The 1st patch fixes one race between resettting failure and remove. The 2nd patch avoids hanging in del_gendisk() if del_gendisk() waits for these submitted writeback requests. Keith, about the 1st patch, as we talked, there are two directions for fixing the 1st issue, I know you may be working towards treating this reset failure as controller dead, and not completed yet. But if nvme_remove_dead_ctrl() is needed in any one of reset failure path, this patch is still required. Seems both two directions aren't contradictorily. Ming Lei (2): nvme: fix race between removing and reseting failure nvme: avoid to hang in remove disk drivers/nvme/host/core.c | 8 ++++++++ drivers/nvme/host/pci.c | 9 ++++++++- 2 files changed, 16 insertions(+), 1 deletion(-) -- 2.9.3
