diff options
Diffstat (limited to 'arch/x86')
58 files changed, 2107 insertions, 1569 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 6a47bb22657f..9037289617ac 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -130,7 +130,7 @@ config SBUS bool config NEED_DMA_MAP_STATE - def_bool (X86_64 || DMAR || DMA_API_DEBUG) + def_bool (X86_64 || INTEL_IOMMU || DMA_API_DEBUG) config NEED_SG_DMA_LENGTH def_bool y @@ -220,7 +220,7 @@ config ARCH_SUPPORTS_DEBUG_PAGEALLOC config HAVE_INTEL_TXT def_bool y - depends on EXPERIMENTAL && DMAR && ACPI + depends on EXPERIMENTAL && INTEL_IOMMU && ACPI config X86_32_SMP def_bool y @@ -279,7 +279,7 @@ config SMP Y to "Enhanced Real Time Clock Support", below. The "Advanced Power Management" code will be disabled if you say Y here. - See also <file:Documentation/i386/IO-APIC.txt>, + See also <file:Documentation/x86/i386/IO-APIC.txt>, <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at <http://www.tldp.org/docs.html#howto>. @@ -287,7 +287,7 @@ config SMP config X86_X2APIC bool "Support x2apic" - depends on X86_LOCAL_APIC && X86_64 && INTR_REMAP + depends on X86_LOCAL_APIC && X86_64 && IRQ_REMAP ---help--- This enables x2apic support on CPUs that have this feature. diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index c0f8a5c88910..bf56e1793272 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -139,7 +139,7 @@ config IOMMU_DEBUG code. When you use it make sure you have a big enough IOMMU/AGP aperture. Most of the options enabled by this can be set more finegrained using the iommu= command line - options. See Documentation/x86_64/boot-options.txt for more + options. See Documentation/x86/x86_64/boot-options.txt for more details. config IOMMU_STRESS diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S index 93e689f4bd86..bdb4d458ec8c 100644 --- a/arch/x86/boot/header.S +++ b/arch/x86/boot/header.S @@ -129,7 +129,7 @@ start_sys_seg: .word SYSSEG # obsolete and meaningless, but just type_of_loader: .byte 0 # 0 means ancient bootloader, newer # bootloaders know to change this. - # See Documentation/i386/boot.txt for + # See Documentation/x86/boot.txt for # assigned ids # flags, unused bits must be zero (RFU) bit within loadflags diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index 22a0dc8e51dd..058a35b8286c 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -67,8 +67,8 @@ CONFIG_CPU_FREQ_GOV_PERFORMANCE=y CONFIG_CPU_FREQ_GOV_ONDEMAND=y CONFIG_X86_ACPI_CPUFREQ=y CONFIG_PCI_MMCONFIG=y -CONFIG_DMAR=y -# CONFIG_DMAR_DEFAULT_ON is not set +CONFIG_INTEL_IOMMU=y +# CONFIG_INTEL_IOMMU_DEFAULT_ON is not set CONFIG_PCIEPORTBUS=y CONFIG_PCCARD=y CONFIG_YENTA=y diff --git a/arch/x86/include/asm/device.h b/arch/x86/include/asm/device.h index 029f230ab637..63a2a03d7d51 100644 --- a/arch/x86/include/asm/device.h +++ b/arch/x86/include/asm/device.h @@ -8,7 +8,7 @@ struct dev_archdata { #ifdef CONFIG_X86_64 struct dma_map_ops *dma_ops; #endif -#if defined(CONFIG_DMAR) || defined(CONFIG_AMD_IOMMU) +#if defined(CONFIG_INTEL_IOMMU) || defined(CONFIG_AMD_IOMMU) void *iommu; /* hook for IOMMU specific extension */ #endif }; diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index d4c419f883a0..ed3065fd6314 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -2,7 +2,7 @@ #define _ASM_X86_DMA_MAPPING_H /* - * IOMMU interface. See Documentation/PCI/PCI-DMA-mapping.txt and + * IOMMU interface. See Documentation/DMA-API-HOWTO.txt and * Documentation/DMA-API.txt for documentation. */ diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 09199052060f..eb92a6ed2be7 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -119,7 +119,7 @@ struct irq_cfg { cpumask_var_t old_domain; u8 vector; u8 move_in_progress : 1; -#ifdef CONFIG_INTR_REMAP +#ifdef CONFIG_IRQ_REMAP struct irq_2_iommu irq_2_iommu; #endif }; diff --git a/arch/x86/include/asm/hyperv.h b/arch/x86/include/asm/hyperv.h index 5df477ac3af7..b80420bcd09d 100644 --- a/arch/x86/include/asm/hyperv.h +++ b/arch/x86/include/asm/hyperv.h @@ -189,5 +189,6 @@ #define HV_STATUS_INVALID_HYPERCALL_CODE 2 #define HV_STATUS_INVALID_HYPERCALL_INPUT 3 #define HV_STATUS_INVALID_ALIGNMENT 4 +#define HV_STATUS_INSUFFICIENT_BUFFERS 19 #endif diff --git a/arch/x86/include/asm/irq_remapping.h b/arch/x86/include/asm/irq_remapping.h index 1c23360fb2d8..47d99934580f 100644 --- a/arch/x86/include/asm/irq_remapping.h +++ b/arch/x86/include/asm/irq_remapping.h @@ -3,7 +3,8 @@ #define IRTE_DEST(dest) ((x2apic_mode) ? dest : dest << 8) -#ifdef CONFIG_INTR_REMAP +#ifdef CONFIG_IRQ_REMAP +static void irq_remap_modify_chip_defaults(struct irq_chip *chip); static inline void prepare_irte(struct irte *irte, int vector, unsigned int dest) { @@ -36,6 +37,9 @@ static inline bool irq_remapped(struct irq_cfg *cfg) { return false; } +static inline void irq_remap_modify_chip_defaults(struct irq_chip *chip) +{ +} #endif #endif /* _ASM_X86_IRQ_REMAPPING_H */ diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index 4886a68f267e..fd3f9f18cf3f 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -22,27 +22,26 @@ void arch_trigger_all_cpu_backtrace(void); #define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace #endif -/* - * Define some priorities for the nmi notifier call chain. - * - * Create a local nmi bit that has a higher priority than - * external nmis, because the local ones are more frequent. - * - * Also setup some default high/normal/low settings for - * subsystems to registers with. Using 4 bits to separate - * the priorities. This can go a lot higher if needed be. - */ - -#define NMI_LOCAL_SHIFT 16 /* randomly picked */ -#define NMI_LOCAL_BIT (1ULL << NMI_LOCAL_SHIFT) -#define NMI_HIGH_PRIOR (1ULL << 8) -#define NMI_NORMAL_PRIOR (1ULL << 4) -#define NMI_LOW_PRIOR (1ULL << 0) -#define NMI_LOCAL_HIGH_PRIOR (NMI_LOCAL_BIT | NMI_HIGH_PRIOR) -#define NMI_LOCAL_NORMAL_PRIOR (NMI_LOCAL_BIT | NMI_NORMAL_PRIOR) -#define NMI_LOCAL_LOW_PRIOR (NMI_LOCAL_BIT | NMI_LOW_PRIOR) +#define NMI_FLAG_FIRST 1 + +enum { + NMI_LOCAL=0, + NMI_UNKNOWN, + NMI_MAX +}; + +#define NMI_DONE 0 +#define NMI_HANDLED 1 + +typedef int (*nmi_handler_t)(unsigned int, struct pt_regs *); + +int register_nmi_handler(unsigned int, nmi_handler_t, unsigned long, + const char *); + +void unregister_nmi_handler(unsigned int, const char *); void stop_nmi(void); void restart_nmi(void); +void local_touch_nmi(void); #endif /* _ASM_X86_NMI_H */ diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index 094fb30817ab..f61c62f7d5d8 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -29,6 +29,9 @@ #define ARCH_PERFMON_EVENTSEL_INV (1ULL << 23) #define ARCH_PERFMON_EVENTSEL_CMASK 0xFF000000ULL +#define AMD_PERFMON_EVENTSEL_GUESTONLY (1ULL << 40) +#define AMD_PERFMON_EVENTSEL_HOSTONLY (1ULL << 41) + #define AMD64_EVENTSEL_EVENT \ (ARCH_PERFMON_EVENTSEL_EVENT | (0x0FULL << 32)) #define INTEL_ARCH_EVENT_MASK \ @@ -43,14 +46,17 @@ #define AMD64_RAW_EVENT_MASK \ (X86_RAW_EVENT_MASK | \ AMD64_EVENTSEL_EVENT) +#define AMD64_NUM_COUNTERS 4 +#define AMD64_NUM_COUNTERS_F15H 6 +#define AMD64_NUM_COUNTERS_MAX AMD64_NUM_COUNTERS_F15H -#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL 0x3c +#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL 0x3c #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK (0x00 << 8) -#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX 0 +#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX 0 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT \ (1 << (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX)) -#define ARCH_PERFMON_BRANCH_MISSES_RETIRED 6 +#define ARCH_PERFMON_BRANCH_MISSES_RETIRED 6 /* * Intel "Architectural Performance Monitoring" CPUID @@ -110,6 +116,35 @@ union cpuid10_edx { */ #define X86_PMC_IDX_FIXED_BTS (X86_PMC_IDX_FIXED + 16) +/* + * IBS cpuid feature detection + */ + +#define IBS_CPUID_FEATURES 0x8000001b + +/* + * Same bit mask as for IBS cpuid feature flags (Fn8000_001B_EAX), but + * bit 0 is used to indicate the existence of IBS. + */ +#define IBS_CAPS_AVAIL (1U<<0) +#define IBS_CAPS_FETCHSAM (1U<<1) +#define IBS_CAPS_OPSAM (1U<<2) +#define IBS_CAPS_RDWROPCNT (1U<<3) +#define IBS_CAPS_OPCNT (1U<<4) +#define IBS_CAPS_BRNTRGT (1U<<5) +#define IBS_CAPS_OPCNTEXT (1U<<6) + +#define IBS_CAPS_DEFAULT (IBS_CAPS_AVAIL \ + | IBS_CAPS_FETCHSAM \ + | IBS_CAPS_OPSAM) + +/* + * IBS APIC setup + */ +#define IBSCTL 0x1cc +#define IBSCTL_LVT_OFFSET_VALID (1ULL<<8) +#define IBSCTL_LVT_OFFSET_MASK 0x0F + /* IbsFetchCtl bits/masks */ #define IBS_FETCH_RAND_EN (1ULL<<57) #define IBS_FETCH_VAL (1ULL<<49) @@ -124,6 +159,8 @@ union cpuid10_edx { #define IBS_OP_MAX_CNT 0x0000FFFFULL #define IBS_OP_MAX_CNT_EXT 0x007FFFFFULL /* not a register bit mask */ +extern u32 get_ibs_caps(void); + #ifdef CONFIG_PERF_EVENTS extern void perf_events_lapic_init(void); @@ -159,7 +196,19 @@ extern unsigned long perf_misc_flags(struct pt_regs *regs); ); \ } +struct perf_guest_switch_msr { + unsigned msr; + u64 host, guest; +}; + +extern struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr); #else +static inline perf_guest_switch_msr *perf_guest_get_msrs(int *nr) +{ + *nr = 0; + return NULL; +} + static inline void perf_events_lapic_init(void) { } #endif diff --git a/arch/x86/include/asm/reboot.h b/arch/x86/include/asm/reboot.h index 3250e3d605d9..92f297069e87 100644 --- a/arch/x86/include/asm/reboot.h +++ b/arch/x86/include/asm/reboot.h @@ -23,7 +23,7 @@ void machine_real_restart(unsigned int type); #define MRR_BIOS 0 #define MRR_APM 1 -typedef void (*nmi_shootdown_cb)(int, struct die_args*); +typedef void (*nmi_shootdown_cb)(int, struct pt_regs*); void nmi_shootdown_cpus(nmi_shootdown_cb callback); #endif /* _ASM_X86_REBOOT_H */ diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 7ff4669580cf..c34f96c2f7a0 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -12,6 +12,7 @@ #include <asm/pgtable.h> #include <xen/interface/xen.h> +#include <xen/grant_table.h> #include <xen/features.h> /* Xen machine address */ @@ -48,14 +49,11 @@ extern unsigned long set_phys_range_identity(unsigned long pfn_s, unsigned long pfn_e); extern int m2p_add_override(unsigned long mfn, struct page *page, - bool clear_pte); + struct gnttab_map_grant_ref *kmap_op); extern int m2p_remove_override(struct page *page, bool clear_pte); extern struct page *m2p_find_override(unsigned long mfn); extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn); -#ifdef CONFIG_XEN_DEBUG_FS -extern int p2m_dump_show(struct seq_file *m, void *v); -#endif static inline unsigned long pfn_to_mfn(unsigned long pfn) { unsigned long mfn; diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 82f2912155a5..8baca3c4871c 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -19,7 +19,7 @@ endif obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o -obj-y += time.o ioport.o ldt.o dumpstack.o +obj-y += time.o ioport.o ldt.o dumpstack.o nmi.o obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-y += probe_roms.o diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c index 8a439d364b94..b1e7c7f7a0af 100644 --- a/arch/x86/kernel/amd_gart_64.c +++ b/arch/x86/kernel/amd_gart_64.c @@ -5,7 +5,7 @@ * This allows to use PCI devices that only support 32bit addresses on systems * with more than 4GB. * - * See Documentation/PCI/PCI-DMA-mapping.txt for the interface specification. + * See Documentation/DMA-API-HOWTO.txt for the interface specification. * * Copyright 2002 Andi Kleen, SuSE Labs. * Subject to the GNU General Public License v2 only. diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 52fa56399a50..a2fd72e0ab35 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1437,27 +1437,21 @@ void enable_x2apic(void) int __init enable_IR(void) { -#ifdef CONFIG_INTR_REMAP +#ifdef CONFIG_IRQ_REMAP if (!intr_remapping_supported()) { pr_debug("intr-remapping not supported\n"); - return 0; + return -1; } if (!x2apic_preenabled && skip_ioapic_setup) { pr_info("Skipped enabling intr-remap because of skipping " "io-apic setup\n"); - return 0; + return -1; } - if (enable_intr_remapping(x2apic_supported())) - return 0; - - pr_info("Enabled Interrupt-remapping\n"); - - return 1; - + return enable_intr_remapping(); #endif - return 0; + return -1; } void __init enable_IR_x2apic(void) @@ -1481,11 +1475,11 @@ void __init enable_IR_x2apic(void) mask_ioapic_entries(); if (dmar_table_init_ret) - ret = 0; + ret = -1; else ret = enable_IR(); - if (!ret) { + if (ret < 0) { /* IR is required if there is APIC ID > 255 even when running * under KVM */ @@ -1499,6 +1493,9 @@ void __init enable_IR_x2apic(void) x2apic_force_phys(); } + if (ret == IRQ_REMAP_XAPIC_MODE) + goto nox2apic; + x2apic_enabled = 1; if (x2apic_supported() && !x2apic_mode) { @@ -1508,19 +1505,21 @@ void __init enable_IR_x2apic(void) } nox2apic: - if (!ret) /* IR enabling failed */ + if (ret < 0) /* IR enabling failed */ restore_ioapic_entries(); legacy_pic->restore_mask(); local_irq_restore(flags); out: - if (x2apic_enabled) + if (x2apic_enabled || !x2apic_supported()) return; if (x2apic_preenabled) panic("x2apic: enabled by BIOS but kernel init failed."); - else if (cpu_has_x2apic) - pr_info("Not enabling x2apic, Intr-remapping init failed.\n"); + else if (ret == IRQ_REMAP_XAPIC_MODE) + pr_info("x2apic not enabled, IRQ remapping is in xapic mode\n"); + else if (ret < 0) + pr_info("x2apic not enabled, IRQ remapping init failed\n"); } #ifdef CONFIG_X86_64 diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c index d5e57db0f7be..31cb9ae992b7 100644 --- a/arch/x86/kernel/apic/hw_nmi.c +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -60,22 +60,10 @@ void arch_trigger_all_cpu_backtrace(void) } static int __kprobes -arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self, - unsigned long cmd, void *__args) +arch_trigger_all_cpu_backtrace_handler(unsigned int cmd, struct pt_regs *regs) { - struct die_args *args = __args; - struct pt_regs *regs; int cpu; - switch (cmd) { - case DIE_NMI: - break; - - default: - return NOTIFY_DONE; - } - - regs = args->regs; cpu = smp_processor_id(); if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { @@ -86,21 +74,16 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self, show_regs(regs); arch_spin_unlock(&lock); cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask)); - return NOTIFY_STOP; + return NMI_HANDLED; } - return NOTIFY_DONE; + return NMI_DONE; } -static __read_mostly struct notifier_block backtrace_notifier = { - .notifier_call = arch_trigger_all_cpu_backtrace_handler, - .next = NULL, - .priority = NMI_LOCAL_LOW_PRIOR, -}; - static int __init register_trigger_all_cpu_backtrace(void) { - register_die_notifier(&backtrace_notifier); + register_nmi_handler(NMI_LOCAL, arch_trigger_all_cpu_backtrace_handler, + 0, "arch_bt"); return 0; } early_initcall(register_trigger_all_cpu_backtrace); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 8eb863e27ea6..229e19f3eb57 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -394,13 +394,21 @@ union entry_union { struct IO_APIC_route_entry entry; }; +static struct IO_APIC_route_entry __ioapic_read_entry(int apic, int pin) +{ + union entry_union eu; + + eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); + eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); + return eu.entry; +} + static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) { union entry_union eu; unsigned long flags; raw_spin_lock_irqsave(&ioapic_lock, flags); - eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); - eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); + eu.entry = __ioapic_read_entry(apic, pin); raw_spin_unlock_irqrestore(&ioapic_lock, flags); return eu.entry; } @@ -529,18 +537,6 @@ static void io_apic_modify_irq(struct irq_cfg *cfg, __io_apic_modify_irq(entry, mask_and, mask_or, final); } -static void __mask_and_edge_IO_APIC_irq(struct irq_pin_list *entry) -{ - __io_apic_modify_irq(entry, ~IO_APIC_REDIR_LEVEL_TRIGGER, - IO_APIC_REDIR_MASKED, NULL); -} - -static void __unmask_and_level_IO_APIC_irq(struct irq_pin_list *entry) -{ - __io_apic_modify_irq(entry, ~IO_APIC_REDIR_MASKED, - IO_APIC_REDIR_LEVEL_TRIGGER, NULL); -} - static void io_apic_sync(struct irq_pin_list *entry) { /* @@ -585,6 +581,66 @@ static void unmask_ioapic_irq(struct irq_data *data) unmask_ioapic(data->chip_data); } +/* + * IO-APIC versions below 0x20 don't support EOI register. + * For the record, here is the information about various versions: + * 0Xh 82489DX + * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant + * 2Xh I/O(x)APIC which is PCI 2.2 Compliant + * 30h-FFh Reserved + * + * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic + * version as 0x2. This is an error with documentation and these ICH chips + * use io-apic's of version 0x20. + * + * For IO-APIC's with EOI register, we use that to do an explicit EOI. + * Otherwise, we simulate the EOI message manually by changing the trigger + * mode to edge and then back to level, with RTE being masked during this. + */ +static void __eoi_ioapic_pin(int apic, int pin, int vector, struct irq_cfg *cfg) +{ + if (mpc_ioapic_ver(apic) >= 0x20) { + /* + * Intr-remapping uses pin number as the virtual vector + * in the RTE. Actual vector is programmed in + * intr-remapping table entry. Hence for the io-apic + * EOI we use the pin number. + */ + if (cfg && irq_remapped(cfg)) + io_apic_eoi(apic, pin); + else + io_apic_eoi(apic, vector); + } else { + struct IO_APIC_route_entry entry, entry1; + + entry = entry1 = __ioapic_read_entry(apic, pin); + + /* + * Mask the entry and change the trigger mode to edge. + */ + entry1.mask = 1; + entry1.trigger = IOAPIC_EDGE; + + __ioapic_write_entry(apic, pin, entry1); + + /* + * Restore the previous level triggered entry. + */ + __ioapic_write_entry(apic, pin, entry); + } +} + +static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) +{ + struct irq_pin_list *entry; + unsigned long flags; + + raw_spin_lock_irqsave(&ioapic_lock, flags); + for_each_irq_pin(entry, cfg->irq_2_pin) + __eoi_ioapic_pin(entry->apic, entry->pin, cfg->vector, cfg); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); +} + static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) { struct IO_APIC_route_entry entry; @@ -593,10 +649,44 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) entry = ioapic_read_entry(apic, pin); if (entry.delivery_mode == dest_SMI) return; + + /* + * Make sure the entry is masked and re-read the contents to check + * if it is a level triggered pin and if the remote-IRR is set. + */ + if (!entry.mask) { + entry.mask = 1; + ioapic_write_entry(apic, pin, entry); + entry = ioapic_read_entry(apic, pin); + } + + if (entry.irr) { + unsigned long flags; + + /* + * Make sure the trigger mode is set to level. Explicit EOI + * doesn't clear the remote-IRR if the trigger mode is not + * set to level. + */ + if (!entry.trigger) { + entry.trigger = IOAPIC_LEVEL; + ioapic_write_entry(apic, pin, entry); + } + + raw_spin_lock_irqsave(&ioapic_lock, flags); + __eoi_ioapic_pin(apic, pin, entry.vector, NULL); + raw_spin_unlock_irqrestore(&ioapic_lock, flags); + } + /* - * Disable it in the IO-APIC irq-routing table: + * Clear the rest of the bits in the IO-APIC RTE except for the mask + * bit. */ ioapic_mask_entry(apic, pin); + entry = ioapic_read_entry(apic, pin); + if (entry.irr) + printk(KERN_ERR "Unable to reset IRR for apic: %d, pin :%d\n", + mpc_ioapic_id(apic), pin); } static void clear_IO_APIC (void) @@ -1202,7 +1292,6 @@ void __setup_vector_irq(int cpu) } static struct irq_chip ioapic_chip; -static struct irq_chip ir_ioapic_chip; #ifdef CONFIG_X86_32 static inline int IO_APIC_irq_trigger(int irq) @@ -1246,7 +1335,7 @@ static void ioapic_register_intr(unsigned int irq, struct irq_cfg *cfg, if (irq_remapped(cfg)) { irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); - chip = &ir_ioapic_chip; + irq_remap_modify_chip_defaults(chip); fasteoi = trigger != 0; } @@ -2255,7 +2344,7 @@ ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, return ret; } -#ifdef CONFIG_INTR_REMAP +#ifdef CONFIG_IRQ_REMAP /* * Migrate the IO-APIC irq in the presence of intr-remapping. @@ -2267,6 +2356,9 @@ ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, * updated vector information), by using a virtual vector (io-apic pin number). * Real vector that is used for interrupting cpu will be coming from * the interrupt-remapping table entry. + * + * As the migration is a simple atomic update of IRTE, the same mechanism + * is used to migrate MSI irq's in the presence of interrupt-remapping. */ static int ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, @@ -2291,10 +2383,16 @@ ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask, irte.dest_id = IRTE_DEST(dest); /* - * Modified the IRTE and flushes the Interrupt entry cache. + * Atomically updates the IRTE with the new destination, vector + * and flushes the interrupt entry cache. */ modify_irte(irq, &irte); + /* + * After this point, all the interrupts will start arriving + * at the new destination. So, time to cleanup the previous + * vector allocation. + */ if (cfg->move_in_progress) send_cleanup_vector(cfg); @@ -2407,48 +2505,6 @@ static void ack_apic_edge(struct irq_data *data) atomic_t irq_mis_count; -/* - * IO-APIC versions below 0x20 don't support EOI register. - * For the record, here is the information about various versions: - * 0Xh 82489DX - * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant - * 2Xh I/O(x)APIC which is PCI 2.2 Compliant - * 30h-FFh Reserved - * - * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic - * version as 0x2. This is an error with documentation and these ICH chips - * use io-apic's of version 0x20. - * - * For IO-APIC's with EOI register, we use that to do an explicit EOI. - * Otherwise, we simulate the EOI message manually by changing the trigger - * mode to edge and then back to level, with RTE being masked during this. -*/ -static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) -{ - struct irq_pin_list *entry; - unsigned long flags; - - raw_spin_lock_irqsave(&ioapic_lock, flags); - for_each_irq_pin(entry, cfg->irq_2_pin) { - if (mpc_ioapic_ver(entry->apic) >= 0x20) { - /* - * Intr-remapping uses pin number as the virtual vector - * in the RTE. Actual vector is programmed in - * intr-remapping table entry. Hence for the io-apic - * EOI we use the pin number. - */ - if (irq_remapped(cfg)) - io_apic_eoi(entry->apic, entry->pin); - else - io_apic_eoi(entry->apic, cfg->vector); - } else { - __mask_and_edge_IO_APIC_irq(entry); - __unmask_and_level_IO_APIC_irq(entry); - } - } - raw_spin_unlock_irqrestore(&ioapic_lock, flags); -} - static void ack_apic_level(struct irq_data *data) { struct irq_cfg *cfg = data->chip_data; @@ -2552,7 +2608,7 @@ static void ack_apic_level(struct irq_data *data) } } -#ifdef CONFIG_INTR_REMAP +#ifdef CONFIG_IRQ_REMAP static void ir_ack_apic_edge(struct irq_data *data) { ack_APIC_irq(); @@ -2563,7 +2619,23 @@ static void ir_ack_apic_level(struct irq_data *data) ack_APIC_irq(); eoi_ioapic_irq(data->irq, data->chip_data); } -#endif /* CONFIG_INTR_REMAP */ + +static void ir_print_prefix(struct irq_data *data, struct seq_file *p) +{ + seq_printf(p, " IR-%s", data->chip->name); +} + +static void irq_remap_modify_chip_defaults(struct irq_chip *chip) +{ + chip->irq_print_chip = ir_print_prefix; + chip->irq_ack = ir_ack_apic_edge; + chip->irq_eoi = ir_ack_apic_level; + +#ifdef CONFIG_SMP + chip->irq_set_affinity = ir_ioapic_set_affinity; +#endif +} +#endif /* CONFIG_IRQ_REMAP */ static struct irq_chip ioapic_chip __read_mostly = { .name = "IO-APIC", @@ -2578,21 +2650,6 @@ static struct irq_chip ioapic_chip __read_mostly = { .irq_retrigger = ioapic_retrigger_irq, }; -static struct irq_chip ir_ioapic_chip __read_mostly = { - .name = "IR-IO-APIC", - .irq_startup = startup_ioapic_irq, - .irq_mask = mask_ioapic_irq, - .irq_unmask = unmask_ioapic_irq, -#ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, - .irq_eoi = ir_ack_apic_level, -#ifdef CONFIG_SMP - .irq_set_affinity = ir_ioapic_set_affinity, -#endif -#endif - .irq_retrigger = ioapic_retrigger_irq, -}; - static inline void init_IO_APIC_traps(void) { struct irq_cfg *cfg; @@ -3144,45 +3201,6 @@ msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) return 0; } -#ifdef CONFIG_INTR_REMAP -/* - * Migrate the MSI irq to another cpumask. This migration is - * done in the process context using interrupt-remapping hardware. - */ -static int -ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, - bool force) -{ - struct irq_cfg *cfg = data->chip_data; - unsigned int dest, irq = data->irq; - struct irte irte; - - if (get_irte(irq, &irte)) - return -1; - - if (__ioapic_set_affinity(data, mask, &dest)) - return -1; - - irte.vector = cfg->vector; - irte.dest_id = IRTE_DEST(dest); - - /* - * atomically update the IRTE with the new destination and vector. - */ - modify_irte(irq, &irte); - - /* - * After this point, all the interrupts will start arriving - * at the new destination. So, time to cleanup the previous - * vector allocation. - */ - if (cfg->move_in_progress) - send_cleanup_vector(cfg); - - return 0; -} - -#endif #endif /* CONFIG_SMP */ /* @@ -3200,19 +3218,6 @@ static struct irq_chip msi_chip = { .irq_retrigger = ioapic_retrigger_irq, }; -static struct irq_chip msi_ir_chip = { - .name = "IR-PCI-MSI", - .irq_unmask = unmask_msi_irq, - .irq_mask = mask_msi_irq, -#ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, -#ifdef CONFIG_SMP - .irq_set_affinity = ir_msi_set_affinity, -#endif -#endif - .irq_retrigger = ioapic_retrigger_irq, -}; - /* * Map the PCI dev to the corresponding remapping hardware unit * and allocate 'nvec' consecutive interrupt-remapping table entries @@ -3255,7 +3260,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) if (irq_remapped(irq_get_chip_data(irq))) { irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); - chip = &msi_ir_chip; + irq_remap_modify_chip_defaults(chip); } irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge"); @@ -3328,7 +3333,7 @@ void native_teardown_msi_irq(unsigned int irq) destroy_irq(irq); } -#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP) +#ifdef CONFIG_DMAR_TABLE #ifdef CONFIG_SMP static int dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask, @@ -3409,19 +3414,6 @@ static int hpet_msi_set_affinity(struct irq_data *data, #endif /* CONFIG_SMP */ -static struct irq_chip ir_hpet_msi_type = { - .name = "IR-HPET_MSI", - .irq_unmask = hpet_msi_unmask, - .irq_mask = hpet_msi_mask, -#ifdef CONFIG_INTR_REMAP - .irq_ack = ir_ack_apic_edge, -#ifdef CONFIG_SMP - .irq_set_affinity = ir_msi_set_affinity, -#endif -#endif - .irq_retrigger = ioapic_retrigger_irq, -}; - static struct irq_chip hpet_msi_type = { .name = "HPET_MSI", .irq_unmask = hpet_msi_unmask, @@ -3458,7 +3450,7 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id) hpet_msi_write(irq_get_handler_data(irq), &msg); irq_set_status_flags(irq, IRQ_MOVE_PCNTXT); if (irq_remapped(irq_get_chip_data(irq))) - chip = &ir_hpet_msi_type; + irq_remap_modify_chip_defaults(chip); irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge"); return 0; diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 34b18594e724..75be00ecfff2 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -672,18 +672,11 @@ void __cpuinit uv_cpu_init(void) /* * When NMI is received, print a stack trace. */ -int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) +int uv_handle_nmi(unsigned int reason, struct pt_regs *regs) { unsigned long real_uv_nmi; int bid; - if (reason != DIE_NMIUNKNOWN) - return NOTIFY_OK; - - if (in_crash_kexec) - /* do nothing if entering the crash kernel */ - return NOTIFY_OK; - /* * Each blade has an MMR that indicates when an NMI has been sent * to cpus on the blade. If an NMI is detected, atomically @@ -704,7 +697,7 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) } if (likely(__get_cpu_var(cpu_last_nmi_count) == uv_blade_info[bid].nmi_count)) - return NOTIFY_DONE; + return NMI_DONE; __get_cpu_var(cpu_last_nmi_count) = uv_blade_info[bid].nmi_count; @@ -717,17 +710,12 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) dump_stack(); spin_unlock(&uv_nmi_lock); - return NOTIFY_STOP; + return NMI_HANDLED; } -static struct notifier_block uv_dump_stack_nmi_nb = { - .notifier_call = uv_handle_nmi, - .priority = NMI_LOCAL_LOW_PRIOR - 1, -}; - void uv_register_nmi_notifier(void) { - if (register_die_notifier(&uv_dump_stack_nmi_nb)) + if (register_nmi_handler(NMI_UNKNOWN, uv_handle_nmi, 0, "uv")) printk(KERN_WARNING "UV NMI handler failed to register\n"); } diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 0371c484bb8a..a46bd383953c 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -249,8 +249,6 @@ extern int (*console_blank_hook)(int); #define APM_MINOR_DEV 134 /* - * See Documentation/Config.help for the configuration options. - * * Various options can be changed at boot time as follows: * (We allow underscores for compatibility with the modules code) * apm=on/off enable/disable APM diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 6042981d0309..fe6eb197f848 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -28,10 +28,15 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o +ifdef CONFIG_PERF_EVENTS +obj-$(CONFIG_CPU_SUP_AMD) += perf_event_amd.o +obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_p4.o perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o +endif + obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ -obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o +obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o perf_event_amd_ibs.o quiet_cmd_mkcapflags = MKCAP $@ cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@ diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c index 0ed633c5048b..6199232161cf 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-inject.c +++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c @@ -78,27 +78,20 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs) static cpumask_var_t mce_inject_cpumask; -static int mce_raise_notify(struct notifier_block *self, - unsigned long val, void *data) +static int mce_raise_notify(unsigned int cmd, struct pt_regs *regs) { - struct die_args *args = (struct die_args *)data; int cpu = smp_processor_id(); struct mce *m = &__get_cpu_var(injectm); - if (val != DIE_NMI || !cpumask_test_cpu(cpu, mce_inject_cpumask)) - return NOTIFY_DONE; + if (!cpumask_test_cpu(cpu, mce_inject_cpumask)) + return NMI_DONE; cpumask_clear_cpu(cpu, mce_inject_cpumask); if (m->inject_flags & MCJ_EXCEPTION) - raise_exception(m, args->regs); + raise_exception(m, regs); else if (m->status) raise_poll(m); - return NOTIFY_STOP; + return NMI_HANDLED; } -static struct notifier_block mce_raise_nb = { - .notifier_call = mce_raise_notify, - .priority = NMI_LOCAL_NORMAL_PRIOR, -}; - /* Inject mce on current CPU */ static int raise_local(void) { @@ -216,7 +209,8 @@ static int inject_init(void) return -ENOMEM; printk(KERN_INFO "Machine check injector initialized\n"); mce_chrdev_ops.write = mce_write; - register_die_notifier(&mce_raise_nb); + register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0, + "mce_notify"); return 0; } diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 08363b042122..fce51ad1f362 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -908,9 +908,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) percpu_inc(mce_exception_count); - if (notify_die(DIE_NMI, "machine check", regs, error_code, - 18, SIGKILL) == NOTIFY_STOP) - goto out; if (!banks) goto out; @@ -1140,6 +1137,15 @@ static void mce_start_timer(unsigned long data) add_timer_on(t, smp_processor_id()); } +/* Must not be called in IRQ context where del_timer_sync() can deadlock */ +static void mce_timer_delete_all(void) +{ + int cpu; + + for_each_online_cpu(cpu) + del_timer_sync(&per_cpu(mce_timer, cpu)); +} + static void mce_do_trigger(struct work_struct *work) { call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT); @@ -1750,7 +1756,6 @@ static struct syscore_ops mce_syscore_ops = { static void mce_cpu_restart(void *data) { - del_timer_sync(&__get_cpu_var(mce_timer)); if (!mce_available(__this_cpu_ptr(&cpu_info))) return; __mcheck_cpu_init_generic(); @@ -1760,16 +1765,15 @@ static void mce_cpu_restart(void *data) /* Reinit MCEs after user configuration changes */ static void mce_restart(void) { + mce_timer_delete_all(); on_each_cpu(mce_cpu_restart, NULL, 1); } /* Toggle features for corrected errors */ -static void mce_disable_ce(void *all) +static void mce_disable_cmci(void *data) { if (!mce_available(__this_cpu_ptr(&cpu_info))) return; - if (all) - del_timer_sync(&__get_cpu_var(mce_timer)); cmci_clear(); } @@ -1852,7 +1856,8 @@ static ssize_t set_ignore_ce(struct sys_device *s, if (mce_ignore_ce ^ !!new) { if (new) { /* disable ce features */ - on_each_cpu(mce_disable_ce, (void *)1, 1); + mce_timer_delete_all(); + on_each_cpu(mce_disable_cmci, NULL, 1); mce_ignore_ce = 1; } else { /* enable ce features */ @@ -1875,7 +1880,7 @@ static ssize_t set_cmci_disabled(struct sys_device *s, if (mce_cmci_disabled ^ !!new) { if (new) { /* disable cmci */ - on_each_cpu(mce_disable_ce, NULL, 1); + on_each_cpu(mce_disable_cmci, NULL, 1); mce_cmci_disabled = 1; } else { /* enable cmci */ diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index 8694ef56459d..38e49bc95ffc 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -28,7 +28,7 @@ static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); * cmci_discover_lock protects against parallel discovery attempts * which could race against each other. */ -static DEFINE_SPINLOCK(cmci_discover_lock); +static DEFINE_RAW_SPINLOCK(cmci_discover_lock); #define CMCI_THRESHOLD 1 @@ -85,7 +85,7 @@ static void cmci_discover(int banks, int boot) int hdr = 0; int i; - spin_lock_irqsave(&cmci_discover_lock, flags); + raw_spin_lock_irqsave(&cmci_discover_lock, flags); for (i = 0; i < banks; i++) { u64 val; @@ -116,7 +116,7 @@ static void cmci_discover(int banks, int boot) WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); } } - spin_unlock_irqrestore(&cmci_discover_lock, flags); + raw_spin_unlock_irqrestore(&cmci_discover_lock, flags); if (hdr) printk(KERN_CONT "\n"); } @@ -150,7 +150,7 @@ void cmci_clear(void) if (!cmci_supported(&banks)) return; - spin_lock_irqsave(&cmci_discover_lock, flags); + raw_spin_lock_irqsave(&cmci_discover_lock, flags); for (i = 0; i < banks; i++) { if (!test_bit(i, __get_cpu_var(mce_banks_owned))) continue; @@ -160,7 +160,7 @@ void cmci_clear(void) wrmsrl(MSR_IA32_MCx_CTL2(i), val); __clear_bit(i, __get_cpu_var(mce_banks_owned)); } - spin_unlock_irqrestore(&cmci_discover_lock, flags); + raw_spin_unlock_irqrestore(&cmci_discover_lock, flags); } /* diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index cfa62ec090ec..640891014b2a 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -32,6 +32,8 @@ #include <asm/smp.h> #include <asm/alternative.h> +#include "perf_event.h" + #if 0 #undef wrmsrl #define wrmsrl(msr, val) \ @@ -43,283 +45,17 @@ do { \ } while (0) #endif -/* - * | NHM/WSM | SNB | - * register ------------------------------- - * | HT | no HT | HT | no HT | - *----------------------------------------- - * offcore | core | core | cpu | core | - * lbr_sel | core | core | cpu | core | - * ld_lat | cpu | core | cpu | core | - *----------------------------------------- - * - * Given that there is a small number of shared regs, - * we can pre-allocate their slot in the per-cpu - * per-core reg tables. - */ -enum extra_reg_type { - EXTRA_REG_NONE = -1, /* not used */ - - EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */ - EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ - - EXTRA_REG_MAX /* number of entries needed */ -}; - -struct event_constraint { - union { - unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - u64 idxmsk64; - }; - u64 code; - u64 cmask; - int weight; -}; - -struct amd_nb { - int nb_id; /* NorthBridge id */ - int refcnt; /* reference count */ - struct perf_event *owners[X86_PMC_IDX_MAX]; - struct event_constraint event_constraints[X86_PMC_IDX_MAX]; -}; - -struct intel_percore; - -#define MAX_LBR_ENTRIES 16 - -struct cpu_hw_events { - /* - * Generic x86 PMC bits - */ - struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ - unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - int enabled; - - int n_events; - int n_added; - int n_txn; - int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ - u64 tags[X86_PMC_IDX_MAX]; - struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ - - unsigned int group_flag; - - /* - * Intel DebugStore bits - */ - struct debug_store *ds; - u64 pebs_enabled; - - /* - * Intel LBR bits - */ - int lbr_users; - void *lbr_context; - struct perf_branch_stack lbr_stack; - struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; - - /* - * manage shared (per-core, per-cpu) registers - * used on Intel NHM/WSM/SNB - */ - struct intel_shared_regs *shared_regs; - - /* - * AMD specific bits - */ - struct amd_nb *amd_nb; -}; - -#define __EVENT_CONSTRAINT(c, n, m, w) {\ - { .idxmsk64 = (n) }, \ - .code = (c), \ - .cmask = (m), \ - .weight = (w), \ -} - -#define EVENT_CONSTRAINT(c, n, m) \ - __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) - -/* - * Constraint on the Event code. - */ -#define INTEL_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT) - -/* - * Constraint on the Event code + UMask + fixed-mask - * - * filter mask to validate fixed counter events. - * the following filters disqualify for fixed counters: - * - inv - * - edge - * - cnt-mask - * The other filters are supported by fixed counters. - * The any-thread option is supported starting with v3. - */ -#define FIXED_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK) - -/* - * Constraint on the Event code + UMask - */ -#define INTEL_UEVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK) - -#define EVENT_CONSTRAINT_END \ - EVENT_CONSTRAINT(0, 0, 0) - -#define for_each_event_constraint(e, c) \ - for ((e) = (c); (e)->weight; (e)++) - -/* - * Per register state. - */ -struct er_account { - raw_spinlock_t lock; /* per-core: protect structure */ - u64 config; /* extra MSR config */ - u64 reg; /* extra MSR number */ - atomic_t ref; /* reference count */ -}; - -/* - * Extra registers for specific events. - * - * Some events need large masks and require external MSRs. - * Those extra MSRs end up being shared for all events on - * a PMU and sometimes between PMU of sibling HT threads. - * In either case, the kernel needs to handle conflicting - * accesses to those extra, shared, regs. The data structure - * to manage those registers is stored in cpu_hw_event. - */ -struct extra_reg { - unsigned int event; - unsigned int msr; - u64 config_mask; - u64 valid_mask; - int idx; /* per_xxx->regs[] reg index */ -}; - -#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \ - .event = (e), \ - .msr = (ms), \ - .config_mask = (m), \ - .valid_mask = (vm), \ - .idx = EXTRA_REG_##i \ - } - -#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \ - EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx) - -#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0) - -union perf_capabilities { - struct { - u64 lbr_format : 6; - u64 pebs_trap : 1; - u64 pebs_arch_reg : 1; - u64 pebs_format : 4; - u64 smm_freeze : 1; - }; - u64 capabilities; -}; - -/* - * struct x86_pmu - generic x86 pmu - */ -struct x86_pmu { - /* - * Generic x86 PMC bits - */ - const char *name; - int version; - int (*handle_irq)(struct pt_regs *); - void (*disable_all)(void); - void (*enable_all)(int added); - void (*enable)(struct perf_event *); - void (*disable)(struct perf_event *); - int (*hw_config)(struct perf_event *event); - int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); - unsigned eventsel; - unsigned perfctr; - u64 (*event_map)(int); - int max_events; - int num_counters; - int num_counters_fixed; - int cntval_bits; - u64 cntval_mask; - int apic; - u64 max_period; - struct event_constraint * - (*get_event_constraints)(struct cpu_hw_events *cpuc, - struct perf_event *event); - - void (*put_event_constraints)(struct cpu_hw_events *cpuc, - struct perf_event *event); - struct event_constraint *event_constraints; - void (*quirks)(void); - int perfctr_second_write; - - int (*cpu_prepare)(int cpu); - void (*cpu_starting)(int cpu); - void (*cpu_dying)(int cpu); - void (*cpu_dead)(int cpu); - - /* - * Intel Arch Perfmon v2+ - */ - u64 intel_ctrl; - union perf_capabilities intel_cap; +struct x86_pmu x86_pmu __read_mostly; - /* - * Intel DebugStore bits - */ - int bts, pebs; - int bts_active, pebs_active; - int pebs_record_size; - void (*drain_pebs)(struct pt_regs *regs); - struct event_constraint *pebs_constraints; - - /* - * Intel LBR - */ - unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */ - int lbr_nr; /* hardware stack size */ - - /* - * Extra registers for events - */ - struct extra_reg *extra_regs; - unsigned int er_flags; -}; - -#define ERF_NO_HT_SHARING 1 -#define ERF_HAS_RSP_1 2 - -static struct x86_pmu x86_pmu __read_mostly; - -static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { +DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; -static int x86_perf_event_set_period(struct perf_event *event); - -/* - * Generalized hw caching related hw_event table, filled - * in on a per model basis. A value of 0 means - * 'not supported', -1 means 'hw_event makes no sense on - * this CPU', any other value means the raw hw_event - * ID. - */ - -#define C(x) PERF_COUNT_HW_CACHE_##x - -static u64 __read_mostly hw_cache_event_ids +u64 __read_mostly hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX]; -static u64 __read_mostly hw_cache_extra_regs +u64 __read_mostly hw_cache_extra_regs [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX]; @@ -329,8 +65,7 @@ static u64 __read_mostly hw_cache_extra_regs * Can only be executed on the CPU where the event is active. * Returns the delta events processed. */ -static u64 -x86_perf_event_update(struct perf_event *event) +u64 x86_perf_event_update(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; int shift = 64 - x86_pmu.cntval_bits; @@ -373,30 +108,6 @@ again: return new_raw_count; } -static inline int x86_pmu_addr_offset(int index) -{ - int offset; - - /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */ - alternative_io(ASM_NOP2, - "shll $1, %%eax", - X86_FEATURE_PERFCTR_CORE, - "=a" (offset), - "a" (index)); - - return offset; -} - -static inline unsigned int x86_pmu_config_addr(int index) -{ - return x86_pmu.eventsel + x86_pmu_addr_offset(index); -} - -static inline unsigned int x86_pmu_event_addr(int index) -{ - return x86_pmu.perfctr + x86_pmu_addr_offset(index); -} - /* * Find and validate any extra registers to set up. */ @@ -532,9 +243,6 @@ msr_fail: return false; } -static void reserve_ds_buffers(void); -static void release_ds_buffers(void); - static void hw_perf_event_destroy(struct perf_event *event) { if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) { @@ -583,7 +291,7 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event) return x86_pmu_extra_regs(val, event); } -static int x86_setup_perfctr(struct perf_event *event) +int x86_setup_perfctr(struct perf_event *event) { struct perf_event_attr *attr = &event->attr; struct hw_perf_event *hwc = &event->hw; @@ -647,7 +355,7 @@ static int x86_setup_perfctr(struct perf_event *event) return 0; } -static int x86_pmu_hw_config(struct perf_event *event) +int x86_pmu_hw_config(struct perf_event *event) { if (event->attr.precise_ip) { int precise = 0; @@ -723,7 +431,7 @@ static int __x86_pmu_event_init(struct perf_event *event) return x86_pmu.hw_config(event); } -static void x86_pmu_disable_all(void) +void x86_pmu_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; @@ -758,15 +466,7 @@ static void x86_pmu_disable(struct pmu *pmu) x86_pmu.disable_all(); } -static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc, - u64 enable_mask) -{ - if (hwc->extra_reg.reg) - wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config); - wrmsrl(hwc->config_base, hwc->config | enable_mask); -} - -static void x86_pmu_enable_all(int added) +void x86_pmu_enable_all(int added) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); int idx; @@ -788,7 +488,7 @@ static inline int is_x86_event(struct perf_event *event) return event->pmu == &pmu; } -static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) +int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) { struct event_constraint *c, *constraints[X86_PMC_IDX_MAX]; unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; @@ -959,7 +659,6 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc, } static void x86_pmu_start(struct perf_event *event, int flags); -static void x86_pmu_stop(struct perf_event *event, int flags); static void x86_pmu_enable(struct pmu *pmu) { @@ -1031,21 +730,13 @@ static void x86_pmu_enable(struct pmu *pmu) x86_pmu.enable_all(added); } -static inline void x86_pmu_disable_event(struct perf_event *event) -{ - struct hw_perf_event *hwc = &event->hw; - - wrmsrl(hwc->config_base, hwc->config); -} - static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left); /* * Set the next IRQ period, based on the hwc->period_left value. * To be called with the event disabled in hw: */ -static int -x86_perf_event_set_period(struct perf_event *event) +int x86_perf_event_set_period(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; s64 left = local64_read(&hwc->period_left); @@ -1105,7 +796,7 @@ x86_perf_event_set_period(struct perf_event *event) return ret; } -static void x86_pmu_enable_event(struct perf_event *event) +void x86_pmu_enable_event(struct perf_event *event) { if (__this_cpu_read(cpu_hw_events.enabled)) __x86_pmu_enable_event(&event->hw, @@ -1244,7 +935,7 @@ void perf_event_print_debug(void) local_irq_restore(flags); } -static void x86_pmu_stop(struct perf_event *event, int flags) +void x86_pmu_stop(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -1297,7 +988,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) perf_event_update_userpage(event); } -static int x86_pmu_handle_irq(struct pt_regs *regs) +int x86_pmu_handle_irq(struct pt_regs *regs) { struct perf_sample_data data; struct cpu_hw_events *cpuc; @@ -1367,109 +1058,28 @@ void perf_events_lapic_init(void) apic_write(APIC_LVTPC, APIC_DM_NMI); } -struct pmu_nmi_state { - unsigned int marked; - int handled; -}; - -static DEFINE_PER_CPU(struct pmu_nmi_state, pmu_nmi); - static int __kprobes -perf_event_nmi_handler(struct notifier_block *self, - unsigned long cmd, void *__args) +perf_event_nmi_handler(unsigned int cmd, struct pt_regs *regs) { - struct die_args *args = __args; - unsigned int this_nmi; - int handled; - if (!atomic_read(&active_events)) - return NOTIFY_DONE; - - switch (cmd) { - case DIE_NMI: - break; - case DIE_NMIUNKNOWN: - this_nmi = percpu_read(irq_stat.__nmi_count); - if (this_nmi != __this_cpu_read(pmu_nmi.marked)) - /* let the kernel handle the unknown nmi */ - return NOTIFY_DONE; - /* - * This one is a PMU back-to-back nmi. Two events - * trigger 'simultaneously' raising two back-to-back - * NMIs. If the first NMI handles both, the latter - * will be empty and daze the CPU. So, we drop it to - * avoid false-positive 'unknown nmi' messages. - */ - return NOTIFY_STOP; - default: - return NOTIFY_DONE; - } - - handled = x86_pmu.handle_irq(args->regs); - if (!handled) - return NOTIFY_DONE; - - this_nmi = percpu_read(irq_stat.__nmi_count); - if ((handled > 1) || - /* the next nmi could be a back-to-back nmi */ - ((__this_cpu_read(pmu_nmi.marked) == this_nmi) && - (__this_cpu_read(pmu_nmi.handled) > 1))) { - /* - * We could have two subsequent back-to-back nmis: The - * first handles more than one counter, the 2nd - * handles only one counter and the 3rd handles no - * counter. - * - * This is the 2nd nmi because the previous was - * handling more than one counter. We will mark the - * next (3rd) and then drop it if unhandled. - */ - __this_cpu_write(pmu_nmi.marked, this_nmi + 1); - __this_cpu_write(pmu_nmi.handled, handled); - } + return NMI_DONE; - return NOTIFY_STOP; + return x86_pmu.handle_irq(regs); } -static __read_mostly struct notifier_block perf_event_nmi_notifier = { - .notifier_call = perf_event_nmi_handler, - .next = NULL, - .priority = NMI_LOCAL_LOW_PRIOR, -}; - -static struct event_constraint unconstrained; -static struct event_constraint emptyconstraint; - -static struct event_constraint * -x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) -{ - struct event_constraint *c; - - if (x86_pmu.event_constraints) { - for_each_event_constraint(c, x86_pmu.event_constraints) { - if ((event->hw.config & c->cmask) == c->code) - return c; - } - } - - return &unconstrained; -} - -#include "perf_event_amd.c" -#include "perf_event_p6.c" -#include "perf_event_p4.c" -#include "perf_event_intel_lbr.c" -#include "perf_event_intel_ds.c" -#include "perf_event_intel.c" +struct event_constraint emptyconstraint; +struct event_constraint unconstrained; static int __cpuinit x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; + struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu); int ret = NOTIFY_OK; switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: + cpuc->kfree_on_online = NULL; if (x86_pmu.cpu_prepare) ret = x86_pmu.cpu_prepare(cpu); break; @@ -1479,6 +1089,10 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) x86_pmu.cpu_starting(cpu); break; + case CPU_ONLINE: + kfree(cpuc->kfree_on_online); + break; + case CPU_DYING: if (x86_pmu.cpu_dying) x86_pmu.cpu_dying(cpu); @@ -1557,7 +1171,7 @@ static int __init init_hw_perf_events(void) ((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED; perf_events_lapic_init(); - register_die_notifier(&perf_event_nmi_notifier); + register_nmi_handler(NMI_LOCAL, perf_event_nmi_handler, 0, "PMI"); unconstrained = (struct event_constraint) __EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1, diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h new file mode 100644 index 000000000000..b9698d40ac4b --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event.h @@ -0,0 +1,505 @@ +/* + * Performance events x86 architecture header + * + * Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2009 Jaswinder Singh Rajput + * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter + * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> + * Copyright (C) 2009 Google, Inc., Stephane Eranian + * + * For licencing details see kernel-base/COPYING + */ + +#include <linux/perf_event.h> + +/* + * | NHM/WSM | SNB | + * register ------------------------------- + * | HT | no HT | HT | no HT | + *----------------------------------------- + * offcore | core | core | cpu | core | + * lbr_sel | core | core | cpu | core | + * ld_lat | cpu | core | cpu | core | + *----------------------------------------- + * + * Given that there is a small number of shared regs, + * we can pre-allocate their slot in the per-cpu + * per-core reg tables. + */ +enum extra_reg_type { + EXTRA_REG_NONE = -1, /* not used */ + + EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */ + EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ + + EXTRA_REG_MAX /* number of entries needed */ +}; + +struct event_constraint { + union { + unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + u64 idxmsk64; + }; + u64 code; + u64 cmask; + int weight; +}; + +struct amd_nb { + int nb_id; /* NorthBridge id */ + int refcnt; /* reference count */ + struct perf_event *owners[X86_PMC_IDX_MAX]; + struct event_constraint event_constraints[X86_PMC_IDX_MAX]; +}; + +/* The maximal number of PEBS events: */ +#define MAX_PEBS_EVENTS 4 + +/* + * A debug store configuration. + * + * We only support architectures that use 64bit fields. + */ +struct debug_store { + u64 bts_buffer_base; + u64 bts_index; + u64 bts_absolute_maximum; + u64 bts_interrupt_threshold; + u64 pebs_buffer_base; + u64 pebs_index; + u64 pebs_absolute_maximum; + u64 pebs_interrupt_threshold; + u64 pebs_event_reset[MAX_PEBS_EVENTS]; +}; + +/* + * Per register state. + */ +struct er_account { + raw_spinlock_t lock; /* per-core: protect structure */ + u64 config; /* extra MSR config */ + u64 reg; /* extra MSR number */ + atomic_t ref; /* reference count */ +}; + +/* + * Per core/cpu state + * + * Used to coordinate shared registers between HT threads or + * among events on a single PMU. + */ +struct intel_shared_regs { + struct er_account regs[EXTRA_REG_MAX]; + int refcnt; /* per-core: #HT threads */ + unsigned core_id; /* per-core: core id */ +}; + +#define MAX_LBR_ENTRIES 16 + +struct cpu_hw_events { + /* + * Generic x86 PMC bits + */ + struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */ + unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; + int enabled; + + int n_events; + int n_added; + int n_txn; + int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */ + u64 tags[X86_PMC_IDX_MAX]; + struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */ + + unsigned int group_flag; + + /* + * Intel DebugStore bits + */ + struct debug_store *ds; + u64 pebs_enabled; + + /* + * Intel LBR bits + */ + int lbr_users; + void *lbr_context; + struct perf_branch_stack lbr_stack; + struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES]; + + /* + * Intel host/guest exclude bits + */ + u64 intel_ctrl_guest_mask; + u64 intel_ctrl_host_mask; + struct perf_guest_switch_msr guest_switch_msrs[X86_PMC_IDX_MAX]; + + /* + * manage shared (per-core, per-cpu) registers + * used on Intel NHM/WSM/SNB + */ + struct intel_shared_regs *shared_regs; + + /* + * AMD specific bits + */ + struct amd_nb *amd_nb; + + void *kfree_on_online; +}; + +#define __EVENT_CONSTRAINT(c, n, m, w) {\ + { .idxmsk64 = (n) }, \ + .code = (c), \ + .cmask = (m), \ + .weight = (w), \ +} + +#define EVENT_CONSTRAINT(c, n, m) \ + __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n)) + +/* + * Constraint on the Event code. + */ +#define INTEL_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT) + +/* + * Constraint on the Event code + UMask + fixed-mask + * + * filter mask to validate fixed counter events. + * the following filters disqualify for fixed counters: + * - inv + * - edge + * - cnt-mask + * The other filters are supported by fixed counters. + * The any-thread option is supported starting with v3. + */ +#define FIXED_EVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK) + +/* + * Constraint on the Event code + UMask + */ +#define INTEL_UEVENT_CONSTRAINT(c, n) \ + EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK) + +#define EVENT_CONSTRAINT_END \ + EVENT_CONSTRAINT(0, 0, 0) + +#define for_each_event_constraint(e, c) \ + for ((e) = (c); (e)->weight; (e)++) + +/* + * Extra registers for specific events. + * + * Some events need large masks and require external MSRs. + * Those extra MSRs end up being shared for all events on + * a PMU and sometimes between PMU of sibling HT threads. + * In either case, the kernel needs to handle conflicting + * accesses to those extra, shared, regs. The data structure + * to manage those registers is stored in cpu_hw_event. + */ +struct extra_reg { + unsigned int event; + unsigned int msr; + u64 config_mask; + u64 valid_mask; + int idx; /* per_xxx->regs[] reg index */ +}; + +#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \ + .event = (e), \ + .msr = (ms), \ + .config_mask = (m), \ + .valid_mask = (vm), \ + .idx = EXTRA_REG_##i \ + } + +#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \ + EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx) + +#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0) + +union perf_capabilities { + struct { + u64 lbr_format:6; + u64 pebs_trap:1; + u64 pebs_arch_reg:1; + u64 pebs_format:4; + u64 smm_freeze:1; + }; + u64 capabilities; +}; + +/* + * struct x86_pmu - generic x86 pmu + */ +struct x86_pmu { + /* + * Generic x86 PMC bits + */ + const char *name; + int version; + int (*handle_irq)(struct pt_regs *); + void (*disable_all)(void); + void (*enable_all)(int added); + void (*enable)(struct perf_event *); + void (*disable)(struct perf_event *); + int (*hw_config)(struct perf_event *event); + int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign); + unsigned eventsel; + unsigned perfctr; + u64 (*event_map)(int); + int max_events; + int num_counters; + int num_counters_fixed; + int cntval_bits; + u64 cntval_mask; + int apic; + u64 max_period; + struct event_constraint * + (*get_event_constraints)(struct cpu_hw_events *cpuc, + struct perf_event *event); + + void (*put_event_constraints)(struct cpu_hw_events *cpuc, + struct perf_event *event); + struct event_constraint *event_constraints; + void (*quirks)(void); + int perfctr_second_write; + + int (*cpu_prepare)(int cpu); + void (*cpu_starting)(int cpu); + void (*cpu_dying)(int cpu); + void (*cpu_dead)(int cpu); + + /* + * Intel Arch Perfmon v2+ + */ + u64 intel_ctrl; + union perf_capabilities intel_cap; + + /* + * Intel DebugStore bits + */ + int bts, pebs; + int bts_active, pebs_active; + int pebs_record_size; + void (*drain_pebs)(struct pt_regs *regs); + struct event_constraint *pebs_constraints; + + /* + * Intel LBR + */ + unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */ + int lbr_nr; /* hardware stack size */ + + /* + * Extra registers for events + */ + struct extra_reg *extra_regs; + unsigned int er_flags; + + /* + * Intel host/guest support (KVM) + */ + struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr); +}; + +#define ERF_NO_HT_SHARING 1 +#define ERF_HAS_RSP_1 2 + +extern struct x86_pmu x86_pmu __read_mostly; + +DECLARE_PER_CPU(struct cpu_hw_events, cpu_hw_events); + +int x86_perf_event_set_period(struct perf_event *event); + +/* + * Generalized hw caching related hw_event table, filled + * in on a per model basis. A value of 0 means + * 'not supported', -1 means 'hw_event makes no sense on + * this CPU', any other value means the raw hw_event + * ID. + */ + +#define C(x) PERF_COUNT_HW_CACHE_##x + +extern u64 __read_mostly hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; +extern u64 __read_mostly hw_cache_extra_regs + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; + +u64 x86_perf_event_update(struct perf_event *event); + +static inline int x86_pmu_addr_offset(int index) +{ + int offset; + + /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */ + alternative_io(ASM_NOP2, + "shll $1, %%eax", + X86_FEATURE_PERFCTR_CORE, + "=a" (offset), + "a" (index)); + + return offset; +} + +static inline unsigned int x86_pmu_config_addr(int index) +{ + return x86_pmu.eventsel + x86_pmu_addr_offset(index); +} + +static inline unsigned int x86_pmu_event_addr(int index) +{ + return x86_pmu.perfctr + x86_pmu_addr_offset(index); +} + +int x86_setup_perfctr(struct perf_event *event); + +int x86_pmu_hw_config(struct perf_event *event); + +void x86_pmu_disable_all(void); + +static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc, + u64 enable_mask) +{ + if (hwc->extra_reg.reg) + wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config); + wrmsrl(hwc->config_base, hwc->config | enable_mask); +} + +void x86_pmu_enable_all(int added); + +int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign); + +void x86_pmu_stop(struct perf_event *event, int flags); + +static inline void x86_pmu_disable_event(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + wrmsrl(hwc->config_base, hwc->config); +} + +void x86_pmu_enable_event(struct perf_event *event); + +int x86_pmu_handle_irq(struct pt_regs *regs); + +extern struct event_constraint emptyconstraint; + +extern struct event_constraint unconstrained; + +#ifdef CONFIG_CPU_SUP_AMD + +int amd_pmu_init(void); + +#else /* CONFIG_CPU_SUP_AMD */ + +static inline int amd_pmu_init(void) +{ + return 0; +} + +#endif /* CONFIG_CPU_SUP_AMD */ + +#ifdef CONFIG_CPU_SUP_INTEL + +int intel_pmu_save_and_restart(struct perf_event *event); + +struct event_constraint * +x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event); + +struct intel_shared_regs *allocate_shared_regs(int cpu); + +int intel_pmu_init(void); + +void init_debug_store_on_cpu(int cpu); + +void fini_debug_store_on_cpu(int cpu); + +void release_ds_buffers(void); + +void reserve_ds_buffers(void); + +extern struct event_constraint bts_constraint; + +void intel_pmu_enable_bts(u64 config); + +void intel_pmu_disable_bts(void); + +int intel_pmu_drain_bts_buffer(void); + +extern struct event_constraint intel_core2_pebs_event_constraints[]; + +extern struct event_constraint intel_atom_pebs_event_constraints[]; + +extern struct event_constraint intel_nehalem_pebs_event_constraints[]; + +extern struct event_constraint intel_westmere_pebs_event_constraints[]; + +extern struct event_constraint intel_snb_pebs_event_constraints[]; + +struct event_constraint *intel_pebs_constraints(struct perf_event *event); + +void intel_pmu_pebs_enable(struct perf_event *event); + +void intel_pmu_pebs_disable(struct perf_event *event); + +void intel_pmu_pebs_enable_all(void); + +void intel_pmu_pebs_disable_all(void); + +void intel_ds_init(void); + +void intel_pmu_lbr_reset(void); + +void intel_pmu_lbr_enable(struct perf_event *event); + +void intel_pmu_lbr_disable(struct perf_event *event); + +void intel_pmu_lbr_enable_all(void); + +void intel_pmu_lbr_disable_all(void); + +void intel_pmu_lbr_read(void); + +void intel_pmu_lbr_init_core(void); + +void intel_pmu_lbr_init_nhm(void); + +void intel_pmu_lbr_init_atom(void); + +int p4_pmu_init(void); + +int p6_pmu_init(void); + +#else /* CONFIG_CPU_SUP_INTEL */ + +static inline void reserve_ds_buffers(void) +{ +} + +static inline void release_ds_buffers(void) +{ +} + +static inline int intel_pmu_init(void) +{ + return 0; +} + +static inline struct intel_shared_regs *allocate_shared_regs(int cpu) +{ + return NULL; +} + +#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index 941caa2e449b..aeefd45697a2 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -1,4 +1,10 @@ -#ifdef CONFIG_CPU_SUP_AMD +#include <linux/perf_event.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <asm/apicdef.h> + +#include "perf_event.h" static __initconst const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] @@ -132,6 +138,19 @@ static int amd_pmu_hw_config(struct perf_event *event) if (ret) return ret; + if (event->attr.exclude_host && event->attr.exclude_guest) + /* + * When HO == GO == 1 the hardware treats that as GO == HO == 0 + * and will count in both modes. We don't want to count in that + * case so we emulate no-counting by setting US = OS = 0. + */ + event->hw.config &= ~(ARCH_PERFMON_EVENTSEL_USR | + ARCH_PERFMON_EVENTSEL_OS); + else if (event->attr.exclude_host) + event->hw.config |= AMD_PERFMON_EVENTSEL_GUESTONLY; + else if (event->attr.exclude_guest) + event->hw.config |= AMD_PERFMON_EVENTSEL_HOSTONLY; + if (event->attr.type != PERF_TYPE_RAW) return 0; @@ -350,7 +369,7 @@ static void amd_pmu_cpu_starting(int cpu) continue; if (nb->nb_id == nb_id) { - kfree(cpuc->amd_nb); + cpuc->kfree_on_online = cpuc->amd_nb; cpuc->amd_nb = nb; break; } @@ -392,7 +411,7 @@ static __initconst const struct x86_pmu amd_pmu = { .perfctr = MSR_K7_PERFCTR0, .event_map = amd_pmu_event_map, .max_events = ARRAY_SIZE(amd_perfmon_event_map), - .num_counters = 4, + .num_counters = AMD64_NUM_COUNTERS, .cntval_bits = 48, .cntval_mask = (1ULL << 48) - 1, .apic = 1, @@ -556,7 +575,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = { .perfctr = MSR_F15H_PERF_CTR, .event_map = amd_pmu_event_map, .max_events = ARRAY_SIZE(amd_perfmon_event_map), - .num_counters = 6, + .num_counters = AMD64_NUM_COUNTERS_F15H, .cntval_bits = 48, .cntval_mask = (1ULL << 48) - 1, .apic = 1, @@ -573,7 +592,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = { #endif }; -static __init int amd_pmu_init(void) +__init int amd_pmu_init(void) { /* Performance-monitoring supported from K7 and later: */ if (boot_cpu_data.x86 < 6) @@ -602,12 +621,3 @@ static __init int amd_pmu_init(void) return 0; } - -#else /* CONFIG_CPU_SUP_AMD */ - -static int amd_pmu_init(void) -{ - return 0; -} - -#endif diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c new file mode 100644 index 000000000000..ab6343d21825 --- /dev/null +++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c @@ -0,0 +1,294 @@ +/* + * Performance events - AMD IBS + * + * Copyright (C) 2011 Advanced Micro Devices, Inc., Robert Richter + * + * For licencing details see kernel-base/COPYING + */ + +#include <linux/perf_event.h> +#include <linux/module.h> +#include <linux/pci.h> + +#include <asm/apic.h> + +static u32 ibs_caps; + +#if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) + +static struct pmu perf_ibs; + +static int perf_ibs_init(struct perf_event *event) +{ + if (perf_ibs.type != event->attr.type) + return -ENOENT; + return 0; +} + +static int perf_ibs_add(struct perf_event *event, int flags) +{ + return 0; +} + +static void perf_ibs_del(struct perf_event *event, int flags) +{ +} + +static struct pmu perf_ibs = { + .event_init= perf_ibs_init, + .add= perf_ibs_add, + .del= perf_ibs_del, +}; + +static __init int perf_event_ibs_init(void) +{ + if (!ibs_caps) + return -ENODEV; /* ibs not supported by the cpu */ + + perf_pmu_register(&perf_ibs, "ibs", -1); + printk(KERN_INFO "perf: AMD IBS detected (0x%08x)\n", ibs_caps); + + return 0; +} + +#else /* defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) */ + +static __init int perf_event_ibs_init(void) { return 0; } + +#endif + +/* IBS - apic initialization, for perf and oprofile */ + +static __init u32 __get_ibs_caps(void) +{ + u32 caps; + unsigned int max_level; + + if (!boot_cpu_has(X86_FEATURE_IBS)) + return 0; + + /* check IBS cpuid feature flags */ + max_level = cpuid_eax(0x80000000); + if (max_level < IBS_CPUID_FEATURES) + return IBS_CAPS_DEFAULT; + + caps = cpuid_eax(IBS_CPUID_FEATURES); + if (!(caps & IBS_CAPS_AVAIL)) + /* cpuid flags not valid */ + return IBS_CAPS_DEFAULT; + + return caps; +} + +u32 get_ibs_caps(void) +{ + return ibs_caps; +} + +EXPORT_SYMBOL(get_ibs_caps); + +static inline int get_eilvt(int offset) +{ + return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1); +} + +static inline int put_eilvt(int offset) +{ + return !setup_APIC_eilvt(offset, 0, 0, 1); +} + +/* + * Check and reserve APIC extended interrupt LVT offset for IBS if available. + */ +static inline int ibs_eilvt_valid(void) +{ + int offset; + u64 val; + int valid = 0; + + preempt_disable(); + + rdmsrl(MSR_AMD64_IBSCTL, val); + offset = val & IBSCTL_LVT_OFFSET_MASK; + + if (!(val & IBSCTL_LVT_OFFSET_VALID)) { + pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n", + smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); + goto out; + } + + if (!get_eilvt(offset)) { + pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n", + smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); + goto out; + } + + valid = 1; +out: + preempt_enable(); + + return valid; +} + +static int setup_ibs_ctl(int ibs_eilvt_off) +{ + struct pci_dev *cpu_cfg; + int nodes; + u32 value = 0; + + nodes = 0; + cpu_cfg = NULL; + do { + cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD, + PCI_DEVICE_ID_AMD_10H_NB_MISC, + cpu_cfg); + if (!cpu_cfg) + break; + ++nodes; + pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off + | IBSCTL_LVT_OFFSET_VALID); + pci_read_config_dword(cpu_cfg, IBSCTL, &value); + if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) { + pci_dev_put(cpu_cfg); + printk(KERN_DEBUG "Failed to setup IBS LVT offset, " + "IBSCTL = 0x%08x\n", value); + return -EINVAL; + } + } while (1); + + if (!nodes) { + printk(KERN_DEBUG "No CPU node configured for IBS\n"); + return -ENODEV; + } + + return 0; +} + +/* + * This runs only on the current cpu. We try to find an LVT offset and + * setup the local APIC. For this we must disable preemption. On + * success we initialize all nodes with this offset. This updates then + * the offset in the IBS_CTL per-node msr. The per-core APIC setup of + * the IBS interrupt vector is handled by perf_ibs_cpu_notifier that + * is using the new offset. + */ +static int force_ibs_eilvt_setup(void) +{ + int offset; + int ret; + + preempt_disable(); + /* find the next free available EILVT entry, skip offset 0 */ + for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) { + if (get_eilvt(offset)) + break; + } + preempt_enable(); + + if (offset == APIC_EILVT_NR_MAX) { + printk(KERN_DEBUG "No EILVT entry available\n"); + return -EBUSY; + } + + ret = setup_ibs_ctl(offset); + if (ret) + goto out; + + if (!ibs_eilvt_valid()) { + ret = -EFAULT; + goto out; + } + + pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset); + pr_err(FW_BUG "workaround enabled for IBS LVT offset\n"); + + return 0; +out: + preempt_disable(); + put_eilvt(offset); + preempt_enable(); + return ret; +} + +static inline int get_ibs_lvt_offset(void) +{ + u64 val; + + rdmsrl(MSR_AMD64_IBSCTL, val); + if (!(val & IBSCTL_LVT_OFFSET_VALID)) + return -EINVAL; + + return val & IBSCTL_LVT_OFFSET_MASK; +} + +static void setup_APIC_ibs(void *dummy) +{ + int offset; + + offset = get_ibs_lvt_offset(); + if (offset < 0) + goto failed; + + if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0)) + return; +failed: + pr_warn("perf: IBS APIC setup failed on cpu #%d\n", + smp_processor_id()); +} + +static void clear_APIC_ibs(void *dummy) +{ + int offset; + + offset = get_ibs_lvt_offset(); + if (offset >= 0) + setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1); +} + +static int __cpuinit +perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_STARTING: + setup_APIC_ibs(NULL); + break; + case CPU_DYING: + clear_APIC_ibs(NULL); + break; + default: + break; + } + + return NOTIFY_OK; +} + +static __init int amd_ibs_init(void) +{ + u32 caps; + int ret; + + caps = __get_ibs_caps(); + if (!caps) + return -ENODEV; /* ibs not supported by the cpu */ + + if (!ibs_eilvt_valid()) { + ret = force_ibs_eilvt_setup(); + if (ret) { + pr_err("Failed to setup IBS, %d\n", ret); + return ret; + } + } + + get_online_cpus(); + ibs_caps = caps; + /* make ibs_caps visible to other cpus: */ + smp_mb(); + perf_cpu_notifier(perf_ibs_cpu_notifier); + smp_call_function(setup_APIC_ibs, NULL, 1); + put_online_cpus(); + + return perf_event_ibs_init(); +} + +/* Since we need the pci subsystem to init ibs we can't do this earlier: */ +device_initcall(amd_ibs_init); diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index f88af2c2a561..e09ca20e86ee 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -1,16 +1,19 @@ -#ifdef CONFIG_CPU_SUP_INTEL - /* * Per core/cpu state * * Used to coordinate shared registers between HT threads or * among events on a single PMU. */ -struct intel_shared_regs { - struct er_account regs[EXTRA_REG_MAX]; - int refcnt; /* per-core: #HT threads */ - unsigned core_id; /* per-core: core id */ -}; + +#include <linux/stddef.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/slab.h> + +#include <asm/hardirq.h> +#include <asm/apic.h> + +#include "perf_event.h" /* * Intel PerfMon, used on Core and later. @@ -746,7 +749,8 @@ static void intel_pmu_enable_all(int added) intel_pmu_pebs_enable_all(); intel_pmu_lbr_enable_all(); - wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); + wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, + x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask); if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) { struct perf_event *event = @@ -869,6 +873,7 @@ static void intel_pmu_disable_fixed(struct hw_perf_event *hwc) static void intel_pmu_disable_event(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) { intel_pmu_disable_bts(); @@ -876,6 +881,9 @@ static void intel_pmu_disable_event(struct perf_event *event) return; } + cpuc->intel_ctrl_guest_mask &= ~(1ull << hwc->idx); + cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx); + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { intel_pmu_disable_fixed(hwc); return; @@ -921,6 +929,7 @@ static void intel_pmu_enable_fixed(struct hw_perf_event *hwc) static void intel_pmu_enable_event(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) { if (!__this_cpu_read(cpu_hw_events.enabled)) @@ -930,6 +939,11 @@ static void intel_pmu_enable_event(struct perf_event *event) return; } + if (event->attr.exclude_host) + cpuc->intel_ctrl_guest_mask |= (1ull << hwc->idx); + if (event->attr.exclude_guest) + cpuc->intel_ctrl_host_mask |= (1ull << hwc->idx); + if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { intel_pmu_enable_fixed(hwc); return; @@ -945,7 +959,7 @@ static void intel_pmu_enable_event(struct perf_event *event) * Save and restart an expired event. Called by NMI contexts, * so it has to be careful about preempting normal event ops: */ -static int intel_pmu_save_and_restart(struct perf_event *event) +int intel_pmu_save_and_restart(struct perf_event *event) { x86_perf_event_update(event); return x86_perf_event_set_period(event); @@ -1197,6 +1211,21 @@ intel_shared_regs_constraints(struct cpu_hw_events *cpuc, return c; } +struct event_constraint * +x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) +{ + struct event_constraint *c; + + if (x86_pmu.event_constraints) { + for_each_event_constraint(c, x86_pmu.event_constraints) { + if ((event->hw.config & c->cmask) == c->code) + return c; + } + } + + return &unconstrained; +} + static struct event_constraint * intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event) { @@ -1284,12 +1313,84 @@ static int intel_pmu_hw_config(struct perf_event *event) return 0; } +struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr) +{ + if (x86_pmu.guest_get_msrs) + return x86_pmu.guest_get_msrs(nr); + *nr = 0; + return NULL; +} +EXPORT_SYMBOL_GPL(perf_guest_get_msrs); + +static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs; + + arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL; + arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask; + arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask; + + *nr = 1; + return arr; +} + +static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs; + int idx; + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + struct perf_event *event = cpuc->events[idx]; + + arr[idx].msr = x86_pmu_config_addr(idx); + arr[idx].host = arr[idx].guest = 0; + + if (!test_bit(idx, cpuc->active_mask)) + continue; + + arr[idx].host = arr[idx].guest = + event->hw.config | ARCH_PERFMON_EVENTSEL_ENABLE; + + if (event->attr.exclude_host) + arr[idx].host &= ~ARCH_PERFMON_EVENTSEL_ENABLE; + else if (event->attr.exclude_guest) + arr[idx].guest &= ~ARCH_PERFMON_EVENTSEL_ENABLE; + } + + *nr = x86_pmu.num_counters; + return arr; +} + +static void core_pmu_enable_event(struct perf_event *event) +{ + if (!event->attr.exclude_host) + x86_pmu_enable_event(event); +} + +static void core_pmu_enable_all(int added) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int idx; + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + struct hw_perf_event *hwc = &cpuc->events[idx]->hw; + + if (!test_bit(idx, cpuc->active_mask) || + cpuc->events[idx]->attr.exclude_host) + continue; + + __x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE); + } +} + static __initconst const struct x86_pmu core_pmu = { .name = "core", .handle_irq = x86_pmu_handle_irq, .disable_all = x86_pmu_disable_all, - .enable_all = x86_pmu_enable_all, - .enable = x86_pmu_enable_event, + .enable_all = core_pmu_enable_all, + .enable = core_pmu_enable_event, .disable = x86_pmu_disable_event, .hw_config = x86_pmu_hw_config, .schedule_events = x86_schedule_events, @@ -1307,9 +1408,10 @@ static __initconst const struct x86_pmu core_pmu = { .get_event_constraints = intel_get_event_constraints, .put_event_constraints = intel_put_event_constraints, .event_constraints = intel_core_event_constraints, + .guest_get_msrs = core_guest_get_msrs, }; -static struct intel_shared_regs *allocate_shared_regs(int cpu) +struct intel_shared_regs *allocate_shared_regs(int cpu) { struct intel_shared_regs *regs; int i; @@ -1362,7 +1464,7 @@ static void intel_pmu_cpu_starting(int cpu) pc = per_cpu(cpu_hw_events, i).shared_regs; if (pc && pc->core_id == core_id) { - kfree(cpuc->shared_regs); + cpuc->kfree_on_online = cpuc->shared_regs; cpuc->shared_regs = pc; break; } @@ -1413,6 +1515,7 @@ static __initconst const struct x86_pmu intel_pmu = { .cpu_prepare = intel_pmu_cpu_prepare, .cpu_starting = intel_pmu_cpu_starting, .cpu_dying = intel_pmu_cpu_dying, + .guest_get_msrs = intel_guest_get_msrs, }; static void intel_clovertown_quirks(void) @@ -1441,7 +1544,7 @@ static void intel_clovertown_quirks(void) x86_pmu.pebs_constraints = NULL; } -static __init int intel_pmu_init(void) +__init int intel_pmu_init(void) { union cpuid10_edx edx; union cpuid10_eax eax; @@ -1597,7 +1700,7 @@ static __init int intel_pmu_init(void) intel_pmu_lbr_init_nhm(); x86_pmu.event_constraints = intel_snb_event_constraints; - x86_pmu.pebs_constraints = intel_snb_pebs_events; + x86_pmu.pebs_constraints = intel_snb_pebs_event_constraints; x86_pmu.extra_regs = intel_snb_extra_regs; /* all extra regs are per-cpu when HT is on */ x86_pmu.er_flags |= ERF_HAS_RSP_1; @@ -1628,16 +1731,3 @@ static __init int intel_pmu_init(void) } return 0; } - -#else /* CONFIG_CPU_SUP_INTEL */ - -static int intel_pmu_init(void) -{ - return 0; -} - -static struct intel_shared_regs *allocate_shared_regs(int cpu) -{ - return NULL; -} -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 1b1ef3addcfd..c0d238f49db8 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -1,7 +1,10 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/bitops.h> +#include <linux/types.h> +#include <linux/slab.h> -/* The maximal number of PEBS events: */ -#define MAX_PEBS_EVENTS 4 +#include <asm/perf_event.h> + +#include "perf_event.h" /* The size of a BTS record in bytes: */ #define BTS_RECORD_SIZE 24 @@ -37,24 +40,7 @@ struct pebs_record_nhm { u64 status, dla, dse, lat; }; -/* - * A debug store configuration. - * - * We only support architectures that use 64bit fields. - */ -struct debug_store { - u64 bts_buffer_base; - u64 bts_index; - u64 bts_absolute_maximum; - u64 bts_interrupt_threshold; - u64 pebs_buffer_base; - u64 pebs_index; - u64 pebs_absolute_maximum; - u64 pebs_interrupt_threshold; - u64 pebs_event_reset[MAX_PEBS_EVENTS]; -}; - -static void init_debug_store_on_cpu(int cpu) +void init_debug_store_on_cpu(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; @@ -66,7 +52,7 @@ static void init_debug_store_on_cpu(int cpu) (u32)((u64)(unsigned long)ds >> 32)); } -static void fini_debug_store_on_cpu(int cpu) +void fini_debug_store_on_cpu(int cpu) { if (!per_cpu(cpu_hw_events, cpu).ds) return; @@ -175,7 +161,7 @@ static void release_ds_buffer(int cpu) kfree(ds); } -static void release_ds_buffers(void) +void release_ds_buffers(void) { int cpu; @@ -194,7 +180,7 @@ static void release_ds_buffers(void) put_online_cpus(); } -static void reserve_ds_buffers(void) +void reserve_ds_buffers(void) { int bts_err = 0, pebs_err = 0; int cpu; @@ -260,10 +246,10 @@ static void reserve_ds_buffers(void) * BTS */ -static struct event_constraint bts_constraint = +struct event_constraint bts_constraint = EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0); -static void intel_pmu_enable_bts(u64 config) +void intel_pmu_enable_bts(u64 config) { unsigned long debugctlmsr; @@ -282,7 +268,7 @@ static void intel_pmu_enable_bts(u64 config) update_debugctlmsr(debugctlmsr); } -static void intel_pmu_disable_bts(void) +void intel_pmu_disable_bts(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); unsigned long debugctlmsr; @@ -299,7 +285,7 @@ static void intel_pmu_disable_bts(void) update_debugctlmsr(debugctlmsr); } -static int intel_pmu_drain_bts_buffer(void) +int intel_pmu_drain_bts_buffer(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct debug_store *ds = cpuc->ds; @@ -361,7 +347,7 @@ static int intel_pmu_drain_bts_buffer(void) /* * PEBS */ -static struct event_constraint intel_core2_pebs_event_constraints[] = { +struct event_constraint intel_core2_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */ @@ -370,14 +356,14 @@ static struct event_constraint intel_core2_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_atom_pebs_event_constraints[] = { +struct event_constraint intel_atom_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */ INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* MISPREDICTED_BRANCH_RETIRED */ INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED.* */ EVENT_CONSTRAINT_END }; -static struct event_constraint intel_nehalem_pebs_event_constraints[] = { +struct event_constraint intel_nehalem_pebs_event_constraints[] = { INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */ INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */ INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */ @@ -392,7 +378,7 @@ static struct event_constraint intel_nehalem_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_westmere_pebs_event_constraints[] = { +struct event_constraint intel_westmere_pebs_event_constraints[] = { INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */ INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */ INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */ @@ -407,7 +393,7 @@ static struct event_constraint intel_westmere_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint intel_snb_pebs_events[] = { +struct event_constraint intel_snb_pebs_event_constraints[] = { INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */ INTEL_UEVENT_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */ INTEL_UEVENT_CONSTRAINT(0x02c2, 0xf), /* UOPS_RETIRED.RETIRE_SLOTS */ @@ -428,8 +414,7 @@ static struct event_constraint intel_snb_pebs_events[] = { EVENT_CONSTRAINT_END }; -static struct event_constraint * -intel_pebs_constraints(struct perf_event *event) +struct event_constraint *intel_pebs_constraints(struct perf_event *event) { struct event_constraint *c; @@ -446,7 +431,7 @@ intel_pebs_constraints(struct perf_event *event) return &emptyconstraint; } -static void intel_pmu_pebs_enable(struct perf_event *event) +void intel_pmu_pebs_enable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -460,7 +445,7 @@ static void intel_pmu_pebs_enable(struct perf_event *event) intel_pmu_lbr_enable(event); } -static void intel_pmu_pebs_disable(struct perf_event *event) +void intel_pmu_pebs_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; @@ -475,7 +460,7 @@ static void intel_pmu_pebs_disable(struct perf_event *event) intel_pmu_lbr_disable(event); } -static void intel_pmu_pebs_enable_all(void) +void intel_pmu_pebs_enable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -483,7 +468,7 @@ static void intel_pmu_pebs_enable_all(void) wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); } -static void intel_pmu_pebs_disable_all(void) +void intel_pmu_pebs_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -576,8 +561,6 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) return 0; } -static int intel_pmu_save_and_restart(struct perf_event *event); - static void __intel_pmu_pebs_event(struct perf_event *event, struct pt_regs *iregs, void *__pebs) { @@ -716,7 +699,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) * BTS, PEBS probe and setup */ -static void intel_ds_init(void) +void intel_ds_init(void) { /* * No support for 32bit formats @@ -749,15 +732,3 @@ static void intel_ds_init(void) } } } - -#else /* CONFIG_CPU_SUP_INTEL */ - -static void reserve_ds_buffers(void) -{ -} - -static void release_ds_buffers(void) -{ -} - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c index d202c1bece1a..3fab3de3ce96 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c +++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c @@ -1,4 +1,10 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> +#include <linux/types.h> + +#include <asm/perf_event.h> +#include <asm/msr.h> + +#include "perf_event.h" enum { LBR_FORMAT_32 = 0x00, @@ -48,7 +54,7 @@ static void intel_pmu_lbr_reset_64(void) } } -static void intel_pmu_lbr_reset(void) +void intel_pmu_lbr_reset(void) { if (!x86_pmu.lbr_nr) return; @@ -59,7 +65,7 @@ static void intel_pmu_lbr_reset(void) intel_pmu_lbr_reset_64(); } -static void intel_pmu_lbr_enable(struct perf_event *event) +void intel_pmu_lbr_enable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -81,7 +87,7 @@ static void intel_pmu_lbr_enable(struct perf_event *event) cpuc->lbr_users++; } -static void intel_pmu_lbr_disable(struct perf_event *event) +void intel_pmu_lbr_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -95,7 +101,7 @@ static void intel_pmu_lbr_disable(struct perf_event *event) __intel_pmu_lbr_disable(); } -static void intel_pmu_lbr_enable_all(void) +void intel_pmu_lbr_enable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -103,7 +109,7 @@ static void intel_pmu_lbr_enable_all(void) __intel_pmu_lbr_enable(); } -static void intel_pmu_lbr_disable_all(void) +void intel_pmu_lbr_disable_all(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -178,7 +184,7 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc) cpuc->lbr_stack.nr = i; } -static void intel_pmu_lbr_read(void) +void intel_pmu_lbr_read(void) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); @@ -191,7 +197,7 @@ static void intel_pmu_lbr_read(void) intel_pmu_lbr_read_64(cpuc); } -static void intel_pmu_lbr_init_core(void) +void intel_pmu_lbr_init_core(void) { x86_pmu.lbr_nr = 4; x86_pmu.lbr_tos = 0x01c9; @@ -199,7 +205,7 @@ static void intel_pmu_lbr_init_core(void) x86_pmu.lbr_to = 0x60; } -static void intel_pmu_lbr_init_nhm(void) +void intel_pmu_lbr_init_nhm(void) { x86_pmu.lbr_nr = 16; x86_pmu.lbr_tos = 0x01c9; @@ -207,12 +213,10 @@ static void intel_pmu_lbr_init_nhm(void) x86_pmu.lbr_to = 0x6c0; } -static void intel_pmu_lbr_init_atom(void) +void intel_pmu_lbr_init_atom(void) { x86_pmu.lbr_nr = 8; x86_pmu.lbr_tos = 0x01c9; x86_pmu.lbr_from = 0x40; x86_pmu.lbr_to = 0x60; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 7809d2bcb209..492bf1358a7c 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -7,9 +7,13 @@ * For licencing details see kernel-base/COPYING */ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> #include <asm/perf_event_p4.h> +#include <asm/hardirq.h> +#include <asm/apic.h> + +#include "perf_event.h" #define P4_CNTR_LIMIT 3 /* @@ -1303,7 +1307,7 @@ static __initconst const struct x86_pmu p4_pmu = { .perfctr_second_write = 1, }; -static __init int p4_pmu_init(void) +__init int p4_pmu_init(void) { unsigned int low, high; @@ -1326,5 +1330,3 @@ static __init int p4_pmu_init(void) return 0; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c index 20c097e33860..c7181befecde 100644 --- a/arch/x86/kernel/cpu/perf_event_p6.c +++ b/arch/x86/kernel/cpu/perf_event_p6.c @@ -1,4 +1,7 @@ -#ifdef CONFIG_CPU_SUP_INTEL +#include <linux/perf_event.h> +#include <linux/types.h> + +#include "perf_event.h" /* * Not sure about some of these @@ -114,7 +117,7 @@ static __initconst const struct x86_pmu p6_pmu = { .event_constraints = p6_event_constraints, }; -static __init int p6_pmu_init(void) +__init int p6_pmu_init(void) { switch (boot_cpu_data.x86_model) { case 1: @@ -138,5 +141,3 @@ static __init int p6_pmu_init(void) return 0; } - -#endif /* CONFIG_CPU_SUP_INTEL */ diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index 764c7c2b1811..13ad89971d47 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -32,15 +32,12 @@ int in_crash_kexec; #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) -static void kdump_nmi_callback(int cpu, struct die_args *args) +static void kdump_nmi_callback(int cpu, struct pt_regs *regs) { - struct pt_regs *regs; #ifdef CONFIG_X86_32 struct pt_regs fixed_regs; #endif - regs = args->regs; - #ifdef CONFIG_X86_32 if (!user_mode_vm(regs)) { crash_fixup_ss_esp(&fixed_regs, regs); diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c index 3fee346ef545..cacdd46d184d 100644 --- a/arch/x86/kernel/jump_label.c +++ b/arch/x86/kernel/jump_label.c @@ -42,7 +42,7 @@ void arch_jump_label_transform(struct jump_entry *entry, put_online_cpus(); } -void arch_jump_label_text_poke_early(jump_label_t addr) +void __init_or_module arch_jump_label_text_poke_early(jump_label_t addr) { text_poke_early((void *)addr, ideal_nops[NOP_ATOMIC5], JUMP_LABEL_NOP_SIZE); diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 00354d4919a9..faba5771acad 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -511,28 +511,37 @@ single_step_cont(struct pt_regs *regs, struct die_args *args) static int was_in_debug_nmi[NR_CPUS]; -static int __kgdb_notify(struct die_args *args, unsigned long cmd) +static int kgdb_nmi_handler(unsigned int cmd, struct pt_regs *regs) { - struct pt_regs *regs = args->regs; - switch (cmd) { - case DIE_NMI: + case NMI_LOCAL: if (atomic_read(&kgdb_active) != -1) { /* KGDB CPU roundup */ kgdb_nmicallback(raw_smp_processor_id(), regs); was_in_debug_nmi[raw_smp_processor_id()] = 1; touch_nmi_watchdog(); - return NOTIFY_STOP; + return NMI_HANDLED; } - return NOTIFY_DONE; + break; - case DIE_NMIUNKNOWN: + case NMI_UNKNOWN: if (was_in_debug_nmi[raw_smp_processor_id()]) { was_in_debug_nmi[raw_smp_processor_id()] = 0; - return NOTIFY_STOP; + return NMI_HANDLED; } - return NOTIFY_DONE; + break; + default: + /* do nothing */ + break; + } + return NMI_DONE; +} + +static int __kgdb_notify(struct die_args *args, unsigned long cmd) +{ + struct pt_regs *regs = args->regs; + switch (cmd) { case DIE_DEBUG: if (atomic_read(&kgdb_cpu_doing_single_step) != -1) { if (user_mode(regs)) @@ -590,11 +599,6 @@ kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr) static struct notifier_block kgdb_notifier = { .notifier_call = kgdb_notify, - - /* - * Lowest-prio notifier priority, we want to be notified last: - */ - .priority = NMI_LOCAL_LOW_PRIOR, }; /** @@ -605,7 +609,31 @@ static struct notifier_block kgdb_notifier = { */ int kgdb_arch_init(void) { - return register_die_notifier(&kgdb_notifier); + int retval; + + retval = register_die_notifier(&kgdb_notifier); + if (retval) + goto out; + + retval = register_nmi_handler(NMI_LOCAL, kgdb_nmi_handler, + 0, "kgdb"); + if (retval) + goto out1; + + retval = register_nmi_handler(NMI_UNKNOWN, kgdb_nmi_handler, + 0, "kgdb"); + + if (retval) + goto out2; + + return retval; + +out2: + unregister_nmi_handler(NMI_LOCAL, "kgdb"); +out1: + unregister_die_notifier(&kgdb_notifier); +out: + return retval; } static void kgdb_hw_overflow_handler(struct perf_event *event, @@ -673,6 +701,8 @@ void kgdb_arch_exit(void) breakinfo[i].pev = NULL; } } + unregister_nmi_handler(NMI_UNKNOWN, "kgdb"); + unregister_nmi_handler(NMI_LOCAL, "kgdb"); unregister_die_notifier(&kgdb_notifier); } diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index f1a6244d7d93..7da647d8b64c 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -75,8 +75,11 @@ DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); /* * Undefined/reserved opcodes, conditional jump, Opcode Extension * Groups, and some special opcodes can not boost. + * This is non-const and volatile to keep gcc from statically + * optimizing it out, as variable_test_bit makes gcc think only + * *(unsigned long*) is used. */ -static const u32 twobyte_is_boostable[256 / 32] = { +static volatile u32 twobyte_is_boostable[256 / 32] = { /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */ /* ---------------------------------------------- */ W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */ diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c new file mode 100644 index 000000000000..7ec5bd140b87 --- /dev/null +++ b/arch/x86/kernel/nmi.c @@ -0,0 +1,433 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + * Copyright (C) 2011 Don Zickus Red Hat, Inc. + * + * Pentium III FXSR, SSE support + * Gareth Hughes <gareth@valinux.com>, May 2000 + */ + +/* + * Handle hardware traps and faults. + */ +#include <linux/spinlock.h> +#include <linux/kprobes.h> +#include <linux/kdebug.h> +#include <linux/nmi.h> +#include <linux/delay.h> +#include <linux/hardirq.h> +#include <linux/slab.h> + +#include <linux/mca.h> + +#if defined(CONFIG_EDAC) +#include <linux/edac.h> +#endif + +#include <linux/atomic.h> +#include <asm/traps.h> +#include <asm/mach_traps.h> +#include <asm/nmi.h> + +#define NMI_MAX_NAMELEN 16 +struct nmiaction { + struct list_head list; + nmi_handler_t handler; + unsigned int flags; + char *name; +}; + +struct nmi_desc { + spinlock_t lock; + struct list_head head; +}; + +static struct nmi_desc nmi_desc[NMI_MAX] = +{ + { + .lock = __SPIN_LOCK_UNLOCKED(&nmi_desc[0].lock), + .head = LIST_HEAD_INIT(nmi_desc[0].head), + }, + { + .lock = __SPIN_LOCK_UNLOCKED(&nmi_desc[1].lock), + .head = LIST_HEAD_INIT(nmi_desc[1].head), + }, + +}; + +struct nmi_stats { + unsigned int normal; + unsigned int unknown; + unsigned int external; + unsigned int swallow; +}; + +static DEFINE_PER_CPU(struct nmi_stats, nmi_stats); + +static int ignore_nmis; + +int unknown_nmi_panic; +/* + * Prevent NMI reason port (0x61) being accessed simultaneously, can + * only be used in NMI handler. + */ +static DEFINE_RAW_SPINLOCK(nmi_reason_lock); + +static int __init setup_unknown_nmi_panic(char *str) +{ + unknown_nmi_panic = 1; + return 1; +} +__setup("unknown_nmi_panic", setup_unknown_nmi_panic); + +#define nmi_to_desc(type) (&nmi_desc[type]) + +static int notrace __kprobes nmi_handle(unsigned int type, struct pt_regs *regs, bool b2b) +{ + struct nmi_desc *desc = nmi_to_desc(type); + struct nmiaction *a; + int handled=0; + + rcu_read_lock(); + + /* + * NMIs are edge-triggered, which means if you have enough + * of them concurrently, you can lose some because only one + * can be latched at any given time. Walk the whole list + * to handle those situations. + */ + list_for_each_entry_rcu(a, &desc->head, list) + handled += a->handler(type, regs); + + rcu_read_unlock(); + + /* return total number of NMI events handled */ + return handled; +} + +static int __setup_nmi(unsigned int type, struct nmiaction *action) +{ + struct nmi_desc *desc = nmi_to_desc(type); + unsigned long flags; + + spin_lock_irqsave(&desc->lock, flags); + + /* + * most handlers of type NMI_UNKNOWN never return because + * they just assume the NMI is theirs. Just a sanity check + * to manage expectations + */ + WARN_ON_ONCE(type == NMI_UNKNOWN && !list_empty(&desc->head)); + + /* + * some handlers need to be executed first otherwise a fake + * event confuses some handlers (kdump uses this flag) + */ + if (action->flags & NMI_FLAG_FIRST) + list_add_rcu(&action->list, &desc->head); + else + list_add_tail_rcu(&action->list, &desc->head); + + spin_unlock_irqrestore(&desc->lock, flags); + return 0; +} + +static struct nmiaction *__free_nmi(unsigned int type, const char *name) +{ + struct nmi_desc *desc = nmi_to_desc(type); + struct nmiaction *n; + unsigned long flags; + + spin_lock_irqsave(&desc->lock, flags); + + list_for_each_entry_rcu(n, &desc->head, list) { + /* + * the name passed in to describe the nmi handler + * is used as the lookup key + */ + if (!strcmp(n->name, name)) { + WARN(in_nmi(), + "Trying to free NMI (%s) from NMI context!\n", n->name); + list_del_rcu(&n->list); + break; + } + } + + spin_unlock_irqrestore(&desc->lock, flags); + synchronize_rcu(); + return (n); +} + +int register_nmi_handler(unsigned int type, nmi_handler_t handler, + unsigned long nmiflags, const char *devname) +{ + struct nmiaction *action; + int retval = -ENOMEM; + + if (!handler) + return -EINVAL; + + action = kzalloc(sizeof(struct nmiaction), GFP_KERNEL); + if (!action) + goto fail_action; + + action->handler = handler; + action->flags = nmiflags; + action->name = kstrndup(devname, NMI_MAX_NAMELEN, GFP_KERNEL); + if (!action->name) + goto fail_action_name; + + retval = __setup_nmi(type, action); + + if (retval) + goto fail_setup_nmi; + + return retval; + +fail_setup_nmi: + kfree(action->name); +fail_action_name: + kfree(action); +fail_action: + + return retval; +} +EXPORT_SYMBOL_GPL(register_nmi_handler); + +void unregister_nmi_handler(unsigned int type, const char *name) +{ + struct nmiaction *a; + + a = __free_nmi(type, name); + if (a) { + kfree(a->name); + kfree(a); + } +} + +EXPORT_SYMBOL_GPL(unregister_nmi_handler); + +static notrace __kprobes void +pci_serr_error(unsigned char reason, struct pt_regs *regs) +{ + pr_emerg("NMI: PCI system error (SERR) for reason %02x on CPU %d.\n", + reason, smp_processor_id()); + + /* + * On some machines, PCI SERR line is used to report memory + * errors. EDAC makes use of it. + */ +#if defined(CONFIG_EDAC) + if (edac_handler_set()) { + edac_atomic_assert_error(); + return; + } +#endif + + if (panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); + + pr_emerg("Dazed and confused, but trying to continue\n"); + + /* Clear and disable the PCI SERR error line. */ + reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_SERR; + outb(reason, NMI_REASON_PORT); +} + +static notrace __kprobes void +io_check_error(unsigned char reason, struct pt_regs *regs) +{ + unsigned long i; + + pr_emerg( + "NMI: IOCK error (debug interrupt?) for reason %02x on CPU %d.\n", + reason, smp_processor_id()); + show_registers(regs); + + if (panic_on_io_nmi) + panic("NMI IOCK error: Not continuing"); + + /* Re-enable the IOCK line, wait for a few seconds */ + reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_IOCHK; + outb(reason, NMI_REASON_PORT); + + i = 20000; + while (--i) { + touch_nmi_watchdog(); + udelay(100); + } + + reason &= ~NMI_REASON_CLEAR_IOCHK; + outb(reason, NMI_REASON_PORT); +} + +static notrace __kprobes void +unknown_nmi_error(unsigned char reason, struct pt_regs *regs) +{ + int handled; + + /* + * Use 'false' as back-to-back NMIs are dealt with one level up. + * Of course this makes having multiple 'unknown' handlers useless + * as only the first one is ever run (unless it can actually determine + * if it caused the NMI) + */ + handled = nmi_handle(NMI_UNKNOWN, regs, false); + if (handled) { + __this_cpu_add(nmi_stats.unknown, handled); + return; + } + + __this_cpu_add(nmi_stats.unknown, 1); + +#ifdef CONFIG_MCA + /* + * Might actually be able to figure out what the guilty party + * is: + */ + if (MCA_bus) { + mca_handle_nmi(); + return; + } +#endif + pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", + reason, smp_processor_id()); + + pr_emerg("Do you have a strange power saving mode enabled?\n"); + if (unknown_nmi_panic || panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); + + pr_emerg("Dazed and confused, but trying to continue\n"); +} + +static DEFINE_PER_CPU(bool, swallow_nmi); +static DEFINE_PER_CPU(unsigned long, last_nmi_rip); + +static notrace __kprobes void default_do_nmi(struct pt_regs *regs) +{ + unsigned char reason = 0; + int handled; + bool b2b = false; + + /* + * CPU-specific NMI must be processed before non-CPU-specific + * NMI, otherwise we may lose it, because the CPU-specific + * NMI can not be detected/processed on other CPUs. + */ + + /* + * Back-to-back NMIs are interesting because they can either + * be two NMI or more than two NMIs (any thing over two is dropped + * due to NMI being edge-triggered). If this is the second half + * of the back-to-back NMI, assume we dropped things and process + * more handlers. Otherwise reset the 'swallow' NMI behaviour + */ + if (regs->ip == __this_cpu_read(last_nmi_rip)) + b2b = true; + else + __this_cpu_write(swallow_nmi, false); + + __this_cpu_write(last_nmi_rip, regs->ip); + + handled = nmi_handle(NMI_LOCAL, regs, b2b); + __this_cpu_add(nmi_stats.normal, handled); + if (handled) { + /* + * There are cases when a NMI handler handles multiple + * events in the current NMI. One of these events may + * be queued for in the next NMI. Because the event is + * already handled, the next NMI will result in an unknown + * NMI. Instead lets flag this for a potential NMI to + * swallow. + */ + if (handled > 1) + __this_cpu_write(swallow_nmi, true); + return; + } + + /* Non-CPU-specific NMI: NMI sources can be processed on any CPU */ + raw_spin_lock(&nmi_reason_lock); + reason = get_nmi_reason(); + + if (reason & NMI_REASON_MASK) { + if (reason & NMI_REASON_SERR) + pci_serr_error(reason, regs); + else if (reason & NMI_REASON_IOCHK) + io_check_error(reason, regs); +#ifdef CONFIG_X86_32 + /* + * Reassert NMI in case it became active + * meanwhile as it's edge-triggered: + */ + reassert_nmi(); +#endif + __this_cpu_add(nmi_stats.external, 1); + raw_spin_unlock(&nmi_reason_lock); + return; + } + raw_spin_unlock(&nmi_reason_lock); + + /* + * Only one NMI can be latched at a time. To handle + * this we may process multiple nmi handlers at once to + * cover the case where an NMI is dropped. The downside + * to this approach is we may process an NMI prematurely, + * while its real NMI is sitting latched. This will cause + * an unknown NMI on the next run of the NMI processing. + * + * We tried to flag that condition above, by setting the + * swallow_nmi flag when we process more than one event. + * This condition is also only present on the second half + * of a back-to-back NMI, so we flag that condition too. + * + * If both are true, we assume we already processed this + * NMI previously and we swallow it. Otherwise we reset + * the logic. + * + * There are scenarios where we may accidentally swallow + * a 'real' unknown NMI. For example, while processing + * a perf NMI another perf NMI comes in along with a + * 'real' unknown NMI. These two NMIs get combined into + * one (as descibed above). When the next NMI gets + * processed, it will be flagged by perf as handled, but + * noone will know that there was a 'real' unknown NMI sent + * also. As a result it gets swallowed. Or if the first + * perf NMI returns two events handled then the second + * NMI will get eaten by the logic below, again losing a + * 'real' unknown NMI. But this is the best we can do + * for now. + */ + if (b2b && __this_cpu_read(swallow_nmi)) + __this_cpu_add(nmi_stats.swallow, 1); + else + unknown_nmi_error(reason, regs); +} + +dotraplinkage notrace __kprobes void +do_nmi(struct pt_regs *regs, long error_code) +{ + nmi_enter(); + + inc_irq_stat(__nmi_count); + + if (!ignore_nmis) + default_do_nmi(regs); + + nmi_exit(); +} + +void stop_nmi(void) +{ + ignore_nmis++; +} + +void restart_nmi(void) +{ + ignore_nmis--; +} + +/* reset the back-to-back NMI logic */ +void local_touch_nmi(void) +{ + __this_cpu_write(last_nmi_rip, 0); +} diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index b49d00da2aed..622872054fbe 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -117,8 +117,8 @@ again: } /* - * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter - * documentation. + * See <Documentation/x86/x86_64/boot-options.txt> for the iommu kernel + * parameter documentation. */ static __init int iommu_setup(char *p) { diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 7a3b65107a27..795b79f984c2 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -57,6 +57,7 @@ #include <asm/idle.h> #include <asm/syscalls.h> #include <asm/debugreg.h> +#include <asm/nmi.h> asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); @@ -107,6 +108,7 @@ void cpu_idle(void) if (cpu_is_offline(cpu)) play_dead(); + local_touch_nmi(); local_irq_disable(); /* Don't trace irqs off for idle */ stop_critical_timings(); @@ -262,7 +264,7 @@ EXPORT_SYMBOL_GPL(start_thread); /* - * switch_to(x,yn) should switch tasks from x to y. + * switch_to(x,y) should switch tasks from x to y. * * We fsave/fwait so that an exception goes off at the right time * (as a call from the fsave or fwait in effect) rather than to diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index f693e44e1bf6..3bd7e6eebf31 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -51,6 +51,7 @@ #include <asm/idle.h> #include <asm/syscalls.h> #include <asm/debugreg.h> +#include <asm/nmi.h> asmlinkage extern void ret_from_fork(void); @@ -133,6 +134,7 @@ void cpu_idle(void) * from here on, until they go to idle. * Otherwise, idle callbacks can misfire. */ + local_touch_nmi(); local_irq_disable(); enter_idle(); /* Don't trace irqs off for idle */ diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 9242436e9937..e334be1182b9 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -464,7 +464,7 @@ static inline void kb_wait(void) } } -static void vmxoff_nmi(int cpu, struct die_args *args) +static void vmxoff_nmi(int cpu, struct pt_regs *regs) { cpu_emergency_vmxoff(); } @@ -736,14 +736,10 @@ static nmi_shootdown_cb shootdown_callback; static atomic_t waiting_for_crash_ipi; -static int crash_nmi_callback(struct notifier_block *self, - unsigned long val, void *data) +static int crash_nmi_callback(unsigned int val, struct pt_regs *regs) { int cpu; - if (val != DIE_NMI) - return NOTIFY_OK; - cpu = raw_smp_processor_id(); /* Don't do anything if this handler is invoked on crashing cpu. @@ -751,10 +747,10 @@ static int crash_nmi_callback(struct notifier_block *self, * an NMI if system was initially booted with nmi_watchdog parameter. */ if (cpu == crashing_cpu) - return NOTIFY_STOP; + return NMI_HANDLED; local_irq_disable(); - shootdown_callback(cpu, (struct die_args *)data); + shootdown_callback(cpu, regs); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -762,7 +758,7 @@ static int crash_nmi_callback(struct notifier_block *self, for (;;) cpu_relax(); - return 1; + return NMI_HANDLED; } static void smp_send_nmi_allbutself(void) @@ -770,12 +766,6 @@ static void smp_send_nmi_allbutself(void) apic->send_IPI_allbutself(NMI_VECTOR); } -static struct notifier_block crash_nmi_nb = { - .notifier_call = crash_nmi_callback, - /* we want to be the first one called */ - .priority = NMI_LOCAL_HIGH_PRIOR+1, -}; - /* Halt all other CPUs, calling the specified function on each of them * * This function can be used to halt all other CPUs on crash @@ -794,7 +784,8 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback) atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); /* Would it be better to replace the trap vector here? */ - if (register_die_notifier(&crash_nmi_nb)) + if (register_nmi_handler(NMI_LOCAL, crash_nmi_callback, + NMI_FLAG_FIRST, "crash")) return; /* return what? */ /* Ensure the new callback function is set before sending * out the NMI diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 6913369c234c..a8e3eb83466c 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -81,15 +81,6 @@ gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, }; DECLARE_BITMAP(used_vectors, NR_VECTORS); EXPORT_SYMBOL_GPL(used_vectors); -static int ignore_nmis; - -int unknown_nmi_panic; -/* - * Prevent NMI reason port (0x61) being accessed simultaneously, can - * only be used in NMI handler. - */ -static DEFINE_RAW_SPINLOCK(nmi_reason_lock); - static inline void conditional_sti(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) @@ -307,152 +298,6 @@ gp_in_kernel: die("general protection fault", regs, error_code); } -static int __init setup_unknown_nmi_panic(char *str) -{ - unknown_nmi_panic = 1; - return 1; -} -__setup("unknown_nmi_panic", setup_unknown_nmi_panic); - -static notrace __kprobes void -pci_serr_error(unsigned char reason, struct pt_regs *regs) -{ - pr_emerg("NMI: PCI system error (SERR) for reason %02x on CPU %d.\n", - reason, smp_processor_id()); - - /* - * On some machines, PCI SERR line is used to report memory - * errors. EDAC makes use of it. - */ -#if defined(CONFIG_EDAC) - if (edac_handler_set()) { - edac_atomic_assert_error(); - return; - } -#endif - - if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); - - pr_emerg("Dazed and confused, but trying to continue\n"); - - /* Clear and disable the PCI SERR error line. */ - reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_SERR; - outb(reason, NMI_REASON_PORT); -} - -static notrace __kprobes void -io_check_error(unsigned char reason, struct pt_regs *regs) -{ - unsigned long i; - - pr_emerg( - "NMI: IOCK error (debug interrupt?) for reason %02x on CPU %d.\n", - reason, smp_processor_id()); - show_registers(regs); - - if (panic_on_io_nmi) - panic("NMI IOCK error: Not continuing"); - - /* Re-enable the IOCK line, wait for a few seconds */ - reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_IOCHK; - outb(reason, NMI_REASON_PORT); - - i = 20000; - while (--i) { - touch_nmi_watchdog(); - udelay(100); - } - - reason &= ~NMI_REASON_CLEAR_IOCHK; - outb(reason, NMI_REASON_PORT); -} - -static notrace __kprobes void -unknown_nmi_error(unsigned char reason, struct pt_regs *regs) -{ - if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == - NOTIFY_STOP) - return; -#ifdef CONFIG_MCA - /* - * Might actually be able to figure out what the guilty party - * is: - */ - if (MCA_bus) { - mca_handle_nmi(); - return; - } -#endif - pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n", - reason, smp_processor_id()); - - pr_emerg("Do you have a strange power saving mode enabled?\n"); - if (unknown_nmi_panic || panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); - - pr_emerg("Dazed and confused, but trying to continue\n"); -} - -static notrace __kprobes void default_do_nmi(struct pt_regs *regs) -{ - unsigned char reason = 0; - - /* - * CPU-specific NMI must be processed before non-CPU-specific - * NMI, otherwise we may lose it, because the CPU-specific - * NMI can not be detected/processed on other CPUs. - */ - if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP) - return; - - /* Non-CPU-specific NMI: NMI sources can be processed on any CPU */ - raw_spin_lock(&nmi_reason_lock); - reason = get_nmi_reason(); - - if (reason & NMI_REASON_MASK) { - if (reason & NMI_REASON_SERR) - pci_serr_error(reason, regs); - else if (reason & NMI_REASON_IOCHK) - io_check_error(reason, regs); -#ifdef CONFIG_X86_32 - /* - * Reassert NMI in case it became active - * meanwhile as it's edge-triggered: - */ - reassert_nmi(); -#endif - raw_spin_unlock(&nmi_reason_lock); - return; - } - raw_spin_unlock(&nmi_reason_lock); - - unknown_nmi_error(reason, regs); -} - -dotraplinkage notrace __kprobes void -do_nmi(struct pt_regs *regs, long error_code) -{ - nmi_enter(); - - inc_irq_stat(__nmi_count); - - if (!ignore_nmis) - default_do_nmi(regs); - - nmi_exit(); -} - -void stop_nmi(void) -{ - ignore_nmis++; -} - -void restart_nmi(void) -{ - ignore_nmis--; -} - /* May run on IST stack. */ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 18ae83dd1cd7..b56c65de384d 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -56,7 +56,7 @@ DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data) = .lock = __SEQLOCK_UNLOCKED(__vsyscall_gtod_data.lock), }; -static enum { EMULATE, NATIVE, NONE } vsyscall_mode = EMULATE; +static enum { EMULATE, NATIVE, NONE } vsyscall_mode = NATIVE; static int __init vsyscall_setup(char *str) { diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 30326443ab81..87488b93a65c 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -63,9 +63,8 @@ static void __init find_early_table_space(unsigned long end, int use_pse, #ifdef CONFIG_X86_32 /* for fixmap */ tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE); - - good_end = max_pfn_mapped << PAGE_SHIFT; #endif + good_end = max_pfn_mapped << PAGE_SHIFT; base = memblock_find_in_range(start, good_end, tables, PAGE_SIZE); if (base == MEMBLOCK_ERROR) diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c index 67421f38a215..de54b9b278a7 100644 --- a/arch/x86/mm/mmio-mod.c +++ b/arch/x86/mm/mmio-mod.c @@ -29,7 +29,6 @@ #include <linux/slab.h> #include <linux/uaccess.h> #include <linux/io.h> -#include <linux/version.h> #include <linux/kallsyms.h> #include <asm/pgtable.h> #include <linux/mmiotrace.h> diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 68894fdc034b..75f9528e0372 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -61,26 +61,15 @@ u64 op_x86_get_ctrl(struct op_x86_model_spec const *model, } -static int profile_exceptions_notify(struct notifier_block *self, - unsigned long val, void *data) +static int profile_exceptions_notify(unsigned int val, struct pt_regs *regs) { - struct die_args *args = (struct die_args *)data; - int ret = NOTIFY_DONE; - - switch (val) { - case DIE_NMI: - if (ctr_running) - model->check_ctrs(args->regs, &__get_cpu_var(cpu_msrs)); - else if (!nmi_enabled) - break; - else - model->stop(&__get_cpu_var(cpu_msrs)); - ret = NOTIFY_STOP; - break; - default: - break; - } - return ret; + if (ctr_running) + model->check_ctrs(regs, &__get_cpu_var(cpu_msrs)); + else if (!nmi_enabled) + return NMI_DONE; + else + model->stop(&__get_cpu_var(cpu_msrs)); + return NMI_HANDLED; } static void nmi_cpu_save_registers(struct op_msrs *msrs) @@ -355,20 +344,14 @@ static void nmi_cpu_setup(void *dummy) int cpu = smp_processor_id(); struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu); nmi_cpu_save_registers(msrs); - spin_lock(&oprofilefs_lock); + raw_spin_lock(&oprofilefs_lock); model->setup_ctrs(model, msrs); nmi_cpu_setup_mux(cpu, msrs); - spin_unlock(&oprofilefs_lock); + raw_spin_unlock(&oprofilefs_lock); per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC); apic_write(APIC_LVTPC, APIC_DM_NMI); } -static struct notifier_block profile_exceptions_nb = { - .notifier_call = profile_exceptions_notify, - .next = NULL, - .priority = NMI_LOCAL_LOW_PRIOR, -}; - static void nmi_cpu_restore_registers(struct op_msrs *msrs) { struct op_msr *counters = msrs->counters; @@ -402,8 +385,6 @@ static void nmi_cpu_shutdown(void *dummy) apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu)); apic_write(APIC_LVTERR, v); nmi_cpu_restore_registers(msrs); - if (model->cpu_down) - model->cpu_down(); } static void nmi_cpu_up(void *dummy) @@ -508,7 +489,8 @@ static int nmi_setup(void) ctr_running = 0; /* make variables visible to the nmi handler: */ smp_mb(); - err = register_die_notifier(&profile_exceptions_nb); + err = register_nmi_handler(NMI_LOCAL, profile_exceptions_notify, + 0, "oprofile"); if (err) goto fail; @@ -538,7 +520,7 @@ static void nmi_shutdown(void) put_online_cpus(); /* make variables visible to the nmi handler: */ smp_mb(); - unregister_die_notifier(&profile_exceptions_nb); + unregister_nmi_handler(NMI_LOCAL, "oprofile"); msrs = &get_cpu_var(cpu_msrs); model->shutdown(msrs); free_msrs(); diff --git a/arch/x86/oprofile/nmi_timer_int.c b/arch/x86/oprofile/nmi_timer_int.c index 720bf5a53c51..7f8052cd6620 100644 --- a/arch/x86/oprofile/nmi_timer_int.c +++ b/arch/x86/oprofile/nmi_timer_int.c @@ -18,32 +18,16 @@ #include <asm/apic.h> #include <asm/ptrace.h> -static int profile_timer_exceptions_notify(struct notifier_block *self, - unsigned long val, void *data) +static int profile_timer_exceptions_notify(unsigned int val, struct pt_regs *regs) { - struct die_args *args = (struct die_args *)data; - int ret = NOTIFY_DONE; - - switch (val) { - case DIE_NMI: - oprofile_add_sample(args->regs, 0); - ret = NOTIFY_STOP; - break; - default: - break; - } - return ret; + oprofile_add_sample(regs, 0); + return NMI_HANDLED; } -static struct notifier_block profile_timer_exceptions_nb = { - .notifier_call = profile_timer_exceptions_notify, - .next = NULL, - .priority = NMI_LOW_PRIOR, -}; - static int timer_start(void) { - if (register_die_notifier(&profile_timer_exceptions_nb)) + if (register_nmi_handler(NMI_LOCAL, profile_timer_exceptions_notify, + 0, "oprofile-timer")) return 1; return 0; } @@ -51,7 +35,7 @@ static int timer_start(void) static void timer_stop(void) { - unregister_die_notifier(&profile_timer_exceptions_nb); + unregister_nmi_handler(NMI_LOCAL, "oprofile-timer"); synchronize_sched(); /* Allow already-started NMIs to complete. */ } diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index 9cbb710dc94b..303f08637826 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -29,8 +29,6 @@ #include "op_x86_model.h" #include "op_counter.h" -#define NUM_COUNTERS 4 -#define NUM_COUNTERS_F15H 6 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX #define NUM_VIRT_COUNTERS 32 #else @@ -70,62 +68,12 @@ static struct ibs_config ibs_config; static struct ibs_state ibs_state; /* - * IBS cpuid feature detection - */ - -#define IBS_CPUID_FEATURES 0x8000001b - -/* - * Same bit mask as for IBS cpuid feature flags (Fn8000_001B_EAX), but - * bit 0 is used to indicate the existence of IBS. - */ -#define IBS_CAPS_AVAIL (1U<<0) -#define IBS_CAPS_FETCHSAM (1U<<1) -#define IBS_CAPS_OPSAM (1U<<2) -#define IBS_CAPS_RDWROPCNT (1U<<3) -#define IBS_CAPS_OPCNT (1U<<4) -#define IBS_CAPS_BRNTRGT (1U<<5) -#define IBS_CAPS_OPCNTEXT (1U<<6) - -#define IBS_CAPS_DEFAULT (IBS_CAPS_AVAIL \ - | IBS_CAPS_FETCHSAM \ - | IBS_CAPS_OPSAM) - -/* - * IBS APIC setup - */ -#define IBSCTL 0x1cc -#define IBSCTL_LVT_OFFSET_VALID (1ULL<<8) -#define IBSCTL_LVT_OFFSET_MASK 0x0F - -/* * IBS randomization macros */ #define IBS_RANDOM_BITS 12 #define IBS_RANDOM_MASK ((1ULL << IBS_RANDOM_BITS) - 1) #define IBS_RANDOM_MAXCNT_OFFSET (1ULL << (IBS_RANDOM_BITS - 5)) -static u32 get_ibs_caps(void) -{ - u32 ibs_caps; - unsigned int max_level; - - if (!boot_cpu_has(X86_FEATURE_IBS)) - return 0; - - /* check IBS cpuid feature flags */ - max_level = cpuid_eax(0x80000000); - if (max_level < IBS_CPUID_FEATURES) - return IBS_CAPS_DEFAULT; - - ibs_caps = cpuid_eax(IBS_CPUID_FEATURES); - if (!(ibs_caps & IBS_CAPS_AVAIL)) - /* cpuid flags not valid */ - return IBS_CAPS_DEFAULT; - - return ibs_caps; -} - /* * 16-bit Linear Feedback Shift Register (LFSR) * @@ -316,81 +264,6 @@ static void op_amd_stop_ibs(void) wrmsrl(MSR_AMD64_IBSOPCTL, 0); } -static inline int get_eilvt(int offset) -{ - return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1); -} - -static inline int put_eilvt(int offset) -{ - return !setup_APIC_eilvt(offset, 0, 0, 1); -} - -static inline int ibs_eilvt_valid(void) -{ - int offset; - u64 val; - int valid = 0; - - preempt_disable(); - - rdmsrl(MSR_AMD64_IBSCTL, val); - offset = val & IBSCTL_LVT_OFFSET_MASK; - - if (!(val & IBSCTL_LVT_OFFSET_VALID)) { - pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n", - smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); - goto out; - } - - if (!get_eilvt(offset)) { - pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n", - smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); - goto out; - } - - valid = 1; -out: - preempt_enable(); - - return valid; -} - -static inline int get_ibs_offset(void) -{ - u64 val; - - rdmsrl(MSR_AMD64_IBSCTL, val); - if (!(val & IBSCTL_LVT_OFFSET_VALID)) - return -EINVAL; - - return val & IBSCTL_LVT_OFFSET_MASK; -} - -static void setup_APIC_ibs(void) -{ - int offset; - - offset = get_ibs_offset(); - if (offset < 0) - goto failed; - - if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0)) - return; -failed: - pr_warn("oprofile: IBS APIC setup failed on cpu #%d\n", - smp_processor_id()); -} - -static void clear_APIC_ibs(void) -{ - int offset; - - offset = get_ibs_offset(); - if (offset >= 0) - setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1); -} - #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX static void op_mux_switch_ctrl(struct op_x86_model_spec const *model, @@ -439,7 +312,7 @@ static int op_amd_fill_in_addresses(struct op_msrs * const msrs) goto fail; } /* both registers must be reserved */ - if (num_counters == NUM_COUNTERS_F15H) { + if (num_counters == AMD64_NUM_COUNTERS_F15H) { msrs->counters[i].addr = MSR_F15H_PERF_CTR + (i << 1); msrs->controls[i].addr = MSR_F15H_PERF_CTL + (i << 1); } else { @@ -504,15 +377,6 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, val |= op_x86_get_ctrl(model, &counter_config[virt]); wrmsrl(msrs->controls[i].addr, val); } - - if (ibs_caps) - setup_APIC_ibs(); -} - -static void op_amd_cpu_shutdown(void) -{ - if (ibs_caps) - clear_APIC_ibs(); } static int op_amd_check_ctrs(struct pt_regs * const regs, @@ -575,86 +439,6 @@ static void op_amd_stop(struct op_msrs const * const msrs) op_amd_stop_ibs(); } -static int setup_ibs_ctl(int ibs_eilvt_off) -{ - struct pci_dev *cpu_cfg; - int nodes; - u32 value = 0; - - nodes = 0; - cpu_cfg = NULL; - do { - cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD, - PCI_DEVICE_ID_AMD_10H_NB_MISC, - cpu_cfg); - if (!cpu_cfg) - break; - ++nodes; - pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off - | IBSCTL_LVT_OFFSET_VALID); - pci_read_config_dword(cpu_cfg, IBSCTL, &value); - if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) { - pci_dev_put(cpu_cfg); - printk(KERN_DEBUG "Failed to setup IBS LVT offset, " - "IBSCTL = 0x%08x\n", value); - return -EINVAL; - } - } while (1); - - if (!nodes) { - printk(KERN_DEBUG "No CPU node configured for IBS\n"); - return -ENODEV; - } - - return 0; -} - -/* - * This runs only on the current cpu. We try to find an LVT offset and - * setup the local APIC. For this we must disable preemption. On - * success we initialize all nodes with this offset. This updates then - * the offset in the IBS_CTL per-node msr. The per-core APIC setup of - * the IBS interrupt vector is called from op_amd_setup_ctrs()/op_- - * amd_cpu_shutdown() using the new offset. - */ -static int force_ibs_eilvt_setup(void) -{ - int offset; - int ret; - - preempt_disable(); - /* find the next free available EILVT entry, skip offset 0 */ - for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) { - if (get_eilvt(offset)) - break; - } - preempt_enable(); - - if (offset == APIC_EILVT_NR_MAX) { - printk(KERN_DEBUG "No EILVT entry available\n"); - return -EBUSY; - } - - ret = setup_ibs_ctl(offset); - if (ret) - goto out; - - if (!ibs_eilvt_valid()) { - ret = -EFAULT; - goto out; - } - - pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset); - pr_err(FW_BUG "workaround enabled for IBS LVT offset\n"); - - return 0; -out: - preempt_disable(); - put_eilvt(offset); - preempt_enable(); - return ret; -} - /* * check and reserve APIC extended interrupt LVT offset for IBS if * available @@ -667,17 +451,6 @@ static void init_ibs(void) if (!ibs_caps) return; - if (ibs_eilvt_valid()) - goto out; - - if (!force_ibs_eilvt_setup()) - goto out; - - /* Failed to setup ibs */ - ibs_caps = 0; - return; - -out: printk(KERN_INFO "oprofile: AMD IBS detected (0x%08x)\n", ibs_caps); } @@ -741,9 +514,9 @@ static int op_amd_init(struct oprofile_operations *ops) ops->create_files = setup_ibs_files; if (boot_cpu_data.x86 == 0x15) { - num_counters = NUM_COUNTERS_F15H; + num_counters = AMD64_NUM_COUNTERS_F15H; } else { - num_counters = NUM_COUNTERS; + num_counters = AMD64_NUM_COUNTERS; } op_amd_spec.num_counters = num_counters; @@ -760,7 +533,6 @@ struct op_x86_model_spec op_amd_spec = { .init = op_amd_init, .fill_in_addresses = &op_amd_fill_in_addresses, .setup_ctrs = &op_amd_setup_ctrs, - .cpu_down = &op_amd_cpu_shutdown, .check_ctrs = &op_amd_check_ctrs, .start = &op_amd_start, .stop = &op_amd_stop, diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 94b745045e45..d90528ea5412 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -28,7 +28,7 @@ static int counter_width = 32; #define MSR_PPRO_EVENTSEL_RESERVED ((0xFFFFFFFFULL<<32)|(1ULL<<21)) -static u64 *reset_value; +static u64 reset_value[OP_MAX_COUNTER]; static void ppro_shutdown(struct op_msrs const * const msrs) { @@ -40,10 +40,6 @@ static void ppro_shutdown(struct op_msrs const * const msrs) release_perfctr_nmi(MSR_P6_PERFCTR0 + i); release_evntsel_nmi(MSR_P6_EVNTSEL0 + i); } - if (reset_value) { - kfree(reset_value); - reset_value = NULL; - } } static int ppro_fill_in_addresses(struct op_msrs * const msrs) @@ -79,13 +75,6 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model, u64 val; int i; - if (!reset_value) { - reset_value = kzalloc(sizeof(reset_value[0]) * num_counters, - GFP_ATOMIC); - if (!reset_value) - return; - } - if (cpu_has_arch_perfmon) { union cpuid10_eax eax; eax.full = cpuid_eax(0xa); @@ -141,13 +130,6 @@ static int ppro_check_ctrs(struct pt_regs * const regs, u64 val; int i; - /* - * This can happen if perf counters are in use when - * we steal the die notifier NMI. - */ - if (unlikely(!reset_value)) - goto out; - for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; @@ -158,7 +140,6 @@ static int ppro_check_ctrs(struct pt_regs * const regs, wrmsrl(msrs->counters[i].addr, -reset_value[i]); } -out: /* Only P6 based Pentium M need to re-unmask the apic vector but it * doesn't hurt other P6 variant */ apic_write(APIC_LVTPC, apic_read(APIC_LVTPC) & ~APIC_LVT_MASKED); @@ -179,8 +160,6 @@ static void ppro_start(struct op_msrs const * const msrs) u64 val; int i; - if (!reset_value) - return; for (i = 0; i < num_counters; ++i) { if (reset_value[i]) { rdmsrl(msrs->controls[i].addr, val); @@ -196,8 +175,6 @@ static void ppro_stop(struct op_msrs const * const msrs) u64 val; int i; - if (!reset_value) - return; for (i = 0; i < num_counters; ++i) { if (!reset_value[i]) continue; @@ -242,7 +219,7 @@ static void arch_perfmon_setup_counters(void) eax.split.bit_width = 40; } - num_counters = eax.split.num_counters; + num_counters = min((int)eax.split.num_counters, OP_MAX_COUNTER); op_arch_perfmon_spec.num_counters = num_counters; op_arch_perfmon_spec.num_controls = num_counters; diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h index 89017fa1fd63..71e8a67337e2 100644 --- a/arch/x86/oprofile/op_x86_model.h +++ b/arch/x86/oprofile/op_x86_model.h @@ -43,7 +43,6 @@ struct op_x86_model_spec { int (*fill_in_addresses)(struct op_msrs * const msrs); void (*setup_ctrs)(struct op_x86_model_spec const *model, struct op_msrs const * const msrs); - void (*cpu_down)(void); int (*check_ctrs)(struct pt_regs * const regs, struct op_msrs const * const msrs); void (*start)(struct op_msrs const * const msrs); diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index 1017c7bee388..492ade8c978e 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -175,8 +175,10 @@ static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) "pcifront-msi-x" : "pcifront-msi", DOMID_SELF); - if (irq < 0) + if (irq < 0) { + ret = irq; goto free; + } i++; } kfree(v); @@ -221,8 +223,10 @@ static int xen_hvm_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) if (msg.data != XEN_PIRQ_MSI_DATA || xen_irq_from_pirq(pirq) < 0) { pirq = xen_allocate_pirq_msi(dev, msidesc); - if (pirq < 0) + if (pirq < 0) { + irq = -ENODEV; goto error; + } xen_msi_compose_msg(dev, pirq, &msg); __write_msi_msg(msidesc, &msg); dev_dbg(&dev->dev, "xen: msi bound to pirq=%d\n", pirq); @@ -244,10 +248,12 @@ static int xen_hvm_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) error: dev_err(&dev->dev, "Xen PCI frontend has not registered MSI/MSI-X support!\n"); - return -ENODEV; + return irq; } #ifdef CONFIG_XEN_DOM0 +static bool __read_mostly pci_seg_supported = true; + static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { int ret = 0; @@ -265,10 +271,11 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) memset(&map_irq, 0, sizeof(map_irq)); map_irq.domid = domid; - map_irq.type = MAP_PIRQ_TYPE_MSI; + map_irq.type = MAP_PIRQ_TYPE_MSI_SEG; map_irq.index = -1; map_irq.pirq = -1; - map_irq.bus = dev->bus->number; + map_irq.bus = dev->bus->number | + (pci_domain_nr(dev->bus) << 16); map_irq.devfn = dev->devfn; if (type == PCI_CAP_ID_MSIX) { @@ -285,7 +292,20 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) map_irq.entry_nr = msidesc->msi_attrib.entry_nr; } - ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, &map_irq); + ret = -EINVAL; + if (pci_seg_supported) + ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, + &map_irq); + if (ret == -EINVAL && !pci_domain_nr(dev->bus)) { + map_irq.type = MAP_PIRQ_TYPE_MSI; + map_irq.index = -1; + map_irq.pirq = -1; + map_irq.bus = dev->bus->number; + ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, + &map_irq); + if (ret != -EINVAL) + pci_seg_supported = false; + } if (ret) { dev_warn(&dev->dev, "xen map irq failed %d for %d domain\n", ret, domid); diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c index 58425adc22c6..e6379526675b 100644 --- a/arch/x86/platform/mrst/mrst.c +++ b/arch/x86/platform/mrst/mrst.c @@ -14,6 +14,8 @@ #include <linux/init.h> #include <linux/kernel.h> +#include <linux/interrupt.h> +#include <linux/scatterlist.h> #include <linux/sfi.h> #include <linux/intel_pmic_gpio.h> #include <linux/spi/spi.h> @@ -392,6 +394,7 @@ static void __init *max3111_platform_data(void *info) struct spi_board_info *spi_info = info; int intr = get_gpio_by_name("max3111_int"); + spi_info->mode = SPI_MODE_0; if (intr == -1) return NULL; spi_info->irq = intr + MRST_IRQ_OFFSET; @@ -678,38 +681,40 @@ static int __init sfi_parse_devs(struct sfi_table_header *table) pentry = (struct sfi_device_table_entry *)sb->pentry; for (i = 0; i < num; i++, pentry++) { - if (pentry->irq != (u8)0xff) { /* native RTE case */ + int irq = pentry->irq; + + if (irq != (u8)0xff) { /* native RTE case */ /* these SPI2 devices are not exposed to system as PCI * devices, but they have separate RTE entry in IOAPIC * so we have to enable them one by one here */ - ioapic = mp_find_ioapic(pentry->irq); + ioapic = mp_find_ioapic(irq); irq_attr.ioapic = ioapic; - irq_attr.ioapic_pin = pentry->irq; + irq_attr.ioapic_pin = irq; irq_attr.trigger = 1; irq_attr.polarity = 1; - io_apic_set_pci_routing(NULL, pentry->irq, &irq_attr); + io_apic_set_pci_routing(NULL, irq, &irq_attr); } else - pentry->irq = 0; /* No irq */ + irq = 0; /* No irq */ switch (pentry->type) { case SFI_DEV_TYPE_IPC: /* ID as IRQ is a hack that will go away */ - pdev = platform_device_alloc(pentry->name, pentry->irq); + pdev = platform_device_alloc(pentry->name, irq); if (pdev == NULL) { pr_err("out of memory for SFI platform device '%s'.\n", pentry->name); continue; } - install_irq_resource(pdev, pentry->irq); + install_irq_resource(pdev, irq); pr_debug("info[%2d]: IPC bus, name = %16.16s, " - "irq = 0x%2x\n", i, pentry->name, pentry->irq); + "irq = 0x%2x\n", i, pentry->name, irq); sfi_handle_ipc_dev(pdev); break; case SFI_DEV_TYPE_SPI: memset(&spi_info, 0, sizeof(spi_info)); strncpy(spi_info.modalias, pentry->name, SFI_NAME_LEN); - spi_info.irq = pentry->irq; + spi_info.irq = irq; spi_info.bus_num = pentry->host_num; spi_info.chip_select = pentry->addr; spi_info.max_speed_hz = pentry->max_freq; @@ -726,7 +731,7 @@ static int __init sfi_parse_devs(struct sfi_table_header *table) memset(&i2c_info, 0, sizeof(i2c_info)); bus = pentry->host_num; strncpy(i2c_info.type, pentry->name, SFI_NAME_LEN); - i2c_info.irq = pentry->irq; + i2c_info.irq = irq; i2c_info.addr = pentry->addr; pr_debug("info[%2d]: I2C bus = %d, name = %16.16s, " "irq = 0x%2x, addr = 0x%x\n", i, bus, diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig index 5cc821cb2e09..26c731a106af 100644 --- a/arch/x86/xen/Kconfig +++ b/arch/x86/xen/Kconfig @@ -25,8 +25,7 @@ config XEN_PRIVILEGED_GUEST config XEN_PVHVM def_bool y - depends on XEN - depends on X86_LOCAL_APIC + depends on XEN && PCI && X86_LOCAL_APIC config XEN_MAX_DOMAIN_MEMORY int @@ -49,11 +48,3 @@ config XEN_DEBUG_FS help Enable statistics output and various tuning options in debugfs. Enabling this option may incur a significant performance overhead. - -config XEN_DEBUG - bool "Enable Xen debug checks" - depends on XEN - default n - help - Enable various WARN_ON checks in the Xen MMU code. - Enabling this option WILL incur a significant performance overhead. diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 2d69617950f7..da8afd576a6b 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -251,6 +251,7 @@ static void __init xen_init_cpuid_mask(void) ~((1 << X86_FEATURE_APIC) | /* disable local APIC */ (1 << X86_FEATURE_ACPI)); /* disable ACPI */ ax = 1; + cx = 0; xen_cpuid(&ax, &bx, &cx, &dx); xsave_mask = diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 3dd53f997b11..87f6673b1207 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -495,41 +495,6 @@ static pte_t xen_make_pte(pteval_t pte) } PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte); -#ifdef CONFIG_XEN_DEBUG -pte_t xen_make_pte_debug(pteval_t pte) -{ - phys_addr_t addr = (pte & PTE_PFN_MASK); - phys_addr_t other_addr; - bool io_page = false; - pte_t _pte; - - if (pte & _PAGE_IOMAP) - io_page = true; - - _pte = xen_make_pte(pte); - - if (!addr) - return _pte; - - if (io_page && - (xen_initial_domain() || addr >= ISA_END_ADDRESS)) { - other_addr = pfn_to_mfn(addr >> PAGE_SHIFT) << PAGE_SHIFT; - WARN_ONCE(addr != other_addr, - "0x%lx is using VM_IO, but it is 0x%lx!\n", - (unsigned long)addr, (unsigned long)other_addr); - } else { - pteval_t iomap_set = (_pte.pte & PTE_FLAGS_MASK) & _PAGE_IOMAP; - other_addr = (_pte.pte & PTE_PFN_MASK); - WARN_ONCE((addr == other_addr) && (!io_page) && (!iomap_set), - "0x%lx is missing VM_IO (and wasn't fixed)!\n", - (unsigned long)addr); - } - - return _pte; -} -PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte_debug); -#endif - static pgd_t xen_make_pgd(pgdval_t pgd) { pgd = pte_pfn_to_mfn(pgd); @@ -1992,9 +1957,6 @@ void __init xen_ident_map_ISA(void) static void __init xen_post_allocator_init(void) { -#ifdef CONFIG_XEN_DEBUG - pv_mmu_ops.make_pte = PV_CALLEE_SAVE(xen_make_pte_debug); -#endif pv_mmu_ops.set_pte = xen_set_pte; pv_mmu_ops.set_pmd = xen_set_pmd; pv_mmu_ops.set_pud = xen_set_pud; @@ -2404,17 +2366,3 @@ out: return err; } EXPORT_SYMBOL_GPL(xen_remap_domain_mfn_range); - -#ifdef CONFIG_XEN_DEBUG_FS -static int p2m_dump_open(struct inode *inode, struct file *filp) -{ - return single_open(filp, p2m_dump_show, NULL); -} - -static const struct file_operations p2m_dump_fops = { - .open = p2m_dump_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; -#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index 58efeb9d5440..1b267e75158d 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -161,7 +161,9 @@ #include <asm/xen/page.h> #include <asm/xen/hypercall.h> #include <asm/xen/hypervisor.h> +#include <xen/grant_table.h> +#include "multicalls.h" #include "xen-ops.h" static void __init m2p_override_init(void); @@ -676,7 +678,8 @@ static unsigned long mfn_hash(unsigned long mfn) } /* Add an MFN override for a particular page */ -int m2p_add_override(unsigned long mfn, struct page *page, bool clear_pte) +int m2p_add_override(unsigned long mfn, struct page *page, + struct gnttab_map_grant_ref *kmap_op) { unsigned long flags; unsigned long pfn; @@ -692,16 +695,28 @@ int m2p_add_override(unsigned long mfn, struct page *page, bool clear_pte) "m2p_add_override: pfn %lx not mapped", pfn)) return -EINVAL; } - - page->private = mfn; + WARN_ON(PagePrivate(page)); + SetPagePrivate(page); + set_page_private(page, mfn); page->index = pfn_to_mfn(pfn); if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) return -ENOMEM; - if (clear_pte && !PageHighMem(page)) - /* Just zap old mapping for now */ - pte_clear(&init_mm, address, ptep); + if (kmap_op != NULL) { + if (!PageHighMem(page)) { + struct multicall_space mcs = + xen_mc_entry(sizeof(*kmap_op)); + + MULTI_grant_table_op(mcs.mc, + GNTTABOP_map_grant_ref, kmap_op, 1); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + } + /* let's use dev_bus_addr to record the old mfn instead */ + kmap_op->dev_bus_addr = page->index; + page->index = (unsigned long) kmap_op; + } spin_lock_irqsave(&m2p_override_lock, flags); list_add(&page->lru, &m2p_overrides[mfn_hash(mfn)]); spin_unlock_irqrestore(&m2p_override_lock, flags); @@ -735,13 +750,56 @@ int m2p_remove_override(struct page *page, bool clear_pte) spin_lock_irqsave(&m2p_override_lock, flags); list_del(&page->lru); spin_unlock_irqrestore(&m2p_override_lock, flags); - set_phys_to_machine(pfn, page->index); + WARN_ON(!PagePrivate(page)); + ClearPagePrivate(page); - if (clear_pte && !PageHighMem(page)) - set_pte_at(&init_mm, address, ptep, - pfn_pte(pfn, PAGE_KERNEL)); - /* No tlb flush necessary because the caller already - * left the pte unmapped. */ + if (clear_pte) { + struct gnttab_map_grant_ref *map_op = + (struct gnttab_map_grant_ref *) page->index; + set_phys_to_machine(pfn, map_op->dev_bus_addr); + if (!PageHighMem(page)) { + struct multicall_space mcs; + struct gnttab_unmap_grant_ref *unmap_op; + + /* + * It might be that we queued all the m2p grant table + * hypercalls in a multicall, then m2p_remove_override + * get called before the multicall has actually been + * issued. In this case handle is going to -1 because + * it hasn't been modified yet. + */ + if (map_op->handle == -1) + xen_mc_flush(); + /* + * Now if map_op->handle is negative it means that the + * hypercall actually returned an error. + */ + if (map_op->handle == GNTST_general_error) { + printk(KERN_WARNING "m2p_remove_override: " + "pfn %lx mfn %lx, failed to modify kernel mappings", + pfn, mfn); + return -1; + } + + mcs = xen_mc_entry( + sizeof(struct gnttab_unmap_grant_ref)); + unmap_op = mcs.args; + unmap_op->host_addr = map_op->host_addr; + unmap_op->handle = map_op->handle; + unmap_op->dev_bus_addr = 0; + + MULTI_grant_table_op(mcs.mc, + GNTTABOP_unmap_grant_ref, unmap_op, 1); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + set_pte_at(&init_mm, address, ptep, + pfn_pte(pfn, PAGE_KERNEL)); + __flush_tlb_single(address); + map_op->host_addr = 0; + } + } else + set_phys_to_machine(pfn, page->index); return 0; } @@ -758,7 +816,7 @@ struct page *m2p_find_override(unsigned long mfn) spin_lock_irqsave(&m2p_override_lock, flags); list_for_each_entry(p, bucket, lru) { - if (p->private == mfn) { + if (page_private(p) == mfn) { ret = p; break; } @@ -782,17 +840,21 @@ unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn) EXPORT_SYMBOL_GPL(m2p_find_override_pfn); #ifdef CONFIG_XEN_DEBUG_FS - -int p2m_dump_show(struct seq_file *m, void *v) +#include <linux/debugfs.h> +#include "debugfs.h" +static int p2m_dump_show(struct seq_file *m, void *v) { static const char * const level_name[] = { "top", "middle", - "entry", "abnormal" }; - static const char * const type_name[] = { "identity", "missing", - "pfn", "abnormal"}; + "entry", "abnormal", "error"}; #define TYPE_IDENTITY 0 #define TYPE_MISSING 1 #define TYPE_PFN 2 #define TYPE_UNKNOWN 3 + static const char * const type_name[] = { + [TYPE_IDENTITY] = "identity", + [TYPE_MISSING] = "missing", + [TYPE_PFN] = "pfn", + [TYPE_UNKNOWN] = "abnormal"}; unsigned long pfn, prev_pfn_type = 0, prev_pfn_level = 0; unsigned int uninitialized_var(prev_level); unsigned int uninitialized_var(prev_type); @@ -856,4 +918,32 @@ int p2m_dump_show(struct seq_file *m, void *v) #undef TYPE_PFN #undef TYPE_UNKNOWN } -#endif + +static int p2m_dump_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, p2m_dump_show, NULL); +} + +static const struct file_operations p2m_dump_fops = { + .open = p2m_dump_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static struct dentry *d_mmu_debug; + +static int __init xen_p2m_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + d_mmu_debug = debugfs_create_dir("mmu", d_xen); + + debugfs_create_file("p2m", 0600, d_mmu_debug, NULL, &p2m_dump_fops); + return 0; +} +fs_initcall(xen_p2m_debugfs); +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 46d6d21dbdbe..38d0af4fefec 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -37,7 +37,10 @@ extern void xen_syscall_target(void); extern void xen_syscall32_target(void); /* Amount of extra memory space we add to the e820 ranges */ -phys_addr_t xen_extra_mem_start, xen_extra_mem_size; +struct xen_memory_region xen_extra_mem[XEN_EXTRA_MEM_MAX_REGIONS] __initdata; + +/* Number of pages released from the initial allocation. */ +unsigned long xen_released_pages; /* * The maximum amount of extra memory compared to the base size. The @@ -51,48 +54,47 @@ phys_addr_t xen_extra_mem_start, xen_extra_mem_size; */ #define EXTRA_MEM_RATIO (10) -static void __init xen_add_extra_mem(unsigned long pages) +static void __init xen_add_extra_mem(u64 start, u64 size) { unsigned long pfn; + int i; - u64 size = (u64)pages * PAGE_SIZE; - u64 extra_start = xen_extra_mem_start + xen_extra_mem_size; - - if (!pages) - return; - - e820_add_region(extra_start, size, E820_RAM); - sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); - - memblock_x86_reserve_range(extra_start, extra_start + size, "XEN EXTRA"); + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) { + /* Add new region. */ + if (xen_extra_mem[i].size == 0) { + xen_extra_mem[i].start = start; + xen_extra_mem[i].size = size; + break; + } + /* Append to existing region. */ + if (xen_extra_mem[i].start + xen_extra_mem[i].size == start) { + xen_extra_mem[i].size += size; + break; + } + } + if (i == XEN_EXTRA_MEM_MAX_REGIONS) + printk(KERN_WARNING "Warning: not enough extra memory regions\n"); - xen_extra_mem_size += size; + memblock_x86_reserve_range(start, start + size, "XEN EXTRA"); - xen_max_p2m_pfn = PFN_DOWN(extra_start + size); + xen_max_p2m_pfn = PFN_DOWN(start + size); - for (pfn = PFN_DOWN(extra_start); pfn <= xen_max_p2m_pfn; pfn++) + for (pfn = PFN_DOWN(start); pfn <= xen_max_p2m_pfn; pfn++) __set_phys_to_machine(pfn, INVALID_P2M_ENTRY); } -static unsigned long __init xen_release_chunk(phys_addr_t start_addr, - phys_addr_t end_addr) +static unsigned long __init xen_release_chunk(unsigned long start, + unsigned long end) { struct xen_memory_reservation reservation = { .address_bits = 0, .extent_order = 0, .domid = DOMID_SELF }; - unsigned long start, end; unsigned long len = 0; unsigned long pfn; int ret; - start = PFN_UP(start_addr); - end = PFN_DOWN(end_addr); - - if (end <= start) - return 0; - for(pfn = start; pfn < end; pfn++) { unsigned long mfn = pfn_to_mfn(pfn); @@ -117,72 +119,52 @@ static unsigned long __init xen_release_chunk(phys_addr_t start_addr, return len; } -static unsigned long __init xen_return_unused_memory(unsigned long max_pfn, - const struct e820map *e820) +static unsigned long __init xen_set_identity_and_release( + const struct e820entry *list, size_t map_size, unsigned long nr_pages) { - phys_addr_t max_addr = PFN_PHYS(max_pfn); - phys_addr_t last_end = ISA_END_ADDRESS; + phys_addr_t start = 0; unsigned long released = 0; - int i; - - /* Free any unused memory above the low 1Mbyte. */ - for (i = 0; i < e820->nr_map && last_end < max_addr; i++) { - phys_addr_t end = e820->map[i].addr; - end = min(max_addr, end); - - if (last_end < end) - released += xen_release_chunk(last_end, end); - last_end = max(last_end, e820->map[i].addr + e820->map[i].size); - } - - if (last_end < max_addr) - released += xen_release_chunk(last_end, max_addr); - - printk(KERN_INFO "released %lu pages of unused memory\n", released); - return released; -} - -static unsigned long __init xen_set_identity(const struct e820entry *list, - ssize_t map_size) -{ - phys_addr_t last = xen_initial_domain() ? 0 : ISA_END_ADDRESS; - phys_addr_t start_pci = last; - const struct e820entry *entry; unsigned long identity = 0; + const struct e820entry *entry; int i; + /* + * Combine non-RAM regions and gaps until a RAM region (or the + * end of the map) is reached, then set the 1:1 map and + * release the pages (if available) in those non-RAM regions. + * + * The combined non-RAM regions are rounded to a whole number + * of pages so any partial pages are accessible via the 1:1 + * mapping. This is needed for some BIOSes that put (for + * example) the DMI tables in a reserved region that begins on + * a non-page boundary. + */ for (i = 0, entry = list; i < map_size; i++, entry++) { - phys_addr_t start = entry->addr; - phys_addr_t end = start + entry->size; + phys_addr_t end = entry->addr + entry->size; - if (start < last) - start = last; + if (entry->type == E820_RAM || i == map_size - 1) { + unsigned long start_pfn = PFN_DOWN(start); + unsigned long end_pfn = PFN_UP(end); - if (end <= start) - continue; + if (entry->type == E820_RAM) + end_pfn = PFN_UP(entry->addr); - /* Skip over the 1MB region. */ - if (last > end) - continue; + if (start_pfn < end_pfn) { + if (start_pfn < nr_pages) + released += xen_release_chunk( + start_pfn, min(end_pfn, nr_pages)); - if ((entry->type == E820_RAM) || (entry->type == E820_UNUSABLE)) { - if (start > start_pci) identity += set_phys_range_identity( - PFN_UP(start_pci), PFN_DOWN(start)); - - /* Without saving 'last' we would gooble RAM too - * at the end of the loop. */ - last = end; - start_pci = end; - continue; + start_pfn, end_pfn); + } + start = end; } - start_pci = min(start, start_pci); - last = end; } - if (last > start_pci) - identity += set_phys_range_identity( - PFN_UP(start_pci), PFN_DOWN(last)); - return identity; + + printk(KERN_INFO "Released %lu pages of unused memory\n", released); + printk(KERN_INFO "Set %ld page(s) to 1-1 mapping\n", identity); + + return released; } static unsigned long __init xen_get_max_pages(void) @@ -197,21 +179,32 @@ static unsigned long __init xen_get_max_pages(void) return min(max_pages, MAX_DOMAIN_PAGES); } +static void xen_align_and_add_e820_region(u64 start, u64 size, int type) +{ + u64 end = start + size; + + /* Align RAM regions to page boundaries. */ + if (type == E820_RAM) { + start = PAGE_ALIGN(start); + end &= ~((u64)PAGE_SIZE - 1); + } + + e820_add_region(start, end - start, type); +} + /** * machine_specific_memory_setup - Hook for machine specific memory setup. **/ char * __init xen_memory_setup(void) { static struct e820entry map[E820MAX] __initdata; - static struct e820entry map_raw[E820MAX] __initdata; unsigned long max_pfn = xen_start_info->nr_pages; unsigned long long mem_end; int rc; struct xen_memory_map memmap; + unsigned long max_pages; unsigned long extra_pages = 0; - unsigned long extra_limit; - unsigned long identity_pages = 0; int i; int op; @@ -237,58 +230,65 @@ char * __init xen_memory_setup(void) } BUG_ON(rc); - memcpy(map_raw, map, sizeof(map)); - e820.nr_map = 0; - xen_extra_mem_start = mem_end; - for (i = 0; i < memmap.nr_entries; i++) { - unsigned long long end; - - /* Guard against non-page aligned E820 entries. */ - if (map[i].type == E820_RAM) - map[i].size -= (map[i].size + map[i].addr) % PAGE_SIZE; - - end = map[i].addr + map[i].size; - if (map[i].type == E820_RAM && end > mem_end) { - /* RAM off the end - may be partially included */ - u64 delta = min(map[i].size, end - mem_end); - - map[i].size -= delta; - end -= delta; - - extra_pages += PFN_DOWN(delta); - /* - * Set RAM below 4GB that is not for us to be unusable. - * This prevents "System RAM" address space from being - * used as potential resource for I/O address (happens - * when 'allocate_resource' is called). - */ - if (delta && - (xen_initial_domain() && end < 0x100000000ULL)) - e820_add_region(end, delta, E820_UNUSABLE); + /* Make sure the Xen-supplied memory map is well-ordered. */ + sanitize_e820_map(map, memmap.nr_entries, &memmap.nr_entries); + + max_pages = xen_get_max_pages(); + if (max_pages > max_pfn) + extra_pages += max_pages - max_pfn; + + /* + * Set P2M for all non-RAM pages and E820 gaps to be identity + * type PFNs. Any RAM pages that would be made inaccesible by + * this are first released. + */ + xen_released_pages = xen_set_identity_and_release( + map, memmap.nr_entries, max_pfn); + extra_pages += xen_released_pages; + + /* + * Clamp the amount of extra memory to a EXTRA_MEM_RATIO + * factor the base size. On non-highmem systems, the base + * size is the full initial memory allocation; on highmem it + * is limited to the max size of lowmem, so that it doesn't + * get completely filled. + * + * In principle there could be a problem in lowmem systems if + * the initial memory is also very large with respect to + * lowmem, but we won't try to deal with that here. + */ + extra_pages = min(EXTRA_MEM_RATIO * min(max_pfn, PFN_DOWN(MAXMEM)), + extra_pages); + + i = 0; + while (i < memmap.nr_entries) { + u64 addr = map[i].addr; + u64 size = map[i].size; + u32 type = map[i].type; + + if (type == E820_RAM) { + if (addr < mem_end) { + size = min(size, mem_end - addr); + } else if (extra_pages) { + size = min(size, (u64)extra_pages * PAGE_SIZE); + extra_pages -= size / PAGE_SIZE; + xen_add_extra_mem(addr, size); + } else + type = E820_UNUSABLE; } - if (map[i].size > 0 && end > xen_extra_mem_start) - xen_extra_mem_start = end; + xen_align_and_add_e820_region(addr, size, type); - /* Add region if any remains */ - if (map[i].size > 0) - e820_add_region(map[i].addr, map[i].size, map[i].type); + map[i].addr += size; + map[i].size -= size; + if (map[i].size == 0) + i++; } - /* Align the balloon area so that max_low_pfn does not get set - * to be at the _end_ of the PCI gap at the far end (fee01000). - * Note that xen_extra_mem_start gets set in the loop above to be - * past the last E820 region. */ - if (xen_initial_domain() && (xen_extra_mem_start < (1ULL<<32))) - xen_extra_mem_start = (1ULL<<32); /* * In domU, the ISA region is normal, usable memory, but we * reserve ISA memory anyway because too many things poke * about in there. - * - * In Dom0, the host E820 information can leave gaps in the - * ISA range, which would cause us to release those pages. To - * avoid this, we unconditionally reserve them here. */ e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS, E820_RESERVED); @@ -305,44 +305,6 @@ char * __init xen_memory_setup(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); - extra_limit = xen_get_max_pages(); - if (max_pfn + extra_pages > extra_limit) { - if (extra_limit > max_pfn) - extra_pages = extra_limit - max_pfn; - else - extra_pages = 0; - } - - extra_pages += xen_return_unused_memory(xen_start_info->nr_pages, &e820); - - /* - * Clamp the amount of extra memory to a EXTRA_MEM_RATIO - * factor the base size. On non-highmem systems, the base - * size is the full initial memory allocation; on highmem it - * is limited to the max size of lowmem, so that it doesn't - * get completely filled. - * - * In principle there could be a problem in lowmem systems if - * the initial memory is also very large with respect to - * lowmem, but we won't try to deal with that here. - */ - extra_limit = min(EXTRA_MEM_RATIO * min(max_pfn, PFN_DOWN(MAXMEM)), - max_pfn + extra_pages); - - if (extra_limit >= max_pfn) - extra_pages = extra_limit - max_pfn; - else - extra_pages = 0; - - xen_add_extra_mem(extra_pages); - - /* - * Set P2M for all non-RAM pages and E820 gaps to be identity - * type PFNs. We supply it with the non-sanitized version - * of the E820. - */ - identity_pages = xen_set_identity(map_raw, memmap.nr_entries); - printk(KERN_INFO "Set %ld page(s) to 1-1 mapping.\n", identity_pages); return "Xen"; } |