Re: [PATCH v3] Enable MSR-BASED TPR shadow even if APICv is inactive

2016-09-21 Thread Wanpeng Li
2016-09-22 0:35 GMT+08:00 Radim Krčmář : > 2016-09-21 10:14+0800, Wanpeng Li: [...] > Could only be "else" without if here. I'll change that when refactoring > of vmx bitmap handling, posting soon (hopefully). > [...] > > Pasto here (disable vs. enable). It's not called with !apicv_active and > t

Re: [PATCH v3] Enable MSR-BASED TPR shadow even if APICv is inactive

2016-09-21 Thread Radim Krčmář
2016-09-21 10:14+0800, Wanpeng Li: > From: Wanpeng Li > > I observed that kvmvapic(to optimize flexpriority=N or AMD) is used > to boost TPR access when testing kvm-unit-test/eventinj.flat tpr case > on my haswell desktop (w/ flexpriority, w/o APICv). Commit (8d14695f9542 > x86, apicv: add virt

Re: [PATCH v3] Enable MSR-BASED TPR shadow even if APICv is inactive

2016-09-20 Thread Wanpeng Li
The subject should be "KVM: VMX: Enable MSR-BASED TPR shadow even if APICv is inactive" 2016-09-21 10:14 GMT+08:00 Wanpeng Li : > From: Wanpeng Li > > I observed that kvmvapic(to optimize flexpriority=N or AMD) is used > to boost TPR access when testing kvm-unit-test/eventinj.flat tpr case > on my

[PATCH v3] Enable MSR-BASED TPR shadow even if APICv is inactive

2016-09-20 Thread Wanpeng Li
From: Wanpeng Li I observed that kvmvapic(to optimize flexpriority=N or AMD) is used to boost TPR access when testing kvm-unit-test/eventinj.flat tpr case on my haswell desktop (w/ flexpriority, w/o APICv). Commit (8d14695f9542 x86, apicv: add virtual x2apic support) disable virtual x2apic mode