Dear all scsi timeout and error handler are based on an assumption that normal completion mustn't do anything on an timeout request. After 12f5b931 (blk-mq: Remove generation seqeunce), we lost this. __blk_mq_complete request could ensure a request won't be completed twice, but it can still complete a timeout request. scsi (even other drivers) have been working on this assumption for many years, it is dangerous to discard it suddenly. This patch set is to regain this. 1st patch - introduce blk_mq_mark_rq_complete in blk_mq_check_expired and blk_mq_complete_request to avoid the race between them. - blk_mq_complete_request cannot complete a timeout request anymore, so have to export __blk_mq_complete_request for LLDD timeout path 2nd ~ 5th patches replace the blk_mq_complete_request to __blk_mq_complete_request. Jianchao Wang(5) blk-mq: prevent normal completion from entering a timeout request nbd: use __blk_mq_complete_request in timeout path null_blk: use __blk_mq_complete_request in timeout path mmc: use __blk_mq_complete_request in timeout path nvme: use __blk_mq_complete_request in timeout path block/blk-mq.c | 22 +++++++++++++++------- drivers/block/nbd.c | 2 +- drivers/block/null_blk.c | 2 +- drivers/mmc/core/queue.c | 2 +- drivers/nvme/host/pci.c | 8 ++++++++ drivers/nvme/host/rdma.c | 1 + drivers/nvme/target/loop.c | 1 + include/linux/blk-mq.h | 1 + include/linux/blkdev.h | 6 ++++++ Thanks Jianchao
