July 2013 - Devel - libvirt List Archives

[libvirt] [PATCH] maint: update to latest gnulib
by Eric Blake 11 Jul '13

11 Jul '13

Future patches need LGPLv2+ versions of some modules that had recent license changes; but separating the gnulib update from the actual use of the modules makes it easier to backport to an older version while avoiding a submodule update (assuming, of course, that the backport is to a system where glibc provides adequate functionaliy without needing the gnulib module). * .gnulib: Update to latest, for modules needed in later patches. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the trivial gnulib rule. * .gnulib b72ff2a...da8d59e (9): > regex: port to --with-included-regex --enable-gcc-warnings non-threaded > accept4, dup3, pipe2: port to Cygwin > mountlist: add support for deallocating returned list entries > stdalign, verify: port to FreeBSD 9.1, to C11, and to C++11 > fnmatch: don't goto over declaration > pipe2: decouple from binary-io a bit > mgetgroups: relax license to LGPLv2+ > xalloc-oversized: relax license to LGPLv2+ > nproc: relax license to LGPLv2+ .gnulib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index b72ff2a..da8d59e 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit b72ff2a45efde544c406804186d37a3254728571 +Subproject commit da8d59ee79138b85daee1ad5b22ea12e4fb6ce47 -- 1.8.1.4

1 0

[libvirt] [PATCH v2 00/10] Re-enable memballoon driver statistics reporting
by John Ferlan 11 Jul '13

11 Jul '13

This patchset replaces: https://www.redhat.com/archives/libvir-list/2013-July/msg00108.html Changes since V1: * Remove the external (eg, qemu_monitor.{c,h}) API's to the qom-list, qom-get, qom-set interfaces * Handle errors in virXPathInt() return checking domain_conf.c * Move the 'period' from _qemuDomainObjPrivate into _qemuMonitor * Move/rename qemuProcessFindBalloonObjectPath from qemu_process.c into qemu_monitor.c as qemuMonitorFindBalloonObjectPath() * Create qemuMonitorJSONSetMemoryStatsPeriod() in qemu_monitor_json.c from parts of prior change to add qemuProcessUpdateBalloonStatsPeriod() in qemu_process.c and hav * Add qemuMonitorSetMemoryStatsPeriod() in qemu_monitor.c to set the collection period. Adjust the code to take parts of prior change and make check for balloon path in this path (set if necessary). * Adjust code in qemuProcessReconnect(), qemuProcessStart(), and qemuProcessAttach() to call qemuMonitorSetMemoryStatsPeriod() * Adjust qemuMonitorJSONGetMemoryStats() to get the balloon stats once the "actual" stat is collected from qemuMonitorJSONGetBalloonInfo(). Reviewer note: The prior code to get "actual" was a duplicate of the BalloonInfo code. Since "balloon-info" is where the actual memory is still stored it still must be called as that data is not present in the balloon stats qom-get "stats" fetch implemented as the new API qemuMonitorJSONGetBalloonStats(). * Avoid overloading virDomainSetMemoryFlags()... * Create/use virDomainSetMemoryStatsPeriodFlags() to set the collection period dynamically from "virsh dommemstats" using the --period qualifier. John Ferlan (10): Add qemuMonitorJSONGetObjectListPaths() method for QMP qom-list command Add qemuMonitorJSONGetObjectProperty() method for QMP qom-get command Add qemuMonitorJSONSetObjectProperty() method for QMP qom-set command Add 'period' for Memballoon statistics gathering capability Determine whether to start balloon memory stats gathering. Add capability to fetch balloon stats Add new public API virDomainSetMemoryStatsPeriodFlags Specify remote protocol for virDomainSetMemoryStatsPeriodFlags Implement the virDomainSetMemoryStatsPeriodFlags for QEMU driver Allow balloon driver collection to be adjusted dynamically docs/formatdomain.html.in | 19 ++ docs/schemas/domaincommon.rng | 7 + include/libvirt/libvirt.h.in | 3 + src/conf/domain_conf.c | 44 +++- src/conf/domain_conf.h | 1 + src/driver.h | 6 + src/libvirt.c | 65 ++++++ src/libvirt_public.syms | 5 + src/qemu/qemu_driver.c | 66 ++++++ src/qemu/qemu_monitor.c | 133 ++++++++++- src/qemu/qemu_monitor.h | 2 + src/qemu/qemu_monitor_json.c | 498 ++++++++++++++++++++++++++++++++---------- src/qemu/qemu_monitor_json.h | 64 ++++++ src/qemu/qemu_process.c | 21 +- src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 15 +- src/remote_protocol-structs | 6 + tests/qemumonitorjsontest.c | 186 ++++++++++++++++ tools/virsh-domain-monitor.c | 66 +++++- 19 files changed, 1085 insertions(+), 123 deletions(-) -- 1.8.1.4

4 32

[libvirt] [PATCH 0/7] Resolve Coverity issues
by John Ferlan 11 Jul '13

11 Jul '13

The recent changes to change 'int i' to 'size_t i' seem to have uncovered a few new Coverity issues - resolve those. Also add a [dead_error_code] tag to a Coverity issue that's been there for a bit. After these patches the result is just back to 1 Coverity error in 'virLXCProcessReboot' regarding the usage of 'conn' which still needs to be eventually resolved, see: https://www.redhat.com/archives/libvir-list/2013-May/msg01218.html and subsequent followups John Ferlan (7): hellolibvirt: Resolve Coverity issues testutils: Resolve Coverity issues virsh-domain-monitor: Resolve Coverity issues virsh-interface: Resolve Coverity issues virsh-nodedev: Resolve Coverity issues storage_backend: Resolve Coverity issue qemu_hostdev: Resolve Coverity issue examples/hellolibvirt/hellolibvirt.c | 7 +++++++ src/qemu/qemu_hostdev.c | 12 +++++++----- src/storage/storage_backend.c | 1 + tests/testutils.c | 3 +++ tools/virsh-domain-monitor.c | 10 ++++++---- tools/virsh-interface.c | 24 +++++++++++++----------- tools/virsh-nodedev.c | 12 +++++++----- 7 files changed, 44 insertions(+), 25 deletions(-) -- 1.8.1.4

4 18

[libvirt] [PATCH RFC]:tolerate numa_node_size64 < 0 because nodeid might start from 1 instead of 0
by hejia hejia 11 Jul '13

11 Jul '13

In nodeGetFreeMemory/nodeGetCellsFreeMemory, they will calculate the free memory of every nodes. They assumed that nodeid of NUMA machine must be continuous and start from 0. But here is a counter-example: # numactl -H available: 1 nodes (1) node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 node 1 size: 16340 MB node 1 free: 11065 MB test results before this patch: #virsh freecell error: internal error Failed to query NUMA free memory #virsh freecell 0 error: internal error Failed to query NUMA free memory for node: 0 after this patch: # virsh freecell Total: 15772580 KiB # virsh freecell 0 0: 0 KiB --- libvirt/src/nodeinfo.c.orig 2013-07-08 04:25:11.970351101 -0500 +++ libvirt/src/nodeinfo.c 2013-07-08 09:00:30.834471495 -0500 @@ -1717,10 +1717,6 @@ nodeGetCellsFreeMemory(unsigned long lon for (numCells = 0, n = startCell; n <= lastCell; n++) { long long mem; if (numa_node_size64(n, &mem) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("Failed to query NUMA free memory for node: %d"), - n); - goto cleanup; } freeMems[numCells++] = mem; } @@ -1743,9 +1739,6 @@ nodeGetFreeMemory(void) for (n = 0; n <= numa_max_node(); n++) { long long mem; if (numa_node_size64(n, &mem) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - "%s", _("Failed to query NUMA free memory")); - goto cleanup; } freeMem += mem; }

3 4

[libvirt] [PATCH 1/2] LXC: hostdev: introduce lxcContainerSetupHostdevCapsMakePath
by Gao feng 11 Jul '13

11 Jul '13

This helper function is used to create parent directroy for the hostdev which will be added to the container. if the parent directory of this hostdev doesn't exist, the mknod of the hostdev will fail. Signed-off-by: Gao feng <gaofeng(a)cn.fujitsu.com> --- src/lxc/lxc_container.c | 17 +++++++++++++++++ src/lxc/lxc_container.h | 2 ++ 2 files changed, 19 insertions(+) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index c8420db..b954107 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1544,6 +1544,23 @@ cleanup: } +int lxcContainerSetupHostdevCapsMakePath(char *dev) +{ + int ret = 0; + char *dir = NULL; + + if ((dir = strrchr(dev, '/'))) { + *dir = '\0'; + if ((virFileMakePath(dev) < 0) && (errno != EEXIST)) + ret = -1; + + *dir = '/'; + } + + return ret; +} + + static int lxcContainerSetupHostdevCapsStorage(virDomainDefPtr vmDef ATTRIBUTE_UNUSED, virDomainHostdevDefPtr def ATTRIBUTE_UNUSED, virSecurityManagerPtr securityDriver ATTRIBUTE_UNUSED) diff --git a/src/lxc/lxc_container.h b/src/lxc/lxc_container.h index ada72f7..f168703 100644 --- a/src/lxc/lxc_container.h +++ b/src/lxc/lxc_container.h @@ -63,6 +63,8 @@ int lxcContainerStart(virDomainDefPtr def, int lxcContainerAvailable(int features); +int lxcContainerSetupHostdevCapsMakePath(char *dev); + virArch lxcContainerGetAlt32bitArch(virArch arch); #endif /* LXC_CONTAINER_H */ -- 1.8.3.1

3 4

[libvirt] [PATCH 0/3] Fix hotplug error messages
by Michal Privoznik 11 Jul '13

11 Jul '13

https://bugzilla.redhat.com/show_bug.cgi?id=982304 The problem is virDomainDeviceDefParse's incompleteness. When doing a hot(un-)plug or update, the passed XML snippet is parsed. However, if virDomainDeviceDefParse doesn't recognize the device (due its implementation incompleteness) an error is returned: error: XML error: unknown device type and the control doesn't even reach the check for supported devices for current operation, where much more correct error message would be printed out: error: Operation not supported: persistent detach of device 'memballoon' is not supported Michar Privoznik (3): conf: Extend virDomainDeviceDefParse handled types conf: Rework virDomainDeviceDefParse qemu: Fix hot (un-)plug error codes and messages src/conf/domain_conf.c | 111 ++++++++++++++++++++++++++++++++++--------------- src/qemu/qemu_driver.c | 42 ++++++++++--------- 2 files changed, 101 insertions(+), 52 deletions(-) -- 1.8.1.5

2 6

[libvirt] [PATCH] Reject duplicate disk targets
by Martin Kletzander 11 Jul '13

11 Jul '13

We never check for disks with duplicate targets connected to the same controller/bus/etc. That means we go ahead, create the same IDs for them and pass them to qemu, which subsequently fails to start. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=968899 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/conf/domain_conf.c | 42 +++++++++++++++++++++++++++++++++++++++++- 1 file changed, 41 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 3398d8b..01720e1 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2629,6 +2629,45 @@ virDomainDeviceInfoIterate(virDomainDefPtr def, static int +virDomainDefRejectDuplicateDiskTargets(virDomainDefPtr def) +{ + char *disk_id = NULL; + int ret = -1; + size_t i = 0; + virHashTablePtr targets = NULL; + + if (!(targets = virHashCreate(def->ndisks, NULL))) + goto cleanup; + + for (i = 0; i < def->ndisks; i++) { + virDomainDiskDefPtr disk = def->disks[i]; + + if (virAsprintf(&disk_id, "%d%s%d%d%d%d", + disk->bus, + NULLSTR(disk->dst), + disk->info.addr.drive.controller, + disk->info.addr.drive.bus, + disk->info.addr.drive.target, + disk->info.addr.drive.unit) < 0) + goto cleanup; + + if (virHashAddEntry(targets, disk_id, NULL) < 0) { + virReportError(VIR_ERR_XML_ERROR, + _("Multiple disks with same address and target ('%s')"), + def->disks[i]->dst); + VIR_FREE(disk_id); + goto cleanup; + } + VIR_FREE(disk_id); + } + + ret = 0; + cleanup: + virHashFree(targets); + return ret; +} + +static int virDomainDefRejectDuplicateControllers(virDomainDefPtr def) { int max_idx[VIR_DOMAIN_CONTROLLER_TYPE_LAST]; @@ -2762,7 +2801,8 @@ virDomainDefPostParseInternal(virDomainDefPtr def, } } - if (virDomainDefRejectDuplicateControllers(def) < 0) + if (virDomainDefRejectDuplicateControllers(def) < 0 || + virDomainDefRejectDuplicateDiskTargets(def) < 0) return -1; return 0; } -- 1.8.3.2

2 6

[libvirt] [PATCH] LXC: remove some incorrect setting ATTRIBUTE_UNUSED
by Gao feng 11 Jul '13

11 Jul '13

these parameters shouldn't be marked as ATTRIBUTE_UNUSED. Signed-off-by: Gao feng <gaofeng(a)cn.fujitsu.com> --- src/lxc/lxc_container.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 257cf93..543e0d1 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1449,9 +1449,9 @@ static int lxcContainerSetupAllDisks(virDomainDefPtr vmDef, } -static int lxcContainerSetupHostdevSubsysUSB(virDomainDefPtr vmDef ATTRIBUTE_UNUSED, - virDomainHostdevDefPtr def ATTRIBUTE_UNUSED, - virSecurityManagerPtr securityDriver ATTRIBUTE_UNUSED) +static int lxcContainerSetupHostdevSubsysUSB(virDomainDefPtr vmDef, + virDomainHostdevDefPtr def, + virSecurityManagerPtr securityDriver) { int ret = -1; char *src = NULL; @@ -1541,9 +1541,9 @@ cleanup: } -static int lxcContainerSetupHostdevCapsStorage(virDomainDefPtr vmDef ATTRIBUTE_UNUSED, - virDomainHostdevDefPtr def ATTRIBUTE_UNUSED, - virSecurityManagerPtr securityDriver ATTRIBUTE_UNUSED) +static int lxcContainerSetupHostdevCapsStorage(virDomainDefPtr vmDef, + virDomainHostdevDefPtr def, + virSecurityManagerPtr securityDriver) { char *src = NULL; int ret = -1; @@ -1603,9 +1603,9 @@ cleanup: } -static int lxcContainerSetupHostdevCapsMisc(virDomainDefPtr vmDef ATTRIBUTE_UNUSED, - virDomainHostdevDefPtr def ATTRIBUTE_UNUSED, - virSecurityManagerPtr securityDriver ATTRIBUTE_UNUSED) +static int lxcContainerSetupHostdevCapsMisc(virDomainDefPtr vmDef, + virDomainHostdevDefPtr def, + virSecurityManagerPtr securityDriver) { char *src = NULL; int ret = -1; @@ -1873,9 +1873,9 @@ static int lxcContainerResolveSymlinks(virDomainDefPtr vmDef) * It removes some capabilities that could be dangerous to * host system, since they are not currently "containerized" */ -static int lxcContainerDropCapabilities(bool keepReboot ATTRIBUTE_UNUSED) -{ #if WITH_CAPNG +static int lxcContainerDropCapabilities(bool keepReboot) +{ int ret; capng_get_caps_process(); @@ -1907,11 +1907,15 @@ static int lxcContainerDropCapabilities(bool keepReboot ATTRIBUTE_UNUSED) * container it is fine for SECURE_NOROOT / SECURE_NO_SETUID_FIXUP to * be unmasked - they can never escape the bounding set. */ + return 0; +} #else +static int lxcContainerDropCapabilities(bool keepReboot ATTRIBUTE_UNUSED) +{ VIR_WARN("libcap-ng support not compiled in, unable to clear capabilities"); -#endif return 0; } +#endif /** -- 1.8.3.1

2 1

[libvirt] [PATCH] storage: return -1 when fs pool can't be mounted
by Ján Tomko 11 Jul '13

11 Jul '13

Don't reuse the return value of virStorageBackendFileSystemIsMounted. If it's 0, we'd return it even if the mount command failed. Also, don't report another error if it's -1, since one has already been reported. Introduced by 258e06c. https://bugzilla.redhat.com/show_bug.cgi?id=981251 --- src/storage/storage_backend_fs.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 8b40f55..d305b06 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -355,6 +355,7 @@ virStorageBackendFileSystemMount(virStoragePoolObjPtr pool) { pool->def->source.format == VIR_STORAGE_POOL_NETFS_GLUSTERFS); virCommandPtr cmd = NULL; int ret = -1; + int rc; if (pool->def->type == VIR_STORAGE_POOL_NETFS) { if (pool->def->source.nhost != 1) { @@ -381,10 +382,12 @@ virStorageBackendFileSystemMount(virStoragePoolObjPtr pool) { } /* Short-circuit if already mounted */ - if ((ret = virStorageBackendFileSystemIsMounted(pool)) != 0) { - virReportError(VIR_ERR_OPERATION_INVALID, - _("Target '%s' is already mounted"), - pool->def->target.path); + if ((rc = virStorageBackendFileSystemIsMounted(pool)) != 0) { + if (rc == 1) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("Target '%s' is already mounted"), + pool->def->target.path); + } return -1; } -- 1.8.1.5

2 2

[libvirt] [PATCH] qemu: Slightly increase memory limit
by Jiri Denemark 11 Jul '13

11 Jul '13

For low-memory domains (roughly under 400MB) our automatic memory limit computation comes up with a limit that's too low. This is because the 0.5 multiplication does not add enough for such small values. Let's increase the constant part of the computation to fix this. --- src/qemu/qemu_domain.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 28b8ee4..c38c86e 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -2197,7 +2197,7 @@ qemuDomainMemoryLimit(virDomainDefPtr def) * limit' has been chosen: * (1 + k) * (domain memory + total video memory) + (32MB for * cache per each disk) + F - * where k = 0.5 and F = 200MB. The cache for disks is important as + * where k = 0.5 and F = 400MB. The cache for disks is important as * kernel cache on the host side counts into the RSS limit. * Moreover, VFIO requires some amount for IO space. Alex Williamson * suggested adding 1GiB for IO space just to be safe (some finer @@ -2212,7 +2212,7 @@ qemuDomainMemoryLimit(virDomainDefPtr def) mem += def->videos[i]->vram; mem *= 1.5; mem += def->ndisks * 32768; - mem += 204800; + mem += 409600; for (i = 0; i < def->nhostdevs; i++) { virDomainHostdevDefPtr hostdev = def->hostdevs[i]; -- 1.8.2.1

2 2