[libvirt] [PATCH v2] qemu: Reject unsupported tuning in session mode
When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever. When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL). Given the above it seems reasonable to reject such unsupported settings. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1023366 Signed-off-by: Martin Kletzander <mkletzan@redhat.com> --- Notes: v2: - Allow CPU pinning since setting affinity should still work. - Remove bogus 'cfg = ...' in SetBlockIoTune. src/qemu/qemu_command.c | 26 ++++++++++++++++++ src/qemu/qemu_driver.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 98 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 18d0a64..611d21d 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7762,6 +7762,32 @@ qemuBuildCommandLine(virConnectPtr conn, emulator = def->emulator; + if (!cfg->privileged) { + /* If we have no cgroups than we can have no tunings that + * require them */ + + if (def->mem.hard_limit || def->mem.soft_limit || + def->mem.min_guarantee || def->mem.swap_hard_limit) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto error; + } + + if (def->blkio.weight || def->blkio.ndevices) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto error; + } + + if (def->cputune.shares || def->cputune.period || + def->cputune.quota || def->cputune.emulator_period || + def->cputune.emulator_quota) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto error; + } + } + for (i = 0; i < def->ngraphics; ++i) { switch (def->graphics[i]->type) { case VIR_DOMAIN_GRAPHICS_TYPE_SDL: diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 4fbcb27..1e50572 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7434,6 +7434,8 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, char *ret = NULL; virDomainObjPtr vm = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverPtr driver = dom->conn->privateData; + virQEMUDriverConfigPtr cfg = NULL; if (!(vm = qemuDomObjFromDomain(dom))) goto cleanup; @@ -7443,6 +7445,13 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, if (virDomainGetSchedulerTypeEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + /* Domain not running, thus no cgroups - return defaults */ if (!virDomainObjIsActive(vm)) { if (nparams) @@ -7469,6 +7478,7 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, cleanup: if (vm) virObjectUnlock(vm); + virObjectUnref(cfg); return ret; } @@ -7683,6 +7693,12 @@ qemuDomainSetBlkioParameters(virDomainPtr dom, if (virDomainSetBlkioParametersEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -7863,6 +7879,7 @@ qemuDomainGetBlkioParameters(virDomainPtr dom, int ret = -1; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -7881,6 +7898,13 @@ qemuDomainGetBlkioParameters(virDomainPtr dom, if (virDomainGetBlkioParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8269,6 +8293,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -8316,6 +8341,12 @@ qemuDomainSetMemoryParameters(virDomainPtr dom, if (virDomainSetMemoryParametersEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8423,6 +8454,7 @@ qemuDomainGetMemoryParameters(virDomainPtr dom, int ret = -1; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -8439,6 +8471,13 @@ qemuDomainGetMemoryParameters(virDomainPtr dom, if (virDomainGetMemoryParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8550,6 +8589,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -8975,6 +9015,13 @@ qemuDomainSetSchedulerParametersFlags(virDomainPtr dom, if (virDomainSetSchedulerParametersFlagsEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -9210,6 +9257,7 @@ qemuDomainGetSchedulerParametersFlags(virDomainPtr dom, virDomainDefPtr persistentDef; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -9226,6 +9274,13 @@ qemuDomainGetSchedulerParametersFlags(virDomainPtr dom, if (virDomainGetSchedulerParametersFlagsEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + if (*nparams > 1) cpu_bw_status = virCgroupSupportsCpuBW(priv->cgroup); @@ -9320,6 +9375,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -15521,11 +15577,17 @@ qemuDomainSetBlockIoTune(virDomainPtr dom, if (virDomainSetBlockIoTuneEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; priv = vm->privateData; - cfg = virQEMUDriverGetConfig(driver); if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto endjob; @@ -15668,6 +15730,7 @@ qemuDomainGetBlockIoTune(virDomainPtr dom, int ret = -1; size_t i; virCapsPtr caps = NULL; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -15682,6 +15745,13 @@ qemuDomainGetBlockIoTune(virDomainPtr dom, if (virDomainGetBlockIoTuneEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -15784,6 +15854,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } -- 1.9.0
On 03/04/2014 07:15 AM, Martin Kletzander wrote:
When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever.
When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL).
Given the above it seems reasonable to reject such unsupported settings.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1023366
Signed-off-by: Martin Kletzander <mkletzan@redhat.com> ---
@@ -7443,6 +7445,13 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, if (virDomainGetSchedulerTypeEnsureACL(dom->conn, vm->def) < 0) goto cleanup;
+ cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + /* Domain not running, thus no cgroups - return defaults */ if (!virDomainObjIsActive(vm)) {
I can understand failing the Set commands if we can't change things; but since we have a fallback for the Get command even for an offline domain, shouldn't we stick to returning the defaults rather than erroring out? Weak ACK; I'd still wait for Dan to weigh in. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On Tue, Mar 04, 2014 at 02:04:05PM -0700, Eric Blake wrote:
On 03/04/2014 07:15 AM, Martin Kletzander wrote:
When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever.
When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL).
Given the above it seems reasonable to reject such unsupported settings.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1023366
Signed-off-by: Martin Kletzander <mkletzan@redhat.com> ---
@@ -7443,6 +7445,13 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, if (virDomainGetSchedulerTypeEnsureACL(dom->conn, vm->def) < 0) goto cleanup;
+ cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + /* Domain not running, thus no cgroups - return defaults */ if (!virDomainObjIsActive(vm)) {
I can understand failing the Set commands if we can't change things; but since we have a fallback for the Get command even for an offline domain, shouldn't we stick to returning the defaults rather than erroring out?
Particularly for SchedulerType, we can allow the Get method, but the reading will still fail as there are no cgroups set up. If we rework our cgroups code for non-privileged daemons, we will have to check the cgroup in which the domain belongs to, which will be the same as other user processes, and doe to that fact the settings won't apply only to that domain, but for everything in that cgroup; that will make the values less useful (e.g. maximum used memory cannot be guaranteed since the cgroup has more tasks inside than only qemu). Our cgroup path assembly functions will then need to differ based on whether it is non-privileged or not. If that's really needed, I'll rework it that way (even though it'll probably need to get changed back when user cgroups will be user-editable), but until then, could we make the error reported now: $ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Requested operation is not valid: cgroup CPU controller is not mounted changed into: $ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Operation not supported: CPU tuning is not available in session mode ??? // I should've added that info into the commit message, I just forgot. Martin
Weak ACK; I'd still wait for Dan to weigh in.
-- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On 03/05/2014 03:37 AM, Martin Kletzander wrote:
On Tue, Mar 04, 2014 at 02:04:05PM -0700, Eric Blake wrote:
On 03/04/2014 07:15 AM, Martin Kletzander wrote:
When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever.
When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL).
Given the above it seems reasonable to reject such unsupported settings.
I can understand failing the Set commands if we can't change things; but since we have a fallback for the Get command even for an offline domain, shouldn't we stick to returning the defaults rather than erroring out?
If that's really needed, I'll rework it that way (even though it'll probably need to get changed back when user cgroups will be user-editable), but until then, could we make the error reported now:
$ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Requested operation is not valid: cgroup CPU controller is not mounted
changed into:
$ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Operation not supported: CPU tuning is not available in session mode
// I should've added that info into the commit message, I just forgot.
Indeed, trading one error for another nicer one is always safe - and mentioning it in the commit message highlights that it is an improvement (we aren't causing an error on any situation where we used to pass). ACK to this patch, then. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On Wed, Mar 05, 2014 at 06:12:41AM -0700, Eric Blake wrote:
On 03/05/2014 03:37 AM, Martin Kletzander wrote:
On Tue, Mar 04, 2014 at 02:04:05PM -0700, Eric Blake wrote:
On 03/04/2014 07:15 AM, Martin Kletzander wrote:
When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever.
When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL).
Given the above it seems reasonable to reject such unsupported settings.
I can understand failing the Set commands if we can't change things; but since we have a fallback for the Get command even for an offline domain, shouldn't we stick to returning the defaults rather than erroring out?
If that's really needed, I'll rework it that way (even though it'll probably need to get changed back when user cgroups will be user-editable), but until then, could we make the error reported now:
$ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Requested operation is not valid: cgroup CPU controller is not mounted
changed into:
$ virsh -c qemu:///session schedinfo dummy Scheduler : Unknown error: Operation not supported: CPU tuning is not available in session mode
// I should've added that info into the commit message, I just forgot.
Indeed, trading one error for another nicer one is always safe - and mentioning it in the commit message highlights that it is an improvement (we aren't causing an error on any situation where we used to pass). ACK to this patch, then.
I added the info into the commit message and pushed, thanks for the review. Martin
participants (2)
-
Eric Blake -
Martin Kletzander