- Devel - libvirt List Archives

[libvirt] [PATCH] typo: fix typo error in virLXCControllerSetupResourceLimits
by zhang bo 27 Apr '15

27 Apr '15

iff -> if rather then -> rather than Signed-off-by: YueWenyuan <yuewenyuan(a)huawei.com> --- src/lxc/lxc_controller.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index e144c2d..7ad9d50 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -745,9 +745,9 @@ static int virLXCControllerSetupResourceLimits(virLXCControllerPtr ctrl) if (mode == VIR_DOMAIN_NUMATUNE_MEM_STRICT && virCgroupControllerAvailable(VIR_CGROUP_CONTROLLER_CPUSET)) { - /* Use virNuma* API iff necessary. Once set and child is exec()-ed, + /* Use virNuma* API if necessary. Once set and child is exec()-ed, * there's no way for us to change it. Rely on cgroups (if available - * and enabled in the config) rather then virNuma*. */ + * and enabled in the config) rather than virNuma*. */ VIR_DEBUG("Relying on CGroups for memory binding"); } else { -- 1.7.12.4

2 1

[libvirt] [PATCH 0/5] qemu: Connect to guest agent after channel hotplug
by Peter Krempa 26 Apr '15

26 Apr '15

Clean up some parts of the agent code, fix a recently added bug and add support for guest agent hotplug, all this while removing more lines than adding. Peter Krempa (5): qemu: agent: Reuse virJSONValueObjectCreateVArgs in qemuAgentMakeCommand qemu: Fix domain object leak in qemuDomainInterfaceAddresses qemu: Reuse qemuDomainAgentAvailable in qemuDomainInterfaceAddresses qemu: agent: Differentiate errors when the agent channel was hotplugged qemu: Connect to guest agent after channel hotplug src/qemu/qemu_agent.c | 73 ++----------------------------------------------- src/qemu/qemu_domain.c | 45 +++++++++++++++++++++++++++--- src/qemu/qemu_domain.h | 2 ++ src/qemu/qemu_driver.c | 56 +++++++++++++++++++++---------------- src/qemu/qemu_process.c | 22 +-------------- src/qemu/qemu_process.h | 2 ++ 6 files changed, 81 insertions(+), 119 deletions(-) -- 2.3.5

2 8

[libvirt] [PATCH] libxl: remove duplicate check for pci subsystem type
by Olaf Hering 24 Apr '15

24 Apr '15

Both attach and detach functions get called only if the type matches. Signed-off-by: Olaf Hering <olaf(a)aepfle.de> Cc: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libxl_driver.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index d76efda..f915f91 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -2901,9 +2901,6 @@ libxlDomainAttachHostPCIDevice(libxlDriverPrivatePtr driver, libxl_device_pci_init(&pcidev); - if (hostdev->source.subsys.type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) - goto cleanup; - if (virDomainHostdevFind(vm->def, hostdev, &found) >= 0) { virReportError(VIR_ERR_OPERATION_FAILED, _("target pci device %.4x:%.2x:%.2x.%.1x already exists"), @@ -3239,9 +3236,6 @@ libxlDomainDetachHostPCIDevice(libxlDriverPrivatePtr driver, libxl_device_pci_init(&pcidev); - if (subsys->type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) - goto cleanup; - idx = virDomainHostdevFind(vm->def, hostdev, &detach); if (idx < 0) { virReportError(VIR_ERR_OPERATION_FAILED,

2 1

[libvirt] [PATCH] libxl: use pci init and dispose functions
by Olaf Hering 24 Apr '15

24 Apr '15

Rearrange code so that the local variable is always initialized and disposed. Signed-off-by: Olaf Hering <olaf(a)aepfle.de> Cc: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libxl_driver.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index 393f8bd..d76efda 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -2893,12 +2893,14 @@ libxlDomainAttachHostPCIDevice(libxlDriverPrivatePtr driver, virDomainHostdevDefPtr hostdev) { libxlDriverConfigPtr cfg = libxlDriverConfigGet(driver); - libxl_device_pci pcidev; + libxl_device_pci pcidev; virDomainHostdevDefPtr found; virHostdevManagerPtr hostdev_mgr = driver->hostdevMgr; virDomainHostdevSubsysPCIPtr pcisrc = &hostdev->source.subsys.u.pci; int ret = -1; + libxl_device_pci_init(&pcidev); + if (hostdev->source.subsys.type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) goto cleanup; @@ -2939,6 +2941,7 @@ libxlDomainAttachHostPCIDevice(libxlDriverPrivatePtr driver, cleanup: virObjectUnref(cfg); + libxl_device_pci_dispose(&pcidev); return ret; } @@ -3234,6 +3237,8 @@ libxlDomainDetachHostPCIDevice(libxlDriverPrivatePtr driver, virHostdevManagerPtr hostdev_mgr = driver->hostdevMgr; int ret = -1; + libxl_device_pci_init(&pcidev); + if (subsys->type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) goto cleanup; @@ -3255,8 +3260,6 @@ libxlDomainDetachHostPCIDevice(libxlDriverPrivatePtr driver, } - libxl_device_pci_init(&pcidev); - if (libxlMakePCI(detach, &pcidev) < 0) goto error; @@ -3269,7 +3272,6 @@ libxlDomainDetachHostPCIDevice(libxlDriverPrivatePtr driver, goto error; } - libxl_device_pci_dispose(&pcidev); virDomainHostdevRemove(vm->def, idx); @@ -3283,6 +3285,7 @@ libxlDomainDetachHostPCIDevice(libxlDriverPrivatePtr driver, cleanup: virObjectUnref(cfg); + libxl_device_pci_dispose(&pcidev); return ret; }

2 1

[libvirt] [PATCH] libxl: initialize sdl.opengl defbool in libxlMakeVfb
by Jim Fehlig 24 Apr '15

24 Apr '15

Commit bf32462b missed initializing sdl.opengl. Without the initialization, libvirtd will be terminated by an assert from libxl: Assertion `!libxl_defbool_is_default(db)' failed. Reported-by: Olaf Hering <olaf(a)aepfle.de> Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- Perhaps bending the "trivial rule" a bit, but this patch is trivial and should have been included in bf32462b. I simply forgot to add initialization of sdl.opengl before pushing bf32462b. I'll push this fix shortly. src/libxl/libxl_conf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/libxl/libxl_conf.c b/src/libxl/libxl_conf.c index 55bd21b..53f327b 100644 --- a/src/libxl/libxl_conf.c +++ b/src/libxl/libxl_conf.c @@ -1243,6 +1243,7 @@ libxlMakeVfb(virPortAllocatorPtr graphicsports, case VIR_DOMAIN_GRAPHICS_TYPE_SDL: libxl_defbool_set(&x_vfb->sdl.enable, 1); libxl_defbool_set(&x_vfb->vnc.enable, 0); + libxl_defbool_set(&x_vfb->sdl.opengl, 0); if (VIR_STRDUP(x_vfb->sdl.display, l_vfb->data.sdl.display) < 0) return -1; if (VIR_STRDUP(x_vfb->sdl.xauthority, l_vfb->data.sdl.xauth) < 0) -- 1.8.4.5

1 0

[libvirt] CfP Virtualization in High-Performance Cloud Computing Workshop (VHPC '15)
by VHPC 15 24 Apr '15

24 Apr '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. Docker et al. OS-level virtualization, with its capability to isolate multiple user-space environments allows for their co-existence within the same OS kernel. It promises to provide many of the advantages of machine virtualization with high levels of responsiveness and performance; coupled with lightweight OSs it forms a potent architecture with promise to become a mainstream environment for HPC workloads. 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; 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. 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 - OS-level virtualization including container runtimes (Docker, rkt et al.) - Lightweight compute node operating systems/VMMs - Optimizations of virtual machine monitor platforms, hypervisors and - 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, 2015 - Acceptance notification October 2, 2015 - Camera-ready version due August 25, 2015 - 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 William Magato, University of Cincinnati, USA 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 0/2] Prefer virDomainObjListFindByUUID wherever possible
by Michal Privoznik 24 Apr '15

24 Apr '15

This should go on the top of [1] so that both virDomainObjListFindByUUID and virDomainObjListFindByName are both O(1). Then, the only function left that is more time complex is virDomainObjListFindByID. Unfortunately, we can't drop it completely, but we can use it less. 1: https://www.redhat.com/archives/libvir-list/2015-April/msg01168.html Michal Privoznik (2): uml: s/virDomainObjListFindByID/virDomainObjListFindByUUID/ libxl: s/virDomainObjListFindByID/virDomainObjListFindByUUID/ src/libxl/libxl_domain.c | 2 +- src/uml/uml_driver.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) -- 2.0.5

3 5

[libvirt] [PATCH v2] parallels: fix crash in prlsdkAddNet in case of CT definition
by Maxim Nestratov 24 Apr '15

24 Apr '15

Since net->model is not defined for containers we shouldn't touch it. In case network adapter model is defined, a warning about ignoring it is shown. Signed-off-by: Maxim Nestratov <mnestratov(a)parallels.com> --- src/parallels/parallels_sdk.c | 29 ++++++++++++++++++----------- 1 files changed, 18 insertions(+), 11 deletions(-) diff --git a/src/parallels/parallels_sdk.c b/src/parallels/parallels_sdk.c index d54f894..c67044c 100644 --- a/src/parallels/parallels_sdk.c +++ b/src/parallels/parallels_sdk.c @@ -2725,7 +2725,8 @@ static const char * prlsdkFormatMac(virMacAddrPtr mac, char *macstr) static int prlsdkAddNet(PRL_HANDLE sdkdom, parallelsConnPtr privconn, - virDomainNetDefPtr net) + virDomainNetDefPtr net, + bool isCt) { PRL_RESULT pret; PRL_HANDLE sdknet = PRL_INVALID_HANDLE; @@ -2757,19 +2758,25 @@ static int prlsdkAddNet(PRL_HANDLE sdkdom, pret = PrlVmDevNet_SetMacAddress(sdknet, macstr); prlsdkCheckRetGoto(pret, cleanup); - if (STREQ(net->model, "rtl8139")) { - pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_RTL); - } else if (STREQ(net->model, "e1000")) { - pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_E1000); - } else if (STREQ(net->model, "virtio")) { - pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_VIRTIO); + if (isCt) { + if (net->model) + VIR_WARN("Setting network adapter for containers is not " + "supported by Parallels Cloud Server."); } else { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + if (STREQ(net->model, "rtl8139")) { + pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_RTL); + } else if (STREQ(net->model, "e1000")) { + pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_E1000); + } else if (STREQ(net->model, "virtio")) { + pret = PrlVmDevNet_SetAdapterType(sdknet, PNT_VIRTIO); + } else { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("Specified network adapter model is not " "supported by Parallels Cloud Server.")); - goto cleanup; + goto cleanup; + } + prlsdkCheckRetGoto(pret, cleanup); } - prlsdkCheckRetGoto(pret, cleanup); if (net->type == VIR_DOMAIN_NET_TYPE_NETWORK) { if (STREQ(net->data.network.name, PARALLELS_DOMAIN_ROUTED_NETWORK_NAME)) { @@ -3153,7 +3160,7 @@ prlsdkDoApplyConfig(virConnectPtr conn, } for (i = 0; i < def->nnets; i++) { - if (prlsdkAddNet(sdkdom, conn->privateData, def->nets[i]) < 0) + if (prlsdkAddNet(sdkdom, conn->privateData, def->nets[i], IS_CT(def)) < 0) goto error; } -- 1.7.1

2 1

[libvirt] [PATCH v2] parallels: implement .domainGetMaxMemory
by Dmitry Guryanov 24 Apr '15

24 Apr '15

Since we haven't implemented balloon parameters tuning we can just return amount of memory in this function. Signed-off-by: Dmitry Guryanov <dguryanov(a)parallels.com> --- src/parallels/parallels_driver.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) Changes in v2: use parallelsDomObjFromDomain function diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c index 07f1311..c791b5b 100644 --- a/src/parallels/parallels_driver.c +++ b/src/parallels/parallels_driver.c @@ -1098,6 +1098,20 @@ static int parallelsDomainAttachDevice(virDomainPtr dom, const char *xml) VIR_DOMAIN_AFFECT_CONFIG | VIR_DOMAIN_AFFECT_LIVE); } +static unsigned long long +parallelsDomainGetMaxMemory(virDomainPtr domain) +{ + virDomainObjPtr dom = NULL; + int ret = -1; + + if (!(dom = parallelsDomObjFromDomain(domain))) + return -1; + + ret = dom->def->mem.max_balloon; + virObjectUnlock(dom); + return ret; +} + static virHypervisorDriver parallelsDriver = { .name = "Parallels", .connectOpen = parallelsConnectOpen, /* 0.10.0 */ @@ -1144,6 +1158,7 @@ static virHypervisorDriver parallelsDriver = { .domainHasManagedSaveImage = parallelsDomainHasManagedSaveImage, /* 1.2.13 */ .domainManagedSave = parallelsDomainManagedSave, /* 1.2.14 */ .domainManagedSaveRemove = parallelsDomainManagedSaveRemove, /* 1.2.14 */ + .domainGetMaxMemory = parallelsDomainGetMaxMemory, /* 1.2.15 */ }; static virConnectDriver parallelsConnectDriver = { -- 2.1.0

2 3

[libvirt] [PATCH] bhyve: fix build in tests
by Roman Bogorodskiy 24 Apr '15

24 Apr '15

Commit ca32929 caused build errors in bhyvexml2argvtest.c because of some undefined variables. Pushing under the build breaker rule. --- tests/bhyvexml2argvtest.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/tests/bhyvexml2argvtest.c b/tests/bhyvexml2argvtest.c index 1cce2aa..1c7bc31 100644 --- a/tests/bhyvexml2argvtest.c +++ b/tests/bhyvexml2argvtest.c @@ -19,7 +19,6 @@ static int testCompareXMLToArgvFiles(const char *xml, const char *ldcmdline, const char *dmcmdline) { - int len; char *actualargv = NULL, *actualld = NULL, *actualdm = NULL; virDomainDefPtr vmdef = NULL; virDomainObj vm; @@ -59,9 +58,6 @@ static int testCompareXMLToArgvFiles(const char *xml, if (virtTestCompareToFile(actualld, ldcmdline) < 0) goto out; - if (virtTestCompareToFile(formatted, xml) < 0) - goto out; - if (virFileExists(dmcmdline) || actualdm) { if (virtTestCompareToFile(actualdm, dmcmdline) < 0) goto out; -- 2.3.5

1 0