linux-kernel - Re: [PATCH 0/4] -ffreestanding/-fno-builtin-* patches

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: <20200818222542.GA3254379@rani.riverdale.lan>
Date:   Tue, 18 Aug 2020 18:25:42 -0400
From:   Arvind Sankar <nivedita@...m.mit.edu>
To:     Nick Desaulniers <ndesaulniers@...gle.com>
Cc:     Masahiro Yamada <masahiroy@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        Michal Marek <michal.lkml@...kovi.net>,
        linux-kbuild@...r.kernel.org, linux-kernel@...r.kernel.org,
        Kees Cook <keescook@...omium.org>,
        Tony Luck <tony.luck@...el.com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Michael Ellerman <mpe@...erman.id.au>,
        Joe Perches <joe@...ches.com>,
        Joel Fernandes <joel@...lfernandes.org>,
        Daniel Axtens <dja@...ens.net>,
        Arvind Sankar <nivedita@...m.mit.edu>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Alexandru Ardelean <alexandru.ardelean@...log.com>,
        Yury Norov <yury.norov@...il.com>, x86@...nel.org,
        "H . Peter Anvin" <hpa@...or.com>,
        Ard Biesheuvel <ardb@...nel.org>,
        "Paul E . McKenney" <paulmck@...nel.org>,
        Daniel Kiper <daniel.kiper@...cle.com>,
        Bruce Ashfield <bruce.ashfield@...il.com>,
        Marco Elver <elver@...gle.com>,
        Vamshi K Sthambamkadi <vamshi.k.sthambamkadi@...il.com>,
        Andi Kleen <ak@...e.de>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Dávid Bolvanský <david.bolvansky@...il.com>,
        Eli Friedman <efriedma@...cinc.com>
Subject: Re: [PATCH 0/4] -ffreestanding/-fno-builtin-* patches

On Mon, Aug 17, 2020 at 03:02:08PM -0700, Nick Desaulniers wrote:
> -ffreestanding typically inhibits "libcall optimizations" where calls to
> certain library functions can be replaced by the compiler in certain
> cases to calls to other library functions that may be more efficient.
> This can be problematic for embedded targets that don't provide full
> libc implementations.
> 
> -ffreestanding inhibits all such optimizations, which is the safe
> choice, but generally we want the optimizations that are performed. The
> Linux kernel does implement a fair amount of libc routines. Instead of
> -ffreestanding (which makes more sense in smaller images like kexec's
> purgatory image), prefer -fno-builtin-* flags to disable the compiler
> from emitting calls to functions which may not be defined.
> 
> If you see a linkage failure due to a missing symbol that's typically
> defined in a libc, and not explicitly called from the source code, then
> the compiler may have done such a transform.  You can either implement
> such a function (ie. in lib/string.c) or disable the transform outright
> via -fno-builtin-* flag (where * is the name of the library routine, ie.
> -fno-builtin-bcmp).
> 
> Patch 1 unbreaks the build with ToT clang, which has been red all
> weekend, by adding -fno-builtin-stpcpy.
> Patch 2 is a revert but adds -fno-builtin-bcmp.
> Patch 3 does the same for x86 purgatory.
> Patch 4 removes -ffreestanding from i386.
> 
> The first patch makes sense for Kbuild, the second maybe akpm@, the
> third and forth for x86.  Not sure who should pick up the series (they
> can be merged out of order, technically) but I really need the first
> patch soon. The 3 latter patches are cleanups.
> 
> Nick Desaulniers (4):
>   Makefile: add -fno-builtin-stpcpy
>   Revert "lib/string.c: implement a basic bcmp"
>   x86/boot: use -fno-builtin-bcmp
>   x86: don't build CONFIG_X86_32 as -ffreestanding
> 
>  Makefile               |  7 +++++++
>  arch/x86/Makefile      |  3 ---
>  arch/x86/boot/Makefile |  1 +
>  arch/x86/boot/string.c |  8 --------
>  include/linux/string.h |  3 ---
>  lib/string.c           | 20 --------------------
>  6 files changed, 8 insertions(+), 34 deletions(-)
> 
> -- 
> 2.28.0.220.ged08abb693-goog
> 

Another thing that needs to be fixed is that at least lib/string.c needs
to be compiled with -ffreestanding.

gcc-10 optimizes the generic memset implementation in there into a call
to memset. Now that's on x86 which doesn't use the generic
implementation, but this is just waiting to bite us.

https://godbolt.org/z/6EhG15