From: Keith Busch <keith.busch@xxxxxxxxx>
Otherwise we're never going to complete a command when it is restarted just
after we completed all other outstanding commands in nvme_clear_queue.
The controller must be disabled prior to completing a presumed lost
command, do this by directly shutting down the controller before
queueing the reset work, and return EH_HANDLED from the timeout handler
after we shut the controller down.
Signed-off-by: Keith Busch <keith.busch@xxxxxxxxx>
[hch: split and rebase]
Signed-off-by: Christoph Hellwig <hch@xxxxxx>
---
drivers/nvme/host/pci.c | 25 ++++++++++++++++---------
1 file changed, 16 insertions(+), 9 deletions(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index febcef5..6082f27 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -81,6 +81,7 @@ static int nvme_reset(struct nvme_dev *dev);
static void nvme_process_cq(struct nvme_queue *nvmeq);
static void nvme_unmap_data(struct nvme_dev *dev, struct nvme_iod *iod);
static void nvme_dead_ctrl(struct nvme_dev *dev);
+static void nvme_dev_shutdown(struct nvme_dev *dev);
struct async_cmd_info {
struct kthread_work work;
@@ -1087,17 +1088,23 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req, bool reserved)
struct nvme_command cmd;
/*
- * Schedule controller reset if the command was already aborted once
- * before and still hasn't been returned to the driver, or if this is
- * the admin queue.
+ * Shutdown the controller immediately and schedule a reset if the
+ * command was already aborted once before and still hasn't been
+ * returned to the driver, or if this is the admin queue.
*/
if (!nvmeq->qid || cmd_rq->aborted) {
- if (queue_work(nvme_workq, &dev->reset_work)) {
- dev_warn(dev->dev,
- "I/O %d QID %d timeout, reset controller\n",
- req->tag, nvmeq->qid);
- }
- return BLK_EH_RESET_TIMER;
+ dev_warn(dev->dev,
+ "I/O %d QID %d timeout, reset controller\n",
+ req->tag, nvmeq->qid);
+ nvme_dev_shutdown(dev);
+ queue_work(nvme_workq, &dev->reset_work);
+
+ /*
+ * Mark the request as handled, since the inline shutdown
+ * forces all outstanding requests to complete.
+ */
+ req->errors = NVME_SC_CANCELLED;
+ return BLK_EH_HANDLED;
}
if (!dev->ctrl.abort_limit)
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-block" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
