From eb619fdb2d4cb8b3d3419e9113921e87e7daf557 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 Nov 2017 08:32:43 -0700 Subject: blk-mq: fix issue with shared tag queue re-running This patch attempts to make the case of hctx re-running on driver tag failure more robust. Without this patch, it's pretty easy to trigger a stall condition with shared tags. An example is using null_blk like this: modprobe null_blk queue_mode=2 nr_devices=4 shared_tags=1 submit_queues=1 hw_queue_depth=1 which sets up 4 devices, sharing the same tag set with a depth of 1. Running a fio job ala: [global] bs=4k rw=randread norandommap direct=1 ioengine=libaio iodepth=4 [nullb0] filename=/dev/nullb0 [nullb1] filename=/dev/nullb1 [nullb2] filename=/dev/nullb2 [nullb3] filename=/dev/nullb3 will inevitably end with one or more threads being stuck waiting for a scheduler tag. That IO is then stuck forever, until someone else triggers a run of the queue. Ensure that we always re-run the hardware queue, if the driver tag we were waiting for got freed before we added our leftover request entries back on the dispatch list. Reviewed-by: Bart Van Assche Tested-by: Bart Van Assche Reviewed-by: Ming Lei Reviewed-by: Omar Sandoval Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux/blk-mq.h') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 674641527da7..4ae987c2352c 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -35,7 +35,7 @@ struct blk_mq_hw_ctx { struct blk_mq_ctx **ctxs; unsigned int nr_ctx; - wait_queue_entry_t dispatch_wait; + wait_queue_entry_t dispatch_wait; atomic_t wait_index; struct blk_mq_tags *tags; @@ -181,8 +181,7 @@ enum { BLK_MQ_S_STOPPED = 0, BLK_MQ_S_TAG_ACTIVE = 1, BLK_MQ_S_SCHED_RESTART = 2, - BLK_MQ_S_TAG_WAITING = 3, - BLK_MQ_S_START_ON_RUN = 4, + BLK_MQ_S_START_ON_RUN = 3, BLK_MQ_MAX_DEPTH = 10240, -- cgit