On Nov 19, 2024 / 21:49, Yi Zhang wrote:
> Skip def_nsid(1) namespace removal during the test as it will be removed
> from _remove_nvmet_ns during _nvmet_target_cleanup phase
>
> $ ./check nvme/052
> nvme/052 (tr=loop) (Test file-ns creation/deletion under one subsystem) [failed]
> runtime 3.273s ... 3.299s
> --- tests/nvme/052.out 2024-11-19 19:29:36.873210200 -0500
> +++ /root/blktests/results/nodev_tr_loop/nvme/052.out.bad 2024-11-19 21:29:26.016088521 -0500
> @@ -1,2 +1,4 @@
> Running nvme/052
> +common/nvme: line 635: /sys/kernel/config/nvmet//subsystems/blktests-subsystem-1/namespaces/1/enable: No such file or directory
> +rmdir: failed to remove '/sys/kernel/config/nvmet//subsystems/blktests-subsystem-1/namespaces/1': No such file or directory
> Test complete
>
> Fixes: 67e25d7 ("nvme/052: do not create namespace when setting up the target")
Oops, thanks for finding this failure. The test case always fail on my system
due to the other kernel issue [1], then I overlooked this failure.
I took a closer look, and found that the failure cause is a missing patch in the
first series of he ANA preparation work [2]. I dropped the patch from the first
series, and planned to include in the third series. But it was required for the
commit 67e25d7 above in the second series. I confirmed that the missing patch
avoids the failure on my system. Will send the patch soon.
[1] https://lore.kernel.org/linux-nvme/tqcy3sveity7p56v7ywp7ssyviwcb3w4623cnxj3knoobfcanq@yxgt2mjkbkam/
[2] https://lore.kernel.org/linux-nvme/20241024010025.2216242-14-shinichiro.kawasaki@xxxxxxx/
> Signed-off-by: Yi Zhang <yi.zhang@xxxxxxxxxx>
> ---
> tests/nvme/052 | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/tests/nvme/052 b/tests/nvme/052
> index 8443c90..2ff9e53 100755
> --- a/tests/nvme/052
> +++ b/tests/nvme/052
> @@ -71,6 +71,8 @@ test() {
> break
> fi
>
> + [ ${nsid} -eq 1 ] && continue
> +
> _remove_nvmet_ns "${def_subsysnqn}" "${nsid}"
>
> # wait until async request is processed and ns is removed
> --
> 2.45.1
>
