linux-kernel - Re: [PATCH 01/13] x86/CPU/AMD: Add ZenX generations flags

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]

Date:   Thu, 30 Nov 2023 12:17:02 -0600
From:   Tom Lendacky <thomas.lendacky@....com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     X86 ML <x86@...nel.org>, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 01/13] x86/CPU/AMD: Add ZenX generations flags

On 11/30/23 11:13, Borislav Petkov wrote:
> On Thu, Nov 30, 2023 at 11:05:14AM -0600, Tom Lendacky wrote:
>> Previously just being family 17h or 19h would get X86_FEATURE_ZEN set. With
>> this, if the model check doesn't match, you won't get any X86_FEATURE_ZEN*
>> set. Should you do set X86_FEATURE_ZEN here, e.g. lowest common denominator
>> for the family?
> 
> My assumption/expectation is that those WARNs should never happen
> because they will be caught early enough in enablement and I will get
> patches.
> 
> Besides, X86_FEATURE_ZEN means only Zen1 now.

There are references to X86_FEATURE_ZEN in arch/x86/kernel/process.c and 
drivers/acpi/resource.c that should probably be vetted.

Maybe having X86_FEATURE_ZEN mean all ZEN (and set for anything family 17h 
or higher) and a separate per generation, e.g. X86_FEATURE_ZEN1, when you 
need to be specific, would work.

Thanks,
Tom

>