linux-kernel - Re: [PATCH] x86/fpu: Fix the os panic issue caused by the XGETBV instruction

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: <1c2e3700-5663-4e6e-9da9-500b64e778a2@zytor.com>
Date: Fri, 17 Jan 2025 19:13:42 -0800
From: "H. Peter Anvin" <hpa@...or.com>
To: Sean Christopherson <seanjc@...gle.com>,
        "Chang S. Bae" <chang.seok.bae@...el.com>
Cc: Tony W Wang-oc <TonyWWang-oc@...oxin.com>, tglx@...utronix.de,
        mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com,
        x86@...nel.org, aruna.ramakrishna@...cle.com, pbonzini@...hat.com,
        levymitchell0@...il.com, attofari@...zon.de,
        linux-kernel@...r.kernel.org, CobeChen@...oxin.com, TimGuo@...oxin.com,
        LeoLiu-oc@...oxin.com, Lyle Li <LyleLi@...oxin.com>
Subject: Re: [PATCH] x86/fpu: Fix the os panic issue caused by the XGETBV
 instruction

On 1/17/25 14:10, Sean Christopherson wrote:
>>
>> Note that XFD is already listed as dependent on XGETBV1.
>>
>> But I doubt the kernel needs to be resilient to deliberately misconfigured
>> or crazy virtual machine setups.
> 
> I don't see anything in the SDM that suggests this is a misconfigured CPU.  Intel
> might not have plans to ship such CPUs, but AFAICT it's not a violation of the
> architecture as defined in the SDM.
> 
> The SDM even explicitly says that protection keys can exist and be used without
> PKU state being supported in XSAVE at all, at which point assuming the existence
> of XGETBV1 is rather nonsensical.
> 

Whether or not a combination is *possible* at all is totally separate 
from whether or not it is worth for Linux to support it. The CPUID 
dependency list exists for exactly that reason -- it defines *Linux 
policy* with regards to feature dependencies.

	-hpa