From: Laibin Qiu qiulaibin@huawei.com
mainline inclusion from mainline-v5.17-rc1 commit 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IAQPKU CVE: NA
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U);
tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx.
Fix this by recalculating wake_batch when inc or dec active_queues.
Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei ming.lei@redhat.com Suggested-by: John Garry john.garry@huawei.com Signed-off-by: Laibin Qiu qiulaibin@huawei.com Reviewed-by: Andy Shevchenko andriy.shevchenko@linux.intel.com Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe axboe@kernel.dk Conflicts: block/blk-mq-tag.c [Context conflicts] Signed-off-by: Yu Kuai yukuai3@huawei.com --- block/blk-mq-tag.c | 42 +++++++++++++++++++++++++++++++++-------- include/linux/sbitmap.h | 11 +++++++++++ lib/sbitmap.c | 25 +++++++++++++++++++++--- 3 files changed, 67 insertions(+), 11 deletions(-)
diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index 24b48a2f7fba..676d56a04094 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -18,6 +18,21 @@
#define BLK_MQ_DTAG_WAIT_EXPIRE (5 * HZ)
+/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -26,18 +41,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_sbitmap_shared(hctx->flags)) { struct request_queue *q = hctx->queue;
- if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } }
+ users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; }
@@ -58,6 +81,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users;
if (blk_mq_is_sbitmap_shared(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -65,13 +89,15 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) if (!test_and_clear_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) return; - atomic_dec(&tags->active_queues); } else { if (!test_and_clear_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) return; - atomic_dec(&tags->active_queues); }
+ users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); + blk_mq_tag_wakeup_all(tags, false); }
diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index 86004ddc60bd..a0c8bd1c83ad 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -381,6 +381,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); }
+/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 267aa7709416..168d8ddb73e0 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -395,10 +395,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node);
-static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i;
if (sbq->wake_batch != wake_batch) { @@ -414,6 +413,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } }
+static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);