KVM: Fixes for 6.10-rcN
Sean Christopherson
reviewerTwo selftests fixes and two KVM fixes for 6.10. Only the kvm_vcpu_on_spin() fix is tagged for stable@. I deliberately didn't tag the mmu_notifier issue for stable@ as I'm 99.9% certain it can't actually cause problems (other than making UBSAN sad).
Sorry for the late-in-the-week request, I have a feeling I missed your window where you're working on stuff for this week's KVM pull request. :-/
The following changes since commit c3f38fa61af77b49866b006939479069cd451173:
Linux 6.10-rc2 (2024-06-02 15:44:56 -0700)
are available in the Git repository at:
https://github.com/kvm-x86/linux.git tags/kvm-x86-fixes-6.10-rcN
for you to fetch changes up to c3f3edf73a8f854f8766a69d2734198a58762e33:
KVM: Stop processing *all* memslots when "null" mmu_notifier handler is found (2024-06-18 08:51:03 -0700)
----------------------------------------------------------------
KVM fixes for 6.10
- Fix a "shift too big" goof in the KVM_SEV_INIT2 selftest.
- Compute the max mappable gfn for KVM selftests on x86 using GuestMaxPhyAddr from KVM's supported CPUID (if it's available).
- Fix a race in kvm_vcpu_on_spin() by ensuring loads and stores are atomic.
- Fix technically benign bug in __kvm_handle_hva_range() where KVM consumes the return from a void-returning function as if it were a boolean.
----------------------------------------------------------------
Babu Moger (1):
KVM: Stop processing *all* memslots when "null" mmu_notifier handler is found
Breno Leitao (1):
KVM: Fix a data race on last_boosted_vcpu in kvm_vcpu_on_spin()
Colin Ian King (1):
KVM: selftests: Fix shift of 32 bit unsigned int more than 32 bits
Tao Su (1):
KVM: selftests: x86: Prioritize getting max_gfn from GuestPhysBits
tools/testing/selftests/kvm/include/x86_64/processor.h | 1 + tools/testing/selftests/kvm/lib/x86_64/processor.c | 15 +++++++++++++-- tools/testing/selftests/kvm/x86_64/sev_init2_tests.c | 4 ++-- virt/kvm/kvm_main.c | 8 +++++--- 4 files changed, 21 insertions(+), 7 deletions(-)
Paolo Bonzini
maintainerQuoted 8 lines · Sean Christopherson
Two selftests fixes and two KVM fixes for 6.10. Only the
kvm_vcpu_on_spin()fix is tagged for stable@. I deliberately didn't tag themmu_notifierissue for stable@ as I'm 99.9% certain it can't actually cause problems (other than making UBSAN sad).Sorry for the late-in-the-week request, I have a feeling I missed your window where you're working on stuff for this week's KVM pull request. :-/
Nope, I read that you were going to send something.
Pulled, thanks.
Paolo
Quoted 44 lines
The following changes since commit
c3f38fa61af77b49866b006939479069cd451173:Linux 6.10-rc2 (2024-06-02 15:44:56 -0700)
are available in the Git repository at:
https://github.com/kvm-x86/linux.git tags/kvm-x86-fixes-6.10-rcN
for you to fetch changes up to
c3f3edf73a8f854f8766a69d2734198a58762e33:KVM: Stop processing *all* memslots when "null"
mmu_notifierhandler is found (2024-06-18 08:51:03 -0700)----------------------------------------------------------------
KVM fixes for 6.10- Fix a "shift too big" goof in the
KVM_SEV_INIT2selftest.- Compute the max mappable gfn for KVM selftests on x86 using GuestMaxPhyAddr from KVM's supported CPUID (if it's available).
- Fix a race in
kvm_vcpu_on_spin()by ensuring loads and stores are atomic.- Fix technically benign bug in
__kvm_handle_hva_range()where KVM consumes the return from a void-returning function as if it were a boolean.----------------------------------------------------------------
Babu Moger (1):
KVM: Stop processing *all* memslots when "null"mmu_notifierhandler is foundBreno Leitao (1):
KVM: Fix a data race onlast_boosted_vcpuinkvm_vcpu_on_spin()Colin Ian King (1):
KVM: selftests: Fix shift of 32 bit unsigned int more than 32 bitsTao Su (1):
KVM: selftests: x86: Prioritize gettingmax_gfnfrom GuestPhysBitstools/testing/selftests/kvm/include/x86_64/processor.h | 1 + tools/testing/selftests/kvm/lib/x86_64/processor.c | 15 +++++++++++++-- tools/testing/selftests/kvm/x86_64/sev_init2_tests.c | 4 ++-- virt/kvm/kvm_main.c | 8 +++++--- 4 files changed, 21 insertions(+), 7 deletions(-)