bcache: fix inaccurate result of unused buckets
authorGuoju Fang <fangguoju@gmail.com>
Wed, 24 Apr 2019 16:48:27 +0000 (00:48 +0800)
committerJens Axboe <axboe@kernel.dk>
Wed, 24 Apr 2019 16:56:27 +0000 (10:56 -0600)
To get the amount of unused buckets in sysfs_priority_stats, the code
count the buckets which GC_SECTORS_USED is zero. It's correct and should
not be overwritten by the count of buckets which prio is zero.

Signed-off-by: Guoju Fang <fangguoju@gmail.com>
Signed-off-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
drivers/md/bcache/sysfs.c

index 17bae9c..6cd44d3 100644 (file)
@@ -996,8 +996,6 @@ SHOW(__bch_cache)
                       !cached[n - 1])
                        --n;
 
-               unused = ca->sb.nbuckets - n;
-
                while (cached < p + n &&
                       *cached == BTREE_PRIO)
                        cached++, n--;