diff options
| author | Christoph Hellwig <hch@lst.de> | 2025-01-31 13:03:47 +0100 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2025-01-31 07:20:08 -0700 |
| commit | 1e1a9cecfab3f22ebef0a976f849c87be8d03c1c (patch) | |
| tree | 34fa7958ed94c56127aa0fc55347bb409574af3a /block/blk-mq.c | |
| parent | 14ef49657ff3b7156952b2eadcf2e5bafd735795 (diff) | |
block: force noio scope in blk_mq_freeze_queue
When block drivers or the core block code perform allocations with a
frozen queue, this could try to recurse into the block device to
reclaim memory and deadlock. Thus all allocations done by a process
that froze a queue need to be done without __GFP_IO and __GFP_FS.
Instead of tying to track all of them down, force a noio scope as
part of freezing the queue.
Note that nvme is a bit of a mess here due to the non-owner freezes,
and they will be addressed separately.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20250131120352.1315351-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block/blk-mq.c')
| -rw-r--r-- | block/blk-mq.c | 21 |
1 files changed, 13 insertions, 8 deletions
diff --git a/block/blk-mq.c b/block/blk-mq.c index da39a1cac702..40490ac88045 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -210,12 +210,12 @@ int blk_mq_freeze_queue_wait_timeout(struct request_queue *q, } EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_wait_timeout); -void blk_mq_freeze_queue(struct request_queue *q) +void blk_mq_freeze_queue_nomemsave(struct request_queue *q) { blk_freeze_queue_start(q); blk_mq_freeze_queue_wait(q); } -EXPORT_SYMBOL_GPL(blk_mq_freeze_queue); +EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_nomemsave); bool __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic) { @@ -236,12 +236,12 @@ bool __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic) return unfreeze; } -void blk_mq_unfreeze_queue(struct request_queue *q) +void blk_mq_unfreeze_queue_nomemrestore(struct request_queue *q) { if (__blk_mq_unfreeze_queue(q, false)) blk_unfreeze_release_lock(q); } -EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue); +EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue_nomemrestore); /* * non_owner variant of blk_freeze_queue_start @@ -4223,13 +4223,14 @@ static void blk_mq_update_tag_set_shared(struct blk_mq_tag_set *set, bool shared) { struct request_queue *q; + unsigned int memflags; lockdep_assert_held(&set->tag_list_lock); list_for_each_entry(q, &set->tag_list, tag_set_list) { - blk_mq_freeze_queue(q); + memflags = blk_mq_freeze_queue(q); queue_set_hctx_shared(q, shared); - blk_mq_unfreeze_queue(q); + blk_mq_unfreeze_queue(q, memflags); } } @@ -4992,6 +4993,7 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, struct request_queue *q; LIST_HEAD(head); int prev_nr_hw_queues = set->nr_hw_queues; + unsigned int memflags; int i; lockdep_assert_held(&set->tag_list_lock); @@ -5003,8 +5005,10 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, if (set->nr_maps == 1 && nr_hw_queues == set->nr_hw_queues) return; + memflags = memalloc_noio_save(); list_for_each_entry(q, &set->tag_list, tag_set_list) - blk_mq_freeze_queue(q); + blk_mq_freeze_queue_nomemsave(q); + /* * Switch IO scheduler to 'none', cleaning up the data associated * with the previous scheduler. We will switch back once we are done @@ -5052,7 +5056,8 @@ switch_back: blk_mq_elv_switch_back(&head, q); list_for_each_entry(q, &set->tag_list, tag_set_list) - blk_mq_unfreeze_queue(q); + blk_mq_unfreeze_queue_nomemrestore(q); + memalloc_noio_restore(memflags); /* Free the excess tags when nr_hw_queues shrink. */ for (i = set->nr_hw_queues; i < prev_nr_hw_queues; i++) |