void __iomem *map;
 	int ret, i, c;
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv04.c
index 8626d0d6cbbc..473c5c03d3c9 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv04.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
 #include <subdev/devinit.h>
@@ -46,10 +47,11 @@ nv04_identify(struct nouveau_device *device)
 	case 0x04:
 		device->cname = "NV04";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv04_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv04_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -63,10 +65,11 @@ nv04_identify(struct nouveau_device *device)
 	case 0x05:
 		device->cname = "NV05";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv05_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv04_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv10.c
index 9c40b0fb23f6..d0774f5bebe1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv10.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -48,10 +49,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV10";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -64,10 +66,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV15";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -82,10 +85,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV16";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -100,10 +104,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "nForce";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv1a_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -118,10 +123,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV11";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -136,10 +142,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV17";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -154,10 +161,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "nForce2";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv1a_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -172,10 +180,11 @@ nv10_identify(struct nouveau_device *device)
 		device->cname = "NV18";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv20.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv20.c
index 74f88f48e1c2..ab920e0dc45b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv20.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -49,10 +50,11 @@ nv20_identify(struct nouveau_device *device)
 		device->cname = "NV20";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv20_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -67,10 +69,11 @@ nv20_identify(struct nouveau_device *device)
 		device->cname = "NV25";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -85,10 +88,11 @@ nv20_identify(struct nouveau_device *device)
 		device->cname = "NV28";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -103,10 +107,11 @@ nv20_identify(struct nouveau_device *device)
 		device->cname = "NV2A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv30.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv30.c
index 0ac1b2c4f61d..5f2110261b04 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv30.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv30.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -49,10 +50,11 @@ nv30_identify(struct nouveau_device *device)
 		device->cname = "NV30";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv30_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -67,10 +69,11 @@ nv30_identify(struct nouveau_device *device)
 		device->cname = "NV35";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv35_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -85,10 +88,11 @@ nv30_identify(struct nouveau_device *device)
 		device->cname = "NV31";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv30_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -104,10 +108,11 @@ nv30_identify(struct nouveau_device *device)
 		device->cname = "NV36";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv36_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
@@ -123,10 +128,11 @@ nv30_identify(struct nouveau_device *device)
 		device->cname = "NV34";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv40.c
index 41d59689a021..f3d55efe9ac9 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv40.c
@@ -24,6 +24,8 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
+#include <subdev/vm.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -50,11 +52,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV40";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv40_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -70,11 +73,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV41";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -90,11 +94,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV42";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -110,11 +115,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV43";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -130,11 +136,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV45";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv40_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -150,11 +157,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "G70";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv47_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -170,11 +178,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "G71";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv49_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -190,11 +199,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "G73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv49_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -210,11 +220,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV44";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv44_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -230,11 +241,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "G72";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -250,11 +262,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "NV44A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv44_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -270,11 +283,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "C61";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -290,11 +304,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "C51";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv4e_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv4e_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -310,11 +325,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "C73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -330,11 +346,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "C67";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
@@ -350,11 +367,12 @@ nv40_identify(struct nouveau_device *device)
 		device->cname = "C68";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/device/nv50.c
index 6ccfd8585ba2..5ed2fa51ddc2 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nv50.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -57,12 +58,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G80";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -79,12 +81,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G84";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -104,12 +107,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G86";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -129,12 +133,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G92";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -154,12 +159,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G94";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -179,12 +185,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G96";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -204,12 +211,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G98";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -229,12 +237,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "G200";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -254,12 +263,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "MCP77/MCP78";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -279,12 +289,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "MCP79/MCP7A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -304,12 +315,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "GT215";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -330,12 +342,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "GT216";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -355,12 +368,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "GT218";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
@@ -380,12 +394,13 @@ nv50_identify(struct nouveau_device *device)
 		device->cname = "MCP89";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/device/nvc0.c
index f0461685a422..4393eb4d6564 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nvc0.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -57,12 +58,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF100";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -85,12 +87,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -113,12 +116,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -141,12 +145,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF114";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -169,12 +174,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF116";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -197,12 +203,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF108";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -225,12 +232,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -253,12 +261,13 @@ nvc0_identify(struct nouveau_device *device)
 		device->cname = "GF119";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -282,4 +291,4 @@ nvc0_identify(struct nouveau_device *device)
 	}
 
 	return 0;
-}
+	}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/device/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/device/nve0.c
index 03a652876e73..5c12391619fd 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/device/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/device/nve0.c
@@ -24,6 +24,7 @@
 
 #include <subdev/device.h>
 #include <subdev/bios.h>
+#include <subdev/bus.h>
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/clock.h>
@@ -56,13 +57,14 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe4:
 		device->cname = "GK104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -84,13 +86,14 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe7:
 		device->cname = "GK107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
@@ -112,13 +115,14 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe6:
 		device->cname = "GK106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nouveau_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
-		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv50.c
index ae7249b09797..4a8577838417 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv50.c
@@ -78,12 +78,13 @@ nv50_devinit_init(struct nouveau_object *object)
 	if (ret)
 		return ret;
 
-	/* if we ran the init tables, execute first script pointer for each
-	 * display table output entry that has a matching dcb entry.
+	/* if we ran the init tables, we have to execute the first script
+	 * pointer of each dcb entry's display encoder table in order
+	 * to properly initialise each encoder.
 	 */
-	while (priv->base.post && ver) {
-		u16 data = nvbios_outp_parse(bios, i++, &ver, &hdr, &cnt, &len, &info);
-		if (data && dcb_outp_match(bios, info.type, info.mask, &ver, &len, &outp)) {
+	while (priv->base.post && dcb_outp_parse(bios, i, &ver, &hdr, &outp)) {
+		if (nvbios_outp_match(bios, outp.hasht, outp.hashm,
+				     &ver, &hdr, &cnt, &len, &info)) {
 			struct nvbios_init init = {
 				.subdev = nv_subdev(priv),
 				.bios = bios,
@@ -95,7 +96,8 @@ nv50_devinit_init(struct nouveau_object *object)
 
 			nvbios_exec(&init);
 		}
-	};
+		i++;
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
index eac236ed19b2..0772ec978165 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
@@ -22,8 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include <core/object.h>
+#include <core/client.h>
 #include <core/enum.h>
+#include <core/engctx.h>
+#include <core/object.h>
 
 #include <subdev/fb.h>
 #include <subdev/bios.h>
@@ -303,17 +305,18 @@ static const struct nouveau_enum vm_client[] = {
 };
 
 static const struct nouveau_enum vm_engine[] = {
-	{ 0x00000000, "PGRAPH", NULL },
-	{ 0x00000001, "PVP", NULL },
+	{ 0x00000000, "PGRAPH", NULL, NVDEV_ENGINE_GR },
+	{ 0x00000001, "PVP", NULL, NVDEV_ENGINE_VP },
 	{ 0x00000004, "PEEPHOLE", NULL },
-	{ 0x00000005, "PFIFO", vm_pfifo_subclients },
+	{ 0x00000005, "PFIFO", vm_pfifo_subclients, NVDEV_ENGINE_FIFO },
 	{ 0x00000006, "BAR", vm_bar_subclients },
-	{ 0x00000008, "PPPP", NULL },
-	{ 0x00000009, "PBSP", NULL },
-	{ 0x0000000a, "PCRYPT", NULL },
+	{ 0x00000008, "PPPP", NULL, NVDEV_ENGINE_PPP },
+	{ 0x00000008, "PMPEG", NULL, NVDEV_ENGINE_MPEG },
+	{ 0x00000009, "PBSP", NULL, NVDEV_ENGINE_BSP },
+	{ 0x0000000a, "PCRYPT", NULL, NVDEV_ENGINE_CRYPT },
 	{ 0x0000000b, "PCOUNTER", NULL },
 	{ 0x0000000c, "SEMAPHORE_BG", NULL },
-	{ 0x0000000d, "PCOPY", NULL },
+	{ 0x0000000d, "PCOPY", NULL, NVDEV_ENGINE_COPY0 },
 	{ 0x0000000e, "PDAEMON", NULL },
 	{}
 };
@@ -335,8 +338,10 @@ static void
 nv50_fb_intr(struct nouveau_subdev *subdev)
 {
 	struct nouveau_device *device = nv_device(subdev);
+	struct nouveau_engine *engine;
 	struct nv50_fb_priv *priv = (void *)subdev;
 	const struct nouveau_enum *en, *cl;
+	struct nouveau_object *engctx = NULL;
 	u32 trap[6], idx, chan;
 	u8 st0, st1, st2, st3;
 	int i;
@@ -367,36 +372,55 @@ nv50_fb_intr(struct nouveau_subdev *subdev)
 	}
 	chan = (trap[2] << 16) | trap[1];
 
-	nv_error(priv, "trapped %s at 0x%02x%04x%04x on channel 0x%08x ",
+	en = nouveau_enum_find(vm_engine, st0);
+
+	if (en && en->data2) {
+		const struct nouveau_enum *orig_en = en;
+		while (en->name && en->value == st0 && en->data2) {
+			engine = nouveau_engine(subdev, en->data2);
+			if (engine) {
+				engctx = nouveau_engctx_get(engine, chan);
+				if (engctx)
+					break;
+			}
+			en++;
+		}
+		if (!engctx)
+			en = orig_en;
+	}
+
+	nv_error(priv, "trapped %s at 0x%02x%04x%04x on channel 0x%08x [%s] ",
 		 (trap[5] & 0x00000100) ? "read" : "write",
-		 trap[5] & 0xff, trap[4] & 0xffff, trap[3] & 0xffff, chan);
+		 trap[5] & 0xff, trap[4] & 0xffff, trap[3] & 0xffff, chan,
+		 nouveau_client_name(engctx));
+
+	nouveau_engctx_put(engctx);
 
-	en = nouveau_enum_find(vm_engine, st0);
 	if (en)
-		printk("%s/", en->name);
+		pr_cont("%s/", en->name);
 	else
-		printk("%02x/", st0);
+		pr_cont("%02x/", st0);
 
 	cl = nouveau_enum_find(vm_client, st2);
 	if (cl)
-		printk("%s/", cl->name);
+		pr_cont("%s/", cl->name);
 	else
-		printk("%02x/", st2);
+		pr_cont("%02x/", st2);
 
 	if      (cl && cl->data) cl = nouveau_enum_find(cl->data, st3);
 	else if (en && en->data) cl = nouveau_enum_find(en->data, st3);
 	else                     cl = NULL;
 	if (cl)
-		printk("%s", cl->name);
+		pr_cont("%s", cl->name);
 	else
-		printk("%02x", st3);
+		pr_cont("%02x", st3);
 
-	printk(" reason: ");
+	pr_cont(" reason: ");
 	en = nouveau_enum_find(vm_fault, st1);
 	if (en)
-		printk("%s\n", en->name);
+		pr_cont("%s\n", en->name);
 	else
-		printk("0x%08x\n", st1);
+		pr_cont("0x%08x\n", st1);
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
index 9fb0f9b92d49..d422acc9af15 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
@@ -102,135 +102,19 @@ nouveau_gpio_get(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line)
 	return ret;
 }
 
-static int
-nouveau_gpio_irq(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line, bool on)
-{
-	struct dcb_gpio_func func;
-	int ret;
-
-	ret = nouveau_gpio_find(gpio, idx, tag, line, &func);
-	if (ret == 0) {
-		if (idx == 0 && gpio->irq_enable)
-			gpio->irq_enable(gpio, func.line, on);
-		else
-			ret = -ENODEV;
-	}
-
-	return ret;
-}
-
-struct gpio_isr {
-	struct nouveau_gpio *gpio;
-	struct list_head head;
-	struct work_struct work;
-	int idx;
-	struct dcb_gpio_func func;
-	void (*handler)(void *, int);
-	void *data;
-	bool inhibit;
-};
-
-static void
-nouveau_gpio_isr_bh(struct work_struct *work)
-{
-	struct gpio_isr *isr = container_of(work, struct gpio_isr, work);
-	struct nouveau_gpio *gpio = isr->gpio;
-	unsigned long flags;
-	int state;
-
-	state = nouveau_gpio_get(gpio, isr->idx, isr->func.func,
-						 isr->func.line);
-	if (state >= 0)
-		isr->handler(isr->data, state);
-
-	spin_lock_irqsave(&gpio->lock, flags);
-	isr->inhibit = false;
-	spin_unlock_irqrestore(&gpio->lock, flags);
-}
-
-static void
-nouveau_gpio_isr_run(struct nouveau_gpio *gpio, int idx, u32 line_mask)
-{
-	struct gpio_isr *isr;
-
-	if (idx != 0)
-		return;
-
-	spin_lock(&gpio->lock);
-	list_for_each_entry(isr, &gpio->isr, head) {
-		if (line_mask & (1 << isr->func.line)) {
-			if (isr->inhibit)
-				continue;
-			isr->inhibit = true;
-			schedule_work(&isr->work);
-		}
-	}
-	spin_unlock(&gpio->lock);
-}
-
-static int
-nouveau_gpio_isr_add(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line,
-		     void (*handler)(void *, int), void *data)
-{
-	struct gpio_isr *isr;
-	unsigned long flags;
-	int ret;
-
-	isr = kzalloc(sizeof(*isr), GFP_KERNEL);
-	if (!isr)
-		return -ENOMEM;
-
-	ret = nouveau_gpio_find(gpio, idx, tag, line, &isr->func);
-	if (ret) {
-		kfree(isr);
-		return ret;
-	}
-
-	INIT_WORK(&isr->work, nouveau_gpio_isr_bh);
-	isr->gpio = gpio;
-	isr->handler = handler;
-	isr->data = data;
-	isr->idx = idx;
-
-	spin_lock_irqsave(&gpio->lock, flags);
-	list_add(&isr->head, &gpio->isr);
-	spin_unlock_irqrestore(&gpio->lock, flags);
-	return 0;
-}
-
-static void
-nouveau_gpio_isr_del(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line,
-		     void (*handler)(void *, int), void *data)
+void
+_nouveau_gpio_dtor(struct nouveau_object *object)
 {
-	struct gpio_isr *isr, *tmp;
-	struct dcb_gpio_func func;
-	unsigned long flags;
-	LIST_HEAD(tofree);
-	int ret;
-
-	ret = nouveau_gpio_find(gpio, idx, tag, line, &func);
-	if (ret == 0) {
-		spin_lock_irqsave(&gpio->lock, flags);
-		list_for_each_entry_safe(isr, tmp, &gpio->isr, head) {
-			if (memcmp(&isr->func, &func, sizeof(func)) ||
-			    isr->idx != idx ||
-			    isr->handler != handler || isr->data != data)
-				continue;
-			list_move_tail(&isr->head, &tofree);
-		}
-		spin_unlock_irqrestore(&gpio->lock, flags);
-
-		list_for_each_entry_safe(isr, tmp, &tofree, head) {
-			flush_work(&isr->work);
-			kfree(isr);
-		}
-	}
+	struct nouveau_gpio *gpio = (void *)object;
+	nouveau_event_destroy(&gpio->events);
+	nouveau_subdev_destroy(&gpio->base);
 }
 
 int
 nouveau_gpio_create_(struct nouveau_object *parent,
 		     struct nouveau_object *engine,
-		     struct nouveau_oclass *oclass, int length, void **pobject)
+		     struct nouveau_oclass *oclass, int lines,
+		     int length, void **pobject)
 {
 	struct nouveau_gpio *gpio;
 	int ret;
@@ -241,15 +125,13 @@ nouveau_gpio_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
+	ret = nouveau_event_create(lines, &gpio->events);
+	if (ret)
+		return ret;
+
 	gpio->find = nouveau_gpio_find;
 	gpio->set  = nouveau_gpio_set;
 	gpio->get  = nouveau_gpio_get;
-	gpio->irq  = nouveau_gpio_irq;
-	gpio->isr_run = nouveau_gpio_isr_run;
-	gpio->isr_add = nouveau_gpio_isr_add;
-	gpio->isr_del = nouveau_gpio_isr_del;
-	INIT_LIST_HEAD(&gpio->isr);
-	spin_lock_init(&gpio->lock);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
index 168d16a9a8e9..76d5d5465ddd 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
@@ -24,7 +24,7 @@
  *
  */
 
-#include <subdev/gpio.h>
+#include "priv.h"
 
 struct nv10_gpio_priv {
 	struct nouveau_gpio base;
@@ -83,27 +83,36 @@ nv10_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 }
 
 static void
-nv10_gpio_irq_enable(struct nouveau_gpio *gpio, int line, bool on)
-{
-	u32 mask = 0x00010001 << line;
-
-	nv_wr32(gpio, 0x001104, mask);
-	nv_mask(gpio, 0x001144, mask, on ? mask : 0);
-}
-
-static void
 nv10_gpio_intr(struct nouveau_subdev *subdev)
 {
 	struct nv10_gpio_priv *priv = (void *)subdev;
 	u32 intr = nv_rd32(priv, 0x001104);
 	u32 hi = (intr & 0x0000ffff) >> 0;
 	u32 lo = (intr & 0xffff0000) >> 16;
+	int i;
 
-	priv->base.isr_run(&priv->base, 0, hi | lo);
+	for (i = 0; (hi | lo) && i < 32; i++) {
+		if ((hi | lo) & (1 << i))
+			nouveau_event_trigger(priv->base.events, i);
+	}
 
 	nv_wr32(priv, 0x001104, intr);
 }
 
+static void
+nv10_gpio_intr_enable(struct nouveau_event *event, int line)
+{
+	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
+	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00010001 << line);
+}
+
+static void
+nv10_gpio_intr_disable(struct nouveau_event *event, int line)
+{
+	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
+	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00000000);
+}
+
 static int
 nv10_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
@@ -112,14 +121,16 @@ nv10_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv10_gpio_priv *priv;
 	int ret;
 
-	ret = nouveau_gpio_create(parent, engine, oclass, &priv);
+	ret = nouveau_gpio_create(parent, engine, oclass, 16, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
 	priv->base.drive = nv10_gpio_drive;
 	priv->base.sense = nv10_gpio_sense;
-	priv->base.irq_enable = nv10_gpio_irq_enable;
+	priv->base.events->priv = priv;
+	priv->base.events->enable = nv10_gpio_intr_enable;
+	priv->base.events->disable = nv10_gpio_intr_disable;
 	nv_subdev(priv)->intr = nv10_gpio_intr;
 	return 0;
 }
@@ -141,8 +152,6 @@ nv10_gpio_init(struct nouveau_object *object)
 	if (ret)
 		return ret;
 
-	nv_wr32(priv, 0x001140, 0x00000000);
-	nv_wr32(priv, 0x001100, 0xffffffff);
 	nv_wr32(priv, 0x001144, 0x00000000);
 	nv_wr32(priv, 0x001104, 0xffffffff);
 	return 0;
@@ -152,7 +161,6 @@ static int
 nv10_gpio_fini(struct nouveau_object *object, bool suspend)
 {
 	struct nv10_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0x001140, 0x00000000);
 	nv_wr32(priv, 0x001144, 0x00000000);
 	return nouveau_gpio_fini(&priv->base, suspend);
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
index bf13a1200f26..bf489dcf46e2 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/gpio.h>
+#include "priv.h"
 
 struct nv50_gpio_priv {
 	struct nouveau_gpio base;
@@ -95,21 +95,12 @@ nv50_gpio_sense(struct nouveau_gpio *gpio, int line)
 }
 
 void
-nv50_gpio_irq_enable(struct nouveau_gpio *gpio, int line, bool on)
-{
-	u32 reg  = line < 16 ? 0xe050 : 0xe070;
-	u32 mask = 0x00010001 << (line & 0xf);
-
-	nv_wr32(gpio, reg + 4, mask);
-	nv_mask(gpio, reg + 0, mask, on ? mask : 0);
-}
-
-void
 nv50_gpio_intr(struct nouveau_subdev *subdev)
 {
 	struct nv50_gpio_priv *priv = (void *)subdev;
 	u32 intr0, intr1 = 0;
 	u32 hi, lo;
+	int i;
 
 	intr0 = nv_rd32(priv, 0xe054) & nv_rd32(priv, 0xe050);
 	if (nv_device(priv)->chipset >= 0x90)
@@ -117,13 +108,35 @@ nv50_gpio_intr(struct nouveau_subdev *subdev)
 
 	hi = (intr0 & 0x0000ffff) | (intr1 << 16);
 	lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-	priv->base.isr_run(&priv->base, 0, hi | lo);
+
+	for (i = 0; (hi | lo) && i < 32; i++) {
+		if ((hi | lo) & (1 << i))
+			nouveau_event_trigger(priv->base.events, i);
+	}
 
 	nv_wr32(priv, 0xe054, intr0);
 	if (nv_device(priv)->chipset >= 0x90)
 		nv_wr32(priv, 0xe074, intr1);
 }
 
+void
+nv50_gpio_intr_enable(struct nouveau_event *event, int line)
+{
+	const u32 addr = line < 16 ? 0xe050 : 0xe070;
+	const u32 mask = 0x00010001 << (line & 0xf);
+	nv_wr32(event->priv, addr + 0x04, mask);
+	nv_mask(event->priv, addr + 0x00, mask, mask);
+}
+
+void
+nv50_gpio_intr_disable(struct nouveau_event *event, int line)
+{
+	const u32 addr = line < 16 ? 0xe050 : 0xe070;
+	const u32 mask = 0x00010001 << (line & 0xf);
+	nv_wr32(event->priv, addr + 0x04, mask);
+	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
+}
+
 static int
 nv50_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
@@ -132,7 +145,9 @@ nv50_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_gpio_priv *priv;
 	int ret;
 
-	ret = nouveau_gpio_create(parent, engine, oclass, &priv);
+	ret = nouveau_gpio_create(parent, engine, oclass,
+				  nv_device(parent)->chipset >= 0x90 ? 32 : 16,
+				  &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -140,7 +155,9 @@ nv50_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->base.reset = nv50_gpio_reset;
 	priv->base.drive = nv50_gpio_drive;
 	priv->base.sense = nv50_gpio_sense;
-	priv->base.irq_enable = nv50_gpio_irq_enable;
+	priv->base.events->priv = priv;
+	priv->base.events->enable = nv50_gpio_intr_enable;
+	priv->base.events->disable = nv50_gpio_intr_disable;
 	nv_subdev(priv)->intr = nv50_gpio_intr;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
index 83e8b8f16e6a..010431e3acec 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
@@ -22,13 +22,13 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/gpio.h>
+#include "priv.h"
 
 struct nvd0_gpio_priv {
 	struct nouveau_gpio base;
 };
 
-static void
+void
 nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 {
 	struct nouveau_bios *bios = nouveau_bios(gpio);
@@ -57,7 +57,7 @@ nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 	}
 }
 
-static int
+int
 nvd0_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 {
 	u32 data = ((dir ^ 1) << 13) | (out << 12);
@@ -66,7 +66,7 @@ nvd0_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 	return 0;
 }
 
-static int
+int
 nvd0_gpio_sense(struct nouveau_gpio *gpio, int line)
 {
 	return !!(nv_rd32(gpio, 0x00d610 + (line * 4)) & 0x00004000);
@@ -80,7 +80,7 @@ nvd0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvd0_gpio_priv *priv;
 	int ret;
 
-	ret = nouveau_gpio_create(parent, engine, oclass, &priv);
+	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -88,7 +88,9 @@ nvd0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->base.reset = nvd0_gpio_reset;
 	priv->base.drive = nvd0_gpio_drive;
 	priv->base.sense = nvd0_gpio_sense;
-	priv->base.irq_enable = nv50_gpio_irq_enable;
+	priv->base.events->priv = priv;
+	priv->base.events->enable = nv50_gpio_intr_enable;
+	priv->base.events->disable = nv50_gpio_intr_disable;
 	nv_subdev(priv)->intr = nv50_gpio_intr;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
new file mode 100644
index 000000000000..16b8c5bf5efa
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
@@ -0,0 +1,131 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+struct nve0_gpio_priv {
+	struct nouveau_gpio base;
+};
+
+void
+nve0_gpio_intr(struct nouveau_subdev *subdev)
+{
+	struct nve0_gpio_priv *priv = (void *)subdev;
+	u32 intr0 = nv_rd32(priv, 0xdc00) & nv_rd32(priv, 0xdc08);
+	u32 intr1 = nv_rd32(priv, 0xdc80) & nv_rd32(priv, 0xdc88);
+	u32 hi = (intr0 & 0x0000ffff) | (intr1 << 16);
+	u32 lo = (intr0 >> 16) | (intr1 & 0xffff0000);
+	int i;
+
+	for (i = 0; (hi | lo) && i < 32; i++) {
+		if ((hi | lo) & (1 << i))
+			nouveau_event_trigger(priv->base.events, i);
+	}
+
+	nv_wr32(priv, 0xdc00, intr0);
+	nv_wr32(priv, 0xdc88, intr1);
+}
+
+void
+nve0_gpio_intr_enable(struct nouveau_event *event, int line)
+{
+	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
+	const u32 mask = 0x00010001 << (line & 0xf);
+	nv_wr32(event->priv, addr + 0x08, mask);
+	nv_mask(event->priv, addr + 0x00, mask, mask);
+}
+
+void
+nve0_gpio_intr_disable(struct nouveau_event *event, int line)
+{
+	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
+	const u32 mask = 0x00010001 << (line & 0xf);
+	nv_wr32(event->priv, addr + 0x08, mask);
+	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
+}
+
+int
+nve0_gpio_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nve0_gpio_priv *priv = (void *)object;
+	nv_wr32(priv, 0xdc08, 0x00000000);
+	nv_wr32(priv, 0xdc88, 0x00000000);
+	return nouveau_gpio_fini(&priv->base, suspend);
+}
+
+int
+nve0_gpio_init(struct nouveau_object *object)
+{
+	struct nve0_gpio_priv *priv = (void *)object;
+	int ret;
+
+	ret = nouveau_gpio_init(&priv->base);
+	if (ret)
+		return ret;
+
+	nv_wr32(priv, 0xdc00, 0xffffffff);
+	nv_wr32(priv, 0xdc80, 0xffffffff);
+	return 0;
+}
+
+void
+nve0_gpio_dtor(struct nouveau_object *object)
+{
+	struct nve0_gpio_priv *priv = (void *)object;
+	nouveau_gpio_destroy(&priv->base);
+}
+
+static int
+nve0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct nve0_gpio_priv *priv;
+	int ret;
+
+	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.reset = nvd0_gpio_reset;
+	priv->base.drive = nvd0_gpio_drive;
+	priv->base.sense = nvd0_gpio_sense;
+	priv->base.events->priv = priv;
+	priv->base.events->enable = nve0_gpio_intr_enable;
+	priv->base.events->disable = nve0_gpio_intr_disable;
+	nv_subdev(priv)->intr = nve0_gpio_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+nve0_gpio_oclass = {
+	.handle = NV_SUBDEV(GPIO, 0xe0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_gpio_ctor,
+		.dtor = nv50_gpio_dtor,
+		.init = nve0_gpio_init,
+		.fini = nve0_gpio_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
new file mode 100644
index 000000000000..2ee1c895c782
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
@@ -0,0 +1,17 @@
+#ifndef __NVKM_GPIO_H__
+#define __NVKM_GPIO_H__
+
+#include <subdev/gpio.h>
+
+void nv50_gpio_dtor(struct nouveau_object *);
+int  nv50_gpio_init(struct nouveau_object *);
+int  nv50_gpio_fini(struct nouveau_object *, bool);
+void nv50_gpio_intr(struct nouveau_subdev *);
+void nv50_gpio_intr_enable(struct nouveau_event *, int line);
+void nv50_gpio_intr_disable(struct nouveau_event *, int line);
+
+void nvd0_gpio_reset(struct nouveau_gpio *, u8);
+int  nvd0_gpio_drive(struct nouveau_gpio *, int, int, int);
+int  nvd0_gpio_sense(struct nouveau_gpio *, int);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
new file mode 100644
index 000000000000..dec94e9d776a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
@@ -0,0 +1,279 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <subdev/i2c.h>
+
+struct anx9805_i2c_port {
+	struct nouveau_i2c_port base;
+	u32 addr;
+	u32 ctrl;
+};
+
+static int
+anx9805_train(struct nouveau_i2c_port *port, int link_nr, int link_bw, bool enh)
+{
+	struct anx9805_i2c_port *chan = (void *)port;
+	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
+	u8 tmp, i;
+
+	nv_wri2cr(mast, chan->addr, 0xa0, link_bw);
+	nv_wri2cr(mast, chan->addr, 0xa1, link_nr | (enh ? 0x80 : 0x00));
+	nv_wri2cr(mast, chan->addr, 0xa2, 0x01);
+	nv_wri2cr(mast, chan->addr, 0xa8, 0x01);
+
+	i = 0;
+	while ((tmp = nv_rdi2cr(mast, chan->addr, 0xa8)) & 0x01) {
+		mdelay(5);
+		if (i++ == 100) {
+			nv_error(port, "link training timed out\n");
+			return -ETIMEDOUT;
+		}
+	}
+
+	if (tmp & 0x70) {
+		nv_error(port, "link training failed: 0x%02x\n", tmp);
+		return -EIO;
+	}
+
+	return 1;
+}
+
+static int
+anx9805_aux(struct nouveau_i2c_port *port, u8 type, u32 addr, u8 *data, u8 size)
+{
+	struct anx9805_i2c_port *chan = (void *)port;
+	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
+	int i, ret = -ETIMEDOUT;
+	u8 tmp;
+
+	tmp = nv_rdi2cr(mast, chan->ctrl, 0x07) & ~0x04;
+	nv_wri2cr(mast, chan->ctrl, 0x07, tmp | 0x04);
+	nv_wri2cr(mast, chan->ctrl, 0x07, tmp);
+	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
+
+	nv_wri2cr(mast, chan->addr, 0xe4, 0x80);
+	for (i = 0; !(type & 1) && i < size; i++)
+		nv_wri2cr(mast, chan->addr, 0xf0 + i, data[i]);
+	nv_wri2cr(mast, chan->addr, 0xe5, ((size - 1) << 4) | type);
+	nv_wri2cr(mast, chan->addr, 0xe6, (addr & 0x000ff) >>  0);
+	nv_wri2cr(mast, chan->addr, 0xe7, (addr & 0x0ff00) >>  8);
+	nv_wri2cr(mast, chan->addr, 0xe8, (addr & 0xf0000) >> 16);
+	nv_wri2cr(mast, chan->addr, 0xe9, 0x01);
+
+	i = 0;
+	while ((tmp = nv_rdi2cr(mast, chan->addr, 0xe9)) & 0x01) {
+		mdelay(5);
+		if (i++ == 32)
+			goto done;
+	}
+
+	if ((tmp = nv_rdi2cr(mast, chan->ctrl, 0xf7)) & 0x01) {
+		ret = -EIO;
+		goto done;
+	}
+
+	for (i = 0; (type & 1) && i < size; i++)
+		data[i] = nv_rdi2cr(mast, chan->addr, 0xf0 + i);
+	ret = 0;
+done:
+	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
+	return ret;
+}
+
+static const struct nouveau_i2c_func
+anx9805_aux_func = {
+	.aux = anx9805_aux,
+	.lnk_ctl = anx9805_train,
+};
+
+static int
+anx9805_aux_chan_ctor(struct nouveau_object *parent,
+		      struct nouveau_object *engine,
+		      struct nouveau_oclass *oclass, void *data, u32 index,
+		      struct nouveau_object **pobject)
+{
+	struct nouveau_i2c_port *mast = (void *)parent;
+	struct anx9805_i2c_port *chan;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_aux_algo, &chan);
+	*pobject = nv_object(chan);
+	if (ret)
+		return ret;
+
+	switch ((oclass->handle & 0xff00) >> 8) {
+	case 0x0d:
+		chan->addr = 0x38;
+		chan->ctrl = 0x39;
+		break;
+	case 0x0e:
+		chan->addr = 0x3c;
+		chan->ctrl = 0x3b;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	if (mast->adapter.algo == &i2c_bit_algo) {
+		struct i2c_algo_bit_data *algo = mast->adapter.algo_data;
+		algo->udelay = max(algo->udelay, 40);
+	}
+
+	chan->base.func = &anx9805_aux_func;
+	return 0;
+}
+
+static struct nouveau_ofuncs
+anx9805_aux_ofuncs = {
+	.ctor =  anx9805_aux_chan_ctor,
+	.dtor = _nouveau_i2c_port_dtor,
+	.init = _nouveau_i2c_port_init,
+	.fini = _nouveau_i2c_port_fini,
+};
+
+static int
+anx9805_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
+{
+	struct anx9805_i2c_port *port = adap->algo_data;
+	struct nouveau_i2c_port *mast = (void *)nv_object(port)->parent;
+	struct i2c_msg *msg = msgs;
+	int ret = -ETIMEDOUT;
+	int i, j, cnt = num;
+	u8 seg = 0x00, off = 0x00, tmp;
+
+	tmp = nv_rdi2cr(mast, port->ctrl, 0x07) & ~0x10;
+	nv_wri2cr(mast, port->ctrl, 0x07, tmp | 0x10);
+	nv_wri2cr(mast, port->ctrl, 0x07, tmp);
+	nv_wri2cr(mast, port->addr, 0x43, 0x05);
+	mdelay(5);
+
+	while (cnt--) {
+		if ( (msg->flags & I2C_M_RD) && msg->addr == 0x50) {
+			nv_wri2cr(mast, port->addr, 0x40, msg->addr << 1);
+			nv_wri2cr(mast, port->addr, 0x41, seg);
+			nv_wri2cr(mast, port->addr, 0x42, off);
+			nv_wri2cr(mast, port->addr, 0x44, msg->len);
+			nv_wri2cr(mast, port->addr, 0x45, 0x00);
+			nv_wri2cr(mast, port->addr, 0x43, 0x01);
+			for (i = 0; i < msg->len; i++) {
+				j = 0;
+				while (nv_rdi2cr(mast, port->addr, 0x46) & 0x10) {
+					mdelay(5);
+					if (j++ == 32)
+						goto done;
+				}
+				msg->buf[i] = nv_rdi2cr(mast, port->addr, 0x47);
+			}
+		} else
+		if (!(msg->flags & I2C_M_RD)) {
+			if (msg->addr == 0x50 && msg->len == 0x01) {
+				off = msg->buf[0];
+			} else
+			if (msg->addr == 0x30 && msg->len == 0x01) {
+				seg = msg->buf[0];
+			} else
+				goto done;
+		} else {
+			goto done;
+		}
+		msg++;
+	}
+
+	ret = num;
+done:
+	nv_wri2cr(mast, port->addr, 0x43, 0x00);
+	return ret;
+}
+
+static u32
+anx9805_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+static const struct i2c_algorithm
+anx9805_i2c_algo = {
+	.master_xfer = anx9805_xfer,
+	.functionality = anx9805_func
+};
+
+static const struct nouveau_i2c_func
+anx9805_i2c_func = {
+};
+
+static int
+anx9805_ddc_port_ctor(struct nouveau_object *parent,
+		      struct nouveau_object *engine,
+		      struct nouveau_oclass *oclass, void *data, u32 index,
+		      struct nouveau_object **pobject)
+{
+	struct nouveau_i2c_port *mast = (void *)parent;
+	struct anx9805_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &anx9805_i2c_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	switch ((oclass->handle & 0xff00) >> 8) {
+	case 0x0d:
+		port->addr = 0x3d;
+		port->ctrl = 0x39;
+		break;
+	case 0x0e:
+		port->addr = 0x3f;
+		port->ctrl = 0x3b;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	if (mast->adapter.algo == &i2c_bit_algo) {
+		struct i2c_algo_bit_data *algo = mast->adapter.algo_data;
+		algo->udelay = max(algo->udelay, 40);
+	}
+
+	port->base.func = &anx9805_i2c_func;
+	return 0;
+}
+
+static struct nouveau_ofuncs
+anx9805_ddc_ofuncs = {
+	.ctor =  anx9805_ddc_port_ctor,
+	.dtor = _nouveau_i2c_port_dtor,
+	.init = _nouveau_i2c_port_init,
+	.fini = _nouveau_i2c_port_fini,
+};
+
+struct nouveau_oclass
+nouveau_anx9805_sclass[] = {
+	{ .handle = NV_I2C_TYPE_EXTDDC(0x0d), .ofuncs = &anx9805_ddc_ofuncs },
+	{ .handle = NV_I2C_TYPE_EXTAUX(0x0d), .ofuncs = &anx9805_aux_ofuncs },
+	{ .handle = NV_I2C_TYPE_EXTDDC(0x0e), .ofuncs = &anx9805_ddc_ofuncs },
+	{ .handle = NV_I2C_TYPE_EXTAUX(0x0e), .ofuncs = &anx9805_aux_ofuncs },
+	{}
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
index dc27e794a851..5de074ad170b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
@@ -24,151 +24,40 @@
 
 #include <subdev/i2c.h>
 
-/******************************************************************************
- * aux channel util functions
- *****************************************************************************/
-#define AUX_DBG(fmt, args...) nv_debug(aux, "AUXCH(%d): " fmt, ch, ##args)
-#define AUX_ERR(fmt, args...) nv_error(aux, "AUXCH(%d): " fmt, ch, ##args)
-
-static void
-auxch_fini(struct nouveau_i2c *aux, int ch)
-{
-	nv_mask(aux, 0x00e4e4 + (ch * 0x50), 0x00310000, 0x00000000);
-}
-
-static int
-auxch_init(struct nouveau_i2c *aux, int ch)
-{
-	const u32 unksel = 1; /* nfi which to use, or if it matters.. */
-	const u32 ureq = unksel ? 0x00100000 : 0x00200000;
-	const u32 urep = unksel ? 0x01000000 : 0x02000000;
-	u32 ctrl, timeout;
-
-	/* wait up to 1ms for any previous transaction to be done... */
-	timeout = 1000;
-	do {
-		ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
-		udelay(1);
-		if (!timeout--) {
-			AUX_ERR("begin idle timeout 0x%08x\n", ctrl);
-			return -EBUSY;
-		}
-	} while (ctrl & 0x03010000);
-
-	/* set some magic, and wait up to 1ms for it to appear */
-	nv_mask(aux, 0x00e4e4 + (ch * 0x50), 0x00300000, ureq);
-	timeout = 1000;
-	do {
-		ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
-		udelay(1);
-		if (!timeout--) {
-			AUX_ERR("magic wait 0x%08x\n", ctrl);
-			auxch_fini(aux, ch);
-			return -EBUSY;
-		}
-	} while ((ctrl & 0x03000000) != urep);
-
-	return 0;
-}
-
-static int
-auxch_tx(struct nouveau_i2c *aux, int ch, u8 type, u32 addr, u8 *data, u8 size)
-{
-	u32 ctrl, stat, timeout, retries;
-	u32 xbuf[4] = {};
-	int ret, i;
-
-	AUX_DBG("%d: 0x%08x %d\n", type, addr, size);
-
-	ret = auxch_init(aux, ch);
-	if (ret)
-		goto out;
-
-	stat = nv_rd32(aux, 0x00e4e8 + (ch * 0x50));
-	if (!(stat & 0x10000000)) {
-		AUX_DBG("sink not detected\n");
-		ret = -ENXIO;
-		goto out;
-	}
-
-	if (!(type & 1)) {
-		memcpy(xbuf, data, size);
-		for (i = 0; i < 16; i += 4) {
-			AUX_DBG("wr 0x%08x\n", xbuf[i / 4]);
-			nv_wr32(aux, 0x00e4c0 + (ch * 0x50) + i, xbuf[i / 4]);
-		}
-	}
-
-	ctrl  = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
-	ctrl &= ~0x0001f0ff;
-	ctrl |= type << 12;
-	ctrl |= size - 1;
-	nv_wr32(aux, 0x00e4e0 + (ch * 0x50), addr);
-
-	/* retry transaction a number of times on failure... */
-	ret = -EREMOTEIO;
-	for (retries = 0; retries < 32; retries++) {
-		/* reset, and delay a while if this is a retry */
-		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x80000000 | ctrl);
-		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00000000 | ctrl);
-		if (retries)
-			udelay(400);
-
-		/* transaction request, wait up to 1ms for it to complete */
-		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00010000 | ctrl);
-
-		timeout = 1000;
-		do {
-			ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
-			udelay(1);
-			if (!timeout--) {
-				AUX_ERR("tx req timeout 0x%08x\n", ctrl);
-				goto out;
-			}
-		} while (ctrl & 0x00010000);
-
-		/* read status, and check if transaction completed ok */
-		stat = nv_mask(aux, 0x00e4e8 + (ch * 0x50), 0, 0);
-		if (!(stat & 0x000f0f00)) {
-			ret = 0;
-			break;
-		}
-
-		AUX_DBG("%02d 0x%08x 0x%08x\n", retries, ctrl, stat);
-	}
-
-	if (type & 1) {
-		for (i = 0; i < 16; i += 4) {
-			xbuf[i / 4] = nv_rd32(aux, 0x00e4d0 + (ch * 0x50) + i);
-			AUX_DBG("rd 0x%08x\n", xbuf[i / 4]);
-		}
-		memcpy(data, xbuf, size);
-	}
-
-out:
-	auxch_fini(aux, ch);
-	return ret;
-}
-
 int
-nv_rdaux(struct nouveau_i2c_port *auxch, u32 addr, u8 *data, u8 size)
+nv_rdaux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
-	return auxch_tx(auxch->i2c, auxch->drive, 9, addr, data, size);
+	if (port->func->aux) {
+		if (port->func->acquire)
+			port->func->acquire(port);
+		return port->func->aux(port, 9, addr, data, size);
+	}
+	return -ENODEV;
 }
 
 int
-nv_wraux(struct nouveau_i2c_port *auxch, u32 addr, u8 *data, u8 size)
+nv_wraux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
-	return auxch_tx(auxch->i2c, auxch->drive, 8, addr, data, size);
+	if (port->func->aux) {
+		if (port->func->acquire)
+			port->func->acquire(port);
+		return port->func->aux(port, 8, addr, data, size);
+	}
+	return -ENODEV;
 }
 
 static int
 aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 {
-	struct nouveau_i2c_port *auxch = (struct nouveau_i2c_port *)adap;
+	struct nouveau_i2c_port *port = adap->algo_data;
 	struct i2c_msg *msg = msgs;
 	int ret, mcnt = num;
 
+	if (!port->func->aux)
+		return -ENODEV;
+	if ( port->func->acquire)
+		port->func->acquire(port);
+
 	while (mcnt--) {
 		u8 remaining = msg->len;
 		u8 *ptr = msg->buf;
@@ -185,8 +74,7 @@ aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 			if (mcnt || remaining > 16)
 				cmd |= 4; /* MOT */
 
-			ret = auxch_tx(auxch->i2c, auxch->drive, cmd,
-				       msg->addr, ptr, cnt);
+			ret = port->func->aux(port, cmd, msg->addr, ptr, cnt);
 			if (ret < 0)
 				return ret;
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
index dbfc2abf0cfe..a114a0ed7e98 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012 Red Hat Inc.
+ * Copyright 2013 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -22,64 +22,136 @@
  * Authors: Ben Skeggs
  */
 
-#include "core/option.h"
+#include <core/option.h>
 
-#include "subdev/i2c.h"
-#include "subdev/vga.h"
+#include <subdev/bios.h>
+#include <subdev/bios/dcb.h>
+#include <subdev/bios/i2c.h>
+#include <subdev/i2c.h>
+#include <subdev/vga.h>
 
-int
-nv_rdi2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg)
+/******************************************************************************
+ * interface to linux i2c bit-banging algorithm
+ *****************************************************************************/
+
+#ifdef CONFIG_NOUVEAU_I2C_INTERNAL_DEFAULT
+#define CSTMSEL true
+#else
+#define CSTMSEL false
+#endif
+
+static int
+nouveau_i2c_pre_xfer(struct i2c_adapter *adap)
 {
-	u8 val;
-	struct i2c_msg msgs[] = {
-		{ .addr = addr, .flags = 0, .len = 1, .buf = &reg },
-		{ .addr = addr, .flags = I2C_M_RD, .len = 1, .buf = &val },
-	};
+	struct i2c_algo_bit_data *bit = adap->algo_data;
+	struct nouveau_i2c_port *port = bit->data;
+	if (port->func->acquire)
+		port->func->acquire(port);
+	return 0;
+}
 
-	int ret = i2c_transfer(&port->adapter, msgs, 2);
-	if (ret != 2)
-		return -EIO;
+static void
+nouveau_i2c_setscl(void *data, int state)
+{
+	struct nouveau_i2c_port *port = data;
+	port->func->drive_scl(port, state);
+}
 
-	return val;
+static void
+nouveau_i2c_setsda(void *data, int state)
+{
+	struct nouveau_i2c_port *port = data;
+	port->func->drive_sda(port, state);
 }
 
-int
-nv_wri2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg, u8 val)
+static int
+nouveau_i2c_getscl(void *data)
 {
-	struct i2c_msg msgs[] = {
-		{ .addr = addr, .flags = 0, .len = 1, .buf = &reg },
-		{ .addr = addr, .flags = 0, .len = 1, .buf = &val },
-	};
+	struct nouveau_i2c_port *port = data;
+	return port->func->sense_scl(port);
+}
 
-	int ret = i2c_transfer(&port->adapter, msgs, 2);
-	if (ret != 2)
-		return -EIO;
+static int
+nouveau_i2c_getsda(void *data)
+{
+	struct nouveau_i2c_port *port = data;
+	return port->func->sense_sda(port);
+}
 
-	return 0;
+/******************************************************************************
+ * base i2c "port" class implementation
+ *****************************************************************************/
+
+void
+_nouveau_i2c_port_dtor(struct nouveau_object *object)
+{
+	struct nouveau_i2c_port *port = (void *)object;
+	i2c_del_adapter(&port->adapter);
+	nouveau_object_destroy(&port->base);
 }
 
-bool
-nv_probe_i2c(struct nouveau_i2c_port *port, u8 addr)
+int
+nouveau_i2c_port_create_(struct nouveau_object *parent,
+			 struct nouveau_object *engine,
+			 struct nouveau_oclass *oclass, u8 index,
+			 const struct i2c_algorithm *algo,
+			 int size, void **pobject)
 {
-	u8 buf[] = { 0 };
-	struct i2c_msg msgs[] = {
-		{
-			.addr = addr,
-			.flags = 0,
-			.len = 1,
-			.buf = buf,
-		},
-		{
-			.addr = addr,
-			.flags = I2C_M_RD,
-			.len = 1,
-			.buf = buf,
-		}
-	};
+	struct nouveau_device *device = nv_device(parent);
+	struct nouveau_i2c *i2c = (void *)engine;
+	struct nouveau_i2c_port *port;
+	int ret;
 
-	return i2c_transfer(&port->adapter, msgs, 2) == 2;
+	ret = nouveau_object_create_(parent, engine, oclass, 0, size, pobject);
+	port = *pobject;
+	if (ret)
+		return ret;
+
+	snprintf(port->adapter.name, sizeof(port->adapter.name),
+		 "nouveau-%s-%d", device->name, index);
+	port->adapter.owner = THIS_MODULE;
+	port->adapter.dev.parent = &device->pdev->dev;
+	port->index = index;
+	i2c_set_adapdata(&port->adapter, i2c);
+
+	if ( algo == &nouveau_i2c_bit_algo &&
+	    !nouveau_boolopt(device->cfgopt, "NvI2C", CSTMSEL)) {
+		struct i2c_algo_bit_data *bit;
+
+		bit = kzalloc(sizeof(*bit), GFP_KERNEL);
+		if (!bit)
+			return -ENOMEM;
+
+		bit->udelay = 10;
+		bit->timeout = usecs_to_jiffies(2200);
+		bit->data = port;
+		bit->pre_xfer = nouveau_i2c_pre_xfer;
+		bit->setsda = nouveau_i2c_setsda;
+		bit->setscl = nouveau_i2c_setscl;
+		bit->getsda = nouveau_i2c_getsda;
+		bit->getscl = nouveau_i2c_getscl;
+
+		port->adapter.algo_data = bit;
+		ret = i2c_bit_add_bus(&port->adapter);
+	} else {
+		port->adapter.algo_data = port;
+		port->adapter.algo = algo;
+		ret = i2c_add_adapter(&port->adapter);
+	}
+
+	/* drop port's i2c subdev refcount, i2c handles this itself */
+	if (ret == 0) {
+		list_add_tail(&port->head, &i2c->ports);
+		atomic_dec(&engine->refcount);
+	}
+
+	return ret;
 }
 
+/******************************************************************************
+ * base i2c subdev class implementation
+ *****************************************************************************/
+
 static struct nouveau_i2c_port *
 nouveau_i2c_find(struct nouveau_i2c *i2c, u8 index)
 {
@@ -103,29 +175,23 @@ nouveau_i2c_find(struct nouveau_i2c *i2c, u8 index)
 
 	list_for_each_entry(port, &i2c->ports, head) {
 		if (port->index == index)
-			break;
+			return port;
 	}
 
-	if (&port->head == &i2c->ports)
-		return NULL;
+	return NULL;
+}
 
-	if (nv_device(i2c)->card_type >= NV_50 && (port->dcb & 0x00000100)) {
-		u32 reg = 0x00e500, val;
-		if (port->type == 6) {
-			reg += port->drive * 0x50;
-			val  = 0x2002;
-		} else {
-			reg += ((port->dcb & 0x1e00) >> 9) * 0x50;
-			val  = 0xe001;
-		}
+static struct nouveau_i2c_port *
+nouveau_i2c_find_type(struct nouveau_i2c *i2c, u16 type)
+{
+	struct nouveau_i2c_port *port;
 
-		/* nfi, but neither auxch or i2c work if it's 1 */
-		nv_mask(i2c, reg + 0x0c, 0x00000001, 0x00000000);
-		/* nfi, but switches auxch vs normal i2c */
-		nv_mask(i2c, reg + 0x00, 0x0000f003, val);
+	list_for_each_entry(port, &i2c->ports, head) {
+		if (nv_hclass(port) == type)
+			return port;
 	}
 
-	return port;
+	return NULL;
 }
 
 static int
@@ -155,109 +221,86 @@ nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 	return -ENODEV;
 }
 
-void
-nouveau_i2c_drive_scl(void *data, int state)
+int
+_nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
 {
-	struct nouveau_i2c_port *port = data;
+	struct nouveau_i2c *i2c = (void *)object;
+	struct nouveau_i2c_port *port;
+	int ret;
 
-	if (port->type == DCB_I2C_NV04_BIT) {
-		u8 val = nv_rdvgac(port->i2c, 0, port->drive);
-		if (state) val |= 0x20;
-		else	   val &= 0xdf;
-		nv_wrvgac(port->i2c, 0, port->drive, val | 0x01);
-	} else
-	if (port->type == DCB_I2C_NV4E_BIT) {
-		nv_mask(port->i2c, port->drive, 0x2f, state ? 0x21 : 0x01);
-	} else
-	if (port->type == DCB_I2C_NVIO_BIT) {
-		if (state) port->state |= 0x01;
-		else	   port->state &= 0xfe;
-		nv_wr32(port->i2c, port->drive, 4 | port->state);
+	list_for_each_entry(port, &i2c->ports, head) {
+		ret = nv_ofuncs(port)->fini(nv_object(port), suspend);
+		if (ret && suspend)
+			goto fail;
 	}
-}
-
-void
-nouveau_i2c_drive_sda(void *data, int state)
-{
-	struct nouveau_i2c_port *port = data;
 
-	if (port->type == DCB_I2C_NV04_BIT) {
-		u8 val = nv_rdvgac(port->i2c, 0, port->drive);
-		if (state) val |= 0x10;
-		else	   val &= 0xef;
-		nv_wrvgac(port->i2c, 0, port->drive, val | 0x01);
-	} else
-	if (port->type == DCB_I2C_NV4E_BIT) {
-		nv_mask(port->i2c, port->drive, 0x1f, state ? 0x11 : 0x01);
-	} else
-	if (port->type == DCB_I2C_NVIO_BIT) {
-		if (state) port->state |= 0x02;
-		else	   port->state &= 0xfd;
-		nv_wr32(port->i2c, port->drive, 4 | port->state);
+	return nouveau_subdev_fini(&i2c->base, suspend);
+fail:
+	list_for_each_entry_continue_reverse(port, &i2c->ports, head) {
+		nv_ofuncs(port)->init(nv_object(port));
 	}
+
+	return ret;
 }
 
 int
-nouveau_i2c_sense_scl(void *data)
+_nouveau_i2c_init(struct nouveau_object *object)
 {
-	struct nouveau_i2c_port *port = data;
-	struct nouveau_device *device = nv_device(port->i2c);
-
-	if (port->type == DCB_I2C_NV04_BIT) {
-		return !!(nv_rdvgac(port->i2c, 0, port->sense) & 0x04);
-	} else
-	if (port->type == DCB_I2C_NV4E_BIT) {
-		return !!(nv_rd32(port->i2c, port->sense) & 0x00040000);
-	} else
-	if (port->type == DCB_I2C_NVIO_BIT) {
-		if (device->card_type < NV_D0)
-			return !!(nv_rd32(port->i2c, port->sense) & 0x01);
-		else
-			return !!(nv_rd32(port->i2c, port->sense) & 0x10);
+	struct nouveau_i2c *i2c = (void *)object;
+	struct nouveau_i2c_port *port;
+	int ret;
+
+	ret = nouveau_subdev_init(&i2c->base);
+	if (ret == 0) {
+		list_for_each_entry(port, &i2c->ports, head) {
+			ret = nv_ofuncs(port)->init(nv_object(port));
+			if (ret)
+				goto fail;
+		}
 	}
 
-	return 0;
+	return ret;
+fail:
+	list_for_each_entry_continue_reverse(port, &i2c->ports, head) {
+		nv_ofuncs(port)->fini(nv_object(port), false);
+	}
+
+	return ret;
 }
 
-int
-nouveau_i2c_sense_sda(void *data)
+void
+_nouveau_i2c_dtor(struct nouveau_object *object)
 {
-	struct nouveau_i2c_port *port = data;
-	struct nouveau_device *device = nv_device(port->i2c);
-
-	if (port->type == DCB_I2C_NV04_BIT) {
-		return !!(nv_rdvgac(port->i2c, 0, port->sense) & 0x08);
-	} else
-	if (port->type == DCB_I2C_NV4E_BIT) {
-		return !!(nv_rd32(port->i2c, port->sense) & 0x00080000);
-	} else
-	if (port->type == DCB_I2C_NVIO_BIT) {
-		if (device->card_type < NV_D0)
-			return !!(nv_rd32(port->i2c, port->sense) & 0x02);
-		else
-			return !!(nv_rd32(port->i2c, port->sense) & 0x20);
+	struct nouveau_i2c *i2c = (void *)object;
+	struct nouveau_i2c_port *port, *temp;
+
+	list_for_each_entry_safe(port, temp, &i2c->ports, head) {
+		nouveau_object_ref(NULL, (struct nouveau_object **)&port);
 	}
 
-	return 0;
+	nouveau_subdev_destroy(&i2c->base);
 }
 
-static const u32 nv50_i2c_port[] = {
-	0x00e138, 0x00e150, 0x00e168, 0x00e180,
-	0x00e254, 0x00e274, 0x00e764, 0x00e780,
-	0x00e79c, 0x00e7b8
+static struct nouveau_oclass *
+nouveau_i2c_extdev_sclass[] = {
+	nouveau_anx9805_sclass,
 };
 
-static int
-nouveau_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-		 struct nouveau_oclass *oclass, void *data, u32 size,
-		 struct nouveau_object **pobject)
+int
+nouveau_i2c_create_(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass,
+		    struct nouveau_oclass *sclass,
+		    int length, void **pobject)
 {
-	struct nouveau_device *device = nv_device(parent);
 	struct nouveau_bios *bios = nouveau_bios(parent);
-	struct nouveau_i2c_port *port;
 	struct nouveau_i2c *i2c;
+	struct nouveau_object *object;
 	struct dcb_i2c_entry info;
-	int ret, i = -1;
+	int ret, i, j, index = -1;
+	struct dcb_output outp;
+	u8  ver, hdr;
+	u32 data;
 
 	ret = nouveau_subdev_create(parent, engine, oclass, 0,
 				    "I2C", "i2c", &i2c);
@@ -266,142 +309,60 @@ nouveau_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		return ret;
 
 	i2c->find = nouveau_i2c_find;
+	i2c->find_type = nouveau_i2c_find_type;
 	i2c->identify = nouveau_i2c_identify;
 	INIT_LIST_HEAD(&i2c->ports);
 
-	while (!dcb_i2c_parse(bios, ++i, &info)) {
+	while (!dcb_i2c_parse(bios, ++index, &info)) {
 		if (info.type == DCB_I2C_UNUSED)
 			continue;
 
-		port = kzalloc(sizeof(*port), GFP_KERNEL);
-		if (!port) {
-			nv_error(i2c, "failed port memory alloc at %d\n", i);
-			break;
-		}
-
-		port->type = info.type;
-		switch (port->type) {
-		case DCB_I2C_NV04_BIT:
-			port->drive = info.drive;
-			port->sense = info.sense;
-			break;
-		case DCB_I2C_NV4E_BIT:
-			port->drive = 0x600800 + info.drive;
-			port->sense = port->drive;
-			break;
-		case DCB_I2C_NVIO_BIT:
-			port->drive = info.drive & 0x0f;
-			if (device->card_type < NV_D0) {
-				if (port->drive >= ARRAY_SIZE(nv50_i2c_port))
-					break;
-				port->drive = nv50_i2c_port[port->drive];
-				port->sense = port->drive;
-			} else {
-				port->drive = 0x00d014 + (port->drive * 0x20);
-				port->sense = port->drive;
+		oclass = sclass;
+		do {
+			ret = -EINVAL;
+			if (oclass->handle == info.type) {
+				ret = nouveau_object_ctor(*pobject, *pobject,
+							  oclass, &info,
+							  index, &object);
 			}
+		} while (ret && (++oclass)->handle);
+	}
+
+	/* in addition to the busses specified in the i2c table, there
+	 * may be ddc/aux channels hiding behind external tmds/dp/etc
+	 * transmitters.
+	 */
+	index = ((index + 0x0f) / 0x10) * 0x10;
+	i = -1;
+	while ((data = dcb_outp_parse(bios, ++i, &ver, &hdr, &outp))) {
+		if (!outp.location || !outp.extdev)
+			continue;
+
+		switch (outp.type) {
+		case DCB_OUTPUT_TMDS:
+			info.type = NV_I2C_TYPE_EXTDDC(outp.extdev);
 			break;
-		case DCB_I2C_NVIO_AUX:
-			port->drive = info.drive & 0x0f;
-			port->sense = port->drive;
-			port->adapter.algo = &nouveau_i2c_aux_algo;
+		case DCB_OUTPUT_DP:
+			info.type = NV_I2C_TYPE_EXTAUX(outp.extdev);
 			break;
 		default:
-			break;
-		}
-
-		if (!port->adapter.algo && !port->drive) {
-			nv_error(i2c, "I2C%d: type %d index %x/%x unknown\n",
-				 i, port->type, port->drive, port->sense);
-			kfree(port);
 			continue;
 		}
 
-		snprintf(port->adapter.name, sizeof(port->adapter.name),
-			 "nouveau-%s-%d", device->name, i);
-		port->adapter.owner = THIS_MODULE;
-		port->adapter.dev.parent = &device->pdev->dev;
-		port->i2c = i2c;
-		port->index = i;
-		port->dcb = info.data;
-		i2c_set_adapdata(&port->adapter, i2c);
-
-		if (port->adapter.algo != &nouveau_i2c_aux_algo) {
-			nouveau_i2c_drive_scl(port, 0);
-			nouveau_i2c_drive_sda(port, 1);
-			nouveau_i2c_drive_scl(port, 1);
-
-#ifdef CONFIG_NOUVEAU_I2C_INTERNAL_DEFAULT
-			if (nouveau_boolopt(device->cfgopt, "NvI2C", true)) {
-#else
-			if (nouveau_boolopt(device->cfgopt, "NvI2C", false)) {
-#endif
-				port->adapter.algo = &nouveau_i2c_bit_algo;
-				ret = i2c_add_adapter(&port->adapter);
-			} else {
-				port->adapter.algo_data = &port->bit;
-				port->bit.udelay = 10;
-				port->bit.timeout = usecs_to_jiffies(2200);
-				port->bit.data = port;
-				port->bit.setsda = nouveau_i2c_drive_sda;
-				port->bit.setscl = nouveau_i2c_drive_scl;
-				port->bit.getsda = nouveau_i2c_sense_sda;
-				port->bit.getscl = nouveau_i2c_sense_scl;
-				ret = i2c_bit_add_bus(&port->adapter);
-			}
-		} else {
-			port->adapter.algo = &nouveau_i2c_aux_algo;
-			ret = i2c_add_adapter(&port->adapter);
-		}
-
-		if (ret) {
-			nv_error(i2c, "I2C%d: failed register: %d\n", i, ret);
-			kfree(port);
-			continue;
+		ret = -ENODEV;
+		j = -1;
+		while (ret && ++j < ARRAY_SIZE(nouveau_i2c_extdev_sclass)) {
+			parent = nv_object(i2c->find(i2c, outp.i2c_index));
+			oclass = nouveau_i2c_extdev_sclass[j];
+			do {
+				if (oclass->handle != info.type)
+					continue;
+				ret = nouveau_object_ctor(parent, *pobject,
+							  oclass, NULL,
+							  index++, &object);
+			} while (ret && (++oclass)->handle);
 		}
-
-		list_add_tail(&port->head, &i2c->ports);
 	}
 
 	return 0;
 }
-
-static void
-nouveau_i2c_dtor(struct nouveau_object *object)
-{
-	struct nouveau_i2c *i2c = (void *)object;
-	struct nouveau_i2c_port *port, *temp;
-
-	list_for_each_entry_safe(port, temp, &i2c->ports, head) {
-		i2c_del_adapter(&port->adapter);
-		list_del(&port->head);
-		kfree(port);
-	}
-
-	nouveau_subdev_destroy(&i2c->base);
-}
-
-static int
-nouveau_i2c_init(struct nouveau_object *object)
-{
-	struct nouveau_i2c *i2c = (void *)object;
-	return nouveau_subdev_init(&i2c->base);
-}
-
-static int
-nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
-{
-	struct nouveau_i2c *i2c = (void *)object;
-	return nouveau_subdev_fini(&i2c->base, suspend);
-}
-
-struct nouveau_oclass
-nouveau_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x00),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nouveau_i2c_ctor,
-		.dtor = nouveau_i2c_dtor,
-		.init = nouveau_i2c_init,
-		.fini = nouveau_i2c_fini,
-	},
-};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
index 1c4c9a5c8e2e..a6e72d3b06b5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
@@ -32,25 +32,25 @@
 static inline void
 i2c_drive_scl(struct nouveau_i2c_port *port, int state)
 {
-	nouveau_i2c_drive_scl(port, state);
+	port->func->drive_scl(port, state);
 }
 
 static inline void
 i2c_drive_sda(struct nouveau_i2c_port *port, int state)
 {
-	nouveau_i2c_drive_sda(port, state);
+	port->func->drive_sda(port, state);
 }
 
 static inline int
 i2c_sense_scl(struct nouveau_i2c_port *port)
 {
-	return nouveau_i2c_sense_scl(port);
+	return port->func->sense_scl(port);
 }
 
 static inline int
 i2c_sense_sda(struct nouveau_i2c_port *port)
 {
-	return nouveau_i2c_sense_sda(port);
+	return port->func->sense_sda(port);
 }
 
 static void
@@ -77,9 +77,8 @@ i2c_start(struct nouveau_i2c_port *port)
 {
 	int ret = 0;
 
-	port->state  = i2c_sense_scl(port);
-	port->state |= i2c_sense_sda(port) << 1;
-	if (port->state != 3) {
+	if (!i2c_sense_scl(port) ||
+	    !i2c_sense_sda(port)) {
 		i2c_drive_scl(port, 0);
 		i2c_drive_sda(port, 1);
 		if (!i2c_raise_scl(port))
@@ -184,10 +183,13 @@ i2c_addr(struct nouveau_i2c_port *port, struct i2c_msg *msg)
 static int
 i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 {
-	struct nouveau_i2c_port *port = (struct nouveau_i2c_port *)adap;
+	struct nouveau_i2c_port *port = adap->algo_data;
 	struct i2c_msg *msg = msgs;
 	int ret = 0, mcnt = num;
 
+	if (port->func->acquire)
+		port->func->acquire(port);
+
 	while (!ret && mcnt--) {
 		u8 remaining = msg->len;
 		u8 *ptr = msg->buf;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
new file mode 100644
index 000000000000..2ad18840fe63
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
@@ -0,0 +1,143 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+#include <subdev/vga.h>
+
+struct nv04_i2c_priv {
+	struct nouveau_i2c base;
+};
+
+struct nv04_i2c_port {
+	struct nouveau_i2c_port base;
+	u8 drive;
+	u8 sense;
+};
+
+static void
+nv04_i2c_drive_scl(struct nouveau_i2c_port *base, int state)
+{
+	struct nv04_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv04_i2c_port *port = (void *)base;
+	u8 val = nv_rdvgac(priv, 0, port->drive);
+	if (state) val |= 0x20;
+	else	   val &= 0xdf;
+	nv_wrvgac(priv, 0, port->drive, val | 0x01);
+}
+
+static void
+nv04_i2c_drive_sda(struct nouveau_i2c_port *base, int state)
+{
+	struct nv04_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv04_i2c_port *port = (void *)base;
+	u8 val = nv_rdvgac(priv, 0, port->drive);
+	if (state) val |= 0x10;
+	else	   val &= 0xef;
+	nv_wrvgac(priv, 0, port->drive, val | 0x01);
+}
+
+static int
+nv04_i2c_sense_scl(struct nouveau_i2c_port *base)
+{
+	struct nv04_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv04_i2c_port *port = (void *)base;
+	return !!(nv_rdvgac(priv, 0, port->sense) & 0x04);
+}
+
+static int
+nv04_i2c_sense_sda(struct nouveau_i2c_port *base)
+{
+	struct nv04_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv04_i2c_port *port = (void *)base;
+	return !!(nv_rdvgac(priv, 0, port->sense) & 0x08);
+}
+
+static const struct nouveau_i2c_func
+nv04_i2c_func = {
+	.drive_scl = nv04_i2c_drive_scl,
+	.drive_sda = nv04_i2c_drive_sda,
+	.sense_scl = nv04_i2c_sense_scl,
+	.sense_sda = nv04_i2c_sense_sda,
+};
+
+static int
+nv04_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv04_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_bit_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	port->base.func = &nv04_i2c_func;
+	port->drive = info->drive;
+	port->sense = info->sense;
+	return 0;
+}
+
+static struct nouveau_oclass
+nv04_i2c_sclass[] = {
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NV04_BIT),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv04_i2c_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = _nouveau_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{}
+};
+
+static int
+nv04_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv04_i2c_priv *priv;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, nv04_i2c_sclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nv04_i2c_oclass = {
+	.handle = NV_SUBDEV(I2C, 0x04),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
new file mode 100644
index 000000000000..f501ae25dbb3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
@@ -0,0 +1,135 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+#include <subdev/vga.h>
+
+struct nv4e_i2c_priv {
+	struct nouveau_i2c base;
+};
+
+struct nv4e_i2c_port {
+	struct nouveau_i2c_port base;
+	u32 addr;
+};
+
+static void
+nv4e_i2c_drive_scl(struct nouveau_i2c_port *base, int state)
+{
+	struct nv4e_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv4e_i2c_port *port = (void *)base;
+	nv_mask(priv, port->addr, 0x2f, state ? 0x21 : 0x01);
+}
+
+static void
+nv4e_i2c_drive_sda(struct nouveau_i2c_port *base, int state)
+{
+	struct nv4e_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv4e_i2c_port *port = (void *)base;
+	nv_mask(priv, port->addr, 0x1f, state ? 0x11 : 0x01);
+}
+
+static int
+nv4e_i2c_sense_scl(struct nouveau_i2c_port *base)
+{
+	struct nv4e_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv4e_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00040000);
+}
+
+static int
+nv4e_i2c_sense_sda(struct nouveau_i2c_port *base)
+{
+	struct nv4e_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv4e_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00080000);
+}
+
+static const struct nouveau_i2c_func
+nv4e_i2c_func = {
+	.drive_scl = nv4e_i2c_drive_scl,
+	.drive_sda = nv4e_i2c_drive_sda,
+	.sense_scl = nv4e_i2c_sense_scl,
+	.sense_sda = nv4e_i2c_sense_sda,
+};
+
+static int
+nv4e_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv4e_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_bit_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	port->base.func = &nv4e_i2c_func;
+	port->addr = 0x600800 + info->drive;
+	return 0;
+}
+
+static struct nouveau_oclass
+nv4e_i2c_sclass[] = {
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NV4E_BIT),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv4e_i2c_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = _nouveau_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{}
+};
+
+static int
+nv4e_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv4e_i2c_priv *priv;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, nv4e_i2c_sclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nv4e_i2c_oclass = {
+	.handle = NV_SUBDEV(I2C, 0x4e),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv4e_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
new file mode 100644
index 000000000000..378dfa324e5f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
@@ -0,0 +1,149 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+void
+nv50_i2c_drive_scl(struct nouveau_i2c_port *base, int state)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	if (state) port->state |= 0x01;
+	else	   port->state &= 0xfe;
+	nv_wr32(priv, port->addr, port->state);
+}
+
+void
+nv50_i2c_drive_sda(struct nouveau_i2c_port *base, int state)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	if (state) port->state |= 0x02;
+	else	   port->state &= 0xfd;
+	nv_wr32(priv, port->addr, port->state);
+}
+
+int
+nv50_i2c_sense_scl(struct nouveau_i2c_port *base)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00000001);
+}
+
+int
+nv50_i2c_sense_sda(struct nouveau_i2c_port *base)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00000002);
+}
+
+static const struct nouveau_i2c_func
+nv50_i2c_func = {
+	.drive_scl = nv50_i2c_drive_scl,
+	.drive_sda = nv50_i2c_drive_sda,
+	.sense_scl = nv50_i2c_sense_scl,
+	.sense_sda = nv50_i2c_sense_sda,
+};
+
+const u32 nv50_i2c_addr[] = {
+	0x00e138, 0x00e150, 0x00e168, 0x00e180,
+	0x00e254, 0x00e274, 0x00e764, 0x00e780,
+	0x00e79c, 0x00e7b8
+};
+const int nv50_i2c_addr_nr = ARRAY_SIZE(nv50_i2c_addr);
+
+static int
+nv50_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv50_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_bit_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	if (info->drive >= nv50_i2c_addr_nr)
+		return -EINVAL;
+
+	port->base.func = &nv50_i2c_func;
+	port->state = 0x00000007;
+	port->addr = nv50_i2c_addr[info->drive];
+	return 0;
+}
+
+int
+nv50_i2c_port_init(struct nouveau_object *object)
+{
+	struct nv50_i2c_priv *priv = (void *)object->engine;
+	struct nv50_i2c_port *port = (void *)object;
+	nv_wr32(priv, port->addr, port->state);
+	return nouveau_i2c_port_init(&port->base);
+}
+
+static struct nouveau_oclass
+nv50_i2c_sclass[] = {
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv50_i2c_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = nv50_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{}
+};
+
+static int
+nv50_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv50_i2c_priv *priv;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, nv50_i2c_sclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nv50_i2c_oclass = {
+	.handle = NV_SUBDEV(I2C, 0x50),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
new file mode 100644
index 000000000000..4e5ba48ebf5a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
@@ -0,0 +1,32 @@
+#ifndef __NV50_I2C_H__
+#define __NV50_I2C_H__
+
+#include <subdev/i2c.h>
+
+struct nv50_i2c_priv {
+	struct nouveau_i2c base;
+};
+
+struct nv50_i2c_port {
+	struct nouveau_i2c_port base;
+	u32 addr;
+	u32 ctrl;
+	u32 data;
+	u32 state;
+};
+
+extern const u32 nv50_i2c_addr[];
+extern const int nv50_i2c_addr_nr;
+int  nv50_i2c_port_init(struct nouveau_object *);
+int  nv50_i2c_sense_scl(struct nouveau_i2c_port *);
+int  nv50_i2c_sense_sda(struct nouveau_i2c_port *);
+void nv50_i2c_drive_scl(struct nouveau_i2c_port *, int state);
+void nv50_i2c_drive_sda(struct nouveau_i2c_port *, int state);
+
+int  nv94_aux_port_ctor(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, void *, u32,
+			struct nouveau_object **);
+void nv94_i2c_acquire(struct nouveau_i2c_port *);
+void nv94_i2c_release(struct nouveau_i2c_port *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
new file mode 100644
index 000000000000..61b771670bfe
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
@@ -0,0 +1,285 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+#define AUX_DBG(fmt, args...) nv_debug(aux, "AUXCH(%d): " fmt, ch, ##args)
+#define AUX_ERR(fmt, args...) nv_error(aux, "AUXCH(%d): " fmt, ch, ##args)
+
+static void
+auxch_fini(struct nouveau_i2c *aux, int ch)
+{
+	nv_mask(aux, 0x00e4e4 + (ch * 0x50), 0x00310000, 0x00000000);
+}
+
+static int
+auxch_init(struct nouveau_i2c *aux, int ch)
+{
+	const u32 unksel = 1; /* nfi which to use, or if it matters.. */
+	const u32 ureq = unksel ? 0x00100000 : 0x00200000;
+	const u32 urep = unksel ? 0x01000000 : 0x02000000;
+	u32 ctrl, timeout;
+
+	/* wait up to 1ms for any previous transaction to be done... */
+	timeout = 1000;
+	do {
+		ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
+		udelay(1);
+		if (!timeout--) {
+			AUX_ERR("begin idle timeout 0x%08x\n", ctrl);
+			return -EBUSY;
+		}
+	} while (ctrl & 0x03010000);
+
+	/* set some magic, and wait up to 1ms for it to appear */
+	nv_mask(aux, 0x00e4e4 + (ch * 0x50), 0x00300000, ureq);
+	timeout = 1000;
+	do {
+		ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
+		udelay(1);
+		if (!timeout--) {
+			AUX_ERR("magic wait 0x%08x\n", ctrl);
+			auxch_fini(aux, ch);
+			return -EBUSY;
+		}
+	} while ((ctrl & 0x03000000) != urep);
+
+	return 0;
+}
+
+int
+nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
+{
+	struct nouveau_i2c *aux = nouveau_i2c(base);
+	struct nv50_i2c_port *port = (void *)base;
+	u32 ctrl, stat, timeout, retries;
+	u32 xbuf[4] = {};
+	int ch = port->addr;
+	int ret, i;
+
+	AUX_DBG("%d: 0x%08x %d\n", type, addr, size);
+
+	ret = auxch_init(aux, ch);
+	if (ret)
+		goto out;
+
+	stat = nv_rd32(aux, 0x00e4e8 + (ch * 0x50));
+	if (!(stat & 0x10000000)) {
+		AUX_DBG("sink not detected\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	if (!(type & 1)) {
+		memcpy(xbuf, data, size);
+		for (i = 0; i < 16; i += 4) {
+			AUX_DBG("wr 0x%08x\n", xbuf[i / 4]);
+			nv_wr32(aux, 0x00e4c0 + (ch * 0x50) + i, xbuf[i / 4]);
+		}
+	}
+
+	ctrl  = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
+	ctrl &= ~0x0001f0ff;
+	ctrl |= type << 12;
+	ctrl |= size - 1;
+	nv_wr32(aux, 0x00e4e0 + (ch * 0x50), addr);
+
+	/* retry transaction a number of times on failure... */
+	ret = -EREMOTEIO;
+	for (retries = 0; retries < 32; retries++) {
+		/* reset, and delay a while if this is a retry */
+		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x80000000 | ctrl);
+		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00000000 | ctrl);
+		if (retries)
+			udelay(400);
+
+		/* transaction request, wait up to 1ms for it to complete */
+		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00010000 | ctrl);
+
+		timeout = 1000;
+		do {
+			ctrl = nv_rd32(aux, 0x00e4e4 + (ch * 0x50));
+			udelay(1);
+			if (!timeout--) {
+				AUX_ERR("tx req timeout 0x%08x\n", ctrl);
+				goto out;
+			}
+		} while (ctrl & 0x00010000);
+
+		/* read status, and check if transaction completed ok */
+		stat = nv_mask(aux, 0x00e4e8 + (ch * 0x50), 0, 0);
+		if (!(stat & 0x000f0f00)) {
+			ret = 0;
+			break;
+		}
+
+		AUX_DBG("%02d 0x%08x 0x%08x\n", retries, ctrl, stat);
+	}
+
+	if (type & 1) {
+		for (i = 0; i < 16; i += 4) {
+			xbuf[i / 4] = nv_rd32(aux, 0x00e4d0 + (ch * 0x50) + i);
+			AUX_DBG("rd 0x%08x\n", xbuf[i / 4]);
+		}
+		memcpy(data, xbuf, size);
+	}
+
+out:
+	auxch_fini(aux, ch);
+	return ret;
+}
+
+void
+nv94_i2c_acquire(struct nouveau_i2c_port *base)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	if (port->ctrl) {
+		nv_mask(priv, port->ctrl + 0x0c, 0x00000001, 0x00000000);
+		nv_mask(priv, port->ctrl + 0x00, 0x0000f003, port->data);
+	}
+}
+
+void
+nv94_i2c_release(struct nouveau_i2c_port *base)
+{
+}
+
+static const struct nouveau_i2c_func
+nv94_i2c_func = {
+	.acquire   = nv94_i2c_acquire,
+	.release   = nv94_i2c_release,
+	.drive_scl = nv50_i2c_drive_scl,
+	.drive_sda = nv50_i2c_drive_sda,
+	.sense_scl = nv50_i2c_sense_scl,
+	.sense_sda = nv50_i2c_sense_sda,
+};
+
+static int
+nv94_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv50_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_bit_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	if (info->drive >= nv50_i2c_addr_nr)
+		return -EINVAL;
+
+	port->base.func = &nv94_i2c_func;
+	port->state = 7;
+	port->addr = nv50_i2c_addr[info->drive];
+	if (info->share != DCB_I2C_UNUSED) {
+		port->ctrl = 0x00e500 + (info->share * 0x50);
+		port->data = 0x0000e001;
+	}
+	return 0;
+}
+
+static const struct nouveau_i2c_func
+nv94_aux_func = {
+	.acquire   = nv94_i2c_acquire,
+	.release   = nv94_i2c_release,
+	.aux       = nv94_aux,
+};
+
+int
+nv94_aux_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv50_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_aux_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	port->base.func = &nv94_aux_func;
+	port->addr = info->drive;
+	if (info->share != DCB_I2C_UNUSED) {
+		port->ctrl = 0x00e500 + (info->drive * 0x50);
+		port->data = 0x00002002;
+	}
+
+	return 0;
+}
+
+static struct nouveau_oclass
+nv94_i2c_sclass[] = {
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv94_i2c_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = nv50_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_AUX),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv94_aux_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = _nouveau_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{}
+};
+
+static int
+nv94_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv50_i2c_priv *priv;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, nv94_i2c_sclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nv94_i2c_oclass = {
+	.handle = NV_SUBDEV(I2C, 0x94),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv94_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
new file mode 100644
index 000000000000..f761b8a610f1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
@@ -0,0 +1,124 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+static int
+nvd0_i2c_sense_scl(struct nouveau_i2c_port *base)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00000010);
+}
+
+static int
+nvd0_i2c_sense_sda(struct nouveau_i2c_port *base)
+{
+	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
+	struct nv50_i2c_port *port = (void *)base;
+	return !!(nv_rd32(priv, port->addr) & 0x00000020);
+}
+
+static const struct nouveau_i2c_func
+nvd0_i2c_func = {
+	.acquire   = nv94_i2c_acquire,
+	.release   = nv94_i2c_release,
+	.drive_scl = nv50_i2c_drive_scl,
+	.drive_sda = nv50_i2c_drive_sda,
+	.sense_scl = nvd0_i2c_sense_scl,
+	.sense_sda = nvd0_i2c_sense_sda,
+};
+
+static int
+nvd0_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct dcb_i2c_entry *info = data;
+	struct nv50_i2c_port *port;
+	int ret;
+
+	ret = nouveau_i2c_port_create(parent, engine, oclass, index,
+				     &nouveau_i2c_bit_algo, &port);
+	*pobject = nv_object(port);
+	if (ret)
+		return ret;
+
+	port->base.func = &nvd0_i2c_func;
+	port->state = 0x00000007;
+	port->addr = 0x00d014 + (info->drive * 0x20);
+	if (info->share != DCB_I2C_UNUSED) {
+		port->ctrl = 0x00e500 + (info->share * 0x50);
+		port->data = 0x0000e001;
+	}
+	return 0;
+}
+
+static struct nouveau_oclass
+nvd0_i2c_sclass[] = {
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nvd0_i2c_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = nv50_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_AUX),
+	  .ofuncs = &(struct nouveau_ofuncs) {
+		  .ctor = nv94_aux_port_ctor,
+		  .dtor = _nouveau_i2c_port_dtor,
+		  .init = _nouveau_i2c_port_init,
+		  .fini = _nouveau_i2c_port_fini,
+	  },
+	},
+	{}
+};
+
+static int
+nvd0_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nv50_i2c_priv *priv;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, nvd0_i2c_sclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nvd0_i2c_oclass = {
+	.handle = NV_SUBDEV(I2C, 0xd0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvd0_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
index 23ebe477a6f0..89da8fa7ea0f 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
@@ -37,7 +37,7 @@ nv04_mc_intr[] = {
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x01000000, NVDEV_ENGINE_DISP },	/* NV04- PCRTC0 */
 	{ 0x02000000, NVDEV_ENGINE_DISP },	/* NV11- PCRTC1 */
-	{ 0x10000000, NVDEV_SUBDEV_GPIO },	/* PBUS */
+	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{}
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
index 8d759f830323..5965add6daee 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
@@ -38,6 +38,7 @@ nv50_mc_intr[] = {
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x00200000, NVDEV_SUBDEV_GPIO },
 	{ 0x04000000, NVDEV_ENGINE_DISP },
+	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0000d101, NVDEV_SUBDEV_FB },
 	{},
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
index ceb5c83f9459..3a80b29dce0f 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
@@ -35,10 +35,12 @@ nv98_mc_intr[] = {
 	{ 0x00001000, NVDEV_ENGINE_GR },
 	{ 0x00004000, NVDEV_ENGINE_CRYPT },	/* NV84:NVA3 */
 	{ 0x00008000, NVDEV_ENGINE_BSP },
+	{ 0x00080000, NVDEV_SUBDEV_THERM },	/* NVA3:NVC0 */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x00200000, NVDEV_SUBDEV_GPIO },
 	{ 0x00400000, NVDEV_ENGINE_COPY0 },	/* NVA3-     */
 	{ 0x04000000, NVDEV_ENGINE_DISP },
+	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0040d101, NVDEV_SUBDEV_FB },
 	{},
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
index 92796682722d..42bbf72023a8 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
@@ -36,11 +36,13 @@ nvc0_mc_intr[] = {
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
 	{ 0x00008000, NVDEV_ENGINE_BSP },
+	{ 0x00040000, NVDEV_SUBDEV_THERM },
 	{ 0x00020000, NVDEV_ENGINE_VP },
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x00200000, NVDEV_SUBDEV_GPIO },
 	{ 0x02000000, NVDEV_SUBDEV_LTCG },
 	{ 0x04000000, NVDEV_ENGINE_DISP },
+	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x40000000, NVDEV_SUBDEV_IBUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{},
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mxm/mxms.c b/drivers/gpu/drm/nouveau/core/subdev/mxm/mxms.c
index 839ca1edc132..4bde7f7f7b81 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mxm/mxms.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mxm/mxms.c
@@ -156,15 +156,15 @@ mxms_foreach(struct nouveau_mxm *mxm, u8 types,
 
 			nv_debug(mxm, "%4s: ", mxms_desc_name[type]);
 			for (j = headerlen - 1; j >= 0; j--)
-				printk("%02x", dump[j]);
-			printk("\n");
+				pr_cont("%02x", dump[j]);
+			pr_cont("\n");
 			dump += headerlen;
 
 			for (i = 0; i < entries; i++, dump += recordlen) {
 				nv_debug(mxm, "      ");
 				for (j = recordlen - 1; j >= 0; j--)
-					printk("%02x", dump[j]);
-				printk("\n");
+					pr_cont("%02x", dump[j]);
+				pr_cont("\n");
 			}
 		}
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
index 1674c74a76c8..f794dc89a3b2 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
@@ -29,6 +29,134 @@
 
 #include "priv.h"
 
+static int
+nouveau_therm_update_trip(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nouveau_therm_trip_point *trip = priv->fan->bios.trip,
+					*cur_trip = NULL,
+					*last_trip = priv->last_trip;
+	u8  temp = therm->temp_get(therm);
+	u16 duty, i;
+
+	/* look for the trip point corresponding to the current temperature */
+	cur_trip = NULL;
+	for (i = 0; i < priv->fan->bios.nr_fan_trip; i++) {
+		if (temp >= trip[i].temp)
+			cur_trip = &trip[i];
+	}
+
+	/* account for the hysteresis cycle */
+	if (last_trip && temp <= (last_trip->temp) &&
+	    temp > (last_trip->temp - last_trip->hysteresis))
+		cur_trip = last_trip;
+
+	if (cur_trip) {
+		duty = cur_trip->fan_duty;
+		priv->last_trip = cur_trip;
+	} else {
+		duty = 0;
+		priv->last_trip = NULL;
+	}
+
+	return duty;
+}
+
+static int
+nouveau_therm_update_linear(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	u8  linear_min_temp = priv->fan->bios.linear_min_temp;
+	u8  linear_max_temp = priv->fan->bios.linear_max_temp;
+	u8  temp = therm->temp_get(therm);
+	u16 duty;
+
+	/* handle the non-linear part first */
+	if (temp < linear_min_temp)
+		return priv->fan->bios.min_duty;
+	else if (temp > linear_max_temp)
+		return priv->fan->bios.max_duty;
+
+	/* we are in the linear zone */
+	duty  = (temp - linear_min_temp);
+	duty *= (priv->fan->bios.max_duty - priv->fan->bios.min_duty);
+	duty /= (linear_max_temp - linear_min_temp);
+	duty += priv->fan->bios.min_duty;
+
+	return duty;
+}
+
+static void
+nouveau_therm_update(struct nouveau_therm *therm, int mode)
+{
+	struct nouveau_timer *ptimer = nouveau_timer(therm);
+	struct nouveau_therm_priv *priv = (void *)therm;
+	unsigned long flags;
+	int duty;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	if (mode < 0)
+		mode = priv->mode;
+	priv->mode = mode;
+
+	switch (mode) {
+	case NOUVEAU_THERM_CTRL_MANUAL:
+		duty = nouveau_therm_fan_get(therm);
+		if (duty < 0)
+			duty = 100;
+		break;
+	case NOUVEAU_THERM_CTRL_AUTO:
+		if (priv->fan->bios.nr_fan_trip)
+			duty = nouveau_therm_update_trip(therm);
+		else
+			duty = nouveau_therm_update_linear(therm);
+		break;
+	case NOUVEAU_THERM_CTRL_NONE:
+	default:
+		goto done;
+	}
+
+	nv_debug(therm, "FAN target request: %d%%\n", duty);
+	nouveau_therm_fan_set(therm, (mode != NOUVEAU_THERM_CTRL_AUTO), duty);
+
+done:
+	if (list_empty(&priv->alarm.head) && (mode == NOUVEAU_THERM_CTRL_AUTO))
+		ptimer->alarm(ptimer, 1000000000ULL, &priv->alarm);
+	spin_unlock_irqrestore(&priv->lock, flags);
+}
+
+static void
+nouveau_therm_alarm(struct nouveau_alarm *alarm)
+{
+	struct nouveau_therm_priv *priv =
+	       container_of(alarm, struct nouveau_therm_priv, alarm);
+	nouveau_therm_update(&priv->base, -1);
+}
+
+int
+nouveau_therm_mode(struct nouveau_therm *therm, int mode)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nouveau_device *device = nv_device(therm);
+	static const char *name[] = {
+		"disabled",
+		"manual",
+		"automatic"
+	};
+
+	/* The default PDAEMON ucode interferes with fan management */
+	if ((mode >= ARRAY_SIZE(name)) ||
+	    (mode != NOUVEAU_THERM_CTRL_NONE && device->card_type >= NV_C0))
+		return -EINVAL;
+
+	if (priv->mode == mode)
+		return 0;
+
+	nv_info(therm, "Thermal management: %s\n", name[mode]);
+	nouveau_therm_update(therm, mode);
+	return 0;
+}
+
 int
 nouveau_therm_attr_get(struct nouveau_therm *therm,
 		       enum nouveau_therm_attr_type type)
@@ -37,11 +165,11 @@ nouveau_therm_attr_get(struct nouveau_therm *therm,
 
 	switch (type) {
 	case NOUVEAU_THERM_ATTR_FAN_MIN_DUTY:
-		return priv->bios_fan.min_duty;
+		return priv->fan->bios.min_duty;
 	case NOUVEAU_THERM_ATTR_FAN_MAX_DUTY:
-		return priv->bios_fan.max_duty;
+		return priv->fan->bios.max_duty;
 	case NOUVEAU_THERM_ATTR_FAN_MODE:
-		return priv->fan.mode;
+		return priv->mode;
 	case NOUVEAU_THERM_ATTR_THRS_FAN_BOOST:
 		return priv->bios_sensor.thrs_fan_boost.temp;
 	case NOUVEAU_THERM_ATTR_THRS_FAN_BOOST_HYST:
@@ -73,42 +201,50 @@ nouveau_therm_attr_set(struct nouveau_therm *therm,
 	case NOUVEAU_THERM_ATTR_FAN_MIN_DUTY:
 		if (value < 0)
 			value = 0;
-		if (value > priv->bios_fan.max_duty)
-			value = priv->bios_fan.max_duty;
-		priv->bios_fan.min_duty = value;
+		if (value > priv->fan->bios.max_duty)
+			value = priv->fan->bios.max_duty;
+		priv->fan->bios.min_duty = value;
 		return 0;
 	case NOUVEAU_THERM_ATTR_FAN_MAX_DUTY:
 		if (value < 0)
 			value = 0;
-		if (value < priv->bios_fan.min_duty)
-			value = priv->bios_fan.min_duty;
-		priv->bios_fan.max_duty = value;
+		if (value < priv->fan->bios.min_duty)
+			value = priv->fan->bios.min_duty;
+		priv->fan->bios.max_duty = value;
 		return 0;
 	case NOUVEAU_THERM_ATTR_FAN_MODE:
-		return nouveau_therm_fan_set_mode(therm, value);
+		return nouveau_therm_mode(therm, value);
 	case NOUVEAU_THERM_ATTR_THRS_FAN_BOOST:
 		priv->bios_sensor.thrs_fan_boost.temp = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_FAN_BOOST_HYST:
 		priv->bios_sensor.thrs_fan_boost.hysteresis = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_DOWN_CLK:
 		priv->bios_sensor.thrs_down_clock.temp = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_DOWN_CLK_HYST:
 		priv->bios_sensor.thrs_down_clock.hysteresis = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_CRITICAL:
 		priv->bios_sensor.thrs_critical.temp = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_CRITICAL_HYST:
 		priv->bios_sensor.thrs_critical.hysteresis = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_SHUTDOWN:
 		priv->bios_sensor.thrs_shutdown.temp = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	case NOUVEAU_THERM_ATTR_THRS_SHUTDOWN_HYST:
 		priv->bios_sensor.thrs_shutdown.hysteresis = value;
+		priv->sensor.program_alarms(therm);
 		return 0;
 	}
 
@@ -116,7 +252,7 @@ nouveau_therm_attr_set(struct nouveau_therm *therm,
 }
 
 int
-nouveau_therm_init(struct nouveau_object *object)
+_nouveau_therm_init(struct nouveau_object *object)
 {
 	struct nouveau_therm *therm = (void *)object;
 	struct nouveau_therm_priv *priv = (void *)therm;
@@ -126,19 +262,69 @@ nouveau_therm_init(struct nouveau_object *object)
 	if (ret)
 		return ret;
 
-	if (priv->fan.percent >= 0)
-		therm->fan_set(therm, priv->fan.percent);
-
+	if (priv->suspend >= 0)
+		nouveau_therm_mode(therm, priv->mode);
+	priv->sensor.program_alarms(therm);
 	return 0;
 }
 
 int
-nouveau_therm_fini(struct nouveau_object *object, bool suspend)
+_nouveau_therm_fini(struct nouveau_object *object, bool suspend)
 {
 	struct nouveau_therm *therm = (void *)object;
 	struct nouveau_therm_priv *priv = (void *)therm;
 
-	priv->fan.percent = therm->fan_get(therm);
+	if (suspend) {
+		priv->suspend = priv->mode;
+		priv->mode = NOUVEAU_THERM_CTRL_NONE;
+	}
 
 	return nouveau_subdev_fini(&therm->base, suspend);
 }
+
+int
+nouveau_therm_create_(struct nouveau_object *parent,
+		      struct nouveau_object *engine,
+		      struct nouveau_oclass *oclass,
+		      int length, void **pobject)
+{
+	struct nouveau_therm_priv *priv;
+	int ret;
+
+	ret = nouveau_subdev_create_(parent, engine, oclass, 0, "PTHERM",
+				     "therm", length, pobject);
+	priv = *pobject;
+	if (ret)
+		return ret;
+
+	nouveau_alarm_init(&priv->alarm, nouveau_therm_alarm);
+	spin_lock_init(&priv->lock);
+	spin_lock_init(&priv->sensor.alarm_program_lock);
+
+	priv->base.fan_get = nouveau_therm_fan_user_get;
+	priv->base.fan_set = nouveau_therm_fan_user_set;
+	priv->base.fan_sense = nouveau_therm_fan_sense;
+	priv->base.attr_get = nouveau_therm_attr_get;
+	priv->base.attr_set = nouveau_therm_attr_set;
+	priv->mode = priv->suspend = -1; /* undefined */
+	return 0;
+}
+
+int
+nouveau_therm_preinit(struct nouveau_therm *therm)
+{
+	nouveau_therm_ic_ctor(therm);
+	nouveau_therm_sensor_ctor(therm);
+	nouveau_therm_fan_ctor(therm);
+
+	nouveau_therm_mode(therm, NOUVEAU_THERM_CTRL_NONE);
+	return 0;
+}
+
+void
+_nouveau_therm_dtor(struct nouveau_object *object)
+{
+	struct nouveau_therm_priv *priv = (void *)object;
+	kfree(priv->fan);
+	nouveau_subdev_destroy(&priv->base.base);
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
index 523178685180..c728380d3d62 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
@@ -27,90 +27,107 @@
 
 #include <core/object.h>
 #include <core/device.h>
+
 #include <subdev/gpio.h>
 #include <subdev/timer.h>
 
-int
-nouveau_therm_fan_get(struct nouveau_therm *therm)
+static int
+nouveau_fan_update(struct nouveau_fan *fan, bool immediate, int target)
 {
+	struct nouveau_therm *therm = fan->parent;
 	struct nouveau_therm_priv *priv = (void *)therm;
-	struct nouveau_gpio *gpio = nouveau_gpio(therm);
-	struct dcb_gpio_func func;
-	int card_type = nv_device(therm)->card_type;
-	u32 divs, duty;
-	int ret;
-
-	if (!priv->fan.pwm_get)
-		return -ENODEV;
+	struct nouveau_timer *ptimer = nouveau_timer(priv);
+	unsigned long flags;
+	int ret = 0;
+	int duty;
+
+	/* update target fan speed, restricting to allowed range */
+	spin_lock_irqsave(&fan->lock, flags);
+	if (target < 0)
+		target = fan->percent;
+	target = max_t(u8, target, fan->bios.min_duty);
+	target = min_t(u8, target, fan->bios.max_duty);
+	if (fan->percent != target) {
+		nv_debug(therm, "FAN target: %d\n", target);
+		fan->percent = target;
+	}
 
-	ret = gpio->find(gpio, 0, DCB_GPIO_PWM_FAN, 0xff, &func);
-	if (ret == 0) {
-		ret = priv->fan.pwm_get(therm, func.line, &divs, &duty);
-		if (ret == 0 && divs) {
-			divs = max(divs, duty);
-			if (card_type <= NV_40 || (func.log[0] & 1))
-				duty = divs - duty;
-			return (duty * 100) / divs;
-		}
+	/* check that we're not already at the target duty cycle */
+	duty = fan->get(therm);
+	if (duty == target)
+		goto done;
+
+	/* smooth out the fanspeed increase/decrease */
+	if (!immediate && duty >= 0) {
+		/* the constant "3" is a rough approximation taken from
+		 * nvidia's behaviour.
+		 * it is meant to bump the fan speed more incrementally
+		 */
+		if (duty < target)
+			duty = min(duty + 3, target);
+		else if (duty > target)
+			duty = max(duty - 3, target);
+	} else {
+		duty = target;
+	}
 
-		return gpio->get(gpio, 0, func.func, func.line) * 100;
+	nv_debug(therm, "FAN update: %d\n", duty);
+	ret = fan->set(therm, duty);
+	if (ret)
+		goto done;
+
+	/* schedule next fan update, if not at target speed already */
+	if (list_empty(&fan->alarm.head) && target != duty) {
+		u16 bump_period = fan->bios.bump_period;
+		u16 slow_down_period = fan->bios.slow_down_period;
+		u64 delay;
+
+		if (duty > target)
+			delay = slow_down_period;
+		else if (duty == target)
+			delay = min(bump_period, slow_down_period) ;
+		else
+			delay = bump_period;
+
+		ptimer->alarm(ptimer, delay * 1000 * 1000, &fan->alarm);
 	}
 
-	return -ENODEV;
+done:
+	spin_unlock_irqrestore(&fan->lock, flags);
+	return ret;
+}
+
+static void
+nouveau_fan_alarm(struct nouveau_alarm *alarm)
+{
+	struct nouveau_fan *fan = container_of(alarm, struct nouveau_fan, alarm);
+	nouveau_fan_update(fan, false, -1);
 }
 
 int
-nouveau_therm_fan_set(struct nouveau_therm *therm, int percent)
+nouveau_therm_fan_get(struct nouveau_therm *therm)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
-	struct nouveau_gpio *gpio = nouveau_gpio(therm);
-	struct dcb_gpio_func func;
-	int card_type = nv_device(therm)->card_type;
-	u32 divs, duty;
-	int ret;
-
-	if (priv->fan.mode == FAN_CONTROL_NONE)
-		return -EINVAL;
-
-	if (!priv->fan.pwm_set)
-		return -ENODEV;
-
-	if (percent < priv->bios_fan.min_duty)
-		percent = priv->bios_fan.min_duty;
-	if (percent > priv->bios_fan.max_duty)
-		percent = priv->bios_fan.max_duty;
-
-	ret = gpio->find(gpio, 0, DCB_GPIO_PWM_FAN, 0xff, &func);
-	if (ret == 0) {
-		divs = priv->bios_perf_fan.pwm_divisor;
-		if (priv->bios_fan.pwm_freq) {
-			divs = 1;
-			if (priv->fan.pwm_clock)
-				divs = priv->fan.pwm_clock(therm);
-			divs /= priv->bios_fan.pwm_freq;
-		}
-
-		duty = ((divs * percent) + 99) / 100;
-		if (card_type <= NV_40 || (func.log[0] & 1))
-			duty = divs - duty;
-
-		ret = priv->fan.pwm_set(therm, func.line, divs, duty);
-		return ret;
-	}
+	return priv->fan->get(therm);
+}
 
-	return -ENODEV;
+int
+nouveau_therm_fan_set(struct nouveau_therm *therm, bool immediate, int percent)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	return nouveau_fan_update(priv->fan, immediate, percent);
 }
 
 int
 nouveau_therm_fan_sense(struct nouveau_therm *therm)
 {
+	struct nouveau_therm_priv *priv = (void *)therm;
 	struct nouveau_timer *ptimer = nouveau_timer(therm);
 	struct nouveau_gpio *gpio = nouveau_gpio(therm);
-	struct dcb_gpio_func func;
 	u32 cycles, cur, prev;
 	u64 start, end, tach;
 
-	if (gpio->find(gpio, 0, DCB_GPIO_FAN_SENSE, 0xff, &func))
+	if (priv->fan->tach.func == DCB_GPIO_UNUSED)
 		return -ENODEV;
 
 	/* Time a complete rotation and extrapolate to RPM:
@@ -118,12 +135,12 @@ nouveau_therm_fan_sense(struct nouveau_therm *therm)
 	 * We get 4 changes (0 -> 1 -> 0 -> 1) per complete rotation.
 	 */
 	start = ptimer->read(ptimer);
-	prev = gpio->get(gpio, 0, func.func, func.line);
+	prev = gpio->get(gpio, 0, priv->fan->tach.func, priv->fan->tach.line);
 	cycles = 0;
 	do {
 		usleep_range(500, 1000); /* supports 0 < rpm < 7500 */
 
-		cur = gpio->get(gpio, 0, func.func, func.line);
+		cur = gpio->get(gpio, 0, priv->fan->tach.func, priv->fan->tach.line);
 		if (prev != cur) {
 			if (!start)
 				start = ptimer->read(ptimer);
@@ -142,34 +159,6 @@ nouveau_therm_fan_sense(struct nouveau_therm *therm)
 }
 
 int
-nouveau_therm_fan_set_mode(struct nouveau_therm *therm,
-			   enum nouveau_therm_fan_mode mode)
-{
-	struct nouveau_therm_priv *priv = (void *)therm;
-
-	if (priv->fan.mode == mode)
-		return 0;
-
-	if (mode < FAN_CONTROL_NONE || mode >= FAN_CONTROL_NR)
-		return -EINVAL;
-
-	switch (mode)
-	{
-	case FAN_CONTROL_NONE:
-		nv_info(therm, "switch fan to no-control mode\n");
-		break;
-	case FAN_CONTROL_MANUAL:
-		nv_info(therm, "switch fan to manual mode\n");
-		break;
-	case FAN_CONTROL_NR:
-		break;
-	}
-
-	priv->fan.mode = mode;
-	return 0;
-}
-
-int
 nouveau_therm_fan_user_get(struct nouveau_therm *therm)
 {
 	return nouveau_therm_fan_get(therm);
@@ -180,55 +169,86 @@ nouveau_therm_fan_user_set(struct nouveau_therm *therm, int percent)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
 
-	if (priv->fan.mode != FAN_CONTROL_MANUAL)
+	if (priv->mode != NOUVEAU_THERM_CTRL_MANUAL)
 		return -EINVAL;
 
-	return nouveau_therm_fan_set(therm, percent);
+	return nouveau_therm_fan_set(therm, true, percent);
 }
 
-void
+static void
 nouveau_therm_fan_set_defaults(struct nouveau_therm *therm)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
 
-	priv->bios_fan.pwm_freq = 0;
-	priv->bios_fan.min_duty = 0;
-	priv->bios_fan.max_duty = 100;
+	priv->fan->bios.pwm_freq = 0;
+	priv->fan->bios.min_duty = 0;
+	priv->fan->bios.max_duty = 100;
+	priv->fan->bios.bump_period = 500;
+	priv->fan->bios.slow_down_period = 2000;
+	priv->fan->bios.linear_min_temp = 40;
+	priv->fan->bios.linear_max_temp = 85;
 }
 
-
 static void
 nouveau_therm_fan_safety_checks(struct nouveau_therm *therm)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
 
-	if (priv->bios_fan.min_duty > 100)
-		priv->bios_fan.min_duty = 100;
-	if (priv->bios_fan.max_duty > 100)
-		priv->bios_fan.max_duty = 100;
+	if (priv->fan->bios.min_duty > 100)
+		priv->fan->bios.min_duty = 100;
+	if (priv->fan->bios.max_duty > 100)
+		priv->fan->bios.max_duty = 100;
 
-	if (priv->bios_fan.min_duty > priv->bios_fan.max_duty)
-		priv->bios_fan.min_duty = priv->bios_fan.max_duty;
-}
-
-int nouveau_fan_pwm_clock_dummy(struct nouveau_therm *therm)
-{
-	return 1;
+	if (priv->fan->bios.min_duty > priv->fan->bios.max_duty)
+		priv->fan->bios.min_duty = priv->fan->bios.max_duty;
 }
 
 int
 nouveau_therm_fan_ctor(struct nouveau_therm *therm)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nouveau_gpio *gpio = nouveau_gpio(therm);
 	struct nouveau_bios *bios = nouveau_bios(therm);
+	struct dcb_gpio_func func;
+	int ret;
 
+	/* attempt to locate a drivable fan, and determine control method */
+	ret = gpio->find(gpio, 0, DCB_GPIO_FAN, 0xff, &func);
+	if (ret == 0) {
+		if (func.log[0] & DCB_GPIO_LOG_DIR_IN) {
+			nv_debug(therm, "GPIO_FAN is in input mode\n");
+			ret = -EINVAL;
+		} else {
+			ret = nouveau_fanpwm_create(therm, &func);
+			if (ret != 0)
+				ret = nouveau_fantog_create(therm, &func);
+		}
+	}
+
+	/* no controllable fan found, create a dummy fan module */
+	if (ret != 0) {
+		ret = nouveau_fannil_create(therm);
+		if (ret)
+			return ret;
+	}
+
+	nv_info(therm, "FAN control: %s\n", priv->fan->type);
+
+	/* attempt to detect a tachometer connection */
+	ret = gpio->find(gpio, 0, DCB_GPIO_FAN_SENSE, 0xff, &priv->fan->tach);
+	if (ret)
+		priv->fan->tach.func = DCB_GPIO_UNUSED;
+
+	/* initialise fan bump/slow update handling */
+	priv->fan->parent = therm;
+	nouveau_alarm_init(&priv->fan->alarm, nouveau_fan_alarm);
+	spin_lock_init(&priv->fan->lock);
+
+	/* other random init... */
 	nouveau_therm_fan_set_defaults(therm);
-	nvbios_perf_fan_parse(bios, &priv->bios_perf_fan);
-	if (nvbios_therm_fan_parse(bios, &priv->bios_fan))
+	nvbios_perf_fan_parse(bios, &priv->fan->perf);
+	if (nvbios_therm_fan_parse(bios, &priv->fan->bios))
 		nv_error(therm, "parsing the thermal table failed\n");
 	nouveau_therm_fan_safety_checks(therm);
-
-	nouveau_therm_fan_set_mode(therm, FAN_CONTROL_NONE);
-
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fannil.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fannil.c
new file mode 100644
index 000000000000..b78c182e1d51
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fannil.c
@@ -0,0 +1,54 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+static int
+nouveau_fannil_get(struct nouveau_therm *therm)
+{
+	return -ENODEV;
+}
+
+static int
+nouveau_fannil_set(struct nouveau_therm *therm, int percent)
+{
+	return -ENODEV;
+}
+
+int
+nouveau_fannil_create(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fan *priv;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	tpriv->fan = priv;
+	if (!priv)
+		return -ENOMEM;
+
+	priv->type = "none / external";
+	priv->get = nouveau_fannil_get;
+	priv->set = nouveau_fannil_set;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fanpwm.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fanpwm.c
new file mode 100644
index 000000000000..5f71db8e8992
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fanpwm.c
@@ -0,0 +1,107 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ * 	    Martin Peres
+ */
+
+#include <core/option.h>
+#include <subdev/gpio.h>
+
+#include "priv.h"
+
+struct nouveau_fanpwm_priv {
+	struct nouveau_fan base;
+	struct dcb_gpio_func func;
+};
+
+static int
+nouveau_fanpwm_get(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fanpwm_priv *priv = (void *)tpriv->fan;
+	struct nouveau_gpio *gpio = nouveau_gpio(therm);
+	int card_type = nv_device(therm)->card_type;
+	u32 divs, duty;
+	int ret;
+
+	ret = therm->pwm_get(therm, priv->func.line, &divs, &duty);
+	if (ret == 0 && divs) {
+		divs = max(divs, duty);
+		if (card_type <= NV_40 || (priv->func.log[0] & 1))
+			duty = divs - duty;
+		return (duty * 100) / divs;
+	}
+
+	return gpio->get(gpio, 0, priv->func.func, priv->func.line) * 100;
+}
+
+static int
+nouveau_fanpwm_set(struct nouveau_therm *therm, int percent)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fanpwm_priv *priv = (void *)tpriv->fan;
+	int card_type = nv_device(therm)->card_type;
+	u32 divs, duty;
+	int ret;
+
+	divs = priv->base.perf.pwm_divisor;
+	if (priv->base.bios.pwm_freq) {
+		divs = 1;
+		if (therm->pwm_clock)
+			divs = therm->pwm_clock(therm);
+		divs /= priv->base.bios.pwm_freq;
+	}
+
+	duty = ((divs * percent) + 99) / 100;
+	if (card_type <= NV_40 || (priv->func.log[0] & 1))
+		duty = divs - duty;
+
+	ret = therm->pwm_set(therm, priv->func.line, divs, duty);
+	if (ret == 0)
+		ret = therm->pwm_ctrl(therm, priv->func.line, true);
+	return ret;
+}
+
+int
+nouveau_fanpwm_create(struct nouveau_therm *therm, struct dcb_gpio_func *func)
+{
+	struct nouveau_device *device = nv_device(therm);
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fanpwm_priv *priv;
+	u32 divs, duty;
+
+	if (!nouveau_boolopt(device->cfgopt, "NvFanPWM", func->param) ||
+	    !therm->pwm_ctrl ||
+	     therm->pwm_get(therm, func->line, &divs, &duty) == -ENODEV)
+		return -ENODEV;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	tpriv->fan = &priv->base;
+	if (!priv)
+		return -ENOMEM;
+
+	priv->base.type = "PWM";
+	priv->base.get = nouveau_fanpwm_get;
+	priv->base.set = nouveau_fanpwm_set;
+	priv->func = *func;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c
new file mode 100644
index 000000000000..e601773ee475
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c
@@ -0,0 +1,115 @@
+/*
+ * Copyright 2012 The Nouveau community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+
+#include "priv.h"
+
+#include <core/object.h>
+#include <core/device.h>
+
+#include <subdev/gpio.h>
+#include <subdev/timer.h>
+
+struct nouveau_fantog_priv {
+	struct nouveau_fan base;
+	struct nouveau_alarm alarm;
+	spinlock_t lock;
+	u32 period_us;
+	u32 percent;
+	struct dcb_gpio_func func;
+};
+
+static void
+nouveau_fantog_update(struct nouveau_fantog_priv *priv, int percent)
+{
+	struct nouveau_therm_priv *tpriv = (void *)priv->base.parent;
+	struct nouveau_timer *ptimer = nouveau_timer(tpriv);
+	struct nouveau_gpio *gpio = nouveau_gpio(tpriv);
+	unsigned long flags;
+	int duty;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	if (percent < 0)
+		percent = priv->percent;
+	priv->percent = percent;
+
+	duty = !gpio->get(gpio, 0, DCB_GPIO_FAN, 0xff);
+	gpio->set(gpio, 0, DCB_GPIO_FAN, 0xff, duty);
+
+	if (list_empty(&priv->alarm.head) && percent != (duty * 100)) {
+		u64 next_change = (percent * priv->period_us) / 100;
+		if (!duty)
+			next_change = priv->period_us - next_change;
+		ptimer->alarm(ptimer, next_change * 1000, &priv->alarm);
+	}
+	spin_unlock_irqrestore(&priv->lock, flags);
+}
+
+static void
+nouveau_fantog_alarm(struct nouveau_alarm *alarm)
+{
+	struct nouveau_fantog_priv *priv =
+	       container_of(alarm, struct nouveau_fantog_priv, alarm);
+	nouveau_fantog_update(priv, -1);
+}
+
+static int
+nouveau_fantog_get(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fantog_priv *priv = (void *)tpriv->fan;
+	return priv->percent;
+}
+
+static int
+nouveau_fantog_set(struct nouveau_therm *therm, int percent)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fantog_priv *priv = (void *)tpriv->fan;
+	if (therm->pwm_ctrl)
+		therm->pwm_ctrl(therm, priv->func.line, false);
+	nouveau_fantog_update(priv, percent);
+	return 0;
+}
+
+int
+nouveau_fantog_create(struct nouveau_therm *therm, struct dcb_gpio_func *func)
+{
+	struct nouveau_therm_priv *tpriv = (void *)therm;
+	struct nouveau_fantog_priv *priv;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	tpriv->fan = &priv->base;
+	if (!priv)
+		return -ENOMEM;
+
+	priv->base.type = "toggle";
+	priv->base.get = nouveau_fantog_get;
+	priv->base.set = nouveau_fantog_set;
+	nouveau_alarm_init(&priv->alarm, nouveau_fantog_alarm);
+	priv->period_us = 100000; /* 10Hz */
+	priv->percent = 100;
+	priv->func = *func;
+	spin_lock_init(&priv->lock);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
index e512ff0aae60..e24090bac195 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
@@ -31,7 +31,7 @@ static bool
 probe_monitoring_device(struct nouveau_i2c_port *i2c,
 			struct i2c_board_info *info)
 {
-	struct nouveau_therm_priv *priv = (void *)nouveau_therm(i2c->i2c);
+	struct nouveau_therm_priv *priv = (void *)nouveau_therm(i2c);
 	struct i2c_client *client;
 
 	request_module("%s%s", I2C_MODULE_PREFIX, info->type);
@@ -53,6 +53,31 @@ probe_monitoring_device(struct nouveau_i2c_port *i2c,
 	return true;
 }
 
+static struct i2c_board_info
+nv_board_infos[] = {
+	{ I2C_BOARD_INFO("w83l785ts", 0x2d) },
+	{ I2C_BOARD_INFO("w83781d", 0x2d) },
+	{ I2C_BOARD_INFO("adt7473", 0x2e) },
+	{ I2C_BOARD_INFO("adt7473", 0x2d) },
+	{ I2C_BOARD_INFO("adt7473", 0x2c) },
+	{ I2C_BOARD_INFO("f75375", 0x2e) },
+	{ I2C_BOARD_INFO("lm99", 0x4c) },
+	{ I2C_BOARD_INFO("lm90", 0x4c) },
+	{ I2C_BOARD_INFO("lm90", 0x4d) },
+	{ I2C_BOARD_INFO("adm1021", 0x18) },
+	{ I2C_BOARD_INFO("adm1021", 0x19) },
+	{ I2C_BOARD_INFO("adm1021", 0x1a) },
+	{ I2C_BOARD_INFO("adm1021", 0x29) },
+	{ I2C_BOARD_INFO("adm1021", 0x2a) },
+	{ I2C_BOARD_INFO("adm1021", 0x2b) },
+	{ I2C_BOARD_INFO("adm1021", 0x4c) },
+	{ I2C_BOARD_INFO("adm1021", 0x4d) },
+	{ I2C_BOARD_INFO("adm1021", 0x4e) },
+	{ I2C_BOARD_INFO("lm63", 0x18) },
+	{ I2C_BOARD_INFO("lm63", 0x4e) },
+	{ }
+};
+
 void
 nouveau_therm_ic_ctor(struct nouveau_therm *therm)
 {
@@ -60,29 +85,6 @@ nouveau_therm_ic_ctor(struct nouveau_therm *therm)
 	struct nouveau_bios *bios = nouveau_bios(therm);
 	struct nouveau_i2c *i2c = nouveau_i2c(therm);
 	struct nvbios_extdev_func extdev_entry;
-	struct i2c_board_info info[] = {
-		{ I2C_BOARD_INFO("w83l785ts", 0x2d) },
-		{ I2C_BOARD_INFO("w83781d", 0x2d) },
-		{ I2C_BOARD_INFO("adt7473", 0x2e) },
-		{ I2C_BOARD_INFO("adt7473", 0x2d) },
-		{ I2C_BOARD_INFO("adt7473", 0x2c) },
-		{ I2C_BOARD_INFO("f75375", 0x2e) },
-		{ I2C_BOARD_INFO("lm99", 0x4c) },
-		{ I2C_BOARD_INFO("lm90", 0x4c) },
-		{ I2C_BOARD_INFO("lm90", 0x4d) },
-		{ I2C_BOARD_INFO("adm1021", 0x18) },
-		{ I2C_BOARD_INFO("adm1021", 0x19) },
-		{ I2C_BOARD_INFO("adm1021", 0x1a) },
-		{ I2C_BOARD_INFO("adm1021", 0x29) },
-		{ I2C_BOARD_INFO("adm1021", 0x2a) },
-		{ I2C_BOARD_INFO("adm1021", 0x2b) },
-		{ I2C_BOARD_INFO("adm1021", 0x4c) },
-		{ I2C_BOARD_INFO("adm1021", 0x4d) },
-		{ I2C_BOARD_INFO("adm1021", 0x4e) },
-		{ I2C_BOARD_INFO("lm63", 0x18) },
-		{ I2C_BOARD_INFO("lm63", 0x4e) },
-		{ }
-	};
 
 	if (!nvbios_extdev_find(bios, NVBIOS_EXTDEV_LM89, &extdev_entry)) {
 		struct i2c_board_info board[] = {
@@ -111,6 +113,6 @@ nouveau_therm_ic_ctor(struct nouveau_therm *therm)
 	/* The vbios doesn't provide the address of an exisiting monitoring
 	   device. Let's try our static list.
 	 */
-	i2c->identify(i2c, NV_I2C_DEFAULT(0), "monitoring device", info,
-		      probe_monitoring_device);
+	i2c->identify(i2c, NV_I2C_DEFAULT(0), "monitoring device",
+		      nv_board_infos, probe_monitoring_device);
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nv40.c
index fcf2cfe731d6..0f5363edb964 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nv40.c
@@ -25,6 +25,10 @@
 
 #include "priv.h"
 
+struct nv40_therm_priv {
+	struct nouveau_therm_priv base;
+};
+
 static int
 nv40_sensor_setup(struct nouveau_therm *therm)
 {
@@ -34,6 +38,7 @@ nv40_sensor_setup(struct nouveau_therm *therm)
 	if (device->chipset >= 0x46) {
 		nv_mask(therm, 0x15b8, 0x80000000, 0);
 		nv_wr32(therm, 0x15b0, 0x80003fff);
+		mdelay(10); /* wait for the temperature to stabilize */
 		return nv_rd32(therm, 0x15b4) & 0x3fff;
 	} else {
 		nv_wr32(therm, 0x15b0, 0xff);
@@ -75,7 +80,20 @@ nv40_temp_get(struct nouveau_therm *therm)
 	return core_temp;
 }
 
-int
+static int
+nv40_fan_pwm_ctrl(struct nouveau_therm *therm, int line, bool enable)
+{
+	u32 mask = enable ? 0x80000000 : 0x0000000;
+	if      (line == 2) nv_mask(therm, 0x0010f0, 0x80000000, mask);
+	else if (line == 9) nv_mask(therm, 0x0015f4, 0x80000000, mask);
+	else {
+		nv_error(therm, "unknown pwm ctrl for gpio %d\n", line);
+		return -ENODEV;
+	}
+	return 0;
+}
+
+static int
 nv40_fan_pwm_get(struct nouveau_therm *therm, int line, u32 *divs, u32 *duty)
 {
 	if (line == 2) {
@@ -101,15 +119,15 @@ nv40_fan_pwm_get(struct nouveau_therm *therm, int line, u32 *divs, u32 *duty)
 	return -EINVAL;
 }
 
-int
+static int
 nv40_fan_pwm_set(struct nouveau_therm *therm, int line, u32 divs, u32 duty)
 {
 	if (line == 2) {
-		nv_wr32(therm, 0x0010f0, 0x80000000 | (duty << 16) | divs);
+		nv_mask(therm, 0x0010f0, 0x7fff7fff, (duty << 16) | divs);
 	} else
 	if (line == 9) {
 		nv_wr32(therm, 0x0015f8, divs);
-		nv_wr32(therm, 0x0015f4, duty | 0x80000000);
+		nv_mask(therm, 0x0015f4, 0x7fffffff, duty);
 	} else {
 		nv_error(therm, "unknown pwm ctrl for gpio %d\n", line);
 		return -ENODEV;
@@ -118,37 +136,51 @@ nv40_fan_pwm_set(struct nouveau_therm *therm, int line, u32 divs, u32 duty)
 	return 0;
 }
 
+static void
+nv40_therm_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_therm *therm = nouveau_therm(subdev);
+	uint32_t stat = nv_rd32(therm, 0x1100);
+
+	/* traitement */
+
+	/* ack all IRQs */
+	nv_wr32(therm, 0x1100, 0x70000);
+
+	nv_error(therm, "THERM received an IRQ: stat = %x\n", stat);
+}
+
 static int
 nv40_therm_ctor(struct nouveau_object *parent,
-		   struct nouveau_object *engine,
-		   struct nouveau_oclass *oclass, void *data, u32 size,
-		   struct nouveau_object **pobject)
+		struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
 {
-	struct nouveau_therm_priv *priv;
-	struct nouveau_therm *therm;
+	struct nv40_therm_priv *priv;
 	int ret;
 
 	ret = nouveau_therm_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
-	therm = (void *) priv;
 	if (ret)
 		return ret;
 
-	nouveau_therm_ic_ctor(therm);
-	nouveau_therm_sensor_ctor(therm);
-	nouveau_therm_fan_ctor(therm);
+	priv->base.base.pwm_ctrl = nv40_fan_pwm_ctrl;
+	priv->base.base.pwm_get = nv40_fan_pwm_get;
+	priv->base.base.pwm_set = nv40_fan_pwm_set;
+	priv->base.base.temp_get = nv40_temp_get;
+	priv->base.sensor.program_alarms = nouveau_therm_program_alarms_polling;
+	nv_subdev(priv)->intr = nv40_therm_intr;
+	return nouveau_therm_preinit(&priv->base.base);
+}
 
-	priv->fan.pwm_get = nv40_fan_pwm_get;
-	priv->fan.pwm_set = nv40_fan_pwm_set;
+static int
+nv40_therm_init(struct nouveau_object *object)
+{
+	struct nouveau_therm *therm = (void *)object;
 
-	therm->temp_get = nv40_temp_get;
-	therm->fan_get = nouveau_therm_fan_user_get;
-	therm->fan_set = nouveau_therm_fan_user_set;
-	therm->fan_sense = nouveau_therm_fan_sense;
-	therm->attr_get = nouveau_therm_attr_get;
-	therm->attr_set = nouveau_therm_attr_set;
+	nv40_sensor_setup(therm);
 
-	return 0;
+	return _nouveau_therm_init(object);
 }
 
 struct nouveau_oclass
@@ -157,7 +189,7 @@ nv40_therm_oclass = {
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv40_therm_ctor,
 		.dtor = _nouveau_therm_dtor,
-		.init = nouveau_therm_init,
-		.fini = nouveau_therm_fini,
+		.init = nv40_therm_init,
+		.fini = _nouveau_therm_fini,
 	},
-};
\ No newline at end of file
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nv50.c
index 9360ddd469e7..86632cbd65ce 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nv50.c
@@ -25,6 +25,10 @@
 
 #include "priv.h"
 
+struct nv50_therm_priv {
+	struct nouveau_therm_priv base;
+};
+
 static int
 pwm_info(struct nouveau_therm *therm, int *line, int *ctrl, int *indx)
 {
@@ -51,6 +55,16 @@ pwm_info(struct nouveau_therm *therm, int *line, int *ctrl, int *indx)
 }
 
 int
+nv50_fan_pwm_ctrl(struct nouveau_therm *therm, int line, bool enable)
+{
+	u32 data = enable ? 0x00000001 : 0x00000000;
+	int ctrl, id, ret = pwm_info(therm, &line, &ctrl, &id);
+	if (ret == 0)
+		nv_mask(therm, ctrl, 0x00010001 << line, data << line);
+	return ret;
+}
+
+int
 nv50_fan_pwm_get(struct nouveau_therm *therm, int line, u32 *divs, u32 *duty)
 {
 	int ctrl, id, ret = pwm_info(therm, &line, &ctrl, &id);
@@ -73,7 +87,6 @@ nv50_fan_pwm_set(struct nouveau_therm *therm, int line, u32 divs, u32 duty)
 	if (ret)
 		return ret;
 
-	nv_mask(therm, ctrl, 0x00010001 << line, 0x00000001 << line);
 	nv_wr32(therm, 0x00e114 + (id * 8), divs);
 	nv_wr32(therm, 0x00e118 + (id * 8), duty | 0x80000000);
 	return 0;
@@ -111,38 +124,178 @@ nv50_temp_get(struct nouveau_therm *therm)
 	return nv_rd32(therm, 0x20400);
 }
 
+static void
+nv50_therm_program_alarms(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nvbios_therm_sensor *sensor = &priv->bios_sensor;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags);
+
+	/* enable RISING and FALLING IRQs for shutdown, THRS 0, 1, 2 and 4 */
+	nv_wr32(therm, 0x20000, 0x000003ff);
+
+	/* shutdown: The computer should be shutdown when reached */
+	nv_wr32(therm, 0x20484, sensor->thrs_shutdown.hysteresis);
+	nv_wr32(therm, 0x20480, sensor->thrs_shutdown.temp);
+
+	/* THRS_1 : fan boost*/
+	nv_wr32(therm, 0x204c4, sensor->thrs_fan_boost.temp);
+
+	/* THRS_2 : critical */
+	nv_wr32(therm, 0x204c0, sensor->thrs_critical.temp);
+
+	/* THRS_4 : down clock */
+	nv_wr32(therm, 0x20414, sensor->thrs_down_clock.temp);
+	spin_unlock_irqrestore(&priv->sensor.alarm_program_lock, flags);
+
+	nv_info(therm,
+		"Programmed thresholds [ %d(%d), %d(%d), %d(%d), %d(%d) ]\n",
+		sensor->thrs_fan_boost.temp, sensor->thrs_fan_boost.hysteresis,
+		sensor->thrs_down_clock.temp,
+		sensor->thrs_down_clock.hysteresis,
+		sensor->thrs_critical.temp, sensor->thrs_critical.hysteresis,
+		sensor->thrs_shutdown.temp, sensor->thrs_shutdown.hysteresis);
+
+}
+
+/* must be called with alarm_program_lock taken ! */
+static void
+nv50_therm_threshold_hyst_emulation(struct nouveau_therm *therm,
+				   uint32_t thrs_reg, u8 status_bit,
+				   const struct nvbios_therm_threshold *thrs,
+				   enum nouveau_therm_thrs thrs_name)
+{
+	enum nouveau_therm_thrs_direction direction;
+	enum nouveau_therm_thrs_state prev_state, new_state;
+	int temp, cur;
+
+	prev_state = nouveau_therm_sensor_get_threshold_state(therm, thrs_name);
+	temp = nv_rd32(therm, thrs_reg);
+
+	/* program the next threshold */
+	if (temp == thrs->temp) {
+		nv_wr32(therm, thrs_reg, thrs->temp - thrs->hysteresis);
+		new_state = NOUVEAU_THERM_THRS_HIGHER;
+	} else {
+		nv_wr32(therm, thrs_reg, thrs->temp);
+		new_state = NOUVEAU_THERM_THRS_LOWER;
+	}
+
+	/* fix the state (in case someone reprogrammed the alarms) */
+	cur = therm->temp_get(therm);
+	if (new_state == NOUVEAU_THERM_THRS_LOWER && cur > thrs->temp)
+		new_state = NOUVEAU_THERM_THRS_HIGHER;
+	else if (new_state == NOUVEAU_THERM_THRS_HIGHER &&
+		cur < thrs->temp - thrs->hysteresis)
+		new_state = NOUVEAU_THERM_THRS_LOWER;
+	nouveau_therm_sensor_set_threshold_state(therm, thrs_name, new_state);
+
+	/* find the direction */
+	if (prev_state < new_state)
+		direction = NOUVEAU_THERM_THRS_RISING;
+	else if (prev_state > new_state)
+		direction = NOUVEAU_THERM_THRS_FALLING;
+	else
+		return;
+
+	/* advertise a change in direction */
+	nouveau_therm_sensor_event(therm, thrs_name, direction);
+}
+
+static void
+nv50_therm_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_therm *therm = nouveau_therm(subdev);
+	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nvbios_therm_sensor *sensor = &priv->bios_sensor;
+	unsigned long flags;
+	uint32_t intr;
+
+	spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags);
+
+	intr = nv_rd32(therm, 0x20100);
+
+	/* THRS_4: downclock */
+	if (intr & 0x002) {
+		nv50_therm_threshold_hyst_emulation(therm, 0x20414, 24,
+						  &sensor->thrs_down_clock,
+						  NOUVEAU_THERM_THRS_DOWNCLOCK);
+		intr &= ~0x002;
+	}
+
+	/* shutdown */
+	if (intr & 0x004) {
+		nv50_therm_threshold_hyst_emulation(therm, 0x20480, 20,
+						   &sensor->thrs_shutdown,
+						   NOUVEAU_THERM_THRS_SHUTDOWN);
+		intr &= ~0x004;
+	}
+
+	/* THRS_1 : fan boost */
+	if (intr & 0x008) {
+		nv50_therm_threshold_hyst_emulation(therm, 0x204c4, 21,
+						   &sensor->thrs_fan_boost,
+						   NOUVEAU_THERM_THRS_FANBOOST);
+		intr &= ~0x008;
+	}
+
+	/* THRS_2 : critical */
+	if (intr & 0x010) {
+		nv50_therm_threshold_hyst_emulation(therm, 0x204c0, 22,
+						   &sensor->thrs_critical,
+						   NOUVEAU_THERM_THRS_CRITICAL);
+		intr &= ~0x010;
+	}
+
+	if (intr)
+		nv_error(therm, "unhandled intr 0x%08x\n", intr);
+
+	/* ACK everything */
+	nv_wr32(therm, 0x20100, 0xffffffff);
+	nv_wr32(therm, 0x1100, 0x10000); /* PBUS */
+
+	spin_unlock_irqrestore(&priv->sensor.alarm_program_lock, flags);
+}
+
 static int
 nv50_therm_ctor(struct nouveau_object *parent,
-		   struct nouveau_object *engine,
-		   struct nouveau_oclass *oclass, void *data, u32 size,
-		   struct nouveau_object **pobject)
+		struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
 {
-	struct nouveau_therm_priv *priv;
-	struct nouveau_therm *therm;
+	struct nv50_therm_priv *priv;
 	int ret;
 
 	ret = nouveau_therm_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
-	therm = (void *) priv;
 	if (ret)
 		return ret;
 
-	nouveau_therm_ic_ctor(therm);
-	nouveau_therm_sensor_ctor(therm);
-	nouveau_therm_fan_ctor(therm);
+	priv->base.base.pwm_ctrl = nv50_fan_pwm_ctrl;
+	priv->base.base.pwm_get = nv50_fan_pwm_get;
+	priv->base.base.pwm_set = nv50_fan_pwm_set;
+	priv->base.base.pwm_clock = nv50_fan_pwm_clock;
+	priv->base.base.temp_get = nv50_temp_get;
+	priv->base.sensor.program_alarms = nv50_therm_program_alarms;
+	nv_subdev(priv)->intr = nv50_therm_intr;
 
-	priv->fan.pwm_get = nv50_fan_pwm_get;
-	priv->fan.pwm_set = nv50_fan_pwm_set;
-	priv->fan.pwm_clock = nv50_fan_pwm_clock;
+	/* init the thresholds */
+	nouveau_therm_sensor_set_threshold_state(&priv->base.base,
+						 NOUVEAU_THERM_THRS_SHUTDOWN,
+						 NOUVEAU_THERM_THRS_LOWER);
+	nouveau_therm_sensor_set_threshold_state(&priv->base.base,
+						 NOUVEAU_THERM_THRS_FANBOOST,
+						 NOUVEAU_THERM_THRS_LOWER);
+	nouveau_therm_sensor_set_threshold_state(&priv->base.base,
+						 NOUVEAU_THERM_THRS_CRITICAL,
+						 NOUVEAU_THERM_THRS_LOWER);
+	nouveau_therm_sensor_set_threshold_state(&priv->base.base,
+						 NOUVEAU_THERM_THRS_DOWNCLOCK,
+						 NOUVEAU_THERM_THRS_LOWER);
 
-	therm->temp_get = nv50_temp_get;
-	therm->fan_get = nouveau_therm_fan_user_get;
-	therm->fan_set = nouveau_therm_fan_user_set;
-	therm->fan_sense = nouveau_therm_fan_sense;
-	therm->attr_get = nouveau_therm_attr_get;
-	therm->attr_set = nouveau_therm_attr_set;
-
-	return 0;
+	return nouveau_therm_preinit(&priv->base.base);
 }
 
 struct nouveau_oclass
@@ -151,7 +304,7 @@ nv50_therm_oclass = {
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_therm_ctor,
 		.dtor = _nouveau_therm_dtor,
-		.init = nouveau_therm_init,
-		.fini = nouveau_therm_fini,
+		.init = _nouveau_therm_init,
+		.fini = _nouveau_therm_fini,
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
new file mode 100644
index 000000000000..2dcc5437116a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
@@ -0,0 +1,99 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/gpio.h>
+
+#include "priv.h"
+
+struct nva3_therm_priv {
+	struct nouveau_therm_priv base;
+};
+
+int
+nva3_therm_fan_sense(struct nouveau_therm *therm)
+{
+	u32 tach = nv_rd32(therm, 0x00e728) & 0x0000ffff;
+	u32 ctrl = nv_rd32(therm, 0x00e720);
+	if (ctrl & 0x00000001)
+		return tach * 60;
+	return -ENODEV;
+}
+
+static int
+nva3_therm_init(struct nouveau_object *object)
+{
+	struct nva3_therm_priv *priv = (void *)object;
+	struct dcb_gpio_func *tach = &priv->base.fan->tach;
+	int ret;
+
+	ret = nouveau_therm_init(&priv->base.base);
+	if (ret)
+		return ret;
+
+	/* enable fan tach, count revolutions per-second */
+	nv_mask(priv, 0x00e720, 0x00000003, 0x00000002);
+	if (tach->func != DCB_GPIO_UNUSED) {
+		nv_wr32(priv, 0x00e724, nv_device(priv)->crystal * 1000);
+		nv_mask(priv, 0x00e720, 0x001f0000, tach->line << 16);
+		nv_mask(priv, 0x00e720, 0x00000001, 0x00000001);
+	}
+	nv_mask(priv, 0x00e720, 0x00000002, 0x00000000);
+
+	return 0;
+}
+
+static int
+nva3_therm_ctor(struct nouveau_object *parent,
+		struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
+{
+	struct nva3_therm_priv *priv;
+	int ret;
+
+	ret = nouveau_therm_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.base.pwm_ctrl = nv50_fan_pwm_ctrl;
+	priv->base.base.pwm_get = nv50_fan_pwm_get;
+	priv->base.base.pwm_set = nv50_fan_pwm_set;
+	priv->base.base.pwm_clock = nv50_fan_pwm_clock;
+	priv->base.base.temp_get = nv50_temp_get;
+	priv->base.base.fan_sense = nva3_therm_fan_sense;
+	priv->base.sensor.program_alarms = nouveau_therm_program_alarms_polling;
+	return nouveau_therm_preinit(&priv->base.base);
+}
+
+struct nouveau_oclass
+nva3_therm_oclass = {
+	.handle = NV_SUBDEV(THERM, 0xa3),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nva3_therm_ctor,
+		.dtor = _nouveau_therm_dtor,
+		.init = nva3_therm_init,
+		.fini = _nouveau_therm_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
new file mode 100644
index 000000000000..d7d30ee8332e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
@@ -0,0 +1,153 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+struct nvd0_therm_priv {
+	struct nouveau_therm_priv base;
+};
+
+static int
+pwm_info(struct nouveau_therm *therm, int line)
+{
+	u32 gpio = nv_rd32(therm, 0x00d610 + (line * 0x04));
+	switch (gpio & 0x000000c0) {
+	case 0x00000000: /* normal mode, possibly pwm forced off by us */
+	case 0x00000040: /* nvio special */
+		switch (gpio & 0x0000001f) {
+		case 0x19: return 1;
+		case 0x1c: return 0;
+		default:
+			break;
+		}
+	default:
+		break;
+	}
+
+	nv_error(therm, "GPIO %d unknown PWM: 0x%08x\n", line, gpio);
+	return -ENODEV;
+}
+
+static int
+nvd0_fan_pwm_ctrl(struct nouveau_therm *therm, int line, bool enable)
+{
+	u32 data = enable ? 0x00000040 : 0x00000000;
+	int indx = pwm_info(therm, line);
+	if (indx < 0)
+		return indx;
+
+	nv_mask(therm, 0x00d610 + (line * 0x04), 0x000000c0, data);
+	return 0;
+}
+
+static int
+nvd0_fan_pwm_get(struct nouveau_therm *therm, int line, u32 *divs, u32 *duty)
+{
+	int indx = pwm_info(therm, line);
+	if (indx < 0)
+		return indx;
+
+	if (nv_rd32(therm, 0x00d610 + (line * 0x04)) & 0x00000040) {
+		*divs = nv_rd32(therm, 0x00e114 + (indx * 8));
+		*duty = nv_rd32(therm, 0x00e118 + (indx * 8));
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static int
+nvd0_fan_pwm_set(struct nouveau_therm *therm, int line, u32 divs, u32 duty)
+{
+	int indx = pwm_info(therm, line);
+	if (indx < 0)
+		return indx;
+
+	nv_wr32(therm, 0x00e114 + (indx * 8), divs);
+	nv_wr32(therm, 0x00e118 + (indx * 8), duty | 0x80000000);
+	return 0;
+}
+
+static int
+nvd0_fan_pwm_clock(struct nouveau_therm *therm)
+{
+	return (nv_device(therm)->crystal * 1000) / 20;
+}
+
+static int
+nvd0_therm_init(struct nouveau_object *object)
+{
+	struct nvd0_therm_priv *priv = (void *)object;
+	int ret;
+
+	ret = nouveau_therm_init(&priv->base.base);
+	if (ret)
+		return ret;
+
+	/* enable fan tach, count revolutions per-second */
+	nv_mask(priv, 0x00e720, 0x00000003, 0x00000002);
+	if (priv->base.fan->tach.func != DCB_GPIO_UNUSED) {
+		nv_mask(priv, 0x00d79c, 0x000000ff, priv->base.fan->tach.line);
+		nv_wr32(priv, 0x00e724, nv_device(priv)->crystal * 1000);
+		nv_mask(priv, 0x00e720, 0x00000001, 0x00000001);
+	}
+	nv_mask(priv, 0x00e720, 0x00000002, 0x00000000);
+
+	return 0;
+}
+
+static int
+nvd0_therm_ctor(struct nouveau_object *parent,
+		struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
+{
+	struct nvd0_therm_priv *priv;
+	int ret;
+
+	ret = nouveau_therm_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.base.pwm_ctrl = nvd0_fan_pwm_ctrl;
+	priv->base.base.pwm_get = nvd0_fan_pwm_get;
+	priv->base.base.pwm_set = nvd0_fan_pwm_set;
+	priv->base.base.pwm_clock = nvd0_fan_pwm_clock;
+	priv->base.base.temp_get = nv50_temp_get;
+	priv->base.base.fan_sense = nva3_therm_fan_sense;
+	priv->base.sensor.program_alarms = nouveau_therm_program_alarms_polling;
+	return nouveau_therm_preinit(&priv->base.base);
+}
+
+struct nouveau_oclass
+nvd0_therm_oclass = {
+	.handle = NV_SUBDEV(THERM, 0xd0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvd0_therm_ctor,
+		.dtor = _nouveau_therm_dtor,
+		.init = nvd0_therm_init,
+		.fini = _nouveau_therm_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
index 1c3cd6abc36e..06b98706b3fc 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
@@ -1,3 +1,6 @@
+#ifndef __NVTHERM_PRIV_H__
+#define __NVTHERM_PRIV_H__
+
 /*
  * Copyright 2012 The Nouveau community
  *
@@ -25,33 +28,81 @@
 #include <subdev/therm.h>
 
 #include <subdev/bios/extdev.h>
+#include <subdev/bios/gpio.h>
 #include <subdev/bios/perf.h>
 #include <subdev/bios/therm.h>
+#include <subdev/timer.h>
+
+struct nouveau_fan {
+	struct nouveau_therm *parent;
+	const char *type;
+
+	struct nvbios_therm_fan bios;
+	struct nvbios_perf_fan perf;
+
+	struct nouveau_alarm alarm;
+	spinlock_t lock;
+	int percent;
+
+	int (*get)(struct nouveau_therm *therm);
+	int (*set)(struct nouveau_therm *therm, int percent);
+
+	struct dcb_gpio_func tach;
+};
+
+enum nouveau_therm_thrs_direction {
+	NOUVEAU_THERM_THRS_FALLING = 0,
+	NOUVEAU_THERM_THRS_RISING = 1
+};
+
+enum nouveau_therm_thrs_state {
+	NOUVEAU_THERM_THRS_LOWER = 0,
+	NOUVEAU_THERM_THRS_HIGHER = 1
+};
+
+enum nouveau_therm_thrs {
+	NOUVEAU_THERM_THRS_FANBOOST = 0,
+	NOUVEAU_THERM_THRS_DOWNCLOCK = 1,
+	NOUVEAU_THERM_THRS_CRITICAL = 2,
+	NOUVEAU_THERM_THRS_SHUTDOWN = 3,
+	NOUVEAU_THERM_THRS_NR
+};
 
 struct nouveau_therm_priv {
 	struct nouveau_therm base;
 
+	/* automatic thermal management */
+	struct nouveau_alarm alarm;
+	spinlock_t lock;
+	struct nouveau_therm_trip_point *last_trip;
+	int mode;
+	int suspend;
+
 	/* bios */
 	struct nvbios_therm_sensor bios_sensor;
-	struct nvbios_therm_fan bios_fan;
-	struct nvbios_perf_fan bios_perf_fan;
 
 	/* fan priv */
+	struct nouveau_fan *fan;
+
+	/* alarms priv */
 	struct {
-		enum nouveau_therm_fan_mode mode;
-		int percent;
+		spinlock_t alarm_program_lock;
+		struct nouveau_alarm therm_poll_alarm;
+		enum nouveau_therm_thrs_state alarm_state[NOUVEAU_THERM_THRS_NR];
+		void (*program_alarms)(struct nouveau_therm *);
+	} sensor;
 
-		int (*pwm_get)(struct nouveau_therm *, int line, u32*, u32*);
-		int (*pwm_set)(struct nouveau_therm *, int line, u32, u32);
-		int (*pwm_clock)(struct nouveau_therm *);
-	} fan;
+	/* what should be done if the card overheats */
+	struct {
+		void (*downclock)(struct nouveau_therm *, bool active);
+		void (*pause)(struct nouveau_therm *, bool active);
+	} emergency;
 
 	/* ic */
 	struct i2c_client *ic;
 };
 
-int nouveau_therm_init(struct nouveau_object *object);
-int nouveau_therm_fini(struct nouveau_object *object, bool suspend);
+int nouveau_therm_mode(struct nouveau_therm *therm, int mode);
 int nouveau_therm_attr_get(struct nouveau_therm *therm,
 		       enum nouveau_therm_attr_type type);
 int nouveau_therm_attr_set(struct nouveau_therm *therm,
@@ -63,11 +114,35 @@ int nouveau_therm_sensor_ctor(struct nouveau_therm *therm);
 
 int nouveau_therm_fan_ctor(struct nouveau_therm *therm);
 int nouveau_therm_fan_get(struct nouveau_therm *therm);
-int nouveau_therm_fan_set(struct nouveau_therm *therm, int percent);
+int nouveau_therm_fan_set(struct nouveau_therm *therm, bool now, int percent);
 int nouveau_therm_fan_user_get(struct nouveau_therm *therm);
 int nouveau_therm_fan_user_set(struct nouveau_therm *therm, int percent);
-int nouveau_therm_fan_set_mode(struct nouveau_therm *therm,
-			   enum nouveau_therm_fan_mode mode);
-
 
 int nouveau_therm_fan_sense(struct nouveau_therm *therm);
+
+int nouveau_therm_preinit(struct nouveau_therm *);
+
+void nouveau_therm_sensor_set_threshold_state(struct nouveau_therm *therm,
+					     enum nouveau_therm_thrs thrs,
+					     enum nouveau_therm_thrs_state st);
+enum nouveau_therm_thrs_state
+nouveau_therm_sensor_get_threshold_state(struct nouveau_therm *therm,
+					 enum nouveau_therm_thrs thrs);
+void nouveau_therm_sensor_event(struct nouveau_therm *therm,
+			        enum nouveau_therm_thrs thrs,
+			        enum nouveau_therm_thrs_direction dir);
+void nouveau_therm_program_alarms_polling(struct nouveau_therm *therm);
+
+int nv50_fan_pwm_ctrl(struct nouveau_therm *, int, bool);
+int nv50_fan_pwm_get(struct nouveau_therm *, int, u32 *, u32 *);
+int nv50_fan_pwm_set(struct nouveau_therm *, int, u32, u32);
+int nv50_fan_pwm_clock(struct nouveau_therm *);
+int nv50_temp_get(struct nouveau_therm *therm);
+
+int nva3_therm_fan_sense(struct nouveau_therm *);
+
+int nouveau_fanpwm_create(struct nouveau_therm *, struct dcb_gpio_func *);
+int nouveau_fantog_create(struct nouveau_therm *, struct dcb_gpio_func *);
+int nouveau_fannil_create(struct nouveau_therm *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
index 204282301fb1..b37624af8297 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
@@ -58,11 +58,171 @@ static void
 nouveau_therm_temp_safety_checks(struct nouveau_therm *therm)
 {
 	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nvbios_therm_sensor *s = &priv->bios_sensor;
 
 	if (!priv->bios_sensor.slope_div)
 		priv->bios_sensor.slope_div = 1;
 	if (!priv->bios_sensor.offset_den)
 		priv->bios_sensor.offset_den = 1;
+
+	/* enforce a minimum hysteresis on thresholds */
+	s->thrs_fan_boost.hysteresis = max_t(u8, s->thrs_fan_boost.hysteresis, 2);
+	s->thrs_down_clock.hysteresis = max_t(u8, s->thrs_down_clock.hysteresis, 2);
+	s->thrs_critical.hysteresis = max_t(u8, s->thrs_critical.hysteresis, 2);
+	s->thrs_shutdown.hysteresis = max_t(u8, s->thrs_shutdown.hysteresis, 2);
+}
+
+/* must be called with alarm_program_lock taken ! */
+void nouveau_therm_sensor_set_threshold_state(struct nouveau_therm *therm,
+					     enum nouveau_therm_thrs thrs,
+					     enum nouveau_therm_thrs_state st)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	priv->sensor.alarm_state[thrs] = st;
+}
+
+/* must be called with alarm_program_lock taken ! */
+enum nouveau_therm_thrs_state
+nouveau_therm_sensor_get_threshold_state(struct nouveau_therm *therm,
+					 enum nouveau_therm_thrs thrs)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	return priv->sensor.alarm_state[thrs];
+}
+
+static void
+nv_poweroff_work(struct work_struct *work)
+{
+	orderly_poweroff(true);
+	kfree(work);
+}
+
+void nouveau_therm_sensor_event(struct nouveau_therm *therm,
+			        enum nouveau_therm_thrs thrs,
+			        enum nouveau_therm_thrs_direction dir)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	bool active;
+	const char *thresolds[] = {
+		"fanboost", "downclock", "critical", "shutdown"
+	};
+	uint8_t temperature = therm->temp_get(therm);
+
+	if (thrs < 0 || thrs > 3)
+		return;
+
+	if (dir == NOUVEAU_THERM_THRS_FALLING)
+		nv_info(therm, "temperature (%u C) went below the '%s' threshold\n",
+			temperature, thresolds[thrs]);
+	else
+		nv_info(therm, "temperature (%u C) hit the '%s' threshold\n",
+			temperature, thresolds[thrs]);
+
+	active = (dir == NOUVEAU_THERM_THRS_RISING);
+	switch (thrs) {
+	case NOUVEAU_THERM_THRS_FANBOOST:
+		if (active) {
+			nouveau_therm_fan_set(therm, true, 100);
+			nouveau_therm_mode(therm, NOUVEAU_THERM_CTRL_AUTO);
+		}
+		break;
+	case NOUVEAU_THERM_THRS_DOWNCLOCK:
+		if (priv->emergency.downclock)
+			priv->emergency.downclock(therm, active);
+		break;
+	case NOUVEAU_THERM_THRS_CRITICAL:
+		if (priv->emergency.pause)
+			priv->emergency.pause(therm, active);
+		break;
+	case NOUVEAU_THERM_THRS_SHUTDOWN:
+		if (active) {
+			struct work_struct *work;
+
+			work = kmalloc(sizeof(*work), GFP_ATOMIC);
+			if (work) {
+				INIT_WORK(work, nv_poweroff_work);
+				schedule_work(work);
+			}
+		}
+		break;
+	case NOUVEAU_THERM_THRS_NR:
+		break;
+	}
+
+}
+
+/* must be called with alarm_program_lock taken ! */
+static void
+nouveau_therm_threshold_hyst_polling(struct nouveau_therm *therm,
+				   const struct nvbios_therm_threshold *thrs,
+				   enum nouveau_therm_thrs thrs_name)
+{
+	enum nouveau_therm_thrs_direction direction;
+	enum nouveau_therm_thrs_state prev_state, new_state;
+	int temp = therm->temp_get(therm);
+
+	prev_state = nouveau_therm_sensor_get_threshold_state(therm, thrs_name);
+
+	if (temp >= thrs->temp && prev_state == NOUVEAU_THERM_THRS_LOWER) {
+		direction = NOUVEAU_THERM_THRS_RISING;
+		new_state = NOUVEAU_THERM_THRS_HIGHER;
+	} else if (temp <= thrs->temp - thrs->hysteresis &&
+			prev_state == NOUVEAU_THERM_THRS_HIGHER) {
+		direction = NOUVEAU_THERM_THRS_FALLING;
+		new_state = NOUVEAU_THERM_THRS_LOWER;
+	} else
+		return; /* nothing to do */
+
+	nouveau_therm_sensor_set_threshold_state(therm, thrs_name, new_state);
+	nouveau_therm_sensor_event(therm, thrs_name, direction);
+}
+
+static void
+alarm_timer_callback(struct nouveau_alarm *alarm)
+{
+	struct nouveau_therm_priv *priv =
+	container_of(alarm, struct nouveau_therm_priv, sensor.therm_poll_alarm);
+	struct nvbios_therm_sensor *sensor = &priv->bios_sensor;
+	struct nouveau_timer *ptimer = nouveau_timer(priv);
+	struct nouveau_therm *therm = &priv->base;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags);
+
+	nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_fan_boost,
+					     NOUVEAU_THERM_THRS_FANBOOST);
+
+	nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_down_clock,
+					     NOUVEAU_THERM_THRS_DOWNCLOCK);
+
+	nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_critical,
+					     NOUVEAU_THERM_THRS_CRITICAL);
+
+	nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_shutdown,
+					     NOUVEAU_THERM_THRS_SHUTDOWN);
+
+	/* schedule the next poll in one second */
+	if (list_empty(&alarm->head))
+		ptimer->alarm(ptimer, 1000 * 1000 * 1000, alarm);
+
+	spin_unlock_irqrestore(&priv->sensor.alarm_program_lock, flags);
+}
+
+void
+nouveau_therm_program_alarms_polling(struct nouveau_therm *therm)
+{
+	struct nouveau_therm_priv *priv = (void *)therm;
+	struct nvbios_therm_sensor *sensor = &priv->bios_sensor;
+
+	nv_info(therm,
+		"programmed thresholds [ %d(%d), %d(%d), %d(%d), %d(%d) ]\n",
+		sensor->thrs_fan_boost.temp, sensor->thrs_fan_boost.hysteresis,
+		sensor->thrs_down_clock.temp,
+		sensor->thrs_down_clock.hysteresis,
+		sensor->thrs_critical.temp, sensor->thrs_critical.hysteresis,
+		sensor->thrs_shutdown.temp, sensor->thrs_shutdown.hysteresis);
+
+	alarm_timer_callback(&priv->sensor.therm_poll_alarm);
 }
 
 int
@@ -71,6 +231,8 @@ nouveau_therm_sensor_ctor(struct nouveau_therm *therm)
 	struct nouveau_therm_priv *priv = (void *)therm;
 	struct nouveau_bios *bios = nouveau_bios(therm);
 
+	nouveau_alarm_init(&priv->sensor.therm_poll_alarm, alarm_timer_callback);
+
 	nouveau_therm_temp_set_defaults(therm);
 	if (nvbios_therm_sensor_parse(bios, NVBIOS_THERM_DOMAIN_CORE,
 				      &priv->bios_sensor))
diff --git a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
index c26ca9bef671..8e1bae4f12e8 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
@@ -79,7 +79,7 @@ nv04_timer_alarm_trigger(struct nouveau_timer *ptimer)
 
 	/* execute any pending alarm handlers */
 	list_for_each_entry_safe(alarm, atemp, &exec, head) {
-		list_del(&alarm->head);
+		list_del_init(&alarm->head);
 		alarm->func(alarm);
 	}
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.h b/drivers/gpu/drm/nouveau/nouveau_acpi.h
index d0da230d7706..74acf0f87785 100644
--- a/drivers/gpu/drm/nouveau/nouveau_acpi.h
+++ b/drivers/gpu/drm/nouveau/nouveau_acpi.h
@@ -3,7 +3,7 @@
 
 #define ROM_BIOS_PAGE 4096
 
-#if defined(CONFIG_ACPI)
+#if defined(CONFIG_ACPI) && defined(CONFIG_X86)
 bool nouveau_is_optimus(void);
 bool nouveau_is_v1_dsm(void);
 void nouveau_register_dsm_handler(void);
diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c
index f65b20a375f6..5d940302d2aa 100644
--- a/drivers/gpu/drm/nouveau/nouveau_backlight.c
+++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c
@@ -84,6 +84,8 @@ nv40_backlight_init(struct drm_connector *connector)
 	props.max_brightness = 31;
 	bd = backlight_device_register("nv_backlight", &connector->kdev, drm,
 				       &nv40_bl_ops, &props);
+	if (IS_ERR(bd))
+		return PTR_ERR(bd);
 	drm->backlight = bd;
 	bd->props.brightness = nv40_get_intensity(bd);
 	backlight_update_status(bd);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 865eddfa30a7..50a6dd02f7c5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -678,23 +678,6 @@ int run_tmds_table(struct drm_device *dev, struct dcb_output *dcbent, int head,
 	return 0;
 }
 
-static void parse_bios_version(struct drm_device *dev, struct nvbios *bios, uint16_t offset)
-{
-	/*
-	 * offset + 0  (8 bits): Micro version
-	 * offset + 1  (8 bits): Minor version
-	 * offset + 2  (8 bits): Chip version
-	 * offset + 3  (8 bits): Major version
-	 */
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	bios->major_version = bios->data[offset + 3];
-	bios->chip_version = bios->data[offset + 2];
-	NV_INFO(drm, "Bios version %02x.%02x.%02x.%02x\n",
-		 bios->data[offset + 3], bios->data[offset + 2],
-		 bios->data[offset + 1], bios->data[offset]);
-}
-
 static void parse_script_table_pointers(struct nvbios *bios, uint16_t offset)
 {
 	/*
@@ -710,12 +693,6 @@ static void parse_script_table_pointers(struct nvbios *bios, uint16_t offset)
 	 */
 
 	bios->init_script_tbls_ptr = ROM16(bios->data[offset]);
-	bios->macro_index_tbl_ptr = ROM16(bios->data[offset + 2]);
-	bios->macro_tbl_ptr = ROM16(bios->data[offset + 4]);
-	bios->condition_tbl_ptr = ROM16(bios->data[offset + 6]);
-	bios->io_condition_tbl_ptr = ROM16(bios->data[offset + 8]);
-	bios->io_flag_condition_tbl_ptr = ROM16(bios->data[offset + 10]);
-	bios->init_function_tbl_ptr = ROM16(bios->data[offset + 12]);
 }
 
 static int parse_bit_A_tbl_entry(struct drm_device *dev, struct nvbios *bios, struct bit_entry *bitentry)
@@ -765,25 +742,6 @@ static int parse_bit_A_tbl_entry(struct drm_device *dev, struct nvbios *bios, st
 	return 0;
 }
 
-static int parse_bit_C_tbl_entry(struct drm_device *dev, struct nvbios *bios, struct bit_entry *bitentry)
-{
-	/*
-	 * offset + 8  (16 bits): PLL limits table pointer
-	 *
-	 * There's more in here, but that's unknown.
-	 */
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	if (bitentry->length < 10) {
-		NV_ERROR(drm, "Do not understand BIT C table\n");
-		return -EINVAL;
-	}
-
-	bios->pll_limit_tbl_ptr = ROM16(bios->data[bitentry->offset + 8]);
-
-	return 0;
-}
-
 static int parse_bit_display_tbl_entry(struct drm_device *dev, struct nvbios *bios, struct bit_entry *bitentry)
 {
 	/*
@@ -821,12 +779,6 @@ static int parse_bit_init_tbl_entry(struct drm_device *dev, struct nvbios *bios,
 	}
 
 	parse_script_table_pointers(bios, bitentry->offset);
-
-	if (bitentry->length >= 16)
-		bios->some_script_ptr = ROM16(bios->data[bitentry->offset + 14]);
-	if (bitentry->length >= 18)
-		bios->init96_tbl_ptr = ROM16(bios->data[bitentry->offset + 16]);
-
 	return 0;
 }
 
@@ -852,8 +804,6 @@ static int parse_bit_i_tbl_entry(struct drm_device *dev, struct nvbios *bios, st
 		return -EINVAL;
 	}
 
-	parse_bios_version(dev, bios, bitentry->offset);
-
 	/*
 	 * bit 4 seems to indicate a mobile bios (doesn't suffer from BMP's
 	 * Quadro identity crisis), other bits possibly as for BMP feature byte
@@ -1078,9 +1028,6 @@ parse_bit_structure(struct nvbios *bios, const uint16_t bitoffset)
 		return ret;
 	if (bios->major_version >= 0x60) /* g80+ */
 		parse_bit_table(bios, bitoffset, &BIT_TABLE('A', A));
-	ret = parse_bit_table(bios, bitoffset, &BIT_TABLE('C', C));
-	if (ret)
-		return ret;
 	parse_bit_table(bios, bitoffset, &BIT_TABLE('D', display));
 	ret = parse_bit_table(bios, bitoffset, &BIT_TABLE('I', init));
 	if (ret)
@@ -1228,8 +1175,6 @@ static int parse_bmp_structure(struct drm_device *dev, struct nvbios *bios, unsi
 	 */
 	bios->feature_byte = bmp[9];
 
-	parse_bios_version(dev, bios, offset + 10);
-
 	if (bmp_version_major < 5 || bmp_version_minor < 0x10)
 		bios->old_style_init = true;
 	legacy_scripts_offset = 18;
@@ -1276,8 +1221,10 @@ static int parse_bmp_structure(struct drm_device *dev, struct nvbios *bios, unsi
 		bios->fp.lvdsmanufacturerpointer = ROM16(bmp[117]);
 		bios->fp.fpxlatemanufacturertableptr = ROM16(bmp[119]);
 	}
+#if 0
 	if (bmplength > 143)
 		bios->pll_limit_tbl_ptr = ROM16(bmp[142]);
+#endif
 
 	if (bmplength > 157)
 		bios->fp.duallink_transition_clk = ROM16(bmp[156]) * 10;
@@ -1522,6 +1469,7 @@ parse_dcb20_entry(struct drm_device *dev, struct dcb_table *dcb,
 	}
 	case DCB_OUTPUT_DP:
 		entry->dpconf.sor.link = (conf & 0x00000030) >> 4;
+		entry->extdev = (conf & 0x0000ff00) >> 8;
 		switch ((conf & 0x00e00000) >> 21) {
 		case 0:
 			entry->dpconf.link_bw = 162000;
@@ -1543,8 +1491,10 @@ parse_dcb20_entry(struct drm_device *dev, struct dcb_table *dcb,
 		}
 		break;
 	case DCB_OUTPUT_TMDS:
-		if (dcb->version >= 0x40)
+		if (dcb->version >= 0x40) {
 			entry->tmdsconf.sor.link = (conf & 0x00000030) >> 4;
+			entry->extdev = (conf & 0x0000ff00) >> 8;
+		}
 		else if (dcb->version >= 0x30)
 			entry->tmdsconf.slave_addr = (conf & 0x00000700) >> 8;
 		else if (dcb->version >= 0x22)
@@ -1937,9 +1887,9 @@ parse_dcb_table(struct drm_device *dev, struct nvbios *bios)
 		if (conn[0] != 0xff) {
 			NV_INFO(drm, "DCB conn %02d: ", idx);
 			if (olddcb_conntab(dev)[3] < 4)
-				printk("%04x\n", ROM16(conn[0]));
+				pr_cont("%04x\n", ROM16(conn[0]));
 			else
-				printk("%08x\n", ROM32(conn[0]));
+				pr_cont("%08x\n", ROM32(conn[0]));
 		}
 	}
 	dcb_fake_connectors(bios);
@@ -2052,45 +2002,29 @@ uint8_t *nouveau_bios_embedded_edid(struct drm_device *dev)
 static bool NVInitVBIOS(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-
-	memset(bios, 0, sizeof(struct nvbios));
-	spin_lock_init(&bios->lock);
-	bios->dev = dev;
-
-	bios->data = nouveau_bios(drm->device)->data;
-	bios->length = nouveau_bios(drm->device)->size;
-	return true;
-}
+	struct nouveau_bios *bios = nouveau_bios(drm->device);
+	struct nvbios *legacy = &drm->vbios;
+
+	memset(legacy, 0, sizeof(struct nvbios));
+	spin_lock_init(&legacy->lock);
+	legacy->dev = dev;
+
+	legacy->data = bios->data;
+	legacy->length = bios->size;
+	legacy->major_version = bios->version.major;
+	legacy->chip_version = bios->version.chip;
+	if (bios->bit_offset) {
+		legacy->type = NVBIOS_BIT;
+		legacy->offset = bios->bit_offset;
+		return !parse_bit_structure(legacy, legacy->offset + 6);
+	} else
+	if (bios->bmp_offset) {
+		legacy->type = NVBIOS_BMP;
+		legacy->offset = bios->bmp_offset;
+		return !parse_bmp_structure(dev, legacy, legacy->offset);
+	}
 
-static int nouveau_parse_vbios_struct(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-	const uint8_t bit_signature[] = { 0xff, 0xb8, 'B', 'I', 'T' };
-	const uint8_t bmp_signature[] = { 0xff, 0x7f, 'N', 'V', 0x0 };
-	int offset;
-
-	offset = findstr(bios->data, bios->length,
-					bit_signature, sizeof(bit_signature));
-	if (offset) {
-		NV_INFO(drm, "BIT BIOS found\n");
-		bios->type = NVBIOS_BIT;
-		bios->offset = offset;
-		return parse_bit_structure(bios, offset + 6);
-	}
-
-	offset = findstr(bios->data, bios->length,
-					bmp_signature, sizeof(bmp_signature));
-	if (offset) {
-		NV_INFO(drm, "BMP BIOS found\n");
-		bios->type = NVBIOS_BMP;
-		bios->offset = offset;
-		return parse_bmp_structure(dev, bios, offset);
-	}
-
-	NV_ERROR(drm, "No known BIOS signature found\n");
-	return -ENODEV;
+	return false;
 }
 
 int
@@ -2146,10 +2080,6 @@ nouveau_bios_init(struct drm_device *dev)
 	if (!NVInitVBIOS(dev))
 		return -ENODEV;
 
-	ret = nouveau_parse_vbios_struct(dev);
-	if (ret)
-		return ret;
-
 	ret = parse_dcb_table(dev, bios);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.h b/drivers/gpu/drm/nouveau/nouveau_bios.h
index f68c54ca422f..7ccd28f11adf 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.h
@@ -107,20 +107,10 @@ struct nvbios {
 	bool old_style_init;
 	uint16_t init_script_tbls_ptr;
 	uint16_t extra_init_script_tbl_ptr;
-	uint16_t macro_index_tbl_ptr;
-	uint16_t macro_tbl_ptr;
-	uint16_t condition_tbl_ptr;
-	uint16_t io_condition_tbl_ptr;
-	uint16_t io_flag_condition_tbl_ptr;
-	uint16_t init_function_tbl_ptr;
-
-	uint16_t pll_limit_tbl_ptr;
+
 	uint16_t ram_restrict_tbl_ptr;
 	uint8_t ram_restrict_group_count;
 
-	uint16_t some_script_ptr; /* BIT I + 14 */
-	uint16_t init96_tbl_ptr; /* BIT I + 16 */
-
 	struct dcb_table dcb;
 
 	struct {
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 1699a9083a2f..11ca82148edc 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -301,17 +301,18 @@ nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype)
 	struct ttm_buffer_object *bo = &nvbo->bo;
 	int ret;
 
+	ret = ttm_bo_reserve(bo, false, false, false, 0);
+	if (ret)
+		goto out;
+
 	if (nvbo->pin_refcnt && !(memtype & (1 << bo->mem.mem_type))) {
 		NV_ERROR(drm, "bo %p pinned elsewhere: 0x%08x vs 0x%08x\n", bo,
 			 1 << bo->mem.mem_type, memtype);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 
 	if (nvbo->pin_refcnt++)
-		return 0;
-
-	ret = ttm_bo_reserve(bo, false, false, false, 0);
-	if (ret)
 		goto out;
 
 	nouveau_bo_placement_set(nvbo, memtype, 0);
@@ -329,10 +330,8 @@ nouveau_bo_pin(struct nouveau_bo *nvbo, uint32_t memtype)
 			break;
 		}
 	}
-	ttm_bo_unreserve(bo);
 out:
-	if (unlikely(ret))
-		nvbo->pin_refcnt--;
+	ttm_bo_unreserve(bo);
 	return ret;
 }
 
@@ -343,13 +342,13 @@ nouveau_bo_unpin(struct nouveau_bo *nvbo)
 	struct ttm_buffer_object *bo = &nvbo->bo;
 	int ret;
 
-	if (--nvbo->pin_refcnt)
-		return 0;
-
 	ret = ttm_bo_reserve(bo, false, false, false, 0);
 	if (ret)
 		return ret;
 
+	if (--nvbo->pin_refcnt)
+		goto out;
+
 	nouveau_bo_placement_set(nvbo, bo->mem.placement, 0);
 
 	ret = nouveau_bo_validate(nvbo, false, false);
@@ -366,6 +365,7 @@ nouveau_bo_unpin(struct nouveau_bo *nvbo)
 		}
 	}
 
+out:
 	ttm_bo_unreserve(bo);
 	return ret;
 }
@@ -562,7 +562,7 @@ nouveau_bo_move_accel_cleanup(struct nouveau_channel *chan,
 	struct nouveau_fence *fence = NULL;
 	int ret;
 
-	ret = nouveau_fence_new(chan, &fence);
+	ret = nouveau_fence_new(chan, false, &fence);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 25ca37989d2c..653dbbbd4fa1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -28,10 +28,11 @@ struct nouveau_bo {
 	struct nouveau_drm_tile *tile;
 
 	struct drm_gem_object *gem;
+
+	/* protect by the ttm reservation lock */
 	int pin_refcnt;
 
 	struct ttm_bo_kmap_obj dma_buf_vmap;
-	int vmapping_count;
 };
 
 static inline struct nouveau_bo *
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index 174300b6a02e..eaa80a2b81ee 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -51,14 +51,15 @@ nouveau_channel_idle(struct nouveau_channel *chan)
 	struct nouveau_fence *fence = NULL;
 	int ret;
 
-	ret = nouveau_fence_new(chan, &fence);
+	ret = nouveau_fence_new(chan, false, &fence);
 	if (!ret) {
 		ret = nouveau_fence_wait(fence, false, false);
 		nouveau_fence_unref(&fence);
 	}
 
 	if (ret)
-		NV_ERROR(cli, "failed to idle channel 0x%08x\n", chan->handle);
+		NV_ERROR(cli, "failed to idle channel 0x%08x [%s]\n",
+			 chan->handle, cli->base.name);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index e620ba8271b4..4dd7ae2ac6c6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -55,8 +55,6 @@ MODULE_PARM_DESC(duallink, "Allow dual-link TMDS (default: enabled)");
 static int nouveau_duallink = 1;
 module_param_named(duallink, nouveau_duallink, int, 0400);
 
-static void nouveau_connector_hotplug(void *, int);
-
 struct nouveau_encoder *
 find_encoder(struct drm_connector *connector, int type)
 {
@@ -100,22 +98,6 @@ static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
-	struct nouveau_gpio *gpio;
-	struct nouveau_drm *drm;
-	struct drm_device *dev;
-
-	if (!nv_connector)
-		return;
-
-	dev  = nv_connector->base.dev;
-	drm  = nouveau_drm(dev);
-	gpio = nouveau_gpio(drm->device);
-
-	if (gpio && nv_connector->hpd != DCB_GPIO_UNUSED) {
-		gpio->isr_del(gpio, 0, nv_connector->hpd, 0xff,
-			      nouveau_connector_hotplug, connector);
-	}
-
 	kfree(nv_connector->edid);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
@@ -130,7 +112,6 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_i2c_port *port = NULL;
 	int i, panel = -ENODEV;
 
@@ -160,8 +141,7 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 			continue;
 		nv_encoder = nouveau_encoder(obj_to_encoder(obj));
 
-		if (nv_encoder->dcb->i2c_index < 0xf)
-			port = i2c->find(i2c, nv_encoder->dcb->i2c_index);
+		port = nv_encoder->i2c;
 		if (port && nv_probe_i2c(port, 0x50)) {
 			*pnv_encoder = nv_encoder;
 			break;
@@ -399,9 +379,10 @@ nouveau_connector_detect_lvds(struct drm_connector *connector, bool force)
 		struct edid *edid =
 			(struct edid *)nouveau_bios_embedded_edid(dev);
 		if (edid) {
-			nv_connector->edid = kmalloc(EDID_LENGTH, GFP_KERNEL);
-			*(nv_connector->edid) = *edid;
-			status = connector_status_connected;
+			nv_connector->edid =
+					kmemdup(edid, EDID_LENGTH, GFP_KERNEL);
+			if (nv_connector->edid)
+				status = connector_status_connected;
 		}
 	}
 
@@ -911,6 +892,37 @@ nouveau_connector_funcs_lvds = {
 	.force = nouveau_connector_force
 };
 
+static void
+nouveau_connector_hotplug_work(struct work_struct *work)
+{
+	struct nouveau_connector *nv_connector =
+		container_of(work, struct nouveau_connector, hpd_work);
+	struct drm_connector *connector = &nv_connector->base;
+	struct drm_device *dev = connector->dev;
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
+	bool plugged = gpio->get(gpio, 0, nv_connector->hpd.func, 0xff);
+
+	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un",
+		 drm_get_connector_name(connector));
+
+	if (plugged)
+		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
+	else
+		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+
+	drm_helper_hpd_irq_event(dev);
+}
+
+static int
+nouveau_connector_hotplug(struct nouveau_eventh *event, int index)
+{
+	struct nouveau_connector *nv_connector =
+		container_of(event, struct nouveau_connector, hpd_func);
+	schedule_work(&nv_connector->hpd_work);
+	return NVKM_EVENT_KEEP;
+}
+
 static int
 drm_conntype_from_dcb(enum dcb_connector_type dcb)
 {
@@ -961,6 +973,7 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		return ERR_PTR(-ENOMEM);
 
 	connector = &nv_connector->base;
+	INIT_WORK(&nv_connector->hpd_work, nouveau_connector_hotplug_work);
 	nv_connector->index = index;
 
 	/* attempt to parse vbios connector type and hotplug gpio */
@@ -975,8 +988,11 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		if (olddcb_conntab(dev)[3] >= 4)
 			entry |= (u32)ROM16(nv_connector->dcb[2]) << 16;
 
-		nv_connector->hpd = ffs((entry & 0x07033000) >> 12);
-		nv_connector->hpd = hpd[nv_connector->hpd];
+		ret = gpio->find(gpio, 0, hpd[ffs((entry & 0x07033000) >> 12)],
+				 DCB_GPIO_UNUSED, &nv_connector->hpd);
+		nv_connector->hpd_func.func = nouveau_connector_hotplug;
+		if (ret)
+			nv_connector->hpd.func = DCB_GPIO_UNUSED;
 
 		nv_connector->type = nv_connector->dcb[0];
 		if (drm_conntype_from_dcb(nv_connector->type) ==
@@ -999,7 +1015,7 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 	} else {
 		nv_connector->type = DCB_CONNECTOR_NONE;
-		nv_connector->hpd = DCB_GPIO_UNUSED;
+		nv_connector->hpd.func = DCB_GPIO_UNUSED;
 	}
 
 	/* no vbios data, or an unknown dcb connector type - attempt to
@@ -1126,31 +1142,9 @@ nouveau_connector_create(struct drm_device *dev, int index)
 	}
 
 	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-	if (gpio && nv_connector->hpd != DCB_GPIO_UNUSED) {
-		ret = gpio->isr_add(gpio, 0, nv_connector->hpd, 0xff,
-				    nouveau_connector_hotplug, connector);
-		if (ret == 0)
-			connector->polled = DRM_CONNECTOR_POLL_HPD;
-	}
+	if (nv_connector->hpd.func != DCB_GPIO_UNUSED)
+		connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	drm_sysfs_connector_add(connector);
 	return connector;
 }
-
-static void
-nouveau_connector_hotplug(void *data, int plugged)
-{
-	struct drm_connector *connector = data;
-	struct drm_device *dev = connector->dev;
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un",
-		 drm_get_connector_name(connector));
-
-	if (plugged)
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
-	else
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
-
-	drm_helper_hpd_irq_event(dev);
-}
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index 20eb84cce9e6..6e399aad491a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -30,6 +30,11 @@
 #include <drm/drm_edid.h>
 #include "nouveau_crtc.h"
 
+#include <core/event.h>
+
+#include <subdev/bios.h>
+#include <subdev/bios/gpio.h>
+
 struct nouveau_i2c_port;
 
 enum nouveau_underscan_type {
@@ -61,7 +66,10 @@ struct nouveau_connector {
 	enum dcb_connector_type type;
 	u8 index;
 	u8 *dcb;
-	u8 hpd;
+
+	struct dcb_gpio_func hpd;
+	struct work_struct hpd_work;
+	struct nouveau_eventh hpd_func;
 
 	int dithering_mode;
 	int dithering_depth;
diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
new file mode 100644
index 000000000000..5392e07edfc6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -0,0 +1,64 @@
+/*
+ * Copyright (C) 2009 Red Hat <bskeggs@redhat.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining
+ * a copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sublicense, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial
+ * portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS SUPPLIERS BE
+ * LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+ * OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+ * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+/*
+ * Authors:
+ *  Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include "nouveau_debugfs.h"
+#include "nouveau_drm.h"
+
+static int
+nouveau_debugfs_vbios_image(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct nouveau_drm *drm = nouveau_drm(node->minor->dev);
+	int i;
+
+	for (i = 0; i < drm->vbios.length; i++)
+		seq_printf(m, "%c", drm->vbios.data[i]);
+	return 0;
+}
+
+static struct drm_info_list nouveau_debugfs_list[] = {
+	{ "vbios.rom", nouveau_debugfs_vbios_image, 0, NULL },
+};
+#define NOUVEAU_DEBUGFS_ENTRIES ARRAY_SIZE(nouveau_debugfs_list)
+
+int
+nouveau_debugfs_init(struct drm_minor *minor)
+{
+	drm_debugfs_create_files(nouveau_debugfs_list, NOUVEAU_DEBUGFS_ENTRIES,
+				 minor->debugfs_root, minor);
+	return 0;
+}
+
+void
+nouveau_debugfs_takedown(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(nouveau_debugfs_list, NOUVEAU_DEBUGFS_ENTRIES,
+				 minor);
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.h b/drivers/gpu/drm/nouveau/nouveau_debugfs.h
new file mode 100644
index 000000000000..a62af6fb5f99
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.h
@@ -0,0 +1,22 @@
+#ifndef __NOUVEAU_DEBUGFS_H__
+#define __NOUVEAU_DEBUGFS_H__
+
+#include <drm/drmP.h>
+
+#if defined(CONFIG_DEBUG_FS)
+extern int  nouveau_debugfs_init(struct drm_minor *);
+extern void nouveau_debugfs_takedown(struct drm_minor *);
+#else
+static inline int
+nouveau_debugfs_init(struct drm_minor *minor)
+{
+       return 0;
+}
+
+static inline void nouveau_debugfs_takedown(struct drm_minor *minor)
+{
+}
+
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 508b00a2ce0d..4610c3a29bbe 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -41,6 +41,8 @@
 #include <subdev/gpio.h>
 #include <engine/disp.h>
 
+#include <core/class.h>
+
 static void
 nouveau_user_framebuffer_destroy(struct drm_framebuffer *drm_fb)
 {
@@ -78,11 +80,6 @@ nouveau_framebuffer_init(struct drm_device *dev,
 	struct drm_framebuffer *fb = &nv_fb->base;
 	int ret;
 
-	ret = drm_framebuffer_init(dev, fb, &nouveau_framebuffer_funcs);
-	if (ret) {
-		return ret;
-	}
-
 	drm_helper_mode_fill_fb_struct(fb, mode_cmd);
 	nv_fb->nvbo = nvbo;
 
@@ -125,6 +122,11 @@ nouveau_framebuffer_init(struct drm_device *dev,
 		}
 	}
 
+	ret = drm_framebuffer_init(dev, fb, &nouveau_framebuffer_funcs);
+	if (ret) {
+		return ret;
+	}
+
 	return 0;
 }
 
@@ -231,8 +233,10 @@ nouveau_display_init(struct drm_device *dev)
 	/* enable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (gpio)
-			gpio->irq(gpio, 0, conn->hpd, 0xff, true);
+		if (gpio && conn->hpd.func != DCB_GPIO_UNUSED) {
+			nouveau_event_get(gpio->events, conn->hpd.line,
+					 &conn->hpd_func);
+		}
 	}
 
 	return ret;
@@ -249,37 +253,20 @@ nouveau_display_fini(struct drm_device *dev)
 	/* disable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (gpio)
-			gpio->irq(gpio, 0, conn->hpd, 0xff, false);
+		if (gpio && conn->hpd.func != DCB_GPIO_UNUSED) {
+			nouveau_event_put(gpio->events, conn->hpd.line,
+					 &conn->hpd_func);
+		}
 	}
 
 	drm_kms_helper_poll_disable(dev);
 	disp->fini(dev);
 }
 
-static void
-nouveau_display_vblank_notify(void *data, int crtc)
-{
-	drm_handle_vblank(data, crtc);
-}
-
-static void
-nouveau_display_vblank_get(void *data, int crtc)
-{
-	drm_vblank_get(data, crtc);
-}
-
-static void
-nouveau_display_vblank_put(void *data, int crtc)
-{
-	drm_vblank_put(data, crtc);
-}
-
 int
 nouveau_display_create(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
 	struct nouveau_display *disp;
 	u32 pclass = dev->pdev->class >> 8;
 	int ret, gen;
@@ -288,11 +275,6 @@ nouveau_display_create(struct drm_device *dev)
 	if (!disp)
 		return -ENOMEM;
 
-	pdisp->vblank.data = dev;
-	pdisp->vblank.notify = nouveau_display_vblank_notify;
-	pdisp->vblank.get = nouveau_display_vblank_get;
-	pdisp->vblank.put = nouveau_display_vblank_put;
-
 	drm_mode_config_init(dev);
 	drm_mode_create_scaling_mode_property(dev);
 	drm_mode_create_dvi_i_properties(dev);
@@ -316,17 +298,13 @@ nouveau_display_create(struct drm_device *dev)
 		drm_property_create_range(dev, 0, "underscan vborder", 0, 128);
 
 	if (gen >= 1) {
+		/* -90..+90 */
 		disp->vibrant_hue_property =
-			drm_property_create(dev, DRM_MODE_PROP_RANGE,
-					    "vibrant hue", 2);
-		disp->vibrant_hue_property->values[0] = 0;
-		disp->vibrant_hue_property->values[1] = 180; /* -90..+90 */
+			drm_property_create_range(dev, 0, "vibrant hue", 0, 180);
 
+		/* -100..+100 */
 		disp->color_vibrance_property =
-			drm_property_create(dev, DRM_MODE_PROP_RANGE,
-					    "color vibrance", 2);
-		disp->color_vibrance_property->values[0] = 0;
-		disp->color_vibrance_property->values[1] = 200; /* -100..+100 */
+			drm_property_create_range(dev, 0, "color vibrance", 0, 200);
 	}
 
 	dev->mode_config.funcs = &nouveau_mode_config_funcs;
@@ -478,39 +456,6 @@ nouveau_display_resume(struct drm_device *dev)
 	}
 }
 
-int
-nouveau_vblank_enable(struct drm_device *dev, int crtc)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-
-	if (device->card_type >= NV_D0)
-		nv_mask(device, 0x6100c0 + (crtc * 0x800), 1, 1);
-	else
-	if (device->card_type >= NV_50)
-		nv_mask(device, NV50_PDISPLAY_INTR_EN_1, 0,
-			NV50_PDISPLAY_INTR_EN_1_VBLANK_CRTC_(crtc));
-	else
-		NVWriteCRTC(dev, crtc, NV_PCRTC_INTR_EN_0,
-			    NV_PCRTC_INTR_0_VBLANK);
-
-	return 0;
-}
-
-void
-nouveau_vblank_disable(struct drm_device *dev, int crtc)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-
-	if (device->card_type >= NV_D0)
-		nv_mask(device, 0x6100c0 + (crtc * 0x800), 1, 0);
-	else
-	if (device->card_type >= NV_50)
-		nv_mask(device, NV50_PDISPLAY_INTR_EN_1,
-			NV50_PDISPLAY_INTR_EN_1_VBLANK_CRTC_(crtc), 0);
-	else
-		NVWriteCRTC(dev, crtc, NV_PCRTC_INTR_EN_0, 0);
-}
-
 static int
 nouveau_page_flip_reserve(struct nouveau_bo *old_bo,
 			  struct nouveau_bo *new_bo)
@@ -595,7 +540,7 @@ nouveau_page_flip_emit(struct nouveau_channel *chan,
 	}
 	FIRE_RING (chan);
 
-	ret = nouveau_fence_new(chan, pfence);
+	ret = nouveau_fence_new(chan, false, pfence);
 	if (ret)
 		goto fail;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index 722548bb3bd3..1ea3e4734b62 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -59,9 +59,6 @@ void nouveau_display_fini(struct drm_device *dev);
 int  nouveau_display_suspend(struct drm_device *dev);
 void nouveau_display_resume(struct drm_device *dev);
 
-int  nouveau_vblank_enable(struct drm_device *dev, int crtc);
-void nouveau_vblank_disable(struct drm_device *dev, int crtc);
-
 int  nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			    struct drm_pending_vblank_event *event);
 int  nouveau_finish_page_flip(struct nouveau_channel *,
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h
index 5c2e22932d1c..690d5930ce32 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.h
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.h
@@ -191,7 +191,7 @@ WIND_RING(struct nouveau_channel *chan)
 #define NV84_SUBCHAN_SEMAPHORE_TRIGGER_WRITE_LONG                    0x00000002
 #define NV84_SUBCHAN_SEMAPHORE_TRIGGER_ACQUIRE_GEQUAL                0x00000004
 #define NVC0_SUBCHAN_SEMAPHORE_TRIGGER_YIELD                         0x00001000
-#define NV84_SUBCHAN_NOTIFY_INTR                                     0x00000020
+#define NV84_SUBCHAN_UEVENT                                          0x00000020
 #define NV84_SUBCHAN_WRCACHE_FLUSH                                   0x00000024
 #define NV10_SUBCHAN_REF_CNT                                         0x00000050
 #define NVSW_SUBCHAN_PAGE_FLIP                                       0x00000054
diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index 59838651ee8f..36fd22500569 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -35,300 +35,6 @@
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 
-/******************************************************************************
- * link training
- *****************************************************************************/
-struct dp_state {
-	struct nouveau_i2c_port *auxch;
-	struct nouveau_object *core;
-	struct dcb_output *dcb;
-	int crtc;
-	u8 *dpcd;
-	int link_nr;
-	u32 link_bw;
-	u8  stat[6];
-	u8  conf[4];
-};
-
-static void
-dp_set_link_config(struct drm_device *dev, struct dp_state *dp)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct dcb_output *dcb = dp->dcb;
-	const u32 or = ffs(dcb->or) - 1, link = !(dcb->sorconf.link & 1);
-	const u32 moff = (dp->crtc << 3) | (link << 2) | or;
-	u8 sink[2];
-	u32 data;
-
-	NV_DEBUG(drm, "%d lanes at %d KB/s\n", dp->link_nr, dp->link_bw);
-
-	/* set desired link configuration on the source */
-	data = ((dp->link_bw / 27000) << 8) | dp->link_nr;
-	if (dp->dpcd[2] & DP_ENHANCED_FRAME_CAP)
-		data |= NV94_DISP_SOR_DP_LNKCTL_FRAME_ENH;
-
-	nv_call(dp->core, NV94_DISP_SOR_DP_LNKCTL + moff, data);
-
-	/* inform the sink of the new configuration */
-	sink[0] = dp->link_bw / 27000;
-	sink[1] = dp->link_nr;
-	if (dp->dpcd[2] & DP_ENHANCED_FRAME_CAP)
-		sink[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
-
-	nv_wraux(dp->auxch, DP_LINK_BW_SET, sink, 2);
-}
-
-static void
-dp_set_training_pattern(struct drm_device *dev, struct dp_state *dp, u8 pattern)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct dcb_output *dcb = dp->dcb;
-	const u32 or = ffs(dcb->or) - 1, link = !(dcb->sorconf.link & 1);
-	const u32 moff = (dp->crtc << 3) | (link << 2) | or;
-	u8 sink_tp;
-
-	NV_DEBUG(drm, "training pattern %d\n", pattern);
-
-	nv_call(dp->core, NV94_DISP_SOR_DP_TRAIN + moff, pattern);
-
-	nv_rdaux(dp->auxch, DP_TRAINING_PATTERN_SET, &sink_tp, 1);
-	sink_tp &= ~DP_TRAINING_PATTERN_MASK;
-	sink_tp |= pattern;
-	nv_wraux(dp->auxch, DP_TRAINING_PATTERN_SET, &sink_tp, 1);
-}
-
-static int
-dp_link_train_commit(struct drm_device *dev, struct dp_state *dp)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct dcb_output *dcb = dp->dcb;
-	const u32 or = ffs(dcb->or) - 1, link = !(dcb->sorconf.link & 1);
-	const u32 moff = (dp->crtc << 3) | (link << 2) | or;
-	int i;
-
-	for (i = 0; i < dp->link_nr; i++) {
-		u8 lane = (dp->stat[4 + (i >> 1)] >> ((i & 1) * 4)) & 0xf;
-		u8 lpre = (lane & 0x0c) >> 2;
-		u8 lvsw = (lane & 0x03) >> 0;
-
-		dp->conf[i] = (lpre << 3) | lvsw;
-		if (lvsw == DP_TRAIN_VOLTAGE_SWING_1200)
-			dp->conf[i] |= DP_TRAIN_MAX_SWING_REACHED;
-		if ((lpre << 3) == DP_TRAIN_PRE_EMPHASIS_9_5)
-			dp->conf[i] |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
-
-		NV_DEBUG(drm, "config lane %d %02x\n", i, dp->conf[i]);
-
-		nv_call(dp->core, NV94_DISP_SOR_DP_DRVCTL(i) + moff, (lvsw << 8) | lpre);
-	}
-
-	return nv_wraux(dp->auxch, DP_TRAINING_LANE0_SET, dp->conf, 4);
-}
-
-static int
-dp_link_train_update(struct drm_device *dev, struct dp_state *dp, u32 delay)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	int ret;
-
-	udelay(delay);
-
-	ret = nv_rdaux(dp->auxch, DP_LANE0_1_STATUS, dp->stat, 6);
-	if (ret)
-		return ret;
-
-	NV_DEBUG(drm, "status %*ph\n", 6, dp->stat);
-	return 0;
-}
-
-static int
-dp_link_train_cr(struct drm_device *dev, struct dp_state *dp)
-{
-	bool cr_done = false, abort = false;
-	int voltage = dp->conf[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
-	int tries = 0, i;
-
-	dp_set_training_pattern(dev, dp, DP_TRAINING_PATTERN_1);
-
-	do {
-		if (dp_link_train_commit(dev, dp) ||
-		    dp_link_train_update(dev, dp, 100))
-			break;
-
-		cr_done = true;
-		for (i = 0; i < dp->link_nr; i++) {
-			u8 lane = (dp->stat[i >> 1] >> ((i & 1) * 4)) & 0xf;
-			if (!(lane & DP_LANE_CR_DONE)) {
-				cr_done = false;
-				if (dp->conf[i] & DP_TRAIN_MAX_SWING_REACHED)
-					abort = true;
-				break;
-			}
-		}
-
-		if ((dp->conf[0] & DP_TRAIN_VOLTAGE_SWING_MASK) != voltage) {
-			voltage = dp->conf[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
-			tries = 0;
-		}
-	} while (!cr_done && !abort && ++tries < 5);
-
-	return cr_done ? 0 : -1;
-}
-
-static int
-dp_link_train_eq(struct drm_device *dev, struct dp_state *dp)
-{
-	bool eq_done, cr_done = true;
-	int tries = 0, i;
-
-	dp_set_training_pattern(dev, dp, DP_TRAINING_PATTERN_2);
-
-	do {
-		if (dp_link_train_update(dev, dp, 400))
-			break;
-
-		eq_done = !!(dp->stat[2] & DP_INTERLANE_ALIGN_DONE);
-		for (i = 0; i < dp->link_nr && eq_done; i++) {
-			u8 lane = (dp->stat[i >> 1] >> ((i & 1) * 4)) & 0xf;
-			if (!(lane & DP_LANE_CR_DONE))
-				cr_done = false;
-			if (!(lane & DP_LANE_CHANNEL_EQ_DONE) ||
-			    !(lane & DP_LANE_SYMBOL_LOCKED))
-				eq_done = false;
-		}
-
-		if (dp_link_train_commit(dev, dp))
-			break;
-	} while (!eq_done && cr_done && ++tries <= 5);
-
-	return eq_done ? 0 : -1;
-}
-
-static void
-dp_link_train_init(struct drm_device *dev, struct dp_state *dp, bool spread)
-{
-	struct dcb_output *dcb = dp->dcb;
-	const u32 or = ffs(dcb->or) - 1, link = !(dcb->sorconf.link & 1);
-	const u32 moff = (dp->crtc << 3) | (link << 2) | or;
-
-	nv_call(dp->core, NV94_DISP_SOR_DP_TRAIN + moff, (spread ?
-			  NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD_ON :
-			  NV94_DISP_SOR_DP_TRAIN_INIT_SPREAD_OFF) |
-			  NV94_DISP_SOR_DP_TRAIN_OP_INIT);
-}
-
-static void
-dp_link_train_fini(struct drm_device *dev, struct dp_state *dp)
-{
-	struct dcb_output *dcb = dp->dcb;
-	const u32 or = ffs(dcb->or) - 1, link = !(dcb->sorconf.link & 1);
-	const u32 moff = (dp->crtc << 3) | (link << 2) | or;
-
-	nv_call(dp->core, NV94_DISP_SOR_DP_TRAIN + moff,
-			  NV94_DISP_SOR_DP_TRAIN_OP_FINI);
-}
-
-static bool
-nouveau_dp_link_train(struct drm_encoder *encoder, u32 datarate,
-		      struct nouveau_object *core)
-{
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
-	struct nouveau_connector *nv_connector =
-		nouveau_encoder_connector_get(nv_encoder);
-	struct drm_device *dev = encoder->dev;
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	const u32 bw_list[] = { 270000, 162000, 0 };
-	const u32 *link_bw = bw_list;
-	struct dp_state dp;
-
-	dp.auxch = i2c->find(i2c, nv_encoder->dcb->i2c_index);
-	if (!dp.auxch)
-		return false;
-
-	dp.core = core;
-	dp.dcb = nv_encoder->dcb;
-	dp.crtc = nv_crtc->index;
-	dp.dpcd = nv_encoder->dp.dpcd;
-
-	/* adjust required bandwidth for 8B/10B coding overhead */
-	datarate = (datarate / 8) * 10;
-
-	/* some sinks toggle hotplug in response to some of the actions
-	 * we take during link training (DP_SET_POWER is one), we need
-	 * to ignore them for the moment to avoid races.
-	 */
-	gpio->irq(gpio, 0, nv_connector->hpd, 0xff, false);
-
-	/* enable down-spreading and execute pre-train script from vbios */
-	dp_link_train_init(dev, &dp, nv_encoder->dp.dpcd[3] & 1);
-
-	/* start off at highest link rate supported by encoder and display */
-	while (*link_bw > nv_encoder->dp.link_bw)
-		link_bw++;
-
-	while (link_bw[0]) {
-		/* find minimum required lane count at this link rate */
-		dp.link_nr = nv_encoder->dp.link_nr;
-		while ((dp.link_nr >> 1) * link_bw[0] > datarate)
-			dp.link_nr >>= 1;
-
-		/* drop link rate to minimum with this lane count */
-		while ((link_bw[1] * dp.link_nr) > datarate)
-			link_bw++;
-		dp.link_bw = link_bw[0];
-
-		/* program selected link configuration */
-		dp_set_link_config(dev, &dp);
-
-		/* attempt to train the link at this configuration */
-		memset(dp.stat, 0x00, sizeof(dp.stat));
-		if (!dp_link_train_cr(dev, &dp) &&
-		    !dp_link_train_eq(dev, &dp))
-			break;
-
-		/* retry at lower rate */
-		link_bw++;
-	}
-
-	/* finish link training */
-	dp_set_training_pattern(dev, &dp, DP_TRAINING_PATTERN_DISABLE);
-
-	/* execute post-train script from vbios */
-	dp_link_train_fini(dev, &dp);
-
-	/* re-enable hotplug detect */
-	gpio->irq(gpio, 0, nv_connector->hpd, 0xff, true);
-	return true;
-}
-
-void
-nouveau_dp_dpms(struct drm_encoder *encoder, int mode, u32 datarate,
-		struct nouveau_object *core)
-{
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nouveau_drm *drm = nouveau_drm(encoder->dev);
-	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
-	struct nouveau_i2c_port *auxch;
-	u8 status;
-
-	auxch = i2c->find(i2c, nv_encoder->dcb->i2c_index);
-	if (!auxch)
-		return;
-
-	if (mode == DRM_MODE_DPMS_ON)
-		status = DP_SET_POWER_D0;
-	else
-		status = DP_SET_POWER_D3;
-
-	nv_wraux(auxch, DP_SET_POWER, &status, 1);
-
-	if (mode == DRM_MODE_DPMS_ON)
-		nouveau_dp_link_train(encoder, datarate, core);
-}
-
 static void
 nouveau_dp_probe_oui(struct drm_device *dev, struct nouveau_i2c_port *auxch,
 		     u8 *dpcd)
@@ -355,12 +61,11 @@ nouveau_dp_detect(struct drm_encoder *encoder)
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct drm_device *dev = encoder->dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_i2c_port *auxch;
 	u8 *dpcd = nv_encoder->dp.dpcd;
 	int ret;
 
-	auxch = i2c->find(i2c, nv_encoder->dcb->i2c_index);
+	auxch = nv_encoder->i2c;
 	if (!auxch)
 		return false;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 5e7aef23825a..d1099365bfc1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -34,6 +34,8 @@
 #include <subdev/device.h>
 #include <subdev/vm.h>
 
+#include <engine/disp.h>
+
 #include "nouveau_drm.h"
 #include "nouveau_irq.h"
 #include "nouveau_dma.h"
@@ -48,6 +50,7 @@
 #include "nouveau_abi16.h"
 #include "nouveau_fbcon.h"
 #include "nouveau_fence.h"
+#include "nouveau_debugfs.h"
 
 MODULE_PARM_DESC(config, "option string to pass to driver core");
 static char *nouveau_config;
@@ -68,6 +71,32 @@ module_param_named(modeset, nouveau_modeset, int, 0400);
 
 static struct drm_driver driver;
 
+static int
+nouveau_drm_vblank_enable(struct drm_device *dev, int head)
+{
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
+	nouveau_event_get(pdisp->vblank, head, &drm->vblank);
+	return 0;
+}
+
+static void
+nouveau_drm_vblank_disable(struct drm_device *dev, int head)
+{
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
+	nouveau_event_put(pdisp->vblank, head, &drm->vblank);
+}
+
+static int
+nouveau_drm_vblank_handler(struct nouveau_eventh *event, int head)
+{
+	struct nouveau_drm *drm =
+		container_of(event, struct nouveau_drm, vblank);
+	drm_handle_vblank(drm->dev, head);
+	return NVKM_EVENT_KEEP;
+}
+
 static u64
 nouveau_name(struct pci_dev *pdev)
 {
@@ -132,7 +161,8 @@ nouveau_accel_init(struct nouveau_drm *drm)
 
 	/* initialise synchronisation routines */
 	if      (device->card_type < NV_10) ret = nv04_fence_create(drm);
-	else if (device->card_type < NV_50) ret = nv10_fence_create(drm);
+	else if (device->chipset   <  0x17) ret = nv10_fence_create(drm);
+	else if (device->card_type < NV_50) ret = nv17_fence_create(drm);
 	else if (device->chipset   <  0x84) ret = nv50_fence_create(drm);
 	else if (device->card_type < NV_C0) ret = nv84_fence_create(drm);
 	else                                ret = nvc0_fence_create(drm);
@@ -262,6 +292,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
 
 	dev->dev_private = drm;
 	drm->dev = dev;
+	drm->vblank.func = nouveau_drm_vblank_handler;
 
 	INIT_LIST_HEAD(&drm->clients);
 	spin_lock_init(&drm->tile.lock);
@@ -401,7 +432,7 @@ nouveau_drm_remove(struct pci_dev *pdev)
 	nouveau_object_debug();
 }
 
-int
+static int
 nouveau_do_suspend(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
@@ -472,7 +503,7 @@ int nouveau_pmops_suspend(struct device *dev)
 	return 0;
 }
 
-int
+static int
 nouveau_do_resume(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
@@ -546,10 +577,11 @@ nouveau_drm_open(struct drm_device *dev, struct drm_file *fpriv)
 	struct pci_dev *pdev = dev->pdev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_cli *cli;
-	char name[16];
+	char name[32], tmpname[TASK_COMM_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "%d", pid_nr(fpriv->pid));
+	get_task_comm(tmpname, current);
+	snprintf(name, sizeof(name), "%s[%d]", tmpname, pid_nr(fpriv->pid));
 
 	ret = nouveau_cli_create(pdev, name, sizeof(*cli), (void **)&cli);
 	if (ret)
@@ -639,22 +671,32 @@ driver = {
 	.postclose = nouveau_drm_postclose,
 	.lastclose = nouveau_vga_lastclose,
 
+#if defined(CONFIG_DEBUG_FS)
+	.debugfs_init = nouveau_debugfs_init,
+	.debugfs_cleanup = nouveau_debugfs_takedown,
+#endif
+
 	.irq_preinstall = nouveau_irq_preinstall,
 	.irq_postinstall = nouveau_irq_postinstall,
 	.irq_uninstall = nouveau_irq_uninstall,
 	.irq_handler = nouveau_irq_handler,
 
 	.get_vblank_counter = drm_vblank_count,
-	.enable_vblank = nouveau_vblank_enable,
-	.disable_vblank = nouveau_vblank_disable,
+	.enable_vblank = nouveau_drm_vblank_enable,
+	.disable_vblank = nouveau_drm_vblank_disable,
 
 	.ioctls = nouveau_ioctls,
 	.fops = &nouveau_driver_fops,
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-	.gem_prime_export = nouveau_gem_prime_export,
-	.gem_prime_import = nouveau_gem_prime_import,
+	.gem_prime_export = drm_gem_prime_export,
+	.gem_prime_import = drm_gem_prime_import,
+	.gem_prime_pin = nouveau_gem_prime_pin,
+	.gem_prime_get_sg_table = nouveau_gem_prime_get_sg_table,
+	.gem_prime_import_sg_table = nouveau_gem_prime_import_sg_table,
+	.gem_prime_vmap = nouveau_gem_prime_vmap,
+	.gem_prime_vunmap = nouveau_gem_prime_vunmap,
 
 	.gem_init_object = nouveau_gem_object_new,
 	.gem_free_object = nouveau_gem_object_del,
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h
index aa89eb938b47..b25df374c901 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.h
@@ -13,6 +13,7 @@
 #define DRIVER_PATCHLEVEL	0
 
 #include <core/client.h>
+#include <core/event.h>
 
 #include <subdev/vm.h>
 
@@ -112,6 +113,7 @@ struct nouveau_drm {
 	struct nvbios vbios;
 	struct nouveau_display *display;
 	struct backlight_device *backlight;
+	struct nouveau_eventh vblank;
 
 	/* power management */
 	struct nouveau_pm *pm;
diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h
index d0d95bd511ab..e24341229d5e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_encoder.h
+++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h
@@ -36,19 +36,12 @@
 
 struct nouveau_i2c_port;
 
-struct dp_train_func {
-	void (*link_set)(struct drm_device *, struct dcb_output *, int crtc,
-			 int nr, u32 bw, bool enhframe);
-	void (*train_set)(struct drm_device *, struct dcb_output *, u8 pattern);
-	void (*train_adj)(struct drm_device *, struct dcb_output *,
-			  u8 lane, u8 swing, u8 preem);
-};
-
 struct nouveau_encoder {
 	struct drm_encoder_slave base;
 
 	struct dcb_output *dcb;
 	int or;
+	struct nouveau_i2c_port *i2c;
 
 	/* different to drm_encoder.crtc, this reflects what's
 	 * actually programmed on the hw, not the proposed crtc */
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index 67a1a069de28..b03531781580 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -251,9 +251,10 @@ nouveau_fbcon_zfill(struct drm_device *dev, struct nouveau_fbdev *fbcon)
 }
 
 static int
-nouveau_fbcon_create(struct nouveau_fbdev *fbcon,
+nouveau_fbcon_create(struct drm_fb_helper *helper,
 		     struct drm_fb_helper_surface_size *sizes)
 {
+	struct nouveau_fbdev *fbcon = (struct nouveau_fbdev *)helper;
 	struct drm_device *dev = fbcon->dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_device *device = nv_device(drm->device);
@@ -388,23 +389,6 @@ out:
 	return ret;
 }
 
-static int
-nouveau_fbcon_find_or_create_single(struct drm_fb_helper *helper,
-				    struct drm_fb_helper_surface_size *sizes)
-{
-	struct nouveau_fbdev *fbcon = (struct nouveau_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = nouveau_fbcon_create(fbcon, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 void
 nouveau_fbcon_output_poll_changed(struct drm_device *dev)
 {
@@ -433,6 +417,7 @@ nouveau_fbcon_destroy(struct drm_device *dev, struct nouveau_fbdev *fbcon)
 		nouveau_fb->nvbo = NULL;
 	}
 	drm_fb_helper_fini(&fbcon->helper);
+	drm_framebuffer_unregister_private(&nouveau_fb->base);
 	drm_framebuffer_cleanup(&nouveau_fb->base);
 	return 0;
 }
@@ -449,7 +434,7 @@ void nouveau_fbcon_gpu_lockup(struct fb_info *info)
 static struct drm_fb_helper_funcs nouveau_fbcon_helper_funcs = {
 	.gamma_set = nouveau_fbcon_gamma_set,
 	.gamma_get = nouveau_fbcon_gamma_get,
-	.fb_probe = nouveau_fbcon_find_or_create_single,
+	.fb_probe = nouveau_fbcon_create,
 };
 
 
@@ -490,6 +475,9 @@ nouveau_fbcon_init(struct drm_device *dev)
 	else
 		preferred_bpp = 32;
 
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	drm_fb_helper_initial_config(&fbcon->helper, preferred_bpp);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 1d049be79f74..6c946837a0aa 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -33,14 +33,14 @@
 #include "nouveau_dma.h"
 #include "nouveau_fence.h"
 
+#include <engine/fifo.h>
+
 void
 nouveau_fence_context_del(struct nouveau_fence_chan *fctx)
 {
 	struct nouveau_fence *fence, *fnext;
 	spin_lock(&fctx->lock);
 	list_for_each_entry_safe(fence, fnext, &fctx->pending, head) {
-		if (fence->work)
-			fence->work(fence->priv, false);
 		fence->channel = NULL;
 		list_del(&fence->head);
 		nouveau_fence_unref(&fence);
@@ -59,17 +59,14 @@ nouveau_fence_context_new(struct nouveau_fence_chan *fctx)
 static void
 nouveau_fence_update(struct nouveau_channel *chan)
 {
-	struct nouveau_fence_priv *priv = chan->drm->fence;
 	struct nouveau_fence_chan *fctx = chan->fence;
 	struct nouveau_fence *fence, *fnext;
 
 	spin_lock(&fctx->lock);
 	list_for_each_entry_safe(fence, fnext, &fctx->pending, head) {
-		if (priv->read(chan) < fence->sequence)
+		if (fctx->read(chan) < fence->sequence)
 			break;
 
-		if (fence->work)
-			fence->work(fence->priv, true);
 		fence->channel = NULL;
 		list_del(&fence->head);
 		nouveau_fence_unref(&fence);
@@ -80,7 +77,6 @@ nouveau_fence_update(struct nouveau_channel *chan)
 int
 nouveau_fence_emit(struct nouveau_fence *fence, struct nouveau_channel *chan)
 {
-	struct nouveau_fence_priv *priv = chan->drm->fence;
 	struct nouveau_fence_chan *fctx = chan->fence;
 	int ret;
 
@@ -88,7 +84,7 @@ nouveau_fence_emit(struct nouveau_fence *fence, struct nouveau_channel *chan)
 	fence->timeout  = jiffies + (3 * DRM_HZ);
 	fence->sequence = ++fctx->sequence;
 
-	ret = priv->emit(fence);
+	ret = fctx->emit(fence);
 	if (!ret) {
 		kref_get(&fence->kref);
 		spin_lock(&fctx->lock);
@@ -107,13 +103,87 @@ nouveau_fence_done(struct nouveau_fence *fence)
 	return !fence->channel;
 }
 
+struct nouveau_fence_uevent {
+	struct nouveau_eventh handler;
+	struct nouveau_fence_priv *priv;
+};
+
+static int
+nouveau_fence_wait_uevent_handler(struct nouveau_eventh *event, int index)
+{
+	struct nouveau_fence_uevent *uevent =
+		container_of(event, struct nouveau_fence_uevent, handler);
+	wake_up_all(&uevent->priv->waiting);
+	return NVKM_EVENT_KEEP;
+}
+
+static int
+nouveau_fence_wait_uevent(struct nouveau_fence *fence, bool intr)
+
+{
+	struct nouveau_channel *chan = fence->channel;
+	struct nouveau_fifo *pfifo = nouveau_fifo(chan->drm->device);
+	struct nouveau_fence_priv *priv = chan->drm->fence;
+	struct nouveau_fence_uevent uevent = {
+		.handler.func = nouveau_fence_wait_uevent_handler,
+		.priv = priv,
+	};
+	int ret = 0;
+
+	nouveau_event_get(pfifo->uevent, 0, &uevent.handler);
+
+	if (fence->timeout) {
+		unsigned long timeout = fence->timeout - jiffies;
+
+		if (time_before(jiffies, fence->timeout)) {
+			if (intr) {
+				ret = wait_event_interruptible_timeout(
+						priv->waiting,
+						nouveau_fence_done(fence),
+						timeout);
+			} else {
+				ret = wait_event_timeout(priv->waiting,
+						nouveau_fence_done(fence),
+						timeout);
+			}
+		}
+
+		if (ret >= 0) {
+			fence->timeout = jiffies + ret;
+			if (time_after_eq(jiffies, fence->timeout))
+				ret = -EBUSY;
+		}
+	} else {
+		if (intr) {
+			ret = wait_event_interruptible(priv->waiting,
+					nouveau_fence_done(fence));
+		} else {
+			wait_event(priv->waiting, nouveau_fence_done(fence));
+		}
+	}
+
+	nouveau_event_put(pfifo->uevent, 0, &uevent.handler);
+	if (unlikely(ret < 0))
+		return ret;
+
+	return 0;
+}
+
 int
 nouveau_fence_wait(struct nouveau_fence *fence, bool lazy, bool intr)
 {
+	struct nouveau_channel *chan = fence->channel;
+	struct nouveau_fence_priv *priv = chan ? chan->drm->fence : NULL;
 	unsigned long sleep_time = NSEC_PER_MSEC / 1000;
 	ktime_t t;
 	int ret = 0;
 
+	while (priv && priv->uevent && lazy && !nouveau_fence_done(fence)) {
+		ret = nouveau_fence_wait_uevent(fence, intr);
+		if (ret < 0)
+			return ret;
+	}
+
 	while (!nouveau_fence_done(fence)) {
 		if (fence->timeout && time_after_eq(jiffies, fence->timeout)) {
 			ret = -EBUSY;
@@ -143,14 +213,14 @@ nouveau_fence_wait(struct nouveau_fence *fence, bool lazy, bool intr)
 int
 nouveau_fence_sync(struct nouveau_fence *fence, struct nouveau_channel *chan)
 {
-	struct nouveau_fence_priv *priv = chan->drm->fence;
+	struct nouveau_fence_chan *fctx = chan->fence;
 	struct nouveau_channel *prev;
 	int ret = 0;
 
 	prev = fence ? fence->channel : NULL;
 	if (prev) {
 		if (unlikely(prev != chan && !nouveau_fence_done(fence))) {
-			ret = priv->sync(fence, prev, chan);
+			ret = fctx->sync(fence, prev, chan);
 			if (unlikely(ret))
 				ret = nouveau_fence_wait(fence, true, false);
 		}
@@ -182,7 +252,8 @@ nouveau_fence_ref(struct nouveau_fence *fence)
 }
 
 int
-nouveau_fence_new(struct nouveau_channel *chan, struct nouveau_fence **pfence)
+nouveau_fence_new(struct nouveau_channel *chan, bool sysmem,
+		  struct nouveau_fence **pfence)
 {
 	struct nouveau_fence *fence;
 	int ret = 0;
@@ -193,13 +264,13 @@ nouveau_fence_new(struct nouveau_channel *chan, struct nouveau_fence **pfence)
 	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
 	if (!fence)
 		return -ENOMEM;
+
+	fence->sysmem = sysmem;
 	kref_init(&fence->kref);
 
-	if (chan) {
-		ret = nouveau_fence_emit(fence, chan);
-		if (ret)
-			nouveau_fence_unref(&fence);
-	}
+	ret = nouveau_fence_emit(fence, chan);
+	if (ret)
+		nouveau_fence_unref(&fence);
 
 	*pfence = fence;
 	return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h
index cdb83acdffe2..c89943407b52 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.h
@@ -7,15 +7,15 @@ struct nouveau_fence {
 	struct list_head head;
 	struct kref kref;
 
+	bool sysmem;
+
 	struct nouveau_channel *channel;
 	unsigned long timeout;
 	u32 sequence;
-
-	void (*work)(void *priv, bool signalled);
-	void *priv;
 };
 
-int  nouveau_fence_new(struct nouveau_channel *, struct nouveau_fence **);
+int  nouveau_fence_new(struct nouveau_channel *, bool sysmem,
+		       struct nouveau_fence **);
 struct nouveau_fence *
 nouveau_fence_ref(struct nouveau_fence *);
 void nouveau_fence_unref(struct nouveau_fence **);
@@ -29,6 +29,13 @@ struct nouveau_fence_chan {
 	struct list_head pending;
 	struct list_head flip;
 
+	int  (*emit)(struct nouveau_fence *);
+	int  (*sync)(struct nouveau_fence *, struct nouveau_channel *,
+		     struct nouveau_channel *);
+	u32  (*read)(struct nouveau_channel *);
+	int  (*emit32)(struct nouveau_channel *, u64, u32);
+	int  (*sync32)(struct nouveau_channel *, u64, u32);
+
 	spinlock_t lock;
 	u32 sequence;
 };
@@ -39,10 +46,9 @@ struct nouveau_fence_priv {
 	void (*resume)(struct nouveau_drm *);
 	int  (*context_new)(struct nouveau_channel *);
 	void (*context_del)(struct nouveau_channel *);
-	int  (*emit)(struct nouveau_fence *);
-	int  (*sync)(struct nouveau_fence *, struct nouveau_channel *,
-		     struct nouveau_channel *);
-	u32  (*read)(struct nouveau_channel *);
+
+	wait_queue_head_t waiting;
+	bool uevent;
 };
 
 #define nouveau_fence(drm) ((struct nouveau_fence_priv *)(drm)->fence)
@@ -60,13 +66,31 @@ u32  nv10_fence_read(struct nouveau_channel *);
 void nv10_fence_context_del(struct nouveau_channel *);
 void nv10_fence_destroy(struct nouveau_drm *);
 int  nv10_fence_create(struct nouveau_drm *);
+
+int  nv17_fence_create(struct nouveau_drm *);
 void nv17_fence_resume(struct nouveau_drm *drm);
 
 int nv50_fence_create(struct nouveau_drm *);
 int nv84_fence_create(struct nouveau_drm *);
 int nvc0_fence_create(struct nouveau_drm *);
-u64 nvc0_fence_crtc(struct nouveau_channel *, int crtc);
 
 int nouveau_flip_complete(void *chan);
 
+struct nv84_fence_chan {
+	struct nouveau_fence_chan base;
+	struct nouveau_vma vma;
+	struct nouveau_vma vma_gart;
+	struct nouveau_vma dispc_vma[4];
+};
+
+struct nv84_fence_priv {
+	struct nouveau_fence_priv base;
+	struct nouveau_bo *bo;
+	struct nouveau_bo *bo_gart;
+	u32 *suspend;
+};
+
+u64  nv84_fence_crtc(struct nouveau_channel *, int);
+int  nv84_fence_context_new(struct nouveau_channel *);
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 8bf695c52f95..b4b4d0c1f4af 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -24,8 +24,6 @@
  *
  */
 
-#include <linux/dma-buf.h>
-
 #include <subdev/fb.h>
 
 #include "nouveau_drm.h"
@@ -205,6 +203,7 @@ nouveau_gem_ioctl_new(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct nouveau_fb *pfb = nouveau_fb(drm->device);
 	struct drm_nouveau_gem_new *req = data;
 	struct nouveau_bo *nvbo = NULL;
@@ -213,7 +212,7 @@ nouveau_gem_ioctl_new(struct drm_device *dev, void *data,
 	drm->ttm.bdev.dev_mapping = drm->dev->dev_mapping;
 
 	if (!pfb->memtype_valid(pfb, req->info.tile_flags)) {
-		NV_ERROR(drm, "bad page flags: 0x%08x\n", req->info.tile_flags);
+		NV_ERROR(cli, "bad page flags: 0x%08x\n", req->info.tile_flags);
 		return -EINVAL;
 	}
 
@@ -315,16 +314,18 @@ validate_init(struct nouveau_channel *chan, struct drm_file *file_priv,
 	      struct drm_nouveau_gem_pushbuf_bo *pbbo,
 	      int nr_buffers, struct validate_op *op)
 {
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct drm_device *dev = chan->drm->dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	uint32_t sequence;
 	int trycnt = 0;
 	int ret, i;
+	struct nouveau_bo *res_bo = NULL;
 
 	sequence = atomic_add_return(1, &drm->ttm.validate_sequence);
 retry:
 	if (++trycnt > 100000) {
-		NV_ERROR(drm, "%s failed and gave up.\n", __func__);
+		NV_ERROR(cli, "%s failed and gave up.\n", __func__);
 		return -EINVAL;
 	}
 
@@ -335,14 +336,19 @@ retry:
 
 		gem = drm_gem_object_lookup(dev, file_priv, b->handle);
 		if (!gem) {
-			NV_ERROR(drm, "Unknown handle 0x%08x\n", b->handle);
+			NV_ERROR(cli, "Unknown handle 0x%08x\n", b->handle);
 			validate_fini(op, NULL);
 			return -ENOENT;
 		}
 		nvbo = gem->driver_private;
+		if (nvbo == res_bo) {
+			res_bo = NULL;
+			drm_gem_object_unreference_unlocked(gem);
+			continue;
+		}
 
 		if (nvbo->reserved_by && nvbo->reserved_by == file_priv) {
-			NV_ERROR(drm, "multiple instances of buffer %d on "
+			NV_ERROR(cli, "multiple instances of buffer %d on "
 				      "validation list\n", b->handle);
 			drm_gem_object_unreference_unlocked(gem);
 			validate_fini(op, NULL);
@@ -352,15 +358,19 @@ retry:
 		ret = ttm_bo_reserve(&nvbo->bo, true, false, true, sequence);
 		if (ret) {
 			validate_fini(op, NULL);
-			if (unlikely(ret == -EAGAIN))
-				ret = ttm_bo_wait_unreserved(&nvbo->bo, true);
-			drm_gem_object_unreference_unlocked(gem);
+			if (unlikely(ret == -EAGAIN)) {
+				sequence = atomic_add_return(1, &drm->ttm.validate_sequence);
+				ret = ttm_bo_reserve_slowpath(&nvbo->bo, true,
+							      sequence);
+				if (!ret)
+					res_bo = nvbo;
+			}
 			if (unlikely(ret)) {
+				drm_gem_object_unreference_unlocked(gem);
 				if (ret != -ERESTARTSYS)
-					NV_ERROR(drm, "fail reserve\n");
+					NV_ERROR(cli, "fail reserve\n");
 				return ret;
 			}
-			goto retry;
 		}
 
 		b->user_priv = (uint64_t)(unsigned long)nvbo;
@@ -376,12 +386,14 @@ retry:
 		if (b->valid_domains & NOUVEAU_GEM_DOMAIN_GART)
 			list_add_tail(&nvbo->entry, &op->gart_list);
 		else {
-			NV_ERROR(drm, "invalid valid domains: 0x%08x\n",
+			NV_ERROR(cli, "invalid valid domains: 0x%08x\n",
 				 b->valid_domains);
 			list_add_tail(&nvbo->entry, &op->both_list);
 			validate_fini(op, NULL);
 			return -EINVAL;
 		}
+		if (nvbo == res_bo)
+			goto retry;
 	}
 
 	return 0;
@@ -407,8 +419,9 @@ validate_sync(struct nouveau_channel *chan, struct nouveau_bo *nvbo)
 }
 
 static int
-validate_list(struct nouveau_channel *chan, struct list_head *list,
-	      struct drm_nouveau_gem_pushbuf_bo *pbbo, uint64_t user_pbbo_ptr)
+validate_list(struct nouveau_channel *chan, struct nouveau_cli *cli,
+	      struct list_head *list, struct drm_nouveau_gem_pushbuf_bo *pbbo,
+	      uint64_t user_pbbo_ptr)
 {
 	struct nouveau_drm *drm = chan->drm;
 	struct drm_nouveau_gem_pushbuf_bo __user *upbbo =
@@ -421,7 +434,7 @@ validate_list(struct nouveau_channel *chan, struct list_head *list,
 
 		ret = validate_sync(chan, nvbo);
 		if (unlikely(ret)) {
-			NV_ERROR(drm, "fail pre-validate sync\n");
+			NV_ERROR(cli, "fail pre-validate sync\n");
 			return ret;
 		}
 
@@ -429,20 +442,20 @@ validate_list(struct nouveau_channel *chan, struct list_head *list,
 					     b->write_domains,
 					     b->valid_domains);
 		if (unlikely(ret)) {
-			NV_ERROR(drm, "fail set_domain\n");
+			NV_ERROR(cli, "fail set_domain\n");
 			return ret;
 		}
 
 		ret = nouveau_bo_validate(nvbo, true, false);
 		if (unlikely(ret)) {
 			if (ret != -ERESTARTSYS)
-				NV_ERROR(drm, "fail ttm_validate\n");
+				NV_ERROR(cli, "fail ttm_validate\n");
 			return ret;
 		}
 
 		ret = validate_sync(chan, nvbo);
 		if (unlikely(ret)) {
-			NV_ERROR(drm, "fail post-validate sync\n");
+			NV_ERROR(cli, "fail post-validate sync\n");
 			return ret;
 		}
 
@@ -478,7 +491,7 @@ nouveau_gem_pushbuf_validate(struct nouveau_channel *chan,
 			     uint64_t user_buffers, int nr_buffers,
 			     struct validate_op *op, int *apply_relocs)
 {
-	struct nouveau_drm *drm = chan->drm;
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	int ret, relocs = 0;
 
 	INIT_LIST_HEAD(&op->vram_list);
@@ -491,32 +504,32 @@ nouveau_gem_pushbuf_validate(struct nouveau_channel *chan,
 	ret = validate_init(chan, file_priv, pbbo, nr_buffers, op);
 	if (unlikely(ret)) {
 		if (ret != -ERESTARTSYS)
-			NV_ERROR(drm, "validate_init\n");
+			NV_ERROR(cli, "validate_init\n");
 		return ret;
 	}
 
-	ret = validate_list(chan, &op->vram_list, pbbo, user_buffers);
+	ret = validate_list(chan, cli, &op->vram_list, pbbo, user_buffers);
 	if (unlikely(ret < 0)) {
 		if (ret != -ERESTARTSYS)
-			NV_ERROR(drm, "validate vram_list\n");
+			NV_ERROR(cli, "validate vram_list\n");
 		validate_fini(op, NULL);
 		return ret;
 	}
 	relocs += ret;
 
-	ret = validate_list(chan, &op->gart_list, pbbo, user_buffers);
+	ret = validate_list(chan, cli, &op->gart_list, pbbo, user_buffers);
 	if (unlikely(ret < 0)) {
 		if (ret != -ERESTARTSYS)
-			NV_ERROR(drm, "validate gart_list\n");
+			NV_ERROR(cli, "validate gart_list\n");
 		validate_fini(op, NULL);
 		return ret;
 	}
 	relocs += ret;
 
-	ret = validate_list(chan, &op->both_list, pbbo, user_buffers);
+	ret = validate_list(chan, cli, &op->both_list, pbbo, user_buffers);
 	if (unlikely(ret < 0)) {
 		if (ret != -ERESTARTSYS)
-			NV_ERROR(drm, "validate both_list\n");
+			NV_ERROR(cli, "validate both_list\n");
 		validate_fini(op, NULL);
 		return ret;
 	}
@@ -545,11 +558,10 @@ u_memcpya(uint64_t user, unsigned nmemb, unsigned size)
 }
 
 static int
-nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
+nouveau_gem_pushbuf_reloc_apply(struct nouveau_cli *cli,
 				struct drm_nouveau_gem_pushbuf *req,
 				struct drm_nouveau_gem_pushbuf_bo *bo)
 {
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_nouveau_gem_pushbuf_reloc *reloc = NULL;
 	int ret = 0;
 	unsigned i;
@@ -565,7 +577,7 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 		uint32_t data;
 
 		if (unlikely(r->bo_index > req->nr_buffers)) {
-			NV_ERROR(drm, "reloc bo index invalid\n");
+			NV_ERROR(cli, "reloc bo index invalid\n");
 			ret = -EINVAL;
 			break;
 		}
@@ -575,7 +587,7 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 			continue;
 
 		if (unlikely(r->reloc_bo_index > req->nr_buffers)) {
-			NV_ERROR(drm, "reloc container bo index invalid\n");
+			NV_ERROR(cli, "reloc container bo index invalid\n");
 			ret = -EINVAL;
 			break;
 		}
@@ -583,7 +595,7 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 
 		if (unlikely(r->reloc_bo_offset + 4 >
 			     nvbo->bo.mem.num_pages << PAGE_SHIFT)) {
-			NV_ERROR(drm, "reloc outside of bo\n");
+			NV_ERROR(cli, "reloc outside of bo\n");
 			ret = -EINVAL;
 			break;
 		}
@@ -592,7 +604,7 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 			ret = ttm_bo_kmap(&nvbo->bo, 0, nvbo->bo.mem.num_pages,
 					  &nvbo->kmap);
 			if (ret) {
-				NV_ERROR(drm, "failed kmap for reloc\n");
+				NV_ERROR(cli, "failed kmap for reloc\n");
 				break;
 			}
 			nvbo->validate_mapped = true;
@@ -617,7 +629,7 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 		ret = ttm_bo_wait(&nvbo->bo, false, false, false);
 		spin_unlock(&nvbo->bo.bdev->fence_lock);
 		if (ret) {
-			NV_ERROR(drm, "reloc wait_idle failed: %d\n", ret);
+			NV_ERROR(cli, "reloc wait_idle failed: %d\n", ret);
 			break;
 		}
 
@@ -633,6 +645,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 			  struct drm_file *file_priv)
 {
 	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct nouveau_abi16_chan *temp;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_nouveau_gem_pushbuf *req = data;
@@ -662,19 +675,19 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 		goto out_next;
 
 	if (unlikely(req->nr_push > NOUVEAU_GEM_MAX_PUSH)) {
-		NV_ERROR(drm, "pushbuf push count exceeds limit: %d max %d\n",
+		NV_ERROR(cli, "pushbuf push count exceeds limit: %d max %d\n",
 			 req->nr_push, NOUVEAU_GEM_MAX_PUSH);
 		return nouveau_abi16_put(abi16, -EINVAL);
 	}
 
 	if (unlikely(req->nr_buffers > NOUVEAU_GEM_MAX_BUFFERS)) {
-		NV_ERROR(drm, "pushbuf bo count exceeds limit: %d max %d\n",
+		NV_ERROR(cli, "pushbuf bo count exceeds limit: %d max %d\n",
 			 req->nr_buffers, NOUVEAU_GEM_MAX_BUFFERS);
 		return nouveau_abi16_put(abi16, -EINVAL);
 	}
 
 	if (unlikely(req->nr_relocs > NOUVEAU_GEM_MAX_RELOCS)) {
-		NV_ERROR(drm, "pushbuf reloc count exceeds limit: %d max %d\n",
+		NV_ERROR(cli, "pushbuf reloc count exceeds limit: %d max %d\n",
 			 req->nr_relocs, NOUVEAU_GEM_MAX_RELOCS);
 		return nouveau_abi16_put(abi16, -EINVAL);
 	}
@@ -692,7 +705,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 	/* Ensure all push buffers are on validate list */
 	for (i = 0; i < req->nr_push; i++) {
 		if (push[i].bo_index >= req->nr_buffers) {
-			NV_ERROR(drm, "push %d buffer not in list\n", i);
+			NV_ERROR(cli, "push %d buffer not in list\n", i);
 			ret = -EINVAL;
 			goto out_prevalid;
 		}
@@ -703,15 +716,15 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 					   req->nr_buffers, &op, &do_reloc);
 	if (ret) {
 		if (ret != -ERESTARTSYS)
-			NV_ERROR(drm, "validate: %d\n", ret);
+			NV_ERROR(cli, "validate: %d\n", ret);
 		goto out_prevalid;
 	}
 
 	/* Apply any relocations that are required */
 	if (do_reloc) {
-		ret = nouveau_gem_pushbuf_reloc_apply(dev, req, bo);
+		ret = nouveau_gem_pushbuf_reloc_apply(cli, req, bo);
 		if (ret) {
-			NV_ERROR(drm, "reloc apply: %d\n", ret);
+			NV_ERROR(cli, "reloc apply: %d\n", ret);
 			goto out;
 		}
 	}
@@ -719,7 +732,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 	if (chan->dma.ib_max) {
 		ret = nouveau_dma_wait(chan, req->nr_push + 1, 16);
 		if (ret) {
-			NV_ERROR(drm, "nv50cal_space: %d\n", ret);
+			NV_ERROR(cli, "nv50cal_space: %d\n", ret);
 			goto out;
 		}
 
@@ -734,7 +747,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 	if (nv_device(drm->device)->chipset >= 0x25) {
 		ret = RING_SPACE(chan, req->nr_push * 2);
 		if (ret) {
-			NV_ERROR(drm, "cal_space: %d\n", ret);
+			NV_ERROR(cli, "cal_space: %d\n", ret);
 			goto out;
 		}
 
@@ -748,7 +761,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 	} else {
 		ret = RING_SPACE(chan, req->nr_push * (2 + NOUVEAU_DMA_SKIPS));
 		if (ret) {
-			NV_ERROR(drm, "jmp_space: %d\n", ret);
+			NV_ERROR(cli, "jmp_space: %d\n", ret);
 			goto out;
 		}
 
@@ -784,9 +797,9 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 		}
 	}
 
-	ret = nouveau_fence_new(chan, &fence);
+	ret = nouveau_fence_new(chan, false, &fence);
 	if (ret) {
-		NV_ERROR(drm, "error fencing pushbuf: %d\n", ret);
+		NV_ERROR(cli, "error fencing pushbuf: %d\n", ret);
 		WIND_RING(chan);
 		goto out;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.h b/drivers/gpu/drm/nouveau/nouveau_gem.h
index 5c1049236d22..8d7a3f0aeb86 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.h
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.h
@@ -35,9 +35,11 @@ extern int nouveau_gem_ioctl_cpu_fini(struct drm_device *, void *,
 extern int nouveau_gem_ioctl_info(struct drm_device *, void *,
 				  struct drm_file *);
 
-extern struct dma_buf *nouveau_gem_prime_export(struct drm_device *dev,
-				struct drm_gem_object *obj, int flags);
-extern struct drm_gem_object *nouveau_gem_prime_import(struct drm_device *dev,
-				struct dma_buf *dma_buf);
+extern int nouveau_gem_prime_pin(struct drm_gem_object *);
+extern struct sg_table *nouveau_gem_prime_get_sg_table(struct drm_gem_object *);
+extern struct drm_gem_object *nouveau_gem_prime_import_sg_table(
+	struct drm_device *, size_t size, struct sg_table *);
+extern void *nouveau_gem_prime_vmap(struct drm_gem_object *);
+extern void nouveau_gem_prime_vunmap(struct drm_gem_object *, void *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c b/drivers/gpu/drm/nouveau/nouveau_pm.c
index a701ff5ffa5b..bb54098c6d97 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_pm.c
@@ -409,6 +409,81 @@ static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, nouveau_hwmon_show_temp,
 						  NULL, 0);
 
 static ssize_t
+nouveau_hwmon_show_temp1_auto_point1_pwm(struct device *d,
+					 struct device_attribute *a, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", 100);
+}
+static SENSOR_DEVICE_ATTR(temp1_auto_point1_pwm, S_IRUGO,
+			  nouveau_hwmon_show_temp1_auto_point1_pwm, NULL, 0);
+
+static ssize_t
+nouveau_hwmon_temp1_auto_point1_temp(struct device *d,
+				     struct device_attribute *a, char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	      therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_FAN_BOOST) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_temp1_auto_point1_temp(struct device *d,
+					 struct device_attribute *a,
+					 const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_FAN_BOOST,
+			value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_auto_point1_temp, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_temp1_auto_point1_temp,
+			  nouveau_hwmon_set_temp1_auto_point1_temp, 0);
+
+static ssize_t
+nouveau_hwmon_temp1_auto_point1_temp_hyst(struct device *d,
+					  struct device_attribute *a, char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	 therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_FAN_BOOST_HYST) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_temp1_auto_point1_temp_hyst(struct device *d,
+					      struct device_attribute *a,
+					      const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_FAN_BOOST_HYST,
+			value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_auto_point1_temp_hyst, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_temp1_auto_point1_temp_hyst,
+			  nouveau_hwmon_set_temp1_auto_point1_temp_hyst, 0);
+
+static ssize_t
 nouveau_hwmon_max_temp(struct device *d, struct device_attribute *a, char *buf)
 {
 	struct drm_device *dev = dev_get_drvdata(d);
@@ -439,6 +514,38 @@ static SENSOR_DEVICE_ATTR(temp1_max, S_IRUGO | S_IWUSR, nouveau_hwmon_max_temp,
 						  0);
 
 static ssize_t
+nouveau_hwmon_max_temp_hyst(struct device *d, struct device_attribute *a,
+			    char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	  therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_DOWN_CLK_HYST) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_max_temp_hyst(struct device *d, struct device_attribute *a,
+						const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_DOWN_CLK_HYST,
+			value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_max_hyst, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_max_temp_hyst,
+			  nouveau_hwmon_set_max_temp_hyst, 0);
+
+static ssize_t
 nouveau_hwmon_critical_temp(struct device *d, struct device_attribute *a,
 							char *buf)
 {
@@ -471,6 +578,107 @@ static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO | S_IWUSR,
 						nouveau_hwmon_set_critical_temp,
 						0);
 
+static ssize_t
+nouveau_hwmon_critical_temp_hyst(struct device *d, struct device_attribute *a,
+							char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	  therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_CRITICAL_HYST) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_critical_temp_hyst(struct device *d,
+				     struct device_attribute *a,
+				     const char *buf,
+				     size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_CRITICAL_HYST,
+			value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_critical_temp_hyst,
+			  nouveau_hwmon_set_critical_temp_hyst, 0);
+static ssize_t
+nouveau_hwmon_emergency_temp(struct device *d, struct device_attribute *a,
+							char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	       therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_SHUTDOWN) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_emergency_temp(struct device *d, struct device_attribute *a,
+							    const char *buf,
+								size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_SHUTDOWN, value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_emergency, S_IRUGO | S_IWUSR,
+					nouveau_hwmon_emergency_temp,
+					nouveau_hwmon_set_emergency_temp,
+					0);
+
+static ssize_t
+nouveau_hwmon_emergency_temp_hyst(struct device *d, struct device_attribute *a,
+							char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+	  therm->attr_get(therm, NOUVEAU_THERM_ATTR_THRS_SHUTDOWN_HYST) * 1000);
+}
+static ssize_t
+nouveau_hwmon_set_emergency_temp_hyst(struct device *d,
+				      struct device_attribute *a,
+				      const char *buf,
+				      size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	long value;
+
+	if (kstrtol(buf, 10, &value) == -EINVAL)
+		return count;
+
+	therm->attr_set(therm, NOUVEAU_THERM_ATTR_THRS_SHUTDOWN_HYST,
+			value / 1000);
+
+	return count;
+}
+static SENSOR_DEVICE_ATTR(temp1_emergency_hyst, S_IRUGO | S_IWUSR,
+					nouveau_hwmon_emergency_temp_hyst,
+					nouveau_hwmon_set_emergency_temp_hyst,
+					0);
+
 static ssize_t nouveau_hwmon_show_name(struct device *dev,
 				      struct device_attribute *attr,
 				      char *buf)
@@ -490,7 +698,7 @@ static SENSOR_DEVICE_ATTR(update_rate, S_IRUGO,
 						NULL, 0);
 
 static ssize_t
-nouveau_hwmon_show_fan0_input(struct device *d, struct device_attribute *attr,
+nouveau_hwmon_show_fan1_input(struct device *d, struct device_attribute *attr,
 			      char *buf)
 {
 	struct drm_device *dev = dev_get_drvdata(d);
@@ -499,7 +707,7 @@ nouveau_hwmon_show_fan0_input(struct device *d, struct device_attribute *attr,
 
 	return snprintf(buf, PAGE_SIZE, "%d\n", therm->fan_sense(therm));
 }
-static SENSOR_DEVICE_ATTR(fan0_input, S_IRUGO, nouveau_hwmon_show_fan0_input,
+static SENSOR_DEVICE_ATTR(fan1_input, S_IRUGO, nouveau_hwmon_show_fan1_input,
 			  NULL, 0);
 
  static ssize_t
@@ -665,14 +873,21 @@ static SENSOR_DEVICE_ATTR(pwm1_max, S_IRUGO | S_IWUSR,
 
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
+	&sensor_dev_attr_temp1_auto_point1_pwm.dev_attr.attr,
+	&sensor_dev_attr_temp1_auto_point1_temp.dev_attr.attr,
+	&sensor_dev_attr_temp1_auto_point1_temp_hyst.dev_attr.attr,
 	&sensor_dev_attr_temp1_max.dev_attr.attr,
+	&sensor_dev_attr_temp1_max_hyst.dev_attr.attr,
 	&sensor_dev_attr_temp1_crit.dev_attr.attr,
+	&sensor_dev_attr_temp1_crit_hyst.dev_attr.attr,
+	&sensor_dev_attr_temp1_emergency.dev_attr.attr,
+	&sensor_dev_attr_temp1_emergency_hyst.dev_attr.attr,
 	&sensor_dev_attr_name.dev_attr.attr,
 	&sensor_dev_attr_update_rate.dev_attr.attr,
 	NULL
 };
 static struct attribute *hwmon_fan_rpm_attributes[] = {
-	&sensor_dev_attr_fan0_input.dev_attr.attr,
+	&sensor_dev_attr_fan1_input.dev_attr.attr,
 	NULL
 };
 static struct attribute *hwmon_pwm_fan_attributes[] = {
@@ -717,7 +932,7 @@ nouveau_hwmon_init(struct drm_device *dev)
 	dev_set_drvdata(hwmon_dev, dev);
 
 	/* default sysfs entries */
-	ret = sysfs_create_group(&dev->pdev->dev.kobj, &hwmon_attrgroup);
+	ret = sysfs_create_group(&hwmon_dev->kobj, &hwmon_attrgroup);
 	if (ret) {
 		if (ret)
 			goto error;
@@ -728,7 +943,7 @@ nouveau_hwmon_init(struct drm_device *dev)
 	 *     the gpio entries for pwm fan control even when there's no
 	 *     actual fan connected to it... therm table? */
 	if (therm->fan_get && therm->fan_get(therm) >= 0) {
-		ret = sysfs_create_group(&dev->pdev->dev.kobj,
+		ret = sysfs_create_group(&hwmon_dev->kobj,
 					 &hwmon_pwm_fan_attrgroup);
 		if (ret)
 			goto error;
@@ -736,7 +951,7 @@ nouveau_hwmon_init(struct drm_device *dev)
 
 	/* if the card can read the fan rpm */
 	if (therm->fan_sense(therm) >= 0) {
-		ret = sysfs_create_group(&dev->pdev->dev.kobj,
+		ret = sysfs_create_group(&hwmon_dev->kobj,
 					 &hwmon_fan_rpm_attrgroup);
 		if (ret)
 			goto error;
@@ -764,10 +979,10 @@ nouveau_hwmon_fini(struct drm_device *dev)
 	struct nouveau_pm *pm = nouveau_pm(dev);
 
 	if (pm->hwmon) {
-		sysfs_remove_group(&dev->pdev->dev.kobj, &hwmon_attrgroup);
-		sysfs_remove_group(&dev->pdev->dev.kobj,
+		sysfs_remove_group(&pm->hwmon->kobj, &hwmon_attrgroup);
+		sysfs_remove_group(&pm->hwmon->kobj,
 				   &hwmon_pwm_fan_attrgroup);
-		sysfs_remove_group(&dev->pdev->dev.kobj,
+		sysfs_remove_group(&pm->hwmon->kobj,
 				   &hwmon_fan_rpm_attrgroup);
 
 		hwmon_device_unregister(pm->hwmon);
diff --git a/drivers/gpu/drm/nouveau/nouveau_prime.c b/drivers/gpu/drm/nouveau/nouveau_prime.c
index b8e05ae38212..f53e10874cae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_prime.c
+++ b/drivers/gpu/drm/nouveau/nouveau_prime.c
@@ -22,126 +22,42 @@
  * Authors: Dave Airlie
  */
 
-#include <linux/dma-buf.h>
-
 #include <drm/drmP.h>
 
 #include "nouveau_drm.h"
 #include "nouveau_gem.h"
 
-static struct sg_table *nouveau_gem_map_dma_buf(struct dma_buf_attachment *attachment,
-					  enum dma_data_direction dir)
+struct sg_table *nouveau_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
-	struct nouveau_bo *nvbo = attachment->dmabuf->priv;
-	struct drm_device *dev = nvbo->gem->dev;
+	struct nouveau_bo *nvbo = nouveau_gem_object(obj);
 	int npages = nvbo->bo.num_pages;
-	struct sg_table *sg;
-	int nents;
-
-	mutex_lock(&dev->struct_mutex);
-	sg = drm_prime_pages_to_sg(nvbo->bo.ttm->pages, npages);
-	nents = dma_map_sg(attachment->dev, sg->sgl, sg->nents, dir);
-	mutex_unlock(&dev->struct_mutex);
-	return sg;
-}
-
-static void nouveau_gem_unmap_dma_buf(struct dma_buf_attachment *attachment,
-				      struct sg_table *sg, enum dma_data_direction dir)
-{
-	dma_unmap_sg(attachment->dev, sg->sgl, sg->nents, dir);
-	sg_free_table(sg);
-	kfree(sg);
-}
-
-static void nouveau_gem_dmabuf_release(struct dma_buf *dma_buf)
-{
-	struct nouveau_bo *nvbo = dma_buf->priv;
-
-	if (nvbo->gem->export_dma_buf == dma_buf) {
-		nvbo->gem->export_dma_buf = NULL;
-		drm_gem_object_unreference_unlocked(nvbo->gem);
-	}
-}
-
-static void *nouveau_gem_kmap_atomic(struct dma_buf *dma_buf, unsigned long page_num)
-{
-	return NULL;
-}
-
-static void nouveau_gem_kunmap_atomic(struct dma_buf *dma_buf, unsigned long page_num, void *addr)
-{
 
-}
-static void *nouveau_gem_kmap(struct dma_buf *dma_buf, unsigned long page_num)
-{
-	return NULL;
+	return drm_prime_pages_to_sg(nvbo->bo.ttm->pages, npages);
 }
 
-static void nouveau_gem_kunmap(struct dma_buf *dma_buf, unsigned long page_num, void *addr)
+void *nouveau_gem_prime_vmap(struct drm_gem_object *obj)
 {
-
-}
-
-static int nouveau_gem_prime_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma)
-{
-	return -EINVAL;
-}
-
-static void *nouveau_gem_prime_vmap(struct dma_buf *dma_buf)
-{
-	struct nouveau_bo *nvbo = dma_buf->priv;
-	struct drm_device *dev = nvbo->gem->dev;
+	struct nouveau_bo *nvbo = nouveau_gem_object(obj);
 	int ret;
 
-	mutex_lock(&dev->struct_mutex);
-	if (nvbo->vmapping_count) {
-		nvbo->vmapping_count++;
-		goto out_unlock;
-	}
-
 	ret = ttm_bo_kmap(&nvbo->bo, 0, nvbo->bo.num_pages,
 			  &nvbo->dma_buf_vmap);
-	if (ret) {
-		mutex_unlock(&dev->struct_mutex);
+	if (ret)
 		return ERR_PTR(ret);
-	}
-	nvbo->vmapping_count = 1;
-out_unlock:
-	mutex_unlock(&dev->struct_mutex);
+
 	return nvbo->dma_buf_vmap.virtual;
 }
 
-static void nouveau_gem_prime_vunmap(struct dma_buf *dma_buf, void *vaddr)
+void nouveau_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
 {
-	struct nouveau_bo *nvbo = dma_buf->priv;
-	struct drm_device *dev = nvbo->gem->dev;
+	struct nouveau_bo *nvbo = nouveau_gem_object(obj);
 
-	mutex_lock(&dev->struct_mutex);
-	nvbo->vmapping_count--;
-	if (nvbo->vmapping_count == 0) {
-		ttm_bo_kunmap(&nvbo->dma_buf_vmap);
-	}
-	mutex_unlock(&dev->struct_mutex);
+	ttm_bo_kunmap(&nvbo->dma_buf_vmap);
 }
 
-static const struct dma_buf_ops nouveau_dmabuf_ops =  {
-	.map_dma_buf = nouveau_gem_map_dma_buf,
-	.unmap_dma_buf = nouveau_gem_unmap_dma_buf,
-	.release = nouveau_gem_dmabuf_release,
-	.kmap = nouveau_gem_kmap,
-	.kmap_atomic = nouveau_gem_kmap_atomic,
-	.kunmap = nouveau_gem_kunmap,
-	.kunmap_atomic = nouveau_gem_kunmap_atomic,
-	.mmap = nouveau_gem_prime_mmap,
-	.vmap = nouveau_gem_prime_vmap,
-	.vunmap = nouveau_gem_prime_vunmap,
-};
-
-static int
-nouveau_prime_new(struct drm_device *dev,
-		  size_t size,
-		  struct sg_table *sg,
-		  struct nouveau_bo **pnvbo)
+struct drm_gem_object *nouveau_gem_prime_import_sg_table(struct drm_device *dev,
+							 size_t size,
+							 struct sg_table *sg)
 {
 	struct nouveau_bo *nvbo;
 	u32 flags = 0;
@@ -150,24 +66,22 @@ nouveau_prime_new(struct drm_device *dev,
 	flags = TTM_PL_FLAG_TT;
 
 	ret = nouveau_bo_new(dev, size, 0, flags, 0, 0,
-			     sg, pnvbo);
+			     sg, &nvbo);
 	if (ret)
-		return ret;
-	nvbo = *pnvbo;
+		return ERR_PTR(ret);
 
 	nvbo->valid_domains = NOUVEAU_GEM_DOMAIN_GART;
 	nvbo->gem = drm_gem_object_alloc(dev, nvbo->bo.mem.size);
 	if (!nvbo->gem) {
-		nouveau_bo_ref(NULL, pnvbo);
-		return -ENOMEM;
+		nouveau_bo_ref(NULL, &nvbo);
+		return ERR_PTR(-ENOMEM);
 	}
 
 	nvbo->gem->driver_private = nvbo;
-	return 0;
+	return nvbo->gem;
 }
 
-struct dma_buf *nouveau_gem_prime_export(struct drm_device *dev,
-				struct drm_gem_object *obj, int flags)
+int nouveau_gem_prime_pin(struct drm_gem_object *obj)
 {
 	struct nouveau_bo *nvbo = nouveau_gem_object(obj);
 	int ret = 0;
@@ -175,52 +89,7 @@ struct dma_buf *nouveau_gem_prime_export(struct drm_device *dev,
 	/* pin buffer into GTT */
 	ret = nouveau_bo_pin(nvbo, TTM_PL_FLAG_TT);
 	if (ret)
-		return ERR_PTR(-EINVAL);
-
-	return dma_buf_export(nvbo, &nouveau_dmabuf_ops, obj->size, flags);
-}
-
-struct drm_gem_object *nouveau_gem_prime_import(struct drm_device *dev,
-				struct dma_buf *dma_buf)
-{
-	struct dma_buf_attachment *attach;
-	struct sg_table *sg;
-	struct nouveau_bo *nvbo;
-	int ret;
-
-	if (dma_buf->ops == &nouveau_dmabuf_ops) {
-		nvbo = dma_buf->priv;
-		if (nvbo->gem) {
-			if (nvbo->gem->dev == dev) {
-				drm_gem_object_reference(nvbo->gem);
-				dma_buf_put(dma_buf);
-				return nvbo->gem;
-			}
-		}
-	}
-	/* need to attach */
-	attach = dma_buf_attach(dma_buf, dev->dev);
-	if (IS_ERR(attach))
-		return ERR_PTR(PTR_ERR(attach));
-
-	sg = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL);
-	if (IS_ERR(sg)) {
-		ret = PTR_ERR(sg);
-		goto fail_detach;
-	}
-
-	ret = nouveau_prime_new(dev, dma_buf->size, sg, &nvbo);
-	if (ret)
-		goto fail_unmap;
-
-	nvbo->gem->import_attach = attach;
-
-	return nvbo->gem;
+		return -EINVAL;
 
-fail_unmap:
-	dma_buf_unmap_attachment(attach, sg, DMA_BIDIRECTIONAL);
-fail_detach:
-	dma_buf_detach(dma_buf, attach);
-	return ERR_PTR(ret);
+	return 0;
 }
-
diff --git a/drivers/gpu/drm/nouveau/nv04_dfp.c b/drivers/gpu/drm/nouveau/nv04_dfp.c
index 39ffc07f906b..7e24cdf1cb39 100644
--- a/drivers/gpu/drm/nouveau/nv04_dfp.c
+++ b/drivers/gpu/drm/nouveau/nv04_dfp.c
@@ -490,8 +490,8 @@ static void nv04_dfp_update_backlight(struct drm_encoder *encoder, int mode)
 	/* BIOS scripts usually take care of the backlight, thanks
 	 * Apple for your consistency.
 	 */
-	if (dev->pci_device == 0x0179 || dev->pci_device == 0x0189 ||
-	    dev->pci_device == 0x0329) {
+	if (dev->pci_device == 0x0174 || dev->pci_device == 0x0179 ||
+	    dev->pci_device == 0x0189 || dev->pci_device == 0x0329) {
 		if (mode == DRM_MODE_DPMS_ON) {
 			nv_mask(device, NV_PBUS_DEBUG_DUALHEAD_CTL, 0, 1 << 31);
 			nv_mask(device, NV_PCRTC_GPIO_EXT, 3, 1);
diff --git a/drivers/gpu/drm/nouveau/nv04_display.c b/drivers/gpu/drm/nouveau/nv04_display.c
index 2cd6fb8c548e..ad48444c385c 100644
--- a/drivers/gpu/drm/nouveau/nv04_display.c
+++ b/drivers/gpu/drm/nouveau/nv04_display.c
@@ -22,6 +22,9 @@
  * Author: Ben Skeggs
  */
 
+#include <core/object.h>
+#include <core/class.h>
+
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 
@@ -31,6 +34,8 @@
 #include "nouveau_encoder.h"
 #include "nouveau_connector.h"
 
+#include <subdev/i2c.h>
+
 int
 nv04_display_early_init(struct drm_device *dev)
 {
@@ -53,6 +58,7 @@ int
 nv04_display_create(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct dcb_table *dcb = &drm->vbios.dcb;
 	struct drm_connector *connector, *ct;
 	struct drm_encoder *encoder;
@@ -71,6 +77,11 @@ nv04_display_create(struct drm_device *dev)
 
 	nouveau_hw_save_vga_fonts(dev, 1);
 
+	ret = nouveau_object_new(nv_object(drm), NVDRM_DEVICE, 0xd1500000,
+				 NV04_DISP_CLASS, NULL, 0, &disp->core);
+	if (ret)
+		return ret;
+
 	nv04_crtc_create(dev, 0);
 	if (nv_two_heads(dev))
 		nv04_crtc_create(dev, 1);
@@ -114,6 +125,11 @@ nv04_display_create(struct drm_device *dev)
 		}
 	}
 
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+		nv_encoder->i2c = i2c->find(i2c, nv_encoder->dcb->i2c_index);
+	}
+
 	/* Save previous state */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 		crtc->funcs->save(crtc);
@@ -140,7 +156,7 @@ nv04_display_destroy(struct drm_device *dev)
 			.crtc = crtc,
 		};
 
-		crtc->funcs->set_config(&modeset);
+		drm_mode_set_config_internal(&modeset);
 	}
 
 	/* Restore state */
diff --git a/drivers/gpu/drm/nouveau/nv04_display.h b/drivers/gpu/drm/nouveau/nv04_display.h
index 45322802e37d..a0a031dad13f 100644
--- a/drivers/gpu/drm/nouveau/nv04_display.h
+++ b/drivers/gpu/drm/nouveau/nv04_display.h
@@ -80,6 +80,7 @@ struct nv04_display {
 	struct nv04_mode_state saved_reg;
 	uint32_t saved_vga_font[4][16384];
 	uint32_t dac_users[4];
+	struct nouveau_object *core;
 };
 
 static inline struct nv04_display *
diff --git a/drivers/gpu/drm/nouveau/nv04_fence.c b/drivers/gpu/drm/nouveau/nv04_fence.c
index a220b94ba9f2..94eadd1dd10a 100644
--- a/drivers/gpu/drm/nouveau/nv04_fence.c
+++ b/drivers/gpu/drm/nouveau/nv04_fence.c
@@ -78,6 +78,9 @@ nv04_fence_context_new(struct nouveau_channel *chan)
 	struct nv04_fence_chan *fctx = kzalloc(sizeof(*fctx), GFP_KERNEL);
 	if (fctx) {
 		nouveau_fence_context_new(&fctx->base);
+		fctx->base.emit = nv04_fence_emit;
+		fctx->base.sync = nv04_fence_sync;
+		fctx->base.read = nv04_fence_read;
 		chan->fence = fctx;
 		return 0;
 	}
@@ -104,8 +107,5 @@ nv04_fence_create(struct nouveau_drm *drm)
 	priv->base.dtor = nv04_fence_destroy;
 	priv->base.context_new = nv04_fence_context_new;
 	priv->base.context_del = nv04_fence_context_del;
-	priv->base.emit = nv04_fence_emit;
-	priv->base.sync = nv04_fence_sync;
-	priv->base.read = nv04_fence_read;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nv04_tv.c b/drivers/gpu/drm/nouveau/nv04_tv.c
index 62e826a139b3..4a69ccdef9b4 100644
--- a/drivers/gpu/drm/nouveau/nv04_tv.c
+++ b/drivers/gpu/drm/nouveau/nv04_tv.c
@@ -184,14 +184,23 @@ static const struct drm_encoder_funcs nv04_tv_funcs = {
 	.destroy = nv04_tv_destroy,
 };
 
+static const struct drm_encoder_helper_funcs nv04_tv_helper_funcs = {
+	.dpms = nv04_tv_dpms,
+	.save = drm_i2c_encoder_save,
+	.restore = drm_i2c_encoder_restore,
+	.mode_fixup = drm_i2c_encoder_mode_fixup,
+	.prepare = nv04_tv_prepare,
+	.commit = nv04_tv_commit,
+	.mode_set = nv04_tv_mode_set,
+	.detect = drm_i2c_encoder_detect,
+};
+
 int
 nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry)
 {
 	struct nouveau_encoder *nv_encoder;
 	struct drm_encoder *encoder;
 	struct drm_device *dev = connector->dev;
-	struct drm_encoder_helper_funcs *hfuncs;
-	struct drm_encoder_slave_funcs *sfuncs;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_i2c_port *port = i2c->find(i2c, entry->i2c_index);
@@ -207,17 +216,11 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry)
 	if (!nv_encoder)
 		return -ENOMEM;
 
-	hfuncs = kzalloc(sizeof(*hfuncs), GFP_KERNEL);
-	if (!hfuncs) {
-		ret = -ENOMEM;
-		goto fail_free;
-	}
-
 	/* Initialize the common members */
 	encoder = to_drm_encoder(nv_encoder);
 
 	drm_encoder_init(dev, encoder, &nv04_tv_funcs, DRM_MODE_ENCODER_TVDAC);
-	drm_encoder_helper_add(encoder, hfuncs);
+	drm_encoder_helper_add(encoder, &nv04_tv_helper_funcs);
 
 	encoder->possible_crtcs = entry->heads;
 	encoder->possible_clones = 0;
@@ -230,30 +233,14 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry)
 	if (ret < 0)
 		goto fail_cleanup;
 
-	/* Fill the function pointers */
-	sfuncs = get_slave_funcs(encoder);
-
-	*hfuncs = (struct drm_encoder_helper_funcs) {
-		.dpms = nv04_tv_dpms,
-		.save = sfuncs->save,
-		.restore = sfuncs->restore,
-		.mode_fixup = sfuncs->mode_fixup,
-		.prepare = nv04_tv_prepare,
-		.commit = nv04_tv_commit,
-		.mode_set = nv04_tv_mode_set,
-		.detect = sfuncs->detect,
-	};
-
 	/* Attach it to the specified connector. */
-	sfuncs->create_resources(encoder, connector);
+	get_slave_funcs(encoder)->create_resources(encoder, connector);
 	drm_mode_connector_attach_encoder(connector, encoder);
 
 	return 0;
 
 fail_cleanup:
 	drm_encoder_cleanup(encoder);
-	kfree(hfuncs);
-fail_free:
 	kfree(nv_encoder);
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nv10_fence.c b/drivers/gpu/drm/nouveau/nv10_fence.c
index 03017f24d593..06f434f03fba 100644
--- a/drivers/gpu/drm/nouveau/nv10_fence.c
+++ b/drivers/gpu/drm/nouveau/nv10_fence.c
@@ -27,18 +27,7 @@
 
 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
-#include "nouveau_fence.h"
-
-struct nv10_fence_chan {
-	struct nouveau_fence_chan base;
-};
-
-struct nv10_fence_priv {
-	struct nouveau_fence_priv base;
-	struct nouveau_bo *bo;
-	spinlock_t lock;
-	u32 sequence;
-};
+#include "nv10_fence.h"
 
 int
 nv10_fence_emit(struct nouveau_fence *fence)
@@ -61,45 +50,6 @@ nv10_fence_sync(struct nouveau_fence *fence,
 	return -ENODEV;
 }
 
-int
-nv17_fence_sync(struct nouveau_fence *fence,
-		struct nouveau_channel *prev, struct nouveau_channel *chan)
-{
-	struct nv10_fence_priv *priv = chan->drm->fence;
-	u32 value;
-	int ret;
-
-	if (!mutex_trylock(&prev->cli->mutex))
-		return -EBUSY;
-
-	spin_lock(&priv->lock);
-	value = priv->sequence;
-	priv->sequence += 2;
-	spin_unlock(&priv->lock);
-
-	ret = RING_SPACE(prev, 5);
-	if (!ret) {
-		BEGIN_NV04(prev, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 4);
-		OUT_RING  (prev, NvSema);
-		OUT_RING  (prev, 0);
-		OUT_RING  (prev, value + 0);
-		OUT_RING  (prev, value + 1);
-		FIRE_RING (prev);
-	}
-
-	if (!ret && !(ret = RING_SPACE(chan, 5))) {
-		BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 4);
-		OUT_RING  (chan, NvSema);
-		OUT_RING  (chan, 0);
-		OUT_RING  (chan, value + 1);
-		OUT_RING  (chan, value + 2);
-		FIRE_RING (chan);
-	}
-
-	mutex_unlock(&prev->cli->mutex);
-	return 0;
-}
-
 u32
 nv10_fence_read(struct nouveau_channel *chan)
 {
@@ -115,39 +65,20 @@ nv10_fence_context_del(struct nouveau_channel *chan)
 	kfree(fctx);
 }
 
-static int
+int
 nv10_fence_context_new(struct nouveau_channel *chan)
 {
-	struct nv10_fence_priv *priv = chan->drm->fence;
 	struct nv10_fence_chan *fctx;
-	int ret = 0;
 
 	fctx = chan->fence = kzalloc(sizeof(*fctx), GFP_KERNEL);
 	if (!fctx)
 		return -ENOMEM;
 
 	nouveau_fence_context_new(&fctx->base);
-
-	if (priv->bo) {
-		struct ttm_mem_reg *mem = &priv->bo->bo.mem;
-		struct nouveau_object *object;
-		u32 start = mem->start * PAGE_SIZE;
-		u32 limit = mem->start + mem->size - 1;
-
-		ret = nouveau_object_new(nv_object(chan->cli), chan->handle,
-					 NvSema, 0x0002,
-					 &(struct nv_dma_class) {
-						.flags = NV_DMA_TARGET_VRAM |
-							 NV_DMA_ACCESS_RDWR,
-						.start = start,
-						.limit = limit,
-					 }, sizeof(struct nv_dma_class),
-					 &object);
-	}
-
-	if (ret)
-		nv10_fence_context_del(chan);
-	return ret;
+	fctx->base.emit = nv10_fence_emit;
+	fctx->base.read = nv10_fence_read;
+	fctx->base.sync = nv10_fence_sync;
+	return 0;
 }
 
 void
@@ -162,18 +93,10 @@ nv10_fence_destroy(struct nouveau_drm *drm)
 	kfree(priv);
 }
 
-void nv17_fence_resume(struct nouveau_drm *drm)
-{
-	struct nv10_fence_priv *priv = drm->fence;
-
-	nouveau_bo_wr32(priv->bo, 0, priv->sequence);
-}
-
 int
 nv10_fence_create(struct nouveau_drm *drm)
 {
 	struct nv10_fence_priv *priv;
-	int ret = 0;
 
 	priv = drm->fence = kzalloc(sizeof(*priv), GFP_KERNEL);
 	if (!priv)
@@ -182,33 +105,6 @@ nv10_fence_create(struct nouveau_drm *drm)
 	priv->base.dtor = nv10_fence_destroy;
 	priv->base.context_new = nv10_fence_context_new;
 	priv->base.context_del = nv10_fence_context_del;
-	priv->base.emit = nv10_fence_emit;
-	priv->base.read = nv10_fence_read;
-	priv->base.sync = nv10_fence_sync;
 	spin_lock_init(&priv->lock);
-
-	if (nv_device(drm->device)->chipset >= 0x17) {
-		ret = nouveau_bo_new(drm->dev, 4096, 0x1000, TTM_PL_FLAG_VRAM,
-				     0, 0x0000, NULL, &priv->bo);
-		if (!ret) {
-			ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM);
-			if (!ret) {
-				ret = nouveau_bo_map(priv->bo);
-				if (ret)
-					nouveau_bo_unpin(priv->bo);
-			}
-			if (ret)
-				nouveau_bo_ref(NULL, &priv->bo);
-		}
-
-		if (ret == 0) {
-			nouveau_bo_wr32(priv->bo, 0x000, 0x00000000);
-			priv->base.sync = nv17_fence_sync;
-			priv->base.resume = nv17_fence_resume;
-		}
-	}
-
-	if (ret)
-		nv10_fence_destroy(drm);
-	return ret;
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nv10_fence.h b/drivers/gpu/drm/nouveau/nv10_fence.h
new file mode 100644
index 000000000000..e5d9204826c2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv10_fence.h
@@ -0,0 +1,19 @@
+#ifndef __NV10_FENCE_H_
+#define __NV10_FENCE_H_
+
+#include <core/os.h>
+#include "nouveau_fence.h"
+#include "nouveau_bo.h"
+
+struct nv10_fence_chan {
+	struct nouveau_fence_chan base;
+};
+
+struct nv10_fence_priv {
+	struct nouveau_fence_priv base;
+	struct nouveau_bo *bo;
+	spinlock_t lock;
+	u32 sequence;
+};
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nv17_fence.c b/drivers/gpu/drm/nouveau/nv17_fence.c
new file mode 100644
index 000000000000..8e47a9bae8c3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv17_fence.c
@@ -0,0 +1,149 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <core/object.h>
+#include <core/class.h>
+
+#include "nouveau_drm.h"
+#include "nouveau_dma.h"
+#include "nv10_fence.h"
+
+int
+nv17_fence_sync(struct nouveau_fence *fence,
+		struct nouveau_channel *prev, struct nouveau_channel *chan)
+{
+	struct nv10_fence_priv *priv = chan->drm->fence;
+	u32 value;
+	int ret;
+
+	if (!mutex_trylock(&prev->cli->mutex))
+		return -EBUSY;
+
+	spin_lock(&priv->lock);
+	value = priv->sequence;
+	priv->sequence += 2;
+	spin_unlock(&priv->lock);
+
+	ret = RING_SPACE(prev, 5);
+	if (!ret) {
+		BEGIN_NV04(prev, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 4);
+		OUT_RING  (prev, NvSema);
+		OUT_RING  (prev, 0);
+		OUT_RING  (prev, value + 0);
+		OUT_RING  (prev, value + 1);
+		FIRE_RING (prev);
+	}
+
+	if (!ret && !(ret = RING_SPACE(chan, 5))) {
+		BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 4);
+		OUT_RING  (chan, NvSema);
+		OUT_RING  (chan, 0);
+		OUT_RING  (chan, value + 1);
+		OUT_RING  (chan, value + 2);
+		FIRE_RING (chan);
+	}
+
+	mutex_unlock(&prev->cli->mutex);
+	return 0;
+}
+
+static int
+nv17_fence_context_new(struct nouveau_channel *chan)
+{
+	struct nv10_fence_priv *priv = chan->drm->fence;
+	struct nv10_fence_chan *fctx;
+	struct ttm_mem_reg *mem = &priv->bo->bo.mem;
+	struct nouveau_object *object;
+	u32 start = mem->start * PAGE_SIZE;
+	u32 limit = mem->start + mem->size - 1;
+	int ret = 0;
+
+	fctx = chan->fence = kzalloc(sizeof(*fctx), GFP_KERNEL);
+	if (!fctx)
+		return -ENOMEM;
+
+	nouveau_fence_context_new(&fctx->base);
+	fctx->base.emit = nv10_fence_emit;
+	fctx->base.read = nv10_fence_read;
+	fctx->base.sync = nv17_fence_sync;
+
+	ret = nouveau_object_new(nv_object(chan->cli), chan->handle,
+				 NvSema, 0x0002,
+				 &(struct nv_dma_class) {
+					.flags = NV_DMA_TARGET_VRAM |
+						 NV_DMA_ACCESS_RDWR,
+					.start = start,
+					.limit = limit,
+				 }, sizeof(struct nv_dma_class),
+				 &object);
+	if (ret)
+		nv10_fence_context_del(chan);
+	return ret;
+}
+
+void
+nv17_fence_resume(struct nouveau_drm *drm)
+{
+	struct nv10_fence_priv *priv = drm->fence;
+
+	nouveau_bo_wr32(priv->bo, 0, priv->sequence);
+}
+
+int
+nv17_fence_create(struct nouveau_drm *drm)
+{
+	struct nv10_fence_priv *priv;
+	int ret = 0;
+
+	priv = drm->fence = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->base.dtor = nv10_fence_destroy;
+	priv->base.resume = nv17_fence_resume;
+	priv->base.context_new = nv17_fence_context_new;
+	priv->base.context_del = nv10_fence_context_del;
+	spin_lock_init(&priv->lock);
+
+	ret = nouveau_bo_new(drm->dev, 4096, 0x1000, TTM_PL_FLAG_VRAM,
+			     0, 0x0000, NULL, &priv->bo);
+	if (!ret) {
+		ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM);
+		if (!ret) {
+			ret = nouveau_bo_map(priv->bo);
+			if (ret)
+				nouveau_bo_unpin(priv->bo);
+		}
+		if (ret)
+			nouveau_bo_ref(NULL, &priv->bo);
+	}
+
+	if (ret) {
+		nv10_fence_destroy(drm);
+		return ret;
+	}
+
+	nouveau_bo_wr32(priv->bo, 0x000, 0x00000000);
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/nv17_tv.c b/drivers/gpu/drm/nouveau/nv17_tv.c
index 2ca276ada507..977e42be2050 100644
--- a/drivers/gpu/drm/nouveau/nv17_tv.c
+++ b/drivers/gpu/drm/nouveau/nv17_tv.c
@@ -768,7 +768,7 @@ static int nv17_tv_set_property(struct drm_encoder *encoder,
 				.crtc = crtc,
 			};
 
-			crtc->funcs->set_config(&modeset);
+			drm_mode_set_config_internal(&modeset);
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 35874085a61e..a6237c9cbbc3 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -43,6 +43,7 @@
 #include <subdev/timer.h>
 #include <subdev/bar.h>
 #include <subdev/fb.h>
+#include <subdev/i2c.h>
 
 #define EVO_DMA_NR 9
 
@@ -128,6 +129,11 @@ struct nv50_dmac {
 	struct nv50_chan base;
 	dma_addr_t handle;
 	u32 *ptr;
+
+	/* Protects against concurrent pushbuf access to this channel, lock is
+	 * grabbed by evo_wait (if the pushbuf reservation is successful) and
+	 * dropped again by evo_kick. */
+	struct mutex lock;
 };
 
 static void
@@ -271,6 +277,8 @@ nv50_dmac_create(struct nouveau_object *core, u32 bclass, u8 head,
 	u32 pushbuf = *(u32 *)data;
 	int ret;
 
+	mutex_init(&dmac->lock);
+
 	dmac->ptr = pci_alloc_consistent(nv_device(core)->pdev, PAGE_SIZE,
 					&dmac->handle);
 	if (!dmac->ptr)
@@ -395,11 +403,13 @@ evo_wait(void *evoc, int nr)
 	struct nv50_dmac *dmac = evoc;
 	u32 put = nv_ro32(dmac->base.user, 0x0000) / 4;
 
+	mutex_lock(&dmac->lock);
 	if (put + nr >= (PAGE_SIZE / 4) - 8) {
 		dmac->ptr[put] = 0x20000000;
 
 		nv_wo32(dmac->base.user, 0x0000, 0x00000000);
 		if (!nv_wait(dmac->base.user, 0x0004, ~0, 0x00000000)) {
+			mutex_unlock(&dmac->lock);
 			NV_ERROR(dmac->base.user, "channel stalled\n");
 			return NULL;
 		}
@@ -415,6 +425,7 @@ evo_kick(u32 *push, void *evoc)
 {
 	struct nv50_dmac *dmac = evoc;
 	nv_wo32(dmac->base.user, 0x0000, (push - dmac->ptr) << 2);
+	mutex_unlock(&dmac->lock);
 }
 
 #define evo_mthd(p,m,s) *((p)++) = (((s) << 18) | (m))
@@ -423,7 +434,10 @@ evo_kick(u32 *push, void *evoc)
 static bool
 evo_sync_wait(void *data)
 {
-	return nouveau_bo_rd32(data, EVO_MAST_NTFY) != 0x00000000;
+	if (nouveau_bo_rd32(data, EVO_MAST_NTFY) != 0x00000000)
+		return true;
+	usleep_range(1, 2);
+	return false;
 }
 
 static int
@@ -502,7 +516,7 @@ nv50_display_flip_next(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		if (ret)
 			return ret;
 
-		if (nv_mclass(chan->object) < NVC0_CHANNEL_IND_CLASS) {
+		if (nv_mclass(chan->object) < NV84_CHANNEL_IND_CLASS) {
 			BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 2);
 			OUT_RING  (chan, NvEvoSema0 + nv_crtc->index);
 			OUT_RING  (chan, sync->sem.offset);
@@ -512,24 +526,36 @@ nv50_display_flip_next(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			OUT_RING  (chan, sync->sem.offset ^ 0x10);
 			OUT_RING  (chan, 0x74b1e000);
 			BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 1);
-			if (nv_mclass(chan->object) < NV84_CHANNEL_DMA_CLASS)
-				OUT_RING  (chan, NvSema);
-			else
-				OUT_RING  (chan, chan->vram);
+			OUT_RING  (chan, NvSema);
+		} else
+		if (nv_mclass(chan->object) < NVC0_CHANNEL_IND_CLASS) {
+			u64 offset = nv84_fence_crtc(chan, nv_crtc->index);
+			offset += sync->sem.offset;
+
+			BEGIN_NV04(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
+			OUT_RING  (chan, upper_32_bits(offset));
+			OUT_RING  (chan, lower_32_bits(offset));
+			OUT_RING  (chan, 0xf00d0000 | sync->sem.value);
+			OUT_RING  (chan, 0x00000002);
+			BEGIN_NV04(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
+			OUT_RING  (chan, upper_32_bits(offset));
+			OUT_RING  (chan, lower_32_bits(offset ^ 0x10));
+			OUT_RING  (chan, 0x74b1e000);
+			OUT_RING  (chan, 0x00000001);
 		} else {
-			u64 offset = nvc0_fence_crtc(chan, nv_crtc->index);
+			u64 offset = nv84_fence_crtc(chan, nv_crtc->index);
 			offset += sync->sem.offset;
 
 			BEGIN_NVC0(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
 			OUT_RING  (chan, upper_32_bits(offset));
 			OUT_RING  (chan, lower_32_bits(offset));
 			OUT_RING  (chan, 0xf00d0000 | sync->sem.value);
-			OUT_RING  (chan, 0x1002);
+			OUT_RING  (chan, 0x00001002);
 			BEGIN_NVC0(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
 			OUT_RING  (chan, upper_32_bits(offset));
 			OUT_RING  (chan, lower_32_bits(offset ^ 0x10));
 			OUT_RING  (chan, 0x74b1e000);
-			OUT_RING  (chan, 0x1001);
+			OUT_RING  (chan, 0x00001001);
 		}
 
 		FIRE_RING (chan);
@@ -1493,9 +1519,6 @@ nv50_dac_disconnect(struct drm_encoder *encoder)
 				evo_mthd(push, 0x0180 + (or * 0x020), 1);
 				evo_data(push, 0x00000000);
 			}
-
-			evo_mthd(push, 0x0080, 1);
-			evo_data(push, 0x00000000);
 			evo_kick(push, mast);
 		}
 	}
@@ -1542,20 +1565,23 @@ static const struct drm_encoder_funcs nv50_dac_func = {
 static int
 nv50_dac_create(struct drm_connector *connector, struct dcb_output *dcbe)
 {
-	struct drm_device *dev = connector->dev;
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_encoder *nv_encoder;
 	struct drm_encoder *encoder;
+	int type = DRM_MODE_ENCODER_DAC;
 
 	nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL);
 	if (!nv_encoder)
 		return -ENOMEM;
 	nv_encoder->dcb = dcbe;
 	nv_encoder->or = ffs(dcbe->or) - 1;
+	nv_encoder->i2c = i2c->find(i2c, dcbe->i2c_index);
 
 	encoder = to_drm_encoder(nv_encoder);
 	encoder->possible_crtcs = dcbe->heads;
 	encoder->possible_clones = 0;
-	drm_encoder_init(dev, encoder, &nv50_dac_func, DRM_MODE_ENCODER_DAC);
+	drm_encoder_init(connector->dev, encoder, &nv50_dac_func, type);
 	drm_encoder_helper_add(encoder, &nv50_dac_hfunc);
 
 	drm_mode_connector_attach_encoder(connector, encoder);
@@ -1664,9 +1690,6 @@ nv50_sor_dpms(struct drm_encoder *encoder, int mode)
 	}
 
 	nv_call(disp->core, NV50_DISP_SOR_PWR + or, (mode == DRM_MODE_DPMS_ON));
-
-	if (nv_encoder->dcb->type == DCB_OUTPUT_DP)
-		nouveau_dp_dpms(encoder, mode, nv_encoder->dp.datarate, disp->core);
 }
 
 static bool
@@ -1709,9 +1732,6 @@ nv50_sor_disconnect(struct drm_encoder *encoder)
 				evo_mthd(push, 0x0200 + (or * 0x20), 1);
 				evo_data(push, 0x00000000);
 			}
-
-			evo_mthd(push, 0x0080, 1);
-			evo_data(push, 0x00000000);
 			evo_kick(push, mast);
 		}
 
@@ -1723,14 +1743,6 @@ nv50_sor_disconnect(struct drm_encoder *encoder)
 }
 
 static void
-nv50_sor_prepare(struct drm_encoder *encoder)
-{
-	nv50_sor_disconnect(encoder);
-	if (nouveau_encoder(encoder)->dcb->type == DCB_OUTPUT_DP)
-		evo_sync(encoder->dev);
-}
-
-static void
 nv50_sor_commit(struct drm_encoder *encoder)
 {
 }
@@ -1825,8 +1837,13 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 	push = evo_wait(nv50_mast(dev), 8);
 	if (push) {
 		if (nv50_vers(mast) < NVD0_DISP_CLASS) {
+			u32 ctrl = (depth << 16) | (proto << 8) | owner;
+			if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+				ctrl |= 0x00001000;
+			if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+				ctrl |= 0x00002000;
 			evo_mthd(push, 0x0600 + (nv_encoder->or * 0x040), 1);
-			evo_data(push, (depth << 16) | (proto << 8) | owner);
+			evo_data(push, ctrl);
 		} else {
 			u32 magic = 0x31ec6000 | (nv_crtc->index << 25);
 			u32 syncs = 0x00000001;
@@ -1862,7 +1879,7 @@ nv50_sor_destroy(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs nv50_sor_hfunc = {
 	.dpms = nv50_sor_dpms,
 	.mode_fixup = nv50_sor_mode_fixup,
-	.prepare = nv50_sor_prepare,
+	.prepare = nv50_sor_disconnect,
 	.commit = nv50_sor_commit,
 	.mode_set = nv50_sor_mode_set,
 	.disable = nv50_sor_disconnect,
@@ -1876,21 +1893,33 @@ static const struct drm_encoder_funcs nv50_sor_func = {
 static int
 nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe)
 {
-	struct drm_device *dev = connector->dev;
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_encoder *nv_encoder;
 	struct drm_encoder *encoder;
+	int type;
+
+	switch (dcbe->type) {
+	case DCB_OUTPUT_LVDS: type = DRM_MODE_ENCODER_LVDS; break;
+	case DCB_OUTPUT_TMDS:
+	case DCB_OUTPUT_DP:
+	default:
+		type = DRM_MODE_ENCODER_TMDS;
+		break;
+	}
 
 	nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL);
 	if (!nv_encoder)
 		return -ENOMEM;
 	nv_encoder->dcb = dcbe;
 	nv_encoder->or = ffs(dcbe->or) - 1;
+	nv_encoder->i2c = i2c->find(i2c, dcbe->i2c_index);
 	nv_encoder->last_dpms = DRM_MODE_DPMS_OFF;
 
 	encoder = to_drm_encoder(nv_encoder);
 	encoder->possible_crtcs = dcbe->heads;
 	encoder->possible_clones = 0;
-	drm_encoder_init(dev, encoder, &nv50_sor_func, DRM_MODE_ENCODER_TMDS);
+	drm_encoder_init(connector->dev, encoder, &nv50_sor_func, type);
 	drm_encoder_helper_add(encoder, &nv50_sor_hfunc);
 
 	drm_mode_connector_attach_encoder(connector, encoder);
@@ -1898,6 +1927,181 @@ nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe)
 }
 
 /******************************************************************************
+ * PIOR
+ *****************************************************************************/
+
+static void
+nv50_pior_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nv50_disp *disp = nv50_disp(encoder->dev);
+	u32 mthd = (nv_encoder->dcb->type << 12) | nv_encoder->or;
+	u32 ctrl = (mode == DRM_MODE_DPMS_ON);
+	nv_call(disp->core, NV50_DISP_PIOR_PWR + mthd, ctrl);
+}
+
+static bool
+nv50_pior_mode_fixup(struct drm_encoder *encoder,
+		     const struct drm_display_mode *mode,
+		     struct drm_display_mode *adjusted_mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_connector *nv_connector;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	if (nv_connector && nv_connector->native_mode) {
+		if (nv_connector->scaling_mode != DRM_MODE_SCALE_NONE) {
+			int id = adjusted_mode->base.id;
+			*adjusted_mode = *nv_connector->native_mode;
+			adjusted_mode->base.id = id;
+		}
+	}
+
+	adjusted_mode->clock *= 2;
+	return true;
+}
+
+static void
+nv50_pior_commit(struct drm_encoder *encoder)
+{
+}
+
+static void
+nv50_pior_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
+		   struct drm_display_mode *adjusted_mode)
+{
+	struct nv50_mast *mast = nv50_mast(encoder->dev);
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
+	struct nouveau_connector *nv_connector;
+	u8 owner = 1 << nv_crtc->index;
+	u8 proto, depth;
+	u32 *push;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	switch (nv_connector->base.display_info.bpc) {
+	case 10: depth = 0x6; break;
+	case  8: depth = 0x5; break;
+	case  6: depth = 0x2; break;
+	default: depth = 0x0; break;
+	}
+
+	switch (nv_encoder->dcb->type) {
+	case DCB_OUTPUT_TMDS:
+	case DCB_OUTPUT_DP:
+		proto = 0x0;
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
+
+	nv50_pior_dpms(encoder, DRM_MODE_DPMS_ON);
+
+	push = evo_wait(mast, 8);
+	if (push) {
+		if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
+			u32 ctrl = (depth << 16) | (proto << 8) | owner;
+			if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+				ctrl |= 0x00001000;
+			if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+				ctrl |= 0x00002000;
+			evo_mthd(push, 0x0700 + (nv_encoder->or * 0x040), 1);
+			evo_data(push, ctrl);
+		}
+
+		evo_kick(push, mast);
+	}
+
+	nv_encoder->crtc = encoder->crtc;
+}
+
+static void
+nv50_pior_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nv50_mast *mast = nv50_mast(encoder->dev);
+	const int or = nv_encoder->or;
+	u32 *push;
+
+	if (nv_encoder->crtc) {
+		nv50_crtc_prepare(nv_encoder->crtc);
+
+		push = evo_wait(mast, 4);
+		if (push) {
+			if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
+				evo_mthd(push, 0x0700 + (or * 0x040), 1);
+				evo_data(push, 0x00000000);
+			}
+			evo_kick(push, mast);
+		}
+	}
+
+	nv_encoder->crtc = NULL;
+}
+
+static void
+nv50_pior_destroy(struct drm_encoder *encoder)
+{
+	drm_encoder_cleanup(encoder);
+	kfree(encoder);
+}
+
+static const struct drm_encoder_helper_funcs nv50_pior_hfunc = {
+	.dpms = nv50_pior_dpms,
+	.mode_fixup = nv50_pior_mode_fixup,
+	.prepare = nv50_pior_disconnect,
+	.commit = nv50_pior_commit,
+	.mode_set = nv50_pior_mode_set,
+	.disable = nv50_pior_disconnect,
+	.get_crtc = nv50_display_crtc_get,
+};
+
+static const struct drm_encoder_funcs nv50_pior_func = {
+	.destroy = nv50_pior_destroy,
+};
+
+static int
+nv50_pior_create(struct drm_connector *connector, struct dcb_output *dcbe)
+{
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
+	struct nouveau_i2c_port *ddc = NULL;
+	struct nouveau_encoder *nv_encoder;
+	struct drm_encoder *encoder;
+	int type;
+
+	switch (dcbe->type) {
+	case DCB_OUTPUT_TMDS:
+		ddc  = i2c->find_type(i2c, NV_I2C_TYPE_EXTDDC(dcbe->extdev));
+		type = DRM_MODE_ENCODER_TMDS;
+		break;
+	case DCB_OUTPUT_DP:
+		ddc  = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(dcbe->extdev));
+		type = DRM_MODE_ENCODER_TMDS;
+		break;
+	default:
+		return -ENODEV;
+	}
+
+	nv_encoder = kzalloc(sizeof(*nv_encoder), GFP_KERNEL);
+	if (!nv_encoder)
+		return -ENOMEM;
+	nv_encoder->dcb = dcbe;
+	nv_encoder->or = ffs(dcbe->or) - 1;
+	nv_encoder->i2c = ddc;
+
+	encoder = to_drm_encoder(nv_encoder);
+	encoder->possible_crtcs = dcbe->heads;
+	encoder->possible_clones = 0;
+	drm_encoder_init(connector->dev, encoder, &nv50_pior_func, type);
+	drm_encoder_helper_add(encoder, &nv50_pior_hfunc);
+
+	drm_mode_connector_attach_encoder(connector, encoder);
+	return 0;
+}
+
+/******************************************************************************
  * Init
  *****************************************************************************/
 void
@@ -1913,7 +2117,7 @@ nv50_display_init(struct drm_device *dev)
 		evo_mthd(push, 0x0088, 1);
 		evo_data(push, NvEvoSync);
 		evo_kick(push, nv50_mast(dev));
-		return evo_sync(dev);
+		return 0;
 	}
 
 	return -EBUSY;
@@ -2019,25 +2223,28 @@ nv50_display_create(struct drm_device *dev)
 		if (IS_ERR(connector))
 			continue;
 
-		if (dcbe->location != DCB_LOC_ON_CHIP) {
-			NV_WARN(drm, "skipping off-chip encoder %d/%d\n",
-				dcbe->type, ffs(dcbe->or) - 1);
-			continue;
+		if (dcbe->location == DCB_LOC_ON_CHIP) {
+			switch (dcbe->type) {
+			case DCB_OUTPUT_TMDS:
+			case DCB_OUTPUT_LVDS:
+			case DCB_OUTPUT_DP:
+				ret = nv50_sor_create(connector, dcbe);
+				break;
+			case DCB_OUTPUT_ANALOG:
+				ret = nv50_dac_create(connector, dcbe);
+				break;
+			default:
+				ret = -ENODEV;
+				break;
+			}
+		} else {
+			ret = nv50_pior_create(connector, dcbe);
 		}
 
-		switch (dcbe->type) {
-		case DCB_OUTPUT_TMDS:
-		case DCB_OUTPUT_LVDS:
-		case DCB_OUTPUT_DP:
-			nv50_sor_create(connector, dcbe);
-			break;
-		case DCB_OUTPUT_ANALOG:
-			nv50_dac_create(connector, dcbe);
-			break;
-		default:
-			NV_WARN(drm, "skipping unsupported encoder %d/%d\n",
-				dcbe->type, ffs(dcbe->or) - 1);
-			continue;
+		if (ret) {
+			NV_WARN(drm, "failed to create encoder %d/%d/%d: %d\n",
+				     dcbe->location, dcbe->type,
+				     ffs(dcbe->or) - 1, ret);
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nv50_fence.c b/drivers/gpu/drm/nouveau/nv50_fence.c
index d889f3ac0d41..f9701e567db8 100644
--- a/drivers/gpu/drm/nouveau/nv50_fence.c
+++ b/drivers/gpu/drm/nouveau/nv50_fence.c
@@ -27,27 +27,16 @@
 
 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
-#include "nouveau_fence.h"
+#include "nv10_fence.h"
 
 #include "nv50_display.h"
 
-struct nv50_fence_chan {
-	struct nouveau_fence_chan base;
-};
-
-struct nv50_fence_priv {
-	struct nouveau_fence_priv base;
-	struct nouveau_bo *bo;
-	spinlock_t lock;
-	u32 sequence;
-};
-
 static int
 nv50_fence_context_new(struct nouveau_channel *chan)
 {
 	struct drm_device *dev = chan->drm->dev;
-	struct nv50_fence_priv *priv = chan->drm->fence;
-	struct nv50_fence_chan *fctx;
+	struct nv10_fence_priv *priv = chan->drm->fence;
+	struct nv10_fence_chan *fctx;
 	struct ttm_mem_reg *mem = &priv->bo->bo.mem;
 	struct nouveau_object *object;
 	int ret, i;
@@ -57,6 +46,9 @@ nv50_fence_context_new(struct nouveau_channel *chan)
 		return -ENOMEM;
 
 	nouveau_fence_context_new(&fctx->base);
+	fctx->base.emit = nv10_fence_emit;
+	fctx->base.read = nv10_fence_read;
+	fctx->base.sync = nv17_fence_sync;
 
 	ret = nouveau_object_new(nv_object(chan->cli), chan->handle,
 				 NvSema, 0x0002,
@@ -91,7 +83,7 @@ nv50_fence_context_new(struct nouveau_channel *chan)
 int
 nv50_fence_create(struct nouveau_drm *drm)
 {
-	struct nv50_fence_priv *priv;
+	struct nv10_fence_priv *priv;
 	int ret = 0;
 
 	priv = drm->fence = kzalloc(sizeof(*priv), GFP_KERNEL);
@@ -99,11 +91,9 @@ nv50_fence_create(struct nouveau_drm *drm)
 		return -ENOMEM;
 
 	priv->base.dtor = nv10_fence_destroy;
+	priv->base.resume = nv17_fence_resume;
 	priv->base.context_new = nv50_fence_context_new;
 	priv->base.context_del = nv10_fence_context_del;
-	priv->base.emit = nv10_fence_emit;
-	priv->base.read = nv10_fence_read;
-	priv->base.sync = nv17_fence_sync;
 	spin_lock_init(&priv->lock);
 
 	ret = nouveau_bo_new(drm->dev, 4096, 0x1000, TTM_PL_FLAG_VRAM,
@@ -119,13 +109,11 @@ nv50_fence_create(struct nouveau_drm *drm)
 			nouveau_bo_ref(NULL, &priv->bo);
 	}
 
-	if (ret == 0) {
-		nouveau_bo_wr32(priv->bo, 0x000, 0x00000000);
-		priv->base.sync = nv17_fence_sync;
-		priv->base.resume = nv17_fence_resume;
+	if (ret) {
+		nv10_fence_destroy(drm);
+		return ret;
 	}
 
-	if (ret)
-		nv10_fence_destroy(drm);
+	nouveau_bo_wr32(priv->bo, 0x000, 0x00000000);
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c
index c686650584b6..9fd475c89820 100644
--- a/drivers/gpu/drm/nouveau/nv84_fence.c
+++ b/drivers/gpu/drm/nouveau/nv84_fence.c
@@ -23,6 +23,7 @@
  */
 
 #include <core/object.h>
+#include <core/client.h>
 #include <core/class.h>
 
 #include <engine/fifo.h>
@@ -33,79 +34,115 @@
 
 #include "nv50_display.h"
 
-struct nv84_fence_chan {
-	struct nouveau_fence_chan base;
-};
-
-struct nv84_fence_priv {
-	struct nouveau_fence_priv base;
-	struct nouveau_gpuobj *mem;
-};
+u64
+nv84_fence_crtc(struct nouveau_channel *chan, int crtc)
+{
+	struct nv84_fence_chan *fctx = chan->fence;
+	return fctx->dispc_vma[crtc].offset;
+}
 
 static int
-nv84_fence_emit(struct nouveau_fence *fence)
+nv84_fence_emit32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
 {
-	struct nouveau_channel *chan = fence->channel;
-	struct nouveau_fifo_chan *fifo = (void *)chan->object;
-	int ret = RING_SPACE(chan, 7);
+	int ret = RING_SPACE(chan, 8);
 	if (ret == 0) {
 		BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 1);
-		OUT_RING  (chan, NvSema);
-		BEGIN_NV04(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
-		OUT_RING  (chan, upper_32_bits(fifo->chid * 16));
-		OUT_RING  (chan, lower_32_bits(fifo->chid * 16));
-		OUT_RING  (chan, fence->sequence);
+		OUT_RING  (chan, chan->vram);
+		BEGIN_NV04(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 5);
+		OUT_RING  (chan, upper_32_bits(virtual));
+		OUT_RING  (chan, lower_32_bits(virtual));
+		OUT_RING  (chan, sequence);
 		OUT_RING  (chan, NV84_SUBCHAN_SEMAPHORE_TRIGGER_WRITE_LONG);
+		OUT_RING  (chan, 0x00000000);
 		FIRE_RING (chan);
 	}
 	return ret;
 }
 
-
 static int
-nv84_fence_sync(struct nouveau_fence *fence,
-		struct nouveau_channel *prev, struct nouveau_channel *chan)
+nv84_fence_sync32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
 {
-	struct nouveau_fifo_chan *fifo = (void *)prev->object;
 	int ret = RING_SPACE(chan, 7);
 	if (ret == 0) {
 		BEGIN_NV04(chan, 0, NV11_SUBCHAN_DMA_SEMAPHORE, 1);
-		OUT_RING  (chan, NvSema);
+		OUT_RING  (chan, chan->vram);
 		BEGIN_NV04(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
-		OUT_RING  (chan, upper_32_bits(fifo->chid * 16));
-		OUT_RING  (chan, lower_32_bits(fifo->chid * 16));
-		OUT_RING  (chan, fence->sequence);
+		OUT_RING  (chan, upper_32_bits(virtual));
+		OUT_RING  (chan, lower_32_bits(virtual));
+		OUT_RING  (chan, sequence);
 		OUT_RING  (chan, NV84_SUBCHAN_SEMAPHORE_TRIGGER_ACQUIRE_GEQUAL);
 		FIRE_RING (chan);
 	}
 	return ret;
 }
 
+static int
+nv84_fence_emit(struct nouveau_fence *fence)
+{
+	struct nouveau_channel *chan = fence->channel;
+	struct nv84_fence_chan *fctx = chan->fence;
+	struct nouveau_fifo_chan *fifo = (void *)chan->object;
+	u64 addr = fifo->chid * 16;
+
+	if (fence->sysmem)
+		addr += fctx->vma_gart.offset;
+	else
+		addr += fctx->vma.offset;
+
+	return fctx->base.emit32(chan, addr, fence->sequence);
+}
+
+static int
+nv84_fence_sync(struct nouveau_fence *fence,
+		struct nouveau_channel *prev, struct nouveau_channel *chan)
+{
+	struct nv84_fence_chan *fctx = chan->fence;
+	struct nouveau_fifo_chan *fifo = (void *)prev->object;
+	u64 addr = fifo->chid * 16;
+
+	if (fence->sysmem)
+		addr += fctx->vma_gart.offset;
+	else
+		addr += fctx->vma.offset;
+
+	return fctx->base.sync32(chan, addr, fence->sequence);
+}
+
 static u32
 nv84_fence_read(struct nouveau_channel *chan)
 {
 	struct nouveau_fifo_chan *fifo = (void *)chan->object;
 	struct nv84_fence_priv *priv = chan->drm->fence;
-	return nv_ro32(priv->mem, fifo->chid * 16);
+	return nouveau_bo_rd32(priv->bo, fifo->chid * 16/4);
 }
 
 static void
 nv84_fence_context_del(struct nouveau_channel *chan)
 {
+	struct drm_device *dev = chan->drm->dev;
+	struct nv84_fence_priv *priv = chan->drm->fence;
 	struct nv84_fence_chan *fctx = chan->fence;
+	int i;
+
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		struct nouveau_bo *bo = nv50_display_crtc_sema(dev, i);
+		nouveau_bo_vma_del(bo, &fctx->dispc_vma[i]);
+	}
+
+	nouveau_bo_vma_del(priv->bo, &fctx->vma_gart);
+	nouveau_bo_vma_del(priv->bo, &fctx->vma);
 	nouveau_fence_context_del(&fctx->base);
 	chan->fence = NULL;
 	kfree(fctx);
 }
 
-static int
+int
 nv84_fence_context_new(struct nouveau_channel *chan)
 {
-	struct drm_device *dev = chan->drm->dev;
 	struct nouveau_fifo_chan *fifo = (void *)chan->object;
+	struct nouveau_client *client = nouveau_client(fifo);
 	struct nv84_fence_priv *priv = chan->drm->fence;
 	struct nv84_fence_chan *fctx;
-	struct nouveau_object *object;
 	int ret, i;
 
 	fctx = chan->fence = kzalloc(sizeof(*fctx), GFP_KERNEL);
@@ -113,44 +150,74 @@ nv84_fence_context_new(struct nouveau_channel *chan)
 		return -ENOMEM;
 
 	nouveau_fence_context_new(&fctx->base);
+	fctx->base.emit = nv84_fence_emit;
+	fctx->base.sync = nv84_fence_sync;
+	fctx->base.read = nv84_fence_read;
+	fctx->base.emit32 = nv84_fence_emit32;
+	fctx->base.sync32 = nv84_fence_sync32;
 
-	ret = nouveau_object_new(nv_object(chan->cli), chan->handle,
-				 NvSema, 0x0002,
-				 &(struct nv_dma_class) {
-					.flags = NV_DMA_TARGET_VRAM |
-						 NV_DMA_ACCESS_RDWR,
-					.start = priv->mem->addr,
-					.limit = priv->mem->addr +
-						 priv->mem->size - 1,
-				 }, sizeof(struct nv_dma_class),
-				 &object);
-
-	/* dma objects for display sync channel semaphore blocks */
-	for (i = 0; !ret && i < dev->mode_config.num_crtc; i++) {
-		struct nouveau_bo *bo = nv50_display_crtc_sema(dev, i);
+	ret = nouveau_bo_vma_add(priv->bo, client->vm, &fctx->vma);
+	if (ret == 0) {
+		ret = nouveau_bo_vma_add(priv->bo_gart, client->vm,
+					&fctx->vma_gart);
+	}
 
-		ret = nouveau_object_new(nv_object(chan->cli), chan->handle,
-					 NvEvoSema0 + i, 0x003d,
-					 &(struct nv_dma_class) {
-						.flags = NV_DMA_TARGET_VRAM |
-							 NV_DMA_ACCESS_RDWR,
-						.start = bo->bo.offset,
-						.limit = bo->bo.offset + 0xfff,
-					 }, sizeof(struct nv_dma_class),
-					 &object);
+	/* map display semaphore buffers into channel's vm */
+	for (i = 0; !ret && i < chan->drm->dev->mode_config.num_crtc; i++) {
+		struct nouveau_bo *bo = nv50_display_crtc_sema(chan->drm->dev, i);
+		ret = nouveau_bo_vma_add(bo, client->vm, &fctx->dispc_vma[i]);
 	}
 
+	nouveau_bo_wr32(priv->bo, fifo->chid * 16/4, 0x00000000);
+
 	if (ret)
 		nv84_fence_context_del(chan);
-	nv_wo32(priv->mem, fifo->chid * 16, 0x00000000);
 	return ret;
 }
 
+static bool
+nv84_fence_suspend(struct nouveau_drm *drm)
+{
+	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
+	struct nv84_fence_priv *priv = drm->fence;
+	int i;
+
+	priv->suspend = vmalloc((pfifo->max + 1) * sizeof(u32));
+	if (priv->suspend) {
+		for (i = 0; i <= pfifo->max; i++)
+			priv->suspend[i] = nouveau_bo_rd32(priv->bo, i*4);
+	}
+
+	return priv->suspend != NULL;
+}
+
+static void
+nv84_fence_resume(struct nouveau_drm *drm)
+{
+	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
+	struct nv84_fence_priv *priv = drm->fence;
+	int i;
+
+	if (priv->suspend) {
+		for (i = 0; i <= pfifo->max; i++)
+			nouveau_bo_wr32(priv->bo, i*4, priv->suspend[i]);
+		vfree(priv->suspend);
+		priv->suspend = NULL;
+	}
+}
+
 static void
 nv84_fence_destroy(struct nouveau_drm *drm)
 {
 	struct nv84_fence_priv *priv = drm->fence;
-	nouveau_gpuobj_ref(NULL, &priv->mem);
+	nouveau_bo_unmap(priv->bo_gart);
+	if (priv->bo_gart)
+		nouveau_bo_unpin(priv->bo_gart);
+	nouveau_bo_ref(NULL, &priv->bo_gart);
+	nouveau_bo_unmap(priv->bo);
+	if (priv->bo)
+		nouveau_bo_unpin(priv->bo);
+	nouveau_bo_ref(NULL, &priv->bo);
 	drm->fence = NULL;
 	kfree(priv);
 }
@@ -160,7 +227,6 @@ nv84_fence_create(struct nouveau_drm *drm)
 {
 	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
 	struct nv84_fence_priv *priv;
-	u32 chan = pfifo->max + 1;
 	int ret;
 
 	priv = drm->fence = kzalloc(sizeof(*priv), GFP_KERNEL);
@@ -168,14 +234,42 @@ nv84_fence_create(struct nouveau_drm *drm)
 		return -ENOMEM;
 
 	priv->base.dtor = nv84_fence_destroy;
+	priv->base.suspend = nv84_fence_suspend;
+	priv->base.resume = nv84_fence_resume;
 	priv->base.context_new = nv84_fence_context_new;
 	priv->base.context_del = nv84_fence_context_del;
-	priv->base.emit = nv84_fence_emit;
-	priv->base.sync = nv84_fence_sync;
-	priv->base.read = nv84_fence_read;
 
-	ret = nouveau_gpuobj_new(drm->device, NULL, chan * 16, 0x1000, 0,
-				&priv->mem);
+	init_waitqueue_head(&priv->base.waiting);
+	priv->base.uevent = true;
+
+	ret = nouveau_bo_new(drm->dev, 16 * (pfifo->max + 1), 0,
+			     TTM_PL_FLAG_VRAM, 0, 0, NULL, &priv->bo);
+	if (ret == 0) {
+		ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM);
+		if (ret == 0) {
+			ret = nouveau_bo_map(priv->bo);
+			if (ret)
+				nouveau_bo_unpin(priv->bo);
+		}
+		if (ret)
+			nouveau_bo_ref(NULL, &priv->bo);
+	}
+
+	if (ret == 0)
+		ret = nouveau_bo_new(drm->dev, 16 * (pfifo->max + 1), 0,
+				     TTM_PL_FLAG_TT, 0, 0, NULL,
+				     &priv->bo_gart);
+	if (ret == 0) {
+		ret = nouveau_bo_pin(priv->bo_gart, TTM_PL_FLAG_TT);
+		if (ret == 0) {
+			ret = nouveau_bo_map(priv->bo_gart);
+			if (ret)
+				nouveau_bo_unpin(priv->bo_gart);
+		}
+		if (ret)
+			nouveau_bo_ref(NULL, &priv->bo_gart);
+	}
+
 	if (ret)
 		nv84_fence_destroy(drm);
 	return ret;
diff --git a/drivers/gpu/drm/nouveau/nvc0_fence.c b/drivers/gpu/drm/nouveau/nvc0_fence.c
index 2a56b1b551cb..9566267fbc42 100644
--- a/drivers/gpu/drm/nouveau/nvc0_fence.c
+++ b/drivers/gpu/drm/nouveau/nvc0_fence.c
@@ -34,203 +34,57 @@
 
 #include "nv50_display.h"
 
-struct nvc0_fence_priv {
-	struct nouveau_fence_priv base;
-	struct nouveau_bo *bo;
-	u32 *suspend;
-};
-
-struct nvc0_fence_chan {
-	struct nouveau_fence_chan base;
-	struct nouveau_vma vma;
-	struct nouveau_vma dispc_vma[4];
-};
-
-u64
-nvc0_fence_crtc(struct nouveau_channel *chan, int crtc)
-{
-	struct nvc0_fence_chan *fctx = chan->fence;
-	return fctx->dispc_vma[crtc].offset;
-}
-
 static int
-nvc0_fence_emit(struct nouveau_fence *fence)
+nvc0_fence_emit32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
 {
-	struct nouveau_channel *chan = fence->channel;
-	struct nvc0_fence_chan *fctx = chan->fence;
-	struct nouveau_fifo_chan *fifo = (void *)chan->object;
-	u64 addr = fctx->vma.offset + fifo->chid * 16;
-	int ret;
-
-	ret = RING_SPACE(chan, 5);
+	int ret = RING_SPACE(chan, 6);
 	if (ret == 0) {
-		BEGIN_NVC0(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
-		OUT_RING  (chan, upper_32_bits(addr));
-		OUT_RING  (chan, lower_32_bits(addr));
-		OUT_RING  (chan, fence->sequence);
+		BEGIN_NVC0(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 5);
+		OUT_RING  (chan, upper_32_bits(virtual));
+		OUT_RING  (chan, lower_32_bits(virtual));
+		OUT_RING  (chan, sequence);
 		OUT_RING  (chan, NV84_SUBCHAN_SEMAPHORE_TRIGGER_WRITE_LONG);
+		OUT_RING  (chan, 0x00000000);
 		FIRE_RING (chan);
 	}
-
 	return ret;
 }
 
 static int
-nvc0_fence_sync(struct nouveau_fence *fence,
-		struct nouveau_channel *prev, struct nouveau_channel *chan)
+nvc0_fence_sync32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
 {
-	struct nvc0_fence_chan *fctx = chan->fence;
-	struct nouveau_fifo_chan *fifo = (void *)prev->object;
-	u64 addr = fctx->vma.offset + fifo->chid * 16;
-	int ret;
-
-	ret = RING_SPACE(chan, 5);
+	int ret = RING_SPACE(chan, 5);
 	if (ret == 0) {
 		BEGIN_NVC0(chan, 0, NV84_SUBCHAN_SEMAPHORE_ADDRESS_HIGH, 4);
-		OUT_RING  (chan, upper_32_bits(addr));
-		OUT_RING  (chan, lower_32_bits(addr));
-		OUT_RING  (chan, fence->sequence);
+		OUT_RING  (chan, upper_32_bits(virtual));
+		OUT_RING  (chan, lower_32_bits(virtual));
+		OUT_RING  (chan, sequence);
 		OUT_RING  (chan, NV84_SUBCHAN_SEMAPHORE_TRIGGER_ACQUIRE_GEQUAL |
 				 NVC0_SUBCHAN_SEMAPHORE_TRIGGER_YIELD);
 		FIRE_RING (chan);
 	}
-
 	return ret;
 }
 
-static u32
-nvc0_fence_read(struct nouveau_channel *chan)
-{
-	struct nouveau_fifo_chan *fifo = (void *)chan->object;
-	struct nvc0_fence_priv *priv = chan->drm->fence;
-	return nouveau_bo_rd32(priv->bo, fifo->chid * 16/4);
-}
-
-static void
-nvc0_fence_context_del(struct nouveau_channel *chan)
-{
-	struct drm_device *dev = chan->drm->dev;
-	struct nvc0_fence_priv *priv = chan->drm->fence;
-	struct nvc0_fence_chan *fctx = chan->fence;
-	int i;
-
-	for (i = 0; i < dev->mode_config.num_crtc; i++) {
-		struct nouveau_bo *bo = nv50_display_crtc_sema(dev, i);
-		nouveau_bo_vma_del(bo, &fctx->dispc_vma[i]);
-	}
-
-	nouveau_bo_vma_del(priv->bo, &fctx->vma);
-	nouveau_fence_context_del(&fctx->base);
-	chan->fence = NULL;
-	kfree(fctx);
-}
-
 static int
 nvc0_fence_context_new(struct nouveau_channel *chan)
 {
-	struct nouveau_fifo_chan *fifo = (void *)chan->object;
-	struct nouveau_client *client = nouveau_client(fifo);
-	struct nvc0_fence_priv *priv = chan->drm->fence;
-	struct nvc0_fence_chan *fctx;
-	int ret, i;
-
-	fctx = chan->fence = kzalloc(sizeof(*fctx), GFP_KERNEL);
-	if (!fctx)
-		return -ENOMEM;
-
-	nouveau_fence_context_new(&fctx->base);
-
-	ret = nouveau_bo_vma_add(priv->bo, client->vm, &fctx->vma);
-	if (ret)
-		nvc0_fence_context_del(chan);
-
-	/* map display semaphore buffers into channel's vm */
-	for (i = 0; !ret && i < chan->drm->dev->mode_config.num_crtc; i++) {
-		struct nouveau_bo *bo = nv50_display_crtc_sema(chan->drm->dev, i);
-		ret = nouveau_bo_vma_add(bo, client->vm, &fctx->dispc_vma[i]);
+	int ret = nv84_fence_context_new(chan);
+	if (ret == 0) {
+		struct nv84_fence_chan *fctx = chan->fence;
+		fctx->base.emit32 = nvc0_fence_emit32;
+		fctx->base.sync32 = nvc0_fence_sync32;
 	}
-
-	nouveau_bo_wr32(priv->bo, fifo->chid * 16/4, 0x00000000);
 	return ret;
 }
 
-static bool
-nvc0_fence_suspend(struct nouveau_drm *drm)
-{
-	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
-	struct nvc0_fence_priv *priv = drm->fence;
-	int i;
-
-	priv->suspend = vmalloc((pfifo->max + 1) * sizeof(u32));
-	if (priv->suspend) {
-		for (i = 0; i <= pfifo->max; i++)
-			priv->suspend[i] = nouveau_bo_rd32(priv->bo, i);
-	}
-
-	return priv->suspend != NULL;
-}
-
-static void
-nvc0_fence_resume(struct nouveau_drm *drm)
-{
-	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
-	struct nvc0_fence_priv *priv = drm->fence;
-	int i;
-
-	if (priv->suspend) {
-		for (i = 0; i <= pfifo->max; i++)
-			nouveau_bo_wr32(priv->bo, i, priv->suspend[i]);
-		vfree(priv->suspend);
-		priv->suspend = NULL;
-	}
-}
-
-static void
-nvc0_fence_destroy(struct nouveau_drm *drm)
-{
-	struct nvc0_fence_priv *priv = drm->fence;
-	nouveau_bo_unmap(priv->bo);
-	if (priv->bo)
-		nouveau_bo_unpin(priv->bo);
-	nouveau_bo_ref(NULL, &priv->bo);
-	drm->fence = NULL;
-	kfree(priv);
-}
-
 int
 nvc0_fence_create(struct nouveau_drm *drm)
 {
-	struct nouveau_fifo *pfifo = nouveau_fifo(drm->device);
-	struct nvc0_fence_priv *priv;
-	int ret;
-
-	priv = drm->fence = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
-
-	priv->base.dtor = nvc0_fence_destroy;
-	priv->base.suspend = nvc0_fence_suspend;
-	priv->base.resume = nvc0_fence_resume;
-	priv->base.context_new = nvc0_fence_context_new;
-	priv->base.context_del = nvc0_fence_context_del;
-	priv->base.emit = nvc0_fence_emit;
-	priv->base.sync = nvc0_fence_sync;
-	priv->base.read = nvc0_fence_read;
-
-	ret = nouveau_bo_new(drm->dev, 16 * (pfifo->max + 1), 0,
-			     TTM_PL_FLAG_VRAM, 0, 0, NULL, &priv->bo);
+	int ret = nv84_fence_create(drm);
 	if (ret == 0) {
-		ret = nouveau_bo_pin(priv->bo, TTM_PL_FLAG_VRAM);
-		if (ret == 0) {
-			ret = nouveau_bo_map(priv->bo);
-			if (ret)
-				nouveau_bo_unpin(priv->bo);
-		}
-		if (ret)
-			nouveau_bo_ref(NULL, &priv->bo);
+		struct nv84_fence_priv *priv = drm->fence;
+		priv->base.context_new = nvc0_fence_context_new;
 	}
-
-	if (ret)
-		nvc0_fence_destroy(drm);
 	return ret;
 }
diff --git a/drivers/staging/omapdrm/Kconfig b/drivers/gpu/drm/omapdrm/Kconfig
index 09f65dc3d2c8..09f65dc3d2c8 100644
--- a/drivers/staging/omapdrm/Kconfig
+++ b/drivers/gpu/drm/omapdrm/Kconfig
diff --git a/drivers/staging/omapdrm/Makefile b/drivers/gpu/drm/omapdrm/Makefile
index d85e058f2845..d85e058f2845 100644
--- a/drivers/staging/omapdrm/Makefile
+++ b/drivers/gpu/drm/omapdrm/Makefile
diff --git a/drivers/gpu/drm/omapdrm/TODO b/drivers/gpu/drm/omapdrm/TODO
new file mode 100644
index 000000000000..4d8c18aa5dd7
--- /dev/null
+++ b/drivers/gpu/drm/omapdrm/TODO
@@ -0,0 +1,23 @@
+TODO
+. Where should we do eviction (detatch_pages())?  We aren't necessarily
+  accessing the pages via a GART, so maybe we need some other threshold
+  to put a cap on the # of pages that can be pin'd.
+  . Use mm_shrinker to trigger unpinning pages.
+  . This is mainly theoretical since most of these devices don't actually
+    have swap or harddrive.
+. GEM/shmem backed pages can have existing mappings (kernel linear map,
+  etc..), which isn't really ideal.
+. Revisit GEM sync object infrastructure.. TTM has some framework for this
+  already.  Possibly this could be refactored out and made more common?
+  There should be some way to do this with less wheel-reinvention.
+  . This can be handled by the dma-buf fence/reservation stuff when it
+    lands
+
+Userspace:
+. git://anongit.freedesktop.org/xorg/driver/xf86-video-omap
+
+Currently tested on
+. OMAP3530 beagleboard
+. OMAP4430 pandaboard
+. OMAP4460 pandaboard
+. OMAP5432 uEVM
diff --git a/drivers/staging/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c
index 8979c80adb5f..c451c41a7a7d 100644
--- a/drivers/staging/omapdrm/omap_connector.c
+++ b/drivers/gpu/drm/omapdrm/omap_connector.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_connector.c
+ * drivers/gpu/drm/omapdrm/omap_connector.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
diff --git a/drivers/staging/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 32109c09357c..bec66a490b8f 100644
--- a/drivers/staging/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_crtc.c
+ * drivers/gpu/drm/omapdrm/omap_crtc.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
@@ -274,17 +274,16 @@ static void page_flip_worker(struct work_struct *work)
 	struct omap_crtc *omap_crtc =
 			container_of(work, struct omap_crtc, page_flip_work);
 	struct drm_crtc *crtc = &omap_crtc->base;
-	struct drm_device *dev = crtc->dev;
 	struct drm_display_mode *mode = &crtc->mode;
 	struct drm_gem_object *bo;
 
-	mutex_lock(&dev->mode_config.mutex);
+	mutex_lock(&crtc->mutex);
 	omap_plane_mode_set(omap_crtc->plane, crtc, crtc->fb,
 			0, 0, mode->hdisplay, mode->vdisplay,
 			crtc->x << 16, crtc->y << 16,
 			mode->hdisplay << 16, mode->vdisplay << 16,
 			vblank_cb, crtc);
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&crtc->mutex);
 
 	bo = omap_framebuffer_bo(crtc->fb, 0);
 	drm_gem_object_unreference_unlocked(bo);
@@ -417,7 +416,7 @@ static void apply_worker(struct work_struct *work)
 	 * the callbacks and list modification all serialized
 	 * with respect to modesetting ioctls from userspace.
 	 */
-	mutex_lock(&dev->mode_config.mutex);
+	mutex_lock(&crtc->mutex);
 	dispc_runtime_get();
 
 	/*
@@ -462,16 +461,15 @@ static void apply_worker(struct work_struct *work)
 
 out:
 	dispc_runtime_put();
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&crtc->mutex);
 }
 
 int omap_crtc_apply(struct drm_crtc *crtc,
 		struct omap_drm_apply *apply)
 {
 	struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
 
-	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	WARN_ON(!mutex_is_locked(&crtc->mutex));
 
 	/* no need to queue it again if it is already queued: */
 	if (apply->queued)
diff --git a/drivers/staging/omapdrm/omap_debugfs.c b/drivers/gpu/drm/omapdrm/omap_debugfs.c
index 2f122e00b51d..c27f59da7f29 100644
--- a/drivers/staging/omapdrm/omap_debugfs.c
+++ b/drivers/gpu/drm/omapdrm/omap_debugfs.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_debugfs.c
+ * drivers/gpu/drm/omapdrm/omap_debugfs.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
@@ -57,21 +57,11 @@ static int fb_show(struct seq_file *m, void *arg)
 	struct drm_device *dev = node->minor->dev;
 	struct omap_drm_private *priv = dev->dev_private;
 	struct drm_framebuffer *fb;
-	int ret;
-
-	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
-	if (ret)
-		return ret;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret) {
-		mutex_unlock(&dev->mode_config.mutex);
-		return ret;
-	}
 
 	seq_printf(m, "fbcon ");
 	omap_framebuffer_describe(priv->fbdev->fb, m);
 
+	mutex_lock(&dev->mode_config.fb_lock);
 	list_for_each_entry(fb, &dev->mode_config.fb_list, head) {
 		if (fb == priv->fbdev->fb)
 			continue;
@@ -79,9 +69,7 @@ static int fb_show(struct seq_file *m, void *arg)
 		seq_printf(m, "user ");
 		omap_framebuffer_describe(fb, m);
 	}
-
-	mutex_unlock(&dev->struct_mutex);
-	mutex_unlock(&dev->mode_config.mutex);
+	mutex_unlock(&dev->mode_config.fb_lock);
 
 	return 0;
 }
diff --git a/drivers/staging/omapdrm/omap_dmm_priv.h b/drivers/gpu/drm/omapdrm/omap_dmm_priv.h
index 58bcd6ae0255..58bcd6ae0255 100644
--- a/drivers/staging/omapdrm/omap_dmm_priv.h
+++ b/drivers/gpu/drm/omapdrm/omap_dmm_priv.h
diff --git a/drivers/staging/omapdrm/omap_dmm_tiler.c b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
index 9b794c933c81..9b794c933c81 100644
--- a/drivers/staging/omapdrm/omap_dmm_tiler.c
+++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
diff --git a/drivers/staging/omapdrm/omap_dmm_tiler.h b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.h
index 4fdd61e54bd2..4fdd61e54bd2 100644
--- a/drivers/staging/omapdrm/omap_dmm_tiler.h
+++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.h
diff --git a/drivers/staging/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 480dc343446c..079c54c6f94c 100644
--- a/drivers/staging/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_drv.c
+ * drivers/gpu/drm/omapdrm/omap_drv.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
@@ -452,9 +452,9 @@ static void dev_lastclose(struct drm_device *dev)
 		}
 	}
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	ret = drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 	if (ret)
 		DBG("failed to restore crtc mode");
 }
diff --git a/drivers/staging/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index f921027e7500..d4f997bb4ac0 100644
--- a/drivers/staging/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_drv.h
+ * drivers/gpu/drm/omapdrm/omap_drv.h
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
@@ -25,8 +25,8 @@
 #include <linux/types.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/omap_drm.h>
 #include <linux/platform_data/omap_drm.h>
-#include "omap_drm.h"
 
 
 #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
diff --git a/drivers/staging/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c
index 25fc0c7b4f6d..21d126d0317e 100644
--- a/drivers/staging/omapdrm/omap_encoder.c
+++ b/drivers/gpu/drm/omapdrm/omap_encoder.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_encoder.c
+ * drivers/gpu/drm/omapdrm/omap_encoder.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
diff --git a/drivers/staging/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index bb4969942148..8031402e7951 100644
--- a/drivers/staging/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_fb.c
+ * drivers/gpu/drm/omapdrm/omap_fb.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
@@ -423,14 +423,6 @@ struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev,
 	}
 
 	fb = &omap_fb->base;
-	ret = drm_framebuffer_init(dev, fb, &omap_framebuffer_funcs);
-	if (ret) {
-		dev_err(dev->dev, "framebuffer init failed: %d\n", ret);
-		goto fail;
-	}
-
-	DBG("create: FB ID: %d (%p)", fb->base.id, fb);
-
 	omap_fb->format = format;
 
 	for (i = 0; i < n; i++) {
@@ -461,6 +453,14 @@ struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev,
 
 	drm_helper_mode_fill_fb_struct(fb, mode_cmd);
 
+	ret = drm_framebuffer_init(dev, fb, &omap_framebuffer_funcs);
+	if (ret) {
+		dev_err(dev->dev, "framebuffer init failed: %d\n", ret);
+		goto fail;
+	}
+
+	DBG("create: FB ID: %d (%p)", fb->base.id, fb);
+
 	return fb;
 
 fail:
diff --git a/drivers/staging/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c
index 70f2d6ed2ed3..b11ce609fcc2 100644
--- a/drivers/staging/omapdrm/omap_fbdev.c
+++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_fbdev.c
+ * drivers/gpu/drm/omapdrm/omap_fbdev.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob@ti.com>
@@ -131,9 +131,6 @@ static struct fb_ops omap_fb_ops = {
 	.fb_pan_display = omap_fbdev_pan_display,
 	.fb_blank = drm_fb_helper_blank,
 	.fb_setcmap = drm_fb_helper_setcmap,
-
-	.fb_debug_enter = drm_fb_helper_debug_enter,
-	.fb_debug_leave = drm_fb_helper_debug_leave,
 };
 
 static int omap_fbdev_create(struct drm_fb_helper *helper,
@@ -275,8 +272,10 @@ fail:
 	if (ret) {
 		if (fbi)
 			framebuffer_release(fbi);
-		if (fb)
+		if (fb) {
+			drm_framebuffer_unregister_private(fb);
 			drm_framebuffer_remove(fb);
+		}
 	}
 
 	return ret;
@@ -294,25 +293,10 @@ static void omap_crtc_fb_gamma_get(struct drm_crtc *crtc,
 	DBG("fbdev: get gamma");
 }
 
-static int omap_fbdev_probe(struct drm_fb_helper *helper,
-		struct drm_fb_helper_surface_size *sizes)
-{
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = omap_fbdev_create(helper, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static struct drm_fb_helper_funcs omap_fb_helper_funcs = {
 	.gamma_set = omap_crtc_fb_gamma_set,
 	.gamma_get = omap_crtc_fb_gamma_get,
-	.fb_probe = omap_fbdev_probe,
+	.fb_probe = omap_fbdev_create,
 };
 
 static struct drm_fb_helper *get_fb(struct fb_info *fbi)
@@ -365,6 +349,10 @@ struct drm_fb_helper *omap_fbdev_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	drm_fb_helper_initial_config(helper, 32);
 
 	priv->fbdev = helper;
@@ -398,8 +386,10 @@ void omap_fbdev_free(struct drm_device *dev)
 	fbdev = to_omap_fbdev(priv->fbdev);
 
 	/* this will free the backing object */
-	if (fbdev->fb)
+	if (fbdev->fb) {
+		drm_framebuffer_unregister_private(fbdev->fb);
 		drm_framebuffer_remove(fbdev->fb);
+	}
 
 	kfree(fbdev);
 
diff --git a/drivers/staging/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c
index 518d03d4d4f3..ebbdf4132e9c 100644
--- a/drivers/staging/omapdrm/omap_gem.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_gem.c
+ * drivers/gpu/drm/omapdrm/omap_gem.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
diff --git a/drivers/staging/omapdrm/omap_gem_dmabuf.c b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
index a3236abfca3d..ac74d1bc67bf 100644
--- a/drivers/staging/omapdrm/omap_gem_dmabuf.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_gem_dmabuf.c
+ * drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
diff --git a/drivers/staging/omapdrm/omap_gem_helpers.c b/drivers/gpu/drm/omapdrm/omap_gem_helpers.c
index ffb8cceaeb46..e4a66a35fc6a 100644
--- a/drivers/staging/omapdrm/omap_gem_helpers.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem_helpers.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_gem_helpers.c
+ * drivers/gpu/drm/omapdrm/omap_gem_helpers.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
diff --git a/drivers/staging/omapdrm/omap_irq.c b/drivers/gpu/drm/omapdrm/omap_irq.c
index 2629ba7be6c8..e01303ee00c3 100644
--- a/drivers/staging/omapdrm/omap_irq.c
+++ b/drivers/gpu/drm/omapdrm/omap_irq.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_irq.c
+ * drivers/gpu/drm/omapdrm/omap_irq.c
  *
  * Copyright (C) 2012 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
diff --git a/drivers/staging/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c
index c063476db3bb..2882cda6ea19 100644
--- a/drivers/staging/omapdrm/omap_plane.c
+++ b/drivers/gpu/drm/omapdrm/omap_plane.c
@@ -1,5 +1,5 @@
 /*
- * drivers/staging/omapdrm/omap_plane.c
+ * drivers/gpu/drm/omapdrm/omap_plane.c
  *
  * Copyright (C) 2011 Texas Instruments
  * Author: Rob Clark <rob.clark@linaro.org>
diff --git a/drivers/staging/omapdrm/tcm-sita.c b/drivers/gpu/drm/omapdrm/tcm-sita.c
index efb609510540..efb609510540 100644
--- a/drivers/staging/omapdrm/tcm-sita.c
+++ b/drivers/gpu/drm/omapdrm/tcm-sita.c
diff --git a/drivers/staging/omapdrm/tcm-sita.h b/drivers/gpu/drm/omapdrm/tcm-sita.h
index 0444f868671c..0444f868671c 100644
--- a/drivers/staging/omapdrm/tcm-sita.h
+++ b/drivers/gpu/drm/omapdrm/tcm-sita.h
diff --git a/drivers/staging/omapdrm/tcm.h b/drivers/gpu/drm/omapdrm/tcm.h
index a8d5ce47686f..a8d5ce47686f 100644
--- a/drivers/staging/omapdrm/tcm.h
+++ b/drivers/gpu/drm/omapdrm/tcm.h
diff --git a/drivers/gpu/drm/radeon/Kconfig b/drivers/gpu/drm/radeon/Kconfig
index ea92bbe3ed37..970f8e92dbb7 100644
--- a/drivers/gpu/drm/radeon/Kconfig
+++ b/drivers/gpu/drm/radeon/Kconfig
@@ -1,31 +1,8 @@
-config DRM_RADEON_KMS
-	bool "Enable modesetting on radeon by default - NEW DRIVER"
+config DRM_RADEON_UMS
+	bool "Enable userspace modesetting on radeon (DEPRECATED)"
 	depends on DRM_RADEON
-	select BACKLIGHT_CLASS_DEVICE
 	help
-	  Choose this option if you want kernel modesetting enabled by default.
+	  Choose this option if you still need userspace modesetting.
 
-	  This is a completely new driver. It's only part of the existing drm
-	  for compatibility reasons. It requires an entirely different graphics
-	  stack above it and works very differently from the old drm stack.
-	  i.e. don't enable this unless you know what you are doing it may
-	  cause issues or bugs compared to the previous userspace driver stack.
-
-	  When kernel modesetting is enabled the IOCTL of radeon/drm
-	  driver are considered as invalid and an error message is printed
-	  in the log and they return failure.
-
-	  KMS enabled userspace will use new API to talk with the radeon/drm
-	  driver. The new API provide functions to create/destroy/share/mmap
-	  buffer object which are then managed by the kernel memory manager
-	  (here TTM). In order to submit command to the GPU the userspace
-	  provide a buffer holding the command stream, along this buffer
-	  userspace have to provide a list of buffer object used by the
-	  command stream. The kernel radeon driver will then place buffer
-	  in GPU accessible memory and will update command stream to reflect
-	  the position of the different buffers.
-
-	  The kernel will also perform security check on command stream
-	  provided by the user, we want to catch and forbid any illegal use
-	  of the GPU such as DMA into random system memory or into memory
-	  not owned by the process supplying the command stream.
+	  Userspace modesetting is deprecated for quite some time now, so
+	  enable this only if you have ancient versions of the DDX drivers.
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index a6598fd66423..bf172522ea68 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -56,8 +56,12 @@ $(obj)/r600_cs.o: $(obj)/r600_reg_safe.h
 
 $(obj)/evergreen_cs.o: $(obj)/evergreen_reg_safe.h $(obj)/cayman_reg_safe.h
 
-radeon-y := radeon_drv.o radeon_cp.o radeon_state.o radeon_mem.o \
-	radeon_irq.o r300_cmdbuf.o r600_cp.o
+radeon-y := radeon_drv.o
+
+# add UMS driver
+radeon-$(CONFIG_DRM_RADEON_UMS)+= radeon_cp.o radeon_state.o radeon_mem.o \
+	radeon_irq.o r300_cmdbuf.o r600_cp.o r600_blit.o
+
 # add KMS driver
 radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
 	radeon_atombios.o radeon_agp.o atombios_crtc.o radeon_combios.o \
@@ -67,7 +71,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
 	radeon_clocks.o radeon_fb.o radeon_gem.o radeon_ring.o radeon_irq_kms.o \
 	radeon_cs.o radeon_bios.o radeon_benchmark.o r100.o r300.o r420.o \
 	rs400.o rs600.o rs690.o rv515.o r520.o r600.o rv770.o radeon_test.o \
-	r200.o radeon_legacy_tv.o r600_cs.o r600_blit.o r600_blit_shaders.o \
+	r200.o radeon_legacy_tv.o r600_cs.o r600_blit_shaders.o \
 	r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
 	evergreen.o evergreen_cs.o evergreen_blit_shaders.o evergreen_blit_kms.o \
 	evergreen_hdmi.o radeon_trace_points.o ni.o cayman_blit_shaders.o \
diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c
index 5ce9bf51a8de..46a9c3772850 100644
--- a/drivers/gpu/drm/radeon/atom.c
+++ b/drivers/gpu/drm/radeon/atom.c
@@ -1238,6 +1238,8 @@ static int atom_iio_len[] = { 1, 2, 3, 3, 3, 3, 4, 4, 4, 3 };
 static void atom_index_iio(struct atom_context *ctx, int base)
 {
 	ctx->iio = kzalloc(2 * 256, GFP_KERNEL);
+	if (!ctx->iio)
+		return;
 	while (CU8(base) == ATOM_IIO_START) {
 		ctx->iio[CU8(base + 1)] = base + 2;
 		base += 2;
@@ -1287,6 +1289,10 @@ struct atom_context *atom_parse(struct card_info *card, void *bios)
 	ctx->cmd_table = CU16(base + ATOM_ROM_CMD_PTR);
 	ctx->data_table = CU16(base + ATOM_ROM_DATA_PTR);
 	atom_index_iio(ctx, CU16(ctx->data_table + ATOM_DATA_IIO_PTR) + 4);
+	if (!ctx->iio) {
+		atom_destroy(ctx);
+		return NULL;
+	}
 
 	str = CSTR(CU16(base + ATOM_ROM_MSG_PTR));
 	while (*str && ((*str == '\n') || (*str == '\r')))
@@ -1335,8 +1341,7 @@ int atom_asic_init(struct atom_context *ctx)
 
 void atom_destroy(struct atom_context *ctx)
 {
-	if (ctx->iio)
-		kfree(ctx->iio);
+	kfree(ctx->iio);
 	kfree(ctx);
 }
 
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 9175615bbd8a..21a892c6ab9c 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -252,8 +252,6 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode)
 		radeon_crtc->enabled = true;
 		/* adjust pm to dpms changes BEFORE enabling crtcs */
 		radeon_pm_compute_clocks(rdev);
-		if (ASIC_IS_DCE6(rdev) && !radeon_crtc->in_mode_set)
-			atombios_powergate_crtc(crtc, ATOM_DISABLE);
 		atombios_enable_crtc(crtc, ATOM_ENABLE);
 		if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev))
 			atombios_enable_crtc_memreq(crtc, ATOM_ENABLE);
@@ -271,8 +269,6 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode)
 			atombios_enable_crtc_memreq(crtc, ATOM_DISABLE);
 		atombios_enable_crtc(crtc, ATOM_DISABLE);
 		radeon_crtc->enabled = false;
-		if (ASIC_IS_DCE6(rdev) && !radeon_crtc->in_mode_set)
-			atombios_powergate_crtc(crtc, ATOM_ENABLE);
 		/* adjust pm to dpms changes AFTER disabling crtcs */
 		radeon_pm_compute_clocks(rdev);
 		break;
@@ -1844,6 +1840,8 @@ static void atombios_crtc_disable(struct drm_crtc *crtc)
 	int i;
 
 	atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+	if (ASIC_IS_DCE6(rdev))
+		atombios_powergate_crtc(crtc, ATOM_ENABLE);
 
 	for (i = 0; i < rdev->num_crtc; i++) {
 		if (rdev->mode_info.crtcs[i] &&
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index a2d478e8692a..3c38ea46531c 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -403,6 +403,19 @@ void evergreen_pm_misc(struct radeon_device *rdev)
 			rdev->pm.current_vddc = voltage->voltage;
 			DRM_DEBUG("Setting: vddc: %d\n", voltage->voltage);
 		}
+
+		/* starting with BTC, there is one state that is used for both
+		 * MH and SH.  Difference is that we always use the high clock index for
+		 * mclk and vddci.
+		 */
+		if ((rdev->pm.pm_method == PM_METHOD_PROFILE) &&
+		    (rdev->family >= CHIP_BARTS) &&
+		    rdev->pm.active_crtc_count &&
+		    ((rdev->pm.profile_index == PM_PROFILE_MID_MH_IDX) ||
+		     (rdev->pm.profile_index == PM_PROFILE_LOW_MH_IDX)))
+			voltage = &rdev->pm.power_state[req_ps_idx].
+				clock_info[rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx].voltage;
+
 		/* 0xff01 is a flag rather then an actual voltage */
 		if (voltage->vddci == 0xff01)
 			return;
@@ -2308,32 +2321,8 @@ int evergreen_mc_init(struct radeon_device *rdev)
 	return 0;
 }
 
-bool evergreen_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+void evergreen_print_gpu_status_regs(struct radeon_device *rdev)
 {
-	u32 srbm_status;
-	u32 grbm_status;
-	u32 grbm_status_se0, grbm_status_se1;
-
-	srbm_status = RREG32(SRBM_STATUS);
-	grbm_status = RREG32(GRBM_STATUS);
-	grbm_status_se0 = RREG32(GRBM_STATUS_SE0);
-	grbm_status_se1 = RREG32(GRBM_STATUS_SE1);
-	if (!(grbm_status & GUI_ACTIVE)) {
-		radeon_ring_lockup_update(ring);
-		return false;
-	}
-	/* force CP activities */
-	radeon_ring_force_activity(rdev, ring);
-	return radeon_ring_test_lockup(rdev, ring);
-}
-
-static void evergreen_gpu_soft_reset_gfx(struct radeon_device *rdev)
-{
-	u32 grbm_reset = 0;
-
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		return;
-
 	dev_info(rdev->dev, "  GRBM_STATUS               = 0x%08X\n",
 		RREG32(GRBM_STATUS));
 	dev_info(rdev->dev, "  GRBM_STATUS_SE0           = 0x%08X\n",
@@ -2342,6 +2331,8 @@ static void evergreen_gpu_soft_reset_gfx(struct radeon_device *rdev)
 		RREG32(GRBM_STATUS_SE1));
 	dev_info(rdev->dev, "  SRBM_STATUS               = 0x%08X\n",
 		RREG32(SRBM_STATUS));
+	dev_info(rdev->dev, "  SRBM_STATUS2              = 0x%08X\n",
+		RREG32(SRBM_STATUS2));
 	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
 		RREG32(CP_STALLED_STAT1));
 	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
@@ -2350,112 +2341,283 @@ static void evergreen_gpu_soft_reset_gfx(struct radeon_device *rdev)
 		RREG32(CP_BUSY_STAT));
 	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
 		RREG32(CP_STAT));
+	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
+		RREG32(DMA_STATUS_REG));
+	if (rdev->family >= CHIP_CAYMAN) {
+		dev_info(rdev->dev, "  R_00D834_DMA_STATUS_REG   = 0x%08X\n",
+			 RREG32(DMA_STATUS_REG + 0x800));
+	}
+}
 
-	/* Disable CP parsing/prefetching */
-	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT);
+bool evergreen_is_display_hung(struct radeon_device *rdev)
+{
+	u32 crtc_hung = 0;
+	u32 crtc_status[6];
+	u32 i, j, tmp;
 
-	/* reset all the gfx blocks */
-	grbm_reset = (SOFT_RESET_CP |
-		      SOFT_RESET_CB |
-		      SOFT_RESET_DB |
-		      SOFT_RESET_PA |
-		      SOFT_RESET_SC |
-		      SOFT_RESET_SPI |
-		      SOFT_RESET_SH |
-		      SOFT_RESET_SX |
-		      SOFT_RESET_TC |
-		      SOFT_RESET_TA |
-		      SOFT_RESET_VC |
-		      SOFT_RESET_VGT);
-
-	dev_info(rdev->dev, "  GRBM_SOFT_RESET=0x%08X\n", grbm_reset);
-	WREG32(GRBM_SOFT_RESET, grbm_reset);
-	(void)RREG32(GRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(GRBM_SOFT_RESET, 0);
-	(void)RREG32(GRBM_SOFT_RESET);
+	for (i = 0; i < rdev->num_crtc; i++) {
+		if (RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]) & EVERGREEN_CRTC_MASTER_EN) {
+			crtc_status[i] = RREG32(EVERGREEN_CRTC_STATUS_HV_COUNT + crtc_offsets[i]);
+			crtc_hung |= (1 << i);
+		}
+	}
 
-	dev_info(rdev->dev, "  GRBM_STATUS               = 0x%08X\n",
-		RREG32(GRBM_STATUS));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE0           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE0));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE1           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE1));
-	dev_info(rdev->dev, "  SRBM_STATUS               = 0x%08X\n",
-		RREG32(SRBM_STATUS));
-	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT1));
-	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT2));
-	dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
-		RREG32(CP_BUSY_STAT));
-	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
-		RREG32(CP_STAT));
+	for (j = 0; j < 10; j++) {
+		for (i = 0; i < rdev->num_crtc; i++) {
+			if (crtc_hung & (1 << i)) {
+				tmp = RREG32(EVERGREEN_CRTC_STATUS_HV_COUNT + crtc_offsets[i]);
+				if (tmp != crtc_status[i])
+					crtc_hung &= ~(1 << i);
+			}
+		}
+		if (crtc_hung == 0)
+			return false;
+		udelay(100);
+	}
+
+	return true;
 }
 
-static void evergreen_gpu_soft_reset_dma(struct radeon_device *rdev)
+static u32 evergreen_gpu_check_soft_reset(struct radeon_device *rdev)
 {
+	u32 reset_mask = 0;
 	u32 tmp;
 
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		return;
+	/* GRBM_STATUS */
+	tmp = RREG32(GRBM_STATUS);
+	if (tmp & (PA_BUSY | SC_BUSY |
+		   SH_BUSY | SX_BUSY |
+		   TA_BUSY | VGT_BUSY |
+		   DB_BUSY | CB_BUSY |
+		   SPI_BUSY | VGT_BUSY_NO_DMA))
+		reset_mask |= RADEON_RESET_GFX;
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	if (tmp & (CF_RQ_PENDING | PF_RQ_PENDING |
+		   CP_BUSY | CP_COHERENCY_BUSY))
+		reset_mask |= RADEON_RESET_CP;
 
-	/* Disable DMA */
-	tmp = RREG32(DMA_RB_CNTL);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL, tmp);
+	if (tmp & GRBM_EE_BUSY)
+		reset_mask |= RADEON_RESET_GRBM | RADEON_RESET_GFX | RADEON_RESET_CP;
 
-	/* Reset dma */
-	WREG32(SRBM_SOFT_RESET, SOFT_RESET_DMA);
-	RREG32(SRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(SRBM_SOFT_RESET, 0);
+	/* DMA_STATUS_REG */
+	tmp = RREG32(DMA_STATUS_REG);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA;
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	/* SRBM_STATUS2 */
+	tmp = RREG32(SRBM_STATUS2);
+	if (tmp & DMA_BUSY)
+		reset_mask |= RADEON_RESET_DMA;
+
+	/* SRBM_STATUS */
+	tmp = RREG32(SRBM_STATUS);
+	if (tmp & (RLC_RQ_PENDING | RLC_BUSY))
+		reset_mask |= RADEON_RESET_RLC;
+
+	if (tmp & IH_BUSY)
+		reset_mask |= RADEON_RESET_IH;
+
+	if (tmp & SEM_BUSY)
+		reset_mask |= RADEON_RESET_SEM;
+
+	if (tmp & GRBM_RQ_PENDING)
+		reset_mask |= RADEON_RESET_GRBM;
+
+	if (tmp & VMC_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
+
+	if (tmp & (MCB_BUSY | MCB_NON_DISPLAY_BUSY |
+		   MCC_BUSY | MCD_BUSY))
+		reset_mask |= RADEON_RESET_MC;
+
+	if (evergreen_is_display_hung(rdev))
+		reset_mask |= RADEON_RESET_DISPLAY;
+
+	/* VM_L2_STATUS */
+	tmp = RREG32(VM_L2_STATUS);
+	if (tmp & L2_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
+
+	return reset_mask;
 }
 
-static int evergreen_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
+static void evergreen_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
 {
 	struct evergreen_mc_save save;
-
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		reset_mask &= ~(RADEON_RESET_GFX | RADEON_RESET_COMPUTE);
-
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		reset_mask &= ~RADEON_RESET_DMA;
+	u32 grbm_soft_reset = 0, srbm_soft_reset = 0;
+	u32 tmp;
 
 	if (reset_mask == 0)
-		return 0;
+		return;
 
 	dev_info(rdev->dev, "GPU softreset: 0x%08X\n", reset_mask);
 
+	evergreen_print_gpu_status_regs(rdev);
+
+	/* Disable CP parsing/prefetching */
+	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT);
+
+	if (reset_mask & RADEON_RESET_DMA) {
+		/* Disable DMA */
+		tmp = RREG32(DMA_RB_CNTL);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL, tmp);
+	}
+
+	udelay(50);
+
 	evergreen_mc_stop(rdev, &save);
 	if (evergreen_mc_wait_for_idle(rdev)) {
 		dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
 	}
 
-	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE))
-		evergreen_gpu_soft_reset_gfx(rdev);
+	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE)) {
+		grbm_soft_reset |= SOFT_RESET_DB |
+			SOFT_RESET_CB |
+			SOFT_RESET_PA |
+			SOFT_RESET_SC |
+			SOFT_RESET_SPI |
+			SOFT_RESET_SX |
+			SOFT_RESET_SH |
+			SOFT_RESET_TC |
+			SOFT_RESET_TA |
+			SOFT_RESET_VC |
+			SOFT_RESET_VGT;
+	}
+
+	if (reset_mask & RADEON_RESET_CP) {
+		grbm_soft_reset |= SOFT_RESET_CP |
+			SOFT_RESET_VGT;
+
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+	}
 
 	if (reset_mask & RADEON_RESET_DMA)
-		evergreen_gpu_soft_reset_dma(rdev);
+		srbm_soft_reset |= SOFT_RESET_DMA;
+
+	if (reset_mask & RADEON_RESET_DISPLAY)
+		srbm_soft_reset |= SOFT_RESET_DC;
+
+	if (reset_mask & RADEON_RESET_RLC)
+		srbm_soft_reset |= SOFT_RESET_RLC;
+
+	if (reset_mask & RADEON_RESET_SEM)
+		srbm_soft_reset |= SOFT_RESET_SEM;
+
+	if (reset_mask & RADEON_RESET_IH)
+		srbm_soft_reset |= SOFT_RESET_IH;
+
+	if (reset_mask & RADEON_RESET_GRBM)
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+
+	if (reset_mask & RADEON_RESET_VMC)
+		srbm_soft_reset |= SOFT_RESET_VMC;
+
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		if (reset_mask & RADEON_RESET_MC)
+			srbm_soft_reset |= SOFT_RESET_MC;
+	}
+
+	if (grbm_soft_reset) {
+		tmp = RREG32(GRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(rdev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~grbm_soft_reset;
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+	}
+
+	if (srbm_soft_reset) {
+		tmp = RREG32(SRBM_SOFT_RESET);
+		tmp |= srbm_soft_reset;
+		dev_info(rdev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~srbm_soft_reset;
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+	}
 
 	/* Wait a little for things to settle down */
 	udelay(50);
 
 	evergreen_mc_resume(rdev, &save);
-	return 0;
+	udelay(50);
+
+	evergreen_print_gpu_status_regs(rdev);
 }
 
 int evergreen_asic_reset(struct radeon_device *rdev)
 {
-	return evergreen_gpu_soft_reset(rdev, (RADEON_RESET_GFX |
-					       RADEON_RESET_COMPUTE |
-					       RADEON_RESET_DMA));
+	u32 reset_mask;
+
+	reset_mask = evergreen_gpu_check_soft_reset(rdev);
+
+	if (reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, true);
+
+	evergreen_gpu_soft_reset(rdev, reset_mask);
+
+	reset_mask = evergreen_gpu_check_soft_reset(rdev);
+
+	if (!reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, false);
+
+	return 0;
+}
+
+/**
+ * evergreen_gfx_is_lockup - Check if the GFX engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the GFX engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool evergreen_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 reset_mask = evergreen_gpu_check_soft_reset(rdev);
+
+	if (!(reset_mask & (RADEON_RESET_GFX |
+			    RADEON_RESET_COMPUTE |
+			    RADEON_RESET_CP))) {
+		radeon_ring_lockup_update(ring);
+		return false;
+	}
+	/* force CP activities */
+	radeon_ring_force_activity(rdev, ring);
+	return radeon_ring_test_lockup(rdev, ring);
+}
+
+/**
+ * evergreen_dma_is_lockup - Check if the DMA engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the async DMA engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool evergreen_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 reset_mask = evergreen_gpu_check_soft_reset(rdev);
+
+	if (!(reset_mask & RADEON_RESET_DMA)) {
+		radeon_ring_lockup_update(ring);
+		return false;
+	}
+	/* force ring activities */
+	radeon_ring_force_activity(rdev, ring);
+	return radeon_ring_test_lockup(rdev, ring);
 }
 
 /* Interrupts */
@@ -3280,14 +3442,14 @@ void evergreen_dma_fence_ring_emit(struct radeon_device *rdev,
 	struct radeon_ring *ring = &rdev->ring[fence->ring];
 	u64 addr = rdev->fence_drv[fence->ring].gpu_addr;
 	/* write the fence */
-	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_FENCE, 0, 0, 0));
+	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_FENCE, 0, 0));
 	radeon_ring_write(ring, addr & 0xfffffffc);
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff));
 	radeon_ring_write(ring, fence->seq);
 	/* generate an interrupt */
-	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_TRAP, 0, 0, 0));
+	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_TRAP, 0, 0));
 	/* flush HDP */
-	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_SRBM_WRITE, 0, 0, 0));
+	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_SRBM_WRITE, 0, 0));
 	radeon_ring_write(ring, (0xf << 16) | (HDP_MEM_COHERENCY_FLUSH_CNTL >> 2));
 	radeon_ring_write(ring, 1);
 }
@@ -3310,7 +3472,7 @@ void evergreen_dma_ring_ib_execute(struct radeon_device *rdev,
 		while ((next_rptr & 7) != 5)
 			next_rptr++;
 		next_rptr += 3;
-		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 1));
+		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 1));
 		radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
 		radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xff);
 		radeon_ring_write(ring, next_rptr);
@@ -3320,8 +3482,8 @@ void evergreen_dma_ring_ib_execute(struct radeon_device *rdev,
 	 * Pad as necessary with NOPs.
 	 */
 	while ((ring->wptr & 7) != 5)
-		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_NOP, 0, 0, 0));
-	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_INDIRECT_BUFFER, 0, 0, 0));
+		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_NOP, 0, 0));
+	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_INDIRECT_BUFFER, 0, 0));
 	radeon_ring_write(ring, (ib->gpu_addr & 0xFFFFFFE0));
 	radeon_ring_write(ring, (ib->length_dw << 12) | (upper_32_bits(ib->gpu_addr) & 0xFF));
 
@@ -3380,7 +3542,7 @@ int evergreen_copy_dma(struct radeon_device *rdev,
 		if (cur_size_in_dw > 0xFFFFF)
 			cur_size_in_dw = 0xFFFFF;
 		size_in_dw -= cur_size_in_dw;
-		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 0, 0, cur_size_in_dw));
+		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 0, cur_size_in_dw));
 		radeon_ring_write(ring, dst_offset & 0xfffffffc);
 		radeon_ring_write(ring, src_offset & 0xfffffffc);
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
@@ -3488,7 +3650,7 @@ static int evergreen_startup(struct radeon_device *rdev)
 	ring = &rdev->ring[R600_RING_TYPE_DMA_INDEX];
 	r = radeon_ring_init(rdev, ring, ring->ring_size, R600_WB_DMA_RPTR_OFFSET,
 			     DMA_RB_RPTR, DMA_RB_WPTR,
-			     2, 0x3fffc, DMA_PACKET(DMA_PACKET_NOP, 0, 0, 0));
+			     2, 0x3fffc, DMA_PACKET(DMA_PACKET_NOP, 0, 0));
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index ee4cff534f10..99fb13286fd0 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -36,9 +36,6 @@
 
 int r600_dma_cs_next_reloc(struct radeon_cs_parser *p,
 			   struct radeon_cs_reloc **cs_reloc);
-static int evergreen_cs_packet_next_reloc(struct radeon_cs_parser *p,
-					  struct radeon_cs_reloc **cs_reloc);
-
 struct evergreen_cs_track {
 	u32			group_size;
 	u32			nbanks;
@@ -1009,223 +1006,35 @@ static int evergreen_cs_track_check(struct radeon_cs_parser *p)
 }
 
 /**
- * evergreen_cs_packet_parse() - parse cp packet and point ib index to next packet
- * @parser:	parser structure holding parsing context.
- * @pkt:	where to store packet informations
- *
- * Assume that chunk_ib_index is properly set. Will return -EINVAL
- * if packet is bigger than remaining ib size. or if packets is unknown.
- **/
-static int evergreen_cs_packet_parse(struct radeon_cs_parser *p,
-			      struct radeon_cs_packet *pkt,
-			      unsigned idx)
-{
-	struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx];
-	uint32_t header;
-
-	if (idx >= ib_chunk->length_dw) {
-		DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
-			  idx, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	header = radeon_get_ib_value(p, idx);
-	pkt->idx = idx;
-	pkt->type = CP_PACKET_GET_TYPE(header);
-	pkt->count = CP_PACKET_GET_COUNT(header);
-	pkt->one_reg_wr = 0;
-	switch (pkt->type) {
-	case PACKET_TYPE0:
-		pkt->reg = CP_PACKET0_GET_REG(header);
-		break;
-	case PACKET_TYPE3:
-		pkt->opcode = CP_PACKET3_GET_OPCODE(header);
-		break;
-	case PACKET_TYPE2:
-		pkt->count = -1;
-		break;
-	default:
-		DRM_ERROR("Unknown packet type %d at %d !\n", pkt->type, idx);
-		return -EINVAL;
-	}
-	if ((pkt->count + 1 + pkt->idx) >= ib_chunk->length_dw) {
-		DRM_ERROR("Packet (%d:%d:%d) end after CS buffer (%d) !\n",
-			  pkt->idx, pkt->type, pkt->count, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	return 0;
-}
-
-/**
- * evergreen_cs_packet_next_reloc() - parse next packet which should be reloc packet3
+ * evergreen_cs_packet_parse_vline() - parse userspace VLINE packet
  * @parser:		parser structure holding parsing context.
- * @data:		pointer to relocation data
- * @offset_start:	starting offset
- * @offset_mask:	offset mask (to align start offset on)
- * @reloc:		reloc informations
  *
- * Check next packet is relocation packet3, do bo validation and compute
- * GPU offset using the provided start.
- **/
-static int evergreen_cs_packet_next_reloc(struct radeon_cs_parser *p,
-					  struct radeon_cs_reloc **cs_reloc)
-{
-	struct radeon_cs_chunk *relocs_chunk;
-	struct radeon_cs_packet p3reloc;
-	unsigned idx;
-	int r;
-
-	if (p->chunk_relocs_idx == -1) {
-		DRM_ERROR("No relocation chunk !\n");
-		return -EINVAL;
-	}
-	*cs_reloc = NULL;
-	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
-	r = evergreen_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return r;
-	}
-	p->idx += p3reloc.count + 2;
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		DRM_ERROR("No packet3 for relocation for packet at %d.\n",
-			  p3reloc.idx);
-		return -EINVAL;
-	}
-	idx = radeon_get_ib_value(p, p3reloc.idx + 1);
-	if (idx >= relocs_chunk->length_dw) {
-		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
-			  idx, relocs_chunk->length_dw);
-		return -EINVAL;
-	}
-	/* FIXME: we assume reloc size is 4 dwords */
-	*cs_reloc = p->relocs_ptr[(idx / 4)];
-	return 0;
-}
-
-/**
- * evergreen_cs_packet_next_is_pkt3_nop() - test if the next packet is NOP
- * @p:		structure holding the parser context.
- *
- * Check if the next packet is a relocation packet3.
- **/
-static bool evergreen_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p)
-{
-	struct radeon_cs_packet p3reloc;
-	int r;
-
-	r = evergreen_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return false;
-	}
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		return false;
-	}
-	return true;
-}
-
-/**
- * evergreen_cs_packet_next_vline() - parse userspace VLINE packet
- * @parser:		parser structure holding parsing context.
- *
- * Userspace sends a special sequence for VLINE waits.
- * PACKET0 - VLINE_START_END + value
- * PACKET3 - WAIT_REG_MEM poll vline status reg
- * RELOC (P3) - crtc_id in reloc.
- *
- * This function parses this and relocates the VLINE START END
- * and WAIT_REG_MEM packets to the correct crtc.
- * It also detects a switched off crtc and nulls out the
- * wait in that case.
+ * This is an Evergreen(+)-specific function for parsing VLINE packets.
+ * Real work is done by r600_cs_common_vline_parse function.
+ * Here we just set up ASIC-specific register table and call
+ * the common implementation function.
  */
 static int evergreen_cs_packet_parse_vline(struct radeon_cs_parser *p)
 {
-	struct drm_mode_object *obj;
-	struct drm_crtc *crtc;
-	struct radeon_crtc *radeon_crtc;
-	struct radeon_cs_packet p3reloc, wait_reg_mem;
-	int crtc_id;
-	int r;
-	uint32_t header, h_idx, reg, wait_reg_mem_info;
-	volatile uint32_t *ib;
-
-	ib = p->ib.ptr;
-
-	/* parse the WAIT_REG_MEM */
-	r = evergreen_cs_packet_parse(p, &wait_reg_mem, p->idx);
-	if (r)
-		return r;
-
-	/* check its a WAIT_REG_MEM */
-	if (wait_reg_mem.type != PACKET_TYPE3 ||
-	    wait_reg_mem.opcode != PACKET3_WAIT_REG_MEM) {
-		DRM_ERROR("vline wait missing WAIT_REG_MEM segment\n");
-		return -EINVAL;
-	}
-
-	wait_reg_mem_info = radeon_get_ib_value(p, wait_reg_mem.idx + 1);
-	/* bit 4 is reg (0) or mem (1) */
-	if (wait_reg_mem_info & 0x10) {
-		DRM_ERROR("vline WAIT_REG_MEM waiting on MEM rather than REG\n");
-		return -EINVAL;
-	}
-	/* waiting for value to be equal */
-	if ((wait_reg_mem_info & 0x7) != 0x3) {
-		DRM_ERROR("vline WAIT_REG_MEM function not equal\n");
-		return -EINVAL;
-	}
-	if ((radeon_get_ib_value(p, wait_reg_mem.idx + 2) << 2) != EVERGREEN_VLINE_STATUS) {
-		DRM_ERROR("vline WAIT_REG_MEM bad reg\n");
-		return -EINVAL;
-	}
-
-	if (radeon_get_ib_value(p, wait_reg_mem.idx + 5) != EVERGREEN_VLINE_STAT) {
-		DRM_ERROR("vline WAIT_REG_MEM bad bit mask\n");
-		return -EINVAL;
-	}
-
-	/* jump over the NOP */
-	r = evergreen_cs_packet_parse(p, &p3reloc, p->idx + wait_reg_mem.count + 2);
-	if (r)
-		return r;
-
-	h_idx = p->idx - 2;
-	p->idx += wait_reg_mem.count + 2;
-	p->idx += p3reloc.count + 2;
 
-	header = radeon_get_ib_value(p, h_idx);
-	crtc_id = radeon_get_ib_value(p, h_idx + 2 + 7 + 1);
-	reg = CP_PACKET0_GET_REG(header);
-	obj = drm_mode_object_find(p->rdev->ddev, crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
-		DRM_ERROR("cannot find crtc %d\n", crtc_id);
-		return -EINVAL;
-	}
-	crtc = obj_to_crtc(obj);
-	radeon_crtc = to_radeon_crtc(crtc);
-	crtc_id = radeon_crtc->crtc_id;
-
-	if (!crtc->enabled) {
-		/* if the CRTC isn't enabled - we need to nop out the WAIT_REG_MEM */
-		ib[h_idx + 2] = PACKET2(0);
-		ib[h_idx + 3] = PACKET2(0);
-		ib[h_idx + 4] = PACKET2(0);
-		ib[h_idx + 5] = PACKET2(0);
-		ib[h_idx + 6] = PACKET2(0);
-		ib[h_idx + 7] = PACKET2(0);
-		ib[h_idx + 8] = PACKET2(0);
-	} else {
-		switch (reg) {
-		case EVERGREEN_VLINE_START_END:
-			header &= ~R600_CP_PACKET0_REG_MASK;
-			header |= (EVERGREEN_VLINE_START_END + radeon_crtc->crtc_offset) >> 2;
-			ib[h_idx] = header;
-			ib[h_idx + 4] = (EVERGREEN_VLINE_STATUS + radeon_crtc->crtc_offset) >> 2;
-			break;
-		default:
-			DRM_ERROR("unknown crtc reloc\n");
-			return -EINVAL;
-		}
-	}
-	return 0;
+	static uint32_t vline_start_end[6] = {
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC0_REGISTER_OFFSET,
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC1_REGISTER_OFFSET,
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		EVERGREEN_VLINE_START_END + EVERGREEN_CRTC5_REGISTER_OFFSET
+	};
+	static uint32_t vline_status[6] = {
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET,
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET,
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		EVERGREEN_VLINE_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET
+	};
+
+	return r600_cs_common_vline_parse(p, vline_start_end, vline_status);
 }
 
 static int evergreen_packet0_check(struct radeon_cs_parser *p,
@@ -1347,7 +1156,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_LSTMP_RING_BASE:
 	case SQ_PSTMP_RING_BASE:
 	case SQ_VSTMP_RING_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1376,7 +1185,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_Z_INFO:
 		track->db_z_info = radeon_get_ib_value(p, idx);
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
 						"0x%04X\n", reg);
@@ -1418,7 +1227,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_dirty = true;
 		break;
 	case DB_Z_READ_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1430,7 +1239,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_dirty = true;
 		break;
 	case DB_Z_WRITE_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1442,7 +1251,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_dirty = true;
 		break;
 	case DB_STENCIL_READ_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1454,7 +1263,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_dirty = true;
 		break;
 	case DB_STENCIL_WRITE_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1477,7 +1286,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case VGT_STRMOUT_BUFFER_BASE_1:
 	case VGT_STRMOUT_BUFFER_BASE_2:
 	case VGT_STRMOUT_BUFFER_BASE_3:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1499,7 +1308,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->streamout_dirty = true;
 		break;
 	case CP_COHER_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "missing reloc for CP_COHER_BASE "
 					"0x%04X\n", reg);
@@ -1563,7 +1372,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		tmp = (reg - CB_COLOR0_INFO) / 0x3c;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
 						"0x%04X\n", reg);
@@ -1581,7 +1390,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		tmp = ((reg - CB_COLOR8_INFO) / 0x1c) + 8;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
 						"0x%04X\n", reg);
@@ -1642,7 +1451,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR5_ATTRIB:
 	case CB_COLOR6_ATTRIB:
 	case CB_COLOR7_ATTRIB:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1670,7 +1479,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR9_ATTRIB:
 	case CB_COLOR10_ATTRIB:
 	case CB_COLOR11_ATTRIB:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1703,7 +1512,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR6_FMASK:
 	case CB_COLOR7_FMASK:
 		tmp = (reg - CB_COLOR0_FMASK) / 0x3c;
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
 			return -EINVAL;
@@ -1720,7 +1529,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR6_CMASK:
 	case CB_COLOR7_CMASK:
 		tmp = (reg - CB_COLOR0_CMASK) / 0x3c;
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
 			return -EINVAL;
@@ -1758,7 +1567,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR5_BASE:
 	case CB_COLOR6_BASE:
 	case CB_COLOR7_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1774,7 +1583,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR9_BASE:
 	case CB_COLOR10_BASE:
 	case CB_COLOR11_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1787,7 +1596,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->cb_dirty = true;
 		break;
 	case DB_HTILE_DATA_BASE:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1905,7 +1714,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_ALU_CONST_CACHE_LS_13:
 	case SQ_ALU_CONST_CACHE_LS_14:
 	case SQ_ALU_CONST_CACHE_LS_15:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1919,7 +1728,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 				 "0x%04X\n", reg);
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONFIG_REG "
 					"0x%04X\n", reg);
@@ -1933,7 +1742,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 				 "0x%04X\n", reg);
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -2018,7 +1827,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			return -EINVAL;
 		}
 
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad SET PREDICATION\n");
 			return -EINVAL;
@@ -2064,7 +1873,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad INDEX_BASE\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad INDEX_BASE\n");
 			return -EINVAL;
@@ -2091,7 +1900,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad DRAW_INDEX\n");
 			return -EINVAL;
@@ -2119,7 +1928,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad DRAW_INDEX_2\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad DRAW_INDEX_2\n");
 			return -EINVAL;
@@ -2210,7 +2019,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad DISPATCH_INDIRECT\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad DISPATCH_INDIRECT\n");
 			return -EINVAL;
@@ -2231,7 +2040,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x10) {
 			uint64_t offset;
 
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad WAIT_REG_MEM\n");
 				return -EINVAL;
@@ -2243,6 +2052,9 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffffc);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
+		} else if (idx_value & 0x100) {
+			DRM_ERROR("cannot use PFP on REG wait\n");
+			return -EINVAL;
 		}
 		break;
 	case PACKET3_CP_DMA:
@@ -2282,7 +2094,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			}
 			/* src address space is memory */
 			if (((info & 0x60000000) >> 29) == 0) {
-				r = evergreen_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
 					DRM_ERROR("bad CP DMA SRC\n");
 					return -EINVAL;
@@ -2320,7 +2132,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				return -EINVAL;
 			}
 			if (((info & 0x00300000) >> 20) == 0) {
-				r = evergreen_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
 					DRM_ERROR("bad CP DMA DST\n");
 					return -EINVAL;
@@ -2354,7 +2166,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		/* 0xffffffff/0x0 is flush all cache flag */
 		if (radeon_get_ib_value(p, idx + 1) != 0xffffffff ||
 		    radeon_get_ib_value(p, idx + 2) != 0) {
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad SURFACE_SYNC\n");
 				return -EINVAL;
@@ -2370,7 +2182,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if (pkt->count) {
 			uint64_t offset;
 
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad EVENT_WRITE\n");
 				return -EINVAL;
@@ -2391,7 +2203,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
@@ -2413,7 +2225,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad EVENT_WRITE_EOS\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad EVENT_WRITE_EOS\n");
 			return -EINVAL;
@@ -2480,7 +2292,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			switch (G__SQ_CONSTANT_TYPE(radeon_get_ib_value(p, idx+1+(i*8)+7))) {
 			case SQ_TEX_VTX_VALID_TEXTURE:
 				/* tex base */
-				r = evergreen_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
 					DRM_ERROR("bad SET_RESOURCE (tex)\n");
 					return -EINVAL;
@@ -2511,13 +2323,13 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 				if ((tex_dim == SQ_TEX_DIM_2D_MSAA || tex_dim == SQ_TEX_DIM_2D_ARRAY_MSAA) &&
 				    !mip_address &&
-				    !evergreen_cs_packet_next_is_pkt3_nop(p)) {
+				    !radeon_cs_packet_next_is_pkt3_nop(p)) {
 					/* MIP_ADDRESS should point to FMASK for an MSAA texture.
 					 * It should be 0 if FMASK is disabled. */
 					moffset = 0;
 					mipmap = NULL;
 				} else {
-					r = evergreen_cs_packet_next_reloc(p, &reloc);
+					r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 					if (r) {
 						DRM_ERROR("bad SET_RESOURCE (tex)\n");
 						return -EINVAL;
@@ -2536,7 +2348,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			{
 				uint64_t offset64;
 				/* vtx base */
-				r = evergreen_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
 					DRM_ERROR("bad SET_RESOURCE (vtx)\n");
 					return -EINVAL;
@@ -2618,7 +2430,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		/* Updating memory at DST_ADDRESS. */
 		if (idx_value & 0x1) {
 			u64 offset;
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
 				return -EINVAL;
@@ -2637,7 +2449,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		/* Reading data from SRC_ADDRESS. */
 		if (((idx_value >> 1) & 0x3) == 2) {
 			u64 offset;
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
 				return -EINVAL;
@@ -2662,7 +2474,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad MEM_WRITE (invalid count)\n");
 			return -EINVAL;
 		}
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("bad MEM_WRITE (missing reloc)\n");
 			return -EINVAL;
@@ -2691,7 +2503,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x1) {
 			u64 offset;
 			/* SRC is memory. */
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad COPY_DW (missing src reloc)\n");
 				return -EINVAL;
@@ -2715,7 +2527,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x2) {
 			u64 offset;
 			/* DST is memory. */
-			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("bad COPY_DW (missing dst reloc)\n");
 				return -EINVAL;
@@ -2819,7 +2631,7 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 		p->track = track;
 	}
 	do {
-		r = evergreen_cs_packet_parse(p, &pkt, p->idx);
+		r = radeon_cs_packet_parse(p, &pkt, p->idx);
 		if (r) {
 			kfree(p->track);
 			p->track = NULL;
@@ -2827,12 +2639,12 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 		}
 		p->idx += pkt.count + 2;
 		switch (pkt.type) {
-		case PACKET_TYPE0:
+		case RADEON_PACKET_TYPE0:
 			r = evergreen_cs_parse_packet0(p, &pkt);
 			break;
-		case PACKET_TYPE2:
+		case RADEON_PACKET_TYPE2:
 			break;
-		case PACKET_TYPE3:
+		case RADEON_PACKET_TYPE3:
 			r = evergreen_packet3_check(p, &pkt);
 			break;
 		default:
@@ -2858,16 +2670,6 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 	return 0;
 }
 
-/*
- *  DMA
- */
-
-#define GET_DMA_CMD(h) (((h) & 0xf0000000) >> 28)
-#define GET_DMA_COUNT(h) ((h) & 0x000fffff)
-#define GET_DMA_T(h) (((h) & 0x00800000) >> 23)
-#define GET_DMA_NEW(h) (((h) & 0x04000000) >> 26)
-#define GET_DMA_MISC(h) (((h) & 0x0700000) >> 20)
-
 /**
  * evergreen_dma_cs_parse() - parse the DMA IB
  * @p:		parser structure holding parsing context.
@@ -2881,9 +2683,9 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 {
 	struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx];
 	struct radeon_cs_reloc *src_reloc, *dst_reloc, *dst2_reloc;
-	u32 header, cmd, count, tiled, new_cmd, misc;
+	u32 header, cmd, count, sub_cmd;
 	volatile u32 *ib = p->ib.ptr;
-	u32 idx, idx_value;
+	u32 idx;
 	u64 src_offset, dst_offset, dst2_offset;
 	int r;
 
@@ -2897,9 +2699,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 		header = radeon_get_ib_value(p, idx);
 		cmd = GET_DMA_CMD(header);
 		count = GET_DMA_COUNT(header);
-		tiled = GET_DMA_T(header);
-		new_cmd = GET_DMA_NEW(header);
-		misc = GET_DMA_MISC(header);
+		sub_cmd = GET_DMA_SUB_CMD(header);
 
 		switch (cmd) {
 		case DMA_PACKET_WRITE:
@@ -2908,19 +2708,27 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				DRM_ERROR("bad DMA_PACKET_WRITE\n");
 				return -EINVAL;
 			}
-			if (tiled) {
+			switch (sub_cmd) {
+			/* tiled */
+			case 8:
 				dst_offset = radeon_get_ib_value(p, idx+1);
 				dst_offset <<= 8;
 
 				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
 				p->idx += count + 7;
-			} else {
+				break;
+			/* linear */
+			case 0:
 				dst_offset = radeon_get_ib_value(p, idx+1);
 				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
 
 				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
 				ib[idx+2] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
 				p->idx += count + 3;
+				break;
+			default:
+				DRM_ERROR("bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n", idx, header);
+				return -EINVAL;
 			}
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
 				dev_warn(p->dev, "DMA write buffer too small (%llu %lu)\n",
@@ -2939,338 +2747,330 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				DRM_ERROR("bad DMA_PACKET_COPY\n");
 				return -EINVAL;
 			}
-			if (tiled) {
-				idx_value = radeon_get_ib_value(p, idx + 2);
-				if (new_cmd) {
-					switch (misc) {
-					case 0:
-						/* L2T, frame to fields */
-						if (idx_value & (1 << 31)) {
-							DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						r = r600_dma_cs_next_reloc(p, &dst2_reloc);
-						if (r) {
-							DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						dst_offset = radeon_get_ib_value(p, idx+1);
-						dst_offset <<= 8;
-						dst2_offset = radeon_get_ib_value(p, idx+2);
-						dst2_offset <<= 8;
-						src_offset = radeon_get_ib_value(p, idx+8);
-						src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, frame to fields src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
-								 dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
-						ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						p->idx += 10;
-						break;
-					case 1:
-						/* L2T, T2L partial */
-						if (p->family < CHIP_CAYMAN) {
-							DRM_ERROR("L2T, T2L Partial is cayman only !\n");
-							return -EINVAL;
-						}
-						/* detile bit */
-						if (idx_value & (1 << 31)) {
-							/* tiled src, linear dst */
-							ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
-
-							ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-						} else {
-							/* linear src, tiled dst */
-							ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-
-							ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						}
-						p->idx += 12;
-						break;
-					case 3:
-						/* L2T, broadcast */
-						if (idx_value & (1 << 31)) {
-							DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						r = r600_dma_cs_next_reloc(p, &dst2_reloc);
-						if (r) {
-							DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						dst_offset = radeon_get_ib_value(p, idx+1);
-						dst_offset <<= 8;
-						dst2_offset = radeon_get_ib_value(p, idx+2);
-						dst2_offset <<= 8;
-						src_offset = radeon_get_ib_value(p, idx+8);
-						src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
-								 dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
-						ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						p->idx += 10;
-						break;
-					case 4:
-						/* L2T, T2L */
-						/* detile bit */
-						if (idx_value & (1 << 31)) {
-							/* tiled src, linear dst */
-							src_offset = radeon_get_ib_value(p, idx+1);
-							src_offset <<= 8;
-							ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
-
-							dst_offset = radeon_get_ib_value(p, idx+7);
-							dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
-							ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-						} else {
-							/* linear src, tiled dst */
-							src_offset = radeon_get_ib_value(p, idx+7);
-							src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
-							ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-
-							dst_offset = radeon_get_ib_value(p, idx+1);
-							dst_offset <<= 8;
-							ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						}
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, T2L src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, T2L dst buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						p->idx += 9;
-						break;
-					case 5:
-						/* T2T partial */
-						if (p->family < CHIP_CAYMAN) {
-							DRM_ERROR("L2T, T2L Partial is cayman only !\n");
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
-						ib[idx+4] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						p->idx += 13;
-						break;
-					case 7:
-						/* L2T, broadcast */
-						if (idx_value & (1 << 31)) {
-							DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						r = r600_dma_cs_next_reloc(p, &dst2_reloc);
-						if (r) {
-							DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						dst_offset = radeon_get_ib_value(p, idx+1);
-						dst_offset <<= 8;
-						dst2_offset = radeon_get_ib_value(p, idx+2);
-						dst2_offset <<= 8;
-						src_offset = radeon_get_ib_value(p, idx+8);
-						src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
-								 dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
-						ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						p->idx += 10;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
+			switch (sub_cmd) {
+			/* Copy L2L, DW aligned */
+			case 0x00:
+				/* L2L, dw */
+				src_offset = radeon_get_ib_value(p, idx+2);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, dw src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, dw dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+				ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 5;
+				break;
+			/* Copy L2T/T2L */
+			case 0x08:
+				/* detile bit */
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					/* tiled src, linear dst */
+					src_offset = radeon_get_ib_value(p, idx+1);
+					src_offset <<= 8;
+					ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
+
+					dst_offset = radeon_get_ib_value(p, idx + 7);
+					dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
+					ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
 				} else {
-					switch (misc) {
-					case 0:
-						/* detile bit */
-						if (idx_value & (1 << 31)) {
-							/* tiled src, linear dst */
-							src_offset = radeon_get_ib_value(p, idx+1);
-							src_offset <<= 8;
-							ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
-
-							dst_offset = radeon_get_ib_value(p, idx+7);
-							dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
-							ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-						} else {
-							/* linear src, tiled dst */
-							src_offset = radeon_get_ib_value(p, idx+7);
-							src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
-							ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-							ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-
-							dst_offset = radeon_get_ib_value(p, idx+1);
-							dst_offset <<= 8;
-							ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
-						}
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						p->idx += 9;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
+					/* linear src, tiled dst */
+					src_offset = radeon_get_ib_value(p, idx+7);
+					src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
+					ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+
+					dst_offset = radeon_get_ib_value(p, idx+1);
+					dst_offset <<= 8;
+					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
 				}
-			} else {
-				if (new_cmd) {
-					switch (misc) {
-					case 0:
-						/* L2L, byte */
-						src_offset = radeon_get_ib_value(p, idx+2);
-						src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
-						dst_offset = radeon_get_ib_value(p, idx+1);
-						dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
-						if ((src_offset + count) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2L, byte src buffer too small (%llu %lu)\n",
-								 src_offset + count, radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + count) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2L, byte dst buffer too small (%llu %lu)\n",
-								 dst_offset + count, radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xffffffff);
-						ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xffffffff);
-						ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-						ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						p->idx += 5;
-						break;
-					case 1:
-						/* L2L, partial */
-						if (p->family < CHIP_CAYMAN) {
-							DRM_ERROR("L2L Partial is cayman only !\n");
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset & 0xffffffff);
-						ib[idx+2] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						ib[idx+4] += (u32)(dst_reloc->lobj.gpu_offset & 0xffffffff);
-						ib[idx+5] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-
-						p->idx += 9;
-						break;
-					case 4:
-						/* L2L, dw, broadcast */
-						r = r600_dma_cs_next_reloc(p, &dst2_reloc);
-						if (r) {
-							DRM_ERROR("bad L2L, dw, broadcast DMA_PACKET_COPY\n");
-							return -EINVAL;
-						}
-						dst_offset = radeon_get_ib_value(p, idx+1);
-						dst_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
-						dst2_offset = radeon_get_ib_value(p, idx+2);
-						dst2_offset |= ((u64)(radeon_get_ib_value(p, idx+5) & 0xff)) << 32;
-						src_offset = radeon_get_ib_value(p, idx+3);
-						src_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
-						if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2L, dw, broadcast src buffer too small (%llu %lu)\n",
-								 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2L, dw, broadcast dst buffer too small (%llu %lu)\n",
-								 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-							return -EINVAL;
-						}
-						if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-							dev_warn(p->dev, "DMA L2L, dw, broadcast dst2 buffer too small (%llu %lu)\n",
-								 dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
-							return -EINVAL;
-						}
-						ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+3] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-						ib[idx+4] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-						ib[idx+5] += upper_32_bits(dst2_reloc->lobj.gpu_offset) & 0xff;
-						ib[idx+6] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-						p->idx += 7;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				p->idx += 9;
+				break;
+			/* Copy L2L, byte aligned */
+			case 0x40:
+				/* L2L, byte */
+				src_offset = radeon_get_ib_value(p, idx+2);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
+				if ((src_offset + count) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, byte src buffer too small (%llu %lu)\n",
+							src_offset + count, radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + count) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, byte dst buffer too small (%llu %lu)\n",
+							dst_offset + count, radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xffffffff);
+				ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xffffffff);
+				ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+				ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 5;
+				break;
+			/* Copy L2L, partial */
+			case 0x41:
+				/* L2L, partial */
+				if (p->family < CHIP_CAYMAN) {
+					DRM_ERROR("L2L Partial is cayman only !\n");
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset & 0xffffffff);
+				ib[idx+2] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				ib[idx+4] += (u32)(dst_reloc->lobj.gpu_offset & 0xffffffff);
+				ib[idx+5] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+
+				p->idx += 9;
+				break;
+			/* Copy L2L, DW aligned, broadcast */
+			case 0x44:
+				/* L2L, dw, broadcast */
+				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
+				if (r) {
+					DRM_ERROR("bad L2L, dw, broadcast DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+				dst2_offset = radeon_get_ib_value(p, idx+2);
+				dst2_offset |= ((u64)(radeon_get_ib_value(p, idx+5) & 0xff)) << 32;
+				src_offset = radeon_get_ib_value(p, idx+3);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, dw, broadcast src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, dw, broadcast dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2L, dw, broadcast dst2 buffer too small (%llu %lu)\n",
+							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+3] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+4] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+				ib[idx+5] += upper_32_bits(dst2_reloc->lobj.gpu_offset) & 0xff;
+				ib[idx+6] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 7;
+				break;
+			/* Copy L2T Frame to Field */
+			case 0x48:
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
+				if (r) {
+					DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset <<= 8;
+				dst2_offset = radeon_get_ib_value(p, idx+2);
+				dst2_offset <<= 8;
+				src_offset = radeon_get_ib_value(p, idx+8);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, frame to fields src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
+							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
+				ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
+				ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 10;
+				break;
+			/* Copy L2T/T2L, partial */
+			case 0x49:
+				/* L2T, T2L partial */
+				if (p->family < CHIP_CAYMAN) {
+					DRM_ERROR("L2T, T2L Partial is cayman only !\n");
+					return -EINVAL;
+				}
+				/* detile bit */
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					/* tiled src, linear dst */
+					ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
+
+					ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
+				} else {
+					/* linear src, tiled dst */
+					ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+
+					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
+				}
+				p->idx += 12;
+				break;
+			/* Copy L2T broadcast */
+			case 0x4b:
+				/* L2T, broadcast */
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
+				if (r) {
+					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset <<= 8;
+				dst2_offset = radeon_get_ib_value(p, idx+2);
+				dst2_offset <<= 8;
+				src_offset = radeon_get_ib_value(p, idx+8);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
+							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
+				ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
+				ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 10;
+				break;
+			/* Copy L2T/T2L (tile units) */
+			case 0x4c:
+				/* L2T, T2L */
+				/* detile bit */
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					/* tiled src, linear dst */
+					src_offset = radeon_get_ib_value(p, idx+1);
+					src_offset <<= 8;
+					ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
+
+					dst_offset = radeon_get_ib_value(p, idx+7);
+					dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
+					ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
 				} else {
-					/* L2L, dw */
-					src_offset = radeon_get_ib_value(p, idx+2);
-					src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+					/* linear src, tiled dst */
+					src_offset = radeon_get_ib_value(p, idx+7);
+					src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
+					ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+					ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+
 					dst_offset = radeon_get_ib_value(p, idx+1);
-					dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
-					if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-						dev_warn(p->dev, "DMA L2L, dw src buffer too small (%llu %lu)\n",
-							 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
-						return -EINVAL;
-					}
-					if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-						dev_warn(p->dev, "DMA L2L, dw dst buffer too small (%llu %lu)\n",
-							 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
-						return -EINVAL;
-					}
-					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
-					ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
-					ib[idx+3] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
-					ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
-					p->idx += 5;
+					dst_offset <<= 8;
+					ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
 				}
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, T2L src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, T2L dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				p->idx += 9;
+				break;
+			/* Copy T2T, partial (tile units) */
+			case 0x4d:
+				/* T2T partial */
+				if (p->family < CHIP_CAYMAN) {
+					DRM_ERROR("L2T, T2L Partial is cayman only !\n");
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
+				ib[idx+4] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
+				p->idx += 13;
+				break;
+			/* Copy L2T broadcast (tile units) */
+			case 0x4f:
+				/* L2T, broadcast */
+				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
+					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
+				if (r) {
+					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					return -EINVAL;
+				}
+				dst_offset = radeon_get_ib_value(p, idx+1);
+				dst_offset <<= 8;
+				dst2_offset = radeon_get_ib_value(p, idx+2);
+				dst2_offset <<= 8;
+				src_offset = radeon_get_ib_value(p, idx+8);
+				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
+				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
+							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
+							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					return -EINVAL;
+				}
+				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
+					dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
+							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					return -EINVAL;
+				}
+				ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
+				ib[idx+2] += (u32)(dst2_reloc->lobj.gpu_offset >> 8);
+				ib[idx+8] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
+				ib[idx+9] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
+				p->idx += 10;
+				break;
+			default:
+				DRM_ERROR("bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n", idx, header);
+				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_CONSTANT_FILL:
@@ -3583,19 +3383,19 @@ int evergreen_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 
 	do {
 		pkt.idx = idx;
-		pkt.type = CP_PACKET_GET_TYPE(ib->ptr[idx]);
-		pkt.count = CP_PACKET_GET_COUNT(ib->ptr[idx]);
+		pkt.type = RADEON_CP_PACKET_GET_TYPE(ib->ptr[idx]);
+		pkt.count = RADEON_CP_PACKET_GET_COUNT(ib->ptr[idx]);
 		pkt.one_reg_wr = 0;
 		switch (pkt.type) {
-		case PACKET_TYPE0:
+		case RADEON_PACKET_TYPE0:
 			dev_err(rdev->dev, "Packet0 not allowed!\n");
 			ret = -EINVAL;
 			break;
-		case PACKET_TYPE2:
+		case RADEON_PACKET_TYPE2:
 			idx += 1;
 			break;
-		case PACKET_TYPE3:
-			pkt.opcode = CP_PACKET3_GET_OPCODE(ib->ptr[idx]);
+		case RADEON_PACKET_TYPE3:
+			pkt.opcode = RADEON_CP_PACKET3_GET_OPCODE(ib->ptr[idx]);
 			ret = evergreen_vm_packet3_check(rdev, ib->ptr, &pkt);
 			idx += pkt.count + 2;
 			break;
@@ -3623,88 +3423,79 @@ int evergreen_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 {
 	u32 idx = 0;
-	u32 header, cmd, count, tiled, new_cmd, misc;
+	u32 header, cmd, count, sub_cmd;
 
 	do {
 		header = ib->ptr[idx];
 		cmd = GET_DMA_CMD(header);
 		count = GET_DMA_COUNT(header);
-		tiled = GET_DMA_T(header);
-		new_cmd = GET_DMA_NEW(header);
-		misc = GET_DMA_MISC(header);
+		sub_cmd = GET_DMA_SUB_CMD(header);
 
 		switch (cmd) {
 		case DMA_PACKET_WRITE:
-			if (tiled)
+			switch (sub_cmd) {
+			/* tiled */
+			case 8:
 				idx += count + 7;
-			else
+				break;
+			/* linear */
+			case 0:
 				idx += count + 3;
+				break;
+			default:
+				DRM_ERROR("bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n", idx, ib->ptr[idx]);
+				return -EINVAL;
+			}
 			break;
 		case DMA_PACKET_COPY:
-			if (tiled) {
-				if (new_cmd) {
-					switch (misc) {
-					case 0:
-						/* L2T, frame to fields */
-						idx += 10;
-						break;
-					case 1:
-						/* L2T, T2L partial */
-						idx += 12;
-						break;
-					case 3:
-						/* L2T, broadcast */
-						idx += 10;
-						break;
-					case 4:
-						/* L2T, T2L */
-						idx += 9;
-						break;
-					case 5:
-						/* T2T partial */
-						idx += 13;
-						break;
-					case 7:
-						/* L2T, broadcast */
-						idx += 10;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
-				} else {
-					switch (misc) {
-					case 0:
-						idx += 9;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
-				}
-			} else {
-				if (new_cmd) {
-					switch (misc) {
-					case 0:
-						/* L2L, byte */
-						idx += 5;
-						break;
-					case 1:
-						/* L2L, partial */
-						idx += 9;
-						break;
-					case 4:
-						/* L2L, dw, broadcast */
-						idx += 7;
-						break;
-					default:
-						DRM_ERROR("bad DMA_PACKET_COPY misc %u\n", misc);
-						return -EINVAL;
-					}
-				} else {
-					/* L2L, dw */
-					idx += 5;
-				}
+			switch (sub_cmd) {
+			/* Copy L2L, DW aligned */
+			case 0x00:
+				idx += 5;
+				break;
+			/* Copy L2T/T2L */
+			case 0x08:
+				idx += 9;
+				break;
+			/* Copy L2L, byte aligned */
+			case 0x40:
+				idx += 5;
+				break;
+			/* Copy L2L, partial */
+			case 0x41:
+				idx += 9;
+				break;
+			/* Copy L2L, DW aligned, broadcast */
+			case 0x44:
+				idx += 7;
+				break;
+			/* Copy L2T Frame to Field */
+			case 0x48:
+				idx += 10;
+				break;
+			/* Copy L2T/T2L, partial */
+			case 0x49:
+				idx += 12;
+				break;
+			/* Copy L2T broadcast */
+			case 0x4b:
+				idx += 10;
+				break;
+			/* Copy L2T/T2L (tile units) */
+			case 0x4c:
+				idx += 9;
+				break;
+			/* Copy T2T, partial (tile units) */
+			case 0x4d:
+				idx += 13;
+				break;
+			/* Copy L2T broadcast (tile units) */
+			case 0x4f:
+				idx += 10;
+				break;
+			default:
+				DRM_ERROR("bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n", idx, ib->ptr[idx]);
+				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_CONSTANT_FILL:
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 327c08b54180..4fdecc2b4040 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -24,6 +24,7 @@
  * Authors: Christian König
  *          Rafał Miłecki
  */
+#include <linux/hdmi.h>
 #include <drm/drmP.h>
 #include <drm/radeon_drm.h>
 #include "radeon.h"
@@ -54,79 +55,18 @@ static void evergreen_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t cloc
 }
 
 /*
- * calculate the crc for a given info frame
- */
-static void evergreen_hdmi_infoframe_checksum(uint8_t packetType,
-					 uint8_t versionNumber,
-					 uint8_t length,
-					 uint8_t *frame)
-{
-	int i;
-	frame[0] = packetType + versionNumber + length;
-	for (i = 1; i <= length; i++)
-		frame[0] += frame[i];
-	frame[0] = 0x100 - frame[0];
-}
-
-/*
  * build a HDMI Video Info Frame
  */
-static void evergreen_hdmi_videoinfoframe(
-	struct drm_encoder *encoder,
-	uint8_t color_format,
-	int active_information_present,
-	uint8_t active_format_aspect_ratio,
-	uint8_t scan_information,
-	uint8_t colorimetry,
-	uint8_t ex_colorimetry,
-	uint8_t quantization,
-	int ITC,
-	uint8_t picture_aspect_ratio,
-	uint8_t video_format_identification,
-	uint8_t pixel_repetition,
-	uint8_t non_uniform_picture_scaling,
-	uint8_t bar_info_data_valid,
-	uint16_t top_bar,
-	uint16_t bottom_bar,
-	uint16_t left_bar,
-	uint16_t right_bar
-)
+static void evergreen_hdmi_update_avi_infoframe(struct drm_encoder *encoder,
+						void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
+	uint8_t *frame = buffer + 3;
 
-	uint8_t frame[14];
-
-	frame[0x0] = 0;
-	frame[0x1] =
-		(scan_information & 0x3) |
-		((bar_info_data_valid & 0x3) << 2) |
-		((active_information_present & 0x1) << 4) |
-		((color_format & 0x3) << 5);
-	frame[0x2] =
-		(active_format_aspect_ratio & 0xF) |
-		((picture_aspect_ratio & 0x3) << 4) |
-		((colorimetry & 0x3) << 6);
-	frame[0x3] =
-		(non_uniform_picture_scaling & 0x3) |
-		((quantization & 0x3) << 2) |
-		((ex_colorimetry & 0x7) << 4) |
-		((ITC & 0x1) << 7);
-	frame[0x4] = (video_format_identification & 0x7F);
-	frame[0x5] = (pixel_repetition & 0xF);
-	frame[0x6] = (top_bar & 0xFF);
-	frame[0x7] = (top_bar >> 8);
-	frame[0x8] = (bottom_bar & 0xFF);
-	frame[0x9] = (bottom_bar >> 8);
-	frame[0xA] = (left_bar & 0xFF);
-	frame[0xB] = (left_bar >> 8);
-	frame[0xC] = (right_bar & 0xFF);
-	frame[0xD] = (right_bar >> 8);
-
-	evergreen_hdmi_infoframe_checksum(0x82, 0x02, 0x0D, frame);
 	/* Our header values (type, version, length) should be alright, Intel
 	 * is using the same. Checksum function also seems to be OK, it works
 	 * fine for audio infoframe. However calculated value is always lower
@@ -154,7 +94,10 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
+	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
+	ssize_t err;
 
 	/* Silent, r600_hdmi_enable will raise WARN for us */
 	if (!dig->afmt->enabled)
@@ -200,9 +143,19 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 
 	WREG32(HDMI_GC + offset, 0); /* unset HDMI_GC_AVMUTE */
 
-	evergreen_hdmi_videoinfoframe(encoder, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-				      0, 0, 0, 0, 0, 0);
+	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	if (err < 0) {
+		DRM_ERROR("failed to setup AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		DRM_ERROR("failed to pack AVI infoframe: %zd\n", err);
+		return;
+	}
 
+	evergreen_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
 	evergreen_hdmi_update_ACR(encoder, mode->clock);
 
 	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h
index 034f4c22e5db..f585be16e2d5 100644
--- a/drivers/gpu/drm/radeon/evergreen_reg.h
+++ b/drivers/gpu/drm/radeon/evergreen_reg.h
@@ -223,6 +223,7 @@
 #define EVERGREEN_CRTC_STATUS                           0x6e8c
 #       define EVERGREEN_CRTC_V_BLANK                   (1 << 0)
 #define EVERGREEN_CRTC_STATUS_POSITION                  0x6e90
+#define EVERGREEN_CRTC_STATUS_HV_COUNT                  0x6ea0
 #define EVERGREEN_MASTER_UPDATE_MODE                    0x6ef8
 #define EVERGREEN_CRTC_UPDATE_LOCK                      0x6ed4
 
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index 0bfd0e9e469b..982d25ad9af3 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -729,6 +729,18 @@
 #define	WAIT_UNTIL					0x8040
 
 #define	SRBM_STATUS				        0x0E50
+#define		RLC_RQ_PENDING 				(1 << 3)
+#define		GRBM_RQ_PENDING 			(1 << 5)
+#define		VMC_BUSY 				(1 << 8)
+#define		MCB_BUSY 				(1 << 9)
+#define		MCB_NON_DISPLAY_BUSY 			(1 << 10)
+#define		MCC_BUSY 				(1 << 11)
+#define		MCD_BUSY 				(1 << 12)
+#define		SEM_BUSY 				(1 << 14)
+#define		RLC_BUSY 				(1 << 15)
+#define		IH_BUSY 				(1 << 17)
+#define	SRBM_STATUS2				        0x0EC4
+#define		DMA_BUSY 				(1 << 5)
 #define	SRBM_SOFT_RESET				        0x0E60
 #define		SRBM_SOFT_RESET_ALL_MASK    	       	0x00FEEFA6
 #define		SOFT_RESET_BIF				(1 << 1)
@@ -924,20 +936,23 @@
 #define CAYMAN_DMA1_CNTL                                  0xd82c
 
 /* async DMA packets */
-#define DMA_PACKET(cmd, t, s, n)	((((cmd) & 0xF) << 28) |	\
-					 (((t) & 0x1) << 23) |		\
-					 (((s) & 0x1) << 22) |		\
-					 (((n) & 0xFFFFF) << 0))
+#define DMA_PACKET(cmd, sub_cmd, n) ((((cmd) & 0xF) << 28) |    \
+                                    (((sub_cmd) & 0xFF) << 20) |\
+                                    (((n) & 0xFFFFF) << 0))
+#define GET_DMA_CMD(h) (((h) & 0xf0000000) >> 28)
+#define GET_DMA_COUNT(h) ((h) & 0x000fffff)
+#define GET_DMA_SUB_CMD(h) (((h) & 0x0ff00000) >> 20)
+
 /* async DMA Packet types */
-#define	DMA_PACKET_WRITE				  0x2
-#define	DMA_PACKET_COPY					  0x3
-#define	DMA_PACKET_INDIRECT_BUFFER			  0x4
-#define	DMA_PACKET_SEMAPHORE				  0x5
-#define	DMA_PACKET_FENCE				  0x6
-#define	DMA_PACKET_TRAP					  0x7
-#define	DMA_PACKET_SRBM_WRITE				  0x9
-#define	DMA_PACKET_CONSTANT_FILL			  0xd
-#define	DMA_PACKET_NOP					  0xf
+#define	DMA_PACKET_WRITE                        0x2
+#define	DMA_PACKET_COPY                         0x3
+#define	DMA_PACKET_INDIRECT_BUFFER              0x4
+#define	DMA_PACKET_SEMAPHORE                    0x5
+#define	DMA_PACKET_FENCE                        0x6
+#define	DMA_PACKET_TRAP                         0x7
+#define	DMA_PACKET_SRBM_WRITE                   0x9
+#define	DMA_PACKET_CONSTANT_FILL                0xd
+#define	DMA_PACKET_NOP                          0xf
 
 /* PCIE link stuff */
 #define PCIE_LC_TRAINING_CNTL                             0xa1 /* PCIE_P */
@@ -980,16 +995,7 @@
 /*
  * PM4
  */
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0xFFFF) << 2)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-#define PACKET0(reg, n)	((PACKET_TYPE0 << 30) |				\
+#define PACKET0(reg, n)	((RADEON_PACKET_TYPE0 << 30) |			\
 			 (((reg) >> 2) & 0xFFFF) |			\
 			 ((n) & 0x3FFF) << 16)
 #define CP_PACKET2			0x80000000
@@ -998,7 +1004,7 @@
 
 #define PACKET2(v)	(CP_PACKET2 | REG_SET(PACKET2_PAD, (v)))
 
-#define PACKET3(op, n)	((PACKET_TYPE3 << 30) |				\
+#define PACKET3(op, n)	((RADEON_PACKET_TYPE3 << 30) |			\
 			 (((op) & 0xFF) << 8) |				\
 			 ((n) & 0x3FFF) << 16)
 
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 835992d8d067..7cead763be9e 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -34,6 +34,8 @@
 #include "ni_reg.h"
 #include "cayman_blit_shaders.h"
 
+extern bool evergreen_is_display_hung(struct radeon_device *rdev);
+extern void evergreen_print_gpu_status_regs(struct radeon_device *rdev);
 extern void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save);
 extern void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save);
 extern int evergreen_mc_wait_for_idle(struct radeon_device *rdev);
@@ -1310,120 +1312,90 @@ void cayman_dma_fini(struct radeon_device *rdev)
 	radeon_ring_fini(rdev, &rdev->ring[CAYMAN_RING_TYPE_DMA1_INDEX]);
 }
 
-static void cayman_gpu_soft_reset_gfx(struct radeon_device *rdev)
+static u32 cayman_gpu_check_soft_reset(struct radeon_device *rdev)
 {
-	u32 grbm_reset = 0;
+	u32 reset_mask = 0;
+	u32 tmp;
 
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		return;
+	/* GRBM_STATUS */
+	tmp = RREG32(GRBM_STATUS);
+	if (tmp & (PA_BUSY | SC_BUSY |
+		   SH_BUSY | SX_BUSY |
+		   TA_BUSY | VGT_BUSY |
+		   DB_BUSY | CB_BUSY |
+		   GDS_BUSY | SPI_BUSY |
+		   IA_BUSY | IA_BUSY_NO_DMA))
+		reset_mask |= RADEON_RESET_GFX;
 
-	dev_info(rdev->dev, "  GRBM_STATUS               = 0x%08X\n",
-		RREG32(GRBM_STATUS));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE0           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE0));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE1           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE1));
-	dev_info(rdev->dev, "  SRBM_STATUS               = 0x%08X\n",
-		RREG32(SRBM_STATUS));
-	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT1));
-	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT2));
-	dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
-		RREG32(CP_BUSY_STAT));
-	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
-		RREG32(CP_STAT));
+	if (tmp & (CF_RQ_PENDING | PF_RQ_PENDING |
+		   CP_BUSY | CP_COHERENCY_BUSY))
+		reset_mask |= RADEON_RESET_CP;
 
-	/* Disable CP parsing/prefetching */
-	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT);
+	if (tmp & GRBM_EE_BUSY)
+		reset_mask |= RADEON_RESET_GRBM | RADEON_RESET_GFX | RADEON_RESET_CP;
 
-	/* reset all the gfx blocks */
-	grbm_reset = (SOFT_RESET_CP |
-		      SOFT_RESET_CB |
-		      SOFT_RESET_DB |
-		      SOFT_RESET_GDS |
-		      SOFT_RESET_PA |
-		      SOFT_RESET_SC |
-		      SOFT_RESET_SPI |
-		      SOFT_RESET_SH |
-		      SOFT_RESET_SX |
-		      SOFT_RESET_TC |
-		      SOFT_RESET_TA |
-		      SOFT_RESET_VGT |
-		      SOFT_RESET_IA);
-
-	dev_info(rdev->dev, "  GRBM_SOFT_RESET=0x%08X\n", grbm_reset);
-	WREG32(GRBM_SOFT_RESET, grbm_reset);
-	(void)RREG32(GRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(GRBM_SOFT_RESET, 0);
-	(void)RREG32(GRBM_SOFT_RESET);
-
-	dev_info(rdev->dev, "  GRBM_STATUS               = 0x%08X\n",
-		RREG32(GRBM_STATUS));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE0           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE0));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE1           = 0x%08X\n",
-		RREG32(GRBM_STATUS_SE1));
-	dev_info(rdev->dev, "  SRBM_STATUS               = 0x%08X\n",
-		RREG32(SRBM_STATUS));
-	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT1));
-	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT2));
-	dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
-		RREG32(CP_BUSY_STAT));
-	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
-		RREG32(CP_STAT));
+	/* DMA_STATUS_REG 0 */
+	tmp = RREG32(DMA_STATUS_REG + DMA0_REGISTER_OFFSET);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA;
 
-}
+	/* DMA_STATUS_REG 1 */
+	tmp = RREG32(DMA_STATUS_REG + DMA1_REGISTER_OFFSET);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA1;
 
-static void cayman_gpu_soft_reset_dma(struct radeon_device *rdev)
-{
-	u32 tmp;
+	/* SRBM_STATUS2 */
+	tmp = RREG32(SRBM_STATUS2);
+	if (tmp & DMA_BUSY)
+		reset_mask |= RADEON_RESET_DMA;
 
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		return;
+	if (tmp & DMA1_BUSY)
+		reset_mask |= RADEON_RESET_DMA1;
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	/* SRBM_STATUS */
+	tmp = RREG32(SRBM_STATUS);
+	if (tmp & (RLC_RQ_PENDING | RLC_BUSY))
+		reset_mask |= RADEON_RESET_RLC;
 
-	/* dma0 */
-	tmp = RREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET, tmp);
+	if (tmp & IH_BUSY)
+		reset_mask |= RADEON_RESET_IH;
 
-	/* dma1 */
-	tmp = RREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET, tmp);
+	if (tmp & SEM_BUSY)
+		reset_mask |= RADEON_RESET_SEM;
 
-	/* Reset dma */
-	WREG32(SRBM_SOFT_RESET, SOFT_RESET_DMA | SOFT_RESET_DMA1);
-	RREG32(SRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(SRBM_SOFT_RESET, 0);
+	if (tmp & GRBM_RQ_PENDING)
+		reset_mask |= RADEON_RESET_GRBM;
+
+	if (tmp & VMC_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
+
+	if (tmp & (MCB_BUSY | MCB_NON_DISPLAY_BUSY |
+		   MCC_BUSY | MCD_BUSY))
+		reset_mask |= RADEON_RESET_MC;
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	if (evergreen_is_display_hung(rdev))
+		reset_mask |= RADEON_RESET_DISPLAY;
 
+	/* VM_L2_STATUS */
+	tmp = RREG32(VM_L2_STATUS);
+	if (tmp & L2_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
+
+	return reset_mask;
 }
 
-static int cayman_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
+static void cayman_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
 {
 	struct evergreen_mc_save save;
-
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		reset_mask &= ~(RADEON_RESET_GFX | RADEON_RESET_COMPUTE);
-
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		reset_mask &= ~RADEON_RESET_DMA;
+	u32 grbm_soft_reset = 0, srbm_soft_reset = 0;
+	u32 tmp;
 
 	if (reset_mask == 0)
-		return 0;
+		return;
 
 	dev_info(rdev->dev, "GPU softreset: 0x%08X\n", reset_mask);
 
+	evergreen_print_gpu_status_regs(rdev);
 	dev_info(rdev->dev, "  VM_CONTEXT0_PROTECTION_FAULT_ADDR   0x%08X\n",
 		 RREG32(0x14F8));
 	dev_info(rdev->dev, "  VM_CONTEXT0_PROTECTION_FAULT_STATUS 0x%08X\n",
@@ -1433,29 +1405,158 @@ static int cayman_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
 	dev_info(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 		 RREG32(0x14DC));
 
+	/* Disable CP parsing/prefetching */
+	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT);
+
+	if (reset_mask & RADEON_RESET_DMA) {
+		/* dma0 */
+		tmp = RREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET, tmp);
+	}
+
+	if (reset_mask & RADEON_RESET_DMA1) {
+		/* dma1 */
+		tmp = RREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET, tmp);
+	}
+
+	udelay(50);
+
 	evergreen_mc_stop(rdev, &save);
 	if (evergreen_mc_wait_for_idle(rdev)) {
 		dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
 	}
 
-	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE))
-		cayman_gpu_soft_reset_gfx(rdev);
+	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE)) {
+		grbm_soft_reset = SOFT_RESET_CB |
+			SOFT_RESET_DB |
+			SOFT_RESET_GDS |
+			SOFT_RESET_PA |
+			SOFT_RESET_SC |
+			SOFT_RESET_SPI |
+			SOFT_RESET_SH |
+			SOFT_RESET_SX |
+			SOFT_RESET_TC |
+			SOFT_RESET_TA |
+			SOFT_RESET_VGT |
+			SOFT_RESET_IA;
+	}
+
+	if (reset_mask & RADEON_RESET_CP) {
+		grbm_soft_reset |= SOFT_RESET_CP | SOFT_RESET_VGT;
+
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+	}
 
 	if (reset_mask & RADEON_RESET_DMA)
-		cayman_gpu_soft_reset_dma(rdev);
+		srbm_soft_reset |= SOFT_RESET_DMA;
+
+	if (reset_mask & RADEON_RESET_DMA1)
+		srbm_soft_reset |= SOFT_RESET_DMA1;
+
+	if (reset_mask & RADEON_RESET_DISPLAY)
+		srbm_soft_reset |= SOFT_RESET_DC;
+
+	if (reset_mask & RADEON_RESET_RLC)
+		srbm_soft_reset |= SOFT_RESET_RLC;
+
+	if (reset_mask & RADEON_RESET_SEM)
+		srbm_soft_reset |= SOFT_RESET_SEM;
+
+	if (reset_mask & RADEON_RESET_IH)
+		srbm_soft_reset |= SOFT_RESET_IH;
+
+	if (reset_mask & RADEON_RESET_GRBM)
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+
+	if (reset_mask & RADEON_RESET_VMC)
+		srbm_soft_reset |= SOFT_RESET_VMC;
+
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		if (reset_mask & RADEON_RESET_MC)
+			srbm_soft_reset |= SOFT_RESET_MC;
+	}
+
+	if (grbm_soft_reset) {
+		tmp = RREG32(GRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(rdev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~grbm_soft_reset;
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+	}
+
+	if (srbm_soft_reset) {
+		tmp = RREG32(SRBM_SOFT_RESET);
+		tmp |= srbm_soft_reset;
+		dev_info(rdev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~srbm_soft_reset;
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+	}
 
 	/* Wait a little for things to settle down */
 	udelay(50);
 
 	evergreen_mc_resume(rdev, &save);
-	return 0;
+	udelay(50);
+
+	evergreen_print_gpu_status_regs(rdev);
 }
 
 int cayman_asic_reset(struct radeon_device *rdev)
 {
-	return cayman_gpu_soft_reset(rdev, (RADEON_RESET_GFX |
-					    RADEON_RESET_COMPUTE |
-					    RADEON_RESET_DMA));
+	u32 reset_mask;
+
+	reset_mask = cayman_gpu_check_soft_reset(rdev);
+
+	if (reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, true);
+
+	cayman_gpu_soft_reset(rdev, reset_mask);
+
+	reset_mask = cayman_gpu_check_soft_reset(rdev);
+
+	if (!reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, false);
+
+	return 0;
+}
+
+/**
+ * cayman_gfx_is_lockup - Check if the GFX engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the GFX engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool cayman_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 reset_mask = cayman_gpu_check_soft_reset(rdev);
+
+	if (!(reset_mask & (RADEON_RESET_GFX |
+			    RADEON_RESET_COMPUTE |
+			    RADEON_RESET_CP))) {
+		radeon_ring_lockup_update(ring);
+		return false;
+	}
+	/* force CP activities */
+	radeon_ring_force_activity(rdev, ring);
+	return radeon_ring_test_lockup(rdev, ring);
 }
 
 /**
@@ -1464,18 +1565,20 @@ int cayman_asic_reset(struct radeon_device *rdev)
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
  *
- * Check if the async DMA engine is locked up (cayman-SI).
+ * Check if the async DMA engine is locked up.
  * Returns true if the engine appears to be locked up, false if not.
  */
 bool cayman_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	u32 dma_status_reg;
+	u32 reset_mask = cayman_gpu_check_soft_reset(rdev);
+	u32 mask;
 
 	if (ring->idx == R600_RING_TYPE_DMA_INDEX)
-		dma_status_reg = RREG32(DMA_STATUS_REG + DMA0_REGISTER_OFFSET);
+		mask = RADEON_RESET_DMA;
 	else
-		dma_status_reg = RREG32(DMA_STATUS_REG + DMA1_REGISTER_OFFSET);
-	if (dma_status_reg & DMA_IDLE) {
+		mask = RADEON_RESET_DMA1;
+
+	if (!(reset_mask & mask)) {
 		radeon_ring_lockup_update(ring);
 		return false;
 	}
@@ -1843,19 +1946,21 @@ uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags)
  * cayman_vm_set_page - update the page tables using the CP
  *
  * @rdev: radeon_device pointer
+ * @ib: indirect buffer to fill with commands
  * @pe: addr of the page entry
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
  * @flags: access flags
  *
- * Update the page tables using the CP (cayman-si).
+ * Update the page tables using the CP (cayman/TN).
  */
-void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
+void cayman_vm_set_page(struct radeon_device *rdev,
+			struct radeon_ib *ib,
+			uint64_t pe,
 			uint64_t addr, unsigned count,
 			uint32_t incr, uint32_t flags)
 {
-	struct radeon_ring *ring = &rdev->ring[rdev->asic->vm.pt_ring_index];
 	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
 	uint64_t value;
 	unsigned ndw;
@@ -1866,9 +1971,9 @@ void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 			if (ndw > 0x3FFF)
 				ndw = 0x3FFF;
 
-			radeon_ring_write(ring, PACKET3(PACKET3_ME_WRITE, ndw));
-			radeon_ring_write(ring, pe);
-			radeon_ring_write(ring, upper_32_bits(pe) & 0xff);
+			ib->ptr[ib->length_dw++] = PACKET3(PACKET3_ME_WRITE, ndw);
+			ib->ptr[ib->length_dw++] = pe;
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
 			for (; ndw > 1; ndw -= 2, --count, pe += 8) {
 				if (flags & RADEON_VM_PAGE_SYSTEM) {
 					value = radeon_vm_map_gart(rdev, addr);
@@ -1880,8 +1985,8 @@ void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 				}
 				addr += incr;
 				value |= r600_flags;
-				radeon_ring_write(ring, value);
-				radeon_ring_write(ring, upper_32_bits(value));
+				ib->ptr[ib->length_dw++] = value;
+				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
 		}
 	} else {
@@ -1891,9 +1996,9 @@ void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 				ndw = 0xFFFFE;
 
 			/* for non-physically contiguous pages (system) */
-			radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 0, ndw));
-			radeon_ring_write(ring, pe);
-			radeon_ring_write(ring, upper_32_bits(pe) & 0xff);
+			ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_WRITE, 0, 0, ndw);
+			ib->ptr[ib->length_dw++] = pe;
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
 			for (; ndw > 0; ndw -= 2, --count, pe += 8) {
 				if (flags & RADEON_VM_PAGE_SYSTEM) {
 					value = radeon_vm_map_gart(rdev, addr);
@@ -1905,10 +2010,12 @@ void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 				}
 				addr += incr;
 				value |= r600_flags;
-				radeon_ring_write(ring, value);
-				radeon_ring_write(ring, upper_32_bits(value));
+				ib->ptr[ib->length_dw++] = value;
+				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
 		}
+		while (ib->length_dw & 0x7)
+			ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_NOP, 0, 0, 0);
 	}
 }
 
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index 48e5022ee921..079dee202a9e 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -49,6 +49,16 @@
 #define		RINGID(x)					(((x) & 0x3) << 0)
 #define		VMID(x)						(((x) & 0x7) << 0)
 #define	SRBM_STATUS				        0x0E50
+#define		RLC_RQ_PENDING 				(1 << 3)
+#define		GRBM_RQ_PENDING 			(1 << 5)
+#define		VMC_BUSY 				(1 << 8)
+#define		MCB_BUSY 				(1 << 9)
+#define		MCB_NON_DISPLAY_BUSY 			(1 << 10)
+#define		MCC_BUSY 				(1 << 11)
+#define		MCD_BUSY 				(1 << 12)
+#define		SEM_BUSY 				(1 << 14)
+#define		RLC_BUSY 				(1 << 15)
+#define		IH_BUSY 				(1 << 17)
 
 #define	SRBM_SOFT_RESET				        0x0E60
 #define		SOFT_RESET_BIF				(1 << 1)
@@ -68,6 +78,10 @@
 #define		SOFT_RESET_REGBB			(1 << 22)
 #define		SOFT_RESET_ORB				(1 << 23)
 
+#define	SRBM_STATUS2				        0x0EC4
+#define		DMA_BUSY 				(1 << 5)
+#define		DMA1_BUSY 				(1 << 6)
+
 #define VM_CONTEXT0_REQUEST_RESPONSE			0x1470
 #define		REQUEST_TYPE(x)					(((x) & 0xf) << 0)
 #define		RESPONSE_TYPE_MASK				0x000000F0
@@ -474,16 +488,7 @@
 /*
  * PM4
  */
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0xFFFF) << 2)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-#define PACKET0(reg, n)	((PACKET_TYPE0 << 30) |				\
+#define PACKET0(reg, n)	((RADEON_PACKET_TYPE0 << 30) |			\
 			 (((reg) >> 2) & 0xFFFF) |			\
 			 ((n) & 0x3FFF) << 16)
 #define CP_PACKET2			0x80000000
@@ -492,7 +497,7 @@
 
 #define PACKET2(v)	(CP_PACKET2 | REG_SET(PACKET2_PAD, (v)))
 
-#define PACKET3(op, n)	((PACKET_TYPE3 << 30) |				\
+#define PACKET3(op, n)	((RADEON_PACKET_TYPE3 << 30) |			\
 			 (((op) & 0xFF) << 8) |				\
 			 ((n) & 0x3FFF) << 16)
 
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 8ff7cac222dc..9db58530be37 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1215,11 +1215,11 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
 	struct radeon_cs_reloc *reloc;
 	u32 value;
 
-	r = r100_cs_packet_next_reloc(p, &reloc);
+	r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 	if (r) {
 		DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 			  idx, reg);
-		r100_cs_dump_packet(p, pkt);
+		radeon_cs_dump_packet(p, pkt);
 		return r;
 	}
 
@@ -1233,7 +1233,7 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
 		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
 			if (reg == RADEON_SRC_PITCH_OFFSET) {
 				DRM_ERROR("Cannot src blit from microtiled surface\n");
-				r100_cs_dump_packet(p, pkt);
+				radeon_cs_dump_packet(p, pkt);
 				return -EINVAL;
 			}
 			tile_flags |= RADEON_DST_TILE_MICRO;
@@ -1263,16 +1263,16 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 	if (c > 16) {
 	    DRM_ERROR("Only 16 vertex buffers are allowed %d\n",
 		      pkt->opcode);
-	    r100_cs_dump_packet(p, pkt);
+	    radeon_cs_dump_packet(p, pkt);
 	    return -EINVAL;
 	}
 	track->num_arrays = c;
 	for (i = 0; i < (c - 1); i+=2, idx+=3) {
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n",
 				  pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		idx_value = radeon_get_ib_value(p, idx);
@@ -1281,11 +1281,11 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 		track->arrays[i + 0].esize = idx_value >> 8;
 		track->arrays[i + 0].robj = reloc->robj;
 		track->arrays[i + 0].esize &= 0x7F;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n",
 				  pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx+2] = radeon_get_ib_value(p, idx + 2) + ((u32)reloc->lobj.gpu_offset);
@@ -1294,11 +1294,11 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 		track->arrays[i + 1].esize &= 0x7F;
 	}
 	if (c & 1) {
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n",
 					  pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		idx_value = radeon_get_ib_value(p, idx);
@@ -1355,67 +1355,6 @@ int r100_cs_parse_packet0(struct radeon_cs_parser *p,
 	return 0;
 }
 
-void r100_cs_dump_packet(struct radeon_cs_parser *p,
-			 struct radeon_cs_packet *pkt)
-{
-	volatile uint32_t *ib;
-	unsigned i;
-	unsigned idx;
-
-	ib = p->ib.ptr;
-	idx = pkt->idx;
-	for (i = 0; i <= (pkt->count + 1); i++, idx++) {
-		DRM_INFO("ib[%d]=0x%08X\n", idx, ib[idx]);
-	}
-}
-
-/**
- * r100_cs_packet_parse() - parse cp packet and point ib index to next packet
- * @parser:	parser structure holding parsing context.
- * @pkt:	where to store packet informations
- *
- * Assume that chunk_ib_index is properly set. Will return -EINVAL
- * if packet is bigger than remaining ib size. or if packets is unknown.
- **/
-int r100_cs_packet_parse(struct radeon_cs_parser *p,
-			 struct radeon_cs_packet *pkt,
-			 unsigned idx)
-{
-	struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx];
-	uint32_t header;
-
-	if (idx >= ib_chunk->length_dw) {
-		DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
-			  idx, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	header = radeon_get_ib_value(p, idx);
-	pkt->idx = idx;
-	pkt->type = CP_PACKET_GET_TYPE(header);
-	pkt->count = CP_PACKET_GET_COUNT(header);
-	switch (pkt->type) {
-	case PACKET_TYPE0:
-		pkt->reg = CP_PACKET0_GET_REG(header);
-		pkt->one_reg_wr = CP_PACKET0_GET_ONE_REG_WR(header);
-		break;
-	case PACKET_TYPE3:
-		pkt->opcode = CP_PACKET3_GET_OPCODE(header);
-		break;
-	case PACKET_TYPE2:
-		pkt->count = -1;
-		break;
-	default:
-		DRM_ERROR("Unknown packet type %d at %d !\n", pkt->type, idx);
-		return -EINVAL;
-	}
-	if ((pkt->count + 1 + pkt->idx) >= ib_chunk->length_dw) {
-		DRM_ERROR("Packet (%d:%d:%d) end after CS buffer (%d) !\n",
-			  pkt->idx, pkt->type, pkt->count, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	return 0;
-}
-
 /**
  * r100_cs_packet_next_vline() - parse userspace VLINE packet
  * @parser:		parser structure holding parsing context.
@@ -1444,7 +1383,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	ib = p->ib.ptr;
 
 	/* parse the wait until */
-	r = r100_cs_packet_parse(p, &waitreloc, p->idx);
+	r = radeon_cs_packet_parse(p, &waitreloc, p->idx);
 	if (r)
 		return r;
 
@@ -1461,7 +1400,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	}
 
 	/* jump over the NOP */
-	r = r100_cs_packet_parse(p, &p3reloc, p->idx + waitreloc.count + 2);
+	r = radeon_cs_packet_parse(p, &p3reloc, p->idx + waitreloc.count + 2);
 	if (r)
 		return r;
 
@@ -1471,7 +1410,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 
 	header = radeon_get_ib_value(p, h_idx);
 	crtc_id = radeon_get_ib_value(p, h_idx + 5);
-	reg = CP_PACKET0_GET_REG(header);
+	reg = R100_CP_PACKET0_GET_REG(header);
 	obj = drm_mode_object_find(p->rdev->ddev, crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_ERROR("cannot find crtc %d\n", crtc_id);
@@ -1506,54 +1445,6 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	return 0;
 }
 
-/**
- * r100_cs_packet_next_reloc() - parse next packet which should be reloc packet3
- * @parser:		parser structure holding parsing context.
- * @data:		pointer to relocation data
- * @offset_start:	starting offset
- * @offset_mask:	offset mask (to align start offset on)
- * @reloc:		reloc informations
- *
- * Check next packet is relocation packet3, do bo validation and compute
- * GPU offset using the provided start.
- **/
-int r100_cs_packet_next_reloc(struct radeon_cs_parser *p,
-			      struct radeon_cs_reloc **cs_reloc)
-{
-	struct radeon_cs_chunk *relocs_chunk;
-	struct radeon_cs_packet p3reloc;
-	unsigned idx;
-	int r;
-
-	if (p->chunk_relocs_idx == -1) {
-		DRM_ERROR("No relocation chunk !\n");
-		return -EINVAL;
-	}
-	*cs_reloc = NULL;
-	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
-	r = r100_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return r;
-	}
-	p->idx += p3reloc.count + 2;
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		DRM_ERROR("No packet3 for relocation for packet at %d.\n",
-			  p3reloc.idx);
-		r100_cs_dump_packet(p, &p3reloc);
-		return -EINVAL;
-	}
-	idx = radeon_get_ib_value(p, p3reloc.idx + 1);
-	if (idx >= relocs_chunk->length_dw) {
-		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
-			  idx, relocs_chunk->length_dw);
-		r100_cs_dump_packet(p, &p3reloc);
-		return -EINVAL;
-	}
-	/* FIXME: we assume reloc size is 4 dwords */
-	*cs_reloc = p->relocs_ptr[(idx / 4)];
-	return 0;
-}
-
 static int r100_get_vtx_size(uint32_t vtx_fmt)
 {
 	int vtx_size;
@@ -1631,7 +1522,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		break;
@@ -1644,11 +1535,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 			return r;
 		break;
 	case RADEON_RB3D_DEPTHOFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->zb.robj = reloc->robj;
@@ -1657,11 +1548,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
 		break;
 	case RADEON_RB3D_COLOROFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->cb[0].robj = reloc->robj;
@@ -1673,11 +1564,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_PP_TXOFFSET_1:
 	case RADEON_PP_TXOFFSET_2:
 		i = (reg - RADEON_PP_TXOFFSET_0) / 24;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -1700,11 +1591,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_PP_CUBIC_OFFSET_T0_3:
 	case RADEON_PP_CUBIC_OFFSET_T0_4:
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T0_0) / 4;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->textures[0].cube_info[i].offset = idx_value;
@@ -1718,11 +1609,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_PP_CUBIC_OFFSET_T1_3:
 	case RADEON_PP_CUBIC_OFFSET_T1_4:
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T1_0) / 4;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->textures[1].cube_info[i].offset = idx_value;
@@ -1736,11 +1627,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_PP_CUBIC_OFFSET_T2_3:
 	case RADEON_PP_CUBIC_OFFSET_T2_4:
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T2_0) / 4;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->textures[2].cube_info[i].offset = idx_value;
@@ -1754,11 +1645,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		track->zb_dirty = true;
 		break;
 	case RADEON_RB3D_COLORPITCH:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -1825,11 +1716,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		track->zb_dirty = true;
 		break;
 	case RADEON_RB3D_ZPASS_ADDR:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
@@ -1986,10 +1877,10 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 			return r;
 		break;
 	case PACKET3_INDX_BUFFER:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx+1] = radeon_get_ib_value(p, idx+1) + ((u32)reloc->lobj.gpu_offset);
@@ -2000,10 +1891,10 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case 0x23:
 		/* 3D_RNDR_GEN_INDX_PRIM on r100/r200 */
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx] = radeon_get_ib_value(p, idx) + ((u32)reloc->lobj.gpu_offset);
@@ -2100,37 +1991,36 @@ int r100_cs_parse(struct radeon_cs_parser *p)
 	r100_cs_track_clear(p->rdev, track);
 	p->track = track;
 	do {
-		r = r100_cs_packet_parse(p, &pkt, p->idx);
+		r = radeon_cs_packet_parse(p, &pkt, p->idx);
 		if (r) {
 			return r;
 		}
 		p->idx += pkt.count + 2;
 		switch (pkt.type) {
-			case PACKET_TYPE0:
-				if (p->rdev->family >= CHIP_R200)
-					r = r100_cs_parse_packet0(p, &pkt,
-								  p->rdev->config.r100.reg_safe_bm,
-								  p->rdev->config.r100.reg_safe_bm_size,
-								  &r200_packet0_check);
-				else
-					r = r100_cs_parse_packet0(p, &pkt,
-								  p->rdev->config.r100.reg_safe_bm,
-								  p->rdev->config.r100.reg_safe_bm_size,
-								  &r100_packet0_check);
-				break;
-			case PACKET_TYPE2:
-				break;
-			case PACKET_TYPE3:
-				r = r100_packet3_check(p, &pkt);
-				break;
-			default:
-				DRM_ERROR("Unknown packet type %d !\n",
-					  pkt.type);
-				return -EINVAL;
+		case RADEON_PACKET_TYPE0:
+			if (p->rdev->family >= CHIP_R200)
+				r = r100_cs_parse_packet0(p, &pkt,
+					p->rdev->config.r100.reg_safe_bm,
+					p->rdev->config.r100.reg_safe_bm_size,
+					&r200_packet0_check);
+			else
+				r = r100_cs_parse_packet0(p, &pkt,
+					p->rdev->config.r100.reg_safe_bm,
+					p->rdev->config.r100.reg_safe_bm_size,
+					&r100_packet0_check);
+			break;
+		case RADEON_PACKET_TYPE2:
+			break;
+		case RADEON_PACKET_TYPE3:
+			r = r100_packet3_check(p, &pkt);
+			break;
+		default:
+			DRM_ERROR("Unknown packet type %d !\n",
+				  pkt.type);
+			return -EINVAL;
 		}
-		if (r) {
+		if (r)
 			return r;
-		}
 	} while (p->idx < p->chunks[p->chunk_ib_idx].length_dw);
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/r100_track.h b/drivers/gpu/drm/radeon/r100_track.h
index 6a603b378adb..eb40888bdfcc 100644
--- a/drivers/gpu/drm/radeon/r100_track.h
+++ b/drivers/gpu/drm/radeon/r100_track.h
@@ -81,10 +81,6 @@ struct r100_cs_track {
 
 int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track);
 void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track);
-int r100_cs_packet_next_reloc(struct radeon_cs_parser *p,
-			      struct radeon_cs_reloc **cs_reloc);
-void r100_cs_dump_packet(struct radeon_cs_parser *p,
-			 struct radeon_cs_packet *pkt);
 
 int r100_cs_packet_parse_vline(struct radeon_cs_parser *p);
 
diff --git a/drivers/gpu/drm/radeon/r100d.h b/drivers/gpu/drm/radeon/r100d.h
index eab91760fae0..f0f8ee69f480 100644
--- a/drivers/gpu/drm/radeon/r100d.h
+++ b/drivers/gpu/drm/radeon/r100d.h
@@ -64,17 +64,6 @@
 			 REG_SET(PACKET3_IT_OPCODE, (op)) |		\
 			 REG_SET(PACKET3_COUNT, (n)))
 
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0x1FFF) << 2)
-#define CP_PACKET0_GET_ONE_REG_WR(h) (((h) >> 15) & 1)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-
 /* Registers */
 #define R_0000F0_RBBM_SOFT_RESET                     0x0000F0
 #define   S_0000F0_SOFT_RESET_CP(x)                    (((x) & 0x1) << 0)
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index 98143a5c5b73..b3807edb1936 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -162,7 +162,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		break;
@@ -175,11 +175,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 			return r;
 		break;
 	case RADEON_RB3D_DEPTHOFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->zb.robj = reloc->robj;
@@ -188,11 +188,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
 		break;
 	case RADEON_RB3D_COLOROFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->cb[0].robj = reloc->robj;
@@ -207,11 +207,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case R200_PP_TXOFFSET_4:
 	case R200_PP_TXOFFSET_5:
 		i = (reg - R200_PP_TXOFFSET_0) / 24;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -260,11 +260,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case R200_PP_CUBIC_OFFSET_F5_5:
 		i = (reg - R200_PP_TXOFFSET_0) / 24;
 		face = (reg - ((i * 24) + R200_PP_TXOFFSET_0)) / 4;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->textures[i].cube_info[face - 1].offset = idx_value;
@@ -278,11 +278,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		track->zb_dirty = true;
 		break;
 	case RADEON_RB3D_COLORPITCH:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 
@@ -355,11 +355,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		track->zb_dirty = true;
 		break;
 	case RADEON_RB3D_ZPASS_ADDR:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index d0ba6023a1f8..c60350e6872d 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -615,7 +615,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		break;
@@ -630,11 +630,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case R300_RB3D_COLOROFFSET2:
 	case R300_RB3D_COLOROFFSET3:
 		i = (reg - R300_RB3D_COLOROFFSET0) >> 2;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->cb[i].robj = reloc->robj;
@@ -643,11 +643,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
 		break;
 	case R300_ZB_DEPTHOFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->zb.robj = reloc->robj;
@@ -672,11 +672,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case R300_TX_OFFSET_0+56:
 	case R300_TX_OFFSET_0+60:
 		i = (reg - R300_TX_OFFSET_0) >> 2;
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 
@@ -745,11 +745,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		/* RB3D_COLORPITCH2 */
 		/* RB3D_COLORPITCH3 */
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-			r = r100_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					  idx, reg);
-				r100_cs_dump_packet(p, pkt);
+				radeon_cs_dump_packet(p, pkt);
 				return r;
 			}
 
@@ -830,11 +830,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case 0x4F24:
 		/* ZB_DEPTHPITCH */
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-			r = r100_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
 				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					  idx, reg);
-				r100_cs_dump_packet(p, pkt);
+				radeon_cs_dump_packet(p, pkt);
 				return r;
 			}
 
@@ -1045,11 +1045,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		track->tex_dirty = true;
 		break;
 	case R300_ZB_ZPASS_ADDR:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 					idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
@@ -1087,11 +1087,11 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		track->cb_dirty = true;
 		break;
 	case R300_RB3D_AARESOLVE_OFFSET:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		track->aa.robj = reloc->robj;
@@ -1156,10 +1156,10 @@ static int r300_packet3_check(struct radeon_cs_parser *p,
 			return r;
 		break;
 	case PACKET3_INDX_BUFFER:
-		r = r100_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
 			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
-			r100_cs_dump_packet(p, pkt);
+			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
 		ib[idx+1] = radeon_get_ib_value(p, idx + 1) + ((u32)reloc->lobj.gpu_offset);
@@ -1257,21 +1257,21 @@ int r300_cs_parse(struct radeon_cs_parser *p)
 	r100_cs_track_clear(p->rdev, track);
 	p->track = track;
 	do {
-		r = r100_cs_packet_parse(p, &pkt, p->idx);
+		r = radeon_cs_packet_parse(p, &pkt, p->idx);
 		if (r) {
 			return r;
 		}
 		p->idx += pkt.count + 2;
 		switch (pkt.type) {
-		case PACKET_TYPE0:
+		case RADEON_PACKET_TYPE0:
 			r = r100_cs_parse_packet0(p, &pkt,
 						  p->rdev->config.r300.reg_safe_bm,
 						  p->rdev->config.r300.reg_safe_bm_size,
 						  &r300_packet0_check);
 			break;
-		case PACKET_TYPE2:
+		case RADEON_PACKET_TYPE2:
 			break;
-		case PACKET_TYPE3:
+		case RADEON_PACKET_TYPE3:
 			r = r300_packet3_check(p, &pkt);
 			break;
 		default:
diff --git a/drivers/gpu/drm/radeon/r300_cmdbuf.c b/drivers/gpu/drm/radeon/r300_cmdbuf.c
index 002ab038d2ab..865e2c9980db 100644
--- a/drivers/gpu/drm/radeon/r300_cmdbuf.c
+++ b/drivers/gpu/drm/radeon/r300_cmdbuf.c
@@ -29,6 +29,8 @@
  *
  * Authors:
  *    Nicolai Haehnle <prefect_@gmx.net>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <drm/drmP.h>
diff --git a/drivers/gpu/drm/radeon/r300d.h b/drivers/gpu/drm/radeon/r300d.h
index 1f519a5ffb8c..ff229a00d273 100644
--- a/drivers/gpu/drm/radeon/r300d.h
+++ b/drivers/gpu/drm/radeon/r300d.h
@@ -65,17 +65,6 @@
 			 REG_SET(PACKET3_IT_OPCODE, (op)) |		\
 			 REG_SET(PACKET3_COUNT, (n)))
 
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0x1FFF) << 2)
-#define CP_PACKET0_GET_ONE_REG_WR(h) (((h) >> 15) & 1)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-
 /* Registers */
 #define R_000148_MC_FB_LOCATION                      0x000148
 #define   S_000148_MC_FB_START(x)                      (((x) & 0xFFFF) << 0)
diff --git a/drivers/gpu/drm/radeon/r500_reg.h b/drivers/gpu/drm/radeon/r500_reg.h
index ec576aaafb73..c0dc8d3ba0bb 100644
--- a/drivers/gpu/drm/radeon/r500_reg.h
+++ b/drivers/gpu/drm/radeon/r500_reg.h
@@ -355,6 +355,7 @@
 #       define AVIVO_D1CRTC_V_BLANK                             (1 << 0)
 #define AVIVO_D1CRTC_STATUS_POSITION                            0x60a0
 #define AVIVO_D1CRTC_FRAME_COUNT                                0x60a4
+#define AVIVO_D1CRTC_STATUS_HV_COUNT                            0x60ac
 #define AVIVO_D1CRTC_STEREO_CONTROL                             0x60c4
 
 #define AVIVO_D1MODE_MASTER_UPDATE_MODE                         0x60e4
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index becb03e8b32f..6d4b5611daf4 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -94,6 +94,12 @@ MODULE_FIRMWARE("radeon/SUMO_me.bin");
 MODULE_FIRMWARE("radeon/SUMO2_pfp.bin");
 MODULE_FIRMWARE("radeon/SUMO2_me.bin");
 
+static const u32 crtc_offsets[2] =
+{
+	0,
+	AVIVO_D2CRTC_H_TOTAL - AVIVO_D1CRTC_H_TOTAL
+};
+
 int r600_debugfs_mc_info_init(struct radeon_device *rdev);
 
 /* r600,rv610,rv630,rv620,rv635,rv670 */
@@ -103,6 +109,19 @@ void r600_fini(struct radeon_device *rdev);
 void r600_irq_disable(struct radeon_device *rdev);
 static void r600_pcie_gen2_enable(struct radeon_device *rdev);
 
+/**
+ * r600_get_xclk - get the xclk
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Returns the reference clock used by the gfx engine
+ * (r6xx, IGPs, APUs).
+ */
+u32 r600_get_xclk(struct radeon_device *rdev)
+{
+	return rdev->clock.spll.reference_freq;
+}
+
 /* get temperature in millidegrees */
 int rv6xx_get_temp(struct radeon_device *rdev)
 {
@@ -1254,169 +1273,301 @@ void r600_vram_scratch_fini(struct radeon_device *rdev)
 	radeon_bo_unref(&rdev->vram_scratch.robj);
 }
 
-/* We doesn't check that the GPU really needs a reset we simply do the
- * reset, it's up to the caller to determine if the GPU needs one. We
- * might add an helper function to check that.
- */
-static void r600_gpu_soft_reset_gfx(struct radeon_device *rdev)
+void r600_set_bios_scratch_engine_hung(struct radeon_device *rdev, bool hung)
 {
-	u32 grbm_busy_mask = S_008010_VC_BUSY(1) | S_008010_VGT_BUSY_NO_DMA(1) |
-				S_008010_VGT_BUSY(1) | S_008010_TA03_BUSY(1) |
-				S_008010_TC_BUSY(1) | S_008010_SX_BUSY(1) |
-				S_008010_SH_BUSY(1) | S_008010_SPI03_BUSY(1) |
-				S_008010_SMX_BUSY(1) | S_008010_SC_BUSY(1) |
-				S_008010_PA_BUSY(1) | S_008010_DB03_BUSY(1) |
-				S_008010_CR_BUSY(1) | S_008010_CB03_BUSY(1) |
-				S_008010_GUI_ACTIVE(1);
-	u32 grbm2_busy_mask = S_008014_SPI0_BUSY(1) | S_008014_SPI1_BUSY(1) |
-			S_008014_SPI2_BUSY(1) | S_008014_SPI3_BUSY(1) |
-			S_008014_TA0_BUSY(1) | S_008014_TA1_BUSY(1) |
-			S_008014_TA2_BUSY(1) | S_008014_TA3_BUSY(1) |
-			S_008014_DB0_BUSY(1) | S_008014_DB1_BUSY(1) |
-			S_008014_DB2_BUSY(1) | S_008014_DB3_BUSY(1) |
-			S_008014_CB0_BUSY(1) | S_008014_CB1_BUSY(1) |
-			S_008014_CB2_BUSY(1) | S_008014_CB3_BUSY(1);
-	u32 tmp;
+	u32 tmp = RREG32(R600_BIOS_3_SCRATCH);
 
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		return;
+	if (hung)
+		tmp |= ATOM_S3_ASIC_GUI_ENGINE_HUNG;
+	else
+		tmp &= ~ATOM_S3_ASIC_GUI_ENGINE_HUNG;
+
+	WREG32(R600_BIOS_3_SCRATCH, tmp);
+}
 
+static void r600_print_gpu_status_regs(struct radeon_device *rdev)
+{
 	dev_info(rdev->dev, "  R_008010_GRBM_STATUS      = 0x%08X\n",
-		RREG32(R_008010_GRBM_STATUS));
+		 RREG32(R_008010_GRBM_STATUS));
 	dev_info(rdev->dev, "  R_008014_GRBM_STATUS2     = 0x%08X\n",
-		RREG32(R_008014_GRBM_STATUS2));
+		 RREG32(R_008014_GRBM_STATUS2));
 	dev_info(rdev->dev, "  R_000E50_SRBM_STATUS      = 0x%08X\n",
-		RREG32(R_000E50_SRBM_STATUS));
+		 RREG32(R_000E50_SRBM_STATUS));
 	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT1));
+		 RREG32(CP_STALLED_STAT1));
 	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT2));
+		 RREG32(CP_STALLED_STAT2));
 	dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
-		RREG32(CP_BUSY_STAT));
+		 RREG32(CP_BUSY_STAT));
 	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
-		RREG32(CP_STAT));
-
-	/* Disable CP parsing/prefetching */
-	WREG32(R_0086D8_CP_ME_CNTL, S_0086D8_CP_ME_HALT(1));
+		 RREG32(CP_STAT));
+	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
+		RREG32(DMA_STATUS_REG));
+}
 
-	/* Check if any of the rendering block is busy and reset it */
-	if ((RREG32(R_008010_GRBM_STATUS) & grbm_busy_mask) ||
-	    (RREG32(R_008014_GRBM_STATUS2) & grbm2_busy_mask)) {
-		tmp = S_008020_SOFT_RESET_CR(1) |
-			S_008020_SOFT_RESET_DB(1) |
-			S_008020_SOFT_RESET_CB(1) |
-			S_008020_SOFT_RESET_PA(1) |
-			S_008020_SOFT_RESET_SC(1) |
-			S_008020_SOFT_RESET_SMX(1) |
-			S_008020_SOFT_RESET_SPI(1) |
-			S_008020_SOFT_RESET_SX(1) |
-			S_008020_SOFT_RESET_SH(1) |
-			S_008020_SOFT_RESET_TC(1) |
-			S_008020_SOFT_RESET_TA(1) |
-			S_008020_SOFT_RESET_VC(1) |
-			S_008020_SOFT_RESET_VGT(1);
-		dev_info(rdev->dev, "  R_008020_GRBM_SOFT_RESET=0x%08X\n", tmp);
-		WREG32(R_008020_GRBM_SOFT_RESET, tmp);
-		RREG32(R_008020_GRBM_SOFT_RESET);
-		mdelay(15);
-		WREG32(R_008020_GRBM_SOFT_RESET, 0);
+static bool r600_is_display_hung(struct radeon_device *rdev)
+{
+	u32 crtc_hung = 0;
+	u32 crtc_status[2];
+	u32 i, j, tmp;
+
+	for (i = 0; i < rdev->num_crtc; i++) {
+		if (RREG32(AVIVO_D1CRTC_CONTROL + crtc_offsets[i]) & AVIVO_CRTC_EN) {
+			crtc_status[i] = RREG32(AVIVO_D1CRTC_STATUS_HV_COUNT + crtc_offsets[i]);
+			crtc_hung |= (1 << i);
+		}
 	}
-	/* Reset CP (we always reset CP) */
-	tmp = S_008020_SOFT_RESET_CP(1);
-	dev_info(rdev->dev, "R_008020_GRBM_SOFT_RESET=0x%08X\n", tmp);
-	WREG32(R_008020_GRBM_SOFT_RESET, tmp);
-	RREG32(R_008020_GRBM_SOFT_RESET);
-	mdelay(15);
-	WREG32(R_008020_GRBM_SOFT_RESET, 0);
 
-	dev_info(rdev->dev, "  R_008010_GRBM_STATUS      = 0x%08X\n",
-		RREG32(R_008010_GRBM_STATUS));
-	dev_info(rdev->dev, "  R_008014_GRBM_STATUS2     = 0x%08X\n",
-		RREG32(R_008014_GRBM_STATUS2));
-	dev_info(rdev->dev, "  R_000E50_SRBM_STATUS      = 0x%08X\n",
-		RREG32(R_000E50_SRBM_STATUS));
-	dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT1));
-	dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
-		RREG32(CP_STALLED_STAT2));
-	dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
-		RREG32(CP_BUSY_STAT));
-	dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
-		RREG32(CP_STAT));
+	for (j = 0; j < 10; j++) {
+		for (i = 0; i < rdev->num_crtc; i++) {
+			if (crtc_hung & (1 << i)) {
+				tmp = RREG32(AVIVO_D1CRTC_STATUS_HV_COUNT + crtc_offsets[i]);
+				if (tmp != crtc_status[i])
+					crtc_hung &= ~(1 << i);
+			}
+		}
+		if (crtc_hung == 0)
+			return false;
+		udelay(100);
+	}
 
+	return true;
 }
 
-static void r600_gpu_soft_reset_dma(struct radeon_device *rdev)
+static u32 r600_gpu_check_soft_reset(struct radeon_device *rdev)
 {
+	u32 reset_mask = 0;
 	u32 tmp;
 
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		return;
+	/* GRBM_STATUS */
+	tmp = RREG32(R_008010_GRBM_STATUS);
+	if (rdev->family >= CHIP_RV770) {
+		if (G_008010_PA_BUSY(tmp) | G_008010_SC_BUSY(tmp) |
+		    G_008010_SH_BUSY(tmp) | G_008010_SX_BUSY(tmp) |
+		    G_008010_TA_BUSY(tmp) | G_008010_VGT_BUSY(tmp) |
+		    G_008010_DB03_BUSY(tmp) | G_008010_CB03_BUSY(tmp) |
+		    G_008010_SPI03_BUSY(tmp) | G_008010_VGT_BUSY_NO_DMA(tmp))
+			reset_mask |= RADEON_RESET_GFX;
+	} else {
+		if (G_008010_PA_BUSY(tmp) | G_008010_SC_BUSY(tmp) |
+		    G_008010_SH_BUSY(tmp) | G_008010_SX_BUSY(tmp) |
+		    G_008010_TA03_BUSY(tmp) | G_008010_VGT_BUSY(tmp) |
+		    G_008010_DB03_BUSY(tmp) | G_008010_CB03_BUSY(tmp) |
+		    G_008010_SPI03_BUSY(tmp) | G_008010_VGT_BUSY_NO_DMA(tmp))
+			reset_mask |= RADEON_RESET_GFX;
+	}
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	if (G_008010_CF_RQ_PENDING(tmp) | G_008010_PF_RQ_PENDING(tmp) |
+	    G_008010_CP_BUSY(tmp) | G_008010_CP_COHERENCY_BUSY(tmp))
+		reset_mask |= RADEON_RESET_CP;
 
-	/* Disable DMA */
-	tmp = RREG32(DMA_RB_CNTL);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL, tmp);
+	if (G_008010_GRBM_EE_BUSY(tmp))
+		reset_mask |= RADEON_RESET_GRBM | RADEON_RESET_GFX | RADEON_RESET_CP;
 
-	/* Reset dma */
-	if (rdev->family >= CHIP_RV770)
-		WREG32(SRBM_SOFT_RESET, RV770_SOFT_RESET_DMA);
-	else
-		WREG32(SRBM_SOFT_RESET, SOFT_RESET_DMA);
-	RREG32(SRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(SRBM_SOFT_RESET, 0);
+	/* DMA_STATUS_REG */
+	tmp = RREG32(DMA_STATUS_REG);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA;
 
-	dev_info(rdev->dev, "  R_00D034_DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	/* SRBM_STATUS */
+	tmp = RREG32(R_000E50_SRBM_STATUS);
+	if (G_000E50_RLC_RQ_PENDING(tmp) | G_000E50_RLC_BUSY(tmp))
+		reset_mask |= RADEON_RESET_RLC;
+
+	if (G_000E50_IH_BUSY(tmp))
+		reset_mask |= RADEON_RESET_IH;
+
+	if (G_000E50_SEM_BUSY(tmp))
+		reset_mask |= RADEON_RESET_SEM;
+
+	if (G_000E50_GRBM_RQ_PENDING(tmp))
+		reset_mask |= RADEON_RESET_GRBM;
+
+	if (G_000E50_VMC_BUSY(tmp))
+		reset_mask |= RADEON_RESET_VMC;
+
+	if (G_000E50_MCB_BUSY(tmp) | G_000E50_MCDZ_BUSY(tmp) |
+	    G_000E50_MCDY_BUSY(tmp) | G_000E50_MCDX_BUSY(tmp) |
+	    G_000E50_MCDW_BUSY(tmp))
+		reset_mask |= RADEON_RESET_MC;
+
+	if (r600_is_display_hung(rdev))
+		reset_mask |= RADEON_RESET_DISPLAY;
+
+	return reset_mask;
 }
 
-static int r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
+static void r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
 {
 	struct rv515_mc_save save;
-
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		reset_mask &= ~(RADEON_RESET_GFX | RADEON_RESET_COMPUTE);
-
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		reset_mask &= ~RADEON_RESET_DMA;
+	u32 grbm_soft_reset = 0, srbm_soft_reset = 0;
+	u32 tmp;
 
 	if (reset_mask == 0)
-		return 0;
+		return;
 
 	dev_info(rdev->dev, "GPU softreset: 0x%08X\n", reset_mask);
 
+	r600_print_gpu_status_regs(rdev);
+
+	/* Disable CP parsing/prefetching */
+	if (rdev->family >= CHIP_RV770)
+		WREG32(R_0086D8_CP_ME_CNTL, S_0086D8_CP_ME_HALT(1) | S_0086D8_CP_PFP_HALT(1));
+	else
+		WREG32(R_0086D8_CP_ME_CNTL, S_0086D8_CP_ME_HALT(1));
+
+	/* disable the RLC */
+	WREG32(RLC_CNTL, 0);
+
+	if (reset_mask & RADEON_RESET_DMA) {
+		/* Disable DMA */
+		tmp = RREG32(DMA_RB_CNTL);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL, tmp);
+	}
+
+	mdelay(50);
+
 	rv515_mc_stop(rdev, &save);
 	if (r600_mc_wait_for_idle(rdev)) {
 		dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
 	}
 
-	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE))
-		r600_gpu_soft_reset_gfx(rdev);
+	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE)) {
+		if (rdev->family >= CHIP_RV770)
+			grbm_soft_reset |= S_008020_SOFT_RESET_DB(1) |
+				S_008020_SOFT_RESET_CB(1) |
+				S_008020_SOFT_RESET_PA(1) |
+				S_008020_SOFT_RESET_SC(1) |
+				S_008020_SOFT_RESET_SPI(1) |
+				S_008020_SOFT_RESET_SX(1) |
+				S_008020_SOFT_RESET_SH(1) |
+				S_008020_SOFT_RESET_TC(1) |
+				S_008020_SOFT_RESET_TA(1) |
+				S_008020_SOFT_RESET_VC(1) |
+				S_008020_SOFT_RESET_VGT(1);
+		else
+			grbm_soft_reset |= S_008020_SOFT_RESET_CR(1) |
+				S_008020_SOFT_RESET_DB(1) |
+				S_008020_SOFT_RESET_CB(1) |
+				S_008020_SOFT_RESET_PA(1) |
+				S_008020_SOFT_RESET_SC(1) |
+				S_008020_SOFT_RESET_SMX(1) |
+				S_008020_SOFT_RESET_SPI(1) |
+				S_008020_SOFT_RESET_SX(1) |
+				S_008020_SOFT_RESET_SH(1) |
+				S_008020_SOFT_RESET_TC(1) |
+				S_008020_SOFT_RESET_TA(1) |
+				S_008020_SOFT_RESET_VC(1) |
+				S_008020_SOFT_RESET_VGT(1);
+	}
+
+	if (reset_mask & RADEON_RESET_CP) {
+		grbm_soft_reset |= S_008020_SOFT_RESET_CP(1) |
+			S_008020_SOFT_RESET_VGT(1);
+
+		srbm_soft_reset |= S_000E60_SOFT_RESET_GRBM(1);
+	}
+
+	if (reset_mask & RADEON_RESET_DMA) {
+		if (rdev->family >= CHIP_RV770)
+			srbm_soft_reset |= RV770_SOFT_RESET_DMA;
+		else
+			srbm_soft_reset |= SOFT_RESET_DMA;
+	}
+
+	if (reset_mask & RADEON_RESET_RLC)
+		srbm_soft_reset |= S_000E60_SOFT_RESET_RLC(1);
+
+	if (reset_mask & RADEON_RESET_SEM)
+		srbm_soft_reset |= S_000E60_SOFT_RESET_SEM(1);
+
+	if (reset_mask & RADEON_RESET_IH)
+		srbm_soft_reset |= S_000E60_SOFT_RESET_IH(1);
+
+	if (reset_mask & RADEON_RESET_GRBM)
+		srbm_soft_reset |= S_000E60_SOFT_RESET_GRBM(1);
+
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		if (reset_mask & RADEON_RESET_MC)
+			srbm_soft_reset |= S_000E60_SOFT_RESET_MC(1);
+	}
+
+	if (reset_mask & RADEON_RESET_VMC)
+		srbm_soft_reset |= S_000E60_SOFT_RESET_VMC(1);
+
+	if (grbm_soft_reset) {
+		tmp = RREG32(R_008020_GRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(rdev->dev, "R_008020_GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(R_008020_GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(R_008020_GRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~grbm_soft_reset;
+		WREG32(R_008020_GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(R_008020_GRBM_SOFT_RESET);
+	}
+
+	if (srbm_soft_reset) {
+		tmp = RREG32(SRBM_SOFT_RESET);
+		tmp |= srbm_soft_reset;
+		dev_info(rdev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+
+		udelay(50);
 
-	if (reset_mask & RADEON_RESET_DMA)
-		r600_gpu_soft_reset_dma(rdev);
+		tmp &= ~srbm_soft_reset;
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+	}
 
 	/* Wait a little for things to settle down */
 	mdelay(1);
 
 	rv515_mc_resume(rdev, &save);
+	udelay(50);
+
+	r600_print_gpu_status_regs(rdev);
+}
+
+int r600_asic_reset(struct radeon_device *rdev)
+{
+	u32 reset_mask;
+
+	reset_mask = r600_gpu_check_soft_reset(rdev);
+
+	if (reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, true);
+
+	r600_gpu_soft_reset(rdev, reset_mask);
+
+	reset_mask = r600_gpu_check_soft_reset(rdev);
+
+	if (!reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, false);
+
 	return 0;
 }
 
-bool r600_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+/**
+ * r600_gfx_is_lockup - Check if the GFX engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the GFX engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool r600_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	u32 srbm_status;
-	u32 grbm_status;
-	u32 grbm_status2;
-
-	srbm_status = RREG32(R_000E50_SRBM_STATUS);
-	grbm_status = RREG32(R_008010_GRBM_STATUS);
-	grbm_status2 = RREG32(R_008014_GRBM_STATUS2);
-	if (!G_008010_GUI_ACTIVE(grbm_status)) {
+	u32 reset_mask = r600_gpu_check_soft_reset(rdev);
+
+	if (!(reset_mask & (RADEON_RESET_GFX |
+			    RADEON_RESET_COMPUTE |
+			    RADEON_RESET_CP))) {
 		radeon_ring_lockup_update(ring);
 		return false;
 	}
@@ -1431,15 +1582,14 @@ bool r600_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
  *
- * Check if the async DMA engine is locked up (r6xx-evergreen).
+ * Check if the async DMA engine is locked up.
  * Returns true if the engine appears to be locked up, false if not.
  */
 bool r600_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	u32 dma_status_reg;
+	u32 reset_mask = r600_gpu_check_soft_reset(rdev);
 
-	dma_status_reg = RREG32(DMA_STATUS_REG);
-	if (dma_status_reg & DMA_IDLE) {
+	if (!(reset_mask & RADEON_RESET_DMA)) {
 		radeon_ring_lockup_update(ring);
 		return false;
 	}
@@ -1448,13 +1598,6 @@ bool r600_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 	return radeon_ring_test_lockup(rdev, ring);
 }
 
-int r600_asic_reset(struct radeon_device *rdev)
-{
-	return r600_gpu_soft_reset(rdev, (RADEON_RESET_GFX |
-					  RADEON_RESET_COMPUTE |
-					  RADEON_RESET_DMA));
-}
-
 u32 r6xx_remap_render_backend(struct radeon_device *rdev,
 			      u32 tiling_pipe_num,
 			      u32 max_rb_num,
@@ -4318,14 +4461,14 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev)
 }
 
 /**
- * r600_get_gpu_clock - return GPU clock counter snapshot
+ * r600_get_gpu_clock_counter - return GPU clock counter snapshot
  *
  * @rdev: radeon_device pointer
  *
  * Fetches a GPU clock counter snapshot (R6xx-cayman).
  * Returns the 64 bit clock counter snapshot.
  */
-uint64_t r600_get_gpu_clock(struct radeon_device *rdev)
+uint64_t r600_get_gpu_clock_counter(struct radeon_device *rdev)
 {
 	uint64_t clock;
 
diff --git a/drivers/gpu/drm/radeon/r600_blit.c b/drivers/gpu/drm/radeon/r600_blit.c
index 77da1f9c0b8e..f651881eb0ae 100644
--- a/drivers/gpu/drm/radeon/r600_blit.c
+++ b/drivers/gpu/drm/radeon/r600_blit.c
@@ -22,6 +22,8 @@
  *
  * Authors:
  *     Alex Deucher <alexander.deucher@amd.com>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 #include <drm/drmP.h>
 #include <drm/radeon_drm.h>
@@ -488,37 +490,6 @@ set_default_state(drm_radeon_private_t *dev_priv)
 	ADVANCE_RING();
 }
 
-/* 23 bits of float fractional data */
-#define I2F_FRAC_BITS  23
-#define I2F_MASK ((1 << I2F_FRAC_BITS) - 1)
-
-/*
- * Converts unsigned integer into 32-bit IEEE floating point representation.
- * Will be exact from 0 to 2^24.  Above that, we round towards zero
- * as the fractional bits will not fit in a float.  (It would be better to
- * round towards even as the fpu does, but that is slower.)
- */
-__pure uint32_t int2float(uint32_t x)
-{
-	uint32_t msb, exponent, fraction;
-
-	/* Zero is special */
-	if (!x) return 0;
-
-	/* Get location of the most significant bit */
-	msb = __fls(x);
-
-	/*
-	 * Use a rotate instead of a shift because that works both leftwards
-	 * and rightwards due to the mod(32) behaviour.  This means we don't
-	 * need to check to see if we are above 2^24 or not.
-	 */
-	fraction = ror32(x, (msb - I2F_FRAC_BITS) & 0x1f) & I2F_MASK;
-	exponent = (127 + msb) << I2F_FRAC_BITS;
-
-	return fraction + exponent;
-}
-
 static int r600_nomm_get_vb(struct drm_device *dev)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index e082dca6feee..9fb5780a552f 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -31,6 +31,37 @@
 #include "r600_blit_shaders.h"
 #include "radeon_blit_common.h"
 
+/* 23 bits of float fractional data */
+#define I2F_FRAC_BITS  23
+#define I2F_MASK ((1 << I2F_FRAC_BITS) - 1)
+
+/*
+ * Converts unsigned integer into 32-bit IEEE floating point representation.
+ * Will be exact from 0 to 2^24.  Above that, we round towards zero
+ * as the fractional bits will not fit in a float.  (It would be better to
+ * round towards even as the fpu does, but that is slower.)
+ */
+__pure uint32_t int2float(uint32_t x)
+{
+	uint32_t msb, exponent, fraction;
+
+	/* Zero is special */
+	if (!x) return 0;
+
+	/* Get location of the most significant bit */
+	msb = __fls(x);
+
+	/*
+	 * Use a rotate instead of a shift because that works both leftwards
+	 * and rightwards due to the mod(32) behaviour.  This means we don't
+	 * need to check to see if we are above 2^24 or not.
+	 */
+	fraction = ror32(x, (msb - I2F_FRAC_BITS) & 0x1f) & I2F_MASK;
+	exponent = (127 + msb) << I2F_FRAC_BITS;
+
+	return fraction + exponent;
+}
+
 /* emits 21 on rv770+, 23 on r600 */
 static void
 set_render_target(struct radeon_device *rdev, int format,
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index be85f75aedda..1c51c08b1fde 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -24,6 +24,8 @@
  * Authors:
  *     Dave Airlie <airlied@redhat.com>
  *     Alex Deucher <alexander.deucher@amd.com>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <linux/module.h>
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 9b2512bf1a46..01a3ec83f284 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -31,12 +31,7 @@
 #include "r600d.h"
 #include "r600_reg_safe.h"
 
-static int r600_cs_packet_next_reloc_mm(struct radeon_cs_parser *p,
-					struct radeon_cs_reloc **cs_reloc);
-static int r600_cs_packet_next_reloc_nomm(struct radeon_cs_parser *p,
-					struct radeon_cs_reloc **cs_reloc);
-typedef int (*next_reloc_t)(struct radeon_cs_parser*, struct radeon_cs_reloc**);
-static next_reloc_t r600_cs_packet_next_reloc = &r600_cs_packet_next_reloc_mm;
+static int r600_nomm;
 extern void r600_cs_legacy_get_tiling_conf(struct drm_device *dev, u32 *npipes, u32 *nbanks, u32 *group_size);
 
 
@@ -784,170 +779,29 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
 }
 
 /**
- * r600_cs_packet_parse() - parse cp packet and point ib index to next packet
- * @parser:	parser structure holding parsing context.
- * @pkt:	where to store packet informations
- *
- * Assume that chunk_ib_index is properly set. Will return -EINVAL
- * if packet is bigger than remaining ib size. or if packets is unknown.
- **/
-static int r600_cs_packet_parse(struct radeon_cs_parser *p,
-			struct radeon_cs_packet *pkt,
-			unsigned idx)
-{
-	struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx];
-	uint32_t header;
-
-	if (idx >= ib_chunk->length_dw) {
-		DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
-			  idx, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	header = radeon_get_ib_value(p, idx);
-	pkt->idx = idx;
-	pkt->type = CP_PACKET_GET_TYPE(header);
-	pkt->count = CP_PACKET_GET_COUNT(header);
-	pkt->one_reg_wr = 0;
-	switch (pkt->type) {
-	case PACKET_TYPE0:
-		pkt->reg = CP_PACKET0_GET_REG(header);
-		break;
-	case PACKET_TYPE3:
-		pkt->opcode = CP_PACKET3_GET_OPCODE(header);
-		break;
-	case PACKET_TYPE2:
-		pkt->count = -1;
-		break;
-	default:
-		DRM_ERROR("Unknown packet type %d at %d !\n", pkt->type, idx);
-		return -EINVAL;
-	}
-	if ((pkt->count + 1 + pkt->idx) >= ib_chunk->length_dw) {
-		DRM_ERROR("Packet (%d:%d:%d) end after CS buffer (%d) !\n",
-			  pkt->idx, pkt->type, pkt->count, ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	return 0;
-}
-
-/**
- * r600_cs_packet_next_reloc_mm() - parse next packet which should be reloc packet3
- * @parser:		parser structure holding parsing context.
- * @data:		pointer to relocation data
- * @offset_start:	starting offset
- * @offset_mask:	offset mask (to align start offset on)
- * @reloc:		reloc informations
- *
- * Check next packet is relocation packet3, do bo validation and compute
- * GPU offset using the provided start.
- **/
-static int r600_cs_packet_next_reloc_mm(struct radeon_cs_parser *p,
-					struct radeon_cs_reloc **cs_reloc)
-{
-	struct radeon_cs_chunk *relocs_chunk;
-	struct radeon_cs_packet p3reloc;
-	unsigned idx;
-	int r;
-
-	if (p->chunk_relocs_idx == -1) {
-		DRM_ERROR("No relocation chunk !\n");
-		return -EINVAL;
-	}
-	*cs_reloc = NULL;
-	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
-	r = r600_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return r;
-	}
-	p->idx += p3reloc.count + 2;
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		DRM_ERROR("No packet3 for relocation for packet at %d.\n",
-			  p3reloc.idx);
-		return -EINVAL;
-	}
-	idx = radeon_get_ib_value(p, p3reloc.idx + 1);
-	if (idx >= relocs_chunk->length_dw) {
-		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
-			  idx, relocs_chunk->length_dw);
-		return -EINVAL;
-	}
-	/* FIXME: we assume reloc size is 4 dwords */
-	*cs_reloc = p->relocs_ptr[(idx / 4)];
-	return 0;
-}
-
-/**
- * r600_cs_packet_next_reloc_nomm() - parse next packet which should be reloc packet3
+ * r600_cs_packet_parse_vline() - parse userspace VLINE packet
  * @parser:		parser structure holding parsing context.
- * @data:		pointer to relocation data
- * @offset_start:	starting offset
- * @offset_mask:	offset mask (to align start offset on)
- * @reloc:		reloc informations
  *
- * Check next packet is relocation packet3, do bo validation and compute
- * GPU offset using the provided start.
- **/
-static int r600_cs_packet_next_reloc_nomm(struct radeon_cs_parser *p,
-					struct radeon_cs_reloc **cs_reloc)
-{
-	struct radeon_cs_chunk *relocs_chunk;
-	struct radeon_cs_packet p3reloc;
-	unsigned idx;
-	int r;
-
-	if (p->chunk_relocs_idx == -1) {
-		DRM_ERROR("No relocation chunk !\n");
-		return -EINVAL;
-	}
-	*cs_reloc = NULL;
-	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
-	r = r600_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return r;
-	}
-	p->idx += p3reloc.count + 2;
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		DRM_ERROR("No packet3 for relocation for packet at %d.\n",
-			  p3reloc.idx);
-		return -EINVAL;
-	}
-	idx = radeon_get_ib_value(p, p3reloc.idx + 1);
-	if (idx >= relocs_chunk->length_dw) {
-		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
-			  idx, relocs_chunk->length_dw);
-		return -EINVAL;
-	}
-	*cs_reloc = p->relocs;
-	(*cs_reloc)->lobj.gpu_offset = (u64)relocs_chunk->kdata[idx + 3] << 32;
-	(*cs_reloc)->lobj.gpu_offset |= relocs_chunk->kdata[idx + 0];
-	return 0;
-}
-
-/**
- * r600_cs_packet_next_is_pkt3_nop() - test if next packet is packet3 nop for reloc
- * @parser:		parser structure holding parsing context.
- *
- * Check next packet is relocation packet3, do bo validation and compute
- * GPU offset using the provided start.
- **/
-static int r600_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p)
+ * This is an R600-specific function for parsing VLINE packets.
+ * Real work is done by r600_cs_common_vline_parse function.
+ * Here we just set up ASIC-specific register table and call
+ * the common implementation function.
+ */
+static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 {
-	struct radeon_cs_packet p3reloc;
-	int r;
+	static uint32_t vline_start_end[2] = {AVIVO_D1MODE_VLINE_START_END,
+					      AVIVO_D2MODE_VLINE_START_END};
+	static uint32_t vline_status[2] = {AVIVO_D1MODE_VLINE_STATUS,
+					   AVIVO_D2MODE_VLINE_STATUS};
 
-	r = r600_cs_packet_parse(p, &p3reloc, p->idx);
-	if (r) {
-		return 0;
-	}
-	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
-		return 0;
-	}
-	return 1;
+	return r600_cs_common_vline_parse(p, vline_start_end, vline_status);
 }
 
 /**
- * r600_cs_packet_next_vline() - parse userspace VLINE packet
+ * r600_cs_common_vline_parse() - common vline parser
  * @parser:		parser structure holding parsing context.
+ * @vline_start_end:    table of vline_start_end registers
+ * @vline_status:       table of vline_status registers
  *
  * Userspace sends a special sequence for VLINE waits.
  * PACKET0 - VLINE_START_END + value
@@ -957,9 +811,16 @@ static int r600_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p)
  * This function parses this and relocates the VLINE START END
  * and WAIT_REG_MEM packets to the correct crtc.
  * It also detects a switched off crtc and nulls out the
- * wait in that case.
+ * wait in that case. This function is common for all ASICs that
+ * are R600 and newer. The parsing algorithm is the same, and only
+ * differs in which registers are used.
+ *
+ * Caller is the ASIC-specific function which passes the parser
+ * context and ASIC-specific register table
  */
-static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
+int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
+			       uint32_t *vline_start_end,
+			       uint32_t *vline_status)
 {
 	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
@@ -973,12 +834,12 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	ib = p->ib.ptr;
 
 	/* parse the WAIT_REG_MEM */
-	r = r600_cs_packet_parse(p, &wait_reg_mem, p->idx);
+	r = radeon_cs_packet_parse(p, &wait_reg_mem, p->idx);
 	if (r)
 		return r;
 
 	/* check its a WAIT_REG_MEM */
-	if (wait_reg_mem.type != PACKET_TYPE3 ||
+	if (wait_reg_mem.type != RADEON_PACKET_TYPE3 ||
 	    wait_reg_mem.opcode != PACKET3_WAIT_REG_MEM) {
 		DRM_ERROR("vline wait missing WAIT_REG_MEM segment\n");
 		return -EINVAL;
@@ -987,7 +848,12 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	wait_reg_mem_info = radeon_get_ib_value(p, wait_reg_mem.idx + 1);
 	/* bit 4 is reg (0) or mem (1) */
 	if (wait_reg_mem_info & 0x10) {
-		DRM_ERROR("vline WAIT_REG_MEM waiting on MEM rather than REG\n");
+		DRM_ERROR("vline WAIT_REG_MEM waiting on MEM instead of REG\n");
+		return -EINVAL;
+	}
+	/* bit 8 is me (0) or pfp (1) */
+	if (wait_reg_mem_info & 0x100) {
+		DRM_ERROR("vline WAIT_REG_MEM waiting on PFP instead of ME\n");
 		return -EINVAL;
 	}
 	/* waiting for value to be equal */
@@ -995,18 +861,18 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 		DRM_ERROR("vline WAIT_REG_MEM function not equal\n");
 		return -EINVAL;
 	}
-	if ((radeon_get_ib_value(p, wait_reg_mem.idx + 2) << 2) != AVIVO_D1MODE_VLINE_STATUS) {
+	if ((radeon_get_ib_value(p, wait_reg_mem.idx + 2) << 2) != vline_status[0]) {
 		DRM_ERROR("vline WAIT_REG_MEM bad reg\n");
 		return -EINVAL;
 	}
 
-	if (radeon_get_ib_value(p, wait_reg_mem.idx + 5) != AVIVO_D1MODE_VLINE_STAT) {
+	if (radeon_get_ib_value(p, wait_reg_mem.idx + 5) != RADEON_VLINE_STAT) {
 		DRM_ERROR("vline WAIT_REG_MEM bad bit mask\n");
 		return -EINVAL;
 	}
 
 	/* jump over the NOP */
-	r = r600_cs_packet_parse(p, &p3reloc, p->idx + wait_reg_mem.count + 2);
+	r = radeon_cs_packet_parse(p, &p3reloc, p->idx + wait_reg_mem.count + 2);
 	if (r)
 		return r;
 
@@ -1016,7 +882,7 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 
 	header = radeon_get_ib_value(p, h_idx);
 	crtc_id = radeon_get_ib_value(p, h_idx + 2 + 7 + 1);
-	reg = CP_PACKET0_GET_REG(header);
+	reg = R600_CP_PACKET0_GET_REG(header);
 
 	obj = drm_mode_object_find(p->rdev->ddev, crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
@@ -1028,7 +894,7 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	crtc_id = radeon_crtc->crtc_id;
 
 	if (!crtc->enabled) {
-		/* if the CRTC isn't enabled - we need to nop out the WAIT_REG_MEM */
+		/* CRTC isn't enabled - we need to nop out the WAIT_REG_MEM */
 		ib[h_idx + 2] = PACKET2(0);
 		ib[h_idx + 3] = PACKET2(0);
 		ib[h_idx + 4] = PACKET2(0);
@@ -1036,20 +902,15 @@ static int r600_cs_packet_parse_vline(struct radeon_cs_parser *p)
 		ib[h_idx + 6] = PACKET2(0);
 		ib[h_idx + 7] = PACKET2(0);
 		ib[h_idx + 8] = PACKET2(0);
-	} else if (crtc_id == 1) {
-		switch (reg) {
-		case AVIVO_D1MODE_VLINE_START_END:
-			header &= ~R600_CP_PACKET0_REG_MASK;
-			header |= AVIVO_D2MODE_VLINE_START_END >> 2;
-			break;
-		default:
-			DRM_ERROR("unknown crtc reloc\n");
-			return -EINVAL;
-		}
+	} else if (reg == vline_start_end[0]) {
+		header &= ~R600_CP_PACKET0_REG_MASK;
+		header |= vline_start_end[crtc_id] >> 2;
 		ib[h_idx] = header;
-		ib[h_idx + 4] = AVIVO_D2MODE_VLINE_STATUS >> 2;
+		ib[h_idx + 4] = vline_status[crtc_id] >> 2;
+	} else {
+		DRM_ERROR("unknown crtc reloc\n");
+		return -EINVAL;
 	}
-
 	return 0;
 }
 
@@ -1155,8 +1016,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		break;
 	case R_028010_DB_DEPTH_INFO:
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS) &&
-		    r600_cs_packet_next_is_pkt3_nop(p)) {
-			r = r600_cs_packet_next_reloc(p, &reloc);
+		    radeon_cs_packet_next_is_pkt3_nop(p)) {
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					 "0x%04X\n", reg);
@@ -1198,7 +1059,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case VGT_STRMOUT_BUFFER_BASE_1:
 	case VGT_STRMOUT_BUFFER_BASE_2:
 	case VGT_STRMOUT_BUFFER_BASE_3:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1221,7 +1082,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->streamout_dirty = true;
 		break;
 	case CP_COHER_BASE:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "missing reloc for CP_COHER_BASE "
 					"0x%04X\n", reg);
@@ -1256,8 +1117,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case R_0280B8_CB_COLOR6_INFO:
 	case R_0280BC_CB_COLOR7_INFO:
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS) &&
-		     r600_cs_packet_next_is_pkt3_nop(p)) {
-			r = r600_cs_packet_next_reloc(p, &reloc);
+		     radeon_cs_packet_next_is_pkt3_nop(p)) {
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
 				return -EINVAL;
@@ -1320,7 +1181,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case R_0280F8_CB_COLOR6_FRAG:
 	case R_0280FC_CB_COLOR7_FRAG:
 		tmp = (reg - R_0280E0_CB_COLOR0_FRAG) / 4;
-		if (!r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!radeon_cs_packet_next_is_pkt3_nop(p)) {
 			if (!track->cb_color_base_last[tmp]) {
 				dev_err(p->dev, "Broken old userspace ? no cb_color0_base supplied before trying to write 0x%08X\n", reg);
 				return -EINVAL;
@@ -1329,7 +1190,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 			track->cb_color_frag_offset[tmp] = track->cb_color_bo_offset[tmp];
 			ib[idx] = track->cb_color_base_last[tmp];
 		} else {
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
 				return -EINVAL;
@@ -1351,7 +1212,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case R_0280D8_CB_COLOR6_TILE:
 	case R_0280DC_CB_COLOR7_TILE:
 		tmp = (reg - R_0280C0_CB_COLOR0_TILE) / 4;
-		if (!r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!radeon_cs_packet_next_is_pkt3_nop(p)) {
 			if (!track->cb_color_base_last[tmp]) {
 				dev_err(p->dev, "Broken old userspace ? no cb_color0_base supplied before trying to write 0x%08X\n", reg);
 				return -EINVAL;
@@ -1360,7 +1221,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 			track->cb_color_tile_offset[tmp] = track->cb_color_bo_offset[tmp];
 			ib[idx] = track->cb_color_base_last[tmp];
 		} else {
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
 				return -EINVAL;
@@ -1395,7 +1256,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR5_BASE:
 	case CB_COLOR6_BASE:
 	case CB_COLOR7_BASE:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1410,7 +1271,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->cb_dirty = true;
 		break;
 	case DB_DEPTH_BASE:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1423,7 +1284,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_dirty = true;
 		break;
 	case DB_HTILE_DATA_BASE:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1493,7 +1354,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_ALU_CONST_CACHE_VS_13:
 	case SQ_ALU_CONST_CACHE_VS_14:
 	case SQ_ALU_CONST_CACHE_VS_15:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONTEXT_REG "
 					"0x%04X\n", reg);
@@ -1502,7 +1363,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
 		break;
 	case SX_MEMORY_EXPORT_BASE:
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			dev_warn(p->dev, "bad SET_CONFIG_REG "
 					"0x%04X\n", reg);
@@ -1788,7 +1649,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			return -EINVAL;
 		}
 
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			DRM_ERROR("bad SET PREDICATION\n");
 			return -EINVAL;
@@ -1829,7 +1690,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			DRM_ERROR("bad DRAW_INDEX\n");
 			return -EINVAL;
@@ -1881,7 +1742,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x10) {
 			uint64_t offset;
 
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad WAIT_REG_MEM\n");
 				return -EINVAL;
@@ -1893,6 +1754,9 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffff0);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
+		} else if (idx_value & 0x100) {
+			DRM_ERROR("cannot use PFP on REG wait\n");
+			return -EINVAL;
 		}
 		break;
 	case PACKET3_CP_DMA:
@@ -1915,7 +1779,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				return -EINVAL;
 			}
 			/* src address space is memory */
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad CP DMA SRC\n");
 				return -EINVAL;
@@ -1945,7 +1809,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				DRM_ERROR("CP DMA DAIC only supported for registers\n");
 				return -EINVAL;
 			}
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad CP DMA DST\n");
 				return -EINVAL;
@@ -1975,7 +1839,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		/* 0xffffffff/0x0 is flush all cache flag */
 		if (radeon_get_ib_value(p, idx + 1) != 0xffffffff ||
 		    radeon_get_ib_value(p, idx + 2) != 0) {
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad SURFACE_SYNC\n");
 				return -EINVAL;
@@ -1991,7 +1855,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if (pkt->count) {
 			uint64_t offset;
 
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad EVENT_WRITE\n");
 				return -EINVAL;
@@ -2012,7 +1876,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
 		}
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			DRM_ERROR("bad EVENT_WRITE\n");
 			return -EINVAL;
@@ -2078,7 +1942,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			switch (G__SQ_VTX_CONSTANT_TYPE(radeon_get_ib_value(p, idx+(i*7)+6+1))) {
 			case SQ_TEX_VTX_VALID_TEXTURE:
 				/* tex base */
-				r = r600_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
 					DRM_ERROR("bad SET_RESOURCE\n");
 					return -EINVAL;
@@ -2092,7 +1956,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				}
 				texture = reloc->robj;
 				/* tex mip base */
-				r = r600_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
 					DRM_ERROR("bad SET_RESOURCE\n");
 					return -EINVAL;
@@ -2113,7 +1977,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			{
 				uint64_t offset64;
 				/* vtx base */
-				r = r600_cs_packet_next_reloc(p, &reloc);
+				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
 					DRM_ERROR("bad SET_RESOURCE\n");
 					return -EINVAL;
@@ -2214,7 +2078,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		{
 			u64 offset;
 
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad STRMOUT_BASE_UPDATE reloc\n");
 				return -EINVAL;
@@ -2258,7 +2122,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		/* Updating memory at DST_ADDRESS. */
 		if (idx_value & 0x1) {
 			u64 offset;
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
 				return -EINVAL;
@@ -2277,7 +2141,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		/* Reading data from SRC_ADDRESS. */
 		if (((idx_value >> 1) & 0x3) == 2) {
 			u64 offset;
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
 				return -EINVAL;
@@ -2302,7 +2166,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad MEM_WRITE (invalid count)\n");
 			return -EINVAL;
 		}
-		r = r600_cs_packet_next_reloc(p, &reloc);
+		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
 			DRM_ERROR("bad MEM_WRITE (missing reloc)\n");
 			return -EINVAL;
@@ -2331,7 +2195,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x1) {
 			u64 offset;
 			/* SRC is memory. */
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad COPY_DW (missing src reloc)\n");
 				return -EINVAL;
@@ -2355,7 +2219,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if (idx_value & 0x2) {
 			u64 offset;
 			/* DST is memory. */
-			r = r600_cs_packet_next_reloc(p, &reloc);
+			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
 				DRM_ERROR("bad COPY_DW (missing dst reloc)\n");
 				return -EINVAL;
@@ -2410,7 +2274,7 @@ int r600_cs_parse(struct radeon_cs_parser *p)
 		p->track = track;
 	}
 	do {
-		r = r600_cs_packet_parse(p, &pkt, p->idx);
+		r = radeon_cs_packet_parse(p, &pkt, p->idx);
 		if (r) {
 			kfree(p->track);
 			p->track = NULL;
@@ -2418,12 +2282,12 @@ int r600_cs_parse(struct radeon_cs_parser *p)
 		}
 		p->idx += pkt.count + 2;
 		switch (pkt.type) {
-		case PACKET_TYPE0:
+		case RADEON_PACKET_TYPE0:
 			r = r600_cs_parse_packet0(p, &pkt);
 			break;
-		case PACKET_TYPE2:
+		case RADEON_PACKET_TYPE2:
 			break;
-		case PACKET_TYPE3:
+		case RADEON_PACKET_TYPE3:
 			r = r600_packet3_check(p, &pkt);
 			break;
 		default:
@@ -2449,17 +2313,7 @@ int r600_cs_parse(struct radeon_cs_parser *p)
 	return 0;
 }
 
-static int r600_cs_parser_relocs_legacy(struct radeon_cs_parser *p)
-{
-	if (p->chunk_relocs_idx == -1) {
-		return 0;
-	}
-	p->relocs = kzalloc(sizeof(struct radeon_cs_reloc), GFP_KERNEL);
-	if (p->relocs == NULL) {
-		return -ENOMEM;
-	}
-	return 0;
-}
+#ifdef CONFIG_DRM_RADEON_UMS
 
 /**
  * cs_parser_fini() - clean parser states
@@ -2485,6 +2339,18 @@ static void r600_cs_parser_fini(struct radeon_cs_parser *parser, int error)
 	kfree(parser->chunks_array);
 }
 
+static int r600_cs_parser_relocs_legacy(struct radeon_cs_parser *p)
+{
+	if (p->chunk_relocs_idx == -1) {
+		return 0;
+	}
+	p->relocs = kzalloc(sizeof(struct radeon_cs_reloc), GFP_KERNEL);
+	if (p->relocs == NULL) {
+		return -ENOMEM;
+	}
+	return 0;
+}
+
 int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp,
 			unsigned family, u32 *ib, int *l)
 {
@@ -2543,9 +2409,11 @@ int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp,
 
 void r600_cs_legacy_init(void)
 {
-	r600_cs_packet_next_reloc = &r600_cs_packet_next_reloc_nomm;
+	r600_nomm = 1;
 }
 
+#endif
+
 /*
  *  DMA
  */
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index ff80efe9cb7d..21ecc0e12dc4 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -23,6 +23,7 @@
  *
  * Authors: Christian König
  */
+#include <linux/hdmi.h>
 #include <drm/drmP.h>
 #include <drm/radeon_drm.h>
 #include "radeon.h"
@@ -121,79 +122,18 @@ static void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 }
 
 /*
- * calculate the crc for a given info frame
- */
-static void r600_hdmi_infoframe_checksum(uint8_t packetType,
-					 uint8_t versionNumber,
-					 uint8_t length,
-					 uint8_t *frame)
-{
-	int i;
-	frame[0] = packetType + versionNumber + length;
-	for (i = 1; i <= length; i++)
-		frame[0] += frame[i];
-	frame[0] = 0x100 - frame[0];
-}
-
-/*
  * build a HDMI Video Info Frame
  */
-static void r600_hdmi_videoinfoframe(
-	struct drm_encoder *encoder,
-	enum r600_hdmi_color_format color_format,
-	int active_information_present,
-	uint8_t active_format_aspect_ratio,
-	uint8_t scan_information,
-	uint8_t colorimetry,
-	uint8_t ex_colorimetry,
-	uint8_t quantization,
-	int ITC,
-	uint8_t picture_aspect_ratio,
-	uint8_t video_format_identification,
-	uint8_t pixel_repetition,
-	uint8_t non_uniform_picture_scaling,
-	uint8_t bar_info_data_valid,
-	uint16_t top_bar,
-	uint16_t bottom_bar,
-	uint16_t left_bar,
-	uint16_t right_bar
-)
+static void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder,
+					   void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
+	uint8_t *frame = buffer + 3;
 
-	uint8_t frame[14];
-
-	frame[0x0] = 0;
-	frame[0x1] =
-		(scan_information & 0x3) |
-		((bar_info_data_valid & 0x3) << 2) |
-		((active_information_present & 0x1) << 4) |
-		((color_format & 0x3) << 5);
-	frame[0x2] =
-		(active_format_aspect_ratio & 0xF) |
-		((picture_aspect_ratio & 0x3) << 4) |
-		((colorimetry & 0x3) << 6);
-	frame[0x3] =
-		(non_uniform_picture_scaling & 0x3) |
-		((quantization & 0x3) << 2) |
-		((ex_colorimetry & 0x7) << 4) |
-		((ITC & 0x1) << 7);
-	frame[0x4] = (video_format_identification & 0x7F);
-	frame[0x5] = (pixel_repetition & 0xF);
-	frame[0x6] = (top_bar & 0xFF);
-	frame[0x7] = (top_bar >> 8);
-	frame[0x8] = (bottom_bar & 0xFF);
-	frame[0x9] = (bottom_bar >> 8);
-	frame[0xA] = (left_bar & 0xFF);
-	frame[0xB] = (left_bar >> 8);
-	frame[0xC] = (right_bar & 0xFF);
-	frame[0xD] = (right_bar >> 8);
-
-	r600_hdmi_infoframe_checksum(0x82, 0x02, 0x0D, frame);
 	/* Our header values (type, version, length) should be alright, Intel
 	 * is using the same. Checksum function also seems to be OK, it works
 	 * fine for audio infoframe. However calculated value is always lower
@@ -215,39 +155,15 @@ static void r600_hdmi_videoinfoframe(
 /*
  * build a Audio Info Frame
  */
-static void r600_hdmi_audioinfoframe(
-	struct drm_encoder *encoder,
-	uint8_t channel_count,
-	uint8_t coding_type,
-	uint8_t sample_size,
-	uint8_t sample_frequency,
-	uint8_t format,
-	uint8_t channel_allocation,
-	uint8_t level_shift,
-	int downmix_inhibit
-)
+static void r600_hdmi_update_audio_infoframe(struct drm_encoder *encoder,
+					     const void *buffer, size_t size)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
-
-	uint8_t frame[11];
-
-	frame[0x0] = 0;
-	frame[0x1] = (channel_count & 0x7) | ((coding_type & 0xF) << 4);
-	frame[0x2] = (sample_size & 0x3) | ((sample_frequency & 0x7) << 2);
-	frame[0x3] = format;
-	frame[0x4] = channel_allocation;
-	frame[0x5] = ((level_shift & 0xF) << 3) | ((downmix_inhibit & 0x1) << 7);
-	frame[0x6] = 0;
-	frame[0x7] = 0;
-	frame[0x8] = 0;
-	frame[0x9] = 0;
-	frame[0xA] = 0;
-
-	r600_hdmi_infoframe_checksum(0x84, 0x01, 0x0A, frame);
+	const u8 *frame = buffer + 3;
 
 	WREG32(HDMI0_AUDIO_INFO0 + offset,
 		frame[0x0] | (frame[0x1] << 8) | (frame[0x2] << 16) | (frame[0x3] << 24));
@@ -320,7 +236,10 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
+	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
+	ssize_t err;
 
 	/* Silent, r600_hdmi_enable will raise WARN for us */
 	if (!dig->afmt->enabled)
@@ -371,9 +290,19 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 
 	WREG32(HDMI0_GC + offset, 0); /* unset HDMI0_GC_AVMUTE */
 
-	r600_hdmi_videoinfoframe(encoder, RGB, 0, 0, 0, 0,
-		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	if (err < 0) {
+		DRM_ERROR("failed to setup AVI infoframe: %zd\n", err);
+		return;
+	}
 
+	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		DRM_ERROR("failed to pack AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	r600_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
 	r600_hdmi_update_ACR(encoder, mode->clock);
 
 	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
@@ -395,8 +324,11 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	struct r600_audio audio = r600_audio_status(rdev);
+	uint8_t buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AUDIO_INFOFRAME_SIZE];
+	struct hdmi_audio_infoframe frame;
 	uint32_t offset;
 	uint32_t iec;
+	ssize_t err;
 
 	if (!dig->afmt || !dig->afmt->enabled)
 		return;
@@ -462,9 +394,21 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 		iec |= 0x5 << 16;
 	WREG32_P(HDMI0_60958_1 + offset, iec, ~0x5000f);
 
-	r600_hdmi_audioinfoframe(encoder, audio.channels - 1, 0, 0, 0, 0, 0, 0,
-				 0);
+	err = hdmi_audio_infoframe_init(&frame);
+	if (err < 0) {
+		DRM_ERROR("failed to setup audio infoframe\n");
+		return;
+	}
+
+	frame.channels = audio.channels;
+
+	err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		DRM_ERROR("failed to pack audio infoframe\n");
+		return;
+	}
 
+	r600_hdmi_update_audio_infoframe(encoder, buffer, sizeof(buffer));
 	r600_hdmi_audio_workaround(encoder);
 }
 
@@ -544,7 +488,6 @@ void r600_hdmi_disable(struct drm_encoder *encoder)
 
 	/* Called for ATOM_ENCODER_MODE_HDMI only */
 	if (!dig || !dig->afmt) {
-		WARN_ON(1);
 		return;
 	}
 	if (!dig->afmt->enabled)
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 4a53402b1852..a42ba11a3bed 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -182,6 +182,8 @@
 #define	CP_COHER_BASE					0x85F8
 #define	CP_DEBUG					0xC1FC
 #define	R_0086D8_CP_ME_CNTL			0x86D8
+#define		S_0086D8_CP_PFP_HALT(x)			(((x) & 1)<<26)
+#define		C_0086D8_CP_PFP_HALT(x)			((x) & 0xFBFFFFFF)
 #define		S_0086D8_CP_ME_HALT(x)			(((x) & 1)<<28)
 #define		C_0086D8_CP_ME_HALT(x)			((x) & 0xEFFFFFFF)
 #define	CP_ME_RAM_DATA					0xC160
@@ -1143,19 +1145,10 @@
 /*
  * PM4
  */
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0xFFFF) << 2)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-#define PACKET0(reg, n)	((PACKET_TYPE0 << 30) |				\
+#define PACKET0(reg, n)	((RADEON_PACKET_TYPE0 << 30) |			\
 			 (((reg) >> 2) & 0xFFFF) |			\
 			 ((n) & 0x3FFF) << 16)
-#define PACKET3(op, n)	((PACKET_TYPE3 << 30) |				\
+#define PACKET3(op, n)	((RADEON_PACKET_TYPE3 << 30) |			\
 			 (((op) & 0xFF) << 8) |				\
 			 ((n) & 0x3FFF) << 16)
 
@@ -1328,6 +1321,7 @@
 #define		G_008010_VC_BUSY(x)			(((x) >> 11) & 1)
 #define		G_008010_DB03_CLEAN(x)			(((x) >> 12) & 1)
 #define		G_008010_CB03_CLEAN(x)			(((x) >> 13) & 1)
+#define		G_008010_TA_BUSY(x)			(((x) >> 14) & 1)
 #define		G_008010_VGT_BUSY_NO_DMA(x)		(((x) >> 16) & 1)
 #define		G_008010_VGT_BUSY(x)			(((x) >> 17) & 1)
 #define		G_008010_TA03_BUSY(x)			(((x) >> 18) & 1)
@@ -1395,6 +1389,7 @@
 #define		G_000E50_MCDW_BUSY(x)			(((x) >> 13) & 1)
 #define		G_000E50_SEM_BUSY(x)			(((x) >> 14) & 1)
 #define		G_000E50_RLC_BUSY(x)			(((x) >> 15) & 1)
+#define		G_000E50_IH_BUSY(x)			(((x) >> 17) & 1)
 #define		G_000E50_BIF_BUSY(x)			(((x) >> 29) & 1)
 #define	R_000E60_SRBM_SOFT_RESET			0x0E60
 #define		S_000E60_SOFT_RESET_BIF(x)		(((x) & 1) << 1)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index a08f657329a0..8263af3fd832 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -136,6 +136,15 @@ extern int radeon_lockup_timeout;
 #define RADEON_RESET_GFX			(1 << 0)
 #define RADEON_RESET_COMPUTE			(1 << 1)
 #define RADEON_RESET_DMA			(1 << 2)
+#define RADEON_RESET_CP				(1 << 3)
+#define RADEON_RESET_GRBM			(1 << 4)
+#define RADEON_RESET_DMA1			(1 << 5)
+#define RADEON_RESET_RLC			(1 << 6)
+#define RADEON_RESET_SEM			(1 << 7)
+#define RADEON_RESET_IH				(1 << 8)
+#define RADEON_RESET_VMC			(1 << 9)
+#define RADEON_RESET_MC				(1 << 10)
+#define RADEON_RESET_DISPLAY			(1 << 11)
 
 /*
  * Errata workarounds.
@@ -341,7 +350,6 @@ struct radeon_bo {
 	struct drm_gem_object		gem_base;
 
 	struct ttm_bo_kmap_obj dma_buf_vmap;
-	int vmapping_count;
 };
 #define gem_to_radeon_bo(gobj) container_of((gobj), struct radeon_bo, gem_base)
 
@@ -771,6 +779,7 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
 		  struct radeon_ib *ib, struct radeon_vm *vm,
 		  unsigned size);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
+void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence);
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 		       struct radeon_ib *const_ib);
 int radeon_ib_pool_init(struct radeon_device *rdev);
@@ -1169,6 +1178,10 @@ struct radeon_asic {
 	bool (*gui_idle)(struct radeon_device *rdev);
 	/* wait for mc_idle */
 	int (*mc_wait_for_idle)(struct radeon_device *rdev);
+	/* get the reference clock */
+	u32 (*get_xclk)(struct radeon_device *rdev);
+	/* get the gpu clock counter */
+	uint64_t (*get_gpu_clock_counter)(struct radeon_device *rdev);
 	/* gart */
 	struct {
 		void (*tlb_flush)(struct radeon_device *rdev);
@@ -1179,7 +1192,9 @@ struct radeon_asic {
 		void (*fini)(struct radeon_device *rdev);
 
 		u32 pt_ring_index;
-		void (*set_page)(struct radeon_device *rdev, uint64_t pe,
+		void (*set_page)(struct radeon_device *rdev,
+				 struct radeon_ib *ib,
+				 uint64_t pe,
 				 uint64_t addr, unsigned count,
 				 uint32_t incr, uint32_t flags);
 	} vm;
@@ -1757,6 +1772,7 @@ void r100_pll_errata_after_index(struct radeon_device *rdev);
 #define ASIC_IS_DCE6(rdev) ((rdev->family >= CHIP_ARUBA))
 #define ASIC_IS_DCE61(rdev) ((rdev->family >= CHIP_ARUBA) && \
 			     (rdev->flags & RADEON_IS_IGP))
+#define ASIC_IS_DCE64(rdev) ((rdev->family == CHIP_OLAND))
 
 /*
  * BIOS helpers.
@@ -1801,7 +1817,7 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
 #define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart.set_page((rdev), (i), (p))
 #define radeon_asic_vm_init(rdev) (rdev)->asic->vm.init((rdev))
 #define radeon_asic_vm_fini(rdev) (rdev)->asic->vm.fini((rdev))
-#define radeon_asic_vm_set_page(rdev, pe, addr, count, incr, flags) ((rdev)->asic->vm.set_page((rdev), (pe), (addr), (count), (incr), (flags)))
+#define radeon_asic_vm_set_page(rdev, ib, pe, addr, count, incr, flags) ((rdev)->asic->vm.set_page((rdev), (ib), (pe), (addr), (count), (incr), (flags)))
 #define radeon_ring_start(rdev, r, cp) (rdev)->asic->ring[(r)].ring_start((rdev), (cp))
 #define radeon_ring_test(rdev, r, cp) (rdev)->asic->ring[(r)].ring_test((rdev), (cp))
 #define radeon_ib_test(rdev, r, cp) (rdev)->asic->ring[(r)].ib_test((rdev), (cp))
@@ -1847,10 +1863,13 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
 #define radeon_post_page_flip(rdev, crtc) (rdev)->asic->pflip.post_page_flip((rdev), (crtc))
 #define radeon_wait_for_vblank(rdev, crtc) (rdev)->asic->display.wait_for_vblank((rdev), (crtc))
 #define radeon_mc_wait_for_idle(rdev) (rdev)->asic->mc_wait_for_idle((rdev))
+#define radeon_get_xclk(rdev) (rdev)->asic->get_xclk((rdev))
+#define radeon_get_gpu_clock_counter(rdev) (rdev)->asic->get_gpu_clock_counter((rdev))
 
 /* Common functions */
 /* AGP */
 extern int radeon_gpu_reset(struct radeon_device *rdev);
+extern void r600_set_bios_scratch_engine_hung(struct radeon_device *rdev, bool hung);
 extern void radeon_agp_disable(struct radeon_device *rdev);
 extern int radeon_modeset_init(struct radeon_device *rdev);
 extern void radeon_modeset_fini(struct radeon_device *rdev);
@@ -1972,6 +1991,19 @@ static inline int radeon_acpi_init(struct radeon_device *rdev) { return 0; }
 static inline void radeon_acpi_fini(struct radeon_device *rdev) { }
 #endif
 
+int radeon_cs_packet_parse(struct radeon_cs_parser *p,
+			   struct radeon_cs_packet *pkt,
+			   unsigned idx);
+bool radeon_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p);
+void radeon_cs_dump_packet(struct radeon_cs_parser *p,
+			   struct radeon_cs_packet *pkt);
+int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p,
+				struct radeon_cs_reloc **cs_reloc,
+				int nomm);
+int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
+			       uint32_t *vline_start_end,
+			       uint32_t *vline_status);
+
 #include "radeon_object.h"
 
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 0b202c07fe50..aba0a893ea98 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -934,6 +934,8 @@ static struct radeon_asic r600_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &r600_mc_wait_for_idle,
+	.get_xclk = &r600_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &r600_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -946,7 +948,7 @@ static struct radeon_asic r600_asic = {
 			.cs_parse = &r600_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &r600_gpu_is_lockup,
+			.is_lockup = &r600_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &r600_dma_ring_ib_execute,
@@ -1018,6 +1020,8 @@ static struct radeon_asic rs780_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &r600_mc_wait_for_idle,
+	.get_xclk = &r600_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &r600_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1030,7 +1034,7 @@ static struct radeon_asic rs780_asic = {
 			.cs_parse = &r600_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &r600_gpu_is_lockup,
+			.is_lockup = &r600_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &r600_dma_ring_ib_execute,
@@ -1102,6 +1106,8 @@ static struct radeon_asic rv770_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &r600_mc_wait_for_idle,
+	.get_xclk = &rv770_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &r600_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1114,7 +1120,7 @@ static struct radeon_asic rv770_asic = {
 			.cs_parse = &r600_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &r600_gpu_is_lockup,
+			.is_lockup = &r600_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &r600_dma_ring_ib_execute,
@@ -1186,6 +1192,8 @@ static struct radeon_asic evergreen_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &rv770_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &evergreen_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1198,7 +1206,7 @@ static struct radeon_asic evergreen_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &evergreen_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &evergreen_dma_ring_ib_execute,
@@ -1207,7 +1215,7 @@ static struct radeon_asic evergreen_asic = {
 			.cs_parse = &evergreen_dma_cs_parse,
 			.ring_test = &r600_dma_ring_test,
 			.ib_test = &r600_dma_ib_test,
-			.is_lockup = &r600_dma_is_lockup,
+			.is_lockup = &evergreen_dma_is_lockup,
 		}
 	},
 	.irq = {
@@ -1270,6 +1278,8 @@ static struct radeon_asic sumo_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &r600_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &evergreen_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1282,7 +1292,7 @@ static struct radeon_asic sumo_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &evergreen_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &evergreen_dma_ring_ib_execute,
@@ -1291,7 +1301,7 @@ static struct radeon_asic sumo_asic = {
 			.cs_parse = &evergreen_dma_cs_parse,
 			.ring_test = &r600_dma_ring_test,
 			.ib_test = &r600_dma_ib_test,
-			.is_lockup = &r600_dma_is_lockup,
+			.is_lockup = &evergreen_dma_is_lockup,
 		}
 	},
 	.irq = {
@@ -1354,6 +1364,8 @@ static struct radeon_asic btc_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &rv770_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &evergreen_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1366,7 +1378,7 @@ static struct radeon_asic btc_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &evergreen_gfx_is_lockup,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
 			.ib_execute = &evergreen_dma_ring_ib_execute,
@@ -1375,7 +1387,7 @@ static struct radeon_asic btc_asic = {
 			.cs_parse = &evergreen_dma_cs_parse,
 			.ring_test = &r600_dma_ring_test,
 			.ib_test = &r600_dma_ib_test,
-			.is_lockup = &r600_dma_is_lockup,
+			.is_lockup = &evergreen_dma_is_lockup,
 		}
 	},
 	.irq = {
@@ -1438,6 +1450,8 @@ static struct radeon_asic cayman_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &rv770_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &cayman_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1445,7 +1459,7 @@ static struct radeon_asic cayman_asic = {
 	.vm = {
 		.init = &cayman_vm_init,
 		.fini = &cayman_vm_fini,
-		.pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
+		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
 		.set_page = &cayman_vm_set_page,
 	},
 	.ring = {
@@ -1457,7 +1471,7 @@ static struct radeon_asic cayman_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP1_INDEX] = {
@@ -1468,7 +1482,7 @@ static struct radeon_asic cayman_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP2_INDEX] = {
@@ -1479,7 +1493,7 @@ static struct radeon_asic cayman_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
@@ -1565,6 +1579,8 @@ static struct radeon_asic trinity_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &r600_get_xclk,
+	.get_gpu_clock_counter = &r600_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &cayman_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1572,7 +1588,7 @@ static struct radeon_asic trinity_asic = {
 	.vm = {
 		.init = &cayman_vm_init,
 		.fini = &cayman_vm_fini,
-		.pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
+		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
 		.set_page = &cayman_vm_set_page,
 	},
 	.ring = {
@@ -1584,7 +1600,7 @@ static struct radeon_asic trinity_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP1_INDEX] = {
@@ -1595,7 +1611,7 @@ static struct radeon_asic trinity_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP2_INDEX] = {
@@ -1606,7 +1622,7 @@ static struct radeon_asic trinity_asic = {
 			.cs_parse = &evergreen_cs_parse,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &evergreen_gpu_is_lockup,
+			.is_lockup = &cayman_gfx_is_lockup,
 			.vm_flush = &cayman_vm_flush,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
@@ -1692,6 +1708,8 @@ static struct radeon_asic si_asic = {
 	.ioctl_wait_idle = r600_ioctl_wait_idle,
 	.gui_idle = &r600_gui_idle,
 	.mc_wait_for_idle = &evergreen_mc_wait_for_idle,
+	.get_xclk = &si_get_xclk,
+	.get_gpu_clock_counter = &si_get_gpu_clock_counter,
 	.gart = {
 		.tlb_flush = &si_pcie_gart_tlb_flush,
 		.set_page = &rs600_gart_set_page,
@@ -1699,7 +1717,7 @@ static struct radeon_asic si_asic = {
 	.vm = {
 		.init = &si_vm_init,
 		.fini = &si_vm_fini,
-		.pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
+		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
 		.set_page = &si_vm_set_page,
 	},
 	.ring = {
@@ -1711,7 +1729,7 @@ static struct radeon_asic si_asic = {
 			.cs_parse = NULL,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &si_gpu_is_lockup,
+			.is_lockup = &si_gfx_is_lockup,
 			.vm_flush = &si_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP1_INDEX] = {
@@ -1722,7 +1740,7 @@ static struct radeon_asic si_asic = {
 			.cs_parse = NULL,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &si_gpu_is_lockup,
+			.is_lockup = &si_gfx_is_lockup,
 			.vm_flush = &si_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_CP2_INDEX] = {
@@ -1733,7 +1751,7 @@ static struct radeon_asic si_asic = {
 			.cs_parse = NULL,
 			.ring_test = &r600_ring_test,
 			.ib_test = &r600_ib_test,
-			.is_lockup = &si_gpu_is_lockup,
+			.is_lockup = &si_gfx_is_lockup,
 			.vm_flush = &si_vm_flush,
 		},
 		[R600_RING_TYPE_DMA_INDEX] = {
@@ -1744,7 +1762,7 @@ static struct radeon_asic si_asic = {
 			.cs_parse = NULL,
 			.ring_test = &r600_dma_ring_test,
 			.ib_test = &r600_dma_ib_test,
-			.is_lockup = &cayman_dma_is_lockup,
+			.is_lockup = &si_dma_is_lockup,
 			.vm_flush = &si_dma_vm_flush,
 		},
 		[CAYMAN_RING_TYPE_DMA1_INDEX] = {
@@ -1755,7 +1773,7 @@ static struct radeon_asic si_asic = {
 			.cs_parse = NULL,
 			.ring_test = &r600_dma_ring_test,
 			.ib_test = &r600_dma_ib_test,
-			.is_lockup = &cayman_dma_is_lockup,
+			.is_lockup = &si_dma_is_lockup,
 			.vm_flush = &si_dma_vm_flush,
 		}
 	},
@@ -1944,9 +1962,13 @@ int radeon_asic_init(struct radeon_device *rdev)
 	case CHIP_TAHITI:
 	case CHIP_PITCAIRN:
 	case CHIP_VERDE:
+	case CHIP_OLAND:
 		rdev->asic = &si_asic;
 		/* set num crtcs */
-		rdev->num_crtc = 6;
+		if (rdev->family == CHIP_OLAND)
+			rdev->num_crtc = 2;
+		else
+			rdev->num_crtc = 6;
 		break;
 	default:
 		/* FIXME: not supported yet */
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 15d70e613076..3535f73ad3e2 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -319,7 +319,7 @@ void r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
 				  bool emit_wait);
 void r600_dma_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 bool r600_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
-bool r600_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
+bool r600_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r600_asic_reset(struct radeon_device *rdev);
 int r600_set_surface_reg(struct radeon_device *rdev, int reg,
 			 uint32_t tiling_flags, uint32_t pitch,
@@ -389,7 +389,8 @@ void r600_kms_blit_copy(struct radeon_device *rdev,
 			unsigned num_gpu_pages,
 			struct radeon_sa_bo *vb);
 int r600_mc_wait_for_idle(struct radeon_device *rdev);
-uint64_t r600_get_gpu_clock(struct radeon_device *rdev);
+u32 r600_get_xclk(struct radeon_device *rdev);
+uint64_t r600_get_gpu_clock_counter(struct radeon_device *rdev);
 
 /*
  * rv770,rv730,rv710,rv740
@@ -407,6 +408,7 @@ int rv770_copy_dma(struct radeon_device *rdev,
 		  uint64_t src_offset, uint64_t dst_offset,
 		  unsigned num_gpu_pages,
 		   struct radeon_fence **fence);
+u32 rv770_get_xclk(struct radeon_device *rdev);
 
 /*
  * evergreen
@@ -422,7 +424,8 @@ int evergreen_init(struct radeon_device *rdev);
 void evergreen_fini(struct radeon_device *rdev);
 int evergreen_suspend(struct radeon_device *rdev);
 int evergreen_resume(struct radeon_device *rdev);
-bool evergreen_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
+bool evergreen_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
+bool evergreen_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int evergreen_asic_reset(struct radeon_device *rdev);
 void evergreen_bandwidth_update(struct radeon_device *rdev);
 void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
@@ -473,13 +476,16 @@ int cayman_vm_init(struct radeon_device *rdev);
 void cayman_vm_fini(struct radeon_device *rdev);
 void cayman_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags);
-void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe,
+void cayman_vm_set_page(struct radeon_device *rdev,
+			struct radeon_ib *ib,
+			uint64_t pe,
 			uint64_t addr, unsigned count,
 			uint32_t incr, uint32_t flags);
 int evergreen_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 void cayman_dma_ring_ib_execute(struct radeon_device *rdev,
 				struct radeon_ib *ib);
+bool cayman_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
 bool cayman_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
 void cayman_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 
@@ -496,23 +502,27 @@ int si_init(struct radeon_device *rdev);
 void si_fini(struct radeon_device *rdev);
 int si_suspend(struct radeon_device *rdev);
 int si_resume(struct radeon_device *rdev);
-bool si_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
+bool si_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
+bool si_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int si_asic_reset(struct radeon_device *rdev);
 void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 int si_irq_set(struct radeon_device *rdev);
 int si_irq_process(struct radeon_device *rdev);
 int si_vm_init(struct radeon_device *rdev);
 void si_vm_fini(struct radeon_device *rdev);
-void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
+void si_vm_set_page(struct radeon_device *rdev,
+		    struct radeon_ib *ib,
+		    uint64_t pe,
 		    uint64_t addr, unsigned count,
 		    uint32_t incr, uint32_t flags);
 void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
-uint64_t si_get_gpu_clock(struct radeon_device *rdev);
 int si_copy_dma(struct radeon_device *rdev,
 		uint64_t src_offset, uint64_t dst_offset,
 		unsigned num_gpu_pages,
 		struct radeon_fence **fence);
 void si_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
+u32 si_get_xclk(struct radeon_device *rdev);
+uint64_t si_get_gpu_clock_counter(struct radeon_device *rdev);
 
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 15f5ded65e0c..d96070bf8388 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -43,6 +43,12 @@ struct atpx_verify_interface {
 	u32 function_bits;	/* supported functions bit vector */
 } __packed;
 
+struct atpx_px_params {
+	u16 size;		/* structure size in bytes (includes size field) */
+	u32 valid_flags;	/* which flags are valid */
+	u32 flags;		/* flags */
+} __packed;
+
 struct atpx_power_control {
 	u16 size;
 	u8 dgpu_state;
@@ -123,9 +129,61 @@ static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mas
 }
 
 /**
+ * radeon_atpx_validate_functions - validate ATPX functions
+ *
+ * @atpx: radeon atpx struct
+ *
+ * Validate that required functions are enabled (all asics).
+ * returns 0 on success, error on failure.
+ */
+static int radeon_atpx_validate(struct radeon_atpx *atpx)
+{
+	/* make sure required functions are enabled */
+	/* dGPU power control is required */
+	atpx->functions.power_cntl = true;
+
+	if (atpx->functions.px_params) {
+		union acpi_object *info;
+		struct atpx_px_params output;
+		size_t size;
+		u32 valid_bits;
+
+		info = radeon_atpx_call(atpx->handle, ATPX_FUNCTION_GET_PX_PARAMETERS, NULL);
+		if (!info)
+			return -EIO;
+
+		memset(&output, 0, sizeof(output));
+
+		size = *(u16 *) info->buffer.pointer;
+		if (size < 10) {
+			printk("ATPX buffer is too small: %zu\n", size);
+			kfree(info);
+			return -EINVAL;
+		}
+		size = min(sizeof(output), size);
+
+		memcpy(&output, info->buffer.pointer, size);
+
+		valid_bits = output.flags & output.valid_flags;
+		/* if separate mux flag is set, mux controls are required */
+		if (valid_bits & ATPX_SEPARATE_MUX_FOR_I2C) {
+			atpx->functions.i2c_mux_cntl = true;
+			atpx->functions.disp_mux_cntl = true;
+		}
+		/* if any outputs are muxed, mux controls are required */
+		if (valid_bits & (ATPX_CRT1_RGB_SIGNAL_MUXED |
+				  ATPX_TV_SIGNAL_MUXED |
+				  ATPX_DFP_SIGNAL_MUXED))
+			atpx->functions.disp_mux_cntl = true;
+
+		kfree(info);
+	}
+	return 0;
+}
+
+/**
  * radeon_atpx_verify_interface - verify ATPX
  *
- * @handle: acpi handle
  * @atpx: radeon atpx struct
  *
  * Execute the ATPX_FUNCTION_VERIFY_INTERFACE ATPX function
@@ -406,8 +464,19 @@ static bool radeon_atpx_pci_probe_handle(struct pci_dev *pdev)
  */
 static int radeon_atpx_init(void)
 {
+	int r;
+
 	/* set up the ATPX handle */
-	return radeon_atpx_verify_interface(&radeon_atpx_priv.atpx);
+	r = radeon_atpx_verify_interface(&radeon_atpx_priv.atpx);
+	if (r)
+		return r;
+
+	/* validate the atpx setup */
+	r = radeon_atpx_validate(&radeon_atpx_priv.atpx);
+	if (r)
+		return r;
+
+	return 0;
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c
index 9143fc45e35b..efc4f6441ef4 100644
--- a/drivers/gpu/drm/radeon/radeon_cp.c
+++ b/drivers/gpu/drm/radeon/radeon_cp.c
@@ -27,6 +27,8 @@
  * Authors:
  *    Kevin E. Martin <martin@valinux.com>
  *    Gareth Hughes <gareth@valinux.com>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <linux/module.h>
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 5407459e56d2..70d38241b083 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -29,9 +29,6 @@
 #include "radeon_reg.h"
 #include "radeon.h"
 
-void r100_cs_dump_packet(struct radeon_cs_parser *p,
-			 struct radeon_cs_packet *pkt);
-
 static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 {
 	struct drm_device *ddev = p->rdev->ddev;
@@ -128,18 +125,6 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority
 	return 0;
 }
 
-static void radeon_cs_sync_to(struct radeon_cs_parser *p,
-			      struct radeon_fence *fence)
-{
-	struct radeon_fence *other;
-
-	if (!fence)
-		return;
-
-	other = p->ib.sync_to[fence->ring];
-	p->ib.sync_to[fence->ring] = radeon_fence_later(fence, other);
-}
-
 static void radeon_cs_sync_rings(struct radeon_cs_parser *p)
 {
 	int i;
@@ -148,7 +133,7 @@ static void radeon_cs_sync_rings(struct radeon_cs_parser *p)
 		if (!p->relocs[i].robj)
 			continue;
 
-		radeon_cs_sync_to(p, p->relocs[i].robj->tbo.sync_obj);
+		radeon_ib_sync_to(&p->ib, p->relocs[i].robj->tbo.sync_obj);
 	}
 }
 
@@ -203,7 +188,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 		p->chunks[i].length_dw = user_chunk.length_dw;
 		p->chunks[i].kdata = NULL;
 		p->chunks[i].chunk_id = user_chunk.chunk_id;
-
+		p->chunks[i].user_ptr = (void __user *)(unsigned long)user_chunk.chunk_data;
 		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) {
 			p->chunk_relocs_idx = i;
 		}
@@ -226,9 +211,6 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 				return -EINVAL;
 		}
 
-		p->chunks[i].length_dw = user_chunk.length_dw;
-		p->chunks[i].user_ptr = (void __user *)(unsigned long)user_chunk.chunk_data;
-
 		cdata = (uint32_t *)(unsigned long)user_chunk.chunk_data;
 		if ((p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) ||
 		    (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS)) {
@@ -478,8 +460,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 		goto out;
 	}
 	radeon_cs_sync_rings(parser);
-	radeon_cs_sync_to(parser, vm->fence);
-	radeon_cs_sync_to(parser, radeon_vm_grab_id(rdev, vm, parser->ring));
+	radeon_ib_sync_to(&parser->ib, vm->fence);
+	radeon_ib_sync_to(&parser->ib, radeon_vm_grab_id(
+		rdev, vm, parser->ring));
 
 	if ((rdev->family >= CHIP_TAHITI) &&
 	    (parser->chunk_const_ib_idx != -1)) {
@@ -648,3 +631,152 @@ u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx)
 	idx_value = ibc->kpage[new_page][pg_offset/4];
 	return idx_value;
 }
+
+/**
+ * radeon_cs_packet_parse() - parse cp packet and point ib index to next packet
+ * @parser:	parser structure holding parsing context.
+ * @pkt:	where to store packet information
+ *
+ * Assume that chunk_ib_index is properly set. Will return -EINVAL
+ * if packet is bigger than remaining ib size. or if packets is unknown.
+ **/
+int radeon_cs_packet_parse(struct radeon_cs_parser *p,
+			   struct radeon_cs_packet *pkt,
+			   unsigned idx)
+{
+	struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx];
+	struct radeon_device *rdev = p->rdev;
+	uint32_t header;
+
+	if (idx >= ib_chunk->length_dw) {
+		DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
+			  idx, ib_chunk->length_dw);
+		return -EINVAL;
+	}
+	header = radeon_get_ib_value(p, idx);
+	pkt->idx = idx;
+	pkt->type = RADEON_CP_PACKET_GET_TYPE(header);
+	pkt->count = RADEON_CP_PACKET_GET_COUNT(header);
+	pkt->one_reg_wr = 0;
+	switch (pkt->type) {
+	case RADEON_PACKET_TYPE0:
+		if (rdev->family < CHIP_R600) {
+			pkt->reg = R100_CP_PACKET0_GET_REG(header);
+			pkt->one_reg_wr =
+				RADEON_CP_PACKET0_GET_ONE_REG_WR(header);
+		} else
+			pkt->reg = R600_CP_PACKET0_GET_REG(header);
+		break;
+	case RADEON_PACKET_TYPE3:
+		pkt->opcode = RADEON_CP_PACKET3_GET_OPCODE(header);
+		break;
+	case RADEON_PACKET_TYPE2:
+		pkt->count = -1;
+		break;
+	default:
+		DRM_ERROR("Unknown packet type %d at %d !\n", pkt->type, idx);
+		return -EINVAL;
+	}
+	if ((pkt->count + 1 + pkt->idx) >= ib_chunk->length_dw) {
+		DRM_ERROR("Packet (%d:%d:%d) end after CS buffer (%d) !\n",
+			  pkt->idx, pkt->type, pkt->count, ib_chunk->length_dw);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/**
+ * radeon_cs_packet_next_is_pkt3_nop() - test if the next packet is P3 NOP
+ * @p:		structure holding the parser context.
+ *
+ * Check if the next packet is NOP relocation packet3.
+ **/
+bool radeon_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p)
+{
+	struct radeon_cs_packet p3reloc;
+	int r;
+
+	r = radeon_cs_packet_parse(p, &p3reloc, p->idx);
+	if (r)
+		return false;
+	if (p3reloc.type != RADEON_PACKET_TYPE3)
+		return false;
+	if (p3reloc.opcode != RADEON_PACKET3_NOP)
+		return false;
+	return true;
+}
+
+/**
+ * radeon_cs_dump_packet() - dump raw packet context
+ * @p:		structure holding the parser context.
+ * @pkt:	structure holding the packet.
+ *
+ * Used mostly for debugging and error reporting.
+ **/
+void radeon_cs_dump_packet(struct radeon_cs_parser *p,
+			   struct radeon_cs_packet *pkt)
+{
+	volatile uint32_t *ib;
+	unsigned i;
+	unsigned idx;
+
+	ib = p->ib.ptr;
+	idx = pkt->idx;
+	for (i = 0; i <= (pkt->count + 1); i++, idx++)
+		DRM_INFO("ib[%d]=0x%08X\n", idx, ib[idx]);
+}
+
+/**
+ * radeon_cs_packet_next_reloc() - parse next (should be reloc) packet
+ * @parser:		parser structure holding parsing context.
+ * @data:		pointer to relocation data
+ * @offset_start:	starting offset
+ * @offset_mask:	offset mask (to align start offset on)
+ * @reloc:		reloc informations
+ *
+ * Check if next packet is relocation packet3, do bo validation and compute
+ * GPU offset using the provided start.
+ **/
+int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p,
+				struct radeon_cs_reloc **cs_reloc,
+				int nomm)
+{
+	struct radeon_cs_chunk *relocs_chunk;
+	struct radeon_cs_packet p3reloc;
+	unsigned idx;
+	int r;
+
+	if (p->chunk_relocs_idx == -1) {
+		DRM_ERROR("No relocation chunk !\n");
+		return -EINVAL;
+	}
+	*cs_reloc = NULL;
+	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
+	r = radeon_cs_packet_parse(p, &p3reloc, p->idx);
+	if (r)
+		return r;
+	p->idx += p3reloc.count + 2;
+	if (p3reloc.type != RADEON_PACKET_TYPE3 ||
+	    p3reloc.opcode != RADEON_PACKET3_NOP) {
+		DRM_ERROR("No packet3 for relocation for packet at %d.\n",
+			  p3reloc.idx);
+		radeon_cs_dump_packet(p, &p3reloc);
+		return -EINVAL;
+	}
+	idx = radeon_get_ib_value(p, p3reloc.idx + 1);
+	if (idx >= relocs_chunk->length_dw) {
+		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
+			  idx, relocs_chunk->length_dw);
+		radeon_cs_dump_packet(p, &p3reloc);
+		return -EINVAL;
+	}
+	/* FIXME: we assume reloc size is 4 dwords */
+	if (nomm) {
+		*cs_reloc = p->relocs;
+		(*cs_reloc)->lobj.gpu_offset =
+			(u64)relocs_chunk->kdata[idx + 3] << 32;
+		(*cs_reloc)->lobj.gpu_offset |= relocs_chunk->kdata[idx + 0];
+	} else
+		*cs_reloc = p->relocs_ptr[(idx / 4)];
+	return 0;
+}
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 0d67674b64b1..b097d5b4ff39 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -246,8 +246,14 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 		int i = 0;
 		struct drm_crtc *crtc_p;
 
-		/* avivo cursor image can't end on 128 pixel boundary or
+		/*
+		 * avivo cursor image can't end on 128 pixel boundary or
 		 * go past the end of the frame if both crtcs are enabled
+		 *
+		 * NOTE: It is safe to access crtc->enabled of other crtcs
+		 * without holding either the mode_config lock or the other
+		 * crtc's lock as long as write access to this flag _always_
+		 * grabs all locks.
 		 */
 		list_for_each_entry(crtc_p, &crtc->dev->mode_config.crtc_list, head) {
 			if (crtc_p->enabled)
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 0d6562bb0c93..44b8034a400d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -93,6 +93,7 @@ static const char radeon_family_name[][16] = {
 	"TAHITI",
 	"PITCAIRN",
 	"VERDE",
+	"OLAND",
 	"LAST",
 };
 
@@ -758,6 +759,11 @@ int radeon_atombios_init(struct radeon_device *rdev)
 	atom_card_info->pll_write = cail_pll_write;
 
 	rdev->mode_info.atom_context = atom_parse(atom_card_info, rdev->bios);
+	if (!rdev->mode_info.atom_context) {
+		radeon_atombios_fini(rdev);
+		return -ENOMEM;
+	}
+
 	mutex_init(&rdev->mode_info.atom_context->mutex);
 	radeon_atom_initialize_bios_scratch_regs(rdev->ddev);
 	atom_allocate_fb_scratch(rdev->mode_info.atom_context);
@@ -777,9 +783,11 @@ void radeon_atombios_fini(struct radeon_device *rdev)
 {
 	if (rdev->mode_info.atom_context) {
 		kfree(rdev->mode_info.atom_context->scratch);
-		kfree(rdev->mode_info.atom_context);
 	}
+	kfree(rdev->mode_info.atom_context);
+	rdev->mode_info.atom_context = NULL;
 	kfree(rdev->mode_info.atom_card_info);
+	rdev->mode_info.atom_card_info = NULL;
 }
 
 /* COMBIOS */
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 05c96fa0b051..e38fd559f1ab 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -1089,12 +1089,12 @@ radeon_framebuffer_init(struct drm_device *dev,
 {
 	int ret;
 	rfb->obj = obj;
+	drm_helper_mode_fill_fb_struct(&rfb->base, mode_cmd);
 	ret = drm_framebuffer_init(dev, &rfb->base, &radeon_fb_funcs);
 	if (ret) {
 		rfb->obj = NULL;
 		return ret;
 	}
-	drm_helper_mode_fill_fb_struct(&rfb->base, mode_cmd);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index d9bf96ee299a..167758488ed6 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -118,20 +118,32 @@ int radeon_mode_dumb_create(struct drm_file *file_priv,
 int radeon_mode_dumb_destroy(struct drm_file *file_priv,
 			     struct drm_device *dev,
 			     uint32_t handle);
-struct dma_buf *radeon_gem_prime_export(struct drm_device *dev,
-					struct drm_gem_object *obj,
-					int flags);
-struct drm_gem_object *radeon_gem_prime_import(struct drm_device *dev,
-					       struct dma_buf *dma_buf);
+struct sg_table *radeon_gem_prime_get_sg_table(struct drm_gem_object *obj);
+struct drm_gem_object *radeon_gem_prime_import_sg_table(struct drm_device *dev,
+							size_t size,
+							struct sg_table *sg);
+int radeon_gem_prime_pin(struct drm_gem_object *obj);
+void *radeon_gem_prime_vmap(struct drm_gem_object *obj);
+void radeon_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
+extern long radeon_kms_compat_ioctl(struct file *filp, unsigned int cmd,
+				    unsigned long arg);
 
 #if defined(CONFIG_DEBUG_FS)
 int radeon_debugfs_init(struct drm_minor *minor);
 void radeon_debugfs_cleanup(struct drm_minor *minor);
 #endif
 
+/* atpx handler */
+#if defined(CONFIG_VGA_SWITCHEROO)
+void radeon_register_atpx_handler(void);
+void radeon_unregister_atpx_handler(void);
+#else
+static inline void radeon_register_atpx_handler(void) {}
+static inline void radeon_unregister_atpx_handler(void) {}
+#endif
 
 int radeon_no_wb;
-int radeon_modeset = -1;
+int radeon_modeset = 1;
 int radeon_dynclks = -1;
 int radeon_r4xx_atom = 0;
 int radeon_agpmode = 0;
@@ -199,6 +211,14 @@ module_param_named(msi, radeon_msi, int, 0444);
 MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (defaul 10000 = 10 seconds, 0 = disable)");
 module_param_named(lockup_timeout, radeon_lockup_timeout, int, 0444);
 
+static struct pci_device_id pciidlist[] = {
+	radeon_PCI_IDS
+};
+
+MODULE_DEVICE_TABLE(pci, pciidlist);
+
+#ifdef CONFIG_DRM_RADEON_UMS
+
 static int radeon_suspend(struct drm_device *dev, pm_message_t state)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
@@ -227,14 +247,6 @@ static int radeon_resume(struct drm_device *dev)
 	return 0;
 }
 
-static struct pci_device_id pciidlist[] = {
-	radeon_PCI_IDS
-};
-
-#if defined(CONFIG_DRM_RADEON_KMS)
-MODULE_DEVICE_TABLE(pci, pciidlist);
-#endif
-
 static const struct file_operations radeon_driver_old_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
@@ -284,6 +296,8 @@ static struct drm_driver driver_old = {
 	.patchlevel = DRIVER_PATCHLEVEL,
 };
 
+#endif
+
 static struct drm_driver kms_driver;
 
 static int radeon_kick_out_firmware_fb(struct pci_dev *pdev)
@@ -397,8 +411,13 @@ static struct drm_driver kms_driver = {
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-	.gem_prime_export = radeon_gem_prime_export,
-	.gem_prime_import = radeon_gem_prime_import,
+	.gem_prime_export = drm_gem_prime_export,
+	.gem_prime_import = drm_gem_prime_import,
+	.gem_prime_pin = radeon_gem_prime_pin,
+	.gem_prime_get_sg_table = radeon_gem_prime_get_sg_table,
+	.gem_prime_import_sg_table = radeon_gem_prime_import_sg_table,
+	.gem_prime_vmap = radeon_gem_prime_vmap,
+	.gem_prime_vunmap = radeon_gem_prime_vunmap,
 
 	.name = DRIVER_NAME,
 	.desc = DRIVER_DESC,
@@ -411,10 +430,12 @@ static struct drm_driver kms_driver = {
 static struct drm_driver *driver;
 static struct pci_driver *pdriver;
 
+#ifdef CONFIG_DRM_RADEON_UMS
 static struct pci_driver radeon_pci_driver = {
 	.name = DRIVER_NAME,
 	.id_table = pciidlist,
 };
+#endif
 
 static struct pci_driver radeon_kms_pci_driver = {
 	.name = DRIVER_NAME,
@@ -427,28 +448,6 @@ static struct pci_driver radeon_kms_pci_driver = {
 
 static int __init radeon_init(void)
 {
-	driver = &driver_old;
-	pdriver = &radeon_pci_driver;
-	driver->num_ioctls = radeon_max_ioctl;
-#ifdef CONFIG_VGA_CONSOLE
-	if (vgacon_text_force() && radeon_modeset == -1) {
-		DRM_INFO("VGACON disable radeon kernel modesetting.\n");
-		driver = &driver_old;
-		pdriver = &radeon_pci_driver;
-		driver->driver_features &= ~DRIVER_MODESET;
-		radeon_modeset = 0;
-	}
-#endif
-	/* if enabled by default */
-	if (radeon_modeset == -1) {
-#ifdef CONFIG_DRM_RADEON_KMS
-		DRM_INFO("radeon defaulting to kernel modesetting.\n");
-		radeon_modeset = 1;
-#else
-		DRM_INFO("radeon defaulting to userspace modesetting.\n");
-		radeon_modeset = 0;
-#endif
-	}
 	if (radeon_modeset == 1) {
 		DRM_INFO("radeon kernel modesetting enabled.\n");
 		driver = &kms_driver;
@@ -456,9 +455,21 @@ static int __init radeon_init(void)
 		driver->driver_features |= DRIVER_MODESET;
 		driver->num_ioctls = radeon_max_kms_ioctl;
 		radeon_register_atpx_handler();
+
+	} else {
+#ifdef CONFIG_DRM_RADEON_UMS
+		DRM_INFO("radeon userspace modesetting enabled.\n");
+		driver = &driver_old;
+		pdriver = &radeon_pci_driver;
+		driver->driver_features &= ~DRIVER_MODESET;
+		driver->num_ioctls = radeon_max_ioctl;
+#else
+		DRM_ERROR("No UMS support in radeon module!\n");
+		return -EINVAL;
+#endif
 	}
-	/* if the vga console setting is enabled still
-	 * let modprobe override it */
+
+	/* let modprobe override vga console setting */
 	return drm_pci_init(driver, pdriver);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index e7fdf163a8ca..b369d42f7de5 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -113,6 +113,9 @@
 #define DRIVER_MINOR		33
 #define DRIVER_PATCHLEVEL	0
 
+/* The rest of the file is DEPRECATED! */
+#ifdef CONFIG_DRM_RADEON_UMS
+
 enum radeon_cp_microcode_version {
 	UCODE_R100,
 	UCODE_R200,
@@ -418,8 +421,6 @@ extern int radeon_driver_open(struct drm_device *dev,
 			      struct drm_file *file_priv);
 extern long radeon_compat_ioctl(struct file *filp, unsigned int cmd,
 				unsigned long arg);
-extern long radeon_kms_compat_ioctl(struct file *filp, unsigned int cmd,
-				    unsigned long arg);
 
 extern int radeon_master_create(struct drm_device *dev, struct drm_master *master);
 extern void radeon_master_destroy(struct drm_device *dev, struct drm_master *master);
@@ -462,15 +463,6 @@ extern void r600_blit_swap(struct drm_device *dev,
 			   int sx, int sy, int dx, int dy,
 			   int w, int h, int src_pitch, int dst_pitch, int cpp);
 
-/* atpx handler */
-#if defined(CONFIG_VGA_SWITCHEROO)
-void radeon_register_atpx_handler(void);
-void radeon_unregister_atpx_handler(void);
-#else
-static inline void radeon_register_atpx_handler(void) {}
-static inline void radeon_unregister_atpx_handler(void) {}
-#endif
-
 /* Flags for stats.boxes
  */
 #define RADEON_BOX_DMA_IDLE      0x1
@@ -2167,4 +2159,6 @@ extern void radeon_commit_ring(drm_radeon_private_t *dev_priv);
 } while (0)
 
 
+#endif				/* CONFIG_DRM_RADEON_UMS */
+
 #endif				/* __RADEON_DRV_H__ */
diff --git a/drivers/gpu/drm/radeon/radeon_family.h b/drivers/gpu/drm/radeon/radeon_family.h
index d1fafeabea09..2d91123f2759 100644
--- a/drivers/gpu/drm/radeon/radeon_family.h
+++ b/drivers/gpu/drm/radeon/radeon_family.h
@@ -91,6 +91,7 @@ enum radeon_family {
 	CHIP_TAHITI,
 	CHIP_PITCAIRN,
 	CHIP_VERDE,
+	CHIP_OLAND,
 	CHIP_LAST,
 };
 
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index cc8489d8c6d1..b1746741bc59 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -187,9 +187,10 @@ out_unref:
 	return ret;
 }
 
-static int radeonfb_create(struct radeon_fbdev *rfbdev,
+static int radeonfb_create(struct drm_fb_helper *helper,
 			   struct drm_fb_helper_surface_size *sizes)
 {
+	struct radeon_fbdev *rfbdev = (struct radeon_fbdev *)helper;
 	struct radeon_device *rdev = rfbdev->rdev;
 	struct fb_info *info;
 	struct drm_framebuffer *fb = NULL;
@@ -293,28 +294,13 @@ out_unref:
 	}
 	if (fb && ret) {
 		drm_gem_object_unreference(gobj);
+		drm_framebuffer_unregister_private(fb);
 		drm_framebuffer_cleanup(fb);
 		kfree(fb);
 	}
 	return ret;
 }
 
-static int radeon_fb_find_or_create_single(struct drm_fb_helper *helper,
-					   struct drm_fb_helper_surface_size *sizes)
-{
-	struct radeon_fbdev *rfbdev = (struct radeon_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = radeonfb_create(rfbdev, sizes);
-		if (ret)
-			return ret;
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 void radeon_fb_output_poll_changed(struct radeon_device *rdev)
 {
 	drm_fb_helper_hotplug_event(&rdev->mode_info.rfbdev->helper);
@@ -339,6 +325,7 @@ static int radeon_fbdev_destroy(struct drm_device *dev, struct radeon_fbdev *rfb
 		rfb->obj = NULL;
 	}
 	drm_fb_helper_fini(&rfbdev->helper);
+	drm_framebuffer_unregister_private(&rfb->base);
 	drm_framebuffer_cleanup(&rfb->base);
 
 	return 0;
@@ -347,7 +334,7 @@ static int radeon_fbdev_destroy(struct drm_device *dev, struct radeon_fbdev *rfb
 static struct drm_fb_helper_funcs radeon_fb_helper_funcs = {
 	.gamma_set = radeon_crtc_fb_gamma_set,
 	.gamma_get = radeon_crtc_fb_gamma_get,
-	.fb_probe = radeon_fb_find_or_create_single,
+	.fb_probe = radeonfb_create,
 };
 
 int radeon_fbdev_init(struct radeon_device *rdev)
@@ -377,6 +364,10 @@ int radeon_fbdev_init(struct radeon_device *rdev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&rfbdev->helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(rdev->ddev);
+
 	drm_fb_helper_initial_config(&rfbdev->helper, bpp_sel);
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 6e24f84755b5..2c1341f63dc5 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -929,6 +929,7 @@ uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr)
  */
 static int radeon_vm_update_pdes(struct radeon_device *rdev,
 				 struct radeon_vm *vm,
+				 struct radeon_ib *ib,
 				 uint64_t start, uint64_t end)
 {
 	static const uint32_t incr = RADEON_VM_PTE_COUNT * 8;
@@ -971,7 +972,7 @@ retry:
 		    ((last_pt + incr * count) != pt)) {
 
 			if (count) {
-				radeon_asic_vm_set_page(rdev, last_pde,
+				radeon_asic_vm_set_page(rdev, ib, last_pde,
 							last_pt, count, incr,
 							RADEON_VM_PAGE_VALID);
 			}
@@ -985,7 +986,7 @@ retry:
 	}
 
 	if (count) {
-		radeon_asic_vm_set_page(rdev, last_pde, last_pt, count,
+		radeon_asic_vm_set_page(rdev, ib, last_pde, last_pt, count,
 					incr, RADEON_VM_PAGE_VALID);
 
 	}
@@ -1009,6 +1010,7 @@ retry:
  */
 static void radeon_vm_update_ptes(struct radeon_device *rdev,
 				  struct radeon_vm *vm,
+				  struct radeon_ib *ib,
 				  uint64_t start, uint64_t end,
 				  uint64_t dst, uint32_t flags)
 {
@@ -1038,7 +1040,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 		if ((last_pte + 8 * count) != pte) {
 
 			if (count) {
-				radeon_asic_vm_set_page(rdev, last_pte,
+				radeon_asic_vm_set_page(rdev, ib, last_pte,
 							last_dst, count,
 							RADEON_GPU_PAGE_SIZE,
 							flags);
@@ -1056,7 +1058,8 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 	}
 
 	if (count) {
-		radeon_asic_vm_set_page(rdev, last_pte,	last_dst, count,
+		radeon_asic_vm_set_page(rdev, ib, last_pte,
+					last_dst, count,
 					RADEON_GPU_PAGE_SIZE, flags);
 	}
 }
@@ -1080,8 +1083,7 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 			    struct ttm_mem_reg *mem)
 {
 	unsigned ridx = rdev->asic->vm.pt_ring_index;
-	struct radeon_ring *ring = &rdev->ring[ridx];
-	struct radeon_semaphore *sem = NULL;
+	struct radeon_ib ib;
 	struct radeon_bo_va *bo_va;
 	unsigned nptes, npdes, ndw;
 	uint64_t addr;
@@ -1124,25 +1126,13 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 		bo_va->valid = false;
 	}
 
-	if (vm->fence && radeon_fence_signaled(vm->fence)) {
-		radeon_fence_unref(&vm->fence);
-	}
-
-	if (vm->fence && vm->fence->ring != ridx) {
-		r = radeon_semaphore_create(rdev, &sem);
-		if (r) {
-			return r;
-		}
-	}
-
 	nptes = radeon_bo_ngpu_pages(bo);
 
 	/* assume two extra pdes in case the mapping overlaps the borders */
 	npdes = (nptes >> RADEON_VM_BLOCK_SIZE) + 2;
 
-	/* estimate number of dw needed */
-	/* semaphore, fence and padding */
-	ndw = 32;
+	/* padding, etc. */
+	ndw = 64;
 
 	if (RADEON_VM_BLOCK_SIZE > 11)
 		/* reserve space for one header for every 2k dwords */
@@ -1161,33 +1151,31 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 	/* reserve space for pde addresses */
 	ndw += npdes * 2;
 
-	r = radeon_ring_lock(rdev, ring, ndw);
-	if (r) {
-		return r;
-	}
+	/* update too big for an IB */
+	if (ndw > 0xfffff)
+		return -ENOMEM;
 
-	if (sem && radeon_fence_need_sync(vm->fence, ridx)) {
-		radeon_semaphore_sync_rings(rdev, sem, vm->fence->ring, ridx);
-		radeon_fence_note_sync(vm->fence, ridx);
-	}
+	r = radeon_ib_get(rdev, ridx, &ib, NULL, ndw * 4);
+	ib.length_dw = 0;
 
-	r = radeon_vm_update_pdes(rdev, vm, bo_va->soffset, bo_va->eoffset);
+	r = radeon_vm_update_pdes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset);
 	if (r) {
-		radeon_ring_unlock_undo(rdev, ring);
+		radeon_ib_free(rdev, &ib);
 		return r;
 	}
 
-	radeon_vm_update_ptes(rdev, vm, bo_va->soffset, bo_va->eoffset,
+	radeon_vm_update_ptes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset,
 			      addr, bo_va->flags);
 
-	radeon_fence_unref(&vm->fence);
-	r = radeon_fence_emit(rdev, &vm->fence, ridx);
+	radeon_ib_sync_to(&ib, vm->fence);
+	r = radeon_ib_schedule(rdev, &ib, NULL);
 	if (r) {
-		radeon_ring_unlock_undo(rdev, ring);
+		radeon_ib_free(rdev, &ib);
 		return r;
 	}
-	radeon_ring_unlock_commit(rdev, ring);
-	radeon_semaphore_free(rdev, &sem, vm->fence);
+	radeon_fence_unref(&vm->fence);
+	vm->fence = radeon_fence_ref(ib.fence);
+	radeon_ib_free(rdev, &ib);
 	radeon_fence_unref(&vm->last_flush);
 
 	return 0;
diff --git a/drivers/gpu/drm/radeon/radeon_irq.c b/drivers/gpu/drm/radeon/radeon_irq.c
index e7710339a6a7..8d68e972789a 100644
--- a/drivers/gpu/drm/radeon/radeon_irq.c
+++ b/drivers/gpu/drm/radeon/radeon_irq.c
@@ -28,6 +28,8 @@
  * Authors:
  *    Keith Whitwell <keith@tungstengraphics.com>
  *    Michel D�zer <michel@daenzer.net>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <drm/drmP.h>
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 9c312f9afb68..c75cb2c6ba71 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -185,11 +185,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	if (info->request == RADEON_INFO_TIMESTAMP) {
 		if (rdev->family >= CHIP_R600) {
 			value_ptr64 = (uint64_t*)((unsigned long)info->value);
-			if (rdev->family >= CHIP_TAHITI) {
-				value64 = si_get_gpu_clock(rdev);
-			} else {
-				value64 = r600_get_gpu_clock(rdev);
-			}
+			value64 = radeon_get_gpu_clock_counter(rdev);
 
 			if (DRM_COPY_TO_USER(value_ptr64, &value64, sizeof(value64))) {
 				DRM_ERROR("copy_to_user %s:%u\n", __func__, __LINE__);
@@ -282,7 +278,10 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		break;
 	case RADEON_INFO_CLOCK_CRYSTAL_FREQ:
 		/* return clock value in KHz */
-		value = rdev->clock.spll.reference_freq * 10;
+		if (rdev->asic->get_xclk)
+			value = radeon_get_xclk(rdev) * 10;
+		else
+			value = rdev->clock.spll.reference_freq * 10;
 		break;
 	case RADEON_INFO_NUM_BACKENDS:
 		if (rdev->family >= CHIP_TAHITI)
diff --git a/drivers/gpu/drm/radeon/radeon_mem.c b/drivers/gpu/drm/radeon/radeon_mem.c
index b9f067241633..d54d2d7c9031 100644
--- a/drivers/gpu/drm/radeon/radeon_mem.c
+++ b/drivers/gpu/drm/radeon/radeon_mem.c
@@ -27,6 +27,8 @@
  *
  * Authors:
  *    Keith Whitwell <keith@tungstengraphics.com>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <drm/drmP.h>
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 0bfa656aa87d..338fd6a74e87 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -169,7 +169,7 @@ static void radeon_set_power_state(struct radeon_device *rdev)
 
 		/* starting with BTC, there is one state that is used for both
 		 * MH and SH.  Difference is that we always use the high clock index for
-		 * mclk.
+		 * mclk and vddci.
 		 */
 		if ((rdev->pm.pm_method == PM_METHOD_PROFILE) &&
 		    (rdev->family >= CHIP_BARTS) &&
diff --git a/drivers/gpu/drm/radeon/radeon_prime.c b/drivers/gpu/drm/radeon/radeon_prime.c
index 26c23bb651c6..4940af7e75e6 100644
--- a/drivers/gpu/drm/radeon/radeon_prime.c
+++ b/drivers/gpu/drm/radeon/radeon_prime.c
@@ -28,199 +28,71 @@
 #include "radeon.h"
 #include <drm/radeon_drm.h>
 
-#include <linux/dma-buf.h>
-
-static struct sg_table *radeon_gem_map_dma_buf(struct dma_buf_attachment *attachment,
-					       enum dma_data_direction dir)
+struct sg_table *radeon_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
-	struct radeon_bo *bo = attachment->dmabuf->priv;
-	struct drm_device *dev = bo->rdev->ddev;
+	struct radeon_bo *bo = gem_to_radeon_bo(obj);
 	int npages = bo->tbo.num_pages;
-	struct sg_table *sg;
-	int nents;
-
-	mutex_lock(&dev->struct_mutex);
-	sg = drm_prime_pages_to_sg(bo->tbo.ttm->pages, npages);
-	nents = dma_map_sg(attachment->dev, sg->sgl, sg->nents, dir);
-	mutex_unlock(&dev->struct_mutex);
-	return sg;
-}
-
-static void radeon_gem_unmap_dma_buf(struct dma_buf_attachment *attachment,
-				     struct sg_table *sg, enum dma_data_direction dir)
-{
-	dma_unmap_sg(attachment->dev, sg->sgl, sg->nents, dir);
-	sg_free_table(sg);
-	kfree(sg);
-}
-
-static void radeon_gem_dmabuf_release(struct dma_buf *dma_buf)
-{
-	struct radeon_bo *bo = dma_buf->priv;
-
-	if (bo->gem_base.export_dma_buf == dma_buf) {
-		DRM_ERROR("unreference dmabuf %p\n", &bo->gem_base);
-		bo->gem_base.export_dma_buf = NULL;
-		drm_gem_object_unreference_unlocked(&bo->gem_base);
-	}
-}
-
-static void *radeon_gem_kmap_atomic(struct dma_buf *dma_buf, unsigned long page_num)
-{
-	return NULL;
-}
-
-static void radeon_gem_kunmap_atomic(struct dma_buf *dma_buf, unsigned long page_num, void *addr)
-{
-
-}
-static void *radeon_gem_kmap(struct dma_buf *dma_buf, unsigned long page_num)
-{
-	return NULL;
-}
-
-static void radeon_gem_kunmap(struct dma_buf *dma_buf, unsigned long page_num, void *addr)
-{
 
+	return drm_prime_pages_to_sg(bo->tbo.ttm->pages, npages);
 }
 
-static int radeon_gem_prime_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma)
+void *radeon_gem_prime_vmap(struct drm_gem_object *obj)
 {
-	return -EINVAL;
-}
-
-static void *radeon_gem_prime_vmap(struct dma_buf *dma_buf)
-{
-	struct radeon_bo *bo = dma_buf->priv;
-	struct drm_device *dev = bo->rdev->ddev;
+	struct radeon_bo *bo = gem_to_radeon_bo(obj);
 	int ret;
 
-	mutex_lock(&dev->struct_mutex);
-	if (bo->vmapping_count) {
-		bo->vmapping_count++;
-		goto out_unlock;
-	}
-
 	ret = ttm_bo_kmap(&bo->tbo, 0, bo->tbo.num_pages,
 			  &bo->dma_buf_vmap);
-	if (ret) {
-		mutex_unlock(&dev->struct_mutex);
+	if (ret)
 		return ERR_PTR(ret);
-	}
-	bo->vmapping_count = 1;
-out_unlock:
-	mutex_unlock(&dev->struct_mutex);
+
 	return bo->dma_buf_vmap.virtual;
 }
 
-static void radeon_gem_prime_vunmap(struct dma_buf *dma_buf, void *vaddr)
+void radeon_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
 {
-	struct radeon_bo *bo = dma_buf->priv;
-	struct drm_device *dev = bo->rdev->ddev;
+	struct radeon_bo *bo = gem_to_radeon_bo(obj);
 
-	mutex_lock(&dev->struct_mutex);
-	bo->vmapping_count--;
-	if (bo->vmapping_count == 0) {
-		ttm_bo_kunmap(&bo->dma_buf_vmap);
-	}
-	mutex_unlock(&dev->struct_mutex);
+	ttm_bo_kunmap(&bo->dma_buf_vmap);
 }
-const static struct dma_buf_ops radeon_dmabuf_ops =  {
-	.map_dma_buf = radeon_gem_map_dma_buf,
-	.unmap_dma_buf = radeon_gem_unmap_dma_buf,
-	.release = radeon_gem_dmabuf_release,
-	.kmap = radeon_gem_kmap,
-	.kmap_atomic = radeon_gem_kmap_atomic,
-	.kunmap = radeon_gem_kunmap,
-	.kunmap_atomic = radeon_gem_kunmap_atomic,
-	.mmap = radeon_gem_prime_mmap,
-	.vmap = radeon_gem_prime_vmap,
-	.vunmap = radeon_gem_prime_vunmap,
-};
-
-static int radeon_prime_create(struct drm_device *dev,
-			       size_t size,
-			       struct sg_table *sg,
-			       struct radeon_bo **pbo)
+
+struct drm_gem_object *radeon_gem_prime_import_sg_table(struct drm_device *dev,
+							size_t size,
+							struct sg_table *sg)
 {
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_bo *bo;
 	int ret;
 
 	ret = radeon_bo_create(rdev, size, PAGE_SIZE, false,
-			       RADEON_GEM_DOMAIN_GTT, sg, pbo);
+			       RADEON_GEM_DOMAIN_GTT, sg, &bo);
 	if (ret)
-		return ret;
-	bo = *pbo;
+		return ERR_PTR(ret);
 	bo->gem_base.driver_private = bo;
 
 	mutex_lock(&rdev->gem.mutex);
 	list_add_tail(&bo->list, &rdev->gem.objects);
 	mutex_unlock(&rdev->gem.mutex);
 
-	return 0;
+	return &bo->gem_base;
 }
 
-struct dma_buf *radeon_gem_prime_export(struct drm_device *dev,
-					struct drm_gem_object *obj,
-					int flags)
+int radeon_gem_prime_pin(struct drm_gem_object *obj)
 {
 	struct radeon_bo *bo = gem_to_radeon_bo(obj);
 	int ret = 0;
 
 	ret = radeon_bo_reserve(bo, false);
 	if (unlikely(ret != 0))
-		return ERR_PTR(ret);
+		return ret;
 
 	/* pin buffer into GTT */
 	ret = radeon_bo_pin(bo, RADEON_GEM_DOMAIN_GTT, NULL);
 	if (ret) {
 		radeon_bo_unreserve(bo);
-		return ERR_PTR(ret);
+		return ret;
 	}
 	radeon_bo_unreserve(bo);
-	return dma_buf_export(bo, &radeon_dmabuf_ops, obj->size, flags);
-}
 
-struct drm_gem_object *radeon_gem_prime_import(struct drm_device *dev,
-					       struct dma_buf *dma_buf)
-{
-	struct dma_buf_attachment *attach;
-	struct sg_table *sg;
-	struct radeon_bo *bo;
-	int ret;
-
-	if (dma_buf->ops == &radeon_dmabuf_ops) {
-		bo = dma_buf->priv;
-		if (bo->gem_base.dev == dev) {
-			drm_gem_object_reference(&bo->gem_base);
-			dma_buf_put(dma_buf);
-			return &bo->gem_base;
-		}
-	}
-
-	/* need to attach */
-	attach = dma_buf_attach(dma_buf, dev->dev);
-	if (IS_ERR(attach))
-		return ERR_CAST(attach);
-
-	sg = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL);
-	if (IS_ERR(sg)) {
-		ret = PTR_ERR(sg);
-		goto fail_detach;
-	}
-
-	ret = radeon_prime_create(dev, dma_buf->size, sg, &bo);
-	if (ret)
-		goto fail_unmap;
-
-	bo->gem_base.import_attach = attach;
-
-	return &bo->gem_base;
-
-fail_unmap:
-	dma_buf_unmap_attachment(attach, sg, DMA_BIDIRECTIONAL);
-fail_detach:
-	dma_buf_detach(dma_buf, attach);
-	return ERR_PTR(ret);
+	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h
index 5d8f735d6aaf..7e2c2b7cf188 100644
--- a/drivers/gpu/drm/radeon/radeon_reg.h
+++ b/drivers/gpu/drm/radeon/radeon_reg.h
@@ -3706,4 +3706,19 @@
 
 #define RV530_GB_PIPE_SELECT2           0x4124
 
+#define RADEON_CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
+#define RADEON_CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
+#define RADEON_CP_PACKET0_GET_ONE_REG_WR(h) (((h) >> 15) & 1)
+#define RADEON_CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
+#define R100_CP_PACKET0_GET_REG(h) (((h) & 0x1FFF) << 2)
+#define R600_CP_PACKET0_GET_REG(h) (((h) & 0xFFFF) << 2)
+#define RADEON_PACKET_TYPE0 0
+#define RADEON_PACKET_TYPE1 1
+#define RADEON_PACKET_TYPE2 2
+#define RADEON_PACKET_TYPE3 3
+
+#define RADEON_PACKET3_NOP 0x10
+
+#define RADEON_VLINE_STAT (1 << 12)
+
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index cd72062d5a91..8d58e268ff6d 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -109,6 +109,25 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
 }
 
 /**
+ * radeon_ib_sync_to - sync to fence before executing the IB
+ *
+ * @ib: IB object to add fence to
+ * @fence: fence to sync to
+ *
+ * Sync to the fence before executing the IB
+ */
+void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence)
+{
+	struct radeon_fence *other;
+
+	if (!fence)
+		return;
+
+	other = ib->sync_to[fence->ring];
+	ib->sync_to[fence->ring] = radeon_fence_later(fence, other);
+}
+
+/**
  * radeon_ib_schedule - schedule an IB (Indirect Buffer) on the ring
  *
  * @rdev: radeon_device pointer
diff --git a/drivers/gpu/drm/radeon/radeon_state.c b/drivers/gpu/drm/radeon/radeon_state.c
index 8e9057b6a365..4d20910899d4 100644
--- a/drivers/gpu/drm/radeon/radeon_state.c
+++ b/drivers/gpu/drm/radeon/radeon_state.c
@@ -25,6 +25,8 @@
  * Authors:
  *    Gareth Hughes <gareth@valinux.com>
  *    Kevin E. Martin <martin@valinux.com>
+ *
+ * ------------------------ This file is DEPRECATED! -------------------------
  */
 
 #include <drm/drmP.h>
diff --git a/drivers/gpu/drm/radeon/rv515d.h b/drivers/gpu/drm/radeon/rv515d.h
index 590309a710b1..6927a200daf4 100644
--- a/drivers/gpu/drm/radeon/rv515d.h
+++ b/drivers/gpu/drm/radeon/rv515d.h
@@ -205,17 +205,6 @@
 			 REG_SET(PACKET3_IT_OPCODE, (op)) |		\
 			 REG_SET(PACKET3_COUNT, (n)))
 
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0x1FFF) << 2)
-#define CP_PACKET0_GET_ONE_REG_WR(h) (((h) >> 15) & 1)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-
 /* Registers */
 #define R_0000F0_RBBM_SOFT_RESET                     0x0000F0
 #define   S_0000F0_SOFT_RESET_CP(x)                    (((x) & 0x1) << 0)
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 1b2444f4d8f4..d63fe1d0f53f 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -43,6 +43,31 @@ static void rv770_gpu_init(struct radeon_device *rdev);
 void rv770_fini(struct radeon_device *rdev);
 static void rv770_pcie_gen2_enable(struct radeon_device *rdev);
 
+#define PCIE_BUS_CLK                10000
+#define TCLK                        (PCIE_BUS_CLK / 10)
+
+/**
+ * rv770_get_xclk - get the xclk
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Returns the reference clock used by the gfx engine
+ * (r7xx-cayman).
+ */
+u32 rv770_get_xclk(struct radeon_device *rdev)
+{
+	u32 reference_clock = rdev->clock.spll.reference_freq;
+	u32 tmp = RREG32(CG_CLKPIN_CNTL);
+
+	if (tmp & MUX_TCLK_TO_XCLK)
+		return TCLK;
+
+	if (tmp & XTALIN_DIVIDE)
+		return reference_clock / 4;
+
+	return reference_clock;
+}
+
 u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
diff --git a/drivers/gpu/drm/radeon/rv770d.h b/drivers/gpu/drm/radeon/rv770d.h
index 20e29d23d348..c55f950a4af7 100644
--- a/drivers/gpu/drm/radeon/rv770d.h
+++ b/drivers/gpu/drm/radeon/rv770d.h
@@ -128,6 +128,10 @@
 #define		GUI_ACTIVE					(1<<31)
 #define	GRBM_STATUS2					0x8014
 
+#define CG_CLKPIN_CNTL                                    0x660
+#       define MUX_TCLK_TO_XCLK                           (1 << 8)
+#       define XTALIN_DIVIDE                              (1 << 9)
+
 #define	CG_MULT_THERMAL_STATUS				0x740
 #define		ASIC_T(x)			        ((x) << 16)
 #define		ASIC_T_MASK			        0x3FF0000
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index ae8b48205a6c..80979ed951eb 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -38,6 +38,7 @@
 #define SI_CE_UCODE_SIZE 2144
 #define SI_RLC_UCODE_SIZE 2048
 #define SI_MC_UCODE_SIZE 7769
+#define OLAND_MC_UCODE_SIZE 7863
 
 MODULE_FIRMWARE("radeon/TAHITI_pfp.bin");
 MODULE_FIRMWARE("radeon/TAHITI_me.bin");
@@ -54,6 +55,11 @@ MODULE_FIRMWARE("radeon/VERDE_me.bin");
 MODULE_FIRMWARE("radeon/VERDE_ce.bin");
 MODULE_FIRMWARE("radeon/VERDE_mc.bin");
 MODULE_FIRMWARE("radeon/VERDE_rlc.bin");
+MODULE_FIRMWARE("radeon/OLAND_pfp.bin");
+MODULE_FIRMWARE("radeon/OLAND_me.bin");
+MODULE_FIRMWARE("radeon/OLAND_ce.bin");
+MODULE_FIRMWARE("radeon/OLAND_mc.bin");
+MODULE_FIRMWARE("radeon/OLAND_rlc.bin");
 
 extern int r600_ih_ring_alloc(struct radeon_device *rdev);
 extern void r600_ih_ring_fini(struct radeon_device *rdev);
@@ -61,6 +67,35 @@ extern void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev);
 extern void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save);
 extern void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save);
 extern u32 evergreen_get_number_of_dram_channels(struct radeon_device *rdev);
+extern void evergreen_print_gpu_status_regs(struct radeon_device *rdev);
+extern bool evergreen_is_display_hung(struct radeon_device *rdev);
+
+#define PCIE_BUS_CLK                10000
+#define TCLK                        (PCIE_BUS_CLK / 10)
+
+/**
+ * si_get_xclk - get the xclk
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Returns the reference clock used by the gfx engine
+ * (SI).
+ */
+u32 si_get_xclk(struct radeon_device *rdev)
+{
+        u32 reference_clock = rdev->clock.spll.reference_freq;
+	u32 tmp;
+
+	tmp = RREG32(CG_CLKPIN_CNTL_2);
+	if (tmp & MUX_TCLK_TO_XCLK)
+		return TCLK;
+
+	tmp = RREG32(CG_CLKPIN_CNTL);
+	if (tmp & XTALIN_DIVIDE)
+		return reference_clock / 4;
+
+	return reference_clock;
+}
 
 /* get temperature in millidegrees */
 int si_get_temp(struct radeon_device *rdev)
@@ -200,6 +235,45 @@ static const u32 verde_io_mc_regs[TAHITI_IO_MC_REGS_SIZE][2] = {
 	{0x0000009f, 0x00a37400}
 };
 
+static const u32 oland_io_mc_regs[TAHITI_IO_MC_REGS_SIZE][2] = {
+	{0x0000006f, 0x03044000},
+	{0x00000070, 0x0480c018},
+	{0x00000071, 0x00000040},
+	{0x00000072, 0x01000000},
+	{0x00000074, 0x000000ff},
+	{0x00000075, 0x00143400},
+	{0x00000076, 0x08ec0800},
+	{0x00000077, 0x040000cc},
+	{0x00000079, 0x00000000},
+	{0x0000007a, 0x21000409},
+	{0x0000007c, 0x00000000},
+	{0x0000007d, 0xe8000000},
+	{0x0000007e, 0x044408a8},
+	{0x0000007f, 0x00000003},
+	{0x00000080, 0x00000000},
+	{0x00000081, 0x01000000},
+	{0x00000082, 0x02000000},
+	{0x00000083, 0x00000000},
+	{0x00000084, 0xe3f3e4f4},
+	{0x00000085, 0x00052024},
+	{0x00000087, 0x00000000},
+	{0x00000088, 0x66036603},
+	{0x00000089, 0x01000000},
+	{0x0000008b, 0x1c0a0000},
+	{0x0000008c, 0xff010000},
+	{0x0000008e, 0xffffefff},
+	{0x0000008f, 0xfff3efff},
+	{0x00000090, 0xfff3efbf},
+	{0x00000094, 0x00101101},
+	{0x00000095, 0x00000fff},
+	{0x00000096, 0x00116fff},
+	{0x00000097, 0x60010000},
+	{0x00000098, 0x10010000},
+	{0x00000099, 0x00006000},
+	{0x0000009a, 0x00001000},
+	{0x0000009f, 0x00a17730}
+};
+
 /* ucode loading */
 static int si_mc_load_microcode(struct radeon_device *rdev)
 {
@@ -228,6 +302,11 @@ static int si_mc_load_microcode(struct radeon_device *rdev)
 		ucode_size = SI_MC_UCODE_SIZE;
 		regs_size = TAHITI_IO_MC_REGS_SIZE;
 		break;
+	case CHIP_OLAND:
+		io_mc_regs = (u32 *)&oland_io_mc_regs;
+		ucode_size = OLAND_MC_UCODE_SIZE;
+		regs_size = TAHITI_IO_MC_REGS_SIZE;
+		break;
 	}
 
 	running = RREG32(MC_SEQ_SUP_CNTL) & RUN_MASK;
@@ -322,6 +401,15 @@ static int si_init_microcode(struct radeon_device *rdev)
 		rlc_req_size = SI_RLC_UCODE_SIZE * 4;
 		mc_req_size = SI_MC_UCODE_SIZE * 4;
 		break;
+	case CHIP_OLAND:
+		chip_name = "OLAND";
+		rlc_chip_name = "OLAND";
+		pfp_req_size = SI_PFP_UCODE_SIZE * 4;
+		me_req_size = SI_PM4_UCODE_SIZE * 4;
+		ce_req_size = SI_CE_UCODE_SIZE * 4;
+		rlc_req_size = SI_RLC_UCODE_SIZE * 4;
+		mc_req_size = OLAND_MC_UCODE_SIZE * 4;
+		break;
 	default: BUG();
 	}
 
@@ -1125,7 +1213,8 @@ static void si_tiling_mode_table_init(struct radeon_device *rdev)
 			}
 			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
 		}
-	} else if (rdev->family == CHIP_VERDE) {
+	} else if ((rdev->family == CHIP_VERDE) ||
+		   (rdev->family == CHIP_OLAND)) {
 		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
 			switch (reg_offset) {
 			case 0:  /* non-AA compressed depth or any compressed stencil */
@@ -1570,6 +1659,23 @@ static void si_gpu_init(struct radeon_device *rdev)
 		rdev->config.si.sc_earlyz_tile_fifo_size = 0x130;
 		gb_addr_config = VERDE_GB_ADDR_CONFIG_GOLDEN;
 		break;
+	case CHIP_OLAND:
+		rdev->config.si.max_shader_engines = 1;
+		rdev->config.si.max_tile_pipes = 4;
+		rdev->config.si.max_cu_per_sh = 6;
+		rdev->config.si.max_sh_per_se = 1;
+		rdev->config.si.max_backends_per_se = 2;
+		rdev->config.si.max_texture_channel_caches = 4;
+		rdev->config.si.max_gprs = 256;
+		rdev->config.si.max_gs_threads = 16;
+		rdev->config.si.max_hw_contexts = 8;
+
+		rdev->config.si.sc_prim_fifo_size_frontend = 0x20;
+		rdev->config.si.sc_prim_fifo_size_backend = 0x40;
+		rdev->config.si.sc_hiz_tile_fifo_size = 0x30;
+		rdev->config.si.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = VERDE_GB_ADDR_CONFIG_GOLDEN;
+		break;
 	}
 
 	/* Initialize HDP */
@@ -2106,154 +2212,275 @@ static int si_cp_resume(struct radeon_device *rdev)
 	return 0;
 }
 
-bool si_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+static u32 si_gpu_check_soft_reset(struct radeon_device *rdev)
 {
-	u32 srbm_status;
-	u32 grbm_status, grbm_status2;
-	u32 grbm_status_se0, grbm_status_se1;
-
-	srbm_status = RREG32(SRBM_STATUS);
-	grbm_status = RREG32(GRBM_STATUS);
-	grbm_status2 = RREG32(GRBM_STATUS2);
-	grbm_status_se0 = RREG32(GRBM_STATUS_SE0);
-	grbm_status_se1 = RREG32(GRBM_STATUS_SE1);
-	if (!(grbm_status & GUI_ACTIVE)) {
-		radeon_ring_lockup_update(ring);
-		return false;
-	}
-	/* force CP activities */
-	radeon_ring_force_activity(rdev, ring);
-	return radeon_ring_test_lockup(rdev, ring);
-}
+	u32 reset_mask = 0;
+	u32 tmp;
 
-static void si_gpu_soft_reset_gfx(struct radeon_device *rdev)
-{
-	u32 grbm_reset = 0;
+	/* GRBM_STATUS */
+	tmp = RREG32(GRBM_STATUS);
+	if (tmp & (PA_BUSY | SC_BUSY |
+		   BCI_BUSY | SX_BUSY |
+		   TA_BUSY | VGT_BUSY |
+		   DB_BUSY | CB_BUSY |
+		   GDS_BUSY | SPI_BUSY |
+		   IA_BUSY | IA_BUSY_NO_DMA))
+		reset_mask |= RADEON_RESET_GFX;
 
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		return;
+	if (tmp & (CF_RQ_PENDING | PF_RQ_PENDING |
+		   CP_BUSY | CP_COHERENCY_BUSY))
+		reset_mask |= RADEON_RESET_CP;
 
-	dev_info(rdev->dev, "  GRBM_STATUS=0x%08X\n",
-		RREG32(GRBM_STATUS));
-	dev_info(rdev->dev, "  GRBM_STATUS2=0x%08X\n",
-		RREG32(GRBM_STATUS2));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE0=0x%08X\n",
-		RREG32(GRBM_STATUS_SE0));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE1=0x%08X\n",
-		RREG32(GRBM_STATUS_SE1));
-	dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
-		RREG32(SRBM_STATUS));
+	if (tmp & GRBM_EE_BUSY)
+		reset_mask |= RADEON_RESET_GRBM | RADEON_RESET_GFX | RADEON_RESET_CP;
 
-	/* Disable CP parsing/prefetching */
-	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT | CP_CE_HALT);
+	/* GRBM_STATUS2 */
+	tmp = RREG32(GRBM_STATUS2);
+	if (tmp & (RLC_RQ_PENDING | RLC_BUSY))
+		reset_mask |= RADEON_RESET_RLC;
 
-	/* reset all the gfx blocks */
-	grbm_reset = (SOFT_RESET_CP |
-		      SOFT_RESET_CB |
-		      SOFT_RESET_DB |
-		      SOFT_RESET_GDS |
-		      SOFT_RESET_PA |
-		      SOFT_RESET_SC |
-		      SOFT_RESET_BCI |
-		      SOFT_RESET_SPI |
-		      SOFT_RESET_SX |
-		      SOFT_RESET_TC |
-		      SOFT_RESET_TA |
-		      SOFT_RESET_VGT |
-		      SOFT_RESET_IA);
-
-	dev_info(rdev->dev, "  GRBM_SOFT_RESET=0x%08X\n", grbm_reset);
-	WREG32(GRBM_SOFT_RESET, grbm_reset);
-	(void)RREG32(GRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(GRBM_SOFT_RESET, 0);
-	(void)RREG32(GRBM_SOFT_RESET);
-
-	dev_info(rdev->dev, "  GRBM_STATUS=0x%08X\n",
-		RREG32(GRBM_STATUS));
-	dev_info(rdev->dev, "  GRBM_STATUS2=0x%08X\n",
-		RREG32(GRBM_STATUS2));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE0=0x%08X\n",
-		RREG32(GRBM_STATUS_SE0));
-	dev_info(rdev->dev, "  GRBM_STATUS_SE1=0x%08X\n",
-		RREG32(GRBM_STATUS_SE1));
-	dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
-		RREG32(SRBM_STATUS));
-}
+	/* DMA_STATUS_REG 0 */
+	tmp = RREG32(DMA_STATUS_REG + DMA0_REGISTER_OFFSET);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA;
 
-static void si_gpu_soft_reset_dma(struct radeon_device *rdev)
-{
-	u32 tmp;
+	/* DMA_STATUS_REG 1 */
+	tmp = RREG32(DMA_STATUS_REG + DMA1_REGISTER_OFFSET);
+	if (!(tmp & DMA_IDLE))
+		reset_mask |= RADEON_RESET_DMA1;
 
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		return;
+	/* SRBM_STATUS2 */
+	tmp = RREG32(SRBM_STATUS2);
+	if (tmp & DMA_BUSY)
+		reset_mask |= RADEON_RESET_DMA;
 
-	dev_info(rdev->dev, "  DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	if (tmp & DMA1_BUSY)
+		reset_mask |= RADEON_RESET_DMA1;
 
-	/* dma0 */
-	tmp = RREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET, tmp);
+	/* SRBM_STATUS */
+	tmp = RREG32(SRBM_STATUS);
 
-	/* dma1 */
-	tmp = RREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET);
-	tmp &= ~DMA_RB_ENABLE;
-	WREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET, tmp);
+	if (tmp & IH_BUSY)
+		reset_mask |= RADEON_RESET_IH;
 
-	/* Reset dma */
-	WREG32(SRBM_SOFT_RESET, SOFT_RESET_DMA | SOFT_RESET_DMA1);
-	RREG32(SRBM_SOFT_RESET);
-	udelay(50);
-	WREG32(SRBM_SOFT_RESET, 0);
+	if (tmp & SEM_BUSY)
+		reset_mask |= RADEON_RESET_SEM;
+
+	if (tmp & GRBM_RQ_PENDING)
+		reset_mask |= RADEON_RESET_GRBM;
+
+	if (tmp & VMC_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
 
-	dev_info(rdev->dev, "  DMA_STATUS_REG   = 0x%08X\n",
-		RREG32(DMA_STATUS_REG));
+	if (tmp & (MCB_BUSY | MCB_NON_DISPLAY_BUSY |
+		   MCC_BUSY | MCD_BUSY))
+		reset_mask |= RADEON_RESET_MC;
+
+	if (evergreen_is_display_hung(rdev))
+		reset_mask |= RADEON_RESET_DISPLAY;
+
+	/* VM_L2_STATUS */
+	tmp = RREG32(VM_L2_STATUS);
+	if (tmp & L2_BUSY)
+		reset_mask |= RADEON_RESET_VMC;
+
+	return reset_mask;
 }
 
-static int si_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
+static void si_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
 {
 	struct evergreen_mc_save save;
-
-	if (!(RREG32(GRBM_STATUS) & GUI_ACTIVE))
-		reset_mask &= ~(RADEON_RESET_GFX | RADEON_RESET_COMPUTE);
-
-	if (RREG32(DMA_STATUS_REG) & DMA_IDLE)
-		reset_mask &= ~RADEON_RESET_DMA;
+	u32 grbm_soft_reset = 0, srbm_soft_reset = 0;
+	u32 tmp;
 
 	if (reset_mask == 0)
-		return 0;
+		return;
 
 	dev_info(rdev->dev, "GPU softreset: 0x%08X\n", reset_mask);
 
+	evergreen_print_gpu_status_regs(rdev);
 	dev_info(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
 		 RREG32(VM_CONTEXT1_PROTECTION_FAULT_ADDR));
 	dev_info(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 		 RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS));
 
+	/* Disable CP parsing/prefetching */
+	WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT | CP_CE_HALT);
+
+	if (reset_mask & RADEON_RESET_DMA) {
+		/* dma0 */
+		tmp = RREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL + DMA0_REGISTER_OFFSET, tmp);
+	}
+	if (reset_mask & RADEON_RESET_DMA1) {
+		/* dma1 */
+		tmp = RREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET);
+		tmp &= ~DMA_RB_ENABLE;
+		WREG32(DMA_RB_CNTL + DMA1_REGISTER_OFFSET, tmp);
+	}
+
+	udelay(50);
+
 	evergreen_mc_stop(rdev, &save);
-	if (radeon_mc_wait_for_idle(rdev)) {
+	if (evergreen_mc_wait_for_idle(rdev)) {
 		dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
 	}
 
-	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE))
-		si_gpu_soft_reset_gfx(rdev);
+	if (reset_mask & (RADEON_RESET_GFX | RADEON_RESET_COMPUTE | RADEON_RESET_CP)) {
+		grbm_soft_reset = SOFT_RESET_CB |
+			SOFT_RESET_DB |
+			SOFT_RESET_GDS |
+			SOFT_RESET_PA |
+			SOFT_RESET_SC |
+			SOFT_RESET_BCI |
+			SOFT_RESET_SPI |
+			SOFT_RESET_SX |
+			SOFT_RESET_TC |
+			SOFT_RESET_TA |
+			SOFT_RESET_VGT |
+			SOFT_RESET_IA;
+	}
+
+	if (reset_mask & RADEON_RESET_CP) {
+		grbm_soft_reset |= SOFT_RESET_CP | SOFT_RESET_VGT;
+
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+	}
 
 	if (reset_mask & RADEON_RESET_DMA)
-		si_gpu_soft_reset_dma(rdev);
+		srbm_soft_reset |= SOFT_RESET_DMA;
+
+	if (reset_mask & RADEON_RESET_DMA1)
+		srbm_soft_reset |= SOFT_RESET_DMA1;
+
+	if (reset_mask & RADEON_RESET_DISPLAY)
+		srbm_soft_reset |= SOFT_RESET_DC;
+
+	if (reset_mask & RADEON_RESET_RLC)
+		grbm_soft_reset |= SOFT_RESET_RLC;
+
+	if (reset_mask & RADEON_RESET_SEM)
+		srbm_soft_reset |= SOFT_RESET_SEM;
+
+	if (reset_mask & RADEON_RESET_IH)
+		srbm_soft_reset |= SOFT_RESET_IH;
+
+	if (reset_mask & RADEON_RESET_GRBM)
+		srbm_soft_reset |= SOFT_RESET_GRBM;
+
+	if (reset_mask & RADEON_RESET_VMC)
+		srbm_soft_reset |= SOFT_RESET_VMC;
+
+	if (reset_mask & RADEON_RESET_MC)
+		srbm_soft_reset |= SOFT_RESET_MC;
+
+	if (grbm_soft_reset) {
+		tmp = RREG32(GRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(rdev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~grbm_soft_reset;
+		WREG32(GRBM_SOFT_RESET, tmp);
+		tmp = RREG32(GRBM_SOFT_RESET);
+	}
+
+	if (srbm_soft_reset) {
+		tmp = RREG32(SRBM_SOFT_RESET);
+		tmp |= srbm_soft_reset;
+		dev_info(rdev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+
+		udelay(50);
+
+		tmp &= ~srbm_soft_reset;
+		WREG32(SRBM_SOFT_RESET, tmp);
+		tmp = RREG32(SRBM_SOFT_RESET);
+	}
 
 	/* Wait a little for things to settle down */
 	udelay(50);
 
 	evergreen_mc_resume(rdev, &save);
-	return 0;
+	udelay(50);
+
+	evergreen_print_gpu_status_regs(rdev);
 }
 
 int si_asic_reset(struct radeon_device *rdev)
 {
-	return si_gpu_soft_reset(rdev, (RADEON_RESET_GFX |
-					RADEON_RESET_COMPUTE |
-					RADEON_RESET_DMA));
+	u32 reset_mask;
+
+	reset_mask = si_gpu_check_soft_reset(rdev);
+
+	if (reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, true);
+
+	si_gpu_soft_reset(rdev, reset_mask);
+
+	reset_mask = si_gpu_check_soft_reset(rdev);
+
+	if (!reset_mask)
+		r600_set_bios_scratch_engine_hung(rdev, false);
+
+	return 0;
+}
+
+/**
+ * si_gfx_is_lockup - Check if the GFX engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the GFX engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool si_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 reset_mask = si_gpu_check_soft_reset(rdev);
+
+	if (!(reset_mask & (RADEON_RESET_GFX |
+			    RADEON_RESET_COMPUTE |
+			    RADEON_RESET_CP))) {
+		radeon_ring_lockup_update(ring);
+		return false;
+	}
+	/* force CP activities */
+	radeon_ring_force_activity(rdev, ring);
+	return radeon_ring_test_lockup(rdev, ring);
+}
+
+/**
+ * si_dma_is_lockup - Check if the DMA engine is locked up
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if the async DMA engine is locked up.
+ * Returns true if the engine appears to be locked up, false if not.
+ */
+bool si_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 reset_mask = si_gpu_check_soft_reset(rdev);
+	u32 mask;
+
+	if (ring->idx == R600_RING_TYPE_DMA_INDEX)
+		mask = RADEON_RESET_DMA;
+	else
+		mask = RADEON_RESET_DMA1;
+
+	if (!(reset_mask & mask)) {
+		radeon_ring_lockup_update(ring);
+		return false;
+	}
+	/* force ring activities */
+	radeon_ring_force_activity(rdev, ring);
+	return radeon_ring_test_lockup(rdev, ring);
 }
 
 /* MC */
@@ -2855,19 +3082,19 @@ int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 
 	do {
 		pkt.idx = idx;
-		pkt.type = CP_PACKET_GET_TYPE(ib->ptr[idx]);
-		pkt.count = CP_PACKET_GET_COUNT(ib->ptr[idx]);
+		pkt.type = RADEON_CP_PACKET_GET_TYPE(ib->ptr[idx]);
+		pkt.count = RADEON_CP_PACKET_GET_COUNT(ib->ptr[idx]);
 		pkt.one_reg_wr = 0;
 		switch (pkt.type) {
-		case PACKET_TYPE0:
+		case RADEON_PACKET_TYPE0:
 			dev_err(rdev->dev, "Packet0 not allowed!\n");
 			ret = -EINVAL;
 			break;
-		case PACKET_TYPE2:
+		case RADEON_PACKET_TYPE2:
 			idx += 1;
 			break;
-		case PACKET_TYPE3:
-			pkt.opcode = CP_PACKET3_GET_OPCODE(ib->ptr[idx]);
+		case RADEON_PACKET_TYPE3:
+			pkt.opcode = RADEON_CP_PACKET3_GET_OPCODE(ib->ptr[idx]);
 			if (ib->is_const_ib)
 				ret = si_vm_packet3_ce_check(rdev, ib->ptr, &pkt);
 			else {
@@ -2920,19 +3147,21 @@ void si_vm_fini(struct radeon_device *rdev)
  * si_vm_set_page - update the page tables using the CP
  *
  * @rdev: radeon_device pointer
+ * @ib: indirect buffer to fill with commands
  * @pe: addr of the page entry
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
  * @flags: access flags
  *
- * Update the page tables using the CP (cayman-si).
+ * Update the page tables using the CP (SI).
  */
-void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
+void si_vm_set_page(struct radeon_device *rdev,
+		    struct radeon_ib *ib,
+		    uint64_t pe,
 		    uint64_t addr, unsigned count,
 		    uint32_t incr, uint32_t flags)
 {
-	struct radeon_ring *ring = &rdev->ring[rdev->asic->vm.pt_ring_index];
 	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
 	uint64_t value;
 	unsigned ndw;
@@ -2943,11 +3172,11 @@ void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 			if (ndw > 0x3FFE)
 				ndw = 0x3FFE;
 
-			radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, ndw));
-			radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
-						 WRITE_DATA_DST_SEL(1)));
-			radeon_ring_write(ring, pe);
-			radeon_ring_write(ring, upper_32_bits(pe));
+			ib->ptr[ib->length_dw++] = PACKET3(PACKET3_WRITE_DATA, ndw);
+			ib->ptr[ib->length_dw++] = (WRITE_DATA_ENGINE_SEL(0) |
+					WRITE_DATA_DST_SEL(1));
+			ib->ptr[ib->length_dw++] = pe;
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
 			for (; ndw > 2; ndw -= 2, --count, pe += 8) {
 				if (flags & RADEON_VM_PAGE_SYSTEM) {
 					value = radeon_vm_map_gart(rdev, addr);
@@ -2959,8 +3188,8 @@ void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 				}
 				addr += incr;
 				value |= r600_flags;
-				radeon_ring_write(ring, value);
-				radeon_ring_write(ring, upper_32_bits(value));
+				ib->ptr[ib->length_dw++] = value;
+				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
 		}
 	} else {
@@ -2972,9 +3201,9 @@ void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 					ndw = 0xFFFFE;
 
 				/* for non-physically contiguous pages (system) */
-				radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 0, ndw));
-				radeon_ring_write(ring, pe);
-				radeon_ring_write(ring, upper_32_bits(pe) & 0xff);
+				ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 0, ndw);
+				ib->ptr[ib->length_dw++] = pe;
+				ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
 				for (; ndw > 0; ndw -= 2, --count, pe += 8) {
 					if (flags & RADEON_VM_PAGE_SYSTEM) {
 						value = radeon_vm_map_gart(rdev, addr);
@@ -2986,8 +3215,8 @@ void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 					}
 					addr += incr;
 					value |= r600_flags;
-					radeon_ring_write(ring, value);
-					radeon_ring_write(ring, upper_32_bits(value));
+					ib->ptr[ib->length_dw++] = value;
+					ib->ptr[ib->length_dw++] = upper_32_bits(value);
 				}
 			}
 		} else {
@@ -3001,20 +3230,22 @@ void si_vm_set_page(struct radeon_device *rdev, uint64_t pe,
 				else
 					value = 0;
 				/* for physically contiguous pages (vram) */
-				radeon_ring_write(ring, DMA_PTE_PDE_PACKET(ndw));
-				radeon_ring_write(ring, pe); /* dst addr */
-				radeon_ring_write(ring, upper_32_bits(pe) & 0xff);
-				radeon_ring_write(ring, r600_flags); /* mask */
-				radeon_ring_write(ring, 0);
-				radeon_ring_write(ring, value); /* value */
-				radeon_ring_write(ring, upper_32_bits(value));
-				radeon_ring_write(ring, incr); /* increment size */
-				radeon_ring_write(ring, 0);
+				ib->ptr[ib->length_dw++] = DMA_PTE_PDE_PACKET(ndw);
+				ib->ptr[ib->length_dw++] = pe; /* dst addr */
+				ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
+				ib->ptr[ib->length_dw++] = r600_flags; /* mask */
+				ib->ptr[ib->length_dw++] = 0;
+				ib->ptr[ib->length_dw++] = value; /* value */
+				ib->ptr[ib->length_dw++] = upper_32_bits(value);
+				ib->ptr[ib->length_dw++] = incr; /* increment size */
+				ib->ptr[ib->length_dw++] = 0;
 				pe += ndw * 4;
 				addr += (ndw / 2) * incr;
 				count -= ndw / 2;
 			}
 		}
+		while (ib->length_dw & 0x7)
+			ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_NOP, 0, 0, 0, 0);
 	}
 }
 
@@ -4378,14 +4609,14 @@ void si_fini(struct radeon_device *rdev)
 }
 
 /**
- * si_get_gpu_clock - return GPU clock counter snapshot
+ * si_get_gpu_clock_counter - return GPU clock counter snapshot
  *
  * @rdev: radeon_device pointer
  *
  * Fetches a GPU clock counter snapshot (SI).
  * Returns the 64 bit clock counter snapshot.
  */
-uint64_t si_get_gpu_clock(struct radeon_device *rdev)
+uint64_t si_get_gpu_clock_counter(struct radeon_device *rdev)
 {
 	uint64_t clock;
 
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index c056aae814f0..23fc08fc8e7f 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -58,9 +58,22 @@
 #define VGA_HDP_CONTROL  				0x328
 #define		VGA_MEMORY_DISABLE				(1 << 4)
 
+#define CG_CLKPIN_CNTL                                    0x660
+#       define XTALIN_DIVIDE                              (1 << 1)
+#define CG_CLKPIN_CNTL_2                                  0x664
+#       define MUX_TCLK_TO_XCLK                           (1 << 8)
+
 #define DMIF_ADDR_CONFIG  				0xBD4
 
 #define	SRBM_STATUS				        0xE50
+#define		GRBM_RQ_PENDING 			(1 << 5)
+#define		VMC_BUSY 				(1 << 8)
+#define		MCB_BUSY 				(1 << 9)
+#define		MCB_NON_DISPLAY_BUSY 			(1 << 10)
+#define		MCC_BUSY 				(1 << 11)
+#define		MCD_BUSY 				(1 << 12)
+#define		SEM_BUSY 				(1 << 14)
+#define		IH_BUSY 				(1 << 17)
 
 #define	SRBM_SOFT_RESET				        0x0E60
 #define		SOFT_RESET_BIF				(1 << 1)
@@ -81,6 +94,10 @@
 #define	CC_SYS_RB_BACKEND_DISABLE			0xe80
 #define	GC_USER_SYS_RB_BACKEND_DISABLE			0xe84
 
+#define	SRBM_STATUS2				        0x0EC4
+#define		DMA_BUSY 				(1 << 5)
+#define		DMA1_BUSY 				(1 << 6)
+
 #define VM_L2_CNTL					0x1400
 #define		ENABLE_L2_CACHE					(1 << 0)
 #define		ENABLE_L2_FRAGMENT_PROCESSING			(1 << 1)
@@ -783,16 +800,7 @@
 /*
  * PM4
  */
-#define	PACKET_TYPE0	0
-#define	PACKET_TYPE1	1
-#define	PACKET_TYPE2	2
-#define	PACKET_TYPE3	3
-
-#define CP_PACKET_GET_TYPE(h) (((h) >> 30) & 3)
-#define CP_PACKET_GET_COUNT(h) (((h) >> 16) & 0x3FFF)
-#define CP_PACKET0_GET_REG(h) (((h) & 0xFFFF) << 2)
-#define CP_PACKET3_GET_OPCODE(h) (((h) >> 8) & 0xFF)
-#define PACKET0(reg, n)	((PACKET_TYPE0 << 30) |				\
+#define PACKET0(reg, n)	((RADEON_PACKET_TYPE0 << 30) |			\
 			 (((reg) >> 2) & 0xFFFF) |			\
 			 ((n) & 0x3FFF) << 16)
 #define CP_PACKET2			0x80000000
@@ -801,7 +809,7 @@
 
 #define PACKET2(v)	(CP_PACKET2 | REG_SET(PACKET2_PAD, (v)))
 
-#define PACKET3(op, n)	((PACKET_TYPE3 << 30) |				\
+#define PACKET3(op, n)	((RADEON_PACKET_TYPE3 << 30) |			\
 			 (((op) & 0xFF) << 8) |				\
 			 ((n) & 0x3FFF) << 16)
 
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index d1d5306ebf24..f6e0b5395051 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -313,9 +313,9 @@ static int shmob_drm_pm_resume(struct device *dev)
 {
 	struct shmob_drm_device *sdev = dev_get_drvdata(dev);
 
-	mutex_lock(&sdev->ddev->mode_config.mutex);
+	drm_modeset_lock_all(sdev->ddev);
 	shmob_drm_crtc_resume(&sdev->crtc);
-	mutex_unlock(&sdev->ddev->mode_config.mutex);
+	drm_modeset_unlock_all(sdev->ddev);
 
 	drm_kms_helper_poll_enable(sdev->ddev);
 	return 0;
diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig
index be1daf7344d3..c92955df0658 100644
--- a/drivers/gpu/drm/tegra/Kconfig
+++ b/drivers/gpu/drm/tegra/Kconfig
@@ -4,6 +4,7 @@ config DRM_TEGRA
 	select DRM_KMS_HELPER
 	select DRM_GEM_CMA_HELPER
 	select DRM_KMS_CMA_HELPER
+	select DRM_HDMI
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index b6679b36700f..de94707b9dbe 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -17,26 +17,257 @@
 #include "drm.h"
 #include "dc.h"
 
-struct tegra_dc_window {
-	fixed20_12 x;
-	fixed20_12 y;
-	fixed20_12 w;
-	fixed20_12 h;
-	unsigned int outx;
-	unsigned int outy;
-	unsigned int outw;
-	unsigned int outh;
-	unsigned int stride;
-	unsigned int fmt;
+struct tegra_plane {
+	struct drm_plane base;
+	unsigned int index;
 };
 
+static inline struct tegra_plane *to_tegra_plane(struct drm_plane *plane)
+{
+	return container_of(plane, struct tegra_plane, base);
+}
+
+static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
+			      struct drm_framebuffer *fb, int crtc_x,
+			      int crtc_y, unsigned int crtc_w,
+			      unsigned int crtc_h, uint32_t src_x,
+			      uint32_t src_y, uint32_t src_w, uint32_t src_h)
+{
+	struct tegra_plane *p = to_tegra_plane(plane);
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	struct tegra_dc_window window;
+	unsigned int i;
+
+	memset(&window, 0, sizeof(window));
+	window.src.x = src_x >> 16;
+	window.src.y = src_y >> 16;
+	window.src.w = src_w >> 16;
+	window.src.h = src_h >> 16;
+	window.dst.x = crtc_x;
+	window.dst.y = crtc_y;
+	window.dst.w = crtc_w;
+	window.dst.h = crtc_h;
+	window.format = tegra_dc_format(fb->pixel_format);
+	window.bits_per_pixel = fb->bits_per_pixel;
+
+	for (i = 0; i < drm_format_num_planes(fb->pixel_format); i++) {
+		struct drm_gem_cma_object *gem = drm_fb_cma_get_gem_obj(fb, i);
+
+		window.base[i] = gem->paddr + fb->offsets[i];
+
+		/*
+		 * Tegra doesn't support different strides for U and V planes
+		 * so we display a warning if the user tries to display a
+		 * framebuffer with such a configuration.
+		 */
+		if (i >= 2) {
+			if (fb->pitches[i] != window.stride[1])
+				DRM_ERROR("unsupported UV-plane configuration\n");
+		} else {
+			window.stride[i] = fb->pitches[i];
+		}
+	}
+
+	return tegra_dc_setup_window(dc, p->index, &window);
+}
+
+static int tegra_plane_disable(struct drm_plane *plane)
+{
+	struct tegra_dc *dc = to_tegra_dc(plane->crtc);
+	struct tegra_plane *p = to_tegra_plane(plane);
+	unsigned long value;
+
+	value = WINDOW_A_SELECT << p->index;
+	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
+
+	value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
+	value &= ~WIN_ENABLE;
+	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+
+	tegra_dc_writel(dc, WIN_A_UPDATE << p->index, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, WIN_A_ACT_REQ << p->index, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
+static void tegra_plane_destroy(struct drm_plane *plane)
+{
+	tegra_plane_disable(plane);
+	drm_plane_cleanup(plane);
+}
+
+static const struct drm_plane_funcs tegra_plane_funcs = {
+	.update_plane = tegra_plane_update,
+	.disable_plane = tegra_plane_disable,
+	.destroy = tegra_plane_destroy,
+};
+
+static const uint32_t plane_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUV420,
+	DRM_FORMAT_YUV422,
+};
+
+static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc)
+{
+	unsigned int i;
+	int err = 0;
+
+	for (i = 0; i < 2; i++) {
+		struct tegra_plane *plane;
+
+		plane = devm_kzalloc(drm->dev, sizeof(*plane), GFP_KERNEL);
+		if (!plane)
+			return -ENOMEM;
+
+		plane->index = 1 + i;
+
+		err = drm_plane_init(drm, &plane->base, 1 << dc->pipe,
+				     &tegra_plane_funcs, plane_formats,
+				     ARRAY_SIZE(plane_formats), false);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
+static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
+			     struct drm_framebuffer *fb)
+{
+	struct drm_gem_cma_object *gem = drm_fb_cma_get_gem_obj(fb, 0);
+	unsigned long value;
+
+	tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER);
+
+	value = fb->offsets[0] + y * fb->pitches[0] +
+		x * fb->bits_per_pixel / 8;
+
+	tegra_dc_writel(dc, gem->paddr + value, DC_WINBUF_START_ADDR);
+	tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE);
+
+	value = GENERAL_UPDATE | WIN_A_UPDATE;
+	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
+
+	value = GENERAL_ACT_REQ | WIN_A_ACT_REQ;
+	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
+void tegra_dc_enable_vblank(struct tegra_dc *dc)
+{
+	unsigned long value, flags;
+
+	spin_lock_irqsave(&dc->lock, flags);
+
+	value = tegra_dc_readl(dc, DC_CMD_INT_MASK);
+	value |= VBLANK_INT;
+	tegra_dc_writel(dc, value, DC_CMD_INT_MASK);
+
+	spin_unlock_irqrestore(&dc->lock, flags);
+}
+
+void tegra_dc_disable_vblank(struct tegra_dc *dc)
+{
+	unsigned long value, flags;
+
+	spin_lock_irqsave(&dc->lock, flags);
+
+	value = tegra_dc_readl(dc, DC_CMD_INT_MASK);
+	value &= ~VBLANK_INT;
+	tegra_dc_writel(dc, value, DC_CMD_INT_MASK);
+
+	spin_unlock_irqrestore(&dc->lock, flags);
+}
+
+static void tegra_dc_finish_page_flip(struct tegra_dc *dc)
+{
+	struct drm_device *drm = dc->base.dev;
+	struct drm_crtc *crtc = &dc->base;
+	struct drm_gem_cma_object *gem;
+	unsigned long flags, base;
+
+	if (!dc->event)
+		return;
+
+	gem = drm_fb_cma_get_gem_obj(crtc->fb, 0);
+
+	/* check if new start address has been latched */
+	tegra_dc_writel(dc, READ_MUX, DC_CMD_STATE_ACCESS);
+	base = tegra_dc_readl(dc, DC_WINBUF_START_ADDR);
+	tegra_dc_writel(dc, 0, DC_CMD_STATE_ACCESS);
+
+	if (base == gem->paddr + crtc->fb->offsets[0]) {
+		spin_lock_irqsave(&drm->event_lock, flags);
+		drm_send_vblank_event(drm, dc->pipe, dc->event);
+		drm_vblank_put(drm, dc->pipe);
+		dc->event = NULL;
+		spin_unlock_irqrestore(&drm->event_lock, flags);
+	}
+}
+
+void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file)
+{
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	struct drm_device *drm = crtc->dev;
+	unsigned long flags;
+
+	spin_lock_irqsave(&drm->event_lock, flags);
+
+	if (dc->event && dc->event->base.file_priv == file) {
+		dc->event->base.destroy(&dc->event->base);
+		drm_vblank_put(drm, dc->pipe);
+		dc->event = NULL;
+	}
+
+	spin_unlock_irqrestore(&drm->event_lock, flags);
+}
+
+static int tegra_dc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+			      struct drm_pending_vblank_event *event)
+{
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	struct drm_device *drm = crtc->dev;
+
+	if (dc->event)
+		return -EBUSY;
+
+	if (event) {
+		event->pipe = dc->pipe;
+		dc->event = event;
+		drm_vblank_get(drm, dc->pipe);
+	}
+
+	tegra_dc_set_base(dc, 0, 0, fb);
+	crtc->fb = fb;
+
+	return 0;
+}
+
 static const struct drm_crtc_funcs tegra_crtc_funcs = {
+	.page_flip = tegra_dc_page_flip,
 	.set_config = drm_crtc_helper_set_config,
 	.destroy = drm_crtc_cleanup,
 };
 
-static void tegra_crtc_dpms(struct drm_crtc *crtc, int mode)
+static void tegra_crtc_disable(struct drm_crtc *crtc)
 {
+	struct drm_device *drm = crtc->dev;
+	struct drm_plane *plane;
+
+	list_for_each_entry(plane, &drm->mode_config.plane_list, head) {
+		if (plane->crtc == crtc) {
+			tegra_plane_disable(plane);
+			plane->crtc = NULL;
+
+			if (plane->fb) {
+				drm_framebuffer_unreference(plane->fb);
+				plane->fb = NULL;
+			}
+		}
+	}
 }
 
 static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
@@ -46,10 +277,11 @@ static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
 	return true;
 }
 
-static inline u32 compute_dda_inc(fixed20_12 inf, unsigned int out, bool v,
+static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v,
 				  unsigned int bpp)
 {
 	fixed20_12 outf = dfixed_init(out);
+	fixed20_12 inf = dfixed_init(in);
 	u32 dda_inc;
 	int max;
 
@@ -79,9 +311,10 @@ static inline u32 compute_dda_inc(fixed20_12 inf, unsigned int out, bool v,
 	return dda_inc;
 }
 
-static inline u32 compute_initial_dda(fixed20_12 in)
+static inline u32 compute_initial_dda(unsigned int in)
 {
-	return dfixed_frac(in);
+	fixed20_12 inf = dfixed_init(in);
+	return dfixed_frac(inf);
 }
 
 static int tegra_dc_set_timings(struct tegra_dc *dc,
@@ -152,18 +385,198 @@ static int tegra_crtc_setup_clk(struct drm_crtc *crtc,
 	return 0;
 }
 
+static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar)
+{
+	switch (format) {
+	case WIN_COLOR_DEPTH_YCbCr422:
+	case WIN_COLOR_DEPTH_YUV422:
+		if (planar)
+			*planar = false;
+
+		return true;
+
+	case WIN_COLOR_DEPTH_YCbCr420P:
+	case WIN_COLOR_DEPTH_YUV420P:
+	case WIN_COLOR_DEPTH_YCbCr422P:
+	case WIN_COLOR_DEPTH_YUV422P:
+	case WIN_COLOR_DEPTH_YCbCr422R:
+	case WIN_COLOR_DEPTH_YUV422R:
+	case WIN_COLOR_DEPTH_YCbCr422RA:
+	case WIN_COLOR_DEPTH_YUV422RA:
+		if (planar)
+			*planar = true;
+
+		return true;
+	}
+
+	return false;
+}
+
+int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
+			  const struct tegra_dc_window *window)
+{
+	unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp;
+	unsigned long value;
+	bool yuv, planar;
+
+	/*
+	 * For YUV planar modes, the number of bytes per pixel takes into
+	 * account only the luma component and therefore is 1.
+	 */
+	yuv = tegra_dc_format_is_yuv(window->format, &planar);
+	if (!yuv)
+		bpp = window->bits_per_pixel / 8;
+	else
+		bpp = planar ? 1 : 2;
+
+	value = WINDOW_A_SELECT << index;
+	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
+
+	tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH);
+	tegra_dc_writel(dc, 0, DC_WIN_BYTE_SWAP);
+
+	value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x);
+	tegra_dc_writel(dc, value, DC_WIN_POSITION);
+
+	value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w);
+	tegra_dc_writel(dc, value, DC_WIN_SIZE);
+
+	h_offset = window->src.x * bpp;
+	v_offset = window->src.y;
+	h_size = window->src.w * bpp;
+	v_size = window->src.h;
+
+	value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size);
+	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
+
+	/*
+	 * For DDA computations the number of bytes per pixel for YUV planar
+	 * modes needs to take into account all Y, U and V components.
+	 */
+	if (yuv && planar)
+		bpp = 2;
+
+	h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp);
+	v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp);
+
+	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
+	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
+
+	h_dda = compute_initial_dda(window->src.x);
+	v_dda = compute_initial_dda(window->src.y);
+
+	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
+	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
+
+	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
+	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
+
+	tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR);
+
+	if (yuv && planar) {
+		tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U);
+		tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V);
+		value = window->stride[1] << 16 | window->stride[0];
+		tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE);
+	} else {
+		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
+	}
+
+	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
+	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
+
+	value = WIN_ENABLE;
+
+	if (yuv) {
+		/* setup default colorspace conversion coefficients */
+		tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF);
+		tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR);
+		tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR);
+		tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG);
+		tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG);
+		tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB);
+
+		value |= CSC_ENABLE;
+	} else if (window->bits_per_pixel < 24) {
+		value |= COLOR_EXPAND;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+
+	/*
+	 * Disable blending and assume Window A is the bottom-most window,
+	 * Window C is the top-most window and Window B is in the middle.
+	 */
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY);
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN);
+
+	switch (index) {
+	case 0:
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 1:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 2:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY);
+		break;
+	}
+
+	tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
+unsigned int tegra_dc_format(uint32_t format)
+{
+	switch (format) {
+	case DRM_FORMAT_XRGB8888:
+		return WIN_COLOR_DEPTH_B8G8R8A8;
+
+	case DRM_FORMAT_RGB565:
+		return WIN_COLOR_DEPTH_B5G6R5;
+
+	case DRM_FORMAT_UYVY:
+		return WIN_COLOR_DEPTH_YCbCr422;
+
+	case DRM_FORMAT_YUV420:
+		return WIN_COLOR_DEPTH_YCbCr420P;
+
+	case DRM_FORMAT_YUV422:
+		return WIN_COLOR_DEPTH_YCbCr422P;
+
+	default:
+		break;
+	}
+
+	WARN(1, "unsupported pixel format %u, using default\n", format);
+	return WIN_COLOR_DEPTH_B8G8R8A8;
+}
+
 static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 			       struct drm_display_mode *mode,
 			       struct drm_display_mode *adjusted,
 			       int x, int y, struct drm_framebuffer *old_fb)
 {
-	struct tegra_framebuffer *fb = to_tegra_fb(crtc->fb);
+	struct drm_gem_cma_object *gem = drm_fb_cma_get_gem_obj(crtc->fb, 0);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
-	unsigned int h_dda, v_dda, bpp;
-	struct tegra_dc_window win;
+	struct tegra_dc_window window;
 	unsigned long div, value;
 	int err;
 
+	drm_vblank_pre_modeset(crtc->dev, dc->pipe);
+
 	err = tegra_crtc_setup_clk(crtc, mode, &div);
 	if (err) {
 		dev_err(dc->dev, "failed to setup clock for CRTC: %d\n", err);
@@ -191,83 +604,33 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 	tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL);
 
 	/* setup window parameters */
-	memset(&win, 0, sizeof(win));
-	win.x.full = dfixed_const(0);
-	win.y.full = dfixed_const(0);
-	win.w.full = dfixed_const(mode->hdisplay);
-	win.h.full = dfixed_const(mode->vdisplay);
-	win.outx = 0;
-	win.outy = 0;
-	win.outw = mode->hdisplay;
-	win.outh = mode->vdisplay;
-
-	switch (crtc->fb->pixel_format) {
-	case DRM_FORMAT_XRGB8888:
-		win.fmt = WIN_COLOR_DEPTH_B8G8R8A8;
-		break;
-
-	case DRM_FORMAT_RGB565:
-		win.fmt = WIN_COLOR_DEPTH_B5G6R5;
-		break;
-
-	default:
-		win.fmt = WIN_COLOR_DEPTH_B8G8R8A8;
-		WARN_ON(1);
-		break;
-	}
-
-	bpp = crtc->fb->bits_per_pixel / 8;
-	win.stride = crtc->fb->pitches[0];
-
-	/* program window registers */
-	value = WINDOW_A_SELECT;
-	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
-
-	tegra_dc_writel(dc, win.fmt, DC_WIN_COLOR_DEPTH);
-	tegra_dc_writel(dc, 0, DC_WIN_BYTE_SWAP);
-
-	value = V_POSITION(win.outy) | H_POSITION(win.outx);
-	tegra_dc_writel(dc, value, DC_WIN_POSITION);
-
-	value = V_SIZE(win.outh) | H_SIZE(win.outw);
-	tegra_dc_writel(dc, value, DC_WIN_SIZE);
-
-	value = V_PRESCALED_SIZE(dfixed_trunc(win.h)) |
-		H_PRESCALED_SIZE(dfixed_trunc(win.w) * bpp);
-	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
-
-	h_dda = compute_dda_inc(win.w, win.outw, false, bpp);
-	v_dda = compute_dda_inc(win.h, win.outh, true, bpp);
-
-	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
-	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
-
-	h_dda = compute_initial_dda(win.x);
-	v_dda = compute_initial_dda(win.y);
-
-	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
-	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
-
-	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
-	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
-
-	tegra_dc_writel(dc, fb->obj->paddr, DC_WINBUF_START_ADDR);
-	tegra_dc_writel(dc, win.stride, DC_WIN_LINE_STRIDE);
-	tegra_dc_writel(dc, dfixed_trunc(win.x) * bpp,
-			DC_WINBUF_ADDR_H_OFFSET);
-	tegra_dc_writel(dc, dfixed_trunc(win.y), DC_WINBUF_ADDR_V_OFFSET);
-
-	value = WIN_ENABLE;
-
-	if (bpp < 24)
-		value |= COLOR_EXPAND;
+	memset(&window, 0, sizeof(window));
+	window.src.x = 0;
+	window.src.y = 0;
+	window.src.w = mode->hdisplay;
+	window.src.h = mode->vdisplay;
+	window.dst.x = 0;
+	window.dst.y = 0;
+	window.dst.w = mode->hdisplay;
+	window.dst.h = mode->vdisplay;
+	window.format = tegra_dc_format(crtc->fb->pixel_format);
+	window.bits_per_pixel = crtc->fb->bits_per_pixel;
+	window.stride[0] = crtc->fb->pitches[0];
+	window.base[0] = gem->paddr;
+
+	err = tegra_dc_setup_window(dc, 0, &window);
+	if (err < 0)
+		dev_err(dc->dev, "failed to enable root plane\n");
 
-	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+	return 0;
+}
 
-	tegra_dc_writel(dc, 0xff00, DC_WIN_BLEND_NOKEY);
-	tegra_dc_writel(dc, 0xff00, DC_WIN_BLEND_1WIN);
+static int tegra_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
+				    struct drm_framebuffer *old_fb)
+{
+	struct tegra_dc *dc = to_tegra_dc(crtc);
 
-	return 0;
+	return tegra_dc_set_base(dc, x, y, crtc->fb);
 }
 
 static void tegra_crtc_prepare(struct drm_crtc *crtc)
@@ -314,31 +677,24 @@ static void tegra_crtc_prepare(struct drm_crtc *crtc)
 	tegra_dc_writel(dc, value, DC_DISP_DISP_MEM_HIGH_PRIORITY_TIMER);
 
 	value = VBLANK_INT | WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT;
-	tegra_dc_writel(dc, value, DC_CMD_INT_MASK);
-
-	value = VBLANK_INT | WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT;
 	tegra_dc_writel(dc, value, DC_CMD_INT_ENABLE);
+
+	value = WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT;
+	tegra_dc_writel(dc, value, DC_CMD_INT_MASK);
 }
 
 static void tegra_crtc_commit(struct drm_crtc *crtc)
 {
 	struct tegra_dc *dc = to_tegra_dc(crtc);
-	unsigned long update_mask;
 	unsigned long value;
 
-	update_mask = GENERAL_ACT_REQ | WIN_A_ACT_REQ;
-
-	tegra_dc_writel(dc, update_mask << 8, DC_CMD_STATE_CONTROL);
+	value = GENERAL_UPDATE | WIN_A_UPDATE;
+	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
 
-	value = tegra_dc_readl(dc, DC_CMD_INT_ENABLE);
-	value |= FRAME_END_INT;
-	tegra_dc_writel(dc, value, DC_CMD_INT_ENABLE);
-
-	value = tegra_dc_readl(dc, DC_CMD_INT_MASK);
-	value |= FRAME_END_INT;
-	tegra_dc_writel(dc, value, DC_CMD_INT_MASK);
+	value = GENERAL_ACT_REQ | WIN_A_ACT_REQ;
+	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
 
-	tegra_dc_writel(dc, update_mask, DC_CMD_STATE_CONTROL);
+	drm_vblank_post_modeset(crtc->dev, dc->pipe);
 }
 
 static void tegra_crtc_load_lut(struct drm_crtc *crtc)
@@ -346,15 +702,16 @@ static void tegra_crtc_load_lut(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_helper_funcs tegra_crtc_helper_funcs = {
-	.dpms = tegra_crtc_dpms,
+	.disable = tegra_crtc_disable,
 	.mode_fixup = tegra_crtc_mode_fixup,
 	.mode_set = tegra_crtc_mode_set,
+	.mode_set_base = tegra_crtc_mode_set_base,
 	.prepare = tegra_crtc_prepare,
 	.commit = tegra_crtc_commit,
 	.load_lut = tegra_crtc_load_lut,
 };
 
-static irqreturn_t tegra_drm_irq(int irq, void *data)
+static irqreturn_t tegra_dc_irq(int irq, void *data)
 {
 	struct tegra_dc *dc = data;
 	unsigned long status;
@@ -373,6 +730,7 @@ static irqreturn_t tegra_drm_irq(int irq, void *data)
 		dev_dbg(dc->dev, "%s(): vertical blank\n", __func__);
 		*/
 		drm_handle_vblank(dc->base.dev, dc->pipe);
+		tegra_dc_finish_page_flip(dc);
 	}
 
 	if (status & (WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT)) {
@@ -587,7 +945,7 @@ static int tegra_dc_show_regs(struct seq_file *s, void *data)
 	DUMP_REG(DC_WIN_BLEND_1WIN);
 	DUMP_REG(DC_WIN_BLEND_2WIN_X);
 	DUMP_REG(DC_WIN_BLEND_2WIN_Y);
-	DUMP_REG(DC_WIN_BLEND32WIN_XY);
+	DUMP_REG(DC_WIN_BLEND_3WIN_XY);
 	DUMP_REG(DC_WIN_HP_FETCH_CONTROL);
 	DUMP_REG(DC_WINBUF_START_ADDR);
 	DUMP_REG(DC_WINBUF_START_ADDR_NS);
@@ -689,13 +1047,17 @@ static int tegra_dc_drm_init(struct host1x_client *client,
 		return err;
 	}
 
+	err = tegra_dc_add_planes(drm, dc);
+	if (err < 0)
+		return err;
+
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
 		err = tegra_dc_debugfs_init(dc, drm->primary);
 		if (err < 0)
 			dev_err(dc->dev, "debugfs setup failed: %d\n", err);
 	}
 
-	err = devm_request_irq(dc->dev, dc->irq, tegra_drm_irq, 0,
+	err = devm_request_irq(dc->dev, dc->irq, tegra_dc_irq, 0,
 			       dev_name(dc->dev), dc);
 	if (err < 0) {
 		dev_err(dc->dev, "failed to request IRQ#%u: %d\n", dc->irq,
@@ -744,6 +1106,7 @@ static int tegra_dc_probe(struct platform_device *pdev)
 	if (!dc)
 		return -ENOMEM;
 
+	spin_lock_init(&dc->lock);
 	INIT_LIST_HEAD(&dc->list);
 	dc->dev = &pdev->dev;
 
diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h
index 99977b5d5c36..79eaec9aac77 100644
--- a/drivers/gpu/drm/tegra/dc.h
+++ b/drivers/gpu/drm/tegra/dc.h
@@ -58,6 +58,8 @@
 #define DC_CMD_SIGNAL_RAISE3			0x03e
 
 #define DC_CMD_STATE_ACCESS			0x040
+#define READ_MUX  (1 << 0)
+#define WRITE_MUX (1 << 2)
 
 #define DC_CMD_STATE_CONTROL			0x041
 #define GENERAL_ACT_REQ (1 <<  0)
@@ -290,8 +292,18 @@
 #define DC_DISP_SD_HW_K_VALUES			0x4dd
 #define DC_DISP_SD_MAN_K_VALUES			0x4de
 
+#define DC_WIN_CSC_YOF				0x611
+#define DC_WIN_CSC_KYRGB			0x612
+#define DC_WIN_CSC_KUR				0x613
+#define DC_WIN_CSC_KVR				0x614
+#define DC_WIN_CSC_KUG				0x615
+#define DC_WIN_CSC_KVG				0x616
+#define DC_WIN_CSC_KUB				0x617
+#define DC_WIN_CSC_KVB				0x618
+
 #define DC_WIN_WIN_OPTIONS			0x700
 #define COLOR_EXPAND (1 <<  6)
+#define CSC_ENABLE   (1 << 18)
 #define WIN_ENABLE   (1 << 30)
 
 #define DC_WIN_BYTE_SWAP			0x701
@@ -359,7 +371,7 @@
 #define DC_WIN_BLEND_1WIN			0x710
 #define DC_WIN_BLEND_2WIN_X			0x711
 #define DC_WIN_BLEND_2WIN_Y			0x712
-#define DC_WIN_BLEND32WIN_XY			0x713
+#define DC_WIN_BLEND_3WIN_XY			0x713
 
 #define DC_WIN_HP_FETCH_CONTROL			0x714
 
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index d980dc75788c..9d452df5bcad 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -39,6 +39,10 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 	if (err < 0)
 		return err;
 
+	err = drm_vblank_init(drm, drm->mode_config.num_crtc);
+	if (err < 0)
+		return err;
+
 	err = tegra_drm_fb_init(drm);
 	if (err < 0)
 		return err;
@@ -88,13 +92,112 @@ static const struct file_operations tegra_drm_fops = {
 	.llseek = noop_llseek,
 };
 
+static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) {
+		struct tegra_dc *dc = to_tegra_dc(crtc);
+
+		if (dc->pipe == pipe)
+			return crtc;
+	}
+
+	return NULL;
+}
+
+static u32 tegra_drm_get_vblank_counter(struct drm_device *dev, int crtc)
+{
+	/* TODO: implement real hardware counter using syncpoints */
+	return drm_vblank_count(dev, crtc);
+}
+
+static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+
+	if (!crtc)
+		return -ENODEV;
+
+	tegra_dc_enable_vblank(dc);
+
+	return 0;
+}
+
+static void tegra_drm_disable_vblank(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+
+	if (crtc)
+		tegra_dc_disable_vblank(dc);
+}
+
+static void tegra_drm_preclose(struct drm_device *drm, struct drm_file *file)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head)
+		tegra_dc_cancel_page_flip(crtc, file);
+}
+
+#ifdef CONFIG_DEBUG_FS
+static int tegra_debugfs_framebuffers(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *)s->private;
+	struct drm_device *drm = node->minor->dev;
+	struct drm_framebuffer *fb;
+
+	mutex_lock(&drm->mode_config.fb_lock);
+
+	list_for_each_entry(fb, &drm->mode_config.fb_list, head) {
+		seq_printf(s, "%3d: user size: %d x %d, depth %d, %d bpp, refcount %d\n",
+			   fb->base.id, fb->width, fb->height, fb->depth,
+			   fb->bits_per_pixel,
+			   atomic_read(&fb->refcount.refcount));
+	}
+
+	mutex_unlock(&drm->mode_config.fb_lock);
+
+	return 0;
+}
+
+static struct drm_info_list tegra_debugfs_list[] = {
+	{ "framebuffers", tegra_debugfs_framebuffers, 0 },
+};
+
+static int tegra_debugfs_init(struct drm_minor *minor)
+{
+	return drm_debugfs_create_files(tegra_debugfs_list,
+					ARRAY_SIZE(tegra_debugfs_list),
+					minor->debugfs_root, minor);
+}
+
+static void tegra_debugfs_cleanup(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(tegra_debugfs_list,
+				 ARRAY_SIZE(tegra_debugfs_list), minor);
+}
+#endif
+
 struct drm_driver tegra_drm_driver = {
 	.driver_features = DRIVER_BUS_PLATFORM | DRIVER_MODESET | DRIVER_GEM,
 	.load = tegra_drm_load,
 	.unload = tegra_drm_unload,
 	.open = tegra_drm_open,
+	.preclose = tegra_drm_preclose,
 	.lastclose = tegra_drm_lastclose,
 
+	.get_vblank_counter = tegra_drm_get_vblank_counter,
+	.enable_vblank = tegra_drm_enable_vblank,
+	.disable_vblank = tegra_drm_disable_vblank,
+
+#if defined(CONFIG_DEBUG_FS)
+	.debugfs_init = tegra_debugfs_init,
+	.debugfs_cleanup = tegra_debugfs_cleanup,
+#endif
+
 	.gem_free_object = drm_gem_cma_free_object,
 	.gem_vm_ops = &drm_gem_cma_vm_ops,
 	.dumb_create = drm_gem_cma_dumb_create,
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 741b5dc2742c..6dd75a2600eb 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -18,16 +18,6 @@
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fixed.h>
 
-struct tegra_framebuffer {
-	struct drm_framebuffer base;
-	struct drm_gem_cma_object *obj;
-};
-
-static inline struct tegra_framebuffer *to_tegra_fb(struct drm_framebuffer *fb)
-{
-	return container_of(fb, struct tegra_framebuffer, base);
-}
-
 struct host1x {
 	struct drm_device *drm;
 	struct device *dev;
@@ -44,7 +34,6 @@ struct host1x {
 	struct list_head clients;
 
 	struct drm_fbdev_cma *fbdev;
-	struct tegra_framebuffer fb;
 };
 
 struct host1x_client;
@@ -75,6 +64,7 @@ struct tegra_output;
 
 struct tegra_dc {
 	struct host1x_client client;
+	spinlock_t lock;
 
 	struct host1x *host1x;
 	struct device *dev;
@@ -94,6 +84,9 @@ struct tegra_dc {
 	struct drm_info_list *debugfs_files;
 	struct drm_minor *minor;
 	struct dentry *debugfs;
+
+	/* page-flip handling */
+	struct drm_pending_vblank_event *event;
 };
 
 static inline struct tegra_dc *host1x_client_to_dc(struct host1x_client *client)
@@ -118,6 +111,34 @@ static inline unsigned long tegra_dc_readl(struct tegra_dc *dc,
 	return readl(dc->regs + (reg << 2));
 }
 
+struct tegra_dc_window {
+	struct {
+		unsigned int x;
+		unsigned int y;
+		unsigned int w;
+		unsigned int h;
+	} src;
+	struct {
+		unsigned int x;
+		unsigned int y;
+		unsigned int w;
+		unsigned int h;
+	} dst;
+	unsigned int bits_per_pixel;
+	unsigned int format;
+	unsigned int stride[2];
+	unsigned long base[3];
+};
+
+/* from dc.c */
+extern unsigned int tegra_dc_format(uint32_t format);
+extern int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
+				 const struct tegra_dc_window *window);
+extern void tegra_dc_enable_vblank(struct tegra_dc *dc);
+extern void tegra_dc_disable_vblank(struct tegra_dc *dc);
+extern void tegra_dc_cancel_page_flip(struct drm_crtc *crtc,
+				      struct drm_file *file);
+
 struct tegra_output_ops {
 	int (*enable)(struct tegra_output *output);
 	int (*disable)(struct tegra_output *output);
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 97993c6835fd..03914953cb1c 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -39,10 +39,6 @@ int tegra_drm_fb_init(struct drm_device *drm)
 	if (IS_ERR(fbdev))
 		return PTR_ERR(fbdev);
 
-#ifndef CONFIG_FRAMEBUFFER_CONSOLE
-	drm_fbdev_cma_restore_mode(fbdev);
-#endif
-
 	host1x->fbdev = fbdev;
 
 	return 0;
diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index d4f3fb9f0c29..bb747f6cd1a4 100644
--- a/drivers/gpu/drm/tegra/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -10,12 +10,15 @@
 #include <linux/clk.h>
 #include <linux/debugfs.h>
 #include <linux/gpio.h>
+#include <linux/hdmi.h>
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/regulator/consumer.h>
 #include <linux/clk/tegra.h>
 
+#include <drm/drm_edid.h>
+
 #include "hdmi.h"
 #include "drm.h"
 #include "dc.h"
@@ -400,54 +403,65 @@ static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi, unsigned int pclk)
 	return 0;
 }
 
-static void tegra_hdmi_write_infopack(struct tegra_hdmi *hdmi,
-				      unsigned int offset, u8 type,
-				      u8 version, void *data, size_t size)
+static inline unsigned long tegra_hdmi_subpack(const u8 *ptr, size_t size)
 {
-	unsigned long value;
-	u8 *ptr = data;
-	u32 subpack[2];
+	unsigned long value = 0;
 	size_t i;
-	u8 csum;
 
-	/* first byte of data is the checksum */
-	csum = type + version + size - 1;
+	for (i = size; i > 0; i--)
+		value = (value << 8) | ptr[i - 1];
 
-	for (i = 1; i < size; i++)
-		csum += ptr[i];
+	return value;
+}
 
-	ptr[0] = 0x100 - csum;
+static void tegra_hdmi_write_infopack(struct tegra_hdmi *hdmi, const void *data,
+				      size_t size)
+{
+	const u8 *ptr = data;
+	unsigned long offset;
+	unsigned long value;
+	size_t i, j;
 
-	value = INFOFRAME_HEADER_TYPE(type) |
-		INFOFRAME_HEADER_VERSION(version) |
-		INFOFRAME_HEADER_LEN(size - 1);
-	tegra_hdmi_writel(hdmi, value, offset);
+	switch (ptr[0]) {
+	case HDMI_INFOFRAME_TYPE_AVI:
+		offset = HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_HEADER;
+		break;
 
-	/* The audio inforame only has one set of subpack registers.  The hdmi
-	 * block pads the rest of the data as per the spec so we have to fixup
-	 * the length before filling in the subpacks.
-	 */
-	if (offset == HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER)
-		size = 6;
+	case HDMI_INFOFRAME_TYPE_AUDIO:
+		offset = HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER;
+		break;
 
-	/* each subpack 7 bytes devided into:
-	 *   subpack_low - bytes 0 - 3
-	 *   subpack_high - bytes 4 - 6 (with byte 7 padded to 0x00)
-	 */
-	for (i = 0; i < size; i++) {
-		size_t index = i % 7;
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		offset = HDMI_NV_PDISP_HDMI_GENERIC_HEADER;
+		break;
+
+	default:
+		dev_err(hdmi->dev, "unsupported infoframe type: %02x\n",
+			ptr[0]);
+		return;
+	}
+
+	value = INFOFRAME_HEADER_TYPE(ptr[0]) |
+		INFOFRAME_HEADER_VERSION(ptr[1]) |
+		INFOFRAME_HEADER_LEN(ptr[2]);
+	tegra_hdmi_writel(hdmi, value, offset);
+	offset++;
 
-		if (index == 0)
-			memset(subpack, 0x0, sizeof(subpack));
+	/*
+	 * Each subpack contains 7 bytes, divided into:
+	 * - subpack_low: bytes 0 - 3
+	 * - subpack_high: bytes 4 - 6 (with byte 7 padded to 0x00)
+	 */
+	for (i = 3, j = 0; i < size; i += 7, j += 8) {
+		size_t rem = size - i, num = min_t(size_t, rem, 4);
 
-		((u8 *)subpack)[index] = ptr[i];
+		value = tegra_hdmi_subpack(&ptr[i], num);
+		tegra_hdmi_writel(hdmi, value, offset++);
 
-		if (index == 6 || (i + 1 == size)) {
-			unsigned int reg = offset + 1 + (i / 7) * 2;
+		num = min_t(size_t, rem - num, 3);
 
-			tegra_hdmi_writel(hdmi, subpack[0], reg);
-			tegra_hdmi_writel(hdmi, subpack[1], reg + 1);
-		}
+		value = tegra_hdmi_subpack(&ptr[i + 4], num);
+		tegra_hdmi_writel(hdmi, value, offset++);
 	}
 }
 
@@ -455,9 +469,8 @@ static void tegra_hdmi_setup_avi_infoframe(struct tegra_hdmi *hdmi,
 					   struct drm_display_mode *mode)
 {
 	struct hdmi_avi_infoframe frame;
-	unsigned int h_front_porch;
-	unsigned int hsize = 16;
-	unsigned int vsize = 9;
+	u8 buffer[17];
+	ssize_t err;
 
 	if (hdmi->dvi) {
 		tegra_hdmi_writel(hdmi, 0,
@@ -465,69 +478,19 @@ static void tegra_hdmi_setup_avi_infoframe(struct tegra_hdmi *hdmi,
 		return;
 	}
 
-	h_front_porch = mode->hsync_start - mode->hdisplay;
-	memset(&frame, 0, sizeof(frame));
-	frame.r = HDMI_AVI_R_SAME;
-
-	switch (mode->vdisplay) {
-	case 480:
-		if (mode->hdisplay == 640) {
-			frame.m = HDMI_AVI_M_4_3;
-			frame.vic = 1;
-		} else {
-			frame.m = HDMI_AVI_M_16_9;
-			frame.vic = 3;
-		}
-		break;
-
-	case 576:
-		if (((hsize * 10) / vsize) > 14) {
-			frame.m = HDMI_AVI_M_16_9;
-			frame.vic = 18;
-		} else {
-			frame.m = HDMI_AVI_M_4_3;
-			frame.vic = 17;
-		}
-		break;
-
-	case 720:
-	case 1470: /* stereo mode */
-		frame.m = HDMI_AVI_M_16_9;
-
-		if (h_front_porch == 110)
-			frame.vic = 4;
-		else
-			frame.vic = 19;
-		break;
-
-	case 1080:
-	case 2205: /* stereo mode */
-		frame.m = HDMI_AVI_M_16_9;
-
-		switch (h_front_porch) {
-		case 88:
-			frame.vic = 16;
-			break;
-
-		case 528:
-			frame.vic = 31;
-			break;
-
-		default:
-			frame.vic = 32;
-			break;
-		}
-		break;
+	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	if (err < 0) {
+		dev_err(hdmi->dev, "failed to setup AVI infoframe: %zd\n", err);
+		return;
+	}
 
-	default:
-		frame.m = HDMI_AVI_M_16_9;
-		frame.vic = 0;
-		break;
+	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		dev_err(hdmi->dev, "failed to pack AVI infoframe: %zd\n", err);
+		return;
 	}
 
-	tegra_hdmi_write_infopack(hdmi, HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_HEADER,
-				  HDMI_INFOFRAME_TYPE_AVI, HDMI_AVI_VERSION,
-				  &frame, sizeof(frame));
+	tegra_hdmi_write_infopack(hdmi, buffer, err);
 
 	tegra_hdmi_writel(hdmi, INFOFRAME_CTRL_ENABLE,
 			  HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_CTRL);
@@ -536,6 +499,8 @@ static void tegra_hdmi_setup_avi_infoframe(struct tegra_hdmi *hdmi,
 static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 {
 	struct hdmi_audio_infoframe frame;
+	u8 buffer[14];
+	ssize_t err;
 
 	if (hdmi->dvi) {
 		tegra_hdmi_writel(hdmi, 0,
@@ -543,14 +508,29 @@ static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 		return;
 	}
 
-	memset(&frame, 0, sizeof(frame));
-	frame.cc = HDMI_AUDIO_CC_2;
+	err = hdmi_audio_infoframe_init(&frame);
+	if (err < 0) {
+		dev_err(hdmi->dev, "failed to initialize audio infoframe: %d\n",
+			err);
+		return;
+	}
+
+	frame.channels = 2;
+
+	err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		dev_err(hdmi->dev, "failed to pack audio infoframe: %zd\n",
+			err);
+		return;
+	}
 
-	tegra_hdmi_write_infopack(hdmi,
-				  HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER,
-				  HDMI_INFOFRAME_TYPE_AUDIO,
-				  HDMI_AUDIO_VERSION,
-				  &frame, sizeof(frame));
+	/*
+	 * The audio infoframe has only one set of subpack registers, so the
+	 * infoframe needs to be truncated. One set of subpack registers can
+	 * contain 7 bytes. Including the 3 byte header only the first 10
+	 * bytes can be programmed.
+	 */
+	tegra_hdmi_write_infopack(hdmi, buffer, min(10, err));
 
 	tegra_hdmi_writel(hdmi, INFOFRAME_CTRL_ENABLE,
 			  HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL);
@@ -558,8 +538,10 @@ static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 
 static void tegra_hdmi_setup_stereo_infoframe(struct tegra_hdmi *hdmi)
 {
-	struct hdmi_stereo_infoframe frame;
+	struct hdmi_vendor_infoframe frame;
 	unsigned long value;
+	u8 buffer[10];
+	ssize_t err;
 
 	if (!hdmi->stereo) {
 		value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_HDMI_GENERIC_CTRL);
@@ -569,22 +551,32 @@ static void tegra_hdmi_setup_stereo_infoframe(struct tegra_hdmi *hdmi)
 	}
 
 	memset(&frame, 0, sizeof(frame));
-	frame.regid0 = 0x03;
-	frame.regid1 = 0x0c;
-	frame.regid2 = 0x00;
-	frame.hdmi_video_format = 2;
+
+	frame.type = HDMI_INFOFRAME_TYPE_VENDOR;
+	frame.version = 0x01;
+	frame.length = 6;
+
+	frame.data[0] = 0x03; /* regid0 */
+	frame.data[1] = 0x0c; /* regid1 */
+	frame.data[2] = 0x00; /* regid2 */
+	frame.data[3] = 0x02 << 5; /* video format */
 
 	/* TODO: 74 MHz limit? */
 	if (1) {
-		frame._3d_structure = 0;
+		frame.data[4] = 0x00 << 4; /* 3D structure */
 	} else {
-		frame._3d_structure = 8;
-		frame._3d_ext_data = 0;
+		frame.data[4] = 0x08 << 4; /* 3D structure */
+		frame.data[5] = 0x00 << 4; /* 3D ext. data */
+	}
+
+	err = hdmi_vendor_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		dev_err(hdmi->dev, "failed to pack vendor infoframe: %zd\n",
+			err);
+		return;
 	}
 
-	tegra_hdmi_write_infopack(hdmi, HDMI_NV_PDISP_HDMI_GENERIC_HEADER,
-				  HDMI_INFOFRAME_TYPE_VENDOR,
-				  HDMI_VENDOR_VERSION, &frame, 6);
+	tegra_hdmi_write_infopack(hdmi, buffer, err);
 
 	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_HDMI_GENERIC_CTRL);
 	value |= GENERIC_CTRL_ENABLE;
diff --git a/drivers/gpu/drm/tegra/hdmi.h b/drivers/gpu/drm/tegra/hdmi.h
index 1477f36eb45a..52ac36e08ccb 100644
--- a/drivers/gpu/drm/tegra/hdmi.h
+++ b/drivers/gpu/drm/tegra/hdmi.h
@@ -10,195 +10,6 @@
 #ifndef TEGRA_HDMI_H
 #define TEGRA_HDMI_H 1
 
-#define HDMI_INFOFRAME_TYPE_VENDOR   0x81
-#define HDMI_INFOFRAME_TYPE_AVI      0x82
-#define HDMI_INFOFRAME_TYPE_SPD      0x83
-#define HDMI_INFOFRAME_TYPE_AUDIO    0x84
-#define HDMI_INFOFRAME_TYPE_MPEG_SRC 0x85
-#define HDMI_INFOFRAME_TYPE_NTSC_VBI 0x86
-
-/* all fields little endian */
-struct hdmi_avi_infoframe {
-	/* PB0 */
-	u8 csum;
-
-	/* PB1 */
-	unsigned s:2; /* scan information */
-	unsigned b:2; /* bar info data valid */
-	unsigned a:1; /* active info present */
-	unsigned y:2; /* RGB or YCbCr */
-	unsigned res1:1;
-
-	/* PB2 */
-	unsigned r:4; /* active format aspect ratio */
-	unsigned m:2; /* picture aspect ratio */
-	unsigned c:2; /* colorimetry */
-
-	/* PB3 */
-	unsigned sc:2;  /* scan information */
-	unsigned q:2;   /* quantization range */
-	unsigned ec:3;  /* extended colorimetry */
-	unsigned itc:1; /* it content */
-
-	/* PB4 */
-	unsigned vic:7; /* video format id code */
-	unsigned res4:1;
-
-	/* PB5 */
-	unsigned pr:4; /* pixel repetition factor */
-	unsigned cn:2; /* it content type*/
-	unsigned yq:2; /* ycc quantization range */
-
-	/* PB6-7 */
-	u16 top_bar_end_line;
-
-	/* PB8-9 */
-	u16 bot_bar_start_line;
-
-	/* PB10-11 */
-	u16 left_bar_end_pixel;
-
-	/* PB12-13 */
-	u16 right_bar_start_pixel;
-} __packed;
-
-#define HDMI_AVI_VERSION 0x02
-
-#define HDMI_AVI_Y_RGB       0x0
-#define HDMI_AVI_Y_YCBCR_422 0x1
-#define HDMI_AVI_Y_YCBCR_444 0x2
-
-#define HDMI_AVI_B_VERT  0x1
-#define HDMI_AVI_B_HORIZ 0x2
-
-#define HDMI_AVI_S_NONE      0x0
-#define HDMI_AVI_S_OVERSCAN  0x1
-#define HDMI_AVI_S_UNDERSCAN 0x2
-
-#define HDMI_AVI_C_NONE     0x0
-#define HDMI_AVI_C_SMPTE    0x1
-#define HDMI_AVI_C_ITU_R    0x2
-#define HDMI_AVI_C_EXTENDED 0x4
-
-#define HDMI_AVI_M_4_3  0x1
-#define HDMI_AVI_M_16_9 0x2
-
-#define HDMI_AVI_R_SAME        0x8
-#define HDMI_AVI_R_4_3_CENTER  0x9
-#define HDMI_AVI_R_16_9_CENTER 0xa
-#define HDMI_AVI_R_14_9_CENTER 0xb
-
-/* all fields little endian */
-struct hdmi_audio_infoframe {
-	/* PB0 */
-	u8 csum;
-
-	/* PB1 */
-	unsigned cc:3; /* channel count */
-	unsigned res1:1;
-	unsigned ct:4; /* coding type */
-
-	/* PB2 */
-	unsigned ss:2; /* sample size */
-	unsigned sf:3; /* sample frequency */
-	unsigned res2:3;
-
-	/* PB3 */
-	unsigned cxt:5; /* coding extention type */
-	unsigned res3:3;
-
-	/* PB4 */
-	u8 ca; /* channel/speaker allocation */
-
-	/* PB5 */
-	unsigned res5:3;
-	unsigned lsv:4; /* level shift value */
-	unsigned dm_inh:1; /* downmix inhibit */
-
-	/* PB6-10 reserved */
-	u8 res6;
-	u8 res7;
-	u8 res8;
-	u8 res9;
-	u8 res10;
-} __packed;
-
-#define HDMI_AUDIO_VERSION 0x01
-
-#define HDMI_AUDIO_CC_STREAM 0x0 /* specified by audio stream */
-#define HDMI_AUDIO_CC_2      0x1
-#define HDMI_AUDIO_CC_3      0x2
-#define HDMI_AUDIO_CC_4      0x3
-#define HDMI_AUDIO_CC_5      0x4
-#define HDMI_AUDIO_CC_6      0x5
-#define HDMI_AUDIO_CC_7      0x6
-#define HDMI_AUDIO_CC_8      0x7
-
-#define HDMI_AUDIO_CT_STREAM  0x0 /* specified by audio stream */
-#define HDMI_AUDIO_CT_PCM     0x1
-#define HDMI_AUDIO_CT_AC3     0x2
-#define HDMI_AUDIO_CT_MPEG1   0x3
-#define HDMI_AUDIO_CT_MP3     0x4
-#define HDMI_AUDIO_CT_MPEG2   0x5
-#define HDMI_AUDIO_CT_AAC_LC  0x6
-#define HDMI_AUDIO_CT_DTS     0x7
-#define HDMI_AUDIO_CT_ATRAC   0x8
-#define HDMI_AUDIO_CT_DSD     0x9
-#define HDMI_AUDIO_CT_E_AC3   0xa
-#define HDMI_AUDIO_CT_DTS_HD  0xb
-#define HDMI_AUDIO_CT_MLP     0xc
-#define HDMI_AUDIO_CT_DST     0xd
-#define HDMI_AUDIO_CT_WMA_PRO 0xe
-#define HDMI_AUDIO_CT_CXT     0xf
-
-#define HDMI_AUDIO_SF_STREAM 0x0 /* specified by audio stream */
-#define HDMI_AUIDO_SF_32K    0x1
-#define HDMI_AUDIO_SF_44_1K  0x2
-#define HDMI_AUDIO_SF_48K    0x3
-#define HDMI_AUDIO_SF_88_2K  0x4
-#define HDMI_AUDIO_SF_96K    0x5
-#define HDMI_AUDIO_SF_176_4K 0x6
-#define HDMI_AUDIO_SF_192K   0x7
-
-#define HDMI_AUDIO_SS_STREAM 0x0 /* specified by audio stream */
-#define HDMI_AUDIO_SS_16BIT  0x1
-#define HDMI_AUDIO_SS_20BIT  0x2
-#define HDMI_AUDIO_SS_24BIT  0x3
-
-#define HDMI_AUDIO_CXT_CT            0x0 /* refer to coding in CT */
-#define HDMI_AUDIO_CXT_HE_AAC        0x1
-#define HDMI_AUDIO_CXT_HE_AAC_V2     0x2
-#define HDMI_AUDIO_CXT_MPEG_SURROUND 0x3
-
-/* all fields little endian */
-struct hdmi_stereo_infoframe {
-	/* PB0 */
-	u8 csum;
-
-	/* PB1 */
-	u8 regid0;
-
-	/* PB2 */
-	u8 regid1;
-
-	/* PB3 */
-	u8 regid2;
-
-	/* PB4 */
-	unsigned res1:5;
-	unsigned hdmi_video_format:3;
-
-	/* PB5 */
-	unsigned res2:4;
-	unsigned _3d_structure:4;
-
-	/* PB6*/
-	unsigned res3:4;
-	unsigned _3d_ext_data:4;
-} __packed;
-
-#define HDMI_VENDOR_VERSION 0x01
-
 /* register definitions */
 #define HDMI_CTXSW						0x00
 
diff --git a/drivers/gpu/drm/tilcdc/Kconfig b/drivers/gpu/drm/tilcdc/Kconfig
new file mode 100644
index 000000000000..d24d04013476
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/Kconfig
@@ -0,0 +1,13 @@
+config DRM_TILCDC
+	tristate "DRM Support for TI LCDC Display Controller"
+	depends on DRM && OF && ARM
+	select DRM_KMS_HELPER
+	select DRM_KMS_CMA_HELPER
+	select DRM_GEM_CMA_HELPER
+	select OF_VIDEOMODE
+	select OF_DISPLAY_TIMING
+	select BACKLIGHT_CLASS_DEVICE
+	help
+	  Choose this option if you have an TI SoC with LCDC display
+	  controller, for example AM33xx in beagle-bone, DA8xx, or
+	  OMAP-L1xx.  This driver replaces the FB_DA8XX fbdev driver.
diff --git a/drivers/gpu/drm/tilcdc/Makefile b/drivers/gpu/drm/tilcdc/Makefile
new file mode 100644
index 000000000000..deda656b10e7
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/Makefile
@@ -0,0 +1,10 @@
+ccflags-y := -Iinclude/drm -Werror
+
+tilcdc-y := \
+	tilcdc_crtc.o \
+	tilcdc_tfp410.o \
+	tilcdc_slave.o \
+	tilcdc_panel.o \
+	tilcdc_drv.o
+
+obj-$(CONFIG_DRM_TILCDC)	+= tilcdc.o
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_crtc.c b/drivers/gpu/drm/tilcdc/tilcdc_crtc.c
new file mode 100644
index 000000000000..5dd3c7d031d5
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_crtc.c
@@ -0,0 +1,602 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/kfifo.h>
+
+#include "tilcdc_drv.h"
+#include "tilcdc_regs.h"
+
+struct tilcdc_crtc {
+	struct drm_crtc base;
+
+	const struct tilcdc_panel_info *info;
+	uint32_t dirty;
+	dma_addr_t start, end;
+	struct drm_pending_vblank_event *event;
+	int dpms;
+	wait_queue_head_t frame_done_wq;
+	bool frame_done;
+
+	/* fb currently set to scanout 0/1: */
+	struct drm_framebuffer *scanout[2];
+
+	/* for deferred fb unref's: */
+	DECLARE_KFIFO_PTR(unref_fifo, struct drm_framebuffer *);
+	struct work_struct work;
+};
+#define to_tilcdc_crtc(x) container_of(x, struct tilcdc_crtc, base)
+
+static void unref_worker(struct work_struct *work)
+{
+	struct tilcdc_crtc *tilcdc_crtc = container_of(work, struct tilcdc_crtc, work);
+	struct drm_device *dev = tilcdc_crtc->base.dev;
+	struct drm_framebuffer *fb;
+
+	mutex_lock(&dev->mode_config.mutex);
+	while (kfifo_get(&tilcdc_crtc->unref_fifo, &fb))
+		drm_framebuffer_unreference(fb);
+	mutex_unlock(&dev->mode_config.mutex);
+}
+
+static void set_scanout(struct drm_crtc *crtc, int n)
+{
+	static const uint32_t base_reg[] = {
+			LCDC_DMA_FB_BASE_ADDR_0_REG, LCDC_DMA_FB_BASE_ADDR_1_REG,
+	};
+	static const uint32_t ceil_reg[] = {
+			LCDC_DMA_FB_CEILING_ADDR_0_REG, LCDC_DMA_FB_CEILING_ADDR_1_REG,
+	};
+	static const uint32_t stat[] = {
+			LCDC_END_OF_FRAME0, LCDC_END_OF_FRAME1,
+	};
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+
+	pm_runtime_get_sync(dev->dev);
+	tilcdc_write(dev, base_reg[n], tilcdc_crtc->start);
+	tilcdc_write(dev, ceil_reg[n], tilcdc_crtc->end);
+	if (tilcdc_crtc->scanout[n]) {
+		if (kfifo_put(&tilcdc_crtc->unref_fifo,
+				(const struct drm_framebuffer **)&tilcdc_crtc->scanout[n])) {
+			struct tilcdc_drm_private *priv = dev->dev_private;
+			queue_work(priv->wq, &tilcdc_crtc->work);
+		} else {
+			dev_err(dev->dev, "unref fifo full!\n");
+			drm_framebuffer_unreference(tilcdc_crtc->scanout[n]);
+		}
+	}
+	tilcdc_crtc->scanout[n] = crtc->fb;
+	drm_framebuffer_reference(tilcdc_crtc->scanout[n]);
+	tilcdc_crtc->dirty &= ~stat[n];
+	pm_runtime_put_sync(dev->dev);
+}
+
+static void update_scanout(struct drm_crtc *crtc)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct drm_framebuffer *fb = crtc->fb;
+	struct drm_gem_cma_object *gem;
+	unsigned int depth, bpp;
+
+	drm_fb_get_bpp_depth(fb->pixel_format, &depth, &bpp);
+	gem = drm_fb_cma_get_gem_obj(fb, 0);
+
+	tilcdc_crtc->start = gem->paddr + fb->offsets[0] +
+			(crtc->y * fb->pitches[0]) + (crtc->x * bpp/8);
+
+	tilcdc_crtc->end = tilcdc_crtc->start +
+			(crtc->mode.vdisplay * fb->pitches[0]);
+
+	if (tilcdc_crtc->dpms == DRM_MODE_DPMS_ON) {
+		/* already enabled, so just mark the frames that need
+		 * updating and they will be updated on vblank:
+		 */
+		tilcdc_crtc->dirty |= LCDC_END_OF_FRAME0 | LCDC_END_OF_FRAME1;
+		drm_vblank_get(dev, 0);
+	} else {
+		/* not enabled yet, so update registers immediately: */
+		set_scanout(crtc, 0);
+		set_scanout(crtc, 1);
+	}
+}
+
+static void start(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+
+	if (priv->rev == 2) {
+		tilcdc_set(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
+		msleep(1);
+		tilcdc_clear(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
+		msleep(1);
+	}
+
+	tilcdc_set(dev, LCDC_DMA_CTRL_REG, LCDC_DUAL_FRAME_BUFFER_ENABLE);
+	tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_PALETTE_LOAD_MODE(DATA_ONLY));
+	tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ENABLE);
+}
+
+static void stop(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+
+	tilcdc_clear(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ENABLE);
+}
+
+static void tilcdc_crtc_destroy(struct drm_crtc *crtc)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+
+	WARN_ON(tilcdc_crtc->dpms == DRM_MODE_DPMS_ON);
+
+	drm_crtc_cleanup(crtc);
+	WARN_ON(!kfifo_is_empty(&tilcdc_crtc->unref_fifo));
+	kfifo_free(&tilcdc_crtc->unref_fifo);
+	kfree(tilcdc_crtc);
+}
+
+static int tilcdc_crtc_page_flip(struct drm_crtc *crtc,
+		struct drm_framebuffer *fb,
+		struct drm_pending_vblank_event *event)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+
+	if (tilcdc_crtc->event) {
+		dev_err(dev->dev, "already pending page flip!\n");
+		return -EBUSY;
+	}
+
+	crtc->fb = fb;
+	tilcdc_crtc->event = event;
+	update_scanout(crtc);
+
+	return 0;
+}
+
+static void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+
+	/* we really only care about on or off: */
+	if (mode != DRM_MODE_DPMS_ON)
+		mode = DRM_MODE_DPMS_OFF;
+
+	if (tilcdc_crtc->dpms == mode)
+		return;
+
+	tilcdc_crtc->dpms = mode;
+
+	pm_runtime_get_sync(dev->dev);
+
+	if (mode == DRM_MODE_DPMS_ON) {
+		pm_runtime_forbid(dev->dev);
+		start(crtc);
+	} else {
+		tilcdc_crtc->frame_done = false;
+		stop(crtc);
+
+		/* if necessary wait for framedone irq which will still come
+		 * before putting things to sleep..
+		 */
+		if (priv->rev == 2) {
+			int ret = wait_event_timeout(
+					tilcdc_crtc->frame_done_wq,
+					tilcdc_crtc->frame_done,
+					msecs_to_jiffies(50));
+			if (ret == 0)
+				dev_err(dev->dev, "timeout waiting for framedone\n");
+		}
+		pm_runtime_allow(dev->dev);
+	}
+
+	pm_runtime_put_sync(dev->dev);
+}
+
+static bool tilcdc_crtc_mode_fixup(struct drm_crtc *crtc,
+		const struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	return true;
+}
+
+static void tilcdc_crtc_prepare(struct drm_crtc *crtc)
+{
+	tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+}
+
+static void tilcdc_crtc_commit(struct drm_crtc *crtc)
+{
+	tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_ON);
+}
+
+static int tilcdc_crtc_mode_set(struct drm_crtc *crtc,
+		struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode,
+		int x, int y,
+		struct drm_framebuffer *old_fb)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	const struct tilcdc_panel_info *info = tilcdc_crtc->info;
+	uint32_t reg, hbp, hfp, hsw, vbp, vfp, vsw;
+	int ret;
+
+	ret = tilcdc_crtc_mode_valid(crtc, mode);
+	if (WARN_ON(ret))
+		return ret;
+
+	if (WARN_ON(!info))
+		return -EINVAL;
+
+	pm_runtime_get_sync(dev->dev);
+
+	/* Configure the Burst Size and fifo threshold of DMA: */
+	reg = tilcdc_read(dev, LCDC_DMA_CTRL_REG) & ~0x00000770;
+	switch (info->dma_burst_sz) {
+	case 1:
+		reg |= LCDC_DMA_BURST_SIZE(LCDC_DMA_BURST_1);
+		break;
+	case 2:
+		reg |= LCDC_DMA_BURST_SIZE(LCDC_DMA_BURST_2);
+		break;
+	case 4:
+		reg |= LCDC_DMA_BURST_SIZE(LCDC_DMA_BURST_4);
+		break;
+	case 8:
+		reg |= LCDC_DMA_BURST_SIZE(LCDC_DMA_BURST_8);
+		break;
+	case 16:
+		reg |= LCDC_DMA_BURST_SIZE(LCDC_DMA_BURST_16);
+		break;
+	default:
+		return -EINVAL;
+	}
+	reg |= (info->fifo_th << 8);
+	tilcdc_write(dev, LCDC_DMA_CTRL_REG, reg);
+
+	/* Configure timings: */
+	hbp = mode->htotal - mode->hsync_end;
+	hfp = mode->hsync_start - mode->hdisplay;
+	hsw = mode->hsync_end - mode->hsync_start;
+	vbp = mode->vtotal - mode->vsync_end;
+	vfp = mode->vsync_start - mode->vdisplay;
+	vsw = mode->vsync_end - mode->vsync_start;
+
+	DBG("%dx%d, hbp=%u, hfp=%u, hsw=%u, vbp=%u, vfp=%u, vsw=%u",
+			mode->hdisplay, mode->vdisplay, hbp, hfp, hsw, vbp, vfp, vsw);
+
+	/* Configure the AC Bias Period and Number of Transitions per Interrupt: */
+	reg = tilcdc_read(dev, LCDC_RASTER_TIMING_2_REG) & ~0x000fff00;
+	reg |= LCDC_AC_BIAS_FREQUENCY(info->ac_bias) |
+		LCDC_AC_BIAS_TRANSITIONS_PER_INT(info->ac_bias_intrpt);
+	if (priv->rev == 2) {
+		reg |= (hfp & 0x300) >> 8;
+		reg |= (hbp & 0x300) >> 4;
+		reg |= (hsw & 0x3c0) << 21;
+	}
+	tilcdc_write(dev, LCDC_RASTER_TIMING_2_REG, reg);
+
+	reg = (((mode->hdisplay >> 4) - 1) << 4) |
+		((hbp & 0xff) << 24) |
+		((hfp & 0xff) << 16) |
+		((hsw & 0x3f) << 10);
+	if (priv->rev == 2)
+		reg |= (((mode->hdisplay >> 4) - 1) & 0x40) >> 3;
+	tilcdc_write(dev, LCDC_RASTER_TIMING_0_REG, reg);
+
+	reg = ((mode->vdisplay - 1) & 0x3ff) |
+		((vbp & 0xff) << 24) |
+		((vfp & 0xff) << 16) |
+		((vsw & 0x3f) << 10);
+	tilcdc_write(dev, LCDC_RASTER_TIMING_1_REG, reg);
+
+	/* Configure display type: */
+	reg = tilcdc_read(dev, LCDC_RASTER_CTRL_REG) &
+		~(LCDC_TFT_MODE | LCDC_MONO_8BIT_MODE | LCDC_MONOCHROME_MODE |
+			LCDC_V2_TFT_24BPP_MODE | LCDC_V2_TFT_24BPP_UNPACK | 0x000ff000);
+	reg |= LCDC_TFT_MODE; /* no monochrome/passive support */
+	if (info->tft_alt_mode)
+		reg |= LCDC_TFT_ALT_ENABLE;
+	if (priv->rev == 2) {
+		unsigned int depth, bpp;
+
+		drm_fb_get_bpp_depth(crtc->fb->pixel_format, &depth, &bpp);
+		switch (bpp) {
+		case 16:
+			break;
+		case 32:
+			reg |= LCDC_V2_TFT_24BPP_UNPACK;
+			/* fallthrough */
+		case 24:
+			reg |= LCDC_V2_TFT_24BPP_MODE;
+			break;
+		default:
+			dev_err(dev->dev, "invalid pixel format\n");
+			return -EINVAL;
+		}
+	}
+	reg |= info->fdd < 12;
+	tilcdc_write(dev, LCDC_RASTER_CTRL_REG, reg);
+
+	if (info->invert_pxl_clk)
+		tilcdc_set(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_PIXEL_CLOCK);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_PIXEL_CLOCK);
+
+	if (info->sync_ctrl)
+		tilcdc_set(dev, LCDC_RASTER_TIMING_2_REG, LCDC_SYNC_CTRL);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_TIMING_2_REG, LCDC_SYNC_CTRL);
+
+	if (info->sync_edge)
+		tilcdc_set(dev, LCDC_RASTER_TIMING_2_REG, LCDC_SYNC_EDGE);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_TIMING_2_REG, LCDC_SYNC_EDGE);
+
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		tilcdc_set(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_HSYNC);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_HSYNC);
+
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		tilcdc_set(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_VSYNC);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_TIMING_2_REG, LCDC_INVERT_VSYNC);
+
+	if (info->raster_order)
+		tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ORDER);
+	else
+		tilcdc_clear(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ORDER);
+
+
+	update_scanout(crtc);
+	tilcdc_crtc_update_clk(crtc);
+
+	pm_runtime_put_sync(dev->dev);
+
+	return 0;
+}
+
+static int tilcdc_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
+		struct drm_framebuffer *old_fb)
+{
+	update_scanout(crtc);
+	return 0;
+}
+
+static void tilcdc_crtc_load_lut(struct drm_crtc *crtc)
+{
+}
+
+static const struct drm_crtc_funcs tilcdc_crtc_funcs = {
+		.destroy        = tilcdc_crtc_destroy,
+		.set_config     = drm_crtc_helper_set_config,
+		.page_flip      = tilcdc_crtc_page_flip,
+};
+
+static const struct drm_crtc_helper_funcs tilcdc_crtc_helper_funcs = {
+		.dpms           = tilcdc_crtc_dpms,
+		.mode_fixup     = tilcdc_crtc_mode_fixup,
+		.prepare        = tilcdc_crtc_prepare,
+		.commit         = tilcdc_crtc_commit,
+		.mode_set       = tilcdc_crtc_mode_set,
+		.mode_set_base  = tilcdc_crtc_mode_set_base,
+		.load_lut       = tilcdc_crtc_load_lut,
+};
+
+int tilcdc_crtc_max_width(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	int max_width = 0;
+
+	if (priv->rev == 1)
+		max_width = 1024;
+	else if (priv->rev == 2)
+		max_width = 2048;
+
+	return max_width;
+}
+
+int tilcdc_crtc_mode_valid(struct drm_crtc *crtc, struct drm_display_mode *mode)
+{
+	struct tilcdc_drm_private *priv = crtc->dev->dev_private;
+	unsigned int bandwidth;
+
+	if (mode->hdisplay > tilcdc_crtc_max_width(crtc))
+		return MODE_VIRTUAL_X;
+
+	/* width must be multiple of 16 */
+	if (mode->hdisplay & 0xf)
+		return MODE_VIRTUAL_X;
+
+	if (mode->vdisplay > 2048)
+		return MODE_VIRTUAL_Y;
+
+	/* filter out modes that would require too much memory bandwidth: */
+	bandwidth = mode->hdisplay * mode->vdisplay * drm_mode_vrefresh(mode);
+	if (bandwidth > priv->max_bandwidth)
+		return MODE_BAD;
+
+	return MODE_OK;
+}
+
+void tilcdc_crtc_set_panel_info(struct drm_crtc *crtc,
+		const struct tilcdc_panel_info *info)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	tilcdc_crtc->info = info;
+}
+
+void tilcdc_crtc_update_clk(struct drm_crtc *crtc)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	int dpms = tilcdc_crtc->dpms;
+	unsigned int lcd_clk, div;
+	int ret;
+
+	pm_runtime_get_sync(dev->dev);
+
+	if (dpms == DRM_MODE_DPMS_ON)
+		tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+
+	/* in raster mode, minimum divisor is 2: */
+	ret = clk_set_rate(priv->disp_clk, crtc->mode.clock * 1000 * 2);
+	if (ret) {
+		dev_err(dev->dev, "failed to set display clock rate to: %d\n",
+				crtc->mode.clock);
+		goto out;
+	}
+
+	lcd_clk = clk_get_rate(priv->clk);
+	div = lcd_clk / (crtc->mode.clock * 1000);
+
+	DBG("lcd_clk=%u, mode clock=%d, div=%u", lcd_clk, crtc->mode.clock, div);
+	DBG("fck=%lu, dpll_disp_ck=%lu", clk_get_rate(priv->clk), clk_get_rate(priv->disp_clk));
+
+	/* Configure the LCD clock divisor. */
+	tilcdc_write(dev, LCDC_CTRL_REG, LCDC_CLK_DIVISOR(div) |
+			LCDC_RASTER_MODE);
+
+	if (priv->rev == 2)
+		tilcdc_set(dev, LCDC_CLK_ENABLE_REG,
+				LCDC_V2_DMA_CLK_EN | LCDC_V2_LIDD_CLK_EN |
+				LCDC_V2_CORE_CLK_EN);
+
+	if (dpms == DRM_MODE_DPMS_ON)
+		tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_ON);
+
+out:
+	pm_runtime_put_sync(dev->dev);
+}
+
+irqreturn_t tilcdc_crtc_irq(struct drm_crtc *crtc)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	uint32_t stat = tilcdc_read_irqstatus(dev);
+
+	if ((stat & LCDC_SYNC_LOST) && (stat & LCDC_FIFO_UNDERFLOW)) {
+		stop(crtc);
+		dev_err(dev->dev, "error: %08x\n", stat);
+		tilcdc_clear_irqstatus(dev, stat);
+		start(crtc);
+	} else if (stat & LCDC_PL_LOAD_DONE) {
+		tilcdc_clear_irqstatus(dev, stat);
+	} else {
+		struct drm_pending_vblank_event *event;
+		unsigned long flags;
+		uint32_t dirty = tilcdc_crtc->dirty & stat;
+
+		tilcdc_clear_irqstatus(dev, stat);
+
+		if (dirty & LCDC_END_OF_FRAME0)
+			set_scanout(crtc, 0);
+
+		if (dirty & LCDC_END_OF_FRAME1)
+			set_scanout(crtc, 1);
+
+		drm_handle_vblank(dev, 0);
+
+		spin_lock_irqsave(&dev->event_lock, flags);
+		event = tilcdc_crtc->event;
+		tilcdc_crtc->event = NULL;
+		if (event)
+			drm_send_vblank_event(dev, 0, event);
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+
+		if (dirty && !tilcdc_crtc->dirty)
+			drm_vblank_put(dev, 0);
+	}
+
+	if (priv->rev == 2) {
+		if (stat & LCDC_FRAME_DONE) {
+			tilcdc_crtc->frame_done = true;
+			wake_up(&tilcdc_crtc->frame_done_wq);
+		}
+		tilcdc_write(dev, LCDC_END_OF_INT_IND_REG, 0);
+	}
+
+	return IRQ_HANDLED;
+}
+
+void tilcdc_crtc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file)
+{
+	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
+	struct drm_pending_vblank_event *event;
+	struct drm_device *dev = crtc->dev;
+	unsigned long flags;
+
+	/* Destroy the pending vertical blanking event associated with the
+	 * pending page flip, if any, and disable vertical blanking interrupts.
+	 */
+	spin_lock_irqsave(&dev->event_lock, flags);
+	event = tilcdc_crtc->event;
+	if (event && event->base.file_priv == file) {
+		tilcdc_crtc->event = NULL;
+		event->base.destroy(&event->base);
+		drm_vblank_put(dev, 0);
+	}
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+}
+
+struct drm_crtc *tilcdc_crtc_create(struct drm_device *dev)
+{
+	struct tilcdc_crtc *tilcdc_crtc;
+	struct drm_crtc *crtc;
+	int ret;
+
+	tilcdc_crtc = kzalloc(sizeof(*tilcdc_crtc), GFP_KERNEL);
+	if (!tilcdc_crtc) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	crtc = &tilcdc_crtc->base;
+
+	tilcdc_crtc->dpms = DRM_MODE_DPMS_OFF;
+	init_waitqueue_head(&tilcdc_crtc->frame_done_wq);
+
+	ret = kfifo_alloc(&tilcdc_crtc->unref_fifo, 16, GFP_KERNEL);
+	if (ret) {
+		dev_err(dev->dev, "could not allocate unref FIFO\n");
+		goto fail;
+	}
+
+	INIT_WORK(&tilcdc_crtc->work, unref_worker);
+
+	ret = drm_crtc_init(dev, crtc, &tilcdc_crtc_funcs);
+	if (ret < 0)
+		goto fail;
+
+	drm_crtc_helper_add(crtc, &tilcdc_crtc_helper_funcs);
+
+	return crtc;
+
+fail:
+	tilcdc_crtc_destroy(crtc);
+	return NULL;
+}
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
new file mode 100644
index 000000000000..c5b592dc1970
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -0,0 +1,611 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* LCDC DRM driver, based on da8xx-fb */
+
+#include "tilcdc_drv.h"
+#include "tilcdc_regs.h"
+#include "tilcdc_tfp410.h"
+#include "tilcdc_slave.h"
+#include "tilcdc_panel.h"
+
+#include "drm_fb_helper.h"
+
+static LIST_HEAD(module_list);
+
+void tilcdc_module_init(struct tilcdc_module *mod, const char *name,
+		const struct tilcdc_module_ops *funcs)
+{
+	mod->name = name;
+	mod->funcs = funcs;
+	INIT_LIST_HEAD(&mod->list);
+	list_add(&mod->list, &module_list);
+}
+
+void tilcdc_module_cleanup(struct tilcdc_module *mod)
+{
+	list_del(&mod->list);
+}
+
+static struct of_device_id tilcdc_of_match[];
+
+static struct drm_framebuffer *tilcdc_fb_create(struct drm_device *dev,
+		struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	return drm_fb_cma_create(dev, file_priv, mode_cmd);
+}
+
+static void tilcdc_fb_output_poll_changed(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	if (priv->fbdev)
+		drm_fbdev_cma_hotplug_event(priv->fbdev);
+}
+
+static const struct drm_mode_config_funcs mode_config_funcs = {
+	.fb_create = tilcdc_fb_create,
+	.output_poll_changed = tilcdc_fb_output_poll_changed,
+};
+
+static int modeset_init(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	struct tilcdc_module *mod;
+
+	drm_mode_config_init(dev);
+
+	priv->crtc = tilcdc_crtc_create(dev);
+
+	list_for_each_entry(mod, &module_list, list) {
+		DBG("loading module: %s", mod->name);
+		mod->funcs->modeset_init(mod, dev);
+	}
+
+	if ((priv->num_encoders = 0) || (priv->num_connectors == 0)) {
+		/* oh nos! */
+		dev_err(dev->dev, "no encoders/connectors found\n");
+		return -ENXIO;
+	}
+
+	dev->mode_config.min_width = 0;
+	dev->mode_config.min_height = 0;
+	dev->mode_config.max_width = tilcdc_crtc_max_width(priv->crtc);
+	dev->mode_config.max_height = 2048;
+	dev->mode_config.funcs = &mode_config_funcs;
+
+	return 0;
+}
+
+#ifdef CONFIG_CPU_FREQ
+static int cpufreq_transition(struct notifier_block *nb,
+				     unsigned long val, void *data)
+{
+	struct tilcdc_drm_private *priv = container_of(nb,
+			struct tilcdc_drm_private, freq_transition);
+	if (val == CPUFREQ_POSTCHANGE) {
+		if (priv->lcd_fck_rate != clk_get_rate(priv->clk)) {
+			priv->lcd_fck_rate = clk_get_rate(priv->clk);
+			tilcdc_crtc_update_clk(priv->crtc);
+		}
+	}
+
+	return 0;
+}
+#endif
+
+/*
+ * DRM operations:
+ */
+
+static int tilcdc_unload(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	struct tilcdc_module *mod, *cur;
+
+	drm_kms_helper_poll_fini(dev);
+	drm_mode_config_cleanup(dev);
+	drm_vblank_cleanup(dev);
+
+	pm_runtime_get_sync(dev->dev);
+	drm_irq_uninstall(dev);
+	pm_runtime_put_sync(dev->dev);
+
+#ifdef CONFIG_CPU_FREQ
+	cpufreq_unregister_notifier(&priv->freq_transition,
+			CPUFREQ_TRANSITION_NOTIFIER);
+#endif
+
+	if (priv->clk)
+		clk_put(priv->clk);
+
+	if (priv->mmio)
+		iounmap(priv->mmio);
+
+	flush_workqueue(priv->wq);
+	destroy_workqueue(priv->wq);
+
+	dev->dev_private = NULL;
+
+	pm_runtime_disable(dev->dev);
+
+	list_for_each_entry_safe(mod, cur, &module_list, list) {
+		DBG("destroying module: %s", mod->name);
+		mod->funcs->destroy(mod);
+	}
+
+	kfree(priv);
+
+	return 0;
+}
+
+static int tilcdc_load(struct drm_device *dev, unsigned long flags)
+{
+	struct platform_device *pdev = dev->platformdev;
+	struct device_node *node = pdev->dev.of_node;
+	struct tilcdc_drm_private *priv;
+	struct resource *res;
+	int ret;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		dev_err(dev->dev, "failed to allocate private data\n");
+		return -ENOMEM;
+	}
+
+	dev->dev_private = priv;
+
+	priv->wq = alloc_ordered_workqueue("tilcdc", 0);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(dev->dev, "failed to get memory resource\n");
+		ret = -EINVAL;
+		goto fail;
+	}
+
+	priv->mmio = ioremap_nocache(res->start, resource_size(res));
+	if (!priv->mmio) {
+		dev_err(dev->dev, "failed to ioremap\n");
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	priv->clk = clk_get(dev->dev, "fck");
+	if (IS_ERR(priv->clk)) {
+		dev_err(dev->dev, "failed to get functional clock\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+
+	priv->disp_clk = clk_get(dev->dev, "dpll_disp_ck");
+	if (IS_ERR(priv->clk)) {
+		dev_err(dev->dev, "failed to get display clock\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+
+#ifdef CONFIG_CPU_FREQ
+	priv->lcd_fck_rate = clk_get_rate(priv->clk);
+	priv->freq_transition.notifier_call = cpufreq_transition;
+	ret = cpufreq_register_notifier(&priv->freq_transition,
+			CPUFREQ_TRANSITION_NOTIFIER);
+	if (ret) {
+		dev_err(dev->dev, "failed to register cpufreq notifier\n");
+		goto fail;
+	}
+#endif
+
+	if (of_property_read_u32(node, "max-bandwidth", &priv->max_bandwidth))
+		priv->max_bandwidth = 1280 * 1024 * 60;
+
+	pm_runtime_enable(dev->dev);
+
+	/* Determine LCD IP Version */
+	pm_runtime_get_sync(dev->dev);
+	switch (tilcdc_read(dev, LCDC_PID_REG)) {
+	case 0x4c100102:
+		priv->rev = 1;
+		break;
+	case 0x4f200800:
+	case 0x4f201000:
+		priv->rev = 2;
+		break;
+	default:
+		dev_warn(dev->dev, "Unknown PID Reg value 0x%08x, "
+				"defaulting to LCD revision 1\n",
+				tilcdc_read(dev, LCDC_PID_REG));
+		priv->rev = 1;
+		break;
+	}
+
+	pm_runtime_put_sync(dev->dev);
+
+	ret = modeset_init(dev);
+	if (ret < 0) {
+		dev_err(dev->dev, "failed to initialize mode setting\n");
+		goto fail;
+	}
+
+	ret = drm_vblank_init(dev, 1);
+	if (ret < 0) {
+		dev_err(dev->dev, "failed to initialize vblank\n");
+		goto fail;
+	}
+
+	pm_runtime_get_sync(dev->dev);
+	ret = drm_irq_install(dev);
+	pm_runtime_put_sync(dev->dev);
+	if (ret < 0) {
+		dev_err(dev->dev, "failed to install IRQ handler\n");
+		goto fail;
+	}
+
+	platform_set_drvdata(pdev, dev);
+
+	priv->fbdev = drm_fbdev_cma_init(dev, 16,
+			dev->mode_config.num_crtc,
+			dev->mode_config.num_connector);
+
+	drm_kms_helper_poll_init(dev);
+
+	return 0;
+
+fail:
+	tilcdc_unload(dev);
+	return ret;
+}
+
+static void tilcdc_preclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+
+	tilcdc_crtc_cancel_page_flip(priv->crtc, file);
+}
+
+static void tilcdc_lastclose(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	drm_fbdev_cma_restore_mode(priv->fbdev);
+}
+
+static irqreturn_t tilcdc_irq(DRM_IRQ_ARGS)
+{
+	struct drm_device *dev = arg;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	return tilcdc_crtc_irq(priv->crtc);
+}
+
+static void tilcdc_irq_preinstall(struct drm_device *dev)
+{
+	tilcdc_clear_irqstatus(dev, 0xffffffff);
+}
+
+static int tilcdc_irq_postinstall(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+
+	/* enable FIFO underflow irq: */
+	if (priv->rev == 1) {
+		tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_V1_UNDERFLOW_INT_ENA);
+	} else {
+		tilcdc_set(dev, LCDC_INT_ENABLE_SET_REG, LCDC_V2_UNDERFLOW_INT_ENA);
+	}
+
+	return 0;
+}
+
+static void tilcdc_irq_uninstall(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+
+	/* disable irqs that we might have enabled: */
+	if (priv->rev == 1) {
+		tilcdc_clear(dev, LCDC_RASTER_CTRL_REG,
+				LCDC_V1_UNDERFLOW_INT_ENA | LCDC_V1_PL_INT_ENA);
+		tilcdc_clear(dev, LCDC_DMA_CTRL_REG, LCDC_V1_END_OF_FRAME_INT_ENA);
+	} else {
+		tilcdc_clear(dev, LCDC_INT_ENABLE_SET_REG,
+			LCDC_V2_UNDERFLOW_INT_ENA | LCDC_V2_PL_INT_ENA |
+			LCDC_V2_END_OF_FRAME0_INT_ENA | LCDC_V2_END_OF_FRAME1_INT_ENA |
+			LCDC_FRAME_DONE);
+	}
+
+}
+
+static void enable_vblank(struct drm_device *dev, bool enable)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	u32 reg, mask;
+
+	if (priv->rev == 1) {
+		reg = LCDC_DMA_CTRL_REG;
+		mask = LCDC_V1_END_OF_FRAME_INT_ENA;
+	} else {
+		reg = LCDC_INT_ENABLE_SET_REG;
+		mask = LCDC_V2_END_OF_FRAME0_INT_ENA |
+			LCDC_V2_END_OF_FRAME1_INT_ENA | LCDC_FRAME_DONE;
+	}
+
+	if (enable)
+		tilcdc_set(dev, reg, mask);
+	else
+		tilcdc_clear(dev, reg, mask);
+}
+
+static int tilcdc_enable_vblank(struct drm_device *dev, int crtc)
+{
+	enable_vblank(dev, true);
+	return 0;
+}
+
+static void tilcdc_disable_vblank(struct drm_device *dev, int crtc)
+{
+	enable_vblank(dev, false);
+}
+
+#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_PM_SLEEP)
+static const struct {
+	const char *name;
+	uint8_t  rev;
+	uint8_t  save;
+	uint32_t reg;
+} registers[] = 	{
+#define REG(rev, save, reg) { #reg, rev, save, reg }
+		/* exists in revision 1: */
+		REG(1, false, LCDC_PID_REG),
+		REG(1, true,  LCDC_CTRL_REG),
+		REG(1, false, LCDC_STAT_REG),
+		REG(1, true,  LCDC_RASTER_CTRL_REG),
+		REG(1, true,  LCDC_RASTER_TIMING_0_REG),
+		REG(1, true,  LCDC_RASTER_TIMING_1_REG),
+		REG(1, true,  LCDC_RASTER_TIMING_2_REG),
+		REG(1, true,  LCDC_DMA_CTRL_REG),
+		REG(1, true,  LCDC_DMA_FB_BASE_ADDR_0_REG),
+		REG(1, true,  LCDC_DMA_FB_CEILING_ADDR_0_REG),
+		REG(1, true,  LCDC_DMA_FB_BASE_ADDR_1_REG),
+		REG(1, true,  LCDC_DMA_FB_CEILING_ADDR_1_REG),
+		/* new in revision 2: */
+		REG(2, false, LCDC_RAW_STAT_REG),
+		REG(2, false, LCDC_MASKED_STAT_REG),
+		REG(2, false, LCDC_INT_ENABLE_SET_REG),
+		REG(2, false, LCDC_INT_ENABLE_CLR_REG),
+		REG(2, false, LCDC_END_OF_INT_IND_REG),
+		REG(2, true,  LCDC_CLK_ENABLE_REG),
+		REG(2, true,  LCDC_INT_ENABLE_SET_REG),
+#undef REG
+};
+#endif
+
+#ifdef CONFIG_DEBUG_FS
+static int tilcdc_regs_show(struct seq_file *m, void *arg)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	unsigned i;
+
+	pm_runtime_get_sync(dev->dev);
+
+	seq_printf(m, "revision: %d\n", priv->rev);
+
+	for (i = 0; i < ARRAY_SIZE(registers); i++)
+		if (priv->rev >= registers[i].rev)
+			seq_printf(m, "%s:\t %08x\n", registers[i].name,
+					tilcdc_read(dev, registers[i].reg));
+
+	pm_runtime_put_sync(dev->dev);
+
+	return 0;
+}
+
+static int tilcdc_mm_show(struct seq_file *m, void *arg)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	return drm_mm_dump_table(m, dev->mm_private);
+}
+
+static struct drm_info_list tilcdc_debugfs_list[] = {
+		{ "regs", tilcdc_regs_show, 0 },
+		{ "mm",   tilcdc_mm_show,   0 },
+		{ "fb",   drm_fb_cma_debugfs_show, 0 },
+};
+
+static int tilcdc_debugfs_init(struct drm_minor *minor)
+{
+	struct drm_device *dev = minor->dev;
+	struct tilcdc_module *mod;
+	int ret;
+
+	ret = drm_debugfs_create_files(tilcdc_debugfs_list,
+			ARRAY_SIZE(tilcdc_debugfs_list),
+			minor->debugfs_root, minor);
+
+	list_for_each_entry(mod, &module_list, list)
+		if (mod->funcs->debugfs_init)
+			mod->funcs->debugfs_init(mod, minor);
+
+	if (ret) {
+		dev_err(dev->dev, "could not install tilcdc_debugfs_list\n");
+		return ret;
+	}
+
+	return ret;
+}
+
+static void tilcdc_debugfs_cleanup(struct drm_minor *minor)
+{
+	struct tilcdc_module *mod;
+	drm_debugfs_remove_files(tilcdc_debugfs_list,
+			ARRAY_SIZE(tilcdc_debugfs_list), minor);
+
+	list_for_each_entry(mod, &module_list, list)
+		if (mod->funcs->debugfs_cleanup)
+			mod->funcs->debugfs_cleanup(mod, minor);
+}
+#endif
+
+static const struct file_operations fops = {
+	.owner              = THIS_MODULE,
+	.open               = drm_open,
+	.release            = drm_release,
+	.unlocked_ioctl     = drm_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl       = drm_compat_ioctl,
+#endif
+	.poll               = drm_poll,
+	.read               = drm_read,
+	.fasync             = drm_fasync,
+	.llseek             = no_llseek,
+	.mmap               = drm_gem_cma_mmap,
+};
+
+static struct drm_driver tilcdc_driver = {
+	.driver_features    = DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET,
+	.load               = tilcdc_load,
+	.unload             = tilcdc_unload,
+	.preclose           = tilcdc_preclose,
+	.lastclose          = tilcdc_lastclose,
+	.irq_handler        = tilcdc_irq,
+	.irq_preinstall     = tilcdc_irq_preinstall,
+	.irq_postinstall    = tilcdc_irq_postinstall,
+	.irq_uninstall      = tilcdc_irq_uninstall,
+	.get_vblank_counter = drm_vblank_count,
+	.enable_vblank      = tilcdc_enable_vblank,
+	.disable_vblank     = tilcdc_disable_vblank,
+	.gem_free_object    = drm_gem_cma_free_object,
+	.gem_vm_ops         = &drm_gem_cma_vm_ops,
+	.dumb_create        = drm_gem_cma_dumb_create,
+	.dumb_map_offset    = drm_gem_cma_dumb_map_offset,
+	.dumb_destroy       = drm_gem_cma_dumb_destroy,
+#ifdef CONFIG_DEBUG_FS
+	.debugfs_init       = tilcdc_debugfs_init,
+	.debugfs_cleanup    = tilcdc_debugfs_cleanup,
+#endif
+	.fops               = &fops,
+	.name               = "tilcdc",
+	.desc               = "TI LCD Controller DRM",
+	.date               = "20121205",
+	.major              = 1,
+	.minor              = 0,
+};
+
+/*
+ * Power management:
+ */
+
+#ifdef CONFIG_PM_SLEEP
+static int tilcdc_pm_suspend(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct tilcdc_drm_private *priv = ddev->dev_private;
+	unsigned i, n = 0;
+
+	drm_kms_helper_poll_disable(ddev);
+
+	/* Save register state: */
+	for (i = 0; i < ARRAY_SIZE(registers); i++)
+		if (registers[i].save && (priv->rev >= registers[i].rev))
+			priv->saved_register[n++] = tilcdc_read(ddev, registers[i].reg);
+
+	return 0;
+}
+
+static int tilcdc_pm_resume(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct tilcdc_drm_private *priv = ddev->dev_private;
+	unsigned i, n = 0;
+
+	/* Restore register state: */
+	for (i = 0; i < ARRAY_SIZE(registers); i++)
+		if (registers[i].save && (priv->rev >= registers[i].rev))
+			tilcdc_write(ddev, registers[i].reg, priv->saved_register[n++]);
+
+	drm_kms_helper_poll_enable(ddev);
+
+	return 0;
+}
+#endif
+
+static const struct dev_pm_ops tilcdc_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(tilcdc_pm_suspend, tilcdc_pm_resume)
+};
+
+/*
+ * Platform driver:
+ */
+
+static int tilcdc_pdev_probe(struct platform_device *pdev)
+{
+	/* bail out early if no DT data: */
+	if (!pdev->dev.of_node) {
+		dev_err(&pdev->dev, "device-tree data is missing\n");
+		return -ENXIO;
+	}
+
+	return drm_platform_init(&tilcdc_driver, pdev);
+}
+
+static int tilcdc_pdev_remove(struct platform_device *pdev)
+{
+	drm_platform_exit(&tilcdc_driver, pdev);
+
+	return 0;
+}
+
+static struct of_device_id tilcdc_of_match[] = {
+		{ .compatible = "ti,am33xx-tilcdc", },
+		{ },
+};
+MODULE_DEVICE_TABLE(of, tilcdc_of_match);
+
+static struct platform_driver tilcdc_platform_driver = {
+	.probe      = tilcdc_pdev_probe,
+	.remove     = tilcdc_pdev_remove,
+	.driver     = {
+		.owner  = THIS_MODULE,
+		.name   = "tilcdc",
+		.pm     = &tilcdc_pm_ops,
+		.of_match_table = tilcdc_of_match,
+	},
+};
+
+static int __init tilcdc_drm_init(void)
+{
+	DBG("init");
+	tilcdc_tfp410_init();
+	tilcdc_slave_init();
+	tilcdc_panel_init();
+	return platform_driver_register(&tilcdc_platform_driver);
+}
+
+static void __exit tilcdc_drm_fini(void)
+{
+	DBG("fini");
+	tilcdc_tfp410_fini();
+	tilcdc_slave_fini();
+	tilcdc_panel_fini();
+	platform_driver_unregister(&tilcdc_platform_driver);
+}
+
+late_initcall(tilcdc_drm_init);
+module_exit(tilcdc_drm_fini);
+
+MODULE_AUTHOR("Rob Clark <robdclark@gmail.com");
+MODULE_DESCRIPTION("TI LCD Controller DRM Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.h b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
new file mode 100644
index 000000000000..8242b5a4307b
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
@@ -0,0 +1,150 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __TILCDC_DRV_H__
+#define __TILCDC_DRV_H__
+
+#include <linux/clk.h>
+#include <linux/cpufreq.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/pm_runtime.h>
+#include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/list.h>
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_fb_cma_helper.h>
+
+struct tilcdc_drm_private {
+	void __iomem *mmio;
+
+	struct clk *disp_clk;    /* display dpll */
+	struct clk *clk;         /* functional clock */
+	int rev;                 /* IP revision */
+
+	/* don't attempt resolutions w/ higher W * H * Hz: */
+	uint32_t max_bandwidth;
+
+	/* register contents saved across suspend/resume: */
+	u32 saved_register[12];
+
+#ifdef CONFIG_CPU_FREQ
+	struct notifier_block freq_transition;
+	unsigned int lcd_fck_rate;
+#endif
+
+	struct workqueue_struct *wq;
+
+	struct drm_fbdev_cma *fbdev;
+
+	struct drm_crtc *crtc;
+
+	unsigned int num_encoders;
+	struct drm_encoder *encoders[8];
+
+	unsigned int num_connectors;
+	struct drm_connector *connectors[8];
+};
+
+/* Sub-module for display.  Since we don't know at compile time what panels
+ * or display adapter(s) might be present (for ex, off chip dvi/tfp410,
+ * hdmi encoder, various lcd panels), the connector/encoder(s) are split into
+ * separate drivers.  If they are probed and found to be present, they
+ * register themselves with tilcdc_register_module().
+ */
+struct tilcdc_module;
+
+struct tilcdc_module_ops {
+	/* create appropriate encoders/connectors: */
+	int (*modeset_init)(struct tilcdc_module *mod, struct drm_device *dev);
+	void (*destroy)(struct tilcdc_module *mod);
+#ifdef CONFIG_DEBUG_FS
+	/* create debugfs nodes (can be NULL): */
+	int (*debugfs_init)(struct tilcdc_module *mod, struct drm_minor *minor);
+	/* cleanup debugfs nodes (can be NULL): */
+	void (*debugfs_cleanup)(struct tilcdc_module *mod, struct drm_minor *minor);
+#endif
+};
+
+struct tilcdc_module {
+	const char *name;
+	struct list_head list;
+	const struct tilcdc_module_ops *funcs;
+};
+
+void tilcdc_module_init(struct tilcdc_module *mod, const char *name,
+		const struct tilcdc_module_ops *funcs);
+void tilcdc_module_cleanup(struct tilcdc_module *mod);
+
+
+/* Panel config that needs to be set in the crtc, but is not coming from
+ * the mode timings.  The display module is expected to call
+ * tilcdc_crtc_set_panel_info() to set this during modeset.
+ */
+struct tilcdc_panel_info {
+
+	/* AC Bias Pin Frequency */
+	uint32_t ac_bias;
+
+	/* AC Bias Pin Transitions per Interrupt */
+	uint32_t ac_bias_intrpt;
+
+	/* DMA burst size */
+	uint32_t dma_burst_sz;
+
+	/* Bits per pixel */
+	uint32_t bpp;
+
+	/* FIFO DMA Request Delay */
+	uint32_t fdd;
+
+	/* TFT Alternative Signal Mapping (Only for active) */
+	bool tft_alt_mode;
+
+	/* Invert pixel clock */
+	bool invert_pxl_clk;
+
+	/* Horizontal and Vertical Sync Edge: 0=rising 1=falling */
+	uint32_t sync_edge;
+
+	/* Horizontal and Vertical Sync: Control: 0=ignore */
+	uint32_t sync_ctrl;
+
+	/* Raster Data Order Select: 1=Most-to-least 0=Least-to-most */
+	uint32_t raster_order;
+
+	/* DMA FIFO threshold */
+	uint32_t fifo_th;
+};
+
+#define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
+
+struct drm_crtc *tilcdc_crtc_create(struct drm_device *dev);
+void tilcdc_crtc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file);
+irqreturn_t tilcdc_crtc_irq(struct drm_crtc *crtc);
+void tilcdc_crtc_update_clk(struct drm_crtc *crtc);
+void tilcdc_crtc_set_panel_info(struct drm_crtc *crtc,
+		const struct tilcdc_panel_info *info);
+int tilcdc_crtc_mode_valid(struct drm_crtc *crtc, struct drm_display_mode *mode);
+int tilcdc_crtc_max_width(struct drm_crtc *crtc);
+
+#endif /* __TILCDC_DRV_H__ */
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_panel.c b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
new file mode 100644
index 000000000000..580b74e2022b
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
@@ -0,0 +1,436 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/pinctrl/pinmux.h>
+#include <linux/pinctrl/consumer.h>
+#include <linux/backlight.h>
+#include <video/display_timing.h>
+#include <video/of_display_timing.h>
+#include <video/videomode.h>
+
+#include "tilcdc_drv.h"
+
+struct panel_module {
+	struct tilcdc_module base;
+	struct tilcdc_panel_info *info;
+	struct display_timings *timings;
+	struct backlight_device *backlight;
+};
+#define to_panel_module(x) container_of(x, struct panel_module, base)
+
+
+/*
+ * Encoder:
+ */
+
+struct panel_encoder {
+	struct drm_encoder base;
+	struct panel_module *mod;
+};
+#define to_panel_encoder(x) container_of(x, struct panel_encoder, base)
+
+
+static void panel_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct panel_encoder *panel_encoder = to_panel_encoder(encoder);
+	drm_encoder_cleanup(encoder);
+	kfree(panel_encoder);
+}
+
+static void panel_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct panel_encoder *panel_encoder = to_panel_encoder(encoder);
+	struct backlight_device *backlight = panel_encoder->mod->backlight;
+
+	if (!backlight)
+		return;
+
+	backlight->props.power = mode == DRM_MODE_DPMS_ON
+				     ? FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN;
+	backlight_update_status(backlight);
+}
+
+static bool panel_encoder_mode_fixup(struct drm_encoder *encoder,
+		const struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	/* nothing needed */
+	return true;
+}
+
+static void panel_encoder_prepare(struct drm_encoder *encoder)
+{
+	struct panel_encoder *panel_encoder = to_panel_encoder(encoder);
+	panel_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
+	tilcdc_crtc_set_panel_info(encoder->crtc, panel_encoder->mod->info);
+}
+
+static void panel_encoder_commit(struct drm_encoder *encoder)
+{
+	panel_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
+}
+
+static void panel_encoder_mode_set(struct drm_encoder *encoder,
+		struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	/* nothing needed */
+}
+
+static const struct drm_encoder_funcs panel_encoder_funcs = {
+		.destroy        = panel_encoder_destroy,
+};
+
+static const struct drm_encoder_helper_funcs panel_encoder_helper_funcs = {
+		.dpms           = panel_encoder_dpms,
+		.mode_fixup     = panel_encoder_mode_fixup,
+		.prepare        = panel_encoder_prepare,
+		.commit         = panel_encoder_commit,
+		.mode_set       = panel_encoder_mode_set,
+};
+
+static struct drm_encoder *panel_encoder_create(struct drm_device *dev,
+		struct panel_module *mod)
+{
+	struct panel_encoder *panel_encoder;
+	struct drm_encoder *encoder;
+	int ret;
+
+	panel_encoder = kzalloc(sizeof(*panel_encoder), GFP_KERNEL);
+	if (!panel_encoder) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	panel_encoder->mod = mod;
+
+	encoder = &panel_encoder->base;
+	encoder->possible_crtcs = 1;
+
+	ret = drm_encoder_init(dev, encoder, &panel_encoder_funcs,
+			DRM_MODE_ENCODER_LVDS);
+	if (ret < 0)
+		goto fail;
+
+	drm_encoder_helper_add(encoder, &panel_encoder_helper_funcs);
+
+	return encoder;
+
+fail:
+	panel_encoder_destroy(encoder);
+	return NULL;
+}
+
+/*
+ * Connector:
+ */
+
+struct panel_connector {
+	struct drm_connector base;
+
+	struct drm_encoder *encoder;  /* our connected encoder */
+	struct panel_module *mod;
+};
+#define to_panel_connector(x) container_of(x, struct panel_connector, base)
+
+
+static void panel_connector_destroy(struct drm_connector *connector)
+{
+	struct panel_connector *panel_connector = to_panel_connector(connector);
+	drm_connector_cleanup(connector);
+	kfree(panel_connector);
+}
+
+static enum drm_connector_status panel_connector_detect(
+		struct drm_connector *connector,
+		bool force)
+{
+	return connector_status_connected;
+}
+
+static int panel_connector_get_modes(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+	struct panel_connector *panel_connector = to_panel_connector(connector);
+	struct display_timings *timings = panel_connector->mod->timings;
+	int i;
+
+	for (i = 0; i < timings->num_timings; i++) {
+		struct drm_display_mode *mode = drm_mode_create(dev);
+		struct videomode vm;
+
+		if (videomode_from_timing(timings, &vm, i))
+			break;
+
+		drm_display_mode_from_videomode(&vm, mode);
+
+		mode->type = DRM_MODE_TYPE_DRIVER;
+
+		if (timings->native_mode == i)
+			mode->type |= DRM_MODE_TYPE_PREFERRED;
+
+		drm_mode_set_name(mode);
+		drm_mode_probed_add(connector, mode);
+	}
+
+	return i;
+}
+
+static int panel_connector_mode_valid(struct drm_connector *connector,
+		  struct drm_display_mode *mode)
+{
+	struct tilcdc_drm_private *priv = connector->dev->dev_private;
+	/* our only constraints are what the crtc can generate: */
+	return tilcdc_crtc_mode_valid(priv->crtc, mode);
+}
+
+static struct drm_encoder *panel_connector_best_encoder(
+		struct drm_connector *connector)
+{
+	struct panel_connector *panel_connector = to_panel_connector(connector);
+	return panel_connector->encoder;
+}
+
+static const struct drm_connector_funcs panel_connector_funcs = {
+	.destroy            = panel_connector_destroy,
+	.dpms               = drm_helper_connector_dpms,
+	.detect             = panel_connector_detect,
+	.fill_modes         = drm_helper_probe_single_connector_modes,
+};
+
+static const struct drm_connector_helper_funcs panel_connector_helper_funcs = {
+	.get_modes          = panel_connector_get_modes,
+	.mode_valid         = panel_connector_mode_valid,
+	.best_encoder       = panel_connector_best_encoder,
+};
+
+static struct drm_connector *panel_connector_create(struct drm_device *dev,
+		struct panel_module *mod, struct drm_encoder *encoder)
+{
+	struct panel_connector *panel_connector;
+	struct drm_connector *connector;
+	int ret;
+
+	panel_connector = kzalloc(sizeof(*panel_connector), GFP_KERNEL);
+	if (!panel_connector) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	panel_connector->encoder = encoder;
+	panel_connector->mod = mod;
+
+	connector = &panel_connector->base;
+
+	drm_connector_init(dev, connector, &panel_connector_funcs,
+			DRM_MODE_CONNECTOR_LVDS);
+	drm_connector_helper_add(connector, &panel_connector_helper_funcs);
+
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+
+	ret = drm_mode_connector_attach_encoder(connector, encoder);
+	if (ret)
+		goto fail;
+
+	drm_sysfs_connector_add(connector);
+
+	return connector;
+
+fail:
+	panel_connector_destroy(connector);
+	return NULL;
+}
+
+/*
+ * Module:
+ */
+
+static int panel_modeset_init(struct tilcdc_module *mod, struct drm_device *dev)
+{
+	struct panel_module *panel_mod = to_panel_module(mod);
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+
+	encoder = panel_encoder_create(dev, panel_mod);
+	if (!encoder)
+		return -ENOMEM;
+
+	connector = panel_connector_create(dev, panel_mod, encoder);
+	if (!connector)
+		return -ENOMEM;
+
+	priv->encoders[priv->num_encoders++] = encoder;
+	priv->connectors[priv->num_connectors++] = connector;
+
+	return 0;
+}
+
+static void panel_destroy(struct tilcdc_module *mod)
+{
+	struct panel_module *panel_mod = to_panel_module(mod);
+
+	if (panel_mod->timings) {
+		display_timings_release(panel_mod->timings);
+		kfree(panel_mod->timings);
+	}
+
+	tilcdc_module_cleanup(mod);
+	kfree(panel_mod->info);
+	kfree(panel_mod);
+}
+
+static const struct tilcdc_module_ops panel_module_ops = {
+		.modeset_init = panel_modeset_init,
+		.destroy = panel_destroy,
+};
+
+/*
+ * Device:
+ */
+
+/* maybe move this somewhere common if it is needed by other outputs? */
+static struct tilcdc_panel_info * of_get_panel_info(struct device_node *np)
+{
+	struct device_node *info_np;
+	struct tilcdc_panel_info *info;
+	int ret = 0;
+
+	if (!np) {
+		pr_err("%s: no devicenode given\n", __func__);
+		return NULL;
+	}
+
+	info_np = of_get_child_by_name(np, "panel-info");
+	if (!info_np) {
+		pr_err("%s: could not find panel-info node\n", __func__);
+		return NULL;
+	}
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info) {
+		pr_err("%s: allocation failed\n", __func__);
+		return NULL;
+	}
+
+	ret |= of_property_read_u32(info_np, "ac-bias", &info->ac_bias);
+	ret |= of_property_read_u32(info_np, "ac-bias-intrpt", &info->ac_bias_intrpt);
+	ret |= of_property_read_u32(info_np, "dma-burst-sz", &info->dma_burst_sz);
+	ret |= of_property_read_u32(info_np, "bpp", &info->bpp);
+	ret |= of_property_read_u32(info_np, "fdd", &info->fdd);
+	ret |= of_property_read_u32(info_np, "sync-edge", &info->sync_edge);
+	ret |= of_property_read_u32(info_np, "sync-ctrl", &info->sync_ctrl);
+	ret |= of_property_read_u32(info_np, "raster-order", &info->raster_order);
+	ret |= of_property_read_u32(info_np, "fifo-th", &info->fifo_th);
+
+	/* optional: */
+	info->tft_alt_mode      = of_property_read_bool(info_np, "tft-alt-mode");
+	info->invert_pxl_clk    = of_property_read_bool(info_np, "invert-pxl-clk");
+
+	if (ret) {
+		pr_err("%s: error reading panel-info properties\n", __func__);
+		kfree(info);
+		return NULL;
+	}
+
+	return info;
+}
+
+static struct of_device_id panel_of_match[];
+
+static int panel_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct panel_module *panel_mod;
+	struct tilcdc_module *mod;
+	struct pinctrl *pinctrl;
+	int ret = -EINVAL;
+
+
+	/* bail out early if no DT data: */
+	if (!node) {
+		dev_err(&pdev->dev, "device-tree data is missing\n");
+		return -ENXIO;
+	}
+
+	panel_mod = kzalloc(sizeof(*panel_mod), GFP_KERNEL);
+	if (!panel_mod)
+		return -ENOMEM;
+
+	mod = &panel_mod->base;
+
+	tilcdc_module_init(mod, "panel", &panel_module_ops);
+
+	pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
+	if (IS_ERR(pinctrl))
+		dev_warn(&pdev->dev, "pins are not configured\n");
+
+
+	panel_mod->timings = of_get_display_timings(node);
+	if (!panel_mod->timings) {
+		dev_err(&pdev->dev, "could not get panel timings\n");
+		goto fail;
+	}
+
+	panel_mod->info = of_get_panel_info(node);
+	if (!panel_mod->info) {
+		dev_err(&pdev->dev, "could not get panel info\n");
+		goto fail;
+	}
+
+	panel_mod->backlight = of_find_backlight_by_node(node);
+	if (panel_mod->backlight)
+		dev_info(&pdev->dev, "found backlight\n");
+
+	return 0;
+
+fail:
+	panel_destroy(mod);
+	return ret;
+}
+
+static int panel_remove(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static struct of_device_id panel_of_match[] = {
+		{ .compatible = "ti,tilcdc,panel", },
+		{ },
+};
+MODULE_DEVICE_TABLE(of, panel_of_match);
+
+struct platform_driver panel_driver = {
+	.probe = panel_probe,
+	.remove = panel_remove,
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = "panel",
+		.of_match_table = panel_of_match,
+	},
+};
+
+int __init tilcdc_panel_init(void)
+{
+	return platform_driver_register(&panel_driver);
+}
+
+void __exit tilcdc_panel_fini(void)
+{
+	platform_driver_unregister(&panel_driver);
+}
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_panel.h b/drivers/gpu/drm/tilcdc/tilcdc_panel.h
new file mode 100644
index 000000000000..7db40aacc74a
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_panel.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __TILCDC_PANEL_H__
+#define __TILCDC_PANEL_H__
+
+/* sub-module for generic lcd panel output */
+
+int tilcdc_panel_init(void);
+void tilcdc_panel_fini(void);
+
+#endif /* __TILCDC_PANEL_H__ */
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_regs.h b/drivers/gpu/drm/tilcdc/tilcdc_regs.h
new file mode 100644
index 000000000000..17fd1b45428a
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_regs.h
@@ -0,0 +1,154 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __TILCDC_REGS_H__
+#define __TILCDC_REGS_H__
+
+/* LCDC register definitions, based on da8xx-fb */
+
+#include <linux/bitops.h>
+
+#include "tilcdc_drv.h"
+
+/* LCDC Status Register */
+#define LCDC_END_OF_FRAME1                       BIT(9)
+#define LCDC_END_OF_FRAME0                       BIT(8)
+#define LCDC_PL_LOAD_DONE                        BIT(6)
+#define LCDC_FIFO_UNDERFLOW                      BIT(5)
+#define LCDC_SYNC_LOST                           BIT(2)
+#define LCDC_FRAME_DONE                          BIT(0)
+
+/* LCDC DMA Control Register */
+#define LCDC_DMA_BURST_SIZE(x)                   ((x) << 4)
+#define LCDC_DMA_BURST_1                         0x0
+#define LCDC_DMA_BURST_2                         0x1
+#define LCDC_DMA_BURST_4                         0x2
+#define LCDC_DMA_BURST_8                         0x3
+#define LCDC_DMA_BURST_16                        0x4
+#define LCDC_V1_END_OF_FRAME_INT_ENA             BIT(2)
+#define LCDC_V2_END_OF_FRAME0_INT_ENA            BIT(8)
+#define LCDC_V2_END_OF_FRAME1_INT_ENA            BIT(9)
+#define LCDC_DUAL_FRAME_BUFFER_ENABLE            BIT(0)
+
+/* LCDC Control Register */
+#define LCDC_CLK_DIVISOR(x)                      ((x) << 8)
+#define LCDC_RASTER_MODE                         0x01
+
+/* LCDC Raster Control Register */
+#define LCDC_PALETTE_LOAD_MODE(x)                ((x) << 20)
+#define PALETTE_AND_DATA                         0x00
+#define PALETTE_ONLY                             0x01
+#define DATA_ONLY                                0x02
+
+#define LCDC_MONO_8BIT_MODE                      BIT(9)
+#define LCDC_RASTER_ORDER                        BIT(8)
+#define LCDC_TFT_MODE                            BIT(7)
+#define LCDC_V1_UNDERFLOW_INT_ENA                BIT(6)
+#define LCDC_V2_UNDERFLOW_INT_ENA                BIT(5)
+#define LCDC_V1_PL_INT_ENA                       BIT(4)
+#define LCDC_V2_PL_INT_ENA                       BIT(6)
+#define LCDC_MONOCHROME_MODE                     BIT(1)
+#define LCDC_RASTER_ENABLE                       BIT(0)
+#define LCDC_TFT_ALT_ENABLE                      BIT(23)
+#define LCDC_STN_565_ENABLE                      BIT(24)
+#define LCDC_V2_DMA_CLK_EN                       BIT(2)
+#define LCDC_V2_LIDD_CLK_EN                      BIT(1)
+#define LCDC_V2_CORE_CLK_EN                      BIT(0)
+#define LCDC_V2_LPP_B10                          26
+#define LCDC_V2_TFT_24BPP_MODE                   BIT(25)
+#define LCDC_V2_TFT_24BPP_UNPACK                 BIT(26)
+
+/* LCDC Raster Timing 2 Register */
+#define LCDC_AC_BIAS_TRANSITIONS_PER_INT(x)      ((x) << 16)
+#define LCDC_AC_BIAS_FREQUENCY(x)                ((x) << 8)
+#define LCDC_SYNC_CTRL                           BIT(25)
+#define LCDC_SYNC_EDGE                           BIT(24)
+#define LCDC_INVERT_PIXEL_CLOCK                  BIT(22)
+#define LCDC_INVERT_HSYNC                        BIT(21)
+#define LCDC_INVERT_VSYNC                        BIT(20)
+
+/* LCDC Block */
+#define LCDC_PID_REG                             0x0
+#define LCDC_CTRL_REG                            0x4
+#define LCDC_STAT_REG                            0x8
+#define LCDC_RASTER_CTRL_REG                     0x28
+#define LCDC_RASTER_TIMING_0_REG                 0x2c
+#define LCDC_RASTER_TIMING_1_REG                 0x30
+#define LCDC_RASTER_TIMING_2_REG                 0x34
+#define LCDC_DMA_CTRL_REG                        0x40
+#define LCDC_DMA_FB_BASE_ADDR_0_REG              0x44
+#define LCDC_DMA_FB_CEILING_ADDR_0_REG           0x48
+#define LCDC_DMA_FB_BASE_ADDR_1_REG              0x4c
+#define LCDC_DMA_FB_CEILING_ADDR_1_REG           0x50
+
+/* Interrupt Registers available only in Version 2 */
+#define LCDC_RAW_STAT_REG                        0x58
+#define LCDC_MASKED_STAT_REG                     0x5c
+#define LCDC_INT_ENABLE_SET_REG                  0x60
+#define LCDC_INT_ENABLE_CLR_REG                  0x64
+#define LCDC_END_OF_INT_IND_REG                  0x68
+
+/* Clock registers available only on Version 2 */
+#define LCDC_CLK_ENABLE_REG                      0x6c
+#define LCDC_CLK_RESET_REG                       0x70
+#define LCDC_CLK_MAIN_RESET                      BIT(3)
+
+
+/*
+ * Helpers:
+ */
+
+static inline void tilcdc_write(struct drm_device *dev, u32 reg, u32 data)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	iowrite32(data, priv->mmio + reg);
+}
+
+static inline u32 tilcdc_read(struct drm_device *dev, u32 reg)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	return ioread32(priv->mmio + reg);
+}
+
+static inline void tilcdc_set(struct drm_device *dev, u32 reg, u32 mask)
+{
+	tilcdc_write(dev, reg, tilcdc_read(dev, reg) | mask);
+}
+
+static inline void tilcdc_clear(struct drm_device *dev, u32 reg, u32 mask)
+{
+	tilcdc_write(dev, reg, tilcdc_read(dev, reg) & ~mask);
+}
+
+/* the register to read/clear irqstatus differs between v1 and v2 of the IP */
+static inline u32 tilcdc_irqstatus_reg(struct drm_device *dev)
+{
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	return (priv->rev == 2) ? LCDC_MASKED_STAT_REG : LCDC_STAT_REG;
+}
+
+static inline u32 tilcdc_read_irqstatus(struct drm_device *dev)
+{
+	return tilcdc_read(dev, tilcdc_irqstatus_reg(dev));
+}
+
+static inline void tilcdc_clear_irqstatus(struct drm_device *dev, u32 mask)
+{
+	tilcdc_write(dev, tilcdc_irqstatus_reg(dev), mask);
+}
+
+#endif /* __TILCDC_REGS_H__ */
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_slave.c b/drivers/gpu/drm/tilcdc/tilcdc_slave.c
new file mode 100644
index 000000000000..568dc1c08e6c
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_slave.c
@@ -0,0 +1,376 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/i2c.h>
+#include <linux/of_i2c.h>
+#include <linux/pinctrl/pinmux.h>
+#include <linux/pinctrl/consumer.h>
+#include <drm/drm_encoder_slave.h>
+
+#include "tilcdc_drv.h"
+
+struct slave_module {
+	struct tilcdc_module base;
+	struct i2c_adapter *i2c;
+};
+#define to_slave_module(x) container_of(x, struct slave_module, base)
+
+static const struct tilcdc_panel_info slave_info = {
+		.bpp                    = 16,
+		.ac_bias                = 255,
+		.ac_bias_intrpt         = 0,
+		.dma_burst_sz           = 16,
+		.fdd                    = 0x80,
+		.tft_alt_mode           = 0,
+		.sync_edge              = 0,
+		.sync_ctrl              = 1,
+		.raster_order           = 0,
+};
+
+
+/*
+ * Encoder:
+ */
+
+struct slave_encoder {
+	struct drm_encoder_slave base;
+	struct slave_module *mod;
+};
+#define to_slave_encoder(x) container_of(to_encoder_slave(x), struct slave_encoder, base)
+
+static inline struct drm_encoder_slave_funcs *
+get_slave_funcs(struct drm_encoder *enc)
+{
+	return to_encoder_slave(enc)->slave_funcs;
+}
+
+static void slave_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct slave_encoder *slave_encoder = to_slave_encoder(encoder);
+	if (get_slave_funcs(encoder))
+		get_slave_funcs(encoder)->destroy(encoder);
+	drm_encoder_cleanup(encoder);
+	kfree(slave_encoder);
+}
+
+static void slave_encoder_prepare(struct drm_encoder *encoder)
+{
+	drm_i2c_encoder_prepare(encoder);
+	tilcdc_crtc_set_panel_info(encoder->crtc, &slave_info);
+}
+
+static const struct drm_encoder_funcs slave_encoder_funcs = {
+		.destroy        = slave_encoder_destroy,
+};
+
+static const struct drm_encoder_helper_funcs slave_encoder_helper_funcs = {
+		.dpms           = drm_i2c_encoder_dpms,
+		.mode_fixup     = drm_i2c_encoder_mode_fixup,
+		.prepare        = slave_encoder_prepare,
+		.commit         = drm_i2c_encoder_commit,
+		.mode_set       = drm_i2c_encoder_mode_set,
+		.save           = drm_i2c_encoder_save,
+		.restore        = drm_i2c_encoder_restore,
+};
+
+static const struct i2c_board_info info = {
+		I2C_BOARD_INFO("tda998x", 0x70)
+};
+
+static struct drm_encoder *slave_encoder_create(struct drm_device *dev,
+		struct slave_module *mod)
+{
+	struct slave_encoder *slave_encoder;
+	struct drm_encoder *encoder;
+	int ret;
+
+	slave_encoder = kzalloc(sizeof(*slave_encoder), GFP_KERNEL);
+	if (!slave_encoder) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	slave_encoder->mod = mod;
+
+	encoder = &slave_encoder->base.base;
+	encoder->possible_crtcs = 1;
+
+	ret = drm_encoder_init(dev, encoder, &slave_encoder_funcs,
+			DRM_MODE_ENCODER_TMDS);
+	if (ret)
+		goto fail;
+
+	drm_encoder_helper_add(encoder, &slave_encoder_helper_funcs);
+
+	ret = drm_i2c_encoder_init(dev, to_encoder_slave(encoder), mod->i2c, &info);
+	if (ret)
+		goto fail;
+
+	return encoder;
+
+fail:
+	slave_encoder_destroy(encoder);
+	return NULL;
+}
+
+/*
+ * Connector:
+ */
+
+struct slave_connector {
+	struct drm_connector base;
+
+	struct drm_encoder *encoder;  /* our connected encoder */
+	struct slave_module *mod;
+};
+#define to_slave_connector(x) container_of(x, struct slave_connector, base)
+
+static void slave_connector_destroy(struct drm_connector *connector)
+{
+	struct slave_connector *slave_connector = to_slave_connector(connector);
+	drm_connector_cleanup(connector);
+	kfree(slave_connector);
+}
+
+static enum drm_connector_status slave_connector_detect(
+		struct drm_connector *connector,
+		bool force)
+{
+	struct drm_encoder *encoder = to_slave_connector(connector)->encoder;
+	return get_slave_funcs(encoder)->detect(encoder, connector);
+}
+
+static int slave_connector_get_modes(struct drm_connector *connector)
+{
+	struct drm_encoder *encoder = to_slave_connector(connector)->encoder;
+	return get_slave_funcs(encoder)->get_modes(encoder, connector);
+}
+
+static int slave_connector_mode_valid(struct drm_connector *connector,
+		  struct drm_display_mode *mode)
+{
+	struct drm_encoder *encoder = to_slave_connector(connector)->encoder;
+	struct tilcdc_drm_private *priv = connector->dev->dev_private;
+	int ret;
+
+	ret = tilcdc_crtc_mode_valid(priv->crtc, mode);
+	if (ret != MODE_OK)
+		return ret;
+
+	return get_slave_funcs(encoder)->mode_valid(encoder, mode);
+}
+
+static struct drm_encoder *slave_connector_best_encoder(
+		struct drm_connector *connector)
+{
+	struct slave_connector *slave_connector = to_slave_connector(connector);
+	return slave_connector->encoder;
+}
+
+static int slave_connector_set_property(struct drm_connector *connector,
+		struct drm_property *property, uint64_t value)
+{
+	struct drm_encoder *encoder = to_slave_connector(connector)->encoder;
+	return get_slave_funcs(encoder)->set_property(encoder,
+			connector, property, value);
+}
+
+static const struct drm_connector_funcs slave_connector_funcs = {
+	.destroy            = slave_connector_destroy,
+	.dpms               = drm_helper_connector_dpms,
+	.detect             = slave_connector_detect,
+	.fill_modes         = drm_helper_probe_single_connector_modes,
+	.set_property       = slave_connector_set_property,
+};
+
+static const struct drm_connector_helper_funcs slave_connector_helper_funcs = {
+	.get_modes          = slave_connector_get_modes,
+	.mode_valid         = slave_connector_mode_valid,
+	.best_encoder       = slave_connector_best_encoder,
+};
+
+static struct drm_connector *slave_connector_create(struct drm_device *dev,
+		struct slave_module *mod, struct drm_encoder *encoder)
+{
+	struct slave_connector *slave_connector;
+	struct drm_connector *connector;
+	int ret;
+
+	slave_connector = kzalloc(sizeof(*slave_connector), GFP_KERNEL);
+	if (!slave_connector) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	slave_connector->encoder = encoder;
+	slave_connector->mod = mod;
+
+	connector = &slave_connector->base;
+
+	drm_connector_init(dev, connector, &slave_connector_funcs,
+			DRM_MODE_CONNECTOR_HDMIA);
+	drm_connector_helper_add(connector, &slave_connector_helper_funcs);
+
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT |
+			DRM_CONNECTOR_POLL_DISCONNECT;
+
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+
+	get_slave_funcs(encoder)->create_resources(encoder, connector);
+
+	ret = drm_mode_connector_attach_encoder(connector, encoder);
+	if (ret)
+		goto fail;
+
+	drm_sysfs_connector_add(connector);
+
+	return connector;
+
+fail:
+	slave_connector_destroy(connector);
+	return NULL;
+}
+
+/*
+ * Module:
+ */
+
+static int slave_modeset_init(struct tilcdc_module *mod, struct drm_device *dev)
+{
+	struct slave_module *slave_mod = to_slave_module(mod);
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+
+	encoder = slave_encoder_create(dev, slave_mod);
+	if (!encoder)
+		return -ENOMEM;
+
+	connector = slave_connector_create(dev, slave_mod, encoder);
+	if (!connector)
+		return -ENOMEM;
+
+	priv->encoders[priv->num_encoders++] = encoder;
+	priv->connectors[priv->num_connectors++] = connector;
+
+	return 0;
+}
+
+static void slave_destroy(struct tilcdc_module *mod)
+{
+	struct slave_module *slave_mod = to_slave_module(mod);
+
+	tilcdc_module_cleanup(mod);
+	kfree(slave_mod);
+}
+
+static const struct tilcdc_module_ops slave_module_ops = {
+		.modeset_init = slave_modeset_init,
+		.destroy = slave_destroy,
+};
+
+/*
+ * Device:
+ */
+
+static struct of_device_id slave_of_match[];
+
+static int slave_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct device_node *i2c_node;
+	struct slave_module *slave_mod;
+	struct tilcdc_module *mod;
+	struct pinctrl *pinctrl;
+	uint32_t i2c_phandle;
+	int ret = -EINVAL;
+
+	/* bail out early if no DT data: */
+	if (!node) {
+		dev_err(&pdev->dev, "device-tree data is missing\n");
+		return -ENXIO;
+	}
+
+	slave_mod = kzalloc(sizeof(*slave_mod), GFP_KERNEL);
+	if (!slave_mod)
+		return -ENOMEM;
+
+	mod = &slave_mod->base;
+
+	tilcdc_module_init(mod, "slave", &slave_module_ops);
+
+	pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
+	if (IS_ERR(pinctrl))
+		dev_warn(&pdev->dev, "pins are not configured\n");
+
+	if (of_property_read_u32(node, "i2c", &i2c_phandle)) {
+		dev_err(&pdev->dev, "could not get i2c bus phandle\n");
+		goto fail;
+	}
+
+	i2c_node = of_find_node_by_phandle(i2c_phandle);
+	if (!i2c_node) {
+		dev_err(&pdev->dev, "could not get i2c bus node\n");
+		goto fail;
+	}
+
+	slave_mod->i2c = of_find_i2c_adapter_by_node(i2c_node);
+	if (!slave_mod->i2c) {
+		dev_err(&pdev->dev, "could not get i2c\n");
+		goto fail;
+	}
+
+	of_node_put(i2c_node);
+
+	return 0;
+
+fail:
+	slave_destroy(mod);
+	return ret;
+}
+
+static int slave_remove(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static struct of_device_id slave_of_match[] = {
+		{ .compatible = "ti,tilcdc,slave", },
+		{ },
+};
+MODULE_DEVICE_TABLE(of, slave_of_match);
+
+struct platform_driver slave_driver = {
+	.probe = slave_probe,
+	.remove = slave_remove,
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = "slave",
+		.of_match_table = slave_of_match,
+	},
+};
+
+int __init tilcdc_slave_init(void)
+{
+	return platform_driver_register(&slave_driver);
+}
+
+void __exit tilcdc_slave_fini(void)
+{
+	platform_driver_unregister(&slave_driver);
+}
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_slave.h b/drivers/gpu/drm/tilcdc/tilcdc_slave.h
new file mode 100644
index 000000000000..2f8504848320
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_slave.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __TILCDC_SLAVE_H__
+#define __TILCDC_SLAVE_H__
+
+/* sub-module for i2c slave encoder output */
+
+int tilcdc_slave_init(void);
+void tilcdc_slave_fini(void);
+
+#endif /* __TILCDC_SLAVE_H__ */
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
new file mode 100644
index 000000000000..58d487ba2414
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
@@ -0,0 +1,419 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/i2c.h>
+#include <linux/of_i2c.h>
+#include <linux/gpio.h>
+#include <linux/of_gpio.h>
+#include <linux/pinctrl/pinmux.h>
+#include <linux/pinctrl/consumer.h>
+
+#include "tilcdc_drv.h"
+
+struct tfp410_module {
+	struct tilcdc_module base;
+	struct i2c_adapter *i2c;
+	int gpio;
+};
+#define to_tfp410_module(x) container_of(x, struct tfp410_module, base)
+
+
+static const struct tilcdc_panel_info dvi_info = {
+		.ac_bias                = 255,
+		.ac_bias_intrpt         = 0,
+		.dma_burst_sz           = 16,
+		.bpp                    = 16,
+		.fdd                    = 0x80,
+		.tft_alt_mode           = 0,
+		.sync_edge              = 0,
+		.sync_ctrl              = 1,
+		.raster_order           = 0,
+};
+
+/*
+ * Encoder:
+ */
+
+struct tfp410_encoder {
+	struct drm_encoder base;
+	struct tfp410_module *mod;
+	int dpms;
+};
+#define to_tfp410_encoder(x) container_of(x, struct tfp410_encoder, base)
+
+
+static void tfp410_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct tfp410_encoder *tfp410_encoder = to_tfp410_encoder(encoder);
+	drm_encoder_cleanup(encoder);
+	kfree(tfp410_encoder);
+}
+
+static void tfp410_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct tfp410_encoder *tfp410_encoder = to_tfp410_encoder(encoder);
+
+	if (tfp410_encoder->dpms == mode)
+		return;
+
+	if (mode == DRM_MODE_DPMS_ON) {
+		DBG("Power on");
+		gpio_direction_output(tfp410_encoder->mod->gpio, 1);
+	} else {
+		DBG("Power off");
+		gpio_direction_output(tfp410_encoder->mod->gpio, 0);
+	}
+
+	tfp410_encoder->dpms = mode;
+}
+
+static bool tfp410_encoder_mode_fixup(struct drm_encoder *encoder,
+		const struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	/* nothing needed */
+	return true;
+}
+
+static void tfp410_encoder_prepare(struct drm_encoder *encoder)
+{
+	tfp410_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
+	tilcdc_crtc_set_panel_info(encoder->crtc, &dvi_info);
+}
+
+static void tfp410_encoder_commit(struct drm_encoder *encoder)
+{
+	tfp410_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
+}
+
+static void tfp410_encoder_mode_set(struct drm_encoder *encoder,
+		struct drm_display_mode *mode,
+		struct drm_display_mode *adjusted_mode)
+{
+	/* nothing needed */
+}
+
+static const struct drm_encoder_funcs tfp410_encoder_funcs = {
+		.destroy        = tfp410_encoder_destroy,
+};
+
+static const struct drm_encoder_helper_funcs tfp410_encoder_helper_funcs = {
+		.dpms           = tfp410_encoder_dpms,
+		.mode_fixup     = tfp410_encoder_mode_fixup,
+		.prepare        = tfp410_encoder_prepare,
+		.commit         = tfp410_encoder_commit,
+		.mode_set       = tfp410_encoder_mode_set,
+};
+
+static struct drm_encoder *tfp410_encoder_create(struct drm_device *dev,
+		struct tfp410_module *mod)
+{
+	struct tfp410_encoder *tfp410_encoder;
+	struct drm_encoder *encoder;
+	int ret;
+
+	tfp410_encoder = kzalloc(sizeof(*tfp410_encoder), GFP_KERNEL);
+	if (!tfp410_encoder) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	tfp410_encoder->dpms = DRM_MODE_DPMS_OFF;
+	tfp410_encoder->mod = mod;
+
+	encoder = &tfp410_encoder->base;
+	encoder->possible_crtcs = 1;
+
+	ret = drm_encoder_init(dev, encoder, &tfp410_encoder_funcs,
+			DRM_MODE_ENCODER_TMDS);
+	if (ret < 0)
+		goto fail;
+
+	drm_encoder_helper_add(encoder, &tfp410_encoder_helper_funcs);
+
+	return encoder;
+
+fail:
+	tfp410_encoder_destroy(encoder);
+	return NULL;
+}
+
+/*
+ * Connector:
+ */
+
+struct tfp410_connector {
+	struct drm_connector base;
+
+	struct drm_encoder *encoder;  /* our connected encoder */
+	struct tfp410_module *mod;
+};
+#define to_tfp410_connector(x) container_of(x, struct tfp410_connector, base)
+
+
+static void tfp410_connector_destroy(struct drm_connector *connector)
+{
+	struct tfp410_connector *tfp410_connector = to_tfp410_connector(connector);
+	drm_connector_cleanup(connector);
+	kfree(tfp410_connector);
+}
+
+static enum drm_connector_status tfp410_connector_detect(
+		struct drm_connector *connector,
+		bool force)
+{
+	struct tfp410_connector *tfp410_connector = to_tfp410_connector(connector);
+
+	if (drm_probe_ddc(tfp410_connector->mod->i2c))
+		return connector_status_connected;
+
+	return connector_status_unknown;
+}
+
+static int tfp410_connector_get_modes(struct drm_connector *connector)
+{
+	struct tfp410_connector *tfp410_connector = to_tfp410_connector(connector);
+	struct edid *edid;
+	int ret = 0;
+
+	edid = drm_get_edid(connector, tfp410_connector->mod->i2c);
+
+	drm_mode_connector_update_edid_property(connector, edid);
+
+	if (edid) {
+		ret = drm_add_edid_modes(connector, edid);
+		kfree(edid);
+	}
+
+	return ret;
+}
+
+static int tfp410_connector_mode_valid(struct drm_connector *connector,
+		  struct drm_display_mode *mode)
+{
+	struct tilcdc_drm_private *priv = connector->dev->dev_private;
+	/* our only constraints are what the crtc can generate: */
+	return tilcdc_crtc_mode_valid(priv->crtc, mode);
+}
+
+static struct drm_encoder *tfp410_connector_best_encoder(
+		struct drm_connector *connector)
+{
+	struct tfp410_connector *tfp410_connector = to_tfp410_connector(connector);
+	return tfp410_connector->encoder;
+}
+
+static const struct drm_connector_funcs tfp410_connector_funcs = {
+	.destroy            = tfp410_connector_destroy,
+	.dpms               = drm_helper_connector_dpms,
+	.detect             = tfp410_connector_detect,
+	.fill_modes         = drm_helper_probe_single_connector_modes,
+};
+
+static const struct drm_connector_helper_funcs tfp410_connector_helper_funcs = {
+	.get_modes          = tfp410_connector_get_modes,
+	.mode_valid         = tfp410_connector_mode_valid,
+	.best_encoder       = tfp410_connector_best_encoder,
+};
+
+static struct drm_connector *tfp410_connector_create(struct drm_device *dev,
+		struct tfp410_module *mod, struct drm_encoder *encoder)
+{
+	struct tfp410_connector *tfp410_connector;
+	struct drm_connector *connector;
+	int ret;
+
+	tfp410_connector = kzalloc(sizeof(*tfp410_connector), GFP_KERNEL);
+	if (!tfp410_connector) {
+		dev_err(dev->dev, "allocation failed\n");
+		return NULL;
+	}
+
+	tfp410_connector->encoder = encoder;
+	tfp410_connector->mod = mod;
+
+	connector = &tfp410_connector->base;
+
+	drm_connector_init(dev, connector, &tfp410_connector_funcs,
+			DRM_MODE_CONNECTOR_DVID);
+	drm_connector_helper_add(connector, &tfp410_connector_helper_funcs);
+
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT |
+			DRM_CONNECTOR_POLL_DISCONNECT;
+
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+
+	ret = drm_mode_connector_attach_encoder(connector, encoder);
+	if (ret)
+		goto fail;
+
+	drm_sysfs_connector_add(connector);
+
+	return connector;
+
+fail:
+	tfp410_connector_destroy(connector);
+	return NULL;
+}
+
+/*
+ * Module:
+ */
+
+static int tfp410_modeset_init(struct tilcdc_module *mod, struct drm_device *dev)
+{
+	struct tfp410_module *tfp410_mod = to_tfp410_module(mod);
+	struct tilcdc_drm_private *priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+
+	encoder = tfp410_encoder_create(dev, tfp410_mod);
+	if (!encoder)
+		return -ENOMEM;
+
+	connector = tfp410_connector_create(dev, tfp410_mod, encoder);
+	if (!connector)
+		return -ENOMEM;
+
+	priv->encoders[priv->num_encoders++] = encoder;
+	priv->connectors[priv->num_connectors++] = connector;
+
+	return 0;
+}
+
+static void tfp410_destroy(struct tilcdc_module *mod)
+{
+	struct tfp410_module *tfp410_mod = to_tfp410_module(mod);
+
+	if (tfp410_mod->i2c)
+		i2c_put_adapter(tfp410_mod->i2c);
+
+	if (!IS_ERR_VALUE(tfp410_mod->gpio))
+		gpio_free(tfp410_mod->gpio);
+
+	tilcdc_module_cleanup(mod);
+	kfree(tfp410_mod);
+}
+
+static const struct tilcdc_module_ops tfp410_module_ops = {
+		.modeset_init = tfp410_modeset_init,
+		.destroy = tfp410_destroy,
+};
+
+/*
+ * Device:
+ */
+
+static struct of_device_id tfp410_of_match[];
+
+static int tfp410_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct device_node *i2c_node;
+	struct tfp410_module *tfp410_mod;
+	struct tilcdc_module *mod;
+	struct pinctrl *pinctrl;
+	uint32_t i2c_phandle;
+	int ret = -EINVAL;
+
+	/* bail out early if no DT data: */
+	if (!node) {
+		dev_err(&pdev->dev, "device-tree data is missing\n");
+		return -ENXIO;
+	}
+
+	tfp410_mod = kzalloc(sizeof(*tfp410_mod), GFP_KERNEL);
+	if (!tfp410_mod)
+		return -ENOMEM;
+
+	mod = &tfp410_mod->base;
+
+	tilcdc_module_init(mod, "tfp410", &tfp410_module_ops);
+
+	pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
+	if (IS_ERR(pinctrl))
+		dev_warn(&pdev->dev, "pins are not configured\n");
+
+	if (of_property_read_u32(node, "i2c", &i2c_phandle)) {
+		dev_err(&pdev->dev, "could not get i2c bus phandle\n");
+		goto fail;
+	}
+
+	i2c_node = of_find_node_by_phandle(i2c_phandle);
+	if (!i2c_node) {
+		dev_err(&pdev->dev, "could not get i2c bus node\n");
+		goto fail;
+	}
+
+	tfp410_mod->i2c = of_find_i2c_adapter_by_node(i2c_node);
+	if (!tfp410_mod->i2c) {
+		dev_err(&pdev->dev, "could not get i2c\n");
+		goto fail;
+	}
+
+	of_node_put(i2c_node);
+
+	tfp410_mod->gpio = of_get_named_gpio_flags(node, "powerdn-gpio",
+			0, NULL);
+	if (IS_ERR_VALUE(tfp410_mod->gpio)) {
+		dev_warn(&pdev->dev, "No power down GPIO\n");
+	} else {
+		ret = gpio_request(tfp410_mod->gpio, "DVI_PDn");
+		if (ret) {
+			dev_err(&pdev->dev, "could not get DVI_PDn gpio\n");
+			goto fail;
+		}
+	}
+
+	return 0;
+
+fail:
+	tfp410_destroy(mod);
+	return ret;
+}
+
+static int tfp410_remove(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static struct of_device_id tfp410_of_match[] = {
+		{ .compatible = "ti,tilcdc,tfp410", },
+		{ },
+};
+MODULE_DEVICE_TABLE(of, tfp410_of_match);
+
+struct platform_driver tfp410_driver = {
+	.probe = tfp410_probe,
+	.remove = tfp410_remove,
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = "tfp410",
+		.of_match_table = tfp410_of_match,
+	},
+};
+
+int __init tilcdc_tfp410_init(void)
+{
+	return platform_driver_register(&tfp410_driver);
+}
+
+void __exit tilcdc_tfp410_fini(void)
+{
+	platform_driver_unregister(&tfp410_driver);
+}
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.h b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.h
new file mode 100644
index 000000000000..5b800f1f6aa5
--- /dev/null
+++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2012 Texas Instruments
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __TILCDC_TFP410_H__
+#define __TILCDC_TFP410_H__
+
+/* sub-module for tfp410 dvi adaptor */
+
+int tilcdc_tfp410_init(void);
+void tilcdc_tfp410_fini(void);
+
+#endif /* __TILCDC_TFP410_H__ */
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 52b20b12c83a..9b07b7d44a58 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -158,7 +158,8 @@ static void ttm_bo_release_list(struct kref *list_kref)
 	ttm_mem_global_free(bdev->glob->mem_glob, acc_size);
 }
 
-int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo, bool interruptible)
+static int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo,
+				  bool interruptible)
 {
 	if (interruptible) {
 		return wait_event_interruptible(bo->event_queue,
@@ -168,7 +169,6 @@ int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo, bool interruptible)
 		return 0;
 	}
 }
-EXPORT_SYMBOL(ttm_bo_wait_unreserved);
 
 void ttm_bo_add_to_lru(struct ttm_buffer_object *bo)
 {
@@ -213,14 +213,13 @@ int ttm_bo_del_from_lru(struct ttm_buffer_object *bo)
 	return put_count;
 }
 
-int ttm_bo_reserve_locked(struct ttm_buffer_object *bo,
+int ttm_bo_reserve_nolru(struct ttm_buffer_object *bo,
 			  bool interruptible,
 			  bool no_wait, bool use_sequence, uint32_t sequence)
 {
-	struct ttm_bo_global *glob = bo->glob;
 	int ret;
 
-	while (unlikely(atomic_read(&bo->reserved) != 0)) {
+	while (unlikely(atomic_xchg(&bo->reserved, 1) != 0)) {
 		/**
 		 * Deadlock avoidance for multi-bo reserving.
 		 */
@@ -241,26 +240,36 @@ int ttm_bo_reserve_locked(struct ttm_buffer_object *bo,
 		if (no_wait)
 			return -EBUSY;
 
-		spin_unlock(&glob->lru_lock);
 		ret = ttm_bo_wait_unreserved(bo, interruptible);
-		spin_lock(&glob->lru_lock);
 
 		if (unlikely(ret))
 			return ret;
 	}
 
-	atomic_set(&bo->reserved, 1);
 	if (use_sequence) {
+		bool wake_up = false;
 		/**
 		 * Wake up waiters that may need to recheck for deadlock,
 		 * if we decreased the sequence number.
 		 */
 		if (unlikely((bo->val_seq - sequence < (1 << 31))
 			     || !bo->seq_valid))
-			wake_up_all(&bo->event_queue);
+			wake_up = true;
 
+		/*
+		 * In the worst case with memory ordering these values can be
+		 * seen in the wrong order. However since we call wake_up_all
+		 * in that case, this will hopefully not pose a problem,
+		 * and the worst case would only cause someone to accidentally
+		 * hit -EAGAIN in ttm_bo_reserve when they see old value of
+		 * val_seq. However this would only happen if seq_valid was
+		 * written before val_seq was, and just means some slightly
+		 * increased cpu usage
+		 */
 		bo->val_seq = sequence;
 		bo->seq_valid = true;
+		if (wake_up)
+			wake_up_all(&bo->event_queue);
 	} else {
 		bo->seq_valid = false;
 	}
@@ -289,17 +298,64 @@ int ttm_bo_reserve(struct ttm_buffer_object *bo,
 	int put_count = 0;
 	int ret;
 
-	spin_lock(&glob->lru_lock);
-	ret = ttm_bo_reserve_locked(bo, interruptible, no_wait, use_sequence,
-				    sequence);
-	if (likely(ret == 0))
+	ret = ttm_bo_reserve_nolru(bo, interruptible, no_wait, use_sequence,
+				   sequence);
+	if (likely(ret == 0)) {
+		spin_lock(&glob->lru_lock);
 		put_count = ttm_bo_del_from_lru(bo);
-	spin_unlock(&glob->lru_lock);
+		spin_unlock(&glob->lru_lock);
+		ttm_bo_list_ref_sub(bo, put_count, true);
+	}
 
-	ttm_bo_list_ref_sub(bo, put_count, true);
+	return ret;
+}
+
+int ttm_bo_reserve_slowpath_nolru(struct ttm_buffer_object *bo,
+				  bool interruptible, uint32_t sequence)
+{
+	bool wake_up = false;
+	int ret;
+
+	while (unlikely(atomic_xchg(&bo->reserved, 1) != 0)) {
+		WARN_ON(bo->seq_valid && sequence == bo->val_seq);
+
+		ret = ttm_bo_wait_unreserved(bo, interruptible);
 
+		if (unlikely(ret))
+			return ret;
+	}
+
+	if ((bo->val_seq - sequence < (1 << 31)) || !bo->seq_valid)
+		wake_up = true;
+
+	/**
+	 * Wake up waiters that may need to recheck for deadlock,
+	 * if we decreased the sequence number.
+	 */
+	bo->val_seq = sequence;
+	bo->seq_valid = true;
+	if (wake_up)
+		wake_up_all(&bo->event_queue);
+
+	return 0;
+}
+
+int ttm_bo_reserve_slowpath(struct ttm_buffer_object *bo,
+			    bool interruptible, uint32_t sequence)
+{
+	struct ttm_bo_global *glob = bo->glob;
+	int put_count, ret;
+
+	ret = ttm_bo_reserve_slowpath_nolru(bo, interruptible, sequence);
+	if (likely(!ret)) {
+		spin_lock(&glob->lru_lock);
+		put_count = ttm_bo_del_from_lru(bo);
+		spin_unlock(&glob->lru_lock);
+		ttm_bo_list_ref_sub(bo, put_count, true);
+	}
 	return ret;
 }
+EXPORT_SYMBOL(ttm_bo_reserve_slowpath);
 
 void ttm_bo_unreserve_locked(struct ttm_buffer_object *bo)
 {
@@ -511,7 +567,7 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 	int ret;
 
 	spin_lock(&glob->lru_lock);
-	ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+	ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 
 	spin_lock(&bdev->fence_lock);
 	(void) ttm_bo_wait(bo, false, false, true);
@@ -604,7 +660,7 @@ static int ttm_bo_cleanup_refs_and_unlock(struct ttm_buffer_object *bo,
 			return ret;
 
 		spin_lock(&glob->lru_lock);
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 
 		/*
 		 * We raced, and lost, someone else holds the reservation now,
@@ -668,7 +724,14 @@ static int ttm_bo_delayed_delete(struct ttm_bo_device *bdev, bool remove_all)
 			kref_get(&nentry->list_kref);
 		}
 
-		ret = ttm_bo_reserve_locked(entry, false, !remove_all, false, 0);
+		ret = ttm_bo_reserve_nolru(entry, false, true, false, 0);
+		if (remove_all && ret) {
+			spin_unlock(&glob->lru_lock);
+			ret = ttm_bo_reserve_nolru(entry, false, false,
+						   false, 0);
+			spin_lock(&glob->lru_lock);
+		}
+
 		if (!ret)
 			ret = ttm_bo_cleanup_refs_and_unlock(entry, false,
 							     !remove_all);
@@ -816,7 +879,7 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
 
 	spin_lock(&glob->lru_lock);
 	list_for_each_entry(bo, &man->lru, lru) {
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 		if (!ret)
 			break;
 	}
@@ -1797,7 +1860,7 @@ static int ttm_bo_swapout(struct ttm_mem_shrink *shrink)
 
 	spin_lock(&glob->lru_lock);
 	list_for_each_entry(bo, &glob->swap_lru, swap) {
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 		if (!ret)
 			break;
 	}
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index cd9e4523dc56..7b90def15674 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -82,22 +82,6 @@ static void ttm_eu_list_ref_sub(struct list_head *list)
 	}
 }
 
-static int ttm_eu_wait_unreserved_locked(struct list_head *list,
-					 struct ttm_buffer_object *bo)
-{
-	struct ttm_bo_global *glob = bo->glob;
-	int ret;
-
-	ttm_eu_del_from_lru_locked(list);
-	spin_unlock(&glob->lru_lock);
-	ret = ttm_bo_wait_unreserved(bo, true);
-	spin_lock(&glob->lru_lock);
-	if (unlikely(ret != 0))
-		ttm_eu_backoff_reservation_locked(list);
-	return ret;
-}
-
-
 void ttm_eu_backoff_reservation(struct list_head *list)
 {
 	struct ttm_validate_buffer *entry;
@@ -145,47 +129,65 @@ int ttm_eu_reserve_buffers(struct list_head *list)
 	entry = list_first_entry(list, struct ttm_validate_buffer, head);
 	glob = entry->bo->glob;
 
-retry:
 	spin_lock(&glob->lru_lock);
 	val_seq = entry->bo->bdev->val_seq++;
 
+retry:
 	list_for_each_entry(entry, list, head) {
 		struct ttm_buffer_object *bo = entry->bo;
 
-retry_this_bo:
-		ret = ttm_bo_reserve_locked(bo, true, true, true, val_seq);
+		/* already slowpath reserved? */
+		if (entry->reserved)
+			continue;
+
+		ret = ttm_bo_reserve_nolru(bo, true, true, true, val_seq);
 		switch (ret) {
 		case 0:
 			break;
 		case -EBUSY:
-			ret = ttm_eu_wait_unreserved_locked(list, bo);
-			if (unlikely(ret != 0)) {
-				spin_unlock(&glob->lru_lock);
-				ttm_eu_list_ref_sub(list);
-				return ret;
-			}
-			goto retry_this_bo;
+			ttm_eu_del_from_lru_locked(list);
+			spin_unlock(&glob->lru_lock);
+			ret = ttm_bo_reserve_nolru(bo, true, false,
+						   true, val_seq);
+			spin_lock(&glob->lru_lock);
+			if (!ret)
+				break;
+
+			if (unlikely(ret != -EAGAIN))
+				goto err;
+
+			/* fallthrough */
 		case -EAGAIN:
 			ttm_eu_backoff_reservation_locked(list);
+
+			/*
+			 * temporarily increase sequence number every retry,
+			 * to prevent us from seeing our old reservation
+			 * sequence when someone else reserved the buffer,
+			 * but hasn't updated the seq_valid/seqno members yet.
+			 */
+			val_seq = entry->bo->bdev->val_seq++;
+
 			spin_unlock(&glob->lru_lock);
 			ttm_eu_list_ref_sub(list);
-			ret = ttm_bo_wait_unreserved(bo, true);
+			ret = ttm_bo_reserve_slowpath_nolru(bo, true, val_seq);
 			if (unlikely(ret != 0))
 				return ret;
+			spin_lock(&glob->lru_lock);
+			entry->reserved = true;
+			if (unlikely(atomic_read(&bo->cpu_writers) > 0)) {
+				ret = -EBUSY;
+				goto err;
+			}
 			goto retry;
 		default:
-			ttm_eu_backoff_reservation_locked(list);
-			spin_unlock(&glob->lru_lock);
-			ttm_eu_list_ref_sub(list);
-			return ret;
+			goto err;
 		}
 
 		entry->reserved = true;
 		if (unlikely(atomic_read(&bo->cpu_writers) > 0)) {
-			ttm_eu_backoff_reservation_locked(list);
-			spin_unlock(&glob->lru_lock);
-			ttm_eu_list_ref_sub(list);
-			return -EBUSY;
+			ret = -EBUSY;
+			goto err;
 		}
 	}
 
@@ -194,6 +196,12 @@ retry_this_bo:
 	ttm_eu_list_ref_sub(list);
 
 	return 0;
+
+err:
+	ttm_eu_backoff_reservation_locked(list);
+	spin_unlock(&glob->lru_lock);
+	ttm_eu_list_ref_sub(list);
+	return ret;
 }
 EXPORT_SYMBOL(ttm_eu_reserve_buffers);
 
diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h
index 87aa5f5d3c88..cc6d90f28c71 100644
--- a/drivers/gpu/drm/udl/udl_drv.h
+++ b/drivers/gpu/drm/udl/udl_drv.h
@@ -75,6 +75,8 @@ struct udl_framebuffer {
 	struct drm_framebuffer base;
 	struct udl_gem_object *obj;
 	bool active_16; /* active on the 16-bit channel */
+	int x1, y1, x2, y2; /* dirty rect */
+	spinlock_t dirty_lock;
 };
 
 #define to_udl_fb(x) container_of(x, struct udl_framebuffer, base)
diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c
index d4ab3beaada0..9f4be3d4a02e 100644
--- a/drivers/gpu/drm/udl/udl_fb.c
+++ b/drivers/gpu/drm/udl/udl_fb.c
@@ -22,9 +22,9 @@
 
 #include <drm/drm_fb_helper.h>
 
-#define DL_DEFIO_WRITE_DELAY    5 /* fb_deferred_io.delay in jiffies */
+#define DL_DEFIO_WRITE_DELAY    (HZ/20) /* fb_deferred_io.delay in jiffies */
 
-static int fb_defio = 1;  /* Optionally enable experimental fb_defio mmap support */
+static int fb_defio = 0;  /* Optionally enable experimental fb_defio mmap support */
 static int fb_bpp = 16;
 
 module_param(fb_bpp, int, S_IWUSR | S_IRUSR | S_IWGRP | S_IRGRP);
@@ -153,6 +153,9 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 	struct urb *urb;
 	int aligned_x;
 	int bpp = (fb->base.bits_per_pixel / 8);
+	int x2, y2;
+	bool store_for_later = false;
+	unsigned long flags;
 
 	if (!fb->active_16)
 		return 0;
@@ -169,8 +172,6 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 		}
 	}
 
-	start_cycles = get_cycles();
-
 	aligned_x = DL_ALIGN_DOWN(x, sizeof(unsigned long));
 	width = DL_ALIGN_UP(width + (x-aligned_x), sizeof(unsigned long));
 	x = aligned_x;
@@ -180,19 +181,53 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 	    (y + height > fb->base.height))
 		return -EINVAL;
 
+	/* if we are in atomic just store the info
+	   can't test inside spin lock */
+	if (in_atomic())
+		store_for_later = true;
+
+	x2 = x + width - 1;
+	y2 = y + height - 1;
+
+	spin_lock_irqsave(&fb->dirty_lock, flags);
+
+	if (fb->y1 < y)
+		y = fb->y1;
+	if (fb->y2 > y2)
+		y2 = fb->y2;
+	if (fb->x1 < x)
+		x = fb->x1;
+	if (fb->x2 > x2)
+		x2 = fb->x2;
+
+	if (store_for_later) {
+		fb->x1 = x;
+		fb->x2 = x2;
+		fb->y1 = y;
+		fb->y2 = y2;
+		spin_unlock_irqrestore(&fb->dirty_lock, flags);
+		return 0;
+	}
+
+	fb->x1 = fb->y1 = INT_MAX;
+	fb->x2 = fb->y2 = 0;
+
+	spin_unlock_irqrestore(&fb->dirty_lock, flags);
+	start_cycles = get_cycles();
+
 	urb = udl_get_urb(dev);
 	if (!urb)
 		return 0;
 	cmd = urb->transfer_buffer;
 
-	for (i = y; i < y + height ; i++) {
+	for (i = y; i <= y2 ; i++) {
 		const int line_offset = fb->base.pitches[0] * i;
 		const int byte_offset = line_offset + (x * bpp);
 		const int dev_byte_offset = (fb->base.width * bpp * i) + (x * bpp);
 		if (udl_render_hline(dev, bpp, &urb,
 				     (char *) fb->obj->vmapping,
 				     &cmd, byte_offset, dev_byte_offset,
-				     width * bpp,
+				     (x2 - x + 1) * bpp,
 				     &bytes_identical, &bytes_sent))
 			goto error;
 	}
@@ -422,7 +457,6 @@ static void udl_user_framebuffer_destroy(struct drm_framebuffer *fb)
 static const struct drm_framebuffer_funcs udlfb_funcs = {
 	.destroy = udl_user_framebuffer_destroy,
 	.dirty = udl_user_framebuffer_dirty,
-	.create_handle = NULL,
 };
 
 
@@ -434,16 +468,18 @@ udl_framebuffer_init(struct drm_device *dev,
 {
 	int ret;
 
+	spin_lock_init(&ufb->dirty_lock);
 	ufb->obj = obj;
-	ret = drm_framebuffer_init(dev, &ufb->base, &udlfb_funcs);
 	drm_helper_mode_fill_fb_struct(&ufb->base, mode_cmd);
+	ret = drm_framebuffer_init(dev, &ufb->base, &udlfb_funcs);
 	return ret;
 }
 
 
-static int udlfb_create(struct udl_fbdev *ufbdev,
+static int udlfb_create(struct drm_fb_helper *helper,
 			struct drm_fb_helper_surface_size *sizes)
 {
+	struct udl_fbdev *ufbdev = (struct udl_fbdev *)helper;
 	struct drm_device *dev = ufbdev->helper.dev;
 	struct fb_info *info;
 	struct device *device = &dev->usbdev->dev;
@@ -521,27 +557,10 @@ out:
 	return ret;
 }
 
-static int udl_fb_find_or_create_single(struct drm_fb_helper *helper,
-					struct drm_fb_helper_surface_size *sizes)
-{
-	struct udl_fbdev *ufbdev = (struct udl_fbdev *)helper;
-	int new_fb = 0;
-	int ret;
-
-	if (!helper->fb) {
-		ret = udlfb_create(ufbdev, sizes);
-		if (ret)
-			return ret;
-
-		new_fb = 1;
-	}
-	return new_fb;
-}
-
 static struct drm_fb_helper_funcs udl_fb_helper_funcs = {
 	.gamma_set = udl_crtc_fb_gamma_set,
 	.gamma_get = udl_crtc_fb_gamma_get,
-	.fb_probe = udl_fb_find_or_create_single,
+	.fb_probe = udlfb_create,
 };
 
 static void udl_fbdev_destroy(struct drm_device *dev,
@@ -556,6 +575,7 @@ static void udl_fbdev_destroy(struct drm_device *dev,
 		framebuffer_release(info);
 	}
 	drm_fb_helper_fini(&ufbdev->helper);
+	drm_framebuffer_unregister_private(&ufbdev->ufb.base);
 	drm_framebuffer_cleanup(&ufbdev->ufb.base);
 	drm_gem_object_unreference_unlocked(&ufbdev->ufb.obj->base);
 }
@@ -583,6 +603,10 @@ int udl_fbdev_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&ufbdev->helper);
+
+	/* disable all the possible outputs/crtcs before entering KMS mode */
+	drm_helper_disable_unused_functions(dev);
+
 	drm_fb_helper_initial_config(&ufbdev->helper, bpp_sel);
 	return 0;
 }
diff --git a/drivers/gpu/drm/udl/udl_transfer.c b/drivers/gpu/drm/udl/udl_transfer.c
index 142fee5f983f..f343db73e095 100644
--- a/drivers/gpu/drm/udl/udl_transfer.c
+++ b/drivers/gpu/drm/udl/udl_transfer.c
@@ -75,15 +75,19 @@ static int udl_trim_hline(const u8 *bback, const u8 **bfront, int *width_bytes)
 }
 #endif
 
-static inline u16 pixel32_to_be16p(const uint8_t *pixel)
+static inline u16 pixel32_to_be16(const uint32_t pixel)
 {
-	uint32_t pix = *(uint32_t *)pixel;
-	u16 retval;
+	return (((pixel >> 3) & 0x001f) |
+		((pixel >> 5) & 0x07e0) |
+		((pixel >> 8) & 0xf800));
+}
 
-	retval =  (((pix >> 3) & 0x001f) |
-		   ((pix >> 5) & 0x07e0) |
-		   ((pix >> 8) & 0xf800));
-	return retval;
+static bool pixel_repeats(const void *pixel, const uint32_t repeat, int bpp)
+{
+	if (bpp == 2)
+		return *(const uint16_t *)pixel == repeat;
+	else
+		return *(const uint32_t *)pixel == repeat;
 }
 
 /*
@@ -152,29 +156,33 @@ static void udl_compress_hline16(
 		prefetch_range((void *) pixel, (cmd_pixel_end - pixel) * bpp);
 
 		while (pixel < cmd_pixel_end) {
-			const u8 * const repeating_pixel = pixel;
-
-			if (bpp == 2)
-				*(uint16_t *)cmd = cpu_to_be16p((uint16_t *)pixel);
-			else if (bpp == 4)
-				*(uint16_t *)cmd = cpu_to_be16(pixel32_to_be16p(pixel));
+			const u8 *const start = pixel;
+			u32 repeating_pixel;
+
+			if (bpp == 2) {
+				repeating_pixel = *(uint16_t *)pixel;
+				*(uint16_t *)cmd = cpu_to_be16(repeating_pixel);
+			} else {
+				repeating_pixel = *(uint32_t *)pixel;
+				*(uint16_t *)cmd = cpu_to_be16(pixel32_to_be16(repeating_pixel));
+			}
 
 			cmd += 2;
 			pixel += bpp;
 
 			if (unlikely((pixel < cmd_pixel_end) &&
-				     (!memcmp(pixel, repeating_pixel, bpp)))) {
+				     (pixel_repeats(pixel, repeating_pixel, bpp)))) {
 				/* go back and fill in raw pixel count */
-				*raw_pixels_count_byte = (((repeating_pixel -
+				*raw_pixels_count_byte = (((start -
 						raw_pixel_start) / bpp) + 1) & 0xFF;
 
-				while ((pixel < cmd_pixel_end)
-				       && (!memcmp(pixel, repeating_pixel, bpp))) {
+				while ((pixel < cmd_pixel_end) &&
+				       (pixel_repeats(pixel, repeating_pixel, bpp))) {
 					pixel += bpp;
 				}
 
 				/* immediately after raw data is repeat byte */
-				*cmd++ = (((pixel - repeating_pixel) / bpp) - 1) & 0xFF;
+				*cmd++ = (((pixel - start) / bpp) - 1) & 0xFF;
 
 				/* Then start another raw pixel span */
 				raw_pixel_start = pixel;
@@ -223,6 +231,8 @@ int udl_render_hline(struct drm_device *dev, int bpp, struct urb **urb_ptr,
 	u8 *cmd = *urb_buf_ptr;
 	u8 *cmd_end = (u8 *) urb->transfer_buffer + urb->transfer_buffer_length;
 
+	BUG_ON(!(bpp == 2 || bpp == 4));
+
 	line_start = (u8 *) (front + byte_offset);
 	next_pixel = line_start;
 	line_end = next_pixel + byte_width;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 161f8b2549aa..07dfd823cc30 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -829,7 +829,7 @@ static void vmw_lastclose(struct drm_device *dev)
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		set.crtc = crtc;
-		ret = crtc->funcs->set_config(&set);
+		ret = drm_mode_set_config_internal(&set);
 		WARN_ON(ret != 0);
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index d9fbbe191071..c509d40c4897 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -131,7 +131,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 	struct vmw_master *vmaster = vmw_master(file_priv->master);
 	struct drm_vmw_rect __user *clips_ptr;
 	struct drm_vmw_rect *clips = NULL;
-	struct drm_mode_object *obj;
+	struct drm_framebuffer *fb;
 	struct vmw_framebuffer *vfb;
 	struct vmw_resource *res;
 	uint32_t num_clips;
@@ -163,19 +163,15 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 		goto out_no_copy;
 	}
 
-	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
-	if (unlikely(ret != 0)) {
-		ret = -ERESTARTSYS;
-		goto out_no_mode_mutex;
-	}
+	drm_modeset_lock_all(dev);
 
-	obj = drm_mode_object_find(dev, arg->fb_id, DRM_MODE_OBJECT_FB);
-	if (!obj) {
+	fb = drm_framebuffer_lookup(dev, arg->fb_id);
+	if (!fb) {
 		DRM_ERROR("Invalid framebuffer id.\n");
 		ret = -EINVAL;
 		goto out_no_fb;
 	}
-	vfb = vmw_framebuffer_to_vfb(obj_to_fb(obj));
+	vfb = vmw_framebuffer_to_vfb(fb);
 
 	ret = ttm_read_lock(&vmaster->lock, true);
 	if (unlikely(ret != 0))
@@ -199,9 +195,9 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 out_no_surface:
 	ttm_read_unlock(&vmaster->lock);
 out_no_ttm_lock:
+	drm_framebuffer_unreference(fb);
 out_no_fb:
-	mutex_unlock(&dev->mode_config.mutex);
-out_no_mode_mutex:
+	drm_modeset_unlock_all(dev);
 out_no_copy:
 	kfree(clips);
 out_clips:
@@ -220,7 +216,7 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 	struct vmw_master *vmaster = vmw_master(file_priv->master);
 	struct drm_vmw_rect __user *clips_ptr;
 	struct drm_vmw_rect *clips = NULL;
-	struct drm_mode_object *obj;
+	struct drm_framebuffer *fb;
 	struct vmw_framebuffer *vfb;
 	uint32_t num_clips;
 	int ret;
@@ -251,24 +247,20 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 		goto out_no_copy;
 	}
 
-	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
-	if (unlikely(ret != 0)) {
-		ret = -ERESTARTSYS;
-		goto out_no_mode_mutex;
-	}
+	drm_modeset_lock_all(dev);
 
-	obj = drm_mode_object_find(dev, arg->fb_id, DRM_MODE_OBJECT_FB);
-	if (!obj) {
+	fb = drm_framebuffer_lookup(dev, arg->fb_id);
+	if (!fb) {
 		DRM_ERROR("Invalid framebuffer id.\n");
 		ret = -EINVAL;
 		goto out_no_fb;
 	}
 
-	vfb = vmw_framebuffer_to_vfb(obj_to_fb(obj));
+	vfb = vmw_framebuffer_to_vfb(fb);
 	if (!vfb->dmabuf) {
 		DRM_ERROR("Framebuffer not dmabuf backed.\n");
 		ret = -EINVAL;
-		goto out_no_fb;
+		goto out_no_ttm_lock;
 	}
 
 	ret = ttm_read_lock(&vmaster->lock, true);
@@ -281,9 +273,9 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 
 	ttm_read_unlock(&vmaster->lock);
 out_no_ttm_lock:
+	drm_framebuffer_unreference(fb);
 out_no_fb:
-	mutex_unlock(&dev->mode_config.mutex);
-out_no_mode_mutex:
+	drm_modeset_unlock_all(dev);
 out_no_copy:
 	kfree(clips);
 out_clips:
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 54743943d8b3..3e3c7ab33ca2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -180,16 +180,29 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	struct vmw_dma_buffer *dmabuf = NULL;
 	int ret;
 
+	/*
+	 * FIXME: Unclear whether there's any global state touched by the
+	 * cursor_set function, especially vmw_cursor_update_position looks
+	 * suspicious. For now take the easy route and reacquire all locks. We
+	 * can do this since the caller in the drm core doesn't check anything
+	 * which is protected by any looks.
+	 */
+	mutex_unlock(&crtc->mutex);
+	drm_modeset_lock_all(dev_priv->dev);
+
 	/* A lot of the code assumes this */
-	if (handle && (width != 64 || height != 64))
-		return -EINVAL;
+	if (handle && (width != 64 || height != 64)) {
+		ret = -EINVAL;
+		goto out;
+	}
 
 	if (handle) {
 		ret = vmw_user_lookup_handle(dev_priv, tfile,
 					     handle, &surface, &dmabuf);
 		if (ret) {
 			DRM_ERROR("failed to find surface or dmabuf: %i\n", ret);
-			return -EINVAL;
+			ret = -EINVAL;
+			goto out;
 		}
 	}
 
@@ -197,7 +210,8 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	if (surface && !surface->snooper.image) {
 		DRM_ERROR("surface not suitable for cursor\n");
 		vmw_surface_unreference(&surface);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 
 	/* takedown old cursor */
@@ -225,14 +239,20 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 					       du->hotspot_x, du->hotspot_y);
 	} else {
 		vmw_cursor_update_position(dev_priv, false, 0, 0);
-		return 0;
+		ret = 0;
+		goto out;
 	}
 
 	vmw_cursor_update_position(dev_priv, true,
 				   du->cursor_x + du->hotspot_x,
 				   du->cursor_y + du->hotspot_y);
 
-	return 0;
+	ret = 0;
+out:
+	drm_modeset_unlock_all(dev_priv->dev);
+	mutex_lock(&crtc->mutex);
+
+	return ret;
 }
 
 int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
@@ -244,10 +264,23 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	du->cursor_x = x + crtc->x;
 	du->cursor_y = y + crtc->y;
 
+	/*
+	 * FIXME: Unclear whether there's any global state touched by the
+	 * cursor_set function, especially vmw_cursor_update_position looks
+	 * suspicious. For now take the easy route and reacquire all locks. We
+	 * can do this since the caller in the drm core doesn't check anything
+	 * which is protected by any looks.
+	 */
+	mutex_unlock(&crtc->mutex);
+	drm_modeset_lock_all(dev_priv->dev);
+
 	vmw_cursor_update_position(dev_priv, shown,
 				   du->cursor_x + du->hotspot_x,
 				   du->cursor_y + du->hotspot_y);
 
+	drm_modeset_unlock_all(dev_priv->dev);
+	mutex_lock(&crtc->mutex);
+
 	return 0;
 }
 
@@ -373,16 +406,6 @@ void vmw_kms_cursor_post_execbuf(struct vmw_private *dev_priv)
  * Generic framebuffer code
  */
 
-int vmw_framebuffer_create_handle(struct drm_framebuffer *fb,
-				  struct drm_file *file_priv,
-				  unsigned int *handle)
-{
-	if (handle)
-		*handle = 0;
-
-	return 0;
-}
-
 /*
  * Surface framebuffer code
  */
@@ -610,7 +633,6 @@ int vmw_framebuffer_surface_dirty(struct drm_framebuffer *framebuffer,
 static struct drm_framebuffer_funcs vmw_framebuffer_surface_funcs = {
 	.destroy = vmw_framebuffer_surface_destroy,
 	.dirty = vmw_framebuffer_surface_dirty,
-	.create_handle = vmw_framebuffer_create_handle,
 };
 
 static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv,
@@ -681,14 +703,10 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv,
 		goto out_err1;
 	}
 
-	ret = drm_framebuffer_init(dev, &vfbs->base.base,
-				   &vmw_framebuffer_surface_funcs);
-	if (ret)
-		goto out_err2;
-
 	if (!vmw_surface_reference(surface)) {
 		DRM_ERROR("failed to reference surface %p\n", surface);
-		goto out_err3;
+		ret = -EINVAL;
+		goto out_err2;
 	}
 
 	/* XXX get the first 3 from the surface info */
@@ -707,10 +725,15 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv,
 
 	*out = &vfbs->base;
 
+	ret = drm_framebuffer_init(dev, &vfbs->base.base,
+				   &vmw_framebuffer_surface_funcs);
+	if (ret)
+		goto out_err3;
+
 	return 0;
 
 out_err3:
-	drm_framebuffer_cleanup(&vfbs->base.base);
+	vmw_surface_unreference(&surface);
 out_err2:
 	kfree(vfbs);
 out_err1:
@@ -960,7 +983,6 @@ int vmw_framebuffer_dmabuf_dirty(struct drm_framebuffer *framebuffer,
 static struct drm_framebuffer_funcs vmw_framebuffer_dmabuf_funcs = {
 	.destroy = vmw_framebuffer_dmabuf_destroy,
 	.dirty = vmw_framebuffer_dmabuf_dirty,
-	.create_handle = vmw_framebuffer_create_handle,
 };
 
 /**
@@ -1053,14 +1075,10 @@ static int vmw_kms_new_framebuffer_dmabuf(struct vmw_private *dev_priv,
 		goto out_err1;
 	}
 
-	ret = drm_framebuffer_init(dev, &vfbd->base.base,
-				   &vmw_framebuffer_dmabuf_funcs);
-	if (ret)
-		goto out_err2;
-
 	if (!vmw_dmabuf_reference(dmabuf)) {
 		DRM_ERROR("failed to reference dmabuf %p\n", dmabuf);
-		goto out_err3;
+		ret = -EINVAL;
+		goto out_err2;
 	}
 
 	vfbd->base.base.bits_per_pixel = mode_cmd->bpp;
@@ -1077,10 +1095,15 @@ static int vmw_kms_new_framebuffer_dmabuf(struct vmw_private *dev_priv,
 	vfbd->base.user_handle = mode_cmd->handle;
 	*out = &vfbd->base;
 
+	ret = drm_framebuffer_init(dev, &vfbd->base.base,
+				   &vmw_framebuffer_dmabuf_funcs);
+	if (ret)
+		goto out_err3;
+
 	return 0;
 
 out_err3:
-	drm_framebuffer_cleanup(&vfbd->base.base);
+	vmw_dmabuf_unreference(&dmabuf);
 out_err2:
 	kfree(vfbd);
 out_err1:
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index e01a17b407b2..16556170fb32 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -959,13 +959,13 @@ void vmw_resource_unreserve(struct vmw_resource *res,
 	if (new_backup && new_backup != res->backup) {
 
 		if (res->backup) {
-			BUG_ON(atomic_read(&res->backup->base.reserved) == 0);
+			BUG_ON(!ttm_bo_is_reserved(&res->backup->base));
 			list_del_init(&res->mob_head);
 			vmw_dmabuf_unreference(&res->backup);
 		}
 
 		res->backup = vmw_dmabuf_reference(new_backup);
-		BUG_ON(atomic_read(&new_backup->base.reserved) == 0);
+		BUG_ON(!ttm_bo_is_reserved(&new_backup->base));
 		list_add_tail(&res->mob_head, &new_backup->res_list);
 	}
 	if (new_backup)
diff --git a/drivers/gpu/stub/Kconfig b/drivers/gpu/stub/Kconfig
deleted file mode 100644
index 419917955bf6..000000000000
--- a/drivers/gpu/stub/Kconfig
+++ /dev/null
@@ -1,18 +0,0 @@
-config STUB_POULSBO
-	tristate "Intel GMA500 Stub Driver"
-	depends on PCI
-	depends on NET # for THERMAL
-	# Poulsbo stub depends on ACPI_VIDEO when ACPI is enabled
-	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
-	select BACKLIGHT_CLASS_DEVICE if ACPI
-	select VIDEO_OUTPUT_CONTROL if ACPI
-	select INPUT if ACPI
-	select ACPI_VIDEO if ACPI
-	select THERMAL if ACPI
-	help
-	  Choose this option if you have a system that has Intel GMA500
-	  (Poulsbo) integrated graphics. If M is selected, the module will
-	  be called Poulsbo. This driver is a stub driver for Poulsbo that
-	  will call poulsbo.ko to enable the acpi backlight control sysfs
-	  entry file because there have no poulsbo native driver can support
-	  intel opregion.
diff --git a/drivers/gpu/stub/Makefile b/drivers/gpu/stub/Makefile
deleted file mode 100644
index cd940cc9d36d..000000000000
--- a/drivers/gpu/stub/Makefile
+++ /dev/null
@@ -1 +0,0 @@
-obj-$(CONFIG_STUB_POULSBO) += poulsbo.o
diff --git a/drivers/gpu/stub/poulsbo.c b/drivers/gpu/stub/poulsbo.c
deleted file mode 100644
index 7edfd27b8dee..000000000000
--- a/drivers/gpu/stub/poulsbo.c
+++ /dev/null
@@ -1,64 +0,0 @@
-/*
- * Intel Poulsbo Stub driver
- *
- * Copyright (C) 2010 Novell <jlee@novell.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- */
-
-#include <linux/module.h>
-#include <linux/pci.h>
-#include <linux/acpi.h>
-#include <acpi/video.h>
-
-#define DRIVER_NAME "poulsbo"
-
-enum {
-	CHIP_PSB_8108 = 0,
-	CHIP_PSB_8109 = 1,
-};
-
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
-	{0x8086, 0x8108, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PSB_8108}, \
-	{0x8086, 0x8109, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PSB_8109}, \
-	{0, 0, 0}
-};
-
-static int poulsbo_probe(struct pci_dev *pdev, const struct pci_device_id *id)
-{
-	return acpi_video_register();
-}
-
-static void poulsbo_remove(struct pci_dev *pdev)
-{
-	acpi_video_unregister();
-}
-
-static struct pci_driver poulsbo_driver = {
-	.name = DRIVER_NAME,
-	.id_table = pciidlist,
-	.probe = poulsbo_probe,
-	.remove = poulsbo_remove,
-};
-
-static int __init poulsbo_init(void)
-{
-	return pci_register_driver(&poulsbo_driver);
-}
-
-static void __exit poulsbo_exit(void)
-{
-	pci_unregister_driver(&poulsbo_driver);
-}
-
-module_init(poulsbo_init);
-module_exit(poulsbo_exit);
-
-MODULE_AUTHOR("Lee, Chun-Yi <jlee@novell.com>");
-MODULE_DESCRIPTION("Poulsbo Stub Driver");
-MODULE_LICENSE("GPL");
-
-MODULE_DEVICE_TABLE(pci, pciidlist);
diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index fa60add0ff63..cf787e1d9322 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -25,6 +25,7 @@
 #include <linux/fb.h>
 
 #include <linux/pci.h>
+#include <linux/console.h>
 #include <linux/vga_switcheroo.h>
 
 #include <linux/vgaarb.h>
@@ -337,8 +338,10 @@ static int vga_switchto_stage2(struct vga_switcheroo_client *new_client)
 
 	if (new_client->fb_info) {
 		struct fb_event event;
+		console_lock();
 		event.info = new_client->fb_info;
 		fb_notifier_call_chain(FB_EVENT_REMAP_ALL_CONSOLE, &event);
+		console_unlock();
 	}
 
 	ret = vgasr_priv.handler->switchto(new_client->id);
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 43d5c8b8e7ad..0099667a397e 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4255,13 +4255,19 @@ static void quirk_iommu_rwbf(struct pci_dev *dev)
 {
 	/*
 	 * Mobile 4 Series Chipset neglects to set RWBF capability,
-	 * but needs it:
+	 * but needs it. Same seems to hold for the desktop versions.
 	 */
 	printk(KERN_INFO "DMAR: Forcing write-buffer flush capability\n");
 	rwbf_quirk = 1;
 }
 
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e00, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e10, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e20, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e30, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e40, quirk_iommu_rwbf);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e90, quirk_iommu_rwbf);
 
 #define GGC 0x52
 #define GGC_MEMORY_SIZE_MASK	(0xf << 8)
diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig
index 3a7965d6ac28..093f10c88cce 100644
--- a/drivers/staging/Kconfig
+++ b/drivers/staging/Kconfig
@@ -112,8 +112,6 @@ source "drivers/staging/media/Kconfig"
 
 source "drivers/staging/net/Kconfig"
 
-source "drivers/staging/omapdrm/Kconfig"
-
 source "drivers/staging/android/Kconfig"
 
 source "drivers/staging/ozwpan/Kconfig"
diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile
index 5971865d0c61..fa41b04cf4cb 100644
--- a/drivers/staging/Makefile
+++ b/drivers/staging/Makefile
@@ -48,7 +48,6 @@ obj-$(CONFIG_SPEAKUP)		+= speakup/
 obj-$(CONFIG_TOUCHSCREEN_CLEARPAD_TM1217)	+= cptm1217/
 obj-$(CONFIG_TOUCHSCREEN_SYNAPTICS_I2C_RMI4)	+= ste_rmi4/
 obj-$(CONFIG_MFD_NVEC)		+= nvec/
-obj-$(CONFIG_DRM_OMAP)		+= omapdrm/
 obj-$(CONFIG_ANDROID)		+= android/
 obj-$(CONFIG_USB_WPAN_HCD)	+= ozwpan/
 obj-$(CONFIG_USB_G_CCG)		+= ccg/
diff --git a/drivers/staging/omapdrm/TODO b/drivers/staging/omapdrm/TODO
deleted file mode 100644
index abeeb00aaa12..000000000000
--- a/drivers/staging/omapdrm/TODO
+++ /dev/null
@@ -1,32 +0,0 @@
-TODO
-. add video decode/encode support (via syslink3 + codec-engine)
-  . NOTE: with dmabuf this probably could be split into different driver
-    so perhaps this TODO doesn't belong here
-. where should we do eviction (detatch_pages())?  We aren't necessarily
-  accessing the pages via a GART, so maybe we need some other threshold
-  to put a cap on the # of pages that can be pin'd.  (It is mostly only
-  of interest in case you have a swap partition/file.. which a lot of
-  these devices do not.. but it doesn't hurt for the driver to do the
-  right thing anyways.)
-  . Use mm_shrinker to trigger unpinning pages.  Need to figure out how
-    to handle next issue first (I think?)
-  . Note TTM already has some mm_shrinker stuff..  maybe an argument to
-    move to TTM?  Or maybe something that could be factored out in common?
-. GEM/shmem backed pages can have existing mappings (kernel linear map,
-  etc..), which isn't really ideal.
-. Revisit GEM sync object infrastructure.. TTM has some framework for this
-  already.  Possibly this could be refactored out and made more common?
-  There should be some way to do this with less wheel-reinvention.
-. Solve PM sequencing on resume.  DMM/TILER must be reloaded before any
-  access is made from any component in the system.  Which means on suspend
-  CRTC's should be disabled, and on resume the LUT should be reprogrammed
-  before CRTC's are re-enabled, to prevent DSS from trying to DMA from a
-  buffer mapped in DMM/TILER before LUT is reloaded.
-
-Userspace:
-. git://github.com/robclark/xf86-video-omap.git
-
-Currently tested on
-. OMAP3530 beagleboard
-. OMAP4430 pandaboard
-. OMAP4460 pandaboard
diff --git a/drivers/staging/omapdrm/omap_drm.h b/drivers/staging/omapdrm/omap_drm.h
deleted file mode 100644
index f0ac34a8973e..000000000000
--- a/drivers/staging/omapdrm/omap_drm.h
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * include/drm/omap_drm.h
- *
- * Copyright (C) 2011 Texas Instruments
- * Author: Rob Clark <rob@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef __OMAP_DRM_H__
-#define __OMAP_DRM_H__
-
-#include <drm/drm.h>
-
-/* Please note that modifications to all structs defined here are
- * subject to backwards-compatibility constraints.
- */
-
-#define OMAP_PARAM_CHIPSET_ID	1	/* ie. 0x3430, 0x4430, etc */
-
-struct drm_omap_param {
-	uint64_t param;			/* in */
-	uint64_t value;			/* in (set_param), out (get_param) */
-};
-
-#define OMAP_BO_SCANOUT		0x00000001	/* scanout capable (phys contiguous) */
-#define OMAP_BO_CACHE_MASK	0x00000006	/* cache type mask, see cache modes */
-#define OMAP_BO_TILED_MASK	0x00000f00	/* tiled mapping mask, see tiled modes */
-
-/* cache modes */
-#define OMAP_BO_CACHED		0x00000000	/* default */
-#define OMAP_BO_WC		0x00000002	/* write-combine */
-#define OMAP_BO_UNCACHED	0x00000004	/* strongly-ordered (uncached) */
-
-/* tiled modes */
-#define OMAP_BO_TILED_8		0x00000100
-#define OMAP_BO_TILED_16	0x00000200
-#define OMAP_BO_TILED_32	0x00000300
-#define OMAP_BO_TILED		(OMAP_BO_TILED_8 | OMAP_BO_TILED_16 | OMAP_BO_TILED_32)
-
-union omap_gem_size {
-	uint32_t bytes;		/* (for non-tiled formats) */
-	struct {
-		uint16_t width;
-		uint16_t height;
-	} tiled;		/* (for tiled formats) */
-};
-
-struct drm_omap_gem_new {
-	union omap_gem_size size;	/* in */
-	uint32_t flags;			/* in */
-	uint32_t handle;		/* out */
-	uint32_t __pad;
-};
-
-/* mask of operations: */
-enum omap_gem_op {
-	OMAP_GEM_READ = 0x01,
-	OMAP_GEM_WRITE = 0x02,
-};
-
-struct drm_omap_gem_cpu_prep {
-	uint32_t handle;		/* buffer handle (in) */
-	uint32_t op;			/* mask of omap_gem_op (in) */
-};
-
-struct drm_omap_gem_cpu_fini {
-	uint32_t handle;		/* buffer handle (in) */
-	uint32_t op;			/* mask of omap_gem_op (in) */
-	/* TODO maybe here we pass down info about what regions are touched
-	 * by sw so we can be clever about cache ops?  For now a placeholder,
-	 * set to zero and we just do full buffer flush..
-	 */
-	uint32_t nregions;
-	uint32_t __pad;
-};
-
-struct drm_omap_gem_info {
-	uint32_t handle;		/* buffer handle (in) */
-	uint32_t pad;
-	uint64_t offset;		/* mmap offset (out) */
-	/* note: in case of tiled buffers, the user virtual size can be
-	 * different from the physical size (ie. how many pages are needed
-	 * to back the object) which is returned in DRM_IOCTL_GEM_OPEN..
-	 * This size here is the one that should be used if you want to
-	 * mmap() the buffer:
-	 */
-	uint32_t size;			/* virtual size for mmap'ing (out) */
-	uint32_t __pad;
-};
-
-#define DRM_OMAP_GET_PARAM		0x00
-#define DRM_OMAP_SET_PARAM		0x01
-/* placeholder for plugin-api
-#define DRM_OMAP_GET_BASE		0x02
-*/
-#define DRM_OMAP_GEM_NEW		0x03
-#define DRM_OMAP_GEM_CPU_PREP		0x04
-#define DRM_OMAP_GEM_CPU_FINI		0x05
-#define DRM_OMAP_GEM_INFO		0x06
-#define DRM_OMAP_NUM_IOCTLS		0x07
-
-#define DRM_IOCTL_OMAP_GET_PARAM	DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GET_PARAM, struct drm_omap_param)
-#define DRM_IOCTL_OMAP_SET_PARAM	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_SET_PARAM, struct drm_omap_param)
-/* placeholder for plugin-api
-#define DRM_IOCTL_OMAP_GET_BASE		DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GET_BASE, struct drm_omap_get_base)
-*/
-#define DRM_IOCTL_OMAP_GEM_NEW		DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GEM_NEW, struct drm_omap_gem_new)
-#define DRM_IOCTL_OMAP_GEM_CPU_PREP	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_GEM_CPU_PREP, struct drm_omap_gem_cpu_prep)
-#define DRM_IOCTL_OMAP_GEM_CPU_FINI	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_GEM_CPU_FINI, struct drm_omap_gem_cpu_fini)
-#define DRM_IOCTL_OMAP_GEM_INFO		DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GEM_INFO, struct drm_omap_gem_info)
-
-#endif /* __OMAP_DRM_H__ */
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 6c4abeaf690f..fbd447b390f7 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -638,7 +638,7 @@ static inline void save_screen(struct vc_data *vc)
  *	Redrawing of screen
  */
 
-static void clear_buffer_attributes(struct vc_data *vc)
+void clear_buffer_attributes(struct vc_data *vc)
 {
 	unsigned short *p = (unsigned short *)vc->vc_origin;
 	int count = vc->vc_screenbuf_size / 2;
@@ -2987,7 +2987,7 @@ int __init vty_init(const struct file_operations *console_fops)
 
 static struct class *vtconsole_class;
 
-static int bind_con_driver(const struct consw *csw, int first, int last,
+static int do_bind_con_driver(const struct consw *csw, int first, int last,
 			   int deflt)
 {
 	struct module *owner = csw->owner;
@@ -2998,7 +2998,7 @@ static int bind_con_driver(const struct consw *csw, int first, int last,
 	if (!try_module_get(owner))
 		return -ENODEV;
 
-	console_lock();
+	WARN_CONSOLE_UNLOCKED();
 
 	/* check if driver is registered */
 	for (i = 0; i < MAX_NR_CON_DRIVER; i++) {
@@ -3083,11 +3083,22 @@ static int bind_con_driver(const struct consw *csw, int first, int last,
 
 	retval = 0;
 err:
-	console_unlock();
 	module_put(owner);
 	return retval;
 };
 
+
+static int bind_con_driver(const struct consw *csw, int first, int last,
+			   int deflt)
+{
+	int ret;
+
+	console_lock();
+	ret = do_bind_con_driver(csw, first, last, deflt);
+	console_unlock();
+	return ret;
+}
+
 #ifdef CONFIG_VT_HW_CONSOLE_BINDING
 static int con_is_graphics(const struct consw *csw, int first, int last)
 {
@@ -3124,6 +3135,18 @@ static int con_is_graphics(const struct consw *csw, int first, int last)
  */
 int unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
 {
+	int retval;
+
+	console_lock();
+	retval = do_unbind_con_driver(csw, first, last, deflt);
+	console_unlock();
+	return retval;
+}
+EXPORT_SYMBOL(unbind_con_driver);
+
+/* unlocked version of unbind_con_driver() */
+int do_unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
+{
 	struct module *owner = csw->owner;
 	const struct consw *defcsw = NULL;
 	struct con_driver *con_driver = NULL, *con_back = NULL;
@@ -3132,7 +3155,7 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
 	if (!try_module_get(owner))
 		return -ENODEV;
 
-	console_lock();
+	WARN_CONSOLE_UNLOCKED();
 
 	/* check if driver is registered and if it is unbindable */
 	for (i = 0; i < MAX_NR_CON_DRIVER; i++) {
@@ -3145,10 +3168,8 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
 		}
 	}
 
-	if (retval) {
-		console_unlock();
+	if (retval)
 		goto err;
-	}
 
 	retval = -ENODEV;
 
@@ -3164,15 +3185,11 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
 		}
 	}
 
-	if (retval) {
-		console_unlock();
+	if (retval)
 		goto err;
-	}
 
-	if (!con_is_bound(csw)) {
-		console_unlock();
+	if (!con_is_bound(csw))
 		goto err;
-	}
 
 	first = max(first, con_driver->first);
 	last = min(last, con_driver->last);
@@ -3199,15 +3216,14 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt)
 	if (!con_is_bound(csw))
 		con_driver->flag &= ~CON_DRIVER_FLAG_INIT;
 
-	console_unlock();
 	/* ignore return value, binding should not fail */
-	bind_con_driver(defcsw, first, last, deflt);
+	do_bind_con_driver(defcsw, first, last, deflt);
 err:
 	module_put(owner);
 	return retval;
 
 }
-EXPORT_SYMBOL(unbind_con_driver);
+EXPORT_SYMBOL_GPL(do_unbind_con_driver);
 
 static int vt_bind(struct con_driver *con)
 {
@@ -3492,28 +3508,18 @@ int con_debug_leave(void)
 }
 EXPORT_SYMBOL_GPL(con_debug_leave);
 
-/**
- * register_con_driver - register console driver to console layer
- * @csw: console driver
- * @first: the first console to take over, minimum value is 0
- * @last: the last console to take over, maximum value is MAX_NR_CONSOLES -1
- *
- * DESCRIPTION: This function registers a console driver which can later
- * bind to a range of consoles specified by @first and @last. It will
- * also initialize the console driver by calling con_startup().
- */
-int register_con_driver(const struct consw *csw, int first, int last)
+static int do_register_con_driver(const struct consw *csw, int first, int last)
 {
 	struct module *owner = csw->owner;
 	struct con_driver *con_driver;
 	const char *desc;
 	int i, retval = 0;
 
+	WARN_CONSOLE_UNLOCKED();
+
 	if (!try_module_get(owner))
 		return -ENODEV;
 
-	console_lock();
-
 	for (i = 0; i < MAX_NR_CON_DRIVER; i++) {
 		con_driver = &registered_con_driver[i];
 
@@ -3566,10 +3572,29 @@ int register_con_driver(const struct consw *csw, int first, int last)
 	}
 
 err:
-	console_unlock();
 	module_put(owner);
 	return retval;
 }
+
+/**
+ * register_con_driver - register console driver to console layer
+ * @csw: console driver
+ * @first: the first console to take over, minimum value is 0
+ * @last: the last console to take over, maximum value is MAX_NR_CONSOLES -1
+ *
+ * DESCRIPTION: This function registers a console driver which can later
+ * bind to a range of consoles specified by @first and @last. It will
+ * also initialize the console driver by calling con_startup().
+ */
+int register_con_driver(const struct consw *csw, int first, int last)
+{
+	int retval;
+
+	console_lock();
+	retval = do_register_con_driver(csw, first, last);
+	console_unlock();
+	return retval;
+}
 EXPORT_SYMBOL(register_con_driver);
 
 /**
@@ -3585,9 +3610,18 @@ EXPORT_SYMBOL(register_con_driver);
  */
 int unregister_con_driver(const struct consw *csw)
 {
-	int i, retval = -ENODEV;
+	int retval;
 
 	console_lock();
+	retval = do_unregister_con_driver(csw);
+	console_unlock();
+	return retval;
+}
+EXPORT_SYMBOL(unregister_con_driver);
+
+int do_unregister_con_driver(const struct consw *csw)
+{
+	int i, retval = -ENODEV;
 
 	/* cannot unregister a bound driver */
 	if (con_is_bound(csw))
@@ -3613,27 +3647,53 @@ int unregister_con_driver(const struct consw *csw)
 		}
 	}
 err:
-	console_unlock();
 	return retval;
 }
-EXPORT_SYMBOL(unregister_con_driver);
+EXPORT_SYMBOL_GPL(do_unregister_con_driver);
 
 /*
  *	If we support more console drivers, this function is used
  *	when a driver wants to take over some existing consoles
  *	and become default driver for newly opened ones.
  *
- *      take_over_console is basically a register followed by unbind
+ *	take_over_console is basically a register followed by unbind
+ */
+int do_take_over_console(const struct consw *csw, int first, int last, int deflt)
+{
+	int err;
+
+	err = do_register_con_driver(csw, first, last);
+	/*
+	 * If we get an busy error we still want to bind the console driver
+	 * and return success, as we may have unbound the console driver
+	 * but not unregistered it.
+	 */
+	if (err == -EBUSY)
+		err = 0;
+	if (!err)
+		do_bind_con_driver(csw, first, last, deflt);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(do_take_over_console);
+
+/*
+ *	If we support more console drivers, this function is used
+ *	when a driver wants to take over some existing consoles
+ *	and become default driver for newly opened ones.
+ *
+ *	take_over_console is basically a register followed by unbind
  */
 int take_over_console(const struct consw *csw, int first, int last, int deflt)
 {
 	int err;
 
 	err = register_con_driver(csw, first, last);
-	/* if we get an busy error we still want to bind the console driver
+	/*
+	 * If we get an busy error we still want to bind the console driver
 	 * and return success, as we may have unbound the console driver
-	 * but not unregistered it.
-	*/
+	 * but not unregistered it.
+	 */
 	if (err == -EBUSY)
 		err = 0;
 	if (!err)
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 80cbd21b483f..4c1546f71d56 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -21,8 +21,6 @@ source "drivers/gpu/vga/Kconfig"
 
 source "drivers/gpu/drm/Kconfig"
 
-source "drivers/gpu/stub/Kconfig"
-
 config VGASTATE
        tristate
        default n
@@ -33,6 +31,30 @@ config VIDEO_OUTPUT_CONTROL
 	  This framework adds support for low-level control of the video 
 	  output switch.
 
+config DISPLAY_TIMING
+       bool
+
+config VIDEOMODE
+       bool
+
+config OF_DISPLAY_TIMING
+	bool "Enable device tree display timing support"
+	depends on OF
+	select DISPLAY_TIMING
+	help
+	  helper to parse display timings from the devicetree
+
+config OF_VIDEOMODE
+	bool "Enable device tree videomode support"
+	depends on OF
+	select VIDEOMODE
+	select OF_DISPLAY_TIMING
+	help
+	  helper to get videomodes from the devicetree
+
+config HDMI
+	bool
+
 menuconfig FB
 	tristate "Support for frame buffer devices"
 	---help---
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 0577f834fdcd..9df387334cb7 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -5,6 +5,7 @@
 # Each configuration option enables a list of files.
 
 obj-$(CONFIG_VGASTATE)            += vgastate.o
+obj-$(CONFIG_HDMI)                += hdmi.o
 obj-y                             += fb_notify.o
 obj-$(CONFIG_FB)                  += fb.o
 fb-y                              := fbmem.o fbmon.o fbcmap.o fbsysfs.o \
@@ -170,3 +171,7 @@ obj-$(CONFIG_FB_VIRTUAL)          += vfb.o
 
 #video output switch sysfs driver
 obj-$(CONFIG_VIDEO_OUTPUT_CONTROL) += output.o
+obj-$(CONFIG_DISPLAY_TIMING) += display_timing.o
+obj-$(CONFIG_OF_DISPLAY_TIMING) += of_display_timing.o
+obj-$(CONFIG_VIDEOMODE) += videomode.o
+obj-$(CONFIG_OF_VIDEOMODE) += of_videomode.o
diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c
index f8a61e210d2e..3cd675927826 100644
--- a/drivers/video/console/fbcon.c
+++ b/drivers/video/console/fbcon.c
@@ -529,6 +529,33 @@ static int search_for_mapped_con(void)
 	return retval;
 }
 
+static int do_fbcon_takeover(int show_logo)
+{
+	int err, i;
+
+	if (!num_registered_fb)
+		return -ENODEV;
+
+	if (!show_logo)
+		logo_shown = FBCON_LOGO_DONTSHOW;
+
+	for (i = first_fb_vc; i <= last_fb_vc; i++)
+		con2fb_map[i] = info_idx;
+
+	err = do_take_over_console(&fb_con, first_fb_vc, last_fb_vc,
+				fbcon_is_default);
+
+	if (err) {
+		for (i = first_fb_vc; i <= last_fb_vc; i++)
+			con2fb_map[i] = -1;
+		info_idx = -1;
+	} else {
+		fbcon_has_console_bind = 1;
+	}
+
+	return err;
+}
+
 static int fbcon_takeover(int show_logo)
 {
 	int err, i;
@@ -815,6 +842,8 @@ static void con2fb_init_display(struct vc_data *vc, struct fb_info *info,
  *
  *	Maps a virtual console @unit to a frame buffer device
  *	@newidx.
+ *
+ *	This should be called with the console lock held.
  */
 static int set_con2fb_map(int unit, int newidx, int user)
 {
@@ -832,7 +861,7 @@ static int set_con2fb_map(int unit, int newidx, int user)
 
 	if (!search_for_mapped_con() || !con_is_bound(&fb_con)) {
 		info_idx = newidx;
-		return fbcon_takeover(0);
+		return do_fbcon_takeover(0);
 	}
 
 	if (oldidx != -1)
@@ -840,7 +869,6 @@ static int set_con2fb_map(int unit, int newidx, int user)
 
 	found = search_fb_in_map(newidx);
 
-	console_lock();
 	con2fb_map[unit] = newidx;
 	if (!err && !found)
  		err = con2fb_acquire_newinfo(vc, info, unit, oldidx);
@@ -867,7 +895,6 @@ static int set_con2fb_map(int unit, int newidx, int user)
 	if (!search_fb_in_map(info_idx))
 		info_idx = newidx;
 
-	console_unlock();
  	return err;
 }
 
@@ -990,7 +1017,7 @@ static const char *fbcon_startup(void)
 	}
 
 	/* Setup default font */
-	if (!p->fontdata) {
+	if (!p->fontdata && !vc->vc_font.data) {
 		if (!fontname[0] || !(font = find_font(fontname)))
 			font = get_default_font(info->var.xres,
 						info->var.yres,
@@ -1000,6 +1027,8 @@ static const char *fbcon_startup(void)
 		vc->vc_font.height = font->height;
 		vc->vc_font.data = (void *)(p->fontdata = font->data);
 		vc->vc_font.charcount = 256; /* FIXME  Need to support more fonts */
+	} else {
+		p->fontdata = vc->vc_font.data;
 	}
 
 	cols = FBCON_SWAP(ops->rotate, info->var.xres, info->var.yres);
@@ -1159,9 +1188,9 @@ static void fbcon_init(struct vc_data *vc, int init)
 	ops->p = &fb_display[fg_console];
 }
 
-static void fbcon_free_font(struct display *p)
+static void fbcon_free_font(struct display *p, bool freefont)
 {
-	if (p->userfont && p->fontdata && (--REFCOUNT(p->fontdata) == 0))
+	if (freefont && p->userfont && p->fontdata && (--REFCOUNT(p->fontdata) == 0))
 		kfree(p->fontdata - FONT_EXTRA_WORDS * sizeof(int));
 	p->fontdata = NULL;
 	p->userfont = 0;
@@ -1173,8 +1202,8 @@ static void fbcon_deinit(struct vc_data *vc)
 	struct fb_info *info;
 	struct fbcon_ops *ops;
 	int idx;
+	bool free_font = true;
 
-	fbcon_free_font(p);
 	idx = con2fb_map[vc->vc_num];
 
 	if (idx == -1)
@@ -1185,6 +1214,8 @@ static void fbcon_deinit(struct vc_data *vc)
 	if (!info)
 		goto finished;
 
+	if (info->flags & FBINFO_MISC_FIRMWARE)
+		free_font = false;
 	ops = info->fbcon_par;
 
 	if (!ops)
@@ -1196,6 +1227,8 @@ static void fbcon_deinit(struct vc_data *vc)
 	ops->flags &= ~FBCON_FLAGS_INIT;
 finished:
 
+	fbcon_free_font(p, free_font);
+
 	if (!con_is_bound(&fb_con))
 		fbcon_exit();
 
@@ -2985,7 +3018,7 @@ static int fbcon_unbind(void)
 {
 	int ret;
 
-	ret = unbind_con_driver(&fb_con, first_fb_vc, last_fb_vc,
+	ret = do_unbind_con_driver(&fb_con, first_fb_vc, last_fb_vc,
 				fbcon_is_default);
 
 	if (!ret)
@@ -3000,6 +3033,7 @@ static inline int fbcon_unbind(void)
 }
 #endif /* CONFIG_VT_HW_CONSOLE_BINDING */
 
+/* called with console_lock held */
 static int fbcon_fb_unbind(int idx)
 {
 	int i, new_idx = -1, ret = 0;
@@ -3026,6 +3060,7 @@ static int fbcon_fb_unbind(int idx)
 	return ret;
 }
 
+/* called with console_lock held */
 static int fbcon_fb_unregistered(struct fb_info *info)
 {
 	int i, idx;
@@ -3058,11 +3093,12 @@ static int fbcon_fb_unregistered(struct fb_info *info)
 		primary_device = -1;
 
 	if (!num_registered_fb)
-		unregister_con_driver(&fb_con);
+		do_unregister_con_driver(&fb_con);
 
 	return 0;
 }
 
+/* called with console_lock held */
 static void fbcon_remap_all(int idx)
 {
 	int i;
@@ -3107,6 +3143,7 @@ static inline void fbcon_select_primary(struct fb_info *info)
 }
 #endif /* CONFIG_FRAMEBUFFER_DETECT_PRIMARY */
 
+/* called with console_lock held */
 static int fbcon_fb_registered(struct fb_info *info)
 {
 	int ret = 0, i, idx;
@@ -3123,7 +3160,7 @@ static int fbcon_fb_registered(struct fb_info *info)
 		}
 
 		if (info_idx != -1)
-			ret = fbcon_takeover(1);
+			ret = do_fbcon_takeover(1);
 	} else {
 		for (i = first_fb_vc; i <= last_fb_vc; i++) {
 			if (con2fb_map_boot[i] == idx)
@@ -3259,6 +3296,7 @@ static int fbcon_event_notify(struct notifier_block *self,
 		ret = fbcon_fb_unregistered(info);
 		break;
 	case FB_EVENT_SET_CONSOLE_MAP:
+		/* called with console lock held */
 		con2fb = event->data;
 		ret = set_con2fb_map(con2fb->console - 1,
 				     con2fb->framebuffer, 1);
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index d449a74d4a31..5855d17d19ac 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -1064,7 +1064,7 @@ static int vgacon_do_font_op(struct vgastate *state,char *arg,int set,int ch512)
 	unsigned short video_port_status = vga_video_port_reg + 6;
 	int font_select = 0x00, beg, i;
 	char *charmap;
-	
+	bool clear_attribs = false;
 	if (vga_video_type != VIDEO_TYPE_EGAM) {
 		charmap = (char *) VGA_MAP_MEM(colourmap, 0);
 		beg = 0x0e;
@@ -1169,12 +1169,6 @@ static int vgacon_do_font_op(struct vgastate *state,char *arg,int set,int ch512)
 
 	/* if 512 char mode is already enabled don't re-enable it. */
 	if ((set) && (ch512 != vga_512_chars)) {
-		/* attribute controller */
-		for (i = 0; i < MAX_NR_CONSOLES; i++) {
-			struct vc_data *c = vc_cons[i].d;
-			if (c && c->vc_sw == &vga_con)
-				c->vc_hi_font_mask = ch512 ? 0x0800 : 0;
-		}
 		vga_512_chars = ch512;
 		/* 256-char: enable intensity bit
 		   512-char: disable intensity bit */
@@ -1185,8 +1179,22 @@ static int vgacon_do_font_op(struct vgastate *state,char *arg,int set,int ch512)
 		   it means, but it works, and it appears necessary */
 		inb_p(video_port_status);
 		vga_wattr(state->vgabase, VGA_AR_ENABLE_DISPLAY, 0);	
+		clear_attribs = true;
 	}
 	raw_spin_unlock_irq(&vga_lock);
+
+	if (clear_attribs) {
+		for (i = 0; i < MAX_NR_CONSOLES; i++) {
+			struct vc_data *c = vc_cons[i].d;
+			if (c && c->vc_sw == &vga_con) {
+				/* force hi font mask to 0, so we always clear
+				   the bit on either transition */
+				c->vc_hi_font_mask = 0x00;
+				clear_buffer_attributes(c);
+				c->vc_hi_font_mask = ch512 ? 0x0800 : 0;
+			}
+		}
+	}
 	return 0;
 }
 
diff --git a/drivers/video/display_timing.c b/drivers/video/display_timing.c
new file mode 100644
index 000000000000..5e1822cef571
--- /dev/null
+++ b/drivers/video/display_timing.c
@@ -0,0 +1,24 @@
+/*
+ * generic display timing functions
+ *
+ * Copyright (c) 2012 Steffen Trumtrar <s.trumtrar@pengutronix.de>, Pengutronix
+ *
+ * This file is released under the GPLv2
+ */
+
+#include <linux/export.h>
+#include <linux/slab.h>
+#include <video/display_timing.h>
+
+void display_timings_release(struct display_timings *disp)
+{
+	if (disp->timings) {
+		unsigned int i;
+
+		for (i = 0; i < disp->num_timings; i++)
+			kfree(disp->timings[i]);
+		kfree(disp->timings);
+	}
+	kfree(disp);
+}
+EXPORT_SYMBOL_GPL(display_timings_release);
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 3ff0105a496a..dc61c12ecf8c 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -1177,8 +1177,10 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 		event.data = &con2fb;
 		if (!lock_fb_info(info))
 			return -ENODEV;
+		console_lock();
 		event.info = info;
 		ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP, &event);
+		console_unlock();
 		unlock_fb_info(info);
 		break;
 	case FBIOBLANK:
@@ -1650,7 +1652,9 @@ static int do_register_framebuffer(struct fb_info *fb_info)
 	event.info = fb_info;
 	if (!lock_fb_info(fb_info))
 		return -ENODEV;
+	console_lock();
 	fb_notifier_call_chain(FB_EVENT_FB_REGISTERED, &event);
+	console_unlock();
 	unlock_fb_info(fb_info);
 	return 0;
 }
@@ -1666,8 +1670,10 @@ static int do_unregister_framebuffer(struct fb_info *fb_info)
 
 	if (!lock_fb_info(fb_info))
 		return -ENODEV;
+	console_lock();
 	event.info = fb_info;
 	ret = fb_notifier_call_chain(FB_EVENT_FB_UNBIND, &event);
+	console_unlock();
 	unlock_fb_info(fb_info);
 
 	if (ret)
@@ -1682,7 +1688,9 @@ static int do_unregister_framebuffer(struct fb_info *fb_info)
 	num_registered_fb--;
 	fb_cleanup_device(fb_info);
 	event.info = fb_info;
+	console_lock();
 	fb_notifier_call_chain(FB_EVENT_FB_UNREGISTERED, &event);
+	console_unlock();
 
 	/* this may free fb info */
 	put_fb_info(fb_info);
@@ -1853,11 +1861,8 @@ int fb_new_modelist(struct fb_info *info)
 	err = 1;
 
 	if (!list_empty(&info->modelist)) {
-		if (!lock_fb_info(info))
-			return -ENODEV;
 		event.info = info;
 		err = fb_notifier_call_chain(FB_EVENT_NEW_MODELIST, &event);
-		unlock_fb_info(info);
 	}
 
 	return err;
diff --git a/drivers/video/fbmon.c b/drivers/video/fbmon.c
index cef65574db6c..94ad0f71383c 100644
--- a/drivers/video/fbmon.c
+++ b/drivers/video/fbmon.c
@@ -31,6 +31,8 @@
 #include <linux/pci.h>
 #include <linux/slab.h>
 #include <video/edid.h>
+#include <video/of_videomode.h>
+#include <video/videomode.h>
 #ifdef CONFIG_PPC_OF
 #include <asm/prom.h>
 #include <asm/pci-bridge.h>
@@ -1373,6 +1375,98 @@ int fb_get_mode(int flags, u32 val, struct fb_var_screeninfo *var, struct fb_inf
 	kfree(timings);
 	return err;
 }
+
+#if IS_ENABLED(CONFIG_VIDEOMODE)
+int fb_videomode_from_videomode(const struct videomode *vm,
+				struct fb_videomode *fbmode)
+{
+	unsigned int htotal, vtotal;
+
+	fbmode->xres = vm->hactive;
+	fbmode->left_margin = vm->hback_porch;
+	fbmode->right_margin = vm->hfront_porch;
+	fbmode->hsync_len = vm->hsync_len;
+
+	fbmode->yres = vm->vactive;
+	fbmode->upper_margin = vm->vback_porch;
+	fbmode->lower_margin = vm->vfront_porch;
+	fbmode->vsync_len = vm->vsync_len;
+
+	/* prevent division by zero in KHZ2PICOS macro */
+	fbmode->pixclock = vm->pixelclock ?
+			KHZ2PICOS(vm->pixelclock / 1000) : 0;
+
+	fbmode->sync = 0;
+	fbmode->vmode = 0;
+	if (vm->dmt_flags & VESA_DMT_HSYNC_HIGH)
+		fbmode->sync |= FB_SYNC_HOR_HIGH_ACT;
+	if (vm->dmt_flags & VESA_DMT_HSYNC_HIGH)
+		fbmode->sync |= FB_SYNC_VERT_HIGH_ACT;
+	if (vm->data_flags & DISPLAY_FLAGS_INTERLACED)
+		fbmode->vmode |= FB_VMODE_INTERLACED;
+	if (vm->data_flags & DISPLAY_FLAGS_DOUBLESCAN)
+		fbmode->vmode |= FB_VMODE_DOUBLE;
+	fbmode->flag = 0;
+
+	htotal = vm->hactive + vm->hfront_porch + vm->hback_porch +
+		 vm->hsync_len;
+	vtotal = vm->vactive + vm->vfront_porch + vm->vback_porch +
+		 vm->vsync_len;
+	/* prevent division by zero */
+	if (htotal && vtotal) {
+		fbmode->refresh = vm->pixelclock / (htotal * vtotal);
+	/* a mode must have htotal and vtotal != 0 or it is invalid */
+	} else {
+		fbmode->refresh = 0;
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(fb_videomode_from_videomode);
+#endif
+
+#if IS_ENABLED(CONFIG_OF_VIDEOMODE)
+static inline void dump_fb_videomode(const struct fb_videomode *m)
+{
+	pr_debug("fb_videomode = %ux%u@%uHz (%ukHz) %u %u %u %u %u %u %u %u %u\n",
+		 m->xres, m->yres, m->refresh, m->pixclock, m->left_margin,
+		 m->right_margin, m->upper_margin, m->lower_margin,
+		 m->hsync_len, m->vsync_len, m->sync, m->vmode, m->flag);
+}
+
+/**
+ * of_get_fb_videomode - get a fb_videomode from devicetree
+ * @np: device_node with the timing specification
+ * @fb: will be set to the return value
+ * @index: index into the list of display timings in devicetree
+ *
+ * DESCRIPTION:
+ * This function is expensive and should only be used, if only one mode is to be
+ * read from DT. To get multiple modes start with of_get_display_timings ond
+ * work with that instead.
+ */
+int of_get_fb_videomode(struct device_node *np, struct fb_videomode *fb,
+			int index)
+{
+	struct videomode vm;
+	int ret;
+
+	ret = of_get_videomode(np, &vm, index);
+	if (ret)
+		return ret;
+
+	fb_videomode_from_videomode(&vm, fb);
+
+	pr_debug("%s: got %dx%d display mode from %s\n",
+		of_node_full_name(np), vm.hactive, vm.vactive, np->name);
+	dump_fb_videomode(fb);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(of_get_fb_videomode);
+#endif
+
 #else
 int fb_parse_edid(unsigned char *edid, struct fb_var_screeninfo *var)
 {
diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c
index a55e3669d135..ef476b02fbe5 100644
--- a/drivers/video/fbsysfs.c
+++ b/drivers/video/fbsysfs.c
@@ -177,6 +177,8 @@ static ssize_t store_modes(struct device *device,
 	if (i * sizeof(struct fb_videomode) != count)
 		return -EINVAL;
 
+	if (!lock_fb_info(fb_info))
+		return -ENODEV;
 	console_lock();
 	list_splice(&fb_info->modelist, &old_list);
 	fb_videomode_to_modelist((const struct fb_videomode *)buf, i,
@@ -188,6 +190,7 @@ static ssize_t store_modes(struct device *device,
 		fb_destroy_modelist(&old_list);
 
 	console_unlock();
+	unlock_fb_info(fb_info);
 
 	return 0;
 }
diff --git a/drivers/video/hdmi.c b/drivers/video/hdmi.c
new file mode 100644
index 000000000000..ab23c9b79143
--- /dev/null
+++ b/drivers/video/hdmi.c
@@ -0,0 +1,308 @@
+/*
+ * Copyright (C) 2012 Avionic Design GmbH
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <linux/hdmi.h>
+#include <linux/string.h>
+
+static void hdmi_infoframe_checksum(void *buffer, size_t size)
+{
+	u8 *ptr = buffer;
+	u8 csum = 0;
+	size_t i;
+
+	/* compute checksum */
+	for (i = 0; i < size; i++)
+		csum += ptr[i];
+
+	ptr[3] = 256 - csum;
+}
+
+/**
+ * hdmi_avi_infoframe_init() - initialize an HDMI AVI infoframe
+ * @frame: HDMI AVI infoframe
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int hdmi_avi_infoframe_init(struct hdmi_avi_infoframe *frame)
+{
+	memset(frame, 0, sizeof(*frame));
+
+	frame->type = HDMI_INFOFRAME_TYPE_AVI;
+	frame->version = 2;
+	frame->length = 13;
+
+	return 0;
+}
+EXPORT_SYMBOL(hdmi_avi_infoframe_init);
+
+/**
+ * hdmi_avi_infoframe_pack() - write HDMI AVI infoframe to binary buffer
+ * @frame: HDMI AVI infoframe
+ * @buffer: destination buffer
+ * @size: size of buffer
+ *
+ * Packs the information contained in the @frame structure into a binary
+ * representation that can be written into the corresponding controller
+ * registers. Also computes the checksum as required by section 5.3.5 of
+ * the HDMI 1.4 specification.
+ *
+ * Returns the number of bytes packed into the binary buffer or a negative
+ * error code on failure.
+ */
+ssize_t hdmi_avi_infoframe_pack(struct hdmi_avi_infoframe *frame, void *buffer,
+				size_t size)
+{
+	u8 *ptr = buffer;
+	size_t length;
+
+	length = HDMI_INFOFRAME_HEADER_SIZE + frame->length;
+
+	if (size < length)
+		return -ENOSPC;
+
+	memset(buffer, 0, length);
+
+	ptr[0] = frame->type;
+	ptr[1] = frame->version;
+	ptr[2] = frame->length;
+	ptr[3] = 0; /* checksum */
+
+	/* start infoframe payload */
+	ptr += HDMI_INFOFRAME_HEADER_SIZE;
+
+	ptr[0] = ((frame->colorspace & 0x3) << 5) | (frame->scan_mode & 0x3);
+
+	if (frame->active_info_valid)
+		ptr[0] |= BIT(4);
+
+	if (frame->horizontal_bar_valid)
+		ptr[0] |= BIT(3);
+
+	if (frame->vertical_bar_valid)
+		ptr[0] |= BIT(2);
+
+	ptr[1] = ((frame->colorimetry & 0x3) << 6) |
+		 ((frame->picture_aspect & 0x3) << 4) |
+		 (frame->active_aspect & 0xf);
+
+	ptr[2] = ((frame->extended_colorimetry & 0x7) << 4) |
+		 ((frame->quantization_range & 0x3) << 2) |
+		 (frame->nups & 0x3);
+
+	if (frame->itc)
+		ptr[2] |= BIT(7);
+
+	ptr[3] = frame->video_code & 0x7f;
+
+	ptr[4] = ((frame->ycc_quantization_range & 0x3) << 6) |
+		 ((frame->content_type & 0x3) << 4) |
+		 (frame->pixel_repeat & 0xf);
+
+	ptr[5] = frame->top_bar & 0xff;
+	ptr[6] = (frame->top_bar >> 8) & 0xff;
+	ptr[7] = frame->bottom_bar & 0xff;
+	ptr[8] = (frame->bottom_bar >> 8) & 0xff;
+	ptr[9] = frame->left_bar & 0xff;
+	ptr[10] = (frame->left_bar >> 8) & 0xff;
+	ptr[11] = frame->right_bar & 0xff;
+	ptr[12] = (frame->right_bar >> 8) & 0xff;
+
+	hdmi_infoframe_checksum(buffer, length);
+
+	return length;
+}
+EXPORT_SYMBOL(hdmi_avi_infoframe_pack);
+
+/**
+ * hdmi_spd_infoframe_init() - initialize an HDMI SPD infoframe
+ * @frame: HDMI SPD infoframe
+ * @vendor: vendor string
+ * @product: product string
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int hdmi_spd_infoframe_init(struct hdmi_spd_infoframe *frame,
+			    const char *vendor, const char *product)
+{
+	memset(frame, 0, sizeof(*frame));
+
+	frame->type = HDMI_INFOFRAME_TYPE_SPD;
+	frame->version = 1;
+	frame->length = 25;
+
+	strncpy(frame->vendor, vendor, sizeof(frame->vendor));
+	strncpy(frame->product, product, sizeof(frame->product));
+
+	return 0;
+}
+EXPORT_SYMBOL(hdmi_spd_infoframe_init);
+
+/**
+ * hdmi_spd_infoframe_pack() - write HDMI SPD infoframe to binary buffer
+ * @frame: HDMI SPD infoframe
+ * @buffer: destination buffer
+ * @size: size of buffer
+ *
+ * Packs the information contained in the @frame structure into a binary
+ * representation that can be written into the corresponding controller
+ * registers. Also computes the checksum as required by section 5.3.5 of
+ * the HDMI 1.4 specification.
+ *
+ * Returns the number of bytes packed into the binary buffer or a negative
+ * error code on failure.
+ */
+ssize_t hdmi_spd_infoframe_pack(struct hdmi_spd_infoframe *frame, void *buffer,
+				size_t size)
+{
+	u8 *ptr = buffer;
+	size_t length;
+
+	length = HDMI_INFOFRAME_HEADER_SIZE + frame->length;
+
+	if (size < length)
+		return -ENOSPC;
+
+	memset(buffer, 0, length);
+
+	ptr[0] = frame->type;
+	ptr[1] = frame->version;
+	ptr[2] = frame->length;
+	ptr[3] = 0; /* checksum */
+
+	/* start infoframe payload */
+	ptr += HDMI_INFOFRAME_HEADER_SIZE;
+
+	memcpy(ptr, frame->vendor, sizeof(frame->vendor));
+	memcpy(ptr + 8, frame->product, sizeof(frame->product));
+
+	ptr[24] = frame->sdi;
+
+	hdmi_infoframe_checksum(buffer, length);
+
+	return length;
+}
+EXPORT_SYMBOL(hdmi_spd_infoframe_pack);
+
+/**
+ * hdmi_audio_infoframe_init() - initialize an HDMI audio infoframe
+ * @frame: HDMI audio infoframe
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int hdmi_audio_infoframe_init(struct hdmi_audio_infoframe *frame)
+{
+	memset(frame, 0, sizeof(*frame));
+
+	frame->type = HDMI_INFOFRAME_TYPE_AUDIO;
+	frame->version = 1;
+	frame->length = 10;
+
+	return 0;
+}
+EXPORT_SYMBOL(hdmi_audio_infoframe_init);
+
+/**
+ * hdmi_audio_infoframe_pack() - write HDMI audio infoframe to binary buffer
+ * @frame: HDMI audio infoframe
+ * @buffer: destination buffer
+ * @size: size of buffer
+ *
+ * Packs the information contained in the @frame structure into a binary
+ * representation that can be written into the corresponding controller
+ * registers. Also computes the checksum as required by section 5.3.5 of
+ * the HDMI 1.4 specification.
+ *
+ * Returns the number of bytes packed into the binary buffer or a negative
+ * error code on failure.
+ */
+ssize_t hdmi_audio_infoframe_pack(struct hdmi_audio_infoframe *frame,
+				  void *buffer, size_t size)
+{
+	unsigned char channels;
+	u8 *ptr = buffer;
+	size_t length;
+
+	length = HDMI_INFOFRAME_HEADER_SIZE + frame->length;
+
+	if (size < length)
+		return -ENOSPC;
+
+	memset(buffer, 0, length);
+
+	if (frame->channels >= 2)
+		channels = frame->channels - 1;
+	else
+		channels = 0;
+
+	ptr[0] = frame->type;
+	ptr[1] = frame->version;
+	ptr[2] = frame->length;
+	ptr[3] = 0; /* checksum */
+
+	/* start infoframe payload */
+	ptr += HDMI_INFOFRAME_HEADER_SIZE;
+
+	ptr[0] = ((frame->coding_type & 0xf) << 4) | (channels & 0x7);
+	ptr[1] = ((frame->sample_frequency & 0x7) << 2) |
+		 (frame->sample_size & 0x3);
+	ptr[2] = frame->coding_type_ext & 0x1f;
+	ptr[3] = frame->channel_allocation;
+	ptr[4] = (frame->level_shift_value & 0xf) << 3;
+
+	if (frame->downmix_inhibit)
+		ptr[4] |= BIT(7);
+
+	hdmi_infoframe_checksum(buffer, length);
+
+	return length;
+}
+EXPORT_SYMBOL(hdmi_audio_infoframe_pack);
+
+/**
+ * hdmi_vendor_infoframe_pack() - write a HDMI vendor infoframe to binary
+ *                                buffer
+ * @frame: HDMI vendor infoframe
+ * @buffer: destination buffer
+ * @size: size of buffer
+ *
+ * Packs the information contained in the @frame structure into a binary
+ * representation that can be written into the corresponding controller
+ * registers. Also computes the checksum as required by section 5.3.5 of
+ * the HDMI 1.4 specification.
+ *
+ * Returns the number of bytes packed into the binary buffer or a negative
+ * error code on failure.
+ */
+ssize_t hdmi_vendor_infoframe_pack(struct hdmi_vendor_infoframe *frame,
+				   void *buffer, size_t size)
+{
+	u8 *ptr = buffer;
+	size_t length;
+
+	length = HDMI_INFOFRAME_HEADER_SIZE + frame->length;
+
+	if (size < length)
+		return -ENOSPC;
+
+	memset(buffer, 0, length);
+
+	ptr[0] = frame->type;
+	ptr[1] = frame->version;
+	ptr[2] = frame->length;
+	ptr[3] = 0; /* checksum */
+
+	memcpy(&ptr[HDMI_INFOFRAME_HEADER_SIZE], frame->data, frame->length);
+
+	hdmi_infoframe_checksum(buffer, length);
+
+	return length;
+}
+EXPORT_SYMBOL(hdmi_vendor_infoframe_pack);
diff --git a/drivers/video/of_display_timing.c b/drivers/video/of_display_timing.c
new file mode 100644
index 000000000000..13ecd9897010
--- /dev/null
+++ b/drivers/video/of_display_timing.c
@@ -0,0 +1,239 @@
+/*
+ * OF helpers for parsing display timings
+ *
+ * Copyright (c) 2012 Steffen Trumtrar <s.trumtrar@pengutronix.de>, Pengutronix
+ *
+ * based on of_videomode.c by Sascha Hauer <s.hauer@pengutronix.de>
+ *
+ * This file is released under the GPLv2
+ */
+#include <linux/export.h>
+#include <linux/of.h>
+#include <linux/slab.h>
+#include <video/display_timing.h>
+#include <video/of_display_timing.h>
+
+/**
+ * parse_timing_property - parse timing_entry from device_node
+ * @np: device_node with the property
+ * @name: name of the property
+ * @result: will be set to the return value
+ *
+ * DESCRIPTION:
+ * Every display_timing can be specified with either just the typical value or
+ * a range consisting of min/typ/max. This function helps handling this
+ **/
+static int parse_timing_property(struct device_node *np, const char *name,
+			  struct timing_entry *result)
+{
+	struct property *prop;
+	int length, cells, ret;
+
+	prop = of_find_property(np, name, &length);
+	if (!prop) {
+		pr_err("%s: could not find property %s\n",
+			of_node_full_name(np), name);
+		return -EINVAL;
+	}
+
+	cells = length / sizeof(u32);
+	if (cells == 1) {
+		ret = of_property_read_u32(np, name, &result->typ);
+		result->min = result->typ;
+		result->max = result->typ;
+	} else if (cells == 3) {
+		ret = of_property_read_u32_array(np, name, &result->min, cells);
+	} else {
+		pr_err("%s: illegal timing specification in %s\n",
+			of_node_full_name(np), name);
+		return -EINVAL;
+	}
+
+	return ret;
+}
+
+/**
+ * of_get_display_timing - parse display_timing entry from device_node
+ * @np: device_node with the properties
+ **/
+static struct display_timing *of_get_display_timing(struct device_node *np)
+{
+	struct display_timing *dt;
+	u32 val = 0;
+	int ret = 0;
+
+	dt = kzalloc(sizeof(*dt), GFP_KERNEL);
+	if (!dt) {
+		pr_err("%s: could not allocate display_timing struct\n",
+			of_node_full_name(np));
+		return NULL;
+	}
+
+	ret |= parse_timing_property(np, "hback-porch", &dt->hback_porch);
+	ret |= parse_timing_property(np, "hfront-porch", &dt->hfront_porch);
+	ret |= parse_timing_property(np, "hactive", &dt->hactive);
+	ret |= parse_timing_property(np, "hsync-len", &dt->hsync_len);
+	ret |= parse_timing_property(np, "vback-porch", &dt->vback_porch);
+	ret |= parse_timing_property(np, "vfront-porch", &dt->vfront_porch);
+	ret |= parse_timing_property(np, "vactive", &dt->vactive);
+	ret |= parse_timing_property(np, "vsync-len", &dt->vsync_len);
+	ret |= parse_timing_property(np, "clock-frequency", &dt->pixelclock);
+
+	dt->dmt_flags = 0;
+	dt->data_flags = 0;
+	if (!of_property_read_u32(np, "vsync-active", &val))
+		dt->dmt_flags |= val ? VESA_DMT_VSYNC_HIGH :
+				VESA_DMT_VSYNC_LOW;
+	if (!of_property_read_u32(np, "hsync-active", &val))
+		dt->dmt_flags |= val ? VESA_DMT_HSYNC_HIGH :
+				VESA_DMT_HSYNC_LOW;
+	if (!of_property_read_u32(np, "de-active", &val))
+		dt->data_flags |= val ? DISPLAY_FLAGS_DE_HIGH :
+				DISPLAY_FLAGS_DE_LOW;
+	if (!of_property_read_u32(np, "pixelclk-active", &val))
+		dt->data_flags |= val ? DISPLAY_FLAGS_PIXDATA_POSEDGE :
+				DISPLAY_FLAGS_PIXDATA_NEGEDGE;
+
+	if (of_property_read_bool(np, "interlaced"))
+		dt->data_flags |= DISPLAY_FLAGS_INTERLACED;
+	if (of_property_read_bool(np, "doublescan"))
+		dt->data_flags |= DISPLAY_FLAGS_DOUBLESCAN;
+
+	if (ret) {
+		pr_err("%s: error reading timing properties\n",
+			of_node_full_name(np));
+		kfree(dt);
+		return NULL;
+	}
+
+	return dt;
+}
+
+/**
+ * of_get_display_timings - parse all display_timing entries from a device_node
+ * @np: device_node with the subnodes
+ **/
+struct display_timings *of_get_display_timings(struct device_node *np)
+{
+	struct device_node *timings_np;
+	struct device_node *entry;
+	struct device_node *native_mode;
+	struct display_timings *disp;
+
+	if (!np) {
+		pr_err("%s: no devicenode given\n", of_node_full_name(np));
+		return NULL;
+	}
+
+	timings_np = of_find_node_by_name(np, "display-timings");
+	if (!timings_np) {
+		pr_err("%s: could not find display-timings node\n",
+			of_node_full_name(np));
+		return NULL;
+	}
+
+	disp = kzalloc(sizeof(*disp), GFP_KERNEL);
+	if (!disp) {
+		pr_err("%s: could not allocate struct disp'\n",
+			of_node_full_name(np));
+		goto dispfail;
+	}
+
+	entry = of_parse_phandle(timings_np, "native-mode", 0);
+	/* assume first child as native mode if none provided */
+	if (!entry)
+		entry = of_get_next_child(np, NULL);
+	/* if there is no child, it is useless to go on */
+	if (!entry) {
+		pr_err("%s: no timing specifications given\n",
+			of_node_full_name(np));
+		goto entryfail;
+	}
+
+	pr_debug("%s: using %s as default timing\n",
+		of_node_full_name(np), entry->name);
+
+	native_mode = entry;
+
+	disp->num_timings = of_get_child_count(timings_np);
+	if (disp->num_timings == 0) {
+		/* should never happen, as entry was already found above */
+		pr_err("%s: no timings specified\n", of_node_full_name(np));
+		goto entryfail;
+	}
+
+	disp->timings = kzalloc(sizeof(struct display_timing *) *
+				disp->num_timings, GFP_KERNEL);
+	if (!disp->timings) {
+		pr_err("%s: could not allocate timings array\n",
+			of_node_full_name(np));
+		goto entryfail;
+	}
+
+	disp->num_timings = 0;
+	disp->native_mode = 0;
+
+	for_each_child_of_node(timings_np, entry) {
+		struct display_timing *dt;
+
+		dt = of_get_display_timing(entry);
+		if (!dt) {
+			/*
+			 * to not encourage wrong devicetrees, fail in case of
+			 * an error
+			 */
+			pr_err("%s: error in timing %d\n",
+				of_node_full_name(np), disp->num_timings + 1);
+			goto timingfail;
+		}
+
+		if (native_mode == entry)
+			disp->native_mode = disp->num_timings;
+
+		disp->timings[disp->num_timings] = dt;
+		disp->num_timings++;
+	}
+	of_node_put(timings_np);
+	/*
+	 * native_mode points to the device_node returned by of_parse_phandle
+	 * therefore call of_node_put on it
+	 */
+	of_node_put(native_mode);
+
+	pr_debug("%s: got %d timings. Using timing #%d as default\n",
+		of_node_full_name(np), disp->num_timings,
+		disp->native_mode + 1);
+
+	return disp;
+
+timingfail:
+	if (native_mode)
+		of_node_put(native_mode);
+	display_timings_release(disp);
+entryfail:
+	kfree(disp);
+dispfail:
+	of_node_put(timings_np);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(of_get_display_timings);
+
+/**
+ * of_display_timings_exist - check if a display-timings node is provided
+ * @np: device_node with the timing
+ **/
+int of_display_timings_exist(struct device_node *np)
+{
+	struct device_node *timings_np;
+
+	if (!np)
+		return -EINVAL;
+
+	timings_np = of_parse_phandle(np, "display-timings", 0);
+	if (!timings_np)
+		return -EINVAL;
+
+	of_node_put(timings_np);
+	return 1;
+}
+EXPORT_SYMBOL_GPL(of_display_timings_exist);
diff --git a/drivers/video/of_videomode.c b/drivers/video/of_videomode.c
new file mode 100644
index 000000000000..5b8066cd397f
--- /dev/null
+++ b/drivers/video/of_videomode.c
@@ -0,0 +1,54 @@
+/*
+ * generic videomode helper
+ *
+ * Copyright (c) 2012 Steffen Trumtrar <s.trumtrar@pengutronix.de>, Pengutronix
+ *
+ * This file is released under the GPLv2
+ */
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <linux/of.h>
+#include <video/display_timing.h>
+#include <video/of_display_timing.h>
+#include <video/of_videomode.h>
+#include <video/videomode.h>
+
+/**
+ * of_get_videomode - get the videomode #<index> from devicetree
+ * @np - devicenode with the display_timings
+ * @vm - set to return value
+ * @index - index into list of display_timings
+ *	    (Set this to OF_USE_NATIVE_MODE to use whatever mode is
+ *	     specified as native mode in the DT.)
+ *
+ * DESCRIPTION:
+ * Get a list of all display timings and put the one
+ * specified by index into *vm. This function should only be used, if
+ * only one videomode is to be retrieved. A driver that needs to work
+ * with multiple/all videomodes should work with
+ * of_get_display_timings instead.
+ **/
+int of_get_videomode(struct device_node *np, struct videomode *vm,
+		     int index)
+{
+	struct display_timings *disp;
+	int ret;
+
+	disp = of_get_display_timings(np);
+	if (!disp) {
+		pr_err("%s: no timings specified\n", of_node_full_name(np));
+		return -EINVAL;
+	}
+
+	if (index == OF_USE_NATIVE_MODE)
+		index = disp->native_mode;
+
+	ret = videomode_from_timing(disp, vm, index);
+	if (ret)
+		return ret;
+
+	display_timings_release(disp);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(of_get_videomode);
diff --git a/drivers/video/via/hw.c b/drivers/video/via/hw.c
index 80233dae358a..22450908306c 100644
--- a/drivers/video/via/hw.c
+++ b/drivers/video/via/hw.c
@@ -1467,10 +1467,10 @@ void viafb_set_vclock(u32 clk, int set_iga)
 	via_write_misc_reg_mask(0x0C, 0x0C); /* select external clock */
 }
 
-struct display_timing var_to_timing(const struct fb_var_screeninfo *var,
+struct via_display_timing var_to_timing(const struct fb_var_screeninfo *var,
 	u16 cxres, u16 cyres)
 {
-	struct display_timing timing;
+	struct via_display_timing timing;
 	u16 dx = (var->xres - cxres) / 2, dy = (var->yres - cyres) / 2;
 
 	timing.hor_addr = cxres;
@@ -1491,7 +1491,7 @@ struct display_timing var_to_timing(const struct fb_var_screeninfo *var,
 void viafb_fill_crtc_timing(const struct fb_var_screeninfo *var,
 	u16 cxres, u16 cyres, int iga)
 {
-	struct display_timing crt_reg = var_to_timing(var,
+	struct via_display_timing crt_reg = var_to_timing(var,
 		cxres ? cxres : var->xres, cyres ? cyres : var->yres);
 
 	if (iga == IGA1)
diff --git a/drivers/video/via/hw.h b/drivers/video/via/hw.h
index a8205754c736..3be073c58b03 100644
--- a/drivers/video/via/hw.h
+++ b/drivers/video/via/hw.h
@@ -637,7 +637,7 @@ extern int viafb_LCD_ON;
 extern int viafb_DVI_ON;
 extern int viafb_hotplug;
 
-struct display_timing var_to_timing(const struct fb_var_screeninfo *var,
+struct via_display_timing var_to_timing(const struct fb_var_screeninfo *var,
 	u16 cxres, u16 cyres);
 void viafb_fill_crtc_timing(const struct fb_var_screeninfo *var,
 	u16 cxres, u16 cyres, int iga);
diff --git a/drivers/video/via/lcd.c b/drivers/video/via/lcd.c
index 980ee1b1dcf3..5d21ff436ec8 100644
--- a/drivers/video/via/lcd.c
+++ b/drivers/video/via/lcd.c
@@ -549,7 +549,7 @@ void viafb_lcd_set_mode(const struct fb_var_screeninfo *var, u16 cxres,
 	int panel_hres = plvds_setting_info->lcd_panel_hres;
 	int panel_vres = plvds_setting_info->lcd_panel_vres;
 	u32 clock;
-	struct display_timing timing;
+	struct via_display_timing timing;
 	struct fb_var_screeninfo panel_var;
 	const struct fb_videomode *mode_crt_table, *panel_crt_table;
 
diff --git a/drivers/video/via/share.h b/drivers/video/via/share.h
index 3158dfc90bed..65c65c611e0a 100644
--- a/drivers/video/via/share.h
+++ b/drivers/video/via/share.h
@@ -319,7 +319,7 @@ struct crt_mode_table {
 	int refresh_rate;
 	int h_sync_polarity;
 	int v_sync_polarity;
-	struct display_timing crtc;
+	struct via_display_timing crtc;
 };
 
 struct io_reg {
diff --git a/drivers/video/via/via_modesetting.c b/drivers/video/via/via_modesetting.c
index 0e431aee17bb..0b414b09b9b4 100644
--- a/drivers/video/via/via_modesetting.c
+++ b/drivers/video/via/via_modesetting.c
@@ -30,9 +30,9 @@
 #include "debug.h"
 
 
-void via_set_primary_timing(const struct display_timing *timing)
+void via_set_primary_timing(const struct via_display_timing *timing)
 {
-	struct display_timing raw;
+	struct via_display_timing raw;
 
 	raw.hor_total = timing->hor_total / 8 - 5;
 	raw.hor_addr = timing->hor_addr / 8 - 1;
@@ -88,9 +88,9 @@ void via_set_primary_timing(const struct display_timing *timing)
 	via_write_reg_mask(VIACR, 0x17, 0x80, 0x80);
 }
 
-void via_set_secondary_timing(const struct display_timing *timing)
+void via_set_secondary_timing(const struct via_display_timing *timing)
 {
-	struct display_timing raw;
+	struct via_display_timing raw;
 
 	raw.hor_total = timing->hor_total - 1;
 	raw.hor_addr = timing->hor_addr - 1;
diff --git a/drivers/video/via/via_modesetting.h b/drivers/video/via/via_modesetting.h
index 06e09fe351ae..f6a6503da3b3 100644
--- a/drivers/video/via/via_modesetting.h
+++ b/drivers/video/via/via_modesetting.h
@@ -33,7 +33,7 @@
 #define VIA_PITCH_MAX	0x3FF8
 
 
-struct display_timing {
+struct via_display_timing {
 	u16 hor_total;
 	u16 hor_addr;
 	u16 hor_blank_start;
@@ -49,8 +49,8 @@ struct display_timing {
 };
 
 
-void via_set_primary_timing(const struct display_timing *timing);
-void via_set_secondary_timing(const struct display_timing *timing);
+void via_set_primary_timing(const struct via_display_timing *timing);
+void via_set_secondary_timing(const struct via_display_timing *timing);
 void via_set_primary_address(u32 addr);
 void via_set_secondary_address(u32 addr);
 void via_set_primary_pitch(u32 pitch);
diff --git a/drivers/video/videomode.c b/drivers/video/videomode.c
new file mode 100644
index 000000000000..21c47a202afa
--- /dev/null
+++ b/drivers/video/videomode.c
@@ -0,0 +1,39 @@
+/*
+ * generic display timing functions
+ *
+ * Copyright (c) 2012 Steffen Trumtrar <s.trumtrar@pengutronix.de>, Pengutronix
+ *
+ * This file is released under the GPLv2
+ */
+
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <video/display_timing.h>
+#include <video/videomode.h>
+
+int videomode_from_timing(const struct display_timings *disp,
+			  struct videomode *vm, unsigned int index)
+{
+	struct display_timing *dt;
+
+	dt = display_timings_get(disp, index);
+	if (!dt)
+		return -EINVAL;
+
+	vm->pixelclock = display_timing_get_value(&dt->pixelclock, TE_TYP);
+	vm->hactive = display_timing_get_value(&dt->hactive, TE_TYP);
+	vm->hfront_porch = display_timing_get_value(&dt->hfront_porch, TE_TYP);
+	vm->hback_porch = display_timing_get_value(&dt->hback_porch, TE_TYP);
+	vm->hsync_len = display_timing_get_value(&dt->hsync_len, TE_TYP);
+
+	vm->vactive = display_timing_get_value(&dt->vactive, TE_TYP);
+	vm->vfront_porch = display_timing_get_value(&dt->vfront_porch, TE_TYP);
+	vm->vback_porch = display_timing_get_value(&dt->vback_porch, TE_TYP);
+	vm->vsync_len = display_timing_get_value(&dt->vsync_len, TE_TYP);
+
+	vm->dmt_flags = dt->dmt_flags;
+	vm->data_flags = dt->data_flags;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(videomode_from_timing);
author	Linus Torvalds <torvalds@linux-foundation.org>	2013-02-25 16:46:44 -0800
committer	Linus Torvalds <torvalds@linux-foundation.org>	2013-02-25 16:46:44 -0800
commit	fffddfd6c8e0c10c42c6e2cc54ba880fcc36ebbb (patch)
tree	71bc5e597124dbaf7550f1e089d675718b3ed5c0 /drivers
parent	69086a78bdc973ec0b722be790b146e84ba8a8c4 (diff)
parent	be88298b0a3f771a4802f20c5e66af74bfd1dff1 (diff)
download	linux-fffddfd6c8e0c10c42c6e2cc54ba880fcc36ebbb.tar.gz