Merge branch 'for-3.15/core' of git://git.kernel.dk/linux-block

Pull core block layer updates from Jens Axboe:
 "This is the pull request for the core block IO bits for the 3.15
  kernel.  It's a smaller round this time, it contains:

   - Various little blk-mq fixes and additions from Christoph and
     myself.

   - Cleanup of the IPI usage from the block layer, and associated
     helper code.  From Frederic Weisbecker and Jan Kara.

   - Duplicate code cleanup in bio-integrity from Gu Zheng.  This will
     give you a merge conflict, but that should be easy to resolve.

   - blk-mq notify spinlock fix for RT from Mike Galbraith.

   - A blktrace partial accounting bug fix from Roman Pen.

   - Missing REQ_SYNC detection fix for blk-mq from Shaohua Li"

* 'for-3.15/core' of git://git.kernel.dk/linux-block: (25 commits)
  blk-mq: add REQ_SYNC early
  rt,blk,mq: Make blk_mq_cpu_notify_lock a raw spinlock
  blk-mq: support partial I/O completions
  blk-mq: merge blk_mq_insert_request and blk_mq_run_request
  blk-mq: remove blk_mq_alloc_rq
  blk-mq: don't dump CPU -> hw queue map on driver load
  blk-mq: fix wrong usage of hctx->state vs hctx->flags
  blk-mq: allow blk_mq_init_commands() to return failure
  block: remove old blk_iopoll_enabled variable
  blktrace: fix accounting of partially completed requests
  smp: Rename __smp_call_function_single() to smp_call_function_single_async()
  smp: Remove wait argument from __smp_call_function_single()
  watchdog: Simplify a little the IPI call
  smp: Move __smp_call_function_single() below its safe version
  smp: Consolidate the various smp_call_function_single() declensions
  smp: Teach __smp_call_function_single() to check for offline cpus
  smp: Remove unused list_head from csd
  smp: Iterate functions through llist_for_each_entry_safe()
  block: Stop abusing rq->csd.list in blk-softirq
  block: Remove useless IPI struct initialization
  ...

14 files changed, 125 insertions, 48 deletions

diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 4e491d9b5292..b6e95b5e262f 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -336,7 +336,7 @@ static void blkg_destroy(struct blkcg_gq *blkg)
 	 * under queue_lock.  If it's not pointing to @blkg now, it never
 	 * will.  Hint assignment itself can race safely.
 	 */
-	if (rcu_dereference_raw(blkcg->blkg_hint) == blkg)
+	if (rcu_access_pointer(blkcg->blkg_hint) == blkg)
 		rcu_assign_pointer(blkcg->blkg_hint, NULL);
 
 	/*
diff --git a/block/blk-core.c b/block/blk-core.c
index bfe16d5af9f9..e45b321cf6a0 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2353,7 +2353,7 @@ bool blk_update_request(struct request *req, int error, unsigned int nr_bytes)
 	if (!req->bio)
 		return false;
 
-	trace_block_rq_complete(req->q, req);
+	trace_block_rq_complete(req->q, req, nr_bytes);
 
 	/*
 	 * For fs requests, rq is just carrier of independent bio's
diff --git a/block/blk-ioc.c b/block/blk-ioc.c
index 242df01413f6..1a27f45ec776 100644
--- a/block/blk-ioc.c
+++ b/block/blk-ioc.c
@@ -68,7 +68,7 @@ static void ioc_destroy_icq(struct io_cq *icq)
 	 * under queue_lock.  If it's not pointing to @icq now, it never
 	 * will.  Hint assignment itself can race safely.
 	 */
-	if (rcu_dereference_raw(ioc->icq_hint) == icq)
+	if (rcu_access_pointer(ioc->icq_hint) == icq)
 		rcu_assign_pointer(ioc->icq_hint, NULL);
 
 	ioc_exit_icq(icq);
diff --git a/block/blk-iopoll.c b/block/blk-iopoll.c
index 1855bf51edb0..c11d24e379e2 100644
--- a/block/blk-iopoll.c
+++ b/block/blk-iopoll.c
@@ -14,9 +14,6 @@
 
 #include "blk.h"
 
-int blk_iopoll_enabled = 1;
-EXPORT_SYMBOL(blk_iopoll_enabled);
-
 static unsigned int blk_iopoll_budget __read_mostly = 256;
 
 static DEFINE_PER_CPU(struct list_head, blk_cpu_iopoll);
diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c
index f8721278601c..097921329619 100644
--- a/block/blk-mq-cpumap.c
+++ b/block/blk-mq-cpumap.c
@@ -9,15 +9,6 @@
 #include "blk.h"
 #include "blk-mq.h"
 
-static void show_map(unsigned int *map, unsigned int nr)
-{
-	int i;
-
-	pr_info("blk-mq: CPU -> queue map\n");
-	for_each_online_cpu(i)
-		pr_info("  CPU%2u -> Queue %u\n", i, map[i]);
-}
-
 static int cpu_to_queue_index(unsigned int nr_cpus, unsigned int nr_queues,
 			      const int cpu)
 {
@@ -85,7 +76,6 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues)
 			map[i] = map[first_sibling];
 	}
 
-	show_map(map, nr_cpus);
 	free_cpumask_var(cpus);
 	return 0;
 }
diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c
index b91ce75bd35d..b0ba264b0522 100644
--- a/block/blk-mq-sysfs.c
+++ b/block/blk-mq-sysfs.c
@@ -244,6 +244,32 @@ static ssize_t blk_mq_hw_sysfs_tags_show(struct blk_mq_hw_ctx *hctx, char *page)
 	return blk_mq_tag_sysfs_show(hctx->tags, page);
 }
 
+static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page)
+{
+	unsigned int i, queue_num, first = 1;
+	ssize_t ret = 0;
+
+	blk_mq_disable_hotplug();
+
+	for_each_online_cpu(i) {
+		queue_num = hctx->queue->mq_map[i];
+		if (queue_num != hctx->queue_num)
+			continue;
+
+		if (first)
+			ret += sprintf(ret + page, "%u", i);
+		else
+			ret += sprintf(ret + page, ", %u", i);
+
+		first = 0;
+	}
+
+	blk_mq_enable_hotplug();
+
+	ret += sprintf(ret + page, "\n");
+	return ret;
+}
+
 static struct blk_mq_ctx_sysfs_entry blk_mq_sysfs_dispatched = {
 	.attr = {.name = "dispatched", .mode = S_IRUGO },
 	.show = blk_mq_sysfs_dispatched_show,
@@ -294,6 +320,10 @@ static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_tags = {
 	.attr = {.name = "tags", .mode = S_IRUGO },
 	.show = blk_mq_hw_sysfs_tags_show,
 };
+static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_cpus = {
+	.attr = {.name = "cpu_list", .mode = S_IRUGO },
+	.show = blk_mq_hw_sysfs_cpus_show,
+};
 
 static struct attribute *default_hw_ctx_attrs[] = {
 	&blk_mq_hw_sysfs_queued.attr,
@@ -302,6 +332,7 @@ static struct attribute *default_hw_ctx_attrs[] = {
 	&blk_mq_hw_sysfs_pending.attr,
 	&blk_mq_hw_sysfs_ipi.attr,
 	&blk_mq_hw_sysfs_tags.attr,
+	&blk_mq_hw_sysfs_cpus.attr,
 	NULL,
 };
 
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 883f72089015..b1bcc619d0ea 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -320,7 +320,7 @@ void __blk_mq_complete_request(struct request *rq)
 		rq->csd.func = __blk_mq_complete_request_remote;
 		rq->csd.info = rq;
 		rq->csd.flags = 0;
-		__smp_call_function_single(ctx->cpu, &rq->csd, 0);
+		smp_call_function_single_async(ctx->cpu, &rq->csd);
 	} else {
 		rq->q->softirq_done_fn(rq);
 	}
@@ -514,7 +514,7 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 	LIST_HEAD(rq_list);
 	int bit, queued;
 
-	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->flags)))
+	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state)))
 		return;
 
 	hctx->run++;
@@ -603,7 +603,7 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 
 void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
 {
-	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->flags)))
+	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state)))
 		return;
 
 	if (!async)
@@ -623,7 +623,7 @@ void blk_mq_run_queues(struct request_queue *q, bool async)
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if ((!blk_mq_hctx_has_pending(hctx) &&
 		    list_empty_careful(&hctx->dispatch)) ||
-		    test_bit(BLK_MQ_S_STOPPED, &hctx->flags))
+		    test_bit(BLK_MQ_S_STOPPED, &hctx->state))
 			continue;
 
 		blk_mq_run_hw_queue(hctx, async);
@@ -994,8 +994,46 @@ static void blk_mq_hctx_notify(void *data, unsigned long action,
 	blk_mq_put_ctx(ctx);
 }
 
-static void blk_mq_init_hw_commands(struct blk_mq_hw_ctx *hctx,
-				    void (*init)(void *, struct blk_mq_hw_ctx *,
+static int blk_mq_init_hw_commands(struct blk_mq_hw_ctx *hctx,
+				   int (*init)(void *, struct blk_mq_hw_ctx *,
+					struct request *, unsigned int),
+				   void *data)
+{
+	unsigned int i;
+	int ret = 0;
+
+	for (i = 0; i < hctx->queue_depth; i++) {
+		struct request *rq = hctx->rqs[i];
+
+		ret = init(data, hctx, rq, i);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+int blk_mq_init_commands(struct request_queue *q,
+			 int (*init)(void *, struct blk_mq_hw_ctx *,
+					struct request *, unsigned int),
+			 void *data)
+{
+	struct blk_mq_hw_ctx *hctx;
+	unsigned int i;
+	int ret = 0;
+
+	queue_for_each_hw_ctx(q, hctx, i) {
+		ret = blk_mq_init_hw_commands(hctx, init, data);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(blk_mq_init_commands);
+
+static void blk_mq_free_hw_commands(struct blk_mq_hw_ctx *hctx,
+				    void (*free)(void *, struct blk_mq_hw_ctx *,
 					struct request *, unsigned int),
 				    void *data)
 {
@@ -1004,12 +1042,12 @@ static void blk_mq_init_hw_commands(struct blk_mq_hw_ctx *hctx,
 	for (i = 0; i < hctx->queue_depth; i++) {
 		struct request *rq = hctx->rqs[i];
 
-		init(data, hctx, rq, i);
+		free(data, hctx, rq, i);
 	}
 }
 
-void blk_mq_init_commands(struct request_queue *q,
-			  void (*init)(void *, struct blk_mq_hw_ctx *,
+void blk_mq_free_commands(struct request_queue *q,
+			  void (*free)(void *, struct blk_mq_hw_ctx *,
 					struct request *, unsigned int),
 			  void *data)
 {
@@ -1017,9 +1055,9 @@ void blk_mq_init_commands(struct request_queue *q,
 	unsigned int i;
 
 	queue_for_each_hw_ctx(q, hctx, i)
-		blk_mq_init_hw_commands(hctx, init, data);
+		blk_mq_free_hw_commands(hctx, free, data);
 }
-EXPORT_SYMBOL(blk_mq_init_commands);
+EXPORT_SYMBOL(blk_mq_free_commands);
 
 static void blk_mq_free_rq_map(struct blk_mq_hw_ctx *hctx)
 {
@@ -1430,6 +1468,16 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb,
 	return NOTIFY_OK;
 }
 
+void blk_mq_disable_hotplug(void)
+{
+	mutex_lock(&all_q_mutex);
+}
+
+void blk_mq_enable_hotplug(void)
+{
+	mutex_unlock(&all_q_mutex);
+}
+
 static int __init blk_mq_init(void)
 {
 	blk_mq_cpu_init();
diff --git a/block/blk-mq.h b/block/blk-mq.h
index 72beba1f9d55..ebbe6bac9d61 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -39,6 +39,8 @@ void blk_mq_init_cpu_notifier(struct blk_mq_cpu_notifier *notifier,
 void blk_mq_register_cpu_notifier(struct blk_mq_cpu_notifier *notifier);
 void blk_mq_unregister_cpu_notifier(struct blk_mq_cpu_notifier *notifier);
 void blk_mq_cpu_init(void);
+void blk_mq_enable_hotplug(void);
+void blk_mq_disable_hotplug(void);
 
 /*
  * CPU -> queue mappings
diff --git a/block/blk-softirq.c b/block/blk-softirq.c
index 57790c1a97eb..ebd6b6f1bdeb 100644
--- a/block/blk-softirq.c
+++ b/block/blk-softirq.c
@@ -30,8 +30,8 @@ static void blk_done_softirq(struct softirq_action *h)
 	while (!list_empty(&local_list)) {
 		struct request *rq;
 
-		rq = list_entry(local_list.next, struct request, csd.list);
-		list_del_init(&rq->csd.list);
+		rq = list_entry(local_list.next, struct request, queuelist);
+		list_del_init(&rq->queuelist);
 		rq->q->softirq_done_fn(rq);
 	}
 }
@@ -45,9 +45,14 @@ static void trigger_softirq(void *data)
 
 	local_irq_save(flags);
 	list = this_cpu_ptr(&blk_cpu_done);
-	list_add_tail(&rq->csd.list, list);
+	/*
+	 * We reuse queuelist for a list of requests to process. Since the
+	 * queuelist is used by the block layer only for requests waiting to be
+	 * submitted to the device it is unused now.
+	 */
+	list_add_tail(&rq->queuelist, list);
 
-	if (list->next == &rq->csd.list)
+	if (list->next == &rq->queuelist)
 		raise_softirq_irqoff(BLOCK_SOFTIRQ);
 
 	local_irq_restore(flags);
@@ -65,7 +70,7 @@ static int raise_blk_irq(int cpu, struct request *rq)
 		data->info = rq;
 		data->flags = 0;
 
-		__smp_call_function_single(cpu, data, 0);
+		smp_call_function_single_async(cpu, data);
 		return 0;
 	}
 
@@ -136,7 +141,7 @@ void __blk_complete_request(struct request *req)
 		struct list_head *list;
 do_local:
 		list = this_cpu_ptr(&blk_cpu_done);
-		list_add_tail(&req->csd.list, list);
+		list_add_tail(&req->queuelist, list);
 
 		/*
 		 * if the list only contains our just added request,
@@ -144,7 +149,7 @@ do_local:
 		 * entries there, someone already raised the irq but it
 		 * hasn't run yet.
 		 */
-		if (list->next == &req->csd.list)
+		if (list->next == &req->queuelist)
 			raise_softirq_irqoff(BLOCK_SOFTIRQ);
 	} else if (raise_blk_irq(ccpu, req))
 		goto do_local;
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 744833b630c6..5873e4ada9eb 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -2367,10 +2367,10 @@ cfq_merged_requests(struct request_queue *q, struct request *rq,
 	 * reposition in fifo if next is older than rq
 	 */
 	if (!list_empty(&rq->queuelist) && !list_empty(&next->queuelist) &&
-	    time_before(rq_fifo_time(next), rq_fifo_time(rq)) &&
+	    time_before(next->fifo_time, rq->fifo_time) &&
 	    cfqq == RQ_CFQQ(next)) {
 		list_move(&rq->queuelist, &next->queuelist);
-		rq_set_fifo_time(rq, rq_fifo_time(next));
+		rq->fifo_time = next->fifo_time;
 	}
 
 	if (cfqq->next_rq == next)
@@ -2814,7 +2814,7 @@ static struct request *cfq_check_fifo(struct cfq_queue *cfqq)
 		return NULL;
 
 	rq = rq_entry_fifo(cfqq->fifo.next);
-	if (time_before(jiffies, rq_fifo_time(rq)))
+	if (time_before(jiffies, rq->fifo_time))
 		rq = NULL;
 
 	cfq_log_cfqq(cfqq->cfqd, cfqq, "fifo=%p", rq);
@@ -3927,7 +3927,7 @@ static void cfq_insert_request(struct request_queue *q, struct request *rq)
 	cfq_log_cfqq(cfqd, cfqq, "insert_request");
 	cfq_init_prio_data(cfqq, RQ_CIC(rq));
 
-	rq_set_fifo_time(rq, jiffies + cfqd->cfq_fifo_expire[rq_is_sync(rq)]);
+	rq->fifo_time = jiffies + cfqd->cfq_fifo_expire[rq_is_sync(rq)];
 	list_add_tail(&rq->queuelist, &cfqq->fifo);
 	cfq_add_rq_rb(rq);
 	cfqg_stats_update_io_add(RQ_CFQG(rq), cfqd->serving_group,
diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c
index 9ef66406c625..a753df2b3fc2 100644
--- a/block/deadline-iosched.c
+++ b/block/deadline-iosched.c
@@ -106,7 +106,7 @@ deadline_add_request(struct request_queue *q, struct request *rq)
 	/*
 	 * set expire time and add to fifo list
 	 */
-	rq_set_fifo_time(rq, jiffies + dd->fifo_expire[data_dir]);
+	rq->fifo_time = jiffies + dd->fifo_expire[data_dir];
 	list_add_tail(&rq->queuelist, &dd->fifo_list[data_dir]);
 }
 
@@ -174,9 +174,9 @@ deadline_merged_requests(struct request_queue *q, struct request *req,
 	 * and move into next position (next will be deleted) in fifo
 	 */
 	if (!list_empty(&req->queuelist) && !list_empty(&next->queuelist)) {
-		if (time_before(rq_fifo_time(next), rq_fifo_time(req))) {
+		if (time_before(next->fifo_time, req->fifo_time)) {
 			list_move(&req->queuelist, &next->queuelist);
-			rq_set_fifo_time(req, rq_fifo_time(next));
+			req->fifo_time = next->fifo_time;
 		}
 	}
 
@@ -230,7 +230,7 @@ static inline int deadline_check_fifo(struct deadline_data *dd, int ddir)
 	/*
 	 * rq is expired!
 	 */
-	if (time_after_eq(jiffies, rq_fifo_time(rq)))
+	if (time_after_eq(jiffies, rq->fifo_time))
 		return 1;
 
 	return 0;
diff --git a/block/partitions/atari.h b/block/partitions/atari.h
index fe2d32a89f36..f2ec43bfeec1 100644
--- a/block/partitions/atari.h
+++ b/block/partitions/atari.h
@@ -11,6 +11,8 @@
  * by Guenther Kelleter (guenther@pool.informatik.rwth-aachen.de)
  */
 
+#include <linux/compiler.h>
+
 struct partition_info
 {
   u8 flg;			/* bit 0: active; bit 7: bootable */
@@ -29,6 +31,6 @@ struct rootsector
   u32 bsl_st;			/* start of bad sector list */
   u32 bsl_cnt;			/* length of bad sector list */
   u16 checksum;			/* checksum for bootable disks */
-} __attribute__((__packed__));
+} __packed;
 
 int atari_partition(struct parsed_partitions *state);
diff --git a/block/partitions/efi.h b/block/partitions/efi.h
index 4efcafba7e64..abd0b19288a6 100644
--- a/block/partitions/efi.h
+++ b/block/partitions/efi.h
@@ -32,6 +32,7 @@
 #include <linux/major.h>
 #include <linux/string.h>
 #include <linux/efi.h>
+#include <linux/compiler.h>
 
 #define MSDOS_MBR_SIGNATURE 0xaa55
 #define EFI_PMBR_OSTYPE_EFI 0xEF
@@ -87,13 +88,13 @@ typedef struct _gpt_header {
 	 *
 	 * uint8_t		reserved2[ BlockSize - 92 ];
 	 */
-} __attribute__ ((packed)) gpt_header;
+} __packed gpt_header;
 
 typedef struct _gpt_entry_attributes {
 	u64 required_to_function:1;
 	u64 reserved:47;
         u64 type_guid_specific:16;
-} __attribute__ ((packed)) gpt_entry_attributes;
+} __packed gpt_entry_attributes;
 
 typedef struct _gpt_entry {
 	efi_guid_t partition_type_guid;
@@ -102,7 +103,7 @@ typedef struct _gpt_entry {
 	__le64 ending_lba;
 	gpt_entry_attributes attributes;
 	efi_char16_t partition_name[72 / sizeof (efi_char16_t)];
-} __attribute__ ((packed)) gpt_entry;
+} __packed gpt_entry;
 
 typedef struct _gpt_mbr_record {
 	u8	boot_indicator; /* unused by EFI, set to 0x80 for bootable */
@@ -124,7 +125,7 @@ typedef struct _legacy_mbr {
 	__le16 unknown;
 	gpt_mbr_record partition_record[4];
 	__le16 signature;
-} __attribute__ ((packed)) legacy_mbr;
+} __packed legacy_mbr;
 
 /* Functions */
 extern int efi_partition(struct parsed_partitions *state);
diff --git a/block/partitions/karma.c b/block/partitions/karma.c
index 0ea19312706b..9721fa589bb1 100644
--- a/block/partitions/karma.c
+++ b/block/partitions/karma.c
@@ -8,6 +8,7 @@
 
 #include "check.h"
 #include "karma.h"
+#include <linux/compiler.h>
 
 int karma_partition(struct parsed_partitions *state)
 {
@@ -26,7 +27,7 @@ int karma_partition(struct parsed_partitions *state)
 		} d_partitions[2];
 		u8 d_blank[208];
 		__le16 d_magic;
-	} __attribute__((packed)) *label;
+	} __packed *label;
 	struct d_partition *p;
 
 	data = read_part_sector(state, 0, &sect);