linux-kernel - Re: [PATCH] x86_64: test xmm/ymm register state after execve(2)

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]

Date:   Mon, 11 Dec 2023 10:58:22 -0800
From:   Dave Hansen <dave.hansen@...el.com>
To:     Alexey Dobriyan <adobriyan@...il.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,
        linux-kernel@...r.kernel.org, "H. Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH] x86_64: test xmm/ymm register state after execve(2)

On 12/5/23 22:15, Alexey Dobriyan wrote:
>>> +".Ltest_xmm:"
>>> +	"movdqu %xmm0, xmm + 16 * 0;"
>>> +	"movdqu %xmm1, xmm + 16 * 1;"
>> Does this work on systems without XMMs?  I know it's not common these
>> days but it's possible, especially in VMs.
> No. But I think all x86_64 systems have SSE2. So it is up to whoever will
> port this test to i386.
I somehow got it in my head that we can't do XMM's unless XSAVE is in
the picture.  But that's wrong.  FXSAVE works on XMM's just fine and
it's a requirement for FPU support, and 64-bit requires the FPU.

It would be nice to spell that out explicitly in the changelog if you
get a chance to resend this.

Also, on the license... You as the copyright holder are, of course, free
to distribute it under any license you want yourself.  But for the copy
in the kernel, it would be nice if you chose something you could stick
an SPDX label on.  That way, a maintainer knows immediately if the
license is OK or not.