* nipa H/W testing suddenly failing on CX7
@ 2026-10-23 14:01 Paolo Abeni
2026-10-23 15:27 ` Jakub Kicinski
0 siblings, 1 reply; 2+ messages in thread
From: Paolo Abeni @ 2026-10-23 14:01 UTC (permalink / raw)
To: Tariq Toukan, Leon Romanovsky, Saeed Mahameed, Mark Bloch
Cc: network dev, Matthieu Baerts, Jakub Kicinski
Hi,
Mat noted that H/W self-tests suddenly started failing on the
mentioned NIC:
https://netdev.bots.linux.dev/flakes.html?min-flip=0&br-cnt=20&br-pfx=net-next-hw&ld-remote=netdev-CX7
Judging from the test output, GRO stopped aggregating and the
driver exposes a single queue?!?
I could not spot anything obviously wrong in the relevant
batch:
https://netdev.bots.linux.dev/branch_deltas/net-next-hw-2026-10-22--00-00.html
Could you please have a look?
Thanks,
Paolo
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: nipa H/W testing suddenly failing on CX7
2026-10-23 14:01 nipa H/W testing suddenly failing on CX7 Paolo Abeni
@ 2026-10-23 15:27 ` Jakub Kicinski
0 siblings, 0 replies; 2+ messages in thread
From: Jakub Kicinski @ 2026-10-23 15:27 UTC (permalink / raw)
To: Paolo Abeni
Cc: Tariq Toukan, Leon Romanovsky, Saeed Mahameed, Mark Bloch,
network dev, Matthieu Baerts
On Fri, 23 Oct 2026 16:01:48 +0200 Paolo Abeni wrote:
> Hi,
>
> Mat noted that H/W self-tests suddenly started failing on the
> mentioned NIC:
>
> https://netdev.bots.linux.dev/flakes.html?min-flip=0&br-cnt=20&br-pfx=net-next-hw&ld-remote=netdev-CX7
>
> Judging from the test output, GRO stopped aggregating and the
> driver exposes a single queue?!?
>
> I could not spot anything obviously wrong in the relevant
> batch:
>
> https://netdev.bots.linux.dev/branch_deltas/net-next-hw-2026-10-22--00-00.html
>
> Could you please have a look?
FWIW I think it's the garbage check added to page pool, which Paolo
already identified. Please ignore.
Side note, I don't think we should be expecting vendors to pay
attention to netdev CI. We can set up a group of community developers
who would be willing / able to jump on such issues if maintainers
are overloaded. Vendor help is obviously welcome if voluntary
but I feel like I implied/promised in the past that the CI will not
be a burden to vendors, and this email goes against that.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-10-23 15:27 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-10-23 14:01 nipa H/W testing suddenly failing on CX7 Paolo Abeni
2026-10-23 15:27 ` Jakub Kicinski
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox