Sashiko discussions
 help / color / mirror / Atom feed
From: Roman Gushchin <roman.gushchin@linux.dev>
To: robin@snyders.xyz
Cc: sashiko@lists.linux.dev
Subject: Re: [Feature request] Support b4 prerequisite-patch-id dependencies when applying patchsets
Date: Mon, 17 Aug 2026 20:20:12 +0000	[thread overview]
Message-ID: <7ia4h5ksebyb.fsf@castle.c.googlers.com> (raw)
In-Reply-To: <b27c34f3-4188-40e4-be65-e90a89890dc1@mail.infomaniak.com> (robin@snyders.xyz's message of "Wed, 12 Aug 2026 11:33:11 +0200")

robin@snyders.xyz writes:

> Hi,
>
> I ran into a Failed To Apply result in Sashiko for a kernel patch series that uses b4 prerequisite dependencies.
>
> Affected Sashiko patchset:
> https://sashiko.dev/#/patchset/20260812-submit-qcom-smbx-send-v1-v2-0-f504b8f9bfad%40snyders.xyz
>
> Lore thread:
> https://lore.kernel.org/r/20260812-submit-qcom-smbx-send-v1-v2-0-f504b8f9bfad@snyders.xyz/
>
> The series is based on:
>
> base-commit: 99b38cda3f4c486cfbc40f3c8ede1703594e9f13
>
> and declares five prerequisite patches using b4 dependency metadata:
>
> prerequisite-patch-id: 3d8a9cf43b5e5eac64a60aeec5e9c90880bcf792
> prerequisite-patch-id: de3e3b494f9644596cfa4cafc362ca0b48288c45
> prerequisite-patch-id: d5e22a0fe7cd55374f60d7c2cb14ed3df8f94e0b
> prerequisite-patch-id: 9eafa41402e97d6a3e5369bec46df4e9db3d2af6
> prerequisite-patch-id: 5638b901cd10773fa9485605f84b25a1dd826a84
>
> Those five prerequisites are the preceding qcom_smbx fixes series:
> https://lore.kernel.org/r/20260812-qcom-smbx-fixes-v1-0-eb48246be599@snyders.xyz/
>
> The SMB5 v2 series intentionally depends on that fixes series because
> both modify overlapping areas of
> drivers/power/supply/qcom_smbx.c. Applying the v2 directly to the base
> tree without first applying the declared prerequisites is therefore
> expected to fail.
>
> On the Sashiko page, the result is Failed To Apply, Baseline shows
> Selected: -, and no AI review is run. From looking at the current
> baseline/application logic, it appears Sashiko applies the patches in
> the current patchset to a candidate baseline but does not resolve and
> apply b4 prerequisite-patch-id dependencies first.
>
> Expected behaviour would be roughly:
>
> Resolve the base commit.
> Resolve each prerequisite-patch-id from public-inbox/Lore.
> Apply the prerequisite patches in dependency order.
> Apply the patchset being reviewed.
> Run the normal review on the resulting tree.
>
> b4 itself supports this dependency metadata, and b4 prep --check-deps can validate that the dependency chain is available and applicable.
>
> If automatically resolving prerequisite-patch-id is intentionally out
> of scope, it would still be useful for Failed To Apply to report that
> unresolved prerequisite metadata was present, rather than making the
> failure look like the submitted series itself does not apply.
>
> Please let me know if you would like any additional reproduction
> details or the exact b4 commands used to prepare the series.

b4 support is in progress:
https://github.com/sashiko-dev/sashiko/pull/389

Thanks

      reply	other threads:[~2026-08-17 20:20 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-12  9:33 [Feature request] Support b4 prerequisite-patch-id dependencies when applying patchsets robin
2026-08-17 20:20 ` Roman Gushchin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7ia4h5ksebyb.fsf@castle.c.googlers.com \
    --to=roman.gushchin@linux.dev \
    --cc=robin@snyders.xyz \
    --cc=sashiko@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox