* Re: [PATCH 6.18.y 0/6] KVM: x86: CPUID faulting fixes and AMD CPUID faulting virtualization
[not found] <20261014212844.11997-1-aaltinay@google.com>
@ 2026-10-14 22:18 ` Anil Altinay
2026-10-15 5:22 ` Greg KH
0 siblings, 1 reply; 2+ messages in thread
From: Anil Altinay @ 2026-10-14 22:18 UTC (permalink / raw)
To: stable; +Cc: Anil Altinay, Jim Mattson, Sean Christopherson, Paolo Bonzini,
kvm
+Cc: Jim, Sean, Paolo, and kvm@vger.kernel.org (omitted from the initial
send due to --suppress-cc=all).
Full thread on lore:
https://lore.kernel.org/stable/20261014212844.11997-1-aaltinay@google.com/T/#t
On Wed, Oct 14, 2026 at 9:28 PM Anil Altinay <aaltinay@google.com> wrote:
>
> Please backport the following 6 upstream commits to the 6.18.y stable tree:
>
> 1. 690dc03859e7 ("KVM: x86: Ignore cpuid faulting in SMM") [v7.1-rc1]
> 2. cf7d65d1d6f5 ("KVM: x86: Consolidate CPUID fault handling for emulator and interception logic") [v7.2-rc1]
> 3. be7fd7c3e8bc ("KVM: x86: Prioritize CPUID faulting over CPUID VM-exits in nested VMX") [v7.2-rc1]
> 4. d1bc99885a59 ("KVM: x86: Remove supports_cpuid_fault() helper") [v7.2-rc1]
> 5. e93a93f11490 ("KVM: x86: Virtualize AMD CPUID faulting") [v7.2-rc1]
> 6. b16c2aca369d ("KVM: selftests: Update hwcr_msr_test for CPUID faulting bit") [v7.2-rc1]
>
> Patches 1 and 3 fix bugs in KVM's CPUID faulting virtualization (Fixes:
> db2336a80489 ("KVM: x86: virtualize cpuid faulting")), ensuring that
> CPUID does not fault in SMM and that CPUID faulting takes precedence
> over CPUID VM-exits in nested VMX. Patch 2 is a non-functional refactoring
> prerequisite for patch 3.
>
> Patches 4-6 extend KVM's CPUID faulting virtualization (already supported
> by KVM on Intel CPUs and by 6.18 host kernels on AMD CPUs via 65f55a301766
> ("x86/CPU/AMD: Add CPUID faulting support")) to AMD guests via
> CPUID.80000021H:EAX.CpuidUserDis[bit 17] and HWCR.CpuidUserDis[bit 35],
> allowing guests and container runtimes that rely on arch_prctl(ARCH_SET_CPUID)
> to work uniformly across Intel and AMD KVM hosts, along with the
> corresponding KVM selftest update.
>
> All patches cherry-pick cleanly to 6.18.y except for a minor context
> conflict in patch 2 (arch/x86/include/asm/kvm_host.h) and a u64 vs.
> uint64_t type adjustment in patch 6 (tools/testing/selftests/kvm/x86/hwcr_msr_test.c).
>
> Jim Mattson (5):
> KVM: x86: Ignore cpuid faulting in SMM
> KVM: x86: Prioritize CPUID faulting over CPUID VM-exits in nested VMX
> KVM: x86: Remove supports_cpuid_fault() helper
> KVM: x86: Virtualize AMD CPUID faulting
> KVM: selftests: Update hwcr_msr_test for CPUID faulting bit
>
> Sean Christopherson (1):
> KVM: x86: Consolidate CPUID fault handling for emulator and
> interception logic
>
> arch/x86/include/asm/kvm_host.h | 1 -
> arch/x86/include/asm/msr-index.h | 1 +
> arch/x86/kvm/cpuid.c | 6 ++-
> arch/x86/kvm/cpuid.h | 14 ++++---
> arch/x86/kvm/emulate.c | 6 +--
> arch/x86/kvm/kvm_emulate.h | 1 +
> arch/x86/kvm/vmx/nested.c | 2 +
> arch/x86/kvm/x86.c | 39 ++++++++++---------
> .../selftests/kvm/include/x86/processor.h | 1 +
> .../testing/selftests/kvm/x86/hwcr_msr_test.c | 9 ++++-
> 10 files changed, 46 insertions(+), 34 deletions(-)
^ permalink raw reply [flat|nested] 2+ messages in thread