summaryrefslogtreecommitdiff
path: root/arch
diff options
context:
space:
mode:
authorWanpeng Li <wanpengli@tencent.com>2019-08-05 10:03:19 +0800
committerPaolo Bonzini <pbonzini@redhat.com>2019-08-05 12:55:47 +0200
commit17e433b54393a6269acbcb792da97791fe1592d8 (patch)
treea8bf7553d3bdb3110f126917f4a548e460f2fb52 /arch
parent046ddeed0461b5d270470c253cbb321103d048b6 (diff)
KVM: Fix leak vCPU's VMCS value into other pCPU
After commit d73eb57b80b (KVM: Boost vCPUs that are delivering interrupts), a five years old bug is exposed. Running ebizzy benchmark in three 80 vCPUs VMs on one 80 pCPUs Skylake server, a lot of rcu_sched stall warning splatting in the VMs after stress testing: INFO: rcu_sched detected stalls on CPUs/tasks: { 4 41 57 62 77} (detected by 15, t=60004 jiffies, g=899, c=898, q=15073) Call Trace: flush_tlb_mm_range+0x68/0x140 tlb_flush_mmu.part.75+0x37/0xe0 tlb_finish_mmu+0x55/0x60 zap_page_range+0x142/0x190 SyS_madvise+0x3cd/0x9c0 system_call_fastpath+0x1c/0x21 swait_active() sustains to be true before finish_swait() is called in kvm_vcpu_block(), voluntarily preempted vCPUs are taken into account by kvm_vcpu_on_spin() loop greatly increases the probability condition kvm_arch_vcpu_runnable(vcpu) is checked and can be true, when APICv is enabled the yield-candidate vCPU's VMCS RVI field leaks(by vmx_sync_pir_to_irr()) into spinning-on-a-taken-lock vCPU's current VMCS. This patch fixes it by checking conservatively a subset of events. Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim Krčmář <rkrcmar@redhat.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Marc Zyngier <Marc.Zyngier@arm.com> Cc: stable@vger.kernel.org Fixes: 98f4a1467 (KVM: add kvm_arch_vcpu_runnable() test to kvm_vcpu_on_spin() loop) Signed-off-by: Wanpeng Li <wanpengli@tencent.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'arch')
-rw-r--r--arch/powerpc/kvm/powerpc.c5
-rw-r--r--arch/x86/include/asm/kvm_host.h1
-rw-r--r--arch/x86/kvm/svm.c6
-rw-r--r--arch/x86/kvm/vmx/vmx.c6
-rw-r--r--arch/x86/kvm/x86.c16
5 files changed, 34 insertions, 0 deletions
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 0dba7eb24f92..3e34d5fa6708 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -50,6 +50,11 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
return !!(v->arch.pending_exceptions) || kvm_request_pending(v);
}
+bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu)
+{
+ return kvm_arch_vcpu_runnable(vcpu);
+}
+
bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
{
return false;
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index e74f0711eaaf..fc046ca89d32 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1175,6 +1175,7 @@ struct kvm_x86_ops {
int (*update_pi_irte)(struct kvm *kvm, unsigned int host_irq,
uint32_t guest_irq, bool set);
void (*apicv_post_state_restore)(struct kvm_vcpu *vcpu);
+ bool (*dy_apicv_has_pending_interrupt)(struct kvm_vcpu *vcpu);
int (*set_hv_timer)(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc,
bool *expired);
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 7eafc6907861..d685491fce4d 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -5190,6 +5190,11 @@ static void svm_deliver_avic_intr(struct kvm_vcpu *vcpu, int vec)
kvm_vcpu_wake_up(vcpu);
}
+static bool svm_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu)
+{
+ return false;
+}
+
static void svm_ir_list_del(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi)
{
unsigned long flags;
@@ -7314,6 +7319,7 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = {
.pmu_ops = &amd_pmu_ops,
.deliver_posted_interrupt = svm_deliver_avic_intr,
+ .dy_apicv_has_pending_interrupt = svm_dy_apicv_has_pending_interrupt,
.update_pi_irte = svm_update_pi_irte,
.setup_mce = svm_setup_mce,
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 074385c86c09..42ed3faa6af8 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -6117,6 +6117,11 @@ static int vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu)
return max_irr;
}
+static bool vmx_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu)
+{
+ return pi_test_on(vcpu_to_pi_desc(vcpu));
+}
+
static void vmx_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
{
if (!kvm_vcpu_apicv_active(vcpu))
@@ -7726,6 +7731,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
.guest_apic_has_interrupt = vmx_guest_apic_has_interrupt,
.sync_pir_to_irr = vmx_sync_pir_to_irr,
.deliver_posted_interrupt = vmx_deliver_posted_interrupt,
+ .dy_apicv_has_pending_interrupt = vmx_dy_apicv_has_pending_interrupt,
.set_tss_addr = vmx_set_tss_addr,
.set_identity_map_addr = vmx_set_identity_map_addr,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c6d951cbd76c..93b0bd45ac73 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9698,6 +9698,22 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
return kvm_vcpu_running(vcpu) || kvm_vcpu_has_events(vcpu);
}
+bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu)
+{
+ if (READ_ONCE(vcpu->arch.pv.pv_unhalted))
+ return true;
+
+ if (kvm_test_request(KVM_REQ_NMI, vcpu) ||
+ kvm_test_request(KVM_REQ_SMI, vcpu) ||
+ kvm_test_request(KVM_REQ_EVENT, vcpu))
+ return true;
+
+ if (vcpu->arch.apicv_active && kvm_x86_ops->dy_apicv_has_pending_interrupt(vcpu))
+ return true;
+
+ return false;
+}
+
bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
{
return vcpu->arch.preempted_in_kernel;