block: add plug for blkdev_issue_discard

Now discard merge works, add plug for blkdev_issue_discard. This will help discard request merge especially for raid0 case. In raid0, a big discard request is split to small requests, and if correct plug is added, such small requests can be merged in low layer.

Signed-off-by Shaohua Li <[email protected]>

---
 block/blk-lib.c |    3 +++
 1 file changed, 3 insertions(+)

Index: linux/block/blk-lib.c
===================================================================

block/blk-lib.c 2012-10-09 08:45:04.358598888 +0800 renamed from block/blk-lib.c 2012-09-28 15:09:38.655327319 +0800 +2 -0

@@ -48,6 +48,7 @@ int blkdev_issue_discard(struct block_de
        struct bio_batch bb;
        struct bio *bio;
        int ret = 0;
+       struct blk_plug plug;
 
        if (!q)
                return -ENXIO;

@@ -81,6 +82,7 @@ int blkdev_issue_discard(struct block_de
        bb.flags = 1 << BIO_UPTODATE;
        bb.wait = &wait;
 
+       blk_start_plug(&plug);
        while (nr_sects) {
                unsigned int req_sects;
                sector_t end_sect;

Quoted 36 lines · Shaohua Li

Now discard merge works, add plug for blkdev_issue_discard. This will help discard request merge especially for raid0 case. In raid0, a big discard request is split to small requests, and if correct plug is added, such small requests can be merged in low layer.

Signed-off-by Shaohua Li <[email protected]>

---
 block/blk-lib.c |    3 +++
 1 file changed, 3 insertions(+)

Index: linux/block/blk-lib.c
===================================================================

block/blk-lib.c 2012-10-09 08:45:04.358598888 +0800 renamed from block/blk-lib.c 2012-09-28 15:09:38.655327319 +0800 +3 -0

@@ -48,6 +48,7 @@ int blkdev_issue_discard(struct block_de
        struct bio_batch bb;
        struct bio *bio;
        int ret = 0;
+       struct blk_plug plug;
 
        if (!q)
                return -ENXIO;

@@ -81,6 +82,7 @@ int blkdev_issue_discard(struct block_de
        bb.flags = 1 << BIO_UPTODATE;
        bb.wait = &wait;
 
+       blk_start_plug(&plug);
        while (nr_sects) {
                unsigned int req_sects;
                sector_t end_sect;

@@ -117,6 +119,7 @@ int blkdev_issue_discard(struct block_de
                atomic_inc(&bb.done);
                submit_bio(type, bio);
        }
+       blk_finish_plug(&plug);
 

block/blk-lib.c 2012-10-09 08:45:04.358598888 +0800 int blkdev_issue_discard(struct block_de · line 125

Quoted 2 lines

block/blk-lib.c 2012-10-09 08:45:04.358598888 +0800 renamed from block/blk-lib.c 2012-09-28 15:09:38.655327319 +0800 +0 -0

        /* Wait for bios in-flight */
        if (!atomic_dec_and_test(&bb.done))

That plug looks rather coarse-grained. Any adverse affects when mkfs discards an entire (large) raid0 device? (I suspect not but figured I'd ask)