From: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
Borislav Petkov <bp@alien8.de>, Xin Li <xin@zytor.com>,
Chao Gao <chao.gao@intel.com>,
Dapeng Mi <dapeng1.mi@linux.intel.com>
Subject: Re: [PATCH 08/28] KVM: nSVM: Use dedicated array of MSRPM offsets to merge L0 and L1 bitmaps
Date: Wed, 4 Jun 2025 08:49:21 -0700 [thread overview]
Message-ID: <aEBrAbhRFkou4Mvj@google.com> (raw)
In-Reply-To: <688ba3c4-bf8a-47f1-ad14-85a23399582c@redhat.com>
On Wed, Jun 04, 2025, Paolo Bonzini wrote:
> On 5/30/25 01:39, Sean Christopherson wrote:
> > Use a dedicated array of MSRPM offsets to merge L0 and L1 bitmaps, i.e. to
> > merge KVM's vmcb01 bitmap with L1's vmcb12 bitmap. This will eventually
> > allow for the removal of direct_access_msrs, as the only path where
> > tracking the offsets is truly justified is the merge for nested SVM, where
> > merging in chunks is an easy way to batch uaccess reads/writes.
> >
> > Opportunistically omit the x2APIC MSRs from the merge-specific array
> > instead of filtering them out at runtime.
> >
> > Note, disabling interception of XSS, EFER, PAT, GHCB, and TSC_AUX is
> > mutually exclusive with nested virtualization, as KVM passes through the
> > MSRs only for SEV-ES guests, and KVM doesn't support nested virtualization
> > for SEV+ guests. Defer removing those MSRs to a future cleanup in order
> > to make this refactoring as benign as possible.
> >
> > Signed-off-by: Sean Christopherson <seanjc@google.com>
> > ---
> > arch/x86/kvm/svm/nested.c | 72 +++++++++++++++++++++++++++++++++------
> > arch/x86/kvm/svm/svm.c | 4 +++
> > arch/x86/kvm/svm/svm.h | 2 ++
> > 3 files changed, 67 insertions(+), 11 deletions(-)
> >
> > diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
> > index 89a77f0f1cc8..e53020939e60 100644
> > --- a/arch/x86/kvm/svm/nested.c
> > +++ b/arch/x86/kvm/svm/nested.c
> > @@ -184,6 +184,64 @@ void recalc_intercepts(struct vcpu_svm *svm)
> > }
> > }
> > +static int nested_svm_msrpm_merge_offsets[9] __ro_after_init;
> > +static int nested_svm_nr_msrpm_merge_offsets __ro_after_init;
> > +
> > +int __init nested_svm_init_msrpm_merge_offsets(void)
> > +{
> > + const u32 merge_msrs[] = {
>
> "static const", please.
Ugh, I was thinking the compiler would be magical enough to not generate code to
fill an on-stack array at runtime, but that's not the case.
AFAICT, tagging it __initdata works, so I'll do this to hopefully ensure the
memory is discarded after module load.
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index fb4808cf4711..af530f45bf64 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -205,7 +205,7 @@ static int svm_msrpm_offset(u32 msr)
int __init nested_svm_init_msrpm_merge_offsets(void)
{
- const u32 merge_msrs[] = {
+ static const u32 __initdata merge_msrs[] = {
MSR_STAR,
MSR_IA32_SYSENTER_CS,
MSR_IA32_SYSENTER_EIP,
next prev parent reply other threads:[~2025-06-04 15:49 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-29 23:39 [PATCH 00/28] KVM: x86: Clean up MSR interception code Sean Christopherson
2025-05-29 23:39 ` [PATCH 01/28] KVM: SVM: Don't BUG if setting up the MSR intercept bitmaps fails Sean Christopherson
2025-06-03 7:17 ` Chao Gao
2025-06-03 15:28 ` Sean Christopherson
2025-05-29 23:39 ` [PATCH 02/28] KVM: SVM: Tag MSR bitmap initialization helpers with __init Sean Christopherson
2025-06-03 7:18 ` Chao Gao
2025-05-29 23:39 ` [PATCH 03/28] KVM: SVM: Use ARRAY_SIZE() to iterate over direct_access_msrs Sean Christopherson
2025-06-03 7:57 ` Chao Gao
2025-05-29 23:39 ` [PATCH 04/28] KVM: SVM: Kill the VM instead of the host if MSR interception is buggy Sean Christopherson
2025-06-03 8:06 ` Chao Gao
2025-06-03 13:26 ` Sean Christopherson
2025-05-29 23:39 ` [PATCH 05/28] KVM: x86: Use non-atomic bit ops to manipulate "shadow" MSR intercepts Sean Christopherson
2025-06-03 2:53 ` Mi, Dapeng
2025-05-29 23:39 ` [PATCH 06/28] KVM: SVM: Massage name and param of helper that merges vmcb01 and vmcb12 MSRPMs Sean Christopherson
2025-06-03 2:32 ` Mi, Dapeng
2025-05-29 23:39 ` [PATCH 07/28] KVM: SVM: Clean up macros related to architectural MSRPM definitions Sean Christopherson
2025-05-29 23:39 ` [PATCH 08/28] KVM: nSVM: Use dedicated array of MSRPM offsets to merge L0 and L1 bitmaps Sean Christopherson
2025-06-04 5:43 ` Chao Gao
2025-06-04 13:56 ` Sean Christopherson
2025-06-05 8:08 ` Chao Gao
2025-06-04 15:35 ` Paolo Bonzini
2025-06-04 15:49 ` Sean Christopherson [this message]
2025-06-04 15:35 ` Paolo Bonzini
2025-05-29 23:39 ` [PATCH 09/28] KVM: nSVM: Omit SEV-ES specific passthrough MSRs from L0+L1 bitmap merge Sean Christopherson
2025-05-29 23:39 ` [PATCH 10/28] KVM: nSVM: Don't initialize vmcb02 MSRPM with vmcb01's "always passthrough" Sean Christopherson
2025-05-29 23:39 ` [PATCH 11/28] KVM: SVM: Add helpers for accessing MSR bitmap that don't rely on offsets Sean Christopherson
2025-06-04 16:11 ` Paolo Bonzini
2025-06-04 17:35 ` Sean Christopherson
2025-06-04 19:13 ` Paolo Bonzini
2025-05-29 23:39 ` [PATCH 12/28] KVM: SVM: Implement and adopt VMX style MSR intercepts APIs Sean Christopherson
2025-06-05 8:19 ` Chao Gao
2025-05-29 23:39 ` [PATCH 13/28] KVM: SVM: Pass through GHCB MSR if and only if VM is an SEV-ES guest Sean Christopherson
2025-05-29 23:39 ` [PATCH 14/28] KVM: SVM: Drop "always" flag from list of possible passthrough MSRs Sean Christopherson
2025-05-29 23:40 ` [PATCH 15/28] KVM: x86: Move definition of X2APIC_MSR() to lapic.h Sean Christopherson
2025-05-29 23:40 ` [PATCH 16/28] KVM: VMX: Manually recalc all MSR intercepts on userspace MSR filter change Sean Christopherson
2025-05-30 23:38 ` Xin Li
2025-06-02 17:10 ` Sean Christopherson
2025-06-03 3:52 ` Mi, Dapeng
2025-06-05 6:59 ` Chao Gao
2025-05-29 23:40 ` [PATCH 17/28] KVM: SVM: " Sean Christopherson
2025-05-31 18:01 ` Francesco Lavra
2025-06-02 17:08 ` Sean Christopherson
2025-06-05 6:24 ` Chao Gao
2025-06-05 16:39 ` Sean Christopherson
2025-05-29 23:40 ` [PATCH 18/28] KVM: x86: Rename msr_filter_changed() => recalc_msr_intercepts() Sean Christopherson
2025-05-30 23:47 ` Xin Li
2025-05-29 23:40 ` [PATCH 19/28] KVM: SVM: Rename init_vmcb_after_set_cpuid() to make it intercepts specific Sean Christopherson
2025-05-29 23:40 ` [PATCH 20/28] KVM: SVM: Fold svm_vcpu_init_msrpm() into its sole caller Sean Christopherson
2025-05-29 23:40 ` [PATCH 21/28] KVM: SVM: Merge "after set CPUID" intercept recalc helpers Sean Christopherson
2025-05-29 23:40 ` [PATCH 22/28] KVM: SVM: Drop explicit check on MSRPM offset when emulating SEV-ES accesses Sean Christopherson
2025-05-29 23:40 ` [PATCH 23/28] KVM: SVM: Move svm_msrpm_offset() to nested.c Sean Christopherson
2025-05-29 23:40 ` [PATCH 24/28] KVM: SVM: Store MSRPM pointer as "void *" instead of "u32 *" Sean Christopherson
2025-05-29 23:40 ` [PATCH 25/28] KVM: nSVM: Access MSRPM in 4-byte chunks only for merging L0 and L1 bitmaps Sean Christopherson
2025-06-04 16:19 ` Paolo Bonzini
2025-06-04 16:28 ` Sean Christopherson
2025-05-29 23:40 ` [PATCH 26/28] KVM: SVM: Return -EINVAL instead of MSR_INVALID to signal out-of-range MSR Sean Christopherson
2025-05-29 23:40 ` [PATCH 27/28] KVM: nSVM: Merge MSRPM in 64-bit chunks on 64-bit kernels Sean Christopherson
2025-05-29 23:40 ` [PATCH 28/28] KVM: selftests: Verify KVM disable interception (for userspace) on filter change Sean Christopherson
2025-06-03 5:47 ` Mi, Dapeng
2025-06-04 4:43 ` Manali Shukla
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=aEBrAbhRFkou4Mvj@google.com \
--to=seanjc@google.com \
--cc=bp@alien8.de \
--cc=chao.gao@intel.com \
--cc=dapeng1.mi@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=xin@zytor.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
all inboxes | Powered by JetHome®