diff options
| author | Tang Junhui <[email protected]> | 2017-09-06 14:25:52 +0800 |
|---|---|---|
| committer | Jens Axboe <[email protected]> | 2017-09-06 08:17:33 -0600 |
| commit | c81ffa32a214c84b08900fbc9d432187bd948eba (patch) | |
| tree | ab00291fe74fbaa552bc3cd7a679492ac74430bc | |
| parent | 4b758df21ee7081ab41448d21d60367efaa625b3 (diff) | |
bcache: fix sequential large write IO bypass
Sequential write IOs were tested with bs=1M by FIO in writeback cache
mode, these IOs were expected to be bypassed, but actually they did not.
We debug the code, and find in check_should_bypass():
if (!congested &&
mode == CACHE_MODE_WRITEBACK &&
op_is_write(bio_op(bio)) &&
(bio-＞bi_opf & REQ_SYNC))
goto rescale
that means, If in writeback mode, a write IO with REQ_SYNC flag will not
be bypassed though it is a sequential large IO, It's not a correct thing
to do actually, so this patch remove these codes.
Signed-off-by: tang.junhui <[email protected]>
Reviewed-by: Kent Overstreet <[email protected]>
Reviewed-by: Eric Wheeler <[email protected]>
Cc: [email protected]
Signed-off-by: Jens Axboe <[email protected]>
| -rw-r--r-- | drivers/md/bcache/request.c | 6 |
1 files changed, 0 insertions, 6 deletions
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 0e1463d0c334..de97d86ddff4 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -400,12 +400,6 @@ static bool check_should_bypass(struct cached_dev *dc, struct bio *bio) if (!congested && !dc->sequential_cutoff) goto rescale; - if (!congested && - mode == CACHE_MODE_WRITEBACK && - op_is_write(bio->bi_opf) && - op_is_sync(bio->bi_opf)) - goto rescale; - spin_lock(&dc->io_lock); hlist_for_each_entry(i, iohash(dc, bio->bi_iter.bi_sector), hash) |