From d75cf5b26969744b257db5de4161daf4f0822928 Mon Sep 17 00:00:00 2001
From: Glauber Costa <glommer@redhat.com>
Date: Tue, 10 Nov 2009 13:52:04 -0200
Subject: [PATCH] 531701 KERNEL v2: allow userspace to adjust kvmclock offset

RH-Author: Glauber Costa <glommer@redhat.com>
Message-id: <1257271183-26516-1-git-send-email-glommer@redhat.com>
Patchwork-id: 3668
O-Subject: [PATCH] 531701 KERNEL v2: allow userspace to adjust kvmclock offset
Bugzilla: 531701
RH-Acked-by: Juan Quintela <quintela@redhat.com>
RH-Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
RH-Acked-by: Rik van Riel <riel@redhat.com>

When we migrate a kvm guest that uses pvclock between two hosts, we may
suffer a large skew. This is because there can be significant differences
between the monotonic clock of the hosts involved. When a new host with
a much larger monotonic time starts running the guest, the view of time
will be significantly impacted.

Situation is much worse when we do the opposite, and migrate to a host with
a smaller monotonic clock.

This proposed ioctl will allow userspace to inform us what is the monotonic
clock value in the source host, so we can keep the time skew short, and
more importantly, never goes backwards. Userspace may also need to trigger
the current data, since from the first migration onwards, it won't be
reflected by a simple call to clock_gettime() anymore.

[v2: initialize flags when doing GET ]

Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
RH-Upstream-Status: kvm/master
RH-Bugzilla: BZ531701
---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/x86.c              |   42 ++++++++++++++++++++++++++++++++++++++-
 include/linux/kvm.h             |   10 +++++++++
 3 files changed, 52 insertions(+), 1 deletions(-)

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/x86.c              |   42 ++++++++++++++++++++++++++++++++++++++-
 include/linux/kvm.h             |   10 +++++++++
 3 files changed, 52 insertions(+), 1 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index aafb359..35dd13b 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -380,6 +380,7 @@ struct kvm_arch{
 	unsigned long irq_sources_bitmap;
 	unsigned long irq_states[KVM_IOAPIC_NUM_PINS];
 	u64 vm_init_tsc;
+	s64 kvmclock_offset;
 };
 
 struct kvm_vm_stat {
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 85a18e6..a562932 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -616,7 +616,8 @@ static void kvm_write_guest_time(struct kvm_vcpu *v)
 	/* With all the info we got, fill in the values */
 
 	vcpu->hv_clock.system_time = ts.tv_nsec +
-				     (NSEC_PER_SEC * (u64)ts.tv_sec);
+				     (NSEC_PER_SEC * (u64)ts.tv_sec) + v->kvm->arch.kvmclock_offset;
+
 	/*
 	 * The interface expects us to write an even number signaling that the
 	 * update is finished. Since the guest won't see the intermediate
@@ -1027,6 +1028,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_REINJECT_CONTROL:
 	case KVM_CAP_IRQ_INJECT_STATUS:
 	case KVM_CAP_ASSIGN_DEV_IRQ:
+ 	case KVM_CAP_ADJUST_CLOCK:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -2007,6 +2009,44 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
+	case KVM_SET_CLOCK: {
+		struct timespec now;
+		struct kvm_clock_data user_ns;
+		u64 now_ns;
+		s64 delta;
+
+		r = -EFAULT;
+		if (copy_from_user(&user_ns, argp, sizeof(user_ns)))
+			goto out;
+
+		r = -EINVAL;
+		if (user_ns.flags)
+			goto out;
+
+		r = 0;
+		ktime_get_ts(&now);
+		now_ns = timespec_to_ns(&now);
+		delta = user_ns.clock - now_ns;
+		kvm->arch.kvmclock_offset = delta;
+		break;
+	}
+	case KVM_GET_CLOCK: {
+		struct timespec now;
+		struct kvm_clock_data user_ns;
+		u64 now_ns;
+
+		ktime_get_ts(&now);
+		now_ns = timespec_to_ns(&now);
+		user_ns.clock = kvm->arch.kvmclock_offset + now_ns;
+		user_ns.flags = 0;
+
+		r = -EFAULT;
+		if (copy_to_user(argp, &user_ns, sizeof(user_ns)))
+			goto out;
+		r = 0;
+		break;
+	}
+
 	default:
 		;
 	}
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 8367c15..56ef826 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -415,6 +415,7 @@ struct kvm_trace_rec {
 #endif
 /* Another bug in KVM_SET_USER_MEMORY_REGION fixed: */
 #define KVM_CAP_JOIN_MEMORY_REGIONS_WORKS 30
+#define KVM_CAP_ADJUST_CLOCK 39
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -500,6 +501,15 @@ struct kvm_irq_routing {
 			_IOW(KVMIO, 0x74, struct kvm_assigned_msix_entry)
 #define KVM_DEASSIGN_DEV_IRQ       _IOW(KVMIO, 0x75, struct kvm_assigned_irq)
 
+struct kvm_clock_data {
+	__u64 clock;
+	__u32 flags;
+	__u32 pad[9];
+};
+
+#define KVM_SET_CLOCK             _IOW(KVMIO, 0x7b, struct kvm_clock_data)
+#define KVM_GET_CLOCK             _IOR(KVMIO, 0x7c, struct kvm_clock_data)
+
 /*
  * ioctls for vcpu fds
  */
-- 
1.6.3.rc4.29.g8146