From 07f18ca7da27748f136dce482ec1ba2be71f0c24 Mon Sep 17 00:00:00 2001 From: Tang Junhui Date: Wed, 6 Sep 2017 14:25:52 +0800 Subject: bcache: fix sequential large write IO bypass MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit c81ffa32a214c84b08900fbc9d432187bd948eba upstream. 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 Reviewed-by: Kent Overstreet Reviewed-by: Eric Wheeler Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman --- drivers/md/bcache/request.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index c53cc00202c8..4b413db99276 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) -- cgit v1.2.3