linux-kernel - Re: Rewording of Setting IA32_XFD[18] (Re: Thoughts of AMX KVM support based on latest kernel)

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: <87bl1oxh26.ffs@tglx>
Date:   Fri, 10 Dec 2021 22:53:37 +0100
From:   Thomas Gleixner <tglx@...utronix.de>
To:     "Nakajima, Jun" <jun.nakajima@...el.com>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        "Tian, Kevin" <kevin.tian@...el.com>,
        "Liu, Jing2" <jing2.liu@...el.com>,
        LKML <linux-kernel@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>,
        "kvm@...r.kernel.org" <kvm@...r.kernel.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Arjan van de Ven <arjan@...ux.intel.com>,
        Jing Liu <jing2.liu@...ux.intel.com>,
        "seanjc@...gle.com" <seanjc@...gle.com>,
        "Cooper, Andrew" <andrew.cooper3@...rix.com>,
        "Bae, Chang Seok" <chang.seok.bae@...el.com>
Subject: Re: Rewording of Setting IA32_XFD[18] (Re: Thoughts of AMX KVM
 support based on latest kernel)

Jun,

On Wed, Dec 08 2021 at 00:50, Jun Nakajima wrote:
>> On Nov 19, 2021, at 7:41 AM, Nakajima, Jun <jun.nakajima@...el.com> wrote:
>> I’ll work with the H/W team on those (i.e. rewording and the component 17 issue).
>> 
>
> The following is a possible documentation update that may convey the
> rewording you requested.  Does this (the last sentence, “In addition,
> “) work for you?
>
> 3.3 RECOMMENDATIONS FOR SYSTEM SOFTWARE
>
> System software may disable use of Intel AMX by clearing XCR0[18:17],
> by clearing CR4.OSXSAVE, or by setting IA32_XFD[18]. System software
> should initialize AMX state (e.g., by executing TILERELEASE) when
> doing so because maintaining AMX state in a non-initialized state may
> have negative power and performance implications. In addition,
> software should not rely on the state of the tile data after setting
> IA32_XFD[18]; software should always reload or reinitialize the tile
> data after clearing IA32_XFD[18].

looks good to me.

Thanks,

        tglx