- Devel - libvirt List Archives

[libvirt] [PATCH] qemu: Adjust size for qcow2/qed if not on sector boundary (rewrite)
by John Ferlan 12 May '14

12 May '14

A post commit id 'e3d66229' review (and followup): http://www.redhat.com/archives/libvir-list/2014-May/msg00268.html noted some issues with the code, so I have adjusted the code accordingly. The difference between this and the commit prior to the change (commit id 'f3be5f0c') will just be the check for qcow2/qed using a non 512 block aligned size will result in a round up of size. If the size is within the last 512 bytes to ULLONG_MAX, then just set it there rather than erroring out. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/qemu/qemu_driver.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 4ff8a2d..8771cae 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -9413,7 +9413,6 @@ qemuDomainBlockResize(virDomainPtr dom, virDomainObjPtr vm; qemuDomainObjPrivatePtr priv; int ret = -1, idx; - unsigned long long size_up; char *device = NULL; virDomainDiskDefPtr disk = NULL; @@ -9434,12 +9433,6 @@ qemuDomainBlockResize(virDomainPtr dom, return -1; } size *= 1024; - size_up = size; - } else { - /* For 'qcow2' and 'qed', qemu resize blocks expects values - * on sector boundary, so round our value up to prepare - */ - size_up = VIR_ROUND_UP(size, 512); } if (!(vm = qemuDomObjFromDomain(dom))) @@ -9466,19 +9459,16 @@ qemuDomainBlockResize(virDomainPtr dom, } disk = vm->def->disks[idx]; - /* qcow2 and qed must be sized appropriately, so be sure our value - * is sized appropriately and will fit + /* qcow2 and qed must be sized on 512 byte blocks/sectors, + * so adjust size if necessary to round up (if possible). */ - if (size != size_up && - (disk->src.format == VIR_STORAGE_FILE_QCOW2 || - disk->src.format == VIR_STORAGE_FILE_QED)) { - if (size_up > ULLONG_MAX) { - virReportError(VIR_ERR_OVERFLOW, - _("size must be less than %llu KiB"), - ULLONG_MAX / 1024); - goto endjob; - } - size = size_up; + if ((disk->src.format == VIR_STORAGE_FILE_QCOW2 || + disk->src.format == VIR_STORAGE_FILE_QED) && + (size % 512)) { + if ((ULLONG_MAX - size) < 512) + size = ULLONG_MAX; + else + size = VIR_ROUND_UP(size, 512); } if (virAsprintf(&device, "%s%s", QEMU_DRIVE_HOST_PREFIX, -- 1.9.0

2 2

[libvirt] [PATCH libvirt-glib 1/2] gconfig: add spiceport chardev
by Marc-André Lureau 12 May '14

12 May '14

From: Marc-André Lureau <marcandre.lureau(a)gmail.com> --- libvirt-gconfig/Makefile.am | 2 + libvirt-gconfig/libvirt-gconfig-domain-channel.h | 1 + ...bvirt-gconfig-domain-chardev-source-spiceport.c | 101 +++++++++++++++++++++ ...bvirt-gconfig-domain-chardev-source-spiceport.h | 71 +++++++++++++++ libvirt-gconfig/libvirt-gconfig.h | 1 + libvirt-gconfig/libvirt-gconfig.sym | 9 ++ 6 files changed, 185 insertions(+) create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.h diff --git a/libvirt-gconfig/Makefile.am b/libvirt-gconfig/Makefile.am index d328ca7..83d521f 100644 --- a/libvirt-gconfig/Makefile.am +++ b/libvirt-gconfig/Makefile.am @@ -28,6 +28,7 @@ GCONFIG_HEADER_FILES = \ libvirt-gconfig-domain-chardev.h \ libvirt-gconfig-domain-chardev-source.h \ libvirt-gconfig-domain-chardev-source-pty.h \ + libvirt-gconfig-domain-chardev-source-spiceport.h \ libvirt-gconfig-domain-chardev-source-spicevmc.h \ libvirt-gconfig-domain-clock.h \ libvirt-gconfig-domain-console.h \ @@ -114,6 +115,7 @@ GCONFIG_SOURCE_FILES = \ libvirt-gconfig-domain-chardev.c \ libvirt-gconfig-domain-chardev-source.c \ libvirt-gconfig-domain-chardev-source-pty.c \ + libvirt-gconfig-domain-chardev-source-spiceport.c \ libvirt-gconfig-domain-chardev-source-spicevmc.c \ libvirt-gconfig-domain-clock.c \ libvirt-gconfig-domain-console.c \ diff --git a/libvirt-gconfig/libvirt-gconfig-domain-channel.h b/libvirt-gconfig/libvirt-gconfig-domain-channel.h index 889b09a..4f1130e 100644 --- a/libvirt-gconfig/libvirt-gconfig-domain-channel.h +++ b/libvirt-gconfig/libvirt-gconfig-domain-channel.h @@ -59,6 +59,7 @@ struct _GVirConfigDomainChannelClass typedef enum { GVIR_CONFIG_DOMAIN_CHANNEL_TARGET_GUESTFWD, GVIR_CONFIG_DOMAIN_CHANNEL_TARGET_VIRTIO, + GVIR_CONFIG_DOMAIN_CHANNEL_TARGET_SPICEPORT, } GVirConfigDomainChannelTargetType; diff --git a/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.c b/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.c new file mode 100644 index 0000000..570e7ea --- /dev/null +++ b/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.c @@ -0,0 +1,101 @@ +/* + * libvirt-gconfig-domain-chardev-source-spiceport.c: libvirt domain chardev spiceport configuration + * + * Copyright (C) 2014 Red Hat, Inc. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + */ + +#include <config.h> + +#include "libvirt-gconfig/libvirt-gconfig.h" +#include "libvirt-gconfig/libvirt-gconfig-private.h" + +#define GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_GET_PRIVATE(obj) \ + (G_TYPE_INSTANCE_GET_PRIVATE((obj), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, GVirConfigDomainChardevSourceSpicePortPrivate)) + +struct _GVirConfigDomainChardevSourceSpicePortPrivate +{ + gboolean unused; +}; + +G_DEFINE_TYPE(GVirConfigDomainChardevSourceSpicePort, gvir_config_domain_chardev_source_spiceport, GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE); + + +static void gvir_config_domain_chardev_source_spiceport_class_init(GVirConfigDomainChardevSourceSpicePortClass *klass) +{ + g_type_class_add_private(klass, sizeof(GVirConfigDomainChardevSourceSpicePortPrivate)); +} + + +static void gvir_config_domain_chardev_source_spiceport_init(GVirConfigDomainChardevSourceSpicePort *source) +{ + g_debug("Init GVirConfigDomainChardevSourceSpicePort=%p", source); + + source->priv = GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_GET_PRIVATE(source); +} + + +GVirConfigDomainChardevSourceSpicePort *gvir_config_domain_chardev_source_spiceport_new(void) +{ + GVirConfigObject *object; + + /* the name of the root node is just a placeholder, it will be + * overwritten when the GVirConfigDomainChardevSourceSpicePort is attached to a + * GVirConfigDomainChardevSourceSpicePort + */ + object = gvir_config_object_new(GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, "dummy", NULL); + gvir_config_object_set_attribute(object, "type", "spiceport", NULL); + return GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(object); +} + + +GVirConfigDomainChardevSourceSpicePort *gvir_config_domain_chardev_source_spiceport_new_from_xml(const gchar *xml, + GError **error) +{ + GVirConfigObject *object; + + /* the name of the root node is just a placeholder, it will be + * overwritten when the GVirConfigDomainChardevSourceSpicePort is attached to a + * GVirConfigDomainChardevSourceSpicePort + */ + object = gvir_config_object_new_from_xml(GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, + "dummy", NULL, xml, error); + if (g_strcmp0(gvir_config_object_get_attribute(object, NULL, "type"), "spiceport") != 0) { + g_object_unref(G_OBJECT(object)); + g_return_val_if_reached(NULL); + } + return GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(object); +} + +void gvir_config_domain_chardev_source_spiceport_set_channel(GVirConfigDomainChardevSourceSpicePort *port, + const char *channel) +{ + g_return_if_fail(GVIR_CONFIG_IS_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(port)); + + gvir_config_object_replace_child_with_attribute(GVIR_CONFIG_OBJECT(port), + "source", + "channel", + channel); + +} + +const gchar * gvir_config_domain_chardev_source_spiceport_get_channel(GVirConfigDomainChardevSourceSpicePort *port) +{ + g_return_val_if_fail(GVIR_CONFIG_IS_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(port), NULL); + + return gvir_config_object_get_attribute(GVIR_CONFIG_OBJECT(port), + "source", "channel"); +} diff --git a/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.h b/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.h new file mode 100644 index 0000000..5ff7794 --- /dev/null +++ b/libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.h @@ -0,0 +1,71 @@ +/* + * libvirt-gconfig-domain-chardev-source-spiceport.h: libvirt domain chardev spiceport configuration + * + * Copyright (C) 2014 Red Hat, Inc. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + */ + +#if !defined(__LIBVIRT_GCONFIG_H__) && !defined(LIBVIRT_GCONFIG_BUILD) +#error "Only <libvirt-gconfig/libvirt-gconfig.h> can be included directly." +#endif + +#ifndef __LIBVIRT_GCONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_H__ +#define __LIBVIRT_GCONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_H__ + +G_BEGIN_DECLS + +#define GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT (gvir_config_domain_chardev_source_spiceport_get_type ()) +#define GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(obj) (G_TYPE_CHECK_INSTANCE_CAST ((obj), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, GVirConfigDomainChardevSourceSpicePort)) +#define GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_CLASS(klass) (G_TYPE_CHECK_CLASS_CAST ((klass), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, GVirConfigDomainChardevSourceSpicePortClass)) +#define GVIR_CONFIG_IS_DOMAIN_CHARDEV_SOURCE_SPICE_PORT(obj) (G_TYPE_CHECK_INSTANCE_TYPE ((obj), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT)) +#define GVIR_CONFIG_IS_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_CLASS(klass) (G_TYPE_CHECK_CLASS_TYPE ((klass), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT)) +#define GVIR_CONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_GET_CLASS(obj) (G_TYPE_INSTANCE_GET_CLASS ((obj), GVIR_CONFIG_TYPE_DOMAIN_CHARDEV_SOURCE_SPICE_PORT, GVirConfigDomainChardevSourceSpicePortClass)) + +typedef struct _GVirConfigDomainChardevSourceSpicePort GVirConfigDomainChardevSourceSpicePort; +typedef struct _GVirConfigDomainChardevSourceSpicePortPrivate GVirConfigDomainChardevSourceSpicePortPrivate; +typedef struct _GVirConfigDomainChardevSourceSpicePortClass GVirConfigDomainChardevSourceSpicePortClass; + +struct _GVirConfigDomainChardevSourceSpicePort +{ + GVirConfigDomainChardevSource parent; + + GVirConfigDomainChardevSourceSpicePortPrivate *priv; + + /* Do not add fields to this struct */ +}; + +struct _GVirConfigDomainChardevSourceSpicePortClass +{ + GVirConfigDomainChardevSourceClass parent_class; + + gpointer padding[20]; +}; + + +GType gvir_config_domain_chardev_source_spiceport_get_type(void); + +GVirConfigDomainChardevSourceSpicePort *gvir_config_domain_chardev_source_spiceport_new(void); +GVirConfigDomainChardevSourceSpicePort *gvir_config_domain_chardev_source_spiceport_new_from_xml(const gchar *xml, + GError **error); + +void gvir_config_domain_chardev_source_spiceport_set_channel(GVirConfigDomainChardevSourceSpicePort *port, + const char *channel); + +const gchar * gvir_config_domain_chardev_source_spiceport_get_channel(GVirConfigDomainChardevSourceSpicePort *port); + +G_END_DECLS + +#endif /* __LIBVIRT_GCONFIG_DOMAIN_CHARDEV_SOURCE_SPICE_PORT_H__ */ diff --git a/libvirt-gconfig/libvirt-gconfig.h b/libvirt-gconfig/libvirt-gconfig.h index 61b5a04..1582109 100644 --- a/libvirt-gconfig/libvirt-gconfig.h +++ b/libvirt-gconfig/libvirt-gconfig.h @@ -44,6 +44,7 @@ #include <libvirt-gconfig/libvirt-gconfig-domain-chardev.h> #include <libvirt-gconfig/libvirt-gconfig-domain-chardev-source.h> #include <libvirt-gconfig/libvirt-gconfig-domain-chardev-source-pty.h> +#include <libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spiceport.h> #include <libvirt-gconfig/libvirt-gconfig-domain-chardev-source-spicevmc.h> #include <libvirt-gconfig/libvirt-gconfig-domain-channel.h> #include <libvirt-gconfig/libvirt-gconfig-domain-clock.h> diff --git a/libvirt-gconfig/libvirt-gconfig.sym b/libvirt-gconfig/libvirt-gconfig.sym index ead313b..fc68050 100644 --- a/libvirt-gconfig/libvirt-gconfig.sym +++ b/libvirt-gconfig/libvirt-gconfig.sym @@ -687,4 +687,13 @@ global: gvir_config_object_new_from_xml; } LIBVIRT_GCONFIG_0.1.7; +LIBVIRT_GCONFIG_0.1.9 { +global: + gvir_config_domain_chardev_source_spiceport_get_channel; + gvir_config_domain_chardev_source_spiceport_get_type; + gvir_config_domain_chardev_source_spiceport_new; + gvir_config_domain_chardev_source_spiceport_new_from_xml; + gvir_config_domain_chardev_source_spiceport_set_channel; +} LIBVIRT_GCONFIG_0.1.8; + # .... define new API here using predicted next version number .... -- 1.8.5.3

2 3

[libvirt] [PATCH] util: Fix return type mismatch for nl_recv
by Wangrui (K) 12 May '14

12 May '14

1. As shown in the definition of nl_recv, its return value is of INT type. int nl_recv(struct nl_handle *handle, struct sockaddr_nl *nla, unsigned char **buf, struct ucred **creds) However, in function virNetlinkCommand, it uses an unsigned int param, respbuflen, to receive its return value. Thus, the branch "*respbuflen < 0" is unreachable, negative return values are handled incorrectly. 2. It's said in nl_recv's description that "The caller is responsible for freeing the buffer allocated * in \c *buf if a positive value is returned." which means, we needn't to free it in case it fails. Additionally, nl_recv has a BUG in handling buf: in the error branch, it frees the buf, but didn't set it to NULL. Freeing it outside in the caller function will cause double free. Thus, we set but(resp) to NULL if nl_recv fails. Signed-off-by: Zhang Bo <oscar.zhangbo(a)huawei.com> --- src/util/virnetlink.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/util/virnetlink.c b/src/util/virnetlink.c index 0cf18f2..1a09567 100644 --- a/src/util/virnetlink.c +++ b/src/util/virnetlink.c @@ -192,6 +192,7 @@ int virNetlinkCommand(struct nl_msg *nl_msg, int n; struct nlmsghdr *nlmsg = nlmsg_hdr(nl_msg); virNetlinkHandle *nlhandle = NULL; + int length = 0; if (protocol >= MAX_LINKS) { virReportSystemError(EINVAL, @@ -257,12 +258,15 @@ int virNetlinkCommand(struct nl_msg *nl_msg, goto error; } - *respbuflen = nl_recv(nlhandle, &nladdr, - (unsigned char **)resp, NULL); + length = nl_recv(nlhandle, &nladdr, + (unsigned char **)resp, NULL); - if (*respbuflen <= 0) { + if (length <= 0) { virReportSystemError(errno, "%s", _("nl_recv failed")); + *resp = NULL; rc = -1; + } else { + *respbuflen = length; } error: if (rc == -1) { -- 1.7.12.4

2 1

[libvirt] [VHPC’14] LAST Call for Papers - Deadline in 4 weeks
by VHPC 14 12 May '14

12 May '14

================================================================= CALL FOR PAPERS 9th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '14) held in conjunction with Euro-Par 2014, August 25-29, Porto, Portugal (Springer LNCS) ================================================================= Date: August 26, 2014 Workshop URL: http://vhpc.org Paper Submission Deadline: June 9, 2014 (extended) Confirmed Keynote Speakers: Ron Brightwell, Sandia National Laboratory Hobbes: Using Virtualization to Enable Exascale Applications and Helge Meinhard, CERN CALL FOR PAPERS Virtualization technologies constitute a key enabling factor for flexible resource management in modern data centers, and particularly in cloud environments. Cloud 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 NICs/HBAs 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; the increasingly adopted paradigm of Software-Defined Networking (SDN) promises to extend this flexibility to the control and data planes of network paths. 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). What is more, in data communications Network Function Virtualization (NFV) is becoming a key technology enabling a shift from supplying hardware-based network functions, to providing them in a software-based and elastic way. In conjunction with (public and private) cloud technologies, NFV may be used for constructing the foundation for cost-effective network functions that can easily and seamlessly adapt to demand, still keeping their major carrier-grade characteristics in terms of QoS and reliability. 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: - Management, deployment and monitoring of virtualized environments - Language-process virtual machines - Performance monitoring for virtualized/cloud workloads - Virtual machine monitor platforms - Topology management and optimization for distributed virtualized applications - Paravirtualized I/O - Improving I/O and network virtualization including use of RDMA, Infiniband, PCIe - Improving performance in VM access to GPUs, GPU clusters, GP-GPUs - HPC storage virtualization - Virtualized systems for big-data and analytics workloads - Optimizations and enhancements to OS virtualization support - Improving OS-level virtualization and its integration within cloud management - Performance modelling for virtualized/cloud applications - Heterogeneous virtualized environments - Parallel virtualized - virtualization aware file systems - Network virtualization - Software defined networking - Network function virtualization - Hypervisor and network virtualization QoS and SLAs - Cloudbursting - Evolved European grid architectures including such based on network virtualization - Workload characterization for VM-based environments - Optimized communication libraries/protocols in the cloud - System and process/bytecode VM convergence - Cloud frameworks and APIs - Checkpointing/migration of VM-based large compute jobs - Job scheduling/control/policy with VMs - Instrumentation interfaces and languages - VMM performance (auto-)tuning on various load types - Cloud reliability, fault-tolerance, and security - Research, industrial and educational use cases - Virtualization in cloud, cluster and grid environments - Cross-layer VM optimizations - Cloud HPC use cases including optimizations - Services in cloud HPC - Hypervisor extensions and tools for cluster and grid computing - Cluster provisioning in the cloud - Performance and cost modelling - Languages for describing highly-distributed compute jobs - VM cloud and cluster distribution algorithms, load balancing - Instrumentation interfaces and languages - Energy-aware virtualization Important Dates Rolling Paper registration June 9, 2014 - Full paper submission (extended) July 4, 2014 - Acceptance notification October 3, 2014 - Camera-ready version due August 26, 2014 - Workshop Date TPC CHAIR Michael Alexander (chair), TU Wien, Austria Anastassios Nanos (co-chair), NTUA, Greece Tommaso Cucinotta (co-chair), Bell Labs, Dublin, Ireland PROGRAM COMMITTEE Costas Bekas, IBM Jakob Blomer, CERN Roberto Canonico, University of Napoli Federico II, Italy Piero Castoldi, Sant'Anna School of Advanced Studies Paolo Costa, MS Research Cambridge, England Jorge Ejarque Artigas, Barcelona Supercomputing Center, Spain William Gardner, University of Guelph, USA Balazs Gerofi, University of Tokyo, Japan Krishna Kant, Temple University, USA Romeo Kinzler, IBM Nectarios Koziris, National Technical University of Athens, Greece Giuseppe Lettieri, University of Pisa, Italy Jean-Marc Menaud, Ecole des Mines de Nantes, France Christine Morin, INRIA, France Dimitrios Nikolopoulos, Queen's University of Belfast, UK Herbert Poetzl, VServer, Austria Luigi Rizzo, University of Pisa, Italy Josh Simons, VMware, USA Borja Sotomayor, University of Chicago, USA Vangelis Tasoulas, Simula Research Lab, Norway Yoshio Turner, HP Labs, USA Kurt Tutschku, Blekinge Institute of Technology, Sweden 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 EasyChair Abstract Submission Link: https://www.easychair.org/conferences/?conf=europar2014ws GENERAL INFORMATION The workshop is one day in length and will be held in conjunction with Euro-Par 2014, 25-29 August, Porto, Portugal

1 0

[libvirt] [PATCH 2/2] util: do not set resp to NULL after VIR_FREE
by Wangrui (K) 12 May '14

12 May '14

Signed-off-by: Zhang Bo <oscar.zhangbo(a)huawei.com> --- src/util/virnetlink.c | 1 - 1 file changed, 1 deletion(-) diff --git a/src/util/virnetlink.c b/src/util/virnetlink.c index 4f4dedc..40912f5 100644 --- a/src/util/virnetlink.c +++ b/src/util/virnetlink.c @@ -271,7 +271,6 @@ int virNetlinkCommand(struct nl_msg *nl_msg, cleanup: if (rc == -1) { VIR_FREE(*resp); - *resp = NULL; *respbuflen = 0; } -- 1.7.12.4

1 0

[libvirt] [PATCH 1/2] util: replace error with cleanup in virNetlinkCommand
by Wangrui (K) 12 May '14

12 May '14

In funtcion virNetlinkCommand, it uses the word "error" as the goto out keyword, but it's not an error-handling branch, so, use "cleanup" instead. Signed-off-by: Zhang Bo <oscar.zhangbo(a)huawei.com> --- src/util/virnetlink.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/util/virnetlink.c b/src/util/virnetlink.c index 1a09567..4f4dedc 100644 --- a/src/util/virnetlink.c +++ b/src/util/virnetlink.c @@ -212,7 +212,7 @@ int virNetlinkCommand(struct nl_msg *nl_msg, _("cannot connect to netlink socket with protocol %d"), protocol); rc = -1; - goto error; + goto cleanup; } fd = nl_socket_get_fd(nlhandle); @@ -220,14 +220,14 @@ int virNetlinkCommand(struct nl_msg *nl_msg, virReportSystemError(errno, "%s", _("cannot get netlink socket fd")); rc = -1; - goto error; + goto cleanup ; } if (groups && nl_socket_add_membership(nlhandle, groups) < 0) { virReportSystemError(errno, "%s", _("cannot add netlink membership")); rc = -1; - goto error; + goto cleanup; } nlmsg_set_dst(nl_msg, &nladdr); @@ -239,7 +239,7 @@ int virNetlinkCommand(struct nl_msg *nl_msg, virReportSystemError(errno, "%s", _("cannot send to netlink socket")); rc = -1; - goto error; + goto cleanup; } memset(fds, 0, sizeof(fds)); @@ -255,7 +255,7 @@ int virNetlinkCommand(struct nl_msg *nl_msg, virReportSystemError(ETIMEDOUT, "%s", _("no valid netlink response was received")); rc = -1; - goto error; + goto cleanup; } length = nl_recv(nlhandle, &nladdr, @@ -268,7 +268,7 @@ int virNetlinkCommand(struct nl_msg *nl_msg, } else { *respbuflen = length; } - error: + cleanup: if (rc == -1) { VIR_FREE(*resp); *resp = NULL; -- 1.7.12.4

1 0

[libvirt] [PATCH 0/2] *** replace error with cleanup in virNetlinkCommand ***
by Wangrui (K) 12 May '14

12 May '14

Zhang Bo (2): util: replace error with cleanup in virNetlinkCommand util: do not set resp to NULL after VIR_FREE src/util/virnetlink.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) -- 1.7.12.4

1 0

[libvirt] [RFC PATCH] Add support for invtsc timer
by Ján Tomko 12 May '14

12 May '14

Not yet merged in upstream QEMU: https://lists.gnu.org/archive/html/qemu-devel/2014-04/msg05024.html Add support for invariant TSC timer running at constant rate in all ACPI P-, C- and T-states. It can be enabled by specifying: <clock> <timer name='invtsc' present='yes'/> </clock> in the domain XML. Migration and saving the domain does not work with this timer. The support for this timer is indicated by bit 8 of EDX after calling CPUID with 0x80000007. It does not show up in /proc/cpuinfo [1] and since we're calling qemu without 'enforce', it doesn't error out if the host doesn't support this. Alternatively, we could expose it in libvirt as a cpu flag: <cpu mode='custom' match='exact'> <model fallback='forbid'>qemu64</model> <feature policy='require' name='invtsc'/> </cpu> or maybe add +invtsc to qemu args when the 'nonstop_tsc' flag is requested? [1]: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/arch/x… --- docs/formatdomain.html.in | 9 ++++++-- docs/schemas/domaincommon.rng | 1 + src/conf/domain_conf.c | 6 +++-- src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 6 +++++ src/qemu/qemu_migration.c | 14 ++++++++++++ .../qemuxml2argv-clock-timer-inv-tsc.args | 5 +++++ .../qemuxml2argv-clock-timer-inv-tsc.xml | 26 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + tests/qemuxml2xmltest.c | 1 + 10 files changed, 66 insertions(+), 4 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.xml diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 4f19473..1d3fd93 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -1368,13 +1368,18 @@ being modified, and can be one of "platform" (currently unsupported), "hpet" (libxl, xen, qemu), "kvmclock" (qemu), - "pit" (qemu), "rtc" (qemu), "tsc" (libxl) or "hypervclock" - (qemu - since 1.2.2). + "pit" (qemu), "rtc" (qemu), "tsc" (libxl), "hypervclock" + (qemu - since 1.2.2) or + "invtsc" (qemu - since 1.2.5). The <code>hypervclock</code> timer adds support for the reference time counter and the reference page for iTSC feature for guests running the Microsoft Windows operating system. + + The <code>invtsc</code> timer adds support for the invariant + TSC. It runs at a constant rate in all ACPI P- C- and T-states. + A guest with this timer enabled cannot be migrated or saved. </dd> <dt><code>track</code></dt> <dd> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 4249ed5..5154826 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -918,6 +918,7 @@ <choice> <value>kvmclock</value> <value>hypervclock</value> + <value>invtsc</value> </choice> </attribute> </group> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 6c3bdad..893d904 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -706,7 +706,8 @@ VIR_ENUM_IMPL(virDomainTimerName, VIR_DOMAIN_TIMER_NAME_LAST, "hpet", "tsc", "kvmclock", - "hypervclock"); + "hypervclock", + "invtsc"); VIR_ENUM_IMPL(virDomainTimerTrack, VIR_DOMAIN_TIMER_TRACK_LAST, "boot", @@ -2931,7 +2932,8 @@ virDomainDefPostParseInternal(virDomainDefPtr def, virDomainTimerDefPtr timer = def->clock.timers[i]; if (timer->name == VIR_DOMAIN_TIMER_NAME_KVMCLOCK || - timer->name == VIR_DOMAIN_TIMER_NAME_HYPERVCLOCK) { + timer->name == VIR_DOMAIN_TIMER_NAME_HYPERVCLOCK || + timer->name == VIR_DOMAIN_TIMER_NAME_INVTSC) { if (timer->tickpolicy != -1) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("timer %s doesn't support setting of " diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index a92f0f3..53c02e6 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1635,6 +1635,7 @@ enum virDomainTimerNameType { VIR_DOMAIN_TIMER_NAME_TSC, VIR_DOMAIN_TIMER_NAME_KVMCLOCK, VIR_DOMAIN_TIMER_NAME_HYPERVCLOCK, + VIR_DOMAIN_TIMER_NAME_INVTSC, VIR_DOMAIN_TIMER_NAME_LAST }; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 6c1e17d..2994427 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6736,6 +6736,11 @@ qemuBuildCpuArgStr(virQEMUDriverPtr driver, have_cpu ? "" : default_model, timer->present ? '+' : '-'); have_cpu = true; + } else if (timer->name == VIR_DOMAIN_TIMER_NAME_INVTSC) { + virBufferAsprintf(&buf, "%s,%cinvtsc", + have_cpu ? "" : default_model, + timer->present ? '+' : '-'); + have_cpu = true; } else if (timer->name == VIR_DOMAIN_TIMER_NAME_HYPERVCLOCK && timer->present) { virBufferAsprintf(&buf, "%s,hv_time", @@ -8063,6 +8068,7 @@ qemuBuildCommandLine(virConnectPtr conn, case VIR_DOMAIN_TIMER_NAME_KVMCLOCK: case VIR_DOMAIN_TIMER_NAME_HYPERVCLOCK: + case VIR_DOMAIN_TIMER_NAME_INVTSC: /* Timers above are handled when building -cpu. */ case VIR_DOMAIN_TIMER_NAME_LAST: break; diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index a9f7fea..c1ffc0f 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1513,6 +1513,20 @@ qemuMigrationIsAllowed(virQEMUDriverPtr driver, virDomainObjPtr vm, return false; } + for (i = 0; i < def->clock.ntimers; i++) { + virDomainTimerDefPtr timer = def->clock.timers[i]; + + if (timer->present != 1) + continue; + + if (timer->name == VIR_DOMAIN_TIMER_NAME_INVTSC) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("domain has '%s' timer"), + virDomainTimerNameTypeToString(timer->name)); + return false; + } + } + return true; } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.args b/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.args new file mode 100644 index 0000000..ae74ae8 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.args @@ -0,0 +1,5 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ +/usr/bin/kvm -S -M pc \ +-cpu qemu32,+invtsc -m 214 -smp 6 \ +-nographic -monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot n -usb -net \ +none -serial none -parallel none diff --git a/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.xml b/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.xml new file mode 100644 index 0000000..b4a82e8 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-clock-timer-inv-tsc.xml @@ -0,0 +1,26 @@ +<domain type='kvm'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219100</memory> + <currentMemory unit='KiB'>219100</currentMemory> + <vcpu placement='static'>6</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='network'/> + </os> + <features> + <pae/> + </features> + <clock offset='utc'> + <timer name='invtsc' present='yes'/> + </clock> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/kvm</emulator> + <controller type='usb' index='0'/> + <controller type='pci' index='0' model='pci-root'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 14482fd..3c5024e 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -641,6 +641,7 @@ mymain(void) DO_TEST("cpu-host-kvmclock", QEMU_CAPS_ENABLE_KVM, QEMU_CAPS_CPU_HOST); DO_TEST("kvmclock", QEMU_CAPS_KVM); DO_TEST("clock-timer-hyperv-rtc", QEMU_CAPS_KVM); + DO_TEST("clock-timer-inv-tsc", QEMU_CAPS_KVM); DO_TEST("cpu-eoi-disabled", QEMU_CAPS_ENABLE_KVM); DO_TEST("cpu-eoi-enabled", QEMU_CAPS_ENABLE_KVM); diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 3ea03e6..a659cd0 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -185,6 +185,7 @@ mymain(void) DO_TEST("clock-catchup"); DO_TEST("kvmclock"); DO_TEST("clock-timer-hyperv-rtc"); + DO_TEST("clock-timer-inv-tsc"); DO_TEST("cpu-eoi-disabled"); DO_TEST("cpu-eoi-enabled"); -- 1.8.3.2

5 7

[libvirt] [PATCH] maint: prohibit empty first lines
by Martin Kletzander 12 May '14

12 May '14

I just found this in one of my branches and I remembered I wanted to send this right after a discussion with Eric, but I forgot, somehow: https://www.redhat.com/archives/libvir-list/2014-March/msg01001.html Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- cfg.mk | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/cfg.mk b/cfg.mk index 3f4bba0..b2d12f9 100644 --- a/cfg.mk +++ b/cfg.mk @@ -929,6 +929,13 @@ sc_prohibit_mixed_case_abbreviations: halt='Use PCI, USB, SCSI, not Pci, Usb, Scsi' \ $(_sc_search_regexp) +sc_prohibit_empty_first_line: + @awk 'BEGIN { fail=0; } \ + FNR == 1 { if ($$0 == "") { print FILENAME:1:; fail=1; } } \ + END { if (fail == 1) { \ + print "$(ME): Prohibited empty first line" > "/dev/stderr"; \ + } exit fail; }' $$($(VC_LIST_EXCEPT) | grep '\.[ch]$$'); + # We don't use this feature of maint.mk. prev_version_file = /dev/null -- 1.9.2

1 0

[libvirt] [PATCH] qemu: Implement a stub cpuArchDriver.compare() handler for arm and aarch64
by Oleg Strikov 12 May '14

12 May '14

Libvirt calls cpuArchDriver.compare() while doing guest migration. We don't have any logic to distinguish between different arm and aarch64 models that's why this patch allows migration to any host. Signed-off-by: Oleg Strikov <oleg.strikov(a)canonical.com> --- src/cpu/cpu_aarch64.c | 9 ++++++++- src/cpu/cpu_arm.c | 9 ++++++++- 2 files changed, 16 insertions(+), 2 deletions(-) diff --git a/src/cpu/cpu_aarch64.c b/src/cpu/cpu_aarch64.c index 3c3e749..7255d9f 100644 --- a/src/cpu/cpu_aarch64.c +++ b/src/cpu/cpu_aarch64.c @@ -108,11 +108,18 @@ AArch64Baseline(virCPUDefPtr *cpus, return cpu; } +static virCPUCompareResult +AArch64Compare(virCPUDefPtr host ATTRIBUTE_UNUSED, + virCPUDefPtr cpu ATTRIBUTE_UNUSED) +{ + return VIR_CPU_COMPARE_IDENTICAL; +} + struct cpuArchDriver cpuDriverAARCH64 = { .name = "aarch64", .arch = archs, .narch = ARRAY_CARDINALITY(archs), - .compare = NULL, + .compare = AArch64Compare, .decode = AArch64Decode, .encode = NULL, .free = AArch64DataFree, diff --git a/src/cpu/cpu_arm.c b/src/cpu/cpu_arm.c index ec80b35..53bf4f4 100644 --- a/src/cpu/cpu_arm.c +++ b/src/cpu/cpu_arm.c @@ -109,11 +109,18 @@ ArmBaseline(virCPUDefPtr *cpus, return cpu; } +static virCPUCompareResult +ArmCompare(virCPUDefPtr host ATTRIBUTE_UNUSED, + virCPUDefPtr cpu ATTRIBUTE_UNUSED) +{ + return VIR_CPU_COMPARE_IDENTICAL; +} + struct cpuArchDriver cpuDriverArm = { .name = "arm", .arch = archs, .narch = ARRAY_CARDINALITY(archs), - .compare = NULL, + .compare = ArmCompare, .decode = ArmDecode, .encode = NULL, .free = ArmDataFree, -- 1.7.9.5

2 1