January 2016 - Devel - libvirt List Archives

[libvirt] [libvirt-glib] events: Mark 'eventlock' as static
by Christophe Fergeau 21 Apr '16

21 Apr '16

It's not used outside of the libvirt-glib-event.c file, so there is no good reason for not having it static. As it was not listed in libvirt-glib.sym, this will make no change to the publicly exported symbols (ie this is not an ABI change). --- libvirt-glib/libvirt-glib-event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt-glib/libvirt-glib-event.c b/libvirt-glib/libvirt-glib-event.c index f8227d6..4548aa6 100644 --- a/libvirt-glib/libvirt-glib-event.c +++ b/libvirt-glib/libvirt-glib-event.c @@ -110,7 +110,7 @@ struct gvir_event_timeout virFreeCallback ff; }; -GMutex *eventlock = NULL; +static GMutex *eventlock = NULL; static int nextwatch = 1; static GPtrArray *handles; -- 2.5.0

1 1

[libvirt] [PATCH v2] add func to set shared drivers after libvirtd init
by Mikhail Feoktistov 08 Apr '16

08 Apr '16

Diff from v1: Remove vz prefix from the title of this letter. Because this is the common case for all drivers in libvirt. Description: Built-in drivers in libvirt are initialized before libvirtd initialization. Libvirt loads shared drivers on libvirtd initialization step. For built-in drivers we can't set shared drivers, because they are not initialized yet. This patch adds function to set shared drivers after libvirtd init. --- daemon/libvirtd.c | 4 ++++ src/libvirt.c | 41 +++++++++++++++++++++++++++++++++++++++++ src/libvirt_internal.h | 1 + src/libvirt_private.syms | 1 + 4 files changed, 47 insertions(+) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 250094b..aac1826 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -431,6 +431,10 @@ static void daemonInitialize(void) bhyveRegister(); # endif #endif +# ifdef WITH_VZ + virAssignSharedDrivers("vz"); + virAssignSharedDrivers("Parallels"); +# endif } diff --git a/src/libvirt.c b/src/libvirt.c index 25a0040..1763be7 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -1433,3 +1433,44 @@ virTypedParameterValidateSet(virConnectPtr conn, } return 0; } + +/** + * virAssignSharedDrivers: + * @name: name of connection driver + * + * This function fills in any empty pointers for shared drivers + * in connect driver structure + * + * Returns 0 in case of success, -1 in case of error +*/ +int +virAssignSharedDrivers(const char *name) +{ + size_t i; + + if (name == NULL) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Driver name must be specified")); + return -1; + } + + for (i = 0; i < virConnectDriverTabCount; i++) { + if (STREQ(virConnectDriverTab[i]->hypervisorDriver->name, name)) { + if (virConnectDriverTab[i]->interfaceDriver == NULL) + virConnectDriverTab[i]->interfaceDriver = virSharedInterfaceDriver; + if (virConnectDriverTab[i]->networkDriver == NULL) + virConnectDriverTab[i]->networkDriver = virSharedNetworkDriver; + if (virConnectDriverTab[i]->nodeDeviceDriver == NULL) + virConnectDriverTab[i]->nodeDeviceDriver = virSharedNodeDeviceDriver; + if (virConnectDriverTab[i]->nwfilterDriver == NULL) + virConnectDriverTab[i]->nwfilterDriver = virSharedNWFilterDriver; + if (virConnectDriverTab[i]->secretDriver == NULL) + virConnectDriverTab[i]->secretDriver = virSharedSecretDriver; + if (virConnectDriverTab[i]->storageDriver == NULL) + virConnectDriverTab[i]->storageDriver = virSharedStorageDriver; + break; + } + } + + return 0; +} diff --git a/src/libvirt_internal.h b/src/libvirt_internal.h index 1313b58..2a7227b 100644 --- a/src/libvirt_internal.h +++ b/src/libvirt_internal.h @@ -289,4 +289,5 @@ virTypedParameterValidateSet(virConnectPtr conn, virTypedParameterPtr params, int nparams); +int virAssignSharedDrivers(const char *name); #endif diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index a835f18..a0fcdf5 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -943,6 +943,7 @@ virFDStreamSetInternalCloseCb; # libvirt_internal.h +virAssignSharedDrivers; virConnectSupportsFeature; virDomainMigrateBegin3; virDomainMigrateBegin3Params; -- 1.8.3.1

2 3

[libvirt] Block replication driver
by Simon Kollberg 24 Mar '16

24 Mar '16

Hi! I'm working on supporting a new FT/HA solution for qemu called COLO (http://wiki.qemu.org/Features/COLO) The part that is currently being focused on for libvirt integration is Block Replication (http://wiki.qemu.org/Features/BlockReplication) which enables guest state synchronization for disks. Right now there are three issues that I'd like to get your input on: 1. As you can see on the block replication wiki-page we need to reference the secondary disk id. Example from the wiki: -drive if=none,driver=raw,file.filename=1.raw,id=colo1 \ -drive if=xxx,driver=replication,mode=secondary,\ ... file.backing.backing=colo1 My initial thought was to manually set the alias of the disk and add a new reference element to the backingStore: <disk type='file' device='disk'> ... <alias name='colo1'/> </disk> <disk type='file' device='disk'> ... <backingStore type='file'> ... <reference name='colo1'/> </backingStore> </disk> Though, I quickly realized that setting the alias is only done by the hypervisor and is therefore not an option with the current code. Would it be bad letting the user set the alias, and if so, do you have any ideas of how to solve the referencing? 2. The format of the disk and the driver type currently shares the same attribute in libvirt (the type attribute on driver XML element). However, with the new replication disk driver you need to be able to set both the disk format and also the driver name. Example from the wiki: -drive if=xxx,driver=replication,mode=secondary,\ file.file.filename=active_disk.qcow2,\ file.driver=qcow2,\ ... I saw that there was a function in libvirt called virStorageFileProbeFormat that could let us get the format of the disk without stating it in the XML. But as I'm sure you know, it's strongly advised not to be used since you can trick the function by modifying the disk file. 3. When using the replication driver the secondary disk is supposed to be added but not attached. Example from the wiki: -drive if=none,driver=raw,file.filename=1.raw,id=colo1 \ -drive if=xxx,driver=replication,mode=secondary,\ ... Clearly, trying to setup a disk without a target is not allowed at the moment. Is there any better way of doing it? Any comments and insights in general are also greatly appreciated! /Simon

4 5

[libvirt] [PATCH lxc_ethernet] lxc:support network type with ethernet
by zhongguocheng1＠163.com 24 Mar '16

24 Mar '16

From: z00209963 <z00209963(a)szxbz472.huaweiobz.com> --- src/lxc/lxc_process.c | 35 +++++++++++++++++++++++++++++++++++ src/lxc/lxc_process.h | 3 +++ 2 files changed, 38 insertions(+) diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index 57e3880..1ffd9ac 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -359,6 +359,38 @@ char *virLXCProcessSetupInterfaceDirect(virConnectPtr conn, return ret; } +char *virLXCProcessSetupInterfaceEthernet(virDomainDefPtr vm, + virDomainNetDefPtr net) +{ + char *ret = NULL; + char *parentVeth; + char *containerVeth = NULL; + + VIR_DEBUG("calling vethCreate()"); + parentVeth = net->ifname; + if (virNetDevVethCreate(&parentVeth, &containerVeth) < 0) + goto cleanup; + VIR_DEBUG("parentVeth: %s, containerVeth: %s", parentVeth, containerVeth); + + if (net->ifname == NULL) + net->ifname = parentVeth; + + if (virNetDevSetMAC(containerVeth, &net->mac) < 0) + goto cleanup; + + if (virNetDevSetOnline(parentVeth, true) < 0) + goto cleanup; + + if (net->filter && + virDomainConfNWFilterInstantiate(vm->uuid, net) < 0) + goto cleanup; + + ret = containerVeth; + + cleanup: + return ret; +} + static const char *nsInfoLocal[VIR_LXC_DOMAIN_NAMESPACE_LAST] = { [VIR_LXC_DOMAIN_NAMESPACE_SHARENET] = "net", [VIR_LXC_DOMAIN_NAMESPACE_SHAREIPC] = "ipc", @@ -559,6 +591,9 @@ static int virLXCProcessSetupInterfaces(virConnectPtr conn, break; case VIR_DOMAIN_NET_TYPE_ETHERNET: + if (!(veth = virLXCProcessSetupInterfaceEthernet(def, + net))) + goto cleanup; break; case VIR_DOMAIN_NET_TYPE_USER: diff --git a/src/lxc/lxc_process.h b/src/lxc/lxc_process.h index b6c8083..cc8e707 100644 --- a/src/lxc/lxc_process.h +++ b/src/lxc/lxc_process.h @@ -53,5 +53,8 @@ char *virLXCProcessSetupInterfaceBridged(virDomainDefPtr vm, char *virLXCProcessSetupInterfaceDirect(virConnectPtr conn, virDomainDefPtr def, virDomainNetDefPtr net); +char *virLXCProcessSetupInterfaceEthernet(virDomainDefPtr vm, + virDomainNetDefPtr net); + #endif /* __LXC_PROCESS_H__ */ -- 2.5.1.windows.1

4 4

[libvirt] Libvirt bite sized tasks
by Michal Privoznik 21 Mar '16

21 Mar '16

Dear lists, I've just started new wiki page which aim is to summarize small and trivial tasks, that starting contributors can take, investigate and implement. The aim is to give them something easy to start with while not scaring them out about complexity of our code. The page can be found here: http://wiki.libvirt.org/page/BiteSizedTasks The name is copied from qemu wiki: http://wiki.qemu.org/BiteSizedTasks Please, feel free to extend the list. I plan to use it in the future when interviewing GSoC candidates. Michal

3 4

[libvirt] [libvirt-glib v2 0/9] gconfig: Add hostdev support
by Christophe Fergeau 12 Mar '16

12 Mar '16

Hey, While reviewing Zeeshan's libvirt-gconfig hostdev series, I fixed up a few things, and added a few additional helpers/getters in order to simplify the code/make the test case more complete. This series contains this new code. The changes I made to Zeeshan's initial patches are in separate fixup! commits to make reviewing them easier (for you Zeeshan ;). I expect these to be squashed in the main commits before being pushed. Zeeshan's commits are unchanged from v1. Christophe

2 14

[libvirt] [PATCH 0/2] RFC: qemu: aarch64 virtio-pci work
by Cole Robinson 11 Mar '16

11 Mar '16

These patches apply on top of the series I just posted: [PATCH 0/9] tests: qemu: unconditionally enable QEMU_CAPS_DEVICE https://www.redhat.com/archives/libvir-list/2016-January/msg01233.html These specific changers were discussed here: https://www.redhat.com/archives/libvir-list/2015-December/msg00217.html We need a way for apps to tell libvirt 'use virtio-pci not virtio-mmio' for aarch64 -M virt. We can't really make use this by default yet since most OS don't actually support it. So we decided the best course of interaction is to allow users to pass a manual pci controller, and libvirt will interpret that to mean it's okay to use virtio-pci. Patch #1 undoes the current behavior of unconditionally adding a q35 style PCIe controller setup for mach-virt that supports it. In the future we will probably re-enable something like this, but maybe not for a year or so until the bits proliferate to distros. Patch #2 tries to implement the manual pci controller bit. Unfortunately it's not that easy get the ideal XML to work, so this isn't ready for committing. See the commit message for more details Cole Robinson (2): qemu: aarch64: Don't add PCIe controller by default HACK: qemu: aarch64: Use virtio-pci if user specifies PCI controller src/qemu/qemu_command.c | 34 ++++++++---- src/qemu/qemu_domain.c | 4 -- ...l2argv-aarch64-pci-manual-nocontroller-fail.xml | 43 ++++++++++++++++ .../qemuxml2argv-aarch64-virtio-pci-default.args | 2 - ...l2argv-aarch64-virtio-pci-manual-addresses.args | 2 - ...ml2argv-aarch64-virtio-pci-manual-addresses.xml | 1 + ...2argv-aarch64-virtio-pci-manual-controller.args | 36 +++++++++++++ ...l2argv-aarch64-virtio-pci-manual-controller.xml | 50 ++++++++++++++++++ tests/qemuxml2argvtest.c | 22 ++++++-- .../qemuxml2xmlout-aarch64-virtio-pci-default.xml | 10 ---- ...2xmlout-aarch64-virtio-pci-manual-addresses.xml | 11 +--- ...xmlout-aarch64-virtio-pci-manual-controller.xml | 60 ++++++++++++++++++++++ tests/qemuxml2xmltest.c | 6 +++ 13 files changed, 239 insertions(+), 42 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-pci-manual-nocontroller-fail.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-virtio-pci-manual-controller.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-virtio-pci-manual-controller.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-aarch64-virtio-pci-manual-controller.xml -- 2.5.0

4 15

[libvirt] [PATCH] ruby-libvirt: Don't crash in leases_wrap() by passing NULLs to rb_str_new2()
by Dan Williams 08 Mar '16

08 Mar '16

Not all lease values are mandatory, and when they aren't supplied by the libvirt driver they get set to NULL. That makes rb_str_new2() bail out. Signed-off-by: Dan Williams <dcbw(a)redhat.com> --- For example using the qemu driver we don't get 'iaid', and that makes ruby unhappy: [{"iface"=>"virbr1", "expirytime"=>1452189569, "type"=>0, "mac"=>"52:54:00:05:2b:6f", "ipaddr"=>"192.168.121.49", "prefix"=>24, "hostname"=>"openshiftdev", "clientid"=>"ff:00:05:2b:6f:00:01:00:01:1e:21:55:ed:52:54:00:05:2b:6f"}] ext/libvirt/network.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/ext/libvirt/network.c b/ext/libvirt/network.c index 7c77d73..c250d7d 100644 --- a/ext/libvirt/network.c +++ b/ext/libvirt/network.c @@ -269,14 +269,20 @@ static VALUE leases_wrap(VALUE arg) rb_hash_aset(hash, rb_str_new2("expirytime"), LL2NUM(lease->expirytime)); rb_hash_aset(hash, rb_str_new2("type"), INT2NUM(lease->type)); - rb_hash_aset(hash, rb_str_new2("mac"), rb_str_new2(lease->mac)); - rb_hash_aset(hash, rb_str_new2("iaid"), rb_str_new2(lease->iaid)); + if (lease->mac) + rb_hash_aset(hash, rb_str_new2("mac"), rb_str_new2(lease->mac)); + if (lease->iaid) + rb_hash_aset(hash, rb_str_new2("iaid"), rb_str_new2(lease->iaid)); rb_hash_aset(hash, rb_str_new2("ipaddr"), rb_str_new2(lease->ipaddr)); rb_hash_aset(hash, rb_str_new2("prefix"), UINT2NUM(lease->prefix)); - rb_hash_aset(hash, rb_str_new2("hostname"), - rb_str_new2(lease->hostname)); - rb_hash_aset(hash, rb_str_new2("clientid"), - rb_str_new2(lease->clientid)); + if (lease->hostname) { + rb_hash_aset(hash, rb_str_new2("hostname"), + rb_str_new2(lease->hostname)); + } + if (lease->clientid) { + rb_hash_aset(hash, rb_str_new2("clientid"), + rb_str_new2(lease->clientid)); + } rb_ary_store(result, i, hash); } -- 2.4.3

3 6

[libvirt] [FOR 1.3.0 PATCH] conf: add net device prefix for Xen
by Jim Fehlig 24 Feb '16

24 Feb '16

In commit d2e5538b1, the libxl driver was changed to copy interface names autogenerated by libxl to the corresponding network def in the domain's virDomainDef object. The copied name is freed when the domain transitions to the shutoff state. But when migrating a domain, the autogenerated name is included in the XML sent to the destination host. It is possible an interface with the same name already exists on the destination host, causing migration to fail. Indeed the Xen project's OSSTEST CI already encountered such a failure. This patch defines another VIR_NET_GENERATED_PREFIX for Xen, allowing the autogenerated names to be excluded when parsing and formatting inactive config. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- This is an alternative approach to Joao's fix for this regression https://www.redhat.com/archives/libvir-list/2015-December/msg00197.html I think it is the same approach used by the qemu driver. My only reservation is that it expands the potential for clashes with user-defined names. I.e. with this change both 'vnet' and 'vif' are reserved prefixes. src/conf/domain_conf.c | 6 ++++-- src/conf/domain_conf.h | 4 ++++ src/libxl/libxl_domain.c | 5 +++-- 3 files changed, 11 insertions(+), 4 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 2f5c0ed..debcf4e 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -8428,7 +8428,8 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, ifname = virXMLPropString(cur, "dev"); if (ifname && (flags & VIR_DOMAIN_DEF_PARSE_INACTIVE) && - STRPREFIX(ifname, VIR_NET_GENERATED_PREFIX)) { + (STRPREFIX(ifname, VIR_NET_GENERATED_PREFIX) || + STRPREFIX(ifname, VIR_NET_GENERATED_PREFIX_XEN))) { /* An auto-generated target name, blank it out */ VIR_FREE(ifname); } @@ -19790,7 +19791,8 @@ virDomainNetDefFormat(virBufferPtr buf, virBufferEscapeString(buf, "<backenddomain name='%s'/>\n", def->domain_name); if (def->ifname && !((flags & VIR_DOMAIN_DEF_FORMAT_INACTIVE) && - (STRPREFIX(def->ifname, VIR_NET_GENERATED_PREFIX)))) { + (STRPREFIX(def->ifname, VIR_NET_GENERATED_PREFIX) || + STRPREFIX(def->ifname, VIR_NET_GENERATED_PREFIX_XEN)))) { /* Skip auto-generated target names for inactive config. */ virBufferEscapeString(buf, "<target dev='%s'/>\n", def->ifname); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 90d8e13..d2cc26f 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1085,6 +1085,10 @@ struct _virDomainNetDef { * by libvirt, and cannot be used for a persistent network name. */ # define VIR_NET_GENERATED_PREFIX "vnet" +/* Used for prefix of ifname of any network name generated dynamically + * by libvirt for Xen, and cannot be used for a persistent network name. */ +# define VIR_NET_GENERATED_PREFIX_XEN "vif" + typedef enum { VIR_DOMAIN_CHR_DEVICE_STATE_DEFAULT = 0, VIR_DOMAIN_CHR_DEVICE_STATE_CONNECTED, diff --git a/src/libxl/libxl_domain.c b/src/libxl/libxl_domain.c index ef92974..c5d84a4 100644 --- a/src/libxl/libxl_domain.c +++ b/src/libxl/libxl_domain.c @@ -734,7 +734,7 @@ libxlDomainCleanup(libxlDriverPrivatePtr driver, for (i = 0; i < vm->def->nnets; i++) { virDomainNetDefPtr net = vm->def->nets[i]; - if (STRPREFIX(net->ifname, "vif")) + if (STRPREFIX(net->ifname, VIR_NET_GENERATED_PREFIX_XEN)) VIR_FREE(net->ifname); } } @@ -918,7 +918,8 @@ libxlDomainCreateIfaceNames(virDomainDefPtr def, libxl_domain_config *d_config) if (net->ifname) continue; - ignore_value(virAsprintf(&net->ifname, "vif%d.%d%s", + ignore_value(virAsprintf(&net->ifname, + VIR_NET_GENERATED_PREFIX_XEN "%d.%d%s", def->id, x_nic->devid, suffix)); } } -- 2.6.1

3 8

[libvirt] [PATCH v3 0/7] migration: add multithread compression
by Nikolay Shirokovskiy 24 Feb '16

24 Feb '16

Add options to turn multithread compression on during migration and to configure its parameters. Changes from v2: 1. Multiple field parameter is used to specify compression methods that should be used during migration. VIR_MIGRATE_COMPRESSED option flag now specifies that default compression methods should be used so we keep backward compatibility. This method is lxbzrle for qemu. 2. Patch that adds monitor commands to configure multithread compression parameters and patch that adds compression options to migrate command are reworked. This series has not patches to configure compression parameters before migration. Do we need it? Also it has not patch to introspect compression methods supported by driver as Jiri suggested because I want to publish this series ealier to check that I'm going in right direction. Eli Qiao (1): qemumonitorjsontest: add test for getting multithread compress params Nikolay Shirokovskiy (4): qemu migration: factor out setting migration option migration: add compress method option qemu migration: add multithread compression options qemu migration: add xbzrle compression options ShaoHe Feng (2): qemu monitor: add multithread compress parameters accessors virsh: add compression options for migration include/libvirt/libvirt-domain.h | 44 ++++++- src/qemu/qemu_domain.c | 4 + src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 116 +++++++++++++++++- src/qemu/qemu_migration.c | 256 +++++++++++++++++++-------------------- src/qemu/qemu_migration.h | 24 ++++ src/qemu/qemu_monitor.c | 31 ++++- src/qemu/qemu_monitor.h | 17 +++ src/qemu/qemu_monitor_json.c | 87 +++++++++++++ src/qemu/qemu_monitor_json.h | 5 + src/qemu/qemu_monitor_text.c | 100 +++++++++++++++ src/qemu/qemu_monitor_text.h | 5 + tests/qemumonitorjsontest.c | 53 ++++++++ tools/virsh-domain.c | 84 +++++++++++++ tools/virsh.pod | 25 +++- 15 files changed, 711 insertions(+), 142 deletions(-) -- 1.8.3.1

3 23