July 2017 - Devel - libvirt List Archives

[libvirt] [PATCH] Avoid hidden cgroup mount points
by Juan Hernandez 12 Jul '17

12 Jul '17

Currently the scan of the /proc/mounts file used to find cgroup mount points doesn't take into account that mount points may hidden by other mount points. For, example in certain Kubernetes environments the /proc/mounts contains the following lines: cgroup /sys/fs/cgroup/net_prio,net_cls cgroup ... tmpfs /sys/fs/cgroup tmpfs ... cgroup /sys/fs/cgroup/net_cls,net_prio cgroup ... In this particular environment the first mount point is hidden by the second one. The correct mount point is the third one, but libvirt will never process it because it only checks the first mount point for each controller (net_cls in this case). So libvirt will try to use the first mount point, which doesn't actually exist, and the complete detection process will fail. To avoid that issue this patch changes the virCgroupDetectMountsFromFile function so that when there are duplicates it takes the information from the last line in /proc/mounts. This requires removing the previous explicit condition to skip duplicates, and adding code to free the memory used by the processing of duplicated lines. Related-To: https://bugzilla.redhat.com/1468214 Related-To: https://github.com/kubevirt/libvirt/issues/4 Signed-off-by: Juan Hernandez <jhernand(a)redhat.com> --- src/util/vircgroup.c | 23 ++++++++++++++--------- tests/vircgroupdata/kubevirt.mounts | 36 ++++++++++++++++++++++++++++++++++++ tests/vircgroupdata/kubevirt.parsed | 10 ++++++++++ tests/vircgrouptest.c | 1 + 4 files changed, 61 insertions(+), 9 deletions(-) create mode 100644 tests/vircgroupdata/kubevirt.mounts create mode 100644 tests/vircgroupdata/kubevirt.parsed diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index 5aa1db5..41d90e7 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -397,6 +397,7 @@ virCgroupDetectMountsFromFile(virCgroupPtr group, const char *typestr = virCgroupControllerTypeToString(i); int typelen = strlen(typestr); char *tmp = entry.mnt_opts; + struct virCgroupController *controller = &group->controllers[i]; while (tmp) { char *next = strchr(tmp, ','); int len; @@ -406,18 +407,21 @@ virCgroupDetectMountsFromFile(virCgroupPtr group, } else { len = strlen(tmp); } - /* NB, the same controller can appear >1 time in mount list - * due to bind mounts from one location to another. Pick the - * first entry only - */ - if (typelen == len && STREQLEN(typestr, tmp, len) && - !group->controllers[i].mountPoint) { + + if (typelen == len && STREQLEN(typestr, tmp, len)) { char *linksrc; struct stat sb; char *tmp2; - if (VIR_STRDUP(group->controllers[i].mountPoint, - entry.mnt_dir) < 0) + /* Note that the lines in /proc/mounts have the same + * order than the mount operations, and that there may + * be duplicates due to bind mounts. This means + * that the same mount point may be processed more than + * once. We need to save the results of the last one, + * and we need to be careful to release the memory used + * by previous processing. */ + VIR_FREE(controller->mountPoint); + if (VIR_STRDUP(controller->mountPoint, entry.mnt_dir) < 0) goto error; tmp2 = strrchr(entry.mnt_dir, '/'); @@ -453,7 +457,8 @@ virCgroupDetectMountsFromFile(virCgroupPtr group, VIR_WARN("Expecting a symlink at %s for controller %s", linksrc, typestr); } else { - group->controllers[i].linkPoint = linksrc; + VIR_FREE(controller->linkPoint); + controller->linkPoint = linksrc; } } } diff --git a/tests/vircgroupdata/kubevirt.mounts b/tests/vircgroupdata/kubevirt.mounts new file mode 100644 index 0000000..b0d31bb --- /dev/null +++ b/tests/vircgroupdata/kubevirt.mounts @@ -0,0 +1,36 @@ +tmpfs /sys/fs/cgroup tmpfs rw,nosuid,nodev,noexec,relatime,mode=755 0 0 +cgroup /sys/fs/cgroup/systemd cgroup rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd 0 0 +cgroup /sys/fs/cgroup/cpuacct,cpu cgroup rw,nosuid,nodev,noexec,relatime,cpuacct,cpu 0 0 +cgroup /sys/fs/cgroup/pids cgroup rw,nosuid,nodev,noexec,relatime,pids 0 0 +cgroup /sys/fs/cgroup/blkio cgroup rw,nosuid,nodev,noexec,relatime,blkio 0 0 +cgroup /sys/fs/cgroup/cpuset cgroup rw,nosuid,nodev,noexec,relatime,cpuset 0 0 +cgroup /sys/fs/cgroup/memory cgroup rw,nosuid,nodev,noexec,relatime,memory 0 0 +cgroup /sys/fs/cgroup/hugetlb cgroup rw,nosuid,nodev,noexec,relatime,hugetlb 0 0 +cgroup /sys/fs/cgroup/devices cgroup rw,nosuid,nodev,noexec,relatime,devices 0 0 +cgroup /sys/fs/cgroup/freezer cgroup rw,nosuid,nodev,noexec,relatime,freezer 0 0 +cgroup /sys/fs/cgroup/perf_event cgroup rw,nosuid,nodev,noexec,relatime,perf_event 0 0 +cgroup /sys/fs/cgroup/net_prio,net_cls cgroup rw,nosuid,nodev,noexec,relatime,net_prio,net_cls 0 0 +tmpfs /host-sys/fs/cgroup tmpfs ro,nosuid,nodev,noexec,relatime,mode=755 0 0 +cgroup /host-sys/fs/cgroup/systemd cgroup rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd 0 0 +cgroup /host-sys/fs/cgroup/cpu,cpuacct cgroup rw,nosuid,nodev,noexec,relatime,cpuacct,cpu 0 0 +cgroup /host-sys/fs/cgroup/pids cgroup rw,nosuid,nodev,noexec,relatime,pids 0 0 +cgroup /host-sys/fs/cgroup/blkio cgroup rw,nosuid,nodev,noexec,relatime,blkio 0 0 +cgroup /host-sys/fs/cgroup/cpuset cgroup rw,nosuid,nodev,noexec,relatime,cpuset 0 0 +cgroup /host-sys/fs/cgroup/memory cgroup rw,nosuid,nodev,noexec,relatime,memory 0 0 +cgroup /host-sys/fs/cgroup/hugetlb cgroup rw,nosuid,nodev,noexec,relatime,hugetlb 0 0 +cgroup /host-sys/fs/cgroup/devices cgroup rw,nosuid,nodev,noexec,relatime,devices 0 0 +cgroup /host-sys/fs/cgroup/freezer cgroup rw,nosuid,nodev,noexec,relatime,freezer 0 0 +cgroup /host-sys/fs/cgroup/perf_event cgroup rw,nosuid,nodev,noexec,relatime,perf_event 0 0 +cgroup /host-sys/fs/cgroup/net_cls,net_prio cgroup rw,nosuid,nodev,noexec,relatime,net_prio,net_cls 0 0 +tmpfs /sys/fs/cgroup tmpfs ro,nosuid,nodev,noexec,relatime,mode=755 0 0 +cgroup /sys/fs/cgroup/systemd cgroup rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd 0 0 +cgroup /sys/fs/cgroup/cpu,cpuacct cgroup rw,nosuid,nodev,noexec,relatime,cpuacct,cpu 0 0 +cgroup /sys/fs/cgroup/pids cgroup rw,nosuid,nodev,noexec,relatime,pids 0 0 +cgroup /sys/fs/cgroup/blkio cgroup rw,nosuid,nodev,noexec,relatime,blkio 0 0 +cgroup /sys/fs/cgroup/cpuset cgroup rw,nosuid,nodev,noexec,relatime,cpuset 0 0 +cgroup /sys/fs/cgroup/memory cgroup rw,nosuid,nodev,noexec,relatime,memory 0 0 +cgroup /sys/fs/cgroup/hugetlb cgroup rw,nosuid,nodev,noexec,relatime,hugetlb 0 0 +cgroup /sys/fs/cgroup/devices cgroup rw,nosuid,nodev,noexec,relatime,devices 0 0 +cgroup /sys/fs/cgroup/freezer cgroup rw,nosuid,nodev,noexec,relatime,freezer 0 0 +cgroup /sys/fs/cgroup/perf_event cgroup rw,nosuid,nodev,noexec,relatime,perf_event 0 0 +cgroup /sys/fs/cgroup/net_cls,net_prio cgroup rw,nosuid,nodev,noexec,relatime,net_prio,net_cls 0 0 diff --git a/tests/vircgroupdata/kubevirt.parsed b/tests/vircgroupdata/kubevirt.parsed new file mode 100644 index 0000000..3377af0 --- /dev/null +++ b/tests/vircgroupdata/kubevirt.parsed @@ -0,0 +1,10 @@ +cpu /sys/fs/cgroup/cpu,cpuacct +cpuacct /sys/fs/cgroup/cpu,cpuacct +cpuset /sys/fs/cgroup/cpuset +memory /sys/fs/cgroup/memory +devices /sys/fs/cgroup/devices +freezer /sys/fs/cgroup/freezer +blkio /sys/fs/cgroup/blkio +net_cls /sys/fs/cgroup/net_cls,net_prio +perf_event /sys/fs/cgroup/perf_event +name=systemd /sys/fs/cgroup/systemd diff --git a/tests/vircgrouptest.c b/tests/vircgrouptest.c index 8af5e2c..b932b1a 100644 --- a/tests/vircgrouptest.c +++ b/tests/vircgrouptest.c @@ -885,6 +885,7 @@ mymain(void) DETECT_MOUNTS("cgroups3"); DETECT_MOUNTS("all-in-one"); DETECT_MOUNTS("no-cgroups"); + DETECT_MOUNTS("kubevirt"); if (virTestRun("New cgroup for self", testCgroupNewForSelf, NULL) < 0) ret = -1; -- 2.9.4

3 2

[libvirt] [PATCH v2 0/5] Misc cleanups for internal.h & fix tests with CLang
by Daniel P. Berrange 12 Jul '17

12 Jul '17

When writing the fix for test suite mocking under CLang I found a bunch of cruft in internal.h The first four patches thus cleanup up internal.h. We then add the extra annotations requird to prevent CLang optimizer breaking mock overrides. Changed in v2: - Fixed version check to find clang - Use 'printf' instead of 'gnu_printf' on clang still - Addd fix for mock functions under clang Daniel P. Berrange (5): Remove duplicate define of __GNUC_PREREQ Require use of GCC 4.4 or CLang compilers Remove network constants out of internal.h Remove incorrectly used TODO macro Prevent more compiler optimization of mockable functions build-aux/mock-noinline.pl | 2 +- config-post.h | 24 ++--- src/check-symfile.pl | 2 +- src/internal.h | 173 ++++++++++----------------------- src/libxl/libxl_conf.c | 1 + src/nwfilter/nwfilter_dhcpsnoop.c | 1 + src/nwfilter/nwfilter_gentech_driver.c | 1 + src/qemu/qemu_capspriv.h | 2 +- src/qemu/qemu_conf.c | 1 + src/rpc/virnetsocket.h | 4 +- src/util/vircommand.h | 2 +- src/util/vircrypto.h | 2 +- src/util/virfile.h | 2 +- src/util/virhostcpu.h | 4 +- src/util/virmacaddr.h | 2 +- src/util/virnetdev.h | 8 +- src/util/virnetdevip.h | 2 +- src/util/virnetdevopenvswitch.h | 2 +- src/util/virnetdevtap.h | 6 +- src/util/virnuma.h | 16 +-- src/util/virrandom.h | 6 +- src/util/virscsi.h | 2 +- src/util/virscsivhost.h | 2 +- src/util/virsocketaddr.h | 16 +++ src/util/virtpm.h | 2 +- src/util/virutil.c | 1 + src/util/virutil.h | 10 +- src/util/viruuid.h | 2 +- src/vz/vz_sdk.c | 1 + src/xen/xen_hypervisor.c | 6 +- src/xen/xend_internal.c | 6 +- 31 files changed, 135 insertions(+), 176 deletions(-) -- 2.9.4

6 19

[libvirt] [PATCH] qemuDomainGetPreservedMountPath: rename @mount
by Michal Privoznik 12 Jul '17

12 Jul '17

Obviously, old gcc-s ale sad when a variable shares the name with a function. And we do have such variable (added in 4d8a914be0): @mount. Rename it to @mountpoint so that compiler's happy again. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Pushed under build-breaker and trivial rules. src/qemu/qemu_domain.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index e146bc88a..ae260de00 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -7578,9 +7578,9 @@ qemuDomainGetHostdevPath(virDomainDefPtr def, * qemuDomainGetPreservedMountPath: * @cfg: driver configuration data * @vm: domain object - * @mount: mount point path to convert + * @mountpoint: mount point path to convert * - * For given @mount point return new path where the mount point + * For given @mountpoint return new path where the mount point * should be moved temporarily whilst building the namespace. * * Returns: allocated string on success which the caller must free, @@ -7589,21 +7589,21 @@ qemuDomainGetHostdevPath(virDomainDefPtr def, static char * qemuDomainGetPreservedMountPath(virQEMUDriverConfigPtr cfg, virDomainObjPtr vm, - const char *mount) + const char *mountpoint) { char *path = NULL; char *tmp; - const char *suffix = mount + strlen(DEVPREFIX); + const char *suffix = mountpoint + strlen(DEVPREFIX); size_t off; - if (STREQ(mount, "/dev")) + if (STREQ(mountpoint, "/dev")) suffix = "dev"; if (virAsprintf(&path, "%s/%s.%s", cfg->stateDir, vm->def->name, suffix) < 0) return NULL; - /* Now consider that @mount is "/dev/blah/blah2". + /* Now consider that @mountpoint is "/dev/blah/blah2". * @suffix then points to "blah/blah2". However, caller * expects all the @paths to be the same depth. The * caller doesn't always do `mkdir -p` but sometimes bare -- 2.13.0

1 0

[libvirt] [PATCH] qemu: Provide non-linux stub for qemuDomainAttachDeviceMknodRecursive
by Michal Privoznik 11 Jul '17

11 Jul '17

The way we create devices under /dev is highly linux specific. For instance we do mknod(), mount(), umount(), etc. Some platforms are even missing some of these functions. Then again, as declared in qemuDomainNamespaceAvailable(): namespaces are linux only. Therefore, to avoid obfuscating the code by trying to make it compile on weird platforms, just provide a non-linux stub for qemuDomainAttachDeviceMknodRecursive(). At the same time, qemuDomainAttachDeviceMknodHelper() which actually calls the non-existent functions is moved under ifdef __linux__ block since its only caller is in that block too. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_domain.c | 38 ++++++++++++++++++++++++++++++-------- 1 file changed, 30 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 080ff336e..eb1a9794b 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -8541,6 +8541,8 @@ struct qemuDomainAttachDeviceMknodData { }; +/* Our way of creating devices is highly linux specific */ +#if defined(__linux__) static int qemuDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, void *opaque) @@ -8638,7 +8640,7 @@ qemuDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, goto cleanup; } -#ifdef WITH_SELINUX +# ifdef WITH_SELINUX if (data->tcon && lsetfilecon_raw(data->file, (VIR_SELINUX_CTX_CONST char *) data->tcon) < 0) { VIR_WARNINGS_NO_WLOGICALOP_EQUAL_EXPR @@ -8650,7 +8652,7 @@ qemuDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, goto cleanup; } } -#endif +# endif /* Finish mount process started earlier. */ if (isReg && @@ -8661,9 +8663,9 @@ qemuDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, cleanup: if (ret < 0 && delDevice) unlink(data->file); -#ifdef WITH_SELINUX +# ifdef WITH_SELINUX freecon(data->tcon); -#endif +# endif virFileFreeACLs(&data->acl); return ret; } @@ -8754,14 +8756,14 @@ qemuDomainAttachDeviceMknodRecursive(virQEMUDriverPtr driver, goto cleanup; } -#ifdef WITH_SELINUX +# ifdef WITH_SELINUX if (lgetfilecon_raw(file, &data.tcon) < 0 && (errno != ENOTSUP && errno != ENODATA)) { virReportSystemError(errno, _("Unable to get SELinux label from %s"), file); goto cleanup; } -#endif +# endif if (STRPREFIX(file, DEVPREFIX)) { size_t i; @@ -8798,9 +8800,9 @@ qemuDomainAttachDeviceMknodRecursive(virQEMUDriverPtr driver, ret = 0; cleanup: -#ifdef WITH_SELINUX +# ifdef WITH_SELINUX freecon(data.tcon); -#endif +# endif virFileFreeACLs(&data.acl); if (isReg && target) umount(target); @@ -8810,6 +8812,26 @@ qemuDomainAttachDeviceMknodRecursive(virQEMUDriverPtr driver, } +#else /* !defined(__linux__) */ + + +static int +qemuDomainAttachDeviceMknodRecursive(virQEMUDriverPtr driver ATTRIBUTE_UNUSED, + virDomainObjPtr vm ATTRIBUTE_UNUSED, + const char *file ATTRIBUTE_UNUSED, + char * const *devMountsPath ATTRIBUTE_UNUSED, + size_t ndevMountsPath ATTRIBUTE_UNUSED, + unsigned int ttl ATTRIBUTE_UNUSED) +{ + virReportSystemError(ENOSYS, "%s", + _("Namespaces are not supported on this platform.")); + return -1; +} + + +#endif /* !defined(__linux__) */ + + static int qemuDomainAttachDeviceMknod(virQEMUDriverPtr driver, virDomainObjPtr vm, -- 2.13.0

2 1

[libvirt] [PATCH 0/6] qemu: Clean up storage file access
by Peter Krempa 11 Jul '17

11 Jul '17

A collection of few cleanups and bugfixes. Peter Krempa (6): lib: Remove misplaced and redundant comments tests: storage: Fully register storage driver storage: Split out virStorageSource accessors to separate file storage: Make virStorageFileReadHeader more universal qemu: Use storage driver APIs in qemuDomainBlockPeek qemu: Support only raw volumes in qemuDomainBlockPeek po/POTFILES.in | 1 + src/Makefile.am | 1 + src/libvirt-domain.c | 4 +- src/qemu/qemu_domain.c | 1 + src/qemu/qemu_driver.c | 35 +- src/security/virt-aa-helper.c | 2 +- src/storage/storage_backend.h | 9 +- src/storage/storage_backend_fs.c | 20 +- src/storage/storage_backend_gluster.c | 37 ++- src/storage/storage_driver.c | 551 +------------------------------ src/storage/storage_driver.h | 28 -- src/storage/storage_source.c | 587 ++++++++++++++++++++++++++++++++++ src/storage/storage_source.h | 54 ++++ tests/virstoragetest.c | 6 +- 14 files changed, 709 insertions(+), 627 deletions(-) create mode 100644 src/storage/storage_source.c create mode 100644 src/storage/storage_source.h -- 2.12.2

2 16

[libvirt] [PATCH] internal: don't use weak symbols for Win32 platform
by Daniel P. Berrange 11 Jul '17

11 Jul '17

The Win32 platform will fail to link if you use weak symbols because it is incompatible with exporting symbols in a DLL: Cannot export virRandomGenerateWWN: symbol wrong type (2 vs 3) We only need weak symbols for our test suite to do LD_PRELOAD and this doesn't work on Win32, so we can just drop the hack for Win32 Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed as a build fix for Win32 src/internal.h | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/src/internal.h b/src/internal.h index 00edd4f..edc3587 100644 --- a/src/internal.h +++ b/src/internal.h @@ -128,10 +128,14 @@ * */ # ifndef ATTRIBUTE_MOCKABLE -# if __GNUC_PREREQ(4, 5) -# define ATTRIBUTE_MOCKABLE __attribute__((__noinline__, __noclone__, __weak__)) +# if defined(WIN32) +# define ATTRIBUTE_MOCKABLE # else -# define ATTRIBUTE_MOCKABLE __attribute__((__noinline__, __weak__)) +# if __GNUC_PREREQ(4, 5) +# define ATTRIBUTE_MOCKABLE __attribute__((__noinline__, __noclone__, __weak__)) +# else +# define ATTRIBUTE_MOCKABLE __attribute__((__noinline__, __weak__)) +# endif # endif # endif -- 2.9.4

1 0

[libvirt] [PATCH] qemu: handle missing bind host/service on chardev hotplug
by Ján Tomko 11 Jul '17

11 Jul '17

On domain startup, bind host or bind service can be omitted and we will format a working command line. Extend this to hotplug as well and specify the service to QEMU even if the host is missing. https://bugzilla.redhat.com/show_bug.cgi?id=1452441 --- src/qemu/qemu_monitor_json.c | 13 ++++++++++--- tests/qemumonitorjsontest.c | 11 +++++++++++ 2 files changed, 21 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 0837290..ca9bb14 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -6429,6 +6429,8 @@ qemuMonitorJSONAttachCharDevCommand(const char *chrID, virJSONValuePtr data = NULL; virJSONValuePtr addr = NULL; const char *backend_type = NULL; + const char *host; + const char *port; char *tlsalias = NULL; bool telnet; @@ -6492,9 +6494,14 @@ qemuMonitorJSONAttachCharDevCommand(const char *chrID, virJSONValueObjectAppend(data, "remote", addr) < 0) goto error; - if (chr->data.udp.bindHost) { - addr = qemuMonitorJSONBuildInetSocketAddress(chr->data.udp.bindHost, - chr->data.udp.bindService); + host = chr->data.udp.bindHost; + port = chr->data.udp.bindService; + if (host || port) { + if (!host) + host = ""; + if (!port) + port = ""; + addr = qemuMonitorJSONBuildInetSocketAddress(host, port); if (!addr || virJSONValueObjectAppend(data, "local", addr) < 0) goto error; diff --git a/tests/qemumonitorjsontest.c b/tests/qemumonitorjsontest.c index e9f9d47..3de901c 100644 --- a/tests/qemumonitorjsontest.c +++ b/tests/qemumonitorjsontest.c @@ -895,6 +895,17 @@ qemuMonitorJSONTestAttachChardev(virDomainXMLOptionPtr xmlopt) "'data':{'host':'localhost'," "'port':'4321'}}}}}"); + chr.data.udp.bindHost = NULL; + chr.data.udp.bindService = (char *) "4321"; + CHECK("udp", false, + "{'id':'alias'," + "'backend':{'type':'udp'," + "'data':{'remote':{'type':'inet'," + "'data':{'host':'example.com'," + "'port':'1234'}}," + "'local':{'type':'inet'," + "'data':{'host':''," + "'port':'4321'}}}}}"); memset(&chr, 0, sizeof(chr)); chr.type = VIR_DOMAIN_CHR_TYPE_UNIX; chr.data.nix.path = (char *) "/path/to/socket"; -- 2.10.2

2 2

[libvirt] [PATCH] tests: add further XML namespace test
by Daniel P. Berrange 11 Jul '17

11 Jul '17

Validate that we can pass QEMU command line options using a default namespace, instead of a prefixed namespace Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- .../qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.args | 25 ++++++++++++++++++ .../qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.xml | 30 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + 3 files changed, 56 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.xml diff --git a/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.args b/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.args new file mode 100644 index 0000000..562b562 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.args @@ -0,0 +1,25 @@ +LC_ALL=C \ +PATH=/bin \ +HOME=/home/test \ +USER=test \ +LOGNAME=test \ +QEMU_AUDIO_DRV=none \ +NS=ns \ +BAR='' \ +/usr/bin/qemu-system-i686 \ +-name QEMUGuest1 \ +-S \ +-M pc \ +-m 214 \ +-smp 1,sockets=1,cores=1,threads=1 \ +-uuid c7a5fdbd-edaf-9455-926a-d65c16db1809 \ +-nographic \ +-nodefaults \ +-monitor unix:/tmp/lib/domain--1-QEMUGuest1/monitor.sock,server,nowait \ +-no-acpi \ +-boot c \ +-usb \ +-drive file=/dev/HostVG/QEMUGuest1,format=raw,if=none,id=drive-ide0-0-0 \ +-device ide-drive,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 \ +-unknown parameter diff --git a/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.xml b/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.xml new file mode 100644 index 0000000..491fc2d --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-qemu-ns-alt.xml @@ -0,0 +1,30 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219136</memory> + <currentMemory unit='KiB'>219136</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-system-i686</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='ide' index='0'/> + </devices> + <commandline xmlns='http://libvirt.org/schemas/domain/qemu/1.0'> + <arg value='-unknown'/> + <arg value='parameter'/> + <env name='NS' value='ns'/> + <env name='BAR'/> + </commandline> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 27eea70..be2a9bc 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -1505,6 +1505,7 @@ mymain(void) DO_TEST("qemu-ns", NONE); DO_TEST("qemu-ns-no-env", NONE); + DO_TEST("qemu-ns-alt", NONE); DO_TEST("smp", NONE); -- 2.9.4

2 3

[libvirt] [PATCH] qemu: Fix qemuDomainGetBlockInfo allocation value setting
by John Ferlan 11 Jul '17

11 Jul '17

https://bugzilla.redhat.com/show_bug.cgi?id=1467826 Commit id 'b9b1aa639' was supposed to add logic to set the allocation for sparse files when wr_highest_offset was zero; however, an unconditional setting was done just prior. For block devices, this means allocation is always returning 0 since 'actual-size' will be zero. Remove the unconditional setting and add the note about it being possible to still be zero for block devices. As soon as the guest starts writing to the volume, the allocation value will then be obtainable from qemu via the wr_highest_offset. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/qemu/qemu_driver.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index cdb727b..66fb80a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -11711,10 +11711,9 @@ qemuDomainGetBlockInfo(virDomainPtr dom, * Additionally, if qemu hasn't written to the file yet, then set the * allocation to whatever qemu returned for physical (e.g. the "actual- * size" from the json query) as that will match the expected allocation - * value for this API. */ + * value for this API. NB: May still be 0 for block. */ if (entry->physical == 0 || info->allocation == 0 || info->allocation == entry->physical) { - info->allocation = entry->physical; if (info->allocation == 0) info->allocation = entry->physical; -- 2.9.4

2 1

[libvirt] How to make gic_version=3 as defailt to qemu on arm64
by Vishnu Pajjuri 11 Jul '17

11 Jul '17

Hi I'm running Openstack which is installed by using devstack. But it is not launching VMs. >From command line with gic_version=3 option it is running. But openstack glance doesn't have any privilege to specify gic version. On my ARM64 board gicv2 is not supported, so i want to make gicv3 as default one to pass to qemu. Kindly suggest any specific version of vibvirt or patch such that libvirt should pass gicv3 as default one. Thanks in Advance -Vishnu.

2 1