int
qemuSetupGlobalCpuCgroup(virDomainObjPtr vm)
{
qemuDomainObjPrivatePtr priv = vm->privateData;
unsigned long long period = vm->def->cputune.global_period;
long long quota = vm->def->cputune.global_quota;
char *mem_mask = NULL;
virDomainNumatuneMemMode mem_mode;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
/*
* 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))
return 0;
if (virDomainNumatuneGetMode(vm->def->numa, -1, &mem_mode) == 0 &&
mem_mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT &&
virDomainNumatuneMaybeFormatNodeset(vm->def->numa,
priv->autoNodeset,
&mem_mask, -1) < 0)
goto cleanup;
if (period || quota) {
if (qemuSetupCgroupVcpuBW(priv->cgroup, period, quota) < 0)
goto cleanup;
}
VIR_FREE(mem_mask);
return 0;
cleanup:
VIR_FREE(mem_mask);
return -1;
}
int
qemuSetupCgroupForVcpu(virDomainObjPtr vm)
{
virCgroupPtr cgroup_vcpu = NULL;
qemuDomainObjPrivatePtr priv = vm->privateData;
virDomainDefPtr def = vm->def;
size_t i, j;
unsigned long long period = vm->def->cputune.period;
long long quota = vm->def->cputune.quota;
char *mem_mask = NULL;
virDomainNumatuneMemMode mem_mode;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
/*
* 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))
return 0;
/* We are trying to setup cgroups for CPU pinning, which can also be done
* with virProcessSetAffinity, thus the lack of cgroups is not fatal here.
*/
if (priv->cgroup == NULL)
return 0;
if (priv->nvcpupids == 0 || priv->vcpupids[0] == vm->pid) {
/* If we don't know VCPU<->PID mapping or all vcpu runs in the same
* thread, we cannot control each vcpu.
*/
return 0;
}
if (virDomainNumatuneGetMode(vm->def->numa, -1, &mem_mode) == 0 &&
mem_mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT &&
virDomainNumatuneMaybeFormatNodeset(vm->def->numa,
priv->autoNodeset,
&mem_mask, -1) < 0)
goto cleanup;
for (i = 0; i < priv->nvcpupids; i++) {
virCgroupFree(&cgroup_vcpu);
if (virCgroupNewThread(priv->cgroup, VIR_CGROUP_THREAD_VCPU, i,
true, &cgroup_vcpu) < 0)
goto cleanup;
/* move the thread for vcpu to sub dir */
if (virCgroupAddTask(cgroup_vcpu, priv->vcpupids[i]) < 0)
goto cleanup;
if (period || quota) {
if (qemuSetupCgroupVcpuBW(cgroup_vcpu, period, quota) < 0)
goto cleanup;
}
/* Set vcpupin in cgroup if vcpupin xml is provided */
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET)) {
virBitmapPtr cpumap = NULL;
if (mem_mask &&
virCgroupSetCpusetMems(cgroup_vcpu, mem_mask) < 0)
goto cleanup;
/* try to use the default cpu maps */
if (vm->def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO)
cpumap = priv->autoCpuset;
else
cpumap = vm->def->cpumask;
/* lookup a more specific pinning info */
for (j = 0; j < def->cputune.nvcpupin; j++) {
if (def->cputune.vcpupin[j]->id == i) {
cpumap = def->cputune.vcpupin[j]->cpumask;
break;
}
}
if (!cpumap)
continue;
if (qemuSetupCgroupCpusetCpus(cgroup_vcpu, cpumap) < 0)
goto cleanup;
}
}
virCgroupFree(&cgroup_vcpu);
VIR_FREE(mem_mask);
return 0;
cleanup:
if (cgroup_vcpu) {
virCgroupRemove(cgroup_vcpu);
virCgroupFree(&cgroup_vcpu);
}
VIR_FREE(mem_mask);
return -1;
}
int
qemuSetupCgroupForIOThreads(virDomainObjPtr vm)
{
virCgroupPtr cgroup_iothread = NULL;
qemuDomainObjPrivatePtr priv = vm->privateData;
virDomainDefPtr def = vm->def;
size_t i;
unsigned long long period = vm->def->cputune.period;
long long quota = vm->def->cputune.quota;
char *mem_mask = NULL;
virDomainNumatuneMemMode mem_mode;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
/*
* 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))
return 0;
/* We are trying to setup cgroups for CPU pinning, which can also be done
* with virProcessSetAffinity, thus the lack of cgroups is not fatal here.
*/
if (priv->cgroup == NULL)
return 0;
if (virDomainNumatuneGetMode(vm->def->numa, -1, &mem_mode) == 0 &&
mem_mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT &&
virDomainNumatuneMaybeFormatNodeset(vm->def->numa,
priv->autoNodeset,
&mem_mask, -1) < 0)
goto cleanup;
for (i = 0; i < def->niothreadids; i++) {
/* IOThreads are numbered 1..n, although the array is 0..n-1,
* so we will account for that here
*/
if (virCgroupNewThread(priv->cgroup, VIR_CGROUP_THREAD_IOTHREAD,
def->iothreadids[i]->iothread_id,
true, &cgroup_iothread) < 0)
goto cleanup;
/* move the thread for iothread to sub dir */
if (virCgroupAddTask(cgroup_iothread,
def->iothreadids[i]->thread_id) < 0)
goto cleanup;
if (period || quota) {
if (qemuSetupCgroupVcpuBW(cgroup_iothread, period, quota) < 0)
goto cleanup;
}
/* Set iothreadpin in cgroup if iothreadpin xml is provided */
if (virCgroupHasController(priv->cgroup,
VIR_CGROUP_CONTROLLER_CPUSET)) {
virBitmapPtr cpumask = NULL;
if (mem_mask &&
virCgroupSetCpusetMems(cgroup_iothread, mem_mask) < 0)
goto cleanup;
if (def->iothreadids[i]->cpumask)
cpumask = def->iothreadids[i]->cpumask;
else if (def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO)
cpumask = priv->autoCpuset;
else
cpumask = def->cpumask;
if (cpumask &&
qemuSetupCgroupCpusetCpus(cgroup_iothread, cpumask) < 0)
goto cleanup;
}
virCgroupFree(&cgroup_iothread);
}
VIR_FREE(mem_mask);
return 0;
cleanup:
if (cgroup_iothread) {
virCgroupRemove(cgroup_iothread);
virCgroupFree(&cgroup_iothread);
}
VIR_FREE(mem_mask);
return -1;
}
int
qemuSetupCgroupForEmulator(virDomainObjPtr vm)
{
virBitmapPtr cpumask = NULL;
virCgroupPtr cgroup_emulator = NULL;
virDomainDefPtr def = vm->def;
qemuDomainObjPrivatePtr priv = vm->privateData;
unsigned long long period = vm->def->cputune.emulator_period;
long long quota = vm->def->cputune.emulator_quota;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
/*
* 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))
return 0;
if (priv->cgroup == NULL)
return 0; /* Not supported, so claim success */
if (virCgroupNewThread(priv->cgroup, VIR_CGROUP_THREAD_EMULATOR, 0,
true, &cgroup_emulator) < 0)
goto cleanup;
if (virCgroupMoveTask(priv->cgroup, cgroup_emulator) < 0)
goto cleanup;
if (def->cputune.emulatorpin)
cpumask = def->cputune.emulatorpin;
else if (def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO)
cpumask = priv->autoCpuset;
else if (def->cpumask)
cpumask = def->cpumask;
if (cpumask) {
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET) &&
qemuSetupCgroupCpusetCpus(cgroup_emulator, cpumask) < 0)
goto cleanup;
}
if (period || quota) {
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU) &&
qemuSetupCgroupVcpuBW(cgroup_emulator, period,
quota) < 0)
goto cleanup;
}
virCgroupFree(&cgroup_emulator);
return 0;
cleanup:
if (cgroup_emulator) {
virCgroupRemove(cgroup_emulator);
virCgroupFree(&cgroup_emulator);
}
return -1;
}
int
qemuSetupCgroupForEmulator(virQEMUDriverPtr driver,
virDomainObjPtr vm,
virBitmapPtr nodemask)
{
virBitmapPtr cpumask = NULL;
virBitmapPtr cpumap = NULL;
virCgroupPtr cgroup_emulator = NULL;
virDomainDefPtr def = vm->def;
qemuDomainObjPrivatePtr priv = vm->privateData;
unsigned long long period = vm->def->cputune.emulator_period;
long long quota = vm->def->cputune.emulator_quota;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
if (priv->cgroup == NULL)
return 0; /* Not supported, so claim success */
if (virCgroupNewEmulator(priv->cgroup, true, &cgroup_emulator) < 0)
goto cleanup;
if (virCgroupMoveTask(priv->cgroup, cgroup_emulator) < 0)
goto cleanup;
if (def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO) {
if (!(cpumap = qemuPrepareCpumap(driver, nodemask)))
goto cleanup;
cpumask = cpumap;
} else if (def->cputune.emulatorpin) {
cpumask = def->cputune.emulatorpin->cpumask;
} else if (def->cpumask) {
cpumask = def->cpumask;
}
if (cpumask) {
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET) &&
qemuSetupCgroupEmulatorPin(cgroup_emulator, cpumask) < 0)
goto cleanup;
}
if (period || quota) {
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU) &&
qemuSetupCgroupVcpuBW(cgroup_emulator, period,
quota) < 0)
goto cleanup;
}
virCgroupFree(&cgroup_emulator);
virBitmapFree(cpumap);
return 0;
cleanup:
virBitmapFree(cpumap);
if (cgroup_emulator) {
virCgroupRemove(cgroup_emulator);
virCgroupFree(&cgroup_emulator);
}
return -1;
}
int
qemuSetupCgroupForVcpu(virDomainObjPtr vm)
{
virCgroupPtr cgroup_vcpu = NULL;
qemuDomainObjPrivatePtr priv = vm->privateData;
virDomainDefPtr def = vm->def;
size_t i, j;
unsigned long long period = vm->def->cputune.period;
long long quota = vm->def->cputune.quota;
if ((period || quota) &&
!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPU)) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("cgroup cpu is required for scheduler tuning"));
return -1;
}
/* We are trying to setup cgroups for CPU pinning, which can also be done
* with virProcessInfoSetAffinity, thus the lack of cgroups is not fatal
* here.
*/
if (priv->cgroup == NULL)
return 0;
if (priv->nvcpupids == 0 || priv->vcpupids[0] == vm->pid) {
/* If we don't know VCPU<->PID mapping or all vcpu runs in the same
* thread, we cannot control each vcpu.
*/
VIR_WARN("Unable to get vcpus' pids.");
return 0;
}
for (i = 0; i < priv->nvcpupids; i++) {
if (virCgroupNewVcpu(priv->cgroup, i, true, &cgroup_vcpu) < 0)
goto cleanup;
/* move the thread for vcpu to sub dir */
if (virCgroupAddTask(cgroup_vcpu, priv->vcpupids[i]) < 0)
goto cleanup;
if (period || quota) {
if (qemuSetupCgroupVcpuBW(cgroup_vcpu, period, quota) < 0)
goto cleanup;
}
/* Set vcpupin in cgroup if vcpupin xml is provided */
if (virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET)) {
/* find the right CPU to pin, otherwise
* qemuSetupCgroupVcpuPin will fail. */
for (j = 0; j < def->cputune.nvcpupin; j++) {
if (def->cputune.vcpupin[j]->vcpuid != i)
continue;
if (qemuSetupCgroupVcpuPin(cgroup_vcpu,
def->cputune.vcpupin,
def->cputune.nvcpupin,
i) < 0)
goto cleanup;
break;
}
}
virCgroupFree(&cgroup_vcpu);
}
return 0;
cleanup:
if (cgroup_vcpu) {
virCgroupRemove(cgroup_vcpu);
virCgroupFree(&cgroup_vcpu);
}
return -1;
}
