[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87tugb3r7w.ffs@tglx>
Date: Wed, 17 Nov 2021 13:24:19 +0100
From: Thomas Gleixner <tglx@...utronix.de>
To: "Tian, Kevin" <kevin.tian@...el.com>,
Paolo Bonzini <pbonzini@...hat.com>,
"Nakajima, Jun" <jun.nakajima@...el.com>
Cc: "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: Thoughts of AMX KVM support based on latest kernel
Tian,
On Wed, Nov 17 2021 at 10:15, Tian, Kevin wrote:
> We are not sure whether such trick is worthwhile, since a sane
> guest shouldn't set XFD[AMX]=1 before storing the AMX state. This
> is why we want to seek SDM change to mark out that the software
> should not assume XTILEDATA is still valid when XFD[AMX]=1.
Yes, please. Anything else is just causing too much hackery for no
value.
Thanks,
tglx
Powered by blists - more mailing lists