| Message ID | 20210731062130.1533893-1-yukuai3@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | blk-mq: clear active_queues before clearing BLK_MQ_F_TAG_QUEUE_SHARED | expand |
On 2021/07/31 14:21, Yu Kuai wrote: > We run a test that delete and recover devcies frequently(two devices on > the same host), and we found that 'active_queues' is super big after a > period of time. > > If device a and device b share a tag set, and a is deleted, then > blk_mq_exit_queue() will clear BLK_MQ_F_TAG_QUEUE_SHARED because there > is only one queue that are using the tag set. However, if b is still > active, the active_queues of b might never be cleared even if b is > deleted. > > Thus clear active_queues before BLK_MQ_F_TAG_QUEUE_SHARED is cleared. > > Signed-off-by: Yu Kuai <yukuai3@huawei.com> > --- > block/blk-mq.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/block/blk-mq.c b/block/blk-mq.c > index 2c4ac51e54eb..2fe396385a4a 100644 > --- a/block/blk-mq.c > +++ b/block/blk-mq.c > @@ -2994,10 +2994,12 @@ static void queue_set_hctx_shared(struct request_queue *q, bool shared) > int i; > > queue_for_each_hw_ctx(q, hctx, i) { > - if (shared) > + if (shared) { > hctx->flags |= BLK_MQ_F_TAG_QUEUE_SHARED; > - else > + } else { > + blk_mq_tag_idle(hctx); > hctx->flags &= ~BLK_MQ_F_TAG_QUEUE_SHARED; > + } > } > } > > ping...
On Sat, Jul 31, 2021 at 02:21:30PM +0800, Yu Kuai wrote: > We run a test that delete and recover devcies frequently(two devices on > the same host), and we found that 'active_queues' is super big after a > period of time. > > If device a and device b share a tag set, and a is deleted, then > blk_mq_exit_queue() will clear BLK_MQ_F_TAG_QUEUE_SHARED because there > is only one queue that are using the tag set. However, if b is still > active, the active_queues of b might never be cleared even if b is > deleted. > > Thus clear active_queues before BLK_MQ_F_TAG_QUEUE_SHARED is cleared. > > Signed-off-by: Yu Kuai <yukuai3@huawei.com> > --- > block/blk-mq.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/block/blk-mq.c b/block/blk-mq.c > index 2c4ac51e54eb..2fe396385a4a 100644 > --- a/block/blk-mq.c > +++ b/block/blk-mq.c > @@ -2994,10 +2994,12 @@ static void queue_set_hctx_shared(struct request_queue *q, bool shared) > int i; > > queue_for_each_hw_ctx(q, hctx, i) { > - if (shared) > + if (shared) { > hctx->flags |= BLK_MQ_F_TAG_QUEUE_SHARED; > - else > + } else { > + blk_mq_tag_idle(hctx); > hctx->flags &= ~BLK_MQ_F_TAG_QUEUE_SHARED; > + } > } > } Looks correct, the only remained queue in tagset has to be idled before clearing BLK_MQ_F_TAG_QUEUE_SHARED: Reviewed-by: Ming Lei <ming.lei@redhat.com> Thanks, Ming
Hi, jens Can you please consider to apply this patch. By the way, sorry about the wrong email I replied. Thanks Kuai On 2021/08/06 10:25, Ming Lei wrote: > On Sat, Jul 31, 2021 at 02:21:30PM +0800, Yu Kuai wrote: >> We run a test that delete and recover devcies frequently(two devices on >> the same host), and we found that 'active_queues' is super big after a >> period of time. >> >> If device a and device b share a tag set, and a is deleted, then >> blk_mq_exit_queue() will clear BLK_MQ_F_TAG_QUEUE_SHARED because there >> is only one queue that are using the tag set. However, if b is still >> active, the active_queues of b might never be cleared even if b is >> deleted. >> >> Thus clear active_queues before BLK_MQ_F_TAG_QUEUE_SHARED is cleared. >> >> Signed-off-by: Yu Kuai <yukuai3@huawei.com> >> --- >> block/blk-mq.c | 6 ++++-- >> 1 file changed, 4 insertions(+), 2 deletions(-) >> >> diff --git a/block/blk-mq.c b/block/blk-mq.c >> index 2c4ac51e54eb..2fe396385a4a 100644 >> --- a/block/blk-mq.c >> +++ b/block/blk-mq.c >> @@ -2994,10 +2994,12 @@ static void queue_set_hctx_shared(struct request_queue *q, bool shared) >> int i; >> >> queue_for_each_hw_ctx(q, hctx, i) { >> - if (shared) >> + if (shared) { >> hctx->flags |= BLK_MQ_F_TAG_QUEUE_SHARED; >> - else >> + } else { >> + blk_mq_tag_idle(hctx); >> hctx->flags &= ~BLK_MQ_F_TAG_QUEUE_SHARED; >> + } >> } >> } > > Looks correct, the only remained queue in tagset has to be idled before > clearing BLK_MQ_F_TAG_QUEUE_SHARED: > > Reviewed-by: Ming Lei <ming.lei@redhat.com> > > > > Thanks, > Ming > > . >
On 7/31/21 12:21 AM, Yu Kuai wrote: > We run a test that delete and recover devcies frequently(two devices on > the same host), and we found that 'active_queues' is super big after a > period of time. > > If device a and device b share a tag set, and a is deleted, then > blk_mq_exit_queue() will clear BLK_MQ_F_TAG_QUEUE_SHARED because there > is only one queue that are using the tag set. However, if b is still > active, the active_queues of b might never be cleared even if b is > deleted. > > Thus clear active_queues before BLK_MQ_F_TAG_QUEUE_SHARED is cleared. Applied, thanks.
diff --git a/block/blk-mq.c b/block/blk-mq.c index 2c4ac51e54eb..2fe396385a4a 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2994,10 +2994,12 @@ static void queue_set_hctx_shared(struct request_queue *q, bool shared) int i; queue_for_each_hw_ctx(q, hctx, i) { - if (shared) + if (shared) { hctx->flags |= BLK_MQ_F_TAG_QUEUE_SHARED; - else + } else { + blk_mq_tag_idle(hctx); hctx->flags &= ~BLK_MQ_F_TAG_QUEUE_SHARED; + } } }
We run a test that delete and recover devcies frequently(two devices on the same host), and we found that 'active_queues' is super big after a period of time. If device a and device b share a tag set, and a is deleted, then blk_mq_exit_queue() will clear BLK_MQ_F_TAG_QUEUE_SHARED because there is only one queue that are using the tag set. However, if b is still active, the active_queues of b might never be cleared even if b is deleted. Thus clear active_queues before BLK_MQ_F_TAG_QUEUE_SHARED is cleared. Signed-off-by: Yu Kuai <yukuai3@huawei.com> --- block/blk-mq.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)