Re: Sashiko patch review for live-patching?

On Thu, May 14, 2026 at 12:21 AM Petr Mladek <[email protected]> wrote:
>
> On Wed 2026-05-13 12:47:13, Josh Poimboeuf wrote:
> > On Wed, May 13, 2026 at 10:17:51AM -0700, Song Liu wrote:
> > > Hi Joe,
> > >
> > > On Wed, May 13, 2026 at 9:13 AM Joe Lawrence <[email protected]> wrote:
> > > >
> > > > Hello live-patching maintainers,
> > > >
> > > > I've noticed several references to the Sashiko (https://sashiko.dev/)
> > > > kernel review bot on this list and was wondering if there is interest in
> > > > adding live-patching to the mailing lists Sashiko tracks.
> > >
> > > I think it is a great idea. AFAICT, these bots add a lot of values in the
> > > code reviews.
> >
> > +1
> >
> > > > Integration appears straightforward: we can submit an MR to add our
> > > > entry to sashiko-k8s.yaml and customize the bot's email behavior in
> > > > email_policy.toml.
> > > >
> > > > Full Sashiko Maintainer documentation is available here:
> > > > https://github.com/sashiko-dev/sashiko/blob/main/MAINTAINERS_GUIDE.md
> > > >
> > > > Personally, I would vote to set reply_to_author.  I don't have a strong
> > > > opinion on the other custom options, provided that the CC list is opt-in
> > > > rather than simply mirrored from the MAINTAINERS::LIVE PATCHING file.
> > > > Either way, I've found the Sashiko web interface very helpful in patch
> > > > review.
> > >
> > > Given the relatively low volume of patches to the livepatch mail list, I
> > > think we can use reply_all. But if folks prefer reply_to_author instead,
> > > we sure can use the cc list.
> >
> > I would vote reply_all.  The signal/noise ratio isn't perfect, but it's
> > high enough to be useful in many cases.  That way the
> > maintainers/reviewers are aware of any potential issues, and it avoids
> > duplicating review work and fragmenting conversations.
>
> I agree. And it might even motivate us to update the subsystem
> specific review prompts so that the review gets improved over time.

+1 for livepatch specific prompts. We sure have some unique constraints
due to the nature of livepatch.

Thanks,
Song