Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block layer fixes from Jens Axboe:
 "Final small batch of fixes to be included before -rc1.  Some general
  cleanups in here as well, but some of the blk-mq fixes we need for the
  NVMe conversion and/or scsi-mq.  The pull request contains:

   - Support for not merging across a specified "chunk size", if set by
     the driver.  Some NVMe devices perform poorly for IO that crosses
     such a chunk, so we need to support it generically as part of
     request merging avoid having to do complicated split logic.  From
     me.

   - Bump max tag depth to 10Ki tags.  Some scsi devices have a huge
     shared tag space.  Before we failed with EINVAL if a too large tag
     depth was specified, now we truncate it and pass back the actual
     value.  From me.

   - Various blk-mq rq init fixes from me and others.

   - A fix for enter on a dying queue for blk-mq from Keith.  This is
     needed to prevent oopsing on hot device removal.

   - Fixup for blk-mq timer addition from Ming Lei.

   - Small round of performance fixes for mtip32xx from Sam Bradshaw.

   - Minor stack leak fix from Rickard Strandqvist.

   - Two __init annotations from Fabian Frederick"

* 'for-linus' of git://git.kernel.dk/linux-block:
  block: add __init to blkcg_policy_register
  block: add __init to elv_register
  block: ensure that bio_add_page() always accepts a page for an empty bio
  blk-mq: add timer in blk_mq_start_request
  blk-mq: always initialize request->start_time
  block: blk-exec.c: Cleaning up local variable address returnd
  mtip32xx: minor performance enhancements
  blk-mq: ->timeout should be cleared in blk_mq_rq_ctx_init()
  blk-mq: don't allow queue entering for a dying queue
  blk-mq: bump max tag depth to 10K tags
  block: add blk_rq_set_block_pc()
  block: add notion of a chunk size for request merging

15 files changed, 30 insertions, 26 deletions

diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 74abd49fabdc..295f3afbbef5 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -39,6 +39,7 @@
 #include <../drivers/ata/ahci.h>
 #include <linux/export.h>
 #include <linux/debugfs.h>
+#include <linux/prefetch.h>
 #include "mtip32xx.h"
 
 #define HW_CMD_SLOT_SZ		(MTIP_MAX_COMMAND_SLOTS * 32)
@@ -2380,6 +2381,8 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	/* Map the scatter list for DMA access */
 	nents = dma_map_sg(&dd->pdev->dev, command->sg, nents, dma_dir);
 
+	prefetch(&port->flags);
+
 	command->scatter_ents = nents;
 
 	/*
@@ -2392,7 +2395,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	fis = command->command;
 	fis->type        = 0x27;
 	fis->opts        = 1 << 7;
-	if (rq_data_dir(rq) == READ)
+	if (dma_dir == DMA_FROM_DEVICE)
 		fis->command = ATA_CMD_FPDMA_READ;
 	else
 		fis->command = ATA_CMD_FPDMA_WRITE;
@@ -2412,7 +2415,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	fis->res3        = 0;
 	fill_command_sg(dd, command, nents);
 
-	if (command->unaligned)
+	if (unlikely(command->unaligned))
 		fis->device |= 1 << 7;
 
 	/* Populate the command header */
@@ -2433,7 +2436,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	 * To prevent this command from being issued
 	 * if an internal command is in progress or error handling is active.
 	 */
-	if (port->flags & MTIP_PF_PAUSE_IO) {
+	if (unlikely(port->flags & MTIP_PF_PAUSE_IO)) {
 		set_bit(rq->tag, port->cmds_to_issue);
 		set_bit(MTIP_PF_ISSUE_CMDS_BIT, &port->flags);
 		return;
@@ -3754,7 +3757,7 @@ static bool mtip_check_unal_depth(struct blk_mq_hw_ctx *hctx,
 	struct driver_data *dd = hctx->queue->queuedata;
 	struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
 
-	if (!dd->unal_qdepth || rq_data_dir(rq) == READ)
+	if (rq_data_dir(rq) == READ || !dd->unal_qdepth)
 		return false;
 
 	/*
@@ -3776,11 +3779,11 @@ static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq)
 {
 	int ret;
 
-	if (mtip_check_unal_depth(hctx, rq))
+	if (unlikely(mtip_check_unal_depth(hctx, rq)))
 		return BLK_MQ_RQ_QUEUE_BUSY;
 
 	ret = mtip_submit_request(hctx, rq);
-	if (!ret)
+	if (likely(!ret))
 		return BLK_MQ_RQ_QUEUE_OK;
 
 	rq->errors = ret;
diff --git a/drivers/block/mtip32xx/mtip32xx.h b/drivers/block/mtip32xx/mtip32xx.h
index 4b9b554234bc..ba1b31ee22ec 100644
--- a/drivers/block/mtip32xx/mtip32xx.h
+++ b/drivers/block/mtip32xx/mtip32xx.h
@@ -493,19 +493,19 @@ struct driver_data {
 
 	struct workqueue_struct *isr_workq;
 
-	struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
-
 	atomic_t irq_workers_active;
 
+	struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
+
 	int isr_binding;
 
 	struct block_device *bdev;
 
-	int unal_qdepth; /* qdepth of unaligned IO queue */
-
 	struct list_head online_list; /* linkage for online list */
 
 	struct list_head remove_list; /* linkage for removing list */
+
+	int unal_qdepth; /* qdepth of unaligned IO queue */
 };
 
 #endif
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index ef166ad2dbad..758ac442c5b5 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -704,6 +704,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 
 	rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
 			     WRITE : READ, __GFP_WAIT);
+	blk_rq_set_block_pc(rq);
 
 	if (cgc->buflen) {
 		ret = blk_rq_map_kern(q, rq, cgc->buffer, cgc->buflen,
@@ -716,7 +717,6 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	memcpy(rq->cmd, cgc->cmd, CDROM_PACKET_SIZE);
 
 	rq->timeout = 60*HZ;
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	if (cgc->quiet)
 		rq->cmd_flags |= REQ_QUIET;
 
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index 2a44767891f5..898b84bba28a 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -2184,6 +2184,7 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 			ret = -ENOMEM;
 			break;
 		}
+		blk_rq_set_block_pc(rq);
 
 		ret = blk_rq_map_user(q, rq, NULL, ubuf, len, GFP_KERNEL);
 		if (ret) {
@@ -2203,7 +2204,6 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		rq->cmd[9] = 0xf8;
 
 		rq->cmd_len = 12;
-		rq->cmd_type = REQ_TYPE_BLOCK_PC;
 		rq->timeout = 60 * HZ;
 		bio = rq->bio;
 
diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
index 5248c888552b..7bcf67eec921 100644
--- a/drivers/scsi/device_handler/scsi_dh_alua.c
+++ b/drivers/scsi/device_handler/scsi_dh_alua.c
@@ -120,6 +120,7 @@ static struct request *get_alua_req(struct scsi_device *sdev,
 			    "%s: blk_get_request failed\n", __func__);
 		return NULL;
 	}
+	blk_rq_set_block_pc(rq);
 
 	if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
 		blk_put_request(rq);
@@ -128,7 +129,6 @@ static struct request *get_alua_req(struct scsi_device *sdev,
 		return NULL;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->retries = ALUA_FAILOVER_RETRIES;
diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c
index e1c8be06de9d..6f07f7fe3aa1 100644
--- a/drivers/scsi/device_handler/scsi_dh_emc.c
+++ b/drivers/scsi/device_handler/scsi_dh_emc.c
@@ -280,6 +280,7 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
 		return NULL;
 	}
 
+	blk_rq_set_block_pc(rq);
 	rq->cmd_len = COMMAND_SIZE(cmd);
 	rq->cmd[0] = cmd;
 
@@ -304,7 +305,6 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
 		break;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->timeout = CLARIION_TIMEOUT;
diff --git a/drivers/scsi/device_handler/scsi_dh_hp_sw.c b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
index 084062bb8ee9..e9d9fea9e272 100644
--- a/drivers/scsi/device_handler/scsi_dh_hp_sw.c
+++ b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
@@ -120,7 +120,7 @@ retry:
 	if (!req)
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			  REQ_FAILFAST_DRIVER;
 	req->cmd_len = COMMAND_SIZE(TEST_UNIT_READY);
@@ -250,7 +250,7 @@ static int hp_sw_start_stop(struct hp_sw_dh_data *h)
 	if (!req)
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			  REQ_FAILFAST_DRIVER;
 	req->cmd_len = COMMAND_SIZE(START_STOP);
diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c
index 4b9cf93f3fb6..826069db9848 100644
--- a/drivers/scsi/device_handler/scsi_dh_rdac.c
+++ b/drivers/scsi/device_handler/scsi_dh_rdac.c
@@ -279,6 +279,7 @@ static struct request *get_rdac_req(struct scsi_device *sdev,
 				"get_rdac_req: blk_get_request failed.\n");
 		return NULL;
 	}
+	blk_rq_set_block_pc(rq);
 
 	if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
 		blk_put_request(rq);
@@ -287,7 +288,6 @@ static struct request *get_rdac_req(struct scsi_device *sdev,
 		return NULL;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->retries = RDAC_RETRIES;
diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c
index bac04c2335aa..5f4cbf0c4759 100644
--- a/drivers/scsi/osd/osd_initiator.c
+++ b/drivers/scsi/osd/osd_initiator.c
@@ -1570,6 +1570,7 @@ static struct request *_make_request(struct request_queue *q, bool has_write,
 		if (unlikely(!req))
 			return ERR_PTR(-ENOMEM);
 
+		blk_rq_set_block_pc(req);
 		return req;
 	}
 }
@@ -1590,7 +1591,6 @@ static int _init_blk_request(struct osd_request *or,
 	}
 
 	or->request = req;
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= REQ_QUIET;
 
 	req->timeout = or->timeout;
@@ -1608,7 +1608,7 @@ static int _init_blk_request(struct osd_request *or,
 				ret = PTR_ERR(req);
 				goto out;
 			}
-			req->cmd_type = REQ_TYPE_BLOCK_PC;
+			blk_rq_set_block_pc(req);
 			or->in.req = or->request->next_rq = req;
 		}
 	} else if (has_in)
diff --git a/drivers/scsi/osst.c b/drivers/scsi/osst.c
index 21883a2d6324..0727ea7cc387 100644
--- a/drivers/scsi/osst.c
+++ b/drivers/scsi/osst.c
@@ -365,7 +365,7 @@ static int osst_execute(struct osst_request *SRpnt, const unsigned char *cmd,
 	if (!req)
 		return DRIVER_ERROR << 24;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_QUIET;
 
 	SRpnt->bio = NULL;
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 47a1ffc4c904..cbe38e5e7955 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1952,6 +1952,8 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 	 */
 	req = blk_get_request(sdev->request_queue, READ, GFP_KERNEL);
 
+	blk_rq_set_block_pc(req);
+
 	req->cmd[0] = ALLOW_MEDIUM_REMOVAL;
 	req->cmd[1] = 0;
 	req->cmd[2] = 0;
@@ -1961,7 +1963,6 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 
 	req->cmd_len = COMMAND_SIZE(req->cmd[0]);
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= REQ_QUIET;
 	req->timeout = 10 * HZ;
 	req->retries = 5;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index be0d5fad999d..f7e316368c99 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -195,6 +195,7 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	req = blk_get_request(sdev->request_queue, write, __GFP_WAIT);
 	if (!req)
 		return ret;
+	blk_rq_set_block_pc(req);
 
 	if (bufflen &&	blk_rq_map_kern(sdev->request_queue, req,
 					buffer, bufflen, __GFP_WAIT))
@@ -206,7 +207,6 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	req->sense_len = 0;
 	req->retries = retries;
 	req->timeout = timeout;
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= flags | REQ_QUIET | REQ_PREEMPT;
 
 	/*
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index df5e961484e1..53268aaba559 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1653,10 +1653,9 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd)
 	if (!rq)
 		return -ENOMEM;
 
+	blk_rq_set_block_pc(rq);
 	memcpy(rq->cmd, cmd, hp->cmd_len);
-
 	rq->cmd_len = hp->cmd_len;
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 
 	srp->rq = rq;
 	rq->end_io_data = srp;
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index afc834e172c6..14eb4b256a03 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -484,7 +484,7 @@ static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
 	if (!req)
 		return DRIVER_ERROR << 24;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_QUIET;
 
 	mdata->null_mapped = 1;
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index 0f199f6a0738..94d00df28f39 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -1055,6 +1055,8 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 			ret = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 			goto fail;
 		}
+
+		blk_rq_set_block_pc(req);
 	} else {
 		BUG_ON(!cmd->data_length);
 
@@ -1071,7 +1073,6 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 		}
 	}
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->end_io = pscsi_req_done;
 	req->end_io_data = cmd;
 	req->cmd_len = scsi_command_size(pt->pscsi_cdb);