October 2025 - Devel - libvirt List Archives

[PATCH v2 0/1] x86: install modules-load.d file to load msr module
by Hector Cao 22 Oct '25

22 Oct '25

This submission is a follow-up (v2) of the discussion (RFC) that was happening at: https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/ADJ4A... On x86, libvirt makes use of the API exposed by the msr kernel module to read the MSR (Model Specific Registers) values. On some Linux distros, this module is not loaded by default so libvirt might fail to read the MSR values and falls back to /dev/kvm to obtain these values. KVM might be behind in term of exposing all the values of the MSRs so using /dev/kvm is not ideal. This submission tries to make libvirt deploy the modules-load.d configuration so that linux loads the msr module at next boot. One caveat, libvirt installs only the file and does not trigger the load that is only done at next reboot, it is up to each distro to trigger immediately the load of the module without a reboot. v2: - only install the file when qemu driver is enabled Hector Cao (1): x86: install modules-load.d file to load msr module meson_options.txt | 1 + src/util/meson.build | 19 +++++++++++++++++++ src/util/modules-load.d/msr.conf | 1 + tools/virt-host-validate-qemu.c | 7 +++++++ 4 files changed, 28 insertions(+) create mode 100644 src/util/modules-load.d/msr.conf -- 2.45.2

2 2

[PATCH pushed] qemumonitorjsontes: Properly free blockstats
by Peter Krempa 22 Oct '25

22 Oct '25

From: Peter Krempa <pkrempa(a)redhat.com> In the patch converting block stats to objects in 58aa005f3e9 I forgot to change the allocation of the hash table in qemumonitorjsontest which doesn't use the wrapper. This problem didn't manifest itself with newer glib versions. Use 'g_object_unref' instead of 'g_free'. Fixes: 58aa005f3e9 Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- tests/qemumonitorjsontest.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/qemumonitorjsontest.c b/tests/qemumonitorjsontest.c index a229a89860..6129dde449 100644 --- a/tests/qemumonitorjsontest.c +++ b/tests/qemumonitorjsontest.c @@ -1446,7 +1446,7 @@ testQemuMonitorJSONqemuMonitorJSONGetAllBlockStatsInfo(const void *opaque) { const testGenericData *data = opaque; virDomainXMLOption *xmlopt = data->xmlopt; - g_autoptr(GHashTable) blockstats = virHashNew(g_free); + g_autoptr(GHashTable) blockstats = virHashNew(g_object_unref); qemuBlockStats *stats; g_autoptr(qemuMonitorTest) test = NULL; -- 2.51.0

3 2

[PATCH v3 0/3] qemu: Implement support for multiple device-pluggable SMMUv3s
by Nathan Chen 21 Oct '25

21 Oct '25

Hi, This is a follow up to the second patchset [0] for supporting multiple device-pluggable vSMMU instances. This patchset implements support for specifying multiple <iommu> devices within the VM definition when smmuv3 IOMMU model is specified with a pciBus driver attribute, and is tested with Shameer's recently accepted qemu series for user-creatable vSMMU devices [1] For instance, specifying the associated emulated device in a VM definition with multiple IOMMUs, configured to be routed to pcie-expander-bus or pcie-root controllers: <devices> <controller type='pci' index='0' model='pcie-root'/> <controller type='pci' index='1' model='pcie-expander-bus'> <model name='pxb-pcie'/> <target busNr='252'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x0'/> </controller> <controller type='pci' index='2' model='pcie-root-port'> <model name='pcie-root-port'/> <target chassis='2' port='0x0'/> <address type='pci' domain='0x0000' bus='0x01' slot='0x00' function='0x0'/> </controller> ... <rng model='virtio'> <backend model='random'>/dev/urandom</backend> <address type='pci' domain='0x0000' bus='0x02' slot='0x00' function='0x0'/> </rng> <rng model='virtio'> <backend model='random'>/dev/urandom</backend> <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/> </rng> <iommu model='smmuv3'> <driver pciBus='0'/> </iommu> <iommu model='smmuv3'> <driver pciBus='1'/> </iommu> </devices> This would get translated to a qemu command line with the arguments below. -device '{"driver":"pxb-pcie","bus_nr":252,"id":"pci.1","bus":"pcie.0","addr":"0x1"}' \ -device '{"driver":"pcie-root-port","port":0,"chassis":2,"id":"pci.2","bus":"pci.1","addr":"0x0"}' \ -device '{"driver":"arm-smmuv3","primary-bus":"pcie.0","id":"smmuv3.0"}' \ -device '{"driver":"arm-smmuv3","primary-bus":"pci.1","id":"smmuv3.1"}' \ ... -object '{"qom-type":"rng-random","id":"objrng0","filename":"/dev/urandom"}' \ -device '{"driver":"virtio-rng-pci","rng":"objrng0","id":"rng0","bus":"pci.2","addr":"0x0"}' \ -object '{"qom-type":"rng-random","id":"objrng1","filename":"/dev/urandom"}' \ -device '{"driver":"virtio-rng-pci","rng":"objrng1","id":"rng1","bus":"pcie.0","addr":"0x4"}' \ Changes from v2: - Make multi-iommu adoption the first commit - Consolidate smmuv3Dev into smmuv3 model - Change 'parentIdx' attribute name to 'pciBus' - Change naming of virDomainDef 'iommu' attribute to 'iommus' - Move multi-iommu check to validation code - Move qemu dev props function to device-pluggable smmuv3 commit - Fix conditional statements per feedback Changes from v1: - Moved parentIdx attribute under <driver> - Revised qemuxmlconfdata test cases to attach emulated devices This series is on Github: https://github.com/NathanChenNVIDIA/libvirt/tree/nestedsmmuv3-no-accel-10-2… Thanks, Nathan [0] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/IF6O… [1] https://lore.kernel.org/qemu-devel/20250829082543.7680-1-skolothumtho@nvidi… Signed-off-by: Nathan Chen <nathanc(a)nvidia.com> Nathan Chen (3): conf: Support multiple device-pluggable smmuv3 IOMMUs qemu: Implement pluggable-device smmuv3 tests: qemuxmlconfdata: provide device-pluggable smmuv3 sample XML and CLI args docs/formatdomain.rst | 4 + src/conf/domain_conf.c | 100 +++++++-- src/conf/domain_conf.h | 10 +- src/conf/domain_validate.c | 61 ++++-- src/conf/schemas/domaincommon.rng | 9 +- src/libvirt_private.syms | 2 + src/qemu/qemu_alias.c | 15 +- src/qemu/qemu_command.c | 197 ++++++++++++------ src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_domain_address.c | 5 +- src/qemu/qemu_driver.c | 8 +- src/qemu/qemu_postparse.c | 11 +- src/qemu/qemu_validate.c | 2 +- .../iommu-smmuv3-pci-bus.aarch64-latest.args | 41 ++++ .../iommu-smmuv3-pci-bus.aarch64-latest.xml | 62 ++++++ .../qemuxmlconfdata/iommu-smmuv3-pci-bus.xml | 49 +++++ tests/qemuxmlconftest.c | 1 + 17 files changed, 455 insertions(+), 124 deletions(-) create mode 100644 tests/qemuxmlconfdata/iommu-smmuv3-pci-bus.aarch64-latest.args create mode 100644 tests/qemuxmlconfdata/iommu-smmuv3-pci-bus.aarch64-latest.xml create mode 100644 tests/qemuxmlconfdata/iommu-smmuv3-pci-bus.xml -- 2.43.0

1 3

Qemu pause-before-switchover and libvirt live migration
by Rogério Vinhal Nunes 21 Oct '25

21 Oct '25

Hello, I’m looking into the libvirt hooks that are available to make use of during a live migration, but there's only hooks on migration begin and migration end. On qemu directly I’m able to use pause-before-switchover migration capability to run a custom operation just before it moves to the destination, but within libvirt it seems that while this is leveraged internally, it’s not currently exposed to the client in a way I can use. Does anyone have any suggestions on how I could expose this to the client in a generic and sensible way? Thanks, Rogerio

3 9

Plans for 11.9.0 release (freeze on 2025-10-29)
by Jiri Denemark 21 Oct '25

21 Oct '25

We are getting close to 11.9.0 release of libvirt. To aim for the release on Monday 03 Nov I suggest entering the freeze on Wednesday 29 Oct and tagging RC2 on Friday 31 Oct. I hope this works for everyone. Jirka

1 0

[RFC PATCH 00/16] qemu: Expose 'limits' and 'timed_stats' block device statistics via bulk stats API
by Peter Krempa 21 Oct '25

21 Oct '25

This series is RFC as there's pending QEMU work enabling either of the groups: For the limits the following qemu patches are needed: https://lists.nongnu.org/archive/html/qemu-block/2025-09/msg00804.html For timed stats, the statistics already exist but can't be enabled without: https://lists.nongnu.org/archive/html/qemu-block/2025-10/msg00040.html Patches 1-8 are cleanups and refactors Patches 9-10 extract and expose the block backing file limits. Patches 11-12 extract and expose the timed statistics Patch 13 is qemu capabilities update with the necessary changes and is not to be merged upstream; will be replaced by a proper update. Patches 14-16 add infrastructure to enable timed stats. Peter Krempa (16): qemu: monitor: Remove qemuMonitorQueryBlockstats qemu_monitor_json.c: Use consistent function hader coding style qemu_monitor_json.h: Use consistent function hader coding style qemu: monitor: Rework qemuBlockStats into a g_object qemuMigrationCookieAddNBD: Use qemuBlockGetNamedNodeData to fetch the capacities qemuMonitorJSONBlockStatsUpdateCapacityData: Merge into caller qemuMonitorJSONGetAllBlockStatsInfo: Directly probe data from 'query-named-block-nodes' Remove qemuMonitorBlockStatsUpdateCapacityBlockdev qemu: monitor: Extract block limit values Expose qemu storage request limits via bulk stats API qemu_monitor: Extract 'timed_stats' of block devices Expose qemu timed block statistics via bulk stats API DO_NOT_MERGE: Update qemu capabilities after adding patches for block limits and timed stats qemu: capabilities: Introduce QEMU_CAPS_DISK_TIMED_STATS conf: Add configuration option for timed disk statistics collection qemu: Add support for enabling timed block device statistics collection docs/formatdomain.rst | 17 + docs/manpages/virsh.rst | 54 + include/libvirt/libvirt-domain.h | 326 ++++ src/conf/domain_conf.c | 34 + src/conf/domain_conf.h | 2 + src/conf/schemas/domaincommon.rng | 11 + src/qemu/qemu_capabilities.c | 5 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 15 + src/qemu/qemu_driver.c | 203 ++- src/qemu/qemu_migration_cookie.c | 17 +- src/qemu/qemu_monitor.c | 58 +- src/qemu/qemu_monitor.h | 1191 ++++++++------ src/qemu/qemu_monitor_json.c | 607 +++++--- src/qemu/qemu_monitor_json.h | 6 - src/qemu/qemu_validate.c | 29 + .../caps_10.2.0_x86_64.replies | 1373 +++++++++-------- .../caps_10.2.0_x86_64.xml | 3 +- tests/qemumonitorjsontest.c | 2 + ...sk-statistics-intervals.x86_64-latest.args | 37 + ...isk-statistics-intervals.x86_64-latest.xml | 57 + .../disk-statistics-intervals.xml | 46 + tests/qemuxmlconftest.c | 1 + 23 files changed, 2686 insertions(+), 1409 deletions(-) create mode 100644 tests/qemuxmlconfdata/disk-statistics-intervals.x86_64-latest.args create mode 100644 tests/qemuxmlconfdata/disk-statistics-intervals.x86_64-latest.xml create mode 100644 tests/qemuxmlconfdata/disk-statistics-intervals.xml -- 2.51.0

2 17

[PATCH v2] qemu: Drop /dev/kvm from default device ACL
by Praveen K Paladugu 20 Oct '25

20 Oct '25

From: Praveen K Paladugu <prapal(a)linux.microsoft.com> A domain that runs with TCG emulation does not need kvm device, so drop it from default device ACL. Dynamically grant access to /dev/kvm based on domain type. Signed-off-by: Praveen K Paladugu <prapal(a)linux.microsoft.com> --- src/qemu/qemu.conf.in | 3 +-- src/qemu/qemu_cgroup.c | 9 +++++++-- src/qemu/qemu_domain.h | 1 + src/qemu/qemu_namespace.c | 9 +++++++-- src/qemu/test_libvirtd_qemu.aug.in | 3 +-- 5 files changed, 17 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu.conf.in b/src/qemu/qemu.conf.in index fc91ba8f08..0a8abd9544 100644 --- a/src/qemu/qemu.conf.in +++ b/src/qemu/qemu.conf.in @@ -618,8 +618,7 @@ #cgroup_device_acl = [ # "/dev/null", "/dev/full", "/dev/zero", # "/dev/random", "/dev/urandom", -# "/dev/ptmx", "/dev/kvm", -# "/dev/userfaultfd" +# "/dev/ptmx", "/dev/userfaultfd" #] # # RDMA migration requires the following extra files to be added to the list: diff --git a/src/qemu/qemu_cgroup.c b/src/qemu/qemu_cgroup.c index f10976c2b0..100604fae5 100644 --- a/src/qemu/qemu_cgroup.c +++ b/src/qemu/qemu_cgroup.c @@ -41,8 +41,7 @@ VIR_LOG_INIT("qemu.qemu_cgroup"); const char *const defaultDeviceACL[] = { "/dev/null", "/dev/full", "/dev/zero", "/dev/random", "/dev/urandom", - "/dev/ptmx", "/dev/kvm", - "/dev/userfaultfd", + "/dev/ptmx", "/dev/userfaultfd", NULL, }; #define DEVICE_PTY_MAJOR 136 @@ -86,6 +85,12 @@ qemuCgroupAllowDevicesPaths(virDomainObj *vm, if (qemuCgroupAllowDevicePath(vm, deviceACL[i], perms, ignoreEacces) < 0) return -1; } + if (vm->def->virtType == VIR_DOMAIN_VIRT_KVM) { + /* KVM requires access to /dev/kvm */ + if (qemuCgroupAllowDevicePath(vm, QEMU_DEV_KVM, VIR_CGROUP_DEVICE_RW, + ignoreEacces) < 0) + return -1; + } return 0; } diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index f4945f598a..fe4ba4fa15 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -89,6 +89,7 @@ struct _qemuDomainUnpluggingDevice { #define QEMU_DEV_SGX_PROVISION "/dev/sgx_provision" #define QEMU_DEVICE_MAPPER_CONTROL_PATH "/dev/mapper/control" #define QEMU_DEV_UDMABUF "/dev/udmabuf" +#define QEMU_DEV_KVM "/dev/kvm" #define QEMU_DOMAIN_AES_IV_LEN 16 /* 16 bytes for 128 bit random */ diff --git a/src/qemu/qemu_namespace.c b/src/qemu/qemu_namespace.c index f72da83929..ca12fcf587 100644 --- a/src/qemu/qemu_namespace.c +++ b/src/qemu/qemu_namespace.c @@ -210,13 +210,18 @@ qemuDomainGetPreservedMounts(virQEMUDriverConfig *cfg, static int qemuDomainPopulateDevices(virQEMUDriverConfig *cfg, + virDomainObj *vm, GSList **paths) { const char *const *devices = (const char *const *) cfg->cgroupDeviceACL; size_t i; - if (!devices) + if (!devices) { devices = defaultDeviceACL; + if (vm->def->virtType == VIR_DOMAIN_VIRT_KVM) { + *paths = g_slist_prepend(*paths, g_strdup(QEMU_DEV_KVM)); + } + } for (i = 0; devices[i]; i++) { *paths = g_slist_prepend(*paths, g_strdup(devices[i])); @@ -694,7 +699,7 @@ qemuDomainBuildNamespace(virQEMUDriverConfig *cfg, return 0; } - if (qemuDomainPopulateDevices(cfg, &paths) < 0) + if (qemuDomainPopulateDevices(cfg, vm, &paths) < 0) return -1; if (qemuDomainSetupAllDisks(vm, &paths) < 0) diff --git a/src/qemu/test_libvirtd_qemu.aug.in b/src/qemu/test_libvirtd_qemu.aug.in index 90012b3f52..82cfec3b4b 100644 --- a/src/qemu/test_libvirtd_qemu.aug.in +++ b/src/qemu/test_libvirtd_qemu.aug.in @@ -76,8 +76,7 @@ module Test_libvirtd_qemu = { "4" = "/dev/random" } { "5" = "/dev/urandom" } { "6" = "/dev/ptmx" } - { "7" = "/dev/kvm" } - { "8" = "/dev/userfaultfd" } + { "7" = "/dev/userfaultfd" } } { "save_image_format" = "raw" } { "dump_image_format" = "raw" } -- 2.51.0

1 0

[PATCH 0/3] virNetDevVlanParse: Fix memleak and refactor
by Peter Krempa 20 Oct '25

20 Oct '25

Peter Krempa (3): virNetDevVlanParse: Don't clear data on failure virNetDevVlanParse: Use g_autofree for temporary variables virNetDevVlanParse: Refactor cleanup src/conf/netdev_vlan_conf.c | 33 ++++++++++++++------------------- 1 file changed, 14 insertions(+), 19 deletions(-) -- 2.51.0

2 5

download.libvirt.org has expired certificate
by Ivan Veloz 20 Oct '25

20 Oct '25

Hi everyone, I just wanted to bring to your attention that the HTTPS certificate for download.libvirt.org has expired on 2025/10/16, last Thursday. At least on my machine, the certificate I am getting has the following attributes: --------------------------------------- Subject Name Common Namedownload.libvirt.org Issuer Name CountryUS OrganizationLet's Encrypt Common NameR10 Validity Not BeforeFri, 18 Jul 2025 09:44:37 GMT Not AfterThu, 16 Oct 2025 09:44:36 GMT Miscellaneous Serial Number05:60:F2:EA:01:61:B7:52:B2:0E:D3:6F:28:BD:54:7C:F8:FE --------------------------------------- The IP address resolving on my machine is: $ nslookup download.libvirt.org Server: 127.0.0.53 Address: 127.0.0.53#53 Non-authoritative answer: Name: download.libvirt.orgAddress: 149.202.84.84 Hopefully the renewal process goes smoothly (or, let me know if something is horribly wrong on my end!). Kind regards, Ivan Veloz

2 1

[PATCH v3 0/8] Enable mshv support for Qemu domains
by Praveen K Paladugu 17 Oct '25

17 Oct '25

"[PATCH v3 00/26] Implementing a MSHV (Microsoft Hypervisor) accelerator" patchset in qemu-devel mailing list introduces MSHV support within Qemu. With above patchset, Qemu guests can be started with `-accel mshv` option to use Microsoft Hypervisor (/dev/mshv) as the hypervisor. Retaining Libvirt's naming convention, this translates to `hyperv` domains. Below is an excerpt of the xml to use with this patchset: ``` <domain type='hyperv'> <name>qemu_msh_testv</name> <cpu mode='host-model'/> <os> <type machine='q35'>hvm</type> </os> ... ``` This patch set introduces the necessary changes within libvirt to support Qemu guest with hyperv (mshv) hypervisor. Chages in v3: - Addressed review comments on v2. - fixed qemucapabilitiestest case - Added a case to qemuxmlconftest Changes in v2: - Addressed review comments on v1. - Added a test to verify mshv capabilities. Praveen K Paladugu (8): qemu: introduce mshv capabilities qemu: use mshv accel for hyperv domains qemu: add /dev/mshv to default cgroup acl qemu: probe mshv capabilities conf: update capability schema tests: Add mshv variant for qemucapabilitiestest tests: Add domaincaps and qemucaps case for mshv tests: add mshv test to qemuxmlconftest src/conf/schemas/capability.rng | 1 + src/qemu/qemu.conf.in | 2 +- src/qemu/qemu_capabilities.c | 61 +- src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_cgroup.c | 2 +- src/qemu/qemu_command.c | 4 +- src/qemu/qemu_monitor.c | 12 + src/qemu/qemu_monitor.h | 12 +- src/qemu/qemu_monitor_json.c | 32 +- src/qemu/qemu_monitor_json.h | 13 +- src/qemu/test_libvirtd_qemu.aug.in | 3 +- .../qemu_10.1.0-q35.x86_64+mshv.xml | 222 + .../qemu_10.1.0-tcg.x86_64+mshv.xml | 1810 + .../qemu_10.1.0.x86_64+mshv.xml | 222 + tests/qemucapabilitiesdata/README.rst | 5 + .../caps_10.1.0_x86_64+mshv.replies | 46912 ++++++++++++++++ .../caps_10.1.0_x86_64+mshv.xml | 4996 ++ .../qemucaps2xmloutdata/caps.x86_64+mshv.xml | 29 + ...86_64-q35-headless.x86_64-latest+mshv.args | 48 + ...x86_64-q35-headless.x86_64-latest+mshv.xml | 87 + ...mshv-x86_64-q35-headless.x86_64-latest.err | 1 + .../mshv-x86_64-q35-headless.xml | 35 + tests/qemuxmlconftest.c | 5 + 23 files changed, 54495 insertions(+), 20 deletions(-) create mode 100644 tests/domaincapsdata/qemu_10.1.0-q35.x86_64+mshv.xml create mode 100644 tests/domaincapsdata/qemu_10.1.0-tcg.x86_64+mshv.xml create mode 100644 tests/domaincapsdata/qemu_10.1.0.x86_64+mshv.xml create mode 100644 tests/qemucapabilitiesdata/caps_10.1.0_x86_64+mshv.replies create mode 100644 tests/qemucapabilitiesdata/caps_10.1.0_x86_64+mshv.xml create mode 100644 tests/qemucaps2xmloutdata/caps.x86_64+mshv.xml create mode 100644 tests/qemuxmlconfdata/mshv-x86_64-q35-headless.x86_64-latest+mshv.args create mode 100644 tests/qemuxmlconfdata/mshv-x86_64-q35-headless.x86_64-latest+mshv.xml create mode 100644 tests/qemuxmlconfdata/mshv-x86_64-q35-headless.x86_64-latest.err create mode 100644 tests/qemuxmlconfdata/mshv-x86_64-q35-headless.xml -- 2.50.1

2 20