- Devel - libvirt List Archives

[libvirt] [PATCH 0/2] improve detection of UNIX path generated by libvirt
by Pavel Hrdina 12 May '17

12 May '17

Pavel Hrdina (2): util: introduce virStringMatch qemu: improve detection of UNIX path generated by libvirt src/libvirt_private.syms | 1 + src/qemu/qemu_domain.c | 78 +++++++++++++++++++--- src/util/virstring.c | 34 ++++++++++ src/util/virstring.h | 3 + .../qemuxml2argv-channel-unix-gen-path1.xml | 17 +++++ .../qemuxml2argv-channel-unix-gen-path2.xml | 17 +++++ .../qemuxml2argv-channel-unix-gen-path3.xml | 17 +++++ .../qemuxml2argv-channel-unix-user-path.xml | 17 +++++ .../qemuxml2xmlout-channel-unix-gen-path1.xml | 32 +++++++++ .../qemuxml2xmlout-channel-unix-gen-path2.xml | 32 +++++++++ .../qemuxml2xmlout-channel-unix-gen-path3.xml | 32 +++++++++ .../qemuxml2xmlout-channel-unix-user-path.xml | 33 +++++++++ tests/qemuxml2xmltest.c | 5 ++ tests/virstringtest.c | 47 +++++++++++++ 14 files changed, 356 insertions(+), 9 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-channel-unix-gen-path1.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-channel-unix-gen-path2.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-channel-unix-gen-path3.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-channel-unix-user-path.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-channel-unix-gen-path1.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-channel-unix-gen-path2.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-channel-unix-gen-path3.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-channel-unix-user-path.xml -- 2.12.2

3 9

[libvirt] [PATCH] conf: Fix resource leak in virCapabilitiesInitCaches
by John Ferlan 12 May '17

12 May '17

The @type from virFileReadValueString needs to be VIR_FREE each time through the loop since it's not saved and since cleanup can be reached prior to decoding it for @kernel_type amd bank->type, the cleanup code needs to also have a VIR_FREE Found by Coverity Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/conf/capabilities.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/conf/capabilities.c b/src/conf/capabilities.c index c36ca40..d699b08 100644 --- a/src/conf/capabilities.c +++ b/src/conf/capabilities.c @@ -1611,10 +1611,10 @@ virCapabilitiesInitCaches(virCapsPtr caps) if (kernel_type < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Unknown cache type '%s'"), type); - VIR_FREE(type); goto cleanup; } bank->type = kernel_type; + VIR_FREE(type); for (i = 0; i < caps->host.ncaches; i++) { if (virCapsHostCacheBankEquals(bank, caps->host.caches[i])) @@ -1637,6 +1637,7 @@ virCapabilitiesInitCaches(virCapsPtr caps) ret = 0; cleanup: + VIR_FREE(type); VIR_FREE(path); virDirClose(&dirp); virCapsHostCacheBankFree(bank); -- 2.9.3

2 1

[libvirt] [PATCH v2] qemu: reduce packet loss rate for vm with macvtap passthrough mode in migration
by ZhiPeng Lu 12 May '17

12 May '17

Before libvirt that calls virNetDevMacVLanCreateWithVPortProfile sets mac address or vlan of a Virtual Function(VF) linked to a macvtap passthrough device of migration destination host in migration start step. If we ping the migrating vm, we get the network does not pass. Because VFs of migration source and destination have the same MAC address. The patch later calling qemuMigrationVPAssociatePortProfiles sets mac address of VF in migration finish step instead of start step. The patch aims to reduce packet loss rate. Signed-off-by: ZhiPeng Lu <lu.zhipeng(a)zte.com.cn> --- src/qemu/qemu_migration.c | 18 ++++++++++++++++-- src/util/virnetdevmacvlan.c | 17 +++++++++++------ 2 files changed, 27 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 09adb04..795ed71 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -5004,7 +5004,7 @@ qemuMigrationPerform(virQEMUDriverPtr driver, } static int -qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def) +qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def, const char *stateDir) { size_t i; int last_good_net = -1; @@ -5013,6 +5013,20 @@ qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def) for (i = 0; i < def->nnets; i++) { net = def->nets[i]; if (virDomainNetGetActualType(net) == VIR_DOMAIN_NET_TYPE_DIRECT) { + if ((!virDomainNetGetActualVirtPortProfile(net) || (virDomainNetGetActualVirtPortProfile(net) && + virDomainNetGetActualVirtPortProfile(net)->virtPortType != VIR_NETDEV_VPORT_PROFILE_8021QBG && + virDomainNetGetActualVirtPortProfile(net)->virtPortType != VIR_NETDEV_VPORT_PROFILE_8021QBH)) && + virDomainNetGetActualDirectMode(net) == + VIR_NETDEV_MACVLAN_MODE_PASSTHRU) { + if (virNetDevSaveNetConfig(virDomainNetGetActualDirectDev(net), + -1, stateDir, false) < 0) { + goto err_exit; + } + if (virNetDevSetNetConfig(virDomainNetGetActualDirectDev(net), + -1, NULL, virDomainNetGetActualVlan(net), &net->mac, false) < 0) { + goto err_exit; + } + } if (virNetDevVPortProfileAssociate(net->ifname, virDomainNetGetActualVirtPortProfile(net), &net->mac, @@ -5187,7 +5201,7 @@ qemuMigrationFinish(virQEMUDriverPtr driver, goto endjob; } - if (qemuMigrationVPAssociatePortProfiles(vm->def) < 0) + if (qemuMigrationVPAssociatePortProfiles(vm->def, cfg->stateDir) < 0) goto endjob; if (mig->network && qemuDomainMigrateOPDRelocate(driver, vm, mig) < 0) diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index 7222b0f..682dcd1 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -1020,12 +1020,17 @@ virNetDevMacVLanCreateWithVPortProfile(const char *ifnameRequested, */ setVlan = false; } - - if (virNetDevSaveNetConfig(linkdev, -1, stateDir, setVlan) < 0) - return -1; - - if (virNetDevSetNetConfig(linkdev, -1, NULL, vlan, macaddress, setVlan) < 0) - return -1; + if (vmOp != VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START && + virtPortProfile && (virtPortProfile->virtPortType == VIR_NETDEV_VPORT_PROFILE_8021QBH || + virtPortProfile->virtPortType == VIR_NETDEV_VPORT_PROFILE_8021QBG)) { + if (virNetDevSaveNetConfig(linkdev, -1, stateDir, setVlan) < 0) { + return -1; + } + if (virNetDevSetNetConfig(linkdev, -1, NULL, vlan, macaddress, + setVlan) < 0) { + return -1; + } + } } if (ifnameRequested) { -- 1.8.3.1

1 0

[libvirt] [PATCH] qemu: reduce packet loss rate for vm with macvtap passthrough mode in migration
by ZhiPeng Lu 12 May '17

12 May '17

Before libvirt that calls virNetDevMacVLanCreateWithVPortProfile sets mac address or vlan of a Virtual Function(VF) linked to a macvtap passthrough device of migration destination host in migration start step. If we ping the migrating vm, we get the network does not pass. Because VFs of migration source and destination have the same MAC address. The patch later calling qemuMigrationVPAssociatePortProfiles sets mac address of VF in migration finish step instead of start step. The patch aims to reduce packet loss rate. Signed-off-by: ZhiPeng Lu <lu.zhipeng(a)zte.com.cn> --- src/qemu/qemu_migration.c | 17 +++++++++++++++-- src/util/virnetdevmacvlan.c | 17 +++++++++++------ 2 files changed, 26 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 09adb04..0308043 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -5004,7 +5004,7 @@ qemuMigrationPerform(virQEMUDriverPtr driver, } static int -qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def) +qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def, const char *stateDir) { size_t i; int last_good_net = -1; @@ -5013,6 +5013,19 @@ qemuMigrationVPAssociatePortProfiles(virDomainDefPtr def) for (i = 0; i < def->nnets; i++) { net = def->nets[i]; if (virDomainNetGetActualType(net) == VIR_DOMAIN_NET_TYPE_DIRECT) { + if (virDomainNetGetActualVirtPortProfile(net) != VIR_NETDEV_VPORT_PROFILE_8021QBG && + virDomainNetGetActualVirtPortProfile(net) != VIR_NETDEV_VPORT_PROFILE_8021QBH && + (virDomainNetGetActualDirectMode(net) == + VIR_NETDEV_MACVLAN_MODE_PASSTHRU)) { + if (virNetDevSaveNetConfig(virDomainNetGetActualDirectDev(net), + -1, stateDir, false) < 0) { + goto err_exit; + } + if (virNetDevSetNetConfig(virDomainNetGetActualDirectDev(net), + -1, NULL, virDomainNetGetActualVlan(net), &net->mac, false) < 0) { + goto err_exit; + } + } if (virNetDevVPortProfileAssociate(net->ifname, virDomainNetGetActualVirtPortProfile(net), &net->mac, @@ -5187,7 +5200,7 @@ qemuMigrationFinish(virQEMUDriverPtr driver, goto endjob; } - if (qemuMigrationVPAssociatePortProfiles(vm->def) < 0) + if (qemuMigrationVPAssociatePortProfiles(vm->def, cfg->stateDir) < 0) goto endjob; if (mig->network && qemuDomainMigrateOPDRelocate(driver, vm, mig) < 0) diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index 7222b0f..fc08c43 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -1020,12 +1020,17 @@ virNetDevMacVLanCreateWithVPortProfile(const char *ifnameRequested, */ setVlan = false; } - - if (virNetDevSaveNetConfig(linkdev, -1, stateDir, setVlan) < 0) - return -1; - - if (virNetDevSetNetConfig(linkdev, -1, NULL, vlan, macaddress, setVlan) < 0) - return -1; + if (vmOp != VIR_NETDEV_VPORT_PROFILE_OP_MIGRATE_IN_START && + (virtPortProfile == VIR_NETDEV_VPORT_PROFILE_8021QBH || + virtPortProfile == VIR_NETDEV_VPORT_PROFILE_8021QBG)) { + if (virNetDevSaveNetConfig(linkdev, -1, stateDir, setVlan) < 0) { + return -1; + } + if (virNetDevSetNetConfig(linkdev, -1, NULL, vlan, macaddress, + setVlan) < 0) { + return -1; + } + } } if (ifnameRequested) { -- 1.8.3.1

1 0

[libvirt] [PATCH v2] Detect VMDK version 3 files
by Daniel P. Berrange 12 May '17

12 May '17

The metadata libvirt cares about is identical for version 3 as for previous versions, so we merely need list the new version number. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/util/virstoragefile.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index 94a77ce..b43acf6 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -113,7 +113,7 @@ enum { BACKING_STORE_ERROR, }; -#define FILE_TYPE_VERSIONS_LAST 2 +#define FILE_TYPE_VERSIONS_LAST 3 struct FileEncryptionInfo { int format; /* Encryption format to assign */ @@ -374,7 +374,7 @@ static struct FileTypeInfo const fileTypeInfo[] = { }, [VIR_STORAGE_FILE_VMDK] = { 0, "KDMV", NULL, - LV_LITTLE_ENDIAN, 4, 4, {1, 2}, + LV_LITTLE_ENDIAN, 4, 4, {1, 2, 3}, 4+4+4, 8, 512, NULL, vmdk4GetBackingStore, NULL }, }; -- 2.9.3

2 1

[libvirt] [PATCH] Add "io" option to virsh attach-disk sub-command.
by Gordon Messmer 12 May '17

12 May '17

--- tools/virsh-domain.c | 14 +++++++++++--- tools/virsh.pod | 2 ++ 2 files changed, 13 insertions(+), 3 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 0d19d0e..d2a2a05 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -267,6 +267,10 @@ static const vshCmdOptDef opts_attach_disk[] = { .type = VSH_OT_STRING, .help = N_("cache mode of disk device") }, + {.name = "io", + .type = VSH_OT_STRING, + .help = N_("io mode of disk device") + }, {.name = "type", .type = VSH_OT_STRING, .help = N_("target device type") @@ -504,8 +508,9 @@ cmdAttachDisk(vshControl *ctl, const vshCmd *cmd) virDomainPtr dom = NULL; const char *source = NULL, *target = NULL, *driver = NULL, *subdriver = NULL, *type = NULL, *mode = NULL, - *iothread = NULL, *cache = NULL, *serial = NULL, - *straddr = NULL, *wwn = NULL, *targetbus = NULL; + *iothread = NULL, *cache = NULL, *io = NULL, + *serial = NULL, *straddr = NULL, *wwn = NULL, + *targetbus = NULL; struct DiskAddress diskAddr; bool isFile = false, functionReturn = false; int ret; @@ -537,6 +542,7 @@ cmdAttachDisk(vshControl *ctl, const vshCmd *cmd) vshCommandOptStringReq(ctl, cmd, "mode", &mode) < 0 || vshCommandOptStringReq(ctl, cmd, "iothread", &iothread) < 0 || vshCommandOptStringReq(ctl, cmd, "cache", &cache) < 0 || + vshCommandOptStringReq(ctl, cmd, "io", &io) < 0 || vshCommandOptStringReq(ctl, cmd, "serial", &serial) < 0 || vshCommandOptStringReq(ctl, cmd, "wwn", &wwn) < 0 || vshCommandOptStringReq(ctl, cmd, "address", &straddr) < 0 || @@ -579,7 +585,7 @@ cmdAttachDisk(vshControl *ctl, const vshCmd *cmd) virBufferAddLit(&buf, ">\n"); virBufferAdjustIndent(&buf, 2); - if (driver || subdriver || iothread || cache) { + if (driver || subdriver || iothread || cache || io) { virBufferAddLit(&buf, "<driver"); if (driver) @@ -590,6 +596,8 @@ cmdAttachDisk(vshControl *ctl, const vshCmd *cmd) virBufferAsprintf(&buf, " iothread='%s'", iothread); if (cache) virBufferAsprintf(&buf, " cache='%s'", cache); + if (io) + virBufferAsprintf(&buf, " io='%s'", io); virBufferAddLit(&buf, "/>\n"); } diff --git a/tools/virsh.pod b/tools/virsh.pod index cd1f25f..9656411 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -2860,6 +2860,8 @@ I<mode> can specify the two specific mode I<readonly> or I<shareable>. I<sourcetype> can indicate the type of source (block|file) I<cache> can be one of "default", "none", "writethrough", "writeback", "directsync" or "unsafe". +I<io> io is "threads", or "native" and selects between pthread based disk +I/O and native Linux AIO. I<iothread> is the number within the range of domain IOThreads to which this disk may be attached (QEMU only). I<serial> is the serial of disk device. I<wwn> is the wwn of disk device. -- 2.9.3

3 8

[libvirt] [PATCH 0/2] spec: Minor improvements
by Andrea Bolognani 11 May '17

11 May '17

*** BLURBY MCBLURBFACE *** Andrea Bolognani (2): spec: Support maintenance releases on mingw spec: Use HTTPS instead of HTTP libvirt.spec.in | 4 ++-- mingw-libvirt.spec.in | 8 ++++++-- 2 files changed, 8 insertions(+), 4 deletions(-) -- 2.7.4

2 3

[libvirt] [PATCH] qemu: Provide a much clearer message on device hot-plug
by Erik Skultety 11 May '17

11 May '17

Adjust the current message to make it clear, that it is the hot-plug operation that is unsupported with the given host device type. https://bugzilla.redhat.com/show_bug.cgi?id=1450072 Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- src/qemu/qemu_hotplug.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 4ca10863c..1b8504b13 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -2674,7 +2674,7 @@ qemuDomainAttachHostDevice(virConnectPtr conn, { if (hostdev->mode != VIR_DOMAIN_HOSTDEV_MODE_SUBSYS) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("hostdev mode '%s' not supported"), + _("hot-plug is not supported for hostdev mode '%s'"), virDomainHostdevModeTypeToString(hostdev->mode)); return -1; } @@ -2705,7 +2705,8 @@ qemuDomainAttachHostDevice(virConnectPtr conn, default: virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("hostdev subsys type '%s' not supported"), + _("hot-plug is not supported for hostdev subsys type " + "'%s'"), virDomainHostdevSubsysTypeToString(hostdev->source.subsys.type)); goto error; } @@ -4820,7 +4821,8 @@ qemuDomainDetachThisHostDevice(virQEMUDriverPtr driver, break; default: virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("hostdev subsys type '%s' not supported"), + _("hot-unplug is not supported for hostdev subsys type " + "'%s'"), virDomainHostdevSubsysTypeToString(detach->source.subsys.type)); return -1; } @@ -4852,7 +4854,7 @@ int qemuDomainDetachHostDevice(virQEMUDriverPtr driver, if (hostdev->mode != VIR_DOMAIN_HOSTDEV_MODE_SUBSYS) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("hostdev mode '%s' not supported"), + _("hot-unplug is not supported for hostdev mode '%s'"), virDomainHostdevModeTypeToString(hostdev->mode)); return -1; } -- 2.12.2

2 1

[libvirt] ANNOUNCE: libvirt 1.3.3.3 maintenance release
by Cole Robinson 11 May '17

11 May '17

libvirt 1.3.3.3 is now available. This is a maintenance release of libvirt 1.3.3 with additional bugfixes that have accumulated upstream since the initial release. This release can be downloaded at: http://libvirt.org/sources/stable_updates/libvirt-1.3.3.3.tar.gz Changes in this version: * virtlogd: Don't stop or restart along with libvirtd * virtlogd.socket: Tie lifecycle to libvirtd.service * schema: Don't validate paths * maint: fix syntax-check sc_prohibit_int_ijk exclude rule * util: bitmap: clarify virBitmapLastSetBit() behavior for empty bitmaps * Fix building with -Og * qemu: Only use memory-backend-file with NUMA if needed For info about past maintenance releases, see: http://wiki.libvirt.org/page/Maintenance_Releases Thanks, Cole

1 0

[libvirt] ANNOUNCE: libvirt 2.2.1 maintenance release
by Cole Robinson 11 May '17

11 May '17

libvirt 2.2.1 is now available. This is a maintenance release of libvirt 2.2 with additional bugfixes that have accumulated upstream since the initial release. This release can be downloaded at: http://libvirt.org/sources/stable_updates/libvirt-2.2.1.tar.xz Changes in this version: * spec: Avoid RPM verification errors on nwfilter XMLs * qemu_process: spice: don't release used port * qemu: Fix crash during qemuStateCleanup * daemon: Fix crash during daemon cleanup * Fix crash on usb-serial hotplug * qemuBuildMemoryBackendStr: Don't crash if no hugetlbfs is mounted * util: fix crash in virClassIsDerivedFrom for CloseCallbacks objects * storage: driver: Remove unavailable transient pools after restart * storage: driver: Split out code fixing pool state after deactivation * qemu: Don't assume secret provided for LUKS encryption * conf: do not steal pointers from the pool source * schema: do not require name for certain pool types * virtlogd: Don't stop or restart along with libvirtd * virtlogd.socket: Tie lifecycle to libvirtd.service * spec: Update version check for maint Source URL * qemu: capabilities: Don't partially reprope caps on process reconnect * network: fix endless loop when starting network with multiple IPs and no dhcp * qemu: allow 32 slots on pcie-expander-bus, not just 1 * qemu: Only use memory-backend-file with NUMA if needed For info about past maintenance releases, see: http://wiki.libvirt.org/page/Maintenance_Releases Thanks, Cole

1 0