March 2026 - Devel - libvirt List Archives

[PATCH] util: Fix secondary bus reset check for multi-root PCI systems
by ehanoc 23 Apr '26

23 Apr '26

The virPCIDeviceReset() function was using a hardcoded check for bus != 0 to determine if a device is on a root bus before attempting a secondary bus reset. This breaks on multi-root systems like Intel Arrow Lake where devices can be on non-zero root buses (e.g., Bus 0x80 for the PCH root complex). Update the logic to use virPCIDeviceIsOnRootBus() which detects root bus attachment via sysfs topology, making it work correctly on any PCI-compliant system regardless of bus numbers. This ensures secondary bus reset is never attempted on ANY root bus (0x00, 0x80, etc.), only on devices behind PCI bridges where it's safe. Signed-off-by: Bruno Martins <ehanoc(a)protonmail.com> --- src/util/virpci.c | 106 +++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 99 insertions(+), 7 deletions(-) diff --git a/src/util/virpci.c b/src/util/virpci.c index 2348a98003..f2ce24e6aa 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -1081,6 +1081,81 @@ virPCIDeviceInit(virPCIDevice *dev, int cfgfd) return 0; } +/* + * Check if a PCI device is directly attached to a root bus (no parent bridge). + * + * Modern systems like Intel Arrow Lake (Core Ultra 200) can have multiple + * root buses (e.g., bus 0x00 for CPU devices and bus 0x80 for PCH devices). + * We detect root bus attachment by examining the sysfs topology: a device + * is on a root bus if its sysfs path is a direct child of a pciXXXX:YY + * directory (e.g., /sys/devices/pci0000:80/0000:80:14.0). + * + * Returns 1 if on root bus, 0 if not, -1 on error. + */ +static int +virPCIDeviceIsOnRootBus(virPCIDevice *dev) +{ + unsigned int domain, bus; + g_autofree char *devPath = NULL; + g_autofree char *realPath = NULL; + g_autofree char *parentDir = NULL; + const char *parentName; + + /* Get the sysfs symlink path for this device */ + devPath = g_strdup_printf(PCI_SYSFS "devices/%s", dev->name); + + /* Resolve the symlink to get the real path in /sys/devices/... */ + if (virFileResolveLink(devPath, &realPath) < 0) { + VIR_DEBUG("%s %s: failed to resolve sysfs path %s", + dev->id, dev->name, devPath); + return -1; + } + + parentDir = g_path_get_dirname(realPath); + if (!parentDir) + return -1; + + parentName = strrchr(parentDir, '/'); + if (!parentName) + return -1; + parentName++; // skip the '/' + + /* + * Detect root bus by parsing the parent directory name using sscanf. + * + * PCI root buses in sysfs follow the naming convention: + * pci<domain>:<bus> + * + * - <domain>: number (0000-ffff, hex) + * - <bus>: number within the domain (00-ff, hex) + * + * This format is defined in: + * https://www.kernel.org/doc/html/latest/PCI/sysfs-pci.html + * + * Examples from Intel Arrow Lake systems: + * - pci0000:00 (CPU root bus) + * - pci0000:80 (PCH root bus - the problematic case on Arrow Lake) + * + * sscanf pattern "pci%x:%x" matches: + * - "pci" : literal prefix + * - "%x" : hex domain (stops at ':') + * - ":" : literal separator + * - "%x" : hex bus (must reach end of string) + * + * Returns 2 if both domain and bus were successfully parsed, + * confirming the device is a direct child of a root bus. + */ + if (sscanf(parentName, "pci%x:%x", &domain, &bus) == 2) { + VIR_DEBUG("%s %s: device is on root bus (parent=%s)", + dev->id, dev->name, parentName); + return 1; + } + + VIR_DEBUG("%s %s: device is not on root bus (parent=%s)", + dev->id, dev->name, parentName); + return 0; +} + int virPCIDeviceReset(virPCIDevice *dev, virPCIDeviceList *activeDevs, @@ -1146,9 +1221,17 @@ virPCIDeviceReset(virPCIDevice *dev, if (dev->has_pm_reset) ret = virPCIDeviceTryPowerManagementReset(dev, fd); - /* Bus reset is not an option with the root bus */ - if (ret < 0 && dev->address.bus != 0) - ret = virPCIDeviceTrySecondaryBusReset(dev, fd, inactiveDevs); + /* Bus reset is not an option with the root bus. On multi-root + * systems like Intel Arrow Lake, root buses can be non-zero (e.g., 0x80), + * so we use sysfs topology detection instead of checking bus == 0. + */ + if (ret < 0) { + int onRootBus = virPCIDeviceIsOnRootBus(dev); + if (onRootBus < 0) + goto cleanup; + if (!onRootBus) + ret = virPCIDeviceTrySecondaryBusReset(dev, fd, inactiveDevs); + } if (ret < 0) { virErrorPtr err = virGetLastError(); @@ -2531,15 +2614,24 @@ static int virPCIDeviceIsBehindSwitchLackingACS(virPCIDevice *dev) { g_autoptr(virPCIDevice) parent = NULL; + int onRootBus; if (virPCIDeviceGetParent(dev, &parent) < 0) return -1; if (!parent) { - /* if we have no parent, and this is the root bus, ACS doesn't come - * into play since devices on the root bus can't P2P without going - * through the root IOMMU. + /* If we have no parent bridge, check if the device is directly + * attached to a root bus. Systems like Intel Arrow Lake can have + * multiple root buses (e.g., 0x00 and 0x80), so we can't just + * check for bus == 0. Instead, use sysfs topology to detect + * root bus attachment. + * + * ACS doesn't come into play for devices on root buses since + * they can't P2P without going through the root IOMMU. */ - if (dev->address.bus == 0) { + onRootBus = virPCIDeviceIsOnRootBus(dev); + if (onRootBus < 0) + return -1; + if (onRootBus) { return 0; } else { virReportError(VIR_ERR_INTERNAL_ERROR, -- 2.47.3

5 8

[PULL 00/32] Next pr patches
by Daniel P. Berrangé 23 Apr '26

23 Apr '26

The following changes since commit 53b41bb78950912ba2d9809eef6b45e4df30c647: Merge tag 'pull-target-arm-20251031' of https://gitlab.com/pm215/qemu into staging (2025-11-01 10:52:48 +0100) are available in the Git repository at: https://gitlab.com/berrange/qemu tags/next-pr-pull-request for you to fetch changes up to 2aaca8c6d22b18786ceff51189704113d0639590: docs: creation of x509 certs compliant with post-quantum crypto (2025-11-03 10:45:55 +0000) ---------------------------------------------------------------- Merge crypto and other misc fixes / features * Increase minimum gnutls to 3.7.5 * Increase minimum libgcrypt to 1.9.4 * Increase minimum nettle to 3.7.3 * Drop obsolete in-tree XTS impl * Fix memory leak when loading certificates * Remove/reduce duplication when loading certifcates * Fix possible crash when certificates are unloaded while an active TLS connection is using when in a TLS handshake operation * Deprecate use of dh-params.pem file * Document how to create certificates with Post-Quantum Cryptography compliant algorithms. * Support loading multiple certificate identities to allow support for Post-Quantum crypto in parallel with traditional RSA/ECC * Add "-run-with exit-with-parent=on" parameter * Flush pending errors when seeing ENOBUFS with a zero-copy send attempt * Fix data buffer parameters in hash & IO channel APIs to use 'void *' ---------------------------------------------------------------- Daniel P. Berrangé (26): crypto: bump min gnutls to 3.7.5 crypto: unconditionally enable gnutls XTS support crypto: bump min libgcrypt to 1.9.4 crypto: bump min nettle to 3.7.3 crypto: drop in-tree XTS cipher mode impl crypto: remove redundant parameter checking CA certs crypto: add missing free of certs array crypto: replace stat() with access() for credential checks crypto: remove redundant access() checks before loading certs crypto: move check for TLS creds 'dir' property crypto: use g_autofree when loading x509 credentials crypto: remove needless indirection via parent_obj field crypto: move release of DH parameters into TLS creds parent crypto: shorten the endpoint == server check in TLS creds crypto: remove duplication loading x509 CA cert crypto: reduce duplication in handling TLS priority strings crypto: introduce method for reloading TLS creds crypto: introduce a wrapper around gnutls credentials crypto: fix lifecycle handling of gnutls credentials objects crypto: make TLS credentials structs private crypto: deprecate use of external dh-params.pem file crypto: avoid loading the CA certs twice crypto: avoid loading the identity certs twice crypto: expand logic to cope with multiple certificate identities crypto: support upto 5 parallel certificate identities docs: creation of x509 certs compliant with post-quantum crypto Manish Mishra (1): io: flush zerocopy socket error queue on sendmsg failure due to ENOBUF Philippe Mathieu-Daudé (2): crypto/hash: Have hashing functions take void * buffer argument io/channel: Have read/write functions take void * buffer argument Richard W.M. Jones (2): Implement -run-with exit-with-parent=on tests/qtest: Use exit-with-parent=on in qtest invocations Tejus GK (1): io: add a "blocking" field to QIOChannelSocket crypto/cipher-gnutls.c.inc | 8 - crypto/cipher-nettle.c.inc | 44 -- crypto/cipher.c | 2 +- crypto/hash.c | 16 +- crypto/hmac.c | 8 +- crypto/meson.build | 10 +- crypto/tlscreds.c | 79 ++-- crypto/tlscredsanon.c | 64 +-- crypto/tlscredsbox.c | 101 +++++ crypto/tlscredsbox.h | 50 +++ crypto/tlscredspriv.h | 36 +- crypto/tlscredspsk.c | 64 ++- crypto/tlscredsx509.c | 593 +++++++++++++++++--------- crypto/tlssession.c | 139 ++---- crypto/trace-events | 1 + crypto/xts.c | 250 ----------- docs/about/deprecated.rst | 9 + docs/system/tls.rst | 134 +++++- include/crypto/hash.h | 8 +- include/crypto/hmac.h | 4 +- include/crypto/tlscreds.h | 26 ++ include/crypto/tlscredsx509.h | 6 + include/crypto/tlssession.h | 4 +- include/crypto/xts.h | 82 ---- include/io/channel-socket.h | 6 + include/io/channel.h | 14 +- include/qemu/exit-with-parent.h | 57 +++ io/channel-socket.c | 86 +++- io/channel-tls.c | 4 +- io/channel.c | 14 +- meson.build | 69 +-- qemu-options.hx | 13 +- system/exit-with-parent.c | 140 ++++++ system/meson.build | 1 + system/vl.c | 13 + tests/qtest/libqtest.c | 22 +- tests/unit/meson.build | 3 - tests/unit/test-crypto-block.c | 3 +- tests/unit/test-crypto-tlscredsx509.c | 8 +- tests/unit/test-crypto-tlssession.c | 4 +- tests/unit/test-crypto-xts.c | 529 ----------------------- ui/vnc.c | 9 +- 42 files changed, 1208 insertions(+), 1525 deletions(-) create mode 100644 crypto/tlscredsbox.c create mode 100644 crypto/tlscredsbox.h delete mode 100644 crypto/xts.c delete mode 100644 include/crypto/xts.h create mode 100644 include/qemu/exit-with-parent.h create mode 100644 system/exit-with-parent.c delete mode 100644 tests/unit/test-crypto-xts.c -- 2.51.1

3 36

[PATCH v2] qemu: fix potential hang in qemuMigrationSrcCancelUnattended during reconnect
by Denis V. Lunev 22 Apr '26

22 Apr '26

When libvirtd reconnects to a running QEMU process that had an in-progress migration, qemuProcessReconnect first connects the monitor and only later recovers the migration job. During this window the async job is VIR_ASYNC_JOB_NONE, so any MIGRATION status events from QEMU are silently dropped by qemuProcessHandleMigrationStatus. If the migration was already cancelled or completed by QEMU during this window, no further events will be emitted. When qemuMigrationSrcCancelUnattended later restores the async job and calls qemuMigrationSrcCancel with wait=true, the wait loop calls qemuDomainObjWait (virCondWait with no timeout) and blocks forever waiting for an event that will never arrive. Fix this by re-querying QEMU migration state with qemuMigrationAnyRefreshStatus after restoring the async job but before calling qemuMigrationSrcCancel. If QEMU has already reached a terminal state, the cancel is skipped. Signed-off-by: Denis V. Lunev <den(a)openvz.org> CC: Jiri Denemark <jdenemar(a)redhat.com> CC: Peter Krempa <pkrempa(a)redhat.com> CC: Michal Privoznik <mprivozn(a)redhat.com> CC: Efim Shevrin <efim.shevrin(a)virtuozzo.com> --- v1 -> v2: Instead of querying QEMU with query-migrate inside qemuMigrationSrcCancel, use qemuMigrationAnyRefreshStatus in qemuMigrationSrcCancelUnattended after restoring the async job to re-check migration state before the actual cancel. src/qemu/qemu_migration.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index fec808ccfb..a4bd7efa09 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -7330,6 +7330,7 @@ int qemuMigrationSrcCancelUnattended(virDomainObj *vm, virDomainJobObj *oldJob) { + virDomainJobStatus migStatus = VIR_DOMAIN_JOB_STATUS_NONE; bool storage = false; size_t i; @@ -7348,11 +7349,20 @@ qemuMigrationSrcCancelUnattended(virDomainObj *vm, VIR_JOB_NONE); } - /* We're inside a MODIFY job and the restored MIGRATION_OUT async job is - * used only for processing migration events from QEMU. Thus we don't want - * to start a nested job for talking to QEMU. + /* Query the actual migration state from QEMU. The state passed to + * qemuProcessRecoverMigrationOut may be stale: QEMU could have + * reached a terminal state between that initial query and the async + * job restore above, with the corresponding event silently dropped. */ - qemuMigrationSrcCancel(vm, VIR_ASYNC_JOB_NONE, true); + qemuMigrationAnyRefreshStatus(vm, VIR_ASYNC_JOB_NONE, &migStatus); + + if (migStatus != VIR_DOMAIN_JOB_STATUS_CANCELED) { + /* We're inside a MODIFY job and the restored MIGRATION_OUT async + * job is used only for processing migration events from QEMU. + * Thus we don't want to start a nested job for talking to QEMU. + */ + qemuMigrationSrcCancel(vm, VIR_ASYNC_JOB_NONE, true); + } virDomainObjEndAsyncJob(vm); -- 2.51.0

3 7

Re: Clarification Needed: virsh snapshot-delete Behavior with External Snapshots (--disk-only)
by Peter Krempa 10 Apr '26

10 Apr '26

On Wed, Jul 23, 2025 at 04:57:57 +0000, P, Rajasekar via Users wrote: > Hello libvirt community, > > I am seeking clarification on the behavior of `virsh snapshot-delete` when used with external snapshots (created with the `--disk-only` option). > > **Environment:** > - libvirt version: 10.5.0 Note that this was released almost a year ago. > - Kernel version: 5.14.0-503.22.1.el9_5.x86_64 > - Disk image format: qcow2 > - Host OS: RHEL 9.5 > > **Steps to reproduce:** > 1. Create an external snapshot: > virsh snapshot-create-as --domain <vm> --name snap1 --disk-only --atomic > 2. Delete the snapshot: > virsh snapshot-delete <vm> --snapshotname snap1 > > **Observed behavior:** > - Changes from the overlay (snapshot) file are merged into the base image. > - The snapshot metadata is removed from libvirt. > - The snapshot file remains on disk (appears orphaned). You mean that the original overlay images remained on disk? e.g. /path/to/originalimage.snap1 ? > **Questions:** > - Is the observed automatic merge of the overlay file into the base image during `snapshot-delete` the intended behavior for external snapshots in recent libvirt versions? Or we need separate `blockcommit` step. Yes it is intended, because deleting a snapshot is effectively meaning "I don't want to return to this point any more" so the extra overlay makes no sense. > - Should the snapshot file be deleted automatically, or is manual cleanup required? If you mean the overlay, then yes it should have been deleted. > - Is there any official documentation that clarifies this behavior for external snapshots, as most documentation seems to focus on internal snapshots? The semantics of external and internal snapshots are supposed to be indentical. I don't think we document to a deep extent the actual behaviour WRT to the disk files.

2 1

[PATCH] virt-aa-helper: Prevent spurious denials for AoE disks
by PUSHKARAJ PATIL 07 Apr '26

07 Apr '26

virt-aa-helper calls virStorageSourceGetMetadata before adding a disk path to a domain's apparmor profile. This probes the device and may trigger an AppArmor denial when the disk is an AoE device under /dev/etherd/. The return value of virStorageSourceGetMetadata is not checked, so the denial has no functional impact but results in noisy dmesg logs. Allow read access to /dev/etherd/e*.* in the virt-aa-helper profile to avoid these spurious denials. Signed-off-by: PUSHKARAJ PATIL <pushkaraj.patil(a)in.ibm.com> --- src/security/apparmor/usr.lib.libvirt.virt-aa-helper.in | 1 + 1 file changed, 1 insertion(+) diff --git a/src/security/apparmor/usr.lib.libvirt.virt-aa-helper.in b/src/security/apparmor/usr.lib.libvirt.virt-aa-helper.in index e209a8bff7..80e9ef2b08 100644 --- a/src/security/apparmor/usr.lib.libvirt.virt-aa-helper.in +++ b/src/security/apparmor/usr.lib.libvirt.virt-aa-helper.in @@ -73,6 +73,7 @@ profile virt-aa-helper @libexecdir@/virt-aa-helper { /**.vhd r, /**.[iI][sS][oO] r, /**/disk{,.*} r, + /dev/etherd/e*.* r, include if exists <local/usr.lib.libvirt.virt-aa-helper> } -- 2.50.1 (Apple Git-155)

3 2

[PATCH v3] qemu: Store tapfd path in domstatus XML
by Wesley Hershberger 07 Apr '26

07 Apr '26

Introduce a read-only `tapfd` element for direct interfaces (macvtap), which contains the path to the backing tapfd for that interface (e.g. `/dev/tapXX`). The element is only included when the domain is being formatted for internal consumption (VIR_DOMAIN_DEF_FORMAT_STATUS) and is not accepted in user-provided XML (!VIR_DOMAIN_DEF_PARSE_INACTIVE). This is used by the AppArmor security driver when re-generating profiles. Partial-Resolves: https://gitlab.com/libvirt/libvirt/-/issues/692 Bug-Ubuntu: https://bugs.launchpad.net/bugs/2126574 Signed-off-by: Wesley Hershberger <wesley.hershberger(a)canonical.com> --- Resending this patch as I've not recieved a response on my previous submission. Fixed the bug URL in the commit message as I missed that feedback item on my last mail. This submission is a partial revision of a previous series with a fix for the macvtap component of gitlab#692 [1][2]. I haven't had bandwidth to resolve the blockcommit component since the complexity there is somewhat higher (and is also lower priority for us). I kept the separate `tapfd` element rather than reusing the existing `backend` element (virDomainNetBackend.tap) to avoid making a user-visible change [3]. I'd be happy to use the existing field instead if you think that would make more sense. I opted not to introduce/modify a security driver API for FD+path as the patch here is sufficient to resolve the bug, but would be willing to do so if that would make this change more palatable. I've opened a MR to libvirt-tck with test cases that demonstrate the bugs that this fixes [4]. apparmor/110-macvtap.t passes with this patch applied. Thanks for the reviews and continued consideration. ~Wesley [1] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/UNNB… [2] https://gitlab.com/libvirt/libvirt/-/issues/692 [3] https://libvirt.org/formatdomain.html#setting-network-backend-specific-opti… [4] https://gitlab.com/libvirt/libvirt-tck/-/merge_requests/73 --- Changes in v3: - Fix buglink in commit message - Link to v2: https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/IPEB… Changes in v2: - Drop `virt-aa-helper: Ask for no deny rule...` as it was applied - Drop `qemu: Store blockcommit permissions...` due to unresolved concerns - Pass tapfd path through netdef instead of resolving from fd - Link to v1: https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/UNNB… --- src/conf/domain_conf.c | 8 ++++++++ src/conf/domain_conf.h | 1 + src/hypervisor/domain_interface.c | 2 +- src/lxc/lxc_process.c | 1 + src/qemu/qemu_interface.c | 1 + src/security/security_apparmor.c | 1 + src/security/virt-aa-helper.c | 5 +++++ src/util/virnetdevmacvlan.c | 18 +++++++++++------- src/util/virnetdevmacvlan.h | 4 +++- 9 files changed, 32 insertions(+), 9 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 03a05366e1..9714a1e141 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2971,6 +2971,7 @@ virDomainNetDefFree(virDomainNetDef *def) g_free(def->virtio); g_free(def->coalesce); g_free(def->sourceDev); + g_free(def->tapfdpath); virNetDevIPInfoClear(&def->guestIP); virNetDevIPInfoClear(&def->hostIP); @@ -10634,6 +10635,10 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt, return NULL; } + if (!(flags & VIR_DOMAIN_DEF_PARSE_INACTIVE)) { + def->tapfdpath = virXPathString("string(./tapfd/@path)", ctxt); + } + if (virNetworkPortOptionsParseXML(ctxt, &def->isolatedPort) < 0) return NULL; @@ -26004,6 +26009,9 @@ virDomainNetDefFormat(virBuffer *buf, if (def->mtu) virBufferAsprintf(buf, "<mtu size='%u'/>\n", def->mtu); + if (def->tapfdpath && (flags & VIR_DOMAIN_DEF_FORMAT_STATUS)) + virBufferAsprintf(buf, "<tapfd path='%s'/>\n", def->tapfdpath); + virDomainNetDefCoalesceFormatXML(buf, def->coalesce); virDomainDeviceInfoFormat(buf, &def->info, flags | VIR_DOMAIN_DEF_FORMAT_ALLOW_BOOT diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index e63230beec..3bb4de6274 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1212,6 +1212,7 @@ struct _virDomainNetDef { char *downscript; char *domain_name; /* backend domain name */ char *ifname; /* interface name on the host (<target dev='x'/>) */ + char *tapfdpath; /* Path in /dev for macvtap (<tapfd path="/dev/tapXX">) */ virTristateBool managed_tap; virNetDevIPInfo hostIP; char *ifname_guest_actual; diff --git a/src/hypervisor/domain_interface.c b/src/hypervisor/domain_interface.c index 5bc698d272..37e3d453a0 100644 --- a/src/hypervisor/domain_interface.c +++ b/src/hypervisor/domain_interface.c @@ -111,7 +111,7 @@ virDomainInterfaceEthernetConnect(virDomainDef *def, if (virNetDevMacVLanIsMacvtap(net->ifname)) { auditdev = net->ifname; - if (virNetDevMacVLanTapOpen(net->ifname, tapfd, tapfdSize) < 0) + if (virNetDevMacVLanTapOpen(net->ifname, tapfd, tapfdSize, &net->tapfdpath) < 0) goto cleanup; if (virNetDevMacVLanTapSetup(tapfd, tapfdSize, virDomainInterfaceIsVnetCompatModel(net)) < 0) { diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index 1bca9e8dae..c731b28871 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -379,6 +379,7 @@ virLXCProcessSetupInterfaceDirect(virLXCDriver *driver, VIR_NETDEV_VPORT_PROFILE_OP_CREATE, cfg->stateDir, NULL, 0, + &net->tapfdpath, macvlan_create_flags) < 0) return NULL; diff --git a/src/qemu/qemu_interface.c b/src/qemu/qemu_interface.c index 23a23d201a..edc53d53b3 100644 --- a/src/qemu/qemu_interface.c +++ b/src/qemu/qemu_interface.c @@ -81,6 +81,7 @@ qemuInterfaceDirectConnect(virDomainDef *def, &res_ifname, vmop, cfg->stateDir, tapfd, tapfdSize, + &net->tapfdpath, macvlan_create_flags) < 0) goto cleanup; diff --git a/src/security/security_apparmor.c b/src/security/security_apparmor.c index 40f13ec1a5..3ff80e1dc9 100644 --- a/src/security/security_apparmor.c +++ b/src/security/security_apparmor.c @@ -156,6 +156,7 @@ load_profile(virSecurityManager *mgr G_GNUC_UNUSED, if (virDomainDefFormatInternal(def, NULL, &buf, VIR_DOMAIN_DEF_FORMAT_SECURE | + VIR_DOMAIN_DEF_FORMAT_STATUS | VIR_DOMAIN_DEF_FORMAT_VOLUME_TRANSLATED) < 0) return -1; diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index e932e79dab..60e03c2ce8 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -1192,6 +1192,11 @@ get_files(vahControl * ctl) vhu->type) != 0) return -1; } + + if (net->tapfdpath) { + if (vah_add_file(&buf, net->tapfdpath, "rwk") != 0) + return -1; + } } for (i = 0; i < ctl->def->nmems; i++) { diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index cde9d70eef..fcf63e08ff 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -152,24 +152,24 @@ int virNetDevMacVLanDelete(const char *ifname) int virNetDevMacVLanTapOpen(const char *ifname, int *tapfd, - size_t tapfdSize) + size_t tapfdSize, + char **tapname) { int retries = 10; int ret = -1; int ifindex; size_t i = 0; - g_autofree char *tapname = NULL; if (virNetDevGetIndex(ifname, &ifindex) < 0) return -1; - tapname = g_strdup_printf("/dev/tap%d", ifindex); + *tapname = g_strdup_printf("/dev/tap%d", ifindex); for (i = 0; i < tapfdSize; i++) { int fd = -1; while (fd < 0) { - if ((fd = open(tapname, O_RDWR)) >= 0) { + if ((fd = open(*tapname, O_RDWR)) >= 0) { tapfd[i] = fd; } else if (retries-- > 0) { /* may need to wait for udev to be done */ @@ -178,7 +178,7 @@ virNetDevMacVLanTapOpen(const char *ifname, /* However, if haven't succeeded, quit. */ virReportSystemError(errno, _("cannot open macvtap tap device %1$s"), - tapname); + *tapname); goto cleanup; } } @@ -188,6 +188,7 @@ virNetDevMacVLanTapOpen(const char *ifname, cleanup: if (ret < 0) { + g_free(*tapname); while (i--) VIR_FORCE_CLOSE(tapfd[i]); } @@ -659,6 +660,7 @@ virNetDevMacVLanCreateWithVPortProfile(const char *ifnameRequested, char *stateDir, int *tapfd, size_t tapfdSize, + char **tapfdpath, unsigned int flags) { g_autofree char *ifname = NULL; @@ -729,7 +731,7 @@ virNetDevMacVLanCreateWithVPortProfile(const char *ifnameRequested, } if (flags & VIR_NETDEV_MACVLAN_CREATE_WITH_TAP) { - if (virNetDevMacVLanTapOpen(ifname, tapfd, tapfdSize) < 0) + if (virNetDevMacVLanTapOpen(ifname, tapfd, tapfdSize, tapfdpath) < 0) goto disassociate_exit; if (virNetDevMacVLanTapSetup(tapfd, tapfdSize, vnet_hdr) < 0) @@ -888,7 +890,8 @@ int virNetDevMacVLanDelete(const char *ifname G_GNUC_UNUSED) int virNetDevMacVLanTapOpen(const char *ifname G_GNUC_UNUSED, int *tapfd G_GNUC_UNUSED, - size_t tapfdSize G_GNUC_UNUSED) + size_t tapfdSize G_GNUC_UNUSED, + char **tapname G_GNUC_UNUSED) { virReportSystemError(ENOSYS, "%s", _("Cannot create macvlan devices on this platform")); @@ -917,6 +920,7 @@ int virNetDevMacVLanCreateWithVPortProfile(const char *ifname G_GNUC_UNUSED, char *stateDir G_GNUC_UNUSED, int *tapfd G_GNUC_UNUSED, size_t tapfdSize G_GNUC_UNUSED, + char **tapfdpath G_GNUC_UNUSED, unsigned int unused_flags G_GNUC_UNUSED) { virReportSystemError(ENOSYS, "%s", diff --git a/src/util/virnetdevmacvlan.h b/src/util/virnetdevmacvlan.h index 31e4804cdc..7424b87965 100644 --- a/src/util/virnetdevmacvlan.h +++ b/src/util/virnetdevmacvlan.h @@ -72,13 +72,15 @@ int virNetDevMacVLanCreateWithVPortProfile(const char *ifname, char *stateDir, int *tapfd, size_t tapfdSize, + char **tapfdpath, unsigned int flags) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(3) ATTRIBUTE_NONNULL(6) ATTRIBUTE_NONNULL(8) ATTRIBUTE_NONNULL(10) G_GNUC_WARN_UNUSED_RESULT; int virNetDevMacVLanTapOpen(const char *ifname, int *tapfd, - size_t tapfdSize) + size_t tapfdSize, + char **tapname) ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) G_GNUC_WARN_UNUSED_RESULT; --- base-commit: 54f5032e576a886cfb8232abbfa08a2f6e9f9962 change-id: 20260105-apparmor-races-d03238ee4d93 Best regards, -- Wesley Hershberger <wesley.hershberger(a)canonical.com>

2 3

[PATCH] qemu: migration: Call hook script on source host at migration start
by Guy Godfroy 07 Apr '26

07 Apr '26

Currently, the qemu hook script with the "migrate" operation is only called on the destination host at the beginning of incoming migration. This makes it impossible for the source host to prepare for migration, for example to change storage locks from exclusive to shared mode when using shared LVM storage with lvmlockd. Add a hook call on the source host at the beginning of outgoing migration. The hook is called with "source" as the extra argument to distinguish it from the destination hook which uses "-". The same change is also applied to the libxl driver for consistency. Resolves: https://gitlab.com/libvirt/libvirt/-/issues/37 Signed-off-by: Guy Godfroy <guy.godfroy(a)gugod.fr> --- docs/hooks.rst | 36 ++++++++++++++++++++++++++++++++---- src/libxl/libxl_migration.c | 17 +++++++++++++++++ src/qemu/qemu_migration.c | 17 +++++++++++++++++ 3 files changed, 66 insertions(+), 4 deletions(-) diff --git a/docs/hooks.rst b/docs/hooks.rst index e1745b8cc7..4c7cac56ba 100644 --- a/docs/hooks.rst +++ b/docs/hooks.rst @@ -244,7 +244,7 @@ operation. There is no specific operation to indicate a "restart" is occurring. - :since:`Since 0.9.11`, the qemu hook script is also called at the beginning - of incoming migration. It is called as: + of incoming migration on the **destination** host. It is called as: :: @@ -257,6 +257,18 @@ operation. There is no specific operation to indicate a "restart" is occurring. is not valid, incoming migration will be canceled. This hook may be used, e.g., to change location of disk images for incoming domains. +- :since:`Since 12.1.0`, the qemu hook script is also called at the beginning + of outgoing migration on the **source** host. It is called as: + + :: + + /etc/libvirt/hooks/qemu guest_name migrate begin source + + with domain XML sent to standard input of the script. Unlike the destination + hook, this script does not act as a filter and any output is ignored. If the + script returns failure, the migration will be canceled. This hook may be used, + e.g., to change storage lock mode from exclusive to shared before migration. + - :since:`Since 1.2.9`, the qemu hook script is also called when restoring a saved image either via the API or automatically when restoring a managed save machine. It is called as: @@ -417,7 +429,7 @@ operation. There is no specific operation to indicate a "restart" is occurring. /etc/libvirt/hooks/libxl guest_name release end - - :since:`Since 2.1.0`, the libxl hook script is also called at the beginning - of incoming migration. It is called as: + of incoming migration on the **destination** host. It is called as: :: @@ -430,6 +442,18 @@ operation. There is no specific operation to indicate a "restart" is occurring. is not valid, incoming migration will be canceled. This hook may be used, e.g., to change location of disk images for incoming domains. +- :since:`Since 12.1.0`, the libxl hook script is also called at the beginning + of outgoing migration on the **source** host. It is called as: + + :: + + /etc/libvirt/hooks/libxl guest_name migrate begin source + + with domain XML sent to standard input of the script. Unlike the destination + hook, this script does not act as a filter and any output is ignored. If the + script returns failure, the migration will be canceled. This hook may be used, + e.g., to change storage lock mode from exclusive to shared before migration. + - :since:`Since 6.5.0`, you can also place several hook scripts in the directory ``/etc/libvirt/hooks/libxl.d/``. They are executed in alphabetical order after main script. In this case each script also acts as filter and can @@ -566,10 +590,14 @@ Migration of a QEMU guest involves running hook scripts on both the source and destination hosts: #. At the beginning of the migration, the *qemu* hook script on the - **destination** host is executed with the "migrate" operation. + **source** host is executed with the "migrate" operation and "source" + as the extra argument (:since:`since 12.1.0`). This allows the source + host to prepare for migration, e.g., changing storage locks. +#. Then, the *qemu* hook script on the **destination** host is executed + with the "migrate" operation. #. Before QEMU process is spawned, the two operations ("prepare" and "start") called for domain start are executed on **destination** host. -#. If both of these hook script executions exit successfully (exit status 0), +#. If all of these hook script executions exit successfully (exit status 0), the migration continues. Any other exit code indicates failure, and the migration is aborted. #. The QEMU guest is then migrated to the destination host. diff --git a/src/libxl/libxl_migration.c b/src/libxl/libxl_migration.c index f5dee7627b..76f3c7ceb7 100644 --- a/src/libxl/libxl_migration.c +++ b/src/libxl/libxl_migration.c @@ -411,6 +411,23 @@ libxlDomainMigrationSrcBegin(virConnectPtr conn, if (!libxlDomainMigrationIsAllowed(def)) goto endjob; + /* Call hook to allow source host to prepare for migration */ + if (virHookPresent(VIR_HOOK_DRIVER_LIBXL)) { + g_autofree char *hookxml = NULL; + int hookret; + + if (!(hookxml = virDomainDefFormat(def, driver->xmlopt, + VIR_DOMAIN_DEF_FORMAT_SECURE))) + goto endjob; + + hookret = virHookCall(VIR_HOOK_DRIVER_LIBXL, def->name, + VIR_HOOK_LIBXL_OP_MIGRATE, VIR_HOOK_SUBOP_BEGIN, + "source", hookxml, NULL); + + if (hookret < 0) + goto endjob; + } + xml = virDomainDefFormat(def, driver->xmlopt, VIR_DOMAIN_DEF_FORMAT_SECURE); /* Valid xml means success! EndJob in the confirm phase */ if (xml) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 6dd022163b..60e370319c 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -2791,6 +2791,23 @@ qemuMigrationSrcBeginPhase(virQEMUDriver *driver, if (!qemuMigrationSrcIsAllowed(vm, true, vm->job->asyncJob, flags)) return NULL; + /* Call hook to allow source host to prepare for migration */ + if (virHookPresent(VIR_HOOK_DRIVER_QEMU)) { + g_autofree char *xml = NULL; + int hookret; + + if (!(xml = qemuDomainDefFormatLive(driver, priv->qemuCaps, vm->def, + priv->origCPU, false, true))) + return NULL; + + hookret = virHookCall(VIR_HOOK_DRIVER_QEMU, vm->def->name, + VIR_HOOK_QEMU_OP_MIGRATE, VIR_HOOK_SUBOP_BEGIN, + "source", xml, NULL); + + if (hookret < 0) + return NULL; + } + if (!(flags & (VIR_MIGRATE_UNSAFE | VIR_MIGRATE_OFFLINE)) && !qemuMigrationSrcIsSafe(vm, migrate_disks, flags)) return NULL; -- 2.34.1

4 18

[PATCH] remote: Move secrets encryption dependency to a systemd drop-in
by Arun Menon 06 Apr '26

06 Apr '26

From: Arun Menon <armenon(a)redhat.com> The monolithic libvirtd.service currently has a dependency on virt-secret-init-encryption.service. This causes libvirtd to fail to start on systems where the secret driver is not installed or enabled, as systemd cannot satisfy the Requires= / After= units or the LoadCredentialEncrypted= path. See below, Requires=virt-secret-init-encryption.service After=virt-secret-init-encryption.service LoadCredentialEncrypted=secrets-encryption-key:@localstatedir@/lib/libvirt/secrets/secrets-encryption-key This patch decouples the secrets encryption logic from the main libvirtd service file. It is moved into a new systemd drop-in (50-libvirtd-secret.conf) which is only installed when libvirt is built with secret driver support. The override snippet is added to the daemon-driver-secret package. Fixes: 97758bc9a0b1fccf8c0009308658f1204b113b89 Signed-off-by: Arun Menon <armenon(a)redhat.com> Fix-Suggested-by: Andrea Bolognani <abologna(a)redhat.com> --- libvirt.spec.in | 2 ++ src/remote/libvirtd-secret.conf.in | 7 +++++++ src/remote/libvirtd.service.in | 4 ---- src/remote/meson.build | 15 +++++++++++++++ 4 files changed, 24 insertions(+), 4 deletions(-) create mode 100644 src/remote/libvirtd-secret.conf.in diff --git a/libvirt.spec.in b/libvirt.spec.in index 00316a03f2..d840c829d1 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -2259,6 +2259,8 @@ exit 0 %{_unitdir}/virtsecretd.socket %{_unitdir}/virtsecretd-ro.socket %{_unitdir}/virtsecretd-admin.socket +%dir %attr(0700, root, root) %{_unitdir}/libvirtd.service.d/ +%{_unitdir}/libvirtd.service.d/50-libvirtd-secret.conf %attr(0755, root, root) %{_sbindir}/virtsecretd %dir %attr(0700, root, root) %{_sysconfdir}/libvirt/secrets/ %dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt/secrets/ diff --git a/src/remote/libvirtd-secret.conf.in b/src/remote/libvirtd-secret.conf.in new file mode 100644 index 0000000000..d64a6cf63e --- /dev/null +++ b/src/remote/libvirtd-secret.conf.in @@ -0,0 +1,7 @@ +[Unit] +Requires=@service@ +After=@service@ + +[Service] +Environment=SECRETS_ENCRYPTION_KEY=%d/secrets-encryption-key +LoadCredentialEncrypted=secrets-encryption-key:@localstatedir@/lib/libvirt/secrets/secrets-encryption-key diff --git a/src/remote/libvirtd.service.in b/src/remote/libvirtd.service.in index 7965010a0a..b0a062e885 100644 --- a/src/remote/libvirtd.service.in +++ b/src/remote/libvirtd.service.in @@ -12,8 +12,6 @@ After=libvirtd.socket After=libvirtd-ro.socket After=libvirtd-admin.socket Requires=virtlogd.socket -Requires=virt-secret-init-encryption.service -After=virt-secret-init-encryption.service Wants=virtlockd.socket After=virtlogd.socket After=virtlockd.socket @@ -31,8 +29,6 @@ Conflicts=xendomains.service Type=notify-reload Environment=LIBVIRTD_ARGS="--timeout 120" EnvironmentFile=-@initconfdir@/libvirtd -Environment=SECRETS_ENCRYPTION_KEY=%d/secrets-encryption-key -LoadCredentialEncrypted=secrets-encryption-key:@localstatedir@/lib/libvirt/secrets/secrets-encryption-key ExecStart=@sbindir@/libvirtd $LIBVIRTD_ARGS ExecReload=/bin/kill -HUP $MAINPID KillMode=process diff --git a/src/remote/meson.build b/src/remote/meson.build index e503263266..f1c521444f 100644 --- a/src/remote/meson.build +++ b/src/remote/meson.build @@ -343,4 +343,19 @@ if conf.has('WITH_SASL') ) endif +# The monolithic libvirt daemon only attempts to load the +# secrets encryption credentials if the secret driver is enabled +if conf.has('WITH_SECRETS') + secret_dropin_conf = configuration_data() + secret_dropin_conf.set('service', 'virt-secret-init-encryption.service') + secret_dropin_conf.set('localstatedir', localstatedir) + + configure_file( + input: 'libvirtd-secret.conf.in', + output: '50-libvirtd-secret.conf', + configuration: secret_dropin_conf, + install_dir: unitdir / 'libvirtd.service.d' + ) +endif + remote_inc_dir = include_directories('.') -- 2.53.0

2 2

Re: [PATCH v3 07/11] ui/dbus: run without Audio interface when audio is disabled
by Markus Armbruster 01 Apr '26

01 Apr '26

Sergei Heifetz <heifetz(a)yandex-team.com> writes: > D-Bus display can be used even when QEMU is configured with > `--disable-audio`. In that case, audio interface will not be available > on `/org/qemu/Display1/Audio`. > > (The current handling of the situation when audio is enabled but > no D-Bus-compatible audio backend is available is different and > hasn’t been changed.) > > Signed-off-by: Sergei Heifetz <heifetz(a)yandex-team.com> > --- > qapi/ui.json | 2 +- > qemu-options.hx | 4 ++++ > ui/dbus.c | 8 ++++++++ > 3 files changed, 13 insertions(+), 1 deletion(-) > > diff --git a/qapi/ui.json b/qapi/ui.json > index e3da77632a8..62eb05c6091 100644 > --- a/qapi/ui.json > +++ b/qapi/ui.json > @@ -1385,7 +1385,7 @@ > 'data' : { '*rendernode' : 'str', > '*addr': 'str', > '*p2p': 'bool', > - '*audiodev': 'str' } } > + '*audiodev': { 'type': 'str', 'if': 'CONFIG_AUDIO' } } } Is this a compatibility break? DisplayDBus is only used in DisplayOptions, which is only used as return type of query-display-options in the QAPI schema. However, vl.c also uses it to parse -display. Removing parameter @audiodev from -display is technically a compatibility break. However, we're removing an optional paramater that would always be rejected when present: it has no valid value, because there are no audiodevs. Removal from query-display-options is similar: it could not be present. Removal affects introspection, which could conceivably confuse client. I don't expect such trouble here, because @audiodev always existed, i.e. there's no real reason to look for it in introspection. I think we're okay. I'm cc'ing the libvirt list just in case. Do we need a release note? Do we need an entry in docs/about/removed-features.rst? Feels a bit odd; we're not removing a feature, we're adding one, namely "can compile out audio". > > ## > # @DisplayGLMode: > diff --git a/qemu-options.hx b/qemu-options.hx > index 29dc865820a..20d15c0004f 100644 > --- a/qemu-options.hx > +++ b/qemu-options.hx > @@ -2232,6 +2232,10 @@ SRST > If no audio backend is specified and the dbus display backend is > specified, the dbus audio backend is used by default. > > + (If audio is disabled at compile time via the ``--disable-audio`` option, > + no audio backend is used, and none of the audio-related D-Bus interfaces > + are exported.) > + Well, audio is either compiled in or it isn't. If it's compiled in, then why tell the user how things behave when it's not? If it's not compiled in, then why document stuff that doesn't work? How much of -audiodev still works then? -audiodev none? Anything else? > The connection is registered with the "org.qemu" name (and queued when > already owned). > [...]

3 4

Re: [PATCH RFC 02/12] migration/qapi: Rename MigrationStats to MigrationRAMStats
by Markus Armbruster 01 Apr '26

01 Apr '26

Cc: correct libvirt list address Peter Xu <peterx(a)redhat.com> writes: > This stats is only about RAM, so make it accurate. This paves way for > statistics for all devices. > > Note that even if this is part of qapi/, this should not be part of ABI of docs/devel/qapi-code-gen.rst section "Compatibility considerations": Since type names are not visible in the Client JSON Protocol, types may be freely renamed. Even certain refactorings are invisible, such as splitting members from one type into a common base type. So, s/should not be/is not/. > at least query-migrate, because the structure is not changed, and this > stats is always reported only under the "ram" section. Why "at least query-migrate"? Do you have other interfaces in mind, or are you just hedging? > Cc: Daniel P. Berrangé <berrange(a)redhat.com> > Cc: Markus Armbruster <armbru(a)redhat.com> > Cc: Libvirt Mailing List <libvir-list(a)redhat.com> > Signed-off-by: Peter Xu <peterx(a)redhat.com> > --- > docs/about/removed-features.rst | 2 +- > qapi/migration.json | 8 ++++---- > migration/migration-stats.h | 2 +- > 3 files changed, 6 insertions(+), 6 deletions(-) > > diff --git a/docs/about/removed-features.rst b/docs/about/removed-features.rst > index 6f4447993c..7c4f4325f7 100644 > --- a/docs/about/removed-features.rst > +++ b/docs/about/removed-features.rst > @@ -699,7 +699,7 @@ was superseded by ``sections``. > ``query-migrate`` return value member ``skipped`` (removed in 9.1) > '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' > > -Member ``skipped`` of the ``MigrationStats`` struct hasn't been used > +Member ``skipped`` of the ``MigrationRAMStats`` struct hasn't been used > for more than 10 years. Removed with no replacement. Could simply use "of the return value". But this is fine, too. > > ``migrate`` command option ``inc`` (removed in 9.1) > diff --git a/qapi/migration.json b/qapi/migration.json > index 7134d4ce47..cfc6ccee26 100644 > --- a/qapi/migration.json > +++ b/qapi/migration.json > @@ -12,7 +12,7 @@ > { 'include': 'sockets.json' } > > ## > -# @MigrationStats: > +# @MigrationRAMStats: > # > # Detailed migration status. > # > @@ -64,7 +64,7 @@ > # > # Since: 0.14 > ## > -{ 'struct': 'MigrationStats', > +{ 'struct': 'MigrationRAMStats', > 'data': {'transferred': 'int', 'remaining': 'int', 'total': 'int' , > 'duplicate': 'int', > 'normal': 'int', > @@ -209,7 +209,7 @@ > # If this field is not returned, no migration process has been > # initiated > # > -# @ram: `MigrationStats` containing detailed migration status, only > +# @ram: `MigrationRAMStats` containing detailed migration status, only > # returned if status is 'active' or 'completed'(since 1.2) This gets rendered like * ram (MigrationStats, *optional*) -- MigrationStats containing detailed migration status, only returned if status is 'active' or 'completed'(since 1.2) Recommend to rephrase to avoid the repetition. Maybe # @ram: Detailed migration RAM status, only returned if status is # 'active' or 'completed' (since 1.2) Also consider replacing "status" by "statistics". > # > # @xbzrle-cache: `XBZRLECacheStats` containing detailed XBZRLE > @@ -309,7 +309,7 @@ > # Since: 0.14 > ## > { 'struct': 'MigrationInfo', > - 'data': {'*status': 'MigrationStatus', '*ram': 'MigrationStats', > + 'data': {'*status': 'MigrationStatus', '*ram': 'MigrationRAMStats', > '*vfio': 'VfioStats', > '*xbzrle-cache': 'XBZRLECacheStats', > '*total-time': 'int', > diff --git a/migration/migration-stats.h b/migration/migration-stats.h > index c0f50144c9..1153520f7a 100644 > --- a/migration/migration-stats.h > +++ b/migration/migration-stats.h > @@ -27,7 +27,7 @@ > > /* > * These are the ram migration statistic counters. It is loosely > - * based on MigrationStats. > + * based on MigrationRAMStats. > */ > typedef struct { > /* Only minor issues, and none added by this patch, so Acked-by: Markus Armbruster <armbru(a)redhat.com>

2 2