Re: [PATCH 09/19] bcache: update bio->bi_opf bypass/writeback REQ_ flag hints
From: Coly Li <hidden>
Date: 2017-07-01 18:49:20
Also in:
linux-bcache
On 2017/7/1 上午4:42, bcache@lists.ewheeler.net wrote:
From: Eric Wheeler <redacted> Bypass if: bio->bi_opf & (REQ_RAHEAD|REQ_BACKGROUND) Writeback if: op_is_sync(bio->bi_opf) || bio->bi_opf & (REQ_META|REQ_PRIO)
Hi Eric, Could you please explain a bit how the above policy is designed ? I'd like to understand it more. Thanks. Coly
quoted hunk ↗ jump to hunk
Signed-off-by: Eric Wheeler <redacted> --- drivers/md/bcache/request.c | 3 +++ drivers/md/bcache/writeback.h | 3 ++- 2 files changed, 5 insertions(+), 1 deletion(-)diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index a95609f..4629b0c 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c@@ -386,6 +386,9 @@ static bool check_should_bypass(struct cached_dev *dc, struct bio *bio) op_is_write(bio_op(bio)))) goto skip; + if (bio->bi_opf & (REQ_RAHEAD|REQ_BACKGROUND)) + goto skip; + /* If the ioprio already exists on the bio, use that. We assume that * the upper layer properly assigned the calling process's ioprio to * the bio being passed to bcache. Otherwise, use current's ioc. */diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h index cd82fe8..ea2f92e 100644 --- a/drivers/md/bcache/writeback.h +++ b/drivers/md/bcache/writeback.h@@ -81,7 +81,8 @@ static inline bool should_writeback(struct cached_dev *dc, struct bio *bio, return true; } - return op_is_sync(bio->bi_opf) || in_use <= CUTOFF_WRITEBACK; + return op_is_sync(bio->bi_opf) || bio->bi_opf & (REQ_META|REQ_PRIO) + || in_use <= CUTOFF_WRITEBACK; } static inline void bch_writeback_queue(struct cached_dev *dc)