- Devel - libvirt List Archives

[libvirt] [PATCH v1] qemu: Convert to virErrorPreserveLast/virErrorRestore
by Syed Humaid 20 May '19

20 May '19

Replaced usage of virSaveLastError and virSetError/virFreeError with virErrorPreserveLast and virErrorRestore respectively. Signed-off-by: Syed Humaid <syedhumaidbinharoon(a)gmail.com> --- src/qemu/qemu_cgroup.c | 7 ++----- src/qemu/qemu_command.c | 2 +- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_driver.c | 28 ++++++++-------------------- 4 files changed, 12 insertions(+), 27 deletions(-) diff --git a/src/qemu/qemu_cgroup.c b/src/qemu/qemu_cgroup.c index ca76c4fdfa..c8d9ae4808 100644 --- a/src/qemu/qemu_cgroup.c +++ b/src/qemu/qemu_cgroup.c @@ -1123,10 +1123,7 @@ qemuSetupCgroupVcpuBW(virCgroupPtr cgroup, if (period) { virErrorPtr saved = virSaveLastError(); ignore_value(virCgroupSetCpuCfsPeriod(cgroup, old_period)); - if (saved) { - virSetError(saved); - virFreeError(saved); - } + virErrorRestore(&saved); } return -1; @@ -1334,7 +1331,7 @@ qemuCgroupEmulatorAllNodesRestore(qemuCgroupEmulatorAllNodesDataPtr data) if (!data) return; - err = virSaveLastError(); + virErrorPreserveLast(&err); virCgroupSetCpusetMems(data->emulatorCgroup, data->emulatorMemMask); virSetError(err); virFreeError(err); diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 1425d97b1e..9772e4f9af 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -9119,7 +9119,7 @@ qemuBuildNetCommandLine(virQEMUDriverPtr driver, error: /* free up any resources in the network driver * but don't overwrite the original error */ - originalError = virSaveLastError(); + virErrorPreserveLast(&originalError); for (i = 0; last_good_net != -1 && i <= last_good_net; i++) virDomainConfNWFilterTeardown(def->nets[i]); virSetError(originalError); diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 960aaff3c7..dff1d7b1a5 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -8042,7 +8042,7 @@ void qemuDomainObjTaint(virQEMUDriverPtr driver, /* We don't care about errors logging taint info, so * preserve original error, and clear any error that * is raised */ - orig_err = virSaveLastError(); + virErrorPreserveLast(&orig_err); if (!(timestamp = virTimeStringNow())) goto cleanup; diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f48d9256e4..51b16f5aa1 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6844,7 +6844,7 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, * must manually kill it and ignore any error related to * the process */ - orig_err = virSaveLastError(); + virErrorPreserveLast(&orig_err); VIR_FORCE_CLOSE(intermediatefd); VIR_FORCE_CLOSE(*fd); } @@ -6855,10 +6855,7 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, } VIR_DEBUG("Decompression binary stderr: %s", NULLSTR(errbuf)); - if (orig_err) { - virSetError(orig_err); - virFreeError(orig_err); - } + virErrorRestore(&orig_err); } VIR_FORCE_CLOSE(intermediatefd); @@ -14398,7 +14395,7 @@ qemuDomainSnapshotFSThaw(virQEMUDriverPtr driver ATTRIBUTE_UNUSED, agent = qemuDomainObjEnterAgent(vm); if (!report) - err = virSaveLastError(); + virErrorPreserveLast(&err); thawed = qemuAgentFSThaw(agent); if (!report) virSetError(err); @@ -15308,7 +15305,7 @@ qemuDomainSnapshotCreateDiskActive(virQEMUDriverPtr driver, error: if (ret < 0) { - orig_err = virSaveLastError(); + virErrorPreserveLast(&orig_err); for (i = 0; i < snapdef->ndisks; i++) { if (!diskdata[i].src) continue; @@ -15351,10 +15348,7 @@ qemuDomainSnapshotCreateDiskActive(virQEMUDriverPtr driver, virJSONValueFree(actions); virObjectUnref(cfg); - if (orig_err) { - virSetError(orig_err); - virFreeError(orig_err); - } + virErrorRestore(&orig_err); return ret; } @@ -17773,7 +17767,7 @@ qemuDomainBlockCopyCommon(virDomainObjPtr vm, if (qemuDomainObjExitMonitor(driver, vm) < 0) ret = -1; if (ret < 0) { - monitor_error = virSaveLastError(); + virErrorPreserveLast(&monitor_error); qemuDomainDiskChainElementRevoke(driver, vm, mirror); goto endjob; } @@ -17795,10 +17789,7 @@ qemuDomainBlockCopyCommon(virDomainObjPtr vm, VIR_WARN("%s", _("unable to remove just-created copy target")); virStorageFileDeinit(mirror); qemuDomainObjEndJob(driver, vm); - if (monitor_error) { - virSetError(monitor_error); - virFreeError(monitor_error); - } + virErrorRestore(&monitor_error); qemuBlockJobStartupFinalize(job); cleanup: @@ -18196,10 +18187,7 @@ qemuDomainBlockCommit(virDomainPtr dom, if (top_parent && top_parent != disk->src) qemuDomainDiskChainElementPrepare(driver, vm, top_parent, true, false); - if (orig_err) { - virSetError(orig_err); - virFreeError(orig_err); - } + virErrorRestore(&orig_err); } qemuBlockJobStartupFinalize(job); qemuDomainObjEndJob(driver, vm); -- 2.20.1

2 1

[libvirt] [PATCH] qemuProcessReconnect: ensure vm xml integrity when save status
by Wang King 20 May '19

20 May '19

From: Dingzhong Weng <wengdingzhong(a)huawei.com> libvirtd starts qemuProcessReconnect threads when libvirtd is starting and there are live vm on the host. but as qemuProcessReconnect threads are simply side job done by libvirtd upon start and are only called for existing vm. these threads are not managed like that of worker threads and event pool. once libvirtd receives SIGTERM signal or any similar command right after it starts, these qemuProcessReconnect threads might still be running and libvirtd goes to clean up and frees qemu_driver and its components. In this short window, qemuProcessReconnect threads might have a NULL qemu_driver->xmlopt.format, skip this function, and only partial vm status can be saved to disk. As a result, vm may fail to recover with missing information. this patch increases qemu_driver->xmlopt ref count during the lifecycle of qemuProcessReconnect so that libvirtd will not release this resource until qemuProcessReconnect threads finish using it. thus make sure all of vm status information can be formatted and thus maintain its integrity in virDomainSaveStatus. no vm is killed in this patch. Signed-off-by: Dingzhong Weng <wengdingzhong(a)huawei.com> --- src/qemu/qemu_process.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 90466771cd..36b9b5fd03 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -8000,6 +8000,7 @@ qemuProcessReconnect(void *opaque) struct qemuProcessReconnectData *data = opaque; virQEMUDriverPtr driver = data->driver; virDomainObjPtr obj = data->obj; + virDomainXMLOptionPtr xmlopt = NULL; qemuDomainObjPrivatePtr priv; qemuDomainJobObj oldjob; int state; @@ -8023,6 +8024,9 @@ qemuProcessReconnect(void *opaque) cfg = virQEMUDriverGetConfig(driver); priv = obj->privateData; + /* need xmlopt later to save status, do not free */ + xmlopt = virObjectRef(driver->xmlopt); + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto error; @@ -8218,7 +8222,7 @@ qemuProcessReconnect(void *opaque) } /* update domain state XML with possibly updated state in virDomainObj */ - if (virDomainSaveStatus(driver->xmlopt, cfg->stateDir, obj, driver->caps) < 0) + if (virDomainSaveStatus(xmlopt, cfg->stateDir, obj, driver->caps) < 0) goto error; /* Run an hook to allow admins to do some magic */ @@ -8253,6 +8257,7 @@ qemuProcessReconnect(void *opaque) virDomainObjEndAPI(&obj); virObjectUnref(cfg); virObjectUnref(caps); + virObjectUnref(xmlopt); virNWFilterUnlockFilterUpdates(); virIdentitySetCurrent(NULL); return; -- 2.18.0.windows.1

2 2

[libvirt] [PATCH v2] qemu: Add entry for balloon stats stat-htlb-pgalloc and stat-htlb-pgfail
by Han Han 20 May '19

20 May '19

Qemu added reporting of virtio balloon new statistics stat-htlb-pgalloc and stat-htlb-pgfail since qemu-3.0 commit b7b12644297. The value of stat-htlb-pgalloc represents the number of successful hugetlb page allocations while stat-htlb-pgfail represents the number of failed ones. Add this statistics reporting to libvirt. To enable this feature for vm, guest kenel >= 4.17 is required because the exporting hugetlb page allocation for virtio balloon is introduced since 6c64fe7f. Signed-off-by: Han Han <hhan(a)redhat.com> --- include/libvirt/libvirt-domain.h | 14 +++++++++++++- src/libvirt-domain.c | 5 ++++- src/qemu/qemu_driver.c | 2 ++ src/qemu/qemu_monitor_json.c | 5 +++++ tools/virsh-domain-monitor.c | 4 ++++ tools/virsh.pod | 2 ++ 6 files changed, 30 insertions(+), 2 deletions(-) diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h index 7d36820b5a..192d25c1db 100644 --- a/include/libvirt/libvirt-domain.h +++ b/include/libvirt/libvirt-domain.h @@ -636,11 +636,23 @@ typedef enum { */ VIR_DOMAIN_MEMORY_STAT_DISK_CACHES = 10, + /* + * The amount of successful hugetlb(Huge Page Tables) allocations via + * virtio balloon. + */ + VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGALLOC = 11, + + /* + * The amount of failed hugetlb(Huge Page Tables) allocations via + * virtio balloon. + */ + VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGFAIL = 12, + /* * The number of statistics supported by this version of the interface. * To add new statistics, add them to the enum and increase this value. */ - VIR_DOMAIN_MEMORY_STAT_NR = 11, + VIR_DOMAIN_MEMORY_STAT_NR = 13, # ifdef VIR_ENUM_SENTINELS VIR_DOMAIN_MEMORY_STAT_LAST = VIR_DOMAIN_MEMORY_STAT_NR diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index 80284a99f0..a95690e8a4 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -5734,7 +5734,10 @@ virDomainGetInterfaceParameters(virDomainPtr domain, * VIR_DOMAIN_MEMORY_STAT_DISK_CACHES * Memory that can be reclaimed without additional I/O, typically disk * caches (in kb). - * + * VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGALLOC + * The amount of successful hugetlb(Huge Page Tables) allocations + * VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGFAIL + * The amount of failed hugetlb(Huge Page Tables) allocations * Returns: The number of stats provided or -1 in case of failure. */ int diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f48d9256e4..2d849beac8 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -20245,6 +20245,8 @@ qemuDomainGetStatsBalloon(virQEMUDriverPtr driver, STORE_MEM_RECORD(LAST_UPDATE, "last-update") STORE_MEM_RECORD(USABLE, "usable") STORE_MEM_RECORD(DISK_CACHES, "disk_caches") + STORE_MEM_RECORD(HUGETLB_PGALLOC, "hugetlb_pgalloc") + STORE_MEM_RECORD(HUGETLB_PGFAIL, "hugetlb_pgfail") } #undef STORE_MEM_RECORD diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 908967f46c..5e3df5e759 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -2150,6 +2150,11 @@ int qemuMonitorJSONGetMemoryStats(qemuMonitorPtr mon, VIR_DOMAIN_MEMORY_STAT_LAST_UPDATE, 1); GET_BALLOON_STATS(statsdata, "stat-disk-caches", VIR_DOMAIN_MEMORY_STAT_DISK_CACHES, 1024); + GET_BALLOON_STATS(statsdata, "stat-htlb-pgalloc", + VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGALLOC, 1); + GET_BALLOON_STATS(statsdata, "stat-htlb-pgfail", + VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGFAIL, 1); + ret = got; cleanup: virJSONValueFree(cmd); diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index d87475f6f6..0092a9786e 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -376,6 +376,10 @@ cmdDomMemStat(vshControl *ctl, const vshCmd *cmd) vshPrint(ctl, "last_update %llu\n", stats[i].val); if (stats[i].tag == VIR_DOMAIN_MEMORY_STAT_DISK_CACHES) vshPrint(ctl, "disk_caches %llu\n", stats[i].val); + if (stats[i].tag == VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGALLOC) + vshPrint(ctl, "hugetlb_pgalloc %llu\n", stats[i].val); + if (stats[i].tag == VIR_DOMAIN_MEMORY_STAT_HUGETLB_PGFAIL) + vshPrint(ctl, "hugetlb_pgfail %llu\n", stats[i].val); } ret = true; diff --git a/tools/virsh.pod b/tools/virsh.pod index afc1684db0..ef27c527d6 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -938,6 +938,8 @@ without causing host swapping (in KiB) last-update - Timestamp of the last update of statistics (in seconds) disk_caches - The amount of memory that can be reclaimed without additional I/O, typically disk caches (in KiB) + hugetlb_pgalloc - The number of successful huge page table allocations + hugetlb_pgfail - The number of failed huge page table allocations For QEMU/KVM with a memory balloon, setting the optional I<--period> to a value larger than 0 in seconds will allow the balloon driver to return -- 2.20.1

2 3

[libvirt] [glib PATCH 0/8] Add machinery forsetting firmware attribute to <os/>
by Fabiano Fidêncio 20 May '19

20 May '19

This series consists in two different things that, together, are needed to set <os firmware="efi|bios"/> as part of libvirt domain's XML. The first patch simply changes the XML adding the bit mentioned above. All the other patches, on the other hand, add enough (and nothing more than that) infra for consumers to check which are the supported firmwares (the ones listed by virConnectGetDomainCapabilities()). Fabiano Fidêncio (8): gconfig: Add _domain_os_set_firmware() gobject: Use the proper return type in _connection_get_capabilities_finish() gconfig: Add GVirConfigDomainCapabilities gconfig: Add GVirConfigDomainCapabilitiesOs gconfig: Add _domain_capabilities_get_os() gconfig: Add _domain_capabilities_os_get_firmwares() tests: Add basic test for domain capabilities gobject: Wrap virConnectGetDomainCapabilities() libvirt-gconfig/Makefile.am | 4 + .../libvirt-gconfig-domain-capabilities-os.c | 115 ++++++++++++ .../libvirt-gconfig-domain-capabilities-os.h | 66 +++++++ .../libvirt-gconfig-domain-capabilities.c | 94 ++++++++++ .../libvirt-gconfig-domain-capabilities.h | 68 +++++++ libvirt-gconfig/libvirt-gconfig-domain-os.c | 17 ++ libvirt-gconfig/libvirt-gconfig-domain-os.h | 7 + libvirt-gconfig/libvirt-gconfig.h | 2 + libvirt-gconfig/libvirt-gconfig.sym | 8 + libvirt-gobject/libvirt-gobject-connection.c | 170 +++++++++++++++++- libvirt-gobject/libvirt-gobject-connection.h | 22 +++ libvirt-gobject/libvirt-gobject.sym | 8 + tests/test-gconfig.c | 34 ++++ tests/xml/gconfig-domain-capabilities-os.xml | 23 +++ tests/xml/gconfig-domain-os.xml | 2 +- 15 files changed, 638 insertions(+), 2 deletions(-) create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.h create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-capabilities.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-capabilities.h create mode 100644 tests/xml/gconfig-domain-capabilities-os.xml -- 2.21.0

2 12

[libvirt] [PATCH 0/7] x86: Multi-key Total Memory Encryption (Intel)
by Larkins Carvalho 17 May '19

17 May '19

This patch series provides support for launching an encrypted guest using Intel's Multi-key Total Memory Encryption (MKTME) feature. The purpose of this review is to get initial feedback and to get on par with libvirt development process. Considering we have not added tests, this is a preliminary patch and based on the community feedback, we expect more updates to follow. Total Memory Encryption (TME) provides the capability to encrypt the entirety of the physical memory of a system. MKTME builds on TME and adds support for multiple encryption keys. High Level flow: 1. Management tool calls virConnectGetDomainCapabilities which calls virNodeGetMKTMEInfo. This returns an XML document that includes the following: <feature> ... <mktme supported='yes'> <keys_supported> </keys_supported> </mktme> </feature> 2. If MKTME is supported management tool requests to start a guest calling virCreateXML(). The xml would include: <launchSecurity type='mktme'> <id>m0</id> <key_type>user</key_type> <key>samplekey</key> <encryption_algorithm>aes-xts-128</encryption_algorithm> </launchSecurity> 3. Libvirt makes system call with the provided information to generate a key handle using linux keyring services. Qemu uses the key handle to launch the workload. 4. Libvirt generate the QEMU cli arg to enable the MKTME feature, a typical args looks like this: # Qemu ...\ -machine pc,memory-encryption=m0 -object mktme-guest,id=m0,handle=${serial} Intel MKTME spec: https://software.intel.com/sites/default/files/managed/a5/16/Multi-Key-Tota… TODO: Add tests for launch security of type mktme. WIP: Qemu and KVM patch to support Intel MKTME are in the process of upstreaming. The complete git tree is available at: https://github.com/larkinscarvalho/libvirt/tree/intel-mktme-patch-series Larkins Carvalho (7): qemu: provide support to query the MKTME capability conf: Add MKTME feature in domain capabilities libvirt: add new public API to get MKTME Info remote: implement the remote protocol for virNodeMKTMEInfo() qemu: Add support to get the MKTME info conf: introduce launchSecurity element type mktme in domain qemu: add support to launch MKTME guest docs/formatdomain.html.in | 64 +- docs/formatdomaincaps.html.in | 20 + docs/schemas/domaincaps.rng | 14 + docs/schemas/domaincommon.rng | 87 +- include/libvirt/libvirt-host.h | 18 + src/conf/domain_capabilities.c | 30 + src/conf/domain_capabilities.h | 12 + src/conf/domain_conf.c | 112 +- src/conf/domain_conf.h | 14 + src/conf/virconftypes.h | 3 + src/driver-hypervisor.h | 7 + src/libvirt-host.c | 49 + src/libvirt_private.syms | 4 + src/libvirt_public.syms | 5 + src/qemu/qemu_capabilities.c | 140 +- src/qemu/qemu_capabilities.h | 4 + src/qemu/qemu_capspriv.h | 4 + src/qemu/qemu_command.c | 40 + src/qemu/qemu_driver.c | 63 + src/qemu/qemu_monitor.c | 10 + src/qemu/qemu_monitor.h | 3 + src/qemu/qemu_monitor_json.c | 61 + src/qemu/qemu_monitor_json.h | 3 + src/remote/remote_daemon_dispatch.c | 44 + src/remote/remote_driver.c | 41 +- src/remote/remote_protocol.x | 21 +- src/remote_protocol-structs | 12 + src/util/Makefile.inc.am | 2 + src/util/virmktme.c | 127 ++ src/util/virmktme.h | 34 + .../bhyve_basic.x86_64.xml | 1 + .../bhyve_fbuf.x86_64.xml | 1 + .../bhyve_uefi.x86_64.xml | 1 + tests/domaincapsschemadata/empty.xml | 1 + tests/domaincapsschemadata/libxl-xenfv.xml | 1 + tests/domaincapsschemadata/libxl-xenpv.xml | 1 + .../qemu_1.7.0.x86_64.xml | 1 + .../qemu_2.12.0-virt.aarch64.xml | 1 + .../qemu_2.12.0.ppc64.xml | 1 + .../qemu_2.12.0.s390x.xml | 1 + .../qemu_2.12.0.x86_64.xml | 1 + .../qemu_2.6.0-virt.aarch64.xml | 1 + .../qemu_2.6.0.aarch64.xml | 1 + .../domaincapsschemadata/qemu_2.6.0.ppc64.xml | 1 + .../qemu_2.6.0.x86_64.xml | 1 + .../domaincapsschemadata/qemu_2.7.0.s390x.xml | 1 + .../qemu_2.8.0-tcg.x86_64.xml | 1 + .../domaincapsschemadata/qemu_2.8.0.s390x.xml | 1 + .../qemu_2.8.0.x86_64.xml | 1 + .../qemu_2.9.0-q35.x86_64.xml | 1 + .../qemu_2.9.0-tcg.x86_64.xml | 1 + .../qemu_2.9.0.x86_64.xml | 1 + .../domaincapsschemadata/qemu_3.0.0.s390x.xml | 1 + .../qemu_3.1.0.x86_64.xml | 1 + .../domaincapsschemadata/qemu_4.0.0.s390x.xml | 1 + .../qemu_4.0.0.x86_64.xml | 1 + .../qemu_5.3.0.x86_64.xml | 164 ++ tests/domaincapstest.c | 5 + .../caps_5.3.0.x86_64.xml | 1377 +++++++++++++++++ 59 files changed, 2579 insertions(+), 41 deletions(-) create mode 100644 src/util/virmktme.c create mode 100644 src/util/virmktme.h create mode 100644 tests/domaincapsschemadata/qemu_5.3.0.x86_64.xml create mode 100644 tests/qemucapabilitiesdata/caps_5.3.0.x86_64.xml

1 7

[libvirt] [PATCH] rpm: remove dependancy from qemu to network/storage drivers
by Daniel P. Berrangé 17 May '19

17 May '19

The libvirt-daemon-driver-qemu RPM has historically had a hard dependency on the libvirt-daemon-driver-network and libvirt-daemon-driver-storage-core packages. This was because the QEMU driver would directly call into APIs that were part of these drivers. The dependency to the storage driver was eliminated in commit 064fec69be4a4673e0df17b007bf781026c3b4b2 Author: Daniel P. Berrangé <berrange(a)redhat.com> Date: Thu Jan 25 09:35:46 2018 +0000 storage: move storage file backend framework into util directory The dependency to the network driver was eliminated in commit 5b13570ab8b43ec3c590399ec5a7644d91082149 Author: Daniel P. Berrangé <berrange(a)redhat.com> Date: Thu Jan 25 09:35:47 2018 +0000 conf: introduce callback registration for domain net device allocation Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Previously sent as patch of the the network refactor but it can in fact be merged now as it doesn't depend on it directly. libvirt.spec.in | 3 --- 1 file changed, 3 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 5c27a44fc5..b159c50d8f 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -720,9 +720,6 @@ parted and more. Summary: QEMU driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} Requires: libvirt-libs = %{version}-%{release} -# There really is a hard cross-driver dependency here -Requires: libvirt-daemon-driver-network = %{version}-%{release} -Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} Requires: /usr/bin/qemu-img # For image compression Requires: gzip -- 2.21.0

2 1

[libvirt] [PATCH] lib: Avoid double free when passing FDs with virCommandPassFD()
by Michal Privoznik 17 May '19

17 May '19

If an FD is passed into a child using: virCommandPassFD(cmd, fd, VIR_COMMAND_PASS_FD_CLOSE_PARENT); then the parent should refrain from touching @fd thereafter. This is even documented in virCommandPassFD() comment. The reason is that either at virCommandRun()/virCommandRunAsync() or virCommandFree() time the @fd will be closed. Closing it earlier, e.g. right after virCommandPassFD() call might result in undesired results. Another thread might open a file and receive the same FD which is then unexpectedly closed by virCommandFree() or virCommandRun(). Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_command.c | 10 ++++++---- src/util/virpolkit.c | 1 + tests/commandtest.c | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index bf1fb539b1..92bd1524db 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -8978,17 +8978,19 @@ qemuBuildInterfaceCommandLine(virQEMUDriverPtr driver, if (qemuSecuritySetTapFDLabel(driver->securityManager, def, tapfd[i]) < 0) goto cleanup; - virCommandPassFD(cmd, tapfd[i], - VIR_COMMAND_PASS_FD_CLOSE_PARENT); if (virAsprintf(&tapfdName[i], "%d", tapfd[i]) < 0) goto cleanup; + virCommandPassFD(cmd, tapfd[i], + VIR_COMMAND_PASS_FD_CLOSE_PARENT); + tapfd[i] = -1; } for (i = 0; i < vhostfdSize; i++) { - virCommandPassFD(cmd, vhostfd[i], - VIR_COMMAND_PASS_FD_CLOSE_PARENT); if (virAsprintf(&vhostfdName[i], "%d", vhostfd[i]) < 0) goto cleanup; + virCommandPassFD(cmd, vhostfd[i], + VIR_COMMAND_PASS_FD_CLOSE_PARENT); + vhostfd[i] = -1; } if (chardev) diff --git a/src/util/virpolkit.c b/src/util/virpolkit.c index 25eaad2c63..634b46e82b 100644 --- a/src/util/virpolkit.c +++ b/src/util/virpolkit.c @@ -187,6 +187,7 @@ virPolkitAgentCreate(void) virCommandSetOutputFD(agent->cmd, &outfd); virCommandSetErrorFD(agent->cmd, &errfd); virCommandPassFD(agent->cmd, pipe_fd[1], VIR_COMMAND_PASS_FD_CLOSE_PARENT); + pipe_fd[1] = -1; if (virCommandRunAsync(agent->cmd, NULL) < 0) goto error; diff --git a/tests/commandtest.c b/tests/commandtest.c index 816a70860f..146cc4c1bf 100644 --- a/tests/commandtest.c +++ b/tests/commandtest.c @@ -1024,6 +1024,7 @@ static int test24(const void *unused ATTRIBUTE_UNUSED) virCommandDaemonize(cmd); virCommandPassFD(cmd, newfd2, VIR_COMMAND_PASS_FD_CLOSE_PARENT); virCommandPassFD(cmd, newfd3, VIR_COMMAND_PASS_FD_CLOSE_PARENT); + newfd2 = newfd3 = -1; virCommandPassListenFDs(cmd); if (virCommandRun(cmd, NULL) < 0) { @@ -1053,7 +1054,6 @@ static int test24(const void *unused ATTRIBUTE_UNUSED) VIR_FREE(prefix); virCommandFree(cmd); VIR_FORCE_CLOSE(newfd1); - /* coverity[double_close] */ VIR_FORCE_CLOSE(newfd2); VIR_FORCE_CLOSE(newfd3); return ret; -- 2.21.0

2 1

[libvirt] [PATCH] examples: Fix installation on Windows
by Andrea Bolognani 17 May '19

17 May '19

We can't rely on $(noinst_PROGRAMS) retaining its original value, so let's use a separate $(EXAMPLES) variable instead. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- examples/Makefile.am | 35 ++++++++++++++++++++++++++++------- 1 file changed, 28 insertions(+), 7 deletions(-) diff --git a/examples/Makefile.am b/examples/Makefile.am index c7688c7c3d..11c9f44917 100644 --- a/examples/Makefile.am +++ b/examples/Makefile.am @@ -34,11 +34,32 @@ LDADD = $(STATIC_BINARIES) $(WARN_CFLAGS) \ $(top_builddir)/src/libvirt.la \ $(top_builddir)/src/libvirt-admin.la -noinst_PROGRAMS=dominfo/info1 dommigrate/dommigrate domsuspend/suspend \ - domtop/domtop hellolibvirt/hellolibvirt object-events/event-test \ - openauth/openauth rename/rename admin/list_servers admin/list_clients \ - admin/threadpool_params admin/client_limits admin/client_info \ - admin/client_close admin/logging +# List of example programs. We need to list them here instead of using +# $(noinst_PROGRAMS) directly because we want to have access to the +# unmodified list during (un)installation, but at the same time automake +# might tweak $(noinst_PROGRAMS) to eg. automatically add the .exe file +# extension when targeting Windows. +EXAMPLES = \ + admin/client_close \ + admin/client_info \ + admin/client_limits \ + admin/list_clients \ + admin/list_servers \ + admin/logging \ + admin/threadpool_params \ + dominfo/info1 \ + dommigrate/dommigrate \ + domsuspend/suspend \ + domtop/domtop \ + hellolibvirt/hellolibvirt \ + object-events/event-test \ + openauth/openauth \ + rename/rename \ + $(NULL) + +noinst_PROGRAMS = \ + $(EXAMPLES) \ + $(NULL) dominfo_info1_SOURCES = dominfo/info1.c dommigrate_dommigrate_SOURCES = dommigrate/dommigrate.c @@ -88,13 +109,13 @@ examplesdir = $(docdir)/examples install-data-local: $(INSTALL_DATA_LOCAL) $(mkinstalldirs) $(DESTDIR)$(examplesdir) - for p in $(noinst_PROGRAMS); do \ + for p in $(EXAMPLES); do \ d=$$(dirname $$p); \ $(mkinstalldirs) $(DESTDIR)$(examplesdir)/$$d; \ $(INSTALL_DATA) $(srcdir)/$${p}.c $(DESTDIR)$(examplesdir)/$$d/; \ done uninstall-local: $(UNINSTALL_LOCAL) - for p in $(noinst_PROGRAMS); do \ + for p in $(EXAMPLES); do \ rm -f $(DESTDIR)$(examplesdir)/$${p}.c; \ done -- 2.21.0

2 6

[libvirt] [PATCH 0/4] virutil: Rework how 'udevadm settle' is called
by Michal Privoznik 17 May '19

17 May '19

Apparently, some build systems have a very minimalistic approach to their default build root. Not even udevadm is there. This renders our virWaitForDevices() to be no-op. But with this rework we can still run udevadm settle just fine with no extra build dependency added. Michal Prívozník (4): virWaitForDevices: Drop untrue part of comment lib: Drop UDEVSETTLE m4: Provide default value fore UDEVADM m4: Drop needless string checks m4/virt-external-programs.m4 | 26 +++++++------------------- src/util/virutil.c | 20 ++++++-------------- 2 files changed, 13 insertions(+), 33 deletions(-) -- 2.21.0

2 8

[libvirt] [PATCH 0/6] qemu: Rework the way we probe machine type properties
by Andrea Bolognani 17 May '19

17 May '19

This is a prerequisite for vSMMUv3 support, which I'm currently working on, but I'm posting it separately because it still makes perfect sense as a standalone series and I think it's a nice cleanup in its own right. Unfortunately there's quite a bit of test suite churn associated with these changes, but hopefully that should be for the most part compensated by the *lack* of test suite churn when I introduce vSMMUv3 support. Andrea Bolognani (6): qemu: Introduce virQEMUCapsProbeQMPMachineProps() qemu: Add -machine suffix automatically qemu: Move call to virQEMUCapsProbeQMPMachineProps() qemu: Probe canonicalized machine type tests: Refresh capabilities for QEMU on ppc64 qemu: Only probe available machine types src/qemu/qemu_capabilities.c | 63 +++- .../caps_2.12.0.aarch64.replies | 48 +-- .../caps_2.12.0.ppc64.replies | 290 ++++++++-------- .../caps_2.12.0.s390x.replies | 52 +-- .../caps_2.12.0.x86_64.replies | 64 ++-- .../caps_3.0.0.ppc64.replies | 300 ++++++++-------- .../caps_3.0.0.riscv32.replies | 40 +-- .../caps_3.0.0.riscv64.replies | 40 +-- .../caps_3.0.0.s390x.replies | 52 +-- .../caps_3.0.0.x86_64.replies | 64 ++-- .../caps_3.1.0.ppc64.replies | 315 +++++++++-------- .../caps_3.1.0.x86_64.replies | 64 ++-- .../caps_4.0.0.aarch64.replies | 48 +-- .../caps_4.0.0.ppc64.replies | 328 ++++++++++-------- .../caps_4.0.0.riscv32.replies | 40 +-- .../caps_4.0.0.riscv64.replies | 40 +-- .../caps_4.0.0.s390x.replies | 52 +-- .../caps_4.0.0.x86_64.replies | 64 ++-- 18 files changed, 938 insertions(+), 1026 deletions(-) -- 2.21.0

2 12