August 2025 - Devel - libvirt List Archives

[PATCH 0/4] docs: Fix live backup kbase page and add support for RST notes/warnings to the CSS
by Peter Krempa 27 Aug '25

27 Aug '25

You can look at the result here: https://pipo.sk.gitlab.io/-/libvirt/-/jobs/11156282021/artifacts/website/kb… (Sorry for the TLS error, my username has a dot in it and the gitlab.io cert doesn't allow that. No I will not change my username) Peter Krempa (4): kbase: live_full_disk_backup: Fix use of '.. note' directive css: Add style for '.. note:' and '.. warning:' rST roles docs: formatsecret: Convert inline emphasis of a warning to the '.. warning:' role kbase: live_full_disk_backup: Improve the document docs/css/generic.css | 26 ++++++++++++++++++++++++++ docs/formatsecret.rst | 7 ++++--- docs/kbase/live_full_disk_backup.rst | 20 ++++++++++++-------- 3 files changed, 42 insertions(+), 11 deletions(-) -- 2.51.0

2 5

[PATCH v5 0/6] qemu: acpi-generic-initiator support
by Andrea Righi 27 Aug '25

27 Aug '25

= Overview = This patch set introduces support for acpi-generic-initiator devices, supported by QEMU [1]. The acpi-generic-initiator object is required to support Multi-Instance GPU (MIG) configurations on NVIDIA GPUs [2]. MIG enables partitioning of GPU resources into multiple isolated instances, each requiring a dedicated NUMA node definition. = Implementation = This patch set implements the libvirt counterpart to the QEMU feature, enabling users to configure acpi-generic-initiator objects within libvirt domain XML. This includes: - adding XML syntax to define acpi-generic-initiator objects, - resolving the acpi-generic-initiator definitions into the proper QEMU command-line arguments, - ensuring compatibility with existing NUMA configuration. = Example = - Domain XML: ``` ... <cpu mode='host-passthrough' check='none'> <numa> <cell id='0' cpus='0-15' memory='8388608' unit='KiB'/> <cell id='1' memory='0' unit='KiB'/> <cell id='2' memory='0' unit='KiB'/> <cell id='3' memory='0' unit='KiB'/> <cell id='4' memory='0' unit='KiB'/> <cell id='5' memory='0' unit='KiB'/> <cell id='6' memory='0' unit='KiB'/> <cell id='7' memory='0' unit='KiB'/> <cell id='8' memory='0' unit='KiB'/> </numa> </cpu> ... <devices> ... <hostdev mode='subsystem' type='pci' managed='no'> <source> <address domain='0x0009' bus='0x01' slot='0x00' function='0x0'/> </source> <address type='pci' domain='0x0000' bus='0x03' slot='0x00' function='0x0'/> </hostdev> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>1</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>2</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>3</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>4</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>5</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>6</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>7</numa-node> </acpi-generic-initiator> <acpi-generic-initiator> <pci-dev>hostdev0</pci-dev> <numa-node>8</numa-node> </acpi-generic-initiator> </devices> ``` - Generated QEMU command line options: ``` ... /usr/bin/qemu-system-aarch64 \ ... -object '{"qom-type":"memory-backend-ram","id":"ram-node0","size":8589934592}' \ -numa node,nodeid=0,cpus=0-15,memdev=ram-node0 \ -numa node,nodeid=1 \ -numa node,nodeid=2 \ -numa node,nodeid=3 \ -numa node,nodeid=4 \ -numa node,nodeid=5 \ -numa node,nodeid=6 \ -numa node,nodeid=7 \ -numa node,nodeid=8 \ ... -device '{"driver":"vfio-pci","host":"0009:01:00.0","id":"hostdev0","bus":"pci.3","addr":"0x0"}' ... -object acpi-generic-initiator,id=gi0,pci-dev=hostdev0,node=1 \ -object acpi-generic-initiator,id=gi1,pci-dev=hostdev0,node=2 \ -object acpi-generic-initiator,id=gi2,pci-dev=hostdev0,node=3 \ -object acpi-generic-initiator,id=gi3,pci-dev=hostdev0,node=4 \ -object acpi-generic-initiator,id=gi4,pci-dev=hostdev0,node=5 \ -object acpi-generic-initiator,id=gi5,pci-dev=hostdev0,node=6 \ -object acpi-generic-initiator,id=gi6,pci-dev=hostdev0,node=7 \ -object acpi-generic-initiator,id=gi7,pci-dev=hostdev0,node=8 ``` = References = [1] https://lore.kernel.org/all/20231225045603.7654-2-ankita@nvidia.com/ [2] https://www.nvidia.com/en-in/technologies/multi-instance-gpu/ ChangeLog v4 -> v5: - Integrate suggestions and changes from Michal's review - Update qemu capabilities - Rebase to v11.6.0 ChangeLog v3 -> v4: - add acpi-generic-initiator documentation - refactor virDomainAcpiInitiatorDef to use info->alias and drop the name attribute - auto-generate alias for the acpi-generic-initiator devices via qemuAssignDeviceAliases() - use g_autoptr() when possible - add a new entry to NEWS.rst ChangeLog v2 -> v3: - replaced <text/> with proper types in the XML schema - avoid mixing g_free() and VIR_FREE() - use virXMLPropString() instead of looping all XML nodes - report proper errors with virReportError() - use virBufferEscapeString() to process strings passed by the user - fix broken formatting of function headers - misc coding style fixes ChangeLog v1 -> v2: - split parser and driver changes in separate patches - introduce a new qemu capability flag - introduce test in qemuxmlconftest Andrea Righi (5): conf: Introduce acpi-generic-initiator device qemu: Allow to define NUMA nodes without memory or CPUs assigned qemu: capabilies: Introduce QEMU_CAPS_ACPI_GENERIC_INITIATOR qemu: Support acpi-generic-initiator NEWS: Mention new acpi-generic-initiator device Michal Prívozník (1): qemu_validate: Validate acpi-generic-initiator NEWS.rst | 8 ++ docs/formatdomain.rst | 36 ++++++ src/ch/ch_domain.c | 1 + src/conf/domain_conf.c | 138 +++++++++++++++++++++ src/conf/domain_conf.h | 14 +++ src/conf/domain_postparse.c | 1 + src/conf/domain_validate.c | 37 ++++++ src/conf/numa_conf.c | 3 + src/conf/schemas/domaincommon.rng | 14 +++ src/conf/virconftypes.h | 2 + src/hyperv/hyperv_driver.c | 1 + src/libxl/libxl_driver.c | 6 + src/lxc/lxc_driver.c | 6 + src/qemu/qemu_alias.c | 11 ++ src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 50 ++++++-- src/qemu/qemu_domain.c | 2 + src/qemu/qemu_domain_address.c | 4 + src/qemu/qemu_driver.c | 3 + src/qemu/qemu_hotplug.c | 5 + src/qemu/qemu_postparse.c | 1 + src/qemu/qemu_validate.c | 18 +++ src/test/test_driver.c | 4 + tests/qemucapabilitiesdata/caps_10.0.0_aarch64.xml | 1 + .../caps_10.0.0_x86_64+amdsev.xml | 1 + tests/qemucapabilitiesdata/caps_10.0.0_x86_64.xml | 1 + .../caps_10.1.0_x86_64+inteltdx.xml | 1 + tests/qemucapabilitiesdata/caps_10.1.0_x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_9.0.0_x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_9.1.0_riscv64.xml | 1 + tests/qemucapabilitiesdata/caps_9.1.0_x86_64.xml | 1 + .../caps_9.2.0_aarch64+hvf.xml | 1 + .../caps_9.2.0_x86_64+amdsev.xml | 1 + tests/qemucapabilitiesdata/caps_9.2.0_x86_64.xml | 1 + .../acpi-generic-initiator.x86_64-latest.args | 55 ++++++++ .../acpi-generic-initiator.x86_64-latest.xml | 1 + tests/qemuxmlconfdata/acpi-generic-initiator.xml | 94 ++++++++++++++ tests/qemuxmlconftest.c | 1 + 39 files changed, 523 insertions(+), 7 deletions(-) create mode 100644 tests/qemuxmlconfdata/acpi-generic-initiator.x86_64-latest.args create mode 120000 tests/qemuxmlconfdata/acpi-generic-initiator.x86_64-latest.xml create mode 100644 tests/qemuxmlconfdata/acpi-generic-initiator.xml

2 7

[PATCH 0/3] Fixes for 11.7
by Peter Krempa 27 Aug '25

27 Aug '25

Patch 1 fixes a segfault in one certain disk hot unplug case. Patch 2 moves decreases the log verbosity VIR_ERR_NO_SUPPORT Patch 3 removes some dead code noticed while browsing the code. All ought to be safe for the release. Peter Krempa (3): qemu: hotplug: Audit device detach before deleting it daemon: Drop log level of VIR_ERR_NO_SUPPORT to debug qemu: monitor: Remove wrappers for 'block-dirty-bitmap-enable'/'block-dirty-bitmap-disable' src/qemu/qemu_hotplug.c | 7 +++++-- src/qemu/qemu_monitor.c | 18 ------------------ src/qemu/qemu_monitor.h | 8 -------- src/qemu/qemu_monitor_json.c | 26 -------------------------- src/remote/remote_daemon.c | 1 + tests/qemumonitorjsontest.c | 2 -- 6 files changed, 6 insertions(+), 56 deletions(-) -- 2.51.0

3 5

[PATCH] NEWS: document bhyve changes for the release
by Roman Bogorodskiy 27 Aug '25

27 Aug '25

Signed-off-by: Roman Bogorodskiy <bogorodskiy(a)gmail.com> --- NEWS.rst | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index e6a74ad699..06cbd8d6c5 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -22,8 +22,23 @@ v11.7.0 (unreleased) Users can now configure the verbosity of Cloud Hypervisor by setting the "log_level" option in ch.conf + * bhyve: experimental NAT networking support + + The bhyve driver now has experimental NAT networking support + using the Packet Filter (pf) firewall. + + * bhyve: domain statistics reporting + + The bhyve driver now supports querying domain block, interface, + and memory statistics. Not all statistics fields are supported though. + * **Improvements** + * bhyve: improve 'efi' configuration autofill + + When a domain is configured with ``<os firmware='efi'/>``, NVRAM configuration + is now autofilled. + * **Bug fixes** -- 2.51.0

2 1

[PATCH 0/2] datatypes: Add missing check to virGetStream and refactor cleanup paths of helpers
by Peter Krempa 27 Aug '25

27 Aug '25

Peter Krempa (2): datatypes: virGetStream: Add missing 'virCheckConnectReturn' check datatypes: Refactor rest of 'virGet*' helpers src/datatypes.c | 175 +++++++++++++----------------------------------- 1 file changed, 45 insertions(+), 130 deletions(-) -- 2.51.0

1 2

[PATCH 0/3] A couple improvements when auto-adding PCI controllers
by Laine Stump 27 Aug '25

27 Aug '25

The 1st patch fixes a tiny omission from a feature added several years ago, which slightly improves the error message when you attempt to add a pcie-to-pci-bridge controller to a domain that has no PCIe bus. The 2nd improves that error message a bit more, along with improving some comments in the code. The 3rd eliminates said error message completely when you're adding a pcie-to-pci-bridge controller to a domain that *does* support PCIe, but doesn't currently have an open pcie-root-port to plug in the pcie-to-pci-bridge (by auto-adding another pcie-root-port). The impetus for these was https://issues.redhat.com/browse/RHEL-62032 Laine Stump (3): conf: add forgotten clause to virDomainPCIControllerConectTypeToModel() conf: improve error message when a PCI controller can't be auto-added conf: auto-add a pcie-root-port when needed while plugging in pcie-to-pci-bridge src/conf/domain_addr.c | 43 +++++++++++++++---- .../pcie-root-port-too-many.x86_64-latest.err | 2 +- 2 files changed, 35 insertions(+), 10 deletions(-) -- 2.51.0

1 3

[PATCH] qemu: support riscv-aia feature for RISC-V KVM
by BillXiang 25 Aug '25

25 Aug '25

riscv-aia feature was introduced in qemu-8.2 to specify the KVM AIA mode. The "riscv-aia" parameter is passed along with --accel in QEMU command-line. 1) "riscv-aia=emul": IMSIC is emulated by hypervisor 2) "riscv-aia=hwaccel": use hardware guest IMSIC 3) "riscv-aia=auto": use the hardware guest IMSICs whenever available otherwise we fallback to software emulation. This patch add the corresponding feature named 'riscv-aia'. Signed-off-by: BillXiang <xiangwencheng(a)lanxincomputing.com> --- NEWS.rst | 10 ++++++++ docs/formatdomain.rst | 2 ++ src/conf/domain_conf.c | 29 ++++++++++++++++++++++ src/conf/domain_conf.h | 2 ++ src/conf/schemas/domaincommon.rng | 12 +++++++++ src/qemu/qemu_command.c | 13 +++++++--- tests/qemuxmlconfdata/kvm-features-off.xml | 1 + tests/qemuxmlconfdata/kvm-features.xml | 1 + 8 files changed, 67 insertions(+), 3 deletions(-) diff --git a/NEWS.rst b/NEWS.rst index 435760e797..7e91f81d64 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -17,6 +17,16 @@ v11.7.0 (unreleased) * **New features** + * Support riscv-aia feature for RISC-V KVM + + riscv-aia feature was introduced in qemu-8.2 to specify the + KVM AIA mode. The "riscv-aia" parameter is passed along with + --accel in QEMU command-line. + 1) "riscv-aia=emul": IMSIC is emulated by hypervisor + 2) "riscv-aia=hwaccel": use hardware guest IMSIC + 3) "riscv-aia=auto": use the hardware guest IMSICs whenever available + otherwise we fallback to software emulation. + * **Improvements** * **Bug fixes** diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 9f7311b6d5..2ccbb1bfb9 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -2085,6 +2085,7 @@ Hypervisors may allow certain CPU / machine features to be toggled on/off. <poll-control state='on'/> <pv-ipi state='off'/> <dirty-ring state='on' size='4096'/> + <riscv-aia state='on' mode='auto'/> </kvm> <xen> <e820_host state='on'/> @@ -2206,6 +2207,7 @@ are: poll-control Decrease IO completion latency by introducing a grace period of busy waiting on, off :since:`6.10.0 (QEMU 4.2)` pv-ipi Paravirtualized send IPIs on, off :since:`7.10.0 (QEMU 3.1)` dirty-ring Enable dirty ring feature on, off; size - must be power of 2, range [1024,65536] :since:`8.0.0 (QEMU 6.1)` + riscv-aia Set riscv KVM AIA mode. Defaults to 'auto'. on, off; mode - optional string emul, hwaccel or auto :since:`11.7.0 (QEMU 8.2)` ============== ============================================================================ ====================================================== ============================ ``xen`` diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 7766e302ec..8ed65d1794 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -224,6 +224,7 @@ VIR_ENUM_IMPL(virDomainKVM, "poll-control", "pv-ipi", "dirty-ring", + "riscv-aia", ); VIR_ENUM_IMPL(virDomainXen, @@ -17183,6 +17184,20 @@ virDomainFeaturesKVMDefParse(virDomainDef *def, return -1; } } + + if (feature == VIR_DOMAIN_KVM_RISCV_AIA && + value == VIR_TRISTATE_SWITCH_ON) { + def->kvm_features->riscv_aia_mode = virXMLPropString(feat, "mode"); + + if (strcmp(def->kvm_features->riscv_aia_mode, "emul") && + strcmp(def->kvm_features->riscv_aia_mode, "hwaccel") && + strcmp(def->kvm_features->riscv_aia_mode, "auto")) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("riscv aia must be emul, hwaccel or auto")); + + return -1; + } + } } def->features[VIR_DOMAIN_FEATURE_KVM] = VIR_TRISTATE_SWITCH_ON; @@ -21697,6 +21712,7 @@ virDomainDefFeaturesCheckABIStability(virDomainDef *src, case VIR_DOMAIN_KVM_POLLCONTROL: case VIR_DOMAIN_KVM_PVIPI: case VIR_DOMAIN_KVM_DIRTY_RING: + case VIR_DOMAIN_KVM_RISCV_AIA: if (src->kvm_features->features[i] != dst->kvm_features->features[i]) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("State of KVM feature '%1$s' differs: source: '%2$s', destination: '%3$s'"), @@ -28752,6 +28768,19 @@ virDomainDefFormatFeatures(virBuffer *buf, } } break; + case VIR_DOMAIN_KVM_RISCV_AIA: + if (def->kvm_features->features[j] != VIR_TRISTATE_SWITCH_ABSENT) { + virBufferAsprintf(&childBuf, "<%s state='%s'", + virDomainKVMTypeToString(j), + virTristateSwitchTypeToString(def->kvm_features->features[j])); + if (def->kvm_features->riscv_aia_mode != NULL) { + virBufferAsprintf(&childBuf, " mode='%s'/>\n", + def->kvm_features->riscv_aia_mode); + } else { + virBufferAddLit(&childBuf, "/>\n"); + } + } + break; case VIR_DOMAIN_KVM_LAST: break; diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index eca820892e..0d749c0f3c 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2270,6 +2270,7 @@ typedef enum { VIR_DOMAIN_KVM_POLLCONTROL, VIR_DOMAIN_KVM_PVIPI, VIR_DOMAIN_KVM_DIRTY_RING, + VIR_DOMAIN_KVM_RISCV_AIA, VIR_DOMAIN_KVM_LAST } virDomainKVM; @@ -2476,6 +2477,7 @@ struct _virDomainFeatureKVM { int features[VIR_DOMAIN_KVM_LAST]; unsigned int dirty_ring_size; /* size of dirty ring for each vCPU, no units */ + char *riscv_aia_mode; }; typedef struct _virDomainFeatureTCG virDomainFeatureTCG; diff --git a/src/conf/schemas/domaincommon.rng b/src/conf/schemas/domaincommon.rng index e369fb6e81..3de5807b1e 100644 --- a/src/conf/schemas/domaincommon.rng +++ b/src/conf/schemas/domaincommon.rng @@ -8192,6 +8192,18 @@ </optional> </element> </optional> + <optional> + <element name="riscv-aia"> + <ref name="featurestate"/> + <optional> + <attribute name="mode"> + <data type="string"> + <param name="pattern">(emul|hwaccel|auto)</param> + </data> + </attribute> + </optional> + </element> + </optional> </interleave> </element> </define> diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index e8de386f30..6db283ef29 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6679,6 +6679,9 @@ qemuBuildCpuCommandLine(virCommand *cmd, case VIR_DOMAIN_KVM_DIRTY_RING: break; + + case VIR_DOMAIN_KVM_RISCV_AIA: + break; case VIR_DOMAIN_KVM_LAST: break; @@ -7314,9 +7317,13 @@ qemuBuildAccelCommandLine(virCommand *cmd, * not that either kvm or tcg can be specified by libvirt * so do not worry about the conflict of specifying both * */ - if (def->features[VIR_DOMAIN_FEATURE_KVM] == VIR_TRISTATE_SWITCH_ON && - def->kvm_features->features[VIR_DOMAIN_KVM_DIRTY_RING] == VIR_TRISTATE_SWITCH_ON) { - virBufferAsprintf(&buf, ",dirty-ring-size=%d", def->kvm_features->dirty_ring_size); + if (def->features[VIR_DOMAIN_FEATURE_KVM] == VIR_TRISTATE_SWITCH_ON) { + if (def->kvm_features->features[VIR_DOMAIN_KVM_DIRTY_RING] == VIR_TRISTATE_SWITCH_ON) { + virBufferAsprintf(&buf, ",dirty-ring-size=%d", def->kvm_features->dirty_ring_size); + } + if (def->kvm_features->features[VIR_DOMAIN_KVM_RISCV_AIA] == VIR_TRISTATE_SWITCH_ON) { + virBufferAsprintf(&buf, ",riscv-aia=%s", def->kvm_features->riscv_aia_mode); + } } break; diff --git a/tests/qemuxmlconfdata/kvm-features-off.xml b/tests/qemuxmlconfdata/kvm-features-off.xml index 3cd4ff18f2..2340a93ce5 100644 --- a/tests/qemuxmlconfdata/kvm-features-off.xml +++ b/tests/qemuxmlconfdata/kvm-features-off.xml @@ -16,6 +16,7 @@ <poll-control state='off'/> <pv-ipi state='off'/> <dirty-ring state='off'/> + <riscv-aia state='off'/> </kvm> </features> <cpu mode='host-passthrough' check='none' migratable='off'/> diff --git a/tests/qemuxmlconfdata/kvm-features.xml b/tests/qemuxmlconfdata/kvm-features.xml index 78091064b1..c06b0aa58b 100644 --- a/tests/qemuxmlconfdata/kvm-features.xml +++ b/tests/qemuxmlconfdata/kvm-features.xml @@ -16,6 +16,7 @@ <poll-control state='on'/> <pv-ipi state='on'/> <dirty-ring state='on' size='4096'/> + <riscv-aia state='on' mode='auto'/> </kvm> </features> <cpu mode='host-passthrough' check='none' migratable='off'/> -- 2.46.2.windows.1

3 4

[PATCH v2 1/2] qemu: support riscv-aia feature for RISC-V KVM
by BillXiang 25 Aug '25

25 Aug '25

From: xiangwencheng <xiangwencheng(a)lanxincomputing.com> riscv-aia feature was introduced in qemu-8.2 to specify the KVM AIA mode. The "riscv-aia" parameter is passed along with -accel in QEMU command-line. 1) "riscv-aia=emul": IMSIC is emulated by hypervisor 2) "riscv-aia=hwaccel": use hardware guest IMSIC 3) "riscv-aia=auto": use the hardware guest IMSICs whenever available otherwise we fallback to software emulation. This patch add the corresponding feature named 'riscv-aia'. Signed-off-by: BillXiang <xiangwencheng(a)lanxincomputing.com> --- v1[1] -> v2: - Separate changes to NEWS.rst into a separate commit. - Remove changes to qemuxmlconfdata files that are used by x86. - Store riscv-aia mode as enum. - Remove VIR_DOMAIN_KVM_RISCV_AIA from virDomainDefFeaturesCheckABIStability beacuse it doesn't support migration currently. [1] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/XBDT… --- docs/formatdomain.rst | 2 ++ src/conf/domain_conf.c | 33 ++++++++++++++++++++++++++++++- src/conf/domain_conf.h | 10 ++++++++++ src/conf/schemas/domaincommon.rng | 12 +++++++++++ src/libvirt_private.syms | 2 ++ src/qemu/qemu_command.c | 13 +++++++++--- 6 files changed, 68 insertions(+), 4 deletions(-) diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 9f7311b6d5..2ccbb1bfb9 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -2085,6 +2085,7 @@ Hypervisors may allow certain CPU / machine features to be toggled on/off. <poll-control state='on'/> <pv-ipi state='off'/> <dirty-ring state='on' size='4096'/> + <riscv-aia state='on' mode='auto'/> </kvm> <xen> <e820_host state='on'/> @@ -2206,6 +2207,7 @@ are: poll-control Decrease IO completion latency by introducing a grace period of busy waiting on, off :since:`6.10.0 (QEMU 4.2)` pv-ipi Paravirtualized send IPIs on, off :since:`7.10.0 (QEMU 3.1)` dirty-ring Enable dirty ring feature on, off; size - must be power of 2, range [1024,65536] :since:`8.0.0 (QEMU 6.1)` + riscv-aia Set riscv KVM AIA mode. Defaults to 'auto'. on, off; mode - optional string emul, hwaccel or auto :since:`11.7.0 (QEMU 8.2)` ============== ============================================================================ ====================================================== ============================ ``xen`` diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 7766e302ec..e109c74658 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -224,6 +224,14 @@ VIR_ENUM_IMPL(virDomainKVM, "poll-control", "pv-ipi", "dirty-ring", + "riscv-aia", +); + +VIR_ENUM_IMPL(virDomainKVMRiscvAIAMode, + VIR_DOMAIN_KVM_TISCV_AIA_MODE_LAST, + "emul", + "hwaccel", + "auto", ); VIR_ENUM_IMPL(virDomainXen, @@ -17142,6 +17150,7 @@ virDomainFeaturesKVMDefParse(virDomainDef *def, { g_autoptr(GPtrArray) feats = virXMLNodeGetSubelementList(node, NULL); size_t i; + g_autofree char *ptval = NULL; if (!def->kvm_features) def->kvm_features = g_new0(virDomainFeatureKVM, 1); @@ -17183,6 +17192,15 @@ virDomainFeaturesKVMDefParse(virDomainDef *def, return -1; } } + + if (feature == VIR_DOMAIN_KVM_RISCV_AIA && + value == VIR_TRISTATE_SWITCH_ON) { + if (virXMLPropEnum(feat, "mode", + virDomainKVMRiscvAIAModeTypeFromString, + VIR_XML_PROP_NONZERO, + &def->kvm_features->kvm_riscv_aia_mode) < 0) + return -1; + } } def->features[VIR_DOMAIN_FEATURE_KVM] = VIR_TRISTATE_SWITCH_ON; @@ -28752,7 +28770,20 @@ virDomainDefFormatFeatures(virBuffer *buf, } } break; - + case VIR_DOMAIN_KVM_RISCV_AIA: + if (def->kvm_features->features[j] != VIR_TRISTATE_SWITCH_ABSENT) { + virBufferAsprintf(&childBuf, "<%s state='%s'", + virDomainKVMTypeToString(j), + virTristateSwitchTypeToString(def->kvm_features->features[j])); + if (def->kvm_features->kvm_riscv_aia_mode >= 0 && + def->kvm_features->kvm_riscv_aia_mode < VIR_DOMAIN_KVM_TISCV_AIA_MODE_LAST) { + virBufferAsprintf(&childBuf, " mode='%s'/>\n", + virDomainKVMRiscvAIAModeTypeToString(def->kvm_features->kvm_riscv_aia_mode)); + } else { + virBufferAddLit(&childBuf, "/>\n"); + } + } + break; case VIR_DOMAIN_KVM_LAST: break; } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index eca820892e..512c6dfc06 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2270,10 +2270,19 @@ typedef enum { VIR_DOMAIN_KVM_POLLCONTROL, VIR_DOMAIN_KVM_PVIPI, VIR_DOMAIN_KVM_DIRTY_RING, + VIR_DOMAIN_KVM_RISCV_AIA, VIR_DOMAIN_KVM_LAST } virDomainKVM; +typedef enum { + VIR_DOMAIN_KVM_TISCV_AIA_MODE_EMUL = 0, + VIR_DOMAIN_KVM_TISCV_AIA_MODE_HWACCEL, + VIR_DOMAIN_KVM_TISCV_AIA_MODE_AUTO, + + VIR_DOMAIN_KVM_TISCV_AIA_MODE_LAST +} virDomainKVMRiscvAIAMode; + typedef enum { VIR_DOMAIN_MSRS_UNKNOWN = 0, @@ -2476,6 +2485,7 @@ struct _virDomainFeatureKVM { int features[VIR_DOMAIN_KVM_LAST]; unsigned int dirty_ring_size; /* size of dirty ring for each vCPU, no units */ + virDomainKVMRiscvAIAMode kvm_riscv_aia_mode; }; typedef struct _virDomainFeatureTCG virDomainFeatureTCG; diff --git a/src/conf/schemas/domaincommon.rng b/src/conf/schemas/domaincommon.rng index e369fb6e81..3de5807b1e 100644 --- a/src/conf/schemas/domaincommon.rng +++ b/src/conf/schemas/domaincommon.rng @@ -8192,6 +8192,18 @@ </optional> </element> </optional> + <optional> + <element name="riscv-aia"> + <ref name="featurestate"/> + <optional> + <attribute name="mode"> + <data type="string"> + <param name="pattern">(emul|hwaccel|auto)</param> + </data> + </attribute> + </optional> + </element> + </optional> </interleave> </element> </define> diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index b846011f0f..f540bdbd5b 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -501,6 +501,8 @@ virDomainIOThreadIDDel; virDomainIOThreadIDFind; virDomainKeyWrapCipherNameTypeFromString; virDomainKeyWrapCipherNameTypeToString; +virDomainKVMRiscvAIAModeTypeFromString; +virDomainKVMRiscvAIAModeTypeToString; virDomainLaunchSecurityTypeFromString; virDomainLaunchSecurityTypeToString; virDomainLeaseDefFree; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index e8de386f30..9f34b5a37d 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6679,6 +6679,9 @@ qemuBuildCpuCommandLine(virCommand *cmd, case VIR_DOMAIN_KVM_DIRTY_RING: break; + + case VIR_DOMAIN_KVM_RISCV_AIA: + break; case VIR_DOMAIN_KVM_LAST: break; @@ -7314,9 +7317,13 @@ qemuBuildAccelCommandLine(virCommand *cmd, * not that either kvm or tcg can be specified by libvirt * so do not worry about the conflict of specifying both * */ - if (def->features[VIR_DOMAIN_FEATURE_KVM] == VIR_TRISTATE_SWITCH_ON && - def->kvm_features->features[VIR_DOMAIN_KVM_DIRTY_RING] == VIR_TRISTATE_SWITCH_ON) { - virBufferAsprintf(&buf, ",dirty-ring-size=%d", def->kvm_features->dirty_ring_size); + if (def->features[VIR_DOMAIN_FEATURE_KVM] == VIR_TRISTATE_SWITCH_ON) { + if (def->kvm_features->features[VIR_DOMAIN_KVM_DIRTY_RING] == VIR_TRISTATE_SWITCH_ON) { + virBufferAsprintf(&buf, ",dirty-ring-size=%d", def->kvm_features->dirty_ring_size); + } + if (def->kvm_features->features[VIR_DOMAIN_KVM_RISCV_AIA] == VIR_TRISTATE_SWITCH_ON) { + virBufferAsprintf(&buf, ",riscv-aia=%s", virDomainKVMRiscvAIAModeTypeToString(def->kvm_features->kvm_riscv_aia_mode)); + } } break; -- 2.34.1

2 3

[PATCH] qemu: fix multiple missing setup/teardown of passt process for interface type='vhostuser'
by Laine Stump 25 Aug '25

25 Aug '25

passt networking support was originally added only for <interface type='user'>, and all of the codepaths leading to qemuPasst*() functions were protected with if (net->type == VIR_DOMAIN_NET_TYPE_USER && net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) When support was later added to use a vhost-user socket to connect between the passt process and qemu process, *some* of the conditionals similar to the above were changed to be if ((net->type == VIR_DOMAIN_NET_TYPE_USER || net->type == VIR_DOMAIN_NET_TYPE_VHOSTUSER) && net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) As a matter of fact, enough of these places were changed to make passt+vhostuser work. However I missed a few places that resulted in the passt process not being properly shutdown/cleaned up when the interface type was vhostuser, and also as far as I can see from examining the code, the passt process wasn't being added to the cgroup for the domain. We could fix these problems by adding the extra condition to all the missing places (checking for either 'user' or 'vhostuser' as well as for backend type of 'passt'), but since validation already guarantees that if backend type='passt' then the interface type MUST be either 'user' or 'vhostuser', it's really just adding extra code for no good purpose (and would leave open the possibility of the same problem recurring in the future if a different interface type begins using passt as well). So the better solution is to not bother checking net->type at all in those locations - if backend type is 'passt' then we call the passt-related code. Resolves: https://issues.redhat.com/browse/RHEL-80285 Resolves: https://issues.redhat.com/browse/RHEL-92842 Fixes: 1e9054b9c79d721a55f413c2983c5370044f8f60 Signed-off-by: Laine Stump <laine(a)redhat.com> --- src/qemu/qemu_extdevice.c | 15 +++------------ src/qemu/qemu_hotplug.c | 23 ++++++++--------------- 2 files changed, 11 insertions(+), 27 deletions(-) diff --git a/src/qemu/qemu_extdevice.c b/src/qemu/qemu_extdevice.c index 8df93a77ce..28cea52980 100644 --- a/src/qemu/qemu_extdevice.c +++ b/src/qemu/qemu_extdevice.c @@ -213,11 +213,6 @@ qemuExtDevicesStart(virQEMUDriver *driver, for (i = 0; i < def->nnets; i++) { virDomainNetDef *net = def->nets[i]; - if (net->type != VIR_DOMAIN_NET_TYPE_USER && - net->type != VIR_DOMAIN_NET_TYPE_VHOSTUSER) { - continue; - } - if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) { if (qemuPasstStart(vm, net) < 0) return -1; @@ -310,10 +305,8 @@ qemuExtDevicesStop(virQEMUDriver *driver, if (slirp) qemuSlirpStop(slirp, vm, driver, net); - if (net->type == VIR_DOMAIN_NET_TYPE_USER && - net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) { + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) qemuPasstStop(vm, net); - } if (actualType == VIR_DOMAIN_NET_TYPE_ETHERNET && net->downscript) virNetDevRunEthernetScript(net->ifname, net->downscript); @@ -373,8 +366,7 @@ qemuExtDevicesHasDevice(virDomainDef *def) if (QEMU_DOMAIN_NETWORK_PRIVATE(net)->slirp) return true; - if (net->type == VIR_DOMAIN_NET_TYPE_USER && - net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) return true; } @@ -455,8 +447,7 @@ qemuExtDevicesSetupCgroup(virQEMUDriver *driver, if (slirp && qemuSlirpSetupCgroup(slirp, cgroup) < 0) return -1; - if (net->type == VIR_DOMAIN_NET_TYPE_USER && - net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST && + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST && qemuPasstSetupCgroup(vm, net, cgroup) < 0) { return -1; } diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index e9568af125..866dbace90 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1298,8 +1298,6 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver, qemuPasstPrepareVhostUser(vm, net); - if (qemuPasstStart(vm, net) < 0) - goto cleanup; } else { if (virNetDevOpenvswitchGetVhostuserIfname(net->data.vhostuser->data.nix.path, net->data.vhostuser->data.nix.listen, @@ -1313,13 +1311,8 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver, break; case VIR_DOMAIN_NET_TYPE_USER: - if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) { - - if (qemuPasstStart(vm, net) < 0) - goto cleanup; - - } else if (!priv->disableSlirp && - virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_DBUS_VMSTATE)) { + if (!priv->disableSlirp && + virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_DBUS_VMSTATE)) { if (qemuInterfacePrepareSlirp(driver, net) < 0) goto cleanup; @@ -1356,6 +1349,10 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver, goto cleanup; } + /* both NET_TYPE_USER and NET_TYPE_VHOSTUSER might use passt */ + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST && qemuPasstStart(vm, net) < 0) + goto cleanup; + /* Set device online immediately */ if (virDomainInterfaceStartDevice(net) < 0) goto cleanup; @@ -1539,10 +1536,8 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver, if (QEMU_DOMAIN_NETWORK_PRIVATE(net)->slirp) qemuSlirpStop(QEMU_DOMAIN_NETWORK_PRIVATE(net)->slirp, vm, driver, net); - if (net->type == VIR_DOMAIN_NET_TYPE_USER && - net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) { + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) qemuPasstStop(vm, net); - } qemuDomainObjEnterMonitor(vm); if (charDevPlugged && @@ -5112,10 +5107,8 @@ qemuDomainRemoveNetDevice(virQEMUDriver *driver, if (QEMU_DOMAIN_NETWORK_PRIVATE(net)->slirp) qemuSlirpStop(QEMU_DOMAIN_NETWORK_PRIVATE(net)->slirp, vm, driver, net); - if (net->type == VIR_DOMAIN_NET_TYPE_USER && - net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) { + if (net->backend.type == VIR_DOMAIN_NET_BACKEND_PASST) qemuPasstStop(vm, net); - } virDomainAuditNet(vm, net, NULL, "detach", true); -- 2.51.0

2 1

[PATCH v2 2/2] NEWS: ch: announce riscv-aia feature for RISC-V KVM
by BillXiang 22 Aug '25

22 Aug '25

From: xiangwencheng <xiangwencheng(a)lanxincomputing.com> Signed-off-by: BillXiang <xiangwencheng(a)lanxincomputing.com> --- NEWS.rst | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index e6a74ad699..b3f709c172 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -22,6 +22,16 @@ v11.7.0 (unreleased) Users can now configure the verbosity of Cloud Hypervisor by setting the "log_level" option in ch.conf + * Support riscv-aia feature for RISC-V KVM + + riscv-aia feature was introduced in qemu-8.2 to specify the + KVM AIA mode. The "riscv-aia" parameter is passed along with + --accel in QEMU command-line. + 1) "riscv-aia=emul": IMSIC is emulated by hypervisor + 2) "riscv-aia=hwaccel": use hardware guest IMSIC + 3) "riscv-aia=auto": use the hardware guest IMSICs whenever available + otherwise we fallback to software emulation. + * **Improvements** * **Bug fixes** -- 2.34.1

1 0