| Message ID | 20250410133029.2487054-14-ming.lei@redhat.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | block: unify elevator changing and fix lockdep warning | expand |
On 4/10/25 7:00 PM, Ming Lei wrote: > Both blk_mq_map_swqueue() and blk_mq_realloc_hw_ctxs() are only called > from queue initialization or updating nr_hw_queues code, in which > elevator switch can't happen any more. > > So remove these ->elevator_lock uses. > But what if blk_mq_map_swqueue runs in parallel, one context from blk_mq_init_allocated_queue and another from blk_mq_update_nr_hw_queues? It seems this is possible due to blk_mq_map_swqueue is invoked right after queue is added in tag-set from blk_mq_init_allocated_queue. Thanks, --Nilay
On Fri, Apr 11, 2025 at 12:37:49AM +0530, Nilay Shroff wrote: > > > On 4/10/25 7:00 PM, Ming Lei wrote: > > Both blk_mq_map_swqueue() and blk_mq_realloc_hw_ctxs() are only called > > from queue initialization or updating nr_hw_queues code, in which > > elevator switch can't happen any more. > > > > So remove these ->elevator_lock uses. > > > But what if blk_mq_map_swqueue runs in parallel, one context from > blk_mq_init_allocated_queue and another from blk_mq_update_nr_hw_queues? > It seems this is possible due to blk_mq_map_swqueue is invoked right > after queue is added in tag-set from blk_mq_init_allocated_queue. Good catch, one simple fix is to swap blk_mq_add_queue_tag_set() with blk_mq_map_swqueue() in blk_mq_init_allocated_queue() since blk_mq_map_swqueue doesn't rely on BLK_MQ_F_TAG_QUEUE_SHARED. Thanks, Ming
diff --git a/block/blk-mq.c b/block/blk-mq.c index 0fb72a698d77..812dfe759b89 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -4095,8 +4095,6 @@ static void blk_mq_map_swqueue(struct request_queue *q) struct blk_mq_ctx *ctx; struct blk_mq_tag_set *set = q->tag_set; - mutex_lock(&q->elevator_lock); - queue_for_each_hw_ctx(q, hctx, i) { cpumask_clear(hctx->cpumask); hctx->nr_ctx = 0; @@ -4201,8 +4199,6 @@ static void blk_mq_map_swqueue(struct request_queue *q) hctx->next_cpu = blk_mq_first_mapped_cpu(hctx); hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH; } - - mutex_unlock(&q->elevator_lock); } /* @@ -4506,16 +4502,9 @@ static void __blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, } static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, - struct request_queue *q, bool lock) + struct request_queue *q) { - if (lock) { - /* protect against switching io scheduler */ - mutex_lock(&q->elevator_lock); - __blk_mq_realloc_hw_ctxs(set, q); - mutex_unlock(&q->elevator_lock); - } else { - __blk_mq_realloc_hw_ctxs(set, q); - } + __blk_mq_realloc_hw_ctxs(set, q); /* unregister cpuhp callbacks for exited hctxs */ blk_mq_remove_hw_queues_cpuhp(q); @@ -4547,7 +4536,7 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, xa_init(&q->hctx_table); - blk_mq_realloc_hw_ctxs(set, q, false); + blk_mq_realloc_hw_ctxs(set, q); if (!q->nr_hw_queues) goto err_hctxs; @@ -4962,7 +4951,7 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, fallback: blk_mq_update_queue_map(set); list_for_each_entry(q, &set->tag_list, tag_set_list) { - blk_mq_realloc_hw_ctxs(set, q, true); + blk_mq_realloc_hw_ctxs(set, q); if (q->nr_hw_queues != set->nr_hw_queues) { int i = prev_nr_hw_queues;
Both blk_mq_map_swqueue() and blk_mq_realloc_hw_ctxs() are only called from queue initialization or updating nr_hw_queues code, in which elevator switch can't happen any more. So remove these ->elevator_lock uses. Signed-off-by: Ming Lei <ming.lei@redhat.com> --- block/blk-mq.c | 19 ++++--------------- 1 file changed, 4 insertions(+), 15 deletions(-)