[PATCH OLK-5.10 v2 0/2] kvm: debugfs: some improvement for
From: chenjiajun <chenjiajun8@huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA some improvement for vcpu_stat: 1. Add fastpath kvm exit handles to vcpu_stat, avoid the big deviation between the statistical data and the actual scene. 2. Export preemption timer kvm exits to statistical data. -- ChangeList: v2: move preemption timer exits to the end of vcpu_stat v1: kvm: debugfs: some improvement for vcpu_stat chenjiajun (2): kvm: debugfs: add fastpath msr_wr exits to debugfs statistics kvm: debugfs: add EXIT_REASON_PREEMPTION_TIMER to vcpu_stat arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/vmx/vmx.c | 3 +++ arch/x86/kvm/x86.c | 1 + 3 files changed, 5 insertions(+) -- 2.29.GIT
From: chenjiajun <chenjiajun8@huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA At present, there is a flaw in the statistics of KVM exits by debugfs, which only counts trigger times of exits processing function in kvm_vmx_exit_handlers. The kvm exits handles in vmx_exit_handlers_fastpath is omitted, so there is a large numerical error in EXIT_REASON_MSR_WRITE statistics sometimes. Signed-off-by: chenjiajun <chenjiajun8@huawei.com> --- arch/x86/kvm/vmx/vmx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 7556a380a148..3a0964619070 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -6595,6 +6595,7 @@ static fastpath_t vmx_exit_handlers_fastpath(struct kvm_vcpu *vcpu) { switch (to_vmx(vcpu)->exit_reason) { case EXIT_REASON_MSR_WRITE: + ++vcpu->stat.msr_wr_exits; return handle_fastpath_set_msr_irqoff(vcpu); case EXIT_REASON_PREEMPTION_TIMER: return handle_fastpath_preemption_timer(vcpu); -- 2.29.GIT
From: chenjiajun <chenjiajun8@huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA Export EXIT_REASON_PREEMPTION_TIMER kvm exits to vcpu_stat debugfs. Add a new column to vcpu_stat, and provide preemption_timer status to virtualization detection tools. Signed-off-by: chenjiajun <chenjiajun8@huawei.com> --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/vmx/vmx.c | 2 ++ arch/x86/kvm/x86.c | 1 + 3 files changed, 4 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 730ffde044b4..13ba97e63457 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1070,6 +1070,7 @@ struct kvm_vcpu_stat { u64 utime; u64 stime; u64 gtime; + u64 preemption_timer_exits; }; struct x86_instruction_info; diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 3a0964619070..93a419fe9509 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -5605,6 +5605,7 @@ static fastpath_t handle_fastpath_preemption_timer(struct kvm_vcpu *vcpu) static int handle_preemption_timer(struct kvm_vcpu *vcpu) { + ++vcpu->stat.preemption_timer_exits; handle_fastpath_preemption_timer(vcpu); return 1; } @@ -6598,6 +6599,7 @@ static fastpath_t vmx_exit_handlers_fastpath(struct kvm_vcpu *vcpu) ++vcpu->stat.msr_wr_exits; return handle_fastpath_set_msr_irqoff(vcpu); case EXIT_REASON_PREEMPTION_TIMER: + ++vcpu->stat.preemption_timer_exits; return handle_fastpath_preemption_timer(vcpu); default: return EXIT_FASTPATH_NONE; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 3b074dacd096..faabdd581366 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -285,6 +285,7 @@ struct dfx_kvm_stats_debugfs_item dfx_debugfs_entries[] = { DFX_STAT("utime", utime), DFX_STAT("stime", stime), DFX_STAT("gtime", gtime), + DFX_STAT("preemption_timer_exits", preemption_timer_exits), { NULL } }; -- 2.29.GIT
participants (1)
-
Jiajun Chen