When Sashiko updates baselines?

When Sashiko updates baselines?

[SeongJae Park]

Hello,


I posted my patch series [1] after rebasing to the latest mm-new
(0cec77cfd5314c0b3b03530abe1a4b32e991f639).  And Sashiko smartly tried to apply
it on the mm-new, but an older version of the mm-new
(7505bb13a9bb1f214310915ccc06643119fdafc9), and therefore failed [2].

I could rebase my series on the older version of mm-new and repost, but
wondering if Sashiko will use the lateset mm-new for the new version and fail
again.  Is there a way to predict what baseline Sashiko will use, or give a
hint [3]?

[1] https://sashiko.dev/#/patchset/20260514140904.119781-1-sj%40kernel.org
[2] https://sashiko.dev/#/log/baseline/20264/2
[3] https://lore.kernel.org/sashiko/20260426204746.93567-1-sj@kernel.org/


Thanks,
SJ

Re: When Sashiko updates baselines?

[Roman Gushchin]

SeongJae Park <sj@kernel.org> writes:

> Hello,
>
>
> I posted my patch series [1] after rebasing to the latest mm-new
> (0cec77cfd5314c0b3b03530abe1a4b32e991f639).  And Sashiko smartly tried to apply
> it on the mm-new, but an older version of the mm-new
> (7505bb13a9bb1f214310915ccc06643119fdafc9), and therefore failed [2].
>
> I could rebase my series on the older version of mm-new and repost, but
> wondering if Sashiko will use the lateset mm-new for the new version and fail
> again.  Is there a way to predict what baseline Sashiko will use, or give a
> hint [3]?

Hey SJ!

Sashiko is periodically fetching from remote repositories, but not on
every patch (that would put too much load on kernel.org). There is some
heuristic in place, but it sounds like it's not perfect.

I'll take a look.

Thanks



>
> [1] https://sashiko.dev/#/patchset/20260514140904.119781-1-sj%40kernel.org
> [2] https://sashiko.dev/#/log/baseline/20264/2
> [3] https://lore.kernel.org/sashiko/20260426204746.93567-1-sj@kernel.org/
>
>
> Thanks,
> SJ