Hi, The current SCSI quiesce isn't safe and easy to trigger I/O deadlock. Once SCSI device is put into QUIESCE, no new request except for RQF_PREEMPT can be dispatched to SCSI successfully, and scsi_device_quiesce() just simply waits for completion of I/Os dispatched to SCSI stack. It isn't enough at all. Because new request still can be comming, but all the allocated requests can't be dispatched successfully, so request pool can be consumed up easily. Then request with RQF_PREEMPT can't be allocated and wait forever, meantime scsi_device_resume() waits for completion of RQF_PREEMPT, then system hangs forever, such as during system suspend or sending SCSI domain alidation. Both IO hang inside system suspend[1] or SCSI domain validation were reported before. This patch introduces preempt freeze, and solves the issue by preempt freezing block queue during SCSI quiesce, and allows to allocate request of RQF_PREEMPT when queue is in this state. Oleksandr verified that V3 does fix the hang during suspend/resume, and Cathy verified that revised V3 fixes hang in sending SCSI domain validation. Both SCSI and SCSI_MQ have this IO deadlock issue, this patch fixes them all by introducing/unifying blk_freeze_queue_preempt() and blk_unfreeze_queue_preempt(), and cleanup is done together. The patchset can be found in the following gitweb: https://github.com/ming1/linux/tree/blk_safe_scsi_quiesce_V4 V4: - reorganize patch order to make it more reasonable - support nested preempt freeze, as required by SCSI transport spi - check preempt freezing in slow path of of blk_queue_enter() - add "SCSI: transport_spi: resume a quiesced device" - wake up freeze queue in setting dying for both blk-mq and legacy - rename blk_mq_[freeze|unfreeze]_queue() in one patch - rename .mq_freeze_wq and .mq_freeze_depth - improve comment V3: - introduce q->preempt_unfreezing to fix one bug of preempt freeze - call blk_queue_enter_live() only when queue is preempt frozen - cleanup a bit on the implementation of preempt freeze - only patch 6 and 7 are changed V2: - drop the 1st patch in V1 because percpu_ref_is_dying() is enough as pointed by Tejun - introduce preempt version of blk_[freeze|unfreeze]_queue - sync between preempt freeze and normal freeze - fix warning from percpu-refcount as reported by Oleksandr [1] https://marc.info/?t=150340250100013&r=3&w=2 Thanks, Ming Ming Lei (10): blk-mq: only run hw queues for blk-mq block: tracking request allocation with q_usage_counter blk-mq: rename blk_mq_[freeze|unfreeze]_queue blk-mq: rename blk_mq_freeze_queue_wait as blk_freeze_queue_wait block: rename .mq_freeze_wq and .mq_freeze_depth block: pass flags to blk_queue_enter() block: introduce preempt version of blk_[freeze|unfreeze]_queue block: allow to allocate req with RQF_PREEMPT when queue is preempt frozen SCSI: transport_spi: resume a quiesced device SCSI: preempt freeze block queue when SCSI device is put into quiesce block/bfq-iosched.c | 2 +- block/blk-cgroup.c | 8 +- block/blk-core.c | 95 ++++++++++++++++---- block/blk-mq.c | 180 ++++++++++++++++++++++++++++---------- block/blk-mq.h | 1 - block/blk-timeout.c | 2 +- block/blk.h | 12 +++ block/elevator.c | 4 +- drivers/block/loop.c | 24 ++--- drivers/block/rbd.c | 2 +- drivers/nvme/host/core.c | 8 +- drivers/scsi/scsi_lib.c | 25 +++++- drivers/scsi/scsi_transport_spi.c | 3 + fs/block_dev.c | 4 +- include/linux/blk-mq.h | 15 ++-- include/linux/blkdev.h | 32 +++++-- 16 files changed, 313 insertions(+), 104 deletions(-) -- 2.9.5
