linux-kernel - Re: [PATCH v2] minmax: substitute local variables using __UNIQUE

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

Date: Thu, 15 Feb 2024 14:07:04 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Yueh-Shun Li <shamrocklee@...teo.net>
Cc: Mark Brown <broonie@...nel.org>, Andy Shevchenko
 <andriy.shevchenko@...ux.intel.com>, Herve Codina
 <herve.codina@...tlin.com>, Christophe Leroy <christophe.leroy@...roup.eu>,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] minmax: substitute local variables using
 __UNIQUE_ID()

On Thu, 15 Feb 2024 18:58:15 +0000 Yueh-Shun Li <shamrocklee@...teo.net> wrote:

> Substitute identifier names of local variables used in macro
> definitions inside minmax.h with those generated by __UNIQUE_ID(prefix)
> to eliminate passible naming collisions.
> 
> Identifier names like __x, __y and __tmp are everywhere inside the
> kernel source. This patch ensures that macros provided by minmax.h
> will work even when identifiers of these names appear in the expanded
> input arguments.

Makes sense I guess.  However I do wonder how far this goes:

# grep typeof include/linux/*.h | wc -l
313

Many of these are locals being defined within macros.  Do they all need
changing?  If so, do we really want to implement this fix for what has
always been, to my knowledge, a non-problem?