Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from phobos.denx.de (phobos.denx.de [85.214.62.61]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 47D9CD26D85 for ; Fri, 9 Jan 2026 18:32:15 +0000 (UTC) Received: from h2850616.stratoserver.net (localhost [IPv6:::1]) by phobos.denx.de (Postfix) with ESMTP id A65D78439E; Fri, 9 Jan 2026 19:32:13 +0100 (CET) Authentication-Results: phobos.denx.de; dmarc=pass (p=none dis=none) header.from=konsulko.com Authentication-Results: phobos.denx.de; spf=pass smtp.mailfrom=u-boot-bounces@lists.denx.de Authentication-Results: phobos.denx.de; dkim=pass (1024-bit key; unprotected) header.d=konsulko.com header.i=@konsulko.com header.b="iyTezkz+"; dkim-atps=neutral Received: by phobos.denx.de (Postfix, from userid 109) id 8C92E843A4; Fri, 9 Jan 2026 19:32:12 +0100 (CET) Received: from mail-oa1-x42.google.com (mail-oa1-x42.google.com [IPv6:2001:4860:4864:20::42]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits)) (No client certificate requested) by phobos.denx.de (Postfix) with ESMTPS id 41B6584390 for ; Fri, 9 Jan 2026 19:32:08 +0100 (CET) Authentication-Results: phobos.denx.de; dmarc=pass (p=none dis=none) header.from=konsulko.com Authentication-Results: phobos.denx.de; spf=pass smtp.mailfrom=trini@konsulko.com Received: by mail-oa1-x42.google.com with SMTP id 586e51a60fabf-3e3dac349easo4098585fac.2 for ; Fri, 09 Jan 2026 10:32:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=konsulko.com; s=google; t=1767983527; x=1768588327; darn=lists.denx.de; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:from:to:cc:subject:date:message-id:reply-to; bh=voBzvUjrDIYf6Rntizgimh0SS0F+wzSXt+4cGvtJmww=; b=iyTezkz+hRhgwGBq/QKHw8Iu8BLpp1a6eGHNGmCOGQpfCP8stjGOAIsa+9MF6vEwx6 fNC762aB7/hRYB8kNFfIa7tUk/dxiKKBmEslR5MVX+jHq3lpeUu2FuDFbzlymanN7hT9 lbeoTu5ZTG7sb1XjyOotQCPAjbD6A8Q9RDs8c= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1767983527; x=1768588327; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:x-gm-gg:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=voBzvUjrDIYf6Rntizgimh0SS0F+wzSXt+4cGvtJmww=; b=hOZ6f0B3ZTDoNFTbV7Lme52ZT3YxHYqw9AfjZn4if3CtSI0JEXNbH2eE1ECu1REsEt VeDmPMKt7Zv3rg2IpsEvxf7Ke2XkZVjyjgZE3S01cr4VwfAakzxwaAo38ZYVndbL5DNb sM1S4b385HB8PyfXM9PM3Di844UryKr47/U0xNemQicauQ2nB7r14or6LCI9J1MsXipN RmPBFoaCosPKHO2iSwqvm1AsZCH1OaKPhUUtZX9L5EWcieuy+PGc0aHzh+RmPZcKGNd7 JRPUACy/JVab0ZcZLQinqrCGjj4DhtMrOBJ/EQxQj0yl/hcbpNyOVMdvcxuomu6cc0BH Stig== X-Gm-Message-State: AOJu0Yz4ng3o5In2VnBQTiK/DuFHQr+fL6Z7Q/2p931vLX0VgWXXEOOD L7t3UZOBVzRILn+Ce/4Ii0Y7VwP3a44MiSvkhvqYqHBjYJ0ACLXefoCBYK6S7I0P7/g= X-Gm-Gg: AY/fxX4d3UEjkjnzEvXlVLweMBoSVYot8qpTzpPq16gSI1Zz7DRGDRMNgoT37WtBn82 ai+rbMLB0ECSFlbMoqkadjFaipKiy6f6NE2rPoglCgupJzB/KnfZzfREBAono9yav5L5ZbXK3Ny z+k8SoJhyDQGAuUBbqNcrPJV+4Gg/ocrO7w1yNClpc6TwrhteyWcQcWlb0Bh7dvcZ3r5tatZ++W 0kLUU/HbmK+6FuTL+1fQUzT4xFG2sVjf1kyZgg9u/W+WBhYcS9M/JX+AUs+YlpJcP/Uo9TVDiGW LkEsauOc2Gpvgg/gJ3mT6OQRGJIe4Sz3/wIMApi6J3cH4Ms0GksSvp5PDzO4OT6can476OSeMn5 8fY19X0sP3GdJK7T/IrmlZChpx58rcE9Dni1du6ljUlVR5Fe/tXqNJJJK9tOMU8xcyE3cCmFj6V L6fJ/GzIbrXqYCMbEAf19llX91d7a8I6NYxX9HTmbjzRPetRWKa2xVfmOxR4Ss5PgSt9L3XkOEY ADVwRuSF3lfieU+7kiAIwhGDQUI5r9uefNDlzzu2PRSTJIpjA== X-Google-Smtp-Source: AGHT+IHzMculg/QI4Xc2aaryoVNPW2/VR28zCHzAImYyhfIjh9TM4/gdGtvSMH6t5nT6CMN+W0Zn2A== X-Received: by 2002:a4a:d888:0:b0:65f:6535:c55f with SMTP id 006d021491bc7-65f6535c932mr2448390eaf.29.1767983526768; Fri, 09 Jan 2026 10:32:06 -0800 (PST) Received: from bill-the-cat (fixed-189-203-103-235.totalplay.net. [189.203.103.235]) by smtp.gmail.com with ESMTPSA id 586e51a60fabf-3ffa50dcd81sm7233423fac.17.2026.01.09.10.32.05 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 09 Jan 2026 10:32:05 -0800 (PST) Date: Fri, 9 Jan 2026 12:32:03 -0600 From: Tom Rini To: Quentin Schulz Cc: u-boot@lists.denx.de, E Shattow , Heinrich Schuchardt Subject: Re: [PATCH] doc/develop/process.rst: Expand on the Custodian workflow Message-ID: <20260109183203.GH3416603@bill-the-cat> References: <20260109172905.1446923-1-trini@konsulko.com> <35fe875d-0bcc-452b-bb66-a2158b96629e@cherry.de> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="rZS7KWIATO1IKE1K" Content-Disposition: inline In-Reply-To: <35fe875d-0bcc-452b-bb66-a2158b96629e@cherry.de> X-Clacks-Overhead: GNU Terry Pratchett X-BeenThere: u-boot@lists.denx.de X-Mailman-Version: 2.1.39 Precedence: list List-Id: U-Boot discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: u-boot-bounces@lists.denx.de Sender: "U-Boot" X-Virus-Scanned: clamav-milter 0.103.8 at phobos.denx.de X-Virus-Status: Clean --rZS7KWIATO1IKE1K Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Jan 09, 2026 at 07:08:40PM +0100, Quentin Schulz wrote: > Hi Tom, >=20 > On 1/9/26 6:29 PM, Tom Rini wrote: > > As seen with a recent pull request, there have been implicit assumptions > > about how much custodians can change patches before applying and merging > > them. This has lead to an unfortunate situation with a contributor being > > understandably upset about the changes. > >=20 > > To avoid this in the future, document a few things: > > - Non-trivial changes to make something apply need to have at least > > [name: what] in the commit message. > > - Only trivial and obviously correct changes can be made by the > > custodian, and ideally this is done in a merge commit and not the > > patch itself. > > - It is the responsibility of the custodian to gather relevant tags that > > have been provided by the community. > >=20 > > Link: https://lore.kernel.org/u-boot/1faefbf2-e8b0-4c94-85f5-025355915e= 0d@gmx.de/ > > Signed-off-by: Tom Rini > > --- > > With this, Heinrich, per E Shattow's comment on IRC, can you please > > revert the commit in question, then apply his patch as-is, and then your > > changes as a new patch in order to show correct history? > >=20 > > And I realize at this point the subsection of this document is a bit > > wordy. Maybe a follow-up of restructuring this part of the document > > would be helpful? I'm not sure what would flow better however. > >=20 > > Cc: E Shattow > > Cc: Heinrich Schuchardt > > Cc: Quentin Schulz > > --- > > doc/develop/process.rst | 9 ++++++++- > > 1 file changed, 8 insertions(+), 1 deletion(-) > >=20 > > diff --git a/doc/develop/process.rst b/doc/develop/process.rst > > index 4bfbf0eb9c63..84ab09132a0b 100644 > > --- a/doc/develop/process.rst > > +++ b/doc/develop/process.rst > > @@ -244,7 +244,12 @@ like this: > > custodian can also accept patches against older code. It can be > > difficult to find the correct balance between putting too much = work on > > the custodian or too much work on an individual submitting a pa= tch when > > - something does not apply cleanly. > > + something does not apply cleanly. If non-trivial changes need to= be made > > + for it to apply the custodian is expected to annotate the commit= message > > + with ``[name: note about changes made]``. Furthermore, only triv= ial and >=20 > We should provide an example, I think. >=20 > commit 66be03b7ee19444b23aae3990a434a7470fc1641 > Author: J=E9r=E9mie Dautheribes > Date: Fri Nov 28 12:03:04 2025 +0100 >=20 > binman: blob_dtb: improve error message when SPL is not found >=20 > When using binman with the '-a spl-dtb=3Dy' flag, if the SPL blob is = not > found, binman throws a cryptic error message: > binman: 'NoneType' object has no attribute 'startswith' >=20 > Let's improve the error message to explicitly state which SPL blob is > missing. > This is particularly useful when binman is used as a standalone tool > outside the U-Boot source tree. >=20 > Signed-off-by: J=E9r=E9mie Dautheribes > [trini: Add '# pragma: no cover' because coverage doesn't seem to like > the documentation about this error] > Signed-off-by: Tom Rini >=20 > is a good one I think. >=20 > You also added your SoB which isn't mentioned as a requirement here. I > believe it must, because whatever you added/changed, you must comply with > DCO to do it. >=20 > I'm tempted to say we maybe should add a link to the original (unmerged) > commit so that people can actually do the diff themselves if they want to? >=20 > For the above example, have >=20 > Link: https://lore.kernel.org/u-boot/20251128110304.63138-1-jeremie.dauth= eribes@bootlin.com/ >=20 > Maybe? Not sure how much friction we want to put on custodians. There's > probably already a lot to do :) >=20 > I would really like also that custodians *tell* people that they have mer= ged > the commits with changes. https://docs.u-boot.org/en/latest/develop/proce= ss.html#work-flow-of-a-custodian > says it's only "ideally" they do for merged patches, but I really think we > should tell people when changes are made to their patches. >=20 > > + obviously correct changes can be made, no other changes should b= e made to > > + the patch itself. Ideally all of these changes will be in a merg= e commit, > > + and not the commit itself, but that is not always possible. >=20 > I have never been a maintainer in public spaces, and in private spaces I > enforce a strict rebase policy so I don't have merge commits ever. I > wouldn't know what this means (but maybe custodians do, and that's the mo= st > important as this isn't something contributors need to know?). How are we > supposed to do that? Note also that not everybody know that git log -c (or > --cc? I use -c but it seems to be showing the diff of the whole merge and > not necessarily the conflicts?) needs to be used to see the content of a = git > merge commit (it won't show with git log -p). >=20 > Can we provide an example here of a good merge commit that highlights what > is expected from custodians? >=20 > I cannot really comment on whether this will help custodians do the right > thing as I am no custodian. >=20 > We need to set the expectations for contributors as well. >=20 > """ > Trivial fixes may be applied by custodians when merging your patches, > sometimes without public notice. The operated changes are explicitly list= ed > at the end of the commit log or in a merge commit by the custodian. > """ >=20 > or something like that? I'm wondering where we should put this though. ht= tps://docs.u-boot.org/en/latest/develop/process.html#review-process-git-tags > or https://docs.u-boot.org/en/latest/develop/sending_patches.html? I think > this kind of change should be in the same commit (or one that follows aft= er, > but is done now). >=20 > We can have some words about proper etiquette too? Like avoiding big seri= es, > waiting a bit between versions, not silently ignoring previous reviews for > newer versions (you can disagree but you need to speak your mind publicly > and "resolve the conflict" before sending a newer version (or at the very > least say why something wasn't done in the cover-letter or under the --- > section in a patch), waiting at least a week (more?) to send a ping on yet > unreviewed patches, ... That's what's coming to mind right now, there's > probably more to add? This is a separate topic and I don't want this to b= e a > blocker for this here. That's a lot of feedback, thank you. I think this means for v2, I need to: - Move most of the existing wording to another section, and reference up to it. - Note that some things we have done before, really shouldn't be (in your good example, maybe I should have pushed back on the contributor to try and figure out how to convince Python it *had* coverage and not just pragma past it, to not have the contribution sitting out even logner). - Incorporate much of the above and try and remove other assumptions I did make / added in my wording above. --=20 Tom --rZS7KWIATO1IKE1K Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iHUEABYKAB0WIQTzzqh0PWDgGS+bTHor4qD1Cr/kCgUCaWFJoAAKCRAr4qD1Cr/k Cn8kAP9eO33ooyEmdbnVE3A0+F9jowNITs9uPptLIJMPFHZJjQEA+H+CNJQv8Ne+ Dcp7LxruNNoOskwqZMrNFtHuwLdDgww= =c3XJ -----END PGP SIGNATURE----- --rZS7KWIATO1IKE1K--