External/libvirt
1
0
Fork 0
mirror of https://gitlab.com/libvirt/libvirt.git synced 2024-12-22 13:45:38 +00:00
Code Issues Packages Projects Releases Wiki Activity
ba3a5604c9
libvirt/src/qemu/qemu_process.c
Jiri Denemark 66643931e7 qemu: Add support for /dev/userfaultfd 
/dev/userfaultfd device is preferred over userfaultfd syscall for
post-copy migrations. Unless qemu driver is configured to disable mount
namespace or to forbid access to /dev/userfaultfd in cgroup_device_acl,
we will copy it to the limited /dev filesystem QEMU will have access to
and label it appropriately. So in the default configuration post-copy
migration will be allowed even without enabling
vm.unprivileged_userfaultfd sysctl.

Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
2024-02-13 17:44:26 +01:00

9701 lines
296 KiB
C
Raw Blame History

/*
* qemu_process.c: QEMU process management
*
* Copyright (C) 2006-2016 Red Hat, Inc.
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library. If not, see
* <http://www.gnu.org/licenses/>.
*
*/
#include <config.h>
#include <fcntl.h>
#include <unistd.h>
#include <signal.h>
#include <sys/stat.h>
#if defined(__linux__)
# include <linux/capability.h>
#elif defined(__FreeBSD__)
# include <sys/param.h>
# include <sys/cpuset.h>
#endif
#include <sys/utsname.h>
#if WITH_CAPNG
# include <cap-ng.h>
#endif
#include "qemu_process.h"
#define LIBVIRT_QEMU_PROCESSPRIV_H_ALLOW
#include "qemu_processpriv.h"
#include "qemu_alias.h"
#include "qemu_block.h"
#include "qemu_domain.h"
#include "qemu_domain_address.h"
#include "qemu_namespace.h"
#include "qemu_cgroup.h"
#include "qemu_capabilities.h"
#include "qemu_monitor.h"
#include "qemu_command.h"
#include "qemu_hostdev.h"
#include "qemu_hotplug.h"
#include "qemu_migration.h"
#include "qemu_migration_params.h"
#include "qemu_interface.h"
#include "qemu_security.h"
#include "qemu_extdevice.h"
#include "qemu_firmware.h"
#include "qemu_backup.h"
#include "qemu_dbus.h"
#include "qemu_snapshot.h"
#include "cpu/cpu.h"
#include "cpu/cpu_x86.h"
#include "datatypes.h"
#include "virlog.h"
#include "virerror.h"
#include "viralloc.h"
#include "virhook.h"
#include "virfile.h"
#include "virpidfile.h"
#include "virhostcpu.h"
#include "domain_audit.h"
#include "domain_cgroup.h"
#include "domain_interface.h"
#include "domain_nwfilter.h"
#include "domain_postparse.h"
#include "domain_validate.h"
#include "locking/domain_lock.h"
#include "viruuid.h"
#include "virprocess.h"
#include "virtime.h"
#include "virnetdevtap.h"
#include "virnetdevopenvswitch.h"
#include "virnetdevmidonet.h"
#include "virbitmap.h"
#include "virnuma.h"
#include "virstring.h"
#include "virhostdev.h"
#include "configmake.h"
#include "netdev_bandwidth_conf.h"
#include "virresctrl.h"
#include "virvsock.h"
#include "viridentity.h"
#include "virthreadjob.h"
#include "virutil.h"
#include "storage_source.h"
#include "backup_conf.h"
#include "logging/log_manager.h"
#include "logging/log_protocol.h"
#define VIR_FROM_THIS VIR_FROM_QEMU
VIR_LOG_INIT("qemu.qemu_process");
/**
* qemuProcessRemoveDomainStatus
*
* remove all state files of a domain from statedir
*/
static void
qemuProcessRemoveDomainStatus(virQEMUDriver *driver,
virDomainObj *vm)
{
g_autofree char *file = NULL;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
file = g_strdup_printf("%s/%s.xml", cfg->stateDir, vm->def->name);
if (unlink(file) < 0 && errno != ENOENT && errno != ENOTDIR)
VIR_WARN("Failed to remove domain XML for %s: %s",
vm->def->name, g_strerror(errno));
if (priv->pidfile &&
unlink(priv->pidfile) < 0 &&
errno != ENOENT)
VIR_WARN("Failed to remove PID file for %s: %s",
vm->def->name, g_strerror(errno));
}
/*
* This is a callback registered with a qemuAgent *instance,
* and to be invoked when the agent console hits an end of file
* condition, or error, thus indicating VM shutdown should be
* performed
*/
static void
qemuProcessHandleAgentEOF(qemuAgent *agent,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv;
virObjectLock(vm);
VIR_DEBUG("Received EOF from agent on %p '%s'", vm, vm->def->name);
priv = vm->privateData;
if (!priv->agent) {
VIR_DEBUG("Agent freed already");
goto unlock;
}
if (priv->beingDestroyed) {
VIR_DEBUG("Domain is being destroyed, agent EOF is expected");
goto unlock;
}
qemuAgentClose(agent);
priv->agent = NULL;
priv->agentError = false;
virObjectUnlock(vm);
return;
unlock:
virObjectUnlock(vm);
return;
}
/*
* This is invoked when there is some kind of error
* parsing data to/from the agent. The VM can continue
* to run, but no further agent commands will be
* allowed
*/
static void
qemuProcessHandleAgentError(qemuAgent *agent G_GNUC_UNUSED,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv;
virObjectLock(vm);
VIR_DEBUG("Received error from agent on %p '%s'", vm, vm->def->name);
priv = vm->privateData;
priv->agentError = true;
virObjectUnlock(vm);
}
static qemuAgentCallbacks agentCallbacks = {
.eofNotify = qemuProcessHandleAgentEOF,
.errorNotify = qemuProcessHandleAgentError,
};
int
qemuConnectAgent(virQEMUDriver *driver, virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
qemuAgent *agent = NULL;
virDomainChrDef *config = qemuFindAgentConfig(vm->def);
if (!config)
return 0;
if (priv->agent)
return 0;
if (config->state != VIR_DOMAIN_CHR_DEVICE_STATE_CONNECTED) {
VIR_DEBUG("Deferring connecting to guest agent");
return 0;
}
if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def) < 0) {
VIR_ERROR(_("Failed to set security context for agent for %1$s"),
vm->def->name);
goto cleanup;
}
agent = qemuAgentOpen(vm,
config->source,
virEventThreadGetContext(priv->eventThread),
&agentCallbacks);
if (!virDomainObjIsActive(vm)) {
qemuAgentClose(agent);
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("guest crashed while connecting to the guest agent"));
return -1;
}
if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0) {
VIR_ERROR(_("Failed to clear security context for agent for %1$s"),
vm->def->name);
qemuAgentClose(agent);
goto cleanup;
}
priv->agent = agent;
if (!priv->agent)
VIR_INFO("Failed to connect agent for %s", vm->def->name);
cleanup:
if (!priv->agent) {
VIR_WARN("Cannot connect to QEMU guest agent for %s", vm->def->name);
priv->agentError = true;
virResetLastError();
}
return 0;
}
/**
* qemuProcessEventSubmit:
* @vm: pointer to the domain object, the function will take an extra reference
* @eventType: the event to be processed
* @action: event specific action to be taken
* @status: event specific status
* @data: additional data for the event processor (the pointer is stolen and it
* will be properly freed
*
* Submits @eventType to be processed by the asynchronous event handling thread.
*/
static void
qemuProcessEventSubmit(virDomainObj *vm,
qemuProcessEventType eventType,
int action,
int status,
void *data)
{
struct qemuProcessEvent *event = g_new0(struct qemuProcessEvent, 1);
virQEMUDriver *driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event->vm = virObjectRef(vm);
event->eventType = eventType;
event->action = action;
event->status = status;
event->data = data;
if (virThreadPoolSendJob(driver->workerPool, 0, event) < 0) {
virObjectUnref(event->vm);
qemuProcessEventFree(event);
}
}
/*
* This is a callback registered with a qemuMonitor *instance,
* and to be invoked when the monitor console hits an end of file
* condition, or error, thus indicating VM shutdown should be
* performed
*/
static void
qemuProcessHandleMonitorEOF(qemuMonitor *mon,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv;
virObjectLock(vm);
VIR_DEBUG("Received EOF on %p '%s'", vm, vm->def->name);
priv = vm->privateData;
if (priv->beingDestroyed) {
VIR_DEBUG("Domain is being destroyed, EOF is expected");
goto cleanup;
}
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_MONITOR_EOF,
0, 0, GINT_TO_POINTER(vm->def->id));
/* We don't want this EOF handler to be called over and over while the
* thread is waiting for a job.
*/
virObjectLock(mon);
qemuMonitorUnregister(mon);
virObjectUnlock(mon);
/* We don't want any cleanup from EOF handler (or any other
* thread) to enter qemu namespace. */
qemuDomainDestroyNamespace(priv->driver, vm);
cleanup:
virObjectUnlock(vm);
}
/*
* This is invoked when there is some kind of error
* parsing data to/from the monitor. The VM can continue
* to run, but no further monitor commands will be
* allowed
*/
static void
qemuProcessHandleMonitorError(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv;
virObjectEvent *event = NULL;
virObjectLock(vm);
VIR_DEBUG("Received error on %p '%s'", vm, vm->def->name);
priv = vm->privateData;
priv->monError = true;
event = virDomainEventControlErrorNewFromObj(vm);
virObjectEventStateQueue(priv->driver->domainEventState, event);
virObjectUnlock(vm);
}
/**
* qemuProcessFindDomainDiskByAliasOrQOM:
* @vm: domain object to search for the disk
* @alias: -drive or -device alias of the disk
* @qomid: QOM tree device name
*
* Looks up a disk in the domain definition of @vm which either matches the
* -drive or -device alias used for the backend and frontend respectively or the
* QOM name. If @alias is empty it's treated as NULL as it's a mandatory field
* in some cases.
*
* Returns a disk from @vm or NULL if it could not be found.
*/
virDomainDiskDef *
qemuProcessFindDomainDiskByAliasOrQOM(virDomainObj *vm,
const char *alias,
const char *qomid)
{
size_t i;
if (alias && *alias == '\0')
alias = NULL;
if (alias)
alias = qemuAliasDiskDriveSkipPrefix(alias);
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
qemuDomainDiskPrivate *diskPriv = QEMU_DOMAIN_DISK_PRIVATE(disk);
if ((disk->info.alias && STREQ_NULLABLE(disk->info.alias, alias)) ||
(diskPriv->qomName && STREQ_NULLABLE(diskPriv->qomName, qomid)))
return disk;
}
virReportError(VIR_ERR_INTERNAL_ERROR,
_("no disk found with alias '%1$s' or id '%2$s'"),
NULLSTR(alias), NULLSTR(qomid));
return NULL;
}
static void
qemuProcessHandleReset(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
qemuDomainObjPrivate *priv;
virDomainState state;
int reason;
virObjectLock(vm);
priv = vm->privateData;
driver = priv->driver;
state = virDomainObjGetState(vm, &reason);
/* ignore reset events on VM startup. Libvirt in certain instances does a
* reset during startup so that the ACPI tables are re-generated */
if (state == VIR_DOMAIN_PAUSED &&
reason == VIR_DOMAIN_PAUSED_STARTING_UP) {
VIR_DEBUG("ignoring reset event during startup");
goto unlock;
}
event = virDomainEventRebootNewFromObj(vm);
if (priv->agent)
qemuAgentNotifyEvent(priv->agent, QEMU_AGENT_EVENT_RESET);
qemuDomainSetFakeReboot(vm, false);
qemuDomainSaveStatus(vm);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_RESET, 0, 0, NULL);
unlock:
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
/*
* Since we have the '-no-shutdown' flag set, the
* QEMU process will currently have guest OS shutdown
* and the CPUS stopped. To fake the reboot, we thus
* want todo a reset of the virtual hardware, followed
* by restart of the CPUs. This should result in the
* guest OS booting up again
*/
static void
qemuProcessFakeReboot(void *opaque)
{
virDomainObj *vm = opaque;
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUDriver *driver = priv->driver;
virDomainRunningReason reason = VIR_DOMAIN_RUNNING_BOOTED;
int ret = -1, rc;
VIR_DEBUG("vm=%p", vm);
virObjectLock(vm);
if (virDomainObjBeginJob(vm, VIR_JOB_MODIFY) < 0)
goto cleanup;
if (!virDomainObjIsActive(vm)) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("guest unexpectedly quit"));
goto endjob;
}
qemuDomainObjEnterMonitor(vm);
rc = qemuMonitorSystemReset(priv->mon);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
goto endjob;
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_CRASHED)
reason = VIR_DOMAIN_RUNNING_CRASHED;
if (qemuProcessStartCPUs(driver, vm,
reason,
VIR_ASYNC_JOB_NONE) < 0) {
if (virGetLastErrorCode() == VIR_ERR_OK)
virReportError(VIR_ERR_INTERNAL_ERROR,
"%s", _("resume operation failed"));
goto endjob;
}
qemuDomainSaveStatus(vm);
ret = 0;
endjob:
virDomainObjEndJob(vm);
cleanup:
priv->pausedShutdown = false;
qemuDomainSetFakeReboot(vm, false);
if (ret == -1)
ignore_value(qemuProcessKill(vm, VIR_QEMU_PROCESS_KILL_FORCE));
virDomainObjEndAPI(&vm);
}
void
qemuProcessShutdownOrReboot(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
if (priv->fakeReboot ||
vm->def->onPoweroff == VIR_DOMAIN_LIFECYCLE_ACTION_RESTART) {
g_autofree char *name = g_strdup_printf("reboot-%s", vm->def->name);
virThread th;
virObjectRef(vm);
if (virThreadCreateFull(&th,
false,
qemuProcessFakeReboot,
name,
false,
vm) < 0) {
VIR_ERROR(_("Failed to create reboot thread, killing domain"));
ignore_value(qemuProcessKill(vm, VIR_QEMU_PROCESS_KILL_NOWAIT));
priv->pausedShutdown = false;
qemuDomainSetFakeReboot(vm, false);
virObjectUnref(vm);
}
} else {
ignore_value(qemuProcessKill(vm, VIR_QEMU_PROCESS_KILL_NOWAIT));
}
}
static void
qemuProcessHandleEvent(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *eventName,
long long seconds,
unsigned int micros,
const char *details)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
VIR_DEBUG("vm=%p", vm);
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainQemuMonitorEventNew(vm->def->id, vm->def->name,
vm->def->uuid, eventName,
seconds, micros, details);
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleShutdown(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
virTristateBool guest_initiated)
{
virQEMUDriver *driver;
qemuDomainObjPrivate *priv;
virObjectEvent *event = NULL;
int detail = 0;
VIR_DEBUG("vm=%p", vm);
virObjectLock(vm);
priv = vm->privateData;
driver = priv->driver;
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_SHUTDOWN) {
VIR_DEBUG("Ignoring repeated SHUTDOWN event from domain %s",
vm->def->name);
goto unlock;
} else if (!virDomainObjIsActive(vm)) {
VIR_DEBUG("Ignoring SHUTDOWN event from inactive domain %s",
vm->def->name);
goto unlock;
}
/* In case of fake reboot qemu shutdown state is transient so don't
* change domain state nor send events. */
if (!priv->fakeReboot &&
vm->def->onPoweroff != VIR_DOMAIN_LIFECYCLE_ACTION_RESTART) {
VIR_DEBUG("Transitioned guest %s to shutdown state",
vm->def->name);
virDomainObjSetState(vm,
VIR_DOMAIN_SHUTDOWN,
VIR_DOMAIN_SHUTDOWN_UNKNOWN);
switch (guest_initiated) {
case VIR_TRISTATE_BOOL_YES:
detail = VIR_DOMAIN_EVENT_SHUTDOWN_GUEST;
break;
case VIR_TRISTATE_BOOL_NO:
detail = VIR_DOMAIN_EVENT_SHUTDOWN_HOST;
break;
case VIR_TRISTATE_BOOL_ABSENT:
case VIR_TRISTATE_BOOL_LAST:
default:
detail = VIR_DOMAIN_EVENT_SHUTDOWN_FINISHED;
break;
}
event = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_SHUTDOWN,
detail);
qemuDomainSaveStatus(vm);
} else {
priv->pausedShutdown = true;
}
if (priv->agent)
qemuAgentNotifyEvent(priv->agent, QEMU_AGENT_EVENT_SHUTDOWN);
qemuProcessShutdownOrReboot(vm);
unlock:
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleStop(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virDomainPausedReason reason;
virDomainEventSuspendedDetailType detail;
qemuDomainObjPrivate *priv = vm->privateData;
virObjectLock(vm);
driver = priv->driver;
reason = priv->pausedReason;
priv->pausedReason = VIR_DOMAIN_PAUSED_UNKNOWN;
/* In case of fake reboot qemu paused state is transient so don't
* reveal it in domain state nor sent events */
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING &&
!priv->pausedShutdown) {
if (vm->job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT) {
if (vm->job->current->status == VIR_DOMAIN_JOB_STATUS_POSTCOPY)
reason = VIR_DOMAIN_PAUSED_POSTCOPY;
else
reason = VIR_DOMAIN_PAUSED_MIGRATION;
}
detail = qemuDomainPausedReasonToSuspendedEvent(reason);
VIR_DEBUG("Transitioned guest %s to paused state, "
"reason %s, event detail %d",
vm->def->name, virDomainPausedReasonTypeToString(reason),
detail);
if (vm->job->current)
ignore_value(virTimeMillisNow(&vm->job->current->stopped));
if (priv->signalStop)
virDomainObjBroadcast(vm);
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, reason);
event = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_SUSPENDED,
detail);
VIR_FREE(priv->lockState);
if (virDomainLockProcessPause(driver->lockManager, vm, &priv->lockState) < 0)
VIR_WARN("Unable to release lease on %s", vm->def->name);
VIR_DEBUG("Preserving lock state '%s'", NULLSTR(priv->lockState));
qemuDomainSaveStatus(vm);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleResume(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
qemuDomainObjPrivate *priv;
virDomainRunningReason reason = VIR_DOMAIN_RUNNING_UNPAUSED;
virDomainEventResumedDetailType eventDetail;
virObjectLock(vm);
priv = vm->privateData;
driver = priv->driver;
if (priv->runningReason != VIR_DOMAIN_RUNNING_UNKNOWN) {
reason = priv->runningReason;
priv->runningReason = VIR_DOMAIN_RUNNING_UNKNOWN;
}
if (virDomainObjGetState(vm, NULL) != VIR_DOMAIN_RUNNING) {
eventDetail = qemuDomainRunningReasonToResumeEvent(reason);
VIR_DEBUG("Transitioned guest %s into running state, reason '%s', "
"event detail %d",
vm->def->name, virDomainRunningReasonTypeToString(reason),
eventDetail);
/* When a domain is running in (failed) post-copy migration on the
* destination host, we need to make sure to set the appropriate reason
* here. */
if (virDomainObjIsPostcopy(vm, vm->job)) {
if (virDomainObjIsFailedPostcopy(vm, vm->job))
reason = VIR_DOMAIN_RUNNING_POSTCOPY_FAILED;
else
reason = VIR_DOMAIN_RUNNING_POSTCOPY;
}
virDomainObjSetState(vm, VIR_DOMAIN_RUNNING, reason);
event = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_RESUMED,
eventDetail);
qemuDomainSaveStatus(vm);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleRTCChange(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
long long offset)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
if (vm->def->clock.offset == VIR_DOMAIN_CLOCK_OFFSET_VARIABLE) {
/* when a basedate is manually given on the qemu commandline
* rather than simply "-rtc base=utc", the offset sent by qemu
* in this event is *not* the new offset from UTC, but is
* instead the new offset from the *original basedate* +
* uptime. For example, if the original offset was 3600 and
* the guest clock has been advanced by 10 seconds, qemu will
* send "10" in the event - this means that the new offset
* from UTC is 3610, *not* 10. If the guest clock is advanced
* by another 10 seconds, qemu will now send "20" - i.e. each
* event is the sum of the most recent change and all previous
* changes since the domain was started. Fortunately, we have
* saved the initial offset in "adjustment0", so to arrive at
* the proper new "adjustment", we just add the most recent
* offset to adjustment0.
*/
offset += vm->def->clock.data.variable.adjustment0;
vm->def->clock.data.variable.adjustment = offset;
qemuDomainSaveStatus(vm);
}
event = virDomainEventRTCChangeNewFromObj(vm, offset);
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleWatchdog(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
int action)
{
virQEMUDriver *driver;
virObjectEvent *watchdogEvent = NULL;
virObjectEvent *lifecycleEvent = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
watchdogEvent = virDomainEventWatchdogNewFromObj(vm, action);
if (action == VIR_DOMAIN_EVENT_WATCHDOG_PAUSE &&
virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
qemuDomainObjPrivate *priv = vm->privateData;
VIR_DEBUG("Transitioned guest %s to paused state due to watchdog", vm->def->name);
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_WATCHDOG);
lifecycleEvent = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_SUSPENDED,
VIR_DOMAIN_EVENT_SUSPENDED_WATCHDOG);
VIR_FREE(priv->lockState);
if (virDomainLockProcessPause(driver->lockManager, vm, &priv->lockState) < 0)
VIR_WARN("Unable to release lease on %s", vm->def->name);
VIR_DEBUG("Preserving lock state '%s'", NULLSTR(priv->lockState));
qemuDomainSaveStatus(vm);
}
if (vm->def->nwatchdogs &&
vm->def->watchdogs[0]->action == VIR_DOMAIN_WATCHDOG_ACTION_DUMP) {
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_WATCHDOG,
VIR_DOMAIN_WATCHDOG_ACTION_DUMP, 0, NULL);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, watchdogEvent);
virObjectEventStateQueue(driver->domainEventState, lifecycleEvent);
}
static void
qemuProcessHandleIOError(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *diskAlias,
const char *nodename,
int action,
const char *reason)
{
virQEMUDriver *driver;
virObjectEvent *ioErrorEvent = NULL;
virObjectEvent *ioErrorEvent2 = NULL;
virObjectEvent *lifecycleEvent = NULL;
const char *srcPath;
const char *devAlias;
virDomainDiskDef *disk;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
if (*diskAlias == '\0')
diskAlias = NULL;
if (diskAlias)
disk = qemuProcessFindDomainDiskByAliasOrQOM(vm, diskAlias, NULL);
else if (nodename)
disk = qemuDomainDiskLookupByNodename(vm->def, NULL, nodename, NULL);
else
disk = NULL;
if (disk) {
srcPath = virDomainDiskGetSource(disk);
devAlias = disk->info.alias;
} else {
srcPath = "";
devAlias = "";
}
ioErrorEvent = virDomainEventIOErrorNewFromObj(vm, srcPath, devAlias, action);
ioErrorEvent2 = virDomainEventIOErrorReasonNewFromObj(vm, srcPath, devAlias, action, reason);
if (action == VIR_DOMAIN_EVENT_IO_ERROR_PAUSE &&
virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
qemuDomainObjPrivate *priv = vm->privateData;
VIR_DEBUG("Transitioned guest %s to paused state due to IO error", vm->def->name);
if (priv->signalIOError)
virDomainObjBroadcast(vm);
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_IOERROR);
lifecycleEvent = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_SUSPENDED,
VIR_DOMAIN_EVENT_SUSPENDED_IOERROR);
VIR_FREE(priv->lockState);
if (virDomainLockProcessPause(driver->lockManager, vm, &priv->lockState) < 0)
VIR_WARN("Unable to release lease on %s", vm->def->name);
VIR_DEBUG("Preserving lock state '%s'", NULLSTR(priv->lockState));
qemuDomainSaveStatus(vm);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, ioErrorEvent);
virObjectEventStateQueue(driver->domainEventState, ioErrorEvent2);
virObjectEventStateQueue(driver->domainEventState, lifecycleEvent);
}
static void
qemuProcessHandleJobStatusChange(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *jobname,
int status)
{
qemuDomainObjPrivate *priv;
qemuBlockJobData *job = NULL;
int jobnewstate;
virObjectLock(vm);
priv = vm->privateData;
VIR_DEBUG("job '%s'(domain: %p,%s) state changed to '%s'(%d)",
jobname, vm, vm->def->name,
qemuMonitorJobStatusTypeToString(status), status);
if ((jobnewstate = qemuBlockjobConvertMonitorStatus(status)) == QEMU_BLOCKJOB_STATE_LAST)
goto cleanup;
if (!(job = virHashLookup(priv->blockjobs, jobname))) {
VIR_DEBUG("job '%s' not registered", jobname);
goto cleanup;
}
job->newstate = jobnewstate;
if (job->synchronous) {
VIR_DEBUG("job '%s' handled synchronously", jobname);
virDomainObjBroadcast(vm);
} else {
VIR_DEBUG("job '%s' handled by event thread", jobname);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_JOB_STATUS_CHANGE,
0, 0, virObjectRef(job));
}
cleanup:
virObjectUnlock(vm);
}
static void
qemuProcessHandleGraphics(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
int phase,
int localFamily,
const char *localNode,
const char *localService,
int remoteFamily,
const char *remoteNode,
const char *remoteService,
const char *authScheme,
const char *x509dname,
const char *saslUsername)
{
virQEMUDriver *driver;
virObjectEvent *event;
virDomainEventGraphicsAddressPtr localAddr = NULL;
virDomainEventGraphicsAddressPtr remoteAddr = NULL;
virDomainEventGraphicsSubjectPtr subject = NULL;
localAddr = g_new0(virDomainEventGraphicsAddress, 1);
localAddr->family = localFamily;
localAddr->service = g_strdup(localService);
localAddr->node = g_strdup(localNode);
remoteAddr = g_new0(virDomainEventGraphicsAddress, 1);
remoteAddr->family = remoteFamily;
remoteAddr->service = g_strdup(remoteService);
remoteAddr->node = g_strdup(remoteNode);
subject = g_new0(virDomainEventGraphicsSubject, 1);
if (x509dname) {
VIR_REALLOC_N(subject->identities, subject->nidentity+1);
subject->nidentity++;
subject->identities[subject->nidentity - 1].type = g_strdup("x509dname");
subject->identities[subject->nidentity - 1].name = g_strdup(x509dname);
}
if (saslUsername) {
VIR_REALLOC_N(subject->identities, subject->nidentity+1);
subject->nidentity++;
subject->identities[subject->nidentity - 1].type = g_strdup("saslUsername");
subject->identities[subject->nidentity - 1].name = g_strdup(saslUsername);
}
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainEventGraphicsNewFromObj(vm, phase, localAddr, remoteAddr, authScheme, subject);
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleTrayChange(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias,
const char *devid,
int reason)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virDomainDiskDef *disk;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
disk = qemuProcessFindDomainDiskByAliasOrQOM(vm, devAlias, devid);
if (disk) {
event = virDomainEventTrayChangeNewFromObj(vm, disk->info.alias, reason);
/* Update disk tray status */
if (reason == VIR_DOMAIN_EVENT_TRAY_CHANGE_OPEN)
disk->tray_status = VIR_DOMAIN_DISK_TRAY_OPEN;
else if (reason == VIR_DOMAIN_EVENT_TRAY_CHANGE_CLOSE)
disk->tray_status = VIR_DOMAIN_DISK_TRAY_CLOSED;
qemuDomainSaveStatus(vm);
virDomainObjBroadcast(vm);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandlePMWakeup(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectEvent *lifecycleEvent = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainEventPMWakeupNewFromObj(vm);
/* Don't set domain status back to running if it wasn't paused
* from guest side, otherwise it can just cause confusion.
*/
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_PMSUSPENDED) {
VIR_DEBUG("Transitioned guest %s from pmsuspended to running "
"state due to QMP wakeup event", vm->def->name);
virDomainObjSetState(vm, VIR_DOMAIN_RUNNING,
VIR_DOMAIN_RUNNING_WAKEUP);
lifecycleEvent = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_STARTED,
VIR_DOMAIN_EVENT_STARTED_WAKEUP);
qemuDomainSaveStatus(vm);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
virObjectEventStateQueue(driver->domainEventState, lifecycleEvent);
}
static void
qemuProcessHandlePMSuspend(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectEvent *lifecycleEvent = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainEventPMSuspendNewFromObj(vm);
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
qemuDomainObjPrivate *priv = vm->privateData;
VIR_DEBUG("Transitioned guest %s to pmsuspended state due to "
"QMP suspend event", vm->def->name);
virDomainObjSetState(vm, VIR_DOMAIN_PMSUSPENDED,
VIR_DOMAIN_PMSUSPENDED_UNKNOWN);
lifecycleEvent =
virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_PMSUSPENDED,
VIR_DOMAIN_EVENT_PMSUSPENDED_MEMORY);
qemuDomainSaveStatus(vm);
if (priv->agent)
qemuAgentNotifyEvent(priv->agent, QEMU_AGENT_EVENT_SUSPEND);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
virObjectEventStateQueue(driver->domainEventState, lifecycleEvent);
}
static void
qemuProcessHandleBalloonChange(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
unsigned long long actual)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
size_t i;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainEventBalloonChangeNewFromObj(vm, actual);
/* We want the balloon size stored in domain definition to
* account for the actual size of virtio-mem too. But the
* balloon size as reported by QEMU (@actual) contains just
* the balloon size without any virtio-mem. Do a wee bit of
* math to fix it. */
VIR_DEBUG("balloon size before fix is %lld", actual);
for (i = 0; i < vm->def->nmems; i++) {
if (vm->def->mems[i]->model == VIR_DOMAIN_MEMORY_MODEL_VIRTIO_MEM)
actual += vm->def->mems[i]->target.virtio_mem.currentsize;
}
VIR_DEBUG("Updating balloon from %lld to %lld kb",
vm->def->mem.cur_balloon, actual);
vm->def->mem.cur_balloon = actual;
qemuDomainSaveStatus(vm);
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandlePMSuspendDisk(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectEvent *lifecycleEvent = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
event = virDomainEventPMSuspendDiskNewFromObj(vm);
if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) {
qemuDomainObjPrivate *priv = vm->privateData;
VIR_DEBUG("Transitioned guest %s to pmsuspended state due to "
"QMP suspend_disk event", vm->def->name);
virDomainObjSetState(vm, VIR_DOMAIN_PMSUSPENDED,
VIR_DOMAIN_PMSUSPENDED_UNKNOWN);
lifecycleEvent =
virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_PMSUSPENDED,
VIR_DOMAIN_EVENT_PMSUSPENDED_DISK);
qemuDomainSaveStatus(vm);
if (priv->agent)
qemuAgentNotifyEvent(priv->agent, QEMU_AGENT_EVENT_SUSPEND);
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
virObjectEventStateQueue(driver->domainEventState, lifecycleEvent);
}
static void
qemuProcessHandleGuestPanic(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
qemuMonitorEventPanicInfo *info)
{
virObjectLock(vm);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_GUESTPANIC,
vm->def->onCrash, 0, info);
virObjectUnlock(vm);
}
void
qemuProcessHandleDeviceDeleted(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias)
{
virObjectLock(vm);
VIR_DEBUG("Device %s removed from domain %p %s",
devAlias, vm, vm->def->name);
if (qemuDomainSignalDeviceRemoval(vm, devAlias,
QEMU_DOMAIN_UNPLUGGING_DEVICE_STATUS_OK))
goto cleanup;
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_DEVICE_DELETED,
0, 0, g_strdup(devAlias));
cleanup:
virObjectUnlock(vm);
}
static void
qemuProcessHandleDeviceUnplugErr(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devPath,
const char *devAlias)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
VIR_DEBUG("Device %s QOM path %s failed to be removed from domain %p %s",
devAlias, devPath, vm, vm->def->name);
/*
* DEVICE_UNPLUG_GUEST_ERROR will always contain the QOM path
* but QEMU will not guarantee that devAlias will be provided.
*
* However, given that all Libvirt devices have a devAlias, we
* can ignore the case where QEMU emitted this event without it.
*/
if (!devAlias)
goto cleanup;
qemuDomainSignalDeviceRemoval(vm, devAlias,
QEMU_DOMAIN_UNPLUGGING_DEVICE_STATUS_GUEST_REJECTED);
event = virDomainEventDeviceRemovalFailedNewFromObj(vm, devAlias);
cleanup:
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
/**
*
* Meaning of fields reported by the event according to the ACPI standard:
* @source:
* 0x00 - 0xff: Notification values, as passed at the request time
* 0x100: Operating System Shutdown Processing
* 0x103: Ejection processing
* 0x200: Insertion processing
* other values are reserved
*
* @status:
* general values
* 0x00: success
* 0x01: non-specific failure
* 0x02: unrecognized notify code
* 0x03 - 0x7f: reserved
* other values are specific to the notification type (see below)
*
* for the 0x100 source the following additional codes are standardized:
* 0x80: OS Shutdown request denied
* 0x81: OS Shutdown in progress
* 0x82: OS Shutdown completed
* 0x83: OS Graceful shutdown not supported
* other higher values are reserved
*
* for the 0x003 (Ejection request) and 0x103 (Ejection processing) source
* the following additional codes are standardized:
* 0x80: Device ejection not supported by OSPM
* 0x81: Device in use by application
* 0x82: Device Busy
* 0x83: Ejection dependency is busy or not supported for ejection by OSPM
* 0x84: Ejection is in progress (pending)
* other higher values are reserved
*
* for the 0x200 source the following additional codes are standardized:
* 0x80: Device insertion in progress (pending)
* 0x81: Device driver load failure
* 0x82: Device insertion not supported by OSPM
* 0x83-0x8F: Reserved
* 0x90-0x9F: Insertion failure - Resources Unavailable as described by the
* following bit encodings:
* Bit [3]: Bus or Segment Numbers
* Bit [2]: Interrupts
* Bit [1]: I/O
* Bit [0]: Memory
* other higher values are reserved
*
* Other fields and semantics are specific to the qemu handling of the event.
* - @alias may be NULL for successful unplug operations
* - @slotType describes the device type a bit more closely, currently the
* only known value is 'DIMM'
* - @slot describes the specific device
*
* Note that qemu does not emit the event for all the documented sources or
* devices.
*/
static void
qemuProcessHandleAcpiOstInfo(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *alias,
const char *slotType,
const char *slot,
unsigned int source,
unsigned int status)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
VIR_DEBUG("ACPI OST info for device %s domain %p %s. "
"slotType='%s' slot='%s' source=%u status=%u",
NULLSTR(alias), vm, vm->def->name, slotType, slot, source, status);
if (!alias)
goto cleanup;
if (STREQ(slotType, "DIMM")) {
if ((source == 0x003 || source == 0x103) &&
(status == 0x01 || (status >= 0x80 && status <= 0x83))) {
qemuDomainSignalDeviceRemoval(vm, alias,
QEMU_DOMAIN_UNPLUGGING_DEVICE_STATUS_GUEST_REJECTED);
event = virDomainEventDeviceRemovalFailedNewFromObj(vm, alias);
}
}
cleanup:
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleBlockThreshold(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *nodename,
unsigned long long threshold,
unsigned long long excess)
{
qemuDomainObjPrivate *priv;
virQEMUDriver *driver;
virObjectEvent *eventSource = NULL;
virObjectEvent *eventDevice = NULL;
virDomainDiskDef *disk;
virStorageSource *src;
const char *path = NULL;
virObjectLock(vm);
priv = vm->privateData;
driver = priv->driver;
VIR_DEBUG("BLOCK_WRITE_THRESHOLD event for block node '%s' in domain %p %s:"
"threshold '%llu' exceeded by '%llu'",
nodename, vm, vm->def->name, threshold, excess);
if ((disk = qemuDomainDiskLookupByNodename(vm->def, priv->backup, nodename, &src))) {
if (virStorageSourceIsLocalStorage(src))
path = src->path;
if (src == disk->src &&
!src->thresholdEventWithIndex) {
g_autofree char *dev = qemuDomainDiskBackingStoreGetName(disk, 0);
eventDevice = virDomainEventBlockThresholdNewFromObj(vm, dev, path,
threshold, excess);
}
if (src->id != 0) {
g_autofree char *dev = qemuDomainDiskBackingStoreGetName(disk, src->id);
eventSource = virDomainEventBlockThresholdNewFromObj(vm, dev, path,
threshold, excess);
}
}
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, eventDevice);
virObjectEventStateQueue(driver->domainEventState, eventSource);
}
static void
qemuProcessHandleNetdevStreamDisconnected(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias)
{
virObjectLock(vm);
VIR_DEBUG("Device %s Netdev Stream Disconnected in domain %p %s",
devAlias, vm, vm->def->name);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED,
0, 0, g_strdup(devAlias));
virObjectUnlock(vm);
}
static void
qemuProcessHandleNicRxFilterChanged(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias)
{
virObjectLock(vm);
VIR_DEBUG("Device %s RX Filter changed in domain %p %s",
devAlias, vm, vm->def->name);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED,
0, 0, g_strdup(devAlias));
virObjectUnlock(vm);
}
static void
qemuProcessHandleSerialChanged(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias,
bool connected)
{
virObjectLock(vm);
VIR_DEBUG("Serial port %s state changed to '%d' in domain %p %s",
devAlias, connected, vm, vm->def->name);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_SERIAL_CHANGED,
connected, 0, g_strdup(devAlias));
virObjectUnlock(vm);
}
static void
qemuProcessHandleSpiceMigrated(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
qemuDomainJobPrivate *jobPriv;
virObjectLock(vm);
VIR_DEBUG("Spice migration completed for domain %p %s",
vm, vm->def->name);
jobPriv = vm->job->privateData;
if (vm->job->asyncJob != VIR_ASYNC_JOB_MIGRATION_OUT) {
VIR_DEBUG("got SPICE_MIGRATE_COMPLETED event without a migration job");
goto cleanup;
}
jobPriv->spiceMigrated = true;
virDomainObjBroadcast(vm);
cleanup:
virObjectUnlock(vm);
}
static void
qemuProcessHandleMigrationStatus(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
int status)
{
qemuDomainObjPrivate *priv;
qemuDomainJobDataPrivate *privJob = NULL;
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virDomainState state;
int reason;
virObjectLock(vm);
VIR_DEBUG("Migration of domain %p %s changed state to %s",
vm, vm->def->name,
qemuMonitorMigrationStatusTypeToString(status));
priv = vm->privateData;
driver = priv->driver;
if (vm->job->asyncJob == VIR_ASYNC_JOB_NONE) {
VIR_DEBUG("got MIGRATION event without a migration job");
goto cleanup;
}
privJob = vm->job->current->privateData;
privJob->stats.mig.status = status;
virDomainObjBroadcast(vm);
state = virDomainObjGetState(vm, &reason);
switch ((qemuMonitorMigrationStatus) status) {
case QEMU_MONITOR_MIGRATION_STATUS_POSTCOPY:
if (vm->job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT &&
state == VIR_DOMAIN_PAUSED &&
reason == VIR_DOMAIN_PAUSED_MIGRATION) {
VIR_DEBUG("Correcting paused state reason for domain %s to %s",
vm->def->name,
virDomainPausedReasonTypeToString(VIR_DOMAIN_PAUSED_POSTCOPY));
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_POSTCOPY);
event = virDomainEventLifecycleNewFromObj(vm,
VIR_DOMAIN_EVENT_SUSPENDED,
VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY);
qemuDomainSaveStatus(vm);
}
break;
case QEMU_MONITOR_MIGRATION_STATUS_POSTCOPY_PAUSED:
if (vm->job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT &&
state == VIR_DOMAIN_PAUSED) {
/* At this point no thread is watching the migration progress on
* the source as it is just waiting for the Finish phase to end.
* Thus we need to handle the event here. */
qemuMigrationSrcPostcopyFailed(vm);
qemuDomainSaveStatus(vm);
}
break;
case QEMU_MONITOR_MIGRATION_STATUS_POSTCOPY_RECOVER:
if (virDomainObjIsFailedPostcopy(vm, vm->job)) {
int eventType = -1;
int eventDetail = -1;
if (state == VIR_DOMAIN_PAUSED) {
reason = VIR_DOMAIN_PAUSED_POSTCOPY;
eventType = VIR_DOMAIN_EVENT_SUSPENDED;
eventDetail = qemuDomainPausedReasonToSuspendedEvent(reason);
} else {
reason = VIR_DOMAIN_RUNNING_POSTCOPY;
eventType = VIR_DOMAIN_EVENT_RESUMED;
eventDetail = qemuDomainRunningReasonToResumeEvent(reason);
}
VIR_DEBUG("Post-copy migration recovered; correcting state for domain '%s' to %s/%s",
vm->def->name,
virDomainStateTypeToString(state),
NULLSTR(virDomainStateReasonToString(state, reason)));
vm->job->asyncPaused = false;
virDomainObjSetState(vm, state, reason);
event = virDomainEventLifecycleNewFromObj(vm, eventType, eventDetail);
qemuDomainSaveStatus(vm);
}
break;
case QEMU_MONITOR_MIGRATION_STATUS_COMPLETED:
/* A post-copy migration marked as failed when reconnecting to a domain
* with running migration may actually still be running, but we're not
* watching it in any thread. Let's make sure the migration is properly
* finished in case we get a "completed" event.
*/
if (virDomainObjIsPostcopy(vm, vm->job) &&
vm->job->phase == QEMU_MIGRATION_PHASE_POSTCOPY_FAILED &&
vm->job->asyncOwner == 0) {
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_UNATTENDED_MIGRATION,
vm->job->asyncJob, status, NULL);
}
break;
case QEMU_MONITOR_MIGRATION_STATUS_INACTIVE:
case QEMU_MONITOR_MIGRATION_STATUS_SETUP:
case QEMU_MONITOR_MIGRATION_STATUS_ACTIVE:
case QEMU_MONITOR_MIGRATION_STATUS_PRE_SWITCHOVER:
case QEMU_MONITOR_MIGRATION_STATUS_DEVICE:
case QEMU_MONITOR_MIGRATION_STATUS_ERROR:
case QEMU_MONITOR_MIGRATION_STATUS_CANCELLING:
case QEMU_MONITOR_MIGRATION_STATUS_CANCELLED:
case QEMU_MONITOR_MIGRATION_STATUS_WAIT_UNPLUG:
case QEMU_MONITOR_MIGRATION_STATUS_LAST:
default:
break;
}
cleanup:
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleMigrationPass(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
int pass)
{
qemuDomainObjPrivate *priv;
virObjectLock(vm);
VIR_DEBUG("Migrating domain %p %s, iteration %d",
vm, vm->def->name, pass);
priv = vm->privateData;
if (vm->job->asyncJob == VIR_ASYNC_JOB_NONE) {
VIR_DEBUG("got MIGRATION_PASS event without a migration job");
goto cleanup;
}
virObjectEventStateQueue(priv->driver->domainEventState,
virDomainEventMigrationIterationNewFromObj(vm, pass));
cleanup:
virObjectUnlock(vm);
}
static void
qemuProcessHandleDumpCompleted(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
int status,
qemuMonitorDumpStats *stats,
const char *error)
{
qemuDomainJobPrivate *jobPriv;
qemuDomainJobDataPrivate *privJobCurrent = NULL;
virObjectLock(vm);
VIR_DEBUG("Dump completed for domain %p %s with stats=%p error='%s'",
vm, vm->def->name, stats, NULLSTR(error));
jobPriv = vm->job->privateData;
if (vm->job->asyncJob == VIR_ASYNC_JOB_NONE) {
VIR_DEBUG("got DUMP_COMPLETED event without a dump_completed job");
goto cleanup;
}
privJobCurrent = vm->job->current->privateData;
jobPriv->dumpCompleted = true;
privJobCurrent->stats.dump = *stats;
vm->job->error = g_strdup(error);
/* Force error if extracting the DUMP_COMPLETED status failed */
if (!error && status < 0) {
vm->job->error = g_strdup(virGetLastErrorMessage());
privJobCurrent->stats.dump.status = QEMU_MONITOR_DUMP_STATUS_FAILED;
}
virDomainObjBroadcast(vm);
cleanup:
virResetLastError();
virObjectUnlock(vm);
}
static void
qemuProcessHandlePRManagerStatusChanged(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *prManager,
bool connected)
{
qemuDomainObjPrivate *priv;
const char *managedAlias = qemuDomainGetManagedPRAlias();
virObjectLock(vm);
VIR_DEBUG("pr-manager %s status changed for domain %p %s connected=%d",
prManager, vm, vm->def->name, connected);
/* Connect events are boring. */
if (connected)
goto cleanup;
/* Disconnect events are more interesting. */
if (STRNEQ(prManager, managedAlias)) {
VIR_DEBUG("pr-manager %s not managed, ignoring event",
prManager);
goto cleanup;
}
priv = vm->privateData;
priv->prDaemonRunning = false;
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_PR_DISCONNECT,
0, 0, NULL);
cleanup:
virObjectUnlock(vm);
}
static void
qemuProcessHandleRdmaGidStatusChanged(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *netdev,
bool gid_status,
unsigned long long subnet_prefix,
unsigned long long interface_id)
{
qemuMonitorRdmaGidStatus *info = NULL;
virObjectLock(vm);
VIR_DEBUG("netdev=%s,gid_status=%d,subnet_prefix=0x%llx,interface_id=0x%llx",
netdev, gid_status, subnet_prefix, interface_id);
info = g_new0(qemuMonitorRdmaGidStatus, 1);
info->netdev = g_strdup(netdev);
info->gid_status = gid_status;
info->subnet_prefix = subnet_prefix;
info->interface_id = interface_id;
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_RDMA_GID_STATUS_CHANGED,
0, 0, info);
virObjectUnlock(vm);
}
static void
qemuProcessHandleGuestCrashloaded(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm)
{
virObjectLock(vm);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_GUEST_CRASHLOADED,
0, 0, NULL);
virObjectUnlock(vm);
}
static void
qemuProcessHandleMemoryFailure(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
qemuMonitorEventMemoryFailure *mfp)
{
virQEMUDriver *driver;
virObjectEvent *event = NULL;
virDomainMemoryFailureRecipientType recipient;
virDomainMemoryFailureActionType action;
unsigned int flags = 0;
virObjectLock(vm);
driver = QEMU_DOMAIN_PRIVATE(vm)->driver;
switch (mfp->recipient) {
case QEMU_MONITOR_MEMORY_FAILURE_RECIPIENT_HYPERVISOR:
recipient = VIR_DOMAIN_EVENT_MEMORY_FAILURE_RECIPIENT_HYPERVISOR;
break;
case QEMU_MONITOR_MEMORY_FAILURE_RECIPIENT_GUEST:
recipient = VIR_DOMAIN_EVENT_MEMORY_FAILURE_RECIPIENT_GUEST;
break;
case QEMU_MONITOR_MEMORY_FAILURE_RECIPIENT_LAST:
default:
return;
}
switch (mfp->action) {
case QEMU_MONITOR_MEMORY_FAILURE_ACTION_IGNORE:
action = VIR_DOMAIN_EVENT_MEMORY_FAILURE_ACTION_IGNORE;
break;
case QEMU_MONITOR_MEMORY_FAILURE_ACTION_INJECT:
action = VIR_DOMAIN_EVENT_MEMORY_FAILURE_ACTION_INJECT;
break;
case QEMU_MONITOR_MEMORY_FAILURE_ACTION_FATAL:
action = VIR_DOMAIN_EVENT_MEMORY_FAILURE_ACTION_FATAL;
break;
case QEMU_MONITOR_MEMORY_FAILURE_ACTION_RESET:
action = VIR_DOMAIN_EVENT_MEMORY_FAILURE_ACTION_RESET;
break;
case QEMU_MONITOR_MEMORY_FAILURE_ACTION_LAST:
default:
return;
}
if (mfp->action_required)
flags |= VIR_DOMAIN_MEMORY_FAILURE_ACTION_REQUIRED;
if (mfp->recursive)
flags |= VIR_DOMAIN_MEMORY_FAILURE_RECURSIVE;
event = virDomainEventMemoryFailureNewFromObj(vm, recipient, action, flags);
virObjectUnlock(vm);
virObjectEventStateQueue(driver->domainEventState, event);
}
static void
qemuProcessHandleMemoryDeviceSizeChange(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm,
const char *devAlias,
unsigned long long size)
{
qemuMonitorMemoryDeviceSizeChange *info = NULL;
virObjectLock(vm);
VIR_DEBUG("Memory device '%s' changed size to '%llu' in domain '%s'",
devAlias, size, vm->def->name);
info = g_new0(qemuMonitorMemoryDeviceSizeChange, 1);
info->devAlias = g_strdup(devAlias);
info->size = size;
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_MEMORY_DEVICE_SIZE_CHANGE,
0, 0, info);
virObjectUnlock(vm);
}
static qemuMonitorCallbacks monitorCallbacks = {
.eofNotify = qemuProcessHandleMonitorEOF,
.errorNotify = qemuProcessHandleMonitorError,
.domainEvent = qemuProcessHandleEvent,
.domainShutdown = qemuProcessHandleShutdown,
.domainStop = qemuProcessHandleStop,
.domainResume = qemuProcessHandleResume,
.domainReset = qemuProcessHandleReset,
.domainRTCChange = qemuProcessHandleRTCChange,
.domainWatchdog = qemuProcessHandleWatchdog,
.domainIOError = qemuProcessHandleIOError,
.domainGraphics = qemuProcessHandleGraphics,
.jobStatusChange = qemuProcessHandleJobStatusChange,
.domainTrayChange = qemuProcessHandleTrayChange,
.domainPMWakeup = qemuProcessHandlePMWakeup,
.domainPMSuspend = qemuProcessHandlePMSuspend,
.domainBalloonChange = qemuProcessHandleBalloonChange,
.domainPMSuspendDisk = qemuProcessHandlePMSuspendDisk,
.domainGuestPanic = qemuProcessHandleGuestPanic,
.domainDeviceDeleted = qemuProcessHandleDeviceDeleted,
.domainNicRxFilterChanged = qemuProcessHandleNicRxFilterChanged,
.domainSerialChange = qemuProcessHandleSerialChanged,
.domainSpiceMigrated = qemuProcessHandleSpiceMigrated,
.domainMigrationStatus = qemuProcessHandleMigrationStatus,
.domainMigrationPass = qemuProcessHandleMigrationPass,
.domainAcpiOstInfo = qemuProcessHandleAcpiOstInfo,
.domainBlockThreshold = qemuProcessHandleBlockThreshold,
.domainDumpCompleted = qemuProcessHandleDumpCompleted,
.domainPRManagerStatusChanged = qemuProcessHandlePRManagerStatusChanged,
.domainRdmaGidStatusChanged = qemuProcessHandleRdmaGidStatusChanged,
.domainGuestCrashloaded = qemuProcessHandleGuestCrashloaded,
.domainMemoryFailure = qemuProcessHandleMemoryFailure,
.domainMemoryDeviceSizeChange = qemuProcessHandleMemoryDeviceSizeChange,
.domainDeviceUnplugError = qemuProcessHandleDeviceUnplugErr,
.domainNetdevStreamDisconnected = qemuProcessHandleNetdevStreamDisconnected,
};
static void
qemuProcessMonitorReportLogError(qemuMonitor *mon,
const char *msg,
void *opaque);
static void
qemuProcessMonitorLogFree(void *opaque)
{
qemuLogContext *logCtxt = opaque;
g_clear_object(&logCtxt);
}
static int
qemuProcessInitMonitor(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
int ret;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
ret = qemuMonitorSetCapabilities(QEMU_DOMAIN_PRIVATE(vm)->mon);
qemuDomainObjExitMonitor(vm);
return ret;
}
static int
qemuConnectMonitor(virQEMUDriver *driver,
virDomainObj *vm,
int asyncJob,
qemuLogContext *logCtxt,
bool reconnect)
{
qemuDomainObjPrivate *priv = vm->privateData;
qemuMonitor *mon = NULL;
if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def) < 0) {
VIR_ERROR(_("Failed to set security context for monitor for %1$s"),
vm->def->name);
return -1;
}
ignore_value(virTimeMillisNow(&priv->monStart));
mon = qemuMonitorOpen(vm,
priv->monConfig,
virEventThreadGetContext(priv->eventThread),
&monitorCallbacks);
if (mon && logCtxt) {
g_object_ref(logCtxt);
qemuMonitorSetDomainLog(mon,
qemuProcessMonitorReportLogError,
logCtxt,
qemuProcessMonitorLogFree);
}
priv->monStart = 0;
priv->mon = mon;
if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0) {
VIR_ERROR(_("Failed to clear security context for monitor for %1$s"),
vm->def->name);
return -1;
}
if (priv->mon == NULL) {
VIR_INFO("Failed to connect monitor for %s", vm->def->name);
return -1;
}
if (qemuProcessInitMonitor(vm, asyncJob) < 0)
return -1;
if (qemuMigrationCapsCheck(vm, asyncJob, reconnect) < 0)
return -1;
return 0;
}
static int
qemuProcessReportLogError(qemuLogContext *logCtxt,
const char *msgprefix)
{
g_autofree char *logmsg = NULL;
/* assume that 1024 chars of qemu log is the right balance */
if (qemuLogContextReadFiltered(logCtxt, &logmsg, 1024) < 0)
return -1;
virResetLastError();
if (virStringIsEmpty(logmsg))
virReportError(VIR_ERR_INTERNAL_ERROR, "%s", msgprefix);
else
virReportError(VIR_ERR_INTERNAL_ERROR, "%s: %s", /* _( silence sc_libvirt_unmarked_diagnostics */
msgprefix, logmsg);
return 0;
}
static void
qemuProcessMonitorReportLogError(qemuMonitor *mon G_GNUC_UNUSED,
const char *msg,
void *opaque)
{
qemuLogContext *logCtxt = opaque;
qemuProcessReportLogError(logCtxt, msg);
}
static int
qemuProcessLookupPTYs(virDomainChrDef **devices,
int count,
GHashTable *info)
{
size_t i;
for (i = 0; i < count; i++) {
g_autofree char *id = NULL;
virDomainChrDef *chr = devices[i];
if (chr->source->type == VIR_DOMAIN_CHR_TYPE_PTY) {
qemuMonitorChardevInfo *entry;
id = g_strdup_printf("char%s", chr->info.alias);
entry = virHashLookup(info, id);
if (!entry || !entry->ptyPath) {
if (chr->source->data.file.path == NULL) {
/* neither the log output nor 'info chardev' had a
* pty path for this chardev, report an error
*/
virReportError(VIR_ERR_INTERNAL_ERROR,
_("no assigned pty for device %1$s"), id);
return -1;
} else {
/* 'info chardev' had no pty path for this chardev,
* but the log output had, so we're fine
*/
continue;
}
}
g_free(chr->source->data.file.path);
chr->source->data.file.path = g_strdup(entry->ptyPath);
}
}
return 0;
}
static int
qemuProcessFindCharDevicePTYsMonitor(virDomainObj *vm,
GHashTable *info)
{
size_t i = 0;
if (qemuProcessLookupPTYs(vm->def->serials, vm->def->nserials, info) < 0)
return -1;
if (qemuProcessLookupPTYs(vm->def->parallels, vm->def->nparallels,
info) < 0)
return -1;
if (qemuProcessLookupPTYs(vm->def->channels, vm->def->nchannels, info) < 0)
return -1;
/* For historical reasons, console[0] can be just an alias
* for serial[0]. That's why we need to update it as well. */
if (vm->def->nconsoles) {
virDomainChrDef *chr = vm->def->consoles[0];
if (vm->def->nserials &&
chr->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CONSOLE &&
chr->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_SERIAL) {
/* yes, the first console is just an alias for serials[0] */
i = 1;
virDomainChrSourceDefCopy(chr->source,
((vm->def->serials[0])->source));
}
}
if (qemuProcessLookupPTYs(vm->def->consoles + i, vm->def->nconsoles - i,
info) < 0)
return -1;
return 0;
}
static void
qemuProcessRefreshChannelVirtioState(virQEMUDriver *driver,
virDomainObj *vm,
GHashTable *info,
int booted)
{
size_t i;
int agentReason = VIR_CONNECT_DOMAIN_EVENT_AGENT_LIFECYCLE_REASON_CHANNEL;
qemuMonitorChardevInfo *entry;
virObjectEvent *event = NULL;
g_autofree char *id = NULL;
if (booted)
agentReason = VIR_CONNECT_DOMAIN_EVENT_AGENT_LIFECYCLE_REASON_DOMAIN_STARTED;
for (i = 0; i < vm->def->nchannels; i++) {
virDomainChrDef *chr = vm->def->channels[i];
if (chr->targetType == VIR_DOMAIN_CHR_CHANNEL_TARGET_TYPE_VIRTIO) {
VIR_FREE(id);
id = g_strdup_printf("char%s", chr->info.alias);
/* port state not reported */
if (!(entry = virHashLookup(info, id)) ||
!entry->state)
continue;
if (entry->state != VIR_DOMAIN_CHR_DEVICE_STATE_DEFAULT &&
STREQ_NULLABLE(chr->target.name, "org.qemu.guest_agent.0") &&
(event = virDomainEventAgentLifecycleNewFromObj(vm, entry->state,
agentReason)))
virObjectEventStateQueue(driver->domainEventState, event);
chr->state = entry->state;
}
}
}
int
qemuRefreshVirtioChannelState(virQEMUDriver *driver,
virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(GHashTable) info = NULL;
int rc;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = qemuMonitorGetChardevInfo(priv->mon, &info);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
qemuProcessRefreshChannelVirtioState(driver, vm, info, false);
return 0;
}
static int
qemuProcessRefreshPRManagerState(virDomainObj *vm,
GHashTable *info)
{
qemuDomainObjPrivate *priv = vm->privateData;
qemuMonitorPRManagerInfo *prManagerInfo;
const char *managedAlias = qemuDomainGetManagedPRAlias();
if (!(prManagerInfo = virHashLookup(info, managedAlias))) {
virReportError(VIR_ERR_OPERATION_FAILED,
_("missing info on pr-manager %1$s"),
managedAlias);
return -1;
}
priv->prDaemonRunning = prManagerInfo->connected;
if (!priv->prDaemonRunning &&
qemuProcessStartManagedPRDaemon(vm) < 0)
return -1;
return 0;
}
static int
qemuRefreshPRManagerState(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(GHashTable) info = NULL;
int rc;
if (!virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_PR_MANAGER_HELPER) ||
!qemuDomainDefHasManagedPR(vm))
return 0;
qemuDomainObjEnterMonitor(vm);
rc = qemuMonitorGetPRManagerInfo(priv->mon, &info);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
return qemuProcessRefreshPRManagerState(vm, info);
}
static int
qemuProcessRefreshFdsetIndex(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(qemuMonitorFdsets) fdsets = NULL;
size_t i;
int rc;
/* if the previous index was in the status XML we don't need to update it */
if (priv->fdsetindexParsed)
return 0;
qemuDomainObjEnterMonitor(vm);
rc = qemuMonitorQueryFdsets(priv->mon, &fdsets);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
for (i = 0; i < fdsets->nfdsets; i++) {
if (fdsets->fdsets[i].id >= priv->fdsetindex)
priv->fdsetindex = fdsets->fdsets[i].id + 1;
}
return 0;
}
static void
qemuRefreshRTC(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
time_t now, then;
struct tm thenbits = { 0 };
long localOffset;
int rv;
if (vm->def->clock.offset != VIR_DOMAIN_CLOCK_OFFSET_VARIABLE)
return;
qemuDomainObjEnterMonitor(vm);
now = time(NULL);
rv = qemuMonitorGetRTCTime(priv->mon, &thenbits);
qemuDomainObjExitMonitor(vm);
if (rv < 0)
return;
thenbits.tm_isdst = -1;
if ((then = mktime(&thenbits)) == (time_t)-1) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Unable to convert time"));
return;
}
/* Thing is, @now is in local TZ but @then in UTC. */
if (virTimeLocalOffsetFromUTC(&localOffset) < 0)
return;
vm->def->clock.data.variable.adjustment = then - now + localOffset;
}
int
qemuProcessRefreshBalloonState(virDomainObj *vm,
int asyncJob)
{
unsigned long long balloon;
size_t i;
int rc;
/* if no ballooning is available, the current size equals to the current
* full memory size */
if (!virDomainDefHasMemballoon(vm->def)) {
vm->def->mem.cur_balloon = virDomainDefGetMemoryTotal(vm->def);
return 0;
}
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = qemuMonitorGetBalloonInfo(qemuDomainGetMonitor(vm), &balloon);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
/* We want the balloon size stored in domain definition to
* account for the actual size of virtio-mem too. But the
* balloon size as reported by QEMU (@balloon) contains just
* the balloon size without any virtio-mem. Do a wee bit of
* math to fix it. */
VIR_DEBUG("balloon size before fix is %lld", balloon);
for (i = 0; i < vm->def->nmems; i++) {
if (vm->def->mems[i]->model == VIR_DOMAIN_MEMORY_MODEL_VIRTIO_MEM)
balloon += vm->def->mems[i]->target.virtio_mem.currentsize;
}
VIR_DEBUG("Updating balloon from %lld to %lld kb",
vm->def->mem.cur_balloon, balloon);
vm->def->mem.cur_balloon = balloon;
return 0;
}
static int
qemuProcessWaitForMonitor(virQEMUDriver *driver,
virDomainObj *vm,
int asyncJob,
qemuLogContext *logCtxt)
{
int ret = -1;
g_autoptr(GHashTable) info = NULL;
qemuDomainObjPrivate *priv = vm->privateData;
VIR_DEBUG("Connect monitor to vm=%p name='%s'", vm, vm->def->name);
if (qemuConnectMonitor(driver, vm, asyncJob, logCtxt, false) < 0)
goto cleanup;
/* Try to get the pty path mappings again via the monitor. This is much more
* reliable if it's available.
* Note that the monitor itself can be on a pty, so we still need to try the
* log output method. */
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
goto cleanup;
ret = qemuMonitorGetChardevInfo(priv->mon, &info);
VIR_DEBUG("qemuMonitorGetChardevInfo returned %i", ret);
qemuDomainObjExitMonitor(vm);
if (ret == 0) {
if ((ret = qemuProcessFindCharDevicePTYsMonitor(vm, info)) < 0)
goto cleanup;
qemuProcessRefreshChannelVirtioState(driver, vm, info, true);
}
cleanup:
if (logCtxt && kill(vm->pid, 0) == -1 && errno == ESRCH) {
qemuProcessReportLogError(logCtxt,
_("process exited while connecting to monitor"));
ret = -1;
}
return ret;
}
static int
qemuProcessDetectIOThreadPIDs(virDomainObj *vm,
int asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
qemuMonitorIOThreadInfo **iothreads = NULL;
int niothreads = 0;
int ret = -1;
size_t i;
/* Get the list of IOThreads from qemu */
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
goto cleanup;
ret = qemuMonitorGetIOThreads(priv->mon, &iothreads, &niothreads);
qemuDomainObjExitMonitor(vm);
if (ret < 0)
goto cleanup;
if (niothreads != vm->def->niothreadids) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("got wrong number of IOThread pids from QEMU monitor. got %1$d, wanted %2$zu"),
niothreads, vm->def->niothreadids);
goto cleanup;
}
/* Nothing to do */
if (niothreads == 0) {
ret = 0;
goto cleanup;
}
for (i = 0; i < niothreads; i++) {
virDomainIOThreadIDDef *iothrid;
if (!(iothrid = virDomainIOThreadIDFind(vm->def,
iothreads[i]->iothread_id))) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("iothread %1$d not found"),
iothreads[i]->iothread_id);
goto cleanup;
}
iothrid->thread_id = iothreads[i]->thread_id;
}
ret = 0;
cleanup:
if (iothreads) {
for (i = 0; i < niothreads; i++)
VIR_FREE(iothreads[i]);
VIR_FREE(iothreads);
}
return ret;
}
static int
qemuProcessGetAllCpuAffinity(virBitmap **cpumapRet)
{
*cpumapRet = NULL;
if (!virHostCPUHasBitmap())
return 0;
if (!(*cpumapRet = virHostCPUGetOnlineBitmap()))
return -1;
return 0;
}
/*
* To be run between fork/exec of QEMU only
*/
#if defined(WITH_SCHED_GETAFFINITY) || defined(WITH_BSD_CPU_AFFINITY)
static int
qemuProcessInitCpuAffinity(virDomainObj *vm)
{
bool settingAll = false;
g_autoptr(virBitmap) cpumapToSet = NULL;
virDomainNumatuneMemMode mem_mode;
qemuDomainObjPrivate *priv = vm->privateData;
if (!vm->pid) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Cannot setup CPU affinity until process is started"));
return -1;
}
/* Here is the deal, we can't set cpuset.mems before qemu is
* started as it clashes with KVM allocation. Therefore, we
* used to let qemu allocate its memory anywhere as we would
* then move the memory to desired NUMA node via CGroups.
* However, that might not be always possible because qemu
* might lock some parts of its memory (e.g. due to VFIO).
* Even if it possible, memory has to be copied between NUMA
* nodes which is suboptimal.
* Solution is to set affinity that matches the best what we
* would have set in CGroups and then fix it later, once qemu
* is already running. */
if (virDomainNumaGetNodeCount(vm->def->numa) <= 1 &&
virDomainNumatuneGetMode(vm->def->numa, -1, &mem_mode) == 0 &&
mem_mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT) {
virBitmap *nodeset = NULL;
if (virDomainNumatuneMaybeGetNodeset(vm->def->numa,
priv->autoNodeset,
&nodeset,
-1) < 0)
return -1;
if (virNumaNodesetToCPUset(nodeset, &cpumapToSet) < 0)
return -1;
} else if (vm->def->cputune.emulatorpin) {
cpumapToSet = virBitmapNewCopy(vm->def->cputune.emulatorpin);
} else {
settingAll = true;
if (qemuProcessGetAllCpuAffinity(&cpumapToSet) < 0)
return -1;
}
/*
* We only want to error out if we failed to set the affinity to
* user-requested mapping. If we are just trying to reset the affinity
* to all CPUs and this fails it can only be an issue if:
* 1) libvirtd does not have CAP_SYS_NICE
* 2) libvirtd does not run on all CPUs
*
* This scenario can easily occur when libvirtd is run inside a
* container with restrictive permissions and CPU pinning.
*
* See also: https://bugzilla.redhat.com/1819801#c2
*/
if (cpumapToSet &&
virProcessSetAffinity(vm->pid, cpumapToSet, settingAll) < 0) {
return -1;
}
return 0;
}
#else /* !defined(WITH_SCHED_GETAFFINITY) && !defined(WITH_BSD_CPU_AFFINITY) */
static int
qemuProcessInitCpuAffinity(virDomainObj *vm G_GNUC_UNUSED)
{
return 0;
}
#endif /* !defined(WITH_SCHED_GETAFFINITY) && !defined(WITH_BSD_CPU_AFFINITY) */
/* set link states to down on interfaces at qemu start */
static int
qemuProcessSetLinkStates(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainDef *def = vm->def;
size_t i;
int ret = -1;
int rv;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
for (i = 0; i < def->nnets; i++) {
if (def->nets[i]->linkstate == VIR_DOMAIN_NET_INTERFACE_LINK_STATE_DOWN) {
if (!def->nets[i]->info.alias) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("missing alias for network device"));
goto cleanup;
}
VIR_DEBUG("Setting link state: %s", def->nets[i]->info.alias);
rv = qemuMonitorSetLink(priv->mon,
def->nets[i]->info.alias,
VIR_DOMAIN_NET_INTERFACE_LINK_STATE_DOWN);
if (rv < 0) {
virReportError(VIR_ERR_OPERATION_FAILED,
_("Couldn't set link state on interface: %1$s"),
def->nets[i]->info.alias);
goto cleanup;
}
}
}
ret = 0;
cleanup:
qemuDomainObjExitMonitor(vm);
return ret;
}
/**
* qemuProcessSetupPid:
*
* This function sets resource properties (affinity, cgroups,
* scheduler) for any PID associated with a domain. It should be used
* to set up emulator PIDs as well as vCPU and I/O thread pids to
* ensure they are all handled the same way.
*
* Returns 0 on success, -1 on error.
*/
static int
qemuProcessSetupPid(virDomainObj *vm,
pid_t pid,
virCgroupThreadName nameval,
int id,
virBitmap *cpumask,
unsigned long long period,
long long quota,
virDomainThreadSchedParam *sched)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainNuma *numatune = vm->def->numa;
virDomainNumatuneMemMode mem_mode;
virCgroup *cgroup = NULL;
virBitmap *use_cpumask = NULL;
virBitmap *affinity_cpumask = NULL;
g_autoptr(virBitmap) hostcpumap = NULL;
g_autofree char *mem_mask = NULL;
int ret = -1;
size_t i;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
goto cleanup;
}
/* Infer which cpumask shall be used. */
if (!(use_cpumask = qemuDomainEvaluateCPUMask(vm->def,
cpumask, priv->autoCpuset))) {
/* You may think this is redundant, but we can't assume libvirtd
* itself is running on all pCPUs, so we need to explicitly set
* the spawned QEMU instance to all pCPUs if no map is given in
* its config file */
if (qemuProcessGetAllCpuAffinity(&hostcpumap) < 0)
goto cleanup;
affinity_cpumask = hostcpumap;
}
/*
* If CPU cgroup controller is not initialized here, then we need
* neither period nor quota settings. And if CPUSET controller is
* not initialized either, then there's nothing to do anyway.
*/
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU) ||
virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET)) {
if (virDomainNumatuneGetMode(numatune, -1, &mem_mode) == 0) {
/* QEMU allocates its memory from the emulator thread. Thus it
* needs to access union of all host nodes configured. */
if (mem_mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT) {
qemuDomainNumatuneMaybeFormatNodesetUnion(vm, NULL, &mem_mask);
} else if (mem_mode == VIR_DOMAIN_NUMATUNE_MEM_RESTRICTIVE &&
virDomainNumatuneMaybeFormatNodeset(numatune,
priv->autoNodeset,
&mem_mask, -1) < 0) {
goto cleanup;
}
}
/* For restrictive numatune mode we need to set cpuset.mems for vCPU
* threads based on the node they are in as there is nothing else uses
* for such restriction (e.g. numa_set_membind). */
if (nameval == VIR_CGROUP_THREAD_VCPU) {
/* Look for the guest NUMA node of this vCPU */
for (i = 0; i < virDomainNumaGetNodeCount(numatune); i++) {
virBitmap *node_cpus = virDomainNumaGetNodeCpumask(numatune, i);
if (!virBitmapIsBitSet(node_cpus, id))
continue;
/* Update the mem_mask for this vCPU if the mode of its node is
* 'restrictive'. */
if (virDomainNumatuneGetMode(numatune, i, &mem_mode) == 0 &&
mem_mode == VIR_DOMAIN_NUMATUNE_MEM_RESTRICTIVE) {
VIR_FREE(mem_mask);
if (virDomainNumatuneMaybeFormatNodeset(numatune,
priv->autoNodeset,
&mem_mask, i) < 0) {
goto cleanup;
}
}
break;
}
}
if (virCgroupNewThread(priv->cgroup, nameval, id, true, &cgroup) < 0)
goto cleanup;
/* Move the thread to the sub dir before changing the settings so that
* all take effect even with cgroupv2. */
if (virCgroupAddThread(cgroup, pid) < 0)
goto cleanup;
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET)) {
if (use_cpumask &&
virDomainCgroupSetupCpusetCpus(cgroup, use_cpumask) < 0)
goto cleanup;
if (mem_mask && virCgroupSetCpusetMems(cgroup, mem_mask) < 0)
goto cleanup;
}
if (virDomainCgroupSetupVcpuBW(cgroup, period, quota) < 0)
goto cleanup;
}
if (!affinity_cpumask)
affinity_cpumask = use_cpumask;
/* Setup legacy affinity.
*
* We only want to error out if we failed to set the affinity to
* user-requested mapping. If we are just trying to reset the affinity
* to all CPUs and this fails it can only be an issue if:
* 1) libvirtd does not have CAP_SYS_NICE
* 2) libvirtd does not run on all CPUs
*
* This scenario can easily occur when libvirtd is run inside a
* container with restrictive permissions and CPU pinning.
*
* See also: https://bugzilla.redhat.com/1819801#c2
*/
if (affinity_cpumask &&
virProcessSetAffinity(pid, affinity_cpumask,
affinity_cpumask == hostcpumap) < 0) {
goto cleanup;
}
/* Set scheduler type and priority, but not for the main thread. */
if (sched &&
nameval != VIR_CGROUP_THREAD_EMULATOR &&
virProcessSetScheduler(pid, sched->policy, sched->priority) < 0)
goto cleanup;
ret = 0;
cleanup:
if (cgroup) {
if (ret < 0)
virCgroupRemove(cgroup);
virCgroupFree(cgroup);
}
return ret;
}
int
qemuProcessSetupEmulator(virDomainObj *vm)
{
return qemuProcessSetupPid(vm, vm->pid, VIR_CGROUP_THREAD_EMULATOR,
0, vm->def->cputune.emulatorpin,
vm->def->cputune.emulator_period,
vm->def->cputune.emulator_quota,
vm->def->cputune.emulatorsched);
}
static int
qemuProcessResctrlCreate(virQEMUDriver *driver,
virDomainObj *vm)
{
size_t i = 0;
g_autoptr(virCaps) caps = NULL;
qemuDomainObjPrivate *priv = vm->privateData;
if (!vm->def->nresctrls)
return 0;
/* Force capability refresh since resctrl info can change
* XXX: move cache info into virresctrl so caps are not needed */
caps = virQEMUDriverGetCapabilities(driver, true);
if (!caps)
return -1;
for (i = 0; i < vm->def->nresctrls; i++) {
size_t j = 0;
if (virResctrlAllocCreate(caps->host.resctrl,
vm->def->resctrls[i]->alloc,
priv->machineName) < 0)
return -1;
for (j = 0; j < vm->def->resctrls[i]->nmonitors; j++) {
virDomainResctrlMonDef *mon = NULL;
mon = vm->def->resctrls[i]->monitors[j];
if (virResctrlMonitorCreate(mon->instance,
priv->machineName) < 0)
return -1;
}
}
return 0;
}
static char *
qemuProcessBuildPRHelperPidfilePathOld(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
const char *prdAlias = qemuDomainGetManagedPRAlias();
return virPidFileBuildPath(priv->libDir, prdAlias);
}
static char *
qemuProcessBuildPRHelperPidfilePath(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autofree char *domname = virDomainDefGetShortName(vm->def);
g_autofree char *prdName = g_strdup_printf("%s-%s", domname, qemuDomainGetManagedPRAlias());
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
return virPidFileBuildPath(cfg->stateDir, prdName);
}
void
qemuProcessKillManagedPRDaemon(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virErrorPtr orig_err;
g_autofree char *pidfile = NULL;
if (!(pidfile = qemuProcessBuildPRHelperPidfilePath(vm))) {
VIR_WARN("Unable to construct pr-helper pidfile path");
return;
}
if (!virFileExists(pidfile)) {
g_free(pidfile);
if (!(pidfile = qemuProcessBuildPRHelperPidfilePathOld(vm))) {
VIR_WARN("Unable to construct pr-helper pidfile path");
return;
}
}
virErrorPreserveLast(&orig_err);
if (virPidFileForceCleanupPath(pidfile) < 0) {
VIR_WARN("Unable to kill pr-helper process");
} else {
priv->prDaemonRunning = false;
}
virErrorRestore(&orig_err);
}
static int
qemuProcessStartPRDaemonHook(void *opaque)
{
virDomainObj *vm = opaque;
size_t i, nfds = 0;
g_autofree int *fds = NULL;
int ret = -1;
if (qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT)) {
virProcessGetNamespaces(vm->pid, &nfds, &fds);
if (nfds > 0 &&
virProcessSetNamespaces(nfds, fds) < 0)
goto cleanup;
}
ret = 0;
cleanup:
for (i = 0; i < nfds; i++)
VIR_FORCE_CLOSE(fds[i]);
return ret;
}
int
qemuProcessStartManagedPRDaemon(virDomainObj *vm)
{
const char *const prHelperDirs[] = {
"/usr/libexec",
NULL,
};
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUDriver *driver = priv->driver;
g_autoptr(virQEMUDriverConfig) cfg = NULL;
int errfd = -1;
g_autofree char *prHelperPath = NULL;
g_autofree char *pidfile = NULL;
g_autofree char *socketPath = NULL;
pid_t cpid = -1;
g_autoptr(virCommand) cmd = NULL;
virTimeBackOffVar timebackoff;
const unsigned long long timeout = 500000; /* ms */
int ret = -1;
cfg = virQEMUDriverGetConfig(driver);
prHelperPath = virFindFileInPathFull(cfg->prHelperName, prHelperDirs);
if (!prHelperPath) {
virReportSystemError(errno, _("'%1$s' is not a suitable pr helper"),
cfg->prHelperName);
goto cleanup;
}
VIR_DEBUG("Using qemu-pr-helper: %s", prHelperPath);
if (!(pidfile = qemuProcessBuildPRHelperPidfilePath(vm)))
goto cleanup;
if (!(socketPath = qemuDomainGetManagedPRSocketPath(priv)))
goto cleanup;
/* Remove stale socket */
if (unlink(socketPath) < 0 &&
errno != ENOENT) {
virReportSystemError(errno,
_("Unable to remove stale socket path: %1$s"),
socketPath);
goto cleanup;
}
if (!(cmd = virCommandNewArgList(prHelperPath,
"-k", socketPath,
NULL)))
goto cleanup;
virCommandDaemonize(cmd);
virCommandSetPidFile(cmd, pidfile);
virCommandSetErrorFD(cmd, &errfd);
/* Place the process into the same namespace and cgroup as
* qemu (so that it shares the same view of the system). */
virCommandSetPreExecHook(cmd, qemuProcessStartPRDaemonHook, vm);
if (cfg->schedCore == QEMU_SCHED_CORE_FULL) {
pid_t cookie_pid = vm->pid;
if (cookie_pid <= 0)
cookie_pid = priv->schedCoreChildPID;
virCommandSetRunAmong(cmd, cookie_pid);
}
if (virCommandRun(cmd, NULL) < 0)
goto cleanup;
if (virPidFileReadPath(pidfile, &cpid) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("pr helper %1$s didn't show up"),
prHelperPath);
goto cleanup;
}
if (virTimeBackOffStart(&timebackoff, 1, timeout) < 0)
goto cleanup;
while (virTimeBackOffWait(&timebackoff)) {
char errbuf[1024] = { 0 };
if (virFileExists(socketPath))
break;
if (virProcessKill(cpid, 0) == 0)
continue;
if (saferead(errfd, errbuf, sizeof(errbuf) - 1) < 0) {
virReportSystemError(errno,
_("pr helper %1$s died unexpectedly"),
prHelperPath);
} else {
virReportError(VIR_ERR_OPERATION_FAILED,
_("pr helper died and reported: %1$s"), errbuf);
}
goto cleanup;
}
if (!virFileExists(socketPath)) {
virReportError(VIR_ERR_OPERATION_TIMEOUT, "%s",
_("pr helper socked did not show up"));
goto cleanup;
}
if (priv->cgroup &&
virCgroupAddMachineProcess(priv->cgroup, cpid) < 0)
goto cleanup;
if (qemuSecurityDomainSetPathLabel(driver, vm, socketPath, true) < 0)
goto cleanup;
priv->prDaemonRunning = true;
ret = 0;
cleanup:
if (ret < 0) {
virCommandAbort(cmd);
if (cpid >= 0)
virProcessKillPainfully(cpid, true);
if (pidfile)
unlink(pidfile);
}
VIR_FORCE_CLOSE(errfd);
return ret;
}
static int
qemuProcessAllowPostCopyMigration(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUDriver *driver = priv->driver;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
const char *const *devices = (const char *const *) cfg->cgroupDeviceACL;
const char *uffd = "/dev/userfaultfd";
int rc;
if (!virFileExists(uffd)) {
VIR_DEBUG("%s is not supported by the host", uffd);
return 0;
}
if (!devices)
devices = defaultDeviceACL;
if (!g_strv_contains(devices, uffd)) {
VIR_DEBUG("%s is not allowed by device ACL", uffd);
return 0;
}
VIR_DEBUG("Labeling %s in mount namespace", uffd);
if ((rc = qemuSecurityDomainSetMountNSPathLabel(driver, vm, uffd)) < 0)
return -1;
if (rc == 1)
VIR_DEBUG("Mount namespace is not enabled, leaving %s as is", uffd);
return 0;
}
static int
qemuProcessInitPasswords(virQEMUDriver *driver,
virDomainObj *vm,
int asyncJob)
{
int ret = 0;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
size_t i;
for (i = 0; i < vm->def->ngraphics; ++i) {
virDomainGraphicsDef *graphics = vm->def->graphics[i];
if (graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC) {
ret = qemuDomainChangeGraphicsPasswords(vm,
VIR_DOMAIN_GRAPHICS_TYPE_VNC,
&graphics->data.vnc.auth,
cfg->vncPassword,
asyncJob);
} else if (graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_SPICE) {
ret = qemuDomainChangeGraphicsPasswords(vm,
VIR_DOMAIN_GRAPHICS_TYPE_SPICE,
&graphics->data.spice.auth,
cfg->spicePassword,
asyncJob);
}
if (ret < 0)
return ret;
}
return ret;
}
static int
qemuProcessCleanupChardevDevice(virDomainDef *def G_GNUC_UNUSED,
virDomainChrDef *dev,
void *opaque G_GNUC_UNUSED)
{
if (dev->source->type == VIR_DOMAIN_CHR_TYPE_UNIX &&
dev->source->data.nix.listen &&
dev->source->data.nix.path)
unlink(dev->source->data.nix.path);
return 0;
}
/**
* Loads and update video memory size for video devices according to QEMU
* process as the QEMU will silently update the values that we pass to QEMU
* through command line. We need to load these updated values and store them
* into the status XML.
*
* We will fail if for some reason the values cannot be loaded from QEMU because
* its mandatory to get the correct video memory size to status XML to not break
* migration.
*/
static int
qemuProcessUpdateVideoRamSize(virQEMUDriver *driver,
virDomainObj *vm,
int asyncJob)
{
int ret = -1;
ssize_t i;
qemuDomainObjPrivate *priv = vm->privateData;
virDomainVideoDef *video = NULL;
g_autoptr(virQEMUDriverConfig) cfg = NULL;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
for (i = 0; i < vm->def->nvideos; i++) {
video = vm->def->videos[i];
switch (video->type) {
case VIR_DOMAIN_VIDEO_TYPE_VGA:
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_VGA_VGAMEM)) {
if (qemuMonitorUpdateVideoMemorySize(priv->mon, video, "VGA") < 0)
goto error;
}
break;
case VIR_DOMAIN_VIDEO_TYPE_QXL:
if (i == 0) {
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_QXL_VGAMEM) &&
qemuMonitorUpdateVideoMemorySize(priv->mon, video,
"qxl-vga") < 0)
goto error;
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_QXL_VRAM64) &&
qemuMonitorUpdateVideoVram64Size(priv->mon, video,
"qxl-vga") < 0)
goto error;
} else {
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_QXL_VGAMEM) &&
qemuMonitorUpdateVideoMemorySize(priv->mon, video,
"qxl") < 0)
goto error;
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_QXL_VRAM64) &&
qemuMonitorUpdateVideoVram64Size(priv->mon, video,
"qxl") < 0)
goto error;
}
break;
case VIR_DOMAIN_VIDEO_TYPE_VMVGA:
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_VMWARE_SVGA_VGAMEM)) {
if (qemuMonitorUpdateVideoMemorySize(priv->mon, video,
"vmware-svga") < 0)
goto error;
}
break;
case VIR_DOMAIN_VIDEO_TYPE_DEFAULT:
case VIR_DOMAIN_VIDEO_TYPE_CIRRUS:
case VIR_DOMAIN_VIDEO_TYPE_XEN:
case VIR_DOMAIN_VIDEO_TYPE_VBOX:
case VIR_DOMAIN_VIDEO_TYPE_PARALLELS:
case VIR_DOMAIN_VIDEO_TYPE_VIRTIO:
case VIR_DOMAIN_VIDEO_TYPE_GOP:
case VIR_DOMAIN_VIDEO_TYPE_NONE:
case VIR_DOMAIN_VIDEO_TYPE_BOCHS:
case VIR_DOMAIN_VIDEO_TYPE_RAMFB:
case VIR_DOMAIN_VIDEO_TYPE_LAST:
break;
}
}
qemuDomainObjExitMonitor(vm);
cfg = virQEMUDriverGetConfig(driver);
ret = virDomainObjSave(vm, driver->xmlopt, cfg->stateDir);
return ret;
error:
qemuDomainObjExitMonitor(vm);
return -1;
}
struct qemuProcessHookData {
virDomainObj *vm;
virQEMUDriver *driver;
virQEMUDriverConfig *cfg;
};
static int qemuProcessHook(void *data)
{
struct qemuProcessHookData *h = data;
qemuDomainObjPrivate *priv = h->vm->privateData;
int ret = -1;
int fd;
virBitmap *nodeset = NULL;
virDomainNumatuneMemMode mode;
/* This method cannot use any mutexes, which are not
* protected across fork()
*/
qemuSecurityPostFork(h->driver->securityManager);
/* Some later calls want pid present */
h->vm->pid = getpid();
VIR_DEBUG("Obtaining domain lock");
/*
* Since we're going to leak the returned FD to QEMU,
* we need to make sure it gets a sensible label.
* This mildly sucks, because there could be other
* sockets the lock driver opens that we don't want
* labelled. So far we're ok though.
*/
if (qemuSecuritySetSocketLabel(h->driver->securityManager, h->vm->def) < 0)
goto cleanup;
if (virDomainLockProcessStart(h->driver->lockManager,
h->cfg->uri,
h->vm,
/* QEMU is always paused initially */
true,
&fd) < 0)
goto cleanup;
if (qemuSecurityClearSocketLabel(h->driver->securityManager, h->vm->def) < 0)
goto cleanup;
if (qemuDomainUnshareNamespace(h->cfg, h->driver->securityManager, h->vm) < 0)
goto cleanup;
if (virDomainNumatuneGetMode(h->vm->def->numa, -1, &mode) == 0) {
if ((mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT ||
mode == VIR_DOMAIN_NUMATUNE_MEM_RESTRICTIVE) &&
h->cfg->cgroupControllers & (1 << VIR_CGROUP_CONTROLLER_CPUSET) &&
virCgroupControllerAvailable(VIR_CGROUP_CONTROLLER_CPUSET)) {
/* Use virNuma* API iff necessary. Once set and child is exec()-ed,
* there's no way for us to change it. Rely on cgroups (if available
* and enabled in the config) rather than virNuma*. */
VIR_DEBUG("Relying on CGroups for memory binding");
} else {
nodeset = virDomainNumatuneGetNodeset(h->vm->def->numa,
priv->autoNodeset, -1);
if (virNumaSetupMemoryPolicy(mode, nodeset) < 0)
goto cleanup;
}
}
ret = 0;
cleanup:
virObjectUnref(h->cfg);
VIR_DEBUG("Hook complete ret=%d", ret);
return ret;
}
int
qemuProcessPrepareMonitorChr(virDomainChrSourceDef *monConfig,
const char *domainDir)
{
monConfig->type = VIR_DOMAIN_CHR_TYPE_UNIX;
monConfig->data.nix.listen = true;
monConfig->data.nix.path = g_strdup_printf("%s/monitor.sock", domainDir);
return 0;
}
/*
* Precondition: vm must be locked, and a job must be active.
* This method will call {Enter,Exit}Monitor
*/
int
qemuProcessStartCPUs(virQEMUDriver *driver, virDomainObj *vm,
virDomainRunningReason reason,
virDomainAsyncJob asyncJob)
{
int ret = -1;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
/* Bring up netdevs before starting CPUs */
if (virDomainInterfaceStartDevices(vm->def) < 0)
return -1;
VIR_DEBUG("Using lock state '%s'", NULLSTR(priv->lockState));
if (virDomainLockProcessResume(driver->lockManager, cfg->uri,
vm, priv->lockState) < 0) {
/* Don't free priv->lockState on error, because we need
* to make sure we have state still present if the user
* tries to resume again
*/
return -1;
}
VIR_FREE(priv->lockState);
priv->runningReason = reason;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
goto release;
ret = qemuMonitorStartCPUs(priv->mon);
qemuDomainObjExitMonitor(vm);
if (ret < 0)
goto release;
/* The RESUME event handler will change the domain state with the reason
* saved in priv->runningReason and it will also emit corresponding domain
* lifecycle event.
*/
return ret;
release:
priv->runningReason = VIR_DOMAIN_RUNNING_UNKNOWN;
if (virDomainLockProcessPause(driver->lockManager, vm, &priv->lockState) < 0)
VIR_WARN("Unable to release lease on %s", vm->def->name);
VIR_DEBUG("Preserving lock state '%s'", NULLSTR(priv->lockState));
return ret;
}
int qemuProcessStopCPUs(virQEMUDriver *driver,
virDomainObj *vm,
virDomainPausedReason reason,
virDomainAsyncJob asyncJob)
{
int ret = -1;
qemuDomainObjPrivate *priv = vm->privateData;
VIR_FREE(priv->lockState);
priv->pausedReason = reason;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
goto cleanup;
ret = qemuMonitorStopCPUs(priv->mon);
qemuDomainObjExitMonitor(vm);
if (ret < 0)
goto cleanup;
/* de-activate netdevs after stopping CPUs */
ignore_value(virDomainInterfaceStopDevices(vm->def));
if (vm->job->current)
ignore_value(virTimeMillisNow(&vm->job->current->stopped));
/* The STOP event handler will change the domain state with the reason
* saved in priv->pausedReason and it will also emit corresponding domain
* lifecycle event.
*/
if (virDomainLockProcessPause(driver->lockManager, vm, &priv->lockState) < 0)
VIR_WARN("Unable to release lease on %s", vm->def->name);
VIR_DEBUG("Preserving lock state '%s'", NULLSTR(priv->lockState));
cleanup:
if (ret < 0)
priv->pausedReason = VIR_DOMAIN_PAUSED_UNKNOWN;
return ret;
}
static void
qemuProcessNotifyNets(virDomainDef *def)
{
size_t i;
g_autoptr(virConnect) conn = NULL;
for (i = 0; i < def->nnets; i++) {
virDomainNetDef *net = def->nets[i];
/* keep others from trying to use the macvtap device name, but
* don't return error if this happens, since that causes the
* domain to be unceremoniously killed, which would be *very*
* impolite.
*/
switch (virDomainNetGetActualType(net)) {
case VIR_DOMAIN_NET_TYPE_DIRECT:
virNetDevReserveName(net->ifname);
break;
case VIR_DOMAIN_NET_TYPE_BRIDGE:
case VIR_DOMAIN_NET_TYPE_NETWORK:
case VIR_DOMAIN_NET_TYPE_ETHERNET:
virNetDevReserveName(net->ifname);
break;
case VIR_DOMAIN_NET_TYPE_USER:
case VIR_DOMAIN_NET_TYPE_VHOSTUSER:
case VIR_DOMAIN_NET_TYPE_SERVER:
case VIR_DOMAIN_NET_TYPE_CLIENT:
case VIR_DOMAIN_NET_TYPE_MCAST:
case VIR_DOMAIN_NET_TYPE_INTERNAL:
case VIR_DOMAIN_NET_TYPE_HOSTDEV:
case VIR_DOMAIN_NET_TYPE_UDP:
case VIR_DOMAIN_NET_TYPE_VDPA:
case VIR_DOMAIN_NET_TYPE_NULL:
case VIR_DOMAIN_NET_TYPE_VDS:
case VIR_DOMAIN_NET_TYPE_LAST:
break;
}
if (net->type == VIR_DOMAIN_NET_TYPE_NETWORK && !conn)
conn = virGetConnectNetwork();
virDomainNetNotifyActualDevice(conn, def, net);
}
}
/* Attempt to instantiate the filters. Ignore failures because it's
* possible that someone deleted a filter binding and the associated
* filter while the guest was running and we don't want that action
* to cause failure to keep the guest running during the reconnection
* processing. Nor do we necessarily want other failures to do the
* same. We'll just log the error conditions other than of course
* ignoreExists possibility (e.g. the true flag) */
static void
qemuProcessFiltersInstantiate(virDomainDef *def)
{
size_t i;
for (i = 0; i < def->nnets; i++) {
virDomainNetDef *net = def->nets[i];
if ((net->filter) && (net->ifname)) {
if (virDomainConfNWFilterInstantiate(def->name, def->uuid, net,
true) < 0) {
VIR_WARN("filter '%s' instantiation for '%s' failed '%s'",
net->filter, net->ifname, virGetLastErrorMessage());
virResetLastError();
}
}
}
}
static int
qemuProcessUpdateState(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainState state;
virDomainPausedReason reason;
virDomainState newState = VIR_DOMAIN_NOSTATE;
int oldReason;
int newReason;
bool running;
g_autofree char *msg = NULL;
int ret;
qemuDomainObjEnterMonitor(vm);
ret = qemuMonitorGetStatus(priv->mon, &running, &reason);
qemuDomainObjExitMonitor(vm);
if (ret < 0)
return -1;
state = virDomainObjGetState(vm, &oldReason);
if (running &&
(state == VIR_DOMAIN_SHUTOFF ||
(state == VIR_DOMAIN_PAUSED &&
oldReason == VIR_DOMAIN_PAUSED_STARTING_UP))) {
newState = VIR_DOMAIN_RUNNING;
newReason = VIR_DOMAIN_RUNNING_BOOTED;
msg = g_strdup("finished booting");
} else if (state == VIR_DOMAIN_PAUSED && running) {
newState = VIR_DOMAIN_RUNNING;
newReason = VIR_DOMAIN_RUNNING_UNPAUSED;
msg = g_strdup("was unpaused");
} else if (state == VIR_DOMAIN_RUNNING && !running) {
if (reason == VIR_DOMAIN_PAUSED_SHUTTING_DOWN) {
newState = VIR_DOMAIN_SHUTDOWN;
newReason = VIR_DOMAIN_SHUTDOWN_UNKNOWN;
msg = g_strdup("shutdown");
} else if (reason == VIR_DOMAIN_PAUSED_CRASHED) {
newState = VIR_DOMAIN_CRASHED;
newReason = VIR_DOMAIN_CRASHED_PANICKED;
msg = g_strdup("crashed");
} else {
newState = VIR_DOMAIN_PAUSED;
newReason = reason;
msg = g_strdup_printf("was paused (%s)",
virDomainPausedReasonTypeToString(reason));
}
}
if (newState != VIR_DOMAIN_NOSTATE) {
VIR_DEBUG("Domain %s %s while its monitor was disconnected;"
" changing state to %s (%s)",
vm->def->name,
NULLSTR(msg),
virDomainStateTypeToString(newState),
virDomainStateReasonToString(newState, newReason));
virDomainObjSetState(vm, newState, newReason);
}
return 0;
}
void
qemuProcessCleanupMigrationJob(virQEMUDriver *driver,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainState state;
int reason;
state = virDomainObjGetState(vm, &reason);
VIR_DEBUG("driver=%p, vm=%s, asyncJob=%s, state=%s, reason=%s",
driver, vm->def->name,
virDomainAsyncJobTypeToString(vm->job->asyncJob),
virDomainStateTypeToString(state),
virDomainStateReasonToString(state, reason));
if (vm->job->asyncJob != VIR_ASYNC_JOB_MIGRATION_IN &&
vm->job->asyncJob != VIR_ASYNC_JOB_MIGRATION_OUT)
return;
virPortAllocatorRelease(priv->migrationPort);
priv->migrationPort = 0;
qemuDomainObjDiscardAsyncJob(vm);
}
static void
qemuProcessRestoreMigrationJob(virDomainObj *vm,
virDomainJobObj *job)
{
qemuDomainJobPrivate *jobPriv = job->privateData;
virDomainJobOperation op;
unsigned long long allowedJobs;
if (job->asyncJob == VIR_ASYNC_JOB_MIGRATION_IN) {
op = VIR_DOMAIN_JOB_OPERATION_MIGRATION_IN;
allowedJobs = VIR_JOB_NONE;
} else {
op = VIR_DOMAIN_JOB_OPERATION_MIGRATION_OUT;
allowedJobs = VIR_JOB_DEFAULT_MASK | JOB_MASK(VIR_JOB_MIGRATION_OP);
}
allowedJobs |= JOB_MASK(VIR_JOB_MODIFY_MIGRATION_SAFE);
qemuDomainObjRestoreAsyncJob(vm, job->asyncJob, job->phase,
job->asyncStarted, op,
QEMU_DOMAIN_JOB_STATS_TYPE_MIGRATION,
VIR_DOMAIN_JOB_STATUS_PAUSED,
allowedJobs);
job->privateData = g_steal_pointer(&vm->job->privateData);
vm->job->privateData = jobPriv;
vm->job->apiFlags = job->apiFlags;
vm->job->asyncPaused = job->asyncPaused;
qemuDomainCleanupAdd(vm, qemuProcessCleanupMigrationJob);
}
/*
* Returns
* -1 on error, the domain will be killed,
* 0 the domain should remain running with the migration job discarded,
* 1 the daemon was restarted during post-copy phase
*/
static int
qemuProcessRecoverMigrationIn(virQEMUDriver *driver,
virDomainObj *vm,
virDomainJobObj *job,
virDomainState state)
{
VIR_DEBUG("Active incoming migration in phase %s",
qemuMigrationJobPhaseTypeToString(job->phase));
switch ((qemuMigrationJobPhase) job->phase) {
case QEMU_MIGRATION_PHASE_NONE:
case QEMU_MIGRATION_PHASE_PERFORM2:
case QEMU_MIGRATION_PHASE_BEGIN3:
case QEMU_MIGRATION_PHASE_PERFORM3:
case QEMU_MIGRATION_PHASE_PERFORM3_DONE:
case QEMU_MIGRATION_PHASE_CONFIRM3_CANCELLED:
case QEMU_MIGRATION_PHASE_CONFIRM3:
case QEMU_MIGRATION_PHASE_BEGIN_RESUME:
case QEMU_MIGRATION_PHASE_PERFORM_RESUME:
case QEMU_MIGRATION_PHASE_CONFIRM_RESUME:
case QEMU_MIGRATION_PHASE_LAST:
/* N/A for incoming migration */
break;
case QEMU_MIGRATION_PHASE_PREPARE:
VIR_DEBUG("Killing unfinished incoming migration for domain %s",
vm->def->name);
return -1;
case QEMU_MIGRATION_PHASE_FINISH2:
/* source domain is already killed so let's just resume the domain
* and hope we are all set */
VIR_DEBUG("Incoming migration finished, resuming domain %s",
vm->def->name);
if (qemuProcessStartCPUs(driver, vm,
VIR_DOMAIN_RUNNING_MIGRATED,
VIR_ASYNC_JOB_NONE) < 0) {
VIR_WARN("Could not resume domain %s", vm->def->name);
}
break;
case QEMU_MIGRATION_PHASE_FINISH3:
/* migration finished, we started resuming the domain but didn't
* confirm success or failure yet; killing it seems safest unless
* we already started guest CPUs or we were in post-copy mode */
if (virDomainObjIsPostcopy(vm, job))
return 1;
if (state != VIR_DOMAIN_RUNNING) {
VIR_DEBUG("Killing migrated domain %s", vm->def->name);
return -1;
}
break;
case QEMU_MIGRATION_PHASE_POSTCOPY_FAILED:
case QEMU_MIGRATION_PHASE_PREPARE_RESUME:
case QEMU_MIGRATION_PHASE_FINISH_RESUME:
return 1;
}
return 0;
}
/*
* Returns
* -1 the domain should be killed (either after a successful migration or
* on error),
* 0 the domain should remain running with the migration job discarded,
* 1 the daemon was restarted during post-copy phase
*/
static int
qemuProcessRecoverMigrationOut(virQEMUDriver *driver,
virDomainObj *vm,
virDomainJobObj *job,
virDomainJobStatus migStatus,
virDomainState state,
int reason,
unsigned int *stopFlags)
{
bool postcopy = virDomainObjIsPostcopy(vm, job);
bool resume = false;
VIR_DEBUG("Active outgoing migration in phase %s",
qemuMigrationJobPhaseTypeToString(job->phase));
switch ((qemuMigrationJobPhase) job->phase) {
case QEMU_MIGRATION_PHASE_NONE:
case QEMU_MIGRATION_PHASE_PREPARE:
case QEMU_MIGRATION_PHASE_FINISH2:
case QEMU_MIGRATION_PHASE_FINISH3:
case QEMU_MIGRATION_PHASE_PREPARE_RESUME:
case QEMU_MIGRATION_PHASE_FINISH_RESUME:
case QEMU_MIGRATION_PHASE_LAST:
/* N/A for outgoing migration */
break;
case QEMU_MIGRATION_PHASE_BEGIN3:
/* nothing happened so far, just forget we were about to migrate the
* domain */
break;
case QEMU_MIGRATION_PHASE_PERFORM2:
case QEMU_MIGRATION_PHASE_PERFORM3:
/* migration is still in progress, let's cancel it and resume the
* domain; we can do so even in post-copy phase as the domain was not
* resumed on the destination host yet
*/
VIR_DEBUG("Cancelling unfinished migration of domain %s",
vm->def->name);
if (qemuMigrationSrcCancelUnattended(vm, job) < 0) {
VIR_WARN("Could not cancel ongoing migration of domain %s",
vm->def->name);
}
resume = true;
break;
case QEMU_MIGRATION_PHASE_PERFORM3_DONE:
/* migration finished but we didn't have a chance to get the result
* of Finish3 step; third party needs to check what to do next; in
* post-copy mode we can use PAUSED_POSTCOPY_FAILED state for this
*/
if (postcopy)
return 1;
break;
case QEMU_MIGRATION_PHASE_CONFIRM3_CANCELLED:
/* Finish3 failed, we need to resume the domain, but once we enter
* post-copy mode there's no way back, so let's just mark the domain
* as broken in that case
*/
if (postcopy)
return 1;
VIR_DEBUG("Resuming domain %s after failed migration",
vm->def->name);
resume = true;
break;
case QEMU_MIGRATION_PHASE_CONFIRM3:
/* migration completed, we need to kill the domain here */
*stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
return -1;
case QEMU_MIGRATION_PHASE_CONFIRM_RESUME:
if (migStatus == VIR_DOMAIN_JOB_STATUS_HYPERVISOR_COMPLETED) {
/* migration completed, we need to kill the domain here */
*stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
return -1;
}
return 1;
case QEMU_MIGRATION_PHASE_POSTCOPY_FAILED:
case QEMU_MIGRATION_PHASE_BEGIN_RESUME:
case QEMU_MIGRATION_PHASE_PERFORM_RESUME:
return 1;
}
if (resume) {
/* resume the domain but only if it was paused as a result of
* migration
*/
if (state == VIR_DOMAIN_PAUSED &&
(reason == VIR_DOMAIN_PAUSED_MIGRATION ||
reason == VIR_DOMAIN_PAUSED_UNKNOWN)) {
if (qemuProcessStartCPUs(driver, vm,
VIR_DOMAIN_RUNNING_MIGRATION_CANCELED,
VIR_ASYNC_JOB_NONE) < 0) {
VIR_WARN("Could not resume domain %s", vm->def->name);
}
}
}
return 0;
}
static int
qemuProcessRecoverMigration(virQEMUDriver *driver,
virDomainObj *vm,
virDomainJobObj *job,
unsigned int *stopFlags)
{
virDomainJobStatus migStatus = VIR_DOMAIN_JOB_STATUS_NONE;
qemuDomainJobPrivate *jobPriv = job->privateData;
qemuDomainObjPrivate *priv = vm->privateData;
virDomainState state;
int reason;
int rc;
state = virDomainObjGetState(vm, &reason);
qemuMigrationAnyRefreshStatus(vm, VIR_ASYNC_JOB_NONE, &migStatus);
if (job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT) {
rc = qemuProcessRecoverMigrationOut(driver, vm, job, migStatus,
state, reason, stopFlags);
} else {
rc = qemuProcessRecoverMigrationIn(driver, vm, job, state);
}
if (rc < 0)
return -1;
if (rc > 0) {
job->phase = QEMU_MIGRATION_PHASE_POSTCOPY_FAILED;
/* Even though we restore the migration async job here, the APIs below
* use VIR_ASYNC_JOB_NONE because we're already in a MODIFY job started
* before we reconnected to the domain. */
qemuProcessRestoreMigrationJob(vm, job);
if (migStatus == VIR_DOMAIN_JOB_STATUS_POSTCOPY) {
VIR_DEBUG("Post-copy migration of domain %s still running, it will be handled as unattended",
vm->def->name);
vm->job->asyncPaused = false;
return 0;
}
if (migStatus != VIR_DOMAIN_JOB_STATUS_HYPERVISOR_COMPLETED) {
if (job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT)
qemuMigrationSrcPostcopyFailed(vm);
else
qemuMigrationDstPostcopyFailed(vm);
/* Set the asyncPaused flag in case we're reconnecting to a domain
* started by an older libvirt. */
vm->job->asyncPaused = true;
return 0;
}
VIR_DEBUG("Post-copy migration of domain %s already finished",
vm->def->name);
if (job->asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT) {
qemuMigrationSrcComplete(driver, vm, VIR_ASYNC_JOB_NONE);
/* No need to stop the restored job as the domain has just been
* destroyed. */
} else {
qemuMigrationDstComplete(driver, vm, true, VIR_ASYNC_JOB_NONE, job);
virDomainObjEndAsyncJob(vm);
}
return 0;
}
qemuMigrationParamsReset(vm, VIR_ASYNC_JOB_NONE,
jobPriv->migParams, job->apiFlags);
qemuDomainSetMaxMemLock(vm, 0, &priv->preMigrationMemlock);
return 0;
}
static void
qemuProcessAbortSnapshotDelete(virDomainObj *vm,
virDomainJobObj *job)
{
size_t i;
qemuDomainObjPrivate *priv = vm->privateData;
qemuDomainJobPrivate *jobPriv = job->privateData;
if (!jobPriv->snapshotDelete)
return;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
g_autoptr(qemuBlockJobData) diskJob = qemuBlockJobDiskGetJob(disk);
if (!diskJob)
continue;
if (diskJob->type != QEMU_BLOCKJOB_TYPE_COMMIT &&
diskJob->type != QEMU_BLOCKJOB_TYPE_ACTIVE_COMMIT) {
continue;
}
qemuBlockJobSyncBegin(diskJob);
qemuDomainObjEnterMonitor(vm);
ignore_value(qemuMonitorBlockJobCancel(priv->mon, diskJob->name, false));
qemuDomainObjExitMonitor(vm);
diskJob->state = QEMU_BLOCKJOB_STATE_ABORTING;
qemuBlockJobSyncEnd(vm, diskJob, VIR_ASYNC_JOB_NONE);
}
}
static int
qemuProcessRecoverJob(virQEMUDriver *driver,
virDomainObj *vm,
virDomainJobObj *job,
unsigned int *stopFlags)
{
virDomainState state;
int reason;
state = virDomainObjGetState(vm, &reason);
VIR_DEBUG("Recovering job for domain %s, state=%s(%s), async=%s, job=%s",
vm->def->name,
virDomainStateTypeToString(state),
virDomainStateReasonToString(state, reason),
virDomainAsyncJobTypeToString(job->asyncJob),
virDomainJobTypeToString(job->active));
switch (job->asyncJob) {
case VIR_ASYNC_JOB_MIGRATION_OUT:
case VIR_ASYNC_JOB_MIGRATION_IN:
if (qemuProcessRecoverMigration(driver, vm, job, stopFlags) < 0)
return -1;
break;
case VIR_ASYNC_JOB_SAVE:
case VIR_ASYNC_JOB_DUMP:
case VIR_ASYNC_JOB_SNAPSHOT:
qemuMigrationSrcCancel(vm, VIR_ASYNC_JOB_NONE, false);
/* resume the domain but only if it was paused as a result of
* running a migration-to-file operation. Although we are
* recovering an async job, this function is run at startup
* and must resume things using sync monitor connections. */
if (state == VIR_DOMAIN_PAUSED &&
((job->asyncJob == VIR_ASYNC_JOB_DUMP &&
reason == VIR_DOMAIN_PAUSED_DUMP) ||
(job->asyncJob == VIR_ASYNC_JOB_SAVE &&
reason == VIR_DOMAIN_PAUSED_SAVE) ||
(job->asyncJob == VIR_ASYNC_JOB_SNAPSHOT &&
(reason == VIR_DOMAIN_PAUSED_SNAPSHOT ||
reason == VIR_DOMAIN_PAUSED_MIGRATION)) ||
reason == VIR_DOMAIN_PAUSED_UNKNOWN)) {
if (qemuProcessStartCPUs(driver, vm,
VIR_DOMAIN_RUNNING_SAVE_CANCELED,
VIR_ASYNC_JOB_NONE) < 0) {
VIR_WARN("Could not resume domain '%s' after migration to file",
vm->def->name);
}
}
qemuProcessAbortSnapshotDelete(vm, job);
break;
case VIR_ASYNC_JOB_START:
/* Already handled in VIR_DOMAIN_PAUSED_STARTING_UP check. */
break;
case VIR_ASYNC_JOB_BACKUP:
/* Restore the config of the async job which is not persisted */
qemuDomainObjRestoreAsyncJob(vm, VIR_ASYNC_JOB_BACKUP, 0,
job->asyncStarted,
VIR_DOMAIN_JOB_OPERATION_BACKUP,
QEMU_DOMAIN_JOB_STATS_TYPE_BACKUP,
VIR_DOMAIN_JOB_STATUS_ACTIVE,
(VIR_JOB_DEFAULT_MASK |
JOB_MASK(VIR_JOB_SUSPEND) |
JOB_MASK(VIR_JOB_MODIFY)));
break;
case VIR_ASYNC_JOB_NONE:
case VIR_ASYNC_JOB_LAST:
break;
}
if (!virDomainObjIsActive(vm))
return -1;
/* In case any special handling is added for job type that has been ignored
* before, VIR_DOMAIN_TRACK_JOBS (from qemu_domain.h) needs to be updated
* for the job to be properly tracked in domain state XML.
*/
switch (job->active) {
case VIR_JOB_QUERY:
/* harmless */
break;
case VIR_JOB_DESTROY:
VIR_DEBUG("Domain %s should have already been destroyed",
vm->def->name);
return -1;
case VIR_JOB_SUSPEND:
/* mostly harmless */
break;
case VIR_JOB_MODIFY:
/* XXX depending on the command we may be in an inconsistent state and
* we should probably fall back to "monitor error" state and refuse to
*/
break;
case VIR_JOB_MODIFY_MIGRATION_SAFE:
/* event handlers, the reconnection code already handles them as we
* might as well just missed the event while we were not running
*/
break;
case VIR_JOB_MIGRATION_OP:
case VIR_JOB_ABORT:
case VIR_JOB_ASYNC:
case VIR_JOB_ASYNC_NESTED:
/* async job was already handled above */
case VIR_JOB_NONE:
case VIR_JOB_LAST:
break;
}
return 0;
}
static int
qemuProcessUpdateDevices(virQEMUDriver *driver,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainDeviceDef dev;
g_auto(GStrv) old = g_steal_pointer(&priv->qemuDevices);
GStrv tmp;
if (qemuDomainUpdateDeviceList(vm, VIR_ASYNC_JOB_NONE) < 0)
return -1;
if (!old)
return 0;
for (tmp = old; *tmp; tmp++) {
if (!g_strv_contains((const char **) priv->qemuDevices, *tmp) &&
virDomainDefFindDevice(vm->def, *tmp, &dev, false) == 0 &&
qemuDomainRemoveDevice(driver, vm, &dev))
return -1;
}
return 0;
}
static int
qemuDomainPerfRestart(virDomainObj *vm)
{
size_t i;
virDomainDef *def = vm->def;
qemuDomainObjPrivate *priv = vm->privateData;
if (!(priv->perf = virPerfNew()))
return -1;
for (i = 0; i < VIR_PERF_EVENT_LAST; i++) {
if (def->perf.events[i] &&
def->perf.events[i] == VIR_TRISTATE_BOOL_YES) {
/* Failure to re-enable the perf event should not be fatal */
if (virPerfEventEnable(priv->perf, i, vm->pid) < 0)
def->perf.events[i] = VIR_TRISTATE_BOOL_NO;
}
}
return 0;
}
static bool
qemuProcessDomainMemoryDefNeedHugepagesPath(const virDomainMemoryDef *mem,
const long system_pagesize)
{
unsigned long long pagesize = 0;
switch (mem->model) {
case VIR_DOMAIN_MEMORY_MODEL_DIMM:
pagesize = mem->source.dimm.pagesize;
break;
case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_MEM:
pagesize = mem->source.virtio_mem.pagesize;
break;
case VIR_DOMAIN_MEMORY_MODEL_NONE:
case VIR_DOMAIN_MEMORY_MODEL_NVDIMM:
case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_PMEM:
case VIR_DOMAIN_MEMORY_MODEL_SGX_EPC:
case VIR_DOMAIN_MEMORY_MODEL_LAST:
/* None of these can be backed by hugepages. */
return false;
}
return pagesize != 0 && pagesize != system_pagesize;
}
static bool
qemuProcessNeedHugepagesPath(virDomainDef *def,
virDomainMemoryDef *mem)
{
const long system_pagesize = virGetSystemPageSizeKB();
size_t i;
switch ((virDomainMemorySource)def->mem.source) {
case VIR_DOMAIN_MEMORY_SOURCE_FILE:
/* This needs a hugetlbfs mount. */
return true;
case VIR_DOMAIN_MEMORY_SOURCE_MEMFD:
/* memfd works without a hugetlbfs mount */
return false;
case VIR_DOMAIN_MEMORY_SOURCE_NONE:
case VIR_DOMAIN_MEMORY_SOURCE_ANONYMOUS:
case VIR_DOMAIN_MEMORY_SOURCE_LAST:
break;
}
for (i = 0; i < def->mem.nhugepages; i++) {
if (def->mem.hugepages[i].size != system_pagesize)
return true;
}
for (i = 0; i < def->nmems; i++) {
if (qemuProcessDomainMemoryDefNeedHugepagesPath(def->mems[i], system_pagesize))
return true;
}
if (mem &&
qemuProcessDomainMemoryDefNeedHugepagesPath(mem, system_pagesize))
return true;
return false;
}
static bool
qemuProcessNeedMemoryBackingPath(virDomainDef *def,
virDomainMemoryDef *mem)
{
size_t i;
size_t numaNodes;
if (def->mem.source == VIR_DOMAIN_MEMORY_SOURCE_FILE ||
def->mem.access != VIR_DOMAIN_MEMORY_ACCESS_DEFAULT)
return true;
numaNodes = virDomainNumaGetNodeCount(def->numa);
for (i = 0; i < numaNodes; i++) {
if (virDomainNumaGetNodeMemoryAccessMode(def->numa, i)
!= VIR_DOMAIN_MEMORY_ACCESS_DEFAULT)
return true;
}
for (i = 0; i < def->nmems; i++) {
if (def->mems[i]->access != VIR_DOMAIN_MEMORY_ACCESS_DEFAULT)
return true;
}
if (mem) {
switch (mem->model) {
case VIR_DOMAIN_MEMORY_MODEL_DIMM:
case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_MEM:
if (mem->access != VIR_DOMAIN_MEMORY_ACCESS_DEFAULT) {
/* No need to check for access mode on the target node,
* it was checked for in the previous loop. */
return true;
}
break;
case VIR_DOMAIN_MEMORY_MODEL_NONE:
case VIR_DOMAIN_MEMORY_MODEL_NVDIMM:
case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_PMEM:
case VIR_DOMAIN_MEMORY_MODEL_SGX_EPC:
case VIR_DOMAIN_MEMORY_MODEL_LAST:
/* Backed by user provided path. Not stored in memory
* backing dir anyway. */
break;
}
}
return false;
}
static int
qemuProcessBuildDestroyMemoryPathsImpl(virQEMUDriver *driver,
virDomainObj *vm,
const char *path,
bool build)
{
if (build) {
if (virFileExists(path))
return 0;
if (g_mkdir_with_parents(path, 0700) < 0) {
virReportSystemError(errno,
_("Unable to create %1$s"),
path);
return -1;
}
if (qemuDomainNamespaceSetupPath(vm, path, NULL) < 0)
return -1;
if (qemuSecurityDomainSetPathLabel(driver, vm, path, true) < 0)
return -1;
} else {
if (virFileDeleteTree(path) < 0)
return -1;
}
return 0;
}
int
qemuProcessBuildDestroyMemoryPaths(virQEMUDriver *driver,
virDomainObj *vm,
virDomainMemoryDef *mem,
bool build)
{
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
size_t i;
bool shouldBuildHP = false;
bool shouldBuildMB = false;
if (build) {
shouldBuildHP = qemuProcessNeedHugepagesPath(vm->def, mem);
shouldBuildMB = qemuProcessNeedMemoryBackingPath(vm->def, mem);
}
if (!build || shouldBuildHP) {
for (i = 0; i < cfg->nhugetlbfs; i++) {
g_autofree char *path = NULL;
path = qemuGetDomainHugepagePath(driver, vm->def, &cfg->hugetlbfs[i]);
if (!path)
return -1;
if (build &&
qemuHugepageMakeBasedir(driver, &cfg->hugetlbfs[i]) < 0)
return -1;
if (qemuProcessBuildDestroyMemoryPathsImpl(driver, vm,
path, build) < 0)
return -1;
}
}
if (!build || shouldBuildMB) {
g_autofree char *path = NULL;
if (qemuGetMemoryBackingDomainPath(driver, vm->def, &path) < 0)
return -1;
if (qemuProcessBuildDestroyMemoryPathsImpl(driver, vm,
path, build) < 0)
return -1;
}
return 0;
}
int
qemuProcessDestroyMemoryBackingPath(virQEMUDriver *driver,
virDomainObj *vm,
virDomainMemoryDef *mem)
{
g_autofree char *path = NULL;
if (qemuGetMemoryBackingPath(driver, vm->def, mem->info.alias, &path) < 0)
return -1;
if (unlink(path) < 0 &&
errno != ENOENT) {
virReportSystemError(errno, _("Unable to remove %1$s"), path);
return -1;
}
return 0;
}
static int
qemuProcessVNCAllocatePorts(virQEMUDriver *driver,
virDomainGraphicsDef *graphics,
bool allocate)
{
unsigned short port;
if (!allocate) {
if (graphics->data.vnc.autoport)
graphics->data.vnc.port = 5900;
return 0;
}
if (graphics->data.vnc.autoport) {
if (virPortAllocatorAcquire(driver->remotePorts, &port) < 0)
return -1;
graphics->data.vnc.port = port;
graphics->data.vnc.portReserved = true;
}
if (graphics->data.vnc.websocket == -1) {
if (virPortAllocatorAcquire(driver->webSocketPorts, &port) < 0)
return -1;
graphics->data.vnc.websocket = port;
graphics->data.vnc.websocketGenerated = true;
graphics->data.vnc.websocketReserved = true;
}
return 0;
}
static int
qemuProcessSPICEAllocatePorts(virQEMUDriver *driver,
virDomainGraphicsDef *graphics,
bool allocate)
{
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
unsigned short port = 0;
unsigned short tlsPort;
size_t i;
int defaultMode = graphics->data.spice.defaultMode;
bool needTLSPort = false;
bool needPort = false;
if (graphics->data.spice.autoport) {
/* check if tlsPort or port need allocation */
for (i = 0; i < VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_LAST; i++) {
switch (graphics->data.spice.channels[i]) {
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_SECURE:
needTLSPort = true;
break;
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_INSECURE:
needPort = true;
break;
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_ANY:
/* default mode will be used */
break;
}
}
switch (defaultMode) {
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_SECURE:
needTLSPort = true;
break;
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_INSECURE:
needPort = true;
break;
case VIR_DOMAIN_GRAPHICS_SPICE_CHANNEL_MODE_ANY:
if (cfg->spiceTLS)
needTLSPort = true;
needPort = true;
break;
}
}
if (!allocate) {
if (needPort || graphics->data.spice.port == -1)
graphics->data.spice.port = 5901;
if (needTLSPort || graphics->data.spice.tlsPort == -1)
graphics->data.spice.tlsPort = 5902;
return 0;
}
if (needPort || graphics->data.spice.port == -1) {
if (virPortAllocatorAcquire(driver->remotePorts, &port) < 0)
return -1;
graphics->data.spice.port = port;
graphics->data.spice.portReserved = true;
}
if (needTLSPort || graphics->data.spice.tlsPort == -1) {
if (!cfg->spiceTLS) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("Auto allocation of spice TLS port requested but spice TLS is disabled in qemu.conf"));
return -1;
}
if (virPortAllocatorAcquire(driver->remotePorts, &tlsPort) < 0)
return -1;
graphics->data.spice.tlsPort = tlsPort;
graphics->data.spice.tlsPortReserved = true;
}
return 0;
}
static int
qemuProcessVerifyHypervFeatures(virDomainDef *def,
virCPUData *cpu)
{
size_t i;
int rc;
for (i = 0; i < VIR_DOMAIN_HYPERV_LAST; i++) {
g_autofree char *cpuFeature = NULL;
/* always supported string property */
if (i == VIR_DOMAIN_HYPERV_VENDOR_ID ||
i == VIR_DOMAIN_HYPERV_SPINLOCKS)
continue;
if (def->hyperv_features[i] != VIR_TRISTATE_SWITCH_ON)
continue;
cpuFeature = g_strdup_printf("hv-%s", virDomainHypervTypeToString(i));
rc = virCPUDataCheckFeature(cpu, cpuFeature);
if (rc < 0) {
return -1;
} else if (rc == 1) {
if (i == VIR_DOMAIN_HYPERV_STIMER) {
if (def->hyperv_stimer_direct != VIR_TRISTATE_SWITCH_ON)
continue;
rc = virCPUDataCheckFeature(cpu, VIR_CPU_x86_HV_STIMER_DIRECT);
if (rc < 0)
return -1;
else if (rc == 1)
continue;
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("host doesn't support hyperv stimer '%1$s' feature"),
"direct");
return -1;
}
continue;
}
switch ((virDomainHyperv) i) {
case VIR_DOMAIN_HYPERV_RELAXED:
case VIR_DOMAIN_HYPERV_VAPIC:
VIR_WARN("host doesn't support hyperv '%s' feature",
virDomainHypervTypeToString(i));
break;
case VIR_DOMAIN_HYPERV_VPINDEX:
case VIR_DOMAIN_HYPERV_RUNTIME:
case VIR_DOMAIN_HYPERV_SYNIC:
case VIR_DOMAIN_HYPERV_STIMER:
case VIR_DOMAIN_HYPERV_RESET:
case VIR_DOMAIN_HYPERV_FREQUENCIES:
case VIR_DOMAIN_HYPERV_REENLIGHTENMENT:
case VIR_DOMAIN_HYPERV_TLBFLUSH:
case VIR_DOMAIN_HYPERV_IPI:
case VIR_DOMAIN_HYPERV_EVMCS:
case VIR_DOMAIN_HYPERV_AVIC:
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("host doesn't support hyperv '%1$s' feature"),
virDomainHypervTypeToString(i));
return -1;
case VIR_DOMAIN_HYPERV_SPINLOCKS:
case VIR_DOMAIN_HYPERV_VENDOR_ID:
case VIR_DOMAIN_HYPERV_LAST:
break;
}
}
return 0;
}
static int
qemuProcessVerifyKVMFeatures(virDomainDef *def,
virCPUData *cpu)
{
int rc = 0;
if (def->features[VIR_DOMAIN_FEATURE_PVSPINLOCK] != VIR_TRISTATE_SWITCH_ON)
return 0;
rc = virCPUDataCheckFeature(cpu, VIR_CPU_x86_KVM_PV_UNHALT);
if (rc <= 0) {
if (rc == 0)
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("host doesn't support paravirtual spinlocks"));
return -1;
}
return 0;
}
static int
qemuProcessVerifyCPUFeatures(virDomainDef *def,
virCPUData *cpu)
{
int rc;
rc = virCPUCheckFeature(def->os.arch, def->cpu, "invtsc");
if (rc < 0) {
return -1;
} else if (rc == 1) {
rc = virCPUDataCheckFeature(cpu, "invtsc");
if (rc <= 0) {
if (rc == 0) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("host doesn't support invariant TSC"));
}
return -1;
}
}
return 0;
}
/* returns the QOM path to the first vcpu */
static const char *
qemuProcessGetVCPUQOMPath(virDomainObj *vm)
{
virDomainVcpuDef *vcpu = virDomainDefGetVcpu(vm->def, 0);
qemuDomainVcpuPrivate *vcpupriv;
if (vcpu &&
(vcpupriv = QEMU_DOMAIN_VCPU_PRIVATE(vcpu)) &&
vcpupriv->qomPath)
return vcpupriv->qomPath;
return "/machine/unattached/device[0]";
}
static int
qemuProcessFetchGuestCPU(virDomainObj *vm,
virDomainAsyncJob asyncJob,
virCPUData **enabled,
virCPUData **disabled)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virCPUData) dataEnabled = NULL;
g_autoptr(virCPUData) dataDisabled = NULL;
const char *cpuQOMPath = qemuProcessGetVCPUQOMPath(vm);
bool generic;
int rc;
*enabled = NULL;
*disabled = NULL;
generic = virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_CPU_UNAVAILABLE_FEATURES);
if (!generic && !ARCH_IS_X86(vm->def->os.arch))
return 0;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
if (generic) {
rc = qemuMonitorGetGuestCPU(priv->mon,
vm->def->os.arch,
cpuQOMPath,
virQEMUCapsCPUFeatureFromQEMU,
&dataEnabled, &dataDisabled);
} else {
rc = qemuMonitorGetGuestCPUx86(priv->mon, cpuQOMPath, &dataEnabled, &dataDisabled);
}
qemuDomainObjExitMonitor(vm);
if (rc == -1)
return -1;
*enabled = g_steal_pointer(&dataEnabled);
*disabled = g_steal_pointer(&dataDisabled);
return 0;
}
static int
qemuProcessVerifyCPU(virDomainObj *vm,
virCPUData *cpu)
{
virDomainDef *def = vm->def;
if (!cpu)
return 0;
if (qemuProcessVerifyKVMFeatures(def, cpu) < 0 ||
qemuProcessVerifyHypervFeatures(def, cpu) < 0)
return -1;
if (!def->cpu ||
(def->cpu->mode == VIR_CPU_MODE_CUSTOM &&
!def->cpu->model))
return 0;
if (qemuProcessVerifyCPUFeatures(def, cpu) < 0)
return -1;
return 0;
}
static int
qemuProcessUpdateLiveGuestCPU(virDomainObj *vm,
virCPUData *enabled,
virCPUData *disabled)
{
virDomainDef *def = vm->def;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virCPUDef) orig = NULL;
int rc;
if (!enabled)
return 0;
if (!def->cpu ||
(def->cpu->mode == VIR_CPU_MODE_CUSTOM &&
!def->cpu->model))
return 0;
orig = virCPUDefCopy(def->cpu);
if ((rc = virCPUUpdateLive(def->os.arch, def->cpu, enabled, disabled)) < 0) {
return -1;
} else if (rc == 0) {
/* Store the original CPU in priv if QEMU changed it and we didn't
* get the original CPU via migration, restore, or snapshot revert.
*/
if (!priv->origCPU && !virCPUDefIsEqual(def->cpu, orig, false))
priv->origCPU = g_steal_pointer(&orig);
def->cpu->check = VIR_CPU_CHECK_FULL;
}
return 0;
}
static int
qemuProcessUpdateAndVerifyCPU(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
g_autoptr(virCPUData) cpu = NULL;
g_autoptr(virCPUData) disabled = NULL;
if (qemuProcessFetchGuestCPU(vm, asyncJob, &cpu, &disabled) < 0)
return -1;
if (qemuProcessVerifyCPU(vm, cpu) < 0)
return -1;
if (qemuProcessUpdateLiveGuestCPU(vm, cpu, disabled) < 0)
return -1;
return 0;
}
static int
qemuProcessFetchCPUDefinitions(virDomainObj *vm,
virDomainAsyncJob asyncJob,
virDomainCapsCPUModels **cpuModels)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virDomainCapsCPUModels) models = NULL;
int rc;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = virQEMUCapsFetchCPUModels(priv->mon, vm->def->os.arch, &models);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
*cpuModels = g_steal_pointer(&models);
return 0;
}
static int
qemuProcessUpdateCPU(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
g_autoptr(virCPUData) cpu = NULL;
g_autoptr(virCPUData) disabled = NULL;
g_autoptr(virDomainCapsCPUModels) models = NULL;
/* The host CPU model comes from host caps rather than QEMU caps so
* fallback must be allowed no matter what the user specified in the XML.
*/
vm->def->cpu->fallback = VIR_CPU_FALLBACK_ALLOW;
if (qemuProcessFetchGuestCPU(vm, asyncJob, &cpu, &disabled) < 0)
return -1;
if (qemuProcessUpdateLiveGuestCPU(vm, cpu, disabled) < 0)
return -1;
if (qemuProcessFetchCPUDefinitions(vm, asyncJob, &models) < 0 ||
virCPUTranslate(vm->def->os.arch, vm->def->cpu, models) < 0)
return -1;
return 0;
}
struct qemuPrepareNVRAMHelperData {
int srcFD;
const char *srcPath;
};
static int
qemuPrepareNVRAMHelper(int dstFD,
const char *dstPath,
const void *opaque)
{
const struct qemuPrepareNVRAMHelperData *data = opaque;
ssize_t r;
do {
char buf[1024];
if ((r = saferead(data->srcFD, buf, sizeof(buf))) < 0) {
virReportSystemError(errno,
_("Unable to read from file '%1$s'"),
data->srcPath);
return -2;
}
if (safewrite(dstFD, buf, r) < 0) {
virReportSystemError(errno,
_("Unable to write to file '%1$s'"),
dstPath);
return -1;
}
} while (r);
return 0;
}
static int
qemuPrepareNVRAM(virQEMUDriver *driver,
virDomainObj *vm,
bool reset_nvram)
{
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
VIR_AUTOCLOSE srcFD = -1;
virDomainLoaderDef *loader = vm->def->os.loader;
struct qemuPrepareNVRAMHelperData data;
if (!loader || !loader->nvram)
return 0;
if (!virStorageSourceIsLocalStorage(loader->nvram)) {
if (!reset_nvram) {
return 0;
} else {
virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s",
_("resetting of nvram is not supported with network backed nvram"));
return -1;
}
}
if (virFileExists(loader->nvram->path) && !reset_nvram)
return 0;
if (!loader->nvramTemplate) {
virReportError(VIR_ERR_OPERATION_FAILED,
_("unable to find any master var store for loader: %1$s"),
loader->path);
return -1;
}
if ((srcFD = virFileOpenAs(loader->nvramTemplate, O_RDONLY,
0, -1, -1, 0)) < 0) {
virReportSystemError(-srcFD,
_("Failed to open file '%1$s'"),
loader->nvramTemplate);
return -1;
}
data.srcFD = srcFD;
data.srcPath = loader->nvramTemplate;
if (virFileRewrite(loader->nvram->path,
S_IRUSR | S_IWUSR,
cfg->user, cfg->group,
qemuPrepareNVRAMHelper,
&data) < 0) {
return -1;
}
return 0;
}
static void
qemuLogOperation(virDomainObj *vm,
const char *msg,
virCommand *cmd,
qemuLogContext *logCtxt)
{
g_autofree char *timestamp = NULL;
qemuDomainObjPrivate *priv = vm->privateData;
int qemuVersion = virQEMUCapsGetVersion(priv->qemuCaps);
const char *package = virQEMUCapsGetPackage(priv->qemuCaps);
g_autofree char *hostname = virGetHostname();
struct utsname uts;
uname(&uts);
if ((timestamp = virTimeStringNow()) == NULL)
return;
if (qemuLogContextWrite(logCtxt,
"%s: %s %s, qemu version: %d.%d.%d%s, kernel: %s, hostname: %s\n",
timestamp, msg, VIR_LOG_VERSION_STRING,
(qemuVersion / 1000000) % 1000,
(qemuVersion / 1000) % 1000,
qemuVersion % 1000,
NULLSTR_EMPTY(package),
uts.release,
NULLSTR_EMPTY(hostname)) < 0)
return;
if (cmd) {
g_autofree char *args = virCommandToString(cmd, true);
qemuLogContextWrite(logCtxt, "%s\n", args);
}
}
void
qemuProcessIncomingDefFree(qemuProcessIncomingDef *inc)
{
if (!inc)
return;
g_free(inc->address);
g_free(inc->uri);
g_free(inc);
}
/*
* This function does not copy @path, the caller is responsible for keeping
* the @path pointer valid during the lifetime of the allocated
* qemuProcessIncomingDef structure.
*
* The caller is responsible for closing @fd, calling
* qemuProcessIncomingDefFree will NOT close it.
*/
qemuProcessIncomingDef *
qemuProcessIncomingDefNew(virQEMUCaps *qemuCaps,
const char *listenAddress,
const char *migrateFrom,
int fd,
const char *path)
{
qemuProcessIncomingDef *inc = NULL;
if (qemuMigrationDstCheckProtocol(qemuCaps, migrateFrom) < 0)
return NULL;
inc = g_new0(qemuProcessIncomingDef, 1);
inc->address = g_strdup(listenAddress);
inc->uri = qemuMigrationDstGetURI(migrateFrom, fd);
if (!inc->uri)
goto error;
inc->fd = fd;
inc->path = path;
return inc;
error:
qemuProcessIncomingDefFree(inc);
return NULL;
}
/*
* This function starts a new VIR_ASYNC_JOB_START async job. The user is
* responsible for calling qemuProcessEndJob to stop this job and for passing
* VIR_ASYNC_JOB_START as @asyncJob argument to any function requiring this
* parameter between qemuProcessBeginJob and qemuProcessEndJob.
*/
int
qemuProcessBeginJob(virDomainObj *vm,
virDomainJobOperation operation,
unsigned int apiFlags)
{
if (virDomainObjBeginAsyncJob(vm, VIR_ASYNC_JOB_START,
operation, apiFlags) < 0)
return -1;
qemuDomainObjSetAsyncJobMask(vm, VIR_JOB_NONE);
return 0;
}
void
qemuProcessEndJob(virDomainObj *vm)
{
virDomainObjEndAsyncJob(vm);
}
static int
qemuProcessStartHook(virQEMUDriver *driver,
virDomainObj *vm,
virHookQemuOpType op,
virHookSubopType subop)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autofree char *xml = NULL;
int ret;
if (!virHookPresent(VIR_HOOK_DRIVER_QEMU))
return 0;
if (!(xml = qemuDomainDefFormatXML(driver, priv->qemuCaps, vm->def, 0)))
return -1;
ret = virHookCall(VIR_HOOK_DRIVER_QEMU, vm->def->name, op, subop,
NULL, xml, NULL);
return ret;
}
static int
qemuProcessGraphicsReservePorts(virDomainGraphicsDef *graphics,
bool reconnect)
{
virDomainGraphicsListenDef *glisten;
if (graphics->nListens <= 0)
return 0;
glisten = &graphics->listens[0];
if (glisten->type != VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_ADDRESS &&
glisten->type != VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NETWORK)
return 0;
switch (graphics->type) {
case VIR_DOMAIN_GRAPHICS_TYPE_VNC:
if (!graphics->data.vnc.autoport ||
reconnect) {
if (virPortAllocatorSetUsed(graphics->data.vnc.port) < 0)
return -1;
graphics->data.vnc.portReserved = true;
}
if (graphics->data.vnc.websocket > 0) {
if (virPortAllocatorSetUsed(graphics->data.vnc.websocket) < 0)
return -1;
graphics->data.vnc.websocketReserved = true;
}
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SPICE:
if (graphics->data.spice.autoport && !reconnect)
return 0;
if (graphics->data.spice.port > 0) {
if (virPortAllocatorSetUsed(graphics->data.spice.port) < 0)
return -1;
graphics->data.spice.portReserved = true;
}
if (graphics->data.spice.tlsPort > 0) {
if (virPortAllocatorSetUsed(graphics->data.spice.tlsPort) < 0)
return -1;
graphics->data.spice.tlsPortReserved = true;
}
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SDL:
case VIR_DOMAIN_GRAPHICS_TYPE_RDP:
case VIR_DOMAIN_GRAPHICS_TYPE_DESKTOP:
case VIR_DOMAIN_GRAPHICS_TYPE_EGL_HEADLESS:
case VIR_DOMAIN_GRAPHICS_TYPE_DBUS:
case VIR_DOMAIN_GRAPHICS_TYPE_LAST:
break;
}
return 0;
}
static int
qemuProcessGraphicsAllocatePorts(virQEMUDriver *driver,
virDomainGraphicsDef *graphics,
bool allocate)
{
virDomainGraphicsListenDef *glisten;
if (graphics->nListens <= 0)
return 0;
glisten = &graphics->listens[0];
if (glisten->type != VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_ADDRESS &&
glisten->type != VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NETWORK)
return 0;
switch (graphics->type) {
case VIR_DOMAIN_GRAPHICS_TYPE_VNC:
if (qemuProcessVNCAllocatePorts(driver, graphics, allocate) < 0)
return -1;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SPICE:
if (qemuProcessSPICEAllocatePorts(driver, graphics, allocate) < 0)
return -1;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SDL:
case VIR_DOMAIN_GRAPHICS_TYPE_RDP:
case VIR_DOMAIN_GRAPHICS_TYPE_DESKTOP:
case VIR_DOMAIN_GRAPHICS_TYPE_EGL_HEADLESS:
case VIR_DOMAIN_GRAPHICS_TYPE_DBUS:
case VIR_DOMAIN_GRAPHICS_TYPE_LAST:
break;
}
return 0;
}
static int
qemuProcessGetNetworkAddress(const char *netname,
char **netaddr)
{
g_autoptr(virConnect) conn = NULL;
g_autoptr(virNetwork) net = NULL;
g_autoptr(virNetworkDef) netdef = NULL;
virNetworkIPDef *ipdef;
virSocketAddr addr;
virSocketAddr *addrptr = NULL;
char *dev_name = NULL;
g_autofree char *xml = NULL;
*netaddr = NULL;
if (!(conn = virGetConnectNetwork()))
return -1;
net = virNetworkLookupByName(conn, netname);
if (!net)
return -1;
xml = virNetworkGetXMLDesc(net, 0);
if (!xml)
return -1;
netdef = virNetworkDefParse(xml, NULL, NULL, false);
if (!netdef)
return -1;
switch ((virNetworkForwardType) netdef->forward.type) {
case VIR_NETWORK_FORWARD_NONE:
case VIR_NETWORK_FORWARD_NAT:
case VIR_NETWORK_FORWARD_ROUTE:
case VIR_NETWORK_FORWARD_OPEN:
ipdef = virNetworkDefGetIPByIndex(netdef, AF_UNSPEC, 0);
if (!ipdef) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("network '%1$s' doesn't have an IP address"),
netdef->name);
return -1;
}
addrptr = &ipdef->address;
break;
case VIR_NETWORK_FORWARD_BRIDGE:
if ((dev_name = netdef->bridge))
break;
/*
* fall through if netdef->bridge wasn't set, since that is
* macvtap bridge mode network.
*/
G_GNUC_FALLTHROUGH;
case VIR_NETWORK_FORWARD_PRIVATE:
case VIR_NETWORK_FORWARD_VEPA:
case VIR_NETWORK_FORWARD_PASSTHROUGH:
if ((netdef->forward.nifs > 0) && netdef->forward.ifs)
dev_name = netdef->forward.ifs[0].device.dev;
if (!dev_name) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("network '%1$s' has no associated interface or bridge"),
netdef->name);
return -1;
}
break;
case VIR_NETWORK_FORWARD_HOSTDEV:
break;
case VIR_NETWORK_FORWARD_LAST:
default:
virReportEnumRangeError(virNetworkForwardType, netdef->forward.type);
return -1;
}
if (dev_name) {
if (virNetDevIPAddrGet(dev_name, &addr) < 0)
return -1;
addrptr = &addr;
}
if (!(addrptr &&
(*netaddr = virSocketAddrFormat(addrptr)))) {
return -1;
}
return 0;
}
static int
qemuProcessGraphicsSetupNetworkAddress(virDomainGraphicsListenDef *glisten,
const char *listenAddr)
{
int rc;
/* TODO: reject configuration without network specified for network listen */
if (!glisten->network) {
glisten->address = g_strdup(listenAddr);
return 0;
}
rc = qemuProcessGetNetworkAddress(glisten->network, &glisten->address);
if (rc <= -2) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("network-based listen isn't possible, network driver isn't present"));
return -1;
}
if (rc < 0)
return -1;
return 0;
}
static int
qemuProcessGraphicsSetupDBus(virQEMUDriver *driver,
virDomainGraphicsDef *graphics,
virDomainObj *vm)
{
if (graphics->type != VIR_DOMAIN_GRAPHICS_TYPE_DBUS)
return 0;
if (!graphics->data.dbus.p2p && !graphics->data.dbus.address) {
graphics->data.dbus.address = qemuDBusGetAddress(driver, vm);
}
return 0;
}
static int
qemuProcessGraphicsSetupListen(virQEMUDriver *driver,
virDomainGraphicsDef *graphics,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
const char *type = virDomainGraphicsTypeToString(graphics->type);
char *listenAddr = NULL;
bool useSocket = false;
size_t i;
switch (graphics->type) {
case VIR_DOMAIN_GRAPHICS_TYPE_VNC:
useSocket = cfg->vncAutoUnixSocket;
listenAddr = cfg->vncListen;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SPICE:
useSocket = cfg->spiceAutoUnixSocket;
listenAddr = cfg->spiceListen;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SDL:
case VIR_DOMAIN_GRAPHICS_TYPE_RDP:
case VIR_DOMAIN_GRAPHICS_TYPE_DESKTOP:
case VIR_DOMAIN_GRAPHICS_TYPE_EGL_HEADLESS:
case VIR_DOMAIN_GRAPHICS_TYPE_DBUS:
case VIR_DOMAIN_GRAPHICS_TYPE_LAST:
break;
}
for (i = 0; i < graphics->nListens; i++) {
virDomainGraphicsListenDef *glisten = &graphics->listens[i];
switch (glisten->type) {
case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_ADDRESS:
if (!glisten->address) {
/* If there is no address specified and qemu.conf has
* *_auto_unix_socket set we should use unix socket as
* default instead of tcp listen. */
if (useSocket) {
memset(glisten, 0, sizeof(*glisten));
glisten->socket = g_strdup_printf("%s/%s.sock", priv->libDir,
type);
glisten->fromConfig = true;
glisten->type = VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_SOCKET;
} else if (listenAddr) {
glisten->address = g_strdup(listenAddr);
glisten->fromConfig = true;
}
}
break;
case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NETWORK:
if (glisten->address || !listenAddr)
continue;
if (qemuProcessGraphicsSetupNetworkAddress(glisten,
listenAddr) < 0)
return -1;
break;
case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_SOCKET:
if (!glisten->socket) {
glisten->socket = g_strdup_printf("%s/%s.sock", priv->libDir,
type);
glisten->autoGenerated = true;
}
break;
case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NONE:
case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_LAST:
break;
}
}
return 0;
}
static int
qemuProcessGraphicsSetupRenderNode(virDomainGraphicsDef *graphics,
virQEMUCaps *qemuCaps)
{
char **rendernode = NULL;
if (!virDomainGraphicsNeedsAutoRenderNode(graphics))
return 0;
/* Don't bother picking a DRM node if QEMU doesn't support it. */
switch (graphics->type) {
case VIR_DOMAIN_GRAPHICS_TYPE_SPICE:
if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_SPICE_RENDERNODE))
return 0;
rendernode = &graphics->data.spice.rendernode;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_EGL_HEADLESS:
if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_EGL_HEADLESS_RENDERNODE))
return 0;
rendernode = &graphics->data.egl_headless.rendernode;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_DBUS:
rendernode = &graphics->data.dbus.rendernode;
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SDL:
case VIR_DOMAIN_GRAPHICS_TYPE_VNC:
case VIR_DOMAIN_GRAPHICS_TYPE_RDP:
case VIR_DOMAIN_GRAPHICS_TYPE_DESKTOP:
case VIR_DOMAIN_GRAPHICS_TYPE_LAST:
virReportEnumRangeError(virDomainGraphicsType, graphics->type);
break;
}
if (!(*rendernode = virHostGetDRMRenderNode()))
return -1;
return 0;
}
static int
qemuProcessSetupGraphics(virQEMUDriver *driver,
virDomainObj *vm,
virQEMUCaps *qemuCaps,
unsigned int flags)
{
virDomainGraphicsDef *graphics;
bool allocate = !(flags & VIR_QEMU_PROCESS_START_PRETEND);
size_t i;
for (i = 0; i < vm->def->ngraphics; i++) {
graphics = vm->def->graphics[i];
if (qemuProcessGraphicsSetupRenderNode(graphics, qemuCaps) < 0)
return -1;
if (qemuProcessGraphicsSetupListen(driver, graphics, vm) < 0)
return -1;
if (qemuProcessGraphicsSetupDBus(driver, graphics, vm) < 0)
return -1;
}
if (allocate) {
for (i = 0; i < vm->def->ngraphics; i++) {
graphics = vm->def->graphics[i];
if (qemuProcessGraphicsReservePorts(graphics, false) < 0)
return -1;
}
}
for (i = 0; i < vm->def->ngraphics; ++i) {
graphics = vm->def->graphics[i];
if (qemuProcessGraphicsAllocatePorts(driver, graphics, allocate) < 0)
return -1;
}
return 0;
}
static int
qemuProcessSetupRawIO(virDomainObj *vm,
virCommand *cmd G_GNUC_UNUSED)
{
bool rawio = false;
size_t i;
int ret = -1;
/* in case a certain disk is desirous of CAP_SYS_RAWIO, add this */
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
if (disk->rawio == VIR_TRISTATE_BOOL_YES) {
rawio = true;
#ifndef CAP_SYS_RAWIO
break;
#endif
}
}
/* If rawio not already set, check hostdevs as well */
if (!rawio) {
for (i = 0; i < vm->def->nhostdevs; i++) {
virDomainHostdevSubsysSCSI *scsisrc;
if (!virHostdevIsSCSIDevice(vm->def->hostdevs[i]))
continue;
scsisrc = &vm->def->hostdevs[i]->source.subsys.u.scsi;
if (scsisrc->rawio == VIR_TRISTATE_BOOL_YES) {
rawio = true;
break;
}
}
}
ret = 0;
if (rawio) {
#ifdef CAP_SYS_RAWIO
virCommandAllowCap(cmd, CAP_SYS_RAWIO);
#else
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("Raw I/O is not supported on this platform"));
ret = -1;
#endif
}
return ret;
}
static int
qemuProcessSetupBalloon(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
unsigned long long balloon = vm->def->mem.cur_balloon;
qemuDomainObjPrivate *priv = vm->privateData;
int ret = -1;
if (!virDomainDefHasMemballoon(vm->def))
return 0;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
if (vm->def->memballoon->period)
qemuMonitorSetMemoryStatsPeriod(priv->mon, vm->def->memballoon,
vm->def->memballoon->period);
if (qemuMonitorSetBalloon(priv->mon, balloon) < 0)
goto cleanup;
ret = 0;
cleanup:
qemuDomainObjExitMonitor(vm);
return ret;
}
static int
qemuProcessMakeDir(virQEMUDriver *driver,
virDomainObj *vm,
const char *path)
{
if (g_mkdir_with_parents(path, 0750) < 0) {
virReportSystemError(errno, _("Cannot create directory '%1$s'"), path);
return -1;
}
if (qemuSecurityDomainSetPathLabel(driver, vm, path, true) < 0)
return -1;
return 0;
}
static void
qemuProcessStartWarnShmem(virDomainObj *vm)
{
size_t i;
bool check_shmem = false;
bool shmem = vm->def->nshmems;
/*
* For vhost-user to work, the domain has to have some type of
* shared memory configured. We're not the proper ones to judge
* whether shared hugepages or shm are enough and will be in the
* future, so we'll just warn in case neither is configured.
* Moreover failing would give the false illusion that libvirt is
* really checking that everything works before running the domain
* and not only we are unable to do that, but it's also not our
* aim to do so.
*/
for (i = 0; i < vm->def->nnets; i++) {
if (virDomainNetGetActualType(vm->def->nets[i]) ==
VIR_DOMAIN_NET_TYPE_VHOSTUSER) {
check_shmem = true;
break;
}
}
if (!check_shmem)
return;
/*
* This check is by no means complete. We merely check
* whether there are *some* hugepages enabled and *some* NUMA
* nodes with shared memory access.
*/
if (!shmem && vm->def->mem.nhugepages) {
for (i = 0; i < virDomainNumaGetNodeCount(vm->def->numa); i++) {
if (virDomainNumaGetNodeMemoryAccessMode(vm->def->numa, i) ==
VIR_DOMAIN_MEMORY_ACCESS_SHARED) {
shmem = true;
break;
}
}
}
if (!shmem) {
VIR_WARN("Detected vhost-user interface without any shared memory, "
"the interface might not be operational");
}
}
static int
qemuProcessStartValidateGraphics(virDomainObj *vm)
{
size_t i;
for (i = 0; i < vm->def->ngraphics; i++) {
virDomainGraphicsDef *graphics = vm->def->graphics[i];
switch (graphics->type) {
case VIR_DOMAIN_GRAPHICS_TYPE_VNC:
case VIR_DOMAIN_GRAPHICS_TYPE_SPICE:
if (graphics->nListens > 1) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("QEMU does not support multiple listens for one graphics device."));
return -1;
}
break;
case VIR_DOMAIN_GRAPHICS_TYPE_SDL:
case VIR_DOMAIN_GRAPHICS_TYPE_RDP:
case VIR_DOMAIN_GRAPHICS_TYPE_DESKTOP:
case VIR_DOMAIN_GRAPHICS_TYPE_EGL_HEADLESS:
case VIR_DOMAIN_GRAPHICS_TYPE_DBUS:
case VIR_DOMAIN_GRAPHICS_TYPE_LAST:
break;
}
}
return 0;
}
static int
qemuProcessStartValidateShmem(virDomainObj *vm)
{
size_t i;
for (i = 0; i < vm->def->nshmems; i++) {
virDomainShmemDef *shmem = vm->def->shmems[i];
if (strchr(shmem->name, '/')) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("shmem name '%1$s' must not contain '/'"),
shmem->name);
return -1;
}
}
return 0;
}
static int
qemuProcessStartValidateDisks(virDomainObj *vm,
virQEMUCaps *qemuCaps)
{
size_t i;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
virStorageSource *src = disk->src;
/* This is a best effort check as we can only check if the command
* option exists, but we cannot determine whether the running QEMU
* was build with '--enable-vxhs'. */
if (src->type == VIR_STORAGE_TYPE_NETWORK &&
src->protocol == VIR_STORAGE_NET_PROTOCOL_VXHS &&
!virQEMUCapsGet(qemuCaps, QEMU_CAPS_VXHS)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("VxHS protocol is not supported with this QEMU binary"));
return -1;
}
/* PowerPC pseries based VMs do not support floppy device */
if (disk->device == VIR_DOMAIN_DISK_DEVICE_FLOPPY &&
qemuDomainIsPSeries(vm->def)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("PowerPC pseries machines do not support floppy device"));
return -1;
}
if (src->type == VIR_STORAGE_TYPE_NVME &&
!virQEMUCapsGet(qemuCaps, QEMU_CAPS_DRIVE_NVME)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("NVMe disks are not supported with this QEMU binary"));
return -1;
}
}
return 0;
}
/* 250 parts per million (ppm) is a half of NTP threshold */
#define TSC_TOLERANCE 250
static int
qemuProcessStartValidateTSC(virQEMUDriver *driver,
virDomainObj *vm)
{
size_t i;
unsigned long long freq = 0;
unsigned long long tolerance;
unsigned long long minFreq;
unsigned long long maxFreq;
virHostCPUTscInfo *tsc;
g_autoptr(virCPUDef) cpu = NULL;
for (i = 0; i < vm->def->clock.ntimers; i++) {
virDomainTimerDef *timer = vm->def->clock.timers[i];
if (timer->name == VIR_DOMAIN_TIMER_NAME_TSC &&
timer->frequency > 0) {
freq = timer->frequency;
break;
}
}
if (freq == 0)
return 0;
VIR_DEBUG("Requested TSC frequency %llu Hz", freq);
cpu = virQEMUDriverGetHostCPU(driver);
if (!cpu || !cpu->tsc) {
VIR_DEBUG("Host TSC frequency could not be probed");
return 0;
}
tsc = cpu->tsc;
tolerance = tsc->frequency * TSC_TOLERANCE / 1000000;
minFreq = tsc->frequency - tolerance;
maxFreq = tsc->frequency + tolerance;
VIR_DEBUG("Host TSC frequency %llu Hz, scaling %s, tolerance +/- %llu Hz",
tsc->frequency, virTristateBoolTypeToString(tsc->scaling),
tolerance);
if (freq >= minFreq && freq <= maxFreq) {
VIR_DEBUG("Requested TSC frequency is within tolerance interval");
return 0;
}
if (tsc->scaling == VIR_TRISTATE_BOOL_YES)
return 0;
if (tsc->scaling == VIR_TRISTATE_BOOL_ABSENT) {
VIR_DEBUG("Requested TSC frequency falls outside tolerance range and "
"scaling support is unknown, QEMU will try and possibly "
"fail later");
return 0;
}
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("Requested TSC frequency %1$llu Hz is outside tolerance range ([%2$llu, %3$llu] Hz) around host frequency %4$llu Hz and TSC scaling is not supported by the host CPU"),
freq, minFreq, maxFreq, tsc->frequency);
return -1;
}
/**
* qemuProcessStartValidate:
* @vm: domain object
* @qemuCaps: emulator capabilities
* @migration: restoration of existing state
*
* This function aggregates checks done prior to start of a VM.
*
* Flag VIR_QEMU_PROCESS_START_PRETEND tells, that we don't want to actually
* start the domain but create a valid qemu command. If some code shouldn't be
* executed in this case, make sure to check this flag.
*/
static int
qemuProcessStartValidate(virQEMUDriver *driver,
virDomainObj *vm,
virQEMUCaps *qemuCaps,
unsigned int flags)
{
if (!(flags & VIR_QEMU_PROCESS_START_PRETEND)) {
if (vm->def->virtType == VIR_DOMAIN_VIRT_KVM) {
VIR_DEBUG("Checking for KVM availability");
if (!virFileExists("/dev/kvm")) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("Domain requires KVM, but it is not available. Check that virtualization is enabled in the host BIOS, and host configuration is setup to load the kvm modules."));
return -1;
}
}
VIR_DEBUG("Checking domain and device security labels");
if (qemuSecurityCheckAllLabel(driver->securityManager, vm->def) < 0)
return -1;
}
if (virDomainDefValidate(vm->def, 0, driver->xmlopt, qemuCaps) < 0)
return -1;
if (qemuProcessStartValidateGraphics(vm) < 0)
return -1;
if (qemuProcessStartValidateShmem(vm) < 0)
return -1;
if (vm->def->cpu) {
if (virCPUValidateFeatures(vm->def->os.arch, vm->def->cpu) < 0)
return -1;
if (ARCH_IS_X86(vm->def->os.arch) &&
!virQEMUCapsGet(qemuCaps, QEMU_CAPS_CPU_UNAVAILABLE_FEATURES)) {
g_auto(GStrv) features = NULL;
int n;
if ((n = virCPUDefCheckFeatures(vm->def->cpu,
virCPUx86FeatureFilterSelectMSR,
NULL,
&features)) < 0)
return -1;
if (n > 0) {
g_autofree char *str = NULL;
str = g_strjoinv(", ", features);
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("Some features cannot be reliably used with this QEMU: %1$s"), str);
return -1;
}
}
}
if (qemuProcessStartValidateDisks(vm, qemuCaps) < 0)
return -1;
if (qemuProcessStartValidateTSC(driver, vm) < 0)
return -1;
VIR_DEBUG("Checking for any possible (non-fatal) issues");
qemuProcessStartWarnShmem(vm);
return 0;
}
static int
qemuProcessStartUpdateCustomCaps(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
qemuDomainXmlNsDef *nsdef = vm->def->namespaceData;
char **next;
int tmp;
if (cfg->capabilityfilters) {
for (next = cfg->capabilityfilters; *next; next++) {
if ((tmp = virQEMUCapsTypeFromString(*next)) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("invalid capability_filters capability '%1$s'"),
*next);
return -1;
}
virQEMUCapsClear(priv->qemuCaps, tmp);
}
}
if (nsdef) {
for (next = nsdef->capsadd; next && *next; next++) {
if ((tmp = virQEMUCapsTypeFromString(*next)) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("invalid qemu namespace capability '%1$s'"),
*next);
return -1;
}
virQEMUCapsSet(priv->qemuCaps, tmp);
}
for (next = nsdef->capsdel; next && *next; next++) {
if ((tmp = virQEMUCapsTypeFromString(*next)) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("invalid qemu namespace capability '%1$s'"),
*next);
return -1;
}
virQEMUCapsClear(priv->qemuCaps, tmp);
}
}
return 0;
}
/**
* qemuProcessPrepareQEMUCaps:
* @vm: domain object
* @qemuCapsCache: cache of QEMU capabilities
*
* Prepare the capabilities of a QEMU process for startup. This includes
* copying the caps to a static cache and potential post-processing depending
* on the configuration of the VM and startup process.
*
* Returns 0 on success, -1 on error.
*/
static int
qemuProcessPrepareQEMUCaps(virDomainObj *vm,
virFileCache *qemuCapsCache)
{
qemuDomainObjPrivate *priv = vm->privateData;
virObjectUnref(priv->qemuCaps);
if (!(priv->qemuCaps = virQEMUCapsCacheLookupCopy(qemuCapsCache,
vm->def->emulator)))
return -1;
/* Update qemu capabilities according to lists passed in via namespace */
if (qemuProcessStartUpdateCustomCaps(vm) < 0)
return -1;
/* re-process capability lockouts since we might have removed capabilities */
virQEMUCapsInitProcessCapsInterlock(priv->qemuCaps);
return 0;
}
/**
* qemuProcessInit:
*
* Prepares the domain up to the point when priv->qemuCaps is initialized. The
* function calls qemuProcessStop when needed.
*
* Flag VIR_QEMU_PROCESS_START_PRETEND tells, that we don't want to actually
* start the domain but create a valid qemu command. If some code shouldn't be
* executed in this case, make sure to check this flag.
*
* Returns 0 on success, -1 on error.
*/
int
qemuProcessInit(virQEMUDriver *driver,
virDomainObj *vm,
virCPUDef *updatedCPU,
virDomainAsyncJob asyncJob,
bool migration,
unsigned int flags)
{
qemuDomainObjPrivate *priv = vm->privateData;
int stopFlags;
virCPUDef *origCPU = NULL;
int ret = -1;
VIR_DEBUG("vm=%p name=%s id=%d migration=%d",
vm, vm->def->name, vm->def->id, migration);
VIR_DEBUG("Beginning VM startup process");
if (virDomainObjIsActive(vm)) {
virReportError(VIR_ERR_OPERATION_INVALID, "%s",
_("VM is already active"));
goto cleanup;
}
/* in case when the post parse callback failed we need to re-run it on the
* old config prior we start the VM */
if (vm->def->postParseFailed) {
VIR_DEBUG("re-running the post parse callback");
/* we don't have the private copy of qemuCaps at this point */
if (virDomainDefPostParse(vm->def, 0, driver->xmlopt, NULL) < 0)
goto cleanup;
}
VIR_DEBUG("Determining emulator version");
if (qemuProcessPrepareQEMUCaps(vm, driver->qemuCapsCache) < 0)
goto cleanup;
if (qemuDomainUpdateCPU(vm, updatedCPU, &origCPU) < 0)
goto cleanup;
if (qemuProcessStartValidate(driver, vm, priv->qemuCaps, flags) < 0)
goto cleanup;
/* Do this upfront, so any part of the startup process can add
* runtime state to vm->def that won't be persisted. This let's us
* report implicit runtime defaults in the XML, like vnc listen/socket
*/
VIR_DEBUG("Setting current domain def as transient");
if (virDomainObjSetDefTransient(driver->xmlopt, vm, priv->qemuCaps) < 0)
goto cleanup;
if (flags & VIR_QEMU_PROCESS_START_PRETEND) {
if (qemuDomainSetPrivatePaths(driver, vm) < 0) {
virDomainObjRemoveTransientDef(vm);
goto cleanup;
}
} else {
vm->def->id = qemuDriverAllocateID(driver);
qemuDomainSetFakeReboot(vm, false);
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_STARTING_UP);
if (g_atomic_int_add(&driver->nactive, 1) == 0 && driver->inhibitCallback)
driver->inhibitCallback(true, driver->inhibitOpaque);
/* Run an early hook to set-up missing devices */
if (qemuProcessStartHook(driver, vm,
VIR_HOOK_QEMU_OP_PREPARE,
VIR_HOOK_SUBOP_BEGIN) < 0)
goto stop;
if (qemuDomainSetPrivatePaths(driver, vm) < 0)
goto stop;
priv->origCPU = g_steal_pointer(&origCPU);
}
ret = 0;
cleanup:
virCPUDefFree(origCPU);
return ret;
stop:
stopFlags = VIR_QEMU_PROCESS_STOP_NO_RELABEL;
if (migration)
stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED, asyncJob, stopFlags);
goto cleanup;
}
static int
qemuProcessPrepareDomainNetwork(virDomainObj *vm)
{
virDomainDef *def = vm->def;
qemuDomainObjPrivate *priv = vm->privateData;
size_t i;
g_autoptr(virConnect) conn = NULL;
for (i = 0; i < def->nnets; i++) {
virDomainNetDef *net = def->nets[i];
virDomainNetType actualType;
/* If appropriate, grab a physical device from the configured
* network's pool of devices, or resolve bridge device name
* to the one defined in the network definition.
*/
if (net->type == VIR_DOMAIN_NET_TYPE_NETWORK) {
if (!conn && !(conn = virGetConnectNetwork()))
return -1;
if (virDomainNetAllocateActualDevice(conn, def, net) < 0)
return -1;
}
actualType = virDomainNetGetActualType(net);
if (actualType == VIR_DOMAIN_NET_TYPE_HOSTDEV &&
net->type == VIR_DOMAIN_NET_TYPE_NETWORK) {
/* Each type='hostdev' network device must also have a
* corresponding entry in the hostdevs array. For netdevs
* that are hardcoded as type='hostdev', this is already
* done by the parser, but for those allocated from a
* network / determined at runtime, we need to do it
* separately.
*/
virDomainHostdevDef *hostdev = virDomainNetGetActualHostdev(net);
virDomainHostdevSubsysPCI *pcisrc = &hostdev->source.subsys.u.pci;
if (virDomainHostdevFind(def, hostdev, NULL) >= 0) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("PCI device %1$04x:%2$02x:%3$02x.%4$x allocated from network %5$s is already in use by domain %6$s"),
pcisrc->addr.domain, pcisrc->addr.bus,
pcisrc->addr.slot, pcisrc->addr.function,
net->data.network.name, def->name);
return -1;
}
/* For hostdev present in qemuProcessPrepareDomain() phase this was
* done already, but this code runs after that, so we have to call
* it ourselves. */
if (qemuDomainPrepareHostdev(hostdev, priv) < 0)
return -1;
if (virDomainHostdevInsert(def, hostdev) < 0)
return -1;
}
}
return 0;
}
static int
qemuProcessPrepareHostNetwork(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
size_t i;
for (i = 0; i < vm->def->nnets; i++) {
virDomainNetDef *net = vm->def->nets[i];
virDomainNetType actualType = virDomainNetGetActualType(net);
if (actualType == VIR_DOMAIN_NET_TYPE_USER &&
net->backend.type == VIR_DOMAIN_NET_BACKEND_DEFAULT &&
!priv->disableSlirp &&
virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_DBUS_VMSTATE)) {
if (qemuInterfacePrepareSlirp(priv->driver, net) < 0)
return -1;
}
}
return 0;
}
struct qemuProcessSetupVcpuSchedCoreHelperData {
pid_t vcpupid;
pid_t dummypid;
};
static int
qemuProcessSetupVcpuSchedCoreHelper(pid_t ppid G_GNUC_UNUSED,
void *opaque)
{
struct qemuProcessSetupVcpuSchedCoreHelperData *data = opaque;
if (virProcessSchedCoreShareFrom(data->dummypid) < 0) {
virReportSystemError(errno,
_("unable to share scheduling cookie from %1$lld"),
(long long) data->dummypid);
return -1;
}
if (virProcessSchedCoreShareTo(data->vcpupid) < 0) {
virReportSystemError(errno,
_("unable to share scheduling cookie to %1$lld"),
(long long) data->vcpupid);
return -1;
}
return 0;
}
/**
* qemuProcessSetupVcpu:
* @vm: domain object
* @vcpuid: id of VCPU to set defaults
* @schedCore: whether to set scheduling group
*
* This function sets resource properties (cgroups, affinity, scheduler) for a
* vCPU. This function expects that the vCPU is online and the vCPU pids were
* correctly detected at the point when it's called.
*
* Returns 0 on success, -1 on error.
*/
int
qemuProcessSetupVcpu(virDomainObj *vm,
unsigned int vcpuid,
bool schedCore)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
pid_t vcpupid = qemuDomainGetVcpuPid(vm, vcpuid);
virDomainVcpuDef *vcpu = virDomainDefGetVcpu(vm->def, vcpuid);
virDomainResctrlMonDef *mon = NULL;
size_t i = 0;
if (qemuProcessSetupPid(vm, vcpupid, VIR_CGROUP_THREAD_VCPU,
vcpuid, vcpu->cpumask,
vm->def->cputune.period,
vm->def->cputune.quota,
&vcpu->sched) < 0)
return -1;
if (schedCore &&
cfg->schedCore == QEMU_SCHED_CORE_VCPUS) {
struct qemuProcessSetupVcpuSchedCoreHelperData data = { .vcpupid = vcpupid,
.dummypid = -1 };
for (i = 0; i < virDomainDefGetVcpusMax(vm->def); i++) {
pid_t temptid = qemuDomainGetVcpuPid(vm, i);
if (temptid > 0) {
data.dummypid = temptid;
break;
}
}
if (data.dummypid == -1) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Unable to find a vCPU that is online"));
return -1;
}
if (virProcessRunInFork(qemuProcessSetupVcpuSchedCoreHelper, &data) < 0)
return -1;
}
for (i = 0; i < vm->def->nresctrls; i++) {
size_t j = 0;
virDomainResctrlDef *ct = vm->def->resctrls[i];
if (virBitmapIsBitSet(ct->vcpus, vcpuid)) {
if (virResctrlAllocAddPID(ct->alloc, vcpupid) < 0)
return -1;
for (j = 0; j < ct->nmonitors; j++) {
mon = ct->monitors[j];
if (virBitmapEqual(ct->vcpus, mon->vcpus) &&
!virResctrlAllocIsEmpty(ct->alloc))
continue;
if (virBitmapIsBitSet(mon->vcpus, vcpuid)) {
if (virResctrlMonitorAddPID(mon->instance, vcpupid) < 0)
return -1;
break;
}
}
break;
}
}
return 0;
}
static int
qemuProcessSetupAllVcpusSchedCoreHelper(pid_t ppid G_GNUC_UNUSED,
void *opaque)
{
virDomainObj *vm = opaque;
size_t i;
/* Since we are setting all vCPU threads at once and from a forked off
* child, we don't need the dummy schedCoreChildPID and can create one on
* our own. */
if (virProcessSchedCoreCreate() < 0) {
virReportSystemError(errno, "%s",
_("Unable to set SCHED_CORE"));
return -1;
}
for (i = 0; i < virDomainDefGetVcpusMax(vm->def); i++) {
pid_t vcpupid = qemuDomainGetVcpuPid(vm, i);
if (vcpupid > 0 &&
virProcessSchedCoreShareTo(vcpupid) < 0) {
virReportSystemError(errno,
_("unable to share scheduling cookie to %1$lld"),
(long long) vcpupid);
return -1;
}
}
return 0;
}
static int
qemuProcessSetupVcpus(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
virDomainVcpuDef *vcpu;
unsigned int maxvcpus = virDomainDefGetVcpusMax(vm->def);
size_t i;
if ((vm->def->cputune.period || vm->def->cputune.quota) &&
!virCgroupHasController(((qemuDomainObjPrivate *) vm->privateData)->cgroup,
VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
if (!qemuDomainHasVcpuPids(vm)) {
/* If any CPU has custom affinity that differs from the
* VM default affinity, we must reject it */
for (i = 0; i < maxvcpus; i++) {
vcpu = virDomainDefGetVcpu(vm->def, i);
if (!vcpu->online)
continue;
if (vcpu->cpumask &&
!virBitmapEqual(vm->def->cpumask, vcpu->cpumask)) {
virReportError(VIR_ERR_OPERATION_INVALID, "%s",
_("cpu affinity is not supported"));
return -1;
}
}
return 0;
}
for (i = 0; i < maxvcpus; i++) {
vcpu = virDomainDefGetVcpu(vm->def, i);
if (!vcpu->online)
continue;
if (qemuProcessSetupVcpu(vm, i, false) < 0)
return -1;
}
if (cfg->schedCore == QEMU_SCHED_CORE_VCPUS &&
virProcessRunInFork(qemuProcessSetupAllVcpusSchedCoreHelper, vm) < 0)
return -1;
return 0;
}
int
qemuProcessSetupIOThread(virDomainObj *vm,
virDomainIOThreadIDDef *iothread)
{
return qemuProcessSetupPid(vm, iothread->thread_id,
VIR_CGROUP_THREAD_IOTHREAD,
iothread->iothread_id,
iothread->cpumask,
vm->def->cputune.iothread_period,
vm->def->cputune.iothread_quota,
&iothread->sched);
}
static int
qemuProcessSetupIOThreads(virDomainObj *vm)
{
size_t i;
for (i = 0; i < vm->def->niothreadids; i++) {
virDomainIOThreadIDDef *info = vm->def->iothreadids[i];
if (qemuProcessSetupIOThread(vm, info) < 0)
return -1;
}
return 0;
}
static int
qemuProcessValidateHotpluggableVcpus(virDomainDef *def)
{
virDomainVcpuDef *vcpu;
virDomainVcpuDef *subvcpu;
qemuDomainVcpuPrivate *vcpupriv;
unsigned int maxvcpus = virDomainDefGetVcpusMax(def);
size_t i = 0;
size_t j;
g_autoptr(virBitmap) ordermap = virBitmapNew(maxvcpus + 1);
/* validate:
* - all hotpluggable entities to be hotplugged have the correct data
* - vcpus belonging to a hotpluggable entity share configuration
* - order of the hotpluggable entities is unique
*/
for (i = 0; i < maxvcpus; i++) {
vcpu = virDomainDefGetVcpu(def, i);
vcpupriv = QEMU_DOMAIN_VCPU_PRIVATE(vcpu);
/* skip over hotpluggable entities */
if (vcpupriv->vcpus == 0)
continue;
if (vcpu->order != 0) {
if (virBitmapIsBitSet(ordermap, vcpu->order)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("duplicate vcpu order '%1$u'"), vcpu->order);
return -1;
}
if (virBitmapSetBit(ordermap, vcpu->order)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("vcpu order '%1$u' exceeds vcpu count"),
vcpu->order);
return -1;
}
}
for (j = i + 1; j < (i + vcpupriv->vcpus); j++) {
subvcpu = virDomainDefGetVcpu(def, j);
if (subvcpu->hotpluggable != vcpu->hotpluggable ||
subvcpu->online != vcpu->online ||
subvcpu->order != vcpu->order) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("vcpus '%1$zu' and '%2$zu' are in the same hotplug group but differ in configuration"),
i, j);
return -1;
}
}
if (vcpu->online && vcpu->hotpluggable == VIR_TRISTATE_BOOL_YES) {
if ((vcpupriv->socket_id == -1 && vcpupriv->core_id == -1 &&
vcpupriv->thread_id == -1 && vcpupriv->node_id == -1) ||
!vcpupriv->type) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("vcpu '%1$zu' is missing hotplug data"), i);
return -1;
}
}
}
return 0;
}
static int
qemuDomainHasHotpluggableStartupVcpus(virDomainDef *def)
{
size_t maxvcpus = virDomainDefGetVcpusMax(def);
virDomainVcpuDef *vcpu;
size_t i;
for (i = 0; i < maxvcpus; i++) {
vcpu = virDomainDefGetVcpu(def, i);
if (vcpu->online && vcpu->hotpluggable == VIR_TRISTATE_BOOL_YES)
return true;
}
return false;
}
static int
qemuProcessVcpusSortOrder(const void *a,
const void *b,
void *opaque G_GNUC_UNUSED)
{
virDomainVcpuDef *vcpua = *((virDomainVcpuDef **)a);
virDomainVcpuDef *vcpub = *((virDomainVcpuDef **)b);
return vcpua->order - vcpub->order;
}
static int
qemuProcessSetupHotpluggableVcpus(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
unsigned int maxvcpus = virDomainDefGetVcpusMax(vm->def);
qemuDomainObjPrivate *priv = vm->privateData;
virCgroupEmulatorAllNodesData *emulatorCgroup = NULL;
virDomainVcpuDef *vcpu;
qemuDomainVcpuPrivate *vcpupriv;
size_t i;
int ret = -1;
int rc;
g_autofree virDomainVcpuDef **bootHotplug = NULL;
size_t nbootHotplug = 0;
for (i = 0; i < maxvcpus; i++) {
vcpu = virDomainDefGetVcpu(vm->def, i);
vcpupriv = QEMU_DOMAIN_VCPU_PRIVATE(vcpu);
if (vcpu->hotpluggable == VIR_TRISTATE_BOOL_YES && vcpu->online &&
vcpupriv->vcpus != 0) {
vcpupriv->alias = g_strdup_printf("vcpu%zu", i);
VIR_APPEND_ELEMENT(bootHotplug, nbootHotplug, vcpu);
}
}
if (nbootHotplug == 0)
return 0;
g_qsort_with_data(bootHotplug, nbootHotplug,
sizeof(*bootHotplug), qemuProcessVcpusSortOrder, NULL);
if (virDomainCgroupEmulatorAllNodesAllow(priv->cgroup, &emulatorCgroup) < 0)
goto cleanup;
for (i = 0; i < nbootHotplug; i++) {
g_autoptr(virJSONValue) vcpuprops = NULL;
vcpu = bootHotplug[i];
if (!(vcpuprops = qemuBuildHotpluggableCPUProps(vcpu)))
goto cleanup;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
goto cleanup;
rc = qemuMonitorAddDeviceProps(qemuDomainGetMonitor(vm), &vcpuprops);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
goto cleanup;
}
ret = 0;
cleanup:
virDomainCgroupEmulatorAllNodesRestore(emulatorCgroup);
return ret;
}
static bool
qemuProcessDropUnknownCPUFeatures(const char *name,
virCPUFeaturePolicy policy,
void *opaque)
{
const char **features = opaque;
if (policy != VIR_CPU_FEATURE_DISABLE &&
policy != VIR_CPU_FEATURE_FORBID)
return true;
if (g_strv_contains(features, name))
return true;
/* Features unknown to QEMU are implicitly disabled, we can just drop them
* from the definition. */
return false;
}
static int
qemuProcessUpdateGuestCPU(virDomainDef *def,
virQEMUCaps *qemuCaps,
virArch hostarch,
unsigned int flags)
{
if (!def->cpu)
return 0;
/* nothing to do if only topology part of CPU def is used */
if (def->cpu->mode == VIR_CPU_MODE_CUSTOM && !def->cpu->model)
return 0;
/* Old libvirt added host CPU model to host-model CPUs for migrations,
* while new libvirt just turns host-model into custom mode. We need
* to fix the mode to maintain backward compatibility and to avoid
* the CPU model to be replaced in virCPUUpdate.
*/
if (!(flags & VIR_QEMU_PROCESS_START_NEW) &&
ARCH_IS_X86(def->os.arch) &&
def->cpu->mode == VIR_CPU_MODE_HOST_MODEL &&
def->cpu->model) {
def->cpu->mode = VIR_CPU_MODE_CUSTOM;
}
if (!virQEMUCapsIsCPUModeSupported(qemuCaps, hostarch, def->virtType,
def->cpu->mode, def->os.machine)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("CPU mode '%1$s' for %2$s %3$s domain on %4$s host is not supported by hypervisor"),
virCPUModeTypeToString(def->cpu->mode),
virArchToString(def->os.arch),
virDomainVirtTypeToString(def->virtType),
virArchToString(hostarch));
return -1;
}
if (virCPUConvertLegacy(hostarch, def->cpu) < 0)
return -1;
if (def->cpu->check != VIR_CPU_CHECK_NONE) {
virCPUDef *host;
host = virQEMUCapsGetHostModel(qemuCaps, def->virtType,
VIR_QEMU_CAPS_HOST_CPU_FULL);
if (host && virCPUCheckForbiddenFeatures(def->cpu, host) < 0)
return -1;
}
/* nothing to update for host-passthrough / maximum */
if (def->cpu->mode != VIR_CPU_MODE_HOST_PASSTHROUGH &&
def->cpu->mode != VIR_CPU_MODE_MAXIMUM) {
g_autoptr(virDomainCapsCPUModels) cpuModels = NULL;
if (def->cpu->check == VIR_CPU_CHECK_PARTIAL &&
virCPUCompare(hostarch,
virQEMUCapsGetHostModel(qemuCaps, def->virtType,
VIR_QEMU_CAPS_HOST_CPU_FULL),
def->cpu, true) < 0)
return -1;
if (virCPUUpdate(def->os.arch, def->cpu,
virQEMUCapsGetHostModel(qemuCaps, def->virtType,
VIR_QEMU_CAPS_HOST_CPU_MIGRATABLE)) < 0)
return -1;
cpuModels = virQEMUCapsGetCPUModels(qemuCaps, def->virtType, NULL, NULL);
if (virCPUTranslate(def->os.arch, def->cpu, cpuModels) < 0)
return -1;
def->cpu->fallback = VIR_CPU_FALLBACK_FORBID;
}
if (virCPUDefFilterFeatures(def->cpu, virQEMUCapsCPUFilterFeatures,
&def->os.arch) < 0)
return -1;
if (ARCH_IS_X86(def->os.arch)) {
g_auto(GStrv) features = NULL;
if (virQEMUCapsGetCPUFeatures(qemuCaps, def->virtType, false, &features) < 0)
return -1;
if (features &&
virCPUDefFilterFeatures(def->cpu, qemuProcessDropUnknownCPUFeatures,
features) < 0)
return -1;
}
return 0;
}
static int
qemuProcessPrepareDomainNUMAPlacement(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autofree char *nodeset = NULL;
g_autoptr(virBitmap) numadNodeset = NULL;
g_autoptr(virBitmap) hostMemoryNodeset = NULL;
g_autoptr(virCapsHostNUMA) caps = NULL;
/* Get the advisory nodeset from numad if 'placement' of
* either <vcpu> or <numatune> is 'auto'.
*/
if (!virDomainDefNeedsPlacementAdvice(vm->def))
return 0;
nodeset = virNumaGetAutoPlacementAdvice(virDomainDefGetVcpus(vm->def),
virDomainDefGetMemoryTotal(vm->def));
if (!nodeset)
return -1;
if (!(hostMemoryNodeset = virNumaGetHostMemoryNodeset()))
return -1;
VIR_DEBUG("Nodeset returned from numad: %s", nodeset);
if (virBitmapParse(nodeset, &numadNodeset, VIR_DOMAIN_CPUMASK_LEN) < 0)
return -1;
if (!(caps = virCapabilitiesHostNUMANewHost()))
return -1;
/* numad may return a nodeset that only contains cpus but cgroups don't play
* well with that. Set the autoCpuset from all cpus from that nodeset, but
* assign autoNodeset only with nodes containing memory. */
if (!(priv->autoCpuset = virCapabilitiesHostNUMAGetCpus(caps, numadNodeset)))
return -1;
virBitmapIntersect(numadNodeset, hostMemoryNodeset);
priv->autoNodeset = g_steal_pointer(&numadNodeset);
return 0;
}
static void
qemuProcessPrepareDeviceBootorder(virDomainDef *def)
{
size_t i;
unsigned int bootCD = 0;
unsigned int bootFloppy = 0;
unsigned int bootDisk = 0;
unsigned int bootNetwork = 0;
if (def->os.nBootDevs == 0)
return;
for (i = 0; i < def->os.nBootDevs; i++) {
switch (def->os.bootDevs[i]) {
case VIR_DOMAIN_BOOT_CDROM:
bootCD = i + 1;
break;
case VIR_DOMAIN_BOOT_FLOPPY:
bootFloppy = i + 1;
break;
case VIR_DOMAIN_BOOT_DISK:
bootDisk = i + 1;
break;
case VIR_DOMAIN_BOOT_NET:
bootNetwork = i + 1;
break;
case VIR_DOMAIN_BOOT_LAST:
default:
break;
}
}
for (i = 0; i < def->ndisks; i++) {
virDomainDiskDef *disk = def->disks[i];
switch (disk->device) {
case VIR_DOMAIN_DISK_DEVICE_CDROM:
disk->info.effectiveBootIndex = bootCD;
bootCD = 0;
break;
case VIR_DOMAIN_DISK_DEVICE_DISK:
case VIR_DOMAIN_DISK_DEVICE_LUN:
disk->info.effectiveBootIndex = bootDisk;
bootDisk = 0;
break;
case VIR_DOMAIN_DISK_DEVICE_FLOPPY:
disk->info.effectiveBootIndex = bootFloppy;
bootFloppy = 0;
break;
case VIR_DOMAIN_DISK_DEVICE_LAST:
default:
break;
}
}
if (def->nnets > 0 && bootNetwork > 0) {
/* If network boot is enabled, the first network device gets enabled. If
* that one is backed by a host device, then we need to find the first
* corresponding host device */
if (virDomainNetGetActualType(def->nets[0]) == VIR_DOMAIN_NET_TYPE_HOSTDEV) {
for (i = 0; i < def->nhostdevs; i++) {
virDomainHostdevDef *hostdev = def->hostdevs[i];
virDomainHostdevSubsys *subsys = &hostdev->source.subsys;
if (hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS &&
subsys->type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI &&
hostdev->info->type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_UNASSIGNED &&
hostdev->parentnet) {
hostdev->info->effectiveBootIndex = bootNetwork;
break;
}
}
} else {
def->nets[0]->info.effectiveBootIndex = bootNetwork;
}
}
}
static int
qemuProcessPrepareDomainStorage(virQEMUDriver *driver,
virDomainObj *vm,
qemuDomainObjPrivate *priv,
virQEMUDriverConfig *cfg,
unsigned int flags)
{
size_t i;
bool cold_boot = flags & VIR_QEMU_PROCESS_START_COLD;
for (i = vm->def->ndisks; i > 0; i--) {
size_t idx = i - 1;
virDomainDiskDef *disk = vm->def->disks[idx];
if (virDomainDiskTranslateSourcePool(disk) < 0) {
if (qemuDomainCheckDiskStartupPolicy(driver, vm, idx, cold_boot) < 0)
return -1;
/* disk source was dropped */
continue;
}
if (qemuDomainPrepareDiskSource(disk, priv, cfg) < 0)
return -1;
}
return 0;
}
static int
qemuProcessPrepareDomainHostdevs(virDomainObj *vm,
qemuDomainObjPrivate *priv)
{
size_t i;
for (i = 0; i < vm->def->nhostdevs; i++) {
virDomainHostdevDef *hostdev = vm->def->hostdevs[i];
if (qemuDomainPrepareHostdev(hostdev, priv) < 0)
return -1;
}
return 0;
}
/**
* qemuProcessRebootAllowed:
* @def: domain definition
*
* This function encapsulates the logic which dictated whether '-no-reboot' was
* used instead of '-no-shutdown' which is used QEMU versions which don't
* support the 'set-action' QMP command.
*/
bool
qemuProcessRebootAllowed(const virDomainDef *def)
{
return def->onReboot != VIR_DOMAIN_LIFECYCLE_ACTION_DESTROY ||
def->onPoweroff != VIR_DOMAIN_LIFECYCLE_ACTION_DESTROY ||
(def->onCrash != VIR_DOMAIN_LIFECYCLE_ACTION_DESTROY &&
def->onCrash != VIR_DOMAIN_LIFECYCLE_ACTION_COREDUMP_DESTROY);
}
static void
qemuProcessPrepareAllowReboot(virDomainObj *vm)
{
virDomainDef *def = vm->def;
qemuDomainObjPrivate *priv = vm->privateData;
/* with 'set-action' QMP command we don't need to keep this around as
* we always update qemu with the proper state */
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_SET_ACTION))
return;
if (priv->allowReboot != VIR_TRISTATE_BOOL_ABSENT)
return;
priv->allowReboot = virTristateBoolFromBool(qemuProcessRebootAllowed(def));
}
static int
qemuProcessUpdateSEVInfo(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUCaps *qemuCaps = priv->qemuCaps;
virDomainSEVDef *sev = &vm->def->sec->data.sev;
virSEVCapability *sevCaps = NULL;
/* if platform specific info like 'cbitpos' and 'reducedPhysBits' have
* not been supplied, we need to autofill them from caps now as both are
* mandatory on QEMU cmdline
*/
sevCaps = virQEMUCapsGetSEVCapabilities(qemuCaps);
if (!sev->haveCbitpos) {
sev->cbitpos = sevCaps->cbitpos;
sev->haveCbitpos = true;
}
if (!sev->haveReducedPhysBits) {
sev->reduced_phys_bits = sevCaps->reduced_phys_bits;
sev->haveReducedPhysBits = true;
}
return 0;
}
/* qemuProcessPrepareChardevSource:
* @def: live domain definition
* @cfg: driver configuration
*
* Iterate through all devices that use virDomainChrSourceDef as backend.
*/
static int
qemuProcessPrepareChardevSource(virDomainDef *def,
virQEMUDriverConfig *cfg)
{
struct qemuDomainPrepareChardevSourceData data = { .cfg = cfg };
return qemuDomainDeviceBackendChardevForeach(def,
qemuDomainPrepareChardevSourceOne,
&data);
}
/**
* qemuProcessPrepareDomain:
* @driver: qemu driver
* @vm: domain object
* @flags: qemuProcessStartFlags
*
* This function groups all code that modifies only live XML of a domain which
* is about to start and it's the only place to do those modifications.
*
* Flag VIR_QEMU_PROCESS_START_PRETEND tells, that we don't want to actually
* start the domain but create a valid qemu command. If some code shouldn't be
* executed in this case, make sure to check this flag.
*
* This function MUST be called before qemuProcessPrepareHost().
*
* TODO: move all XML modification from qemuBuildCommandLine into this function
*/
int
qemuProcessPrepareDomain(virQEMUDriver *driver,
virDomainObj *vm,
unsigned int flags)
{
size_t i;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
priv->machineName = qemuDomainGetMachineName(vm);
if (!priv->machineName)
return -1;
if (!(flags & VIR_QEMU_PROCESS_START_PRETEND)) {
/* If you are using a SecurityDriver with dynamic labelling,
then generate a security label for isolation */
VIR_DEBUG("Generating domain security label (if required)");
if (qemuSecurityGenLabel(driver->securityManager, vm->def) < 0) {
virDomainAuditSecurityLabel(vm, false);
return -1;
}
virDomainAuditSecurityLabel(vm, true);
if (qemuProcessPrepareDomainNUMAPlacement(vm) < 0)
return -1;
}
/* Whether we should use virtlogd as stdio handler for character
* devices source backend. */
priv->chardevStdioLogd = cfg->stdioLogD;
/* Track if this domain remembers original owner */
priv->rememberOwner = cfg->rememberOwner;
qemuProcessPrepareAllowReboot(vm);
/*
* Normally PCI addresses are assigned in the virDomainCreate
* or virDomainDefine methods. We might still need to assign
* some here to cope with the question of upgrades. Regardless
* we also need to populate the PCI address set cache for later
* use in hotplug
*/
VIR_DEBUG("Assigning domain PCI addresses");
if ((qemuDomainAssignAddresses(vm->def, priv->qemuCaps, driver, vm,
!!(flags & VIR_QEMU_PROCESS_START_NEW))) < 0) {
return -1;
}
if (qemuAssignDeviceAliases(vm->def) < 0)
return -1;
qemuProcessPrepareDeviceBootorder(vm->def);
VIR_DEBUG("Setting graphics devices");
if (qemuProcessSetupGraphics(driver, vm, priv->qemuCaps, flags) < 0)
return -1;
VIR_DEBUG("Create domain masterKey");
if (qemuDomainMasterKeyCreate(vm) < 0)
return -1;
VIR_DEBUG("Setting up storage");
if (qemuProcessPrepareDomainStorage(driver, vm, priv, cfg, flags) < 0)
return -1;
VIR_DEBUG("Setting up host devices");
if (qemuProcessPrepareDomainHostdevs(vm, priv) < 0)
return -1;
VIR_DEBUG("Setting up network devices");
if (qemuProcessPrepareDomainNetwork(vm) < 0)
return -1;
VIR_DEBUG("Prepare chardev source backends");
if (qemuProcessPrepareChardevSource(vm->def, cfg) < 0)
return -1;
VIR_DEBUG("Prepare device secrets");
if (qemuDomainSecretPrepare(driver, vm) < 0)
return -1;
VIR_DEBUG("Prepare bios/uefi paths");
if (qemuFirmwareFillDomain(driver, vm->def, false) < 0)
return -1;
if (qemuDomainInitializePflashStorageSource(vm, cfg) < 0)
return -1;
VIR_DEBUG("Preparing external devices");
if (qemuExtDevicesPrepareDomain(driver, vm) < 0)
return -1;
if (flags & VIR_QEMU_PROCESS_START_NEW) {
VIR_DEBUG("Aligning guest memory");
if (qemuDomainAlignMemorySizes(vm->def) < 0)
return -1;
}
for (i = 0; i < vm->def->nchannels; i++) {
if (qemuDomainPrepareChannel(vm->def->channels[i],
priv->channelTargetDir) < 0)
return -1;
}
if (!(priv->monConfig = virDomainChrSourceDefNew(driver->xmlopt)))
return -1;
VIR_DEBUG("Preparing monitor state");
if (qemuProcessPrepareMonitorChr(priv->monConfig, priv->libDir) < 0)
return -1;
priv->monError = false;
priv->monStart = 0;
priv->runningReason = VIR_DOMAIN_RUNNING_UNKNOWN;
priv->pausedReason = VIR_DOMAIN_PAUSED_UNKNOWN;
VIR_DEBUG("Updating guest CPU definition");
if (qemuProcessUpdateGuestCPU(vm->def, priv->qemuCaps, driver->hostarch, flags) < 0)
return -1;
for (i = 0; i < vm->def->nshmems; i++)
qemuDomainPrepareShmemChardev(vm->def->shmems[i]);
if (vm->def->sec &&
vm->def->sec->sectype == VIR_DOMAIN_LAUNCH_SECURITY_SEV) {
VIR_DEBUG("Updating SEV platform info");
if (qemuProcessUpdateSEVInfo(vm) < 0)
return -1;
}
return 0;
}
static int
qemuProcessSEVCreateFile(virDomainObj *vm,
const char *name,
const char *data)
{
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUDriver *driver = priv->driver;
g_autofree char *configFile = NULL;
if (!(configFile = virFileBuildPath(priv->libDir, name, ".base64")))
return -1;
if (virFileRewriteStr(configFile, S_IRUSR | S_IWUSR, data) < 0) {
virReportSystemError(errno, _("failed to write data to config '%1$s'"),
configFile);
return -1;
}
if (qemuSecurityDomainSetPathLabel(driver, vm, configFile, true) < 0)
return -1;
return 0;
}
static int
qemuProcessPrepareSEVGuestInput(virDomainObj *vm)
{
virDomainSEVDef *sev = &vm->def->sec->data.sev;
VIR_DEBUG("Preparing SEV guest");
if (sev->dh_cert) {
if (qemuProcessSEVCreateFile(vm, "dh_cert", sev->dh_cert) < 0)
return -1;
}
if (sev->session) {
if (qemuProcessSEVCreateFile(vm, "session", sev->session) < 0)
return -1;
}
return 0;
}
static int
qemuProcessPrepareLaunchSecurityGuestInput(virDomainObj *vm)
{
virDomainSecDef *sec = vm->def->sec;
if (!sec)
return 0;
switch ((virDomainLaunchSecurity) sec->sectype) {
case VIR_DOMAIN_LAUNCH_SECURITY_SEV:
return qemuProcessPrepareSEVGuestInput(vm);
case VIR_DOMAIN_LAUNCH_SECURITY_PV:
return 0;
case VIR_DOMAIN_LAUNCH_SECURITY_NONE:
case VIR_DOMAIN_LAUNCH_SECURITY_LAST:
virReportEnumRangeError(virDomainLaunchSecurity, sec->sectype);
return -1;
}
return 0;
}
static int
qemuProcessPrepareHostStorageSourceVDPA(virStorageSource *src,
qemuDomainObjPrivate *priv)
{
qemuDomainStorageSourcePrivate *srcpriv = NULL;
virStorageType actualType = virStorageSourceGetActualType(src);
int vdpafd = -1;
if (actualType != VIR_STORAGE_TYPE_VHOST_VDPA)
return 0;
if ((vdpafd = qemuVDPAConnect(src->vdpadev)) < 0)
return -1;
srcpriv = qemuDomainStorageSourcePrivateFetch(src);
srcpriv->fdpass = qemuFDPassNew(qemuBlockStorageSourceGetStorageNodename(src), priv);
qemuFDPassAddFD(srcpriv->fdpass, &vdpafd, "-vdpa");
return 0;
}
/**
* See qemuProcessPrepareHostStorageSourceChain
*/
int
qemuProcessPrepareHostStorageSource(virDomainObj *vm,
virStorageSource *src)
{
/* connect to any necessary vdpa block devices */
if (qemuProcessPrepareHostStorageSourceVDPA(src, vm->privateData) < 0)
return -1;
return 0;
}
/**
* qemuProcessPrepareHostStorageSourceChain:
*
* @vm: domain object
* @chain: source chain
*
* Prepare the host side of a disk for use with the VM. Note that this function
* accesses host resources.
*/
int
qemuProcessPrepareHostStorageSourceChain(virDomainObj *vm,
virStorageSource *chain)
{
virStorageSource *n;
for (n = chain; virStorageSourceIsBacking(n); n = n->backingStore) {
if (qemuProcessPrepareHostStorageSource(vm, n) < 0)
return -1;
}
return 0;
}
/**
* qemuProcessPrepareHostStorageDisk:
*
* @vm: domain object
* @disk: disk definition object
*
* Prepare the host side of a disk for use with the VM. Note that this function
* accesses host resources.
*
* Note that this function does not call qemuDomainDetermineDiskChain as that is
* needed in qemuProcessPrepareHostStorage to remove disks based on the startup
* policy, thus other callers need to call it explicitly.
*/
int
qemuProcessPrepareHostStorageDisk(virDomainObj *vm,
virDomainDiskDef *disk)
{
if (qemuProcessPrepareHostStorageSourceChain(vm, disk->src) < 0)
return -1;
return 0;
}
static int
qemuProcessPrepareHostStorage(virQEMUDriver *driver,
virDomainObj *vm,
unsigned int flags)
{
size_t i;
bool cold_boot = flags & VIR_QEMU_PROCESS_START_COLD;
for (i = vm->def->ndisks; i > 0; i--) {
size_t idx = i - 1;
virDomainDiskDef *disk = vm->def->disks[idx];
if (virStorageSourceIsEmpty(disk->src))
continue;
/* backing chain needs to be redetected if we aren't using blockdev */
if (qemuDiskBusIsSD(disk->bus))
virStorageSourceBackingStoreClear(disk->src);
/*
* Go to applying startup policy for optional disk with nonexistent
* source file immediately as determining chain will surely fail
* and we don't want noisy error notice in logs for this case.
*/
if (qemuDomainDiskIsMissingLocalOptional(disk) && cold_boot)
VIR_INFO("optional disk '%s' source file is missing, "
"skip checking disk chain", disk->dst);
else if (qemuDomainDetermineDiskChain(driver, vm, disk, NULL) >= 0)
continue;
if (qemuDomainCheckDiskStartupPolicy(driver, vm, idx, cold_boot) >= 0)
continue;
return -1;
}
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
if (qemuProcessPrepareHostStorageDisk(vm, disk) < 0)
return -1;
}
return 0;
}
int
qemuProcessOpenVhostVsock(virDomainVsockDef *vsock)
{
qemuDomainVsockPrivate *priv = (qemuDomainVsockPrivate *)vsock->privateData;
const char *vsock_path = "/dev/vhost-vsock";
int fd;
if ((fd = open(vsock_path, O_RDWR)) < 0) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
"%s", _("unable to open vhost-vsock device"));
return -1;
}
if (vsock->auto_cid == VIR_TRISTATE_BOOL_YES) {
if (virVsockAcquireGuestCid(fd, &vsock->guest_cid) < 0)
goto error;
} else {
if (virVsockSetGuestCid(fd, vsock->guest_cid) < 0)
goto error;
}
priv->vhostfd = fd;
return 0;
error:
VIR_FORCE_CLOSE(fd);
return -1;
}
static int
qemuProcessPrepareHostBackendChardevFileHelper(const char *path,
virTristateSwitch append,
int *fd,
virLogManager *logManager,
virSecurityManager *secManager,
virQEMUDriverConfig *cfg,
const virDomainDef *def)
{
if (logManager) {
int flags = 0;
if (append == VIR_TRISTATE_SWITCH_ABSENT ||
append == VIR_TRISTATE_SWITCH_OFF)
flags |= VIR_LOG_MANAGER_PROTOCOL_DOMAIN_OPEN_LOG_FILE_TRUNCATE;
if ((*fd = virLogManagerDomainOpenLogFile(logManager,
"qemu",
def->uuid,
def->name,
path,
flags,
NULL, NULL)) < 0)
return -1;
} else {
int oflags = O_CREAT | O_WRONLY;
switch (append) {
case VIR_TRISTATE_SWITCH_ABSENT:
case VIR_TRISTATE_SWITCH_OFF:
oflags |= O_TRUNC;
break;
case VIR_TRISTATE_SWITCH_ON:
oflags |= O_APPEND;
break;
case VIR_TRISTATE_SWITCH_LAST:
break;
}
if ((*fd = qemuDomainOpenFile(cfg, def, path, oflags, NULL)) < 0)
return -1;
if (qemuSecuritySetImageFDLabel(secManager, (virDomainDef*)def, *fd) < 0) {
VIR_FORCE_CLOSE(*fd);
return -1;
}
}
return 0;
}
struct qemuProcessPrepareHostBackendChardevData {
qemuDomainObjPrivate *priv;
virLogManager *logManager;
virQEMUDriverConfig *cfg;
virDomainDef *def;
const char *fdprefix;
};
static int
qemuProcessPrepareHostBackendChardevOne(virDomainDeviceDef *dev,
virDomainChrSourceDef *chardev,
void *opaque)
{
struct qemuProcessPrepareHostBackendChardevData *data = opaque;
qemuDomainChrSourcePrivate *charpriv = QEMU_DOMAIN_CHR_SOURCE_PRIVATE(chardev);
const char *devalias = NULL;
/* this function is also called for the monitor backend which doesn't have
* a 'dev' */
if (dev) {
virDomainDeviceInfo *info = virDomainDeviceGetInfo(dev);
devalias = info->alias;
/* vhost-user disk doesn't use FD passing */
if (dev->type == VIR_DOMAIN_DEVICE_DISK)
return 0;
if (dev->type == VIR_DOMAIN_DEVICE_NET) {
/* due to a historical bug in qemu we don't use FD passtrhough for
* vhost-sockets for network devices */
return 0;
}
/* TPMs FD passing setup is special and handled separately */
if (dev->type == VIR_DOMAIN_DEVICE_TPM)
return 0;
} else {
devalias = data->fdprefix;
}
switch ((virDomainChrType) chardev->type) {
case VIR_DOMAIN_CHR_TYPE_NULL:
case VIR_DOMAIN_CHR_TYPE_VC:
case VIR_DOMAIN_CHR_TYPE_PTY:
case VIR_DOMAIN_CHR_TYPE_DEV:
case VIR_DOMAIN_CHR_TYPE_PIPE:
case VIR_DOMAIN_CHR_TYPE_STDIO:
case VIR_DOMAIN_CHR_TYPE_UDP:
case VIR_DOMAIN_CHR_TYPE_TCP:
case VIR_DOMAIN_CHR_TYPE_SPICEVMC:
case VIR_DOMAIN_CHR_TYPE_SPICEPORT:
case VIR_DOMAIN_CHR_TYPE_QEMU_VDAGENT:
case VIR_DOMAIN_CHR_TYPE_DBUS:
break;
case VIR_DOMAIN_CHR_TYPE_FILE: {
VIR_AUTOCLOSE sourcefd = -1;
if (qemuProcessPrepareHostBackendChardevFileHelper(chardev->data.file.path,
chardev->data.file.append,
&sourcefd,
data->logManager,
data->priv->driver->securityManager,
data->cfg,
data->def) < 0)
return -1;
charpriv->sourcefd = qemuFDPassNew(devalias, data->priv);
qemuFDPassAddFD(charpriv->sourcefd, &sourcefd, "-source");
}
break;
case VIR_DOMAIN_CHR_TYPE_UNIX:
if (chardev->data.nix.listen) {
g_autofree char *name = g_strdup_printf("%s-source", devalias);
VIR_AUTOCLOSE sourcefd = -1;
if (qemuSecuritySetSocketLabel(data->priv->driver->securityManager, data->def) < 0)
return -1;
sourcefd = qemuOpenChrChardevUNIXSocket(chardev);
if (qemuSecurityClearSocketLabel(data->priv->driver->securityManager, data->def) < 0 ||
sourcefd < 0)
return -1;
charpriv->directfd = qemuFDPassDirectNew(name, &sourcefd);
}
break;
case VIR_DOMAIN_CHR_TYPE_NMDM:
case VIR_DOMAIN_CHR_TYPE_LAST:
default:
virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
_("unsupported chardev '%1$s'"),
virDomainChrTypeToString(chardev->type));
return -1;
}
if (chardev->logfile) {
VIR_AUTOCLOSE logfd = -1;
if (qemuProcessPrepareHostBackendChardevFileHelper(chardev->logfile,
chardev->logappend,
&logfd,
data->logManager,
data->priv->driver->securityManager,
data->cfg,
data->def) < 0)
return -1;
charpriv->logfd = qemuFDPassNew(devalias, data->priv);
qemuFDPassAddFD(charpriv->logfd, &logfd, "-log");
}
return 0;
}
/* prepare the chardev backends for various devices:
* serial/parallel/channel chardevs, vhost-user disks, vhost-user network
* interfaces, smartcards, shared memory, and redirdevs */
static int
qemuProcessPrepareHostBackendChardev(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
struct qemuProcessPrepareHostBackendChardevData data = {
.priv = priv,
.logManager = NULL,
.cfg = cfg,
.def = vm->def,
};
g_autoptr(virLogManager) logManager = NULL;
if (cfg->stdioLogD) {
if (!(logManager = data.logManager = virLogManagerNew(priv->driver->privileged)))
return -1;
}
if (qemuDomainDeviceBackendChardevForeach(vm->def,
qemuProcessPrepareHostBackendChardevOne,
&data) < 0)
return -1;
data.fdprefix = "monitor";
if (qemuProcessPrepareHostBackendChardevOne(NULL, priv->monConfig, &data) < 0)
return -1;
return 0;
}
int
qemuProcessPrepareHostBackendChardevHotplug(virDomainObj *vm,
virDomainDeviceDef *dev)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(priv->driver);
struct qemuProcessPrepareHostBackendChardevData data = {
.priv = priv,
.logManager = NULL,
.cfg = cfg,
.def = vm->def,
};
g_autoptr(virLogManager) logManager = NULL;
if (cfg->stdioLogD) {
if (!(logManager = data.logManager = virLogManagerNew(priv->driver->privileged)))
return -1;
}
if (qemuDomainDeviceBackendChardevForeachOne(dev,
qemuProcessPrepareHostBackendChardevOne,
&data) < 0)
return -1;
return 0;
}
/**
* qemuProcessPrepareHost:
* @driver: qemu driver
* @vm: domain object
* @flags: qemuProcessStartFlags
*
* This function groups all code that modifies host system (which also may
* update live XML) to prepare environment for a domain which is about to start
* and it's the only place to do those modifications.
*
* This function MUST be called only after qemuProcessPrepareDomain().
*
* TODO: move all host modification from qemuBuildCommandLine into this function
*/
int
qemuProcessPrepareHost(virQEMUDriver *driver,
virDomainObj *vm,
unsigned int flags)
{
unsigned int hostdev_flags = 0;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
/*
* Create all per-domain directories in order to make sure domain
* with any possible seclabels can access it.
*/
if (qemuProcessMakeDir(driver, vm, priv->libDir) < 0 ||
qemuProcessMakeDir(driver, vm, priv->channelTargetDir) < 0)
return -1;
if (qemuPrepareNVRAM(driver, vm, !!(flags & VIR_QEMU_PROCESS_START_RESET_NVRAM)) < 0)
return -1;
if (vm->def->vsock) {
if (qemuProcessOpenVhostVsock(vm->def->vsock) < 0)
return -1;
}
VIR_DEBUG("Preparing network devices");
if (qemuProcessPrepareHostNetwork(vm) < 0)
return -1;
/* Must be run before security labelling */
VIR_DEBUG("Preparing host devices");
if (!cfg->relaxedACS)
hostdev_flags |= VIR_HOSTDEV_STRICT_ACS_CHECK;
if (flags & VIR_QEMU_PROCESS_START_NEW)
hostdev_flags |= VIR_HOSTDEV_COLD_BOOT;
if (qemuHostdevPrepareDomainDevices(driver, vm->def, hostdev_flags) < 0)
return -1;
VIR_DEBUG("Preparing chr device backends");
if (qemuProcessPrepareHostBackendChardev(vm) < 0)
return -1;
if (qemuProcessBuildDestroyMemoryPaths(driver, vm, NULL, true) < 0)
return -1;
/* Ensure no historical cgroup for this VM is lying around bogus
* settings */
VIR_DEBUG("Ensuring no historical cgroup is lying around");
virDomainCgroupRemoveCgroup(vm, priv->cgroup, priv->machineName);
if (g_mkdir_with_parents(cfg->logDir, 0777) < 0) {
virReportSystemError(errno,
_("cannot create log directory %1$s"),
cfg->logDir);
return -1;
}
VIR_FREE(priv->pidfile);
if (!(priv->pidfile = virPidFileBuildPath(cfg->stateDir, vm->def->name))) {
virReportSystemError(errno,
"%s", _("Failed to build pidfile path."));
return -1;
}
if (unlink(priv->pidfile) < 0 &&
errno != ENOENT) {
virReportSystemError(errno,
_("Cannot remove stale PID file %1$s"),
priv->pidfile);
return -1;
}
VIR_DEBUG("Write domain masterKey");
if (qemuDomainWriteMasterKeyFile(driver, vm) < 0)
return -1;
VIR_DEBUG("Preparing disks (host)");
if (qemuProcessPrepareHostStorage(driver, vm, flags) < 0)
return -1;
VIR_DEBUG("Preparing external devices");
if (qemuExtDevicesPrepareHost(driver, vm) < 0)
return -1;
if (qemuProcessPrepareLaunchSecurityGuestInput(vm) < 0)
return -1;
return 0;
}
/**
* qemuProcessGenID:
* @vm: Pointer to domain object
* @flags: qemuProcessStartFlags
*
* If this domain is requesting to use genid, then update the GUID
* value if the VIR_QEMU_PROCESS_START_GEN_VMID flag is set. This
* flag is set on specific paths during domain start processing when
* there is the possibility that the VM is potentially re-executing
* something that has already been executed before.
*/
static int
qemuProcessGenID(virDomainObj *vm,
unsigned int flags)
{
if (!vm->def->genidRequested)
return 0;
/* If we are coming from a path where we must provide a new gen id
* value regardless of whether it was previously generated or provided,
* then generate a new GUID value before we build the command line. */
if (flags & VIR_QEMU_PROCESS_START_GEN_VMID) {
if (virUUIDGenerate(vm->def->genid) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("failed to regenerate genid"));
return -1;
}
}
return 0;
}
/**
* qemuProcessSetupDiskThrottling:
*
* Sets up disk trottling for -blockdev via block_set_io_throttle monitor
* command. This hack should be replaced by proper use of the 'throttle'
* blockdev driver in qemu once it will support changing of the throttle group.
* Same hack is done in qemuDomainAttachDiskGeneric.
*/
static int
qemuProcessSetupDiskThrottling(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
size_t i;
int ret = -1;
VIR_DEBUG("Setting up disk throttling for -blockdev via block_set_io_throttle");
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
/* Setting throttling for empty drives fails */
if (virStorageSourceIsEmpty(disk->src))
continue;
if (!qemuDiskConfigBlkdeviotuneEnabled(disk))
continue;
if (qemuMonitorSetBlockIoThrottle(qemuDomainGetMonitor(vm),
QEMU_DOMAIN_DISK_PRIVATE(disk)->qomName,
&disk->blkdeviotune) < 0)
goto cleanup;
}
ret = 0;
cleanup:
qemuDomainObjExitMonitor(vm);
return ret;
}
static int
qemuProcessEnableDomainNamespaces(virQEMUDriver *driver,
virDomainObj *vm)
{
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
const char *state = "disabled";
if (virBitmapIsBitSet(cfg->namespaces, QEMU_DOMAIN_NS_MOUNT) &&
qemuDomainEnableNamespace(vm, QEMU_DOMAIN_NS_MOUNT) < 0)
return -1;
if (qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT))
state = "enabled";
VIR_DEBUG("Mount namespace for domain name=%s is %s",
vm->def->name, state);
return 0;
}
static int
qemuProcessEnablePerf(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
size_t i;
if (!(priv->perf = virPerfNew()))
return -1;
for (i = 0; i < VIR_PERF_EVENT_LAST; i++) {
if (vm->def->perf.events[i] == VIR_TRISTATE_BOOL_YES &&
virPerfEventEnable(priv->perf, i, vm->pid) < 0)
return -1;
}
return 0;
}
static int
qemuProcessSetupDisksTransientSnapshot(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
g_autoptr(qemuSnapshotDiskContext) snapctxt = NULL;
g_autoptr(GHashTable) blockNamedNodeData = NULL;
size_t i;
if (!(blockNamedNodeData = qemuBlockGetNamedNodeData(vm, asyncJob)))
return -1;
snapctxt = qemuSnapshotDiskContextNew(vm->def->ndisks, vm, asyncJob);
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *domdisk = vm->def->disks[i];
g_autoptr(virDomainSnapshotDiskDef) snapdisk = NULL;
if (!domdisk->transient ||
domdisk->transientShareBacking == VIR_TRISTATE_BOOL_YES)
continue;
/* validation code makes sure that we do this only for local disks
* with a file source */
if (!(snapdisk = qemuSnapshotGetTransientDiskDef(domdisk, vm->def->name)))
return -1;
if (qemuSnapshotDiskPrepareOne(snapctxt, domdisk, snapdisk,
blockNamedNodeData,
false,
false) < 0)
return -1;
}
if (qemuSnapshotDiskCreate(snapctxt) < 0)
return -1;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *domdisk = vm->def->disks[i];
if (!domdisk->transient ||
domdisk->transientShareBacking == VIR_TRISTATE_BOOL_YES)
continue;
QEMU_DOMAIN_DISK_PRIVATE(domdisk)->transientOverlayCreated = true;
}
return 0;
}
static int
qemuProcessSetupDisksTransientHotplug(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
bool hasHotpluggedDisk = false;
size_t i;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *domdisk = vm->def->disks[i];
if (!domdisk->transient ||
domdisk->transientShareBacking != VIR_TRISTATE_BOOL_YES)
continue;
if (qemuDomainAttachDiskGeneric(vm, domdisk, asyncJob) < 0)
return -1;
hasHotpluggedDisk = true;
}
/* in order to allow booting from such disks we need to issue a system-reset
* so that the firmware tables recording bootable devices are regerated */
if (hasHotpluggedDisk) {
int rc;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = qemuMonitorSystemReset(priv->mon);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
}
return 0;
}
static int
qemuProcessSetupDisksTransient(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
if (qemuProcessSetupDisksTransientSnapshot(vm, asyncJob) < 0)
return -1;
if (qemuProcessSetupDisksTransientHotplug(vm, asyncJob) < 0)
return -1;
return 0;
}
static int
qemuProcessSetupLifecycleActions(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
int rc;
if (!(virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_SET_ACTION)))
return 0;
/* for now we handle only onReboot->destroy here as an alternative to
* '-no-reboot' on the commandline */
if (vm->def->onReboot != VIR_DOMAIN_LIFECYCLE_ACTION_DESTROY)
return 0;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = qemuMonitorSetAction(priv->mon,
QEMU_MONITOR_ACTION_SHUTDOWN_KEEP,
QEMU_MONITOR_ACTION_REBOOT_SHUTDOWN,
QEMU_MONITOR_ACTION_WATCHDOG_KEEP,
QEMU_MONITOR_ACTION_PANIC_KEEP);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
return 0;
}
int
qemuProcessDeleteThreadContext(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
GSList *next = priv->threadContextAliases;
int ret = -1;
if (!next)
return 0;
for (; next; next = next->next) {
if (qemuMonitorDelObject(priv->mon, next->data, true) < 0)
goto cleanup;
}
ret = 0;
cleanup:
g_slist_free_full(g_steal_pointer(&priv->threadContextAliases), g_free);
return ret;
}
static int
qemuProcessDeleteThreadContextHelper(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
int ret = -1;
if (!priv->threadContextAliases)
return 0;
VIR_DEBUG("Deleting thread context objects");
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
ret = qemuProcessDeleteThreadContext(vm);
qemuDomainObjExitMonitor(vm);
return ret;
}
/**
* qemuProcessLaunch:
*
* Launch a new QEMU process with stopped virtual CPUs.
*
* The caller is supposed to call qemuProcessStop with appropriate
* flags in case of failure.
*
* Returns 0 on success,
* -1 on error which happened before devices were labeled and thus
* there is no need to restore them,
* -2 on error requesting security labels to be restored.
*/
int
qemuProcessLaunch(virConnectPtr conn,
virQEMUDriver *driver,
virDomainObj *vm,
virDomainAsyncJob asyncJob,
qemuProcessIncomingDef *incoming,
virDomainMomentObj *snapshot,
virNetDevVPortProfileOp vmop,
unsigned int flags)
{
int ret = -1;
int rv;
int logfile = -1;
g_autoptr(qemuLogContext) logCtxt = NULL;
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virCommand) cmd = NULL;
struct qemuProcessHookData hookData;
g_autoptr(virQEMUDriverConfig) cfg = NULL;
size_t nnicindexes = 0;
g_autofree int *nicindexes = NULL;
unsigned long long maxMemLock = 0;
bool incomingMigrationExtDevices = false;
VIR_DEBUG("conn=%p driver=%p vm=%p name=%s id=%d asyncJob=%d "
"incoming.uri=%s "
"incoming.fd=%d incoming.path=%s "
"snapshot=%p vmop=%d flags=0x%x",
conn, driver, vm, vm->def->name, vm->def->id, asyncJob,
NULLSTR(incoming ? incoming->uri : NULL),
incoming ? incoming->fd : -1,
NULLSTR(incoming ? incoming->path : NULL),
snapshot, vmop, flags);
/* Okay, these are just internal flags,
* but doesn't hurt to check */
virCheckFlags(VIR_QEMU_PROCESS_START_COLD |
VIR_QEMU_PROCESS_START_PAUSED |
VIR_QEMU_PROCESS_START_AUTODESTROY |
VIR_QEMU_PROCESS_START_NEW |
VIR_QEMU_PROCESS_START_GEN_VMID |
VIR_QEMU_PROCESS_START_RESET_NVRAM, -1);
cfg = virQEMUDriverGetConfig(driver);
if (flags & VIR_QEMU_PROCESS_START_AUTODESTROY) {
if (!conn) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Domain autodestroy requires a connection handle"));
return -1;
}
if (driver->embeddedRoot) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Domain autodestroy not supported for embedded drivers yet"));
return -1;
}
}
hookData.vm = vm;
hookData.driver = driver;
/* We don't increase cfg's reference counter here. */
hookData.cfg = cfg;
VIR_DEBUG("Creating domain log file");
if (!(logCtxt = qemuLogContextNew(driver, vm, vm->def->name))) {
virLastErrorPrefixMessage("%s", _("can't connect to virtlogd"));
goto cleanup;
}
logfile = qemuLogContextGetWriteFD(logCtxt);
if (qemuProcessGenID(vm, flags) < 0)
goto cleanup;
if (qemuDomainSchedCoreStart(cfg, vm) < 0)
goto cleanup;
/* For external devices the rules of incoming migration are a bit stricter,
* than plain @incoming != NULL. They need to differentiate between
* incoming migration and restore from a save file. */
incomingMigrationExtDevices = incoming &&
vmop == VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START;
if (qemuExtDevicesStart(driver, vm, incomingMigrationExtDevices) < 0)
goto cleanup;
if (!(cmd = qemuBuildCommandLine(vm,
incoming ? "defer" : NULL,
snapshot, vmop,
&nnicindexes, &nicindexes)))
goto cleanup;
if (incoming && incoming->fd != -1)
virCommandPassFD(cmd, incoming->fd, 0);
/* now that we know it is about to start call the hook if present */
if (qemuProcessStartHook(driver, vm,
VIR_HOOK_QEMU_OP_START,
VIR_HOOK_SUBOP_BEGIN) < 0)
goto cleanup;
qemuLogOperation(vm, "starting up", cmd, logCtxt);
qemuDomainObjCheckTaint(driver, vm, logCtxt, incoming != NULL);
qemuLogContextMarkPosition(logCtxt);
if (qemuProcessEnableDomainNamespaces(driver, vm) < 0)
goto cleanup;
VIR_DEBUG("Setting up raw IO");
if (qemuProcessSetupRawIO(vm, cmd) < 0)
goto cleanup;
virCommandSetPreExecHook(cmd, qemuProcessHook, &hookData);
virCommandSetUmask(cmd, 0x002);
VIR_DEBUG("Setting up process limits");
/* In some situations, eg. VFIO passthrough, QEMU might need to lock a
* significant amount of memory, so we need to set the limit accordingly */
maxMemLock = qemuDomainGetMemLockLimitBytes(vm->def);
/* For all these settings, zero indicates that the limit should
* not be set explicitly and the default/inherited limit should
* be applied instead */
if (maxMemLock > 0)
virCommandSetMaxMemLock(cmd, maxMemLock);
if (cfg->maxProcesses > 0)
virCommandSetMaxProcesses(cmd, cfg->maxProcesses);
if (cfg->maxFiles > 0)
virCommandSetMaxFiles(cmd, cfg->maxFiles);
if (cfg->schedCore == QEMU_SCHED_CORE_EMULATOR ||
cfg->schedCore == QEMU_SCHED_CORE_FULL)
virCommandSetRunAmong(cmd, priv->schedCoreChildPID);
/* In this case, however, zero means that core dumps should be
* disabled, and so we always need to set the limit explicitly */
virCommandSetMaxCoreSize(cmd, cfg->maxCore);
VIR_DEBUG("Setting up security labelling");
if (qemuSecuritySetChildProcessLabel(driver->securityManager,
vm->def, false, cmd) < 0)
goto cleanup;
virCommandSetOutputFD(cmd, &logfile);
virCommandSetErrorFD(cmd, &logfile);
virCommandNonblockingFDs(cmd);
virCommandSetPidFile(cmd, priv->pidfile);
virCommandDaemonize(cmd);
virCommandRequireHandshake(cmd);
if (qemuSecurityPreFork(driver->securityManager) < 0)
goto cleanup;
rv = virCommandRun(cmd, NULL);
qemuSecurityPostFork(driver->securityManager);
/* wait for qemu process to show up */
if (rv == 0) {
if ((rv = virPidFileReadPath(priv->pidfile, &vm->pid)) < 0) {
virReportSystemError(-rv,
_("Domain %1$s didn't show up"),
vm->def->name);
goto cleanup;
}
VIR_DEBUG("QEMU vm=%p name=%s running with pid=%lld",
vm, vm->def->name, (long long)vm->pid);
} else {
VIR_DEBUG("QEMU vm=%p name=%s failed to spawn",
vm, vm->def->name);
goto cleanup;
}
VIR_DEBUG("Writing early domain status to disk");
if (virDomainObjSave(vm, driver->xmlopt, cfg->stateDir) < 0)
goto cleanup;
VIR_DEBUG("Waiting for handshake from child");
if (virCommandHandshakeWait(cmd) < 0) {
/* Read errors from child that occurred between fork and exec. */
qemuProcessReportLogError(logCtxt,
_("Process exited prior to exec"));
goto cleanup;
}
VIR_DEBUG("Building domain mount namespace (if required)");
if (qemuDomainBuildNamespace(cfg, vm) < 0)
goto cleanup;
VIR_DEBUG("Setting up domain cgroup (if required)");
if (qemuSetupCgroup(vm, nnicindexes, nicindexes) < 0)
goto cleanup;
VIR_DEBUG("Setting up domain perf (if required)");
if (qemuProcessEnablePerf(vm) < 0)
goto cleanup;
/* This must be done after cgroup placement to avoid resetting CPU
* affinity */
if (qemuProcessInitCpuAffinity(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting emulator tuning/settings");
if (qemuProcessSetupEmulator(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting cgroup for external devices (if required)");
if (qemuSetupCgroupForExtDevices(vm, driver) < 0)
goto cleanup;
VIR_DEBUG("Setting up resctrl");
if (qemuProcessResctrlCreate(driver, vm) < 0)
goto cleanup;
VIR_DEBUG("Setting up managed PR daemon");
if (virDomainDefHasManagedPR(vm->def) &&
qemuProcessStartManagedPRDaemon(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting up permissions to allow post-copy migration");
if (qemuProcessAllowPostCopyMigration(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting domain security labels");
if (qemuSecuritySetAllLabel(driver,
vm,
incoming ? incoming->path : NULL,
incoming != NULL) < 0)
goto cleanup;
/* Security manager labeled all devices, therefore
* if any operation from now on fails, we need to ask the caller to
* restore labels.
*/
ret = -2;
if (incoming && incoming->fd != -1) {
/* if there's an fd to migrate from, and it's a pipe, put the
* proper security label on it
*/
struct stat stdin_sb;
VIR_DEBUG("setting security label on pipe used for migration");
if (fstat(incoming->fd, &stdin_sb) < 0) {
virReportSystemError(errno,
_("cannot stat fd %1$d"), incoming->fd);
goto cleanup;
}
if (S_ISFIFO(stdin_sb.st_mode) &&
qemuSecuritySetImageFDLabel(driver->securityManager,
vm->def, incoming->fd) < 0)
goto cleanup;
}
VIR_DEBUG("Labelling done, completing handshake to child");
if (virCommandHandshakeNotify(cmd) < 0)
goto cleanup;
VIR_DEBUG("Handshake complete, child running");
if (qemuDomainObjStartWorker(vm) < 0)
goto cleanup;
VIR_DEBUG("Waiting for monitor to show up");
if (qemuProcessWaitForMonitor(driver, vm, asyncJob, logCtxt) < 0)
goto cleanup;
if (qemuConnectAgent(driver, vm) < 0)
goto cleanup;
VIR_DEBUG("setting up hotpluggable cpus");
if (qemuDomainHasHotpluggableStartupVcpus(vm->def)) {
if (qemuDomainRefreshVcpuInfo(vm, asyncJob, false) < 0)
goto cleanup;
if (qemuProcessValidateHotpluggableVcpus(vm->def) < 0)
goto cleanup;
if (qemuProcessSetupHotpluggableVcpus(vm, asyncJob) < 0)
goto cleanup;
}
VIR_DEBUG("Refreshing VCPU info");
if (qemuDomainRefreshVcpuInfo(vm, asyncJob, false) < 0)
goto cleanup;
if (qemuDomainValidateVcpuInfo(vm) < 0)
goto cleanup;
qemuDomainVcpuPersistOrder(vm->def);
VIR_DEBUG("Verifying and updating provided guest CPU");
if (qemuProcessUpdateAndVerifyCPU(vm, asyncJob) < 0)
goto cleanup;
VIR_DEBUG("Detecting IOThread PIDs");
if (qemuProcessDetectIOThreadPIDs(vm, asyncJob) < 0)
goto cleanup;
VIR_DEBUG("Setting global CPU cgroup (if required)");
if (virDomainCgroupSetupGlobalCpuCgroup(vm, priv->cgroup) < 0)
goto cleanup;
VIR_DEBUG("Setting vCPU tuning/settings");
if (qemuProcessSetupVcpus(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting IOThread tuning/settings");
if (qemuProcessSetupIOThreads(vm) < 0)
goto cleanup;
VIR_DEBUG("Setting emulator scheduler");
if (vm->def->cputune.emulatorsched &&
virProcessSetScheduler(vm->pid,
vm->def->cputune.emulatorsched->policy,
vm->def->cputune.emulatorsched->priority) < 0)
goto cleanup;
VIR_DEBUG("Setting any required VM passwords");
if (qemuProcessInitPasswords(driver, vm, asyncJob) < 0)
goto cleanup;
/* set default link states */
/* qemu doesn't support setting this on the command line, so
* enter the monitor */
VIR_DEBUG("Setting network link states");
if (qemuProcessSetLinkStates(vm, asyncJob) < 0)
goto cleanup;
VIR_DEBUG("Setting initial memory amount");
if (qemuProcessSetupBalloon(vm, asyncJob) < 0)
goto cleanup;
if (qemuProcessSetupDiskThrottling(vm, asyncJob) < 0)
goto cleanup;
/* Since CPUs were not started yet, the balloon could not return the memory
* to the host and thus cur_balloon needs to be updated so that GetXMLdesc
* and friends return the correct size in case they can't grab the job */
if (!incoming && !snapshot &&
qemuProcessRefreshBalloonState(vm, asyncJob) < 0)
goto cleanup;
if (flags & VIR_QEMU_PROCESS_START_AUTODESTROY)
virCloseCallbacksDomainAdd(vm, conn, qemuProcessAutoDestroy);
if (!incoming && !snapshot) {
VIR_DEBUG("Setting up transient disk");
if (qemuProcessSetupDisksTransient(vm, asyncJob) < 0)
goto cleanup;
}
VIR_DEBUG("Setting handling of lifecycle actions");
if (qemuProcessSetupLifecycleActions(vm, asyncJob) < 0)
goto cleanup;
if (qemuProcessDeleteThreadContextHelper(vm, asyncJob) < 0)
goto cleanup;
ret = 0;
cleanup:
qemuDomainSchedCoreStop(priv);
qemuDomainStartupCleanup(vm);
return ret;
}
static int
qemuProcessRefreshRxFilters(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
size_t i;
for (i = 0; i < vm->def->nnets; i++) {
virDomainNetDef *def = vm->def->nets[i];
if (!virDomainNetGetActualTrustGuestRxFilters(def))
continue;
/* rx-filters are supported only for virtio model and TUN/TAP based
* types. */
if (def->model != VIR_DOMAIN_NET_MODEL_VIRTIO)
continue;
switch (virDomainNetGetActualType(def)) {
case VIR_DOMAIN_NET_TYPE_ETHERNET:
case VIR_DOMAIN_NET_TYPE_NETWORK:
case VIR_DOMAIN_NET_TYPE_BRIDGE:
case VIR_DOMAIN_NET_TYPE_DIRECT:
break;
case VIR_DOMAIN_NET_TYPE_USER:
case VIR_DOMAIN_NET_TYPE_VHOSTUSER:
case VIR_DOMAIN_NET_TYPE_SERVER:
case VIR_DOMAIN_NET_TYPE_CLIENT:
case VIR_DOMAIN_NET_TYPE_MCAST:
case VIR_DOMAIN_NET_TYPE_INTERNAL:
case VIR_DOMAIN_NET_TYPE_HOSTDEV:
case VIR_DOMAIN_NET_TYPE_UDP:
case VIR_DOMAIN_NET_TYPE_VDPA:
case VIR_DOMAIN_NET_TYPE_NULL:
case VIR_DOMAIN_NET_TYPE_VDS:
case VIR_DOMAIN_NET_TYPE_LAST:
default:
continue;
}
if (qemuDomainSyncRxFilter(vm, def, asyncJob) < 0)
return -1;
}
return 0;
}
/**
* qemuProcessRefreshState:
* @driver: qemu driver data
* @vm: domain to refresh
* @asyncJob: async job type
*
* This function gathers calls to refresh qemu state after startup. This
* function is called after a deferred migration finishes so that we can update
* state influenced by the migration stream.
*/
int
qemuProcessRefreshState(virQEMUDriver *driver,
virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
VIR_DEBUG("Fetching list of active devices");
if (qemuDomainUpdateDeviceList(vm, asyncJob) < 0)
return -1;
VIR_DEBUG("Updating info of memory devices");
if (qemuDomainUpdateMemoryDeviceInfo(vm, asyncJob) < 0)
return -1;
VIR_DEBUG("Detecting actual memory size for video device");
if (qemuProcessUpdateVideoRamSize(driver, vm, asyncJob) < 0)
return -1;
VIR_DEBUG("Updating disk data");
if (qemuProcessRefreshDisks(vm, asyncJob) < 0)
return -1;
VIR_DEBUG("Updating rx-filter data");
if (qemuProcessRefreshRxFilters(vm, asyncJob) < 0)
return -1;
return 0;
}
/**
* qemuProcessFinishStartup:
*
* Finish starting a new domain.
*/
int
qemuProcessFinishStartup(virQEMUDriver *driver,
virDomainObj *vm,
virDomainAsyncJob asyncJob,
bool startCPUs,
virDomainPausedReason pausedReason)
{
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
if (startCPUs) {
VIR_DEBUG("Starting domain CPUs");
if (qemuProcessStartCPUs(driver, vm,
VIR_DOMAIN_RUNNING_BOOTED,
asyncJob) < 0) {
if (virGetLastErrorCode() == VIR_ERR_OK)
virReportError(VIR_ERR_OPERATION_FAILED, "%s",
_("resume operation failed"));
return -1;
}
} else {
virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, pausedReason);
}
VIR_DEBUG("Writing domain status to disk");
if (virDomainObjSave(vm, driver->xmlopt, cfg->stateDir) < 0)
return -1;
if (qemuProcessStartHook(driver, vm,
VIR_HOOK_QEMU_OP_STARTED,
VIR_HOOK_SUBOP_BEGIN) < 0)
return -1;
return 0;
}
int
qemuProcessStart(virConnectPtr conn,
virQEMUDriver *driver,
virDomainObj *vm,
virCPUDef *updatedCPU,
virDomainAsyncJob asyncJob,
const char *migrateFrom,
int migrateFd,
const char *migratePath,
virDomainMomentObj *snapshot,
virNetDevVPortProfileOp vmop,
unsigned int flags)
{
qemuDomainObjPrivate *priv = vm->privateData;
qemuProcessIncomingDef *incoming = NULL;
unsigned int stopFlags;
bool relabel = false;
bool relabelSavedState = false;
int ret = -1;
int rv;
VIR_DEBUG("conn=%p driver=%p vm=%p name=%s id=%d asyncJob=%s "
"migrateFrom=%s migrateFd=%d migratePath=%s "
"snapshot=%p vmop=%d flags=0x%x",
conn, driver, vm, vm->def->name, vm->def->id,
virDomainAsyncJobTypeToString(asyncJob),
NULLSTR(migrateFrom), migrateFd, NULLSTR(migratePath),
snapshot, vmop, flags);
virCheckFlagsGoto(VIR_QEMU_PROCESS_START_COLD |
VIR_QEMU_PROCESS_START_PAUSED |
VIR_QEMU_PROCESS_START_AUTODESTROY |
VIR_QEMU_PROCESS_START_GEN_VMID |
VIR_QEMU_PROCESS_START_RESET_NVRAM, cleanup);
if (!migrateFrom && !snapshot)
flags |= VIR_QEMU_PROCESS_START_NEW;
if (qemuProcessInit(driver, vm, updatedCPU,
asyncJob, !!migrateFrom, flags) < 0)
goto cleanup;
if (migrateFrom) {
incoming = qemuProcessIncomingDefNew(priv->qemuCaps, NULL, migrateFrom,
migrateFd, migratePath);
if (!incoming)
goto stop;
}
if (qemuProcessPrepareDomain(driver, vm, flags) < 0)
goto stop;
if (qemuProcessPrepareHost(driver, vm, flags) < 0)
goto stop;
if (migratePath) {
if (qemuSecuritySetSavedStateLabel(driver->securityManager,
vm->def, migratePath) < 0)
goto cleanup;
relabelSavedState = true;
}
if ((rv = qemuProcessLaunch(conn, driver, vm, asyncJob, incoming,
snapshot, vmop, flags)) < 0) {
if (rv == -2)
relabel = true;
goto stop;
}
relabel = true;
if (incoming) {
if (qemuMigrationDstRun(vm, incoming->uri, asyncJob) < 0)
goto stop;
} else {
/* Refresh state of devices from QEMU. During migration this happens
* in qemuMigrationDstFinish to ensure that state information is fully
* transferred. */
if (qemuProcessRefreshState(driver, vm, asyncJob) < 0)
goto stop;
}
if (qemuProcessFinishStartup(driver, vm, asyncJob,
!(flags & VIR_QEMU_PROCESS_START_PAUSED),
incoming ?
VIR_DOMAIN_PAUSED_MIGRATION :
VIR_DOMAIN_PAUSED_USER) < 0)
goto stop;
if (!incoming) {
/* Keep watching qemu log for errors during incoming migration, otherwise
* unset reporting errors from qemu log. */
qemuMonitorSetDomainLog(priv->mon, NULL, NULL, NULL);
}
ret = 0;
cleanup:
if (relabelSavedState &&
qemuSecurityRestoreSavedStateLabel(driver->securityManager,
vm->def, migratePath) < 0)
VIR_WARN("failed to restore save state label on %s", migratePath);
qemuProcessIncomingDefFree(incoming);
return ret;
stop:
stopFlags = 0;
if (!relabel)
stopFlags |= VIR_QEMU_PROCESS_STOP_NO_RELABEL;
if (migrateFrom)
stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
if (priv->mon)
qemuMonitorSetDomainLog(priv->mon, NULL, NULL, NULL);
qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED, asyncJob, stopFlags);
goto cleanup;
}
/**
* qemuProcessStartWithMemoryState:
* @conn: connection object
* @driver: qemu driver object
* @vm: domain object
* @fd: FD pointer of memory state file
* @path: path to memory state file
* @snapshot: internal snapshot to load when starting QEMU process or NULL
* @data: data from memory state file or NULL
* @asyncJob: type of asynchronous job
* @start_flags: flags to start QEMU process with
* @reason: audit log reason
* @started: boolean to store if QEMU process was started
*
* Start VM with existing memory state. Make sure that the stored memory state
* is correctly decompressed so it can be loaded by QEMU process.
*
* When reverting to internal snapshot caller needs to pass @snapshot
* to correctly start QEMU process, @fd, @path, @data needs to be NULL.
*
* When restoring VM from saved image caller needs to pass @fd, @path and
* @data to correctly start QEMU process, @snapshot needs to be NULL.
*
* For audit purposes the expected @reason is one of `restored` or `from-snapshot`.
*
* Returns 0 on success, -1 on error.
*/
int
qemuProcessStartWithMemoryState(virConnectPtr conn,
virQEMUDriver *driver,
virDomainObj *vm,
int *fd,
const char *path,
virDomainMomentObj *snapshot,
virQEMUSaveData *data,
virDomainAsyncJob asyncJob,
unsigned int start_flags,
const char *reason,
bool *started)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(qemuDomainSaveCookie) cookie = NULL;
VIR_AUTOCLOSE intermediatefd = -1;
g_autoptr(virCommand) cmd = NULL;
g_autofree char *errbuf = NULL;
const char *migrateFrom = NULL;
int rc = 0;
if (data) {
if (virSaveCookieParseString(data->cookie, (virObject **)&cookie,
virDomainXMLOptionGetSaveCookie(driver->xmlopt)) < 0)
return -1;
if (qemuSaveImageDecompressionStart(data, fd, &intermediatefd,
&errbuf, &cmd) < 0) {
return -1;
}
migrateFrom = "stdio";
}
/* No cookie means libvirt which saved the domain was too old to mess up
* the CPU definitions.
*/
if (cookie &&
qemuDomainFixupCPUs(vm, &cookie->cpu) < 0)
return -1;
if (cookie && !cookie->slirpHelper)
priv->disableSlirp = true;
if (qemuProcessStart(conn, driver, vm, cookie ? cookie->cpu : NULL,
asyncJob, migrateFrom, *fd, path, snapshot,
VIR_NETDEV_VPORT_PROFILE_OP_RESTORE,
start_flags) == 0)
*started = true;
if (data) {
rc = qemuSaveImageDecompressionStop(cmd, fd, &intermediatefd, errbuf,
*started, path);
}
virDomainAuditStart(vm, reason, *started);
if (!*started || rc < 0)
return -1;
/* qemuProcessStart doesn't unset the qemu error reporting infrastructure
* in case of migration (which is used in this case) so we need to reset it
* so that the handle to virtlogd is not held open unnecessarily */
qemuMonitorSetDomainLog(qemuDomainGetMonitor(vm), NULL, NULL, NULL);
return 0;
}
int
qemuProcessCreatePretendCmdPrepare(virQEMUDriver *driver,
virDomainObj *vm,
const char *migrateURI,
unsigned int flags)
{
virCheckFlags(VIR_QEMU_PROCESS_START_COLD |
VIR_QEMU_PROCESS_START_PAUSED |
VIR_QEMU_PROCESS_START_AUTODESTROY, -1);
flags |= VIR_QEMU_PROCESS_START_PRETEND;
if (!migrateURI)
flags |= VIR_QEMU_PROCESS_START_NEW;
if (qemuProcessInit(driver, vm, NULL, VIR_ASYNC_JOB_NONE,
!!migrateURI, flags) < 0)
return -1;
if (qemuProcessPrepareDomain(driver, vm, flags) < 0)
return -1;
return 0;
}
virCommand *
qemuProcessCreatePretendCmdBuild(virDomainObj *vm,
const char *migrateURI)
{
return qemuBuildCommandLine(vm,
migrateURI,
NULL,
VIR_NETDEV_VPORT_PROFILE_OP_NO_OP,
NULL,
NULL);
}
int
qemuProcessKill(virDomainObj *vm, unsigned int flags)
{
VIR_DEBUG("vm=%p name=%s pid=%lld flags=0x%x",
vm, vm->def->name,
(long long)vm->pid, flags);
if (!(flags & VIR_QEMU_PROCESS_KILL_NOCHECK)) {
if (!virDomainObjIsActive(vm)) {
VIR_DEBUG("VM '%s' not active", vm->def->name);
return 0;
}
}
if (flags & VIR_QEMU_PROCESS_KILL_NOWAIT) {
virProcessKill(vm->pid,
(flags & VIR_QEMU_PROCESS_KILL_FORCE) ?
SIGKILL : SIGTERM);
return 0;
}
/* Request an extra delay of two seconds per current nhostdevs
* to be safe against stalls by the kernel freeing up the resources */
return virProcessKillPainfullyDelay(vm->pid,
!!(flags & VIR_QEMU_PROCESS_KILL_FORCE),
vm->def->nhostdevs * 2,
false);
}
/**
* qemuProcessBeginStopJob:
*
* Stop all current jobs by killing the domain and start a new one for
* qemuProcessStop.
*/
int
qemuProcessBeginStopJob(virDomainObj *vm,
virDomainJob job,
bool forceKill)
{
qemuDomainObjPrivate *priv = vm->privateData;
unsigned int killFlags = forceKill ? VIR_QEMU_PROCESS_KILL_FORCE : 0;
int ret = -1;
/* We need to prevent monitor EOF callback from doing our work (and
* sending misleading events) while the vm is unlocked inside
* BeginJob/ProcessKill API
*/
priv->beingDestroyed = true;
if (qemuProcessKill(vm, killFlags) < 0)
goto cleanup;
/* Wake up anything waiting on domain condition */
VIR_DEBUG("waking up all jobs waiting on the domain condition");
virDomainObjBroadcast(vm);
if (virDomainObjBeginJob(vm, job) < 0)
goto cleanup;
ret = 0;
cleanup:
priv->beingDestroyed = false;
return ret;
}
void qemuProcessStop(virQEMUDriver *driver,
virDomainObj *vm,
virDomainShutoffReason reason,
virDomainAsyncJob asyncJob,
unsigned int flags)
{
int ret;
int retries = 0;
qemuDomainObjPrivate *priv = vm->privateData;
virErrorPtr orig_err;
virDomainDef *def = vm->def;
size_t i;
g_autofree char *timestamp = NULL;
g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver);
bool outgoingMigration;
VIR_DEBUG("Shutting down vm=%p name=%s id=%d pid=%lld, "
"reason=%s, asyncJob=%s, flags=0x%x",
vm, vm->def->name, vm->def->id,
(long long)vm->pid,
virDomainShutoffReasonTypeToString(reason),
virDomainAsyncJobTypeToString(asyncJob),
flags);
/* This method is routinely used in clean up paths. Disable error
* reporting so we don't squash a legit error. */
virErrorPreserveLast(&orig_err);
if (asyncJob != VIR_ASYNC_JOB_NONE) {
if (virDomainObjBeginNestedJob(vm, asyncJob) < 0)
goto cleanup;
} else if (vm->job->asyncJob != VIR_ASYNC_JOB_NONE &&
vm->job->asyncOwner == virThreadSelfID() &&
vm->job->active != VIR_JOB_ASYNC_NESTED) {
VIR_WARN("qemuProcessStop called without a nested job (async=%s)",
virDomainAsyncJobTypeToString(asyncJob));
}
if (!virDomainObjIsActive(vm)) {
VIR_DEBUG("VM '%s' not active", vm->def->name);
goto endjob;
}
qemuProcessBuildDestroyMemoryPaths(driver, vm, NULL, false);
if (!!g_atomic_int_dec_and_test(&driver->nactive) && driver->inhibitCallback)
driver->inhibitCallback(false, driver->inhibitOpaque);
if ((timestamp = virTimeStringNow()) != NULL) {
qemuDomainLogAppendMessage(driver, vm, "%s: shutting down, reason=%s\n",
timestamp,
virDomainShutoffReasonTypeToString(reason));
}
/* Clear network bandwidth */
virDomainClearNetBandwidth(vm->def);
virDomainConfVMNWFilterTeardown(vm);
if (cfg->macFilter) {
for (i = 0; i < def->nnets; i++) {
virDomainNetDef *net = def->nets[i];
if (net->ifname == NULL)
continue;
ignore_value(ebtablesRemoveForwardAllowIn(driver->ebtables,
net->ifname,
&net->mac));
}
}
virPortAllocatorRelease(priv->nbdPort);
priv->nbdPort = 0;
if (priv->agent) {
g_clear_pointer(&priv->agent, qemuAgentClose);
}
priv->agentError = false;
if (priv->mon) {
g_clear_pointer(&priv->mon, qemuMonitorClose);
}
if (priv->monConfig) {
if (priv->monConfig->type == VIR_DOMAIN_CHR_TYPE_UNIX)
unlink(priv->monConfig->data.nix.path);
g_clear_pointer(&priv->monConfig, virObjectUnref);
}
qemuDomainObjStopWorker(vm);
/* Remove the master key */
qemuDomainMasterKeyRemove(priv);
/* Do this before we delete the tree and remove pidfile. */
qemuProcessKillManagedPRDaemon(vm);
ignore_value(virDomainChrDefForeach(vm->def,
false,
qemuProcessCleanupChardevDevice,
NULL));
/* shut it off for sure */
ignore_value(qemuProcessKill(vm,
VIR_QEMU_PROCESS_KILL_FORCE|
VIR_QEMU_PROCESS_KILL_NOCHECK));
/* Its namespace is also gone then. */
qemuDomainDestroyNamespace(driver, vm);
qemuDomainCleanupRun(driver, vm);
outgoingMigration = (flags & VIR_QEMU_PROCESS_STOP_MIGRATED) &&
(asyncJob == VIR_ASYNC_JOB_MIGRATION_OUT);
qemuExtDevicesStop(driver, vm, outgoingMigration);
qemuDBusStop(driver, vm);
vm->def->id = -1;
/* Wake up anything waiting on domain condition */
virDomainObjBroadcast(vm);
virFileDeleteTree(priv->libDir);
virFileDeleteTree(priv->channelTargetDir);
/* Stop autodestroy in case guest is restarted */
virCloseCallbacksDomainRemove(vm, NULL, qemuProcessAutoDestroy);
/* now that we know it's stopped call the hook if present */
if (virHookPresent(VIR_HOOK_DRIVER_QEMU)) {
g_autofree char *xml = qemuDomainDefFormatXML(driver, NULL, vm->def, 0);
/* we can't stop the operation even if the script raised an error */
ignore_value(virHookCall(VIR_HOOK_DRIVER_QEMU, vm->def->name,
VIR_HOOK_QEMU_OP_STOPPED, VIR_HOOK_SUBOP_END,
NULL, xml, NULL));
}
/* Reset Security Labels unless caller don't want us to */
if (!(flags & VIR_QEMU_PROCESS_STOP_NO_RELABEL))
qemuSecurityRestoreAllLabel(driver, vm,
!!(flags & VIR_QEMU_PROCESS_STOP_MIGRATED));
/* Clear out dynamically assigned labels */
for (i = 0; i < vm->def->nseclabels; i++) {
if (vm->def->seclabels[i]->type == VIR_DOMAIN_SECLABEL_DYNAMIC)
VIR_FREE(vm->def->seclabels[i]->label);
VIR_FREE(vm->def->seclabels[i]->imagelabel);
}
qemuHostdevReAttachDomainDevices(driver, vm->def);
for (i = 0; i < def->nnets; i++) {
virDomainNetDef *net = def->nets[i];
virDomainInterfaceDeleteDevice(def,
net,
QEMU_DOMAIN_NETWORK_PRIVATE(net)->created,
cfg->stateDir);
}
retry:
if ((ret = virDomainCgroupRemoveCgroup(vm, priv->cgroup, priv->machineName)) < 0) {
if (ret == -EBUSY && (retries++ < 5)) {
g_usleep(200*1000);
goto retry;
}
VIR_WARN("Failed to remove cgroup for %s",
vm->def->name);
}
/* Remove resctrl allocation after cgroups are cleaned up which makes it
* kind of safer (although removing the allocation should work even with
* pids in tasks file */
for (i = 0; i < vm->def->nresctrls; i++) {
size_t j = 0;
for (j = 0; j < vm->def->resctrls[i]->nmonitors; j++) {
virDomainResctrlMonDef *mon = NULL;
mon = vm->def->resctrls[i]->monitors[j];
virResctrlMonitorRemove(mon->instance);
}
virResctrlAllocRemove(vm->def->resctrls[i]->alloc);
}
qemuProcessRemoveDomainStatus(driver, vm);
/* Remove VNC and Spice ports from port reservation bitmap, but only if
they were reserved by the driver (autoport=yes)
*/
for (i = 0; i < vm->def->ngraphics; ++i) {
virDomainGraphicsDef *graphics = vm->def->graphics[i];
if (graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC) {
if (graphics->data.vnc.portReserved) {
virPortAllocatorRelease(graphics->data.vnc.port);
graphics->data.vnc.portReserved = false;
}
if (graphics->data.vnc.websocketReserved) {
virPortAllocatorRelease(graphics->data.vnc.websocket);
graphics->data.vnc.websocketReserved = false;
}
if (graphics->data.vnc.websocketGenerated) {
graphics->data.vnc.websocketGenerated = false;
graphics->data.vnc.websocket = -1;
}
}
if (graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_SPICE) {
if (graphics->data.spice.portReserved) {
virPortAllocatorRelease(graphics->data.spice.port);
graphics->data.spice.portReserved = false;
}
if (graphics->data.spice.tlsPortReserved) {
virPortAllocatorRelease(graphics->data.spice.tlsPort);
graphics->data.spice.tlsPortReserved = false;
}
}
}
for (i = 0; i < vm->ndeprecations; i++)
g_free(vm->deprecations[i]);
g_clear_pointer(&vm->deprecations, g_free);
vm->ndeprecations = 0;
vm->taint = 0;
vm->pid = 0;
virDomainObjSetState(vm, VIR_DOMAIN_SHUTOFF, reason);
for (i = 0; i < vm->def->niothreadids; i++)
vm->def->iothreadids[i]->thread_id = 0;
/* clean up a possible backup job */
if (priv->backup)
qemuBackupJobTerminate(vm, VIR_DOMAIN_JOB_STATUS_CANCELED);
/* Do this explicitly after vm->pid is reset so that security drivers don't
* try to enter the domain's namespace which is non-existent by now as qemu
* is no longer running. */
if (!(flags & VIR_QEMU_PROCESS_STOP_NO_RELABEL)) {
for (i = 0; i < def->ndisks; i++) {
virDomainDiskDef *disk = def->disks[i];
if (disk->mirror) {
if (qemuSecurityRestoreImageLabel(driver, vm, disk->mirror, false) < 0)
VIR_WARN("Unable to restore security label on %s", disk->dst);
if (virStorageSourceChainHasNVMe(disk->mirror))
qemuHostdevReAttachOneNVMeDisk(driver, vm->def->name, disk->mirror);
}
qemuBlockRemoveImageMetadata(driver, vm, disk->dst, disk->src);
/* for now transient disks are forbidden with migration so they
* can be handled here */
if (disk->transient &&
QEMU_DOMAIN_DISK_PRIVATE(disk)->transientOverlayCreated) {
VIR_DEBUG("Removing transient overlay '%s' of disk '%s'",
disk->src->path, disk->dst);
if (qemuDomainStorageFileInit(driver, vm, disk->src, NULL) >= 0) {
virStorageSourceUnlink(disk->src);
virStorageSourceDeinit(disk->src);
}
}
}
}
qemuSecurityReleaseLabel(driver->securityManager, vm->def);
/* clear all private data entries which are no longer needed */
qemuDomainObjPrivateDataClear(priv);
/* The "release" hook cleans up additional resources */
if (virHookPresent(VIR_HOOK_DRIVER_QEMU)) {
g_autofree char *xml = qemuDomainDefFormatXML(driver, NULL, vm->def, 0);
/* we can't stop the operation even if the script raised an error */
virHookCall(VIR_HOOK_DRIVER_QEMU, vm->def->name,
VIR_HOOK_QEMU_OP_RELEASE, VIR_HOOK_SUBOP_END,
NULL, xml, NULL);
}
virDomainObjRemoveTransientDef(vm);
endjob:
if (asyncJob != VIR_ASYNC_JOB_NONE)
virDomainObjEndJob(vm);
cleanup:
virErrorRestore(&orig_err);
}
void
qemuProcessAutoDestroy(virDomainObj *dom,
virConnectPtr conn)
{
qemuDomainObjPrivate *priv = dom->privateData;
virQEMUDriver *driver = priv->driver;
virObjectEvent *event = NULL;
unsigned int stopFlags = 0;
VIR_DEBUG("vm=%s, conn=%p", dom->def->name, conn);
if (dom->job->asyncJob == VIR_ASYNC_JOB_MIGRATION_IN)
stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
if (dom->job->asyncJob) {
VIR_DEBUG("vm=%s has long-term job active, cancelling",
dom->def->name);
qemuDomainObjDiscardAsyncJob(dom);
}
VIR_DEBUG("Killing domain");
if (qemuProcessBeginStopJob(dom, VIR_JOB_DESTROY, true) < 0)
return;
qemuProcessStop(driver, dom, VIR_DOMAIN_SHUTOFF_DESTROYED,
VIR_ASYNC_JOB_NONE, stopFlags);
virDomainAuditStop(dom, "destroyed");
event = virDomainEventLifecycleNewFromObj(dom,
VIR_DOMAIN_EVENT_STOPPED,
VIR_DOMAIN_EVENT_STOPPED_DESTROYED);
qemuDomainRemoveInactive(driver, dom, 0, false);
virDomainObjEndJob(dom);
virObjectEventStateQueue(driver->domainEventState, event);
}
void
qemuProcessRefreshDiskProps(virDomainDiskDef *disk,
struct qemuDomainDiskInfo *info)
{
qemuDomainDiskPrivate *diskpriv = QEMU_DOMAIN_DISK_PRIVATE(disk);
if (info->removable) {
if (info->empty)
virDomainDiskEmptySource(disk);
if (info->tray) {
if (info->tray_open)
disk->tray_status = VIR_DOMAIN_DISK_TRAY_OPEN;
else
disk->tray_status = VIR_DOMAIN_DISK_TRAY_CLOSED;
}
}
diskpriv->removable = info->removable;
diskpriv->tray = info->tray;
}
int
qemuProcessRefreshDisks(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
virQEMUDriver *driver = priv->driver;
g_autoptr(GHashTable) table = NULL;
size_t i;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) == 0) {
table = qemuMonitorGetBlockInfo(priv->mon);
qemuDomainObjExitMonitor(vm);
}
if (!table)
return -1;
for (i = 0; i < vm->def->ndisks; i++) {
virDomainDiskDef *disk = vm->def->disks[i];
qemuDomainDiskPrivate *diskpriv = QEMU_DOMAIN_DISK_PRIVATE(disk);
struct qemuDomainDiskInfo *info;
const char *entryname = disk->info.alias;
virDomainDiskTray old_tray_status = disk->tray_status;
if (diskpriv->qomName)
entryname = diskpriv->qomName;
if (!(info = virHashLookup(table, entryname)))
continue;
qemuProcessRefreshDiskProps(disk, info);
if (diskpriv->tray &&
old_tray_status != disk->tray_status) {
virDomainEventTrayChangeReason reason = VIR_DOMAIN_EVENT_TRAY_CHANGE_OPEN;
virObjectEvent *event;
if (disk->tray_status == VIR_DOMAIN_DISK_TRAY_CLOSED)
reason = VIR_DOMAIN_EVENT_TRAY_CHANGE_CLOSE;
event = virDomainEventTrayChangeNewFromObj(vm, disk->info.alias, reason);
virObjectEventStateQueue(driver->domainEventState, event);
}
}
return 0;
}
static int
qemuProcessRefreshCPUMigratability(virDomainObj *vm,
virDomainAsyncJob asyncJob)
{
qemuDomainObjPrivate *priv = vm->privateData;
virDomainDef *def = vm->def;
const char *cpuQOMPath = qemuProcessGetVCPUQOMPath(vm);
bool migratable;
int rc;
if (def->cpu->mode != VIR_CPU_MODE_HOST_PASSTHROUGH &&
def->cpu->mode != VIR_CPU_MODE_MAXIMUM)
return 0;
/* If the cpu.migratable capability is present, the migratable attribute
* is set correctly. */
if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_CPU_MIGRATABLE))
return 0;
if (!ARCH_IS_X86(def->os.arch))
return 0;
if (qemuDomainObjEnterMonitorAsync(vm, asyncJob) < 0)
return -1;
rc = qemuMonitorGetCPUMigratable(priv->mon, cpuQOMPath, &migratable);
qemuDomainObjExitMonitor(vm);
if (rc < 0)
return -1;
if (rc == 1)
migratable = false;
/* Libvirt 6.5.0 would set migratable='off' for running domains even though
* the actual default used by QEMU was 'on'. */
if (def->cpu->migratable == VIR_TRISTATE_SWITCH_OFF && migratable) {
VIR_DEBUG("Fixing CPU migratable attribute");
def->cpu->migratable = VIR_TRISTATE_SWITCH_ON;
}
if (def->cpu->migratable == VIR_TRISTATE_SWITCH_ABSENT)
def->cpu->migratable = virTristateSwitchFromBool(migratable);
return 0;
}
static int
qemuProcessRefreshCPU(virQEMUDriver *driver,
virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
g_autoptr(virCPUDef) host = NULL;
g_autoptr(virCPUDef) hostmig = NULL;
g_autoptr(virCPUDef) cpu = NULL;
if (!virQEMUCapsGuestIsNative(driver->hostarch, vm->def->os.arch))
return 0;
if (!vm->def->cpu)
return 0;
if (qemuProcessRefreshCPUMigratability(vm, VIR_ASYNC_JOB_NONE) < 0)
return -1;
if (!(host = virQEMUDriverGetHostCPU(driver))) {
virResetLastError();
return 0;
}
/* If the domain with a host-model CPU was started by an old libvirt
* (< 2.3) which didn't replace the CPU with a custom one, let's do it now
* since the rest of our code does not really expect a host-model CPU in a
* running domain.
*/
if (vm->def->cpu->mode == VIR_CPU_MODE_HOST_MODEL) {
/*
* PSeries domains are able to run with host-model CPU by design,
* even on Libvirt newer than 2.3, never replacing host-model with
* custom in the virCPUUpdate() call. It is not needed to call
* virCPUUpdate() and qemuProcessUpdateCPU() in this case.
*/
if (qemuDomainIsPSeries(vm->def))
return 0;
if (!(hostmig = virCPUCopyMigratable(host->arch, host)))
return -1;
cpu = virCPUDefCopyWithoutModel(hostmig);
virCPUDefCopyModelFilter(cpu, hostmig, false, virQEMUCapsCPUFilterFeatures,
&host->arch);
if (virCPUUpdate(vm->def->os.arch, vm->def->cpu, cpu) < 0)
return -1;
if (qemuProcessUpdateCPU(vm, VIR_ASYNC_JOB_NONE) < 0)
return -1;
} else if (!virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_QUERY_CPU_MODEL_EXPANSION)) {
/* We only try to fix CPUs when the libvirt/QEMU combo used to start
* the domain did not know about query-cpu-model-expansion in which
* case the host-model is known to not contain features which QEMU
* doesn't know about.
*/
if (qemuDomainFixupCPUs(vm, &priv->origCPU) < 0)
return -1;
}
return 0;
}
/**
* qemuProcessReloadMachineTypes:
*
* Reload machine type information into the 'qemuCaps' object from the current
* qemu.
*/
static int
qemuProcessReloadMachineTypes(virDomainObj *vm)
{
qemuDomainObjPrivate *priv = vm->privateData;
bool fail = false;
qemuDomainObjEnterMonitor(vm);
if (virQEMUCapsInitQMPArch(priv->qemuCaps, priv->mon) < 0)
fail = true;
if (!fail &&
virQEMUCapsProbeQMPMachineTypes(priv->qemuCaps,
vm->def->virtType,
priv->mon) < 0)
fail = true;
qemuDomainObjExitMonitor(vm);
if (fail)
return -1;
return 0;
}
struct qemuProcessReconnectData {
virQEMUDriver *driver;
virDomainObj *obj;
virIdentity *identity;
};
/*
* Open an existing VM's monitor, re-detect VCPU threads
* and re-reserve the security labels in use
*
* This function also inherits a locked and ref'd domain object.
*
* This function needs to:
* 1. Enter job
* 1. just before monitor reconnect do lightweight MonitorEnter
* (increase VM refcount and unlock VM)
* 2. reconnect to monitor
* 3. do lightweight MonitorExit (lock VM)
* 4. continue reconnect process
* 5. EndJob
*
* We can't do normal MonitorEnter & MonitorExit because these two lock the
* monitor lock, which does not exists in this early phase.
*/
static void
qemuProcessReconnect(void *opaque)
{
struct qemuProcessReconnectData *data = opaque;
virQEMUDriver *driver = data->driver;
virDomainObj *obj = data->obj;
qemuDomainObjPrivate *priv;
g_auto(virDomainJobObj) oldjob = {
.cb = NULL,
};
int state;
int reason;
g_autoptr(virQEMUDriverConfig) cfg = NULL;
size_t i;
unsigned int stopFlags = 0;
bool jobStarted = false;
bool tryMonReconn = false;
virIdentitySetCurrent(data->identity);
g_clear_object(&data->identity);
VIR_FREE(data);
cfg = virQEMUDriverGetConfig(driver);
priv = obj->privateData;
virDomainObjPreserveJob(obj->job, &oldjob);
if (oldjob.asyncJob == VIR_ASYNC_JOB_MIGRATION_IN)
stopFlags |= VIR_QEMU_PROCESS_STOP_MIGRATED;
if (oldjob.asyncJob == VIR_ASYNC_JOB_BACKUP && priv->backup)
priv->backup->apiFlags = oldjob.apiFlags;
if (virDomainObjBeginJob(obj, VIR_JOB_MODIFY) < 0)
goto error;
jobStarted = true;
/* XXX If we ever gonna change pid file pattern, come up with
* some intelligence here to deal with old paths. */
if (!(priv->pidfile = virPidFileBuildPath(cfg->stateDir, obj->def->name)))
goto error;
/* Restore the masterKey */
if (qemuDomainMasterKeyReadFile(priv) < 0)
goto error;
if (qemuExtDevicesInitPaths(driver, obj->def) < 0)
goto error;
/* If we are connecting to a guest started by old libvirt there is no
* allowReboot in status XML and we need to initialize it. */
qemuProcessPrepareAllowReboot(obj);
if (qemuHostdevUpdateActiveDomainDevices(driver, obj->def) < 0)
goto error;
if (qemuDomainObjStartWorker(obj) < 0)
goto error;
VIR_DEBUG("Reconnect monitor to def=%p name='%s'", obj, obj->def->name);
tryMonReconn = true;
/* XXX check PID liveliness & EXE path */
if (qemuConnectMonitor(driver, obj, VIR_ASYNC_JOB_NONE, NULL, true) < 0)
goto error;
priv->machineName = qemuDomainGetMachineName(obj);
if (!priv->machineName)
goto error;
if (virDomainCgroupConnectCgroup("qemu",
obj,
&priv->cgroup,
cfg->cgroupControllers,
priv->driver->privileged,
priv->machineName) < 0)
goto error;
if (qemuDomainPerfRestart(obj) < 0)
goto error;
for (i = 0; i < obj->def->ndisks; i++) {
virDomainDiskDef *disk = obj->def->disks[i];
if (virDomainDiskTranslateSourcePool(disk) < 0)
goto error;
}
for (i = 0; i < obj->def->ngraphics; i++) {
if (qemuProcessGraphicsReservePorts(obj->def->graphics[i], true) < 0)
goto error;
}
if (qemuProcessUpdateState(obj) < 0)
goto error;
state = virDomainObjGetState(obj, &reason);
if (state == VIR_DOMAIN_SHUTOFF ||
(state == VIR_DOMAIN_PAUSED &&
reason == VIR_DOMAIN_PAUSED_STARTING_UP)) {
VIR_DEBUG("Domain '%s' wasn't fully started yet, killing it",
obj->def->name);
goto error;
}
if (!priv->qemuCaps) {
virReportError(VIR_ERR_INTERNAL_ERROR,
_("domain '%1$s' has no capabilities recorded"),
obj->def->name);
goto error;
}
/* Reload and populate machine type data into 'qemuCaps' as that is not
* serialized into the status XML. */
if (qemuProcessReloadMachineTypes(obj) < 0)
goto error;
if (qemuDomainAssignAddresses(obj->def, priv->qemuCaps,
driver, obj, false) < 0) {
goto error;
}
/* In case the domain shutdown or fake reboot while we were not running,
* we need to finish the shutdown or fake reboot process. And we need to
* do it after we have virQEMUCaps filled in.
*/
if (state == VIR_DOMAIN_SHUTDOWN ||
(state == VIR_DOMAIN_PAUSED &&
reason == VIR_DOMAIN_PAUSED_SHUTTING_DOWN) ||
(priv->fakeReboot && state == VIR_DOMAIN_PAUSED &&
reason == VIR_DOMAIN_PAUSED_USER)) {
VIR_DEBUG("Finishing shutdown sequence for domain %s",
obj->def->name);
qemuProcessShutdownOrReboot(obj);
goto cleanup;
}
/* if domain requests security driver we haven't loaded, report error, but
* do not kill the domain
*/
ignore_value(qemuSecurityCheckAllLabel(driver->securityManager,
obj->def));
if (qemuDomainRefreshVcpuInfo(obj, VIR_ASYNC_JOB_NONE, true) < 0)
goto error;
qemuDomainVcpuPersistOrder(obj->def);
if (qemuProcessRefreshCPU(driver, obj) < 0)
goto error;
if (qemuDomainUpdateMemoryDeviceInfo(obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
if (qemuProcessDetectIOThreadPIDs(obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
if (qemuSecurityReserveLabel(driver->securityManager, obj->def, obj->pid) < 0)
goto error;
if (qemuProcessRefreshRxFilters(obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
qemuProcessNotifyNets(obj->def);
qemuProcessFiltersInstantiate(obj->def);
if (qemuProcessRefreshDisks(obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
/* At this point we've already checked that the startup of the VM was
* completed successfully before, thus that also implies that all transient
* disk overlays were created. */
for (i = 0; i < obj->def->ndisks; i++) {
virDomainDiskDef *disk = obj->def->disks[i];
if (disk->transient)
QEMU_DOMAIN_DISK_PRIVATE(disk)->transientOverlayCreated = true;
}
if (qemuRefreshVirtioChannelState(driver, obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
/* If querying of guest's RTC failed, report error, but do not kill the domain. */
qemuRefreshRTC(obj);
if (qemuProcessRefreshBalloonState(obj, VIR_ASYNC_JOB_NONE) < 0)
goto error;
if (qemuProcessRecoverJob(driver, obj, &oldjob, &stopFlags) < 0)
goto error;
if (qemuBlockJobRefreshJobs(obj) < 0)
goto error;
if (qemuProcessUpdateDevices(driver, obj) < 0)
goto error;
if (qemuRefreshPRManagerState(obj) < 0)
goto error;
if (qemuProcessRefreshFdsetIndex(obj) < 0)
goto error;
if (qemuConnectAgent(driver, obj) < 0)
goto error;
for (i = 0; i < obj->def->nresctrls; i++) {
size_t j = 0;
if (virResctrlAllocDeterminePath(obj->def->resctrls[i]->alloc,
priv->machineName) < 0)
goto error;
for (j = 0; j < obj->def->resctrls[i]->nmonitors; j++) {
virDomainResctrlMonDef *mon = NULL;
mon = obj->def->resctrls[i]->monitors[j];
if (virResctrlMonitorDeterminePath(mon->instance,
priv->machineName) < 0)
goto error;
}
}
for (i = 0; i < obj->def->ndisks; i++)
if (qemuNbdkitStorageSourceManageProcess(obj->def->disks[i]->src, obj) < 0)
goto error;
if (obj->def->os.loader && obj->def->os.loader->nvram)
if (qemuNbdkitStorageSourceManageProcess(obj->def->os.loader->nvram, obj) < 0)
goto error;
/* update domain state XML with possibly updated state in virDomainObj */
if (virDomainObjSave(obj, driver->xmlopt, cfg->stateDir) < 0)
goto error;
/* Run an hook to allow admins to do some magic */
if (virHookPresent(VIR_HOOK_DRIVER_QEMU)) {
g_autofree char *xml = qemuDomainDefFormatXML(driver,
priv->qemuCaps,
obj->def, 0);
int hookret;
hookret = virHookCall(VIR_HOOK_DRIVER_QEMU, obj->def->name,
VIR_HOOK_QEMU_OP_RECONNECT, VIR_HOOK_SUBOP_BEGIN,
NULL, xml, NULL);
/*
* If the script raised an error abort the launch
*/
if (hookret < 0)
goto error;
}
if (g_atomic_int_add(&driver->nactive, 1) == 0 && driver->inhibitCallback)
driver->inhibitCallback(true, driver->inhibitOpaque);
cleanup:
if (jobStarted)
virDomainObjEndJob(obj);
if (!virDomainObjIsActive(obj))
qemuDomainRemoveInactive(driver, obj, 0, false);
virDomainObjEndAPI(&obj);
virIdentitySetCurrent(NULL);
return;
error:
if (virDomainObjIsActive(obj)) {
/* We can't get the monitor back, so must kill the VM
* to remove danger of it ending up running twice if
* user tries to start it again later.
*
* If we cannot get to the monitor when the QEMU command
* line used -no-shutdown, then we can safely say that the
* domain crashed; otherwise, if the monitor was started,
* then we can blame ourselves, else we failed before the
* monitor started so we don't really know. */
if (!priv->mon && tryMonReconn &&
(priv->allowReboot == VIR_TRISTATE_BOOL_YES ||
virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_SET_ACTION)))
state = VIR_DOMAIN_SHUTOFF_CRASHED;
else if (priv->mon)
state = VIR_DOMAIN_SHUTOFF_DAEMON;
else
state = VIR_DOMAIN_SHUTOFF_UNKNOWN;
/* If BeginJob failed, we jumped here without a job, let's hope another
* thread didn't have a chance to start playing with the domain yet
* (it's all we can do anyway).
*/
qemuProcessStop(driver, obj, state, VIR_ASYNC_JOB_NONE, stopFlags);
}
goto cleanup;
}
static int
qemuProcessReconnectHelper(virDomainObj *obj,
void *opaque)
{
virThread thread;
struct qemuProcessReconnectData *src = opaque;
struct qemuProcessReconnectData *data;
g_autofree char *name = NULL;
/* If the VM was inactive, we don't need to reconnect */
if (obj->pid == 0)
return 0;
data = g_new0(struct qemuProcessReconnectData, 1);
memcpy(data, src, sizeof(*data));
data->obj = obj;
data->identity = virIdentityGetCurrent();
/* this lock and reference will be eventually transferred to the thread
* that handles the reconnect */
virObjectLock(obj);
virObjectRef(obj);
name = g_strdup_printf("init-%s", obj->def->name);
if (virThreadCreateFull(&thread, false, qemuProcessReconnect,
name, false, data) < 0) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("Could not create thread. QEMU initialization might be incomplete"));
/* We can't spawn a thread and thus connect to monitor. Kill qemu.
* It's safe to call qemuProcessStop without a job here since there
* is no thread that could be doing anything else with the same domain
* object.
*/
qemuProcessStop(src->driver, obj, VIR_DOMAIN_SHUTOFF_FAILED,
VIR_ASYNC_JOB_NONE, 0);
qemuDomainRemoveInactiveLocked(src->driver, obj);
virDomainObjEndAPI(&obj);
g_clear_object(&data->identity);
VIR_FREE(data);
return -1;
}
return 0;
}
/**
* qemuProcessReconnectAll
*
* Try to re-open the resources for live VMs that we care
* about.
*/
void
qemuProcessReconnectAll(virQEMUDriver *driver)
{
struct qemuProcessReconnectData data = {.driver = driver};
virDomainObjListForEach(driver->domains, true,
qemuProcessReconnectHelper, &data);
}
static void virQEMUCapsMonitorNotify(qemuMonitor *mon G_GNUC_UNUSED,
virDomainObj *vm G_GNUC_UNUSED)
{
}
static qemuMonitorCallbacks callbacks = {
.eofNotify = virQEMUCapsMonitorNotify,
.errorNotify = virQEMUCapsMonitorNotify,
};
static void
qemuProcessQMPStop(qemuProcessQMP *proc)
{
if (proc->mon) {
virObjectUnlock(proc->mon);
g_clear_pointer(&proc->mon, qemuMonitorClose);
}
if (proc->cmd) {
virCommandAbort(proc->cmd);
g_clear_pointer(&proc->cmd, virCommandFree);
}
if (proc->monpath)
unlink(proc->monpath);
virDomainObjEndAPI(&proc->vm);
if (proc->pid != 0) {
VIR_DEBUG("Killing QMP caps process %lld", (long long)proc->pid);
virProcessKillPainfully(proc->pid, true);
virResetLastError();
proc->pid = 0;
}
if (proc->pidfile)
unlink(proc->pidfile);
if (proc->uniqDir)
rmdir(proc->uniqDir);
}
/**
* qemuProcessQMPFree:
* @proc: Stores process and connection state
*
* Kill QEMU process and free process data structure.
*/
void
qemuProcessQMPFree(qemuProcessQMP *proc)
{
if (!proc)
return;
qemuProcessQMPStop(proc);
g_object_unref(proc->eventThread);
g_free(proc->binary);
g_free(proc->libDir);
g_free(proc->uniqDir);
g_free(proc->monpath);
g_free(proc->monarg);
g_free(proc->pidfile);
g_free(proc->stdErr);
g_free(proc);
}
/**
* qemuProcessQMPNew:
* @binary: QEMU binary
* @libDir: Directory for process and connection artifacts
* @runUid: UserId for QEMU process
* @runGid: GroupId for QEMU process
* @forceTCG: Force TCG mode if true
*
* Allocate and initialize domain structure encapsulating QEMU process state
* and monitor connection for completing QMP queries.
*/
qemuProcessQMP *
qemuProcessQMPNew(const char *binary,
const char *libDir,
uid_t runUid,
gid_t runGid,
bool forceTCG)
{
g_autoptr(qemuProcessQMP) proc = NULL;
const char *threadSuffix;
g_autofree char *threadName = NULL;
VIR_DEBUG("exec=%s, libDir=%s, runUid=%u, runGid=%u, forceTCG=%d",
binary, libDir, runUid, runGid, forceTCG);
proc = g_new0(qemuProcessQMP, 1);
proc->binary = g_strdup(binary);
proc->libDir = g_strdup(libDir);
proc->runUid = runUid;
proc->runGid = runGid;
proc->forceTCG = forceTCG;
threadSuffix = strrchr(binary, '-');
if (threadSuffix)
threadSuffix++;
else
threadSuffix = binary;
threadName = g_strdup_printf("qmp-%s", threadSuffix);
if (!(proc->eventThread = virEventThreadNew(threadName)))
return NULL;
return g_steal_pointer(&proc);
}
static int
qemuProcessQEMULabelUniqPath(qemuProcessQMP *proc)
{
/* We cannot use the security driver here, but we should not need to. */
if (chown(proc->uniqDir, proc->runUid, -1) < 0) {
virReportSystemError(errno,
_("Cannot chown uniq path: %1$s"),
proc->uniqDir);
return -1;
}
return 0;
}
static int
qemuProcessQMPInit(qemuProcessQMP *proc)
{
g_autofree char *template = NULL;
VIR_DEBUG("proc=%p, emulator=%s", proc, proc->binary);
template = g_strdup_printf("%s/qmp-XXXXXX", proc->libDir);
if (!(proc->uniqDir = g_mkdtemp(template))) {
virReportSystemError(errno,
_("Failed to create unique directory with template '%1$s' for probing QEMU"),
template);
return -1;
}
/* if g_mkdtemp succeeds, proc->uniqDir is now the owner of
* the string. Set template to NULL to avoid freeing
* the memory in this case */
template = NULL;
if (qemuProcessQEMULabelUniqPath(proc) < 0)
return -1;
proc->monpath = g_strdup_printf("%s/%s", proc->uniqDir, "qmp.monitor");
proc->monarg = g_strdup_printf("unix:%s,server=on,wait=off", proc->monpath);
/*
* Normally we'd use runDir for pid files, but because we're using
* -daemonize we need QEMU to be allowed to create them, rather
* than libvirtd. So we're using libDir which QEMU can write to
*/
proc->pidfile = g_strdup_printf("%s/%s", proc->uniqDir, "qmp.pid");
return 0;
}
#if defined(__linux__)
# define hwaccel "kvm:tcg"
#elif defined(__APPLE__)
# define hwaccel "hvf:tcg"
#else
# define hwaccel "tcg"
#endif
static int
qemuProcessQMPLaunch(qemuProcessQMP *proc)
{
const char *machine;
int status = 0;
int rc;
if (proc->forceTCG)
machine = "none,accel=tcg";
else
machine = "none,accel=" hwaccel;
VIR_DEBUG("Try to probe capabilities of '%s' via QMP, machine %s",
proc->binary, machine);
/*
* We explicitly need to use -daemonize here, rather than
* virCommandDaemonize, because we need to synchronize
* with QEMU creating its monitor socket API. Using
* daemonize guarantees control won't return to libvirt
* until the socket is present.
*/
proc->cmd = virCommandNewArgList(proc->binary,
"-S",
"-no-user-config",
"-nodefaults",
"-nographic",
"-machine", machine,
"-qmp", proc->monarg,
"-pidfile", proc->pidfile,
"-daemonize",
NULL);
virCommandAddEnvPassCommon(proc->cmd);
virCommandClearCaps(proc->cmd);
#if WITH_CAPNG
/* QEMU might run into permission issues, e.g. /dev/sev (0600), override
* them just for the purpose of probing */
if (geteuid() == 0)
virCommandAllowCap(proc->cmd, CAP_DAC_OVERRIDE);
#endif
virCommandSetGID(proc->cmd, proc->runGid);
virCommandSetUID(proc->cmd, proc->runUid);
virCommandSetErrorBuffer(proc->cmd, &(proc->stdErr));
if (virCommandRun(proc->cmd, &status) < 0)
return -1;
if (status != 0) {
VIR_DEBUG("QEMU %s exited with status %d", proc->binary, status);
virReportError(VIR_ERR_INTERNAL_ERROR,
_("Failed to start QEMU binary %1$s for probing: %2$s"),
proc->binary,
proc->stdErr ? proc->stdErr : _("unknown error"));
return -1;
}
if ((rc = virPidFileReadPath(proc->pidfile, &proc->pid)) < 0) {
virReportSystemError(-rc, _("Failed to read pidfile %1$s"), proc->pidfile);
return -1;
}
return 0;
}
int
qemuProcessQMPInitMonitor(qemuMonitor *mon)
{
if (qemuMonitorSetCapabilities(mon) < 0) {
VIR_DEBUG("Failed to set monitor capabilities %s",
virGetLastErrorMessage());
return -1;
}
return 0;
}
static int
qemuProcessQMPConnectMonitor(qemuProcessQMP *proc)
{
g_autoptr(virDomainXMLOption) xmlopt = NULL;
virDomainChrSourceDef monConfig;
VIR_DEBUG("proc=%p, emulator=%s, proc->pid=%lld",
proc, proc->binary, (long long)proc->pid);
monConfig.type = VIR_DOMAIN_CHR_TYPE_UNIX;
monConfig.data.nix.path = proc->monpath;
monConfig.data.nix.listen = false;
if (!(xmlopt = virDomainXMLOptionNew(NULL, NULL, NULL, NULL, NULL, NULL)) ||
!(proc->vm = virDomainObjNew(xmlopt)) ||
!(proc->vm->def = virDomainDefNew(xmlopt)))
return -1;
proc->vm->pid = proc->pid;
if (!(proc->mon = qemuMonitorOpen(proc->vm, &monConfig,
virEventThreadGetContext(proc->eventThread),
&callbacks)))
return -1;
virObjectLock(proc->mon);
if (qemuProcessQMPInitMonitor(proc->mon) < 0)
return -1;
return 0;
}
/**
* qemuProcessQMPStart:
* @proc: QEMU process and connection state created by qemuProcessQMPNew()
*
* Start and connect to QEMU binary so QMP queries can be made.
*
* Usage:
* proc = qemuProcessQMPNew(binary, libDir, runUid, runGid, forceTCG);
* qemuProcessQMPStart(proc);
* ** Send QMP Queries to QEMU using monitor (proc->mon) **
* qemuProcessQMPFree(proc);
*
* Process error output (proc->stdErr) remains available in qemuProcessQMP
* struct until qemuProcessQMPFree is called.
*/
int
qemuProcessQMPStart(qemuProcessQMP *proc)
{
VIR_DEBUG("proc=%p, emulator=%s", proc, proc->binary);
if (qemuProcessQMPInit(proc) < 0)
return -1;
if (qemuProcessQMPLaunch(proc) < 0)
return -1;
if (qemuProcessQMPConnectMonitor(proc) < 0)
return -1;
return 0;
}
void
qemuProcessHandleNbdkitExit(qemuNbdkitProcess *nbdkit,
virDomainObj *vm)
{
virObjectLock(vm);
VIR_DEBUG("nbdkit process %i died", nbdkit->pid);
qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_NBDKIT_EXITED, 0, 0, nbdkit);
virObjectUnlock(vm);
}
Reference in New Issue View Git Blame Copy Permalink