Re: [PATCH net 0/22] pull-request: can 2026-09-28

On 28.09.2026 18:09:21, Jakub Kicinski wrote:

On Mon, 28 Sep 2026 20:45:07 +0200 Marc Kleine-Budde wrote:

quoted

this is a pull request of 22 patches for net/main.

Oof, Linus just shouted at us for pulling too many LLM-driver fixes
late in the release cycle. Could you whittle this down a little bit?

Sure

At a glance the skb freeing from IRQ, and propagating impossible init
time failures (patch 8) can probably go? 9, 10, 13, 16 too?

yes - I've moved the ones w/o sashiko regressions to the can-next queue.

Driver init error paths and remove races are things I commonly apply
to net-next these days, 'cause those patches usually come from LLM
jockeys and nobody cares in practice.

sigh :/

I pushed the Sashiko review to high prio, so pls wait for it to pass
judgment before you respin:

https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260928193312.553632-1-mkl%40pengutronix.de

Thanks - it stopped on patch #9, which is "In progress..." for several
hours now. While the header shows a big red #Failed.

Up to you what you keep and what you toss, FWIW, I mostly want to avoid
obvious egg-on-face conditions, which I think the driver init error
paths could bring about.

Here's the updated PR:
https://lore.kernel.org/all/20260929210700.1183036-1-mkl@pengutronix.de/ (local)

regards,
Marc

-- 
Pengutronix e.K.                 | Marc Kleine-Budde          |
Embedded Linux                   | https://www.pengutronix.de |
Vertretung Nürnberg              | Phone: +49-5121-206917-129 |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-9   |