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

Pull block-related fixes from Jens Axboe:

 - Improvements to the buffered and direct write IO plugging from
   Fengguang.

 - Abstract out the mapping of a bio in a request, and use that to
   provide a blk_bio_map_sg() helper.  Useful for mapping just a bio
   instead of a full request.

 - Regression fix from Hugh, fixing up a patch that went into the
   previous release cycle (and marked stable, too) attempting to prevent
   a loop in __getblk_slow().

 - Updates to discard requests, fixing up the sizing and how we align
   them.  Also a change to disallow merging of discard requests, since
   that doesn't really work properly yet.

 - A few drbd fixes.

 - Documentation updates.

* 'for-linus' of git://git.kernel.dk/linux-block:
  block: replace __getblk_slow misfix by grow_dev_page fix
  drbd: Write all pages of the bitmap after an online resize
  drbd: Finish requests that completed while IO was frozen
  drbd: fix drbd wire compatibility for empty flushes
  Documentation: update tunable options in block/cfq-iosched.txt
  Documentation: update tunable options in block/cfq-iosched.txt
  Documentation: update missing index files in block/00-INDEX
  block: move down direct IO plugging
  block: remove plugging at buffered write time
  block: disable discard request merge temporarily
  bio: Fix potential memory leak in bio_find_or_create_slab()
  block: Don't use static to define "void *p" in show_partition_start()
  block: Add blk_bio_map_sg() helper
  block: Introduce __blk_segment_map_sg() helper
  fs/block-dev.c:fix performance regression in O_DIRECT writes to md block devices
  block: split discard into aligned requests
  block: reorganize rounding of max_discard_sectors

Documentation/block/00-INDEX

@@ -3,15 +3,21 @@
 biodoc.txt
 	- Notes on the Generic Block Layer Rewrite in Linux 2.5
 capability.txt
-	- Generic Block Device Capability (/sys/block/<disk>/capability)
+	- Generic Block Device Capability (/sys/block/<device>/capability)
+cfq-iosched.txt
+	- CFQ IO scheduler tunables
+data-integrity.txt
+	- Block data integrity
 deadline-iosched.txt
 	- Deadline IO scheduler tunables
 ioprio.txt
 	- Block io priorities (in CFQ scheduler)
+queue-sysfs.txt
+	- Queue's sysfs entries
 request.txt
 	- The members of struct request (in include/linux/blkdev.h)
 stat.txt
-	- Block layer statistics in /sys/block/<dev>/stat
+	- Block layer statistics in /sys/block/<device>/stat
 switching-sched.txt
 	- Switching I/O schedulers at runtime
 writeback_cache_control.txt

Documentation/block/cfq-iosched.txt

+CFQ (Complete Fairness Queueing)
+===============================
+
+The main aim of CFQ scheduler is to provide a fair allocation of the disk
+I/O bandwidth for all the processes which requests an I/O operation.
+
+CFQ maintains the per process queue for the processes which request I/O
+operation(syncronous requests). In case of asynchronous requests, all the
+requests from all the processes are batched together according to their
+process's I/O priority.
+
 CFQ ioscheduler tunables
 ========================
 
@@ -25,6 +36,72 @@ there are multiple spindles behind single LUN (Host based hardware RAID
 controller or for storage arrays), setting slice_idle=0 might end up in better
 throughput and acceptable latencies.
 
+back_seek_max
+-------------
+This specifies, given in Kbytes, the maximum "distance" for backward seeking.
+The distance is the amount of space from the current head location to the
+sectors that are backward in terms of distance.
+
+This parameter allows the scheduler to anticipate requests in the "backward"
+direction and consider them as being the "next" if they are within this
+distance from the current head location.
+
+back_seek_penalty
+-----------------
+This parameter is used to compute the cost of backward seeking. If the
+backward distance of request is just 1/back_seek_penalty from a "front"
+request, then the seeking cost of two requests is considered equivalent.
+
+So scheduler will not bias toward one or the other request (otherwise scheduler
+will bias toward front request). Default value of back_seek_penalty is 2.
+
+fifo_expire_async
+-----------------
+This parameter is used to set the timeout of asynchronous requests. Default
+value of this is 248ms.
+
+fifo_expire_sync
+----------------
+This parameter is used to set the timeout of synchronous requests. Default
+value of this is 124ms. In case to favor synchronous requests over asynchronous
+one, this value should be decreased relative to fifo_expire_async.
+
+slice_async
+-----------
+This parameter is same as of slice_sync but for asynchronous queue. The
+default value is 40ms.
+
+slice_async_rq
+--------------
+This parameter is used to limit the dispatching of asynchronous request to
+device request queue in queue's slice time. The maximum number of request that
+are allowed to be dispatched also depends upon the io priority. Default value
+for this is 2.
+
+slice_sync
+----------
+When a queue is selected for execution, the queues IO requests are only
+executed for a certain amount of time(time_slice) before switching to another
+queue. This parameter is used to calculate the time slice of synchronous
+queue.
+
+time_slice is computed using the below equation:-
+time_slice = slice_sync + (slice_sync/5 * (4 - prio)). To increase the
+time_slice of synchronous queue, increase the value of slice_sync. Default
+value is 100ms.
+
+quantum
+-------
+This specifies the number of request dispatched to the device queue. In a
+queue's time slice, a request will not be dispatched if the number of request
+in the device exceeds this parameter. This parameter is used for synchronous
+request.
+
+In case of storage with several disk, this setting can limit the parallel
+processing of request. Therefore, increasing the value can imporve the
+performace although this can cause the latency of some I/O to increase due
+to more number of requests.
+
 CFQ IOPS Mode for group scheduling
 ===================================
 Basic CFQ design is to provide priority based time slices. Higher priority

Documentation/block/queue-sysfs.txt

@@ -9,20 +9,71 @@ These files are the ones found in the /sys/block/xxx/queue/ directory.
 Files denoted with a RO postfix are readonly and the RW postfix means
 read-write.
 
+add_random (RW)
+----------------
+This file allows to trun off the disk entropy contribution. Default
+value of this file is '1'(on).
+
+discard_granularity (RO)
+-----------------------
+This shows the size of internal allocation of the device in bytes, if
+reported by the device. A value of '0' means device does not support
+the discard functionality.
+
+discard_max_bytes (RO)
+----------------------
+Devices that support discard functionality may have internal limits on
+the number of bytes that can be trimmed or unmapped in a single operation.
+The discard_max_bytes parameter is set by the device driver to the maximum
+number of bytes that can be discarded in a single operation. Discard
+requests issued to the device must not exceed this limit. A discard_max_bytes
+value of 0 means that the device does not support discard functionality.
+
+discard_zeroes_data (RO)
+------------------------
+When read, this file will show if the discarded block are zeroed by the
+device or not. If its value is '1' the blocks are zeroed otherwise not.
+
 hw_sector_size (RO)
 -------------------
 This is the hardware sector size of the device, in bytes.
 
+iostats (RW)
+-------------
+This file is used to control (on/off) the iostats accounting of the
+disk.
+
+logical_block_size (RO)
+-----------------------
+This is the logcal block size of the device, in bytes.
+
 max_hw_sectors_kb (RO)
 ----------------------
 This is the maximum number of kilobytes supported in a single data transfer.
 
+max_integrity_segments (RO)
+---------------------------
+When read, this file shows the max limit of integrity segments as
+set by block layer which a hardware controller can handle.
+
 max_sectors_kb (RW)
 -------------------
 This is the maximum number of kilobytes that the block layer will allow
 for a filesystem request. Must be smaller than or equal to the maximum
 size allowed by the hardware.
 
+max_segments (RO)
+-----------------
+Maximum number of segments of the device.
+
+max_segment_size (RO)
+---------------------
+Maximum segment size of the device.
+
+minimum_io_size (RO)
+--------------------
+This is the smallest preferred io size reported by the device.
+
 nomerges (RW)
 -------------
 This enables the user to disable the lookup logic involved with IO
@@ -45,11 +96,24 @@ per-block-cgroup request pool.  IOW, if there are N block cgroups,
 each request queue may have upto N request pools, each independently
 regulated by nr_requests.
 
+optimal_io_size (RO)
+--------------------
+This is the optimal io size reported by the device.
+
+physical_block_size (RO)
+------------------------
+This is the physical block size of device, in bytes.
+
 read_ahead_kb (RW)
 ------------------
 Maximum number of kilobytes to read-ahead for filesystems on this block
 device.
 
+rotational (RW)
+---------------
+This file is used to stat if the device is of rotational type or
+non-rotational type.
+
 rq_affinity (RW)
 ----------------
 If this option is '1', the block layer will migrate request completions to the

block/blk-lib.c

@@ -44,6 +44,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	struct request_queue *q = bdev_get_queue(bdev);
 	int type = REQ_WRITE | REQ_DISCARD;
 	unsigned int max_discard_sectors;
+	unsigned int granularity, alignment, mask;
 	struct bio_batch bb;
 	struct bio *bio;
 	int ret = 0;
@@ -54,18 +55,20 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	if (!blk_queue_discard(q))
 		return -EOPNOTSUPP;
 
+	/* Zero-sector (unknown) and one-sector granularities are the same.  */
+	granularity = max(q->limits.discard_granularity >> 9, 1U);
+	mask = granularity - 1;
+	alignment = (bdev_discard_alignment(bdev) >> 9) & mask;
+
 	/*
 	 * Ensure that max_discard_sectors is of the proper
-	 * granularity
+	 * granularity, so that requests stay aligned after a split.
 	 */
 	max_discard_sectors = min(q->limits.max_discard_sectors, UINT_MAX >> 9);
+	max_discard_sectors = round_down(max_discard_sectors, granularity);
 	if (unlikely(!max_discard_sectors)) {
 		/* Avoid infinite loop below. Being cautious never hurts. */
 		return -EOPNOTSUPP;
-	} else if (q->limits.discard_granularity) {
-		unsigned int disc_sects = q->limits.discard_granularity >> 9;
-
-		max_discard_sectors &= ~(disc_sects - 1);
 	}
 
 	if (flags & BLKDEV_DISCARD_SECURE) {
@@ -79,25 +82,37 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	bb.wait = &wait;
 
 	while (nr_sects) {
+		unsigned int req_sects;
+		sector_t end_sect;
+
 		bio = bio_alloc(gfp_mask, 1);
 		if (!bio) {
 			ret = -ENOMEM;
 			break;
 		}
 
+		req_sects = min_t(sector_t, nr_sects, max_discard_sectors);
+
+		/*
+		 * If splitting a request, and the next starting sector would be
+		 * misaligned, stop the discard at the previous aligned sector.
+		 */
+		end_sect = sector + req_sects;
+		if (req_sects < nr_sects && (end_sect & mask) != alignment) {
+			end_sect =
+				round_down(end_sect - alignment, granularity)
+				+ alignment;
+			req_sects = end_sect - sector;
+		}
+
 		bio->bi_sector = sector;
 		bio->bi_end_io = bio_batch_end_io;
 		bio->bi_bdev = bdev;
 		bio->bi_private = &bb;
 
-		if (nr_sects > max_discard_sectors) {
-			bio->bi_size = max_discard_sectors << 9;
-			nr_sects -= max_discard_sectors;
-			sector += max_discard_sectors;
-		} else {
-			bio->bi_size = nr_sects << 9;
-			nr_sects = 0;
-		}
+		bio->bi_size = req_sects << 9;
+		nr_sects -= req_sects;
+		sector = end_sect;
 
 		atomic_inc(&bb.done);
 		submit_bio(type, bio);

block/blk-merge.c

@@ -110,6 +110,49 @@ static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio,
 	return 0;
 }
 
+static void
+__blk_segment_map_sg(struct request_queue *q, struct bio_vec *bvec,
+		     struct scatterlist *sglist, struct bio_vec **bvprv,
+		     struct scatterlist **sg, int *nsegs, int *cluster)
+{
+
+	int nbytes = bvec->bv_len;
+
+	if (*bvprv && *cluster) {
+		if ((*sg)->length + nbytes > queue_max_segment_size(q))
+			goto new_segment;
+
+		if (!BIOVEC_PHYS_MERGEABLE(*bvprv, bvec))
+			goto new_segment;
+		if (!BIOVEC_SEG_BOUNDARY(q, *bvprv, bvec))
+			goto new_segment;
+
+		(*sg)->length += nbytes;
+	} else {
+new_segment:
+		if (!*sg)
+			*sg = sglist;
+		else {
+			/*
+			 * If the driver previously mapped a shorter
+			 * list, we could see a termination bit
+			 * prematurely unless it fully inits the sg
+			 * table on each mapping. We KNOW that there
+			 * must be more entries here or the driver
+			 * would be buggy, so force clear the
+			 * termination bit to avoid doing a full
+			 * sg_init_table() in drivers for each command.
+			 */
+			(*sg)->page_link &= ~0x02;
+			*sg = sg_next(*sg);
+		}
+
+		sg_set_page(*sg, bvec->bv_page, nbytes, bvec->bv_offset);
+		(*nsegs)++;
+	}
+	*bvprv = bvec;
+}
+
 /*
  * map a request to scatterlist, return number of sg entries setup. Caller
  * must make sure sg can hold rq->nr_phys_segments entries
@@ -131,41 +174,8 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
 	bvprv = NULL;
 	sg = NULL;
 	rq_for_each_segment(bvec, rq, iter) {
-		int nbytes = bvec->bv_len;
-
-		if (bvprv && cluster) {
-			if (sg->length + nbytes > queue_max_segment_size(q))
-				goto new_segment;
-
-			if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec))
-				goto new_segment;
-			if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec))
-				goto new_segment;
-
-			sg->length += nbytes;
-		} else {
-new_segment:
-			if (!sg)
-				sg = sglist;
-			else {
-				/*
-				 * If the driver previously mapped a shorter
-				 * list, we could see a termination bit
-				 * prematurely unless it fully inits the sg
-				 * table on each mapping. We KNOW that there
-				 * must be more entries here or the driver
-				 * would be buggy, so force clear the
-				 * termination bit to avoid doing a full
-				 * sg_init_table() in drivers for each command.
-				 */
-				sg->page_link &= ~0x02;
-				sg = sg_next(sg);
-			}
-
-			sg_set_page(sg, bvec->bv_page, nbytes, bvec->bv_offset);
-			nsegs++;
-		}
-		bvprv = bvec;
+		__blk_segment_map_sg(q, bvec, sglist, &bvprv, &sg,
+				     &nsegs, &cluster);
 	} /* segments in rq */
 
 
@@ -199,6 +209,43 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
 }
 EXPORT_SYMBOL(blk_rq_map_sg);
 
+/**
+ * blk_bio_map_sg - map a bio to a scatterlist
+ * @q: request_queue in question
+ * @bio: bio being mapped
+ * @sglist: scatterlist being mapped
+ *
+ * Note:
+ *    Caller must make sure sg can hold bio->bi_phys_segments entries
+ *
+ * Will return the number of sg entries setup
+ */
+int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
+		   struct scatterlist *sglist)
+{
+	struct bio_vec *bvec, *bvprv;
+	struct scatterlist *sg;
+	int nsegs, cluster;
+	unsigned long i;
+
+	nsegs = 0;
+	cluster = blk_queue_cluster(q);
+
+	bvprv = NULL;
+	sg = NULL;
+	bio_for_each_segment(bvec, bio, i) {
+		__blk_segment_map_sg(q, bvec, sglist, &bvprv, &sg,
+				     &nsegs, &cluster);
+	} /* segments in bio */
+
+	if (sg)
+		sg_mark_end(sg);
+
+	BUG_ON(bio->bi_phys_segments && nsegs > bio->bi_phys_segments);
+	return nsegs;
+}
+EXPORT_SYMBOL(blk_bio_map_sg);
+
 static inline int ll_new_hw_segment(struct request_queue *q,
 				    struct request *req,
 				    struct bio *bio)

block/genhd.c

@@ -835,7 +835,7 @@ static void disk_seqf_stop(struct seq_file *seqf, void *v)
 
 static void *show_partition_start(struct seq_file *seqf, loff_t *pos)
 {
-	static void *p;
+	void *p;
 
 	p = disk_seqf_start(seqf, pos);
 	if (!IS_ERR_OR_NULL(p) && !*pos)

drivers/block/drbd/drbd_bitmap.c

@@ -889,6 +889,7 @@ struct bm_aio_ctx {
 	unsigned int done;
 	unsigned flags;
 #define BM_AIO_COPY_PAGES	1
+#define BM_WRITE_ALL_PAGES	2
 	int error;
 	struct kref kref;
 };
@@ -1059,7 +1060,8 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
 		if (lazy_writeout_upper_idx && i == lazy_writeout_upper_idx)
 			break;
 		if (rw & WRITE) {
-			if (bm_test_page_unchanged(b->bm_pages[i])) {
+			if (!(flags & BM_WRITE_ALL_PAGES) &&
+			    bm_test_page_unchanged(b->bm_pages[i])) {
 				dynamic_dev_dbg(DEV, "skipped bm write for idx %u\n", i);
 				continue;
 			}
@@ -1140,6 +1142,17 @@ int drbd_bm_write(struct drbd_conf *mdev) __must_hold(local)
 	return bm_rw(mdev, WRITE, 0, 0);
 }
 
+/**
+ * drbd_bm_write_all() - Write the whole bitmap to its on disk location.
+ * @mdev:	DRBD device.
+ *
+ * Will write all pages.
+ */
+int drbd_bm_write_all(struct drbd_conf *mdev) __must_hold(local)
+{
+	return bm_rw(mdev, WRITE, BM_WRITE_ALL_PAGES, 0);
+}
+
 /**
  * drbd_bm_lazy_write_out() - Write bitmap pages 0 to @upper_idx-1, if they have changed.
  * @mdev:	DRBD device.

drivers/block/drbd/drbd_int.h

@@ -1469,6 +1469,7 @@ extern int  drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr);
 extern int  drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(local);
 extern int  drbd_bm_read(struct drbd_conf *mdev) __must_hold(local);
 extern int  drbd_bm_write(struct drbd_conf *mdev) __must_hold(local);
+extern int drbd_bm_write_all(struct drbd_conf *mdev) __must_hold(local);
 extern int  drbd_bm_write_copy_pages(struct drbd_conf *mdev) __must_hold(local);
 extern unsigned long drbd_bm_ALe_set_all(struct drbd_conf *mdev,
 		unsigned long al_enr);

drivers/block/drbd/drbd_main.c

@@ -79,6 +79,7 @@ static int w_md_sync(struct drbd_conf *mdev, struct drbd_work *w, int unused);
 static void md_sync_timer_fn(unsigned long data);
 static int w_bitmap_io(struct drbd_conf *mdev, struct drbd_work *w, int unused);
 static int w_go_diskless(struct drbd_conf *mdev, struct drbd_work *w, int unused);
+static void _tl_clear(struct drbd_conf *mdev);
 
 MODULE_AUTHOR("Philipp Reisner <phil@linbit.com>, "
 	      "Lars Ellenberg <lars@linbit.com>");
@@ -432,19 +433,10 @@ static void _tl_restart(struct drbd_conf *mdev, enum drbd_req_event what)
 
 	/* Actions operating on the disk state, also want to work on
 	   requests that got barrier acked. */
-	switch (what) {
-	case fail_frozen_disk_io:
-	case restart_frozen_disk_io:
-		list_for_each_safe(le, tle, &mdev->barrier_acked_requests) {
-			req = list_entry(le, struct drbd_request, tl_requests);
-			_req_mod(req, what);
-		}
 
-	case connection_lost_while_pending:
-	case resend:
-		break;
-	default:
-		dev_err(DEV, "what = %d in _tl_restart()\n", what);
+	list_for_each_safe(le, tle, &mdev->barrier_acked_requests) {
+		req = list_entry(le, struct drbd_request, tl_requests);
+		_req_mod(req, what);
 	}
 }
 
@@ -458,12 +450,17 @@ static void _tl_restart(struct drbd_conf *mdev, enum drbd_req_event what)
  * receiver thread and the worker thread.
  */
 void tl_clear(struct drbd_conf *mdev)
+{
+	spin_lock_irq(&mdev->req_lock);
+	_tl_clear(mdev);
+	spin_unlock_irq(&mdev->req_lock);
+}
+
+static void _tl_clear(struct drbd_conf *mdev)
 {
 	struct list_head *le, *tle;
 	struct drbd_request *r;
 
-	spin_lock_irq(&mdev->req_lock);
-
 	_tl_restart(mdev, connection_lost_while_pending);
 
 	/* we expect this list to be empty. */
@@ -482,7 +479,6 @@ void tl_clear(struct drbd_conf *mdev)
 
 	memset(mdev->app_reads_hash, 0, APP_R_HSIZE*sizeof(void *));
 
-	spin_unlock_irq(&mdev->req_lock);
 }
 
 void tl_restart(struct drbd_conf *mdev, enum drbd_req_event what)
@@ -1476,12 +1472,12 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	if (ns.susp_fen) {
 		/* case1: The outdate peer handler is successful: */
 		if (os.pdsk > D_OUTDATED  && ns.pdsk <= D_OUTDATED) {
-			tl_clear(mdev);
 			if (test_bit(NEW_CUR_UUID, &mdev->flags)) {
 				drbd_uuid_new_current(mdev);
 				clear_bit(NEW_CUR_UUID, &mdev->flags);
 			}
 			spin_lock_irq(&mdev->req_lock);
+			_tl_clear(mdev);
 			_drbd_set_state(_NS(mdev, susp_fen, 0), CS_VERBOSE, NULL);
 			spin_unlock_irq(&mdev->req_lock);
 		}

drivers/block/drbd/drbd_nl.c

@@ -674,8 +674,8 @@ enum determine_dev_size drbd_determine_dev_size(struct drbd_conf *mdev, enum dds
 			 la_size_changed && md_moved ? "size changed and md moved" :
 			 la_size_changed ? "size changed" : "md moved");
 		/* next line implicitly does drbd_suspend_io()+drbd_resume_io() */
-		err = drbd_bitmap_io(mdev, &drbd_bm_write,
-				"size changed", BM_LOCKED_MASK);
+		err = drbd_bitmap_io(mdev, md_moved ? &drbd_bm_write_all : &drbd_bm_write,
+				     "size changed", BM_LOCKED_MASK);
 		if (err) {
 			rv = dev_size_error;
 			goto out;

drivers/block/drbd/drbd_req.c

@@ -695,6 +695,12 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
 		break;
 
 	case resend:
+		/* Simply complete (local only) READs. */
+		if (!(req->rq_state & RQ_WRITE) && !req->w.cb) {
+			_req_may_be_done(req, m);
+			break;
+		}
+
 		/* If RQ_NET_OK is already set, we got a P_WRITE_ACK or P_RECV_ACK
 		   before the connection loss (B&C only); only P_BARRIER_ACK was missing.
 		   Trowing them out of the TL here by pretending we got a BARRIER_ACK
@@ -834,7 +840,15 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns
 		req->private_bio = NULL;
 	}
 	if (rw == WRITE) {
-		remote = 1;
+		/* Need to replicate writes.  Unless it is an empty flush,
+		 * which is better mapped to a DRBD P_BARRIER packet,
+		 * also for drbd wire protocol compatibility reasons. */
+		if (unlikely(size == 0)) {
+			/* The only size==0 bios we expect are empty flushes. */
+			D_ASSERT(bio->bi_rw & REQ_FLUSH);
+			remote = 0;
+		} else
+			remote = 1;
 	} else {
 		/* READ || READA */
 		if (local) {
@@ -870,8 +884,11 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns
 	 * extent.  This waits for any resync activity in the corresponding
 	 * resync extent to finish, and, if necessary, pulls in the target
 	 * extent into the activity log, which involves further disk io because
-	 * of transactional on-disk meta data updates. */
-	if (rw == WRITE && local && !test_bit(AL_SUSPENDED, &mdev->flags)) {
+	 * of transactional on-disk meta data updates.
+	 * Empty flushes don't need to go into the activity log, they can only
+	 * flush data for pending writes which are already in there. */
+	if (rw == WRITE && local && size
+	&& !test_bit(AL_SUSPENDED, &mdev->flags)) {
 		req->rq_state |= RQ_IN_ACT_LOG;
 		drbd_al_begin_io(mdev, sector);
 	}
@@ -994,7 +1011,10 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns
 	if (rw == WRITE && _req_conflicts(req))
 		goto fail_conflicting;
 
-	list_add_tail(&req->tl_requests, &mdev->newest_tle->requests);
+	/* no point in adding empty flushes to the transfer log,
+	 * they are mapped to drbd barriers already. */
+	if (likely(size!=0))
+		list_add_tail(&req->tl_requests, &mdev->newest_tle->requests);
 
 	/* NOTE remote first: to get the concurrent write detection right,
 	 * we must register the request before start of local IO.  */
@@ -1014,6 +1034,14 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns
 	    mdev->net_conf->on_congestion != OC_BLOCK && mdev->agreed_pro_version >= 96)
 		maybe_pull_ahead(mdev);
 
+	/* If this was a flush, queue a drbd barrier/start a new epoch.
+	 * Unless the current epoch was empty anyways, or we are not currently
+	 * replicating, in which case there is no point. */
+	if (unlikely(bio->bi_rw & REQ_FLUSH)
+		&& mdev->newest_tle->n_writes
+		&& drbd_should_do_remote(mdev->state))
+		queue_barrier(mdev);
+
 	spin_unlock_irq(&mdev->req_lock);
 	kfree(b); /* if someone else has beaten us to it... */
 

fs/bio.c

@@ -73,7 +73,7 @@ static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size)
 {
 	unsigned int sz = sizeof(struct bio) + extra_size;
 	struct kmem_cache *slab = NULL;
-	struct bio_slab *bslab;
+	struct bio_slab *bslab, *new_bio_slabs;
 	unsigned int i, entry = -1;
 
 	mutex_lock(&bio_slab_lock);
@@ -97,11 +97,12 @@ static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size)
 
 	if (bio_slab_nr == bio_slab_max && entry == -1) {
 		bio_slab_max <<= 1;
-		bio_slabs = krealloc(bio_slabs,
-				     bio_slab_max * sizeof(struct bio_slab),
-				     GFP_KERNEL);
-		if (!bio_slabs)
+		new_bio_slabs = krealloc(bio_slabs,
+					 bio_slab_max * sizeof(struct bio_slab),
+					 GFP_KERNEL);
+		if (!new_bio_slabs)
 			goto out_unlock;
+		bio_slabs = new_bio_slabs;
 	}
 	if (entry == -1)
 		entry = bio_slab_nr++;

fs/block_dev.c

@@ -1578,10 +1578,12 @@ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov,
 			 unsigned long nr_segs, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
+	struct blk_plug plug;
 	ssize_t ret;
 
 	BUG_ON(iocb->ki_pos != pos);
 
+	blk_start_plug(&plug);
 	ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos);
 	if (ret > 0 || ret == -EIOCBQUEUED) {
 		ssize_t err;
@@ -1590,6 +1592,7 @@ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov,
 		if (err < 0 && ret > 0)
 			ret = err;
 	}
+	blk_finish_plug(&plug);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(blkdev_aio_write);

fs/buffer.c

@@ -914,7 +914,7 @@ link_dev_buffers(struct page *page, struct buffer_head *head)
 /*
  * Initialise the state of a blockdev page's buffers.
  */ 
-static void
+static sector_t
 init_page_buffers(struct page *page, struct block_device *bdev,
 			sector_t block, int size)
 {
@@ -936,33 +936,41 @@ init_page_buffers(struct page *page, struct block_device *bdev,
 		block++;
 		bh = bh->b_this_page;
 	} while (bh != head);
+
+	/*
+	 * Caller needs to validate requested block against end of device.
+	 */
+	return end_block;
 }
 
 /*
  * Create the page-cache page that contains the requested block.
  *
- * This is user purely for blockdev mappings.
+ * This is used purely for blockdev mappings.
  */
-static struct page *
+static int
 grow_dev_page(struct block_device *bdev, sector_t block,
-		pgoff_t index, int size)
+		pgoff_t index, int size, int sizebits)
 {
 	struct inode *inode = bdev->bd_inode;
 	struct page *page;
 	struct buffer_head *bh;
+	sector_t end_block;
+	int ret = 0;		/* Will call free_more_memory() */
 
 	page = find_or_create_page(inode->i_mapping, index,
 		(mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS)|__GFP_MOVABLE);
 	if (!page)
-		return NULL;
+		return ret;
 
 	BUG_ON(!PageLocked(page));
 
 	if (page_has_buffers(page)) {
 		bh = page_buffers(page);
 		if (bh->b_size == size) {
-			init_page_buffers(page, bdev, block, size);
-			return page;
+			end_block = init_page_buffers(page, bdev,
+						index << sizebits, size);
+			goto done;
 		}
 		if (!try_to_free_buffers(page))
 			goto failed;
@@ -982,14 +990,14 @@ grow_dev_page(struct block_device *bdev, sector_t block,
 	 */
 	spin_lock(&inode->i_mapping->private_lock);
 	link_dev_buffers(page, bh);
-	init_page_buffers(page, bdev, block, size);
+	end_block = init_page_buffers(page, bdev, index << sizebits, size);
 	spin_unlock(&inode->i_mapping->private_lock);
-	return page;
-
+done:
+	ret = (block < end_block) ? 1 : -ENXIO;
 failed:
 	unlock_page(page);
 	page_cache_release(page);
-	return NULL;
+	return ret;
 }
 
 /*
@@ -999,7 +1007,6 @@ grow_dev_page(struct block_device *bdev, sector_t block,
 static int
 grow_buffers(struct block_device *bdev, sector_t block, int size)
 {
-	struct page *page;
 	pgoff_t index;
 	int sizebits;
 
@@ -1023,22 +1030,14 @@ grow_buffers(struct block_device *bdev, sector_t block, int size)
 			bdevname(bdev, b));
 		return -EIO;
 	}
-	block = index << sizebits;
+
 	/* Create a page with the proper size buffers.. */
-	page = grow_dev_page(bdev, block, index, size);
-	if (!page)
-		return 0;
-	unlock_page(page);
-	page_cache_release(page);
-	return 1;
+	return grow_dev_page(bdev, block, index, size, sizebits);
 }
 
 static struct buffer_head *
 __getblk_slow(struct block_device *bdev, sector_t block, int size)
 {
-	int ret;
-	struct buffer_head *bh;
-
 	/* Size must be multiple of hard sectorsize */
 	if (unlikely(size & (bdev_logical_block_size(bdev)-1) ||
 			(size < 512 || size > PAGE_SIZE))) {
@@ -1051,21 +1050,20 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size)
 		return NULL;
 	}
 
-retry:
-	bh = __find_get_block(bdev, block, size);
-	if (bh)
-		return bh;
+	for (;;) {
+		struct buffer_head *bh;
+		int ret;
 
-	ret = grow_buffers(bdev, block, size);
-	if (ret == 0) {
-		free_more_memory();
-		goto retry;
-	} else if (ret > 0) {
 		bh = __find_get_block(bdev, block, size);
 		if (bh)
 			return bh;
+
+		ret = grow_buffers(bdev, block, size);
+		if (ret < 0)
+			return NULL;
+		if (ret == 0)
+			free_more_memory();
 	}
-	return NULL;
 }
 
 /*
@@ -1321,10 +1319,6 @@ EXPORT_SYMBOL(__find_get_block);
  * which corresponds to the passed block_device, block and size. The
  * returned buffer has its reference count incremented.
  *
- * __getblk() cannot fail - it just keeps trying.  If you pass it an
- * illegal block number, __getblk() will happily return a buffer_head
- * which represents the non-existent block.  Very weird.
- *
  * __getblk() will lock up the machine if grow_dev_page's try_to_free_buffers()
  * attempt is failing.  FIXME, perhaps?
  */

fs/direct-io.c

@@ -1062,6 +1062,7 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	unsigned long user_addr;
 	size_t bytes;
 	struct buffer_head map_bh = { 0, };
+	struct blk_plug plug;
 
 	if (rw & WRITE)
 		rw = WRITE_ODIRECT;
@@ -1177,6 +1178,8 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 				PAGE_SIZE - user_addr / PAGE_SIZE);
 	}
 
+	blk_start_plug(&plug);
+
 	for (seg = 0; seg < nr_segs; seg++) {
 		user_addr = (unsigned long)iov[seg].iov_base;
 		sdio.size += bytes = iov[seg].iov_len;
@@ -1235,6 +1238,8 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	if (sdio.bio)
 		dio_bio_submit(dio, &sdio);
 
+	blk_finish_plug(&plug);
+
 	/*
 	 * It is possible that, we return short IO due to end of file.
 	 * In that case, we need to release all the pages we got hold on.

include/linux/blkdev.h

@@ -601,7 +601,7 @@ static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
  * it already be started by driver.
  */
 #define RQ_NOMERGE_FLAGS	\
-	(REQ_NOMERGE | REQ_STARTED | REQ_SOFTBARRIER | REQ_FLUSH | REQ_FUA)
+	(REQ_NOMERGE | REQ_STARTED | REQ_SOFTBARRIER | REQ_FLUSH | REQ_FUA | REQ_DISCARD)
 #define rq_mergeable(rq)	\
 	(!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && \
 	 (((rq)->cmd_flags & REQ_DISCARD) || \
@@ -894,6 +894,8 @@ extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
 
 extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
+extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
+			  struct scatterlist *sglist);
 extern void blk_dump_rq_flags(struct request *, char *);
 extern long nr_blockdev_pages(void);
 
@@ -1139,6 +1141,16 @@ static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector
 		& (lim->discard_granularity - 1);
 }
 
+static inline int bdev_discard_alignment(struct block_device *bdev)
+{
+	struct request_queue *q = bdev_get_queue(bdev);
+
+	if (bdev != bdev->bd_contains)
+		return bdev->bd_part->discard_alignment;
+
+	return q->limits.discard_alignment;
+}
+
 static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
 {
 	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)

mm/filemap.c

@@ -1412,12 +1412,8 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 			retval = filemap_write_and_wait_range(mapping, pos,
 					pos + iov_length(iov, nr_segs) - 1);
 			if (!retval) {
-				struct blk_plug plug;
-
-				blk_start_plug(&plug);
 				retval = mapping->a_ops->direct_IO(READ, iocb,
 							iov, pos, nr_segs);
-				blk_finish_plug(&plug);
 			}
 			if (retval > 0) {
 				*ppos = pos + retval;
@@ -2527,14 +2523,12 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_mapping->host;
-	struct blk_plug plug;
 	ssize_t ret;
 
 	BUG_ON(iocb->ki_pos != pos);
 
 	sb_start_write(inode->i_sb);
 	mutex_lock(&inode->i_mutex);
-	blk_start_plug(&plug);
 	ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos);
 	mutex_unlock(&inode->i_mutex);
 
@@ -2545,7 +2539,6 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 		if (err < 0 && ret > 0)
 			ret = err;
 	}
-	blk_finish_plug(&plug);
 	sb_end_write(inode->i_sb);
 	return ret;
 }