kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: "Linux regression tracking (Thorsten Leemhuis)"
	<regressions@leemhuis.info>,
	Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, rcu@vger.kernel.org,
	linux-kernel@vger.kernel.org, Kevin Tian <kevin.tian@intel.com>,
	Yan Zhao <yan.y.zhao@intel.com>, Yiwei Zhang <zzyiwei@google.com>,
	Lai Jiangshan <jiangshanlai@gmail.com>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	Josh Triplett <josh@joshtriplett.org>,
	Gerd Hoffmann <kraxel@redhat.com>,
	Linux kernel regressions list <regressions@lists.linux.dev>
Subject: Re: [PATCH 5/5] KVM: VMX: Always honor guest PAT on CPUs that support self-snoop
Date: Mon, 07 Oct 2024 15:38:15 +0200	[thread overview]
Message-ID: <87iku4ghiw.fsf@redhat.com> (raw)
In-Reply-To: <df3c6560-dd37-4ec2-9b7e-1ad4c3ceba07@leemhuis.info>

"Linux regression tracking (Thorsten Leemhuis)"
<regressions@leemhuis.info> writes:

> On 30.08.24 11:35, Vitaly Kuznetsov wrote:
>> Sean Christopherson <seanjc@google.com> writes:
>> 
>>> Unconditionally honor guest PAT on CPUs that support self-snoop, as
>>> Intel has confirmed that CPUs that support self-snoop always snoop caches
>>> and store buffers.  I.e. CPUs with self-snoop maintain cache coherency
>>> even in the presence of aliased memtypes, thus there is no need to trust
>>> the guest behaves and only honor PAT as a last resort, as KVM does today.
>>>
>>> Honoring guest PAT is desirable for use cases where the guest has access
>>> to non-coherent DMA _without_ bouncing through VFIO, e.g. when a virtual
>>> (mediated, for all intents and purposes) GPU is exposed to the guest, along
>>> with buffers that are consumed directly by the physical GPU, i.e. which
>>> can't be proxied by the host to ensure writes from the guest are performed
>>> with the correct memory type for the GPU.
>> 
>> Necroposting!
>> 
>> Turns out that this change broke "bochs-display" driver in QEMU even
>> when the guest is modern (don't ask me 'who the hell uses bochs for
>> modern guests', it was basically a configuration error :-). E.g:
>> [...]
>
> This regression made it to the list of tracked regressions. It seems
> this thread stalled a while ago. Was this ever fixed? Does not look like
> it, but I might have missed something. Or is this a regression I should
> just ignore for one reason or another?
>

The regression was addressed in by reverting 377b2f359d1f in 6.11

commit 9d70f3fec14421e793ffbc0ec2f739b24e534900
Author: Paolo Bonzini <pbonzini@redhat.com>
Date:   Sun Sep 15 02:49:33 2024 -0400

    Revert "KVM: VMX: Always honor guest PAT on CPUs that support self-snoop"

Also, there's a (pending) DRM patch fixing it from the guest's side:
https://gitlab.freedesktop.org/drm/misc/kernel/-/commit/9388ccf69925223223c87355a417ba39b13a5e8e

-- 
Vitaly


  reply	other threads:[~2024-10-07 13:38 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-09  1:09 [PATCH 0/5] KVM: VMX: Drop MTRR virtualization, honor guest PAT Sean Christopherson
2024-03-09  1:09 ` [PATCH 1/5] KVM: x86: Remove VMX support for virtualizing guest MTRR memtypes Sean Christopherson
2024-03-11  7:44   ` Yan Zhao
2024-03-12  0:08     ` Sean Christopherson
2024-03-12  1:10   ` Dongli Zhang
2024-03-12 17:08     ` Sean Christopherson
2024-03-14 10:31       ` Dongli Zhang
2024-03-14 14:47         ` Sean Christopherson
2024-03-09  1:09 ` [PATCH 2/5] KVM: VMX: Drop support for forcing UC memory when guest CR0.CD=1 Sean Christopherson
2024-03-09  1:09 ` [PATCH 3/5] srcu: Add an API for a memory barrier after SRCU read lock Sean Christopherson
2024-03-09  1:09 ` [PATCH 4/5] KVM: x86: Ensure a full memory barrier is emitted in the VM-Exit path Sean Christopherson
2024-06-20 22:38   ` Paolo Bonzini
2024-06-20 23:42     ` Paul E. McKenney
2024-06-21  0:52     ` Yan Zhao
2024-03-09  1:09 ` [PATCH 5/5] KVM: VMX: Always honor guest PAT on CPUs that support self-snoop Sean Christopherson
2024-03-11  1:16   ` Yan Zhao
2024-03-12  0:25     ` Sean Christopherson
2024-03-12  7:30       ` Tian, Kevin
2024-03-12 16:07         ` Sean Christopherson
2024-03-13  1:18           ` Yan Zhao
2024-03-13  8:52             ` Tian, Kevin
2024-03-13  8:55               ` Yan Zhao
2024-03-13 15:09                 ` Sean Christopherson
2024-03-14  0:12                   ` Yan Zhao
2024-03-14  1:00                     ` Sean Christopherson
2024-03-25  3:43   ` Chao Gao
2024-04-01 22:29     ` Sean Christopherson
2024-08-30  9:35   ` Vitaly Kuznetsov
2024-08-30 11:05     ` Gerd Hoffmann
2024-08-30 13:47       ` Vitaly Kuznetsov
2024-08-30 13:52         ` Sean Christopherson
2024-08-30 14:06           ` Vitaly Kuznetsov
2024-08-30 14:37             ` Vitaly Kuznetsov
2024-08-30 16:13               ` Sean Christopherson
2024-09-02  8:23                 ` Gerd Hoffmann
2024-09-02  1:44         ` Yan Zhao
2024-09-02  9:49           ` Vitaly Kuznetsov
2024-09-03  0:25             ` Yan Zhao
2024-09-03 15:30             ` Sean Christopherson
2024-09-03 16:20               ` Vitaly Kuznetsov
2024-09-04  2:28                 ` Yan Zhao
2024-09-04 12:17                   ` Yan Zhao
2024-09-05  0:41                     ` Sean Christopherson
2024-09-05  9:43                       ` Yan Zhao
2024-09-09  5:30                         ` Yan Zhao
2024-09-09 13:24                           ` Paolo Bonzini
2024-09-09 16:04                             ` Sean Christopherson
2024-09-10  1:05                             ` Yan Zhao
2024-09-04 11:47                 ` Vitaly Kuznetsov
2024-10-07 13:28     ` Linux regression tracking (Thorsten Leemhuis)
2024-10-07 13:38       ` Vitaly Kuznetsov [this message]
2024-10-07 14:04         ` Linux regression tracking (Thorsten Leemhuis)
2024-03-22  9:29 ` [PATCH 0/5] KVM: VMX: Drop MTRR virtualization, honor guest PAT Ma, Yongwei
2024-03-22 13:08 ` Yan Zhao
2024-03-25  6:56   ` Ma, XiangfeiX
2024-03-25  8:02     ` Ma, XiangfeiX
2024-06-05 23:20 ` Sean Christopherson
2024-06-06  0:03   ` Paul E. McKenney
  -- strict thread matches above, loose matches on Subject: below --
2025-04-10  1:13 [PATCH 5/5] KVM: VMX: Always honor guest PAT on CPUs that support self-snoop Myrsky Lintu
2025-04-10  5:12 ` Yan Zhao
2025-04-10 10:05   ` Myrsky Lintu

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=87iku4ghiw.fsf@redhat.com \
    --to=vkuznets@redhat.com \
    --cc=jiangshanlai@gmail.com \
    --cc=josh@joshtriplett.org \
    --cc=kevin.tian@intel.com \
    --cc=kraxel@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulmck@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=rcu@vger.kernel.org \
    --cc=regressions@leemhuis.info \
    --cc=regressions@lists.linux.dev \
    --cc=seanjc@google.com \
    --cc=yan.y.zhao@intel.com \
    --cc=zzyiwei@google.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).