linux-kernel - Re: [PATCH] x86: fix user address masking non-canonical speculation issue

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: <CAHk-=wh2FoBH63Sap4=DY+QNwZ5WULYWKojVn8YzKmYv0xpDpQ@mail.gmail.com>
Date: Fri, 25 Oct 2024 10:41:41 -0700
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Maciej Wieczor-Retman <maciej.wieczor-retman@...el.com>
Cc: Linux Kernel Mailing List <linux-kernel@...r.kernel.org>, x86@...nel.org, 
	Andrew Cooper <andrew.cooper3@...rix.com>, Josh Poimboeuf <jpoimboe@...nel.org>, 
	Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH] x86: fix user address masking non-canonical speculation issue

On Fri, 25 Oct 2024 at 07:16, Maciej Wieczor-Retman
<maciej.wieczor-retman@...el.com> wrote:
>
> I tested the patch on a Sierra Forest Xeon that is LAM capable and I didn't find
> any issues.
>
> To test the LAM related portion I uncommented the above if statement
> and enabled LAM_SUP during the boot process as well as enabled LAM for the user
> process with a syscall.

Thanks. I've applied the patch (with the LAM case still commented
out), and let's see if something comes up.

I guess that backporting it might be slightly painful due to the
runtime-const use, but I'm not even convinced it's needed for stable.

I'd expect that even on Zen 5 it's not actually a _practical_
information leak due to the whole GP exception thing going to mess
with any leaking signal.

          Linus