linux-kernel - Re: [PATCH v2 2/6] riscv: Deduplicate IRQ stack switching

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 24 Aug 2023 12:03:00 -0700
From:   Sami Tolvanen <samitolvanen@...gle.com>
To:     Clément Léger <cleger@...osinc.com>
Cc:     Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Kees Cook <keescook@...omium.org>, Guo Ren <guoren@...nel.org>,
        Deepak Gupta <debug@...osinc.com>,
        Nathan Chancellor <nathan@...nel.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Fangrui Song <maskray@...gle.com>,
        linux-riscv@...ts.infradead.org, llvm@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 2/6] riscv: Deduplicate IRQ stack switching

Hi Clément,

On Thu, Aug 24, 2023 at 1:12 AM Clément Léger <cleger@...osinc.com> wrote:
>
> Some defines for stack frame offsets could be added in asm-offsets for
> the offsets:
>
>         DEFINE(STACKFRAME_SIZE_ON_STACK, ALIGN(sizeof(struct stackframe),
> STACK_ALIGN));
>         OFFSET(STACKFRAME_FP, stackframe, fp);
>         OFFSET(STACKFRAME_RA, stackframe, ra);
>
> And you can probably increment the stack at once (saving two addi in
> prologue/epilogue) for both RA and FP and reuse the asm-offsets defines:
>
>         addi    sp, sp, -STACKFRAME_SIZE_ON_STACK
>         REG_S   ra, STACKFRAME_RA(sp)
>         REG_S   s0, STACKFRAME_FP(sp)
>         addi    s0, sp, STACKFRAME_SIZE_ON_STACK

Thanks for taking a look!

I just copied the existing inline assembly here, but I do agree that
defining stack frame offsets and avoiding the extra addis is a cleaner
approach. I'll change this in v3.

Sami