Re: [GIT PULL] KVM: x86: SVM+SEV changes

[Paolo Bonzini]

On Sat, Apr 11, 2026 at 1:58 AM Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
>
> This is the full set of SVM+SEV changes.  The end goal of the SEV changes, after
> fixing a few fatal bugs, is to add a lockdep assertion to ensure that kvm->lock
> is held when checking if the VM is an SEV+ guest.  This is at least the second
> fatal bug we've had due to SEV+ state being unwound on failure, and lack of
> formal-ish rules makes it hard to reason about the safety of any related code,
> e.g. when reviewing new code.
>
> This has a superficial (I can't even figure out why git treats it as a conflict,
> I think it's both deleting white space or something?) syntactic conflict with the
> "vmxon" PULL request; just take this one.

Yes, it's deleting whitespace.  I pulled the full version and will
review it over the week, leaving this PR last just in case.

Paolo

> There's a syntactic conflict with the "nested" PULL request (this is what I
> see when merging the "nested" one first):
>
> @@@ -870,8 -881,8 +886,8 @@@ void svm_enable_lbrv(struct kvm_vcpu *v
>
>   static void __svm_disable_lbrv(struct kvm_vcpu *vcpu)
>   {
> -       KVM_BUG_ON(sev_es_guest(vcpu->kvm), vcpu->kvm);
> +       KVM_BUG_ON(is_sev_es_guest(vcpu), vcpu->kvm);
>  -      to_svm(vcpu)->vmcb->control.virt_ext &= ~LBR_CTL_ENABLE_MASK;
>  +      to_svm(vcpu)->vmcb->control.misc_ctl2 &= ~SVM_MISC2_ENABLE_V_LBR;
>   }
>
> and a semantic conflict with kvm/master due to the CR8 interception fix:
>
> diff --cc arch/x86/kvm/svm/avic.c
> index 2885c5993ebc,7056c4891f93..adf211860949
> --- a/arch/x86/kvm/svm/avic.c
> +++ b/arch/x86/kvm/svm/avic.c
> @@@ -226,9 -237,6 +237,9 @@@ static void avic_deactivate_vmcb(struc
>         vmcb->control.int_ctl &= ~(AVIC_ENABLE_MASK | X2APIC_MODE_MASK);
>         vmcb->control.avic_physical_id &= ~AVIC_PHYSICAL_MAX_INDEX_MASK;
>
> -       if (!sev_es_guest(svm->vcpu.kvm))
> ++      if (!is_sev_es_guest(&svm->vcpu))
>  +              svm_set_intercept(svm, INTERCEPT_CR8_WRITE);
>  +
>         /*
>          * If running nested and the guest uses its own MSR bitmap, there
>          * is no need to update L0's msr bitmap
>
>
> The following changes since commit 11439c4635edd669ae435eec308f4ab8a0804808:
>
>   Linux 7.0-rc2 (2026-03-01 15:39:31 -0800)
>
> are available in the Git repository at:
>
>   https://github.com/kvm-x86/linux.git tags/kvm-x86-svm-7.1
>
> for you to fetch changes up to bc0932cf9b9917e826871db947398aa2b62789b2:
>
>   KVM: SEV: Goto an existing error label if charging misc_cg for an ASID fails (2026-04-09 12:00:24 -0700)
>
> ----------------------------------------------------------------
> KVM SVM changes for 7.1
>
>  - Fix and optimize IRQ window inhibit handling for AVIC (the tracking needs to
>    be per-vCPU, e.g. so that KVM doesn't prematurely re-enable AVIC if multiple
>    vCPUs have to-be-injected IRQs).
>
>  - Fix an undefined behavior warning where a crafty userspace can read the
>    "avic" module param before it's fully initialized.
>
>  - Fix a (likely benign) bug in the "OS-visible workarounds" handling, where
>    KVM could clobber state when enabling virtualization on multiple CPUs in
>    parallel, and clean up and optimize the code.
>
>  - Drop a WARN in KVM_MEMORY_ENCRYPT_REG_REGION where KVM complains about a
>    "too large" size based purely on user input, and clean up and harden the
>    related pinning code.
>
>  - Disallow synchronizing a VMSA of an already-launched/encrypted vCPU, as
>    doing so for an SNP guest will trigger an RMP violation #PF and crash the
>    host.
>
>  - Protect all of sev_mem_enc_register_region() with kvm->lock to ensure
>    sev_guest() is stable for the entire of the function.
>
>  - Lock all vCPUs when synchronizing VMSAs for SNP guests to ensure the VMSA
>    page isn't actively being used.
>
>  - Overhaul KVM's APIs for detecting SEV+ guests so that VM-scoped queries are
>    required to hold kvm->lock (KVM has had multiple bugs due "is SEV?" checks
>    becoming stale), enforced by lockdep.  Add and use vCPU-scoped APIs when
>    possible/appropriate, as all checks that originate from a vCPU are
>    guaranteed to be stable.
>
>  - Convert a pile of kvm->lock SEV code to guard().
>
> ----------------------------------------------------------------
> Carlos López (5):
>       KVM: SEV: use mutex guard in snp_launch_update()
>       KVM: SEV: use mutex guard in sev_mem_enc_ioctl()
>       KVM: SEV: use mutex guard in sev_mem_enc_unregister_region()
>       KVM: SEV: use mutex guard in snp_handle_guest_req()
>       KVM: SVM: Move lock-protected allocation of SEV ASID into a separate helper
>
> Gal Pressman (1):
>       KVM: SVM: Fix UBSAN warning when reading avic parameter
>
> Li RongQing (1):
>       KVM: SVM: Mark module parameters as __ro_after_init for security and performance
>
> Sean Christopherson (30):
>       KVM: SVM: Fix clearing IRQ window inhibit with nested guests
>       KVM: SVM: Fix IRQ window inhibit handling across multiple vCPUs
>       KVM: SVM: Optimize IRQ window inhibit handling
>       KVM: Isolate apicv_update_lock and apicv_nr_irq_window_req in a cacheline
>       KVM: SVM: Serialize updates to global OS-Visible Workarounds variables
>       KVM: SVM: Skip OSVW MSR reads if KVM is treating all errata as present
>       KVM: SVM: Extract OS-visible workarounds setup to helper function
>       KVM: SVM: Skip OSVW variable updates if current CPU's errata are a subset
>       KVM: SVM: Skip OSVW MSR reads if current CPU doesn't support the feature
>       KVM: SEV: Drop WARN on large size for KVM_MEMORY_ENCRYPT_REG_REGION
>       KVM: SEV: Drop useless sanity checks in sev_mem_enc_register_region()
>       KVM: SEV: Disallow pinning more pages than exist in the system
>       KVM: SEV: Use PFN_DOWN() to simplify "number of pages" math when pinning memory
>       KVM: SEV: Use kvzalloc_objs() when pinning userpages
>       KVM: selftests: Remove duplicate LAUNCH_UPDATE_VMSA call in SEV-ES migrate test
>       KVM: SEV: Reject attempts to sync VMSA of an already-launched/encrypted vCPU
>       KVM: SEV: Protect *all* of sev_mem_enc_register_region() with kvm->lock
>       KVM: SEV: Disallow LAUNCH_FINISH if vCPUs are actively being created
>       KVM: SEV: Lock all vCPUs when synchronzing VMSAs for SNP launch finish
>       KVM: SEV: Lock all vCPUs for the duration of SEV-ES VMSA synchronization
>       KVM: SEV: Provide vCPU-scoped accessors for detecting SEV+ guests
>       KVM: SEV: Add quad-underscore version of VM-scoped APIs to detect SEV+ guests
>       KVM: SEV: Document the SEV-ES check when querying SMM support as "safe"
>       KVM: SEV: Move standard VM-scoped helpers to detect SEV+ guests to sev.c
>       KVM: SEV: Move SEV-specific VM initialization to sev.c
>       KVM: SEV: WARN on unhandled VM type when initializing VM
>       KVM: SEV: Hide "struct kvm_sev_info" behind CONFIG_KVM_AMD_SEV=y
>       KVM: SEV: Document that checking for SEV+ guests when reclaiming memory is "safe"
>       KVM: SEV: Assert that kvm->lock is held when querying SEV+ support
>       KVM: SEV: Goto an existing error label if charging misc_cg for an ASID fails
>
>  arch/x86/include/asm/kvm_host.h                    |  29 +-
>  arch/x86/kvm/svm/avic.c                            |  17 +-
>  arch/x86/kvm/svm/sev.c                             | 374 ++++++++++++---------
>  arch/x86/kvm/svm/svm.c                             | 270 ++++++++-------
>  arch/x86/kvm/svm/svm.h                             |  37 +-
>  arch/x86/kvm/x86.c                                 |  45 ++-
>  include/linux/kvm_host.h                           |   7 +
>  .../testing/selftests/kvm/x86/sev_migrate_tests.c  |   2 -
>  8 files changed, 487 insertions(+), 294 deletions(-)
>