Thanks, applied to nvme-4.12 after fixing up the changelog, and
adding back the (now slightly modified) WARN_ON for a wrong state
in nvme_reset_work.
V2: 1. Use cancel_work_sync instead of introducing new state (Sagi)
2. Skip using NVME_CTRL_SCHED_RESET state (Sagi)
V1: nvme_probe set RESETTING state and check for NVME_CTRL_RESETTING
state in nvme_should_reset instead of work_busy (hch)
This type of information should come after
Commit c5f6ce97c1210 tries to address multiple resets but fails as
work_busy doesn't involve any synchronization and can fail. This is
reproducible easily as can be seen by WARNING below which is triggered
with line:
WARN_ON(dev->ctrl.state == NVME_CTRL_RESETTING)
Allowing multiple resets can re
3 matches
Mail list logo
