- Devel - libvirt List Archives

[libvirt] [PATCH] conf: Don't output <cpu> tag if it contains no information.
by Andrea Bolognani 13 Apr '15

13 Apr '15

The tag is already marked as optional in the schema, so no changes are needed there. RHBZ: https://bugzilla.redhat.com/show_bug.cgi?id=1202606 --- Hi everyone, this is my first contribution to libvirt so I wholeheartedly welcome any criticism, suggestion or recommendation :) Cheers. src/conf/cpu_conf.c | 33 ++++++++++++++++------ tests/qemuxml2argvdata/qemuxml2argv-cpu-empty.xml | 23 +++++++++++++++ .../qemuxml2xmlout-cpu-empty.xml | 21 ++++++++++++++ tests/qemuxml2xmltest.c | 1 + 4 files changed, 69 insertions(+), 9 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-cpu-empty.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-cpu-empty.xml diff --git a/src/conf/cpu_conf.c b/src/conf/cpu_conf.c index cd3882d..8f65d55 100644 --- a/src/conf/cpu_conf.c +++ b/src/conf/cpu_conf.c @@ -435,13 +435,14 @@ virCPUDefFormatBufFull(virBufferPtr buf, bool updateCPU) { int ret = -1; + virBuffer attributeBuf = VIR_BUFFER_INITIALIZER; virBuffer childrenBuf = VIR_BUFFER_INITIALIZER; int indent = virBufferGetIndent(buf, false); if (!def) return 0; - virBufferAddLit(buf, "<cpu"); + /* Format attributes */ if (def->type == VIR_CPU_TYPE_GUEST) { const char *tmp; @@ -451,7 +452,7 @@ virCPUDefFormatBufFull(virBufferPtr buf, _("Unexpected CPU mode %d"), def->mode); goto cleanup; } - virBufferAsprintf(buf, " mode='%s'", tmp); + virBufferAsprintf(&attributeBuf, " mode='%s'", tmp); } if (def->model && @@ -463,10 +464,11 @@ virCPUDefFormatBufFull(virBufferPtr buf, def->match); goto cleanup; } - virBufferAsprintf(buf, " match='%s'", tmp); + virBufferAsprintf(&attributeBuf, " match='%s'", tmp); } } + /* Format children */ virBufferAdjustIndent(&childrenBuf, indent + 2); if (def->arch) virBufferAsprintf(&childrenBuf, "<arch>%s</arch>\n", @@ -477,16 +479,29 @@ virCPUDefFormatBufFull(virBufferPtr buf, if (virDomainNumaDefCPUFormat(&childrenBuf, numa) < 0) goto cleanup; - if (virBufferUse(&childrenBuf)) { - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childrenBuf); - virBufferAddLit(buf, "</cpu>\n"); - } else { - virBufferAddLit(buf, "/>\n"); + /* Put it all together */ + if (virBufferUse(&attributeBuf) || virBufferUse(&childrenBuf)) { + + /* Opening tag */ + virBufferAddLit(buf, "<cpu"); + + /* Attributes (if any) */ + if (virBufferUse(&attributeBuf)) + virBufferAddBuffer(buf, &attributeBuf); + + /* Children (if any) and closing tag */ + if (virBufferUse(&childrenBuf)) { + virBufferAddLit(buf, ">\n"); + virBufferAddBuffer(buf, &childrenBuf); + virBufferAddLit(buf, "</cpu>\n"); + } else { + virBufferAddLit(buf, "/>\n"); + } } ret = 0; cleanup: + virBufferFreeAndReset(&attributeBuf); virBufferFreeAndReset(&childrenBuf); return ret; } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-cpu-empty.xml b/tests/qemuxml2argvdata/qemuxml2argv-cpu-empty.xml new file mode 100644 index 0000000..2a79826 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-cpu-empty.xml @@ -0,0 +1,23 @@ +<domain type='kvm'> + <name>cpu-empty</name> + <uuid>1aed4c39-ad6e-4a78-9264-4ce996290d17</uuid> + <memory unit='KiB'>4000768</memory> + <currentMemory unit='KiB'>1048576</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='x86_64' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <cpu> + </cpu> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-kvm</emulator> + <controller type='usb' index='0'/> + <controller type='pci' index='0' model='pci-root'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2xmloutdata/qemuxml2xmlout-cpu-empty.xml b/tests/qemuxml2xmloutdata/qemuxml2xmlout-cpu-empty.xml new file mode 100644 index 0000000..e678607 --- /dev/null +++ b/tests/qemuxml2xmloutdata/qemuxml2xmlout-cpu-empty.xml @@ -0,0 +1,21 @@ +<domain type='kvm'> + <name>cpu-empty</name> + <uuid>1aed4c39-ad6e-4a78-9264-4ce996290d17</uuid> + <memory unit='KiB'>4000768</memory> + <currentMemory unit='KiB'>1048576</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='x86_64' 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-kvm</emulator> + <controller type='usb' index='0'/> + <controller type='pci' index='0' model='pci-root'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 817e408..cd0b280 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -361,6 +361,7 @@ mymain(void) DO_TEST("clock-utc"); DO_TEST("clock-localtime"); + DO_TEST_DIFFERENT("cpu-empty"); DO_TEST("cpu-kvmclock"); DO_TEST("cpu-host-kvmclock"); DO_TEST("cpu-host-passthrough-features"); -- 2.1.0

2 2

[libvirt] An implementation bug in qemuMigrationWaitForCompletion that introduces an unnecessary sleep of 50 ms when a migration completes
by Xing Lin 13 Apr '15

13 Apr '15

Hi, It seems that I found a bug in the qemuMigrationWaitForCompletion(). It will do a sleep of 50 ms, even when qemuMigrationUpdateJobStatus() detects that a migration has completed. Here is the original implementation of the while loop. while (jobInfo->type == VIR_DOMAIN_JOB_UNBOUNDED) { /* Poll every 50ms for progress & to allow cancellation */ struct timespec ts = { .tv_sec = 0, .tv_nsec = 50 * 1000 * 1000ull }; if (qemuMigrationUpdateJobStatus(driver, vm, job, asyncJob) == -1) break; /* cancel migration if disk I/O error is emitted while migrating */ if (abort_on_error && virDomainObjGetState(vm, &pauseReason) == VIR_DOMAIN_PAUSED && pauseReason == VIR_DOMAIN_PAUSED_IOERROR) { virReportError(VIR_ERR_OPERATION_FAILED, _("%s: %s"), job, _("failed due to I/O error")); break; } if (dconn && virConnectIsAlive(dconn) <= 0) { virReportError(VIR_ERR_OPERATION_FAILED, "%s", _("Lost connection to destination host")); break; } virObjectUnlock(vm); nanosleep(&ts, NULL); virObjectLock(vm); } Even when qemuMigrationUpdateJobStatus() detects a migration has completed, the nanosleep() will be called again. When the while condition is checked, then the program breaks from the while loop. This introduces an unnecessary sleep of 50 ms which can be avoided by doing a sleep first. The code should likely be structured as following. while (jobInfo->type == VIR_DOMAIN_JOB_UNBOUNDED) { /* Poll every 50ms for progress & to allow cancellation */ struct timespec ts = { .tv_sec = 0, .tv_nsec = 50 * 1000 * 1000ull }; // do sleep here. virObjectUnlock(vm); nanosleep(&ts, NULL); virObjectLock(vm); if (qemuMigrationUpdateJobStatus(driver, vm, job, asyncJob) == -1) break; /* cancel migration if disk I/O error is emitted while migrating */ if (abort_on_error && virDomainObjGetState(vm, &pauseReason) == VIR_DOMAIN_PAUSED && pauseReason == VIR_DOMAIN_PAUSED_IOERROR) { virReportError(VIR_ERR_OPERATION_FAILED, _("%s: %s"), job, _("failed due to I/O error")); break; } if (dconn && virConnectIsAlive(dconn) <= 0) { virReportError(VIR_ERR_OPERATION_FAILED, "%s", _("Lost connection to destination host")); break; } } I have filed a bug report at redhat bugzilla, with a bug id 1210502. The link is below. https://bugzilla.redhat.com/show_bug.cgi?id=1210502 I also prepared a patch for fixing it. It is attached. Any comments?

3 4

[libvirt] [PATCH 0/3] Fix usb device version parsing in redir filters
by Ján Tomko 13 Apr '15

13 Apr '15

https://bugzilla.redhat.com/show_bug.cgi?id=1210650 Ján Tomko (3): Do xml->xml test for usb-redir-filter Fix usb device version parsing issues Rewrite usb device version parsing src/conf/domain_conf.c | 68 +++++++--------------- .../qemuxml2argv-usb-redir-filter-version.args | 19 ++++++ .../qemuxml2argv-usb-redir-filter-version.xml | 46 +++++++++++++++ tests/qemuxml2argvtest.c | 6 ++ .../qemuxml2xmlout-usb-redir-filter-version.xml | 46 +++++++++++++++ .../qemuxml2xmlout-usb-redir-filter.xml | 45 ++++++++++++++ tests/qemuxml2xmltest.c | 2 + 7 files changed, 185 insertions(+), 47 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-redir-filter-version.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-redir-filter-version.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-usb-redir-filter-version.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-usb-redir-filter.xml -- 2.0.5

2 6

[libvirt] [PATCH 0/2] virsh vcpupin cleanups
by Ján Tomko 13 Apr '15

13 Apr '15

Ján Tomko (2): Rewrite vshPrintPinInfo Rewrite vshParseCPUList tools/virsh-domain.c | 153 ++++++++++----------------------------------------- 1 file changed, 29 insertions(+), 124 deletions(-) -- 2.0.5

2 4

[libvirt] [PATCH] Do not require virtio-pci devices when using virtio-mmio
by Ján Tomko 13 Apr '15

13 Apr '15

On arm, we probe for virtio-*-pci devices, but use their virtio-*-device variants. Set the capabilities based on the -device variants as well, to make them work with qemus with the PCI devices compiled out. --- src/qemu/qemu_capabilities.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index ce6767c..3b49271 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -1499,6 +1499,7 @@ struct virQEMUCapsStringFlags virQEMUCapsObjectTypes[] = { { "virtio-scsi-pci", QEMU_CAPS_VIRTIO_SCSI }, { "virtio-scsi-s390", QEMU_CAPS_VIRTIO_SCSI }, { "virtio-scsi-ccw", QEMU_CAPS_VIRTIO_SCSI }, + { "virtio-scsi-device", QEMU_CAPS_VIRTIO_SCSI }, { "megasas", QEMU_CAPS_SCSI_MEGASAS }, { "spicevmc", QEMU_CAPS_DEVICE_SPICEVMC }, { "qxl-vga", QEMU_CAPS_DEVICE_QXL_VGA }, @@ -1515,6 +1516,7 @@ struct virQEMUCapsStringFlags virQEMUCapsObjectTypes[] = { { "virtio-rng-pci", QEMU_CAPS_DEVICE_VIRTIO_RNG }, { "virtio-rng-s390", QEMU_CAPS_DEVICE_VIRTIO_RNG }, { "virtio-rng-ccw", QEMU_CAPS_DEVICE_VIRTIO_RNG }, + { "virtio-rng-device", QEMU_CAPS_DEVICE_VIRTIO_RNG }, { "rng-random", QEMU_CAPS_OBJECT_RNG_RANDOM }, { "rng-egd", QEMU_CAPS_OBJECT_RNG_EGD }, { "spapr-nvram", QEMU_CAPS_DEVICE_NVRAM }, -- 2.0.5

2 1

[libvirt] [libvirt-glib] Support setting of compat XML node
by Richa Sehgal 12 Apr '15

12 Apr '15

This change adds support for setting of compat XML node in libvirt gconfig storage volumes target --- libvirt-gconfig/libvirt-gconfig-storage-vol-target.c | 13 +++++++++++++ libvirt-gconfig/libvirt-gconfig-storage-vol-target.h | 2 ++ libvirt-gconfig/libvirt-gconfig.sym | 5 +++++ libvirt-gconfig/tests/test-domain-create.c | 1 + 4 files changed, 21 insertions(+) diff --git a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c index d3151d1..b72b304 100644 --- a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c +++ b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c @@ -99,3 +99,16 @@ void gvir_config_storage_vol_target_set_permissions(GVirConfigStorageVolTarget * "permissions", GVIR_CONFIG_OBJECT(perms)); } + +/** + * gvir_config_storage_vol_target_set_compat: + * @compat: (allow-none): + */ +void gvir_config_storage_vol_target_set_compat(GVirConfigStorageVolTarget *target, + const char *compat) +{ + g_return_if_fail(GVIR_CONFIG_IS_STORAGE_VOL_TARGET(target)); + + gvir_config_object_set_node_content(GVIR_CONFIG_OBJECT(target), + "compat", compat); +} diff --git a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h index b572381..c165e2b 100644 --- a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h +++ b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h @@ -67,6 +67,8 @@ void gvir_config_storage_vol_target_set_format(GVirConfigStorageVolTarget *targe const char *format); void gvir_config_storage_vol_target_set_permissions(GVirConfigStorageVolTarget *target, GVirConfigStoragePermissions *perms); +void gvir_config_storage_vol_target_set_compat(GVirConfigStorageVolTarget *target, + const char *compat); G_END_DECLS diff --git a/libvirt-gconfig/libvirt-gconfig.sym b/libvirt-gconfig/libvirt-gconfig.sym index 8614126..407a52f 100644 --- a/libvirt-gconfig/libvirt-gconfig.sym +++ b/libvirt-gconfig/libvirt-gconfig.sym @@ -714,4 +714,9 @@ global: gvir_config_domain_cpu_set_model; } LIBVIRT_GCONFIG_0.1.8; +LIBVIRT_GCONFIG_0.2.0 { +global: + gvir_config_storage_vol_target_set_compat; +} LIBVIRT_GCONFIG_0.1.9; + # .... define new API here using predicted next version number .... diff --git a/libvirt-gconfig/tests/test-domain-create.c b/libvirt-gconfig/tests/test-domain-create.c index eb4b945..66f618b 100644 --- a/libvirt-gconfig/tests/test-domain-create.c +++ b/libvirt-gconfig/tests/test-domain-create.c @@ -482,6 +482,7 @@ int main(int argc, char **argv) vol_target = gvir_config_storage_vol_target_new(); gvir_config_storage_vol_target_set_format(vol_target, "qcow2"); gvir_config_storage_vol_target_set_permissions(vol_target, perms); + gvir_config_storage_vol_target_set_compat(vol_target, "1.1"); g_object_unref(G_OBJECT(perms)); gvir_config_storage_vol_set_target(vol, vol_target); g_object_unref(G_OBJECT(vol_target)); -- 1.9.1

1 0

[libvirt] [libvirt-glib] Support setting of compat XML node
by Richa Sehgal 12 Apr '15

12 Apr '15

This change adds support for setting of compat XML node in libvirt gconfig storage volumes target --- libvirt-gconfig/libvirt-gconfig-storage-vol-target.c | 13 +++++++++++++ libvirt-gconfig/libvirt-gconfig-storage-vol-target.h | 2 ++ libvirt-gconfig/libvirt-gconfig.sym | 5 +++++ libvirt-gconfig/tests/test-domain-create.c | 1 + 4 files changed, 21 insertions(+) diff --git a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c index d3151d1..b72b304 100644 --- a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c +++ b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.c @@ -99,3 +99,16 @@ void gvir_config_storage_vol_target_set_permissions(GVirConfigStorageVolTarget * "permissions", GVIR_CONFIG_OBJECT(perms)); } + +/** + * gvir_config_storage_vol_target_set_compat: + * @compat: (allow-none): + */ +void gvir_config_storage_vol_target_set_compat(GVirConfigStorageVolTarget *target, + const char *compat) +{ + g_return_if_fail(GVIR_CONFIG_IS_STORAGE_VOL_TARGET(target)); + + gvir_config_object_set_node_content(GVIR_CONFIG_OBJECT(target), + "compat", compat); +} diff --git a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h index b572381..c165e2b 100644 --- a/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h +++ b/libvirt-gconfig/libvirt-gconfig-storage-vol-target.h @@ -67,6 +67,8 @@ void gvir_config_storage_vol_target_set_format(GVirConfigStorageVolTarget *targe const char *format); void gvir_config_storage_vol_target_set_permissions(GVirConfigStorageVolTarget *target, GVirConfigStoragePermissions *perms); +void gvir_config_storage_vol_target_set_compat(GVirConfigStorageVolTarget *target, + const char *compat); G_END_DECLS diff --git a/libvirt-gconfig/libvirt-gconfig.sym b/libvirt-gconfig/libvirt-gconfig.sym index 8614126..407a52f 100644 --- a/libvirt-gconfig/libvirt-gconfig.sym +++ b/libvirt-gconfig/libvirt-gconfig.sym @@ -714,4 +714,9 @@ global: gvir_config_domain_cpu_set_model; } LIBVIRT_GCONFIG_0.1.8; +LIBVIRT_GCONFIG_0.2.0 { +global: + gvir_config_storage_vol_target_set_compat; +} LIBVIRT_GCONFIG_0.1.9; + # .... define new API here using predicted next version number .... diff --git a/libvirt-gconfig/tests/test-domain-create.c b/libvirt-gconfig/tests/test-domain-create.c index eb4b945..66f618b 100644 --- a/libvirt-gconfig/tests/test-domain-create.c +++ b/libvirt-gconfig/tests/test-domain-create.c @@ -482,6 +482,7 @@ int main(int argc, char **argv) vol_target = gvir_config_storage_vol_target_new(); gvir_config_storage_vol_target_set_format(vol_target, "qcow2"); gvir_config_storage_vol_target_set_permissions(vol_target, perms); + gvir_config_storage_vol_target_set_compat(vol_target, "1.1"); g_object_unref(G_OBJECT(perms)); gvir_config_storage_vol_set_target(vol, vol_target); g_object_unref(G_OBJECT(vol_target)); -- 1.9.1

1 0

[libvirt] [PATCH] AUTHORS: Add my email address to AUTHORS.in
by Dmitry Guryanov 10 Apr '15

10 Apr '15

--- AUTHORS.in | 1 + 1 file changed, 1 insertion(+) diff --git a/AUTHORS.in b/AUTHORS.in index 6673b3f..7e4359f 100644 --- a/AUTHORS.in +++ b/AUTHORS.in @@ -14,6 +14,7 @@ Claudio Bley <claudio.bley(a)gmail.com> Cole Robinson <crobinso(a)redhat.com> Daniel Berrange <berrange(a)redhat.com> Daniel Veillard <veillard(a)redhat.com> +Dmitry Guryanov <dguryanov(a)parallels.com> Doug Goldstein <cardoe(a)gentoo.org> Eric Blake <eblake(a)redhat.com> Erik Skultety <eskultet(a)redhat.com> -- 2.1.0

1 0

[libvirt] [PATCH v2 0/5] Refactor cgroup and qemuDomainHotplugVcpus
by John Ferlan 10 Apr '15

10 Apr '15

v1 here: http://www.redhat.com/archives/libvir-list/2015-April/msg00126.html John Ferlan (5): cgroup: Introduce virCgroupNewThread cgroup: Use virCgroupNewThread qemu: qemuDomainHotplugVcpus - separate out the add cgroup qemu: qemuDomainHotplugVcpus - separate out the del cgroup and pin qemu: qemuDomainHotplugVcpus - separate out pin adjustment code src/libvirt_private.syms | 4 +- src/qemu/qemu_cgroup.c | 22 +++-- src/qemu/qemu_driver.c | 233 ++++++++++++++++++++++++++++++----------------- src/util/vircgroup.c | 144 +++++++---------------------- src/util/vircgroup.h | 30 +++--- 5 files changed, 208 insertions(+), 225 deletions(-) -- 2.1.0

2 13

Re: [libvirt] [openstack-dev] [nova] The risk of hanging when shutdown instance.
by zhang bo 10 Apr '15

10 Apr '15

On 2015/3/28 18:06, Rui Chen wrote: > Thank you for reply, Chris. > > > 2015-03-27 23:15 GMT+08:00 Chris Friesen <chris.friesen(a)windriver.com <mailto:chris.friesen@windriver.com>>: > > On 03/26/2015 07:44 PM, Rui Chen wrote: > > Yes, you are right, but we found our instance hang at first dom.shutdown() call, > if the dom.shutdown() don't return, there is no chance to execute dom.destroy(), > right? > > > Correct. The code is written assuming dom.shutdown() cannot block indefinitely. > > The libvirt docs at "https://libvirt.org/html/__libvirt-libvirt-domain.html#__virDomainShutdown <https://libvirt.org/html/libvirt-libvirt-domain.html#virDomainShutdown>" say "...this command returns as soon as the shutdown request is issued rather than blocking until the guest is no longer running." > > If dom.shutdown() blocks indefinitely, then that's a libvirt bug. > > > Chris The API virDomainShutdown's description is out of date, it's not correct. In fact, virDomainShutdown would block or not, depending on its mode. If it's in mode *agent*, then it would be blocked until qemu founds that the guest actually got down. Otherwise, if it's in mode *acpi*, then it would return immediately. Thus, maybe further more work need to be done in Openstack. What's your opinions, Michal and Daniel (from libvirt.org) and Chris (from openstack.org) :)

5 5