July 2019 - Devel - libvirt List Archives

[libvirt] [PATCH v2 0/7] PCI Multifunction hotplug/unplug, part 1
by Daniel Henrique Barboza 16 Aug '19

16 Aug '19

v2: - rebased without 3 patches from Michal that are under review in his NVMe patch series This is the first part of the feature discussed at [1]. These patches are mostly cleanup and fixes, thus it is beneficial to have them upstream right away. The whole feature can be checked out at [2]. All patches survives unit testing. The feature was stress tested with hundreds of consecutive hotplug/unplugs of a Broadcom BCM5719 multifunction network card in a guest running in a Power 8 server. Hopefully I'll find a suitable x86 env to stress test the feature there too. [1] https://www.redhat.com/archives/libvir-list/2019-June/msg00703.html [2] https://github.com/danielhb/libvirt/tree/multifunction_latest Daniel Henrique Barboza (1): util/virhostdev: enhance VFIO device is in use detection Shivaprasad G Bhat (6): tests: Fix the iommu group path in mock pci tests: pci: Mock the iommu groups and vfio virpcitest: Change the stub driver to vfio from pci-stub virpcimock: Mock the SRIOV Virtual functions tests: qemu: Add test case for pci-hostdev hotplug tests: Add a baseline test for multifunction pci device use case src/util/virhostdev.c | 74 +++++-- src/util/virprocess.h | 2 +- tests/Makefile.am | 7 + tests/qemuhotplugtest.c | 42 +++- .../qemuhotplug-hostdev-pci.xml | 6 + .../qemuhotplug-base-live+hostdev-pci.xml | 58 +++++ ...uhotplug-pseries-base-live+hostdev-pci.xml | 51 +++++ .../qemuhotplug-pseries-base-live.xml | 43 ++++ .../hostdev-pci-multifunction.args | 35 +++ .../hostdev-pci-multifunction.xml | 59 +++++ .../hostdev-vfio-multidomain.args | 2 +- .../hostdev-vfio-multidomain.xml | 2 +- tests/qemuxml2argvdata/hostdev-vfio.args | 2 +- tests/qemuxml2argvdata/hostdev-vfio.xml | 2 +- tests/qemuxml2argvdata/net-hostdev-fail.xml | 2 +- tests/qemuxml2argvdata/net-hostdev-vfio.args | 2 +- tests/qemuxml2argvdata/net-hostdev-vfio.xml | 2 +- tests/qemuxml2argvtest.c | 3 + .../hostdev-pci-multifunction.xml | 79 +++++++ tests/qemuxml2xmloutdata/hostdev-vfio.xml | 2 +- tests/qemuxml2xmloutdata/net-hostdev-vfio.xml | 2 +- tests/qemuxml2xmltest.c | 1 + tests/virhostdevtest.c | 4 +- tests/virpcimock.c | 201 ++++++++++++++++-- tests/virpcitest.c | 12 +- tests/virpcitestdata/0000-06-12.0.config | Bin 0 -> 256 bytes tests/virpcitestdata/0000-06-12.1.config | Bin 0 -> 256 bytes tests/virpcitestdata/0000-06-12.2.config | Bin 0 -> 256 bytes tests/virprocessmock.c | 28 +++ 29 files changed, 668 insertions(+), 55 deletions(-) create mode 100644 tests/qemuhotplugtestdevices/qemuhotplug-hostdev-pci.xml create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-base-live+hostdev-pci.xml create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-pseries-base-live+hostdev-pci.xml create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-pseries-base-live.xml create mode 100644 tests/qemuxml2argvdata/hostdev-pci-multifunction.args create mode 100644 tests/qemuxml2argvdata/hostdev-pci-multifunction.xml create mode 100644 tests/qemuxml2xmloutdata/hostdev-pci-multifunction.xml create mode 100644 tests/virpcitestdata/0000-06-12.0.config create mode 100644 tests/virpcitestdata/0000-06-12.1.config create mode 100644 tests/virpcitestdata/0000-06-12.2.config create mode 100644 tests/virprocessmock.c -- 2.21.0

2 14

[libvirt] [PATCH] util: storage: Fix parsing of 'exportname' from legacy NBD strings
by Peter Krempa 15 Aug '19

15 Aug '19

If the nbd export name contains a colon, our parser would not parse it properly as we split the string by colons. Modify the code to look up the exportname and copy any trailing characters as the export name is supposed to be at the end of the string. https://bugzilla.redhat.com/show_bug.cgi?id=1733044 Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- src/util/virstoragefile.c | 6 ++++-- tests/virstoragetest.c | 8 ++++++++ 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index a6de6a1e45..ffafeef9d8 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -2928,6 +2928,7 @@ virStorageSourceParseNBDColonString(const char *nbdstr, virStorageSourcePtr src) { VIR_AUTOSTRINGLIST backing = NULL; + const char *exportname; if (!(backing = virStringSplit(nbdstr, ":", 0))) return -1; @@ -2975,8 +2976,9 @@ virStorageSourceParseNBDColonString(const char *nbdstr, return -1; } - if (backing[3] && STRPREFIX(backing[3], "exportname=")) { - if (VIR_STRDUP(src->path, backing[3] + strlen("exportname=")) < 0) + if ((exportname = strstr(nbdstr, "exportname="))) { + exportname += strlen("exportname="); + if (VIR_STRDUP(src->path, exportname) < 0) return -1; } diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c index ef16b3c6e0..f558eea622 100644 --- a/tests/virstoragetest.c +++ b/tests/virstoragetest.c @@ -1262,6 +1262,14 @@ mymain(void) "<source protocol='nbd' name='blah'>\n" " <host name='example.org' port='6000'/>\n" "</source>\n"); + TEST_BACKING_PARSE("nbd:example.org:6000:exportname=:", + "<source protocol='nbd' name=':'>\n" + " <host name='example.org' port='6000'/>\n" + "</source>\n"); + TEST_BACKING_PARSE("nbd:example.org:6000:exportname=:test", + "<source protocol='nbd' name=':test'>\n" + " <host name='example.org' port='6000'/>\n" + "</source>\n"); TEST_BACKING_PARSE("nbd://example.org:1234", "<source protocol='nbd'>\n" " <host name='example.org' port='1234'/>\n" -- 2.21.0

2 2

[libvirt] [PATCH v2]daemon: Fix a crash during virNetlinkEventServiceStopAll
by Liu Haitao 13 Aug '19

13 Aug '19

When reboot the host, a core dump file would be generated. The call traces are: Note.In this case, the main thread is thread 5. (gdb) thread 5 [Switching to thread 5 (LWP 4142)] (gdb) bt 0 0x00007f00a6838273 in futex_wait_cancelable (private=<optimized out>, expected=0, futex_word=0x7f004c0125c0) at /usr/src/debug/glibc/2.24-r0/git/sysdeps/unix/sysv/linux/futex-internal.h:88 1 __pthread_cond_wait_common (abstime=0x0, mutex=0x7f004c012540, cond=0x7f004c012598) at /usr/src/debug/glibc/2.24-r0/git/nptl/pthread_cond_wait.c:502 2 __pthread_cond_wait (cond=0x7f004c012598, mutex=0x7f004c012540) at /usr/src/debug/glibc/2.24-r0/git/nptl/pthread_cond_wait.c:655 3 0x00007f00aa467246 in virCondWait (c=<optimized out>, m=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virthread.c:154 4 0x00007f00aa467eb0 in virThreadPoolFree (pool=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virthreadpool.c:286 5 0x00007f0074349f9d in qemuStateCleanup () at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/qemu/qemu_driver.c:1036 6 0x00007f00aa5e9486 in virStateCleanup () at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/libvirt.c:682 7 0x000055a687ab86a4 in main (argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/remote/remote_daemon.c:1473 (gdb) thread 1 [Switching to thread 1 (LWP 4403)] (gdb) bt 0 __GI___pthread_mutex_lock (mutex=mutex@entry=0x0) at /usr/src/debug/glibc/2.24-r0/git/nptl/pthread_mutex_lock.c:67 1 0x00007f00aa467165 in virMutexLock (m=m@entry=0x0) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virthread.c:89 2 0x00007f00aa43c0f9 in virNetlinkEventServerLock (driver=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virnetlink.c:799 3 virNetlinkEventRemoveClient (watch=watch@entry=0, macaddr=macaddr@entry=0x7f0088014944, protocol=protocol@entry=0) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virnetlink.c:1197 4 0x00007f00aa4341df in virNetDevMacVLanDeleteWithVPortProfile ( ifname=<optimized out>, macaddr=macaddr@entry=0x7f0088014944, linkdev=0x7f0088014920 "eth1", mode=mode@entry=1, virtPortProfile=virtPortProfile@entry=0x0, stateDir=stateDir@entry=0x7f004c12fa90 "/var/run/libvirt/qemu") at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virnetdevmacvlan.c:1112 5 0x00007f0074312251 in qemuProcessStop (driver=driver@entry=0x7f004c0ecef0, vm=vm@entry=0x7f0088000b00, reason=reason@entry=VIR_DOMAIN_SHUTOFF_SHUTDOWN, asyncJob=asyncJob@entry=QEMU_ASYNC_JOB_NONE, flags=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/qemu/qemu_process.c:7291 6 0x00007f007437a5ea in processMonitorEOFEvent (vm=0x7f0088000b00, driver=0x7f004c0ecef0) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/qemu/qemu_driver.c:4756 7 qemuProcessEventHandler (data=0x55a687d6df10, opaque=0x7f004c0ecef0) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/qemu/qemu_driver.c:4859 8 0x00007f00aa467c5b in virThreadPoolWorker ( opaque=opaque@entry=0x55a687d6c110) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virthreadpool.c:163 9 0x00007f00aa466fe8 in virThreadHelper (data=<optimized out>) at /usr/src/debug/libvirt/5.3.0-r0/libvirt-5.3.0/src/util/virthread.c:206 10 0x00007f00a68323f4 in start_thread (arg=0x7f00699df700) at /usr/src/debug/glibc/2.24-r0/git/nptl/pthread_create.c:456 11 0x00007f00a616e10f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:105 1. The execution flow of main thread (Thread 5 LWP 4142): main() -->virNetDaemonRun() -->virNetDaemonClose(dmn) //cleanup -->virNetlinkEventServiceStopAll() -->virStateCleanup() -->qemuStateCleanup() -->virThreadPoolFree() -->__pthread_cond_wait() virNetDaemonRun() -->virEventRunDefaultImpl -->virEventPollRunOnce -->virEventPollDispatchHandles -->qemuMonitorIO -->qemuProcessHandleMonitorEOF -->processEvent->eventType = QEMU_PROCESS_EVENT_MONITOR_EOF -->virThreadPoolSendJob() After typing reboot command on the host, the main thread would send an event message to another thread. Here it would let thread 1 to handle the shutdown of qemu process. But it could not be executed immediately. virNetlinkEventServiceStopAll() --> virNetlinkEventServiceStop() --> server[protocol] = NULL; // set server to null IN virNetlinkEventServiceStopAll(), some variables related to network are freed, like (static virNetlinkEventSrvPrivatePtr server). virStateCleanup() -->qemuStateCleanup() -->virThreadPoolFree() -->__pthread_cond_wait() In virThreadPoolFree() it will wait other thread to end up. 2. The execution flow of thread 5 (LWP 4403): qemuProcessStop() -->virNetDevMacVLanDeleteWithVPortProfile() -->virNetlinkEventRemoveClient() --> srv = server[protocol] Although the main thread had sent the message to thread 1(4403), it could not be run instantly. It means that the virNetlinkEventServiceStopAll() might be executed earlier than virNetlinkEventRemoveClient(). We could get it from the log file. "" 2019-06-12 00:10:09.230+0000: 4142: info : virNetlinkEventServiceStopAll:941 : stopping all netlink event services 2019-06-12 00:10:09.230+0000: 4142: info : virNetlinkEventServiceStop:904 : stopping netlink event service 2019-06-12 00:10:21.165+0000: 4403: debug : virNetlinkEventRemoveClient:1190 : removing client watch=0, mac=0x7f0088014944. " In virNetlinkEventRemoveClient() the variable server is used again, but now it is null that is freed by virNetlinkEventServiceStopAll().So it would case a crash . The virNetlinkEventServiceStopAll() should be executed behind virStateCleanup(), Signed-off-by: Liu Haitao <haitao.liu(a)windriver.com> --- src/remote/remote_daemon.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/remote/remote_daemon.c b/src/remote/remote_daemon.c index c3782971f1..7da20a6644 100644 --- a/src/remote/remote_daemon.c +++ b/src/remote/remote_daemon.c @@ -1464,8 +1464,6 @@ int main(int argc, char **argv) { /* Keep cleanup order in inverse order of startup */ virNetDaemonClose(dmn); - virNetlinkEventServiceStopAll(); - if (driversInitialized) { /* NB: Possible issue with timing window between driversInitialized * setting if virNetlinkEventServerStart fails */ @@ -1473,6 +1471,8 @@ int main(int argc, char **argv) { virStateCleanup(); } + virNetlinkEventServiceStopAll(); + virObjectUnref(adminProgram); virObjectUnref(srvAdm); virObjectUnref(qemuProgram); -- 2.21.0

2 3

[libvirt] [PATCH v2] snapshot: Store both config and live XML in the snapshot domain
by Maxiwell S. Garcia 09 Aug '19

09 Aug '19

The snapshot-create operation of running guests saves the live XML and uses it to replace the active and inactive domain in case of revert. So, the config XML is ignored by the snapshot process. This commit changes it and adds the config XML in the snapshot XML as the <inactiveDomain> entry. In case of offline guest, the behavior remains the same and the config XML is saved in the snapshot XML as <domain> entry. The behavior of older snapshots of running guests, that don't have the new <inactiveDomain>, remains the same too. The revert, in this case, overrides both active and inactive domain with the <domain> entry. So, the <inactiveDomain> in the snapshot XML is not required to snapshot work, but it's useful to preserve the config XML of running guests. Signed-off-by: Maxiwell S. Garcia <maxiwell(a)linux.ibm.com> --- src/conf/moment_conf.c | 1 + src/conf/moment_conf.h | 11 +++++++++ src/conf/snapshot_conf.c | 48 +++++++++++++++++++++++++++++++++++----- src/qemu/qemu_driver.c | 27 +++++++++++++++++----- src/util/virxml.c | 45 +++++++++++++++++++++++++++++++++++++ src/util/virxml.h | 8 +++++++ 6 files changed, 130 insertions(+), 10 deletions(-) diff --git a/src/conf/moment_conf.c b/src/conf/moment_conf.c index fea13f0f97..f54a44b33e 100644 --- a/src/conf/moment_conf.c +++ b/src/conf/moment_conf.c @@ -66,6 +66,7 @@ virDomainMomentDefDispose(void *obj) VIR_FREE(def->description); VIR_FREE(def->parent_name); virDomainDefFree(def->dom); + virDomainDefFree(def->inactiveDom); } /* Provide defaults for creation time and moment name after parsing XML */ diff --git a/src/conf/moment_conf.h b/src/conf/moment_conf.h index 9fdbef2172..70cc47bd70 100644 --- a/src/conf/moment_conf.h +++ b/src/conf/moment_conf.h @@ -36,7 +36,18 @@ struct _virDomainMomentDef { char *parent_name; long long creationTime; /* in seconds */ + /* + * Store the active domain definition in case of online + * guest and the inactive domain definition in case of + * offline guest + */ virDomainDefPtr dom; + + /* + * Store the inactive domain definition in case of online + * guest and leave NULL in case of offline guest + */ + virDomainDefPtr inactiveDom; }; virClassPtr virClassForDomainMomentDef(void); diff --git a/src/conf/snapshot_conf.c b/src/conf/snapshot_conf.c index 324901a560..8aeac9ab20 100644 --- a/src/conf/snapshot_conf.c +++ b/src/conf/snapshot_conf.c @@ -243,6 +243,8 @@ virDomainSnapshotDefParse(xmlXPathContextPtr ctxt, char *memoryFile = NULL; bool offline = !!(flags & VIR_DOMAIN_SNAPSHOT_PARSE_OFFLINE); virSaveCookieCallbacksPtr saveCookie = virDomainXMLOptionGetSaveCookie(xmlopt); + int domainflags = VIR_DOMAIN_DEF_PARSE_INACTIVE | + VIR_DOMAIN_DEF_PARSE_SKIP_VALIDATE; if (!(def = virDomainSnapshotDefNew())) return NULL; @@ -292,8 +294,6 @@ virDomainSnapshotDefParse(xmlXPathContextPtr ctxt, * clients will have to decide between best effort * initialization or outright failure. */ if ((tmp = virXPathString("string(./domain/@type)", ctxt))) { - int domainflags = VIR_DOMAIN_DEF_PARSE_INACTIVE | - VIR_DOMAIN_DEF_PARSE_SKIP_VALIDATE; xmlNodePtr domainNode = virXPathNode("./domain", ctxt); VIR_FREE(tmp); @@ -309,6 +309,20 @@ virDomainSnapshotDefParse(xmlXPathContextPtr ctxt, } else { VIR_WARN("parsing older snapshot that lacks domain"); } + + /* /inactiveDomain entry saves the config XML present in a running + * VM. In case of absent, leave parent.inactiveDom NULL and use + * parent.dom for config and live XML. */ + if (virXPathString("string(./inactiveDomain/@type)", ctxt)) { + xmlNodePtr domainNode = virXPathNode("./inactiveDomain", ctxt); + + if (domainNode) { + def->parent.inactiveDom = virDomainDefParseNode(ctxt->node->doc, domainNode, + caps, xmlopt, NULL, domainflags); + if (!def->parent.inactiveDom) + goto cleanup; + } + } } else if (virDomainXMLOptionRunMomentPostParse(xmlopt, &def->parent) < 0) { goto cleanup; } @@ -845,6 +859,10 @@ virDomainSnapshotDefFormatInternal(virBufferPtr buf, { size_t i; int domainflags = VIR_DOMAIN_DEF_FORMAT_INACTIVE; + virBuffer inactivedom_buf = VIR_BUFFER_INITIALIZER; + xmlXPathContextPtr inactivedom_ctxt = NULL; + char *inactivedom_str = NULL; + int ret = -1; if (flags & VIR_DOMAIN_SNAPSHOT_FORMAT_SECURE) domainflags |= VIR_DOMAIN_DEF_FORMAT_SECURE; @@ -903,6 +921,20 @@ virDomainSnapshotDefFormatInternal(virBufferPtr buf, virBufferAddLit(buf, "</domain>\n"); } + if (def->parent.inactiveDom) { + if (virDomainDefFormatInternal(def->parent.inactiveDom, caps, + domainflags, &inactivedom_buf, xmlopt) < 0) + goto error; + + inactivedom_ctxt = virXPathBuildContext(&inactivedom_buf); + if (!(inactivedom_str = virXPathRenameNode("/domain", "inactiveDomain", + inactivedom_ctxt))) + goto error; + + virBufferAddStr(buf, inactivedom_str); + virBufferAddLit(buf, "\n"); + } + if (virSaveCookieFormatBuf(buf, def->cookie, virDomainXMLOptionGetSaveCookie(xmlopt)) < 0) goto error; @@ -917,11 +949,17 @@ virDomainSnapshotDefFormatInternal(virBufferPtr buf, if (virBufferCheckError(buf) < 0) goto error; - return 0; + ret = 0; error: - virBufferFreeAndReset(buf); - return -1; + VIR_FREE(inactivedom_str); + xmlXPathFreeContext(inactivedom_ctxt); + virBufferFreeAndReset(&inactivedom_buf); + + if (ret < 0) + virBufferFreeAndReset(buf); + + return ret; } diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 482f915b67..9b95e9b766 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -15697,6 +15697,13 @@ qemuDomainSnapshotCreateXML(virDomainPtr domain, VIR_DOMAIN_DEF_PARSE_SKIP_VALIDATE))) goto endjob; + if (vm->newDef) { + def->parent.inactiveDom = virDomainDefCopy(vm->newDef, caps, + driver->xmlopt, NULL, true); + if (!def->parent.inactiveDom) + goto endjob; + } + if (flags & VIR_DOMAIN_SNAPSHOT_CREATE_DISK_ONLY) { align_location = VIR_DOMAIN_SNAPSHOT_LOCATION_EXTERNAL; align_match = false; @@ -16231,6 +16238,7 @@ qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, qemuDomainObjPrivatePtr priv; int rc; virDomainDefPtr config = NULL; + virDomainDefPtr inactiveConfig = NULL; virQEMUDriverConfigPtr cfg = NULL; virCapsPtr caps = NULL; bool was_stopped = false; @@ -16331,17 +16339,22 @@ qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, * in the failure cases where we know there was no change? */ } - /* Prepare to copy the snapshot inactive xml as the config of this - * domain. - * - * XXX Should domain snapshots track live xml rather - * than inactive xml? */ + /* Prepare to copy the snapshot inactive domain as the config XML + * and the snapshot domain as the live XML. In case of inactive domain + * NULL, both config and live XML will be copied from snapshot domain. + */ if (snap->def->dom) { config = virDomainDefCopy(snap->def->dom, caps, driver->xmlopt, NULL, true); if (!config) goto endjob; } + if (snap->def->inactiveDom) { + inactiveConfig = virDomainDefCopy(snap->def->inactiveDom, caps, + driver->xmlopt, NULL, true); + if (!inactiveConfig) + goto endjob; + } cookie = (qemuDomainSaveCookiePtr) snapdef->cookie; @@ -16592,6 +16605,10 @@ qemuDomainRevertToSnapshot(virDomainSnapshotPtr snapshot, goto endjob; } + if (inactiveConfig) { + virDomainDefFree(vm->newDef); + VIR_STEAL_PTR(vm->newDef, inactiveConfig); + } ret = 0; endjob: diff --git a/src/util/virxml.c b/src/util/virxml.c index f55b9a362c..756c0eedbc 100644 --- a/src/util/virxml.c +++ b/src/util/virxml.c @@ -1408,3 +1408,48 @@ virXPathContextNodeRestore(virXPathContextNodeSavePtr save) save->ctxt->node = save->node; } + + +/** + * virXPathBuildContext: convert an parent buffer to an + * XPath context ptr. The caller has to free the ptr. + */ +xmlXPathContextPtr +virXPathBuildContext(virBufferPtr root) +{ + xmlDocPtr doc; + + if (!root) + return NULL; + + doc = virXMLParseString(virBufferCurrentContent(root), NULL); + if (!doc) + return NULL; + + return xmlXPathNewContext(doc); +} + + +/** + * virXPathRenameNode: get the XML node using the 'xpath' and + * rename it with the 'newname' string. + * + * Returns the XML string of the node found by 'xpath' or NULL + * on error. The caller has to free the string. + */ +char * +virXPathRenameNode(const char *xpath, + const char *newname, + xmlXPathContextPtr ctxt) +{ + xmlNodePtr node; + + if (!xpath || !newname || !ctxt) + return NULL; + + if (!(node = virXPathNode(xpath, ctxt))) + return NULL; + + xmlNodeSetName(node, (xmlChar *) newname); + return virXMLNodeToString(node->doc, node); +} diff --git a/src/util/virxml.h b/src/util/virxml.h index 6208977dd1..48a507c3c1 100644 --- a/src/util/virxml.h +++ b/src/util/virxml.h @@ -220,6 +220,14 @@ virXMLFormatElement(virBufferPtr buf, virBufferPtr childBuf) ATTRIBUTE_RETURN_CHECK; +xmlXPathContextPtr +virXPathBuildContext(virBufferPtr root); + +char * +virXPathRenameNode(const char *xpath, + const char *newname, + xmlXPathContextPtr ctxt); + struct _virXPathContextNodeSave { xmlXPathContextPtr ctxt; xmlNodePtr node; -- 2.20.1

4 7

[libvirt] [dockerfiles PATCH v2 0/4] Add libosinfo Dockerfiles
by Fabiano Fidêncio 08 Aug '19

08 Aug '19

This patch series aims to add libosinfo Dockerfiles to this project, so those containers can be used by libosinfo projects' gitlab CI. Please, take a look at each patch for more context of what has been done. Changes since v1: https://www.redhat.com/archives/libvir-list/2019-July/msg01296.html - Update libosinfo Dockerfiles after having libvirt-jenkins-ci's patches related to libosinfo/osinfo-db-tools dependencies merged. Fabiano Fidêncio (4): refresh: Fix typo: ingores -> ignored refresh: Learn how to deal with the project's name refresh: Add libosinfo project Add Dockerfiles for libosinfo project buildenv-libosinfo-centos-7.Dockerfile | 43 +++++++++++++ ...bosinfo-debian-10-cross-aarch64.Dockerfile | 64 +++++++++++++++++++ ...ibosinfo-debian-10-cross-armv6l.Dockerfile | 64 +++++++++++++++++++ ...ibosinfo-debian-10-cross-armv7l.Dockerfile | 64 +++++++++++++++++++ ...-libosinfo-debian-10-cross-i686.Dockerfile | 64 +++++++++++++++++++ ...-libosinfo-debian-10-cross-mips.Dockerfile | 64 +++++++++++++++++++ ...osinfo-debian-10-cross-mips64el.Dockerfile | 64 +++++++++++++++++++ ...ibosinfo-debian-10-cross-mipsel.Dockerfile | 64 +++++++++++++++++++ ...bosinfo-debian-10-cross-ppc64le.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-10-cross-s390x.Dockerfile | 64 +++++++++++++++++++ buildenv-libosinfo-debian-10.Dockerfile | 50 +++++++++++++++ ...ibosinfo-debian-9-cross-aarch64.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-9-cross-armv6l.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-9-cross-armv7l.Dockerfile | 64 +++++++++++++++++++ ...v-libosinfo-debian-9-cross-mips.Dockerfile | 64 +++++++++++++++++++ ...bosinfo-debian-9-cross-mips64el.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-9-cross-mipsel.Dockerfile | 64 +++++++++++++++++++ ...ibosinfo-debian-9-cross-ppc64le.Dockerfile | 64 +++++++++++++++++++ ...-libosinfo-debian-9-cross-s390x.Dockerfile | 64 +++++++++++++++++++ buildenv-libosinfo-debian-9.Dockerfile | 50 +++++++++++++++ ...osinfo-debian-sid-cross-aarch64.Dockerfile | 64 +++++++++++++++++++ ...bosinfo-debian-sid-cross-armv6l.Dockerfile | 64 +++++++++++++++++++ ...bosinfo-debian-sid-cross-armv7l.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-sid-cross-i686.Dockerfile | 64 +++++++++++++++++++ ...libosinfo-debian-sid-cross-mips.Dockerfile | 64 +++++++++++++++++++ ...sinfo-debian-sid-cross-mips64el.Dockerfile | 64 +++++++++++++++++++ ...bosinfo-debian-sid-cross-mipsel.Dockerfile | 64 +++++++++++++++++++ ...osinfo-debian-sid-cross-ppc64le.Dockerfile | 64 +++++++++++++++++++ ...ibosinfo-debian-sid-cross-s390x.Dockerfile | 64 +++++++++++++++++++ buildenv-libosinfo-debian-sid.Dockerfile | 50 +++++++++++++++ buildenv-libosinfo-fedora-29.Dockerfile | 49 ++++++++++++++ buildenv-libosinfo-fedora-30.Dockerfile | 49 ++++++++++++++ buildenv-libosinfo-fedora-rawhide.Dockerfile | 63 ++++++++++++++++++ buildenv-libosinfo-ubuntu-16.Dockerfile | 50 +++++++++++++++ buildenv-libosinfo-ubuntu-18.Dockerfile | 50 +++++++++++++++ ...le => buildenv-libvirt-centos-7.Dockerfile | 0 ...libvirt-debian-10-cross-aarch64.Dockerfile | 0 ...-libvirt-debian-10-cross-armv6l.Dockerfile | 0 ...-libvirt-debian-10-cross-armv7l.Dockerfile | 0 ...nv-libvirt-debian-10-cross-i686.Dockerfile | 0 ...nv-libvirt-debian-10-cross-mips.Dockerfile | 0 ...ibvirt-debian-10-cross-mips64el.Dockerfile | 0 ...-libvirt-debian-10-cross-mipsel.Dockerfile | 0 ...libvirt-debian-10-cross-ppc64le.Dockerfile | 0 ...v-libvirt-debian-10-cross-s390x.Dockerfile | 0 ...e => buildenv-libvirt-debian-10.Dockerfile | 0 ...-libvirt-debian-9-cross-aarch64.Dockerfile | 0 ...v-libvirt-debian-9-cross-armv6l.Dockerfile | 0 ...v-libvirt-debian-9-cross-armv7l.Dockerfile | 0 ...env-libvirt-debian-9-cross-mips.Dockerfile | 0 ...libvirt-debian-9-cross-mips64el.Dockerfile | 0 ...v-libvirt-debian-9-cross-mipsel.Dockerfile | 0 ...-libvirt-debian-9-cross-ppc64le.Dockerfile | 0 ...nv-libvirt-debian-9-cross-s390x.Dockerfile | 0 ...le => buildenv-libvirt-debian-9.Dockerfile | 0 ...ibvirt-debian-sid-cross-aarch64.Dockerfile | 0 ...libvirt-debian-sid-cross-armv6l.Dockerfile | 0 ...libvirt-debian-sid-cross-armv7l.Dockerfile | 0 ...v-libvirt-debian-sid-cross-i686.Dockerfile | 0 ...v-libvirt-debian-sid-cross-mips.Dockerfile | 0 ...bvirt-debian-sid-cross-mips64el.Dockerfile | 0 ...libvirt-debian-sid-cross-mipsel.Dockerfile | 0 ...ibvirt-debian-sid-cross-ppc64le.Dockerfile | 0 ...-libvirt-debian-sid-cross-s390x.Dockerfile | 0 ... => buildenv-libvirt-debian-sid.Dockerfile | 0 ...e => buildenv-libvirt-fedora-29.Dockerfile | 0 ...e => buildenv-libvirt-fedora-30.Dockerfile | 0 ...buildenv-libvirt-fedora-rawhide.Dockerfile | 0 ...e => buildenv-libvirt-ubuntu-16.Dockerfile | 0 ...e => buildenv-libvirt-ubuntu-18.Dockerfile | 0 refresh | 60 +++++++++++++---- 71 files changed, 2166 insertions(+), 12 deletions(-) create mode 100644 buildenv-libosinfo-centos-7.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-aarch64.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-armv6l.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-armv7l.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-i686.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-mips.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-mips64el.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-mipsel.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-ppc64le.Dockerfile create mode 100644 buildenv-libosinfo-debian-10-cross-s390x.Dockerfile create mode 100644 buildenv-libosinfo-debian-10.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-aarch64.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-armv6l.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-armv7l.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-mips.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-mips64el.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-mipsel.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-ppc64le.Dockerfile create mode 100644 buildenv-libosinfo-debian-9-cross-s390x.Dockerfile create mode 100644 buildenv-libosinfo-debian-9.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-aarch64.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-armv6l.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-armv7l.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-i686.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-mips.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-mips64el.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-mipsel.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-ppc64le.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid-cross-s390x.Dockerfile create mode 100644 buildenv-libosinfo-debian-sid.Dockerfile create mode 100644 buildenv-libosinfo-fedora-29.Dockerfile create mode 100644 buildenv-libosinfo-fedora-30.Dockerfile create mode 100644 buildenv-libosinfo-fedora-rawhide.Dockerfile create mode 100644 buildenv-libosinfo-ubuntu-16.Dockerfile create mode 100644 buildenv-libosinfo-ubuntu-18.Dockerfile rename buildenv-centos-7.Dockerfile => buildenv-libvirt-centos-7.Dockerfile (100%) rename buildenv-debian-10-cross-aarch64.Dockerfile => buildenv-libvirt-debian-10-cross-aarch64.Dockerfile (100%) rename buildenv-debian-10-cross-armv6l.Dockerfile => buildenv-libvirt-debian-10-cross-armv6l.Dockerfile (100%) rename buildenv-debian-10-cross-armv7l.Dockerfile => buildenv-libvirt-debian-10-cross-armv7l.Dockerfile (100%) rename buildenv-debian-10-cross-i686.Dockerfile => buildenv-libvirt-debian-10-cross-i686.Dockerfile (100%) rename buildenv-debian-10-cross-mips.Dockerfile => buildenv-libvirt-debian-10-cross-mips.Dockerfile (100%) rename buildenv-debian-10-cross-mips64el.Dockerfile => buildenv-libvirt-debian-10-cross-mips64el.Dockerfile (100%) rename buildenv-debian-10-cross-mipsel.Dockerfile => buildenv-libvirt-debian-10-cross-mipsel.Dockerfile (100%) rename buildenv-debian-10-cross-ppc64le.Dockerfile => buildenv-libvirt-debian-10-cross-ppc64le.Dockerfile (100%) rename buildenv-debian-10-cross-s390x.Dockerfile => buildenv-libvirt-debian-10-cross-s390x.Dockerfile (100%) rename buildenv-debian-10.Dockerfile => buildenv-libvirt-debian-10.Dockerfile (100%) rename buildenv-debian-9-cross-aarch64.Dockerfile => buildenv-libvirt-debian-9-cross-aarch64.Dockerfile (100%) rename buildenv-debian-9-cross-armv6l.Dockerfile => buildenv-libvirt-debian-9-cross-armv6l.Dockerfile (100%) rename buildenv-debian-9-cross-armv7l.Dockerfile => buildenv-libvirt-debian-9-cross-armv7l.Dockerfile (100%) rename buildenv-debian-9-cross-mips.Dockerfile => buildenv-libvirt-debian-9-cross-mips.Dockerfile (100%) rename buildenv-debian-9-cross-mips64el.Dockerfile => buildenv-libvirt-debian-9-cross-mips64el.Dockerfile (100%) rename buildenv-debian-9-cross-mipsel.Dockerfile => buildenv-libvirt-debian-9-cross-mipsel.Dockerfile (100%) rename buildenv-debian-9-cross-ppc64le.Dockerfile => buildenv-libvirt-debian-9-cross-ppc64le.Dockerfile (100%) rename buildenv-debian-9-cross-s390x.Dockerfile => buildenv-libvirt-debian-9-cross-s390x.Dockerfile (100%) rename buildenv-debian-9.Dockerfile => buildenv-libvirt-debian-9.Dockerfile (100%) rename buildenv-debian-sid-cross-aarch64.Dockerfile => buildenv-libvirt-debian-sid-cross-aarch64.Dockerfile (100%) rename buildenv-debian-sid-cross-armv6l.Dockerfile => buildenv-libvirt-debian-sid-cross-armv6l.Dockerfile (100%) rename buildenv-debian-sid-cross-armv7l.Dockerfile => buildenv-libvirt-debian-sid-cross-armv7l.Dockerfile (100%) rename buildenv-debian-sid-cross-i686.Dockerfile => buildenv-libvirt-debian-sid-cross-i686.Dockerfile (100%) rename buildenv-debian-sid-cross-mips.Dockerfile => buildenv-libvirt-debian-sid-cross-mips.Dockerfile (100%) rename buildenv-debian-sid-cross-mips64el.Dockerfile => buildenv-libvirt-debian-sid-cross-mips64el.Dockerfile (100%) rename buildenv-debian-sid-cross-mipsel.Dockerfile => buildenv-libvirt-debian-sid-cross-mipsel.Dockerfile (100%) rename buildenv-debian-sid-cross-ppc64le.Dockerfile => buildenv-libvirt-debian-sid-cross-ppc64le.Dockerfile (100%) rename buildenv-debian-sid-cross-s390x.Dockerfile => buildenv-libvirt-debian-sid-cross-s390x.Dockerfile (100%) rename buildenv-debian-sid.Dockerfile => buildenv-libvirt-debian-sid.Dockerfile (100%) rename buildenv-fedora-29.Dockerfile => buildenv-libvirt-fedora-29.Dockerfile (100%) rename buildenv-fedora-30.Dockerfile => buildenv-libvirt-fedora-30.Dockerfile (100%) rename buildenv-fedora-rawhide.Dockerfile => buildenv-libvirt-fedora-rawhide.Dockerfile (100%) rename buildenv-ubuntu-16.Dockerfile => buildenv-libvirt-ubuntu-16.Dockerfile (100%) rename buildenv-ubuntu-18.Dockerfile => buildenv-libvirt-ubuntu-18.Dockerfile (100%) -- 2.21.0

3 7

[libvirt] [PATCH 0/2] test_driver: implement get/set BlockIoTune APIs
by Ilias Stamatis 08 Aug '19

08 Aug '19

Ilias Stamatis (2): test_driver: implement virDomainGetBlockIoTune test_driver: implement virDomainSetBlockIoTune src/test/test_driver.c | 286 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 286 insertions(+) -- 2.22.0

2 9

[libvirt] [PATCH] build: Solve mingw build clash with DATADIR
by Eric Blake 08 Aug '19

08 Aug '19

Commit fed58d83 was a hack to fix a mingw build failure due to header inclusion order resulting in a clash over the use of DATADIR, by repeating a trick made several other times in the past of tweaking inclusion order until it goes away. Better is to revert that, and instead use pragmas to avoid the clash in the first place, regardless of header ordering, solving it for everyone in the future. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- I tested that both gcc and clang on F29 support this; but it will take a full CI run to see if everywhere else is okay with it. Thus, it is not 5.6 material. src/util/viratomic.h | 3 +++ src/conf/checkpoint_conf.c | 2 -- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/util/viratomic.h b/src/util/viratomic.h index 35800dafcd..c6e7668324 100644 --- a/src/util/viratomic.h +++ b/src/util/viratomic.h @@ -218,7 +218,10 @@ VIR_STATIC unsigned int virAtomicIntXor(volatile unsigned int *atomic, # ifdef VIR_ATOMIC_OPS_WIN32 +# pragma push_macro("DATADIR") /* If "configmake.h" was included first */ +# undef DATADIR # include <winsock2.h> +# pragma pop_macro("DATADIR") # include <windows.h> # include <intrin.h> # if !defined(_M_AMD64) && !defined (_M_IA64) && !defined(_M_X64) diff --git a/src/conf/checkpoint_conf.c b/src/conf/checkpoint_conf.c index 5ce4cc4853..5f4c275dd8 100644 --- a/src/conf/checkpoint_conf.c +++ b/src/conf/checkpoint_conf.c @@ -21,8 +21,6 @@ #include <config.h> -#include <unistd.h> - #include "configmake.h" #include "internal.h" #include "virbitmap.h" -- 2.20.1

3 4

[libvirt] [PATCH 0/7] qemu: use domCaps for validation
by Cole Robinson 06 Aug '19

06 Aug '19

I'm trying to remove some hurdles and pitfalls WRT extending domaincapabilities data. One issue is that it's too easy to add invalid data to it, or let the data become out of date. For example the first two patches of this series add <rng model=X> domcaps reporting. The logic to fill in the domcaps data from qemuCaps is nearly identical to the logic we use to validate rng->model in qemuDomainRNGDefValidate. If just those patches are added, and later a new qemu rng model was introduced, a future patch could easily miss updated domaincapabilities output. This series aims to set up a pattern to prevent these types of issues from sneaking in. A function virDomainCapsDeviceDefValidate is added which will use domcaps data to perform validation against a devicedef. The existing qemu <rng> model validation is moved there. This ensures that any future <rng> model additions, if they want to work in the qemu driver, effectively need to extend domaincapabilities as well. It's also theoretically useful for other drivers too. One issue is that at DomainDefValidate time we don't have domCaps handy, or any cache layer for domCaps assembling. Patch #4 adds a domCapsCache hashtable to the virQEMUCaps class for caching domCaps builds based on the full tuple of emulator+machine+arch+virttype. If qemuCaps need to be regenerated, the domCaps cache is wiped out for us so we don't need to worry about the data being stale, it's tied to the lifetime of a qemuCaps instance. Cole Robinson (7): conf: domcaps: Report device <rng> qemu: capabilities: fill in domcaps <rng> qemu: conf: add virQEMUDriverGetDomainCapabilities qemu: conf: Cache domCaps in qemuCaps conf: domcaps: Add virDomainCapsDeviceDefValidate qemu: domain: Call virDomainCapsDeviceDefValidate qemu: Move rng model validation to domcaps docs/formatdomaincaps.html.in | 35 ++++++++ docs/schemas/domaincaps.rng | 10 +++ src/conf/domain_capabilities.c | 83 ++++++++++++++++++ src/conf/domain_capabilities.h | 14 ++++ src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 41 +++++++++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_conf.c | 84 +++++++++++++++++++ src/qemu/qemu_conf.h | 7 ++ src/qemu/qemu_domain.c | 38 +++------ src/qemu/qemu_driver.c | 18 +--- .../qemu_1.7.0.x86_64.xml | 9 ++ .../qemu_2.12.0-virt.aarch64.xml | 11 +++ .../qemu_2.12.0.ppc64.xml | 11 +++ .../qemu_2.12.0.s390x.xml | 11 +++ .../qemu_2.12.0.x86_64.xml | 11 +++ .../qemu_2.6.0-virt.aarch64.xml | 11 +++ .../qemu_2.6.0.aarch64.xml | 11 +++ .../domaincapsschemadata/qemu_2.6.0.ppc64.xml | 11 +++ .../qemu_2.6.0.x86_64.xml | 11 +++ .../domaincapsschemadata/qemu_2.7.0.s390x.xml | 11 +++ .../qemu_2.8.0-tcg.x86_64.xml | 11 +++ .../domaincapsschemadata/qemu_2.8.0.s390x.xml | 11 +++ .../qemu_2.8.0.x86_64.xml | 11 +++ .../qemu_2.9.0-q35.x86_64.xml | 11 +++ .../qemu_2.9.0-tcg.x86_64.xml | 11 +++ .../qemu_2.9.0.x86_64.xml | 11 +++ .../domaincapsschemadata/qemu_3.0.0.s390x.xml | 11 +++ .../qemu_4.0.0.x86_64.xml | 11 +++ 29 files changed, 488 insertions(+), 40 deletions(-) -- 2.21.0

3 16

[libvirt] [PATCH] daemon: improve Xen support in systemd service
by Jim Fehlig 05 Aug '19

05 Aug '19

The xencommons service provides all the essential services such as xenstored, xenconsoled, etc. needed by the libvirt Xen driver, so libvirtd should be started after xencommons. The xendomains service uses Xen's xl tool to operate on any domains it finds running, even those managed by libvirt. Add a conflicts on the xendomains service to ensure it is not enabled when libvirtd is enabled. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- src/remote/libvirtd.service.in | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/remote/libvirtd.service.in b/src/remote/libvirtd.service.in index 3ddf0e229b..5dbe7cbe72 100644 --- a/src/remote/libvirtd.service.in +++ b/src/remote/libvirtd.service.in @@ -15,6 +15,8 @@ After=local-fs.target After=remote-fs.target After=systemd-logind.service After=systemd-machined.service +After=xencommons.service +Conflicts=xendomains.service Documentation=man:libvirtd(8) Documentation=https://libvirt.org -- 2.22.0

2 2

[libvirt] [PATCH 0/4] test_driver: implement FS-related APIs
by Ilias Stamatis 04 Aug '19

04 Aug '19

Ilias Stamatis (4): test_driver: introduce domain-private data test_driver: implement virDomainFSFreeze test_driver: implement virDomainFSThaw test_driver: implement virDomainFSTrim src/test/test_driver.c | 191 ++++++++++++++++++++++++++++++++++++++++- 1 file changed, 190 insertions(+), 1 deletion(-) -- 2.22.0

2 10