dm cache: always split discards on cache block boundaries
authorHeinz Mauelshagen <heinzm@redhat.com>
Fri, 23 May 2014 18:10:01 +0000 (14:10 -0400)
committerJiri Slaby <jslaby@suse.cz>
Fri, 20 Jun 2014 15:34:00 +0000 (17:34 +0200)
commit f1daa838e861ae1a0fb7cd9721a21258430fcc8c upstream.

The DM cache target cannot cope with discards that span multiple cache
blocks, so each discard bio that spans more than one cache block must
get split by the DM core.

Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
drivers/md/dm-cache-target.c

index d64bf7d6c8fef58bb9775359028cbcdd95f396b7..0cf3700bfe9e3048fed549da8bc38e1dcd80b86f 100644 (file)
@@ -1932,6 +1932,8 @@ static int cache_create(struct cache_args *ca, struct cache **result)
        ti->num_discard_bios = 1;
        ti->discards_supported = true;
        ti->discard_zeroes_data_unsupported = true;
+       /* Discard bios must be split on a block boundary */
+       ti->split_discard_bios = true;
 
        cache->features = ca->features;
        ti->per_bio_data_size = get_per_bio_data_size(cache);