From: Sean Christopherson <seanjc@google.com>
To: Tom Lendacky <thomas.lendacky@amd.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
x86@kernel.org, Paolo Bonzini <pbonzini@redhat.com>,
Borislav Petkov <bp@alien8.de>,
Dave Hansen <dave.hansen@linux.intel.com>,
Ingo Molnar <mingo@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Michael Roth <michael.roth@amd.com>
Subject: Re: [PATCH] KVM: SVM: Fix SNP AP destroy race with VMRUN
Date: Mon, 17 Mar 2025 10:28:34 -0700 [thread overview]
Message-ID: <Z9hbwkqwDKlyPsqv@google.com> (raw)
In-Reply-To: <48899db8-c506-b4d1-06cd-6ba9041437f7@amd.com>
On Mon, Mar 17, 2025, Tom Lendacky wrote:
> On 3/17/25 12:20, Tom Lendacky wrote:
> > An AP destroy request for a target vCPU is typically followed by an
> > RMPADJUST to remove the VMSA attribute from the page currently being
> > used as the VMSA for the target vCPU. This can result in a vCPU that
> > is about to VMRUN to exit with #VMEXIT_INVALID.
> >
> > This usually does not happen as APs are typically sitting in HLT when
> > being destroyed and therefore the vCPU thread is not running at the time.
> > However, if HLT is allowed inside the VM, then the vCPU could be about to
> > VMRUN when the VMSA attribute is removed from the VMSA page, resulting in
> > a #VMEXIT_INVALID when the vCPU actually issues the VMRUN and causing the
> > guest to crash. An RMPADJUST against an in-use (already running) VMSA
> > results in a #NPF for the vCPU issuing the RMPADJUST, so the VMSA
> > attribute cannot be changed until the VMRUN for target vCPU exits. The
> > Qemu command line option '-overcommit cpu-pm=on' is an example of allowing
> > HLT inside the guest.
> >
> > Use kvm_test_request() to ensure that the target vCPU sees the AP destroy
> > request before returning to the initiating vCPU.
> >
> > Fixes: e366f92ea99e ("KVM: SEV: Support SEV-SNP AP Creation NAE event")
> > Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
>
> Sean,
>
> If you're ok with this approach for the fix, this patch may need to be
> adjusted given your series around AP creation fixes, unless you want to
> put this as an early patch in your series. Let me know what you'd like
> to do.
This is unsafe as it requires userspace to do KVM_RUN _and_ for the vCPU to get
far enough along to consume the request.
Very off-the-cuff, but I assume KVM_REQ_UPDATE_PROTECTED_GUEST_STATE just needs
to be annotated with KVM_REQUEST_WAIT.
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 04e6c5604bc3..67abfe97c600 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -124,7 +124,8 @@
KVM_ARCH_REQ_FLAGS(31, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
#define KVM_REQ_HV_TLB_FLUSH \
KVM_ARCH_REQ_FLAGS(32, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
-#define KVM_REQ_UPDATE_PROTECTED_GUEST_STATE KVM_ARCH_REQ(34)
+#define KVM_REQ_UPDATE_PROTECTED_GUEST_STATE \
+ KVM_ARCH_REQ_FLAGS(34, KVM_REQUEST_WAIT)
#define CR0_RESERVED_BITS \
(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
next prev parent reply other threads:[~2025-03-17 17:28 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-17 17:20 [PATCH] KVM: SVM: Fix SNP AP destroy race with VMRUN Tom Lendacky
2025-03-17 17:23 ` Tom Lendacky
2025-03-17 17:28 ` Sean Christopherson [this message]
2025-03-17 17:36 ` Tom Lendacky
2025-03-18 12:43 ` Tom Lendacky
2025-03-18 13:47 ` Tom Lendacky
2025-03-21 16:52 ` Tom Lendacky
2025-03-21 23:17 ` Sean Christopherson
2025-03-25 17:49 ` Tom Lendacky
2025-03-26 15:34 ` Tom Lendacky
2025-03-26 17:17 ` Sean Christopherson
-- strict thread matches above, loose matches on Subject: below --
2025-03-21 16:20 Tom Lendacky
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=Z9hbwkqwDKlyPsqv@google.com \
--to=seanjc@google.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=michael.roth@amd.com \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=tglx@linutronix.de \
--cc=thomas.lendacky@amd.com \
--cc=x86@kernel.org \
/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.