linux-kernel - Re: [PATCH v4] x86, mem: move memmove to out of line assembler

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: <CAKwvOd=QsUGZ2ZF_3xHb0dG0JF=f9e3_vHvU+rsj0eXfAQP86g@mail.gmail.com>
Date:   Fri, 30 Sep 2022 09:43:27 -0700
From:   Nick Desaulniers <ndesaulniers@...gle.com>
To:     David Laight <David.Laight@...lab.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        "x86@...nel.org" <x86@...nel.org>,
        "H . Peter Anvin" <hpa@...or.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Kees Cook <keescook@...omium.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        "llvm@...ts.linux.dev" <llvm@...ts.linux.dev>,
        Andy Lutomirski <luto@...nel.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>
Subject: Re: [PATCH v4] x86, mem: move memmove to out of line assembler

On Fri, Sep 30, 2022 at 2:55 AM David Laight <David.Laight@...lab.com> wrote:
>
> Oh - and why do all the labels have 'byteswap' in them?

.Lbyteswap is swapping single bytes at a time.
.L4_byteswap is swapping 4 bytes at a time.
.L8_byteswap is swapping 8 bytes at a time.
.L16_byteswap, .L16_byteswap_backwards_loop, and
.L16_byteswap_forwards_loop are swapping 16 bytes at a time.

When doing a move that has to be delicate with respect to overlap, I
thought it was helpful to know how many bytes are being swapped in a
given run of instructions. The names were chosen were based on the
comparison/guards/conditional jmps.

---

Linus pointed out that this 32b memmove looks similar to 64b
memcpy_orig.  Would you prefer me use those labels?

s/.Lbyteswap/.Lstore_1byte/
s/.L4_byteswap/.Lless_3bytes/
s/.L8_byteswap/.Lless_8bytes/
s/.L16_byteswap/.Lless_16bytes/
s/.L16_byteswap_backwards_loop/.Lcopy_backward_loop/
s/.L16_byteswap_forwards_loop/.Lcopy_forward_loop/

or something else?
-- 
Thanks,
~Nick Desaulniers