Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
cfq-iosched: fix RCU problem in cfq_cic_lookup()
block: make blktrace use per-cpu buffers for message notes
Added in elevator switch message to blktrace stream
Added in MESSAGE notes for blktraces
block: reorder cfq_queue to save space on 64bit builds
block: Move the second call to get_request to the end of the loop
splice: handle try_to_release_page() failure
splice: fix sendfile() issue with relay

block/blk-core.c

···
 	rq = get_request(q, rw_flags, bio, GFP_NOIO);
 	while (!rq) {
 		DEFINE_WAIT(wait);
+		struct io_context *ioc;
 		struct request_list *rl = &q->rq;
 
 		prepare_to_wait_exclusive(&rl->wait[rw], &wait,
 				TASK_UNINTERRUPTIBLE);
 
-		rq = get_request(q, rw_flags, bio, GFP_NOIO);
+		blk_add_trace_generic(q, bio, rw, BLK_TA_SLEEPRQ);
 
-		if (!rq) {
-			struct io_context *ioc;
+		__generic_unplug_device(q);
+		spin_unlock_irq(q->queue_lock);
+		io_schedule();
 
-			blk_add_trace_generic(q, bio, rw, BLK_TA_SLEEPRQ);
+		/*
+		 * After sleeping, we become a "batching" process and
+		 * will be able to allocate at least one request, and
+		 * up to a big batch of them for a small period time.
+		 * See ioc_batching, ioc_set_batching
+		 */
+		ioc = current_io_context(GFP_NOIO, q->node);
+		ioc_set_batching(q, ioc);
 
-			__generic_unplug_device(q);
-			spin_unlock_irq(q->queue_lock);
-			io_schedule();
-
-			/*
-			 * After sleeping, we become a "batching" process and
-			 * will be able to allocate at least one request, and
-			 * up to a big batch of them for a small period time.
-			 * See ioc_batching, ioc_set_batching
-			 */
-			ioc = current_io_context(GFP_NOIO, q->node);
-			ioc_set_batching(q, ioc);
-
-			spin_lock_irq(q->queue_lock);
-		}
+		spin_lock_irq(q->queue_lock);
 		finish_wait(&rl->wait[rw], &wait);
-	}
+
+		rq = get_request(q, rw_flags, bio, GFP_NOIO);
+	};
 
 	return rq;
 }

block/blktrace.c

···
 	local_irq_restore(flags);
 }
 
+void __trace_note_message(struct blk_trace *bt, const char *fmt, ...)
+{
+	int n;
+	va_list args;
+	char *buf;
+
+	preempt_disable();
+	buf = per_cpu_ptr(bt->msg_data, smp_processor_id());
+	va_start(args, fmt);
+	n = vscnprintf(buf, BLK_TN_MAX_MSG, fmt, args);
+	va_end(args);
+
+	trace_note(bt, 0, BLK_TN_MESSAGE, buf, n);
+	preempt_enable();
+}
+EXPORT_SYMBOL_GPL(__trace_note_message);
+
 static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector,
 			 pid_t pid)
 {
···
 	debugfs_remove(bt->dropped_file);
 	blk_remove_tree(bt->dir);
 	free_percpu(bt->sequence);
+	free_percpu(bt->msg_data);
 	kfree(bt);
 }
 
···
 	if (!bt->sequence)
 		goto err;
 
+	bt->msg_data = __alloc_percpu(BLK_TN_MAX_MSG);
+	if (!bt->msg_data)
+		goto err;
+
 	ret = -ENOENT;
 	dir = blk_create_tree(buts->name);
 	if (!dir)
···
 		if (bt->dropped_file)
 			debugfs_remove(bt->dropped_file);
 		free_percpu(bt->sequence);
+		free_percpu(bt->msg_data);
 		if (bt->rchan)
 			relay_close(bt->rchan);
 		kfree(bt);

block/cfq-iosched.c

···
 struct cfq_queue {
 	/* reference count */
 	atomic_t ref;
+	/* various state flags, see below */
+	unsigned int flags;
 	/* parent cfq_data */
 	struct cfq_data *cfqd;
 	/* service_tree member */
···
 	int queued[2];
 	/* currently allocated requests */
 	int allocated[2];
-	/* pending metadata requests */
-	int meta_pending;
 	/* fifo list of requests in sort_list */
 	struct list_head fifo;
 
 	unsigned long slice_end;
 	long slice_resid;
 
+	/* pending metadata requests */
+	int meta_pending;
 	/* number of requests that are on the dispatch list or inside driver */
 	int dispatched;
 
···
 	unsigned short ioprio, org_ioprio;
 	unsigned short ioprio_class, org_ioprio_class;
 
-	/* various state flags, see below */
-	unsigned int flags;
 };
 
 enum cfqq_state_flags {
···
 	kmem_cache_free(cfq_pool, cfqq);
 }
 
+/*
+ * Must always be called with the rcu_read_lock() held
+ */
 static void
 __call_for_each_cic(struct io_context *ioc,
 		    void (*func)(struct io_context *, struct cfq_io_context *))
···
 	cfq_cic_free(cic);
 }
 
+/*
+ * Must be called with rcu_read_lock() held or preemption otherwise disabled.
+ * Only two callers of this - ->dtor() which is called with the rcu_read_lock(),
+ * and ->trim() which is called with the task lock held
+ */
 static void cfq_free_io_context(struct io_context *ioc)
 {
 	/*
···
 cfq_cic_lookup(struct cfq_data *cfqd, struct io_context *ioc)
 {
 	struct cfq_io_context *cic;
+	unsigned long flags;
 	void *k;
 
 	if (unlikely(!ioc))
 		return NULL;
 
+	rcu_read_lock();
+
 	/*
 	 * we maintain a last-hit cache, to avoid browsing over the tree
 	 */
 	cic = rcu_dereference(ioc->ioc_data);
-	if (cic && cic->key == cfqd)
+	if (cic && cic->key == cfqd) {
+		rcu_read_unlock();
 		return cic;
+	}
 
 	do {
-		rcu_read_lock();
 		cic = radix_tree_lookup(&ioc->radix_root, (unsigned long) cfqd);
 		rcu_read_unlock();
 		if (!cic)
···
 		k = cic->key;
 		if (unlikely(!k)) {
 			cfq_drop_dead_cic(cfqd, ioc, cic);
+			rcu_read_lock();
 			continue;
 		}
 
+		spin_lock_irqsave(&ioc->lock, flags);
 		rcu_assign_pointer(ioc->ioc_data, cic);
+		spin_unlock_irqrestore(&ioc->lock, flags);
 		break;
 	} while (1);
 
···
 
 static void cfq_slab_kill(void)
 {
+	/*
+	 * Caller already ensured that pending RCU callbacks are completed,
+	 * so we should have no busy allocations at this point.
+	 */
 	if (cfq_pool)
 		kmem_cache_destroy(cfq_pool);
 	if (cfq_ioc_pool)
···
 	ioc_gone = &all_gone;
 	/* ioc_gone's update must be visible before reading ioc_count */
 	smp_wmb();
+
+	/*
+	 * this also protects us from entering cfq_slab_kill() with
+	 * pending RCU callbacks
+	 */
 	if (elv_ioc_count_read(ioc_count))
 		wait_for_completion(ioc_gone);
 	cfq_slab_kill();

block/elevator.c

···
 	queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
 	spin_unlock_irq(q->queue_lock);
 
+	blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name);
+
 	return 1;
 
 fail_register:

fs/splice.c

···
 		 */
 		wait_on_page_writeback(page);
 
-		if (PagePrivate(page))
-			try_to_release_page(page, GFP_KERNEL);
+		if (PagePrivate(page) && !try_to_release_page(page, GFP_KERNEL))
+			goto out_unlock;
 
 		/*
 		 * If we succeeded in removing the mapping, set LRU flag
···
 	 * Raced with truncate or failed to remove page from current
 	 * address space, unlock and return failure.
 	 */
+out_unlock:
 	unlock_page(page);
 	return 1;
 }
···
 
 	while (len) {
 		size_t read_len;
-		loff_t pos = sd->pos;
+		loff_t pos = sd->pos, prev_pos = pos;
 
 		ret = do_splice_to(in, &pos, pipe, len, flags);
 		if (unlikely(ret <= 0))
···
 		 * could get stuck data in the internal pipe:
 		 */
 		ret = actor(pipe, sd);
-		if (unlikely(ret <= 0))
+		if (unlikely(ret <= 0)) {
+			sd->pos = prev_pos;
 			goto out_release;
+		}
 
 		bytes += ret;
 		len -= ret;
 		sd->pos = pos;
 
-		if (ret < read_len)
+		if (ret < read_len) {
+			sd->pos = prev_pos + ret;
 			goto out_release;
+		}
 	}
 
 done:
···
 
 	ret = splice_direct_to_actor(in, &sd, direct_splice_actor);
 	if (ret > 0)
-		*ppos += ret;
+		*ppos = sd.pos;
 
 	return ret;
 }

include/linux/blktrace_api.h

···
 enum blktrace_notify {
 	__BLK_TN_PROCESS = 0,		/* establish pid/name mapping */
 	__BLK_TN_TIMESTAMP,		/* include system clock */
+	__BLK_TN_MESSAGE,		/* Character string message */
 };
 
 
···
 
 #define BLK_TN_PROCESS		(__BLK_TN_PROCESS | BLK_TC_ACT(BLK_TC_NOTIFY))
 #define BLK_TN_TIMESTAMP	(__BLK_TN_TIMESTAMP | BLK_TC_ACT(BLK_TC_NOTIFY))
+#define BLK_TN_MESSAGE		(__BLK_TN_MESSAGE | BLK_TC_ACT(BLK_TC_NOTIFY))
 
 #define BLK_IO_TRACE_MAGIC	0x65617400
 #define BLK_IO_TRACE_VERSION	0x07
···
 	int trace_state;
 	struct rchan *rchan;
 	unsigned long *sequence;
+	unsigned char *msg_data;
 	u16 act_mask;
 	u64 start_lba;
 	u64 end_lba;
···
 extern void __blk_add_trace(struct blk_trace *, sector_t, int, int, u32, int, int, void *);
 extern int do_blk_trace_setup(struct request_queue *q,
 	char *name, dev_t dev, struct blk_user_trace_setup *buts);
+extern void __trace_note_message(struct blk_trace *, const char *fmt, ...);
 
+/**
+ * blk_add_trace_msg - Add a (simple) message to the blktrace stream
+ * @q:		queue the io is for
+ * @fmt:	format to print message in
+ * args...	Variable argument list for format
+ *
+ * Description:
+ *     Records a (simple) message onto the blktrace stream.
+ *
+ *     NOTE: BLK_TN_MAX_MSG characters are output at most.
+ *     NOTE: Can not use 'static inline' due to presence of var args...
+ *
+ **/
+#define blk_add_trace_msg(q, fmt, ...)					\
+	do {								\
+		struct blk_trace *bt = (q)->blk_trace;			\
+		if (unlikely(bt))					\
+			__trace_note_message(bt, fmt, ##__VA_ARGS__);	\
+	} while (0)
+#define BLK_TN_MAX_MSG		128
 
 /**
  * blk_add_trace_rq - Add a trace for a request oriented action
···
 #define blk_trace_setup(q, name, dev, arg)	(-ENOTTY)
 #define blk_trace_startstop(q, start)		(-ENOTTY)
 #define blk_trace_remove(q)			(-ENOTTY)
+#define blk_add_trace_msg(q, fmt, ...)		do { } while (0)
+
 #endif /* CONFIG_BLK_DEV_IO_TRACE */
 #endif /* __KERNEL__ */
 #endif

kernel/relay.c

···
 	ret = 0;
 	spliced = 0;
 
-	while (len) {
+	while (len && !spliced) {
 		ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret);
 		if (ret < 0)
 			break;