linux-hardening - Re: __fortify

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: <202405290950.7E61121D80@keescook>
Date: Wed, 29 May 2024 09:55:32 -0700
From: Kees Cook <keescook@...omium.org>
To: Jeff Johnson <quic_jjohnson@...cinc.com>
Cc: linux-hardening@...r.kernel.org
Subject: Re: __fortify_panic() question

On Wed, May 29, 2024 at 07:36:25AM -0700, Jeff Johnson wrote:
> 'make W=1 C=1' on x86 gives the warning:
> arch/x86/boot/compressed/misc.c:535:6: warning: symbol '__fortify_panic' was not declared. Should it be static?

Hm, I can't reproduce this currently (but yes, it looks like arm vs x86
is mismatched). What tree is this?

> Looking at this I see for ARM there is a prototype for __fortify_panic() in
> arch/arm/boot/compressed/misc.h
> And there is a matching implementation in arch/arm/boot/compressed/misc.c
> 
> But for x86 there is only the implementation in
> arch/x86/boot/compressed/misc.c
> There is not a prototype in arch/x86/boot/compressed/misc.h.
> 
> The easy fix for this would be to add a prototype to
> arch/x86/boot/compressed/misc.h.

Yeah, I think this is the right solution.

> But it seems strange to me to add a prototype to a header file that is only 
> for the benefit of the callee and is not the prototype/header used by the 
> caller, in this case the one in include/linux/fortify-string.h

The stuff in boot/ doesn't tend to include fortify-string.h (since it's
sort of "outside" the kernel), hence the need for additional prototypes.

-- 
Kees Cook