Re: [MAINTAINERS SUMMIT] The place of AI code review in the Linux Kernel process

On Wed, Jul 15, 2026 at 07:51:33PM +0200, Miguel Ojeda wrote:
>On Wed, Jul 15, 2026 at 6:55 PM Roman Gushchin <[email protected]> wrote:
>>
>> * Review of LTS patches and kernel releases.
>
>For the backporting case, I think this could help, and there are
>certain things we could ask for, i.e. custom prompts.
>
>It may also have less noise, in the sense that the AI has the original
>commit in mainline as a baseline, i.e. it is not a completely new
>patch of unknown quality.
>
>I imagine Sasha et al. have experience with this.

I currently utilize AI to review backports that are being sent to stable@,
which has helped with catching many issues that would otherwise be just merged
in the LTS trees.

One piece that is missing is identifying whether a patch is correct when
cherry-picked to an older tree.

I've been (very slowly due to time constraints) working on extending Sashiko to
help with this. You can see an example of that work here:

	http://20.172.224.249:51261/#/backports

But I definitely need to sync more time into this.

-- 
Thanks,
Sasha