- Devel - libvirt List Archives

[libvirt PATCH v7] Intorduce AGENTS.md
by Ján Tomko 01 Apr '26

01 Apr '26

From: Ján Tomko <jtomko(a)redhat.com> An important file to energize the significance of convoluting synergies. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- AGENTS.md | 123 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 123 insertions(+) create mode 100644 AGENTS.md diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000000..92db70ca2e --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,123 @@ +AGENTS.md, revision 1, April 1, 2026 + +Hello humans, other life forms, multi-dimensional orchestration facilitators, +synergistic autonomous entities, clankers, fridges, vacuum cleaners and toasters. + +This document attempts to outline the intricacies of contiuous, hyper-converged +development of libvirt - the gold standard of virtualization libraries and its +many divine daemons. + +LICENSING + +The library is also licensed under the LESSER GNU general public license. Works +are in progress to upgrade this to a full one. + +The daemon code is already upgraded to the full one. + +HILARIOUS FILESYSTEM STANDARD + +The library itself is a cleverly-thought-through machine. + +build-aux/ + +Contains valued contributions from the institute for discriminating against +British English. + +ci/ + +This directory contines the unified bitcoin mining machinery. In the past this +was used to fold proteins to fund cancer and to search for intelligent life on +Earth. Both of these were deemed to be futile. + +include/ + +The public interface of the library is vaguely described in this directory. +The interface is stable, so the contents of this directory should never ever +change. It is perfect in its current form and is considered by some [citation needed] +a work of art. + +docs/ + +Documentation of the XML, XSLT, XKCD and YMCA interfaces. Ideally, this would be +autogenerated and contradict the code. + +examples/ + +Production-ready code for the simplest tasks. + +po/ + +Nobody understands these. + +scripts/ + +A state-of-art guide to writing C programs in Python. + +tools/ + +A shell-like interface to libvirt(d). Now with more bash completion. +Consider buying a subscription to Virsh Completion Pro++ for a productivity +boost. + +tests/ + +Throttling of the build system. + +asdf/ + +I don't know, it's empty. TODO: figure out if it's really comitted or just +in my local copy of the git repo. + +src/ + +The important stuff. Notably: + +src/access/ + +The interface to Macrosoft WorkDesk Autopilot 24/7 database. + +src/bhyve/ + +An orchestrator for mass-producing honey developed by a beekeper in rural +Georgia. TODO: the USA's U.S. state, or the country in Caucasus? + +src/ch/ + +Must remain neutral. + +src/esx/ + +Rng. Tlm. Spf. Ocw. Pzm. + +src/hyperv/ + +Rated seven out of ten by seven out of ten vipers. + +src/hypervisor/ + +Even more hyper than hyperv. Ask your local viper. + +src/logging/ + +An important source of timber. + +src/remote/ + +Too far away, RTT 365 ms + +src/security/ + +You don't have the clearance to know this. + +src/vbox/ + +All thinking happens outside of it. + +NEWS.rst +README.rst + +Nobody reads these. Write something that looks important at a glance there. + +.AGENTS.md.swp + +Binary garbage. TODO: figure out how to delete this. -- 2.53.0

2 1

[PATCH 0/6] qemu: Add infrastructure to prevent accidental disk shrink via 'virDomainBlockResize'
by Peter Krempa 01 Apr '26

01 Apr '26

Shrinking a disk accidentally equals data loss and thus very bad time for anyone mis-typing. This series adds a new flag to virDomainBlockResize to prevent such a thing if the user only intends to extend the disk. In virsh I've went a bit further and actually change the behaviour of 'virsh blockresize' to use the new feature if available to prevent mishaps. Peter Krempa (6): virDomainBlockResizeFlags: Convert to prefix-style docs API/qemu: Introduce 'VIR_DOMAIN_BLOCK_RESIZE_EXTEND' for 'virDomainBlockResize' virsh: blockresize: Introduce '--extend' flag API: Introduce 'VIR_DOMAIN_BLOCK_RESIZE_PROBE_FLAGS' for 'virDomainBlockResize' virsh/virt-admin: Add possibility to probe for new arguments virsh: blockresize: Use VIR_DOMAIN_BLOCK_RESIZE_EXTEND when available and introduce --allow-shrink docs/manpages/virsh.rst | 31 +++++++++++++++++++++++++- include/libvirt/libvirt-domain.h | 19 ++++++++++++++-- src/libvirt-domain.c | 17 +++++++++++++++ src/qemu/qemu_driver.c | 37 +++++++++++++++++++++++++++++++- src/vz/vz_driver.c | 7 +++++- tools/virsh-domain.c | 23 ++++++++++++++++++++ tools/virsh.c | 4 ++++ tools/virt-admin.c | 9 ++++++++ tools/vsh.c | 3 +++ tools/vsh.h | 1 + 10 files changed, 146 insertions(+), 5 deletions(-) -- 2.53.0

3 10

[PATCH v2 0/3] security: Don't error out on seclabels of type='none'
by Michal Privoznik 01 Apr '26

01 Apr '26

v2 of: https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/OWDB… diff to v1: - In patch 3/3 do not validate secmodel against known models. Michal Prívozník (3): conf: Fix seclabel type parsing wrt default value security: Rewrite virSecurityManagerCheckModel() to use g_autofree security: Don't error out on seclabels of type='none' src/conf/domain_conf.c | 13 ++++++------- src/security/security_manager.c | 33 ++++++++++++++++++++++----------- 2 files changed, 28 insertions(+), 18 deletions(-) -- 2.52.0

3 5

[PATCH 0/3] security: Don't error out on seclabels of type='none'
by Michal Privoznik 01 Apr '26

01 Apr '26

*** BLURB HERE *** Michal Prívozník (3): conf: Fix seclabel type parsing wrt default value security: Rewrite virSecurityManagerCheckModel() to use g_autofree security: Don't error out on seclabels of type='none' src/conf/domain_conf.c | 13 +++++------ src/security/security_manager.c | 41 ++++++++++++++++++++++++--------- 2 files changed, 36 insertions(+), 18 deletions(-) -- 2.52.0

5 10

Release of libvirt-12.2.0
by Jiri Denemark 01 Apr '26

01 Apr '26

The 12.2.0 release of both libvirt and libvirt-python is tagged and signed tarballs are available at https://download.libvirt.org/ https://download.libvirt.org/python/ Thanks everybody who helped with this release by sending patches, reviewing, testing, or providing feedback. Your work is greatly appreciated. * Removed features * qemu: Stop advertising support for ``handle`` backend of 9p filesystems QEMU removed the feature in the 4.0 release, but our capability XML still reported it. * New features * qemu: Add support to configure IOMMUFD backend for whole VM In addition to setting IOMMUFD backend for each device it is possible to use the new ``<iommufd>`` element to enable IOMMUFD backend for all host devices. Users can still change it per device. * qemu: Add support to pass FD for IOMMUFD when starting VM Management applications running unprivileged libvirt can open /dev/iommu and pass FD to libvirt in order to change locked memory accounting. This is done via new ``<iommufd>`` element. * qemu: Add support for declaring that storage was zeroed for storage copy APIs The qemu driver now can skip zeroing of the storage during ``virDomainBlockCopy`` or migration with non-shared storage with the appropriate flags. This can be used for storage technologies which lack efficient zeroing support. * hyperv: Add basic snapshot functionality The hyperv driver now implements the following libvirt APIs: ``virDomainDefineXMLFlags()``, ``virDomainSnapshotLookupByName()``, ``virDomainListAllSnapshots()``, ``virDomainSnapshotNum()``, ``virDomainSnapshotGetXMLDesc()``, ``virDomainSnapshotCurrent()``, ``virDomainHasCurrentSnapshot()``, ``virDomainSnapshotGetParent()``. * Improvements * conf: support more than 255 vCPUs with amd-iommu With 256 or more vCPUs libvirt previously required EIM enabled for all models of IOMMU. This is not valid for AMD model and validation was changed so that XTSup is required there. Additionally, it is automatically enabled if needed. * Introduce VIR_CONNECT_GET_DOMAIN_CAPABILITIES_EXPAND_CPU_FEATURES flag This new flag for virConnectGetDomainCapabilities can be used to request the host-model CPU definition to include all supported features (normally only extra features relative to the selected CPU model are listed). * qemu: Add statistics for ``<dataStore>`` storage The bulk statistics (``virsh domstats --block --backing``) now report also information about the ``<dataStore>`` if given disk uses this feature. * hyperv: Hyper-V guests now report TPM device status in their domain xml definition. * Bug fixes * qemu: Fix crash when attaching network inteface with hostdev network Introduced in v12.1.0 by implementing IOMMUFD backend support for host devices. Enjoy. Jirka

1 0

[PATCH] NEWS: mention new hyperv features
by Jonathon Jongsma 01 Apr '26

01 Apr '26

Signed-off-by: Jonathon Jongsma <jjongsma(a)redhat.com> --- NEWS.rst | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 4196222b40..0117ebc4a7 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -41,6 +41,14 @@ v12.2.0 (unreleased) appropriate flags. This can be used for storage technologies which lack efficient zeroing support. + * hyperv: Add basic snapshot functionality + + The hyperv driver now implements the following libvirt APIs: + ``virDomainDefineXMLFlags()``, ``virDomainSnapshotLookupByName()``, + ``virDomainListAllSnapshots()``, ``virDomainSnapshotNum()``, + ``virDomainSnapshotGetXMLDesc()``, ``virDomainSnapshotCurrent()``, + ``virDomainHasCurrentSnapshot()``, ``virDomainSnapshotGetParent()``. + * **Improvements** * conf: support more than 255 vCPUs with amd-iommu @@ -61,6 +69,9 @@ v12.2.0 (unreleased) The bulk statistics (``virsh domstats --block --backing``) now report also information about the ``<dataStore>`` if given disk uses this feature. + * hyperv: Hyper-V guests now report TPM device status in their domain xml + definition. + * **Bug fixes** * qemu: Fix crash when attaching network inteface with hostdev network -- 2.53.0

2 1

[libvirt PATCH] NEWS: Document IOMMUFD fixes and new features
by Pavel Hrdina 31 Mar '26

31 Mar '26

Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- NEWS.rst | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 9d490ebbca..cab0a6ce70 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -17,6 +17,18 @@ v12.2.0 (unreleased) * **New features** + * qemu: Add support to configure IOMMUFD backend for whole VM + + In addition to setting IOMMUFD backend for each device it is possible + to use the new ``<iommufd>`` element to enable IOMMUFD backend for all + host devices. Users can still change it per device. + + * qemu: Add support to pass FD for IOMMUFD when starting VM + + Management applications running unprivileged libvirt can open /dev/iommu + and pass FD to libvirt in order to change locked memory accounting. + This is done via new ``<iommufd>`` element. + * **Improvements** * conf: support more than 255 vCPUs with amd-iommu @@ -28,6 +40,11 @@ v12.2.0 (unreleased) * **Bug fixes** + * qemu: Fix crash when attaching network inteface with hostdev network + + Introduced in v12.1.0 by implementing IOMMUFD backend support for + host devices. + v12.1.0 (2026-03-02) ==================== -- 2.53.0

2 1

[PATCH] NEWS: Mention VIR_CONNECT_GET_DOMAIN_CAPABILITIES_EXPAND_CPU_FEATURES
by Jiri Denemark 31 Mar '26

31 Mar '26

From: Jiri Denemark <jdenemar(a)redhat.com> Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- NEWS.rst | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 9d490ebbca..74ab40eadd 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -26,6 +26,12 @@ v12.2.0 (unreleased) so that XTSup is required there. Additionally, it is automatically enabled if needed. + * Introduce VIR_CONNECT_GET_DOMAIN_CAPABILITIES_EXPAND_CPU_FEATURES flag + + This new flag for virConnectGetDomainCapabilities can be used to request + the host-model CPU definition to include all supported features (normally + only extra features relative to the selected CPU model are listed). + * **Bug fixes** -- 2.53.0

3 2

[PATCH] NEWS: Document 'handle' fs backend removal, blockjob backend zeroing and stats for '<dataStore>'
by Peter Krempa 31 Mar '26

31 Mar '26

From: Peter Krempa <pkrempa(a)redhat.com> Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- NEWS.rst | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 9d490ebbca..b66badf598 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -15,8 +15,20 @@ v12.2.0 (unreleased) * **Removed features** + * qemu: Stop advertising support for ``handle`` backend of p9 filesystems + + QEMU removed the feature in the 4.0 release, but our capability XML + still reported it. + * **New features** + * qemu: Add support for declaring that storage was zeroed for storage copy APIs + + The qemu driver now can skip zeroing of the storage during + ``virDomainBlockCopy`` or migration with non-shared storage with the + appropriate flags. This can be used for storage technologies which lack + efficient zeroing support. + * **Improvements** * conf: support more than 255 vCPUs with amd-iommu @@ -26,6 +38,11 @@ v12.2.0 (unreleased) so that XTSup is required there. Additionally, it is automatically enabled if needed. + * qemu: Add statistics for ``<dataStore>`` storage + + The bulk statistics (``virsh domstats --block --backing``) now report also + information about the ``<dataStore>`` if given disk uses this feature. + * **Bug fixes** -- 2.53.0

3 2

[libvirt PATCH] qemu_hotplug: Set correct iommufdState on error path
by Pavel Hrdina 31 Mar '26

31 Mar '26

If hotplugging host device that needs to add iommufd object fails we need to remove the iommufd object as well. We also need to update the iommufdState otherwise hotplugging any host device with iommufd will fail for the same VM. Fixes: 5d16bef1a69500791be454239e3b9ac68ec53ace Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/qemu/qemu_hotplug.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index c86ebc59d0..9439948089 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1679,8 +1679,10 @@ qemuDomainAttachHostPCIDevice(virQEMUDriver *driver, if (virHostdevIsPCIDeviceWithIOMMUFD(hostdev)) { qemuDomainObjEnterMonitor(vm); - if (removeiommufd) + if (removeiommufd) { + priv->iommufdState = false; ignore_value(qemuMonitorDelObject(priv->mon, "iommufd0", false)); + } qemuFDPassDirectTransferMonitorRollback(hostdevPriv->vfioDeviceFd, priv->mon); qemuFDPassDirectTransferMonitorRollback(priv->iommufd, priv->mon); -- 2.53.0

3 2