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
  ...

6 files changed, 37 insertions, 23 deletions

diff --git a/include/linux/blk-iopoll.h b/include/linux/blk-iopoll.h
index 308734d3d4a2..77ae77c0b704 100644
--- a/include/linux/blk-iopoll.h
+++ b/include/linux/blk-iopoll.h
@@ -43,6 +43,4 @@ extern void __blk_iopoll_complete(struct blk_iopoll *);
 extern void blk_iopoll_enable(struct blk_iopoll *);
 extern void blk_iopoll_disable(struct blk_iopoll *);
 
-extern int blk_iopoll_enabled;
-
 #endif
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index 2ff2e8d982be..0120451545d8 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -109,7 +109,7 @@ enum {
 	BLK_MQ_F_SHOULD_SORT	= 1 << 1,
 	BLK_MQ_F_SHOULD_IPI	= 1 << 2,
 
-	BLK_MQ_S_STOPPED	= 1 << 0,
+	BLK_MQ_S_STOPPED	= 0,
 
 	BLK_MQ_MAX_DEPTH	= 2048,
 };
@@ -117,7 +117,8 @@ enum {
 struct request_queue *blk_mq_init_queue(struct blk_mq_reg *, void *);
 int blk_mq_register_disk(struct gendisk *);
 void blk_mq_unregister_disk(struct gendisk *);
-void blk_mq_init_commands(struct request_queue *, void (*init)(void *data, struct blk_mq_hw_ctx *, struct request *, unsigned int), void *data);
+int blk_mq_init_commands(struct request_queue *, int (*init)(void *data, struct blk_mq_hw_ctx *, struct request *, unsigned int), void *data);
+void blk_mq_free_commands(struct request_queue *, void (*free)(void *data, struct blk_mq_hw_ctx *, struct request *, unsigned int), void *data);
 
 void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4afa4f8f6090..1e1fa3f93d5f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -99,6 +99,7 @@ struct request {
 	union {
 		struct call_single_data csd;
 		struct work_struct mq_flush_work;
+		unsigned long fifo_time;
 	};
 
 	struct request_queue *q;
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 306dd8cd0b6f..df63bd3a8cf1 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -202,17 +202,8 @@ enum {
 #define rq_end_sector(rq)	(blk_rq_pos(rq) + blk_rq_sectors(rq))
 #define rb_entry_rq(node)	rb_entry((node), struct request, rb_node)
 
-/*
- * Hack to reuse the csd.list list_head as the fifo time holder while
- * the request is in the io scheduler. Saves an unsigned long in rq.
- */
-#define rq_fifo_time(rq)	((unsigned long) (rq)->csd.list.next)
-#define rq_set_fifo_time(rq,exp)	((rq)->csd.list.next = (void *) (exp))
 #define rq_entry_fifo(ptr)	list_entry((ptr), struct request, queuelist)
-#define rq_fifo_clear(rq)	do {		\
-	list_del_init(&(rq)->queuelist);	\
-	INIT_LIST_HEAD(&(rq)->csd.list);	\
-	} while (0)
+#define rq_fifo_clear(rq)	list_del_init(&(rq)->queuelist)
 
 #else /* CONFIG_BLOCK */
 
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 6ae004e437ea..633f5edd7470 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -17,10 +17,7 @@ extern void cpu_idle(void);
 
 typedef void (*smp_call_func_t)(void *info);
 struct call_single_data {
-	union {
-		struct list_head list;
-		struct llist_node llist;
-	};
+	struct llist_node llist;
 	smp_call_func_t func;
 	void *info;
 	u16 flags;
@@ -53,8 +50,7 @@ void on_each_cpu_cond(bool (*cond_func)(int cpu, void *info),
 		smp_call_func_t func, void *info, bool wait,
 		gfp_t gfp_flags);
 
-void __smp_call_function_single(int cpuid, struct call_single_data *data,
-				int wait);
+int smp_call_function_single_async(int cpu, struct call_single_data *csd);
 
 #ifdef CONFIG_SMP
 
diff --git a/include/trace/events/block.h b/include/trace/events/block.h
index e76ae19a8d6f..e8a5eca1dbe5 100644
--- a/include/trace/events/block.h
+++ b/include/trace/events/block.h
@@ -132,6 +132,7 @@ DEFINE_EVENT(block_rq_with_error, block_rq_requeue,
  * block_rq_complete - block IO operation completed by device driver
  * @q: queue containing the block operation request
  * @rq: block operations request
+ * @nr_bytes: number of completed bytes
  *
  * The block_rq_complete tracepoint event indicates that some portion
  * of operation request has been completed by the device driver.  If
@@ -139,11 +140,37 @@ DEFINE_EVENT(block_rq_with_error, block_rq_requeue,
  * do for the request. If @rq->bio is non-NULL then there is
  * additional work required to complete the request.
  */
-DEFINE_EVENT(block_rq_with_error, block_rq_complete,
+TRACE_EVENT(block_rq_complete,
 
-	TP_PROTO(struct request_queue *q, struct request *rq),
+	TP_PROTO(struct request_queue *q, struct request *rq,
+		 unsigned int nr_bytes),
 
-	TP_ARGS(q, rq)
+	TP_ARGS(q, rq, nr_bytes),
+
+	TP_STRUCT__entry(
+		__field(  dev_t,	dev			)
+		__field(  sector_t,	sector			)
+		__field(  unsigned int,	nr_sector		)
+		__field(  int,		errors			)
+		__array(  char,		rwbs,	RWBS_LEN	)
+		__dynamic_array( char,	cmd,	blk_cmd_buf_len(rq)	)
+	),
+
+	TP_fast_assign(
+		__entry->dev	   = rq->rq_disk ? disk_devt(rq->rq_disk) : 0;
+		__entry->sector    = blk_rq_pos(rq);
+		__entry->nr_sector = nr_bytes >> 9;
+		__entry->errors    = rq->errors;
+
+		blk_fill_rwbs(__entry->rwbs, rq->cmd_flags, nr_bytes);
+		blk_dump_cmd(__get_str(cmd), rq);
+	),
+
+	TP_printk("%d,%d %s (%s) %llu + %u [%d]",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->rwbs, __get_str(cmd),
+		  (unsigned long long)__entry->sector,
+		  __entry->nr_sector, __entry->errors)
 );
 
 DECLARE_EVENT_CLASS(block_rq,