linux-kernel - Re: [PATCH 2/2] tools/nolibc: x86-64: Fix startup code bug

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]

Date:   Mon, 18 Oct 2021 13:53:29 +0700
From:   Ammar Faizi <ammar.faizi@...dents.amikom.ac.id>
To:     Willy Tarreau <w@....eu>
Cc:     Ammar Faizi <ammar.faizi@...dents.amikom.ac.id>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, x86@...nel.org,
        "H. Peter Anvin" <hpa@...or.com>,
        David Laight <David.Laight@...LAB.COM>,
        Peter Cordes <peter@...des.ca>,
        Bedirhan KURT <windowz414@...weeb.org>,
        Louvian Lyndal <louvianlyndal@...il.com>
Subject: Re: [PATCH 2/2] tools/nolibc: x86-64: Fix startup code bug

On Mon, Oct 18, 2021 at 11:59 AM Willy Tarreau <w@....eu> wrote:
> > Extra fixes:
> >   - Use NR_exit_group instead of NR_exit.
>
> Please, this is independent on the fix above so it must be subject
> of a different patch with its own justification. Also it should cover
> all supported architectures, otherwise programs will start to behave
> differently on different targets.
>
> >   - Use `mov %eax,%edi` instead of `movzbq %al,%rdi`. This makes the
> >     exit code more observable from strace. While the exit code is
> >     only 8-bit, the kernel has taken care of that, so no need to
> >     worry about it.
>
> I'm fine with this one as well, but similarly, it should be in its own
> patch and applied to all architectures.
>
[...]
>
> I find the whole thing much less readable here, as asm code tends to
> be read as visual groups of patterns. I'm suggesting that you place a
> multi-line comment before the asm statement indicating the general
> rules (e.g. lowest stack frame must be zero, rsp+8 must be multiple of
> 16 etc), then only comment each instruction on the same line as it was
> before.

Got it, agree with that. I will address your review and resend this as a
patchset v2 soon.

-- 
Ammar Faizi