From 30a80f8d4d6777b1eb2ef961097037b64bc434ec Mon Sep 17 00:00:00 2001
From: Glauber Costa <glommer@redhat.com>
Date: Wed, 9 Feb 2011 17:56:47 -0200
Subject: [RHEL6 qemu-kvm PATCH 01/11] kvm: x86: Introduce kvmclock device to save/restore its state
RH-Author: Glauber Costa <glommer@redhat.com>
Message-id: <1297274207-9911-1-git-send-email-glommer@redhat.com>
Patchwork-id: 17930
O-Subject: [PATCH RHEL6.1] kvm: x86: Introduce kvmclock device to save/restore its state
Bugzilla: 658467
RH-Acked-by: Michael S. Tsirkin <mst@redhat.com>
RH-Acked-by: Juan Quintela <quintela@redhat.com>
RH-Acked-by: Markus Armbruster <armbru@redhat.com>
RH-Acked-by: Rik van Riel <riel@redhat.com>
RH-Author: Glauber Costa
RH-Bugzilla: 658467
RH-Upstream-Status: Posted
If kvmclock is used, which implies the kernel supports it, register a
kvmclock device with the sysbus. Its main purpose is to save and restore
the kernel state on migration, but this will also allow to visualize it
one day.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Glauber Costa <glommer@redhat.com>
---
Upstream diff: I decided not to register a new qdev device for kvmclock,
in order not to complicate our migration paths. Alternatively, I just got
the code motion to hw/kvmclock.c, and the pre_save function.
We also do not have the CONFIG_KVM_PARA switch in our tree.
---
Makefile.target | 3 +-
hw/kvmclock.c | 102 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
hw/pc.c | 3 ++
qemu-kvm-x86.c | 35 -------------------
4 files changed, 107 insertions(+), 36 deletions(-)
create mode 100644 hw/kvmclock.c
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
Makefile.target | 3 +-
hw/kvmclock.c | 102 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
hw/pc.c | 3 ++
qemu-kvm-x86.c | 35 -------------------
4 files changed, 107 insertions(+), 36 deletions(-)
create mode 100644 hw/kvmclock.c
diff --git a/Makefile.target b/Makefile.target
index a303fdb..93bff38c 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -29,7 +29,7 @@ PROGS=$(QEMU_PROG)
LIBS+=-lm
-kvm.o kvm-all.o: QEMU_CFLAGS+=$(KVM_CFLAGS)
+kvm.o kvm-all.o kvmclock.o: QEMU_CFLAGS+=$(KVM_CFLAGS)
CFLAGS += $(KVM_CFLAGS)
@@ -237,6 +237,7 @@ obj-i386-y += device-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += extboot.o
obj-i386-y += ne2000-isa.o
obj-i386-y += testdev.o
+obj-i386-y += kvmclock.o
obj-i386-$(CONFIG_KVM_PIT) += i8254-kvm.o
obj-i386-$(CONFIG_KVM_DEVICE_ASSIGNMENT) += device-assignment.o
diff --git a/hw/kvmclock.c b/hw/kvmclock.c
new file mode 100644
index 0000000..0c85ce0
--- /dev/null
+++ b/hw/kvmclock.c
@@ -0,0 +1,102 @@
+/*
+ * QEMU KVM support, paravirtual clock device
+ *
+ * Copyright (C) 2011 Siemens AG
+ *
+ * Authors:
+ * Jan Kiszka <jan.kiszka@siemens.com>
+ *
+ * This work is licensed under the terms of the GNU GPL version 2.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "sysemu.h"
+#include "sysbus.h"
+#include "kvm.h"
+#include "kvmclock.h"
+
+#if defined(KVM_CAP_ADJUST_CLOCK)
+
+#include <linux/kvm.h>
+#include <linux/kvm_para.h>
+
+typedef struct KVMClockState {
+ uint64_t clock;
+ bool clock_valid;
+} KVMClockState;
+
+static KVMClockState kvmclock_state;
+
+static void kvmclock_pre_save(void *opaque)
+{
+ KVMClockState *s = opaque;
+ struct kvm_clock_data data;
+ int ret;
+
+ if (s->clock_valid) {
+ return;
+ }
+
+ ret = kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, &data);
+ if (ret < 0) {
+ fprintf(stderr, "KVM_GET_CLOCK failed: %s\n", strerror(ret));
+ data.clock = 0;
+ }
+ s->clock = data.clock;
+ /*
+ * If the VM is stopped, declare the clock state valid to avoid re-reading
+ * it on next vmsave (which would return a different value). Will be reset
+ * when the VM is continued.
+ */
+ s->clock_valid = !vm_running;
+}
+
+static int kvmclock_post_load(void *opaque, int version_id)
+{
+ KVMClockState *s = opaque;
+ struct kvm_clock_data data;
+
+ data.clock = s->clock;
+ data.flags = 0;
+ return kvm_vm_ioctl(kvm_state, KVM_SET_CLOCK, &data);
+}
+
+static void kvmclock_vm_state_change(void *opaque, int running, int reason)
+{
+ KVMClockState *s = opaque;
+
+ if (running) {
+ s->clock_valid = false;
+ }
+}
+
+static const VMStateDescription kvmclock_vmsd = {
+ .name = "kvmclock",
+ .version_id = 1,
+ .minimum_version_id = 1,
+ .minimum_version_id_old = 1,
+ .pre_save = kvmclock_pre_save,
+ .post_load = kvmclock_post_load,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT64(clock, KVMClockState),
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+/* Note: Must be called after VCPU initialization. */
+void kvmclock_create(void)
+{
+ if (kvm_enabled() && kvm_check_extension(kvm_state, KVM_CAP_ADJUST_CLOCK) &&
+ first_cpu->cpuid_kvm_features & (1ULL << KVM_FEATURE_CLOCKSOURCE)) {
+ vmstate_register(NULL, 0, &kvmclock_vmsd, &kvmclock_state);
+ qemu_add_vm_change_state_handler(kvmclock_vm_state_change, &kvmclock_state);
+ }
+}
+
+#else /* !(KVM_CAP_ADJUST_CLOCK) */
+void kvmclock_create(void)
+{
+}
+#endif /* !(KVM_CAP_ADJUST_CLOCK) */
diff --git a/hw/pc.c b/hw/pc.c
index 18f405a..7a6bef8 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -47,6 +47,7 @@
#include "device-assignment.h"
#include "qemu-kvm.h"
#include "ui/qemu-spice.h"
+#include "kvmclock.h"
/* output Bochs bios info messages */
//#define DEBUG_BIOS
@@ -1098,6 +1099,8 @@ static void pc_init1(ram_addr_t ram_size,
#endif
vmport_init();
+ kvmclock_create();
+
/* allocate RAM */
if (fake_machine) {
/* If user boots with -m 1000 We don't actually want to
diff --git a/qemu-kvm-x86.c b/qemu-kvm-x86.c
index 2d581cc..8d8956d 100644
--- a/qemu-kvm-x86.c
+++ b/qemu-kvm-x86.c
@@ -739,37 +739,6 @@ int kvm_qemu_destroy_memory_alias(uint64_t phys_start)
return kvm_destroy_memory_alias(kvm_context, phys_start);
}
-#ifdef KVM_CAP_ADJUST_CLOCK
-static struct kvm_clock_data kvmclock_data;
-
-static void kvmclock_pre_save(void *opaque)
-{
- struct kvm_clock_data *cl = opaque;
-
- kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, cl);
-}
-
-static int kvmclock_post_load(void *opaque, int version_id)
-{
- struct kvm_clock_data *cl = opaque;
-
- return kvm_vm_ioctl(kvm_state, KVM_SET_CLOCK, cl);
-}
-
-static const VMStateDescription vmstate_kvmclock= {
- .name = "kvmclock",
- .version_id = 1,
- .minimum_version_id = 1,
- .minimum_version_id_old = 1,
- .pre_save = kvmclock_pre_save,
- .post_load = kvmclock_post_load,
- .fields = (VMStateField []) {
- VMSTATE_U64(clock, struct kvm_clock_data),
- VMSTATE_END_OF_LIST()
- }
-};
-#endif
-
int kvm_arch_qemu_create_context(void)
{
int i;
@@ -791,10 +760,6 @@ int kvm_arch_qemu_create_context(void)
kvm_has_vm_hsave_pa = 1;
}
-#ifdef KVM_CAP_ADJUST_CLOCK
- if (kvm_check_extension(kvm_state, KVM_CAP_ADJUST_CLOCK))
- vmstate_register(NULL, 0, &vmstate_kvmclock, &kvmclock_data);
-#endif
return 0;
}
--
1.7.3.2