June 2026 - Devel - libvirt List Archives

[PATCH 00/20] qemu: Fix virtio video devcice frontend selection logic for migrations
by Peter Krempa 16 Jun '26

16 Jun '26

TL;DR: We've historically made selection of the specific virtio frontend device based on qemu capabilities but didn't record the choice. Migrations break if your qemu deployment differs. For more in-depth analysis see 16/20. This series conists of: -- refactor of removing detected qemu capabilities via config or the QMEU namespace -- fix to capability detection if host has no qemu -- addition of testing covering various possibilities for the virtio video devices which show specifics -- addition of XML infra for the new property holding the data -- refactors of the virtio selection code -- selecting the model in post parse callbacks -- fixing of validation code -- fixing of commandline generator Peter Krempa (20): qemu: capabilities: Apply 'capability_filters' configration option on all capabilities qemu: capabilities: Export 'virQEMUCapsNewCopy' outside of 'qemu_capspriv' qemu: Allow reuse of 'qemuProcessStartUpdateCustomCaps' qemu: validate: Validate VM config with qemuCaps influenced by <qemu:capabilities> qemu: postparse: Process VM config with qemuCaps influenced by <qemu:capabilities> virQEMUCapsCacheLookupDefault: Fix error message when no emulators are installed qemuxmlconfdata: un-symlink 'video-virtio-vga-gpu-gl' output qemuxmlconftest: Add 'video-virtio-vga' invocation with QEMU_CAPS_DEVICE_VIRTIO_VGA disabled qemuxmlconftest: Add test cases for configs asking for 'virtio-gpu-gl' or 'virtio-vga-gl' without the capability qemuxmlconftest: Add invocation of 'video-virtio-vga-gpu-gl' with missing caps and VIR_DOMAIN_DEF_PARSE_ABI_UPDATE qemustatusxml2xml: Add test case capturing virtio video device virDomainVideoDefFormat: Use 'virXMLFormatElement' instead of custom formatter conf: Add fields for recording actually-selected virtio video device qemuxmlconftest: Add test case for specifying 'virtio-gpu' where 'virtio-vga' would be picked qemuDeviceVideoGetModel: Directly return picked model qemu: postparse: Fill in selected virtio video frondend device in the XML qemuValidateDomainDeviceDefVideo: Fix checks of virtio video devices qemuDeviceVideoGetModel: Remove logic for selecting 'virtio' devices qemuDeviceVideoGetModel: Simplify by relying on checks from 'qemuValidateDomainDeviceDefVideo' qemu: Remove 'qemuDomainSupportsVideoVga' docs/formatdomain.rst | 8 + src/conf/domain_conf.c | 114 +++-- src/conf/domain_conf.h | 15 + src/conf/schemas/domaincommon.rng | 18 +- src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 26 +- src/qemu/qemu_capabilities.h | 5 +- src/qemu/qemu_capspriv.h | 5 +- src/qemu/qemu_command.c | 163 +++--- src/qemu/qemu_domain.c | 96 +++- src/qemu/qemu_domain.h | 9 +- src/qemu/qemu_driver.c | 23 +- src/qemu/qemu_postparse.c | 108 +++- src/qemu/qemu_process.c | 55 +-- src/qemu/qemu_validate.c | 110 +++-- tests/qemucapsprobe.c | 3 +- .../virtio-vga-upgrade-in.xml | 466 ++++++++++++++++++ .../virtio-vga-upgrade-missing-in.xml | 464 +++++++++++++++++ .../virtio-vga-upgrade-missing-out.xml | 457 +++++++++++++++++ .../virtio-vga-upgrade-out.xml | 459 +++++++++++++++++ tests/qemuxmlactivetest.c | 6 + .../aarch64-video-default.aarch64-latest.xml | 2 +- ...64-video-virtio-gpu-pci.aarch64-latest.xml | 2 +- ...fault-models.aarch64-latest.abi-update.xml | 2 +- ...h64-virt-default-models.aarch64-latest.xml | 2 +- .../aarch64-virt-graphics.aarch64-latest.xml | 2 +- ...ault-video-type-aarch64.aarch64-latest.xml | 2 +- ...eo-type-loongarch64.loongarch64-latest.xml | 2 +- ...ault-video-type-riscv64.riscv64-latest.xml | 2 +- .../default-video-type-s390x.s390x-latest.xml | 2 +- ...t-models.loongarch64-latest.abi-update.xml | 2 +- ...virt-default-models.loongarch64-latest.xml | 2 +- ...rch64-virt-graphics.loongarch64-latest.xml | 2 +- .../q35-pcie-autoadd.x86_64-latest.xml | 2 +- .../q35-pcie.x86_64-latest.xml | 2 +- ...fault-models.riscv64-latest.abi-update.xml | 2 +- ...v64-virt-default-models.riscv64-latest.xml | 2 +- .../riscv64-virt-graphics.riscv64-latest.xml | 2 +- ...default-models.s390x-latest.abi-update.xml | 2 +- .../s390x-ccw-default-models.s390x-latest.xml | 2 +- .../s390x-ccw-graphics.s390x-latest.xml | 2 +- ...vhost-user-gpu-secondary.x86_64-latest.xml | 4 +- .../vhost-user-vga.x86_64-latest.xml | 2 +- .../video-virtio-blob-off.x86_64-latest.xml | 4 +- .../video-virtio-blob-on.x86_64-latest.xml | 4 +- .../video-virtio-edid-none.x86_64-latest.xml | 2 +- .../video-virtio-edid-off.x86_64-latest.xml | 2 +- .../video-virtio-edid-on.x86_64-latest.xml | 2 +- ...video-virtio-gpu-ccw-auto.s390x-latest.xml | 2 +- .../video-virtio-gpu-ccw.s390x-latest.xml | 4 +- .../video-virtio-gpu-device.x86_64-latest.xml | 2 +- .../video-virtio-gpu-sdl-gl.x86_64-latest.xml | 2 +- ...deo-virtio-gpu-secondary.x86_64-latest.xml | 4 +- ...ideo-virtio-gpu-spice-gl.x86_64-latest.xml | 2 +- .../video-virtio-gpu-virgl.x86_64-latest.xml | 2 +- ...io-vga-device-downgrade.x86_64-latest.args | 37 ++ ...tio-vga-device-downgrade.x86_64-latest.xml | 44 ++ .../video-virtio-vga-device-downgrade.xml | 33 ++ ..._VIRTIO_GPU_GL_PCI-disabled-ABI_UPDATE.err | 1 + ....QEMU_CAPS_VIRTIO_GPU_GL_PCI-disabled.args | 35 ++ ...t.QEMU_CAPS_VIRTIO_GPU_GL_PCI-disabled.xml | 46 ++ ...APS_VIRTIO_VGA_GL-disabled-ABI_UPDATE.args | 35 ++ ...CAPS_VIRTIO_VGA_GL-disabled-ABI_UPDATE.xml | 46 ++ ...test.QEMU_CAPS_VIRTIO_VGA_GL-disabled.args | 35 ++ ...atest.QEMU_CAPS_VIRTIO_VGA_GL-disabled.xml | 46 ++ .../video-virtio-vga-gpu-gl.x86_64-latest.xml | 47 +- ....QEMU_CAPS_DEVICE_VIRTIO_VGA-disabled.args | 37 ++ ...t.QEMU_CAPS_DEVICE_VIRTIO_VGA-disabled.xml | 44 ++ .../video-virtio-vga.x86_64-latest.xml | 2 +- ...virtio-options-video-ats.x86_64-latest.xml | 4 +- ...rtio-options-video-iommu.x86_64-latest.xml | 4 +- ...tio-options-video-packed.x86_64-latest.xml | 4 +- tests/qemuxmlconfdata/virtio-options.xml | 2 +- tests/qemuxmlconftest.c | 29 ++ tests/testutilsqemu.c | 3 +- 75 files changed, 2908 insertions(+), 315 deletions(-) create mode 100644 tests/qemustatusxml2xmldata/virtio-vga-upgrade-in.xml create mode 100644 tests/qemustatusxml2xmldata/virtio-vga-upgrade-missing-in.xml create mode 100644 tests/qemustatusxml2xmldata/virtio-vga-upgrade-missing-out.xml create mode 100644 tests/qemustatusxml2xmldata/virtio-vga-upgrade-out.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-device-downgrade.x86_64-latest.args create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-device-downgrade.x86_64-latest.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-device-downgrade.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_GPU_GL_PCI-disabled-ABI_UPDATE.err create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_GPU_GL_PCI-disabled.args create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_GPU_GL_PCI-disabled.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_VGA_GL-disabled-ABI_UPDATE.args create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_VGA_GL-disabled-ABI_UPDATE.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_VGA_GL-disabled.args create mode 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.QEMU_CAPS_VIRTIO_VGA_GL-disabled.xml mode change 120000 => 100644 tests/qemuxmlconfdata/video-virtio-vga-gpu-gl.x86_64-latest.xml create mode 100644 tests/qemuxmlconfdata/video-virtio-vga.x86_64-latest.QEMU_CAPS_DEVICE_VIRTIO_VGA-disabled.args create mode 100644 tests/qemuxmlconfdata/video-virtio-vga.x86_64-latest.QEMU_CAPS_DEVICE_VIRTIO_VGA-disabled.xml -- 2.54.0

2 21

[PATCH] bhyve: implement reboot using agent
by Roman Bogorodskiy 16 Jun '26

16 Jun '26

Implement domain reboot using the guest agent. Implementation is very similar to the domain shutdown added earlier. Also, change the VIR_DOMAIN_REBOOT_ACPI_POWER_BTN flag to VIR_DOMAIN_REBOOT_SIGNAL. Even though bhyve emulates the ACPI button, it's triggered by sending a signal to the bhyve process, so VIR_DOMAIN_REBOOT_SIGNAL looks like a more accurate description of communication between libvirt and bhyve. Signed-off-by: Roman Bogorodskiy <bogorodskiy(a)gmail.com> --- src/bhyve/bhyve_driver.c | 38 +++++++++++++++++++++++++++++++++++--- 1 file changed, 35 insertions(+), 3 deletions(-) diff --git a/src/bhyve/bhyve_driver.c b/src/bhyve/bhyve_driver.c index 4b28c698e7..45e8aad5b5 100644 --- a/src/bhyve/bhyve_driver.c +++ b/src/bhyve/bhyve_driver.c @@ -1085,6 +1085,12 @@ bhyveDomainShutdownFlagsAgent(virDomainObj *vm, return ret; } +static int +bhyveDomainRebootAgent(virDomainObj *vm, bool isReboot, bool reportError) +{ + return bhyveDomainShutdownFlagsAgent(vm, isReboot, reportError); +} + static int bhyveDomainShutdownFlags(virDomainPtr dom, unsigned int flags) { @@ -1160,10 +1166,15 @@ bhyveDomainReboot(virDomainPtr dom, unsigned int flags) { virConnectPtr conn = dom->conn; virDomainObj *vm; - bool isReboot = true; + bhyveDomainObjPrivate *priv; int ret = -1; + bool isReboot = true; + bool useAgent = false; + bool agentRequested, signalRequested; + bool agentForced; - virCheckFlags(VIR_DOMAIN_REBOOT_ACPI_POWER_BTN, -1); + virCheckFlags(VIR_DOMAIN_REBOOT_SIGNAL | + VIR_DOMAIN_REBOOT_GUEST_AGENT, -1); if (!(vm = bhyveDomObjFromDomain(dom))) goto cleanup; @@ -1174,13 +1185,34 @@ bhyveDomainReboot(virDomainPtr dom, unsigned int flags) VIR_INFO("Domain on_reboot setting overridden, shutting down"); } + priv = vm->privateData; + agentRequested = flags & VIR_DOMAIN_REBOOT_GUEST_AGENT; + signalRequested = flags & VIR_DOMAIN_REBOOT_SIGNAL; + + /* Prefer agent unless we were requested to not to. */ + if (agentRequested || !flags) + useAgent = true; + if (virDomainRebootEnsureACL(conn, vm->def, flags) < 0) goto cleanup; if (virDomainObjCheckActive(vm) < 0) goto cleanup; - ret = bhyveDomainShutdownSignal(vm, isReboot); + agentForced = agentRequested && !signalRequested; + if (useAgent) { + ret = bhyveDomainRebootAgent(vm, isReboot, agentForced); + if (((ret < 0) || (priv->agent != NULL)) && agentForced) + goto cleanup; + } + + /* If we are not enforced to use just an agent, try signal + * reboot as well in case agent did not succeed. + */ + if (!useAgent || (((ret < 0) || + (priv->agent != NULL)) && (signalRequested || !flags))) { + ret = bhyveDomainShutdownSignal(vm, isReboot); + } cleanup: virDomainObjEndAPI(&vm); -- 2.52.0

2 5

[PATCH 1/1] qemu: allow read-only HMP info monitor commands
by Siddhi Katage 15 Jun '26

15 Jun '26

Read-only connections currently cannot use qemu-monitor-command even for HMP "info" commands, which only query QEMU state. Allow HMP commands whose first token is exactly "info" on read-only connections. Keep QMP commands, file descriptor passing, and all other HMP commands restricted to read-write connections. Signed-off-by: Siddhi Katage <siddhi.katage(a)oracle.com> --- src/libvirt-qemu.c | 13 +++++++++++-- src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 21 +++++++++++++++++++-- src/util/virqemu.c | 21 +++++++++++++++++++++ src/util/virqemu.h | 2 ++ 5 files changed, 54 insertions(+), 4 deletions(-) diff --git a/src/libvirt-qemu.c b/src/libvirt-qemu.c index 2898b71983..d38b747b44 100644 --- a/src/libvirt-qemu.c +++ b/src/libvirt-qemu.c @@ -25,6 +25,7 @@ #include "virlog.h" #include "viruuid.h" #include "datatypes.h" +#include "virqemu.h" #define VIR_FROM_THIS VIR_FROM_NONE @@ -77,8 +78,12 @@ virDomainQemuMonitorCommand(virDomainPtr domain, const char *cmd, virCheckDomainReturn(domain, -1); conn = domain->conn; + virCheckNonNullArgGoto(cmd, error); virCheckNonNullArgGoto(result, error); - virCheckReadOnlyGoto(conn->flags, error); + + if (!(flags & VIR_DOMAIN_QEMU_MONITOR_COMMAND_HMP) || + !virQEMUHMPCommandIsInfo(cmd)) + virCheckReadOnlyGoto(conn->flags, error); if (conn->driver->domainQemuMonitorCommand) { int ret; @@ -170,7 +175,11 @@ virDomainQemuMonitorCommandWithFiles(virDomainPtr domain, } virCheckNonNullArgGoto(result, error); - virCheckReadOnlyGoto(conn->flags, error); + + if (ninfiles > 0 || outfiles || + !(flags & VIR_DOMAIN_QEMU_MONITOR_COMMAND_HMP) || + !virQEMUHMPCommandIsInfo(cmd)) + virCheckReadOnlyGoto(conn->flags, error); if (conn->driver->domainQemuMonitorCommandWithFiles) { int ret; diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 30c4564456..6f7505ecde 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -3351,6 +3351,7 @@ virQEMUBuildBufferEscapeComma; virQEMUBuildCommandLineJSON; virQEMUBuildCommandLineJSONArrayNumbered; virQEMUBuildCommandLineJSONArrayObjectsStr; +virQEMUHMPCommandIsInfo; # util/virrandom.h diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 544955ecf9..2f542e0b8a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -100,6 +100,7 @@ #include "vircgroup.h" #include "virperf.h" #include "virnuma.h" +#include "virqemu.h" #include "netdev_bandwidth_conf.h" #include "virdomainsnapshotobjlist.h" #include "virenum.h" @@ -13752,6 +13753,14 @@ qemuDomainBackupGetXMLDesc(virDomainPtr domain, #define QEMU_DOMAIN_MONITOR_COMMAND_FLAGS \ VIR_DOMAIN_QEMU_MONITOR_COMMAND_HMP +static int +qemuMonitorDomainGetInfo(virConnectPtr conn, + virDomainDef *def) +{ + return virDomainGetInfoEnsureACL(conn, def); +} + + static int qemuDomainQemuMonitorCommandWithFiles(virDomainPtr domain, const char *cmd, @@ -13789,8 +13798,16 @@ qemuDomainQemuMonitorCommandWithFiles(virDomainPtr domain, if (!(vm = qemuDomainObjFromDomain(domain))) goto cleanup; - if (virDomainQemuMonitorCommandWithFilesEnsureACL(domain->conn, vm->def) < 0) - goto cleanup; + /* HMP "info" commands only query QEMU state, so read access is enough. */ + if (ninfds == 0 && !outfds && + (flags & VIR_DOMAIN_QEMU_MONITOR_COMMAND_HMP) && + virQEMUHMPCommandIsInfo(cmd)) { + if (qemuMonitorDomainGetInfo(domain->conn, vm->def) < 0) + goto cleanup; + } else { + if (virDomainQemuMonitorCommandWithFilesEnsureACL(domain->conn, vm->def) < 0) + goto cleanup; + } if (virDomainObjBeginJob(vm, VIR_JOB_QUERY) < 0) goto cleanup; diff --git a/src/util/virqemu.c b/src/util/virqemu.c index d2a69026c8..74f3729b67 100644 --- a/src/util/virqemu.c +++ b/src/util/virqemu.c @@ -32,6 +32,27 @@ VIR_LOG_INIT("util.qemu"); + +bool +virQEMUHMPCommandIsInfo(const char *cmd) +{ + g_autofree char *trimmed = g_strdup(cmd); + const char *args; + + if (!trimmed) + return false; + + g_strstrip(trimmed); + + if (!STRPREFIX(trimmed, "info")) + return false; + + args = trimmed + strlen("info"); + + return *args == '\0' || g_ascii_isspace(*args); +} + + struct virQEMUCommandLineJSONIteratorData { const char *prefix; virBuffer *buf; diff --git a/src/util/virqemu.h b/src/util/virqemu.h index e5d36b95c4..b9e79fb9b6 100644 --- a/src/util/virqemu.h +++ b/src/util/virqemu.h @@ -42,3 +42,5 @@ int virQEMUBuildCommandLineJSON(virJSONValue *value, virQEMUBuildCommandLineJSONArrayFormatFunc array); void virQEMUBuildBufferEscapeComma(virBuffer *buf, const char *str); + +bool virQEMUHMPCommandIsInfo(const char *cmd); -- 2.47.3

3 2

[PATCH] bhyve: fix bhyveConnectAgent()
by Roman Bogorodskiy 15 Jun '26

15 Jun '26

The bhyveConnectAgent() function calls qemuAgentOpen() to open an agent connection. If it fails, e.g. because of insufficient permissions to open the socket, it returns NULL. Currently, if that happens, bhyveConnectAgent() just sets agentError to true and exits with 0. This does not match the contract of bhyveDomainEnsureAgent(), which should either provide an agent connection or fail. Fix that by returning -1 when qemuAgentOpen() fails. To make intent clearer, add a documentation for bhyveConnectAgent(). Signed-off-by: Roman Bogorodskiy <bogorodskiy(a)gmail.com> --- src/bhyve/bhyve_process.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/src/bhyve/bhyve_process.c b/src/bhyve/bhyve_process.c index e92bdeb416..ed7e4e7725 100644 --- a/src/bhyve/bhyve_process.c +++ b/src/bhyve/bhyve_process.c @@ -238,6 +238,18 @@ static qemuAgentCallbacks agentCallbacks = { .errorNotify = bhyveProcessHandleAgentError, }; +/** + * bhyveConnectAgent: + * @driver: driver object + * @vm: domain object + * + * Connect to the guest agent via the UNIX socket specified in the domain + * definition. On success, bhyveDomainObjPrivate's agent field should + * point to the agent instance. + * + * Returns: 0 on success, + * -1 on error + */ int bhyveConnectAgent(struct _bhyveConn *driver G_GNUC_UNUSED, virDomainObj *vm) { @@ -270,8 +282,7 @@ bhyveConnectAgent(struct _bhyveConn *driver G_GNUC_UNUSED, virDomainObj *vm) priv->agent = agent; if (!priv->agent) { VIR_WARN("Cannot connect to QEMU guest agent for %s", vm->def->name); - priv->agentError = true; - virResetLastError(); + return -1; } return 0; -- 2.52.0

2 1

[PATCH] qemu: use proper job type for qemuDomainAuthorizedSSHKeysSet()
by Roman Bogorodskiy 15 Jun '26

15 Jun '26

As the qemuDomainAuthorizedSSHKeysSet() call modifies the domain, use the VIR_AGENT_JOB_MODIFY job type, not VIR_AGENT_JOB_QUERY. Signed-off-by: Roman Bogorodskiy <bogorodskiy(a)gmail.com> --- src/qemu/qemu_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 544955ecf9..c3eb209f70 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -20375,7 +20375,7 @@ qemuDomainAuthorizedSSHKeysSet(virDomainPtr dom, if (virDomainAuthorizedSshKeysSetEnsureACL(dom->conn, vm->def) < 0) goto cleanup; - if (virDomainObjBeginAgentJob(vm, VIR_AGENT_JOB_QUERY) < 0) + if (virDomainObjBeginAgentJob(vm, VIR_AGENT_JOB_MODIFY) < 0) goto cleanup; if (!qemuDomainAgentAvailable(vm, true)) -- 2.52.0

2 2

[PATCH] util: virHostCPUGetStatsLinux: add guest_nice to VIR_NODE_CPU_STATS_GUEST
by Claudio Fontana 15 Jun '26

15 Jun '26

Fix VIR_NODE_CPU_STATS_GUEST to also account for guest nice time, similarly to how VIR_NODE_CPU_STATS_USER also accounts for nice time. VIR_NODE_CPU_STATS_GUEST was introduced in: b5878727c714 ("util: virHostCPUGetStatsLinux: support VIR_NODE_CPU_STATS_GUEST") Extend the tests artificially to capture this specific case. Fixes: b5878727c714d813c820ad4a1b695fbbb5ffc84e Signed-off-by: Claudio Fontana <cfontana(a)suse.de> --- src/util/virhostcpu.c | 2 +- tests/virhostcpudata/linux-cpustat-24cpu.out | 8 ++++---- tests/virhostcpudata/linux-cpustat-24cpu.stat | 4 ++-- 3 files changed, 7 insertions(+), 7 deletions(-) diff --git a/src/util/virhostcpu.c b/src/util/virhostcpu.c index 1db096ba24..f2b25e940b 100644 --- a/src/util/virhostcpu.c +++ b/src/util/virhostcpu.c @@ -882,7 +882,7 @@ virHostCPUGetStatsLinux(FILE *procstat, return -1; if (virHostCPUStatsAssign(&params[4], VIR_NODE_CPU_STATS_GUEST, - guest * TICK_TO_NSEC) < 0) + (guest + guest_nice) * TICK_TO_NSEC) < 0) return -1; return 0; } diff --git a/tests/virhostcpudata/linux-cpustat-24cpu.out b/tests/virhostcpudata/linux-cpustat-24cpu.out index f3183f5c41..a975702a06 100644 --- a/tests/virhostcpudata/linux-cpustat-24cpu.out +++ b/tests/virhostcpudata/linux-cpustat-24cpu.out @@ -1,16 +1,16 @@ cpu: kernel: 8751170 user: 14128079 -idle: 1816344522 +idle: 1816344521 iowait: 81323 -guest: 5880634 +guest: 5880635 cpu0: kernel: 447603 user: 749021 -idle: 75399242 +idle: 75399241 iowait: 5295 -guest: 331814 +guest: 331815 cpu1: kernel: 167215 diff --git a/tests/virhostcpudata/linux-cpustat-24cpu.stat b/tests/virhostcpudata/linux-cpustat-24cpu.stat index bc9d449e1c..e13035b5f0 100644 --- a/tests/virhostcpudata/linux-cpustat-24cpu.stat +++ b/tests/virhostcpudata/linux-cpustat-24cpu.stat @@ -1,5 +1,5 @@ -cpu 14126233 1846 7764352 1816344522 81323 395581 591237 0 5880634 0 -cpu0 748997 24 320851 75399242 5295 22050 104702 0 331814 0 +cpu 14126233 1846 7764352 1816344521 81323 395581 591237 0 5880634 1 +cpu0 748997 24 320851 75399241 5295 22050 104702 0 331814 1 cpu1 337325 1 140909 76178612 1121 8962 17344 0 166726 0 cpu2 666860 29 269302 75649696 4298 18473 21155 0 272094 0 cpu3 328387 77 211400 76131634 1219 9701 6573 0 115551 0 -- 2.35.3

3 3

[PULL 028/106] vhost: drop backend_features field
by Michael S. Tsirkin 14 Jun '26

14 Jun '26

From: Vladimir Sementsov-Ogievskiy <vsementsov(a)yandex-team.ru> This field is mostly unused and sometimes confusing. We even have a TODO-like comment to drop it, the comment is removed in this commit. The field is used to held VHOST_USER_F_PROTOCOL_FEATURES for vhost-user and/or VHOST_NET_F_VIRTIO_NET_HDR for vhost-net (which may be vhost-user-net). But we can simply recalculate these two flags in place from hdev->features, and from net-client for VHOST_NET_F_VIRTIO_NET_HDR. Note: removing field from x-query-virtio-status result is incompatible change. We can do it because the command is unstable. Cc: devel(a)lists.libvirt.org Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov(a)yandex-team.ru> Acked-by: Markus Armbruster <armbru(a)redhat.com> Reviewed-by: Daniil Tatianin <d-tatianin(a)yandex-team.ru> Acked-by: Raphael Norwitz <raphael.s.norwitz(a)gmail.com> Reviewed-by: Michael S. Tsirkin <mst(a)redhat.com> Tested-by: Lei Yang <leiyang(a)redhat.com> Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Message-Id: <20260420200339.708640-3-vsementsov(a)yandex-team.ru> --- hw/block/vhost-user-blk.c | 1 - hw/net/vhost_net.c | 26 ++++++++++++++------------ hw/scsi/vhost-scsi.c | 1 - hw/scsi/vhost-user-scsi.c | 1 - hw/virtio/vdpa-dev.c | 1 - hw/virtio/vhost-user.c | 17 ++++++++--------- hw/virtio/virtio-hmp-cmds.c | 2 -- hw/virtio/virtio-qmp.c | 4 ---- include/hw/virtio/vhost.h | 7 ------- qapi/virtio.json | 3 --- 10 files changed, 22 insertions(+), 41 deletions(-) diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c index c151e83677..dc99de0263 100644 --- a/hw/block/vhost-user-blk.c +++ b/hw/block/vhost-user-blk.c @@ -348,7 +348,6 @@ static int vhost_user_blk_connect(DeviceState *dev, Error **errp) s->dev.nvqs = s->num_queues; s->dev.vqs = s->vhost_vqs; s->dev.vq_index = 0; - s->dev.backend_features = 0; vhost_dev_set_config_notifier(&s->dev, &blk_ops); diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c index 25e9f1fd24..fda90e231e 100644 --- a/hw/net/vhost_net.c +++ b/hw/net/vhost_net.c @@ -52,8 +52,14 @@ int vhost_net_set_config(struct vhost_net *net, const uint8_t *data, void vhost_net_ack_features_ex(struct vhost_net *net, const uint64_t *features) { - virtio_features_copy(net->dev.acked_features_ex, - net->dev.backend_features_ex); + virtio_features_clear(net->dev.acked_features_ex); + if (net->backend == -1) { + net->dev.acked_features = + net->dev.features & (1ULL << VHOST_USER_F_PROTOCOL_FEATURES); + } else if (!qemu_has_vnet_hdr(net->nc)) { + net->dev.acked_features = 1ULL << VHOST_NET_F_VIRTIO_NET_HDR; + } + vhost_ack_features_ex(&net->dev, net->feature_bits, features); } @@ -258,12 +264,9 @@ struct vhost_net *vhost_net_init(VhostNetOptions *options) if (r < 0) { goto fail; } - net->dev.backend_features = qemu_has_vnet_hdr(options->net_backend) - ? 0 : (1ULL << VHOST_NET_F_VIRTIO_NET_HDR); net->backend = r; net->dev.protocol_features = 0; } else { - virtio_features_clear(net->dev.backend_features_ex); net->dev.protocol_features = 0; net->backend = -1; @@ -284,13 +287,12 @@ struct vhost_net *vhost_net_init(VhostNetOptions *options) net->dev.features &= ~(1ULL << VIRTIO_NET_F_MRG_RXBUF); } - if (virtio_features_andnot(missing_features, - net->dev.backend_features_ex, - net->dev.features_ex)) { - fprintf(stderr, "vhost lacks feature mask 0x" VIRTIO_FEATURES_FMT - " for backend\n", VIRTIO_FEATURES_PR(missing_features)); - goto fail; - } + if (!qemu_has_vnet_hdr(options->net_backend) && + (~net->dev.features & (1ULL << VHOST_NET_F_VIRTIO_NET_HDR))) { + fprintf(stderr, "vhost lacks feature mask 0x%llx for backend\n", + ~net->dev.features & (1ULL << VHOST_NET_F_VIRTIO_NET_HDR)); + goto fail; + } } /* Set sane init value. Override when guest acks. */ diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c index 699863cc10..859888fdba 100644 --- a/hw/scsi/vhost-scsi.c +++ b/hw/scsi/vhost-scsi.c @@ -275,7 +275,6 @@ static void vhost_scsi_realize(DeviceState *dev, Error **errp) vqs = g_new0(struct vhost_virtqueue, vsc->dev.nvqs); vsc->dev.vqs = vqs; vsc->dev.vq_index = 0; - vsc->dev.backend_features = 0; ret = vhost_dev_init(&vsc->dev, (void *)(uintptr_t)vhostfd, VHOST_BACKEND_TYPE_KERNEL, 0, errp); diff --git a/hw/scsi/vhost-user-scsi.c b/hw/scsi/vhost-user-scsi.c index 3612897d4b..5070178dc2 100644 --- a/hw/scsi/vhost-user-scsi.c +++ b/hw/scsi/vhost-user-scsi.c @@ -159,7 +159,6 @@ static int vhost_user_scsi_connect(DeviceState *dev, Error **errp) vsc->dev.nvqs = VIRTIO_SCSI_VQ_NUM_FIXED + vs->conf.num_queues; vsc->dev.vqs = s->vhost_vqs; vsc->dev.vq_index = 0; - vsc->dev.backend_features = 0; ret = vhost_dev_init(&vsc->dev, &s->vhost_user, VHOST_BACKEND_TYPE_USER, 0, errp); diff --git a/hw/virtio/vdpa-dev.c b/hw/virtio/vdpa-dev.c index 4532d63653..06179e402e 100644 --- a/hw/virtio/vdpa-dev.c +++ b/hw/virtio/vdpa-dev.c @@ -104,7 +104,6 @@ static void vhost_vdpa_device_realize(DeviceState *dev, Error **errp) v->dev.vqs = vqs; v->dev.vq_index = 0; v->dev.vq_index_end = v->dev.nvqs; - v->dev.backend_features = 0; v->started = false; ret = vhost_vdpa_get_iova_range(v->vhostfd, &iova_range); diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c index 13c73e7b17..7ce17c5b2d 100644 --- a/hw/virtio/vhost-user.c +++ b/hw/virtio/vhost-user.c @@ -1465,14 +1465,15 @@ static int vhost_user_set_features(struct vhost_dev *dev, int ret; /* - * We need to include any extra backend only feature bits that - * might be needed by our device. Currently this includes the - * VHOST_USER_F_PROTOCOL_FEATURES bit for enabling protocol - * features. + * Don't lose VHOST_USER_F_PROTOCOL_FEATURES, which is vhost-user + * specific. */ - ret = vhost_user_set_u64(dev, VHOST_USER_SET_FEATURES, - features | dev->backend_features, - log_enabled); + if (virtio_has_feature(dev->features, VHOST_USER_F_PROTOCOL_FEATURES)) { + features |= 1ULL << VHOST_USER_F_PROTOCOL_FEATURES; + } + + ret = vhost_user_set_u64(dev, VHOST_USER_SET_FEATURES, features, + log_enabled); if (virtio_has_feature(dev->protocol_features, VHOST_USER_PROTOCOL_F_STATUS)) { @@ -2192,8 +2193,6 @@ static int vhost_user_backend_init(struct vhost_dev *dev, void *opaque, (dev->config_ops && dev->config_ops->vhost_dev_config_notifier); uint64_t protocol_features; - dev->backend_features |= 1ULL << VHOST_USER_F_PROTOCOL_FEATURES; - err = vhost_user_get_u64(dev, VHOST_USER_GET_PROTOCOL_FEATURES, &protocol_features); if (err < 0) { diff --git a/hw/virtio/virtio-hmp-cmds.c b/hw/virtio/virtio-hmp-cmds.c index 1daae482d3..4bf9a3109d 100644 --- a/hw/virtio/virtio-hmp-cmds.c +++ b/hw/virtio/virtio-hmp-cmds.c @@ -176,8 +176,6 @@ void hmp_virtio_status(Monitor *mon, const QDict *qdict) hmp_virtio_dump_features(mon, s->vhost_dev->features); monitor_printf(mon, " Acked features:\n"); hmp_virtio_dump_features(mon, s->vhost_dev->acked_features); - monitor_printf(mon, " Backend features:\n"); - hmp_virtio_dump_features(mon, s->vhost_dev->backend_features); monitor_printf(mon, " Protocol features:\n"); hmp_virtio_dump_protocols(mon, s->vhost_dev->protocol_features); } diff --git a/hw/virtio/virtio-qmp.c b/hw/virtio/virtio-qmp.c index f9cdca50d9..71717c56a4 100644 --- a/hw/virtio/virtio-qmp.c +++ b/hw/virtio/virtio-qmp.c @@ -710,8 +710,6 @@ VirtioStatus *qmp_x_query_virtio_status(const char *path, Error **errp) vdev->guest_features_ex); status->host_features = qmp_decode_features(vdev->device_id, vdev->host_features_ex); - status->backend_features = qmp_decode_features(vdev->device_id, - vdev->backend_features_ex); switch (vdev->device_endian) { case VIRTIO_DEVICE_ENDIAN_LITTLE: @@ -752,8 +750,6 @@ VirtioStatus *qmp_x_query_virtio_status(const char *path, Error **errp) qmp_decode_features(vdev->device_id, hdev->features_ex); status->vhost_dev->acked_features = qmp_decode_features(vdev->device_id, hdev->acked_features_ex); - status->vhost_dev->backend_features = - qmp_decode_features(vdev->device_id, hdev->backend_features_ex); status->vhost_dev->protocol_features = qmp_decode_protocols(hdev->protocol_features); diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h index 3dfec5e15c..e139e05985 100644 --- a/include/hw/virtio/vhost.h +++ b/include/hw/virtio/vhost.h @@ -100,16 +100,9 @@ struct vhost_dev { * * @features: available features provided by the backend * @acked_features: final negotiated features with front-end driver - * - * @backend_features: this is used in a couple of places to either - * store VHOST_USER_F_PROTOCOL_FEATURES to apply to - * VHOST_USER_SET_FEATURES or VHOST_NET_F_VIRTIO_NET_HDR. Its - * future use should be discouraged and the variable retired as - * its easy to confuse with the VirtIO backend_features. */ VIRTIO_DECLARE_FEATURES(features); VIRTIO_DECLARE_FEATURES(acked_features); - VIRTIO_DECLARE_FEATURES(backend_features); /** * @protocol_features: is the vhost-user only feature set by diff --git a/qapi/virtio.json b/qapi/virtio.json index 09dd0e6d05..671f1ad793 100644 --- a/qapi/virtio.json +++ b/qapi/virtio.json @@ -85,8 +85,6 @@ # # @acked-features: vhost_dev acked_features # -# @backend-features: vhost_dev backend_features -# # @protocol-features: vhost_dev protocol_features # # @max-queues: vhost_dev max_queues @@ -106,7 +104,6 @@ 'vq-index': 'int', 'features': 'VirtioDeviceFeatures', 'acked-features': 'VirtioDeviceFeatures', - 'backend-features': 'VirtioDeviceFeatures', 'protocol-features': 'VhostDeviceProtocols', 'max-queues': 'uint64', 'backend-cap': 'uint64', -- MST

1 0

[PATCH 0/4] Remove deprecated Power8E and 8NVL CPUs
by Aditya Gupta 12 Jun '26

12 Jun '26

Overview ======== Power8E and Power8NVL were deprecated since QEMU 10.1, with commit 264a604e7163 ("target/ppc: Deprecate Power8E and Power8NVL") Hence, remove usage (chip and core types) of Power8E and 8NVL from PowerNV, PSeries as well as testcases. NOTE: PRE-REQUISITE: Requires following "Testcase improvements for PowerNV" to be merged first: https://lore.kernel.org/qemu-devel/20260610090235.3665904-1-adityag@linux.i… Aditya Gupta (4): ppc/pnv: Replace Power8E with Power11 for 'none' machine test ppc/pnv: Remove Power8E and Power8NVL pnv chips ppc/pnv: Remove Power8E and Power8NVL spapr cores ppc/pnv: Remove Power8E and Power8NVL CPUs docs/about/deprecated.rst | 9 ----- docs/system/ppc/powernv.rst | 2 +- docs/system/ppc/pseries.rst | 2 +- hw/ppc/pnv.c | 62 --------------------------------- hw/ppc/pnv_core.c | 2 -- hw/ppc/spapr_cpu_core.c | 2 -- include/hw/ppc/pnv.h | 8 ----- target/ppc/cpu-models.c | 6 ---- target/ppc/cpu-models.h | 4 --- target/ppc/cpu_init.c | 6 ---- target/ppc/kvm.c | 4 +-- tests/qtest/machine-none-test.c | 2 +- tests/qtest/pnv-xscom.h | 9 ----- 13 files changed, 4 insertions(+), 114 deletions(-) base-commit: de5d8bfd6105d3dd3ae668df9762df244a6d1506 -- 2.54.0

3 13

[PATCH] ci: refresh with 'lcitool manifest'
by Michal Privoznik 12 Jun '26

12 Jun '26

From: Michal Privoznik <mprivozn(a)redhat.com> This switches Alpine from 3.23 to 3.24. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Greenish pipeline: https://gitlab.com/MichalPrivoznik/libvirt/-/pipelines/2596746869 ci/buildenv/{alpine-323.sh => alpine-324.sh} | 0 .../{alpine-323.Dockerfile => alpine-324.Dockerfile} | 2 +- ci/gitlab/builds.yml | 8 ++++---- ci/gitlab/containers.yml | 4 ++-- ci/manifest.yml | 2 +- 5 files changed, 8 insertions(+), 8 deletions(-) rename ci/buildenv/{alpine-323.sh => alpine-324.sh} (100%) rename ci/containers/{alpine-323.Dockerfile => alpine-324.Dockerfile} (98%) diff --git a/ci/buildenv/alpine-323.sh b/ci/buildenv/alpine-324.sh similarity index 100% rename from ci/buildenv/alpine-323.sh rename to ci/buildenv/alpine-324.sh diff --git a/ci/containers/alpine-323.Dockerfile b/ci/containers/alpine-324.Dockerfile similarity index 98% rename from ci/containers/alpine-323.Dockerfile rename to ci/containers/alpine-324.Dockerfile index ee1f194808..e2022db8db 100644 --- a/ci/containers/alpine-323.Dockerfile +++ b/ci/containers/alpine-324.Dockerfile @@ -4,7 +4,7 @@ # # https://gitlab.com/libvirt/libvirt-ci -FROM docker.io/library/alpine:3.23 +FROM docker.io/library/alpine:3.24 RUN apk update && \ apk upgrade && \ diff --git a/ci/gitlab/builds.yml b/ci/gitlab/builds.yml index d947170ebe..c11c834f60 100644 --- a/ci/gitlab/builds.yml +++ b/ci/gitlab/builds.yml @@ -59,15 +59,15 @@ x86_64-almalinux-10-clang: TARGET_BASE_IMAGE: docker.io/library/almalinux:10 -x86_64-alpine-323: +x86_64-alpine-324: extends: .native_build_job needs: - - job: x86_64-alpine-323-container + - job: x86_64-alpine-324-container optional: true allow_failure: false variables: - NAME: alpine-323 - TARGET_BASE_IMAGE: docker.io/library/alpine:3.23 + NAME: alpine-324 + TARGET_BASE_IMAGE: docker.io/library/alpine:3.24 x86_64-alpine-edge: diff --git a/ci/gitlab/containers.yml b/ci/gitlab/containers.yml index c74443f565..a4b9a5f03b 100644 --- a/ci/gitlab/containers.yml +++ b/ci/gitlab/containers.yml @@ -21,11 +21,11 @@ x86_64-almalinux-10-container: NAME: almalinux-10 -x86_64-alpine-323-container: +x86_64-alpine-324-container: extends: .container_job allow_failure: false variables: - NAME: alpine-323 + NAME: alpine-324 x86_64-alpine-edge-container: diff --git a/ci/manifest.yml b/ci/manifest.yml index d36e3130d8..67a53bb3d2 100644 --- a/ci/manifest.yml +++ b/ci/manifest.yml @@ -32,7 +32,7 @@ targets: RPM: skip CC: clang - alpine-323: x86_64 + alpine-324: x86_64 alpine-edge: jobs: -- 2.53.0

2 1

[PATCH] lxc: Assign aliases to filesystem devices
by rsmigiel＠redhat.com 11 Jun '26

11 Jun '26

From: Radoslaw Smigielski <rsmigiel(a)redhat.com> From: Radoslaw Smigielski <rsmigiel(a)redhat.com> LXC domains did not assign device aliases to filesystem devices, unlike other device types (consoles, network interfaces, etc.). This change assigns aliases in the format 'fs0', 'fs1', etc. to all filesystem devices during domain startup, following the same pattern used for console devices. Related: https://gitlab.com/libvirt/libvirt/-/work_items/63 Signed-off-by: Radoslaw Smigielski <rsmigiel(a)redhat.com> --- src/lxc/lxc_process.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index 2c0bcb9dd3aa..aae9fcc9dfd1 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -1350,6 +1350,12 @@ int virLXCProcessStart(virLXCDriver * driver, vm->def->consoles[i]->info.alias = g_strdup_printf("console%zu", i); } + VIR_DEBUG("Setting up filesystem aliases"); + for (i = 0; i < vm->def->nfss; i++) { + g_free(vm->def->fss[i]->info.alias); + vm->def->fss[i]->info.alias = g_strdup_printf("fs%zu", i); + } + VIR_DEBUG("Setting up Interfaces"); if (virLXCProcessSetupInterfaces(driver, vm->def, &veths) < 0) goto cleanup; -- 2.54.0

2 1