| Message ID | 20190521032555.31993-1-bob.liu@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [RESEND,v4] blk-mq: fix hang caused by freeze/unfreeze sequence | expand |
On 5/21/19 5:25 AM, Bob Liu wrote: > The following is a description of a hang in blk_mq_freeze_queue_wait(). > The hang happens on attempt to freeze a queue while another task does > queue unfreeze. > > The root cause is an incorrect sequence of percpu_ref_resurrect() and > percpu_ref_kill() and as a result those two can be swapped: > > CPU#0 CPU#1 > ---------------- ----------------- > q1 = blk_mq_init_queue(shared_tags) > > q2 = blk_mq_init_queue(shared_tags): > blk_mq_add_queue_tag_set(shared_tags): > blk_mq_update_tag_set_depth(shared_tags): > list_for_each_entry() > blk_mq_freeze_queue(q1) > > percpu_ref_kill() > > blk_mq_freeze_queue_wait() > > blk_cleanup_queue(q1) > blk_mq_freeze_queue(q1) > > percpu_ref_kill() > ^^^^^^ freeze_depth can't guarantee the order > > blk_mq_unfreeze_queue() > > percpu_ref_resurrect() > > > blk_mq_freeze_queue_wait() > ^^^^^^ Hang here!!!! > > This wrong sequence raises kernel warning: > percpu_ref_kill_and_confirm called more than once on blk_queue_usage_counter_release! > WARNING: CPU: 0 PID: 11854 at lib/percpu-refcount.c:336 percpu_ref_kill_and_confirm+0x99/0xb0 > > But the most unpleasant effect is a hang of a blk_mq_freeze_queue_wait(), > which waits for a zero of a q_usage_counter, which never happens > because percpu-ref was reinited (instead of being killed) and stays in > PERCPU state forever. > > How to reproduce: > - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" > - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 > - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 > > Script.py: > ------ > #!/usr/bin/python3 > > import os > import sys > > while True: > on = "echo 1 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] > off = "echo 0 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] > os.system(on) > os.system(off) > ------ > > This bug was first reported and fixed by Roman, previous discussion: > [1] Message id: 1443287365-4244-7-git-send-email-akinobu.mita@gmail.com > [2] Message id: 1443563240-29306-6-git-send-email-tj@kernel.org > [3] https://patchwork.kernel.org/patch/9268199/ > > Signed-off-by: Roman Pen <roman.penyaev@profitbricks.com> > Signed-off-by: Bob Liu <bob.liu@oracle.com> > Reviewed-by: Ming Lei <ming.lei@redhat.com> > Reviewed-by: Bart Van Assche <bvanassche@acm.org> > --- > block/blk-core.c | 3 ++- > block/blk-mq.c | 19 ++++++++++--------- > include/linux/blkdev.h | 7 ++++++- > 3 files changed, 18 insertions(+), 11 deletions(-) > Reviewed-by: Hannes Reinecke <hare@suse.com> Cheers, Hannes
On Tue, May 21, 2019 at 11:25:55AM +0800, Bob Liu wrote: > How to reproduce: > - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" > - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 > - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 > > Script.py: Can you wire this up for blktests? Otherwise looks fine: Reviewed-by: Christoph Hellwig <hch@lst.de>
On 5/20/19 9:25 PM, Bob Liu wrote: > The following is a description of a hang in blk_mq_freeze_queue_wait(). > The hang happens on attempt to freeze a queue while another task does > queue unfreeze. > > The root cause is an incorrect sequence of percpu_ref_resurrect() and > percpu_ref_kill() and as a result those two can be swapped: > > CPU#0 CPU#1 > ---------------- ----------------- > q1 = blk_mq_init_queue(shared_tags) > > q2 = blk_mq_init_queue(shared_tags): > blk_mq_add_queue_tag_set(shared_tags): > blk_mq_update_tag_set_depth(shared_tags): > list_for_each_entry() > blk_mq_freeze_queue(q1) > > percpu_ref_kill() > > blk_mq_freeze_queue_wait() > > blk_cleanup_queue(q1) > blk_mq_freeze_queue(q1) > > percpu_ref_kill() > ^^^^^^ freeze_depth can't guarantee the order > > blk_mq_unfreeze_queue() > > percpu_ref_resurrect() > > > blk_mq_freeze_queue_wait() > ^^^^^^ Hang here!!!! > > This wrong sequence raises kernel warning: > percpu_ref_kill_and_confirm called more than once on blk_queue_usage_counter_release! > WARNING: CPU: 0 PID: 11854 at lib/percpu-refcount.c:336 percpu_ref_kill_and_confirm+0x99/0xb0 > > But the most unpleasant effect is a hang of a blk_mq_freeze_queue_wait(), > which waits for a zero of a q_usage_counter, which never happens > because percpu-ref was reinited (instead of being killed) and stays in > PERCPU state forever. > > How to reproduce: > - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" > - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 > - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 > > Script.py: > ------ > #!/usr/bin/python3 > > import os > import sys > > while True: > on = "echo 1 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] > off = "echo 0 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] > os.system(on) > os.system(off) > ------ > > This bug was first reported and fixed by Roman, previous discussion: > [1] Message id: 1443287365-4244-7-git-send-email-akinobu.mita@gmail.com > [2] Message id: 1443563240-29306-6-git-send-email-tj@kernel.org > [3] https://patchwork.kernel.org/patch/9268199/ Applied, thanks.
diff --git a/block/blk-core.c b/block/blk-core.c index 419d600..1bf83a0 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -413,7 +413,7 @@ int blk_queue_enter(struct request_queue *q, blk_mq_req_flags_t flags) smp_rmb(); wait_event(q->mq_freeze_wq, - (atomic_read(&q->mq_freeze_depth) == 0 && + (!q->mq_freeze_depth && (pm || (blk_pm_request_resume(q), !blk_queue_pm_only(q)))) || blk_queue_dying(q)); @@ -503,6 +503,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) spin_lock_init(&q->queue_lock); init_waitqueue_head(&q->mq_freeze_wq); + mutex_init(&q->mq_freeze_lock); /* * Init percpu_ref in atomic mode so that it's faster to shutdown. diff --git a/block/blk-mq.c b/block/blk-mq.c index 08a6248..32b8ad3 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -144,13 +144,14 @@ void blk_mq_in_flight_rw(struct request_queue *q, struct hd_struct *part, void blk_freeze_queue_start(struct request_queue *q) { - int freeze_depth; - - freeze_depth = atomic_inc_return(&q->mq_freeze_depth); - if (freeze_depth == 1) { + mutex_lock(&q->mq_freeze_lock); + if (++q->mq_freeze_depth == 1) { percpu_ref_kill(&q->q_usage_counter); + mutex_unlock(&q->mq_freeze_lock); if (queue_is_mq(q)) blk_mq_run_hw_queues(q, false); + } else { + mutex_unlock(&q->mq_freeze_lock); } } EXPORT_SYMBOL_GPL(blk_freeze_queue_start); @@ -199,14 +200,14 @@ EXPORT_SYMBOL_GPL(blk_mq_freeze_queue); void blk_mq_unfreeze_queue(struct request_queue *q) { - int freeze_depth; - - freeze_depth = atomic_dec_return(&q->mq_freeze_depth); - WARN_ON_ONCE(freeze_depth < 0); - if (!freeze_depth) { + mutex_lock(&q->mq_freeze_lock); + q->mq_freeze_depth--; + WARN_ON_ONCE(q->mq_freeze_depth < 0); + if (!q->mq_freeze_depth) { percpu_ref_resurrect(&q->q_usage_counter); wake_up_all(&q->mq_freeze_wq); } + mutex_unlock(&q->mq_freeze_lock); } EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 1aafeb9..592669b 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -542,7 +542,7 @@ struct request_queue { struct list_head unused_hctx_list; spinlock_t unused_hctx_lock; - atomic_t mq_freeze_depth; + int mq_freeze_depth; #if defined(CONFIG_BLK_DEV_BSG) struct bsg_class_device bsg_dev; @@ -554,6 +554,11 @@ struct request_queue { #endif struct rcu_head rcu_head; wait_queue_head_t mq_freeze_wq; + /* + * Protect concurrent access to q_usage_counter by + * percpu_ref_kill() and percpu_ref_reinit(). + */ + struct mutex mq_freeze_lock; struct percpu_ref q_usage_counter; struct blk_mq_tag_set *tag_set;
The following is a description of a hang in blk_mq_freeze_queue_wait(). The hang happens on attempt to freeze a queue while another task does queue unfreeze. The root cause is an incorrect sequence of percpu_ref_resurrect() and percpu_ref_kill() and as a result those two can be swapped: CPU#0 CPU#1 ---------------- ----------------- q1 = blk_mq_init_queue(shared_tags) q2 = blk_mq_init_queue(shared_tags): blk_mq_add_queue_tag_set(shared_tags): blk_mq_update_tag_set_depth(shared_tags): list_for_each_entry() blk_mq_freeze_queue(q1) > percpu_ref_kill() > blk_mq_freeze_queue_wait() blk_cleanup_queue(q1) blk_mq_freeze_queue(q1) > percpu_ref_kill() ^^^^^^ freeze_depth can't guarantee the order blk_mq_unfreeze_queue() > percpu_ref_resurrect() > blk_mq_freeze_queue_wait() ^^^^^^ Hang here!!!! This wrong sequence raises kernel warning: percpu_ref_kill_and_confirm called more than once on blk_queue_usage_counter_release! WARNING: CPU: 0 PID: 11854 at lib/percpu-refcount.c:336 percpu_ref_kill_and_confirm+0x99/0xb0 But the most unpleasant effect is a hang of a blk_mq_freeze_queue_wait(), which waits for a zero of a q_usage_counter, which never happens because percpu-ref was reinited (instead of being killed) and stays in PERCPU state forever. How to reproduce: - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 Script.py: ------ #!/usr/bin/python3 import os import sys while True: on = "echo 1 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] off = "echo 0 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] os.system(on) os.system(off) ------ This bug was first reported and fixed by Roman, previous discussion: [1] Message id: 1443287365-4244-7-git-send-email-akinobu.mita@gmail.com [2] Message id: 1443563240-29306-6-git-send-email-tj@kernel.org [3] https://patchwork.kernel.org/patch/9268199/ Signed-off-by: Roman Pen <roman.penyaev@profitbricks.com> Signed-off-by: Bob Liu <bob.liu@oracle.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> --- block/blk-core.c | 3 ++- block/blk-mq.c | 19 ++++++++++--------- include/linux/blkdev.h | 7 ++++++- 3 files changed, 18 insertions(+), 11 deletions(-)