From: Paolo Bonzini <pbonzini@redhat.com>
To: Maxim Levitsky <mlevitsk@redhat.com>, kvm@vger.kernel.org
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>,
Borislav Petkov <bp@alien8.de>, Bandan Das <bsd@redhat.com>,
open list <linux-kernel@vger.kernel.org>,
Joerg Roedel <joro@8bytes.org>, Ingo Molnar <mingo@redhat.com>,
Wei Huang <wei.huang2@amd.com>,
Sean Christopherson <seanjc@google.com>,
"open list:KERNEL SELFTEST FRAMEWORK"
<linux-kselftest@vger.kernel.org>,
"H. Peter Anvin" <hpa@zytor.com>,
Jim Mattson <jmattson@google.com>,
Thomas Gleixner <tglx@linutronix.de>,
"maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)"
<x86@kernel.org>, Shuah Khan <shuah@kernel.org>,
Wanpeng Li <wanpengli@tencent.com>
Subject: Re: [PATCH 06/14] KVM: x86: SVM: don't set VMLOAD/VMSAVE intercepts on vCPU reset
Date: Thu, 23 Sep 2021 18:40:51 +0200 [thread overview]
Message-ID: <7289d070-152b-aece-1302-33e5a461657c@redhat.com> (raw)
In-Reply-To: <20210914154825.104886-7-mlevitsk@redhat.com>
On 14/09/21 17:48, Maxim Levitsky wrote:
> commit adc2a23734ac ("KVM: nSVM: improve SYSENTER emulation on AMD"),
> made init_vmcb set vmload/vmsave intercepts unconditionally,
> and relied on svm_vcpu_after_set_cpuid to clear them when possible.
>
> However init_vmcb is also called when the vCPU is reset, and it is
> not followed by another call to svm_vcpu_after_set_cpuid because
> the CPUID is already set.
>
> This mistake makes the VMSAVE/VMLOAD intercept to be set when
> it is not needed, and harms performance of the nested guest.
>
> Fixes: adc2a23734ac ("KVM: nSVM: improve SYSENTER emulation on AMD")
>
> Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
> ---
> arch/x86/kvm/svm/svm.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
> index 6645542df9bd..861ac9f74331 100644
> --- a/arch/x86/kvm/svm/svm.c
> +++ b/arch/x86/kvm/svm/svm.c
> @@ -1199,8 +1199,6 @@ static void init_vmcb(struct kvm_vcpu *vcpu)
> svm_set_intercept(svm, INTERCEPT_SHUTDOWN);
> svm_set_intercept(svm, INTERCEPT_VMRUN);
> svm_set_intercept(svm, INTERCEPT_VMMCALL);
> - svm_set_intercept(svm, INTERCEPT_VMLOAD);
> - svm_set_intercept(svm, INTERCEPT_VMSAVE);
> svm_set_intercept(svm, INTERCEPT_STGI);
> svm_set_intercept(svm, INTERCEPT_CLGI);
> svm_set_intercept(svm, INTERCEPT_SKINIT);
> @@ -1377,6 +1375,10 @@ static int svm_create_vcpu(struct kvm_vcpu *vcpu)
> svm->guest_state_loaded = false;
>
> svm_switch_vmcb(svm, &svm->vmcb01);
> +
> + svm_set_intercept(svm, INTERCEPT_VMLOAD);
> + svm_set_intercept(svm, INTERCEPT_VMSAVE);
> +
> init_vmcb(vcpu);
>
> svm_vcpu_init_msrpm(vcpu, svm->msrpm);
>
This needs to be redone after the latest refactoring of svm_vcpu_reset.
I'll send a patch myself.
Paolo
next prev parent reply other threads:[~2021-09-23 16:41 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-14 15:48 [PATCH 00/14] nSVM fixes and optional features Maxim Levitsky
2021-09-14 15:48 ` [PATCH 01/14] KVM: x86: nSVM: restore int_vector in svm_clear_vintr Maxim Levitsky
2021-09-23 14:04 ` Paolo Bonzini
2021-09-14 15:48 ` [PATCH 02/14] KVM: x86: selftests: test simultaneous uses of V_IRQ from L1 and L0 Maxim Levitsky
2021-09-23 14:05 ` Paolo Bonzini
2021-09-14 15:48 ` [PATCH 03/14] KVM: x86: nSVM: test eax for 4K alignment for GP errata workaround Maxim Levitsky
2021-09-23 14:05 ` Paolo Bonzini
2021-09-14 15:48 ` [PATCH 04/14] KVM: x86: nSVM: don't copy pause related settings Maxim Levitsky
2021-09-14 15:48 ` [PATCH 05/14] KVM: x86: nSVM: don't copy virt_ext from vmcb12 Maxim Levitsky
2021-09-23 14:06 ` Paolo Bonzini
2021-09-14 15:48 ` [PATCH 06/14] KVM: x86: SVM: don't set VMLOAD/VMSAVE intercepts on vCPU reset Maxim Levitsky
2021-09-23 16:40 ` Paolo Bonzini [this message]
2021-09-14 15:48 ` [PATCH 07/14] KVM: x86: SVM: add warning for CVE-2021-3656 Maxim Levitsky
2021-09-23 16:44 ` Paolo Bonzini
2021-10-12 0:21 ` Sean Christopherson
2021-10-11 17:30 ` Xiaoyao Li
2021-10-12 7:51 ` Maxim Levitsky
2021-09-14 15:48 ` [PATCH 08/14] KVM: x86: SVM: add module param to control LBR virtualization Maxim Levitsky
2021-09-14 15:48 ` [PATCH 09/14] KVM: x86: nSVM: correctly virtualize LBR msrs when L2 is running Maxim Levitsky
2021-09-14 15:48 ` [PATCH 10/14] KVM: x86: nSVM: implement nested LBR virtualization Maxim Levitsky
2021-09-14 15:48 ` [PATCH 11/14] KVM: x86: nSVM: implement nested VMLOAD/VMSAVE Maxim Levitsky
2021-09-14 15:48 ` [PATCH 12/14] KVM: x86: SVM: add module param to control TSC scaling Maxim Levitsky
2021-09-14 15:48 ` [PATCH 13/14] KVM: x86: nSVM: implement nested " Maxim Levitsky
2021-09-14 15:48 ` [PATCH 14/14] KVM: x86: nSVM: support PAUSE filter threshold and count Maxim Levitsky
2021-09-23 16:54 ` [PATCH 00/14] nSVM fixes and optional features Paolo Bonzini
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=7289d070-152b-aece-1302-33e5a461657c@redhat.com \
--to=pbonzini@redhat.com \
--cc=bp@alien8.de \
--cc=bsd@redhat.com \
--cc=hpa@zytor.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=mlevitsk@redhat.com \
--cc=seanjc@google.com \
--cc=shuah@kernel.org \
--cc=tglx@linutronix.de \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.com \
--cc=wei.huang2@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 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).