June 2015 - Devel - libvirt List Archives

[libvirt] [v2] storage-pool: API to get/set autostart flag
by Zeeshan Ali (Khattak) 03 Jun '15

03 Jun '15

Add binding for virStoragePoolGetAutostart & virStoragePoolSetAutostart. --- libvirt-gobject/libvirt-gobject-storage-pool.c | 51 ++++++++++++++++++++++++++ libvirt-gobject/libvirt-gobject-storage-pool.h | 5 +++ libvirt-gobject/libvirt-gobject.sym | 6 +++ 3 files changed, 62 insertions(+) diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.c b/libvirt-gobject/libvirt-gobject-storage-pool.c index f3eac0d..7f26b1b 100644 --- a/libvirt-gobject/libvirt-gobject-storage-pool.c +++ b/libvirt-gobject/libvirt-gobject-storage-pool.c @@ -1048,6 +1048,57 @@ gboolean gvir_storage_pool_delete (GVirStoragePool *pool, return TRUE; } +/** + * gvir_storage_pool_get_autostart: + * @pool: the storage pool + * @err: return location for any #GError + * + * Return value: #True if autostart is enabled, #False otherwise. + */ +gboolean gvir_storage_pool_get_autostart(GVirStoragePool *pool, + GError **err) +{ + int ret; + + g_return_val_if_fail(GVIR_IS_STORAGE_POOL(pool), FALSE); + g_return_val_if_fail(err == NULL || *err == NULL, FALSE); + + if (virStoragePoolGetAutostart(pool->priv->handle, &ret)) { + gvir_set_error_literal(err, GVIR_STORAGE_POOL_ERROR, + 0, + "Failed to get autostart flag from storage pool"); + } + + return !!ret; +} + +/** + * gvir_storage_pool_set_autostart: + * @pool: the storage pool + * @autostart: Whether or not to autostart + * @err: return location for any #GError + * + * Sets whether or not storage pool @pool is started automatically on boot. + * + * Return value: #TRUE on success, #FALSE otherwise. + */ +gboolean gvir_storage_pool_set_autostart(GVirStoragePool *pool, + gboolean autostart, + GError **err) +{ + g_return_val_if_fail(GVIR_IS_STORAGE_POOL(pool), FALSE); + g_return_val_if_fail(err == NULL || *err == NULL, FALSE); + + if (virStoragePoolSetAutostart(pool->priv->handle, autostart)) { + gvir_set_error_literal(err, GVIR_STORAGE_POOL_ERROR, + 0, + "Failed to set autostart flag on storage pool"); + return FALSE; + } + + return TRUE; +} + static void gvir_storage_pool_delete_helper(GSimpleAsyncResult *res, GObject *object, diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.h b/libvirt-gobject/libvirt-gobject-storage-pool.h index f8529f0..f7f879c 100644 --- a/libvirt-gobject/libvirt-gobject-storage-pool.h +++ b/libvirt-gobject/libvirt-gobject-storage-pool.h @@ -166,6 +166,11 @@ void gvir_storage_pool_delete_async (GVirStoragePool *pool, gboolean gvir_storage_pool_delete_finish(GVirStoragePool *pool, GAsyncResult *result, GError **err); +gboolean gvir_storage_pool_get_autostart(GVirStoragePool *pool, + GError **err); +gboolean gvir_storage_pool_set_autostart(GVirStoragePool *pool, + gboolean autostart, + GError **err); G_END_DECLS diff --git a/libvirt-gobject/libvirt-gobject.sym b/libvirt-gobject/libvirt-gobject.sym index 927cad9..dcda675 100644 --- a/libvirt-gobject/libvirt-gobject.sym +++ b/libvirt-gobject/libvirt-gobject.sym @@ -265,4 +265,10 @@ LIBVIRT_GOBJECT_0.2.0 { gvir_domain_open_graphics_fd; } LIBVIRT_GOBJECT_0.1.9; +LIBVIRT_GOBJECT_0.2.1 { + global: + gvir_storage_pool_get_autostart; + gvir_storage_pool_set_autostart; +} LIBVIRT_GOBJECT_0.2.0; + # .... define new API here using predicted next version number .... -- 2.4.2

1 1

[libvirt] [PATCH v2] check if console/channel PTY is null before attempting to open
by Shivaprasad G Bhat 03 Jun '15

03 Jun '15

Console/channel devices have their pty devices assigned when the emulator is actually started. If time is spent in guest preparation, someone attempts to open the console/channel, the libvirt crashes in virChrdevLockFilePath(). The patch attempts to fix the crash by adding a check before attempting to open. Signed-off-by: Shivaprasad G Bhat <sbhat(a)linux.vnet.ibm.com> --- src/conf/virchrdev.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/conf/virchrdev.c b/src/conf/virchrdev.c index 5f28f29..701b326 100644 --- a/src/conf/virchrdev.c +++ b/src/conf/virchrdev.c @@ -350,6 +350,11 @@ int virChrdevOpen(virChrdevsPtr devs, switch (source->type) { case VIR_DOMAIN_CHR_TYPE_PTY: path = source->data.file.path; + if (!path) { + virReportError(VIR_ERR_OPERATION_FAILED, "%s", + _("PTY device is not yet assigned")); + return -1; + } break; case VIR_DOMAIN_CHR_TYPE_UNIX: path = source->data.nix.path;

2 1

[libvirt] [PATCH 0/2] lxc: allow using 802.11 wireless interfaces in LXC domains
by Lubomir Rintel 03 Jun '15

03 Jun '15

Hi, I'm using libvirt-lxc with mac80211_hwsim to construct virtual wireless network for NetworkManager testing. Currently, I have to move the interfaces to the domains manually, as sharing wan 802.11 wireless link with a LXC namespace does not work: <hostdev mode='capabilities' type='net'> <source><interface>wlan0</interface></source> </hostdev> The wireless links can only be associated to namespaces by moving the underlying PHY instead of the network interface. I'm following up with patches that fix the problem for me; please take a look. Thank you Lubo

2 7

[libvirt] [PATCH] util: Clear output broadcast address before filling it in
by Martin Kletzander 03 Jun '15

03 Jun '15

Since commit 55ace7c4789c8a7408139460f4b639cee00e5125, the sockettest fails without VIR_TEST_DEBUG set. The problem is found by test number 42 (co-incidence?), which tests range '192.168.122.1' - '192.168.122.255' in network '192.168.122.0/24'. That is supposed to fail because the end address is equal to the broadcast address. When comparing these two in 'virSocketAddrEqual(end, &broadcast)', there is a check for sin_addr as well as for sin_port. That port, however, is different when we do not enable test debugging. With the testing enabled, the port is 0 (correctly initialized), but without that it has a random number there. And that's because the structure is not initialized anywhere. By zeroing the structure before filling in the info, we make sure we return only the address and not any information that was not requested. And the test work once again. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Notes: Pushed as a build-breaker. src/util/virsocketaddr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/util/virsocketaddr.c b/src/util/virsocketaddr.c index 34f2af31b2e7..81539b3730d4 100644 --- a/src/util/virsocketaddr.c +++ b/src/util/virsocketaddr.c @@ -505,6 +505,8 @@ virSocketAddrBroadcast(const virSocketAddr *addr, const virSocketAddr *netmask, virSocketAddrPtr broadcast) { + memset(broadcast, 0, sizeof(*broadcast)); + if ((addr->data.stor.ss_family != AF_INET) || (netmask->data.stor.ss_family != AF_INET)) { broadcast->data.stor.ss_family = AF_UNSPEC; -- 2.4.2

2 1

[libvirt] [PATCH] util: process: Refactor and fix virProcessSetAffinity
by Peter Krempa 03 Jun '15

03 Jun '15

Refactor the function to return the bitmap instead of an integer and the inner workings so that they make more sense. This patch also fixes possible segfault on old systems that was introduced by commit: commit f1a43a8e4139b028257ef4ed05a81cfb5f8a8741 Author: Hu Tao <hutao(a)cn.fujitsu.com> Date: Fri Sep 14 15:46:59 2012 +0800 use virBitmap to store cpu affinity info --- Pushed after review on the libvirt-security list. src/qemu/qemu_driver.c | 5 +-- src/util/virprocess.c | 102 ++++++++++++++++++++++--------------------------- src/util/virprocess.h | 4 +- 3 files changed, 48 insertions(+), 63 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f32b87e..63001b1 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1457,8 +1457,7 @@ qemuDomainHelperGetVcpus(virDomainObjPtr vm, virVcpuInfoPtr info, int maxinfo, unsigned char *cpumap = VIR_GET_CPUMAP(cpumaps, maplen, v); virBitmapPtr map = NULL; - if (virProcessGetAffinity(priv->vcpupids[v], - &map, hostcpus) < 0) + if (!(map = virProcessGetAffinity(priv->vcpupids[v]))) return -1; virBitmapToDataBuf(map, cpumap, maplen); @@ -5727,7 +5726,7 @@ qemuDomainGetIOThreadsLive(virQEMUDriverPtr driver, goto endjob; info_ret[i]->iothread_id = iothreads[i]->iothread_id; - if (virProcessGetAffinity(iothreads[i]->thread_id, &map, hostcpus) < 0) + if (!(map = virProcessGetAffinity(iothreads[i]->thread_id))) goto endjob; if (virBitmapToData(map, &info_ret[i]->cpumap, diff --git a/src/util/virprocess.c b/src/util/virprocess.c index 501569f..a38cb75 100644 --- a/src/util/virprocess.c +++ b/src/util/virprocess.c @@ -468,71 +468,60 @@ int virProcessSetAffinity(pid_t pid, virBitmapPtr map) return 0; } -int virProcessGetAffinity(pid_t pid, - virBitmapPtr *map, - int maxcpu) +virBitmapPtr +virProcessGetAffinity(pid_t pid) { size_t i; -# ifdef CPU_ALLOC - /* New method dynamically allocates cpu mask, allowing unlimted cpus */ - int numcpus = 1024; - size_t masklen; cpu_set_t *mask; + size_t masklen; + virBitmapPtr ret = NULL; - /* Not only may the statically allocated cpu_set_t be too small, - * but there is no way to ask the kernel what size is large enough. - * So you have no option but to pick a size, try, catch EINVAL, - * enlarge, and re-try. - * - * http://lkml.org/lkml/2009/7/28/620 - */ - realloc: - masklen = CPU_ALLOC_SIZE(numcpus); - mask = CPU_ALLOC(numcpus); +# ifdef CPU_ALLOC + /* 262144 cpus ought to be enough for anyone */ + masklen = CPU_ALLOC_SIZE(1024 << 8); + mask = CPU_ALLOC(1024 << 8); if (!mask) { virReportOOMError(); - return -1; + return NULL; } CPU_ZERO_S(masklen, mask); +# else + if (VIR_ALLOC(mask) < 0) + return NULL; + + masklen = sizeof(*mask); + CPU_ZERO(mask); +# endif + if (sched_getaffinity(pid, masklen, mask) < 0) { - CPU_FREE(mask); - if (errno == EINVAL && - numcpus < (1024 << 8)) { /* 262144 cpus ought to be enough for anyone */ - numcpus = numcpus << 2; - goto realloc; - } virReportSystemError(errno, _("cannot get CPU affinity of process %d"), pid); - return -1; + goto cleanup; } - *map = virBitmapNew(maxcpu); - if (!*map) - return -1; + if (!(ret = virBitmapNew(masklen * 8))) + goto cleanup; - for (i = 0; i < maxcpu; i++) + for (i = 0; i < masklen * 8; i++) { +# ifdef CPU_ALLOC if (CPU_ISSET_S(i, masklen, mask)) - ignore_value(virBitmapSetBit(*map, i)); - CPU_FREE(mask); + ignore_value(virBitmapSetBit(ret, i)); # else - /* Legacy method uses a fixed size cpu mask, only allows up to 1024 cpus */ - cpu_set_t mask; - - CPU_ZERO(&mask); - if (sched_getaffinity(pid, sizeof(mask), &mask) < 0) { - virReportSystemError(errno, - _("cannot get CPU affinity of process %d"), pid); - return -1; + if (CPU_ISSET(i, mask)) + ignore_value(virBitmapSetBit(ret, i)); +# endif } - for (i = 0; i < maxcpu; i++) - if (CPU_ISSET(i, &mask)) - ignore_value(virBitmapSetBit(*map, i)); + cleanup: +# ifdef CPU_ALLOC + CPU_FREE(mask); +# else + VIR_FREE(mask); # endif - return 0; + return ret; } #elif defined(HAVE_BSD_CPU_AFFINITY) @@ -562,29 +551,29 @@ int virProcessSetAffinity(pid_t pid, return 0; } -int virProcessGetAffinity(pid_t pid, - virBitmapPtr *map, - int maxcpu) +virBitmapPtr +virProcessGetAffinity(pid_t pid) { size_t i; cpuset_t mask; - - if (!(*map = virBitmapNew(maxcpu))) - return -1; + virBitmapPtr ret = NULL; CPU_ZERO(&mask); if (cpuset_getaffinity(CPU_LEVEL_WHICH, CPU_WHICH_PID, pid, sizeof(mask), &mask) != 0) { virReportSystemError(errno, _("cannot get CPU affinity of process %d"), pid); - return -1; + return NULL; } - for (i = 0; i < maxcpu; i++) + if (!(*map = virBitmapNew(sizeof(mask) * 8))) + return NULL; + + for (i = 0; i < sizeof(mask) * 8; i++) if (CPU_ISSET(i, &mask)) - ignore_value(virBitmapSetBit(*map, i)); + ignore_value(virBitmapSetBit(ret, i)); - return 0; + return ret; } #else /* HAVE_SCHED_GETAFFINITY */ @@ -597,13 +586,12 @@ int virProcessSetAffinity(pid_t pid ATTRIBUTE_UNUSED, return -1; } -int virProcessGetAffinity(pid_t pid ATTRIBUTE_UNUSED, - virBitmapPtr *map ATTRIBUTE_UNUSED, - int maxcpu ATTRIBUTE_UNUSED) +virBitmapPtr +virProcessGetAffinity(pid_t pid ATTRIBUTE_UNUSED) { virReportSystemError(ENOSYS, "%s", _("Process CPU affinity is not supported on this platform")); - return -1; + return NULL; } #endif /* HAVE_SCHED_GETAFFINITY */ diff --git a/src/util/virprocess.h b/src/util/virprocess.h index c812882..1b98493 100644 --- a/src/util/virprocess.h +++ b/src/util/virprocess.h @@ -58,9 +58,7 @@ int virProcessKillPainfully(pid_t pid, bool force); int virProcessSetAffinity(pid_t pid, virBitmapPtr map); -int virProcessGetAffinity(pid_t pid, - virBitmapPtr *map, - int maxcpu); +virBitmapPtr virProcessGetAffinity(pid_t pid); int virProcessGetStartTime(pid_t pid, unsigned long long *timestamp); -- 2.4.1

1 0

[libvirt] [PATCH] virsh: Fix Ctrl-C behavior when watching a job
by Jiri Denemark 03 Jun '15

03 Jun '15

When watching a job (save, managedsave, dump, migrate) virsh spawns a thread to call the appropriate API and waits for the result while watching for interruption signals (SIGINT, Ctrl-C on the terminal). Whenever such signal is caught, virsh calls virDomainAbortJob, stops waiting for the job, and returns the result of virDomainAbortJob. This is wrong because the job might have finished in the meantime or it might have been cancelled by someone else and virsh would just report the failure to abort the job. However, we are not interested in the virDomainAbortJob's result at all, we need to keep waiting for the main job to finish and report its result instead. https://bugzilla.redhat.com/show_bug.cgi?id=1131755 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- tools/virsh-domain.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index fc4d1fc..044b088 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -4234,12 +4234,9 @@ vshWatchJob(vshControl *ctl, if (ret > 0) { if (pollfd[1].revents & POLLIN && saferead(STDIN_FILENO, &retchar, sizeof(retchar)) > 0) { - if (vshTTYIsInterruptCharacter(ctl, retchar)) { + if (vshTTYIsInterruptCharacter(ctl, retchar)) virDomainAbortJob(dom); - goto cleanup; - } else { - continue; - } + continue; } if (pollfd[0].revents & POLLIN && @@ -4259,9 +4256,8 @@ vshWatchJob(vshControl *ctl, if (intCaught) { virDomainAbortJob(dom); intCaught = 0; - } else { - continue; } + continue; } goto cleanup; } -- 2.4.1

2 2

[libvirt] [PATCH 0/8 v3] parallels: add vz driver
by Maxim Nestratov 03 Jun '15

03 Jun '15

v2-v3 change: * Patch "parallels: set virtType depending on driver name" reworked. Implemented ability to set virtType depending on driver name. v1-v2 change: * Only "parallels: accept vz as a driver uri and name" patch is changed. Fixed ability to connect to 'parallels' driver and deal with 'Paralells' domain virt type. Maxim Nestratov (8): parallels: introduce vz driver constant and string parallels: use newly introduced VIR_DOMAIN_VIRT_VZ parallels: add new guest capabilities assigned to vz driver parallels: accept vz as a driver uri and name parallels: add a new vz connection driver and hypervisor structures parallels: increment the number of connection drivers parallels: recommend to connect to vz:///system when connection fails parallels: set virtType depending on driver name src/conf/domain_conf.c | 19 ++++++++---- src/conf/domain_conf.h | 1 + src/libvirt.c | 2 +- src/parallels/parallels_driver.c | 63 +++++++++++++++++++++++++++++++++++---- src/parallels/parallels_network.c | 3 +- src/parallels/parallels_sdk.c | 6 +++- src/parallels/parallels_storage.c | 3 +- src/parallels/parallels_utils.h | 1 + 8 files changed, 82 insertions(+), 16 deletions(-) -- 2.1.0

3 10

[libvirt] [PATCH] conf:audit: fix no audit log when start a vm with iothread
by Luyao Huang 02 Jun '15

02 Jun '15

Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/conf/domain_audit.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/conf/domain_audit.c b/src/conf/domain_audit.c index 4ea10d2..1900039 100644 --- a/src/conf/domain_audit.c +++ b/src/conf/domain_audit.c @@ -885,6 +885,8 @@ virDomainAuditStart(virDomainObjPtr vm, const char *reason, bool success) virDomainAuditMemory(vm, 0, vm->def->mem.cur_balloon, "start", true); virDomainAuditVcpu(vm, 0, vm->def->vcpus, "start", true); + if (vm->def->iothreads) + virDomainAuditIOThread(vm, 0, vm->def->iothreads, "start", true); virDomainAuditLifecycle(vm, "start", reason, success); } -- 1.8.3.1

3 2

[libvirt] [PATCH] qemu: fix forget pass the return value to variable @ret
by Luyao Huang 02 Jun '15

02 Jun '15

If we do not pass the return value from qemuDomainRemoveRNGDevice() function to variable @ret, qemuDomainRemoveDevice() functiuon will always fail when recive rng device unplug event from qemu monitor. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_hotplug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 170768b..f596730 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3271,7 +3271,7 @@ qemuDomainRemoveDevice(virQEMUDriverPtr driver, ret = qemuDomainRemoveChrDevice(driver, vm, dev->data.chr); break; case VIR_DOMAIN_DEVICE_RNG: - qemuDomainRemoveRNGDevice(driver, vm, dev->data.rng); + ret = qemuDomainRemoveRNGDevice(driver, vm, dev->data.rng); break; case VIR_DOMAIN_DEVICE_MEMORY: -- 1.8.3.1

3 2

[libvirt] [PATCH] qemu: fix wrong call addressrelease function when attach RNG device success
by Luyao Huang 02 Jun '15

02 Jun '15

Add a return value check to avoid the wrong address release. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_hotplug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 170768b..f70e3d8 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1695,7 +1695,7 @@ qemuDomainAttachRNGDevice(virQEMUDriverPtr driver, audit: virDomainAuditRNG(vm, NULL, rng, "attach", ret == 0); cleanup: - if (vm) + if (ret < 0 && vm) qemuDomainReleaseDeviceAddress(vm, &rng->info, NULL); VIR_FREE(charAlias); VIR_FREE(objAlias); -- 1.8.3.1

3 2