The headliner is Like's patches to fix a bug where KVM's sleeps with IRQs
disabled due to attempting to pause and reprogram PMU counters in the
VM-Exit fast path.
Patches 1 and 2 (mine) fix mostly theoretical issues found when reviewing
Like's code, and would conflict with Like's patches if posted separately.
This is _very_ lightly tested, borderline RFC, but Like's updated PMU
KUT tests pass, as do the filter selftests, and I'm feeling lucky.
And I'm offline until Monday, so a traditional Friday patch bomb is in
order ;-)
Like Xu (2):
KVM: x86/pmu: Defer reprogram_counter() to kvm_pmu_handle_event()
KVM: x86/pmu: Defer counter emulated overflow via pmc->prev_counter
Sean Christopherson (2):
KVM: x86/pmu: Force reprogramming of all counters on PMU filter change
KVM: x86/pmu: Clear "reprogram" bit if counter is disabled or
disallowed
arch/x86/include/asm/kvm_host.h | 17 ++++--
arch/x86/kvm/pmu.c | 92 +++++++++++++++++++++------------
arch/x86/kvm/pmu.h | 6 ++-
arch/x86/kvm/svm/pmu.c | 4 +-
arch/x86/kvm/vmx/pmu_intel.c | 10 ++--
5 files changed, 86 insertions(+), 43 deletions(-)
base-commit: efc8a6befa592de7e2fd4927b1b9a56a5ec7d8cf
--
2.37.3.998.g577e59143f-goog
