- Devel - libvirt List Archives

[libvirt] [tck PATCH] require libguestfs-tool package
by Laine Stump 26 Mar '18

26 Mar '18

The nwfilter tests use virt-builder to download and configure a guest appliance, but we don't require the libguestfs-tools package. Up until now it's just happened that it was always already installed on the test systems, so it wasn't a problem, but it's annoying to have to type the one extra install command when setting up a new machine to run the tests. Signed-off-by: Laine Stump <laine(a)laine.org> --- perl-Sys-Virt-TCK.spec.PL | 1 + 1 file changed, 1 insertion(+) diff --git a/perl-Sys-Virt-TCK.spec.PL b/perl-Sys-Virt-TCK.spec.PL index 7c83be9..9694227 100644 --- a/perl-Sys-Virt-TCK.spec.PL +++ b/perl-Sys-Virt-TCK.spec.PL @@ -75,6 +75,7 @@ Requires: perl(TAP::Formatter::JUnit) Requires: perl(TAP::Harness::Archive) Requires: perl(Net::OpenSSH) Requires: perl(IO::Pty) +Requires: libguestfs-tools Requires: /usr/bin/mkisofs BuildArchitectures: noarch -- 2.14.3

2 1

[libvirt] [PATCH 0/2] Fix migration with implicit controllers (user-alias saga)
by Ján Tomko 26 Mar '18

26 Mar '18

*** GLUTEN-FREE BLURB *** Ján Tomko (2): virDomainDeviceAliasIsUserAlias: tolerate NULL qemu: do not drop implicit controllers with non-implicit attributes src/conf/domain_conf.c | 2 +- src/qemu/qemu_domain.c | 7 +++++-- 2 files changed, 6 insertions(+), 3 deletions(-) -- 2.13.6

2 4

[libvirt] persistence of pcihole64 over live migration
by Richard, Joseph 25 Mar '18

25 Mar '18

Hello all, In order to reserve 64-bit address space for hotplugging, we are using the pcihole64 attribute on the root pci-controller. When doing a live-migration using libvirt/qemu, the pci-root controller is dropped and re-added on the destination side[1], which results in the pci controller's pcihole64 attribute that we configured when launching the VM being lost. Is there any reason to not keep this (by not dropping it in qemuDomainDefFormatBufInternal when pcihole64 is set on the controller) over a live migration in this case? [1] https://github.com/libvirt/libvirt/commit/b33eb0dca1175bc0e3f796f18e31adc1d… Thanks, Joseph Richard

3 2

[libvirt] [PATCH] build: Remove --with-xml-catalog-file option
by Rainer Müller 25 Mar '18

25 Mar '18

After validation against XHTML 1.0 was dropped in f802c9de0, the XML_CATALOG_FILE is not in use anymore. Therefore the checks in configure can be removed. --- configure.ac | 3 --- m4/virt-xml-catalog.m4 | 33 --------------------------------- 2 files changed, 36 deletions(-) delete mode 100644 m4/virt-xml-catalog.m4 diff --git a/configure.ac b/configure.ac index 56a7ad780..115eb6088 100644 --- a/configure.ac +++ b/configure.ac @@ -459,7 +459,6 @@ dnl LIBVIRT_ARG_DEBUG LIBVIRT_ARG_DTRACE LIBVIRT_ARG_NUMAD -LIBVIRT_ARG_XML_CATALOG LIBVIRT_ARG_INIT_SCRIPT LIBVIRT_ARG_CHRDEV_LOCK_FILES LIBVIRT_ARG_DEFAULT_EDITOR @@ -473,7 +472,6 @@ LIBVIRT_ARG_SYSCTL_CONFIG LIBVIRT_CHECK_DEBUG LIBVIRT_CHECK_DTRACE LIBVIRT_CHECK_NUMAD -LIBVIRT_CHECK_XML_CATALOG LIBVIRT_CHECK_INIT_SCRIPT LIBVIRT_CHECK_CHRDEV_LOCK_FILES LIBVIRT_CHECK_DEFAULT_EDITOR @@ -1032,7 +1030,6 @@ AC_MSG_NOTICE([ Use -Werror: $enable_werror]) AC_MSG_NOTICE([ Warning Flags: $WARN_CFLAGS]) LIBVIRT_RESULT_DTRACE LIBVIRT_RESULT_NUMAD -LIBVIRT_RESULT_XML_CATALOG LIBVIRT_RESULT_INIT_SCRIPT LIBVIRT_RESULT_CHRDEV_LOCK_FILES LIBVIRT_RESULT_DEFAULT_EDITOR diff --git a/m4/virt-xml-catalog.m4 b/m4/virt-xml-catalog.m4 deleted file mode 100644 index a25ae89f6..000000000 --- a/m4/virt-xml-catalog.m4 +++ /dev/null @@ -1,33 +0,0 @@ -dnl The XML catalog file check -dnl -dnl Copyright (C) 2016 Red Hat, Inc. -dnl -dnl This library is free software; you can redistribute it and/or -dnl modify it under the terms of the GNU Lesser General Public -dnl License as published by the Free Software Foundation; either -dnl version 2.1 of the License, or (at your option) any later version. -dnl -dnl This library is distributed in the hope that it will be useful, -dnl but WITHOUT ANY WARRANTY; without even the implied warranty of -dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU -dnl Lesser General Public License for more details. -dnl -dnl You should have received a copy of the GNU Lesser General Public -dnl License along with this library. If not, see -dnl <http://www.gnu.org/licenses/>. -dnl - -AC_DEFUN([LIBVIRT_ARG_XML_CATALOG], [ - LIBVIRT_ARG_WITH([XML_CATALOG_FILE], - [path to XML catalog file for validating generated html], - ['/etc/xml/catalog']) -]) - -AC_DEFUN([LIBVIRT_CHECK_XML_CATALOG], [ - dnl Specific XML catalog file for validation of generated html - AC_SUBST([XML_CATALOG_FILE], [$with_xml_catalog_file]) -]) - -AC_DEFUN([LIBVIRT_RESULT_XML_CATALOG], [ - AC_MSG_NOTICE([ XML Catalog: $with_xml_catalog_file]) -]) -- 2.16.2

2 3

[libvirt] [PATCH] polkit: reintroduce check for pkcheck
by Ján Tomko 23 Mar '18

23 Mar '18

Commit 2499d1a0 was too eager and possibly enabled polkit on all platforms with D-Bus, regardless of whether they use polkit. Reintroduce the usage of pkcheck as a witness for --with-polkit=check, but do not require it for --with-polkit=yes. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> Reported-by: Jiří Denemark <jdenemar(a)redhat.com> --- m4/virt-polkit.m4 | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/m4/virt-polkit.m4 b/m4/virt-polkit.m4 index fd729dc98..16889e163 100644 --- a/m4/virt-polkit.m4 +++ b/m4/virt-polkit.m4 @@ -25,8 +25,17 @@ AC_DEFUN([LIBVIRT_ARG_POLKIT], [ AC_DEFUN([LIBVIRT_CHECK_POLKIT], [ AC_REQUIRE([LIBVIRT_CHECK_DBUS]) + if test "x$with_polkit" = "xcheck"; then + dnl For --with-polkit=check, also require the pkcheck binary, even + dnl though we talk to polkit directly over D-Bus. + AC_PATH_PROG([PKCHECK_PATH], [pkcheck], [], [$LIBVIRT_SBIN_PATH]) + if test "x$PKCHECK_PATH" = "x" ; then + with_polkit="no" + fi + fi + if test "x$with_polkit" = "xyes" || test "x$with_polkit" = "xcheck"; then - dnl All we need to talk to polkit is D-Bus. + dnl For --with-polkit=yes, all we need is D-Bus. if test "x$with_dbus" = "xyes" ; then AC_DEFINE_UNQUOTED([WITH_POLKIT], 1, [use PolicyKit for UNIX socket access checks]) -- 2.16.1

2 1

[libvirt] [PATCH v3 0/6] Add support for video and input devices on S390
by Farhan Ali 23 Mar '18

23 Mar '18

Hi, This patch series adds Libvirt support for video and input devices for QEMU guests on S390. QEMU v2.11.0 added support for the virtio-gpu-ccw device [1] and virtio-{keyboard, mouse, tablet}-ccw devices [2], which can be used as video and input devices respectively. Thanks Farhan [1] https://git.qemu.org/?p=qemu.git;a=commit;h=1f8ad88935f5cb5a2968909e392dbee… [2] https://git.qemu.org/?p=qemu.git;a=commit;h=3382cf1fabbf722dce931846853dae7… ChangeLog --------- v2 -> v3 - Update 2.12 capabilities files (patch 1 and 4). - Move video model change to it's own separate patch (patch 3). - Separate virtio-gpu-ccw tests into appropriate patches as per Jan's suggestions (patch 1, 2 and 3). - Squash virtio input ccw tests with functionality patch (patch 5). v1 -> v2 - Split virtio-gpu-ccw capability (patch 1) and functionality (patch 2) into 2 patches. - Code changes per John's comments (patch 2). - Split capability for input devices (patch 4) and functionality (patch 5) into 2 patches. - Modify news update as per John's suggestion (patch 7). Farhan Ali (6): qemu: Introduce a new capability for virtio-gpu-ccw qemu: Add support for virtio-gpu-ccw video device on S390 qemu: Change default video model type to virtio for S390 qemu: Introduce capabilities for virtio input ccw devices qemu: Add support for virtio input ccw devices news: Update for virtio-gpu-ccw and virtio input ccw devices docs/formatdomain.html.in | 5 ++ docs/news.xml | 10 +++ src/qemu/qemu_capabilities.c | 13 ++++ src/qemu/qemu_capabilities.h | 6 ++ src/qemu/qemu_command.c | 27 +++++-- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_domain_address.c | 12 +++- src/qemu/qemu_process.c | 5 +- .../qemucapabilitiesdata/caps_2.11.0.s390x.replies | 83 +++++++++++++++++++--- tests/qemucapabilitiesdata/caps_2.11.0.s390x.xml | 6 +- .../qemucapabilitiesdata/caps_2.12.0.s390x.replies | 81 ++++++++++++++++++--- tests/qemucapabilitiesdata/caps_2.12.0.s390x.xml | 4 ++ tests/qemuxml2argvdata/input-virtio-ccw.args | 26 +++++++ tests/qemuxml2argvdata/input-virtio-ccw.xml | 29 ++++++++ .../qemuxml2argvdata/video-virtio-gpu-ccw-auto.xml | 18 +++++ tests/qemuxml2argvdata/video-virtio-gpu-ccw.args | 25 +++++++ tests/qemuxml2argvdata/video-virtio-gpu-ccw.xml | 35 +++++++++ tests/qemuxml2argvtest.c | 15 ++++ tests/qemuxml2xmloutdata/input-virtio-ccw.xml | 37 ++++++++++ .../video-virtio-gpu-ccw-auto.xml | 35 +++++++++ tests/qemuxml2xmloutdata/video-virtio-gpu-ccw.xml | 39 ++++++++++ tests/qemuxml2xmltest.c | 22 ++++++ 22 files changed, 508 insertions(+), 27 deletions(-) create mode 100644 tests/qemuxml2argvdata/input-virtio-ccw.args create mode 100644 tests/qemuxml2argvdata/input-virtio-ccw.xml create mode 100644 tests/qemuxml2argvdata/video-virtio-gpu-ccw-auto.xml create mode 100644 tests/qemuxml2argvdata/video-virtio-gpu-ccw.args create mode 100644 tests/qemuxml2argvdata/video-virtio-gpu-ccw.xml create mode 100644 tests/qemuxml2xmloutdata/input-virtio-ccw.xml create mode 100644 tests/qemuxml2xmloutdata/video-virtio-gpu-ccw-auto.xml create mode 100644 tests/qemuxml2xmloutdata/video-virtio-gpu-ccw.xml -- 2.7.4

2 7

[libvirt] [PATCH 0/3] tests: Improvements and cleanups to capabilities testing
by Andrea Bolognani 23 Mar '18

23 Mar '18

*** BLURB NOW OR FOREVER HOLD YOUR PEACE *** Andrea Bolognani (3): tests: domaincaps: Order by architecture tests: domaincaps: Add QEMU 2.12 tests: qemucapabilities: Drop mostly duplicated data ...rt.aarch64.xml => qemu_2.12.0-virt.aarch64.xml} | 5 +- ...icv2-virt.aarch64.xml => qemu_2.12.0.ppc64.xml} | 51 +- ...icv2-virt.aarch64.xml => qemu_2.12.0.s390x.xml} | 50 +- tests/domaincapsschemadata/qemu_2.12.0.x86_64.xml | 143 + ...irt.aarch64.xml => qemu_2.6.0-virt.aarch64.xml} | 0 tests/domaincapstest.c | 44 +- .../caps_2.10.0-gicv3.aarch64.replies | 16495 ----------------- .../caps_2.10.0-gicv3.aarch64.xml | 310 - ...aarch64.replies => caps_2.10.0.aarch64.replies} | 0 ...0-gicv2.aarch64.xml => caps_2.10.0.aarch64.xml} | 0 .../caps_2.12.0-gicv2.aarch64.replies | 18551 ------------------- .../caps_2.12.0-gicv2.aarch64.xml | 322 - ...aarch64.replies => caps_2.12.0.aarch64.replies} | 0 ...0-gicv3.aarch64.xml => caps_2.12.0.aarch64.xml} | 0 .../caps_2.6.0-gicv3.aarch64.replies | 13035 ------------- .../caps_2.6.0-gicv3.aarch64.xml | 287 - ....aarch64.replies => caps_2.6.0.aarch64.replies} | 0 ....0-gicv2.aarch64.xml => caps_2.6.0.aarch64.xml} | 0 tests/qemucapabilitiestest.c | 9 +- 19 files changed, 192 insertions(+), 49110 deletions(-) rename tests/domaincapsschemadata/{qemu_2.6.0-gicv3-virt.aarch64.xml => qemu_2.12.0-virt.aarch64.xml} (95%) copy tests/domaincapsschemadata/{qemu_2.6.0-gicv2-virt.aarch64.xml => qemu_2.12.0.ppc64.xml} (50%) copy tests/domaincapsschemadata/{qemu_2.6.0-gicv2-virt.aarch64.xml => qemu_2.12.0.s390x.xml} (50%) create mode 100644 tests/domaincapsschemadata/qemu_2.12.0.x86_64.xml rename tests/domaincapsschemadata/{qemu_2.6.0-gicv2-virt.aarch64.xml => qemu_2.6.0-virt.aarch64.xml} (100%) delete mode 100644 tests/qemucapabilitiesdata/caps_2.10.0-gicv3.aarch64.replies delete mode 100644 tests/qemucapabilitiesdata/caps_2.10.0-gicv3.aarch64.xml rename tests/qemucapabilitiesdata/{caps_2.10.0-gicv2.aarch64.replies => caps_2.10.0.aarch64.replies} (100%) rename tests/qemucapabilitiesdata/{caps_2.10.0-gicv2.aarch64.xml => caps_2.10.0.aarch64.xml} (100%) delete mode 100644 tests/qemucapabilitiesdata/caps_2.12.0-gicv2.aarch64.replies delete mode 100644 tests/qemucapabilitiesdata/caps_2.12.0-gicv2.aarch64.xml rename tests/qemucapabilitiesdata/{caps_2.12.0-gicv3.aarch64.replies => caps_2.12.0.aarch64.replies} (100%) rename tests/qemucapabilitiesdata/{caps_2.12.0-gicv3.aarch64.xml => caps_2.12.0.aarch64.xml} (100%) delete mode 100644 tests/qemucapabilitiesdata/caps_2.6.0-gicv3.aarch64.replies delete mode 100644 tests/qemucapabilitiesdata/caps_2.6.0-gicv3.aarch64.xml rename tests/qemucapabilitiesdata/{caps_2.6.0-gicv2.aarch64.replies => caps_2.6.0.aarch64.replies} (100%) rename tests/qemucapabilitiesdata/{caps_2.6.0-gicv2.aarch64.xml => caps_2.6.0.aarch64.xml} (100%) -- 2.14.3

2 4

[libvirt] [PATCH] conf: avoid reporting errors when network driver is disabled
by Daniel P. Berrangé 23 Mar '18

23 Mar '18

In previous releases all these methods were a no-op if the network driver is disabled. These helper methods are called unconditionally for all types of network interface, so must be no-ops if missing. Other code will already generate an error if the network driver is disabled and a NIC with type=network is used. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/conf/domain_conf.c | 23 +++++++++-------------- 1 file changed, 9 insertions(+), 14 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index c8d051fa9f..79d6bd378e 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -28979,10 +28979,13 @@ int virDomainNetAllocateActualDevice(virDomainDefPtr dom, virDomainNetDefPtr iface) { + /* Just silently ignore if network driver isn't present. If something + * has tried to use a NIC with type=network, other code will already + * cause an error. This ensures type=bridge doesn't break when + * network driver is compiled out. + */ if (!netAllocate) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("Network device allocation not available")); - return -1; + return 0; } return netAllocate(dom, iface); @@ -28993,8 +28996,6 @@ virDomainNetNotifyActualDevice(virDomainDefPtr dom, virDomainNetDefPtr iface) { if (!netNotify) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("Network device notification not available")); return; } @@ -29007,9 +29008,7 @@ virDomainNetReleaseActualDevice(virDomainDefPtr dom, virDomainNetDefPtr iface) { if (!netRelease) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("Network device release not available")); - return -1; + return 0; } return netRelease(dom, iface); @@ -29020,9 +29019,7 @@ virDomainNetBandwidthChangeAllowed(virDomainNetDefPtr iface, virNetDevBandwidthPtr newBandwidth) { if (!netBandwidthChangeAllowed) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("Network device bandwidth change query not available")); - return -1; + return 0; } return netBandwidthChangeAllowed(iface, newBandwidth); @@ -29033,9 +29030,7 @@ virDomainNetBandwidthUpdate(virDomainNetDefPtr iface, virNetDevBandwidthPtr newBandwidth) { if (!netBandwidthUpdate) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("Network device bandwidth update not available")); - return -1; + return 0; } return netBandwidthUpdate(iface, newBandwidth); -- 2.14.3

2 2

[libvirt] [PATCH python] Disable Python 2 on future RHEL and Fedora > 29
by Daniel P. Berrangé 23 Mar '18

23 Mar '18

The end of Python 2 is nearing, so declare when it will be removed from libvirt Python binding RPMs. NB, this doesn't imply we'll be removing py2 support from upstream libvirt-python on the same timeframe - we'll consider what todo wrt that when upstream Python 2 finally goes EOL. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- libvirt-python.spec.in | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/libvirt-python.spec.in b/libvirt-python.spec.in index 7953c82..9697ea8 100644 --- a/libvirt-python.spec.in +++ b/libvirt-python.spec.in @@ -13,6 +13,10 @@ %endif %define _with_python2 1 +%if 0%{?fedora} > 29 || 0%{?rhel} > 7 +%define _with_python2 0 +%endif + %define _with_python3 0 %if 0%{?fedora} || 0%{?rhel} > 7 %define _with_python3 1 -- 2.14.3

2 1

[libvirt] [dbus PATCH 00/10] *** Use API calls supporting flags ***
by Katerina Koukiou 23 Mar '18

23 Mar '18

* Replace all API calls to support flags where possible. * Replace libvirt-dbus API method names to follow naming of libvirt API. * Change type of virtDBusDomainGetVcpus from property to method. Katerina Koukiou (10): virtDBusDomainGetVcpus: Should be implemented as method and not as property GetVcpus API method: Renamed to GetVcpusFlags virtDBusDomainShutdown: Use virDomainShutdownFlags instead of virDomainShutdown Shutdown API method: Renamed to ShutdownFlags virtDBusDomainCreate: Use virDomainCreateWithFlags instead of virDomainCreate Create API method: Renamed to CreateWithFlags virtDBusDomainUndefine: Use virDomainUndefineFlags instead of virDomainUndefine Undefine API method: Renamed to UndefineFlags virtDBusDomainDestroy: Use virDomainDestroyFlags instead of virDomainDestroy Destroy API method: Renamed to DestroyFlags data/org.libvirt.Domain.xml | 21 +++++-- src/domain.c | 134 +++++++++++++++++++++++++------------------- test/test_domain.py | 22 +++++--- 3 files changed, 106 insertions(+), 71 deletions(-) -- 2.15.0

3 14

[libvirt] [PATCH] qemu: Build smartcard command line more wisely
by Michal Privoznik 23 Mar '18

23 Mar '18

https://bugzilla.redhat.com/show_bug.cgi?id=1558317 Similarly to b133fac356324c3 we need to look up alias of CCID controller when constructing smartcard command line instead of relying on broken assumption it will always be 'ccid0'. After user aliases it can be anything. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_command.c | 38 +++++++++++++++++++++++++++++++- tests/qemuxml2argvdata/user-aliases.args | 3 +++ tests/qemuxml2argvdata/user-aliases.xml | 11 +++++++++ tests/qemuxml2argvtest.c | 3 ++- 4 files changed, 53 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index a8afbd14fa..937a7d1b7b 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -8701,6 +8701,32 @@ qemuBuildNetCommandLine(virQEMUDriverPtr driver, } +static const char * +qemuBuildSmartcardFindCCIDController(const virDomainDef *def, + const virDomainSmartcardDef *smartcard) +{ + size_t i; + + /* Should never happen. But doesn't hurt to check. */ + if (smartcard->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCID) + return NULL; + + for (i = 0; i < def->ncontrollers; i++) { + const virDomainControllerDef *tmp = def->controllers[i]; + + if (tmp->type != VIR_DOMAIN_CONTROLLER_TYPE_CCID) + continue; + + if (tmp->idx != smartcard->info.addr.ccid.controller) + continue; + + return tmp->info.alias; + } + + return NULL; +} + + static int qemuBuildSmartcardCommandLine(virLogManagerPtr logManager, virCommandPtr cmd, @@ -8714,6 +8740,7 @@ qemuBuildSmartcardCommandLine(virLogManagerPtr logManager, char *devstr; virBuffer opt = VIR_BUFFER_INITIALIZER; const char *database; + const char *contAlias = NULL; if (!def->nsmartcards) return 0; @@ -8811,8 +8838,17 @@ qemuBuildSmartcardCommandLine(virLogManagerPtr logManager, virBufferFreeAndReset(&opt); return -1; } + + if (!(contAlias = qemuBuildSmartcardFindCCIDController(def, + smartcard))) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unable to find controller for %s"), + smartcard->info.alias); + return -1; + } + virCommandAddArg(cmd, "-device"); - virBufferAsprintf(&opt, ",id=%s,bus=ccid0.0", smartcard->info.alias); + virBufferAsprintf(&opt, ",id=%s,bus=%s.0", smartcard->info.alias, contAlias); virCommandAddArgBuffer(cmd, &opt); return 0; diff --git a/tests/qemuxml2argvdata/user-aliases.args b/tests/qemuxml2argvdata/user-aliases.args index 4c9c951cef..ad93947104 100644 --- a/tests/qemuxml2argvdata/user-aliases.args +++ b/tests/qemuxml2argvdata/user-aliases.args @@ -31,6 +31,8 @@ server,nowait \ -global PIIX4_PM.disable_s4=0 \ -boot cd \ -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x8 \ +-device usb-ccid,id=ua-myCCID,bus=ua-SomeWeirdController.0,port=1 \ +-device usb-ccid,id=ua-myCCID2,bus=ua-SomeWeirdController.0,port=2 \ -usb \ -drive file=/var/lib/libvirt/images/fd.img,format=raw,if=none,\ id=drive-ua-myDisk1,cache=none \ @@ -55,6 +57,7 @@ bus=pci.0,addr=0x9 \ -device rtl8139,vlan=2,id=ua-AndAlsoClientMode,mac=52:54:00:8c:b1:f8,bus=pci.0,\ addr=0xa \ -net socket,connect=127.0.0.1:1234,vlan=2,name=hostua-AndAlsoClientMode \ +-device ccid-card-emulated,backend=nss-emulated,id=smartcard0,bus=ua-myCCID.0 \ -chardev pty,id=charserial0 \ -device isa-serial,chardev=charserial0,id=serial0 \ -chardev pty,id=charserial1 \ diff --git a/tests/qemuxml2argvdata/user-aliases.xml b/tests/qemuxml2argvdata/user-aliases.xml index 52132a82d7..9ce123b477 100644 --- a/tests/qemuxml2argvdata/user-aliases.xml +++ b/tests/qemuxml2argvdata/user-aliases.xml @@ -85,6 +85,14 @@ <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </controller> <controller type='fdc' index='0'/> + <controller type='ccid' index='0'> + <alias name='ua-myCCID'/> + <address type='usb' bus='0' port='1'/> + </controller> + <controller type='ccid' index='1'> + <alias name='ua-myCCID2'/> + <address type='usb' bus='0' port='2'/> + </controller> <interface type='ethernet'> <mac address='52:54:00:d6:c0:0b'/> <model type='virtio'/> @@ -109,6 +117,9 @@ <alias name='ua-AndAlsoClientMode'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x0a' function='0x0'/> </interface> + <smartcard mode='host'> + <address type='ccid' controller='0' slot='0'/> + </smartcard> <serial type='pty'> <target type='isa-serial' port='0'> <model name='isa-serial'/> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 5ece3f0cc6..064fd3b767 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -2995,7 +2995,8 @@ mymain(void) QEMU_CAPS_OBJECT_MEMORY_FILE, QEMU_CAPS_PIIX_DISABLE_S3, QEMU_CAPS_PIIX_DISABLE_S4, QEMU_CAPS_VNC, QEMU_CAPS_DEVICE_ISA_SERIAL, - QEMU_CAPS_HDA_DUPLEX); + QEMU_CAPS_HDA_DUPLEX, + QEMU_CAPS_CCID_EMULATED); DO_TEST("user-aliases2", QEMU_CAPS_DEVICE_IOH3420, QEMU_CAPS_ICH9_AHCI); DO_TEST("user-aliases-usb", QEMU_CAPS_KVM, QEMU_CAPS_PIIX_DISABLE_S3, QEMU_CAPS_PIIX_DISABLE_S4, -- 2.16.1

2 1

[libvirt] [PATCH] qemu: domain: Drop declaration of qemuDomainDefValidateDisk
by Peter Krempa 23 Mar '18

23 Mar '18

There is no such function in our code. Commit abca72faa49 added it spuriously. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- Pushed as trivial. src/qemu/qemu_domain.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 6d3e6eb5e3..63d9fb6d21 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -888,9 +888,6 @@ bool qemuDomainVcpuHotplugIsInOrder(virDomainDefPtr def) void qemuDomainVcpuPersistOrder(virDomainDefPtr def) ATTRIBUTE_NONNULL(1); -int qemuDomainDefValidateDisk(const virDomainDiskDef *disk, - virQEMUCapsPtr qemuCaps); - int qemuDomainCheckMonitor(virQEMUDriverPtr driver, virDomainObjPtr vm, qemuDomainAsyncJob asyncJob); -- 2.16.2

1 0

[libvirt] [PATCH 0/4] Be consistent with vir*Obj*Remove* APIs
by John Ferlan 23 Mar '18

23 Mar '18

Using comments from recent changes/reviews w/ libxl and usage of the virDomainObjListRemove which note that it's "a problem" that the Remove API expects a locked @obj on input, but then returns the @obj unlocked. Although adjusting domainobjlist is a "future" task, we can at least modify the other drivers/vir*obj modules to be consistent at least with respect to expecting and returning a locked (and reffed) object. Thus, modify interface, secret, nodedev, and storage pool. The network is already done this way and nwfilter appears to be a lost cause as we cannot come to a concensus over how to use hash objects and thus the code still uses forward linked list object manipulation. Changes for virdomainobjlist require a number of other changes to happen first with respect to consistent usage of Find and Add API's to return locked and ref counted objects, then being able to have the virDomainObjListRemove be consistent as well. John Ferlan (4): secret: Return with locked obj from virSecretObjListRemove interface: Return with locked obj from virInterfaceObjListRemove nodedev: Return with locked obj from virNodeDeviceObjListRemove storagepool: Return with locked obj from virStoragePoolObjRemove src/conf/virinterfaceobj.c | 13 ++++++++++++- src/conf/virnodedeviceobj.c | 13 ++++++++++++- src/conf/virsecretobj.c | 15 ++++++++------- src/conf/virstorageobj.c | 17 ++++++++++++++--- src/node_device/node_device_hal.c | 4 ++-- src/node_device/node_device_udev.c | 2 +- src/secret/secret_driver.c | 4 ---- src/storage/storage_driver.c | 12 +----------- src/test/test_driver.c | 20 +++++--------------- 9 files changed, 55 insertions(+), 45 deletions(-) -- 2.13.6

2 6

[libvirt] [PATCH] lxc: report error message raised by the failing function
by Prafull 23 Mar '18

23 Mar '18

The code that calls VIR_WARN after a function fails, doesn't report the error message raised by the failing function. Such error messages are now reported in lxc/lxc_driver.c Signed-off-by: Prafullkumar T <talep158(a)gmail.com> --- src/lxc/lxc_driver.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 4f6b93b..4f600f3 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -3915,8 +3915,8 @@ lxcDomainAttachDeviceDiskLive(virLXCDriverPtr driver, major(sb.st_rdev), minor(sb.st_rdev), perms) < 0) - VIR_WARN("cannot deny device %s for domain %s", - src, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + src, vm->def->name, virGetLastErrorMessage()); goto cleanup; } @@ -4011,8 +4011,8 @@ lxcDomainAttachDeviceNetLive(virConnectPtr conn, goto cleanup; } else { VIR_WARN("setting bandwidth on interfaces of " - "type '%s' is not implemented yet", - virDomainNetTypeToString(actualType)); + "type '%s' is not implemented yet: %s", + virDomainNetTypeToString(actualType), virGetLastErrorMessage()); } } @@ -4116,8 +4116,8 @@ lxcDomainAttachDeviceHostdevSubsysUSBLive(virLXCDriverPtr driver, if (virUSBDeviceFileIterate(usb, virLXCTeardownHostUSBDeviceCgroup, priv->cgroup) < 0) - VIR_WARN("cannot deny device %s for domain %s", - src, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + src, vm->def->name, virGetLastErrorMessage()); goto cleanup; } @@ -4190,8 +4190,8 @@ lxcDomainAttachDeviceHostdevStorageLive(virLXCDriverPtr driver, major(sb.st_rdev), minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) - VIR_WARN("cannot deny device %s for domain %s", - def->source.caps.u.storage.block, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + def->source.caps.u.storage.block, vm->def->name, virGetLastErrorMessage()); goto cleanup; } @@ -4262,8 +4262,8 @@ lxcDomainAttachDeviceHostdevMiscLive(virLXCDriverPtr driver, major(sb.st_rdev), minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) - VIR_WARN("cannot deny device %s for domain %s", - def->source.caps.u.storage.block, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + def->source.caps.u.storage.block, vm->def->name, virGetLastErrorMessage()); goto cleanup; } @@ -4434,8 +4434,8 @@ lxcDomainDetachDeviceDiskLive(virDomainObjPtr vm, if (virCgroupDenyDevicePath(priv->cgroup, src, VIR_CGROUP_DEVICE_RWM, false) != 0) - VIR_WARN("cannot deny device %s for domain %s", - src, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + src, vm->def->name, virGetLastErrorMessage()); virDomainDiskRemove(vm->def, idx); virDomainDiskDefFree(def); @@ -4567,8 +4567,8 @@ lxcDomainDetachDeviceHostdevUSBLive(virLXCDriverPtr driver, if (virUSBDeviceFileIterate(usb, virLXCTeardownHostUSBDeviceCgroup, priv->cgroup) < 0) - VIR_WARN("cannot deny device %s for domain %s", - dst, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + dst, vm->def->name, virGetLastErrorMessage()); virObjectLock(hostdev_mgr->activeUSBHostdevs); virUSBDeviceListDel(hostdev_mgr->activeUSBHostdevs, usb); @@ -4623,8 +4623,8 @@ lxcDomainDetachDeviceHostdevStorageLive(virDomainObjPtr vm, if (virCgroupDenyDevicePath(priv->cgroup, def->source.caps.u.storage.block, VIR_CGROUP_DEVICE_RWM, false) != 0) - VIR_WARN("cannot deny device %s for domain %s", - def->source.caps.u.storage.block, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + def->source.caps.u.storage.block, vm->def->name, virGetLastErrorMessage()); virDomainHostdevRemove(vm->def, idx); virDomainHostdevDefFree(def); @@ -4673,8 +4673,8 @@ lxcDomainDetachDeviceHostdevMiscLive(virDomainObjPtr vm, if (virCgroupDenyDevicePath(priv->cgroup, def->source.caps.u.misc.chardev, VIR_CGROUP_DEVICE_RWM, false) != 0) - VIR_WARN("cannot deny device %s for domain %s", - def->source.caps.u.misc.chardev, vm->def->name); + VIR_WARN("cannot deny device %s for domain %s: %s", + def->source.caps.u.misc.chardev, vm->def->name, virGetLastErrorMessage()); virDomainHostdevRemove(vm->def, idx); virDomainHostdevDefFree(def); -- 2.1.4

3 2

[libvirt] [PATCH] remote: remove some __sun conditionals
by Daniel P. Berrangé 23 Mar '18

23 Mar '18

The libvirtd daemon has some arbitrary logic to drop privileges, but only on Solaris platforms. This was added during Xen days, when Xen was the only driver running in libvirtd. There's no expectation or testing that this works with the new libxl stack, nor whether dropping privileges breaks any of the secondary drivers. Finally, we'll be splitting drivers out into their own independant daemons, so this won't be applicable to libvirtd in future anyway. The remote driver client meanwhile arbitrarily disables daemon auto-spawn when connecting as non-root, breaking a key feature of libvirt unprivileged connections. Since we've not had any contributions for Solaris since circa 2012 and we don't do any CI testing we should consider this platform unmaintained and thus reasonable to remove this cruft. If someone steps forward to maintain Solaris again, this code would need re-evaluating to come up with something more targetted. There's various __sun conditionals in the Xen driver code, but those are not touched. This is all for the legacy Xen driver, which will be entirely removed at some point in future, so not benefit to hacking out just the Solaris parts. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/remote/remote_daemon.c | 33 --------------------------------- src/remote/remote_driver.c | 2 -- 2 files changed, 35 deletions(-) diff --git a/src/remote/remote_daemon.c b/src/remote/remote_daemon.c index 2f9dfad443..31c6ce1b61 100644 --- a/src/remote/remote_daemon.c +++ b/src/remote/remote_daemon.c @@ -675,30 +675,6 @@ daemonVersion(const char *argv0) printf("%s (%s) %s\n", argv0, PACKAGE_NAME, PACKAGE_VERSION); } -#ifdef __sun -static int -daemonSetupPrivs(void) -{ - chown("/var/run/libvirt", SYSTEM_UID, SYSTEM_UID); - - if (__init_daemon_priv(PU_RESETGROUPS | PU_CLEARLIMITSET, - SYSTEM_UID, SYSTEM_UID, PRIV_XVM_CONTROL, NULL)) { - VIR_ERROR(_("additional privileges are required")); - return -1; - } - - if (priv_set(PRIV_OFF, PRIV_ALLSETS, PRIV_FILE_LINK_ANY, PRIV_PROC_INFO, - PRIV_PROC_SESSION, PRIV_PROC_EXEC, PRIV_PROC_FORK, NULL)) { - VIR_ERROR(_("failed to set reduced privileges")); - return -1; - } - - return 0; -} -#else -# define daemonSetupPrivs() 0 -#endif - static void daemonShutdownHandler(virNetDaemonPtr dmn, siginfo_t *sig ATTRIBUTE_UNUSED, @@ -1336,15 +1312,6 @@ int main(int argc, char **argv) { goto cleanup; } - /* Beyond this point, nothing should rely on using - * getuid/geteuid() == 0, for privilege level checks. - */ - VIR_DEBUG("Dropping privileges (if required)"); - if (daemonSetupPrivs() < 0) { - ret = VIR_DAEMON_ERR_PRIVS; - goto cleanup; - } - daemonInitialize(); remoteProcs[REMOTE_PROC_AUTH_LIST].needAuth = false; diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index f765c09416..34e5ced0a9 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -1350,7 +1350,6 @@ remoteConnectOpen(virConnectPtr conn, */ if (!conn->uri) { VIR_DEBUG("Auto-probe remote URI"); -#ifndef __sun if (geteuid() > 0) { VIR_DEBUG("Auto-spawn user daemon instance"); rflags |= VIR_DRV_OPEN_REMOTE_USER; @@ -1359,7 +1358,6 @@ remoteConnectOpen(virConnectPtr conn, STRNEQ(autostart, "0"))) rflags |= VIR_DRV_OPEN_REMOTE_AUTOSTART; } -#endif } ret = doRemoteOpen(conn, priv, auth, conf, rflags); -- 2.14.3

2 1

[libvirt] [dbus PATCH v2] Convert testsuite from unittest to pytest
by Katerina Koukiou 23 Mar '18

23 Mar '18

Move to pytest to take advantage of: * tests parametrization * pdb support * more compact tests - not forcing large test classes Signed-off-by: Katerina Koukiou <kkoukiou(a)redhat.com> --- HACKING.md | 2 +- test/Makefile.am | 7 ++++--- test/conftest.py | 15 +++++++++++++++ test/libvirttest.py | 49 +++++++++++++++++++++++++++++-------------------- test/test_connect.py | 33 +++++++++++++++++---------------- test/test_domain.py | 37 +++++++++++++++++++------------------ test/travis-run | 2 +- 7 files changed, 86 insertions(+), 59 deletions(-) create mode 100644 test/conftest.py diff --git a/HACKING.md b/HACKING.md index 75aa6d0..bb22fd6 100644 --- a/HACKING.md +++ b/HACKING.md @@ -35,7 +35,7 @@ Running from git repository make check ``` - The test tool requires python3 and python3-dbus. + The test tool requires python3, python3-pytest and python3-dbus. * To run libvirt-dbus directly from the build dir without installing it diff --git a/test/Makefile.am b/test/Makefile.am index d3997f3..acb2d33 100644 --- a/test/Makefile.am +++ b/test/Makefile.am @@ -1,17 +1,18 @@ test_helpers = \ - libvirttest.py + libvirttest.py \ + conftest.py test_programs = \ test_connect.py \ test_domain.py -TESTS = $(test_programs) - EXTRA_DIST = \ $(test_helpers) \ $(test_programs) \ travis-run +TESTS = $(test_programs) + TESTS_ENVIRONMENT = \ abs_top_builddir=$(abs_top_builddir) \ VIRT_DBUS_INTERFACES_DIR=$(abs_top_srcdir)/data diff --git a/test/conftest.py b/test/conftest.py new file mode 100644 index 0000000..3bce7b7 --- /dev/null +++ b/test/conftest.py @@ -0,0 +1,15 @@ +import os +import pytest +import subprocess + + +(a)pytest.fixture(scope="session", autouse=True) +def dbus_daemon_session(): + """Fixture to start D-Bus message bus session daemon for use in the whole test suite. + """ + dbus_daemon = subprocess.Popen(['dbus-daemon', '--session', '--print-address'], + stdout=subprocess.PIPE, universal_newlines=True) + os.environ['DBUS_SESSION_BUS_ADDRESS'] = dbus_daemon.stdout.readline().strip() + yield + dbus_daemon.terminate() + dbus_daemon.wait(timeout=10) diff --git a/test/libvirttest.py b/test/libvirttest.py index 6a00aea..23fe536 100644 --- a/test/libvirttest.py +++ b/test/libvirttest.py @@ -1,34 +1,37 @@ -#!/usr/bin/python3 - from dbus.mainloop.glib import DBusGMainLoop from gi.repository import GLib import dbus import os +import pytest import subprocess +import sys import time -import unittest + root = os.environ.get('abs_top_builddir', os.path.dirname(os.path.dirname(__file__))) exe = os.path.join(root, 'src', 'libvirt-dbus') DBusGMainLoop(set_as_default=True) -class TestCase(unittest.TestCase): - @classmethod - def setUpClass(cls): - cls.bus = subprocess.Popen(['dbus-daemon', '--session', '--print-address'], - stdout=subprocess.PIPE, universal_newlines=True) - os.environ['DBUS_SESSION_BUS_ADDRESS'] = cls.bus.stdout.readline().strip() - @classmethod - def tearDownClass(cls): - cls.bus.terminate() - cls.bus.wait(timeout=10) +def run(): + exit(pytest.main(sys.argv)) - def setUp(self): - os.environ['LIBVIRT_DEBUG'] = '3' - self.daemon = subprocess.Popen([exe]) +class BaseTestClass(): + """ Base test class for whole test suite + """ + connect = None + bus = None + libvirt_dbus = None + loop = False + + @pytest.fixture(autouse=True) + def libvirt_dbus_setup(self, request): + """Start libvirt-dbus for each test function + """ + os.environ['LIBVIRT_DEBUG'] = '3' + self.libvirt_dbus = subprocess.Popen([exe]) self.bus = dbus.SessionBus() for i in range(10): @@ -41,12 +44,18 @@ class TestCase(unittest.TestCase): obj = self.bus.get_object('org.libvirt', '/org/libvirt/Test') self.connect = dbus.Interface(obj, 'org.libvirt.Connect') - def tearDown(self): - self.daemon.terminate() - self.daemon.wait(timeout=10) + @pytest.fixture(autouse=True) + def libvirt_dbus_teardown(self): + """Terminate libvirt-dbus at the teardown of each test + """ + yield + self.libvirt_dbus.terminate() + self.libvirt_dbus.wait(timeout=10) def main_loop(self): - self.assertFalse(getattr(self, 'loop', False)) + """Initializes the mainloop + """ + assert getattr(self, 'loop', False) is False def timeout(): self.loop.quit() diff --git a/test/test_connect.py b/test/test_connect.py index 4ec3fe0..a52140c 100755 --- a/test/test_connect.py +++ b/test/test_connect.py @@ -2,9 +2,10 @@ import dbus import libvirttest -import unittest -minimal_xml = ''' + +class TestConnect(libvirttest.BaseTestClass): + minimal_xml = ''' <domain type="test"> <name>foo</name> <memory>1024</memory> @@ -12,16 +13,15 @@ minimal_xml = ''' <type>hvm</type> </os> </domain> -''' + ''' -class TestConnect(libvirttest.TestCase): def test_list_domains(self): domains = self.connect.ListDomains(0) - self.assertEqual(type(domains), dbus.Array) - self.assertEqual(len(domains), 1) + assert isinstance(domains, dbus.Array) + assert len(domains) == 1 for path in domains: - self.assertEqual(type(path), dbus.ObjectPath) + assert isinstance(path, dbus.ObjectPath) domain = self.bus.get_object('org.libvirt', path) # ensure the path exists by calling Introspect on it @@ -29,29 +29,30 @@ class TestConnect(libvirttest.TestCase): def test_create(self): def domain_started(name, path): - self.assertEqual(name, 'foo') - self.assertEqual(type(path), dbus.ObjectPath) + assert name == 'foo' + assert isinstance(path, dbus.ObjectPath) self.loop.quit() self.connect.connect_to_signal('DomainStarted', domain_started) - path = self.connect.CreateXML(minimal_xml, 0) - self.assertEqual(type(path), dbus.ObjectPath) + path = self.connect.CreateXML(self.minimal_xml, 0) + assert isinstance(path, dbus.ObjectPath) self.main_loop() def test_define(self): def domain_defined(name, path): - self.assertEqual(name, 'foo') - self.assertEqual(type(path), dbus.ObjectPath) + assert name == 'foo' + assert isinstance(path, dbus.ObjectPath) self.loop.quit() self.connect.connect_to_signal('DomainDefined', domain_defined) - path = self.connect.DefineXML(minimal_xml) - self.assertEqual(type(path), dbus.ObjectPath) + path = self.connect.DefineXML(self.minimal_xml) + assert isinstance(path, dbus.ObjectPath) self.main_loop() + if __name__ == '__main__': - unittest.main(verbosity=2) + libvirttest.run() diff --git a/test/test_domain.py b/test/test_domain.py index b1ab7a5..1bf9c1b 100755 --- a/test/test_domain.py +++ b/test/test_domain.py @@ -2,9 +2,9 @@ import dbus import libvirttest -import unittest -class TestDomain(libvirttest.TestCase): + +class TestDomain(libvirttest.BaseTestClass): def domain(self): path = self.connect.ListDomains(0)[0] obj = self.bus.get_object('org.libvirt', path) @@ -14,21 +14,21 @@ class TestDomain(libvirttest.TestCase): obj, domain = self.domain() props = obj.GetAll('org.libvirt.Domain', dbus_interface=dbus.PROPERTIES_IFACE) - self.assertEqual(type(props['Name']), dbus.String) - self.assertEqual(type(props['UUID']), dbus.String) - self.assertEqual(type(props['Id']), dbus.UInt32) - self.assertEqual(type(props['Vcpus']), dbus.UInt32) - self.assertEqual(type(props['OSType']), dbus.String) - self.assertEqual(type(props['Active']), dbus.Boolean) - self.assertEqual(type(props['Persistent']), dbus.Boolean) - self.assertEqual(type(props['State']), dbus.String) - self.assertEqual(type(props['Autostart']), dbus.Boolean) + assert isinstance(props['Name'], dbus.String) + assert isinstance(props['UUID'], dbus.String) + assert isinstance(props['Id'], dbus.UInt32) + assert isinstance(props['Vcpus'], dbus.UInt32) + assert isinstance(props['OSType'], dbus.String) + assert isinstance(props['Active'], dbus.Boolean) + assert isinstance(props['Persistent'], dbus.Boolean) + assert isinstance(props['State'], dbus.String) + assert isinstance(props['Autostart'], dbus.Boolean) # Call all methods except Reset and GetStats, because the test backend # doesn't support those xml = domain.GetXMLDesc(0) - self.assertEqual(type(xml), dbus.String) + assert isinstance(xml, dbus.String) domain.Reboot(0) domain.Shutdown() @@ -38,8 +38,8 @@ class TestDomain(libvirttest.TestCase): def test_shutdown(self): def domain_stopped(name, path): - self.assertEqual(name, 'test') - self.assertEqual(type(path), dbus.ObjectPath) + assert name == 'test' + assert isinstance(path, dbus.ObjectPath) self.loop.quit() self.connect.connect_to_signal('DomainStopped', domain_stopped) @@ -48,14 +48,14 @@ class TestDomain(libvirttest.TestCase): domain.Shutdown() state = obj.Get('org.libvirt.Domain', 'State', dbus_interface=dbus.PROPERTIES_IFACE) - self.assertEqual(state, 'shutoff') + assert state == 'shutoff' self.main_loop() def test_undefine(self): def domain_undefined(name, path): - self.assertEqual(name, 'test') - self.assertEqual(type(path), dbus.ObjectPath) + assert name == 'test' + assert isinstance(path, dbus.ObjectPath) self.loop.quit() self.connect.connect_to_signal('DomainUndefined', domain_undefined) @@ -66,5 +66,6 @@ class TestDomain(libvirttest.TestCase): self.main_loop() + if __name__ == '__main__': - unittest.main(verbosity=2) + libvirttest.run() diff --git a/test/travis-run b/test/travis-run index 7577253..80b6dec 100755 --- a/test/travis-run +++ b/test/travis-run @@ -22,7 +22,7 @@ sudo chroot "$CHROOT" << EOF set -ex # install build deps apt-get update -apt-get install -y dh-autoreconf pkg-config libvirt-dev libglib2.0-dev libtool python3-gi python3-dbus dbus +apt-get install -y dh-autoreconf pkg-config libvirt-dev libglib2.0-dev libtool python3-gi python3-dbus python3-pytest dbus # run build and tests as user chown -R buildd:buildd /build -- 2.15.0

3 2

[libvirt] [PATCH v2 0/2] introduce libvirt-dbus binding
by Pavel Hrdina 23 Mar '18

23 Mar '18

Pavel Hrdina (2): docs: fix a typo in docs.html page docs: introduce libvirt-dbus binding docs/bindings.html.in | 13 ++++++- docs/dbus.html.in | 102 +++++++++++++++++++++++++++++++++++++++++++++++++ docs/docs.html.in | 8 ++-- docs/downloads.html.in | 15 ++++++++ 4 files changed, 134 insertions(+), 4 deletions(-) create mode 100644 docs/dbus.html.in -- 2.14.3

2 4

[libvirt] [PATCH RFC] storage: Remove 'cow' disk format
by Peter Krempa 23 Mar '18

23 Mar '18

There's no tests for this format and nothing seems to specifically care about this format. QEMU does not even recognize it. Remove it completely. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- docs/schemas/storagecommon.rng | 1 - src/util/virstoragefile.c | 31 +------------------------------ src/util/virstoragefile.h | 3 +-- 3 files changed, 2 insertions(+), 33 deletions(-) diff --git a/docs/schemas/storagecommon.rng b/docs/schemas/storagecommon.rng index edee1b0845..310ed758ff 100644 --- a/docs/schemas/storagecommon.rng +++ b/docs/schemas/storagecommon.rng @@ -78,7 +78,6 @@ we know how to follow backing chains, and all others --> <define name='storageFormatBacking'> <choice> - <value>cow</value> <value>qcow</value> <value>qcow2</value> <value>qed</value> diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index 67b9ec71ac..dd07f09d2c 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -67,7 +67,7 @@ VIR_ENUM_IMPL(virStorageFileFormat, /* Not direct file formats, but used for various drivers */ "fat", "vhd", "ploop", /* Formats with backing file below here */ - "cow", "qcow", "qcow2", "qed", "vmdk") + "qcow", "qcow2", "qed", "vmdk") VIR_ENUM_IMPL(virStorageFileFeature, VIR_STORAGE_FILE_FEATURE_LAST, @@ -170,8 +170,6 @@ struct FileTypeInfo { }; -static int cowGetBackingStore(char **, int *, - const char *, size_t); static int qcow1GetBackingStore(char **, int *, const char *, size_t); static int qcow2GetBackingStore(char **, int *, @@ -348,11 +346,6 @@ static struct FileTypeInfo const fileTypeInfo[] = { PLOOP_SIZE_MULTIPLIER, NULL, NULL, NULL }, /* All formats with a backing store probe below here */ - [VIR_STORAGE_FILE_COW] = { - 0, "OOOM", NULL, - LV_BIG_ENDIAN, 4, 4, {2}, - 4+4+1024+4, 8, 1, NULL, cowGetBackingStore, NULL - }, [VIR_STORAGE_FILE_QCOW] = { 0, "QFI", NULL, LV_BIG_ENDIAN, 4, 4, {1}, @@ -397,28 +390,6 @@ static const int qcow2CompatibleFeatureArray[] = { verify(ARRAY_CARDINALITY(qcow2CompatibleFeatureArray) == QCOW2_COMPATIBLE_FEATURE_LAST); -static int -cowGetBackingStore(char **res, - int *format, - const char *buf, - size_t buf_size) -{ -#define COW_FILENAME_MAXLEN 1024 - *res = NULL; - *format = VIR_STORAGE_FILE_AUTO; - - if (buf_size < 4+4+ COW_FILENAME_MAXLEN) - return BACKING_STORE_INVALID; - if (buf[4+4] == '\0') { /* cow_header_v2.backing_file[0] */ - *format = VIR_STORAGE_FILE_NONE; - return BACKING_STORE_OK; - } - - if (VIR_STRNDUP(*res, (const char*)buf + 4 + 4, COW_FILENAME_MAXLEN) < 0) - return BACKING_STORE_ERROR; - return BACKING_STORE_OK; -} - static int qcow2GetBackingStoreFormat(int *format, diff --git a/src/util/virstoragefile.h b/src/util/virstoragefile.h index 596746ccb7..c84e013f46 100644 --- a/src/util/virstoragefile.h +++ b/src/util/virstoragefile.h @@ -82,8 +82,7 @@ typedef enum { * libvirt support for following a backing chain */ VIR_STORAGE_FILE_BACKING, - VIR_STORAGE_FILE_COW = VIR_STORAGE_FILE_BACKING, - VIR_STORAGE_FILE_QCOW, + VIR_STORAGE_FILE_QCOW = VIR_STORAGE_FILE_BACKING, VIR_STORAGE_FILE_QCOW2, VIR_STORAGE_FILE_QED, VIR_STORAGE_FILE_VMDK, -- 2.16.2

2 2

[libvirt] [PATCH] docs: introduce libvirt-dbus binding
by Pavel Hrdina 23 Mar '18

23 Mar '18

libvirt-dbus is a new binding that wraps libvirt API into D-Bus calls. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- docs/bindings.html.in | 4 ++ docs/dbus.html.in | 102 +++++++++++++++++++++++++++++++++++++++++++++++++ docs/docs.html.in | 1 + docs/downloads.html.in | 15 ++++++++ 4 files changed, 122 insertions(+) create mode 100644 docs/dbus.html.in diff --git a/docs/bindings.html.in b/docs/bindings.html.in index 733c89bc20..236bdcf2ae 100644 --- a/docs/bindings.html.in +++ b/docs/bindings.html.in @@ -18,6 +18,10 @@ <strong>Go</strong>: Daniel Berrange develops <a href="https://godoc.org/github.com/libvirt/libvirt-go">Go bindings</a>. </li> + <li> + <strong>D-Bus</strong>: Pavel Hrdina develops + <a href="dbus.html">D-Bus bindings</a>. + </li> <li> <strong>Java</strong>: Daniel Veillard develops <a href="java.html">Java bindings</a>. diff --git a/docs/dbus.html.in b/docs/dbus.html.in new file mode 100644 index 0000000000..0da8096122 --- /dev/null +++ b/docs/dbus.html.in @@ -0,0 +1,102 @@ +<?xml version="1.0" encoding="UTF-8"?> +<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> +<html xmlns="http://www.w3.org/1999/xhtml"> + <body> + <h1>D-Bus API bindings</h1> + + <ul id="toc"></ul> + + <h2><a name="description">Description</a></h2> + + <p> + libvirt-dbus wraps libvirt API to provide a high-level object-oriented + API better suited for dbus-based applications. + </p> + + <h2><a name="git">GIT source repository</a></h2> + <p> + The D-Bus bindings source code is maintained in a + <a href="https://git-scm.com/">git</a> repository available on + <a href="https://libvirt.org/git/">libvirt.org</a>: + </p> + +<pre> +git clone https://libvirt.org/git/libvirt-dbus.git +</pre> + + <p> + They can also be browsed online: + </p> + +<pre> +<a href="https://libvirt.org/git/?p=libvirt-dbus.git">https://libvirt.org/git/?p=libvirt-dbus.git</a> +</pre> + + <h2><a name="usage">Usage</a></h2> + + <p> + libvirt-dbus exports libvirt API using D-Bus objects with methods and + properties described by interfaces. Currently only local connection + to libvirt is exported and the list of supported drivers depends + on the type of the bus connection (session or system). + </p> + + <p> + The name of the libvirt-dbus service is <code>org.libvirt</code>. + libvirt-dbus distributes an interface XML descriptions which can be + usually found at <code>/usr/share/dbus-1/interfaces/</code>. + </p> + + <p> + By default unprivileged user has access only to the session D-Bus + connection. In order to allow specific user "foo" to access the system + D-Bus connection you need to create a file + <code>/etc/dbus-1/system.d/org.libvirt.conf</code> that contains: + </p> + +<pre> +<?xml version="1.0"?> +<!DOCTYPE busconfig PUBLIC "-//freedesktop//DTD D-BUS Bus Configuration 1.0//EN" + "http://www.freedesktop.org/standards/dbus/1.0/busconfig.dtd"> + +<busconfig> + + <policy user="foo"> + <allow send_destination="org.libvirt"/> + </policy> + +</busconfig> +</pre> + + <p> + To get a list of supported drivers for the specific bus connection + you can run these commands (not all drivers may be available on + the host): + </p> + +<pre> +gdbus introspect --xml --session --dest org.libvirt --object-path /org/libvirt +gdbus introspect --xml --system --dest org.libvirt --object-path /org/libvirt +</pre> + + <p> + Every object is introspectable so you can get a list of available + interfaces with methods, signals and properties running this command: + </p> + +<pre> +gdbus introspect --xml --system --dest org.libvirt --object-path /org/libvirt/QEMU +</pre> + + <p> + To get a list of domains for specific connection driver you can run + this command: + </p> + +<pre> +gdbus call --system --dest org.libvirt --object-path /org/libvirt/QEMU \ + --method org.libvirt.Connect.ListDomains 0 +</pre> + + </body> +</html> diff --git a/docs/docs.html.in b/docs/docs.html.in index a8d544f83f..abd344ca2e 100644 --- a/docs/docs.html.in +++ b/docs/docs.html.in @@ -57,6 +57,7 @@ <dt><a href="bindings.html">Language bindings</a></dt> <dd>Bindings of the libvirt API for <a href="csharp.html">c#</a>, + <a href="dbus.html">D-Bus</a>, <a href="https://godoc.org/github.com/libvirt/libvirt-go">go</a>, <a href="java.html">java</a>, <a href="https://libvirt.org/ocaml/">ocaml</a>. diff --git a/docs/downloads.html.in b/docs/downloads.html.in index 7c28e2ba9b..c2bc7af3e9 100644 --- a/docs/downloads.html.in +++ b/docs/downloads.html.in @@ -60,6 +60,21 @@ </td> <td></td> </tr> + <tr> + <td>D-Bus</td> + <td> + <a href="ftp://libvirt.org/libvirt/dbus/">ftp</a> + <a href="https://libvirt.org/sources/dbus/">https</a> + </td> + <td> + <a href="https://libvirt.org/git/?p=libvirt-dbus.git;a=summary">libvirt</a> + </td> + <td> + <a href="https://gitlab.com/libvirt/libvirt-dbus">gitlab</a> + <a href="https://github.com/libvirt/libvirt-dbus">github</a> + </td> + <td></td> + </tr> <tr> <td>Go</td> <td> -- 2.14.3

3 6

[libvirt] [PATCH 0/1] Add KVM on Z to Planet virttools
by Stefan Raspl 23 Mar '18

23 Mar '18

Stefan Raspl (1): Add KVM on Z to Planet virttools updater/virt-tools/config.ini | 3 +++ 1 file changed, 3 insertions(+) -- 2.13.5

3 3

[libvirt] ANNOUNCE: libvirt-dbus 0.0.1 released
by Pavel Hrdina 23 Mar '18

23 Mar '18

I'm happy to announce the release of libvirt-dbus 0.0.1. libvirt-dbus wraps libvirt API to provide high-level object-oriented API better suited for dbus-based applications. This is a first release which should be considered as unstable release since we don't promise API stability. You can download it here: https://libvirt.org/sources/dbus/libvirt-dbus-0.0.1.tar.gz This release contains a limited support of libvirt APIs as proof of concept, but we are planning to cover all relevant APIs. Thanks everybody who contributed! Pavel

1 0

[libvirt] [PATCH v4 0/5] Apparmor support for less common devices
by Christian Ehrhardt 23 Mar '18

23 Mar '18

So far users added manual rules for most of these uncommon devices, but recent changes made some of the callbacks mandatory for hotplug so we should take shot at implementing them as those callbacks as well as for the initial start of a guest via virt-aa-helper. *Updates since v1* - Set(Memory|Input)Label: remove seclabel check already done in reload_profile - virt-aa-helper: check pointers before accessing them - add tests for new virt-aa-helper supported xml elements - extend tests to check for expected content (new patch in series) *Updates since v2* - Restore(Memory|Input)Label: drop secdef/relabel check - Set(Memory|Input)Label: check more pointers to be valid before using them *Updates since v3* - added the Acked-by of Jamie Strandboge on patches 1-4 - reuse the already existing tmpdir in virt-aa-helper-test for better cleanup Christian Ehrhardt (5): security, apparmor: add (Set|Restore)MemoryLabel security, apparmor: add (Set|Restore)InputLabel virt-aa-helper: generate rules for passthrough input devices virt-aa-helper: generate rules for nvdimm memory virt-aa-helper: test: check for expected profile content src/security/security_apparmor.c | 94 ++++++++++++++++++++++++++++++++++++++++ src/security/virt-aa-helper.c | 16 +++++++ tests/virt-aa-helper-test | 87 ++++++++++++++++++++++--------------- 3 files changed, 163 insertions(+), 34 deletions(-) -- 2.7.4

2 8

[libvirt] [PATCH] tools: fix variable scope in in check_guests_shutdown
by Christian Ehrhardt 23 Mar '18

23 Mar '18

libvirt-guests.sh when run with more active guests than requested to shut down in parallel will run until it times out only shutting down the first set of guests. This patch fixes parallel shutdown by fixing a variable scope issue where check_guests_shutdown unintentionally reset $guests which prevented further progress. Fixes: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1688508 Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com> --- tools/libvirt-guests.sh.in | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/libvirt-guests.sh.in b/tools/libvirt-guests.sh.in index d5e68e5..fcada31 100644 --- a/tools/libvirt-guests.sh.in +++ b/tools/libvirt-guests.sh.in @@ -333,11 +333,11 @@ guest_count() check_guests_shutdown() { uri=$1 - guests=$2 + guests_to_check=$2 guests_shutting_down= for guest in $guests; do - if ! guest_is_on "$uri" "$guest" >/dev/null 2>&1; then + if ! guest_is_on "$uri" "$guests_to_check" >/dev/null 2>&1; then eval_gettext "Failed to determine state of guest: \$guest. Not tracking it anymore." echo continue -- 2.7.4

3 3

[libvirt] [PATCH v2] test_driver: Add testDomainDestroyFlags
by Katerina Koukiou 22 Mar '18

22 Mar '18

Adding this for completeness Signed-off-by: Katerina Koukiou <kkoukiou(a)redhat.com> --- Needed by the libvirt-dbus which is using test_driver for tests. src/test/test_driver.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 4de0cc533..99c27cc0a 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -1782,13 +1782,16 @@ static int testConnectListDomains(virConnectPtr conn, NULL, NULL); } -static int testDomainDestroy(virDomainPtr domain) +static int testDomainDestroyFlags(virDomainPtr domain, + unsigned int flags) { testDriverPtr privconn = domain->conn->privateData; virDomainObjPtr privdom; virObjectEventPtr event = NULL; int ret = -1; + virCheckFlags(VIR_DOMAIN_DESTROY_GRACEFUL, -1); + if (!(privdom = testDomObjFromDomain(domain))) goto cleanup; @@ -1813,6 +1816,11 @@ static int testDomainDestroy(virDomainPtr domain) return ret; } +static int testDomainDestroy(virDomainPtr domain) +{ + return testDomainDestroyFlags(domain, 0); +} + static int testDomainResume(virDomainPtr domain) { testDriverPtr privconn = domain->conn->privateData; @@ -6872,6 +6880,7 @@ static virHypervisorDriver testHypervisorDriver = { .domainShutdownFlags = testDomainShutdownFlags, /* 0.9.10 */ .domainReboot = testDomainReboot, /* 0.1.1 */ .domainDestroy = testDomainDestroy, /* 0.1.1 */ + .domainDestroyFlags = testDomainDestroyFlags, /* 4.2.0 */ .domainGetOSType = testDomainGetOSType, /* 0.1.9 */ .domainGetMaxMemory = testDomainGetMaxMemory, /* 0.1.4 */ .domainSetMaxMemory = testDomainSetMaxMemory, /* 0.1.1 */ -- 2.15.0

2 1

[libvirt] [PATCH 00/11] tests: Add QAPI schema validation (blockdev-add saga)
by Peter Krempa 22 Mar '18

22 Mar '18

The new test APIs allow to validate JSON objects against the QAPI schema. The infrastructure was developed as part of not-yet-posted blockdev JSON property generator. It is quite useful for developing generators for complex objects against the schema. Since it's kind of useful by itself I decided to post it prior to finishing the blockdev work. It at least makes the 'simple' test cases in the qemumonitorjsontest slightly useful. Peter Krempa (11): util: buffer: Tolerate NULL 'buf' in virBufferStrcat qemu: caps: Move QAPI schema related code into separate file qemu: qapi: Fix naming of moved functions qemu: monitor: Move processing of QMP schema to the new file util: json: Add accessor for looking up JSON value type qemu: qapi: Return correct entry in virQEMUQapiSchemaTraverse tests: Add data file with QEMU QAPI schema tests: qemu: Add infrastructure for QAPI schema testing tests: qemumonitorjson: Fix few arguments of test cases tests: qemumonitor: Allow testing schema for fake monitor interactions tests: qemumonitorjson: Do some useful testing in the 'simple' tests po/POTFILES.in | 1 + src/libvirt_private.syms | 1 + src/qemu/Makefile.inc.am | 2 + src/qemu/qemu_capabilities.c | 171 +- src/qemu/qemu_monitor.c | 2 +- src/qemu/qemu_monitor.h | 2 +- src/qemu/qemu_monitor_json.c | 41 +- src/qemu/qemu_monitor_json.h | 2 +- src/qemu/qemu_qapi.c | 237 + src/qemu/qemu_qapi.h | 40 + src/util/virbuffer.c | 3 + src/util/virjson.c | 7 + src/util/virjson.h | 2 + tests/Makefile.am | 3 + tests/qemuhotplugtest.c | 3 +- tests/qemumonitorjsontest.c | 117 +- tests/qemumonitortestutils.c | 85 +- tests/qemumonitortestutils.h | 7 +- tests/qemuqapischema.json | 11364 +++++++++++++++++++++++++++++++++++++++++ tests/testutilsqemuschema.c | 536 ++ tests/testutilsqemuschema.h | 30 + 21 files changed, 12446 insertions(+), 210 deletions(-) create mode 100644 src/qemu/qemu_qapi.c create mode 100644 src/qemu/qemu_qapi.h create mode 100644 tests/qemuqapischema.json create mode 100644 tests/testutilsqemuschema.c create mode 100644 tests/testutilsqemuschema.h -- 2.16.2

2 22