diff options
author | Wanpeng Li <wanpengli@tencent.com> | 2019-08-05 10:03:19 +0800 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2019-08-16 10:12:53 +0200 |
commit | 2bc73d91411423dd7092596f9c0f91d3ea5a9e26 (patch) | |
tree | 702ce1062a484e4bfda4d7635706e9050c171d8e /arch | |
parent | d1489f0b4de73a06f10d3cc9c7dddc5f89784abb (diff) | |
download | linux-stable-2bc73d91411423dd7092596f9c0f91d3ea5a9e26.tar.gz linux-stable-2bc73d91411423dd7092596f9c0f91d3ea5a9e26.tar.bz2 linux-stable-2bc73d91411423dd7092596f9c0f91d3ea5a9e26.zip |
KVM: Fix leak vCPU's VMCS value into other pCPU
commit 17e433b54393a6269acbcb792da97791fe1592d8 upstream.
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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'arch')
-rw-r--r-- | arch/powerpc/kvm/powerpc.c | 5 | ||||
-rw-r--r-- | arch/x86/include/asm/kvm_host.h | 1 | ||||
-rw-r--r-- | arch/x86/kvm/svm.c | 6 | ||||
-rw-r--r-- | arch/x86/kvm/vmx.c | 6 | ||||
-rw-r--r-- | arch/x86/kvm/x86.c | 16 |
5 files changed, 34 insertions, 0 deletions
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 578174a33d22..51cd66dc1bb0 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -61,6 +61,11 @@ 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 2877e1fbadd8..3245b95ad2d9 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1113,6 +1113,7 @@ 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); void (*cancel_hv_timer)(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index ea454d3f7763..0f33f00aa4df 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -5146,6 +5146,11 @@ 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; @@ -7203,6 +7208,7 @@ 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.c b/arch/x86/kvm/vmx.c index 4cf16378dffe..2e310ea62d60 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -10411,6 +10411,11 @@ static u8 vmx_has_apicv_interrupt(struct kvm_vcpu *vcpu) return ((rvi & 0xf0) > (vppr & 0xf0)); } +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)) @@ -14387,6 +14392,7 @@ 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 cea6568667c4..e10a7a42449b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -9336,6 +9336,22 @@ 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; |