- Devel - libvirt List Archives

[libvirt] [PATCH] domain_conf: fix crash in virDomainObjListFindByUUIDInternal
by Michael Chapman 11 Mar '15

11 Mar '15

If a domain object is being removed and looked up concurrently we must ensure we unlock the object before unreferencing it, since the latter might free the object. Signed-off-by: Michael Chapman <mike(a)very.puzzling.org> --- src/conf/domain_conf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index cc8616b..ce0e173 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -1068,9 +1068,9 @@ virDomainObjListFindByUUIDInternal(virDomainObjListPtr doms, if (obj) { virObjectLock(obj); if (obj->removing) { + virObjectUnlock(obj); if (ref) virObjectUnref(obj); - virObjectUnlock(obj); obj = NULL; } } -- 2.1.0

2 1

[libvirt] [PATCH 0/3] virBitmapGetBit cleanups
by Ján Tomko 11 Mar '15

11 Mar '15

Return bool instead of ignoring the return value almost everywhere. Ján Tomko (3): Use virBitmapNextClearBit in networkNextClassID Return bool in virBitmapGetBit Reverse the logic in virbitmaptest src/conf/domain_conf.c | 4 +--- src/conf/node_device_conf.c | 4 +--- src/conf/snapshot_conf.c | 6 ++---- src/conf/storage_conf.c | 4 +--- src/libxl/libxl_domain.c | 4 +--- src/libxl/libxl_driver.c | 5 +---- src/network/bridge_driver.c | 8 +++----- src/nodeinfo.c | 6 +----- src/qemu/qemu_capabilities.c | 8 +++----- src/qemu/qemu_command.c | 12 +----------- src/qemu/qemu_driver.c | 15 +++------------ src/qemu/qemu_process.c | 13 ++++--------- src/storage/storage_backend.c | 4 +--- src/test/test_driver.c | 5 +---- src/util/virbitmap.c | 13 +++++-------- src/util/virbitmap.h | 6 +++--- src/util/vircgroup.c | 4 +--- src/util/virdnsmasq.c | 7 ++----- src/util/virportallocator.c | 14 ++------------ src/util/virprocess.c | 9 ++------- src/xen/xen_driver.c | 4 +--- src/xen/xend_internal.c | 5 +---- tests/virbitmaptest.c | 18 ++++-------------- tools/virsh-domain.c | 6 ++---- 24 files changed, 47 insertions(+), 137 deletions(-) -- 2.0.5

3 8

[libvirt] [PATCH] {domain, network}_conf: disable autostart when deleting config
by Michael Chapman 11 Mar '15

11 Mar '15

Undefining a running, autostarted domain removes the autostart link, but dom->autostart is not cleared. If the domain is subsequently redefined, libvirt thinks it is already autostarted and will not create the link even if requested: # virsh dominfo example | grep Autostart Autostart: enable # ls /etc/libvirt/qemu/autostart/example.xml /etc/libvirt/qemu/autostart/example.xml # virsh undefine example Domain example has been undefined # virsh define example.xml Domain example defined from example.xml # virsh dominfo example | grep Autostart Autostart: enable # virsh autostart example Domain example marked as autostarted # ls /etc/libvirt/qemu/autostart/example.xml ls: cannot access /etc/libvirt/qemu/autostart/example.xml: No such file or directory This commit ensures dom->autostart is cleared whenever the config and autostart link (if present) are removed. The bridge network driver cleared this flag itself in networkUndefine. This commit moves this into virNetworkDeleteConfig for symmetry with virDomainDeleteConfig, and to ensure it is not missed in future network drivers. Signed-off-by: Michael Chapman <mike(a)very.puzzling.org> --- src/conf/domain_conf.c | 1 + src/conf/network_conf.c | 1 + src/network/bridge_driver.c | 1 - 3 files changed, 2 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index cc8616b..a8f4ce2 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -20998,6 +20998,7 @@ virDomainDeleteConfig(const char *configDir, /* Not fatal if this doesn't work */ unlink(autostartLink); + dom->autostart = 0; if (unlink(configFile) < 0 && errno != ENOENT) { diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index 9c1d578..779a08a 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -3077,6 +3077,7 @@ int virNetworkDeleteConfig(const char *configDir, /* Not fatal if this doesn't work */ unlink(autostartLink); + net->autostart = 0; if (unlink(configFile) < 0) { virReportSystemError(errno, diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 5752acb..5158078 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -3050,7 +3050,6 @@ networkUndefine(virNetworkPtr net) driver->networkAutostartDir, network) < 0) goto cleanup; - network->autostart = 0; event = virNetworkEventLifecycleNew(network->def->name, network->def->uuid, -- 2.1.0

2 1

[libvirt] CfP 10th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '15)
by VHPC 15 11 Mar '15

11 Mar '15

================================================================= CALL FOR PAPERS 10th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '15) held in conjunction with Euro-Par 2015, August 24-28, Vienna, Austria (Springer LNCS) ================================================================= Date: August 25, 2015 Workshop URL: http://vhpc.org Paper Submission Deadline: May 22, 2015 CALL FOR PAPERS Virtualization technologies constitute a key enabling factor for flexible resource management in modern data centers, cloud environments, and increasingly in HPC as well. Providers need to dynamically manage complex infrastructures in a seamless fashion for varying workloads and hosted applications, independently of the customers deploying software or users submitting highly dynamic and heterogeneous workloads. Thanks to virtualization, we have the ability to manage vast computing and networking resources dynamically and close to the marginal cost of providing the services, which is unprecedented in the history of scientific and commercial computing. Various virtualization technologies contribute to the overall picture in different ways: machine virtualization, with its capability to enable consolidation of multiple under-utilized servers with heterogeneous software and operating systems (OSes), and its capability to live-migrate a fully operating virtual machine (VM) with a very short downtime, enables novel and dynamic ways to manage physical servers; OS-level virtualization, with its capability to isolate multiple user-space environments and to allow for their co-existence within the same OS kernel, promises to provide many of the advantages of machine virtualization with high levels of responsiveness and performance; I/O Virtualization allows physical network adapters to take traffic from multiple VMs; network virtualization, with its capability to create logical network overlays that are independent of the underlying physical topology and IP addressing, provides the fundamental ground on top of which evolved network services can be realized with an unprecedented level of dynamicity and flexibility; These technologies have to be inter-mixed and integrated in an intelligent way, to support workloads that are increasingly demanding in terms of absolute performance, responsiveness and interactivity, and have to respect well-specified Service- Level Agreements (SLAs), as needed for industrial-grade provided services. Indeed, among emerging and increasingly interesting application domains for virtualization, we can find big-data application workloads in cloud infrastructures, interactive and real-time multimedia services in the cloud, including real-time big-data streaming platforms such as used in real-time analytics supporting nowadays a plethora of application domains. Distributed cloud infrastructures promise to offer unprecedented responsiveness levels for hosted applications, but that is only possible if the underlying virtualization technologies can overcome most of the latency impairments typical of current virtualized infrastructures (e.g., far worse tail-latency). The Workshop on Virtualization in High-Performance Cloud Computing (VHPC) aims to bring together researchers and industrial practitioners facing the challenges posed by virtualization in order to foster discussion, collaboration, mutual exchange of knowledge and experience, enabling research to ultimately provide novel solutions for virtualized computing systems of tomorrow. The workshop will be one day in length, composed of 20 min paper presentations, each followed by 10 min discussion sections, and lightning talks, limited to 5 minutes. Presentations may be accompanied by interactive demonstrations. TOPICS Topics of interest include, but are not limited to: - Virtualization in supercomputing environments, HPC clusters, cloud HPC and grids - Optimizations of virtual machine monitor platforms, hypervisors and OS-level virtualization - Hypervisor and network virtualization QoS and SLAs - Cloud based network and system management for SDN and NFV - Management, deployment and monitoring of virtualized environments - Performance measurement, modelling and monitoring of virtualized/cloud workloads - Programming models for virtualized environments - Cloud reliability, fault-tolerance, high-availability and security - Heterogeneous virtualized environments, virtualized accelerators, GPUs and co-processors - Optimized communication libraries/protocols in the cloud and for HPC in the cloud - Topology management and optimization for distributed virtualized applications - Cluster provisioning in the cloud and cloud bursting - Adaptation of emerging HPC technologies (high performance networks, RDMA, etc..) - I/O and storage virtualization, virtualization aware file systems - Job scheduling/control/policy in virtualized environments - Checkpointing and migration of VM-based large compute jobs - Cloud frameworks and APIs - Energy-efficient / power-aware virtualization Important Dates April 29, 2015 - Abstract registration May 22, 2015 - Full paper submission June 19, 2014 - Acceptance notification October 2, 2014 - Camera-ready version due August 25, 2014 - Workshop Date TPC CHAIR Michael Alexander (chair), TU Wien, Austria Anastassios Nanos (co-chair), NTUA, Greece Balazs Gerofi (co-chair), RIKEN Advanced Institute for Computational Science, Japan PROGRAM COMMITTEE Stergios Anastasiadis, University of Ioannina, Greece Costas Bekas, IBM Zurich Research Laboratory, Switzerland Jakob Blomer, CERN Ron Brightwell, Sandia National Laboratories, USA Roberto Canonico, University of Napoli Federico II, Italy Julian Chesterfield, OnApp, UK Patrick Dreher, MIT, USA William Gardner, University of Guelph, Canada Kyle Hale, Northwestern University, USA Marcus Hardt, Karlsruhe Institute of Technology, Germany Iftekhar Hussain, Infinera, USA Krishna Kant, Temple University, USA Eiji Kawai, National Institute of Information and Communications Technology, Japan Romeo Kinzler, IBM, Switzerland Kornilios Kourtis, ETH, Switzerland Nectarios Koziris, National Technical University of Athens, Greece Massimo Lamanna, CERN Che-Rung Roger Lee, National Tsing Hua University, Taiwan Helge Meinhard, CERN Jean-Marc Menaud, Ecole des Mines de Nantes France Christine Morin, INRIA, France Amer Qouneh, University of Florida, USA Seetharami Seelam, IBM Watson Research Center, USA Josh Simons, VMWare, USA Borja Sotomayor, University of Chicago, USA Kurt Tutschku, Blekinge Institute of Technology, Sweden Yasuhiro Watashiba, Osaka University, Japan Chao-Tung Yang, Tunghai University, Taiwan PAPER SUBMISSION-PUBLICATION Papers submitted to the workshop will be reviewed by at least two members of the program committee and external reviewers. Submissions should include abstract, key words, the e-mail address of the corresponding author, and must not exceed 10 pages, including tables and figures at a main font size no smaller than 11 point. Submission of a paper should be regarded as a commitment that, should the paper be accepted, at least one of the authors will register and attend the conference to present the work. Accepted papers will be published in the Springer LNCS series - the format must be according to the Springer LNCS Style. Initial submissions are in PDF; authors of accepted papers will be requested to provide source files. Format Guidelines: http://www.springer.de/comp/lncs/authors.html Submission Link: https://easychair.org/conferences/?conf=europar2015ws GENERAL INFORMATION The workshop is one day in length and will be held in conjunction with Euro-Par 2015, 24-28 August, Vienna, Austria

1 0

[libvirt] [PATCH] qemu: fix memory leak in qemuAgentGetFSInfo
by Chen Fan 10 Mar '15

10 Mar '15

in virDomainFSInfoFree(), don't free the virDomainFSInfo data. ==10670== 80 bytes in 2 blocks are definitely lost in loss record 576 of 793 ==10670== at 0x4A06BC3: calloc (vg_replace_malloc.c:618) ==10670== by 0x509DEBD: virAlloc (viralloc.c:144) ==10670== by 0x19FBD558: qemuAgentGetFSInfo (qemu_agent.c:1837) ==10670== by 0x1A03CF91: qemuDomainGetFSInfo (qemu_driver.c:19238) Signed-off-by: Chen Fan <chen.fan.fnst(a)cn.fujitsu.com> --- src/libvirt-domain.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index 04545fd..7f8a7ce 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -11337,4 +11337,6 @@ virDomainFSInfoFree(virDomainFSInfoPtr info) for (i = 0; i < info->ndevAlias; i++) VIR_FREE(info->devAlias[i]); VIR_FREE(info->devAlias); + + VIR_FREE(info); } -- 1.9.3

4 5

[libvirt] [PATCH] virQEMUCapsInitQMP: Don't dispose locked @vm
by Michal Privoznik 10 Mar '15

10 Mar '15

When creating qemu capabilities, a dummy virDomainObj is created just because our monitor code expects that. However, the object is created locked already. Then, under cleanup label, we simply unref the object which results in whole domain object to be disposed. The object lock is destroyed subsequently, but hey - it's still locked: ==24845== Thread #14's call to pthread_mutex_destroy failed ==24845== with error code 16 (EBUSY: Device or resource busy) ==24845== at 0x4C3024E: pthread_mutex_destroy (in /usr/lib64/valgrind/vgpreload_helgrind-amd64-linux.so) ==24845== by 0x531F72E: virMutexDestroy (virthread.c:83) ==24845== by 0x5302977: virObjectLockableDispose (virobject.c:237) ==24845== by 0x5302A89: virObjectUnref (virobject.c:265) ==24845== by 0x1DD37866: virQEMUCapsInitQMP (qemu_capabilities.c:3397) ==24845== by 0x1DD37CC6: virQEMUCapsNewForBinary (qemu_capabilities.c:3481) ==24845== by 0x1DD381E2: virQEMUCapsCacheLookup (qemu_capabilities.c:3609) ==24845== by 0x1DD30F8A: virQEMUCapsInitGuest (qemu_capabilities.c:744) ==24845== by 0x1DD31889: virQEMUCapsInit (qemu_capabilities.c:1020) ==24845== by 0x1DD7DD36: virQEMUDriverCreateCapabilities (qemu_conf.c:888) ==24845== by 0x1DDC57C0: qemuStateInitialize (qemu_driver.c:803) ==24845== by 0x53DC743: virStateInitialize (libvirt.c:777) ==24845== Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_capabilities.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 8193805..dce40e0 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -3394,7 +3394,10 @@ virQEMUCapsInitQMP(virQEMUCapsPtr qemuCaps, if (monpath) ignore_value(unlink(monpath)); VIR_FREE(monpath); - virObjectUnref(vm); + if (vm) { + virObjectUnlock(vm); + virObjectUnref(vm); + } virObjectUnref(xmlopt); if (pid != 0) { -- 2.0.5

2 1

[libvirt] [PATCH 0/6] Follow-up patches for IOThread API/display
by John Ferlan 10 Mar '15

10 Mar '15

Based on jtomko's review of the IOThread series and my foray into the libvirt-perl bindings, these patches make the following adjustments: Patch 1 - Found while generating the virsh iothreadpin description, the vcpupin description had a few missing words Patch 2 - While working on the libvirt-perl bindings - I found I was a bit overaggressive with my GetIOThreadsInfo interface with regard to checking ReadOnly unnecessarily Patch 3 - Adjust the IOThread CPU Affnity algorithm based on jtomko's review comments Patch 4 - Fallout because I ran the patches through my Coverity checker. Patch 5 - Similar to IOThread - adjust the GetVcpuInfo CPU Affinity algorithm for the returned cpumap Patch 5 - Similar to IOThread - adjust the GetEmulatorInfo CPU Affinity algorithm for the returned cpumap John Ferlan (6): Fix syntax for vcpupin description Remove ReadOnly check for GetIOThreadsInfo qemu: Change/Fix IOThread CPU affinity bitmap manipulation qemu: Resolve Coverity CHECKED_RETURN issue qemu: Change qemuDomainGetVcpuPinInfo bitmap manipulation qemu: Change qemuDomainGetEmulatorPinInfo bitmap manipulation src/libvirt-domain.c | 1 - src/qemu/qemu_driver.c | 177 +++++++++++++++++++++---------------------------- tools/virsh.pod | 4 +- 3 files changed, 76 insertions(+), 106 deletions(-) -- 2.1.0

2 10

[libvirt] [PATCH] qemu: don't fill in nicindexes for session mode libvirtd
by Laine Stump 10 Mar '15

10 Mar '15

Commit 4bbe1029f fixed a problem in commit f7afeddc by moving the call to virNetDevGetIndex() to a location common to all interface types (so that the niceindex array would be filled in for macvtap as well as tap interfaces), but the location was *too* common, as the original call to virNetDevGetIndex() had been in a section qualified by "if (cfg->privileged)". The result was that the "fixed" libvirtd would try to call virNetDevGetIndex() even for session mode libvirtd, and end up failing with the log message: Unable to open control socket: Operation not permitted To remedy that, this patch qualifies the call to virNetDevGetIndex() in its new location with cfg->privileged. This resolves https://bugzilla.redhat.com/show_bug.cgi?id=1198244 --- If someone (Rich?) needs this pushed before I am awake, please feel free to push it. (also push to the 1.2.13-maint branch if you do) src/qemu/qemu_command.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 1510797..3d1483e 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7861,6 +7861,7 @@ qemuBuildInterfaceCommandLine(virCommandPtr cmd, char **tapfdName = NULL; char **vhostfdName = NULL; int actualType = virDomainNetGetActualType(net); + virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); virNetDevBandwidthPtr actualBandwidth; size_t i; @@ -7936,7 +7937,7 @@ qemuBuildInterfaceCommandLine(virCommandPtr cmd, /* network and bridge use a tap device, and direct uses a * macvtap device */ - if (nicindexes && nnicindexes && net->ifname) { + if (cfg->privileged && nicindexes && nnicindexes && net->ifname) { if (virNetDevGetIndex(net->ifname, &nicindex) < 0 || VIR_APPEND_ELEMENT(*nicindexes, *nnicindexes, nicindex) < 0) goto cleanup; -- 2.1.0

4 4

[libvirt] [PATCH] virnet*: Don't unlock object in dispose
by Michal Privoznik 10 Mar '15

10 Mar '15

As of bba93d40 all of our RPC objects are derived from virObjectLockable. However, during rewrite some errors sneaked in. For instance, the dispose functions to virNetClient and virNetServerClient objects were not only freeing allocated memory, but unlocking themselves. This is wrong. Object should never disappear while locked. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/rpc/virnetclient.c | 2 -- src/rpc/virnetserverclient.c | 1 - 2 files changed, 3 deletions(-) diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c index d7455b5..7fca055 100644 --- a/src/rpc/virnetclient.c +++ b/src/rpc/virnetclient.c @@ -622,8 +622,6 @@ void virNetClientDispose(void *obj) #endif virNetMessageClear(&client->msg); - - virObjectUnlock(client); } diff --git a/src/rpc/virnetserverclient.c b/src/rpc/virnetserverclient.c index b2a4fdf..f5259c2 100644 --- a/src/rpc/virnetserverclient.c +++ b/src/rpc/virnetserverclient.c @@ -866,7 +866,6 @@ void virNetServerClientDispose(void *obj) virObjectUnref(client->tlsCtxt); #endif virObjectUnref(client->sock); - virObjectUnlock(client); } -- 2.0.5

2 1

[libvirt] [PATCH v3 00/10] Network enhancements and other fixes
by Maxim Nestratov 10 Mar '15

10 Mar '15

Maxim Nestratov (9): parallels: don't forget to unlock domain if unregister fails parallels: fix home directory for VMs parallels: minor cleanup parallels: introduce and use string constants for network types and names parallels: fix parallelsLoadNetworks parallels: better bridge network interface support parallels: set network adapter device status to connected parallels: make E1000 network adapter type default parallels: switch off offline management feature Mikhail Feoktistov (1): parallels: set cpu mode when applying xml configuration

1 12