- Devel - libvirt List Archives

[libvirt] [PATCH] rpc : fix a access for null pointer
by Peng Hao 24 Jul '17

24 Jul '17

virNetSocketRemoveIOCallback get sock's ObjectLock and will call virNetSocketEventFree. virNetSocketEventFree may be free sock object and virNetSocketRemoveIOCallback will access a null pointer in release sock's ObjectLock. Signed-off-by: Liu Yun <liu.yunh(a)zte.com.cn> Signed-off-by: Peng Hao <peng.hao2(a)zte.com.cn> --- src/rpc/virnetsocket.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c index d228c8a..8b550e8 100644 --- a/src/rpc/virnetsocket.c +++ b/src/rpc/virnetsocket.c @@ -2140,14 +2140,12 @@ static void virNetSocketEventFree(void *opaque) virFreeCallback ff; void *eopaque; - virObjectLock(sock); ff = sock->ff; eopaque = sock->opaque; sock->func = NULL; sock->ff = NULL; sock->opaque = NULL; - virObjectUnlock(sock); - + if (ff) ff(eopaque); @@ -2207,6 +2205,7 @@ void virNetSocketUpdateIOCallback(virNetSocketPtr sock, void virNetSocketRemoveIOCallback(virNetSocketPtr sock) { + virObjectRef(sock); virObjectLock(sock); if (sock->watch < 0) { @@ -2220,6 +2219,7 @@ void virNetSocketRemoveIOCallback(virNetSocketPtr sock) sock->watch = -1; virObjectUnlock(sock); + virObjectRef(sock); } void virNetSocketClose(virNetSocketPtr sock) -- 1.8.3.1

3 5

[libvirt] [PATCH] conf: Fix backwards migration of pSeries guests
by Andrea Bolognani 24 Jul '17

24 Jul '17

Recent commits made it so that pci-root controllers for pSeries guests are automatically assigned the spapr-pci-host-bridge model name; however, that prevents guests to migrate to older versions of libvirt which don't know about that model name at all, which at the moment is all of them :) To avoid the issue, just strip the model name from PHBs when formatting the migratable XML; guests that use more than one PHB are not going to be migratable anyway. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- src/conf/domain_conf.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 9320794..21bd7c7 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -21919,7 +21919,20 @@ virDomainControllerDefFormat(virBufferPtr buf, } if (def->type == VIR_DOMAIN_CONTROLLER_TYPE_PCI) { - if (def->opts.pciopts.modelName != VIR_DOMAIN_CONTROLLER_PCI_MODEL_NAME_NONE) { + bool formatModelName = true; + + if (def->opts.pciopts.modelName == VIR_DOMAIN_CONTROLLER_PCI_MODEL_NAME_NONE) + formatModelName = false; + + /* Don't format the model name for PHBs when migrating so that + * guests that only use the default one can be migrated to older + * libvirt version which don't know about PHBs at all */ + if (virDomainControllerIsPCIHostBridge(def) && + flags & VIR_DOMAIN_DEF_FORMAT_MIGRATABLE) { + formatModelName = false; + } + + if (formatModelName) { modelName = virDomainControllerPCIModelNameTypeToString(def->opts.pciopts.modelName); if (!modelName) { virReportError(VIR_ERR_INTERNAL_ERROR, -- 2.7.5

2 7

[libvirt] [PATCH 0/2] docs: Fix enum documentation
by Michal Privoznik 24 Jul '17

24 Jul '17

So this patch sent to the list got me roll up my sleeves and get working: https://www.redhat.com/archives/libvir-list/2017-July/msg00835.html It wasn't that bad after all. Michal Privoznik (2): apibuild.py: Handle enum comments properly docs: Span cells if there's not doc text for enum val docs/apibuild.py | 8 +++++++- docs/newapi.xsl | 25 +++++++++++++++---------- 2 files changed, 22 insertions(+), 11 deletions(-) -- 2.13.0

3 6

[libvirt] An API to copy files.
by Julio Faracco 24 Jul '17

24 Jul '17

Hi guys, Unfortunately, my company does not allow access to IRC. I need to ask question by email. Why doesn't libvirt have any method to copy files? A kind of virFileCopy(src, dst) for example. Libvirt has a lots of helper functions but it does not have a specific method to copy a file. There is a similar method inside openvz_conf.c. But I don't see any method inside virfile.c -- Julio Cesar Faracco

3 2

[libvirt] [PATCH v2 0/6] Convert port to unsigned integer
by Peter Krempa 24 Jul '17

24 Jul '17

Since Michal did not want to finis this I did. I also found a few bugs in my previous patches which are addressed here. Peter Krempa (6): qemu: command: Rename and move qemuNetworkDriveGetPort util: uri: Convert port number to unsigned integer qemu: command: Remove condition to use default sheepdog port util: storage: fill in default ports when parsing backing chain conf: domain: Split up virDomainStorageHostParse and rename it virStorageNetHostDef: Turn @port into integer src/conf/domain_conf.c | 133 +++++++++++++++++++--------------- src/libvirt_private.syms | 1 + src/libxl/libxl_conf.c | 2 +- src/qemu/qemu_block.c | 7 +- src/qemu/qemu_command.c | 30 ++------ src/qemu/qemu_parse_command.c | 13 ++-- src/storage/storage_backend_gluster.c | 17 +---- src/storage/storage_driver.c | 7 +- src/util/virstoragefile.c | 68 +++++++---------- src/util/virstoragefile.h | 4 +- src/util/virstring.c | 37 ++++++++++ src/util/virstring.h | 4 + src/util/viruri.h | 2 +- src/xenconfig/xen_xl.c | 2 +- tests/virstoragetest.c | 14 ++-- 15 files changed, 182 insertions(+), 159 deletions(-) -- 2.13.2

3 9

[libvirt] [PATCH v3 0/2] qemu: Add AAVMF32 to the list of known UEFIs
by dann frazier 24 Jul '17

24 Jul '17

Changes since v2: * Include updates to qemu.conf * Update tests for parity w/ AAVMF * Drop reference to 'experimental' in commit message Changes since v1: * Add a virt-aa-helper-test case. dann frazier (2): qemu: Add AAVMF32 to the list of known UEFIs virt-aa-helper-test: Add test for aarch32 UEFI image path src/qemu/qemu.conf | 3 ++- src/qemu/qemu_conf.c | 12 ++++++++---- src/qemu/test_libvirtd_qemu.aug.in | 1 + tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.7.0.s390x.xml | 1 + tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.8.0.s390x.xml | 1 + tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.9.0-q35.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml | 1 + tests/domaincapstest.c | 1 + tests/virt-aa-helper-test | 1 + 18 files changed, 26 insertions(+), 5 deletions(-) -- 2.13.2

3 7

[libvirt] [PATCH v5 0/3] Make virNodeDeviceObj and virNodeDeviceObjList private
by John Ferlan 24 Jul '17

24 Jul '17

NB: Kept same title for consistency, but it's less true as what's left deals with using the now provide ObjList in order to change from a forward linked list to a hash lookup (thus conceivably speeding up searches *and* removing the need for driver level locks since the ObjList is self lockable. v4: https://www.redhat.com/archives/libvir-list/2017-July/msg00034.html Changes since v4: -> Patches 1-12 from v4 pushed -> Patch 1 (NEW) - Fix the virNodeDeviceObjListGetParentHost to not pass @def, rather pass the individual names being searched since the Deletion code needs to unlock the object and thus @def prior to calling and that could result in a possible, but improbable udevAddOneDevice change event that would have replaced @def. -> Patch 2 (former patch 13) - No change -> Patch 3 (former patch 14) - No change John Ferlan (3): nodedev: Fix usage of virNodeDeviceObjListGetParentHost nodedev: Convert virNodeDeviceObjListPtr to use hash tables nodedev: Remove driver locks around object list mgmt code src/conf/virnodedeviceobj.c | 620 ++++++++++++++++++++++++----------- src/conf/virnodedeviceobj.h | 8 +- src/node_device/node_device_driver.c | 101 +++--- src/node_device/node_device_hal.c | 16 +- src/node_device/node_device_udev.c | 13 +- src/test/test_driver.c | 41 ++- 6 files changed, 523 insertions(+), 276 deletions(-) -- 2.9.4

2 10

[libvirt] [PATCH] qemu_capabilities: Honour caps values formatting
by Michal Privoznik 24 Jul '17

24 Jul '17

So the way we format this huge virQEMUCaps enum is we group the values in groups of five. And then at the beginning of each group we have a small comment that says what's the number of the first item in the group. Well, the last commit of 11b2ebf3e152 does not follow this formatting. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_capabilities.c | 2 ++ src/qemu/qemu_capabilities.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index c3f49a975..5fb6fd4ba 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -431,6 +431,8 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "virtio.ats", "loadparm", "spapr-pci-host-bridge", + + /* 265 */ "spapr-pci-host-bridge.numa_node", ); diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 19f59a6d4..746745e39 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -417,6 +417,8 @@ typedef enum { QEMU_CAPS_VIRTIO_PCI_ATS, /* virtio-*-pci.ats */ QEMU_CAPS_LOADPARM, /* -machine loadparm */ QEMU_CAPS_DEVICE_SPAPR_PCI_HOST_BRIDGE, /* -device spapr-pci-host-bridge */ + + /* 265 */ QEMU_CAPS_SPAPR_PCI_HOST_BRIDGE_NUMA_NODE, /* spapr-pci-host-bridge.numa_node= */ QEMU_CAPS_LAST /* this must always be the last item */ -- 2.13.0

2 1

[libvirt] Error starting RBD storage pool
by Oscar Segarra 23 Jul '17

23 Jul '17

Hi, In my environment I have created an RBD (ceph) pool as follows: <pool type='rbd'> <name>vdic-mgmt-pool</name> <uuid>58d9aab0-85cc-4dcf-884e-86349db38e96</uuid> <capacity unit='bytes'>42843746304</capacity> <allocation unit='bytes'>0</allocation> <available unit='bytes'>41737977856</available> <source> <host name='vdicnode01.local' port='6789'/> <name>vdicmgmtpool</name> <auth type='ceph' username='libvirt'> <secret uuid='3289cb24-0032-49d8-b377-c86d04b800b0'/> </auth> </source> </pool> virsh pool-define vdic-mgmt-pool.xml virsh pool-autostart vdic-mgmt-pool Note: I've tried with username='libvirt' and also username='admin' But I'm not able to start-up the storage pool: [root@vdicnode01 ceph]# virsh pool-start vdic-mgmt-pool error: Failed to start pool vdic-mgmt-pool error: An error occurred, but the cause is unknown I attach the full log (in debug mode): https://paste.fedoraproject.org/paste/KD5TnWdKfB6o9a8AutI4qg Any help will be welcome! Thanks a lot everybody!

1 0

[libvirt] [PATCH 0/3] implement migrate-getmaxdowntime command
by seg＠us.ibm.com 21 Jul '17

21 Jul '17

Currently, the maximum tolerable downtime for a domain being migrated is write-only. This patch implements a way to query that value nondestructively. Signed-off-by: Scott Garfinkle <seg(a)us.ibm.com>

1 3