linux-kernel - Re: [PATCH] selftests: riscv: fix v_exec_initval

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: <Z8lhoTd3BwPWQ0f_@ghost>
Date: Thu, 6 Mar 2025 00:49:37 -0800
From: Charlie Jenkins <charlie@...osinc.com>
To: Ignacio Encinas Rubio <ignacio@...cinas.com>
Cc: linux-kernel-mentees@...ts.linux.dev, skhan@...uxfoundation.org,
	Paul Walmsley <paul.walmsley@...ive.com>,
	Palmer Dabbelt <palmer@...belt.com>,
	Alexandre Ghiti <alex@...ti.fr>, linux-kselftest@...r.kernel.org,
	linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] selftests: riscv: fix v_exec_initval_nolibc.c

On Thu, Mar 06, 2025 at 07:31:22AM +0100, Ignacio Encinas Rubio wrote:
> 
> 
> On 5/3/25 22:49, Charlie Jenkins wrote:
> > On Wed, Mar 05, 2025 at 05:39:28PM +0100, Ignacio Encinas wrote:
> >> Vector registers are zero initialized by the kernel. Stop accepting
> >> "all ones" as a clean value.
> >>
> >> Note that this was not working as expected given that
> >> 	value == 0xff
> >> can be assumed to be always false by the compiler as value's range is
> >> [-128, 127]. Both GCC (-Wtype-limits) and clang
> >> (-Wtautological-constant-out-of-range-compare) warn about this.
> > 
> > This check was included because the "dirty" value is an implementation
> > detail that I believe is not strongly defined in the ABI. Since linux
> > does always set this value to zero (currently) we can safely remove this
> > check. 
> 
> Thanks for the review. Just after sending the patch I noticed it should
> also remove some code that becomes useless after this change: 
> _prev_value_ and _first_ variables were only needed because two "clean" 
> values were supported.
> 
> I'll send a v2 tomorrow. I'm guessing keeping your "Reviewed-by" and
> "Tested-by" is the appropriate thing to do as the changes are very
> simple. Let me know if that's not the case.
> 
> Thanks again!

Yes, those changes seem small so you can keep the tags :)

- Charlie