On 6/15/2022 9:42 PM, Shinichiro Kawasaki wrote:

On Jun 16, 2022 / 07:13, Yi Zhang wrote:

quoted

On Thu, Jun 16, 2022 at 6:01 AM Chaitanya Kulkarni
[off-list ref] wrote:

quoted

On 6/15/22 12:47, Bjorn Helgaas wrote:

quoted

On Tue, Jun 14, 2022 at 04:00:45AM +0000, Shinichiro Kawasaki wrote:

quoted

On Jun 14, 2022 / 02:38, Chaitanya Kulkarni wrote:

quoted

Shinichiro,

[snip]

quoted

quoted

quoted

quoted

quoted

I think it is worth adding a testcase to blktests to make sure
these future releases will test this.

Yeah, this WARN is confusing for us then it would be valuable to
test by blktests not to repeat it. One point I wonder is: which test
group the test case will it fall in? The nvme group could be the
group to add, probably.

since this issue been discovered with nvme rescan and revmoe,
it should be added to the nvme category.

We already have nvme/032 which tests nvme rescan/reset/remove and the
issue was reported by running this one, do we still need one more?

That is a point. Current nvme/032 checks nvme pci adapter rescan/reset/remove
during I/O to catch problems in nvme driver and block layer, but actually it
can catch the problem in pci sub-system also. I think Chaitanya's motivation
for the new test case is to distinguish those two.

Yes, exactly.

If we have the new test case, its code will be similar and duplicated as
nvme/032 code. To avoid such duplication, it would be good to improve nvme/032
to have two steps. The 1st step checks that nvme pci adapter rescan/reset/remove
without I/O causes no kernel WARN (or any other unexpected kernel messages). Any
issue found in this step is reported as a pci sub-system issue. The 2nd step
checks nvme pci adapter rescan/reset/remove during I/O, as the current nvme/032
does. With this, we don't need the new test case, but still we can distinguish
the problems in nvme/block sub-system and pci sub-system.

Totally agree with this.

quoted

quoted

quoted

quoted

Another point I wonder is other kernel test suite than blktests.
Don't we have more appropriate test suite to check PCI device
rescan/remove race ? Such a test sounds more like a PCI bus
sub-system test than block/storage test.

I don't think so we could have caught it long time back,
but we clearly did not.

I see, then it looks that blktests is the test suite to test it.

-ck