linux-kernel - Re: [PATCH] kbuild: Fully disable -Wenum-{compare-conditional,enum-conversion}

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20241016212427.GA695927@thelio-3990X>
Date: Wed, 16 Oct 2024 14:24:27 -0700
From: Nathan Chancellor <nathan@...nel.org>
To: Arnd Bergmann <arnd@...db.de>
Cc: Masahiro Yamada <masahiroy@...nel.org>,
	Nicolas Schier <nicolas@...sle.eu>,
	Nick Desaulniers <ndesaulniers@...gle.com>,
	Bill Wendling <morbo@...gle.com>,
	Justin Stitt <justinstitt@...gle.com>,
	Aleksei Vetrov <vvvvvv@...gle.com>, linux-kbuild@...r.kernel.org,
	linux-kernel@...r.kernel.org, llvm@...ts.linux.dev,
	patches@...ts.linux.dev, stable@...r.kernel.org
Subject: Re: [PATCH] kbuild: Fully disable
 -Wenum-{compare-conditional,enum-conversion}

On Wed, Oct 16, 2024 at 06:31:27PM +0000, Arnd Bergmann wrote:
> On Wed, Oct 16, 2024, at 18:01, Nathan Chancellor wrote:
> > -Wenum-enum-conversion and -Wenum-compare-conditional were strengthened
> > in clang-19 to warn in C mode, which caused the kernel to move them to
> > W=1 in commit 75b5ab134bb5 ("kbuild: Move
> > -Wenum-{compare-conditional,enum-conversion} into W=1") because there
> > were numerous instances of each that would break builds with -Werror.
> > Unfortunately, this is not a full solution, as more and more developers,
> > subsystems, and distributors are building with W=1 as well, so they
> > continue to see the numerous instances of these warnings.
> >
> > Since the move to W=1, there have not been many new instances that have
> > appeared through various build reports and the ones that have appeared
> > seem to be following similar existing patterns, suggesting that most
> > instances of these warnings will not be real issues. The only
> > alternatives for silencing these warnings are adding casts (which is
> > generally seen as an ugly practice) or refactoring the enums to macro
> > defines or a unified enum (which may be undesirable because of type
> > safety in other parts of the code).
> >
> > Disable the warnings altogether so that W=1 users do not see them.
> 
> I don't think we have to go all the way of completely disabling
> the warnings here, they are still potentially useful. I can see
> three ways of being less aggressive with them:
> 
> - keep -Wno-enum-compare-conditional in W=1 and fix up the
>   remaining warnings for that, iirc the Wno-enum-enum-conversion
>   is the one that causes the problems.
> 
> - Move them to W=2 instead of always disabled
> 
> - Leave the warnings enabled for clang-18 and older.

Arnd and I talked about this offline in the ClangBuiltLinux meeting
today. I am going to run my usual test matrix against a tree with
-Wenum-compare-conditional turned on to see how many instances of these
warnings are in the tree and how difficult it would be to silence them
to address the first point above. I will move -Wenum-enum-conversion to
W=2 and send that as v2 soon to satisfy point two, which should clear up
the blockage for the Android folks.

While disabling the warnings for clang-19 and newer and leaving them on
for clang-18 and older would technically address the issue at hand, it
won't result in increased coverage because the whole point of the change
that caused this in clang-19 is enabling the warning for C code, so
clang-18 and older won't ever emit these warnings.

Cheers,
Nathan