linux-hardening - Re: [GIT PULL] FORTIFY_SOURCE updates for v5.18-rc1

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: <202203280854.C36F2EC@keescook>
Date:   Mon, 28 Mar 2022 09:01:42 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        George Burgess IV <gbiv@...gle.com>,
        linux-hardening@...r.kernel.org, llvm@...ts.linux.dev,
        Miguel Ojeda <ojeda@...nel.org>,
        Nathan Chancellor <nathan@...nel.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>
Subject: Re: [GIT PULL] FORTIFY_SOURCE updates for v5.18-rc1

On Sat, Mar 26, 2022 at 12:40:18PM -0700, Linus Torvalds wrote:
> On Sat, Mar 26, 2022 at 12:29 PM Linus Torvalds
> <torvalds@...ux-foundation.org> wrote:
> >
> > Because if all the compiler issues and warnings have been sorted out,
> > it sounds to me like the compile-time side could/should be done
> > unconditionally if there are no runtime downsides.

Yeah, I'd like to do this. The way the header files are currently split
up makes this slightly weird, and there have been issues with some
arch/compiler combinations, so it's not quite as cut-and-dried as I'd
like. I'll investigate what it could look like.

> .. or do the existing compiler warnings for the builtins already cover
> all cases, and the only reason the fortify-source code has
> compile-time warnings is that the option takes over the builtins?

This mostly depends on the compiler version, and they often overlap, but
the new FORTIFY logic tends to be more strict (where possible) and is more
consistent; I view the two diagnostic capabilities as complementary.

-- 
Kees Cook