- Devel - libvirt List Archives

[libvirt] [PATCH v1] libxl: reverse defaults on HVM net device attach
by Joao Martins 15 Dec '16

15 Dec '16

libvirt libxl picks its own default with respect to the default NIC to use. libxlMakeNic is the one responsible for this and on boot it picks LIBXL_NIC_TYPE_VIF_IOEMU for HVM domains such that it accomodates both PV and emulated one. The good behaving guest at boot will then select the pv and unplug the emulated device. Now, on HVM when attaching an interface it will pick the same default that is LIBXL_NIC_TYPE_VIF_IOEMU which as a result will fail the attach (see xen commit 32e9d0f ("libxl: nic type defaults to vif in hotplug for hvm guest"). Xen doesn't yet support the hotplug of emulated devices, but we don't want to rule out that case either, which might get support in the future. Hence we simply reverse the defaults when we are attaching the interface which allows libvirt to prefer the PV nic first without adding "model='netfront'" following the same pattern as above commit. Also to avoid ruling out the emulated one we set to LIBXL_NIC_TYPE_IOEMU when setting a model type that is not 'netfront'. Signed-off-by: Joao Martins <joao.m.martins(a)oracle.com> Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- Since RFC: - Joao: Add mention to HVM domains in the first paragraph of commit message. - Jim: Add comment about libxl_device_nic type inner workings - Jim: Prevent attach on PV when model != netfront - Joao: Remove now unused ioemu_nic variable - Joao: Append one small paragraph about why we mimic xl/libxl --- src/libxl/libxl_conf.c | 39 +++++++++++++++++++++++++++++++-------- src/libxl/libxl_conf.h | 3 ++- src/libxl/libxl_driver.c | 2 +- 3 files changed, 34 insertions(+), 10 deletions(-) diff --git a/src/libxl/libxl_conf.c b/src/libxl/libxl_conf.c index dcf8e7e..9515e2a 100644 --- a/src/libxl/libxl_conf.c +++ b/src/libxl/libxl_conf.c @@ -881,9 +881,9 @@ libxlMakeDiskList(virDomainDefPtr def, libxl_domain_config *d_config) int libxlMakeNic(virDomainDefPtr def, virDomainNetDefPtr l_nic, - libxl_device_nic *x_nic) + libxl_device_nic *x_nic, + bool attach) { - bool ioemu_nic = def->os.type == VIR_DOMAIN_OSTYPE_HVM; virDomainNetType actual_type = virDomainNetGetActualType(l_nic); virNetworkPtr network = NULL; virConnectPtr conn = NULL; @@ -907,16 +907,39 @@ libxlMakeNic(virDomainDefPtr def, virMacAddrGetRaw(&l_nic->mac, x_nic->mac); - if (ioemu_nic) - x_nic->nictype = LIBXL_NIC_TYPE_VIF_IOEMU; - else - x_nic->nictype = LIBXL_NIC_TYPE_VIF; - + /* + * The nictype field of libxl_device_nic structure tells Xen which type of + * NIC device to create for the domain. LIBXL_NIC_TYPE_VIF specifies a + * PV NIC. LIBXL_NIC_TYPE_VIF_IOEMU specifies a PV and emulated NIC, + * allowing the domain to choose which NIC to use and unplug the unused + * one. LIBXL_NIC_TYPE_VIF_IOEMU is only valid for HVM domains. Further, + * if hotplugging the NIC, emulated NICs are currently not supported. + * Alternatively one could set LIBXL_NIC_TYPE_UNKNOWN and let libxl decide, + * but its behaviour might not be consistent across all libvirt supported + * versions. The other nictype values are well established already, hence + * we manually select our own default and mimic xl/libxl behaviour starting + * xen commit 32e9d0f ("libxl: nic type defaults to vif in hotplug for + * hvm guest"). + */ if (l_nic->model) { + if (def->os.type == VIR_DOMAIN_OSTYPE_XEN && + STRNEQ(l_nic->model, "netfront")) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("only model 'netfront' is supported for " + "Xen PV domains")); + return -1; + } if (VIR_STRDUP(x_nic->model, l_nic->model) < 0) goto cleanup; if (STREQ(l_nic->model, "netfront")) x_nic->nictype = LIBXL_NIC_TYPE_VIF; + else + x_nic->nictype = LIBXL_NIC_TYPE_VIF_IOEMU; + } else { + if (def->os.type == VIR_DOMAIN_OSTYPE_HVM && !attach) + x_nic->nictype = LIBXL_NIC_TYPE_VIF_IOEMU; + else + x_nic->nictype = LIBXL_NIC_TYPE_VIF; } if (VIR_STRDUP(x_nic->ifname, l_nic->ifname) < 0) @@ -1047,7 +1070,7 @@ libxlMakeNicList(virDomainDefPtr def, libxl_domain_config *d_config) if (virDomainNetGetActualType(l_nics[i]) == VIR_DOMAIN_NET_TYPE_HOSTDEV) continue; - if (libxlMakeNic(def, l_nics[i], &x_nics[nvnics])) + if (libxlMakeNic(def, l_nics[i], &x_nics[nvnics], false)) goto error; /* * The devid (at least right now) will not get initialized by diff --git a/src/libxl/libxl_conf.h b/src/libxl/libxl_conf.h index 0ea76b4..851f3af 100644 --- a/src/libxl/libxl_conf.h +++ b/src/libxl/libxl_conf.h @@ -178,7 +178,8 @@ libxlMakeDisk(virDomainDiskDefPtr l_dev, libxl_device_disk *x_dev); int libxlMakeNic(virDomainDefPtr def, virDomainNetDefPtr l_nic, - libxl_device_nic *x_nic); + libxl_device_nic *x_nic, + bool attach); int libxlMakeVfb(virPortAllocatorPtr graphicsports, virDomainGraphicsDefPtr l_vfb, libxl_device_vfb *x_vfb); diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index b19b17e..7e5d9b6 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -3379,7 +3379,7 @@ libxlDomainAttachNetDevice(libxlDriverPrivatePtr driver, goto cleanup; } - if (libxlMakeNic(vm->def, net, &nic) < 0) + if (libxlMakeNic(vm->def, net, &nic, true) < 0) goto cleanup; if (libxl_device_nic_add(cfg->ctx, vm->def->id, &nic, 0)) { -- 2.1.4

2 2

[libvirt] [PATCH v2 00/10] admin: Introduce runtime logging APIs
by Erik Skultety 15 Dec '16

15 Dec '16

v1 here https://www.redhat.com/archives/libvir-list/2016-November/msg00009.html since v1: - incorporated notes raised during review - allowed passing of NULL via the public APIs * behaves the same way as an empty string, but lead to even more code reduction in 'daemonSetupLogging' for practically no extra code to enable such feature - added forgotten documentation - updated NEWS file Erik Skultety (10): virlog: Introduce virLog{Get,Set}DefaultOutput admin: Allow passing NULL to virLogSetFilters daemon: Hook up the virLog{Get,Set}DefaultOutput to the daemon's init routine admin: Introduce virAdmConnectGetLoggingOutputs admin: Introduce virAdmConnectGetLoggingFilters admin: Introduce virAdmConnectSetLoggingOutputs admin: Introduce virAdmConnectSetLoggingFilters virt-admin: Wire-up the logging APIs admin: Add an example demonstrating how to use the logging APIs admin: Update the news file to include the new logging features .gitignore | 1 + daemon/admin.c | 104 ++++++++++++++++++++++++++ daemon/libvirtd.c | 96 ++++-------------------- docs/news.html.in | 4 + examples/Makefile.am | 3 +- examples/admin/logging.c | 102 ++++++++++++++++++++++++++ include/libvirt/libvirt-admin.h | 16 ++++ src/admin/admin_protocol.x | 50 ++++++++++++- src/admin/admin_remote.c | 70 ++++++++++++++++++ src/admin_protocol-structs | 26 +++++++ src/libvirt-admin.c | 157 ++++++++++++++++++++++++++++++++++++++++ src/libvirt_admin_private.syms | 6 ++ src/libvirt_admin_public.syms | 8 ++ src/libvirt_private.syms | 2 + src/locking/lock_daemon.c | 90 ++++------------------- src/logging/log_daemon.c | 92 ++++------------------- src/util/virlog.c | 100 ++++++++++++++++++++++++- src/util/virlog.h | 4 +- tools/virt-admin.c | 120 ++++++++++++++++++++++++++++++ tools/virt-admin.pod | 90 +++++++++++++++++++++++ 20 files changed, 899 insertions(+), 242 deletions(-) create mode 100644 examples/admin/logging.c -- 2.5.5

2 29

[libvirt] [PATCH v2 00/21] Run qemu under its own namespace
by Michal Privoznik 15 Dec '16

15 Dec '16

v1 posted here: https://www.redhat.com/archives/libvir-list/2016-November/msg01208.html diff to v1: - I've dropped the patches for hugepages which are posted separately [1] - I've reworked some parts according to Dan's suggestions - Filled missing impl for virSCSIVHostDevice which was merged meanwhile Please note that patches 1-5, 7 were ACKed already. You can also find the patches on my github: https://github.com/zippy2/libvirt/tree/qemu_container_v3 (Yeah, _v2 branch is not good enough.) 1: https://www.redhat.com/archives/libvir-list/2016-November/msg01415.html Michal Privoznik (21): virprocess: Introduce virProcessSetupPrivateMountNS virfile: Introduce virFileSetupDev virfile: Introduce ACL helpers virusb: Introduce virUSBDeviceGetPath virscsi: Introduce virSCSIDeviceGetPath virscsivhost: Introduce virSCSIVHostDeviceGetPath qemu_cgroup: Expose defaultDeviceACL qemu: Spawn qemu under mount namespace qemu: Prepare disks when starting a domain qemu: Prepare hostdevs when starting a domain qemu: Prepare chardevs when starting a domain qemu: Prepare TPM when starting a domain qemu: Prepare inputs when starting a domain qemu: Prepare RNGs when starting a domain qemu: Enter the namespace on relabelling qemu: Manage /dev entry on disk hotplug qemu: Manage /dev entry on hostdev hotplug qemu: Manage /dev entry on chardev hotplug qemu: Manage /dev entry on RNG hotplug qemu: Let users opt-out from containerization qemu: Enable mount namespace config-post.h | 2 + configure.ac | 12 +- src/Makefile.am | 7 +- src/libvirt_private.syms | 10 + src/lxc/lxc_container.c | 20 +- src/lxc/lxc_controller.c | 32 +- src/qemu/libvirtd_qemu.aug | 1 + src/qemu/qemu.conf | 8 + src/qemu/qemu_cgroup.c | 2 +- src/qemu/qemu_cgroup.h | 1 + src/qemu/qemu_conf.c | 33 + src/qemu/qemu_conf.h | 2 + src/qemu/qemu_domain.c | 1244 +++++++++++++++++++++++++++++++++++- src/qemu/qemu_domain.h | 52 ++ src/qemu/qemu_driver.c | 5 +- src/qemu/qemu_hotplug.c | 90 ++- src/qemu/qemu_process.c | 28 +- src/qemu/qemu_security.c | 198 ++++++ src/qemu/qemu_security.h | 55 ++ src/qemu/test_libvirtd_qemu.aug.in | 3 + src/util/virfile.c | 153 +++++ src/util/virfile.h | 17 + src/util/virprocess.c | 38 ++ src/util/virprocess.h | 2 + src/util/virscsi.c | 6 + src/util/virscsi.h | 1 + src/util/virscsivhost.c | 7 + src/util/virscsivhost.h | 1 + src/util/virusb.c | 5 + src/util/virusb.h | 1 + 30 files changed, 1951 insertions(+), 85 deletions(-) create mode 100644 src/qemu/qemu_security.c create mode 100644 src/qemu/qemu_security.h -- 2.11.0

2 48

[libvirt] [PATCH 0/2] qemu: fix few image/resource locking problems
by Peter Krempa 15 Dec '16

15 Dec '16

The problem from patch 2/2 made me to look at the locking infrastructure in libvirt. There are a few broken places. This fixes two mistakes but more patches will come. Peter Krempa (2): qemu: monitor: Don't resume lockspaces in resume event handler qemu: snapshot: Don't attempt to resume cpus if they were not paused src/qemu/qemu_driver.c | 4 ++-- src/qemu/qemu_process.c | 11 ----------- 2 files changed, 2 insertions(+), 13 deletions(-) -- 2.11.0

2 3

[libvirt] [PATCH] virsh:remove the dead code
by sochin jiang 15 Dec '16

15 Dec '16

This patch remove the dead code I found. Signed-off-by: sochin jiang <sochin.jiang(a)huawei.com> --- tools/virsh-domain.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 3a6fa5c..6dae5cc 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -1005,7 +1005,6 @@ cmdAttachInterface(vshControl *ctl, const vshCmd *cmd) vshError(ctl, _("No support for %s in command 'attach-interface'"), type); goto cleanup; - break; } if (target != NULL) @@ -7365,7 +7364,6 @@ cmdCPUCompare(vshControl *ctl, const vshCmd *cmd) vshPrint(ctl, _("CPU described in %s is incompatible with host CPU\n"), from); goto cleanup; - break; case VIR_CPU_COMPARE_IDENTICAL: vshPrint(ctl, _("CPU described in %s is identical to host CPU\n"), -- 1.8.3.1

3 2

[libvirt] [PATCH 0/2] qemu_cgroup: allow access to /dev/dri/render*
by Ján Tomko 14 Dec '16

14 Dec '16

Technically a v2, but v1 is already pushed. This version is based on the <gl enable> in <spice> instead of accel3d="yes" in <video><model type="virtio". It also only allows access to the render* devices, instead of all of them. https://bugzilla.redhat.com/show_bug.cgi?id=1337290 Ján Tomko (2): Revert "qemu_cgroup: allow access to /dev/dri for virtio-vga" qemu_cgroup: allow access to /dev/dri/render* src/qemu/qemu_cgroup.c | 71 ++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 57 insertions(+), 14 deletions(-) -- 2.7.3

6 17

[libvirt] [PATCH v4 0/4] gluster: cache glfs connection object per volume
by Prasanna Kumar Kalever 14 Dec '16

14 Dec '16

v4: Address review comments from Peter on v3 Patch 1: changes to fix test failures (qemuargv2xmltest, qemuxml2argvtest, qemuxml2xmltest and virstoragetest) Patch 2: fix hung in qemuxml2argvtest Patch 3 v3: Split this Patch to 2 parts as below Patch 3 v4: don't free metadata used for storage driver access Patch 4: * update variable naming as per standards * virStorageBackendGlusterPreopened* -> virStorageBackendGlusterCache* Thanks Peter! v3: Address comments by Daniel and Peter on v2 * Split the patch to 3 parts Patch 1: change the virStorageNetHostDef type Patch 2: optimize calls to virStorageFileInit and friends Patch 3: add the caching for glfs * Thanks to Daniel, this version make all the methods as thread safe * Thanks to Peter for pointing use of virObjectLockable, this had simplified rest of the parts a lot. v2: Address review comments from Peter on v1 * Rebased on latest master * Changes to commit msg * Introduce storage API's for Register and Unregister of volume * During qemu process Start/Stop and snapshot create * Check Transport and Port type * Atomic element add/del to list and ref counting Pending: Treating IP and FQDN belong to same host v1: Initial patch Prasanna Kumar Kalever (4): util: change the virStorageNetHostDef type storage: optimize calls to virStorageFileInit and friends virStorageFileDeinit: don't free metadata used for storage driver access gluster: cache glfs connection object per volume src/conf/domain_conf.c | 52 +++--- src/qemu/qemu_command.c | 68 ++++---- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_domain.h | 5 + src/qemu/qemu_driver.c | 40 +++-- src/qemu/qemu_parse_command.c | 44 +++-- src/qemu/qemu_process.c | 47 ++++++ src/qemu/qemu_process.h | 4 + src/storage/storage_backend_fs.c | 2 + src/storage/storage_backend_gluster.c | 299 ++++++++++++++++++++++++++++++---- src/storage/storage_driver.c | 23 +-- src/util/virstoragefile.c | 85 +++++----- src/util/virstoragefile.h | 20 ++- tests/virstoragetest.c | 2 +- 14 files changed, 512 insertions(+), 181 deletions(-) -- 2.7.4

3 17

[libvirt] [PATCH v2 0/2] libxl QED support
by Cédric Bosdonnat 14 Dec '16

14 Dec '16

Hi all, Just added on patch before the QED one to add default: cases in xenParseXLDisk to avoid build failures when building on more uptodate xen. Cédric Bosdonnat (2): xenconfig: add default in xenParseXLDisk()'s switches libxl: add QED disk format support src/libxl/libxl_conf.c | 11 +++++++++++ src/xenconfig/xen_xl.c | 15 +++++++++++++++ 2 files changed, 26 insertions(+) -- 2.11.0

2 4

[libvirt] [PATCH] ZFS: Add vol-resize support
by Sergei A Mamonov 14 Dec '16

14 Dec '16

Add virStorageBackendZFSResizeVol method for storage_backend_zfs.c --- src/storage/storage_backend_zfs.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/src/storage/storage_backend_zfs.c b/src/storage/storage_backend_zfs.c index 3a43be4..8e57d72 100644 --- a/src/storage/storage_backend_zfs.c +++ b/src/storage/storage_backend_zfs.c @@ -454,6 +454,33 @@ virStorageBackendZFSDeletePool(virConnectPtr conn ATTRIBUTE_UNUSED, return ret; } +static int +virStorageBackendZFSResizeVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool, + virStorageVolDefPtr vol, + unsigned long long capacity, + unsigned int flags) +{ + int ret = -1; + virCommandPtr resize_cmd = NULL; + + virCheckFlags(VIR_STORAGE_VOL_RESIZE_SHRINK, -1); + + resize_cmd = virCommandNewArgList(ZFS, "set", NULL); + + virCommandAddArgFormat(resize_cmd, "volsize=%llu", capacity); + virCommandAddArgFormat(resize_cmd, "%s/%s", + pool->def->source.name, vol->name); + + if (virCommandRun(resize_cmd, NULL) < 0) + goto cleanup; + + ret = 0; + cleanup: + virCommandFree(resize_cmd); + return ret; +} + virStorageBackend virStorageBackendZFS = { .type = VIR_STORAGE_POOL_ZFS, @@ -465,4 +492,5 @@ virStorageBackend virStorageBackendZFS = { .deletePool = virStorageBackendZFSDeletePool, .uploadVol = virStorageBackendVolUploadLocal, .downloadVol = virStorageBackendVolDownloadLocal, + .resizeVol = virStorageBackendZFSResizeVol, }; -- 1.8.3.1

2 1

[libvirt] [PATCH v5 0/2] List only online cpus for vcpupin/emulatorpin when vcpu placement static
by Nitesh Konkar 14 Dec '16

14 Dec '16

Currently when the vcpu placement is static and cpuset is not specified, CPU Affinity shows 0.. CPUMAX. This patchset will result in display of only online CPU's under CPU Affinity on linux. Fixes the following Bug: virsh dumpxml Fedora <domain type='kvm' id='4'> <name>Fedora</name> <uuid>aecf3e5e-6f9a-42a3-9d6a-223a75569a66</uuid> <maxMemory slots='32' unit='KiB'>3145728</maxMemory> <memory unit='KiB'>524288</memory> <currentMemory unit='KiB'>524288</currentMemory> <vcpu placement='static' current='8'>160</vcpu> <resource> <partition>/machine</partition> </resource> ..................... ....................... ......................... <memballoon model='virtio'> <alias name='balloon0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/> </memballoon> </devices> <seclabel type='dynamic' model='dac' relabel='yes'> <label>+0:+0</label> <imagelabel>+0:+0</imagelabel> </seclabel> </domain> lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 8 On-line CPU(s) list: 0-2,4-7 Off-line CPU(s) list: 3 Thread(s) per core: 1 Core(s) per socket: 7 Socket(s): 1 .......... .......... NUMA node0 CPU(s): 0-2,4-7 NUMA node1 CPU(s): cat /sys/devices/system/cpu/online 0-2,4-7 Before Patch virsh vcpupin Fedora VCPU: CPU Affinity ---------------------------------- 0: 0-7 1: 0-7 ... ... 158: 0-7 159: 0-7 virsh emulatorpin Fedora emulator: CPU Affinity ---------------------------------- *: 0-7 After Patch virsh vcpupin Fedora VCPU: CPU Affinity ---------------------------------- 0: 0-2,4-7 1: 0-2,4-7 ... ... 158: 0-2,4-7 159: 0-2,4-7 virsh emulatorpin Fedora emulator: CPU Affinity ---------------------------------- *: 0-2,4-7 Nitesh Konkar (2): conf: List only online cpus for virsh vcpupin conf: List only online cpus for virsh emulatorpin src/conf/domain_conf.c | 6 ++++++ src/qemu/qemu_driver.c | 5 +++++ 2 files changed, 11 insertions(+) -- 2.1.0

3 5