]> git.baikalelectronics.ru Git - kernel.git/commit
KVM: nVMX: Open a window for pending nested VMX preemption timer
authorSean Christopherson <sean.j.christopherson@intel.com>
Thu, 23 Apr 2020 02:25:39 +0000 (19:25 -0700)
committerPaolo Bonzini <pbonzini@redhat.com>
Wed, 13 May 2020 16:14:27 +0000 (12:14 -0400)
commit733f55fb39a13fcd27fa018d3f3d85a7eedfe891
tree860fb89ee17052748e003aa336eab08b74826c57
parente83b9c7cb85222a84174366caf033a806e15a6f4
KVM: nVMX: Open a window for pending nested VMX preemption timer

Add a kvm_x86_ops hook to detect a nested pending "hypervisor timer" and
use it to effectively open a window for servicing the expired timer.
Like pending SMIs on VMX, opening a window simply means requesting an
immediate exit.

This fixes a bug where an expired VMX preemption timer (for L2) will be
delayed and/or lost if a pending exception is injected into L2.  The
pending exception is rightly prioritized by vmx_check_nested_events()
and injected into L2, with the preemption timer left pending.  Because
no window opened, L2 is free to run uninterrupted.

Fixes: fb3f609212639 ("KVM: nVMX: Fully emulate preemption timer")
Reported-by: Jim Mattson <jmattson@google.com>
Cc: Oliver Upton <oupton@google.com>
Cc: Peter Shier <pshier@google.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200423022550.15113-3-sean.j.christopherson@intel.com>
[Check it in kvm_vcpu_has_events too, to ensure that the preemption
 timer is serviced promptly even if the vCPU is halted and L1 is not
 intercepting HLT. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/include/asm/kvm_host.h
arch/x86/kvm/vmx/nested.c
arch/x86/kvm/x86.c