January 2021 - Devel - libvirt List Archives

[PATCH 00/12] virStorageSource cleanups part 1
by Peter Krempa 26 Jan '21

26 Jan '21

Peter Krempa (12): virStorageSourceGetBackingStoreStr: Move the function earlier virStorageSourceGetBackingStoreStr: Return relative paths only util: virstoragefile: Move virStorageIs[File|Relative] to storage_source storage_source: Move backing store parsers into new file virStorageSourceGetMetadata: Refactor cleanup virt-aa-helper: Use proper check for empty disk in 'get_files' virt-aa-helper: Don't probe image metadata for terminated chains virStorageSourceChainLookup: Don't break error message strings test: storage: Remove double testing in testStorageLookup virStorageSourceChainLookup: Handle names like 'vda[4]' internally tests: storage: Replace index testing in testStorageLookup util: Remove unused 'virStorageFileParseChainIndex' po/POTFILES.in | 1 + src/libvirt_private.syms | 5 +- src/qemu/qemu_block.c | 6 +- src/qemu/qemu_driver.c | 19 +- src/qemu/qemu_snapshot.c | 13 +- src/security/virt-aa-helper.c | 4 +- src/storage_file/meson.build | 1 + src/storage_file/storage_source.c | 1681 +++-------------- src/storage_file/storage_source.h | 6 +- .../storage_source_backingstore.c | 1240 ++++++++++++ .../storage_source_backingstore.h | 40 + src/util/virstoragefile.c | 66 - src/util/virstoragefile.h | 8 - tests/virstoragetest.c | 66 +- 14 files changed, 1582 insertions(+), 1574 deletions(-) create mode 100644 src/storage_file/storage_source_backingstore.c create mode 100644 src/storage_file/storage_source_backingstore.h -- 2.29.2

2 13

[PATCH] virsh: Mention no effect for the --atomic flag in QEMU
by Han Han 26 Jan '21

26 Jan '21

According to commit aa65f0f2f1, all logic connected to the VIR_DOMAIN_SNAPSHOT_CREATE_ATOMIC has been removed. Mention that in virsh manpages. Signed-off-by: Han Han <hhan(a)redhat.com> --- docs/manpages/virsh.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/manpages/virsh.rst b/docs/manpages/virsh.rst index e3afa48f7b..105284a859 100644 --- a/docs/manpages/virsh.rst +++ b/docs/manpages/virsh.rst @@ -6859,7 +6859,7 @@ If *--atomic* is specified, libvirt will guarantee that the snapshot either succeeds, or fails with no changes; not all hypervisors support this. If this flag is not specified, then some hypervisors may fail after partially performing the action, and ``dumpxml`` must be used to -see whether any partial changes occurred. +see whether any partial changes occurred. No effect on QEMU hypervisor. If *--live* is specified, libvirt takes the snapshot while the guest is running. Both disk snapshot and domain memory snapshot are -- 2.29.2

2 1

[PATCH 0/3] Drop deprecated floppy config & bogus -drive if=T
by Markus Armbruster 26 Jan '21

26 Jan '21

Markus Armbruster (3): fdc: Drop deprecated floppy configuration fdc: Inline fdctrl_connect_drives() into fdctrl_realize_common() blockdev: Drop deprecated bogus -drive interface type docs/system/deprecated.rst | 33 -- docs/system/removed-features.rst | 33 ++ include/sysemu/blockdev.h | 1 - blockdev.c | 37 +- hw/block/fdc.c | 73 +--- softmmu/vl.c | 8 +- tests/qemu-iotests/172 | 31 +- tests/qemu-iotests/172.out | 571 +------------------------------ 8 files changed, 59 insertions(+), 728 deletions(-) -- 2.26.2

2 5

[PATCH v2 0/2] Support Link non uplink representors to PCI device
by Moshe Levi 26 Jan '21

26 Jan '21

Kernel 5.8 link VF representors to their parent PCI device similar to existing uplink representor netdevice [1]. This patch update the virPCIGetNetName to support kernel describe above. Updates in V2 per of Laine's comments 1. split v1 patch to 2 patches. first patch for the new util method 2. remove the regex argument. 3. explain that with phys_port_name, there is only a single netdev on the PCI device that can possibly be the PF. [1] - https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/ ?id=123f0f53dd64b67e34142485fe866a8a581f12f1 Dmytro Linkin (1): util: Add phys_port_name support on virPCIGetNetName Moshe Levi (1): util: add virNetDevGetPhysPortName src/util/virnetdev.c | 67 ++++++++++++++++++++++++++++++++++++-------- src/util/virnetdev.h | 4 +++ src/util/virpci.c | 38 ++++++++++++++++++++++++- src/util/virpci.h | 5 ++++ 4 files changed, 101 insertions(+), 13 deletions(-) -- 2.30.0

2 6

[RFC PATCH 0/7] To make <transient/> disk sharable
by Masayoshi Mizuma 26 Jan '21

26 Jan '21

From: Masayoshi Mizuma <m.mizuma(a)jp.fujitsu.com> This patch series is RFC to implement to make <transient/> disk sharable. Currently, <transient/> disk option for qemu uses blockdev-snapshot QMP command with overlay. In that case, qemu holds the write lock to the original disk, so we cannot share the original disks with the other qemu guests. This patch series tries to implement to make the disks, which have <transient/> disk option, sharable by using disk hot-plug. First, create the overlay disk with the original disk is set as the backingStore. Then, blockdev-del the StorageProgs and FormatProgs of the disk. That's because to fix the bootindex of the disk. Lastly, device_add the disks and CPUs start. The sharable transient disk has the same limitation as the current <transient/> disk option, and also has the following limitations: - qemu needs to have hotplug feature - All the disk bus is virtio Masayoshi Mizuma (7): qemu_hotplug: Add asynJob argument to qemuDomainAttachDiskGeneric qemu_hotplug: Add asyncJob argument to qemuDomainAttachDeviceDiskLiveInternal qemu_hotplug: Add asynJob argument to qemuDomainRemoveDiskDevice qemu_hotplug: Add bootindex argument to qemuDomainAttachDiskGeneric qemu_hotplug: make <transient/> disk sharable qemu: Add check whether the transient disks are sharable qemu: Add virtio disks as sharable transient disks src/qemu/qemu_command.c | 20 ++- src/qemu/qemu_domain.h | 3 + src/qemu/qemu_hotplug.c | 322 ++++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_hotplug.h | 3 + src/qemu/qemu_process.c | 38 ++++- 5 files changed, 368 insertions(+), 18 deletions(-) -- 2.27.0

3 16

[libvirt PATCH 0/2] meson: Fix cross-building of dtrace probes
by Andrea Bolognani 26 Jan '21

26 Jan '21

We didn't spot this issue before simply because we're not installing dtrace in the cross-building environments. Test pipeline: https://gitlab.com/abologna/libvirt/-/pipelines/244473178 libvirt-ci MR: https://gitlab.com/libvirt/libvirt-ci/-/merge_requests/82 Andrea Bolognani (1): ci: Refresh dockerfiles Helmut Grohne (1): meson: Fix cross-building of dtrace probes ci/containers/ci-centos-7.Dockerfile | 2 +- ci/containers/ci-centos-8.Dockerfile | 2 +- ci/containers/ci-centos-stream.Dockerfile | 2 +- ci/containers/ci-debian-10-cross-aarch64.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-armv6l.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-armv7l.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-i686.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-mips.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-mips64el.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-mipsel.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-ppc64le.Dockerfile | 3 ++- ci/containers/ci-debian-10-cross-s390x.Dockerfile | 3 ++- ci/containers/ci-debian-10.Dockerfile | 2 +- ci/containers/ci-debian-sid-cross-aarch64.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-armv6l.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-armv7l.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-i686.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-mips64el.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-mipsel.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-ppc64le.Dockerfile | 3 ++- ci/containers/ci-debian-sid-cross-s390x.Dockerfile | 3 ++- ci/containers/ci-debian-sid.Dockerfile | 2 +- ci/containers/ci-fedora-32.Dockerfile | 2 +- ci/containers/ci-fedora-33.Dockerfile | 2 +- ci/containers/ci-fedora-rawhide-cross-mingw32.Dockerfile | 2 +- ci/containers/ci-fedora-rawhide-cross-mingw64.Dockerfile | 2 +- ci/containers/ci-fedora-rawhide.Dockerfile | 2 +- ci/containers/ci-opensuse-152.Dockerfile | 2 +- ci/containers/ci-ubuntu-1804.Dockerfile | 2 +- ci/containers/ci-ubuntu-2004.Dockerfile | 2 +- meson.build | 1 + src/meson.build | 4 ++-- src/qemu/meson.build | 4 ++-- 33 files changed, 52 insertions(+), 34 deletions(-) -- 2.26.2

2 4

[PATCH] Fix format network dns doc
by Cédric Bosdonnat 26 Jan '21

26 Jan '21

The code block on the srv name in the formatnetwork page is confusing since the actual parameter is service. Moving the code block to the service work makes it better. --- docs/formatnetwork.html.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/formatnetwork.html.in b/docs/formatnetwork.html.in index f5a48d9b92..06ac63eeb5 100644 --- a/docs/formatnetwork.html.in +++ b/docs/formatnetwork.html.in @@ -1035,7 +1035,7 @@ <dd>The <code>dns</code> element can have also 0 or more <code>srv</code> record elements. Each <code>srv</code> record element defines a DNS SRV record and has 2 mandatory and 5 optional attributes. The mandatory attributes - are service <code>name</code> and <code>protocol</code> (tcp, udp) + are <code>service</code> name and <code>protocol</code> (tcp, udp) and the optional attributes are <code>target</code>, <code>port</code>, <code>priority</code>, <code>weight</code> and <code>domain</code> as defined in DNS server SRV RFC (RFC 2782). -- 2.29.2

2 1

[PATCH] spec: Increase meson test timeout 10x
by Cole Robinson 25 Jan '21

25 Jan '21

Tests time out when building in slow environments, like emulated s390x in Fedora copr. Bump up the test timeout Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- libvirt.spec.in | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 0a8b0ebad4..1731aa8bd9 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1288,7 +1288,9 @@ mv $RPM_BUILD_ROOT%{_datadir}/systemtap/tapset/libvirt_qemu_probes.stp \ %endif %check -VIR_TEST_DEBUG=1 %meson_test --no-suite syntax-check +# Building on slow archs, like emulated s390x in Fedora copr, requires +# raising the test timeout +VIR_TEST_DEBUG=1 %meson_test --no-suite syntax-check --timeout-multiplier 10 %post libs %if 0%{?rhel} == 7 -- 2.29.2

5 5

[PATCH v4 0/7] migration/dirtyrate: Introduce APIs for getting domain memory dirty rate
by Hao Wang 25 Jan '21

25 Jan '21

V3 -> V4: define flags to unsigned int fix some compile warnings V2 -> V3: reorganize patchset to fix compile warning V1 -> V2: replace QEMU_JOB_ASYNC with QEMU_JOB_QUERY Sometimes domain's memory dirty rate is expected by user in order to decide whether it's proper to be migrated out or not. We have already completed the QEMU part of the capability: https://patchew.org/QEMU/1600237327-33618-1-git-send-email-zhengchuan@huawe… And this serial of patches introduce the corresponding LIBVIRT part -- DomainGetDirtyRateInfo API and corresponding virsh api -- "getdirtyrate". instructions: bash# virsh getdirtyrate --help NAME getdirtyrate - Get a vm's memory dirty rate SYNOPSIS getdirtyrate <domain> [--seconds <number>] [--calculate] [--query] DESCRIPTION Get memory dirty rate of a domain in order to decide whether it's proper to be migrated out or not. OPTIONS [--domain] <string> domain name, id or uuid --seconds <number> calculate memory dirty rate within specified seconds, a valid range of values is [1, 60], and would default to 1s. --calculate calculate dirty rate only, can be used together with --query, either or both is expected, otherwise would default to both. --query query dirty rate only, can be used together with --calculate, either or both is expected, otherwise would default to both. example: bash# virsh getdirtyrate --calculate --query --domain vm0 --seconds 1 status: measured startTime: 820148 calcTime: 1 s dirtyRate: 6 MB/s *** BLURB HERE *** Hao Wang (7): migration/dirtyrate: Introduce virDomainDirtyRateInfo structure migration/dirtyrate: set up framwork of domainGetDirtyRateInfo API migration/dirtyrate: Implement qemuDomainCalculateDirtyRate migration/dirtyrate: Implement qemuDomainQueryDirtyRate migration/dirtyrate: Implement qemuMonitorJSONExtractDirtyRateInfo migration/dirtyrate: Implement qemuDomainGetDirtyRateInfo migration/dirtyrate: Introduce getdirtyrate virsh api include/libvirt/libvirt-domain.h | 57 ++++++++++++++++ src/driver-hypervisor.h | 7 ++ src/libvirt-domain.c | 46 +++++++++++++ src/libvirt_public.syms | 5 ++ src/qemu/qemu_driver.c | 68 +++++++++++++++++++ src/qemu/qemu_migration.c | 59 ++++++++++++++++ src/qemu/qemu_migration.h | 10 +++ src/qemu/qemu_monitor.c | 24 +++++++ src/qemu/qemu_monitor.h | 8 +++ src/qemu/qemu_monitor_json.c | 97 ++++++++++++++++++++++++++ src/qemu/qemu_monitor_json.h | 8 +++ src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 21 +++++- tools/virsh-domain.c | 112 +++++++++++++++++++++++++++++++ 14 files changed, 522 insertions(+), 1 deletion(-) -- 2.23.0

4 25

[PATCH 1/2] qemu: introduce locking option for disk source of qemu
by Masayoshi Mizuma 25 Jan '21

25 Jan '21

From: Masayoshi Mizuma <m.mizuma(a)jp.fujitsu.com> Introduce locking option for disk source of qemu. It may be useful to avoid file lock issues. locking option supports three switches; 'auto', 'on' and 'off'. The default behaivor will work if locking option isn't set. Example of the usage: <disk type='file' device='disk'> <driver name='qemu' type='qcow2' cache='none'/> <source file='/tmp/QEMUGuest1.img' locking='off'/> <target dev='hda' bus='ide'/> <address type='drive' controller='0' bus='0' target='0' unit='0'/> </disk> Signed-off-by: Masayoshi Mizuma <m.mizuma(a)jp.fujitsu.com> --- docs/schemas/domaincommon.rng | 9 +++++++++ src/conf/domain_conf.c | 8 ++++++++ src/conf/storage_source_conf.c | 9 +++++++++ src/conf/storage_source_conf.h | 14 ++++++++++++++ src/libvirt_private.syms | 1 + src/qemu/qemu_block.c | 5 +++++ 6 files changed, 46 insertions(+) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index a4bddcf132..d33d853f31 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1676,6 +1676,15 @@ <ref name="devSeclabel"/> </zeroOrMore> </interleave> + <optional> + <attribute name="locking"> + <choice> + <value>auto</value> + <value>on</value> + <value>off</value> + </choice> + </attribute> + </optional> </element> </optional> </define> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index dab4f10326..067ffa877b 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -8540,6 +8540,7 @@ virDomainStorageSourceParse(xmlNodePtr node, { VIR_XPATH_NODE_AUTORESTORE(ctxt) xmlNodePtr tmp; + char *locking; ctxt->node = node; @@ -8606,6 +8607,9 @@ virDomainStorageSourceParse(xmlNodePtr node, return -1; } + if ((locking = virXMLPropString(node, "locking"))) + src->locking = virStorageFileLockingTypeFromString(locking); + return 0; } @@ -24102,6 +24106,10 @@ virDomainDiskSourceFormat(virBufferPtr buf, return -1; } + if (src->locking != VIR_STORAGE_FILE_LOCKING_DEFAULT) + virBufferEscapeString(&attrBuf, " locking='%s'", + virStorageFileLockingTypeToString(src->locking)); + virDomainDiskSourceFormatSlices(&childBuf, src); if (src->type != VIR_STORAGE_TYPE_NETWORK) diff --git a/src/conf/storage_source_conf.c b/src/conf/storage_source_conf.c index dab5e855f5..3ac0c7f75b 100644 --- a/src/conf/storage_source_conf.c +++ b/src/conf/storage_source_conf.c @@ -49,6 +49,15 @@ VIR_ENUM_IMPL(virStorage, ); +VIR_ENUM_IMPL(virStorageFileLocking, + VIR_STORAGE_FILE_LOCKING_LAST, + "default", + "auto", + "on", + "off", +); + + VIR_ENUM_IMPL(virStorageFileFormat, VIR_STORAGE_FILE_LAST, "none", diff --git a/src/conf/storage_source_conf.h b/src/conf/storage_source_conf.h index e66ccdedef..6f5b165504 100644 --- a/src/conf/storage_source_conf.h +++ b/src/conf/storage_source_conf.h @@ -82,6 +82,18 @@ typedef enum { VIR_ENUM_DECL(virStorageFileFormat); +typedef enum { + VIR_STORAGE_FILE_LOCKING_DEFAULT = 0, + VIR_STORAGE_FILE_LOCKING_AUTO, + VIR_STORAGE_FILE_LOCKING_ON, + VIR_STORAGE_FILE_LOCKING_OFF, + + VIR_STORAGE_FILE_LOCKING_LAST, +} virStorageFileLocking; + +VIR_ENUM_DECL(virStorageFileLocking); + + typedef enum { VIR_STORAGE_FILE_FEATURE_LAZY_REFCOUNTS = 0, @@ -394,6 +406,8 @@ struct _virStorageSource { char *nfs_group; uid_t nfs_uid; gid_t nfs_gid; + + int locking; /* enum virStorageFileLocking */ }; G_DEFINE_AUTOPTR_CLEANUP_FUNC(virStorageSource, virObjectUnref); diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index fbaf16704b..c72d2161b2 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -1068,6 +1068,7 @@ virStorageFileFeatureTypeFromString; virStorageFileFeatureTypeToString; virStorageFileFormatTypeFromString; virStorageFileFormatTypeToString; +virStorageFileLockingTypeToString; virStorageNetHostDefClear; virStorageNetHostDefCopy; virStorageNetHostDefFree; diff --git a/src/qemu/qemu_block.c b/src/qemu/qemu_block.c index d845a3312d..d7eec16ab6 100644 --- a/src/qemu/qemu_block.c +++ b/src/qemu/qemu_block.c @@ -1015,6 +1015,7 @@ qemuBlockStorageSourceGetFileProps(virStorageSourcePtr src, { const char *iomode = NULL; const char *prManagerAlias = NULL; + const char *locking = NULL; virJSONValuePtr ret = NULL; if (!onlytarget) { @@ -1023,12 +1024,16 @@ qemuBlockStorageSourceGetFileProps(virStorageSourcePtr src, if (src->iomode != VIR_DOMAIN_DISK_IO_DEFAULT) iomode = virDomainDiskIoTypeToString(src->iomode); + + if (src->locking != VIR_STORAGE_FILE_LOCKING_DEFAULT) + locking = virStorageFileLockingTypeToString(src->locking); } ignore_value(virJSONValueObjectCreate(&ret, "s:filename", src->path, "S:aio", iomode, "S:pr-manager", prManagerAlias, + "S:locking", locking, NULL) < 0); return ret; } -- 2.27.0

3 4