June 2020 - Devel - libvirt List Archives

Handling merge requests from Weblate translation
by Daniel P. Berrangé 05 Jun '20

05 Jun '20

The libvirt project is now converted to use Weblate for translation. Weblate has its own fork of libvirt git repos. With weblate managing translations, NEVER, under any circumstances should anyone change the .po files in libvirt.git directly. This will result in merge conflicts with Weblate that need tedious manual resolution. Any .po file changes must always be done via the Weblate UI. Changes made in the weblate UI are not committed to its fork immediately. Weblate will wait upto 24 hours before committing. It will keep changes by each translator as separate commits, and will add SoB lines to them. When it has commits ready, it opens a merge request in GitLab. If it has more commits later and the merge request is still open, it will just push to the pre-existing merge request instead of opening a new one. What this means is that if we immediately approve & merge a weblate merge request, it'll likely open a new one the very next day. I don't think we want to be opening & closing 30 merge requests a month for translations. Thus I'm suggesting that we leave weblate merge requests open to accumulate new commits. Only approve & merge them during the freeze period. Any time we update the libvirt.pot, weblate will msgmerge all the .po files and submit them back to libvirt. These are going to be quite large commits. My intention is to only update the libvirt.pot at the time that we start the release freeze period. This leaves a week to receive new translations, during which time libvirt strings shouldn't be changing much anyway, since we're in bugfix only mode. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|

1 0

[libvirt PATCH] qemu: fix typo s/readaehad/readahead/
by Daniel P. Berrangé 05 Jun '20

05 Jun '20

Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed as trivial typo fix src/qemu/qemu_domain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 10f16e9b3b..3239ac1a52 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -5281,7 +5281,7 @@ qemuDomainValidateStorageSource(virStorageSourcePtr src, src->protocol != VIR_STORAGE_NET_PROTOCOL_FTP && src->protocol != VIR_STORAGE_NET_PROTOCOL_FTPS)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("readaehad is supported only with HTTP(S)/FTP(s) protocols")); + _("readahead is supported only with HTTP(S)/FTP(s) protocols")); return -1; } -- 2.26.2

1 0

[PATCH 0/5] qemu: Prefer -numa cpu over -numa node,cpus=
by Michal Privoznik 05 Jun '20

05 Jun '20

Another round of patches to catch up with the latest command line building recommendations. This time, the winner is: -numa cpu Thing is, the way vCPUs are assigned to NUMA nodes is not very deterministic right now. QEMU relies on this weird mapping where firstly threads are iterated over, then cores, then (possibly) dies and then sockets. So they came up with a better schema: -numa cpu,node-id=N,socket-id=S,die-id=D,core-id=C,thread-id=T where everything is specified on the command line. Long story short, some QEMU code is copied over to keep status quo (or, whatever you want). It was buried down down (include/hw/i386/topology.h). I wanted to allow users configure the vCPU topology too, but then decided not to because neither Libvirt nor QEMU can really guarantee the topology will look the same in the guest (at least in my testing with cpu mode='host-passthrough' it didn't). Or, what you're proposing? I can post them as a follow up if needed. Michal Prívozník (5): qemuBuildNumaArgStr: Move vars into loops qemuBuildNumaArgStr: Use g_autofree on @nodeBackends qemuBuildNumaArgStr: Separate out old style of building CPU list qemu: Prefer -numa cpu over -numa node,cpus= virCPUDefParseXML: Parse uint using virXPathUInt src/conf/cpu_conf.c | 14 +- src/qemu/qemu_command.c | 168 +++++++++++++++--- .../hugepages-nvdimm.x86_64-latest.args | 4 +- ...memory-default-hugepage.x86_64-latest.args | 10 +- .../memfd-memory-numa.x86_64-latest.args | 10 +- ...y-hotplug-nvdimm-access.x86_64-latest.args | 4 +- ...ry-hotplug-nvdimm-align.x86_64-latest.args | 4 +- ...ry-hotplug-nvdimm-label.x86_64-latest.args | 4 +- ...ory-hotplug-nvdimm-pmem.x86_64-latest.args | 4 +- ...ory-hotplug-nvdimm-ppc64.ppc64-latest.args | 4 +- ...hotplug-nvdimm-readonly.x86_64-latest.args | 4 +- .../memory-hotplug-nvdimm.x86_64-latest.args | 4 +- ...vhost-user-fs-fd-memory.x86_64-latest.args | 4 +- ...vhost-user-fs-hugepages.x86_64-latest.args | 4 +- ...host-user-gpu-secondary.x86_64-latest.args | 3 +- .../vhost-user-vga.x86_64-latest.args | 3 +- 16 files changed, 195 insertions(+), 53 deletions(-) -- 2.26.2

2 14

[PATCH] numa: forbid '-numa node, mem' for 5.1 and newer machine types
by Igor Mammedov 05 Jun '20

05 Jun '20

Deprecation period is run out and it's a time to flip the switch introduced by cd5ff8333a. Disable legacy option for new machine types (since 5.1) and amend documentation. '-numa node,memdev' shall be used instead of disabled option with new machine types. Signed-off-by: Igor Mammedov <imammedo(a)redhat.com> --- - rebased on top of current master - move compat mode from 4.2 to 5.0 CC: peter.maydell(a)linaro.org CC: ehabkost(a)redhat.com CC: marcel.apfelbaum(a)gmail.com CC: mst(a)redhat.com CC: pbonzini(a)redhat.com CC: rth(a)twiddle.net CC: david(a)gibson.dropbear.id.au CC: libvir-list(a)redhat.com CC: qemu-arm(a)nongnu.org CC: qemu-ppc(a)nongnu.org --- docs/system/deprecated.rst | 17 ----------------- hw/arm/virt.c | 2 +- hw/core/numa.c | 6 ++++++ hw/i386/pc.c | 1 - hw/i386/pc_piix.c | 1 + hw/i386/pc_q35.c | 1 + hw/ppc/spapr.c | 2 +- qemu-options.hx | 9 +++++---- 8 files changed, 15 insertions(+), 24 deletions(-) diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index f0061f94aa..57edc075c2 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -101,23 +101,6 @@ error in the future. The ``-realtime mlock=on|off`` argument has been replaced by the ``-overcommit mem-lock=on|off`` argument. -``-numa node,mem=``\ *size* (since 4.1) -''''''''''''''''''''''''''''''''''''''' - -The parameter ``mem`` of ``-numa node`` is used to assign a part of -guest RAM to a NUMA node. But when using it, it's impossible to manage specified -RAM chunk on the host side (like bind it to a host node, setting bind policy, ...), -so guest end-ups with the fake NUMA configuration with suboptiomal performance. -However since 2014 there is an alternative way to assign RAM to a NUMA node -using parameter ``memdev``, which does the same as ``mem`` and adds -means to actualy manage node RAM on the host side. Use parameter ``memdev`` -with *memory-backend-ram* backend as an replacement for parameter ``mem`` -to achieve the same fake NUMA effect or a properly configured -*memory-backend-file* backend to actually benefit from NUMA configuration. -In future new machine versions will not accept the option but it will still -work with old machine types. User can check QAPI schema to see if the legacy -option is supported by looking at MachineInfo::numa-mem-supported property. - ``-numa`` node (without memory specified) (since 4.1) ''''''''''''''''''''''''''''''''''''''''''''''''''''' diff --git a/hw/arm/virt.c b/hw/arm/virt.c index 37462a6f78..063d4703f7 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -2262,7 +2262,6 @@ static void virt_machine_class_init(ObjectClass *oc, void *data) hc->pre_plug = virt_machine_device_pre_plug_cb; hc->plug = virt_machine_device_plug_cb; hc->unplug_request = virt_machine_device_unplug_request_cb; - mc->numa_mem_supported = true; mc->nvdimm_supported = true; mc->auto_enable_numa_with_memhp = true; mc->default_ram_id = "mach-virt.ram"; @@ -2375,6 +2374,7 @@ DEFINE_VIRT_MACHINE_AS_LATEST(5, 1) static void virt_machine_5_0_options(MachineClass *mc) { virt_machine_5_1_options(mc); + mc->numa_mem_supported = true; } DEFINE_VIRT_MACHINE(5, 0) diff --git a/hw/core/numa.c b/hw/core/numa.c index 316bc50d75..05be412e59 100644 --- a/hw/core/numa.c +++ b/hw/core/numa.c @@ -117,6 +117,12 @@ static void parse_numa_node(MachineState *ms, NumaNodeOptions *node, } if (node->has_mem) { + if (!mc->numa_mem_supported) { + error_setg(errp, "Parameter -numa node,mem is not supported by this" + " machine type. Use -numa node,memdev instead"); + return; + } + numa_info[nodenr].node_mem = node->mem; if (!qtest_enabled()) { warn_report("Parameter -numa node,mem is deprecated," diff --git a/hw/i386/pc.c b/hw/i386/pc.c index 2128f3d6fe..a86136069c 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -1960,7 +1960,6 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) hc->unplug = pc_machine_device_unplug_cb; mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE; mc->nvdimm_supported = true; - mc->numa_mem_supported = true; mc->default_ram_id = "pc.ram"; object_class_property_add(oc, PC_MACHINE_DEVMEM_REGION_SIZE, "int", diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index f66e1d73ce..8955436efa 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -438,6 +438,7 @@ static void pc_i440fx_5_0_machine_options(MachineClass *m) pc_i440fx_5_1_machine_options(m); m->alias = NULL; m->is_default = false; + m->numa_mem_supported = true; compat_props_add(m->compat_props, hw_compat_5_0, hw_compat_5_0_len); compat_props_add(m->compat_props, pc_compat_5_0, pc_compat_5_0_len); } diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 4ba8ac8774..af6b32adb7 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -366,6 +366,7 @@ static void pc_q35_5_0_machine_options(MachineClass *m) { pc_q35_5_1_machine_options(m); m->alias = NULL; + m->numa_mem_supported = true; compat_props_add(m->compat_props, hw_compat_5_0, hw_compat_5_0_len); compat_props_add(m->compat_props, pc_compat_5_0, pc_compat_5_0_len); } diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c index 3b1a5ed865..210d61d1b3 100644 --- a/hw/ppc/spapr.c +++ b/hw/ppc/spapr.c @@ -4511,7 +4511,6 @@ static void spapr_machine_class_init(ObjectClass *oc, void *data) * in which LMBs are represented and hot-added */ mc->numa_mem_align_shift = 28; - mc->numa_mem_supported = true; mc->auto_enable_numa = true; smc->default_caps.caps[SPAPR_CAP_HTM] = SPAPR_CAP_OFF; @@ -4599,6 +4598,7 @@ static void spapr_machine_5_0_class_options(MachineClass *mc) { spapr_machine_5_1_class_options(mc); compat_props_add(mc->compat_props, hw_compat_5_0, hw_compat_5_0_len); + mc->numa_mem_supported = true; } DEFINE_SPAPR_MACHINE(5_0, "5.0", false); diff --git a/qemu-options.hx b/qemu-options.hx index 93bde2bbc8..196f468786 100644 --- a/qemu-options.hx +++ b/qemu-options.hx @@ -239,10 +239,11 @@ SRST -numa node,nodeid=0 -numa node,nodeid=1 \ -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 - '\ ``mem``\ ' assigns a given RAM amount to a node. '\ ``memdev``\ ' - assigns RAM from a given memory backend device to a node. If - '\ ``mem``\ ' and '\ ``memdev``\ ' are omitted in all nodes, RAM is - split equally between them. + Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported + for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from + a given memory backend device to a node. If '\ ``mem``\ ' and + '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them. + '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive. Furthermore, if one node uses '\ ``memdev``\ ', all of them have to -- 2.18.4

3 3

Disable some unit tests
by Ani Sinha 05 Jun '20

05 Jun '20

Hi : Is there a recommended way to disable some unit tests in Libvirt ? We some broken tests which we want to fix incrementally but while we do that, we want to keep running other non-broken tests. thanks Ani

1 0

[libvirt PATCH] gitlab: fix potfile job for new location
by Daniel P. Berrangé 04 Jun '20

04 Jun '20

Since we're storing the libvirt.pot file in git now, we should pull from the source dir, not the build dir. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed as CI fix .gitlab-ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 149334ed6f..7113a1283c 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -210,7 +210,7 @@ potfile: - $MAKE -C src generated-sources - $MAKE -C po libvirt.pot - cd .. - - mv build/po/libvirt.pot libvirt.pot + - cp po/libvirt.pot libvirt.pot image: quay.io/libvirt/buildenv-libvirt-centos-8:latest artifacts: expose_as: 'Potfile' -- 2.26.2

1 0

[libvirt PATCH] qemu: don't reject interface update when switching to/from bridged network
by Laine Stump 04 Jun '20

04 Jun '20

If virDomainUpdateDeviceFlags() was used to update an <interface>, and the interface type changed from type='network' where the network was an unmanaged bridge (so actualType == bridge) to type='bridge' (i.e. actualType *also* == bridge), the update would fail due to the perceived change in type. In practice it is okay to switch between any interface types that end up using a tap device, since libvirt just needs to attach the device to a new bridge. But in this case we were erroneously rejecting it due to a conditional that was too restrictive. This is what the code was doing: if (old->type != new->type) [allow update] else if ((oldActual == bridge and newActual == network) || (oldActual == network and newActual == bridge)) { [allow update] else [error] In the case described above though, old->type and new->type don't match, but oldActual and newActual are both 'bridge', so we get an error. This patch changes the inner conditional so that any combination of 'network' and 'bridge' for oldActual and newActual, since they both use a tap device connected to a bridge. Signed-off-by: Laine Stump <laine(a)redhat.com> --- src/qemu/qemu_hotplug.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index a5d57702eb..dc3bd8245f 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3770,15 +3770,15 @@ qemuDomainChangeNet(virQEMUDriverPtr driver, * where this can only require a minor (or even no) change, * but in most cases we need to do a full reconnection. * - * If we switch (in either direction) between type='bridge' - * and type='network' (for a traditional managed virtual - * network that uses a host bridge, i.e. forward - * mode='route|nat'), we just need to change the bridge. + * As long as both the new and old types use a tap device + * connected to a host bridge (ie VIR_DOMAIN_NET_TYPE_NETWORK + * or VIR_DOMAIN_NET_TYPE_BRIDGE), we just need to connect to + * the new bridge. */ - if ((oldType == VIR_DOMAIN_NET_TYPE_NETWORK && - newType == VIR_DOMAIN_NET_TYPE_BRIDGE) || - (oldType == VIR_DOMAIN_NET_TYPE_BRIDGE && - newType == VIR_DOMAIN_NET_TYPE_NETWORK)) { + if ((oldType == VIR_DOMAIN_NET_TYPE_NETWORK || + oldType == VIR_DOMAIN_NET_TYPE_BRIDGE) && + (newType == VIR_DOMAIN_NET_TYPE_NETWORK || + newType == VIR_DOMAIN_NET_TYPE_BRIDGE)) { needBridgeChange = true; -- 2.25.4

3 2

[libvirt PATCH] conf: fix alignment in virObjectEvent to be 8 not 4 bytes
by Daniel P. Berrangé 04 Jun '20

04 Jun '20

The previous fix accidentally picked up a debug change that put alignment back at 4, not 8, bytes as it claimed: commit 37ae0426420a2f682cc96ee457a8cae9c0db27d3 Author: Daniel P. Berrangé <berrange(a)redhat.com> Date: Wed Jun 3 11:18:23 2020 +0100 conf: force 8 byte alignment for virObjectEvent Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed under the face-palm build breaker fix rule src/conf/object_event_private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/conf/object_event_private.h b/src/conf/object_event_private.h index 126464a9a5..80af043b9b 100644 --- a/src/conf/object_event_private.h +++ b/src/conf/object_event_private.h @@ -42,7 +42,7 @@ typedef void virConnectObjectEventGenericCallback cb, void *cbopaque); -struct __attribute__((aligned(4))) _virObjectEvent { +struct __attribute__((aligned(8))) _virObjectEvent { virObject parent; int eventID; virObjectMeta meta; -- 2.26.2

1 0

[PATCH] conf: force 8 byte alignment for virObjectEvent
by Daniel P. Berrangé 04 Jun '20

04 Jun '20

We need to be able to cast from virObjectEventPtr to one of its many subclasses. Some of these subclasses have 8 byte alignment on 32-bit platforms, but virObjectEventPtr only has 4 byte alignment. Previously the virObject base class had 8 byte alignment but this dropped to 4 byte when converted to inherit from GObject. This introduces cast alignment warnings on 32-bit: ../../src/conf/domain_event.c: In function 'virDomainEventDispatchDefaultFunc': ../../src/conf/domain_event.c:1656:30: error: cast increases required alignment of target type [-Werror=cast-align] 1656 | rtcChangeEvent = (virDomainEventRTCChangePtr)event; | ^ ../../src/conf/domain_event.c:1785:34: error: cast increases required alignment of target type [-Werror=cast-align] 1785 | balloonChangeEvent = (virDomainEventBalloonChangePtr)event; | ^ ../../src/conf/domain_event.c:1896:35: error: cast increases required alignment of target type [-Werror=cast-align] 1896 | blockThresholdEvent = (virDomainEventBlockThresholdPtr)event; | ^ ../../src/conf/domain_event.c: In function 'virDomainQemuMonitorEventDispatchFunc': ../../src/conf/domain_event.c:1974:24: error: cast increases required alignment of target type [-Werror=cast-align] 1974 | qemuMonitorEvent = (virDomainQemuMonitorEventPtr)event; | ^ ../../src/conf/domain_event.c: In function 'virDomainQemuMonitorEventFilter': ../../src/conf/domain_event.c:2179:20: error: cast increases required alignment of target type [-Werror=cast-align] 2179 | monitorEvent = (virDomainQemuMonitorEventPtr) event; | ^ Forcing 8-byte alignment on virObjectEventPtr removes the alignment increase during casts to subclasses. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Technically a build-breaker, but since we don't have any existing usage of __attribute__((aligned)), I wanted to get a second opinion on this approach. One alternative approach would be to switch one of the current "int" fields in virObjectEvent to "long long". src/conf/object_event_private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/conf/object_event_private.h b/src/conf/object_event_private.h index b31441c53a..126464a9a5 100644 --- a/src/conf/object_event_private.h +++ b/src/conf/object_event_private.h @@ -42,7 +42,7 @@ typedef void virConnectObjectEventGenericCallback cb, void *cbopaque); -struct _virObjectEvent { +struct __attribute__((aligned(4))) _virObjectEvent { virObject parent; int eventID; virObjectMeta meta; -- 2.24.1

5 4

[libvirt PATCH] gitdm: Add missing entries
by Andrea Bolognani 04 Jun '20

04 Jun '20

A few new individuals have contributed to libvirt since the last time the gitdm configuration was updated. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- docs/gitdm/groups/unaffiliated | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/docs/gitdm/groups/unaffiliated b/docs/gitdm/groups/unaffiliated index a58731a683..6e9de7084a 100644 --- a/docs/gitdm/groups/unaffiliated +++ b/docs/gitdm/groups/unaffiliated @@ -14,6 +14,7 @@ gmx.de googlemail.com hotmail.com mail.ru +outlook.com pobox.com protonmail.com riseup.net @@ -39,6 +40,8 @@ bigon(a)bigon.be bugzilla.redhat.simon(a)arlott.org cardoe(a)cardoe.com charles(a)dyfis.net +cky(a)cky.nz +contact(a)puzio.waw.pl d.herrendoerfer(a)herrendoerfer.name dan(a)danny.cz debfx(a)fobos.de @@ -53,6 +56,7 @@ ibaldo(a)adinet.com.uy igor47(a)moomers.org infos(a)nafets.de intrigeri(a)boum.org +jaak(a)ristioja.ee james410(a)cowgill.org.uk james(a)shubin.ca jasper(a)humppa.nl -- 2.25.4

2 1