linux-kernel - Re: [PATCH v1 5/6] vsnprintf: Mark pointer() with _

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]

Message-ID: <87msdeebyh.fsf@prevas.dk>
Date: Fri, 21 Mar 2025 14:43:18 +0100
From: Rasmus Villemoes <ravi@...vas.dk>
To: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: Petr Mladek <pmladek@...e.com>,  Christophe JAILLET
 <christophe.jaillet@...adoo.fr>,  Kees Cook <kees@...nel.org>,  Steven
 Rostedt <rostedt@...dmis.org>,  "Masami Hiramatsu (Google)"
 <mhiramat@...nel.org>,  linux-kernel@...r.kernel.org,
  linux-hardening@...r.kernel.org,  linux-trace-kernel@...r.kernel.org,
  John Ogness <john.ogness@...utronix.de>,  Sergey Senozhatsky
 <senozhatsky@...omium.org>,  Andy Shevchenko <andy@...nel.org>,  Mathieu
 Desnoyers <mathieu.desnoyers@...icios.com>,  Andrew Morton
 <akpm@...ux-foundation.org>
Subject: Re: [PATCH v1 5/6] vsnprintf: Mark pointer() with __printf() attribute

On Thu, Mar 20 2025, Andy Shevchenko <andriy.shevchenko@...ux.intel.com> wrote:

> pointer() is using printf() type of format, and GCC compiler
> (Debian 14.2.0-17) is not happy about this:
>
> lib/vsprintf.c:2466:17: error: function ‘pointer’ might be a candidate for ‘gnu_printf’ format attribute [-Werror=suggest-attribute=format]
>
> Fix the compilation errors (`make W=1` when CONFIG_WERROR=y, which is default)
> by adding __printf() attribute.
>

I had quite a bit of trouble reproducing, until I realized I had to
apply your patches in reverse order, because adding the attribute to one
function will then "taint" its callers.

So this one seems to be self-inflicted pain by the annotation of
va_format (which is completely broken, I'll reply separately to that
one). This doesn't solve the false warning for va_format(), but how
about we at least do

 static char *va_format(char *buf, char *end, struct va_format *va_fmt,
-                      struct printf_spec spec, const char *fmt)
+                      struct printf_spec spec)
 {

        case 'V':
-               return va_format(buf, end, ptr, spec, fmt);
+               return va_format(buf, end, ptr, spec);
        case 'K':

because va_format() doesn't use that fmt argument at all.

Rasmus