KVM: PIT: provide an option to disable interrupt reinjection

Certain clocks (such as TSC) in older 2.6 guests overaccount for lost
ticks, causing severe time drift. Interrupt reinjection magnifies the
problem.

Provide an option to disable it.

[avi: allow room for expansion in case we want to disable reinjection
      of other timers]

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>

diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h
index 32eb96c7ca2780d9d8407e9b94cc1256491930e3..54bcf2281526f508162cb17211ec5332359100b3 100644 (file)
--- a/arch/x86/include/asm/kvm.h
+++ b/arch/x86/include/asm/kvm.h
@@ -233,4 +233,9 @@ struct kvm_guest_debug_arch {
 struct kvm_pit_state {
        struct kvm_pit_channel_state channels[3];
 };
+
+struct kvm_reinject_control {
+       __u8 pit_reinject;
+       __u8 reserved[31];
+};
 #endif /* _ASM_X86_KVM_H */

diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index 72bd275a9b5cee632b393448e325f371610f860b..528daadeba4953251e3bc5916a839b2637858b6a 100644 (file)
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -201,6 +201,9 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps)
        if (!atomic_inc_and_test(&pt->pending))
                set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests);
 
+       if (!pt->reinject)
+               atomic_set(&pt->pending, 1);
+
        if (vcpu0 && waitqueue_active(&vcpu0->wq))
                wake_up_interruptible(&vcpu0->wq);
 
@@ -580,6 +583,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm)
        pit_state->irq_ack_notifier.gsi = 0;
        pit_state->irq_ack_notifier.irq_acked = kvm_pit_ack_irq;
        kvm_register_irq_ack_notifier(kvm, &pit_state->irq_ack_notifier);
+       pit_state->pit_timer.reinject = true;
        mutex_unlock(&pit->pit_state.lock);
 
        kvm_pit_reset(pit);

diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h
index 4178022b97aac8d66b4f3cfb535102b8bd47140d..76959c4b500e5b46dbb39fff8e1ad717cf5779b8 100644 (file)
--- a/arch/x86/kvm/i8254.h
+++ b/arch/x86/kvm/i8254.h
@@ -9,6 +9,7 @@ struct kvm_kpit_timer {
        s64 period; /* unit: ns */
        s64 scheduled;
        atomic_t pending;
+       bool reinject;
 };
 
 struct kvm_kpit_channel_state {

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c3fbe8c55c136e7b494832fdcdd8ae9d82a82878..a1f14611f4b926cac053409764da40acff06d307 100644 (file)
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -993,6 +993,7 @@ int kvm_dev_ioctl_check_extension(long ext)
        case KVM_CAP_NOP_IO_DELAY:
        case KVM_CAP_MP_STATE:
        case KVM_CAP_SYNC_MMU:
+       case KVM_CAP_REINJECT_CONTROL:
                r = 1;
                break;
        case KVM_CAP_COALESCED_MMIO:
@@ -1728,6 +1729,15 @@ static int kvm_vm_ioctl_set_pit(struct kvm *kvm, struct kvm_pit_state *ps)
        return r;
 }
 
+static int kvm_vm_ioctl_reinject(struct kvm *kvm,
+                                struct kvm_reinject_control *control)
+{
+       if (!kvm->arch.vpit)
+               return -ENXIO;
+       kvm->arch.vpit->pit_state.pit_timer.reinject = control->pit_reinject;
+       return 0;
+}
+
 /*
  * Get (and clear) the dirty memory log for a memory slot.
  */
@@ -1925,6 +1935,17 @@ long kvm_arch_vm_ioctl(struct file *filp,
                r = 0;
                break;
        }
+       case KVM_REINJECT_CONTROL: {
+               struct kvm_reinject_control control;
+               r =  -EFAULT;
+               if (copy_from_user(&control, argp, sizeof(control)))
+                       goto out;
+               r = kvm_vm_ioctl_reinject(kvm, &control);
+               if (r)
+                       goto out;
+               r = 0;
+               break;
+       }
        default:
                ;
        }

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 11e3e6197c83c3191725ca815672ae6cadc4470d..ae7a12c77427a3200c8091537e362263201e0825 100644 (file)
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -396,6 +396,9 @@ struct kvm_trace_rec {
 #if defined(CONFIG_X86)
 #define KVM_CAP_SET_GUEST_DEBUG 23
 #endif
+#if defined(CONFIG_X86)
+#define KVM_CAP_REINJECT_CONTROL 24
+#endif
 
 /*
  * ioctls for VM fds
@@ -429,6 +432,7 @@ struct kvm_trace_rec {
                                   struct kvm_assigned_pci_dev)
 #define KVM_ASSIGN_IRQ _IOR(KVMIO, 0x70, \
                            struct kvm_assigned_irq)
+#define KVM_REINJECT_CONTROL      _IO(KVMIO, 0x71)
 
 /*
  * ioctls for vcpu fds