A second experiment I ran was to detect and copy the REQ_WRITE_SAME flag in the raid1 make_request. I bench-tested this with a quick mkfs / mount / copy and didn't see any ill effects. I'm sure there must be more to adding support (like that sector math just north of the bio clone), but I figured I'd post it as part of my "what I did today" post. Thanks, -- Joe >From 1a77ac87ef15b751f0c7fd74e2d5443d4c30c16f Mon Sep 17 00:00:00 2001 From: Joe Lawrence <joe.lawrence@xxxxxxxxxxx> Date: Mon, 10 Dec 2012 17:38:18 -0500 Subject: [PATCH] MD Raid1 make_request needs to copy REQ_WRITE_SAME into bio clones. MD Raid1 doesn't apply the REQ_WRITE_SAME bit to cloned bio writes. The block and SCSI layers now support this flag, with the former utilitizing it if the device queue limits contain a non-zero max_write_same_sectors value. Signed-off-by: Joe Lawrence <joe.lawrence@xxxxxxxxxxx> --- drivers/md/raid1.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 636bae0..d77d272 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1001,6 +1001,7 @@ static void make_request(struct mddev *mddev, struct bio * bio) const unsigned long do_flush_fua = (bio->bi_rw & (REQ_FLUSH | REQ_FUA)); const unsigned long do_discard = (bio->bi_rw & (REQ_DISCARD | REQ_SECURE)); + const unsigned long do_same = (bio->bi_rw & REQ_WRITE_SAME); struct md_rdev *blocked_rdev; struct blk_plug_cb *cb; struct raid1_plug_cb *plug = NULL; @@ -1302,7 +1303,7 @@ read_again: conf->mirrors[i].rdev->data_offset); mbio->bi_bdev = conf->mirrors[i].rdev->bdev; mbio->bi_end_io = raid1_end_write_request; - mbio->bi_rw = WRITE | do_flush_fua | do_sync | do_discard; + mbio->bi_rw = WRITE | do_flush_fua | do_sync | do_discard | do_same; mbio->bi_private = r1_bio; atomic_inc(&r1_bio->remaining); -- 1.7.11.7 -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
