diff options
author | William Butler <wab@google.com> | 2024-01-10 21:28:55 +0300 |
---|---|---|
committer | Keith Busch <kbusch@kernel.org> | 2024-01-10 21:36:12 +0300 |
commit | 06c59d427017fcde3107c236177fcc74c9db7909 (patch) | |
tree | 2698b549131f5f196eaa266395e698358bb480e1 /drivers | |
parent | 9a1abc24850eb759e36a2f8869161c3b7254c904 (diff) | |
download | linux-06c59d427017fcde3107c236177fcc74c9db7909.tar.xz |
nvme-pci: set doorbell config before unquiescing
During resets, if queues are unquiesced first, then the host can submit
IOs to the controller using shadow doorbell logic but the controller
won't be aware. This can lead to necessary MMIO doorbells from being
not issued, causing requests to be delayed and timed-out.
Signed-off-by: William Butler <wab@google.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Diffstat (limited to 'drivers')
-rw-r--r-- | drivers/nvme/host/pci.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 75e763ce09aa..46d3897b8986 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2746,10 +2746,10 @@ static void nvme_reset_work(struct work_struct *work) * controller around but remove all namespaces. */ if (dev->online_queues > 1) { + nvme_dbbuf_set(dev); nvme_unquiesce_io_queues(&dev->ctrl); nvme_wait_freeze(&dev->ctrl); nvme_pci_update_nr_queues(dev); - nvme_dbbuf_set(dev); nvme_unfreeze(&dev->ctrl); } else { dev_warn(dev->ctrl.device, "IO queues lost\n"); |