- Devel - libvirt List Archives

[libvirt] [RFC] qemu: monitor: do not report error on shutdown
by Christian Ehrhardt 16 May '17

16 May '17

If a shutdown is expected because it was triggered via libvirt we can also expect the monitor to close. In those cases do not report an internal error like: "internal error: End of file from qemu monitor" Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com> --- src/qemu/qemu_monitor.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index d71f84c..352cc8a 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -63,6 +63,7 @@ struct _qemuMonitor { int fd; int watch; int hasSendFD; + int willhangup; virDomainObjPtr vm; @@ -691,8 +692,10 @@ qemuMonitorIO(int watch, int fd, int events, void *opaque) if (events & VIR_EVENT_HANDLE_HANGUP) { hangup = true; if (!error) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("End of file from qemu monitor")); + if (!mon->willhangup) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("End of file from qemu monitor")); + } eof = true; events &= ~VIR_EVENT_HANDLE_HANGUP; } @@ -731,7 +734,7 @@ qemuMonitorIO(int watch, int fd, int events, void *opaque) if (mon->lastError.code != VIR_ERR_OK) { /* Already have an error, so clear any new error */ virResetLastError(); - } else { + } else if (!mon->willhangup) { virErrorPtr err = virGetLastError(); if (!err) virReportError(VIR_ERR_INTERNAL_ERROR, "%s", @@ -1282,6 +1285,7 @@ qemuMonitorEmitShutdown(qemuMonitorPtr mon) { int ret = -1; VIR_DEBUG("mon=%p", mon); + mon->willhangup = 1; QEMU_MONITOR_CALLBACK(mon, ret, domainShutdown, mon->vm); return ret; -- 2.7.4

2 2

[libvirt] [libvirt-python PATCH v2] spec: Install egg-info with rpm package
by Martin Kletzander 15 May '17

15 May '17

This was being done due to now deprecated policy and that file should be installed so that pip can recognize that the packages is already installed in the system. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- v2: - Put each egg-info ito its respective RPM package libvirt-python.spec.in | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libvirt-python.spec.in b/libvirt-python.spec.in index 5ad029281c52..fc30564133fe 100644 --- a/libvirt-python.spec.in +++ b/libvirt-python.spec.in @@ -65,7 +65,6 @@ CFLAGS="$RPM_OPT_FLAGS" %{__python3} setup.py build %if %{with_python3} %{__python3} setup.py install --skip-build --root=%{buildroot} %endif -rm -f %{buildroot}%{_libdir}/python*/site-packages/*egg-info %check %{__python} setup.py test @@ -80,6 +79,7 @@ rm -f %{buildroot}%{_libdir}/python*/site-packages/*egg-info %{_libdir}/python2*/site-packages/libvirt_qemu.py* %{_libdir}/python2*/site-packages/libvirt_lxc.py* %{_libdir}/python2*/site-packages/libvirtmod* +%{_libdir}/python2*/site-packages/*egg-info %if %{with_python3} %files -n libvirt-python3 @@ -94,6 +94,7 @@ rm -f %{buildroot}%{_libdir}/python*/site-packages/*egg-info %{_libdir}/python3*/site-packages/__pycache__/libvirt_lxc.cpython-*.py* %{_libdir}/python3*/site-packages/__pycache__/libvirtaio.cpython-*.py* %{_libdir}/python3*/site-packages/libvirtmod* +%{_libdir}/python3*/site-packages/*egg-info %endif %changelog -- 2.13.0

2 1

[libvirt] [PATCH] remove hack for debian etch limits.h
by Daniel P. Berrange 15 May '17

15 May '17

The debian etch distro was end-of-life a long time ago so we no longer need the ULLONG_MAX hack. In any case gnulib now provides an equivalent fix by default, and so our definition now triggers syntax-check rule failure src/internal.h:# define ULLONG_MAX ULONG_LONG_MAX maint.mk: define the above via some gnulib .h file maint.mk:843: recipe for target 'sc_prohibit_always-defined_macros' failed Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed as a build break fix src/internal.h | 5 ----- 1 file changed, 5 deletions(-) diff --git a/src/internal.h b/src/internal.h index 713734c..5a5a430 100644 --- a/src/internal.h +++ b/src/internal.h @@ -114,11 +114,6 @@ # define __GNUC_PREREQ(maj, min) 0 # endif -/* Work around broken limits.h on debian etch */ -# if defined _GCC_LIMITS_H_ && ! defined ULLONG_MAX -# define ULLONG_MAX ULONG_LONG_MAX -# endif - # endif /* __GNUC__ */ /** -- 2.9.3

1 0

[libvirt] [libvirt-python PATCH] spec: Install egg-info with rpm package
by Martin Kletzander 15 May '17

15 May '17

This was being done due to now deprecated policy and that file should be installed so that pip can recognize that the packages is already installed in the system. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- libvirt-python.spec.in | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libvirt-python.spec.in b/libvirt-python.spec.in index 5ad029281c52..132183c93c02 100644 --- a/libvirt-python.spec.in +++ b/libvirt-python.spec.in @@ -65,7 +65,6 @@ CFLAGS="$RPM_OPT_FLAGS" %{__python3} setup.py build %if %{with_python3} %{__python3} setup.py install --skip-build --root=%{buildroot} %endif -rm -f %{buildroot}%{_libdir}/python*/site-packages/*egg-info %check %{__python} setup.py test @@ -81,6 +80,8 @@ rm -f %{buildroot}%{_libdir}/python*/site-packages/*egg-info %{_libdir}/python2*/site-packages/libvirt_lxc.py* %{_libdir}/python2*/site-packages/libvirtmod* +%{_libdir}/python*/site-packages/*egg-info + %if %{with_python3} %files -n libvirt-python3 %defattr(-,root,root) -- 2.13.0

2 2

[libvirt] [PATCH] maint: update to latest gnulib
by Daniel P. Berrange 15 May '17

15 May '17

This pulls in the fixes for poll() on Win32 which finally makes the remote driver work again. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- .gnulib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index 94386a1..da830b5 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 94386a13667c645fd42544a7fd302cdddd39fcdf +Subproject commit da830b5146cb553ac2a4bcfe76caeb57bda24cc3 -- 2.9.3

2 1

[libvirt] [PATCH] vz: support virDomainGetBlockInfo in driver
by Nikolay Shirokovskiy 15 May '17

15 May '17

Actually physical size is not available in vz sdk right now so let's set it to allocation as an estimation in non sparse case. --- src/vz/vz_driver.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/vz/vz_sdk.c | 23 +++++++++++++++++++++++ src/vz/vz_sdk.h | 1 + 3 files changed, 74 insertions(+) diff --git a/src/vz/vz_driver.c b/src/vz/vz_driver.c index 9a429f4..7b32558 100644 --- a/src/vz/vz_driver.c +++ b/src/vz/vz_driver.c @@ -4005,6 +4005,55 @@ vzDomainBlockResize(virDomainPtr domain, return ret; } +static int +vzDomainGetBlockInfo(virDomainPtr domain, + const char *path, + virDomainBlockInfoPtr info, + unsigned int flags) +{ + virDomainObjPtr dom; + virDomainDiskDefPtr disk; + long long allocation; + bool job = false; + int ret = -1; + + virCheckFlags(0, -1); + + if (!(dom = vzDomObjFromDomainRef(domain))) + return -1; + + if (virDomainGetBlockInfoEnsureACL(domain->conn, dom->def) < 0) + goto cleanup; + + if (vzDomainObjBeginJob(dom) < 0) + goto cleanup; + job = true; + + if (vzEnsureDomainExists(dom) < 0) + goto cleanup; + + if (!(disk = virDomainDiskByName(dom->def, path, false))) { + virReportError(VIR_ERR_INVALID_ARG, + _("invalid path %s not assigned to domain"), path); + goto cleanup; + } + + if ((allocation = prlsdkGetDiskAllocation(dom, disk)) < 0) + goto cleanup; + + info->capacity = disk->src->capacity; + info->allocation = allocation; + info->physical = allocation; + + ret = 0; + + cleanup: + if (job) + vzDomainObjEndJob(dom); + virDomainObjEndAPI(&dom); + return ret; +} + static virHypervisorDriver vzHypervisorDriver = { .name = "vz", .connectOpen = vzConnectOpen, /* 0.10.0 */ @@ -4106,6 +4155,7 @@ static virHypervisorDriver vzHypervisorDriver = { .domainAbortJob = vzDomainAbortJob, /* 3.1.0 */ .domainReset = vzDomainReset, /* 3.1.0 */ .domainBlockResize = vzDomainBlockResize, /* 3.3.0 */ + .domainGetBlockInfo = vzDomainGetBlockInfo, /* 3.4.0 */ }; static virConnectDriver vzConnectDriver = { diff --git a/src/vz/vz_sdk.c b/src/vz/vz_sdk.c index 4d2c6b0..b8f561c 100644 --- a/src/vz/vz_sdk.c +++ b/src/vz/vz_sdk.c @@ -4962,3 +4962,26 @@ int prlsdkResizeImage(virDomainObjPtr dom, virDomainDiskDefPtr disk, PrlHandle_Free(prldisk); return ret; } + +long long prlsdkGetDiskAllocation(virDomainObjPtr dom, + virDomainDiskDefPtr disk) +{ + vzDomObjPtr privdom = dom->privateData; + PRL_HANDLE job; + PRL_HANDLE prldisk; + PRL_UINT32 size; + PRL_RESULT pret; + + job = PrlVm_RefreshConfig(privdom->sdkdom); + if (waitDomainJob(job, dom)) + return -1; + + prldisk = prlsdkGetDisk(privdom->sdkdom, disk); + if (prldisk == PRL_INVALID_HANDLE) + return -1; + + pret = PrlVmDevHd_GetSizeOnDisk(prldisk, &size); + prlsdkCheckRetExit(pret, -1); + + return ((unsigned long long)size) << 20; +} diff --git a/src/vz/vz_sdk.h b/src/vz/vz_sdk.h index 0a77431..6b73067 100644 --- a/src/vz/vz_sdk.h +++ b/src/vz/vz_sdk.h @@ -91,3 +91,4 @@ PRL_HANDLE prlsdkSdkDomainLookupByName(vzDriverPtr driver, const char *name); int prlsdkCancelJob(virDomainObjPtr dom); int prlsdkResizeImage(virDomainObjPtr dom, virDomainDiskDefPtr disk, unsigned long long newsize); +long long prlsdkGetDiskAllocation(virDomainObjPtr dom, virDomainDiskDefPtr disk); -- 1.8.3.1

1 0

[libvirt] [PATCH 0/5] qemu: Unbreak aarch64/virt TCG guests
by Andrea Bolognani 15 May '17

15 May '17

This issue was reported on the list a while ago, and now bug reports are popping up in Bugzilla as well. Since the proper fix (implementing the MSI controller for GICv3 in QEMU) has been confirmed to be a long way off, let's deal with it in libvirt. Andrea Bolognani (5): qemu: Use qemuDomainMachineIsVirt() more tests: Check default GIC version for aarch64/virt TCG guests qemu: Use GICv2 for aarch64/virt TCG guests gic: Remove VIR_GIC_VERSION_DEFAULT news: Update for GIC version on TCG changes docs/news.xml | 11 ++++++ src/qemu/qemu_capabilities.c | 7 +--- src/qemu/qemu_command.c | 6 +-- src/qemu/qemu_domain.c | 44 +++++++++++++++------- src/util/virgic.h | 3 -- .../qemuxml2argv-aarch64-gic-none-tcg.args | 19 ++++++++++ .../qemuxml2argv-aarch64-gic-none-tcg.xml | 17 +++++++++ tests/qemuxml2argvtest.c | 3 ++ .../qemuxml2xmlout-aarch64-gic-none-tcg.xml | 25 ++++++++++++ tests/qemuxml2xmltest.c | 1 + 10 files changed, 110 insertions(+), 26 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-gic-none-tcg.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-gic-none-tcg.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-aarch64-gic-none-tcg.xml -- 2.7.4

2 11

[libvirt] [PATCH] Add support for VNC autoport feature for bhyve hypervisor.
by Alexander Nusov 15 May '17

15 May '17

This patch adds support for automatic VNC port assignment for bhyve guests. --- src/bhyve/bhyve_command.c | 9 +++++++++ src/bhyve/bhyve_driver.c | 5 +++++ src/bhyve/bhyve_utils.h | 3 +++ 3 files changed, 17 insertions(+) diff --git a/src/bhyve/bhyve_command.c b/src/bhyve/bhyve_command.c index eae5cb3ca..4d35a05c5 100644 --- a/src/bhyve/bhyve_command.c +++ b/src/bhyve/bhyve_command.c @@ -339,6 +339,9 @@ bhyveBuildGraphicsArgStr(const virDomainDef *def ATTRIBUTE_UNUSED, virBuffer opt = VIR_BUFFER_INITIALIZER; virDomainGraphicsListenDefPtr glisten = NULL; bool escapeAddr; + unsigned short port; + + bhyveConnPtr driver = conn->privateData; if (!(bhyveDriverGetCaps(conn) & BHYVE_CAP_LPC_BOOTROM) || def->os.bootloader || @@ -401,6 +404,12 @@ bhyveBuildGraphicsArgStr(const virDomainDef *def ATTRIBUTE_UNUSED, virBufferAdd(&opt, glisten->address, -1); } + if (graphics->data.vnc.autoport) { + if (virPortAllocatorAcquire(driver->remotePorts, &port) < 0) + return -1; + graphics->data.vnc.port = port; + } + virBufferAsprintf(&opt, ":%d", graphics->data.vnc.port); break; default: diff --git a/src/bhyve/bhyve_driver.c b/src/bhyve/bhyve_driver.c index ed2221a35..bffeea7d9 100644 --- a/src/bhyve/bhyve_driver.c +++ b/src/bhyve/bhyve_driver.c @@ -52,6 +52,7 @@ #include "viraccessapicheck.h" #include "virhostcpu.h" #include "virhostmem.h" +#include "virportallocator.h" #include "conf/domain_capabilities.h" #include "bhyve_conf.h" @@ -1219,6 +1220,7 @@ bhyveStateCleanup(void) virObjectUnref(bhyve_driver->closeCallbacks); virObjectUnref(bhyve_driver->domainEventState); virObjectUnref(bhyve_driver->config); + virObjectUnref(bhyve_driver->remotePorts); virMutexDestroy(&bhyve_driver->lock); VIR_FREE(bhyve_driver); @@ -1265,6 +1267,9 @@ bhyveStateInitialize(bool privileged, if (!(bhyve_driver->domainEventState = virObjectEventStateNew())) goto cleanup; + if (!(bhyve_driver->remotePorts = virPortAllocatorNew(_("display"), 5900, 65535, 0))) + goto cleanup; + bhyve_driver->hostsysinfo = virSysinfoRead(); if (!(bhyve_driver->config = virBhyveDriverConfigNew())) diff --git a/src/bhyve/bhyve_utils.h b/src/bhyve/bhyve_utils.h index db50e012a..8ad2698d4 100644 --- a/src/bhyve/bhyve_utils.h +++ b/src/bhyve/bhyve_utils.h @@ -28,6 +28,7 @@ # include "virdomainobjlist.h" # include "virthread.h" # include "virclosecallbacks.h" +# include "virportallocator.h" # define BHYVE_AUTOSTART_DIR SYSCONFDIR "/libvirt/bhyve/autostart" # define BHYVE_CONFIG_DIR SYSCONFDIR "/libvirt/bhyve" @@ -58,6 +59,8 @@ struct _bhyveConn { virCloseCallbacksPtr closeCallbacks; + virPortAllocatorPtr remotePorts; + unsigned bhyvecaps; unsigned grubcaps; }; -- 2.12.1

2 3

[libvirt] [PATCHv4 0/6] Support more options for intel-iommu
by Ján Tomko 15 May '17

15 May '17

https://bugzilla.redhat.com/show_bug.cgi?id=1427005 v1: https://www.redhat.com/archives/libvir-list/2017-March/msg01072.html v2: https://www.redhat.com/archives/libvir-list/2017-April/msg00932.html v3: https://www.redhat.com/archives/libvir-list/2017-April/msg01248.html new in v4: * mention VT-d in the docs * rename irqchip mode to ioapic driver * rename caching to caching_mode * mention intremap's dependency on ioapic in the intermap section * add a separate versioned capability for kernel-irqchip=split Ján Tomko (6): conf: add <ioapic driver> to <features> qemu: format kernel_irqchip on the command line conf: add <driver intremap> to <iommu> qemu: format intel-iommu,intremap on the command line conf: add caching_mode attribute to iommu device qemu: format caching-mode on iommu command line docs/formatdomain.html.in | 41 +++++++++- docs/schemas/domaincommon.rng | 29 +++++++ src/conf/domain_conf.c | 89 ++++++++++++++++++++-- src/conf/domain_conf.h | 13 ++++ src/qemu/qemu_capabilities.c | 18 +++++ src/qemu/qemu_capabilities.h | 6 ++ src/qemu/qemu_command.c | 54 +++++++++++++ tests/qemucapabilitiesdata/caps_1.5.3.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_1.6.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_1.7.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.1.1.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_2.4.0.x86_64.replies | 22 ++++-- tests/qemucapabilitiesdata/caps_2.4.0.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_2.5.0.x86_64.replies | 24 ++++-- tests/qemucapabilitiesdata/caps_2.5.0.x86_64.xml | 1 + .../caps_2.6.0-gicv2.aarch64.xml | 2 + .../caps_2.6.0-gicv3.aarch64.xml | 2 + tests/qemucapabilitiesdata/caps_2.6.0.ppc64le.xml | 2 + .../qemucapabilitiesdata/caps_2.6.0.x86_64.replies | 24 ++++-- tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 2 + tests/qemucapabilitiesdata/caps_2.7.0.s390x.xml | 2 + .../qemucapabilitiesdata/caps_2.7.0.x86_64.replies | 28 +++++-- tests/qemucapabilitiesdata/caps_2.7.0.x86_64.xml | 3 + tests/qemucapabilitiesdata/caps_2.8.0.s390x.xml | 2 + .../qemucapabilitiesdata/caps_2.8.0.x86_64.replies | 37 +++++++-- tests/qemucapabilitiesdata/caps_2.8.0.x86_64.xml | 3 + .../qemucapabilitiesdata/caps_2.9.0.x86_64.replies | 49 +++++++++--- tests/qemucapabilitiesdata/caps_2.9.0.x86_64.xml | 4 + .../qemuxml2argv-intel-iommu-caching-mode.args | 25 ++++++ .../qemuxml2argv-intel-iommu-caching-mode.xml | 50 ++++++++++++ .../qemuxml2argv-intel-iommu-ioapic.args | 19 +++++ .../qemuxml2argv-intel-iommu-ioapic.xml | 31 ++++++++ tests/qemuxml2argvtest.c | 17 +++++ .../qemuxml2xmlout-intel-iommu-caching-mode.xml | 1 + .../qemuxml2xmlout-intel-iommu-ioapic.xml | 1 + tests/qemuxml2xmltest.c | 2 + 36 files changed, 559 insertions(+), 49 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-caching-mode.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-caching-mode.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-ioapic.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-ioapic.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-intel-iommu-caching-mode.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-intel-iommu-ioapic.xml -- 2.10.2

4 20

[libvirt] libvirt-python RPM installation not recognized by pip
by Cleber Rosa 15 May '17

15 May '17

Hello, When using the standard "requirements.txt" files for installation package dependencies, I noticed that "libvirt-python" would attempt to be installed by "pip" even when the equivalent RPM package is already installed. For instance, on a Fedora 25 system: $ rpm -q libvirt-python libvirt-python-2.2.0-1.fc25.x86_64 $ python -e 'import pkg_resources; pkg_resources.get_distribution("libvirt-python")' ... pkg_resources.DistributionNotFound: The 'libvirt-python' distribution was not found and is required by the application The provider (the actual module) is actually present, but (rightfully so) under the name "libvirt": $ python -c 'import pkg_resources; print pkg_resources.get_provider("libvirt")' <pkg_resources.DefaultProvider instance at 0x7fdfbc67a488> At first sight, this seems to be caused by the lack of an "egg-info" file, such as: $PYTHON_SITE_PACKAGES/libvirt_python-2.2.0-py2.7.egg-info I'm reporting a supposedly packaging issue here since the libvirt-python setup.py file itself includes support for a custom rpm command. Please advise if any of this is intentional, and/or whether this should indeed be reported here or on downstream only. Regards, -- Cleber Rosa [ Sr Software Engineer - Virtualization Team - Red Hat ] [ Avocado Test Framework - avocado-framework.github.io ] [ 7ABB 96EB 8B46 B94D 5E0F E9BB 657E 8D33 A5F2 09F3 ]

3 2