linux-kernel - Re: [PATCH] x86/boot: add prototype for __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: <20240531214545.GPZlpFCaXtTGinbcfl@fat_crate.local>
Date: Fri, 31 May 2024 23:45:45 +0200
From: Borislav Petkov <bp@...en8.de>
To: Kees Cook <kees@...nel.org>
Cc: Jeff Johnson <quic_jjohnson@...cinc.com>,
	Nikolay Borisov <nik.borisov@...e.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org,
	"H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org,
	linux-hardening@...r.kernel.org, kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] x86/boot: add prototype for __fortify_panic()

On Fri, May 31, 2024 at 02:34:07PM -0700, Kees Cook wrote:
> On Fri, May 31, 2024 at 11:20:09PM +0200, Borislav Petkov wrote:
> > So I get an allergic reaction everytime we wag the dog - i.e., fix the
> > code because some tool or option can't handle it even if it is
> > a perfectly fine code. In that case it is an unused symbol.
> > 
> > And frankly, I'd prefer the silly warning to denote that fortify doesn't
> > need to do any checking there vs shutting it up just because.
> 
> If we want to declare that x86 boot will never perform string handling
> on strings with unknown lengths, we could just delete the boot/
> implementation of __fortify_panic(), and make it a hard failure if such
> cases are introduced in the future. This hasn't been a particularly
> friendly solution in the past, though, as the fortify routines do tend
> to grow additional coverage over time, so there may be future cases that
> do trip the runtime checking...

Yes, and we should not do anything right now either.

As said, I'd prefer the warning which actually says that fortify
routines are not used, which in itself is useful information vs shutting
it up.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette