diff options
| author | Ming Lei <ming.lei@redhat.com> | 2021-12-03 16:15:32 +0300 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2021-12-04 00:51:29 +0300 |
| commit | 704b914f15fb7daaf517e3acc4bed472b50ca19e (patch) | |
| tree | 5fffc1140fef4ddceeaae58e44fbb65333c33861 /block/blk-mq.h | |
| parent | 2a904d00855f94cb85751e45fa494f225d44ae0d (diff) | |
| download | linux-704b914f15fb7daaf517e3acc4bed472b50ca19e.tar.xz | |
blk-mq: move srcu from blk_mq_hw_ctx to request_queue
In case of BLK_MQ_F_BLOCKING, per-hctx srcu is used to protect dispatch
critical area. However, this srcu instance stays at the end of hctx, and
it often takes standalone cacheline, often cold.
Inside srcu_read_lock() and srcu_read_unlock(), WRITE is always done on
the indirect percpu variable which is allocated from heap instead of
being embedded, srcu->srcu_idx is read only in srcu_read_lock(). It
doesn't matter if srcu structure stays in hctx or request queue.
So switch to per-request-queue srcu for protecting dispatch, and this
way simplifies quiesce a lot, not mention quiesce is always done on the
request queue wide.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211203131534.3668411-3-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block/blk-mq.h')
| -rw-r--r-- | block/blk-mq.h | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/block/blk-mq.h b/block/blk-mq.h index e4c396204928..792f0b29c6eb 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -385,9 +385,9 @@ do { \ int srcu_idx; \ \ might_sleep(); \ - srcu_idx = srcu_read_lock((hctx)->srcu); \ + srcu_idx = srcu_read_lock((hctx)->queue->srcu); \ (dispatch_ops); \ - srcu_read_unlock((hctx)->srcu, srcu_idx); \ + srcu_read_unlock((hctx)->queue->srcu, srcu_idx); \ } \ } while (0) |