All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Tian, Kevin" <kevin.tian@intel.com>
To: Paolo Bonzini <pbonzini@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	"Wang, Wei W" <wei.w.wang@intel.com>,
	"quintela@redhat.com" <quintela@redhat.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
	"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	Jing Liu <jing2.liu@linux.intel.com>,
	"Zhong, Yang" <yang.zhong@intel.com>,
	"x86@kernel.org" <x86@kernel.org>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>, "Christopherson,,
	Sean" <seanjc@google.com>,
	"Nakajima, Jun" <jun.nakajima@intel.com>,
	"Zeng, Guang" <guang.zeng@intel.com>
Subject: RE: [patch 5/6] x86/fpu: Provide fpu_update_guest_xcr0/xfd()
Date: Thu, 16 Dec 2021 01:04:01 +0000	[thread overview]
Message-ID: <BN9PR11MB5276E2165EB86520520D54FD8C779@BN9PR11MB5276.namprd11.prod.outlook.com> (raw)
In-Reply-To: <a4fbf9f8-8876-f58c-d2b6-15add35bedd0@redhat.com>

> From: Paolo Bonzini <paolo.bonzini@gmail.com> On Behalf Of Paolo Bonzini
> Sent: Wednesday, December 15, 2021 6:28 PM
> 
> On 12/15/21 11:09, Thomas Gleixner wrote:
> > Lets assume the restore order is XSTATE, XCR0, XFD:
> >
> >       XSTATE has everything in init state, which means the default
> >       buffer is good enough
> >
> >       XCR0 has everything enabled including AMX, so the buffer is
> >       expanded
> >
> >       XFD has AMX disable set, which means the buffer expansion was
> >       pointless
> >
> > If we go there, then we can just use a full expanded buffer for KVM
> > unconditionally and be done with it. That spares a lot of code.
> 
> If we decide to use a full expanded buffer as soon as KVM_SET_CPUID2 is
> done, that would work for me.  Basically KVM_SET_CPUID2 would:
> 
> - check bits from CPUID[0xD] against the prctl requested with GUEST_PERM
> 
> - return with -ENXIO or whatever if any dynamic bits were not requested
> 
> - otherwise call fpstate_realloc if there are any dynamic bits requested
> 
> Considering that in practice all Linux guests with AMX would have XFD
> passthrough (because if there's no prctl, Linux keeps AMX disabled in
> XFD), this removes the need to do all the #NM handling too.  Just make

#NM trap is for XFD_ERR thus still required.

> XFD passthrough if it can ever be set to a nonzero value.  This costs an
> RDMSR per vmexit even if neither the host nor the guest ever use AMX.

Well, we can still trap WRMSR(XFD) in the start and then disable interception
after the 1st trap.

Thanks
Kevin

  parent reply	other threads:[~2021-12-16  1:04 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-14  2:50 [patch 0/6] x86/fpu: Preparatory changes for guest AMX support Thomas Gleixner
2021-12-14  2:50 ` [patch 1/6] x86/fpu: Extend fpu_xstate_prctl() with guest permissions Thomas Gleixner
2021-12-14  5:13   ` Tian, Kevin
2021-12-14 10:37     ` Paolo Bonzini
2021-12-14  2:50 ` [patch 2/6] x86/fpu: Prepare guest FPU for dynamically enabled FPU features Thomas Gleixner
2021-12-14  2:50 ` [patch 3/6] x86/fpu: Make XFD initialization in __fpstate_reset() a function argument Thomas Gleixner
2021-12-14  2:50 ` [patch 4/6] x86/fpu: Add guest support to xfd_enable_feature() Thomas Gleixner
2021-12-14  6:05   ` Tian, Kevin
2021-12-14 10:21     ` Paolo Bonzini
2021-12-14 13:15       ` Thomas Gleixner
2021-12-15  5:46         ` Tian, Kevin
2021-12-15  9:53           ` Thomas Gleixner
2021-12-15 10:02             ` Tian, Kevin
2021-12-14  2:50 ` [patch 5/6] x86/fpu: Provide fpu_update_guest_xcr0/xfd() Thomas Gleixner
2021-12-14  6:25   ` Tian, Kevin
2021-12-14 15:09   ` Wang, Wei W
2021-12-14 15:40     ` Thomas Gleixner
2021-12-14 16:11       ` Wang, Wei W
2021-12-14 18:04         ` Thomas Gleixner
2021-12-14 19:07           ` Juan Quintela
2021-12-14 20:28             ` Thomas Gleixner
2021-12-14 21:35               ` Juan Quintela
2021-12-15  2:17                 ` Wang, Wei W
2021-12-15 10:09                   ` Thomas Gleixner
2021-12-15 10:27                     ` Paolo Bonzini
2021-12-15 10:41                       ` Paolo Bonzini
2021-12-16  1:00                         ` Tian, Kevin
2021-12-16  5:36                         ` Tian, Kevin
2021-12-16 21:07                           ` Paolo Bonzini
2021-12-16 10:21                         ` Tian, Kevin
2021-12-16 10:24                           ` Paolo Bonzini
2021-12-16 10:26                           ` Paolo Bonzini
2021-12-16 13:00                         ` Tian, Kevin
2021-12-16  1:04                       ` Tian, Kevin [this message]
2021-12-16  9:34                         ` Thomas Gleixner
2021-12-16  9:59                           ` Tian, Kevin
2021-12-16 14:12                             ` Thomas Gleixner
2021-12-17 15:33                               ` Tian, Kevin
2021-12-15  6:14   ` Tian, Kevin
2021-12-14  2:50 ` [patch 6/6] x86/fpu: Provide kvm_sync_guest_vmexit_xfd_state() Thomas Gleixner
2021-12-15  6:35   ` Liu, Jing2
2021-12-15  9:49     ` Thomas Gleixner
2021-12-14  6:50 ` [patch 0/6] x86/fpu: Preparatory changes for guest AMX support Tian, Kevin
2021-12-14  6:52 ` Liu, Jing2
2021-12-14  7:54   ` Tian, Kevin
2021-12-14 10:42 ` Paolo Bonzini
2021-12-14 13:24   ` Thomas Gleixner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=BN9PR11MB5276E2165EB86520520D54FD8C779@BN9PR11MB5276.namprd11.prod.outlook.com \
    --to=kevin.tian@intel.com \
    --cc=dgilbert@redhat.com \
    --cc=guang.zeng@intel.com \
    --cc=jing2.liu@linux.intel.com \
    --cc=jun.nakajima@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=quintela@redhat.com \
    --cc=seanjc@google.com \
    --cc=tglx@linutronix.de \
    --cc=wei.w.wang@intel.com \
    --cc=x86@kernel.org \
    --cc=yang.zhong@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.