- Devel - libvirt List Archives

[libvirt] [PATCH v3 0/3] Honor blacklist for modprobe command
by John Ferlan 04 Feb '14

04 Feb '14

Rework changes from code review v2 here: http://www.redhat.com/archives/libvir-list/2014-January/msg01473.html v3 changes over v2: * Changed configure.ac for "RMMOD" definition * Use "virKMod*()" instead of "virModprobe*()" for function names * Adjust virKModLoad() to take boolean parameter for using -b or not * Add virKModIsBlacklisted() to perform the blacklist check * Adjust virkmodtest.c to use the virCommandSetDryRun() John Ferlan (3): utils: Introduce functions for modprobe tests: Add test for new virkmod functions Honor blacklist for modprobe command .gitignore | 2 + configure.ac | 6 ++ src/Makefile.am | 1 + src/libvirt_private.syms | 7 ++ src/util/virkmod.c | 182 ++++++++++++++++++++++++++++++++++++++++++++++ src/util/virkmod.h | 34 +++++++++ src/util/virpci.c | 28 +++++--- tests/Makefile.am | 5 ++ tests/virkmodtest.c | 183 +++++++++++++++++++++++++++++++++++++++++++++++ 9 files changed, 439 insertions(+), 9 deletions(-) create mode 100644 src/util/virkmod.c create mode 100644 src/util/virkmod.h create mode 100644 tests/virkmodtest.c -- 1.8.4.2

3 16

[libvirt] [PATCH] build: Fix 'make rpm' in VPATH with wireshark
by Martin Kletzander 04 Feb '14

04 Feb '14

Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- tools/wireshark/Makefile.am | 1 + tools/wireshark/src/Makefile.am | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/wireshark/Makefile.am b/tools/wireshark/Makefile.am index b6fa57c..28e6ed8 100644 --- a/tools/wireshark/Makefile.am +++ b/tools/wireshark/Makefile.am @@ -20,3 +20,4 @@ if WITH_WIRESHARK_DISSECTOR SUBDIRS = src endif WITH_WIRESHARK_DISSECTOR +EXTRA_DIST = util/genxdrstub.pl util/make-dissector-reg diff --git a/tools/wireshark/src/Makefile.am b/tools/wireshark/src/Makefile.am index afe9975..44f22be 100644 --- a/tools/wireshark/src/Makefile.am +++ b/tools/wireshark/src/Makefile.am @@ -21,7 +21,7 @@ INCLUDES = -I$(top_srcdir) ws_plugin_LTLIBRARIES = libvirt.la -libvirt_la_SOURCES = packet-libvirt.c plugin.c +libvirt_la_SOURCES = packet-libvirt.h packet-libvirt.c plugin.c libvirt_la_CPPFLAGS = $(WS_DISSECTOR_CPPFLAGS) libvirt_la_LDFLAGS = -avoid-version -module -- 1.8.5.3

4 4

[libvirt] [PATCH] event: improve public API docs
by Eric Blake 04 Feb '14

04 Feb '14

Since libvirt 0.9.3, the entire virevent.c file has been a public API, so improve the documentation in this file. Also, fix a potential core dump - it could only be triggered by bogus use of the API and would only affect the caller (not libvirtd), but we might as well be nice. * src/libvirt.c (virConnectDomainEventRegister) (virConnectDomainEventRegisterAny) (virConnectNetworkEventRegisterAny): Document event loop requirement. * src/util/virevent.c (virEventAddHandle, virEventRemoveHandle) (virEventAddTimeout, virEventRemoveTimeout): Likewise. (virEventUpdateHandle, virEventUpdateTimeout): Likewise, and avoid core dump if caller didn't register handler. (virEventRunDefaultImpl): Expand example, and set up code block in html docs. (virEventRegisterImpl, virEventRegisterDefaultImpl): Document more on the use of the event loop. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/libvirt.c | 24 ++++++++++------ src/util/virevent.c | 82 +++++++++++++++++++++++++++++++++++------------------ 2 files changed, 70 insertions(+), 36 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index 66841c8..f43718d 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -16121,11 +16121,13 @@ error: * @freecb: optional function to deallocate opaque when not used anymore * * Adds a callback to receive notifications of domain lifecycle events - * occurring on a connection + * occurring on a connection. This function requires that an event loop + * has been previously registered with virEventRegisterImpl() or + * virEventRegisterDefaultImpl(). * * Use of this method is no longer recommended. Instead applications * should try virConnectDomainEventRegisterAny() which has a more flexible - * API contract + * API contract. * * The virDomainPtr object handle passed into the callback upon delivery * of an event is only valid for the duration of execution of the callback. @@ -16134,7 +16136,7 @@ error: * The reference can be released once the object is no longer required * by calling virDomainFree. * - * Returns 0 on success, -1 on failure + * Returns 0 on success, -1 on failure. */ int virConnectDomainEventRegister(virConnectPtr conn, @@ -19064,10 +19066,12 @@ error: * @freecb: optional function to deallocate opaque when not used anymore * * Adds a callback to receive notifications of arbitrary domain events - * occurring on a domain. + * occurring on a domain. This function requires that an event loop + * has been previously registered with virEventRegisterImpl() or + * virEventRegisterDefaultImpl(). * * If @dom is NULL, then events will be monitored for any domain. If @dom - * is non-NULL, then only the specific domain will be monitored + * is non-NULL, then only the specific domain will be monitored. * * Most types of event have a callback providing a custom set of parameters * for the event. When registering an event, it is thus necessary to use @@ -19085,7 +19089,7 @@ error: * for the callback. To unregister a callback, this callback ID should * be passed to the virConnectDomainEventDeregisterAny() method. * - * Returns a callback identifier on success, -1 on failure + * Returns a callback identifier on success, -1 on failure. */ int virConnectDomainEventRegisterAny(virConnectPtr conn, @@ -19183,10 +19187,12 @@ error: * @freecb: optional function to deallocate opaque when not used anymore * * Adds a callback to receive notifications of arbitrary network events - * occurring on a network. + * occurring on a network. This function requires that an event loop + * has been previously registered with virEventRegisterImpl() or + * virEventRegisterDefaultImpl(). * * If @net is NULL, then events will be monitored for any network. If @net - * is non-NULL, then only the specific network will be monitored + * is non-NULL, then only the specific network will be monitored. * * Most types of event have a callback providing a custom set of parameters * for the event. When registering an event, it is thus necessary to use @@ -19204,7 +19210,7 @@ error: * for the callback. To unregister a callback, this callback ID should * be passed to the virConnectNetworkEventDeregisterAny() method. * - * Returns a callback identifier on success, -1 on failure + * Returns a callback identifier on success, -1 on failure. */ int virConnectNetworkEventRegisterAny(virConnectPtr conn, diff --git a/src/util/virevent.c b/src/util/virevent.c index fde29a2..0c94946 100644 --- a/src/util/virevent.c +++ b/src/util/virevent.c @@ -37,6 +37,16 @@ static virEventAddTimeoutFunc addTimeoutImpl = NULL; static virEventUpdateTimeoutFunc updateTimeoutImpl = NULL; static virEventRemoveTimeoutFunc removeTimeoutImpl = NULL; + +/***************************************************** + * + * Below this point are *PUBLIC* APIs for event + * loop integration with applications using libvirt. + * These API contracts cannot be changed. + * + *****************************************************/ + + /** * virEventAddHandle: * @@ -46,10 +56,12 @@ static virEventRemoveTimeoutFunc removeTimeoutImpl = NULL; * @opaque: user data to pass to callback * @ff: callback to free opaque when handle is removed * - * Register a callback for monitoring file handle events. + * Register a callback for monitoring file handle events. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * * Returns -1 if the file handle cannot be registered, otherwise a handle - * watch number to be used for updating and unregistering for events + * watch number to be used for updating and unregistering for events. */ int virEventAddHandle(int fd, @@ -70,14 +82,17 @@ virEventAddHandle(int fd, * @watch: watch whose file handle to update * @events: bitset of events to watch from virEventHandleType constants * - * Change event set for a monitored file handle. + * Change event set for a monitored file handle. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * - * Will not fail if fd exists + * Will not fail if fd exists. */ void virEventUpdateHandle(int watch, int events) { - updateHandleImpl(watch, events); + if (updateHandleImpl) + updateHandleImpl(watch, events); } /** @@ -85,7 +100,9 @@ virEventUpdateHandle(int watch, int events) * * @watch: watch whose file handle to remove * - * Unregister a callback from a file handle. + * Unregister a callback from a file handle. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * * Returns -1 if the file handle was not registered, 0 upon success. */ @@ -106,7 +123,9 @@ virEventRemoveHandle(int watch) * @opaque: user data to pass to callback * @ff: callback to free opaque when timeout is removed * - * Register a callback for a timer event. + * Register a callback for a timer event. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * * Setting timeout to -1 will disable the timer. Setting the timeout * to zero will cause it to fire on every event loop iteration. @@ -132,17 +151,20 @@ virEventAddTimeout(int timeout, * @timer: timer id to change * @timeout: time between events in milliseconds * - * Change frequency for a timer. + * Change frequency for a timer. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * * Setting frequency to -1 will disable the timer. Setting the frequency * to zero will cause it to fire on every event loop iteration. * - * Will not fail if timer exists + * Will not fail if timer exists. */ void virEventUpdateTimeout(int timer, int timeout) { - updateTimeoutImpl(timer, timeout); + if (updateTimeoutImpl) + updateTimeoutImpl(timer, timeout); } /** @@ -150,7 +172,9 @@ virEventUpdateTimeout(int timer, int timeout) * * @timer: the timer id to remove * - * Unregister a callback for a timer. + * Unregister a callback for a timer. This function + * requires that an event loop has previously been registered with + * virEventRegisterImpl() or virEventRegisterDefaultImpl(). * * Returns -1 if the timer was not registered, 0 upon success. */ @@ -164,14 +188,6 @@ virEventRemoveTimeout(int timer) } -/***************************************************** - * - * Below this point are 3 *PUBLIC* APIs for event - * loop integration with applications using libvirt. - * These API contracts cannot be changed. - * - *****************************************************/ - /** * virEventRegisterImpl: * @addHandle: the callback to add fd handles @@ -186,9 +202,14 @@ virEventRemoveTimeout(int timer) * to integrate with the libglib2 event loop, or libevent * or the QT event loop. * + * Use of the virEventAddHandle() and similar APIs require that the + * corresponding handler be registered. Use of the + * virConnectDomainEventRegisterAny() and similar APIs requires that + * the three timeout handlers be registered. + * * If an application does not need to integrate with an * existing event loop implementation, then the - * virEventRegisterDefaultImpl method can be used to setup + * virEventRegisterDefaultImpl() method can be used to setup * the generic libvirt implementation. */ void virEventRegisterImpl(virEventAddHandleFunc addHandle, @@ -220,9 +241,12 @@ void virEventRegisterImpl(virEventAddHandleFunc addHandle, * not have a need to integrate with an external event * loop impl. * - * Once registered, the application has to invoke virEventRunDefaultImpl in + * Once registered, the application has to invoke virEventRunDefaultImpl() in * a loop to process events. Failure to do so may result in connections being - * closed unexpectedly as a result of keepalive timeout. + * closed unexpectedly as a result of keepalive timeout. The default + * event loop fully supports handle and and timeout events, but only + * wakes up on events registered by libvirt API calls such as + * virEventAddHandle() or virConnectDomainEventRegisterAny(). * * Returns 0 on success, -1 on failure. */ @@ -255,14 +279,18 @@ int virEventRegisterDefaultImpl(void) * * Run one iteration of the event loop. Applications * will generally want to have a thread which invokes - * this method in an infinite loop + * this method in an infinite loop. Furthermore, it is wise + * to set up a pipe-to-self handler (via virEventAddHandle()) + * or a timeout (via virEventAddTimeout()) before calling this + * function, as it will block forever if there are no + * registered events. * - * static bool quit = false; + * static bool quit = false; * - * while (!quit) { - * if (virEventRunDefaultImpl() < 0) + * while (!quit) { + * if (virEventRunDefaultImpl() < 0) * ...print error... - * } + * } * * Returns 0 on success, -1 on failure. */ -- 1.8.4.2

4 8

[libvirt] [RFC PATCH] libvirt support to force convergence of live guest migration
by Chegu Vinod 04 Feb '14

04 Feb '14

Busy enterprise workloads hosted on large sized VM's tend to dirty memory faster than the transfer rate achieved via live guest migration. Despite some good recent improvements (& using dedicated 10Gig NICs between hosts) the live migration may NOT converge. Recently support was added in qemu (version 1.6) to allow a user to choose if they wish to force convergence of their migration via a new migration capability : "auto-converge". This feature allows for qemu to auto-detect lack of convergence and trigger a throttle-down of the VCPUs. This RFC patch includes the libvirt support needed to trigger this feature. (Testing is still in progress) Signed-off-by: Chegu Vinod <chegu_vinod(a)hp.com> --- include/libvirt/libvirt.h.in | 1 + src/qemu/qemu_migration.c | 44 ++++++++++++++++++++++++++++++++++++++++++ src/qemu/qemu_migration.h | 1 + src/qemu/qemu_monitor.c | 2 +- src/qemu/qemu_monitor.h | 1 + tools/virsh-domain.c | 7 ++++++ 6 files changed, 55 insertions(+), 1 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 146a59b..13b0bfc 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -1192,6 +1192,7 @@ typedef enum { VIR_MIGRATE_OFFLINE = (1 << 10), /* offline migrate */ VIR_MIGRATE_COMPRESSED = (1 << 11), /* compress data during migration */ VIR_MIGRATE_ABORT_ON_ERROR = (1 << 12), /* abort migration on I/O errors happened during migration */ + VIR_MIGRATE_AUTO_CONVERGE = (1 << 13), /* force auto-convergence during during migration */ } virDomainMigrateFlags; diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index e87ea85..8cc0c56 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1565,6 +1565,40 @@ cleanup: } static int +qemuMigrationSetAutoConverge(virQEMUDriverPtr driver, + virDomainObjPtr vm, + enum qemuDomainAsyncJob job) +{ + qemuDomainObjPrivatePtr priv = vm->privateData; + int ret; + + if (qemuDomainObjEnterMonitorAsync(driver, vm, job) < 0) + return -1; + + ret = qemuMonitorGetMigrationCapability( + priv->mon, + QEMU_MONITOR_MIGRATION_CAPS_AUTO_CONVERGE); + + if (ret < 0) { + goto cleanup; + } else if (ret == 0) { + virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, "%s", + _("Auto-Converge migration is not supported by " + "QEMU binary")); + ret = -1; + goto cleanup; + } + + ret = qemuMonitorSetMigrationCapability( + priv->mon, + QEMU_MONITOR_MIGRATION_CAPS_AUTO_CONVERGE); + +cleanup: + qemuDomainObjExitMonitor(driver, vm); + return ret; +} + +static int qemuMigrationWaitForSpice(virQEMUDriverPtr driver, virDomainObjPtr vm) { @@ -2389,6 +2423,11 @@ qemuMigrationPrepareAny(virQEMUDriverPtr driver, QEMU_ASYNC_JOB_MIGRATION_IN) < 0) goto stop; + if (flags & VIR_MIGRATE_AUTO_CONVERGE && + qemuMigrationSetAutoConverge(driver, vm, + QEMU_ASYNC_JOB_MIGRATION_IN) < 0) + goto stop; + if (mig->lockState) { VIR_DEBUG("Received lockstate %s", mig->lockState); VIR_FREE(priv->lockState); @@ -3181,6 +3220,11 @@ qemuMigrationRun(virQEMUDriverPtr driver, QEMU_ASYNC_JOB_MIGRATION_OUT) < 0) goto cleanup; + if (flags & VIR_MIGRATE_AUTO_CONVERGE && + qemuMigrationSetAutoConverge(driver, vm, + QEMU_ASYNC_JOB_MIGRATION_OUT) < 0) + goto cleanup; + if (qemuDomainObjEnterMonitorAsync(driver, vm, QEMU_ASYNC_JOB_MIGRATION_OUT) < 0) goto cleanup; diff --git a/src/qemu/qemu_migration.h b/src/qemu/qemu_migration.h index cafa2a2..c4258a1 100644 --- a/src/qemu/qemu_migration.h +++ b/src/qemu/qemu_migration.h @@ -39,6 +39,7 @@ VIR_MIGRATE_UNSAFE | \ VIR_MIGRATE_OFFLINE | \ VIR_MIGRATE_COMPRESSED | \ + VIR_MIGRATE_AUTO_CONVERGE | \ VIR_MIGRATE_ABORT_ON_ERROR) /* All supported migration parameters and their types. */ diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 1514715..780a29a 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -118,7 +118,7 @@ VIR_ENUM_IMPL(qemuMonitorMigrationStatus, VIR_ENUM_IMPL(qemuMonitorMigrationCaps, QEMU_MONITOR_MIGRATION_CAPS_LAST, - "xbzrle") + "xbzrle", "auto-converge") VIR_ENUM_IMPL(qemuMonitorVMStatus, QEMU_MONITOR_VM_STATUS_LAST, diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index eabf000..95e70ab 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -440,6 +440,7 @@ int qemuMonitorGetSpiceMigrationStatus(qemuMonitorPtr mon, typedef enum { QEMU_MONITOR_MIGRATION_CAPS_XBZRLE, + QEMU_MONITOR_MIGRATION_CAPS_AUTO_CONVERGE, QEMU_MONITOR_MIGRATION_CAPS_LAST } qemuMonitorMigrationCaps; diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 1fe138c..d94a81b 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -8532,6 +8532,10 @@ static const vshCmdOptDef opts_migrate[] = { .type = VSH_OT_BOOL, .help = N_("compress repeated pages during live migration") }, + {.name = "auto-converge", + .type = VSH_OT_BOOL, + .help = N_("force auto convergence during live migration") + }, {.name = "abort-on-error", .type = VSH_OT_BOOL, .help = N_("abort on soft errors during migration") @@ -8676,6 +8680,9 @@ doMigrate(void *opaque) if (vshCommandOptBool(cmd, "compressed")) flags |= VIR_MIGRATE_COMPRESSED; + if (vshCommandOptBool(cmd, "auto-converge")) + flags |= VIR_MIGRATE_AUTO_CONVERGE; + if (vshCommandOptBool(cmd, "offline")) { flags |= VIR_MIGRATE_OFFLINE; } -- 1.7.1

4 7

[libvirt] [PATCH v1 0/2] Network hooks
by Michal Privoznik 04 Feb '14

04 Feb '14

With the latest noise I made on the list with QoS I'm coming forward with this idea. Network hooks can fill some holes in our functionality. For instance if somebody is willing to set a QoS that is not covered by libvirt, or just want to insert some additional iptables rules, etc. You name it. Michal Privoznik (2): networkStartNetwork: Be more verbose network: Introduce start and shutdown hooks docs/hooks.html.in | 43 ++++++++++++++++++++---------- src/network/bridge_driver.c | 64 +++++++++++++++++++++++++++++++++------------ src/util/virhook.c | 10 ++++++- src/util/virhook.h | 8 ++++++ 4 files changed, 94 insertions(+), 31 deletions(-) -- 1.8.5.2

3 7

[libvirt] [PATCH v2 00/18] LXC configuration conversion to domain XML
by Cédric Bosdonnat 04 Feb '14

04 Feb '14

This new version of the patchset, fixes all comments from Daniel. It also adds two more commits: one extending virConf to use it later in lxc_native.c and one for some documentation on the new feature and its limitations. Several RFEs have been filesd for this patchset: they are mentioned in the appropriate commit log, and I still hope to be able to provide patches for them later. Cédric Bosdonnat (18): Improve virConf parse to handle LXC config format LXC driver: started implementing connectDomainXMLFromNative LXC from native: import rootfs LXC from native: migrate fstab and lxc.mount.entry LXC from native: implement no network conversion LXC from native: migrate veth network configuration LXC from native: convert phys network types to net hostdev devices LXC from native: convert lxc.tty to console devices LXC from native: convert macvlan network configuration LXC from native: convert lxc.id_map into <idmap> LXC from native: migrate memory tuning LXC from native: map lxc.cgroup.cpu.* LXC from native: map lxc.cgroup.cpuset.* LXC from native: add lxc.cgroup.blkio.* mapping LXC from native: map lxc.arch to /domain/os/type@arch LXC from native: map block filesystems LXC from native: map vlan network type LXC: added some doc on domxml-from-native with mention of limitations .gitignore | 1 + docs/drvlxc.html.in | 34 +- po/POTFILES.in | 1 + src/Makefile.am | 1 + src/libvirt_private.syms | 1 + src/lxc/lxc_container.c | 2 +- src/lxc/lxc_container.h | 2 + src/lxc/lxc_driver.c | 31 + src/lxc/lxc_native.c | 883 +++++++++++++++++++++ src/lxc/lxc_native.h | 32 + src/util/virconf.c | 46 +- src/util/virconf.h | 10 + tests/Makefile.am | 7 +- tests/lxcconf2xmldata/lxcconf2xml-blkiotune.config | 7 + tests/lxcconf2xmldata/lxcconf2xml-blkiotune.xml | 35 + .../lxcconf2xmldata/lxcconf2xml-cpusettune.config | 6 + tests/lxcconf2xmldata/lxcconf2xml-cpusettune.xml | 27 + tests/lxcconf2xmldata/lxcconf2xml-cputune.config | 7 + tests/lxcconf2xmldata/lxcconf2xml-cputune.xml | 29 + tests/lxcconf2xmldata/lxcconf2xml-fstab.config | 37 + tests/lxcconf2xmldata/lxcconf2xml-idmap.config | 5 + tests/lxcconf2xmldata/lxcconf2xml-idmap.xml | 28 + .../lxcconf2xml-macvlannetwork.config | 13 + .../lxcconf2xmldata/lxcconf2xml-macvlannetwork.xml | 26 + tests/lxcconf2xmldata/lxcconf2xml-memtune.config | 10 + tests/lxcconf2xmldata/lxcconf2xml-memtune.xml | 29 + .../lxcconf2xmldata/lxcconf2xml-nonenetwork.config | 4 + tests/lxcconf2xmldata/lxcconf2xml-nonenetwork.xml | 21 + tests/lxcconf2xmldata/lxcconf2xml-nonetwork.config | 3 + tests/lxcconf2xmldata/lxcconf2xml-nonetwork.xml | 24 + .../lxcconf2xmldata/lxcconf2xml-physnetwork.config | 6 + tests/lxcconf2xmldata/lxcconf2xml-physnetwork.xml | 26 + tests/lxcconf2xmldata/lxcconf2xml-simple.config | 41 + tests/lxcconf2xmldata/lxcconf2xml-simple.xml | 41 + .../lxcconf2xmldata/lxcconf2xml-vlannetwork.config | 12 + tests/lxcconf2xmldata/lxcconf2xml-vlannetwork.xml | 26 + tests/lxcconf2xmltest.c | 131 +++ 37 files changed, 1640 insertions(+), 5 deletions(-) create mode 100644 src/lxc/lxc_native.c create mode 100644 src/lxc/lxc_native.h create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-blkiotune.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-blkiotune.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-cpusettune.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-cpusettune.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-cputune.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-cputune.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-fstab.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-idmap.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-idmap.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-macvlannetwork.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-macvlannetwork.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-memtune.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-memtune.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-nonenetwork.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-nonenetwork.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-nonetwork.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-nonetwork.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-physnetwork.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-physnetwork.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-simple.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-simple.xml create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-vlannetwork.config create mode 100644 tests/lxcconf2xmldata/lxcconf2xml-vlannetwork.xml create mode 100644 tests/lxcconf2xmltest.c -- 1.8.5.2

3 38

[libvirt] [libvirt-glib 1/5] gconfig: Fix gvir_config_domain_os_get_boot_devices() API doc
by Christophe Fergeau 04 Feb '14

04 Feb '14

The elements of the returned list are integer enum values, so they cannot be unreffed. --- libvirt-gconfig/libvirt-gconfig-domain-os.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/libvirt-gconfig/libvirt-gconfig-domain-os.c b/libvirt-gconfig/libvirt-gconfig-domain-os.c index 03c8a85..b922a0e 100644 --- a/libvirt-gconfig/libvirt-gconfig-domain-os.c +++ b/libvirt-gconfig/libvirt-gconfig-domain-os.c @@ -267,8 +267,7 @@ static gboolean add_boot_device(xmlNodePtr node, gpointer opaque) * @os: a #GVirConfigDomainOs * * Gets the list of devices attached to @os. The returned list should be - * freed with g_list_free(), after its elements have been unreffed with - * g_object_unref(). + * freed with g_list_free(). * * Returns: (element-type LibvirtGConfig.DomainOsBootDevice) (transfer full): * a newly allocated #GList of #GVirConfigDomainOsBootDevice. -- 1.8.4.2

2 12

[libvirt] [PATCH] The numa node numbers can be non-sequential.
by Shivaprasad G Bhat 04 Feb '14

04 Feb '14

On PowerNV platform the numa node numbers can be non-sequential, #numactl --hardware node distances: node 0 1 16 17 0: 10 40 40 40 1: 40 10 40 40 16: 40 40 10 40 17: 40 40 40 10 The numa node number is used as index and vice versa which can cause the libvirt to crash. Signed-off-by: Shivaprasad G Bhat <sbhat(a)linux.vnet.ibm.com> Signed-off-by: Pradipta Kumar Banerjee <bpradip(a)in.ibm.com> --- src/conf/capabilities.c | 12 ++++++++++-- src/qemu/qemu_driver.c | 5 +++-- src/qemu/qemu_process.c | 2 +- 3 files changed, 14 insertions(+), 5 deletions(-) diff --git a/src/conf/capabilities.c b/src/conf/capabilities.c index 726ec3f..78f65cb 100644 --- a/src/conf/capabilities.c +++ b/src/conf/capabilities.c @@ -1000,10 +1000,18 @@ virCapabilitiesGetCpusForNode(virCapsPtr caps, size_t node, virBitmapPtr cpumask) { - virCapsHostNUMACellPtr cell = caps->host.numaCell[node]; + virCapsHostNUMACellPtr cell = NULL; size_t cpu; + size_t index; + /* The numa node numbers can be non-contiguous. Ex: 0,1,16,17. */ + for (index = 0; index < caps->host.nnumaCell; index++) { + if (caps->host.numaCell[index]->num == node) { + cell = caps->host.numaCell[index]; + break; + } + } - for (cpu = 0; cpu < cell->ncpus; cpu++) { + for (cpu = 0; cell && cpu < cell->ncpus; cpu++) { if (virBitmapSetBit(cpumask, cell->cpus[cpu].id) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Cpu '%u' in node '%zu' is out of range " diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 6b3825a..ae36fbe 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8532,12 +8532,13 @@ qemuDomainSetNumaParamsLive(virDomainObjPtr vm, for (i = 0; i < caps->host.nnumaCell; i++) { bool result; - if (virBitmapGetBit(nodeset, i, &result) < 0) { + virCapsHostNUMACellPtr cell = caps->host.numaCell[i]; + if (virBitmapGetBit(nodeset, cell->num, &result) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to get cpuset bit values")); goto cleanup; } - if (result && (virBitmapSetBit(temp_nodeset, i) < 0)) { + if (result && (virBitmapSetBit(temp_nodeset, cell->num) < 0)) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to set temporary cpuset bit values")); goto cleanup; diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 8bcd98e..c28f84d 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -2002,7 +2002,7 @@ qemuPrepareCpumap(virQEMUDriverPtr driver, size_t j; int cur_ncpus = caps->host.numaCell[i]->ncpus; bool result; - if (virBitmapGetBit(nodemask, i, &result) < 0) { + if (virBitmapGetBit(nodemask, caps->host.numaCell[i]->num, &result) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to convert nodeset to cpuset")); virBitmapFree(cpumap);

2 1

[libvirt] [PATCH] qemu: be sure we're using the updated value of backend during hotplug
by Laine Stump 04 Feb '14

04 Feb '14

Probable fix for https://bugzilla.redhat.com/show_bug.cgi?id=1056360 commit f094aaac changed qemuPrepareHostdevPCIDevices() such that it may modify the "backend" (vfio vs. legacy kvm) setting in the virHostdevDef. However, qemuDomainAttachHostPciDevice() (used by hotplug) copies the backend setting into a local *before* calling qemuPrepareHostdevPCIDevices(), and then later makes a decision based on that pre-change value. The result is that, if the backend had been set to "default" (i.e. not specified in the config) and was later updated to "VFIO" by qemuPrepareHostdevPCIDevices(), the qemu process' MacMemLock is not increased (as is required for VFIO device assignment). This patch delays making the local copy of backend until after its potential modification. --- src/qemu/qemu_hotplug.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 4315df2..4a885d2 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1,7 +1,7 @@ /* * qemu_hotplug.h: QEMU device hotplug management * - * Copyright (C) 2006-2013 Red Hat, Inc. + * Copyright (C) 2006-2014 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1152,7 +1152,7 @@ qemuDomainAttachHostPciDevice(virQEMUDriverPtr driver, bool releaseaddr = false; bool teardowncgroup = false; bool teardownlabel = false; - int backend = hostdev->source.subsys.u.pci.backend; + int backend; unsigned long long memKB; if (VIR_REALLOC_N(vm->def->hostdevs, vm->def->nhostdevs + 1) < 0) @@ -1162,6 +1162,9 @@ qemuDomainAttachHostPciDevice(virQEMUDriverPtr driver, &hostdev, 1, priv->qemuCaps) < 0) return -1; + /* this could have been changed by qemuPrepareHostdevPCIDevices */ + backend = hostdev->source.subsys.u.pci.backend; + switch ((virDomainHostdevSubsysPciBackendType) backend) { case VIR_DOMAIN_HOSTDEV_PCI_BACKEND_VFIO: if (!virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_DEVICE_VFIO_PCI)) { -- 1.8.5.3

2 1

[libvirt] [PATCH] Write up the project governance process
by Daniel P. Berrange 04 Feb '14

04 Feb '14

The project has historically operated as a meritocratic consensus based community. Formally document what has always been an unwritten assumption amongst the community participants. Also include an explicit code of conduct to prempt any potential, but unlikely, future problems. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- docs/governance.html.in | 292 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 292 insertions(+) create mode 100644 docs/governance.html.in At FOSDEM this past weekend I was asked what the libvirt governance process was. While I believe our community members already understand all this, and it can be infered from behaviour on lists, it will help future new contributors to understand how we operate if we actually document it. This is likely to be particularly helpful for other companies wondering how to get involved in the libvirt project. diff --git a/docs/governance.html.in b/docs/governance.html.in new file mode 100644 index 0000000..8bc4e51 --- /dev/null +++ b/docs/governance.html.in @@ -0,0 +1,292 @@ +<?xml version="1.0" encoding="UTF-8"?> +<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> +<html xmlns="http://www.w3.org/1999/xhtml"> + <body> + <h1>Project governance</h1> + + <ul id="toc"></ul> + + <p> + The libvirt project operates as a meritocratic, consensus-based community. + Anyone with an interest in the project can join the community, contributing + to the ongoing development of the project's work. This pages describes how + that participation takes place and how contributors earn merit, and thus + influence, within the community. + </p> + + <h2><a name="codeofconduct">Code of conduct</a></h2> + + <p> + The libvirt project community covers people from a wide variety of + countries, backgrounds and positions. This global diversity is a great + strength of the project, but can also lead to communication issues, + which may in turn cause unhappiness. To maximise happiness of the + project community taken as a whole, all members (whether users, + contributors or committers) are expected to abide by the project's + code of conduct. At a high level the code can be summarized as + <em>"be excellant to each other"</em>. Expanding on this + </p> + + <ul> + <li><strong>Be respectful:</strong> disagreements between people are to + be expected and are usually the sign of healthy debate and engagement. + Disagreements can lead to frustration and even anger for some members. + Turning to personal insults, intimidation or threatening behaviour does + not improve the situation though. Participants should thus take care to + ensure all communications / interactions stay professional at all times.</li> + + <li><strong>Be considerate:</strong> remember that the community has members + with a diverse background many of whom have English as a second language. + What might appear impolite, may simply be a result of a lack of knowledge + of the English language. Bear in mind that actions will have an impact + on other community members and the project as a whole, so take potential + consequences into account before persuing a course of action.</li> + + <li><strong>Be forgiving:</strong> humans are fallible and as such prone + to make mistakes and inexplicably change their positions at times. Don't + assume that other members are acting with malicious intent. Be prepared + to forgive people who make mistakes and assist each other in learning + from them. Playing a blame game doesn't help anyone.</li> + </ul> + + <h2><a name="roles">Roles and responsibilities</a></h2> + + <h3><a href="users">Users</a></h3> + + <p> + The users are anyone who has a need for the output of the project. + There are no rules or requirements to become a user of libvirt. Even + if the software does not yet work on their OS platform, a person can + be considered a potential future user and welcomed to participate. + </p> + + <p> + Participation by users is key to ensuring the project moves in the + right direction, satisfying their real world needs. Users are + encouraged in participate in the broader libvirt community in any + number of ways: + </p> + + <ul> + <li>Evangelism: spread the word about what libvirt is doing, how it + helps solve your problems. This can be via blog articles, social + media postings, video blogs, user group / conference presentations + and any other method of diseminating information</li> + <li>Feedback: let the developers know about what does and does not + work with the project. Talk to developers on the project's + IRC channel and mailing list, or find them at conferences. Tell + them what gaps the project has or where they should look for + future development</li> + <li>Moral support: developers live for recognition of the positive + impact their work has on users' lives. Give thanks to the developers + when evangelising the project, or when meeting them at user groups, + conferences, etc.</li> + </ul> + + <p> + The above is not an exhaustive list of things users can do to + participate in the project. Further ideas and suggestions are + welcome. Users are encouraged to take their participation + further and become contributors to the project in any of the + ways listed in the next section. + </p> + + <h3><a name="contributors">Contributors</a></h3> + + <p> + The contributors are community members who have some concrete impact + to the ongoing development of the project. There are many ways in which + members can contribute, with no requirement to be a software engineer. + Many users can in fact consider themselves contributors merely by + engaging in evangelism for the project. + </p> + + <ul> + <li>Bug reporting: improve the quality of the project by reporting + any problems found either to the project's own bug tracker, or to + that of the OS vendor shipping the libvirt code.</li> + <li>User help: join the IRC channel or mailing list to assist or advice + other users in troubleshooting the problems they face.</li> + <li>Feature requests: help set the direction for future work by + reporting details of features which are missing to the project's + own bug tracker or mailing lists.</li> + <li>Graphical design: contribute to the development of the project's + websites / wiki brand with improved graphics, styling or layout.</li> + <li>Code development: write and submit patches to address bugs or implement + new features</li> + <li>Architectural design: improve the usefulness of the project + by providing feedback on the design of proposed features, to + ensure they satisfy the broadest applicable needs and an survive + the long term</li> + <li>Code review: look at patches which are submitted and critique + the code to identify bugs, potential design problems or other + issues which should be addressed before the code is accepted</li> + <li>Documentation: contribute to content on personal blogs, the + website, wiki, code comments, or any of the formal documentation + efforts.</li> + <li>Translation: join the Fedora transifex community to improve the + quality of translations needed by the libvirt project.</li> + </ul> + + <p> + The above is not an exhaustive list of things members can do to + contribute to the project. Further ideas and suggestions are + welcome. + </p> + + <p> + There are no special requirements to becoming a contributor other + than having the interest and ability to provide a contribution. The + libvirt project <strong>does not require</strong> any + <em>"Contributor License Agreement"</em> + to be signed prior to engagement with the community. + </p> + + <p> + In making a contribution to the project, the community member is + implicitly stating that they accept the terms of the license under + which the work they are contributing to is distributed. They are + also implicitly stating that they have the legal right to make the + contribution, if doing so on behalf of a broader organization / + company. Most of the project's code is distributed under the GNU + Lesser General Public License, version 2 or later. Details of the + exact license under which contributions will be presumed to be + covered are found in the source repositories, or website in question. + </p> + + <h3><a name="committers">Committers</a></h3> + + <p> + The committers are the subset of contributors who have direct access + to commit code to the project's primary source code repositories, which + are currently using the GIT software. The committers are chosen based + on the quality of their contributions over a period of time. This includes + both the quality of code they submit, as well as the quality of reviews + they provide on other contributors' submissions and a demonstration that + they understand day-to-day operation of the projet and its goals. There + is no minimum level of contribution required in order to become a committer, + though 2-3 months worth of quality contribution would be a rough guide. + </p> + + <p> + There are no special requirements to becoming a committer other than to + have shown a willingness and ability to contribute to the project over + an extended period of time. Proposals for elevating contributors to + committers are typically made by existing comitters, though contributors + are also welcome to make proposals. The decision to approve the elevation + of a contributor to a committer is made through "rough consensus" between + the existing committers. + </p> + + <p> + The aim in elevating contributors to committers is to ensure that there + is a broad base of experience and expertize across all areas of the + project's work. Committers are not required to have knowledge across + all areas of the project's work. While an approved committer has the + technical ability to commit code to any area of the project, by convention + they will only commit to areas they feel themselves to be qualified to + evaluate the contribution. If in doubt, committers will defer to the + opinion of other committers with greater expertize in an area. + </p> + + <p> + The committers hold the ultimate control over what contributions are + accepted by the project, however, this does not mean they have the + right to do whatever they want. Where there is debate and disagreement + between contributors, committers are expected to look at the issues with + an unbiased point of view and help achieve a "rough consensus". If the + committer has a conflict of interest in the discussion, for example due + to their position of employment, they are expected to put the needs of + the community project first. If they cannot put the community project + first, they must declare their conflict of interest, and allow other + non-conflicted committers to make any final decision. + </p> + + <p> + The committers are expected to monitor contributions to areas of the + project where they have expertize and ensure that either some form of + feedback is provided to the contributor, or to accept their contribution. + There is no formal minimum level of approval required to accept a + contribution. Positive review by any committer experienced in the area + of work is considered to be enough to justify acceptance in normal + circumstances. Where one committer explicitly rejects a contribution, + however, other committers should not override that rejection without + first establishing a "rough consensus" amongst the broader group of + committers. + </p> + + <p> + Being a committer is a privilege, not a right. In exceptional + circumstances, the privilege may be removed from an active + contributor. Such decisions will be taken based on "rough + consensus" amongst other committers. In the event that a committer + is no longer able to participate in the project, after some period + of inactivity passes, they may be asked to confirm that they wish + to retain their rights as a committer. + </p> + + <h3><a name="secteam">Security team</a></h3> + + <p> + The security team consists of a subset of the project committers + along with representatives from vendors shipping the project's + software. The subset of project committers is chosen to be the + minimal size neccessary to provide expertise spanning most of + the project's work. Further project committers may be requested + to engage in resolving specific security issues on a case by + case basis. Any vendor who is shipping the project's software + may submit a request for one or more of their representatives + to join the security team. Such requests must by approved by + existing members of the team vouching for the integrity of + the nominated person or organization. + </p> + + <p> + Members of the security team are responsible for triaging and + resolving any security issues that are reported to the project. + They are expected to abide by the project's documented + <a href="securityprocess.html">security process</a>. In particular + they must respect any embargo period agreed amongst the team + before disclosing a private issue. + </p> + + <h2><a name="roughconsensus">Rough consensus</a></h2> + + <p> + A core concept for governance of the project described above is + that of "rough consensus". To expand on this, it is a process + of decision making that involves the following steps + </p> + + <ul> + <li>Proposal</li> + <li>Discussion</li> + <li>Vote (exceptional circumstances only)</li> + <li>Decision</li> + </ul> + + <p> + To put this into words, any contributor is welcome to make a proposal + for consideration. Any contributor may participate in the discussions + around the proposal. The discussion will usually result in agreement + between the interested parties, or at least agreement between the + committers. Only in the very exceptional circumstance where there + is disagreement between committers, would a vote be considered. + Even in these exceptional circumstances, it is usually found to be + obvious what the majority opinion of the committers is. In the event + that even a formal vote is be tied, the committers will have to hold + ongoing discussions until the stalemate is resolved or the proposal + withdrawn. + </p> + + <p> + The overall goal of the "rough consensus" process is to ensure that + decisions can be made within the project, with a minimum level of + bureaucracy and process. Implicit in this is that any person who does + not explicitly reject to a proposal is assumed to be supportive, or + at least agnostic. + </p> + + + </body> +</html> -- 1.8.5.3

4 5

[libvirt] [PATCH 00/15] Introduce virDomain{Resume,Suspend}Flags
by Michal Privoznik 04 Feb '14

04 Feb '14

So far, nothing interesting is happening here. The real work is yet to come :) Michal Privoznik (15): Introduce virDomainSuspendFlags Introduce virDomainResumeFlags esxDriver: Adapt to virDomain{Resume,Suspend}Flags hypervDriver: Adapt to virDomain{Resume,Suspend}Flags libxlDriver: Adapt to virDomain{Resume,Suspend}Flags lxcDriver: Adapt to virDomain{Resume,Suspend}Flags openvzDriver: Adapt to virDomain{Resume,Suspend}Flags parallelsDriver: Adapt to virDomain{Resume,Suspend}Flags qemuDriver: Adapt to virDomain{Resume,Suspend}Flags testDriver: Adapt to virDomain{Resume,Suspend}Flags vboxDriver: Adapt to virDomain{Resume,Suspend}Flags vmwareDriver: Adapt to virDomain{Resume,Suspend}Flags xenUnifiedDriver: Adapt to virDomain{Resume,Suspend}Flags xenapiDriver: Adapt to virDomain{Resume,Suspend}Flags virsh: Adapt to virDomain{Resume,Suspend}Flags include/libvirt/libvirt.h.in | 4 ++ src/driver.h | 10 ++++ src/esx/esx_driver.c | 28 ++++++++++- src/hyperv/hyperv_driver.c | 28 ++++++++++- src/libvirt.c | 104 ++++++++++++++++++++++++++++++++++----- src/libvirt_public.syms | 6 +++ src/libxl/libxl_driver.c | 28 +++++++++-- src/lxc/lxc_driver.c | 30 +++++++++-- src/openvz/openvz_driver.c | 28 ++++++++++- src/parallels/parallels_driver.c | 24 ++++++++- src/qemu/qemu_driver.c | 35 +++++++++++-- src/remote/remote_driver.c | 2 + src/remote/remote_protocol.x | 24 ++++++++- src/remote_protocol-structs | 10 ++++ src/test/test_driver.c | 26 +++++++++- src/vbox/vbox_tmpl.c | 28 ++++++++++- src/vmware/vmware_driver.c | 24 ++++++++- src/xen/xen_driver.c | 28 +++++++++-- src/xenapi/xenapi_driver.c | 42 ++++++++++++++-- tools/virsh-domain.c | 52 +++++++++++++++----- 20 files changed, 502 insertions(+), 59 deletions(-) -- 1.8.5.2

4 28

[libvirt] [PATCHv2 0/2] Fix problems with "forwardPlainNames"
by Laine Stump 04 Feb '14

04 Feb '14

changes from V1: * Fix the test cases * more information in commit log * Add 2nd patch to make default forwardPlainNames='yes', as suggested by danpb (since that was the original behavior). Here is the thread of the original patch: https://www.redhat.com/archives/libvir-list/2013-December/msg00397.html (continued here in January): https://www.redhat.com/archives/libvir-list/2014-January/msg01521.html I believe I've addressed Daniel's concerns except the ability to prevent forwarding of unresolved names in the configured domain, but that was never the intended behavior to begin with. These patches fix the bugs in the currently advertised behavior; we can create followup patches for this "new" behavior if there is demand (so far there have only been people saying that behavior should be *removed* :-) Laine Stump (2): network: only prevent forwarding of DNS requests for unqualified names network: change default of forwardPlainNames to 'yes' src/conf/network_conf.c | 32 +++++++++++++++------- src/conf/network_conf.h | 17 ++++++++++-- src/network/bridge_driver.c | 17 +++++------- tests/networkxml2confdata/dhcp6-nat-network.conf | 2 -- tests/networkxml2confdata/dhcp6-network.conf | 2 -- .../dhcp6host-routed-network.conf | 2 -- tests/networkxml2confdata/isolated-network.conf | 2 -- .../nat-network-dns-forwarders.conf | 2 -- .../networkxml2confdata/nat-network-dns-hosts.conf | 4 +-- .../nat-network-dns-srv-record-minimal.conf | 2 -- .../nat-network-dns-srv-record.conf | 2 -- .../nat-network-dns-txt-record.conf | 2 -- tests/networkxml2confdata/nat-network.conf | 2 -- tests/networkxml2confdata/netboot-network.conf | 2 -- .../networkxml2confdata/netboot-proxy-network.conf | 2 -- tests/networkxml2confdata/routed-network.conf | 2 -- tests/networkxml2xmlout/nat-network-dns-hosts.xml | 2 +- .../nat-network-dns-more-hosts.xml | 2 +- .../nat-network-no-hosts.xml | 1 + 19 files changed, 49 insertions(+), 50 deletions(-) -- 1.8.5.3

3 6

[libvirt] [PATCH] Host cpu offline/online removes the cpus available for vms
by Shivaprasad G Bhat 04 Feb '14

04 Feb '14

Online/Offline operations on the host cpus removes the machine/cpuset.cpus which are never added back. The guests with vcpu pinning can fail to boot unless the xml is edited. If the possibility that the offlined cpus are onlined back, the cpuset.cpus can be updated upon start of the guest thus allowing the guests to boot back without erroring out. Signed-off-by: Shivaprasad G Bhat <sbhat(a)linux.vnet.ibm.com> --- src/util/vircgroup.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index a6d60c5..52575c9 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -72,6 +72,7 @@ typedef enum { * before creating subcgroups and * attaching tasks */ + VIR_CGROUP_CPUS_HIERACHY = 1 << 1, /* call virCgroupCpuSetInherit */ } virCgroupFlags; @@ -891,6 +892,7 @@ virCgroupMakeGroup(virCgroupPtr parent, VIR_DEBUG("Make group %s", group->path); for (i = 0; i < VIR_CGROUP_CONTROLLER_LAST; i++) { char *path = NULL; + int inheritCpuset = 0; /* We must never mkdir() in systemd's hierarchy */ if (i == VIR_CGROUP_CONTROLLER_SYSTEMD) { @@ -933,15 +935,7 @@ virCgroupMakeGroup(virCgroupPtr parent, goto cleanup; } } - if (group->controllers[VIR_CGROUP_CONTROLLER_CPUSET].mountPoint != NULL && - (i == VIR_CGROUP_CONTROLLER_CPUSET || - STREQ(group->controllers[i].mountPoint, - group->controllers[VIR_CGROUP_CONTROLLER_CPUSET].mountPoint))) { - if (virCgroupCpuSetInherit(parent, group) < 0) { - VIR_FREE(path); - goto cleanup; - } - } + inheritCpuset = 1; /* * Note that virCgroupSetMemoryUseHierarchy should always be * called prior to creating subcgroups and attaching tasks. @@ -958,6 +952,19 @@ virCgroupMakeGroup(virCgroupPtr parent, } } + if (inheritCpuset || (flags & VIR_CGROUP_CPUS_HIERACHY)) + { + if (group->controllers[VIR_CGROUP_CONTROLLER_CPUSET].mountPoint != NULL && + (i == VIR_CGROUP_CONTROLLER_CPUSET || + STREQ(group->controllers[i].mountPoint, + group->controllers[VIR_CGROUP_CONTROLLER_CPUSET].mountPoint))) { + if (virCgroupCpuSetInherit(parent, group) < 0) { + VIR_FREE(path); + goto cleanup; + } + } + } + VIR_FREE(path); } @@ -1275,7 +1282,7 @@ virCgroupNewPartition(const char *path, if (virCgroupNew(-1, parentPath, NULL, controllers, &parent) < 0) goto cleanup; - if (virCgroupMakeGroup(parent, *group, create, VIR_CGROUP_NONE) < 0) { + if (virCgroupMakeGroup(parent, *group, create, VIR_CGROUP_CPUS_HIERACHY) < 0) { virCgroupRemove(*group); goto cleanup; }

3 3

[libvirt] [PATCH] man: shm-merge-across-nodes is optional
by Ján Tomko 04 Feb '14

04 Feb '14

Mark the shm-merge-across-nodes parameter of node-memory-tune command as optional in the virsh man page. --- tools/virsh.pod | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Pushed as trivial. diff --git a/tools/virsh.pod b/tools/virsh.pod index 9e670da..f221475 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -307,7 +307,7 @@ in seconds for which the host has to be suspended, it should be at least 60 seconds. =item B<node-memory-tune> [I<shm-pages-to-scan>] [I<shm-sleep-millisecs>] -I<shm-merge-across-nodes> +[I<shm-merge-across-nodes>] Allows you to display or set the node memory parameters. I<shm-pages-to-scan> can be used to set the number of pages to scan -- 1.8.3.2

1 0

[libvirt] 'make rpm' failing in VPATH build because of wireshark
by Eric Blake 04 Feb '14

04 Feb '14

I'm out of time to investigate today, but inside a VPATH build directory, 'make rpm' fails with: Making all in src make[4]: Entering directory `/home/eblake/rpmbuild/BUILD/libvirt-1.2.2/tools/wireshark/src' make[4]: *** No rule to make target `packet-libvirt.h', needed by `packet-libvirt.c'. Stop. If no one beats me to it, I'll try and fix that tomorrow. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

1 0

[libvirt] [PATCH] qemu: minor cleanups
by Martin Kletzander 03 Feb '14

03 Feb '14

Some cleanups around serial chardevs and miscellaneous things I've found inconsistent or not very clean. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/conf/domain_conf.c | 43 ++++++++++++++++++++++++++----------------- src/qemu/qemu_capabilities.c | 4 ++-- src/qemu/qemu_command.c | 10 ++++++++-- 3 files changed, 36 insertions(+), 21 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 28e24f9..fa1ecb5 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -1559,7 +1559,7 @@ virDomainChrSourceDefIsEqual(const virDomainChrSourceDef *src, if (tgt->type != src->type) return false; - switch (src->type) { + switch ((enum virDomainChrType)src->type) { case VIR_DOMAIN_CHR_TYPE_PTY: case VIR_DOMAIN_CHR_TYPE_DEV: case VIR_DOMAIN_CHR_TYPE_FILE: @@ -1583,16 +1583,22 @@ virDomainChrSourceDefIsEqual(const virDomainChrSourceDef *src, STREQ_NULLABLE(src->data.nix.path, tgt->data.nix.path); break; + case VIR_DOMAIN_CHR_TYPE_NULL: case VIR_DOMAIN_CHR_TYPE_VC: case VIR_DOMAIN_CHR_TYPE_STDIO: case VIR_DOMAIN_CHR_TYPE_SPICEVMC: + case VIR_DOMAIN_CHR_TYPE_LAST: /* nada */ return true; } - /* This should happen only on new, - * yet unhandled type */ + /* Even though gcc is able to detect that all possible values are + * handled in the switch above, it is not capable of realizing + * there isn't any possibility of escaping that switch without + * return. So for the sake of clean compilation, there has to be + * a return here */ + /* coverity[dead_error_begin] */ return false; } @@ -6415,7 +6421,7 @@ error: } #define NET_MODEL_CHARS \ - "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ091234567890_-" + "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_-" /* Parse the XML definition for a network interface * @param node XML nodeset to parse for net definition @@ -7182,11 +7188,11 @@ virDomainChrSourceDefParseXML(virDomainChrSourceDefPtr def, } switch ((enum virDomainChrType) def->type) { - case VIR_DOMAIN_CHR_TYPE_LAST: case VIR_DOMAIN_CHR_TYPE_NULL: + case VIR_DOMAIN_CHR_TYPE_VC: case VIR_DOMAIN_CHR_TYPE_STDIO: case VIR_DOMAIN_CHR_TYPE_SPICEVMC: - case VIR_DOMAIN_CHR_TYPE_VC: + case VIR_DOMAIN_CHR_TYPE_LAST: break; case VIR_DOMAIN_CHR_TYPE_PTY: @@ -15573,11 +15579,13 @@ virDomainChrSourceDefFormat(virBufferPtr buf, } virBufferAddLit(buf, ">\n"); - switch (def->type) { + virBufferAdjustIndent(buf, 6); + switch ((enum virDomainChrType)def->type) { case VIR_DOMAIN_CHR_TYPE_NULL: case VIR_DOMAIN_CHR_TYPE_VC: case VIR_DOMAIN_CHR_TYPE_STDIO: case VIR_DOMAIN_CHR_TYPE_SPICEVMC: + case VIR_DOMAIN_CHR_TYPE_LAST: /* nada */ break; @@ -15588,7 +15596,7 @@ virDomainChrSourceDefFormat(virBufferPtr buf, if (def->type != VIR_DOMAIN_CHR_TYPE_PTY || (def->data.file.path && !(flags & VIR_DOMAIN_XML_INACTIVE))) { - virBufferEscapeString(buf, " <source path='%s'/>\n", + virBufferEscapeString(buf, "<source path='%s'/>\n", def->data.file.path); } break; @@ -15597,53 +15605,54 @@ virDomainChrSourceDefFormat(virBufferPtr buf, if (def->data.udp.bindService && def->data.udp.bindHost) { virBufferAsprintf(buf, - " <source mode='bind' host='%s' " + "<source mode='bind' host='%s' " "service='%s'/>\n", def->data.udp.bindHost, def->data.udp.bindService); } else if (def->data.udp.bindHost) { - virBufferAsprintf(buf, " <source mode='bind' host='%s'/>\n", + virBufferAsprintf(buf, "<source mode='bind' host='%s'/>\n", def->data.udp.bindHost); } else if (def->data.udp.bindService) { - virBufferAsprintf(buf, " <source mode='bind' service='%s'/>\n", + virBufferAsprintf(buf, "<source mode='bind' service='%s'/>\n", def->data.udp.bindService); } if (def->data.udp.connectService && def->data.udp.connectHost) { virBufferAsprintf(buf, - " <source mode='connect' host='%s' " + "<source mode='connect' host='%s' " "service='%s'/>\n", def->data.udp.connectHost, def->data.udp.connectService); } else if (def->data.udp.connectHost) { - virBufferAsprintf(buf, " <source mode='connect' host='%s'/>\n", + virBufferAsprintf(buf, "<source mode='connect' host='%s'/>\n", def->data.udp.connectHost); } else if (def->data.udp.connectService) { virBufferAsprintf(buf, - " <source mode='connect' service='%s'/>\n", + "<source mode='connect' service='%s'/>\n", def->data.udp.connectService); } break; case VIR_DOMAIN_CHR_TYPE_TCP: virBufferAsprintf(buf, - " <source mode='%s' host='%s' service='%s'/>\n", + "<source mode='%s' host='%s' service='%s'/>\n", def->data.tcp.listen ? "bind" : "connect", def->data.tcp.host, def->data.tcp.service); - virBufferAsprintf(buf, " <protocol type='%s'/>\n", + virBufferAsprintf(buf, "<protocol type='%s'/>\n", virDomainChrTcpProtocolTypeToString( def->data.tcp.protocol)); break; case VIR_DOMAIN_CHR_TYPE_UNIX: - virBufferAsprintf(buf, " <source mode='%s'", + virBufferAsprintf(buf, "<source mode='%s'", def->data.nix.listen ? "bind" : "connect"); virBufferEscapeString(buf, " path='%s'", def->data.nix.path); virBufferAddLit(buf, "/>\n"); break; } + virBufferAdjustIndent(buf, -6); return 0; } diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 8420047..8aec293 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -1,7 +1,7 @@ /* * qemu_capabilities.c: QEMU capabilities generation * - * Copyright (C) 2006-2013 Red Hat, Inc. + * Copyright (C) 2006-2014 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -247,7 +247,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "boot-strict", /* 160 */ "pvpanic", "enable-fips", - "spice-file-xfer-disable" + "spice-file-xfer-disable", ); struct _virQEMUCaps { diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 96b8825..2db745a 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1,7 +1,7 @@ /* * qemu_command.c: QEMU command generation * - * Copyright (C) 2006-2013 Red Hat, Inc. + * Copyright (C) 2006-2014 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -6004,7 +6004,7 @@ qemuBuildChrArgStr(virDomainChrSourceDefPtr dev, const char *prefix) if (prefix) virBufferAdd(&buf, prefix, strlen(prefix)); - switch (dev->type) { + switch ((enum virDomainChrType)dev->type) { case VIR_DOMAIN_CHR_TYPE_NULL: virBufferAddLit(&buf, "null"); break; @@ -6071,6 +6071,12 @@ qemuBuildChrArgStr(virDomainChrSourceDefPtr dev, const char *prefix) dev->data.nix.path, dev->data.nix.listen ? ",server,nowait" : ""); break; + + case VIR_DOMAIN_CHR_TYPE_SPICEVMC: + /* spicevmc doesn't have any '-serial' compatible option */ + case VIR_DOMAIN_CHR_TYPE_LAST: + /* coverity[dead_error_begin] */ + break; } if (virBufferError(&buf)) { -- 1.8.5.3

3 5

[libvirt] [PATCH] virnetdevbandwidthtest: Link with libxml2
by Michal Privoznik 03 Feb '14

03 Feb '14

I've received a notice over IRC that on some systems, the virnetdevbandwidthtest is not linked with libxml: /usr/bin/ld: virnetdevbandwidthtest.o: undefined reference to symbol 'xmlStrEqual@@LIBXML2_2.4.30' /usr/lib/x86_64-linux-gnu/libxml2.so.2: error adding symbols: DSO missing from command line Trivial way avoiding this is to add LIBXML_LIBS to virnetdevbandwidthtest_LDADD. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index cfc0905..91c385e 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -658,7 +658,7 @@ commandhelper_LDFLAGS = -static virnetdevbandwidthtest_SOURCES = \ virnetdevbandwidthtest.c testutils.h testutils.c -virnetdevbandwidthtest_LDADD = $(LDADDS) +virnetdevbandwidthtest_LDADD = $(LDADDS) $(LIBXML_LIBS) if WITH_LIBVIRTD libvirtdconftest_SOURCES = \ -- 1.8.5.2

3 2

[libvirt] [PATCHv7 0/2] Implement RBD storage pool support
by Adam Walters 03 Feb '14

03 Feb '14

Here is a re-based and fixed up re-submission of my patches to implement RBD storage pool support for QEMU domains. Nothing in it has changed much, other than it has been rebased against the lastest libvirt and two of my patches have been squashed in order to pass a make, make check, and make syntax-check after each individual patch. The code here still works well for me, but I only have the one host to test this on. If possible, I would like to try and get this merged for the 1.2.2 release cycle, so that other users can benefit from the added storage pool support. As always, if you find any problems with my patches, please let me know, and I will fix them up as soon as time permits. Adam Walters (2): qemu: conf Implement domain RBD storage pool support domain: conf: Fix secret type checking for network volumes src/conf/domain_conf.c | 6 +++-- src/qemu/qemu_conf.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 65 insertions(+), 3 deletions(-) -- 1.8.5.2

2 8

[libvirt] [PATCH V3]Sheepdog: Auto Adding volume and pool refresh.
by joel SIMOES 03 Feb '14

03 Feb '14

From: Joel <joel.simoes(a)laposte.net> Fix bug 1055479 https://bugzilla.redhat.com/show_bug.cgi?id=1055479 Libvirt lose sheepdogs volumes on pool refresh or restart. When restarting sheepdog pool, all volumes are missing. This patch add automatically all volume from the added pool. --- src/storage/storage_backend_sheepdog.c | 74 +++++++++++++++++++++++++++++++++- 1 file changed, 72 insertions(+), 2 deletions(-) diff --git a/src/storage/storage_backend_sheepdog.c b/src/storage/storage_backend_sheepdog.c index a6981ce..a836f27 100644 --- a/src/storage/storage_backend_sheepdog.c +++ b/src/storage/storage_backend_sheepdog.c @@ -109,6 +109,70 @@ virStorageBackendSheepdogAddHostArg(virCommandPtr cmd, virCommandAddArgFormat(cmd, "%d", port); } +static int +virStorageBackendSheepdogRefreshAllVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool) +{ + int ret; + char *output = NULL; + + virCommandPtr cmd = virCommandNewArgList(COLLIE, "vdi", "list", "-r", NULL); + virStorageBackendSheepdogAddHostArg(cmd, pool); + virCommandSetOutputBuffer(cmd, &output); + ret = virCommandRun(cmd, NULL); + char** lines; + char** cells; + + if (ret < 0) + goto cleanup; + + ret = -1; + lines = virStringSplit(output, "\n", 0); + size_t i; + for (i = 0; *(lines + i); i++) { + char *line = *(lines + i); + cells = virStringSplit(line, " ", 0); + size_t j; + for (j = 0; *(cells + j); j++) { + + char *cell = *(cells + j); + if (j == 1) { + virStorageVolDefPtr vol = NULL; + if (VIR_ALLOC(vol) < 0) + goto cleanup; + + if (VIR_STRDUP(vol->name, cell) < 0) + goto cleanup; + + vol->type = VIR_STORAGE_VOL_BLOCK; + + if (VIR_EXPAND_N(pool->volumes.objs, pool->volumes.count, 1) < 0) + goto cleanup; + pool->volumes.objs[pool->volumes.count - 1] = vol; + + if (virStorageBackendSheepdogRefreshVol(conn, pool, vol) < 0) + goto cleanup; + + vol = NULL; + } + + VIR_FREE(*(cells + j)); + } + + VIR_FREE(cells); + VIR_FREE(*(lines + i)); + } + + + VIR_FREE(lines); + ret = 0; + +cleanup: + virCommandFree(cmd); + VIR_FREE(lines); + VIR_FREE(cells); + return ret; +} static int virStorageBackendSheepdogRefreshPool(virConnectPtr conn ATTRIBUTE_UNUSED, @@ -122,9 +186,15 @@ virStorageBackendSheepdogRefreshPool(virConnectPtr conn ATTRIBUTE_UNUSED, virStorageBackendSheepdogAddHostArg(cmd, pool); virCommandSetOutputBuffer(cmd, &output); ret = virCommandRun(cmd, NULL); - if (ret == 0) - ret = virStorageBackendSheepdogParseNodeInfo(pool->def, output); + if (ret < 0) + goto cleanup; + ret = virStorageBackendSheepdogParseNodeInfo(pool->def, output); + if (ret < 0) + goto cleanup; + ret = virStorageBackendSheepdogRefreshAllVol(conn, pool); + +cleanup: virCommandFree(cmd); VIR_FREE(output); return ret; -- 1.8.3.2

3 2

[libvirt] [PATCH v4] bhyve: add a basic driver
by Roman Bogorodskiy 03 Feb '14

03 Feb '14

At this point it has a limited functionality and is highly experimental. Supported domain operations are: * define * start * destroy * dumpxml * dominfo It's only possible to have only one disk device and only one network, which should be of type bridge. --- configure.ac | 7 + daemon/libvirtd.c | 9 + include/libvirt/virterror.h | 1 + m4/virt-driver-bhyve.m4 | 57 +++++ po/POTFILES.in | 3 + src/Makefile.am | 30 +++ src/bhyve/bhyve_command.c | 276 +++++++++++++++++++++ src/bhyve/bhyve_command.h | 41 ++++ src/bhyve/bhyve_driver.c | 566 ++++++++++++++++++++++++++++++++++++++++++++ src/bhyve/bhyve_driver.h | 28 +++ src/bhyve/bhyve_process.c | 216 +++++++++++++++++ src/bhyve/bhyve_process.h | 36 +++ src/bhyve/bhyve_utils.h | 48 ++++ src/conf/domain_conf.c | 3 +- src/conf/domain_conf.h | 1 + src/driver.h | 1 + src/libvirt.c | 3 + src/util/virerror.c | 1 + 18 files changed, 1326 insertions(+), 1 deletion(-) create mode 100644 m4/virt-driver-bhyve.m4 create mode 100644 src/bhyve/bhyve_command.c create mode 100644 src/bhyve/bhyve_command.h create mode 100644 src/bhyve/bhyve_driver.c create mode 100644 src/bhyve/bhyve_driver.h create mode 100644 src/bhyve/bhyve_process.c create mode 100644 src/bhyve/bhyve_process.h create mode 100644 src/bhyve/bhyve_utils.h diff --git a/configure.ac b/configure.ac index 1670a41..35c52d5 100644 --- a/configure.ac +++ b/configure.ac @@ -1031,6 +1031,12 @@ fi AM_CONDITIONAL([WITH_PARALLELS], [test "$with_parallels" = "yes"]) dnl +dnl Checks for bhyve driver +dnl + +LIBVIRT_DRIVER_CHECK_BHYVE + +dnl dnl check for shell that understands <> redirection without truncation, dnl needed by src/qemu/qemu_monitor_{text,json}.c. dnl @@ -2687,6 +2693,7 @@ AC_MSG_NOTICE([ PHYP: $with_phyp]) AC_MSG_NOTICE([ ESX: $with_esx]) AC_MSG_NOTICE([ Hyper-V: $with_hyperv]) AC_MSG_NOTICE([Parallels: $with_parallels]) +LIBIVRT_DRIVER_RESULT_BHYVE AC_MSG_NOTICE([ Test: $with_test]) AC_MSG_NOTICE([ Remote: $with_remote]) AC_MSG_NOTICE([ Network: $with_network]) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 49c42ad..b27c6fd 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -77,6 +77,9 @@ # ifdef WITH_VBOX # include "vbox/vbox_driver.h" # endif +# ifdef WITH_BHYVE +# include "bhyve/bhyve_driver.h" +# endif # ifdef WITH_NETWORK # include "network/bridge_driver.h" # endif @@ -405,6 +408,9 @@ static void daemonInitialize(void) # ifdef WITH_VBOX virDriverLoadModule("vbox"); # endif +# ifdef WITH_BHYVE + virDriverLoadModule("bhyve"); +# endif #else # ifdef WITH_NETWORK networkRegister(); @@ -442,6 +448,9 @@ static void daemonInitialize(void) # ifdef WITH_VBOX vboxRegister(); # endif +# ifdef WITH_BHYVE + bhyveRegister(); +# endif #endif } diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index e31e9c4..7915bbb 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -121,6 +121,7 @@ typedef enum { VIR_FROM_ACCESS = 55, /* Error from access control manager */ VIR_FROM_SYSTEMD = 56, /* Error from systemd code */ + VIR_FROM_BHYVE = 57, /* Error from bhyve driver */ # ifdef VIR_ENUM_SENTINELS VIR_ERR_DOMAIN_LAST # endif diff --git a/m4/virt-driver-bhyve.m4 b/m4/virt-driver-bhyve.m4 new file mode 100644 index 0000000..969133a --- /dev/null +++ b/m4/virt-driver-bhyve.m4 @@ -0,0 +1,57 @@ +dnl The bhyve driver +dnl +dnl Copyright (C) 2014 Roman Bogorodskiy +dnl +dnl This library is free software; you can redistribute it and/or +dnl modify it under the terms of the GNU Lesser General Public +dnl License as published by the Free Software Foundation; either +dnl version 2.1 of the License, or (at your option) any later version. +dnl +dnl This library is distributed in the hope that it will be useful, +dnl but WITHOUT ANY WARRANTY; without even the implied warranty of +dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +dnl Lesser General Public License for more details. +dnl +dnl You should have received a copy of the GNU Lesser General Public +dnl License along with this library. If not, see +dnl <http://www.gnu.org/licenses/>. +dnl + +AC_DEFUN([LIBVIRT_DRIVER_CHECK_BHYVE],[ + AC_ARG_WITH([bhyve], + [AS_HELP_STRING([--with-bhyve], + [add BHyVe support @<:@default=check@:>@])]) + m4_divert_text([DEFAULTS], [with_bhyve=check]) + + if test "$with_bhyve" != "no"; then + AC_PATH_PROG([BHYVE], [bhyve], [], [$PATH:/usr/sbin]) + AC_PATH_PROG([BHYVECTL], [bhyvectl], [$PATH:/usr/sbin]) + AC_PATH_PROG([BHYVELOAD], [bhyveload], [$PATH:/usr/sbin/]) + + if test -z "$BHYVE" || test -z "$BHYVECTL" \ + test -z "$BHYVELOAD" || test "$with_freebsd" = "no"; then + if test "$with_bhyve" = "check"; then + with_bhyve="no" + else + AC_MSG_ERROR([The bhyve driver cannot be enabled]) + fi + else + with_bhyve="yes" + fi + fi + + if test "$with_bhyve" = "yes"; then + AC_DEFINE_UNQUOTED([WITH_BHYVE], 1, [whether bhyve driver is enabled]) + AC_DEFINE_UNQUOTED([BHYVE], ["$BHYVE"], + [Location of the bhyve tool]) + AC_DEFINE_UNQUOTED([BHYVECTL], ["$BHYVECTL"], + [Location of the bhyvectl tool]) + AC_DEFINE_UNQUOTED([BHYVELOAD], ["$BHYVELOAD"], + [Location of the bhyveload tool]) + fi + AM_CONDITIONAL([WITH_BHYVE], [test "$with_bhyve" = "yes"]) +]) + +AC_DEFUN([LIBIVRT_DRIVER_RESULT_BHYVE],[ + AC_MSG_NOTICE([ Bhyve: $with_bhyve]) +]) diff --git a/po/POTFILES.in b/po/POTFILES.in index 0359b2f..5c9bc7e 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -8,6 +8,9 @@ gnulib/lib/gai_strerror.c gnulib/lib/regcomp.c src/access/viraccessdriverpolkit.c src/access/viraccessmanager.c +src/bhyve/bhyve_command.c +src/bhyve/bhyve_driver.c +src/bhyve/bhyve_process.c src/conf/capabilities.c src/conf/cpu_conf.c src/conf/device_conf.c diff --git a/src/Makefile.am b/src/Makefile.am index abe0a51..8408063 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -515,6 +515,7 @@ DRIVER_SOURCE_FILES = \ $(NULL) STATEFUL_DRIVER_SOURCE_FILES = \ + $(BHYVE_DRIVER_SOURCES) \ $(INTERFACE_DRIVER_SOURCES) \ $(LIBXL_DRIVER_SOURCES) \ $(LXC_DRIVER_SOURCES) \ @@ -774,6 +775,15 @@ PARALLELS_DRIVER_SOURCES = \ parallels/parallels_storage.c \ parallels/parallels_network.c +BHYVE_DRIVER_SOURCES = \ + bhyve/bhyve_command.c \ + bhyve/bhyve_command.h \ + bhyve/bhyve_driver.h \ + bhyve/bhyve_driver.c \ + bhyve/bhyve_process.c \ + bhyve/bhyve_process.h \ + bhyve/bhyve_utils.h + NETWORK_DRIVER_SOURCES = \ network/bridge_driver.h network/bridge_driver.c \ network/bridge_driver_platform.h \ @@ -1310,6 +1320,25 @@ libvirt_driver_parallels_la_CFLAGS = \ libvirt_driver_parallels_la_SOURCES = $(PARALLELS_DRIVER_SOURCES) endif WITH_PARALLELS +if WITH_BHYVE +noinst_LTLIBRARIES += libvirt_driver_bhyve_impl.la +libvirt_driver_bhyve_la_SOURCES = +libvirt_driver_bhyve_la_LIBADD = libvirt_driver_bhyve_impl.la +if WITH_DRIVER_MODULES +mod_LTLIBRARIES += libvirt_driver_bhyve.la +libvirt_driver_bhyve_la_LIBADD += ../gnulib/lib/libgnu.la +libvirt_driver_bhyve_la_LDFLAGS = -module -avoid-version $(AM_LDFLAGS) +else ! WITH_DRIVER_MODULES +noinst_LTLIBRARIES += libvirt_driver_bhyve.la +endif ! WITH_DRIVER_MODULES + +libvirt_driver_bhyve_impl_la_CFLAGS = \ + -I$(top_srcdir)/src/conf \ + $(AM_CFLAGS) +libvirt_driver_bhyve_impl_la_LDFLAGS = $(AM_LDFLAGS) +libvirt_driver_bhyve_impl_la_SOURCES = $(BHYVE_DRIVER_SOURCES) +endif WITH_BHYVE + if WITH_NETWORK noinst_LTLIBRARIES += libvirt_driver_network_impl.la libvirt_driver_network_la_SOURCES = @@ -1644,6 +1673,7 @@ EXTRA_DIST += \ $(HYPERV_DRIVER_SOURCES) \ $(HYPERV_DRIVER_EXTRA_DIST) \ $(PARALLELS_DRIVER_SOURCES) \ + $(BHYVE_DRIVER_SOURCES) \ $(NETWORK_DRIVER_SOURCES) \ $(INTERFACE_DRIVER_SOURCES) \ $(STORAGE_DRIVER_SOURCES) \ diff --git a/src/bhyve/bhyve_command.c b/src/bhyve/bhyve_command.c new file mode 100644 index 0000000..3a444a7 --- /dev/null +++ b/src/bhyve/bhyve_command.c @@ -0,0 +1,276 @@ +/* + * bhyve_process.c: bhyve command generation + * + * Copyright (C) 2014 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + */ + +#include <config.h> + +#include <fcntl.h> +#include <sys/types.h> +#include <dirent.h> +#include <sys/ioctl.h> +#include <net/if.h> +#include <net/if_tap.h> + +#include "bhyve_command.h" +#include "viralloc.h" +#include "virfile.h" +#include "virstring.h" +#include "virlog.h" +#include "virnetdev.h" +#include "virnetdevbridge.h" +#include "virnetdevtap.h" + +#define VIR_FROM_THIS VIR_FROM_BHYVE + +static char* +virBhyveTapGetRealDeviceName(char *name) +{ + /* This is an ugly hack, because if we rename + * tap device to vnet%d, its device name will be + * still /dev/tap%d, and bhyve tries too open /dev/tap%d, + * so we have to find the real name + */ + char *ret = NULL; + struct dirent *dp; + char *devpath = NULL; + int fd; + + DIR *dirp = opendir("/dev"); + if (dirp == NULL) { + virReportSystemError(errno, + _("Failed to opendir path '%s'"), + "/dev"); + return NULL; + } + + while ((dp = readdir(dirp)) != NULL) { + if (STRPREFIX(dp->d_name, "tap")) { + struct ifreq ifr; + if (virAsprintf(&devpath, "/dev/%s", dp->d_name) < 0) { + goto cleanup; + } + if ((fd = open(devpath, O_RDWR)) < 0) { + virReportSystemError(errno, _("Unable to open '%s'"), devpath); + goto cleanup; + } + + if (ioctl(fd, TAPGIFNAME, (void *)&ifr) < 0) { + virReportSystemError(errno, "%s", + _("Unable to query tap interface name")); + goto cleanup; + } + + if (STREQ(name, ifr.ifr_name)) { + /* we can ignore the return value + * because we still have nothing + * to do but return; + */ + ignore_value(VIR_STRDUP(ret, dp->d_name)); + goto cleanup; + } + + VIR_FREE(devpath); + VIR_FORCE_CLOSE(fd); + } + } + +cleanup: + VIR_FREE(devpath); + VIR_FORCE_CLOSE(fd); + closedir(dirp); + return ret; +} + +static int +bhyveBuildNetArgStr(const virDomainDef *def, virCommandPtr cmd) +{ + virDomainNetDefPtr net = NULL; + char *brname = NULL; + char *realifname = NULL; + int *tapfd = NULL; + + if (def->nnets != 1) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("domain should have one and only one net defined")); + return -1; + } + + net = def->nets[0]; + + if (net != NULL) { + int actualType = virDomainNetGetActualType(net); + + if (actualType == VIR_DOMAIN_NET_TYPE_BRIDGE) { + if (VIR_STRDUP(brname, virDomainNetGetActualBridgeName(net)) < 0) + return -1L; + } else { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("Network type %d is not supported"), + virDomainNetGetActualType(net)); + return -1; + } + + if (!net->ifname || + STRPREFIX(net->ifname, VIR_NET_GENERATED_PREFIX) || + strchr(net->ifname, '%')) { + VIR_FREE(net->ifname); + if (VIR_STRDUP(net->ifname, VIR_NET_GENERATED_PREFIX "%d") < 0) { + VIR_FREE(brname); + return -1; + } + } + + if (virNetDevTapCreateInBridgePort(brname, &net->ifname, &net->mac, + def->uuid, tapfd, 1, + virDomainNetGetActualVirtPortProfile(net), + virDomainNetGetActualVlan(net), + VIR_NETDEV_TAP_CREATE_IFUP | VIR_NETDEV_TAP_CREATE_PERSIST) < 0) { + VIR_FREE(net->ifname); + VIR_FREE(brname); + return -1; + } + } + + realifname = virBhyveTapGetRealDeviceName(net->ifname); + + if (realifname == NULL) { + VIR_FREE(net->ifname); + VIR_FREE(brname); + return -1; + } + + VIR_DEBUG("%s -> %s", net->ifname, realifname); + /* hack on top of other hack: we need to set + * interface to 'UP' again after re-opening to find its + * name + */ + if (virNetDevSetOnline(net->ifname, true) != 0) { + VIR_FREE(net->ifname); + VIR_FREE(brname); + return -1; + } + + virCommandAddArg(cmd, "-s"); + virCommandAddArg(cmd, "0:0,hostbridge"); + virCommandAddArg(cmd, "-s"); + virCommandAddArgFormat(cmd, "1:0,virtio-net,%s", realifname); + + return 0; +} + +static int +bhyveBuildDiskArgStr(const virDomainDef *def, virCommandPtr cmd) +{ + virDomainDiskDefPtr disk; + + if (def->ndisks != 1) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("domain should have one and only one disk defined")); + return -1; + } + + disk = def->disks[0]; + + virCommandAddArg(cmd, "-s"); + virCommandAddArgFormat(cmd, "2:0,ahci-hd,%s", disk->src); + + return 0; +} + +virCommandPtr +virBhyveProcessBuildBhyveCmd(bhyveConnPtr driver ATTRIBUTE_UNUSED, + virDomainObjPtr vm) +{ + /* + * /usr/sbin/bhyve -c 2 -m 256 -AI -H -P \ + * -s 0:0,hostbridge \ + * -s 1:0,virtio-net,tap0 \ + * -s 2:0,ahci-hd,${IMG} \ + * -S 31,uart,stdio \ + * vm0 + */ + virCommandPtr cmd = virCommandNew(BHYVE); + + /* CPUs */ + virCommandAddArg(cmd, "-c"); + virCommandAddArgFormat(cmd, "%d", vm->def->vcpus); + + /* Memory */ + virCommandAddArg(cmd, "-m"); + virCommandAddArgFormat(cmd, "%llu", + VIR_DIV_UP(vm->def->mem.max_balloon, 1024)); + + /* Options */ + virCommandAddArg(cmd, "-A"); /* Create an ACPI table */ + virCommandAddArg(cmd, "-I"); /* Present ioapic to the guest */ + virCommandAddArg(cmd, "-H"); /* vmexit from guest on hlt */ + virCommandAddArg(cmd, "-P"); /* vmexit from guest on pause */ + + /* Devices */ + if (bhyveBuildNetArgStr(vm->def, cmd) < 0) + goto error; + if (bhyveBuildDiskArgStr(vm->def, cmd) < 0) + goto error; + virCommandAddArg(cmd, "-S"); + virCommandAddArg(cmd, "31,uart"); + virCommandAddArg(cmd, vm->def->name); + + return cmd; + +error: + virCommandFree(cmd); + return NULL; +} + +virCommandPtr +virBhyveProcessBuildDestroyCmd(bhyveConnPtr driver ATTRIBUTE_UNUSED, + virDomainObjPtr vm) +{ + virCommandPtr cmd = virCommandNew(BHYVECTL); + + virCommandAddArg(cmd, "--destroy"); + virCommandAddArgPair(cmd, "--vm", vm->def->name); + + return cmd; +} + +virCommandPtr +virBhyveProcessBuildLoadCmd(bhyveConnPtr driver ATTRIBUTE_UNUSED, + virDomainObjPtr vm) +{ + virCommandPtr cmd; + virDomainDiskDefPtr disk = vm->def->disks[0]; + + cmd = virCommandNew(BHYVELOAD); + + /* Memory */ + virCommandAddArg(cmd, "-m"); + virCommandAddArgFormat(cmd, "%llu", + VIR_DIV_UP(vm->def->mem.max_balloon, 1024)); + + /* Image path */ + virCommandAddArg(cmd, "-d"); + virCommandAddArgFormat(cmd, disk->src); + + /* VM name */ + virCommandAddArg(cmd, vm->def->name); + + return cmd; +} diff --git a/src/bhyve/bhyve_command.h b/src/bhyve/bhyve_command.h new file mode 100644 index 0000000..8326971 --- /dev/null +++ b/src/bhyve/bhyve_command.h @@ -0,0 +1,41 @@ +/* + * bhyve_process.c: bhyve command generation + * + * Copyright (C) 2014 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + */ + +#ifndef __BHYVE_COMMAND_H__ +# define __BHYVE_COMMAND_H__ + +# include "bhyve_utils.h" + +# include "domain_conf.h" +# include "vircommand.h" + +virCommandPtr virBhyveProcessBuildBhyveCmd(bhyveConnPtr, + virDomainObjPtr vm); + +virCommandPtr +virBhyveProcessBuildDestroyCmd(bhyveConnPtr driver, + virDomainObjPtr vm); + +virCommandPtr +virBhyveProcessBuildLoadCmd(bhyveConnPtr driver, + virDomainObjPtr vm); + +#endif /* __BHYVE_COMMAND_H__ */ diff --git a/src/bhyve/bhyve_driver.c b/src/bhyve/bhyve_driver.c new file mode 100644 index 0000000..deeb794 --- /dev/null +++ b/src/bhyve/bhyve_driver.c @@ -0,0 +1,566 @@ +/* + * bhyve_driver.c: core driver methods for managing bhyve guests + * + * Copyright (C) 2013 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Roman Bogorodskiy + */ + +#include <config.h> + +#include "virerror.h" +#include "datatypes.h" +#include "virbuffer.h" +#include "viruuid.h" +#include "capabilities.h" +#include "configmake.h" +#include "viralloc.h" +#include "network_conf.h" +#include "interface_conf.h" +#include "domain_audit.h" +#include "domain_conf.h" +#include "snapshot_conf.h" +#include "fdstream.h" +#include "storage_conf.h" +#include "node_device_conf.h" +#include "virxml.h" +#include "virthread.h" +#include "virlog.h" +#include "virfile.h" +#include "virtypedparam.h" +#include "virrandom.h" +#include "virstring.h" +#include "cpu/cpu.h" + +#include "bhyve_driver.h" +#include "bhyve_process.h" +#include "bhyve_utils.h" + +#define VIR_FROM_THIS VIR_FROM_BHYVE + +bhyveConnPtr bhyve_driver = NULL; + +void +bhyveDriverLock(bhyveConnPtr driver) +{ + virMutexLock(&driver->lock); +} + +void +bhyveDriverUnlock(bhyveConnPtr driver) +{ + virMutexUnlock(&driver->lock); +} + +static virCapsPtr +bhyveBuildCapabilities(void) +{ + virCapsPtr caps; + virCapsGuestPtr guest; + + if ((caps = virCapabilitiesNew(virArchFromHost(), + 0, 0)) == NULL) + return NULL; + + if ((guest = virCapabilitiesAddGuest(caps, "hvm", + VIR_ARCH_X86_64, + "bhyve", + NULL, 0, NULL)) == NULL) + goto error; + + if (virCapabilitiesAddGuestDomain(guest, + "bhyve", NULL, NULL, 0, NULL) == NULL) + goto error; + + return caps; + +error: + virObjectUnref(caps); + return NULL; +} + +static char * +bhyveConnectGetCapabilities(virConnectPtr conn) +{ + bhyveConnPtr privconn = conn->privateData; + char *xml; + + bhyveDriverLock(privconn); + if ((xml = virCapabilitiesFormatXML(privconn->caps)) == NULL) + virReportOOMError(); + bhyveDriverUnlock(privconn); + + return xml; +} + +static virDomainObjPtr +bhyveDomObjFromDomain(virDomainPtr domain) +{ + virDomainObjPtr vm; + bhyveConnPtr privconn = domain->conn->privateData; + char uuidstr[VIR_UUID_STRING_BUFLEN]; + + vm = virDomainObjListFindByUUID(privconn->domains, domain->uuid); + if (!vm) { + virUUIDFormat(domain->uuid, uuidstr); + virReportError(VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s' (%s)"), + uuidstr, domain->name); + return NULL; + } + + return vm; +} + +static virDrvOpenStatus +bhyveConnectOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + unsigned int flags) +{ + virCheckFlags(VIR_CONNECT_RO, VIR_DRV_OPEN_ERROR); + + if (!conn->uri) + return VIR_DRV_OPEN_DECLINED; + + if (!conn->uri->scheme || STRNEQ(conn->uri->scheme, "bhyve")) + return VIR_DRV_OPEN_DECLINED; + + if (conn->uri->server) + return VIR_DRV_OPEN_DECLINED; + + if (!STREQ_NULLABLE(conn->uri->path, "/system")) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unexpected bhyve URI path '%s', try bhyve:///system"), + conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } + + if (bhyve_driver == NULL) { + virReportError(VIR_ERR_INTERNAL_ERROR, + "%s", _("bhyve state driver is not active")); + return VIR_DRV_OPEN_ERROR; + } + + conn->privateData = bhyve_driver; + + return VIR_DRV_OPEN_SUCCESS; +} + +static int +bhyveConnectClose(virConnectPtr conn) +{ + bhyveConnPtr privconn = conn->privateData; + + bhyveDriverLock(privconn); + virObjectUnref(privconn->caps); + virObjectUnref(privconn->xmlopt); + virObjectUnref(privconn->domains); + conn->privateData = NULL; + + bhyveDriverUnlock(privconn); + virMutexDestroy(&privconn->lock); + + VIR_FREE(privconn); + return 0; +} + +static char * +bhyveConnectGetHostname(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + return virGetHostname(); +} + +static int +bhyveConnectGetVersion(virConnectPtr conn ATTRIBUTE_UNUSED, unsigned long *hvVer) +{ + if (virParseVersionString("0.0.1", hvVer, true) < 0) { + return -1; + } + + return 0; +} + +static int +bhyveDomainGetInfo(virDomainPtr domain, virDomainInfoPtr info) +{ + virDomainObjPtr vm; + int ret = -1; + + if (!(vm = bhyveDomObjFromDomain(domain))) + goto cleanup; + + info->maxMem = vm->def->mem.max_balloon; + info->nrVirtCpu = vm->def->vcpus; + ret = 0; + +cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static int +bhyveDomainGetState(virDomainPtr domain, + int *state, + int *reason, + unsigned int flags) +{ + virDomainObjPtr vm; + int ret = -1; + + virCheckFlags(0, -1); + + if (!(vm = bhyveDomObjFromDomain(domain))) + goto cleanup; + + *state = virDomainObjGetState(vm, reason); + ret = 0; + +cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static char * +bhyveDomainGetXMLDesc(virDomainPtr domain, unsigned int flags) +{ + virDomainObjPtr vm; + char *ret = NULL; + + if (!(vm = bhyveDomObjFromDomain(domain))) + goto cleanup; + + ret = virDomainDefFormat(vm->def, flags); + +cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static virDomainPtr +bhyveDomainDefineXML(virConnectPtr conn, const char *xml) +{ + bhyveConnPtr privconn = conn->privateData; + virDomainPtr dom = NULL; + virDomainDefPtr def = NULL; + virDomainDefPtr oldDef = NULL; + virDomainObjPtr vm = NULL; + + if ((def = virDomainDefParseString(xml, privconn->caps, privconn->xmlopt, + 1 << VIR_DOMAIN_VIRT_BHYVE, + VIR_DOMAIN_XML_INACTIVE)) == NULL) { + virReportError(VIR_ERR_INVALID_ARG, "%s", + _("Can't parse XML desc")); + goto cleanup; + } + + if (!(vm = virDomainObjListAdd(privconn->domains, def, + privconn->xmlopt, + 0, &oldDef))) + goto cleanup; + + VIR_INFO("Creating domain '%s'", vm->def->name); + dom = virGetDomain(conn, vm->def->name, vm->def->uuid); + if (dom) dom->id = vm->def->id; + + if (virDomainSaveConfig(BHYVE_CONFIG_DIR, vm->def) < 0) { + goto cleanup; + } + +cleanup: + virDomainDefFree(def); + if (vm) + virObjectUnlock(vm); + + return dom; +} + +static int +bhyveConnectListDomains(virConnectPtr conn, int *ids, int maxids) +{ + bhyveConnPtr privconn = conn->privateData; + int n; + + n = virDomainObjListGetActiveIDs(privconn->domains, ids, maxids, + NULL, NULL); + + return n; +} + +static int +bhyveConnectNumOfDomains(virConnectPtr conn) +{ + bhyveConnPtr privconn = conn->privateData; + int count; + + count = virDomainObjListNumOfDomains(privconn->domains, true, + NULL, NULL); + + return count; +} + +static int +bhyveConnectListDefinedDomains(virConnectPtr conn, char **const names, + int maxnames) +{ + bhyveConnPtr privconn = conn->privateData; + int n; + + memset(names, 0, sizeof(*names) * maxnames); + n = virDomainObjListGetInactiveNames(privconn->domains, names, + maxnames, NULL, NULL); + + return n; +} + +static int +bhyveConnectNumOfDefinedDomains(virConnectPtr conn) +{ + bhyveConnPtr privconn = conn->privateData; + int count; + + count = virDomainObjListNumOfDomains(privconn->domains, false, + NULL, NULL); + + return count; +} + +static int +bhyveConnectListAllDomains(virConnectPtr conn, + virDomainPtr **domains, + unsigned int flags) +{ + bhyveConnPtr privconn = conn->privateData; + int ret = -1; + + virCheckFlags(VIR_CONNECT_LIST_DOMAINS_FILTERS_ALL, -1); + + ret = virDomainObjListExport(privconn->domains, conn, domains, + NULL, flags); + + return ret; +} + +static virDomainPtr +bhyveDomainLookupByUUID(virConnectPtr conn, + const unsigned char *uuid) +{ + bhyveConnPtr privconn = conn->privateData; + virDomainObjPtr vm; + virDomainPtr dom = NULL; + + vm = virDomainObjListFindByUUID(privconn->domains, uuid); + + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(uuid, uuidstr); + virReportError(VIR_ERR_NO_DOMAIN, + _("No domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + dom = virGetDomain(conn, vm->def->name, vm->def->uuid); + if (dom) + dom->id = vm->def->id; + +cleanup: + if (vm) + virObjectUnlock(vm); + return dom; +} + +static virDomainPtr bhyveDomainLookupByName(virConnectPtr conn, + const char *name) +{ + bhyveConnPtr privconn = conn->privateData; + virDomainObjPtr vm; + virDomainPtr dom = NULL; + + vm = virDomainObjListFindByName(privconn->domains, name); + + if (!vm) { + virReportError(VIR_ERR_NO_DOMAIN, + _("no domain with matching name '%s'"), name); + goto cleanup; + } + dom = virGetDomain(conn, vm->def->name, vm->def->uuid); + if (dom) + dom->id = vm->def->id; + +cleanup: + if (vm) + virObjectUnlock(vm); + return dom; +} + +static int +bhyveDomainCreate(virDomainPtr dom) +{ + bhyveConnPtr privconn = dom->conn->privateData; + virDomainObjPtr vm; + int ret = -1; + + if (!(vm = bhyveDomObjFromDomain(dom))) + goto cleanup; + + if (virDomainObjIsActive(vm)) { + virReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("Domain is already running")); + goto cleanup; + } + + ret = virBhyveProcessStart(dom->conn, privconn, vm, + VIR_DOMAIN_RUNNING_BOOTED); + +cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static int +bhyveDomainDestroy(virDomainPtr dom) +{ + bhyveConnPtr privconn = dom->conn->privateData; + virDomainObjPtr vm; + int ret = -1; + + if (!(vm = bhyveDomObjFromDomain(dom))) + goto cleanup; + + ret = virBhyveProcessStop(privconn, vm, VIR_DOMAIN_SHUTOFF_DESTROYED); + +cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + +static int +bhyveStateCleanup(void) +{ + VIR_INFO("bhyve state cleanup"); + + if (bhyve_driver == NULL) + return -1; + + virObjectUnref(bhyve_driver->domains); + + virMutexDestroy(&bhyve_driver->lock); + VIR_FREE(bhyve_driver); + + return 0; +} + +static int +bhyveStateInitialize(bool priveleged ATTRIBUTE_UNUSED, + virStateInhibitCallback callback ATTRIBUTE_UNUSED, + void *opaque ATTRIBUTE_UNUSED) +{ + if (!priveleged) { + VIR_INFO("Not running priveleged, disabling driver"); + return 0; + } + + if (VIR_ALLOC(bhyve_driver) < 0) { + return -1; + } + + if (virMutexInit(&bhyve_driver->lock) < 0) { + VIR_FREE(bhyve_driver); + return -1; + } + + if (!(bhyve_driver->caps = bhyveBuildCapabilities())) + goto cleanup; + + if (!(bhyve_driver->xmlopt = virDomainXMLOptionNew(NULL, NULL, NULL))) + goto cleanup; + + if (!(bhyve_driver->domains = virDomainObjListNew())) + goto cleanup; + + if (virFileMakePath(BHYVE_LOG_DIR) < 0) { + virReportSystemError(errno, + _("Failed to mkdir %s"), + BHYVE_LOG_DIR); + goto cleanup; + } + + if (virFileMakePath(BHYVE_STATE_DIR) < 0) { + virReportSystemError(errno, + _("Failed to mkdir %s"), + BHYVE_LOG_DIR); + goto cleanup; + } + + if (virDomainObjListLoadAllConfigs(bhyve_driver->domains, + BHYVE_CONFIG_DIR, + NULL, 0, + bhyve_driver->caps, + bhyve_driver->xmlopt, + 1 << VIR_DOMAIN_VIRT_BHYVE, + NULL, NULL) < 0) + goto cleanup; + + return 0; + +cleanup: + bhyveStateCleanup(); + return -1; +} + + +static virDriver bhyveDriver = { + .no = VIR_DRV_BHYVE, + .name = "bhyve", + .connectOpen = bhyveConnectOpen, /* 0.1.0 */ + .connectClose = bhyveConnectClose, /* 0.1.0 */ + .connectGetVersion = bhyveConnectGetVersion, /* 0.1.0 */ + .connectGetHostname = bhyveConnectGetHostname, /* 0.1.0 */ + .domainGetInfo = bhyveDomainGetInfo, /* 0.1.0 */ + .domainGetState = bhyveDomainGetState, /* 0.1.0 */ + .connectGetCapabilities = bhyveConnectGetCapabilities, /* 0.1.0 */ + .connectListDomains = bhyveConnectListDomains, /* 0.1.0 */ + .connectNumOfDomains = bhyveConnectNumOfDomains, /* 0.1.0 */ + .connectListAllDomains = bhyveConnectListAllDomains, /* 0.1.0 */ + .connectListDefinedDomains = bhyveConnectListDefinedDomains, /* 0.1.0 */ + .connectNumOfDefinedDomains = bhyveConnectNumOfDefinedDomains, /* 0.1.0 */ + .domainCreate = bhyveDomainCreate, /* 0.1.0 */ + .domainDestroy = bhyveDomainDestroy, /* 0.1.0 */ + .domainLookupByUUID = bhyveDomainLookupByUUID, /* 0.1.0 */ + .domainLookupByName = bhyveDomainLookupByName, /* 0.1.0 */ + .domainDefineXML = bhyveDomainDefineXML, /* 0.1.0 */ + .domainGetXMLDesc = bhyveDomainGetXMLDesc, /* 0.1.0 */ +}; + + +static virStateDriver bhyveStateDriver = { + .name = "bhyve", + .stateInitialize = bhyveStateInitialize, + .stateCleanup = bhyveStateCleanup, +}; + +int +bhyveRegister(void) +{ + virRegisterDriver(&bhyveDriver); + virRegisterStateDriver(&bhyveStateDriver); + return 0; +} diff --git a/src/bhyve/bhyve_driver.h b/src/bhyve/bhyve_driver.h new file mode 100644 index 0000000..ffe91e5 --- /dev/null +++ b/src/bhyve/bhyve_driver.h @@ -0,0 +1,28 @@ +/* + * bhyve_driver.h: core driver methods for managing bhyve guests + * + * Copyright (C) 2013 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Roman Bogorodskiy <bogorodskiy(a)gmail.com> + */ + +#ifndef __BHYVE_DRIVER_H__ +# define __BHYVE_DRIVER_H__ + +int bhyveRegister(void); + +#endif /* __BHYVE_DRIVER_H__ */ diff --git a/src/bhyve/bhyve_process.c b/src/bhyve/bhyve_process.c new file mode 100644 index 0000000..89c8a27 --- /dev/null +++ b/src/bhyve/bhyve_process.c @@ -0,0 +1,216 @@ +/* + * bhyve_process.c: bhyve process management + * + * Copyright (C) 2013 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + */ + +#include <config.h> + +#include <fcntl.h> +#include <sys/types.h> +#include <sys/ioctl.h> +#include <net/if.h> +#include <net/if_tap.h> + +#include "bhyve_process.h" +#include "bhyve_command.h" +#include "datatypes.h" +#include "virerror.h" +#include "virlog.h" +#include "virfile.h" +#include "viralloc.h" +#include "vircommand.h" +#include "virstring.h" +#include "virpidfile.h" +#include "virprocess.h" +#include "virnetdev.h" +#include "virnetdevbridge.h" +#include "virnetdevtap.h" + +#define VIR_FROM_THIS VIR_FROM_BHYVE + +int +virBhyveProcessStart(virConnectPtr conn, + bhyveConnPtr driver, + virDomainObjPtr vm, + virDomainRunningReason reason) +{ + char *logfile = NULL; + int logfd = -1; + off_t pos = -1; + char ebuf[1024]; + virCommandPtr cmd = NULL; + bhyveConnPtr privconn = conn->privateData; + int ret = -1, status; + + if (virAsprintf(&logfile, "%s/%s.log", + BHYVE_LOG_DIR, vm->def->name) < 0) + return -1; + + + if ((logfd = open(logfile, O_WRONLY | O_APPEND | O_CREAT, + S_IRUSR|S_IWUSR)) < 0) { + virReportSystemError(errno, + _("Failed to open '%s'"), + logfile); + goto cleanup; + } + + /* Call bhyveload to load a VM */ + if (!(cmd = virBhyveProcessBuildLoadCmd(driver, + vm))) + goto cleanup; + virCommandSetOutputFD(cmd, &logfd); + virCommandSetErrorFD(cmd, &logfd); + + /* Log generated command line */ + virCommandWriteArgLog(cmd, logfd); + if ((pos = lseek(logfd, 0, SEEK_END)) < 0) + VIR_WARN("Unable to seek to end of logfile: %s", + virStrerror(errno, ebuf, sizeof(ebuf))); + + VIR_INFO("Loading domain '%s'", vm->def->name); + if (virCommandRun(cmd, &status) < 0) + goto cleanup; + + if (status != 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Guest failed to load: %d"), status); + goto cleanup; + } + + virCommandFree(cmd); + + VIR_FREE(privconn->pidfile); + if (!(privconn->pidfile = virPidFileBuildPath(BHYVE_STATE_DIR, + vm->def->name))) { + virReportSystemError(errno, + "%s", _("Failed to build pidfile path.")); + goto cleanup; + } + + if (unlink(privconn->pidfile) < 0 && + errno != ENOENT) { + virReportSystemError(errno, + _("Cannot remove state PID file %s"), + privconn->pidfile); + goto cleanup; + } + + /* Call bhyve to start the VM */ + if (!(cmd = virBhyveProcessBuildBhyveCmd(driver, + vm))) + goto cleanup; + + virCommandSetOutputFD(cmd, &logfd); + virCommandSetErrorFD(cmd, &logfd); + virCommandWriteArgLog(cmd, logfd); + virCommandSetPidFile(cmd, privconn->pidfile); + virCommandDaemonize(cmd); + + VIR_INFO("Starting domain '%s'", vm->def->name); + ret = virCommandRun(cmd, NULL); + + if (ret == 0) { + if (virPidFileReadPath(privconn->pidfile, &vm->pid) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Domain %s didn't show up"), vm->def->name); + goto cleanup; + } + + vm->def->id = vm->pid; + virDomainObjSetState(vm, VIR_DOMAIN_RUNNING, reason); + } else { + goto cleanup; + } + +cleanup: + if (ret < 0) { + virCommandPtr destroy_cmd; + if ((destroy_cmd = virBhyveProcessBuildDestroyCmd(driver, vm)) != NULL) + ignore_value(virCommandRun(cmd, NULL)); + } + + VIR_FREE(logfile); + return ret; +} + +int +virBhyveProcessStop(bhyveConnPtr driver, + virDomainObjPtr vm, + virDomainShutoffReason reason ATTRIBUTE_UNUSED) +{ + size_t i; + int ret = -1; + int status; + virCommandPtr cmd = NULL; + + if (!virDomainObjIsActive(vm)) { + VIR_DEBUG("VM '%s' not active", vm->def->name); + return 0; + } + + if (vm->pid <= 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Invalid PID %d for VM"), + (int)vm->pid); + return -1; + } + + /* First, try to kill 'bhyve' process */ + if (virProcessKillPainfully(vm->pid, true) != 0) + VIR_WARN("Failed to gracefully stop bhyve VM '%s' (pid: %d)", + vm->def->name, + (int)vm->pid); + + for (i = 0; i < vm->def->nnets; i++) { + virDomainNetDefPtr net = vm->def->nets[i]; + int actualType = virDomainNetGetActualType(net); + + if (actualType == VIR_DOMAIN_NET_TYPE_BRIDGE) { + ignore_value(virNetDevBridgeRemovePort( + virDomainNetGetActualBridgeName(net), + net->ifname)); + ignore_value(virNetDevTapDelete(net->ifname)); + } + } + + /* No matter if shutdown was successful or not, we + * need to unload the VM */ + if (!(cmd = virBhyveProcessBuildDestroyCmd(driver, vm))) + goto cleanup; + + if (virCommandRun(cmd, &status) < 0) + goto cleanup; + + if (status != 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Guest failed to stop: %d"), status); + goto cleanup; + } + + ret = 0; + + virDomainObjSetState(vm, VIR_DOMAIN_SHUTOFF, reason); + vm->pid = -1; + vm->def->id = -1; + +cleanup: + virCommandFree(cmd); + return ret; +} diff --git a/src/bhyve/bhyve_process.h b/src/bhyve/bhyve_process.h new file mode 100644 index 0000000..70afe0e --- /dev/null +++ b/src/bhyve/bhyve_process.h @@ -0,0 +1,36 @@ +/* + * bhyve_process.h: bhyve process management + * + * Copyright (C) 2013 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + */ + +#ifndef __BHYVE_PROCESS_H__ +# define __BHYVE_PROCESS_H__ + +# include "bhyve_utils.h" + +int virBhyveProcessStart(virConnectPtr conn, + bhyveConnPtr driver, + virDomainObjPtr vm, + virDomainRunningReason reason); + +int virBhyveProcessStop(bhyveConnPtr driver, + virDomainObjPtr vm, + virDomainShutoffReason reason); + +#endif /* __BHYVE_PROCESS_H__ */ diff --git a/src/bhyve/bhyve_utils.h b/src/bhyve/bhyve_utils.h new file mode 100644 index 0000000..ed503cd --- /dev/null +++ b/src/bhyve/bhyve_utils.h @@ -0,0 +1,48 @@ +/* + * bhyve_utils.h: bhyve utils + * + * Copyright (C) 2013 Roman Bogorodskiy + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + */ + +#ifndef __BHYVE_UTILS_H__ +# define __BHYVE_UTILS_H__ + +# include "driver.h" +# include "domain_conf.h" +# include "configmake.h" +# include "virthread.h" + +# define BHYVE_CONFIG_DIR (SYSCONFDIR "/libvirt/bhyve") +# define BHYVE_STATE_DIR (LOCALSTATEDIR "/run/libvirt/bhyve") +# define BHYVE_LOG_DIR (LOCALSTATEDIR "/log/libvirt/bhyve") + +struct _bhyveConn { + virMutex lock; + virDomainObjListPtr domains; + virCapsPtr caps; + virDomainXMLOptionPtr xmlopt; + char *pidfile; +}; + +typedef struct _bhyveConn bhyveConn; +typedef struct _bhyveConn *bhyveConnPtr; + +void bhyveDriverLock(bhyveConnPtr driver); +void bhyveDriverUnlock(bhyveConnPtr driver); + +#endif /* __BHYVE_UTILS_H__ */ diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 28e24f9..26f97d3 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -122,7 +122,8 @@ VIR_ENUM_IMPL(virDomainVirt, VIR_DOMAIN_VIRT_LAST, "hyperv", "vbox", "phyp", - "parallels") + "parallels", + "bhyve") VIR_ENUM_IMPL(virDomainBoot, VIR_DOMAIN_BOOT_LAST, "fd", diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index d8f2e49..6d73de6 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -196,6 +196,7 @@ enum virDomainVirtType { VIR_DOMAIN_VIRT_VBOX, VIR_DOMAIN_VIRT_PHYP, VIR_DOMAIN_VIRT_PARALLELS, + VIR_DOMAIN_VIRT_BHYVE, VIR_DOMAIN_VIRT_LAST }; diff --git a/src/driver.h b/src/driver.h index 5f4cd8d..fbfaac4 100644 --- a/src/driver.h +++ b/src/driver.h @@ -47,6 +47,7 @@ typedef enum { VIR_DRV_LIBXL = 14, VIR_DRV_HYPERV = 15, VIR_DRV_PARALLELS = 16, + VIR_DRV_BHYVE = 17, } virDrvNo; diff --git a/src/libvirt.c b/src/libvirt.c index 9cc5b1c..54ecdc8 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -96,6 +96,9 @@ #ifdef WITH_PARALLELS # include "parallels/parallels_driver.h" #endif +#ifdef WITH_BHYVE +# include "bhyve/bhyve_driver.h" +#endif #define VIR_FROM_THIS VIR_FROM_NONE diff --git a/src/util/virerror.c b/src/util/virerror.c index f0c159f..74c6807 100644 --- a/src/util/virerror.c +++ b/src/util/virerror.c @@ -124,6 +124,7 @@ VIR_ENUM_IMPL(virErrorDomain, VIR_ERR_DOMAIN_LAST, "Access Manager", /* 55 */ "Systemd", + "Bhyve", ) -- 1.8.4.3

1 0

[libvirt] [PATCH] spice: don't force user to specify spicevmc channel
by Martin Kletzander 03 Feb '14

03 Feb '14

We support only one spicevmc channel name anyway and the code is prepared to use the default one, there's only one check missing. I'm not adding it to documentation in case there is another channel name aded in the future, but this helps people using virsh for defining domains with spice vdagent. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- I extended the context to see what I meant by "the code is already prepared to use the default one". src/qemu/qemu_command.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 2db745a..2124477 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6127,30 +6127,31 @@ qemuBuildVirtioSerialPortDevStr(virDomainChrDefPtr dev, "%s", _("virtio serial device has invalid address type")); goto error; } virBufferAsprintf(&buf, ",bus=" QEMU_VIRTIO_SERIAL_PREFIX "%d.%d", dev->info.addr.vioserial.controller, dev->info.addr.vioserial.bus); virBufferAsprintf(&buf, ",nr=%d", dev->info.addr.vioserial.port); } if (dev->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL && dev->source.type == VIR_DOMAIN_CHR_TYPE_SPICEVMC && + dev->target.name && STRNEQ_NULLABLE(dev->target.name, "com.redhat.spice.0")) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Unsupported spicevmc target name '%s'"), dev->target.name); goto error; } if (!(dev->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL && dev->source.type == VIR_DOMAIN_CHR_TYPE_SPICEVMC && virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_SPICEVMC))) { virBufferAsprintf(&buf, ",chardev=char%s,id=%s", dev->info.alias, dev->info.alias); if (dev->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL) { virBufferAsprintf(&buf, ",name=%s", dev->target.name ? dev->target.name : "com.redhat.spice.0"); -- 1.8.5.3

2 2

[libvirt] [PATCH v2] Resolve Coverity dead_error_begin
by John Ferlan 31 Jan '14

31 Jan '14

Coverity complains about default: label in libxl_driver.c not be able to be reached. It's by design for the code and since it's not necessary in the code nor does it elicit any compiler/make check warnings - just remove it rather than adding a coverity[dead_error_begin] tag. While I'm at it, lxc_driver.c and nodeinfo.c have the same design, so I removed the default labels and the existing coverity tags. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/libxl/libxl_driver.c | 3 --- src/lxc/lxc_driver.c | 15 --------------- src/nodeinfo.c | 4 ---- 3 files changed, 22 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index fc0efa2..cb3deec 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -4167,9 +4167,6 @@ libxlDomainGetNumaParameters(virDomainPtr dom, nodeset = NULL; break; - - default: - break; } } diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 73a2986..138c706 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -950,11 +950,6 @@ lxcDomainGetMemoryParameters(virDomainPtr dom, VIR_TYPED_PARAM_ULLONG, val) < 0) goto cleanup; break; - - /* coverity[dead_error_begin] */ - default: - break; - /* should not hit here */ } } @@ -2626,11 +2621,6 @@ lxcDomainGetBlkioParameters(virDomainPtr dom, param->value.s) < 0) goto cleanup; break; - - /* coverity[dead_error_begin] */ - default: - break; - /* should not hit here */ } } } else if (flags & VIR_DOMAIN_AFFECT_CONFIG) { @@ -2818,11 +2808,6 @@ lxcDomainGetBlkioParameters(virDomainPtr dom, goto cleanup; } break; - - /* coverity[dead_error_begin] */ - default: - break; - /* should not hit here */ } } } diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 6ebfb4b..22b139f 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -1478,10 +1478,6 @@ nodeGetMemoryParameters(virTypedParameterPtr params ATTRIBUTE_UNUSED, return -1; break; - - /* coverity[dead_error_begin] */ - default: - break; } } -- 1.8.4.2

2 1

[libvirt] [PATCH] libxl: Resove Coverity dead_error_begin
by John Ferlan 31 Jan '14

31 Jan '14

Coverity complains about default: label not be able to be reached. It's by design for the code, so just add a Coverity tag for dead_error_begin Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/libxl/libxl_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index fc0efa2..513a597 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -4168,6 +4168,7 @@ libxlDomainGetNumaParameters(virDomainPtr dom, break; + /* coverity[dead_error_begin] */ default: break; } -- 1.8.4.2

2 3

[libvirt] [PATCH] virnetdevbandwidthtest: Introduce some more tests
by Michal Privoznik 31 Jan '14

31 Jan '14

And while doing this, fix one error raised by coverity. With current code, @actual_cmd is allowed to be NULL for the whole run of testVirNetDevBandwidthSet. However, if something else was expected, the @actal_cmd is passed to virtTestDifference which dereference it immediately. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/virnetdevbandwidthtest.c | 29 ++++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c index 5bcab12..073fdf8 100644 --- a/tests/virnetdevbandwidthtest.c +++ b/tests/virnetdevbandwidthtest.c @@ -92,7 +92,9 @@ testVirNetDevBandwidthSet(const void *data) } if (STRNEQ_NULLABLE(info->exp_cmd, actual_cmd)) { - virtTestDifference(stderr, info->exp_cmd, actual_cmd); + virtTestDifference(stderr, + NULLSTR(info->exp_cmd), + NULLSTR(actual_cmd)); goto cleanup; } @@ -121,6 +123,31 @@ mymain(void) } while (0) + DO_TEST_SET(NULL, NULL); + + DO_TEST_SET(("<bandwidth/>"), + (TC " qdisc del dev eth0 root\n" + TC " qdisc del dev eth0 ingress\n")); + + DO_TEST_SET(("<bandwidth>" + " <inbound average='1024'/>" + "</bandwidth>"), + (TC " qdisc del dev eth0 root\n" + TC " qdisc del dev eth0 ingress\n" + TC " qdisc add dev eth0 root handle 1: htb default 1\n" + TC " class add dev eth0 parent 1: classid 1:1 htb rate 1024kbps\n" + TC " qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n" + TC " filter add dev eth0 parent 1:0 protocol ip handle 1 fw flowid 1\n")); + + DO_TEST_SET(("<bandwidth>" + " <outbound average='1024'/>" + "</bandwidth>"), + (TC " qdisc del dev eth0 root\n" + TC " qdisc del dev eth0 ingress\n" + TC " qdisc add dev eth0 ingress\n" + TC " filter add dev eth0 parent ffff: protocol ip u32 match ip src 0.0.0.0/0 " + "police rate 1024kbps burst 1024kb mtu 64kb drop flowid :1\n")); + DO_TEST_SET(("<bandwidth>" " <inbound average='1' peak='2' floor='3' burst='4'/>" " <outbound average='5' peak='6' burst='7'/>" -- 1.8.5.2

3 2

[libvirt] Warnings messages in logs from DHCP snooping code
by Daniel P. Berrange 31 Jan '14

31 Jan '14

I'm noticing that VMs which have nwfilters associated with them will apparently always cause the following errors to appear in the logs upon VM shutdown: 2013-02-27 17:02:18.709+0000: 9669: error : virNWFilterDHCPSnoopEnd:2131 : internal error ifname "vnet0" not in key map 2013-02-27 17:02:18.715+0000: 9669: error : virNetDevGetIndex:653 : Unable to get index for interface vnet0: No such device Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

3 2

[libvirt] [PATCH] virnetdevbandwidthtest: fix hard coded /sbin/tc
by Cédric Bosdonnat 31 Jan '14

31 Jan '14

On openSuse, (and possibly other distros), tc isn't located in /sbin/tc. To get rid of that problem, use TC constant instead of hard coded /sbin/tc in the expected string --- tests/virnetdevbandwidthtest.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c index 609deb8..5bcab12 100644 --- a/tests/virnetdevbandwidthtest.c +++ b/tests/virnetdevbandwidthtest.c @@ -125,14 +125,14 @@ mymain(void) " <inbound average='1' peak='2' floor='3' burst='4'/>" " <outbound average='5' peak='6' burst='7'/>" "</bandwidth>"), - ("/sbin/tc qdisc del dev eth0 root\n" - "/sbin/tc qdisc del dev eth0 ingress\n" - "/sbin/tc qdisc add dev eth0 root handle 1: htb default 1\n" - "/sbin/tc class add dev eth0 parent 1: classid 1:1 htb rate 1kbps ceil 2kbps burst 4kb\n" - "/sbin/tc qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n" - "/sbin/tc filter add dev eth0 parent 1:0 protocol ip handle 1 fw flowid 1\n" - "/sbin/tc qdisc add dev eth0 ingress\n" - "/sbin/tc filter add dev eth0 parent ffff: protocol ip u32 match ip src 0.0.0.0/0 " + (TC " qdisc del dev eth0 root\n" + TC " qdisc del dev eth0 ingress\n" + TC " qdisc add dev eth0 root handle 1: htb default 1\n" + TC " class add dev eth0 parent 1: classid 1:1 htb rate 1kbps ceil 2kbps burst 4kb\n" + TC " qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n" + TC " filter add dev eth0 parent 1:0 protocol ip handle 1 fw flowid 1\n" + TC " qdisc add dev eth0 ingress\n" + TC " filter add dev eth0 parent ffff: protocol ip u32 match ip src 0.0.0.0/0 " "police rate 5kbps burst 7kb mtu 64kb drop flowid :1\n")); return ret; -- 1.8.5.2

2 1

[libvirt] [PATCH] [RFC] libxl: add discard support to libxl_device_disk
by Olaf Hering 31 Jan '14

31 Jan '14

Translate libvirt discard settings into libxl-4.5 discard settings. It makes use of upcoming changes for xen-4.5: http://lists.xenproject.org/archives/html/xen-devel/2014-01/msg02632.html This patch is not compile tested. How is a config file supposed to look like to make use of a discard=on|off flag for a given disk image? Signed-off-by: Olaf Hering <olaf(a)aepfle.de> Cc: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libxl_conf.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/src/libxl/libxl_conf.c b/src/libxl/libxl_conf.c index 4cefadf..8fc5e75 100644 --- a/src/libxl/libxl_conf.c +++ b/src/libxl/libxl_conf.c @@ -707,6 +707,26 @@ error: return -1; } +static void +libxlDiskSetDiscard(libxl_device_disk *x_disk, int discard) +{ + if (!x_disk->readwrite) + return; +#if defined(LIBXL_HAVE_LIBXL_DEVICE_DISK_DISCARD_ENABLE) + /* libxl 4.5 */ + switch (discard) { + case VIR_DOMAIN_DISK_DISCARD_DEFAULT: + case VIR_DOMAIN_DISK_DISCARD_UNMAP: + x_disk->discard_enable = 1; + break; + default: + case VIR_DOMAIN_DISK_DISCARD_IGNORE: + x_disk->discard_enable = 0; + break; + } +#endif +} + int libxlMakeDisk(virDomainDiskDefPtr l_disk, libxl_device_disk *x_disk) { @@ -817,6 +837,7 @@ libxlMakeDisk(virDomainDiskDefPtr l_disk, libxl_device_disk *x_disk) x_disk->removable = 1; x_disk->readwrite = !l_disk->readonly; x_disk->is_cdrom = l_disk->device == VIR_DOMAIN_DISK_DEVICE_CDROM ? 1 : 0; + libxlDiskSetDiscard(x_disk, l_disk->discard); if (l_disk->transient) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("libxenlight does not support transient disks"));

2 2

[libvirt] [PATCH 0/2] Fix the scsi util test failure
by Osier Yang 31 Jan '14

31 Jan '14

Detected by Pavel reported. Osier Yang (2): util: Accept test data path for scsi device's sg_path tests: Modify the scsi util tests src/util/virscsi.c | 3 ++- tests/virscsidata/1:0:0:0/block/sdh/dev | 1 + tests/virscsidata/1:0:0:0/block/sr0/dev | 1 - tests/virscsidata/1:0:0:0/scsi_generic/sg1/dev | 1 - tests/virscsidata/1:0:0:0/scsi_generic/sg8/dev | 1 + tests/virscsidata/sg0 | 0 tests/virscsidata/sg8 | 0 tests/virscsitest.c | 4 ++-- 8 files changed, 6 insertions(+), 5 deletions(-) create mode 100644 tests/virscsidata/1:0:0:0/block/sdh/dev delete mode 100644 tests/virscsidata/1:0:0:0/block/sr0/dev delete mode 100644 tests/virscsidata/1:0:0:0/scsi_generic/sg1/dev create mode 100644 tests/virscsidata/1:0:0:0/scsi_generic/sg8/dev create mode 100644 tests/virscsidata/sg0 create mode 100644 tests/virscsidata/sg8 -- 1.8.1.4

2 6

[libvirt] [PATCH v3 0/4] Fix deadlock in nwfilter code
by Daniel P. Berrange 30 Jan '14

30 Jan '14

Since we introduced fine grained locking into the QEMU driver so that VM start can run in parallel, we appear to have caused a race with the nwfilter code. In particular since we no longer hold the global QEMU driver lock for the duration of VM startup we have a lock ordering flaw. This results in deadlock when nwfilter operations happen in parallel with VM startup. This also affects the LXC driver. This patch series attempts to address the problem https://bugzilla.redhat.com/show_bug.cgi?id=929412 the removal of the windows thread impl isn't strictly required, I just didn't want to waste time creating a read/write lock impl for Windows threads. See also this callgraph http://berrange.fedorapeople.org/nwfilter.ps In v3: - Removed addition of extra lock for serializing iptables since it was overkill - Added patch to make libvirt compile with mingw32-pthreads from Fedora 19 and earlier - Fix errno handling with rwlock creation In v2: - Re-ordered patches and squashed two together - Add missing locking in state reload function - Don't remove locking from virNWFilterInstantiateFilterLate *** BLURB HERE *** Daniel P. Berrange (4): Add a read/write lock implementation Fix pthread_sigmask check for mingw32 without winpthreads Remove windows thread implementation in favour of pthreads Push nwfilter update locking up to top level configure.ac | 21 +- src/Makefile.am | 4 - src/conf/nwfilter_conf.c | 23 +- src/conf/nwfilter_conf.h | 3 +- src/libvirt_private.syms | 8 +- src/lxc/lxc_driver.c | 6 + src/nwfilter/nwfilter_driver.c | 10 +- src/nwfilter/nwfilter_gentech_driver.c | 6 +- src/qemu/qemu_driver.c | 6 + src/uml/uml_driver.c | 4 + src/util/virthread.c | 293 +++++++++++++++++++++++- src/util/virthread.h | 51 ++++- src/util/virthreadpthread.c | 278 ----------------------- src/util/virthreadpthread.h | 49 ---- src/util/virthreadwin32.c | 396 --------------------------------- src/util/virthreadwin32.h | 53 ----- 16 files changed, 388 insertions(+), 823 deletions(-) delete mode 100644 src/util/virthreadpthread.c delete mode 100644 src/util/virthreadpthread.h delete mode 100644 src/util/virthreadwin32.c delete mode 100644 src/util/virthreadwin32.h -- 1.8.4.2

3 7

Re: [libvirt] [Xen-devel] [PATCH 00/12] libxl: fork: SIGCHLD flexibility
by Jim Fehlig 30 Jan '14

30 Jan '14

[Adding libvirt list...] Ian Jackson wrote: > Jim Fehlig writes ("Re: [Xen-devel] [PATCH 00/12] libxl: fork: SIGCHLD flexibility"): > >> BTW, I only see the crash when the save/restore script is running. I >> stopped the other scripts and domains, running only save/restore on a >> single domain, and see the crash rather quickly (within 10 iterations). >> > > I'll look at the libvirt code, but: > > With a recurring timeout, how can you ever know it's cancelled ? > There might be threads out there, which don't hold any locks, which > are in the process of executing a callback for a timeout. That might > be arbitrarily delayed from the pov of the rest of the program. > > E.g.: > > Thread A Thread B > > invoke some libxl operation > X do some libxl stuff > X register timeout (libxl) > XV record timeout info > X do some more libxl stuff > ... > X do some more libxl stuff > X deregister timeout (libxl internal) > X converted to request immediate timeout > XV record new timeout info > X release libvirt event loop lock > entering libvirt event loop > V observe timeout is immediate > V need to do callback > call libxl driver > > entering libvirt event loop > V observe timeout is immediate > V need to do callback > call libxl driver > call libxl > X libxl sees timeout is live > X libxl does libxl stuff > libxl driver deregisters > V record lack of timeout > free driver's timeout struct > call libxl > X libxl sees timeout is dead > X libxl does nothing > libxl driver deregisters > V CRASH due to deregistering > V already-deregistered timeout > > If this is how things are, then I think there is no sane way to use > libvirt's timeouts (!) > Looking at the libvirt code again, it seems a single thread services the event loop. See virNetServerRun() in src/util/virnetserver.c. Indeed, I see the same thread ID in all the timer and fd callbacks. One of the libvirt core devs can correct me if I'm wrong. Regards, Jim > In principle I guess the driver could keep its per-timeout structs > around forever and remember whether they've been deregistered or not. > Each one would have to have a lock in it. > > But if you think about it, if you have 10 threads all running the > event loop and you set a timeout to zero, doesn't that mean that every > thread's event loop should do the timeout callback as fast as it can ? > That could be a lot of wasted effort. > > The best solution would appear to be to provide a non-recurring > callback. > > >> I'm not so thrilled with the timeout handling code in the libvirt libxl >> driver. The driver maintains a list of active timeouts because IIRC, >> there were cases when the driver received timeout deregistrations when >> calling libxl_ctx_free, at which point some of the associated structures >> were freed. The idea was to call libxl_osevent_occurred_timeout on any >> active timeouts before freeing libxlDomainObjPrivate and its contents. >> > > libxl does deregister fd callbacks in libxl_ctx_free. > > But libxl doesn't currently "deregister" any timeouts in > libxl_ctx_free; indeed it would be a bit daft for it to do so as at > libxl_ctx_free there are no aos running so there would be nothing to > time out. > > But there is a difficulty with timeouts which libxl has set to occur > immediately but which have not yet actually had the callback. The the > application cannot call libxl_ctx_free with such timeouts outstanding, > because that would imply later calling back into libxl with a stale > ctx. > > (Looking at the code I see that the "nexi" are never actually freed. > Bah.) > > Thanks, > Ian. > >

3 8

[libvirt] [PATCH] docs/page.xls: remove unnecessary namespace attribute
by Pavel Hrdina 30 Jan '14

30 Jan '14

It breaks the build on RHEL-5.10 and because it's only optional we could remove it from the code. The default namespace will be used. This hunk was introduced by commit 237a088ba44633b08e0528fe5103d99b1e213b71. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- docs/page.xsl | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/page.xsl b/docs/page.xsl index d560c09..7de238e 100644 --- a/docs/page.xsl +++ b/docs/page.xsl @@ -187,7 +187,7 @@ </xsl:template> <xsl:template match="html:h2 | html:h3 | html:h4 | html:h5 | html:h6" mode="content"> - <xsl:element name="{name()}" namespace="{namespace-uri()}"> + <xsl:element name="{name()}"> <xsl:apply-templates mode="copy" /> <xsl:if test="./html:a/@name"> <a class="headerlink" href="#{html:a/@name}" title="Permalink to this headline">¶</a> @@ -200,7 +200,7 @@ </xsl:template> <xsl:template match="node()" mode="copy"> - <xsl:element name="{name()}" namespace="{namespace-uri()}"> + <xsl:element name="{name()}"> <xsl:copy-of select="./@*"/> <xsl:apply-templates mode="copy" /> </xsl:element> -- 1.8.3.1

2 2

[libvirt] [PATCH] network: only prevent forwarding of DNS requests for unqualified names
by Laine Stump 30 Jan '14

30 Jan '14

In commit f386825 we began adding the option "--local=/$mydomain/" to all dnsmasq commandlines (later changed to "local=/$mydomain/" when we moved the options from the commandline to a conf file) with the stated reason of preventing forwarding of DNS queries for names that weren't fully qualified domain names ("FQDN", i.e. a name that included some "."s and a domain name). The original patch on the list, and discussion about it, is here: https://www.redhat.com/archives/libvir-list/2012-August/msg01594.html When a domain name isn't specified (no <domain> element in the network definition), the corresponding addition of "local=//" will prevent forwarding of domain-less requests to the virtualization host's DNS resolver, but if a domain *is* specified, the addition of "local=/domain/" will prevent forwarding of any requests for names within that domain that aren't resolvable by libvirt's dnsmasq itself. An example of the problems this causes: let's say a network is defined with: <domain name='example.com'/> <dhcp> .. <host mac='52:54:00:11:22:33' ip='1.2.3.4' name='myguest'/> </dhcp> This results in "local=/example.com/" being added to the dnsmasq options. If a guest requests "myguest" or "myguest.example.com", that will be resolved by dnsmasq. If the guest asks for "www.example.com", dnsmasq will not know the answer, but instead of forwarding it to the host, it will return NOT FOUND to the guest. In most cases that isn't the behavior an admin is looking for. Really we should have been just including the option "--local=//" in all cases, so that (unresolvable by dnsmasq) requests for names without a domain would be treated as "local to dnsmasq" and not forwarded, but all other requests would be forwarded. That's what this patch does. --- I'm debating whether there is any value at all to maintaining the previous behavior of "don't forward unresolved requests for hosts in the network's defined domain", or if it should just be considered purely a bug. If so, I think it shouldn't be the default behavior, and should be controlled by a new attribute to the <domain> element, something like this: <domain name='example.com' forwardUnresolved='no'/> (this would default to yes). Any opinions on 1) whether or not this is needed, and 2) if so, any better name for the option? (it would be nice if it could default to 'no' or 'local-only' (which was == 0) or something else that didn't require a non-0 default or a strange double-negative name). src/network/bridge_driver.c | 12 +++++------- tests/networkxml2confdata/dhcp6-network.conf | 2 +- tests/networkxml2confdata/nat-network-dns-hosts.conf | 2 +- tests/networkxml2confdata/netboot-network.conf | 2 +- tests/networkxml2confdata/netboot-proxy-network.conf | 2 +- 5 files changed, 9 insertions(+), 11 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 80c5acb..b1382e4 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -723,14 +723,12 @@ networkDnsmasqConfContents(virNetworkObjPtr network, network->def->domain); } - if (network->def->domain || !network->def->dns.forwardPlainNames) { - /* need to specify local even if no domain specified, unless - * the config says we should forward "plain" names (i.e. not - * fully qualified, no '.' characters) + if (!network->def->dns.forwardPlainNames) { + /* need to specify local=// whether or not a domain is + * specified, unless the config says we should forward "plain" + * names (i.e. not fully qualified, no '.' characters) */ - virBufferAsprintf(&configbuf, - "local=/%s/\n", - network->def->domain ? network->def->domain : ""); + virBufferAsprintf(&configbuf, "local=//\n"); } if (pidfile) diff --git a/tests/networkxml2confdata/dhcp6-network.conf b/tests/networkxml2confdata/dhcp6-network.conf index 5fde07f..24ac7fa 100644 --- a/tests/networkxml2confdata/dhcp6-network.conf +++ b/tests/networkxml2confdata/dhcp6-network.conf @@ -8,7 +8,7 @@ strict-order domain-needed domain=mynet expand-hosts -local=/mynet/ +local=// except-interface=lo bind-dynamic interface=virbr0 diff --git a/tests/networkxml2confdata/nat-network-dns-hosts.conf b/tests/networkxml2confdata/nat-network-dns-hosts.conf index 2577882..bb1d5f4 100644 --- a/tests/networkxml2confdata/nat-network-dns-hosts.conf +++ b/tests/networkxml2confdata/nat-network-dns-hosts.conf @@ -8,7 +8,7 @@ strict-order domain-needed domain=example.com expand-hosts -local=/example.com/ +local=// except-interface=lo bind-dynamic interface=virbr0 diff --git a/tests/networkxml2confdata/netboot-network.conf b/tests/networkxml2confdata/netboot-network.conf index b6f3c23..37fef40 100644 --- a/tests/networkxml2confdata/netboot-network.conf +++ b/tests/networkxml2confdata/netboot-network.conf @@ -8,7 +8,7 @@ strict-order domain-needed domain=example.com expand-hosts -local=/example.com/ +local=// except-interface=lo bind-interfaces listen-address=192.168.122.1 diff --git a/tests/networkxml2confdata/netboot-proxy-network.conf b/tests/networkxml2confdata/netboot-proxy-network.conf index 1e969fa..ae34aa0 100644 --- a/tests/networkxml2confdata/netboot-proxy-network.conf +++ b/tests/networkxml2confdata/netboot-proxy-network.conf @@ -8,7 +8,7 @@ strict-order domain-needed domain=example.com expand-hosts -local=/example.com/ +local=// except-interface=lo bind-interfaces listen-address=192.168.122.1 -- 1.8.3.1

2 4

[libvirt] [libvirt-designer][PATCH] gvir_designer_domain_add_disk_full: Adapt to latest libvirt-glib
by Michal Privoznik 30 Jan '14

30 Jan '14

With libvirt-glib update, I've noticed a deprecation warning: libvirt-designer-domain.c: In function 'gvir_designer_domain_add_disk_full': libvirt-designer-domain.c:1610:9: error: 'gvir_config_domain_disk_set_driver_type' is deprecated (declared at /usr/include/libvirt-gconfig-1.0/libvirt-gconfig/libvirt-gconfig-domain-disk.h:145): Use 'gvir_config_domain_disk_set_driver_format' instead [-Werror=deprecated-declarations] gvir_config_domain_disk_set_driver_type(disk, format); ^ cc1: all warnings being treated as errors Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- libvirt-designer/Makefile.am | 1 + libvirt-designer/libvirt-designer-domain.c | 22 +++++++++++-- libvirt-designer/libvirt-designer-internal.c | 46 ++++++++++++++++++++++++++++ libvirt-designer/libvirt-designer-internal.h | 2 ++ 4 files changed, 68 insertions(+), 3 deletions(-) create mode 100644 libvirt-designer/libvirt-designer-internal.c diff --git a/libvirt-designer/Makefile.am b/libvirt-designer/Makefile.am index 7becfd2..bda4240 100644 --- a/libvirt-designer/Makefile.am +++ b/libvirt-designer/Makefile.am @@ -25,6 +25,7 @@ DESIGNER_HEADER_FILES = \ libvirt-designer-domain.h \ $(NULL) DESIGNER_SOURCE_FILES = \ + libvirt-designer-internal.c \ libvirt-designer-main.c \ libvirt-designer-domain.c \ $(NULL) diff --git a/libvirt-designer/libvirt-designer-domain.c b/libvirt-designer/libvirt-designer-domain.c index 848dfea..5475c79 100644 --- a/libvirt-designer/libvirt-designer-domain.c +++ b/libvirt-designer/libvirt-designer-domain.c @@ -1580,6 +1580,7 @@ gvir_designer_domain_add_disk_full(GVirDesignerDomain *design, { GVirDesignerDomainPrivate *priv = design->priv; GVirConfigDomainDisk *disk = NULL; + GVirConfigDomainDiskDriver *driver = NULL; GVirConfigDomainDiskBus bus; gchar *target_gen = NULL; const char *driver_name; @@ -1602,12 +1603,27 @@ gvir_designer_domain_add_disk_full(GVirDesignerDomain *design, break; } + driver = gvir_config_domain_disk_driver_new(); + gvir_config_domain_disk_driver_set_name(driver, driver_name); + if (format) { + int fmt; + + fmt = gvir_designer_genum_get_value(GVIR_CONFIG_TYPE_DOMAIN_DISK_FORMAT, + format, -1); + + if (fmt == -1) { + g_set_error(error, GVIR_DESIGNER_DOMAIN_ERROR, 0, + "Unknown disk format: %s", format); + goto error; + } + + gvir_config_domain_disk_driver_set_format(driver, fmt); + } + disk = gvir_config_domain_disk_new(); gvir_config_domain_disk_set_type(disk, type); gvir_config_domain_disk_set_source(disk, path); - gvir_config_domain_disk_set_driver_name(disk, driver_name); - if (format) - gvir_config_domain_disk_set_driver_type(disk, format); + gvir_config_domain_disk_set_driver(disk, driver); controller = gvir_designer_domain_get_preferred_disk_controller(design, NULL); if (controller == NULL) diff --git a/libvirt-designer/libvirt-designer-internal.c b/libvirt-designer/libvirt-designer-internal.c new file mode 100644 index 0000000..bdfa5d1 --- /dev/null +++ b/libvirt-designer/libvirt-designer-internal.c @@ -0,0 +1,46 @@ +/* + * libvirt-designer-internal.c: libvirt-designer helpers + * + * Copyright (C) 2014 Red Hat, Inc. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Authors: Michal Privoznik <mprivozn(a)redhat.com> + */ + +#include <config.h> + +#include "libvirt-designer/libvirt-designer.h" +#include "libvirt-designer/libvirt-designer-internal.h" + +G_GNUC_INTERNAL int +gvir_designer_genum_get_value(GType enum_type, const char *nick, + gint default_value) +{ + GEnumClass *enum_class; + GEnumValue *enum_value; + + g_return_val_if_fail(G_TYPE_IS_ENUM(enum_type), default_value); + g_return_val_if_fail(nick != NULL, default_value); + + enum_class = g_type_class_ref(enum_type); + enum_value = g_enum_get_value_by_nick(enum_class, nick); + g_type_class_unref(enum_class); + + if (enum_value != NULL) + return enum_value->value; + + g_return_val_if_reached(default_value); +} diff --git a/libvirt-designer/libvirt-designer-internal.h b/libvirt-designer/libvirt-designer-internal.h index e95edfc..fd3db2c 100644 --- a/libvirt-designer/libvirt-designer-internal.h +++ b/libvirt-designer/libvirt-designer-internal.h @@ -24,4 +24,6 @@ #ifndef __LIBVIRT_DESIGNER_INTERNAL_H__ #define __LIBVIRT_DESIGNER_INTERNAL_H__ +int gvir_designer_genum_get_value(GType enum_type, const char *nick, gint default_value); + #endif /* __LIBVIRT_DESIGNER_INTERNAL_H__ */ -- 1.8.5.2

2 1

[libvirt] [PATCH v2 0/3] Honor blacklist for modprobe command
by John Ferlan 30 Jan '14

30 Jan '14

Original patch here: http://www.redhat.com/archives/libvir-list/2014-January/msg01408.html Changes since v1: * Add virkmod.{c,h} to perform the various modprobe commands virModprobeConfig() -> modprobe -c virModprobeLoad() -> modprobe driver virModprobeUnload() -> modprobe -r driver virModprobeUseBlacklist() -> modprobe -b driver * virkmodtest.c to test the functions Caveat emptor... The config is the only test run by default. Loading and Unloading modules is something to be done carefully. * virpci.c Use the virModprobe*() API's instead of a direct MODPROBE * .gitignore Added the /tests/virkmodtest and /examples/domain-events/ which seems to have appeared recently. I can separate them out if desired. NOTE: Since I have no messages in virkmod.c, modifying po/POTFILES.in was deemed unnecessary by a 'make check-syntax' John Ferlan (3): utils: Introduce functions for modprobe tests: Add test for new virkmod functions Honor blacklist for modprobe command .gitignore | 2 + src/Makefile.am | 1 + src/libvirt_private.syms | 7 ++ src/util/virkmod.c | 138 ++++++++++++++++++++++++++++++++ src/util/virkmod.h | 35 +++++++++ src/util/virpci.c | 49 +++++++++--- tests/Makefile.am | 5 ++ tests/virkmodtest.c | 201 +++++++++++++++++++++++++++++++++++++++++++++++ 8 files changed, 429 insertions(+), 9 deletions(-) create mode 100644 src/util/virkmod.c create mode 100644 src/util/virkmod.h create mode 100644 tests/virkmodtest.c -- 1.8.4.2

3 9

[libvirt] [libvirt-designer][PATCH] virtxml: Rename to virt-designer
by Michal Privoznik 30 Jan '14

30 Jan '14

Our example utility is called virtxml. However with Cole's latest work, where he introduced virt-xml binary for editing libvirt XML on CLI [1]. It would be better if our example is called virt-designer so we don't confuse users. 1: https://www.redhat.com/archives/libvir-list/2014-January/msg01226.html Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- .gitignore | 2 +- cfg.mk | 2 +- configure.ac | 2 +- examples/Makefile.am | 10 +++++----- examples/{virtxml.c => virt-designer.c} | 32 ++++++++++++++++---------------- libvirt-designer.spec.in | 4 ++-- 6 files changed, 26 insertions(+), 26 deletions(-) rename examples/{virtxml.c => virt-designer.c} (96%) diff --git a/.gitignore b/.gitignore index 6b05db9..09e15a5 100644 --- a/.gitignore +++ b/.gitignore @@ -15,7 +15,7 @@ Makefile.in /config.log /config.status /configure -/examples/virtxml +/examples/virt-designer /libtool /libvirt-designer-1.0.pc /libvirt-designer.spec diff --git a/cfg.mk b/cfg.mk index 1e4b31d..f9ab008 100644 --- a/cfg.mk +++ b/cfg.mk @@ -104,7 +104,7 @@ sc_copyright_format: prev_version_file = /dev/null -exclude_file_name_regexp--sc_bindtextdomain = ^libvirt-designer/(test|libvirt-designer-main\.c)|^examples/virtxml\.c$$ +exclude_file_name_regexp--sc_bindtextdomain = ^libvirt-designer/(test|libvirt-designer-main\.c)|^examples/virt-designer\.c$$ exclude_file_name_regexp--sc_preprocessor_indentation = ^*/*.[ch]$$ diff --git a/configure.ac b/configure.ac index bad199b..595fea4 100644 --- a/configure.ac +++ b/configure.ac @@ -96,7 +96,7 @@ LIBVIRT_DESIGNER_COVERAGE LIBVIRT_DESIGNER_INTROSPECTION AC_ARG_ENABLE([examples], - AS_HELP_STRING([--enable-examples], [enable virtxml example. Default is check, meaning it is enabled as long as libvirt-gobject is installed]), + AS_HELP_STRING([--enable-examples], [enable virt-designer example. Default is check, meaning it is enabled as long as libvirt-gobject is installed]), [],[enable_examples=check]) if test "x$enable_examples" != "xno" ; then diff --git a/examples/Makefile.am b/examples/Makefile.am index ae2667e..1d8947f 100644 --- a/examples/Makefile.am +++ b/examples/Makefile.am @@ -2,10 +2,10 @@ INCLUDES = \ -I$(top_builddir)/libvirt-designer \ -I$(top_srcdir) -virtxml_LDADD = \ +virt_designer_LDADD = \ $(top_builddir)/libvirt-designer/libvirt-designer-1.0.la -virtxml_CFLAGS = \ +virt_designer_CFLAGS = \ $(COVERAGE_CFLAGS) \ $(LIBOSINFO_CFLAGS) \ $(LIBVIRT_GCONFIG_CFLAGS) \ @@ -13,7 +13,7 @@ virtxml_CFLAGS = \ $(LIBVIRT_GOBJECT_CFLAGS) \ $(NULL) -virtxml_LDFLAGS = \ +virt_designer_LDFLAGS = \ $(LIBOSINFO_LIBS) \ $(LIBVIRT_GCONFIG_LIBS) \ $(LIBVIRT_GOBJECT_LIBS) @@ -24,9 +24,9 @@ POD2MAN = pod2man -c "Virtualization Support" -r "$(PACKAGE)-$(VERSION)" $(AM_V_GEN)$(POD2MAN) $< $@ if WITH_EXAMPLES -bin_PROGRAMS = virtxml +bin_PROGRAMS = virt-designer -man1_MANS = virtxml.1 +man1_MANS = virt-designer.1 CLEANFILES = $(man1_MANS) endif diff --git a/examples/virtxml.c b/examples/virt-designer.c similarity index 96% rename from examples/virtxml.c rename to examples/virt-designer.c index 6fb51d5..7628449 100644 --- a/examples/virtxml.c +++ b/examples/virt-designer.c @@ -1,7 +1,7 @@ /* - * virtxml.c: produce an domain XML + * virt-designer.c: produce an domain XML * - * Copyright (C) 2012 Red Hat, Inc. + * Copyright (C) 2012, 2014 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -750,26 +750,26 @@ cleanup: =head1 NAME -virtxml - Generate domain XML +virt-designer - Generate domain XML =head1 SYNOPSIS -B<virtxml> [I<OPTION>]... +B<virt-designer> [I<OPTION>]... =head1 DESCRIPTION -B<virtxml> is a command line tool for generating XML documents for +B<virt-designer> is a command line tool for generating XML documents for libvirt domains. However, it cooperates with libosinfo database to guess the correct combination of attributes (e.g. disk driver, NIC model). -B<virtxml> does not feed libvirt with generated XML though. For now, +B<virt-designer> does not feed libvirt with generated XML though. For now, it's a proof of concept. =head1 OPTIONS -The basic structure of arguments passed to B<virtxml> is: +The basic structure of arguments passed to B<virt-designer> is: - virtxml [-c URI] [OPTION] [OPTION] ... + virt-designer [-c URI] [OPTION] [OPTION] ... However, arguments have no pre-defined order so users can type them in any order they like. @@ -844,7 +844,7 @@ the I<recommended> is used. Usually, both B<--os> and B<--platform> are required as they are needed to make the right decision on driver, model, ... when adding a new device. However, when adding a disk which is an installation medium (e.g. a CD-ROM or -DVD), B<virtxml> tries to guess the OS. Something similar is done with +DVD), B<virt-designer> tries to guess the OS. Something similar is done with platform. Usually, the platform is guessed from the connection URI. =head1 EXAMPLES @@ -852,16 +852,16 @@ platform. Usually, the platform is guessed from the connection URI. Domain with Fedora 17 from locally stored ISO and one NIC with mac 00:11:22:33:44:55 and link set down: - # virtxml -C Fedora-17-x86_64-Live-KDE.iso \ - -i default,mac=00:11:22:33:44:55,link=down + # virt-designer -C Fedora-17-x86_64-Live-KDE.iso \ + -i default,mac=00:11:22:33:44:55,link=down To add multiple devices just use appropriate argument multiple times: - # virtxml -d /tmp/home.img,qcow2 \ - -d /var/lib/libvirt/images/f17.img,qcow2 \ - -i default,mac=00:11:22:33:44:55,link=down \ - -i blue_network \ - -r minimal + # virt-designer -d /tmp/home.img,qcow2 \ + -d /var/lib/libvirt/images/f17.img,qcow2 \ + -i default,mac=00:11:22:33:44:55,link=down \ + -i blue_network \ + -r minimal =head1 AUTHORS diff --git a/libvirt-designer.spec.in b/libvirt-designer.spec.in index 894bcb8..fb5c70a 100644 --- a/libvirt-designer.spec.in +++ b/libvirt-designer.spec.in @@ -78,8 +78,8 @@ rm -rf $RPM_BUILD_ROOT %files %defattr(-,root,root,-) -%{_bindir}/virtxml -%{_mandir}/man1/virtxml.1* +%{_bindir}/virt-designer +%{_mandir}/man1/virt-designer.1* %files libs %defattr(-,root,root,-) -- 1.8.5.2

2 1

[libvirt] [PATCH] qemu: Fix the error message for scsi host device's shareable checking
by Osier Yang 30 Jan '14

30 Jan '14

This fixes the wrong argument order. --- Pushed under trivial rule. --- src/qemu/qemu_hostdev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_hostdev.c b/src/qemu/qemu_hostdev.c index 2b11d6c..1b16386 100644 --- a/src/qemu/qemu_hostdev.c +++ b/src/qemu/qemu_hostdev.c @@ -1135,8 +1135,8 @@ qemuPrepareHostdevSCSIDevices(virQEMUDriverPtr driver, virReportError(VIR_ERR_OPERATION_INVALID, _("SCSI device %s is already in use by " "other domain(s) as '%s'"), - tmp_shareable ? "shareable" : "non-shareable", - virSCSIDeviceGetName(tmp)); + virSCSIDeviceGetName(tmp), + tmp_shareable ? "shareable" : "non-shareable"); goto error; } -- 1.8.1.4

2 2

[libvirt] [PATCH v2]Sheepdog: Auto Adding volume and correct refresh volume problem.
by joel SIMOES 30 Jan '14

30 Jan '14

From: Joel SIMOES <joel.simoes(a)laposte.net> Correct ret return 0 on success --- src/storage/storage_backend_sheepdog.c | 92 ++++++++++++++++++++++++++++++---- 1 file changed, 82 insertions(+), 10 deletions(-) diff --git a/src/storage/storage_backend_sheepdog.c b/src/storage/storage_backend_sheepdog.c index a6981ce..75304b1 100644 --- a/src/storage/storage_backend_sheepdog.c +++ b/src/storage/storage_backend_sheepdog.c @@ -86,7 +86,8 @@ virStorageBackendSheepdogParseNodeInfo(virStoragePoolDefPtr pool, pool->available = pool->capacity - pool->allocation; return 0; - } while ((p = next)); + } + while ((p = next)); return -1; } @@ -109,6 +110,70 @@ virStorageBackendSheepdogAddHostArg(virCommandPtr cmd, virCommandAddArgFormat(cmd, "%d", port); } +static int +virStorageBackendSheepdogRefreshAllVol(virConnectPtr conn ATTRIBUTE_UNUSED, + virStoragePoolObjPtr pool) +{ + int ret; + char *output = NULL; + + virCommandPtr cmd = virCommandNewArgList(COLLIE, "vdi", "list", "-r", NULL); + virStorageBackendSheepdogAddHostArg(cmd, pool); + virCommandSetOutputBuffer(cmd, &output); + ret = virCommandRun(cmd, NULL); + char** lines; + char** cells; + + if (ret < 0) + goto cleanup; + + ret = -1; + lines = virStringSplit(output, "\n", 0); + size_t i; + for (i = 0; *(lines + i); i++) { + char *line = *(lines + i); + cells = virStringSplit(line, " ", 0); + size_t j; + for (j = 0; *(cells + j); j++) { + + char *cell = *(cells + j); + if (j == 1) { + virStorageVolDefPtr vol = NULL; + if (VIR_ALLOC(vol) < 0) + goto cleanup; + + if (VIR_STRDUP(vol->name, cell) < 0) + goto cleanup; + + vol->type = VIR_STORAGE_VOL_BLOCK; + + if (VIR_EXPAND_N(pool->volumes.objs, pool->volumes.count, 1) < 0) + goto cleanup; + pool->volumes.objs[pool->volumes.count - 1] = vol; + + if (virStorageBackendSheepdogRefreshVol(conn, pool, vol) < 0) + goto cleanup; + + vol = NULL; + } + + VIR_FREE(*(cells + j)); + } + + VIR_FREE(cells); + VIR_FREE(*(lines + i)); + } + + + VIR_FREE(lines); + ret = 0; + +cleanup: + virCommandFree(cmd); + VIR_FREE(lines); + VIR_FREE(cells); + return ret; +} static int virStorageBackendSheepdogRefreshPool(virConnectPtr conn ATTRIBUTE_UNUSED, @@ -122,15 +187,18 @@ virStorageBackendSheepdogRefreshPool(virConnectPtr conn ATTRIBUTE_UNUSED, virStorageBackendSheepdogAddHostArg(cmd, pool); virCommandSetOutputBuffer(cmd, &output); ret = virCommandRun(cmd, NULL); - if (ret == 0) - ret = virStorageBackendSheepdogParseNodeInfo(pool->def, output); + if (ret < 0) + goto cleanup; + ret = virStorageBackendSheepdogParseNodeInfo(pool->def, output); + if (ret < 0) + goto cleanup; + ret = virStorageBackendSheepdogRefreshAllVol(conn, pool); +cleanup: virCommandFree(cmd); - VIR_FREE(output); return ret; } - static int virStorageBackendSheepdogDeleteVol(virConnectPtr conn ATTRIBUTE_UNUSED, virStoragePoolObjPtr pool, @@ -143,12 +211,14 @@ virStorageBackendSheepdogDeleteVol(virConnectPtr conn ATTRIBUTE_UNUSED, virCommandPtr cmd = virCommandNewArgList(COLLIE, "vdi", "delete", vol->name, NULL); virStorageBackendSheepdogAddHostArg(cmd, pool); int ret = virCommandRun(cmd, NULL); + if (ret < 0) + goto cleanup; +cleanup: virCommandFree(cmd); return ret; } - static int virStorageBackendSheepdogCreateVol(virConnectPtr conn ATTRIBUTE_UNUSED, virStoragePoolObjPtr pool, @@ -174,7 +244,6 @@ virStorageBackendSheepdogCreateVol(virConnectPtr conn ATTRIBUTE_UNUSED, return 0; } - static int virStorageBackendSheepdogBuildVol(virConnectPtr conn, virStoragePoolObjPtr pool, @@ -195,12 +264,12 @@ virStorageBackendSheepdogBuildVol(virConnectPtr conn, goto cleanup; ret = 0; + cleanup: virCommandFree(cmd); return ret; } - int virStorageBackendSheepdogParseVdiList(virStorageVolDefPtr vol, char *output) @@ -257,7 +326,8 @@ virStorageBackendSheepdogParseVdiList(virStorageVolDefPtr vol, return -1; return 0; - } while ((p = next)); + } + while ((p = next)); return -1; } @@ -295,7 +365,6 @@ cleanup: return ret; } - static int virStorageBackendSheepdogResizeVol(virConnectPtr conn ATTRIBUTE_UNUSED, virStoragePoolObjPtr pool, @@ -310,7 +379,10 @@ virStorageBackendSheepdogResizeVol(virConnectPtr conn ATTRIBUTE_UNUSED, virCommandAddArgFormat(cmd, "%llu", capacity); virStorageBackendSheepdogAddHostArg(cmd, pool); int ret = virCommandRun(cmd, NULL); + if (ret < 0) + goto cleanup; +cleanup: virCommandFree(cmd); return ret; -- 1.8.3.2

2 1

[libvirt] [PATCH 0/2 v4] qemu: Don't fail if the SCSI host device is shareable between domains
by Osier Yang 30 Jan '14

30 Jan '14

*** BLURB HERE *** Osier Yang (2): qemu: Don't fail if the SCSI host device is shareable between domains tests: Add tests for scsi utils docs/formatdomain.html.in | 5 ++ src/libvirt_private.syms | 2 +- src/qemu/qemu_cgroup.c | 3 +- src/qemu/qemu_command.c | 3 +- src/qemu/qemu_command.h | 3 +- src/qemu/qemu_conf.c | 12 ++- src/qemu/qemu_hostdev.c | 86 ++++++++++-------- src/security/security_apparmor.c | 3 +- src/security/security_dac.c | 6 +- src/security/security_selinux.c | 6 +- src/util/virscsi.c | 70 +++++++++++---- src/util/virscsi.h | 16 ++-- tests/Makefile.am | 14 +++ tests/testutilsqemu.c | 3 +- tests/virscsitest.c | 188 +++++++++++++++++++++++++++++++++++++++ 15 files changed, 344 insertions(+), 76 deletions(-) create mode 100644 tests/virscsitest.c -- 1.8.1.4

2 6

[libvirt] [PATCH] maint: add configure checks for BSD CPU affinity
by Roman Bogorodskiy 29 Jan '14

29 Jan '14

Check for presence of sys/cpuset.h header and cpuset_getaffinity() in configure instead of just using #ifdef __FreeBSD__ for that code. --- configure.ac | 10 ++++++++++ src/util/virprocess.c | 7 +++++-- 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index 3a70375..1670a41 100644 --- a/configure.ac +++ b/configure.ac @@ -2604,6 +2604,16 @@ AC_CHECK_DECLS([BRDGSFD, BRDGADD, BRDGDEL], #include <net/if_bridgevar.h> ]) +# Check for BSD CPU affinity availability +AC_CHECK_DECLS([cpuset_getaffinity], + [AC_DEFINE([HAVE_BSD_CPU_AFFINITY], + [1], + [whether BSD CPU affinity management is available])], + [], + [#include <sys/param.h> + #include <sys/cpuset.h> + ]) + # Check if we need to look for ifconfig if test "$want_ifconfig" = "yes"; then AC_PATH_PROG([IFCONFIG_PATH], [ifconfig]) diff --git a/src/util/virprocess.c b/src/util/virprocess.c index 827143c..83d0679 100644 --- a/src/util/virprocess.c +++ b/src/util/virprocess.c @@ -34,11 +34,14 @@ #ifdef __FreeBSD__ # include <sys/param.h> -# include <sys/cpuset.h> # include <sys/sysctl.h> # include <sys/user.h> #endif +#ifdef HAVE_BSD_CPU_AFFINITY +# include <sys/cpuset.h> +#endif + #include "viratomic.h" #include "virprocess.h" #include "virerror.h" @@ -457,7 +460,7 @@ realloc: return 0; } -#elif defined(__FreeBSD__) +#elif defined(HAVE_BSD_CPU_AFFINITY) int virProcessSetAffinity(pid_t pid ATTRIBUTE_UNUSED, virBitmapPtr map) -- 1.8.4.3

2 1

[libvirt] [PATCH] Revert "networkAllocateActualDevice: Set QoS for bridgeless networks too"
by Michal Privoznik 29 Jan '14

29 Jan '14

This reverts commit 2996e6be19a13199ded7c2aa21039cca97318e01 and some parts of 2636dc8c4de83cd37bc0680a6fbc3f6d25023bd7. Conflicts: tests/virnetdevbandwidthtest.c: New test has been introduced since then. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/libvirt_private.syms | 1 - src/network/bridge_driver.c | 25 +----------- src/util/virnetdevbandwidth.c | 84 -------------------------------------- src/util/virnetdevbandwidth.h | 6 --- tests/virnetdevbandwidthtest.c | 91 ------------------------------------------ 5 files changed, 2 insertions(+), 205 deletions(-) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 7655247..45f3117 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -1470,7 +1470,6 @@ virNetDevBandwidthClear; virNetDevBandwidthCopy; virNetDevBandwidthEqual; virNetDevBandwidthFree; -virNetDevBandwidthMinimal; virNetDevBandwidthPlug; virNetDevBandwidthSet; virNetDevBandwidthUnplug; diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 6bdd1d6..0b43a67 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -3284,30 +3284,9 @@ networkAllocateActualDevice(virDomainNetDefPtr iface) else if (portgroup && portgroup->bandwidth) bandwidth = portgroup->bandwidth; - /* For bridgeless networks we must take into account any QoS set to them. - * This is, however, a bit tricky. With bridged network there are two - * points where QoS can be set: domain's interface and the bridge. The - * limits on those differ (in general) in which case the lower limit gets - * into effect. For example, if domain's interface average is 10Mbps but - * the network's is just 1Mbps, the domain will not be able to send data - * any faster than 1Mbps. However, on bridgeless networks we can't just set - * those two points and let kernel do its job since we have only single - * point. Therefore, we must combine the QoS settings from both domain's - * interface and network and choose the minimal value from pairs. - */ - if (netdef->forward.type == VIR_NETWORK_FORWARD_PRIVATE || - netdef->forward.type == VIR_NETWORK_FORWARD_VEPA || - netdef->forward.type == VIR_NETWORK_FORWARD_PASSTHROUGH || - netdef->forward.type == VIR_NETWORK_FORWARD_HOSTDEV) { - if (virNetDevBandwidthMinimal(&iface->data.network.actual->bandwidth, - bandwidth, - netdef->bandwidth) < 0) - goto error; - } else if (bandwidth && - virNetDevBandwidthCopy(&iface->data.network.actual->bandwidth, - bandwidth) < 0) { + if (bandwidth && virNetDevBandwidthCopy(&iface->data.network.actual->bandwidth, + bandwidth) < 0) goto error; - } /* copy appropriate vlan info to actualNet */ if (iface->vlan.nTags > 0) diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c index 273c2db..ed6a19d 100644 --- a/src/util/virnetdevbandwidth.c +++ b/src/util/virnetdevbandwidth.c @@ -27,7 +27,6 @@ #include "viralloc.h" #include "virerror.h" #include "virstring.h" -#include "virutil.h" #define VIR_FROM_THIS VIR_FROM_NONE @@ -589,86 +588,3 @@ cleanup: VIR_FREE(ceil); return ret; } - -static void -virNetDevBandwidthRateMinimal(virNetDevBandwidthRatePtr result, - virNetDevBandwidthRatePtr band1, - virNetDevBandwidthRatePtr band2) -{ - if (!band1 || !band2) { - memcpy(result, band1 ? band1 : band2, sizeof(*result)); - return; - } - - /* We can't do a simple MIN() here, because zero value doesn't mean the - * narrowest limit, but an unset value. Hence we need symmetric F(a, b) - * so that: - * F(a, 0) = F(0, a) = a; special corner case: F(0, 0) = 0 - * F(a, b) = MIN(a, b) for a != 0 and b != 0 - */ -#define NON_ZERO_MIN(a, b) \ - ((a) && (b) ? MIN(a, b) : (a) ? (a) : (b)) - - result->average = NON_ZERO_MIN(band1->average, band2->average); - result->peak = NON_ZERO_MIN(band1->peak, band2->peak); - result->floor = NON_ZERO_MIN(band1->floor, band2->floor); - result->burst = NON_ZERO_MIN(band1->burst, band2->burst); -} - -/** - * virNetDevBandwidthMinimal: - * @result: the minimal intersect of @band1 and @band2 - * @band1: the first bandwidth - * @band2: the second bandwidth - * - * Combine the two bandwidths into one with choosing the minimal value for each - * pair of items within bandwidth structure. The resulting bandwidth is stored - * into @result. In case of both @band1 and @band2 being NULL pointers, the - * @ret is set to NULL as well as there's no bandwidth to calculate (this is - * not considered an error). The caller is responsible for freeing @result when - * no longer needed. - * - * Returns 0 on success, -1 otherwise. - */ -int -virNetDevBandwidthMinimal(virNetDevBandwidthPtr *result, - virNetDevBandwidthPtr band1, - virNetDevBandwidthPtr band2) -{ - int ret = -1; - - if (!band1 && !band2) { - /* Nothing to compute */ - *result = NULL; - return 0; - } - - if (!band1 || !band2) { - /* Sweet, one of the args is NULL. The non-NULL one - * is our minimum then. */ - return virNetDevBandwidthCopy(result, band1 ? band1 : band2); - } - - if (VIR_ALLOC(*result) < 0) - goto cleanup; - - if (band1->in || band2->in) { - if (VIR_ALLOC((*result)->in) < 0) - goto cleanup; - - virNetDevBandwidthRateMinimal((*result)->in, band1->in, band2->in); - } - - if (band1->out || band2->out) { - if (VIR_ALLOC((*result)->out) < 0) - goto cleanup; - - virNetDevBandwidthRateMinimal((*result)->out, band1->out, band2->out); - } - - ret = 0; -cleanup: - if (ret < 0) - VIR_FREE(*result); - return ret; -} diff --git a/src/util/virnetdevbandwidth.h b/src/util/virnetdevbandwidth.h index ea40df4..4606a07 100644 --- a/src/util/virnetdevbandwidth.h +++ b/src/util/virnetdevbandwidth.h @@ -74,10 +74,4 @@ int virNetDevBandwidthUpdateRate(const char *ifname, unsigned long long new_rate) ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_RETURN_CHECK; - -int -virNetDevBandwidthMinimal(virNetDevBandwidthPtr *result, - virNetDevBandwidthPtr band1, - virNetDevBandwidthPtr band2) - ATTRIBUTE_NONNULL(1) ATTRIBUTE_RETURN_CHECK; #endif /* __VIR_NETDEV_BANDWIDTH_H__ */ diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c index 345a0dd..609deb8 100644 --- a/tests/virnetdevbandwidthtest.c +++ b/tests/virnetdevbandwidthtest.c @@ -62,56 +62,6 @@ struct testSetStruct { } while (0) static int -testVirNetDevBandwidthMinimal(const void *data) -{ - int ret = -1; - const struct testMinimalStruct *info = data; - virNetDevBandwidthPtr expected_result = NULL, result = NULL, - band1 = NULL, band2 = NULL; - - - /* Parse given XMLs */ - PARSE(info->expected_result, expected_result); - PARSE(info->band1, band1); - PARSE(info->band2, band2); - - if (virNetDevBandwidthMinimal(&result, band1, band2) < 0) - goto cleanup; - - if (!virNetDevBandwidthEqual(expected_result, result)) { - virBuffer exp_buf = VIR_BUFFER_INITIALIZER, - res_buf = VIR_BUFFER_INITIALIZER; - char *exp = NULL, *res = NULL; - - fprintf(stderr, "expected_result != result"); - - if (virNetDevBandwidthFormat(expected_result, &exp_buf) < 0 || - virNetDevBandwidthFormat(result, &res_buf) < 0 || - !(exp = virBufferContentAndReset(&exp_buf)) || - !(res = virBufferContentAndReset(&res_buf))) { - fprintf(stderr, "Failed to fail"); - virBufferFreeAndReset(&exp_buf); - virBufferFreeAndReset(&res_buf); - VIR_FREE(exp); - VIR_FREE(res); - goto cleanup; - } - - virtTestDifference(stderr, exp, res); - VIR_FREE(exp); - VIR_FREE(res); - } - - ret = 0; -cleanup: - virNetDevBandwidthFree(expected_result); - virNetDevBandwidthFree(result); - virNetDevBandwidthFree(band1); - virNetDevBandwidthFree(band2); - return ret; -} - -static int testVirNetDevBandwidthSet(const void *data) { int ret = -1; @@ -159,15 +109,6 @@ mymain(void) { int ret = 0; -#define DO_TEST_MINIMAL(r, ...) \ - do { \ - struct testMinimalStruct data = {r, __VA_ARGS__}; \ - if (virtTestRun("virNetDevBandwidthMinimal", \ - testVirNetDevBandwidthMinimal, \ - &data) < 0) \ - ret = -1; \ - } while (0) - #define DO_TEST_SET(Band, Exp_cmd, ...) \ do { \ struct testSetStruct data = {.band = Band, \ @@ -180,38 +121,6 @@ mymain(void) } while (0) - DO_TEST_MINIMAL(NULL, NULL, NULL); - - DO_TEST_MINIMAL("<bandwidth>" - " <inbound average='1000' peak='5000' burst='5120'/>" - " <outbound average='128' peak='256' burst='256'/>" - "</bandwidth>", - .band1 = "<bandwidth>" - " <inbound average='1000' peak='5000' burst='5120'/>" - " <outbound average='128' peak='256' burst='256'/>" - "</bandwidth>"); - - DO_TEST_MINIMAL("<bandwidth>" - " <inbound average='1000' peak='5000' burst='5120'/>" - " <outbound average='128' peak='256' burst='256'/>" - "</bandwidth>", - .band2 = "<bandwidth>" - " <inbound average='1000' peak='5000' burst='5120'/>" - " <outbound average='128' peak='256' burst='256'/>" - "</bandwidth>"); - DO_TEST_MINIMAL("<bandwidth>" - " <inbound average='1' peak='2' floor='3' burst='4'/>" - " <outbound average='5' peak='6' burst='7'/>" - "</bandwidth>", - "<bandwidth>" - " <inbound average='1' peak='2' burst='4'/>" - " <outbound average='0' burst='7'/>" - "</bandwidth>", - "<bandwidth>" - " <inbound average='1' peak='2' floor='3'/>" - " <outbound average='5' peak='6'/>" - "</bandwidth>"); - DO_TEST_SET(("<bandwidth>" " <inbound average='1' peak='2' floor='3' burst='4'/>" " <outbound average='5' peak='6' burst='7'/>" -- 1.8.5.2

2 3

[libvirt] [PATCH 0/4] Improve QoS and unit-test it
by Michal Privoznik 29 Jan '14

29 Jan '14

*** BLURB HERE *** Michal Privoznik (4): networkAllocateActualDevice: Set QoS for bridgeless networks too tests: Introduce virnetdevbandwidthtest virnetdevbandwidthtest: Introduce mocking virnetdevbandwidthtest: Introduce testVirNetDevBandwidthSet src/libvirt_private.syms | 1 + src/network/bridge_driver.c | 25 ++++- src/util/virnetdevbandwidth.c | 88 +++++++++++++++ src/util/virnetdevbandwidth.h | 6 + tests/Makefile.am | 14 +++ tests/virnetdevbandwidthmock.c | 106 ++++++++++++++++++ tests/virnetdevbandwidthtest.c | 244 +++++++++++++++++++++++++++++++++++++++++ 7 files changed, 482 insertions(+), 2 deletions(-) create mode 100644 tests/virnetdevbandwidthmock.c create mode 100644 tests/virnetdevbandwidthtest.c -- 1.8.5.2

3 13

[libvirt] [libvirt-glib 0/3] Add unit tests
by Christophe Fergeau 29 Jan '14

29 Jan '14

This patch series adds support for GTester tests to libvirt-glib, and starts implementing some unit tests for libvirt-gconfig, and for a very specific part of the event handling code. This series https://www.redhat.com/archives/libvir-list/2014-January/msg00939.html is needed for the gconfig test to build and successfully pass. This series https://www.redhat.com/archives/libvir-list/2014-January/msg01359.html is needed too if you don't want the event test to fail. Christophe

2 10

[libvirt] ANNOUNCE: virt-xml: Edit libvirt XML with command line options
by Cole Robinson 29 Jan '14

29 Jan '14

I've just pushed a new tool to virt-manager.git called virt-xml. virt-xml uses virt-install's command line options to allow building and editing domain XML. Best way to describe what it can do is with a bunch of examples: See a list of all suboptions that --disk and --network take # virt-xml --disk=? --network=? Change the <description> of domain 'example': # virt-xml example --edit --metadata description="my new description" # Enable the boot device menu for domain 'example': # virt-xml example --edit --boot bootmenu=on Clear the previous <cpu> definition of domain 'winxp', change it to 'host-model', but interactively confirm the diff before saving: # virt-xml winxp --edit --cpu host-model,clearxml=yes --confirm Change the sound cards with model=ac97 to model=ich6 on 'fedora19', but only output the diff: # virt-xml fedora19 --edit model=ac97 --soundhw model=ich6 --print-diff Update the first graphics password to 'foo' of the running VM 'rhel6': # virt-xml rhel6 --edit --graphics password=foo --update Remove the disk path from disk device hdc: # virt-xml rhel6 --edit target=hdc --disk path= Change all disk devices of type 'disk' to use cache=none, using XML from stdin, printing the new XML to stdout. # cat <xmlfile> | virt-xml --edit device=disk --disk cache=none Change all host devices to use driver_name=vfio for VM 'fedora20' on the remote connection # virt-xml --connect qemu+ssh://remotehost/system \ fedora20 --edit all --host-device driver_name=vfio Hotplug host USB device 001.003 to running domain 'fedora19': # virt-xml f19 --add-device --host-device 001.003 --update Add a spicevmc channel to the domain 'winxp', that will be available after the next VM shutdown. # virt-xml winxp --add-device --channel spicevmc Create a 10G qcow2 disk image and attach it to 'fedora18' for the next VM startup: # virt-xml fedora18 --add-device \ --disk /var/lib/libvirt/images/newimage.qcow2,format=qcow2,size=10 Remove the disk vdb from the running domain 'rhel7': # virt-xml rhel7 --remove-device --disk target=vdb Remove all graphics devices from the VM 'rhel7' after the next shutdown: # virt-xml rhel7 --remove-device --graphics all Generate XML for a virtio console device and print it to stdout: # virt-xml --build-xml --console pty,target_type=virtio The man page and --help output have more info, man page at: https://git.fedorahosted.org/cgit/virt-manager.git/tree/man/virt-xml.pod If you are interested in trying it out before it hits any distro, it's really simple: git clone git://git.fedorahosted.org/virt-manager.git cd virt-manager ./virt-xml --help Nothing is set in stone yet, so if anyone has suggestions about the command line syntax or any of the above, please speak up: I plan on cutting a new virt-manager release with virt-xml within the next few weeks. Thanks, Cole

4 6

[libvirt] [PATCH v2 0/4] Fix deadlock in nwfilter code
by Daniel P. Berrange 29 Jan '14

29 Jan '14

Since we introduced fine grained locking into the QEMU driver so that VM start can run in parallel, we appear to have caused a race with the nwfilter code. In particular since we no longer hold the global QEMU driver lock for the duration of VM startup we have a lock ordering flaw. This results in deadlock when nwfilter operations happen in parallel with VM startup. This also affects the LXC driver. This patch series attempts to address the problem https://bugzilla.redhat.com/show_bug.cgi?id=929412 the removal of the windows thread impl isn't strictly required, I just didn't want to waste time creating a read/write lock impl for Windows threads. See also this callgraph http://berrange.fedorapeople.org/nwfilter.ps In v2: - Re-ordered patches and squashed two together - Add missing locking in state reload function - Don't remove locking from virNWFilterInstantiateFilterLate Daniel P. Berrange (4): Add a read/write lock implementation Remove windows thread implementation in favour of pthreads Add a mutex to serialize updates to firewall Push nwfilter update locking up to top level configure.ac | 15 +- src/Makefile.am | 4 - src/conf/nwfilter_conf.c | 23 +- src/conf/nwfilter_conf.h | 3 +- src/libvirt_private.syms | 8 +- src/lxc/lxc_driver.c | 6 + src/nwfilter/nwfilter_driver.c | 15 +- src/nwfilter/nwfilter_gentech_driver.c | 23 +- src/nwfilter/nwfilter_gentech_driver.h | 2 +- src/qemu/qemu_driver.c | 6 + src/uml/uml_driver.c | 4 + src/util/virthread.c | 291 +++++++++++++++++++++++- src/util/virthread.h | 51 ++++- src/util/virthreadpthread.c | 278 ----------------------- src/util/virthreadpthread.h | 49 ---- src/util/virthreadwin32.c | 396 --------------------------------- src/util/virthreadwin32.h | 53 ----- 17 files changed, 404 insertions(+), 823 deletions(-) delete mode 100644 src/util/virthreadpthread.c delete mode 100644 src/util/virthreadpthread.h delete mode 100644 src/util/virthreadwin32.c delete mode 100644 src/util/virthreadwin32.h -- 1.8.4.2

3 21

[libvirt] [RFC PATCHv2 0/4] Adding 'config' driver
by Adam Walters 29 Jan '14

29 Jan '14

This patchset adds a driver named 'config' which provides access to configuration data, such as secret and storage definitions, without requiring a connection to a hypervisor. This complements my previous patchset, which resolves the race condition on libvirtd startup. The rationale behind this idea is that there exist circumstances under which a driver may require access to things such as secrets during a time at which there is no active connection to a hypervisor. Currently, that data can not be accessed. An example of this in libvirt today is that the storage driver requires a connection in order to access secret data to auto-start an RBD storage pool that uses CephX authentication. My previous patchset breaks the ability to auto-start that type of storage pool, and this patchset fixes it again. This driver is technically what one may call a hypervisor driver, but it does not implement any domain operations. It simply exists to handle requests by drivers for access to information that would otherwise by unattainable. The URI provided by this driver is 'config:///' and has been tested working on four different machines of mine, running three different Linux distributions (Archlinux, Gentoo, and CentOS). Being a very simple driver, I would expect it to work pretty much anywhere. I welcome comments and suggestions related to this driver. At the very least, this patchset combined with my previous patchset resolves the current race condition present on startup of libvirtd without any loss of functionality. Adam Walters (4): config: Adding config driver configure: Implement config driver libvirtd: Reorder load of secrets driver storage: Change hardcoded QEMU connection to use config driver configure.ac | 11 ++ daemon/libvirtd.c | 21 ++-- include/libvirt/virterror.h | 2 + po/POTFILES.in | 1 + src/Makefile.am | 25 +++++ src/config/config_driver.c | 238 +++++++++++++++++++++++++++++++++++++++++++ src/config/config_driver.h | 44 ++++++++ src/storage/storage_driver.c | 13 +-- src/util/virerror.c | 2 + 9 files changed, 345 insertions(+), 12 deletions(-) create mode 100644 src/config/config_driver.c create mode 100644 src/config/config_driver.h -- 1.8.5.2

2 7

[libvirt] [PATCHv2 0/9] Snapshots on gluster
by Peter Krempa 29 Jan '14

29 Jan '14

Peter Krempa (9): storage: Add new argument for createVol backend API storage: gluster: Introduce dummy functions for creating a volume storage: Add internal API to create temporary storage pools and vols snapshot: Add support for specifying snapshot disk backing type snapshot: Test snapshot disk type specification storage: Implement ephemeral storage APIs in the generic storage driver storage: gluster: Support conversion of gluster volumes to temp volumes qemu: snapshot: Switch snapshot file deletion to the new storage API qemu: snapshot: Add support for external active snapshots on gluster docs/formatsnapshot.html.in | 15 + docs/hvsupport.pl | 3 + docs/schemas/domainsnapshot.rng | 76 +++- src/Makefile.am | 3 +- src/check-aclrules.pl | 3 + src/check-drivername.pl | 1 + src/conf/snapshot_conf.c | 42 ++- src/conf/snapshot_conf.h | 15 +- src/driver.h | 13 + src/libvirt_internal.c | 70 ++++ src/libvirt_internal.h | 22 ++ src/libvirt_private.syms | 3 + src/qemu/qemu_command.c | 2 +- src/qemu/qemu_command.h | 9 + src/qemu/qemu_conf.c | 3 - src/qemu/qemu_driver.c | 227 ++++++++++-- src/storage/storage_backend.h | 3 +- src/storage/storage_backend_disk.c | 3 +- src/storage/storage_backend_fs.c | 30 +- src/storage/storage_backend_gluster.c | 119 +++++- src/storage/storage_backend_logical.c | 6 +- src/storage/storage_backend_rbd.c | 3 +- src/storage/storage_backend_sheepdog.c | 3 +- src/storage/storage_driver.c | 405 ++++++++++++++++++++- tests/domainsnapshotxml2xmlin/disk_snapshot.xml | 18 + .../disk_driver_name_null.xml | 2 +- tests/domainsnapshotxml2xmlout/disk_snapshot.xml | 22 +- .../disk_snapshot_redefine.xml | 6 +- 28 files changed, 1007 insertions(+), 120 deletions(-) create mode 100644 src/libvirt_internal.c -- 1.8.5.2

2 22

[libvirt] [libvirt-glib 0/3] Fixes to glib/libvirt mainloop integration
by Christophe Fergeau 29 Jan '14

29 Jan '14

While reviewing danpb's win32 patch for the mainloop integration code, I found 2 more issues in the existing code (one leak, one bug). These patches fix this (first patch is just some cleanup). Christophe

2 6

[libvirt] [PATCH] Honor blacklist for modprobe command
by John Ferlan 29 Jan '14

29 Jan '14

https://bugzilla.redhat.com/show_bug.cgi?id=1045124 When loading modules, libvirt does not honor the modprobe blacklist. By adding a "-b" to the modprobe command libvirt will fail to load a module if it's on the blacklist Check if the failure to load a driver was due to it being on the blacklist using the output of a "modprobe -c" searching for "blacklist <driver_Name>" where driver_Name is possibly a modified string of the input driver changing all '-' into '_' since that's what modprobe does. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/util/virpci.c | 48 ++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 42 insertions(+), 6 deletions(-) diff --git a/src/util/virpci.c b/src/util/virpci.c index e2d222e..18b85f2 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -979,6 +979,10 @@ virPCIProbeStubDriver(const char *driver) { char *drvpath = NULL; bool probed = false; + size_t i; + char *drvblklst = NULL; + char *outbuf = NULL; + virCommandPtr cmd = NULL; recheck: if (virPCIDriverDir(&drvpath, driver) == 0 && virFileExists(drvpath)) { @@ -990,18 +994,49 @@ recheck: VIR_FREE(drvpath); if (!probed) { - const char *const probecmd[] = { MODPROBE, driver, NULL }; + cmd = virCommandNewArgList(MODPROBE, "-b", driver, NULL); probed = true; - if (virRun(probecmd, NULL) < 0) { + if (virCommandRun(cmd, NULL) < 0) { char ebuf[1024]; VIR_WARN("failed to load driver %s: %s", driver, virStrerror(errno, ebuf, sizeof(ebuf))); - return -1; + goto cleanup; } + virCommandFree(cmd); + cmd = NULL; goto recheck; } + /* All error path code - purpose is to determine whether the failure + * occurs because device is on blacklist in order to add an error + * message to help detect why load failed + */ + if (virAsprintfQuiet(&drvblklst, "blacklist %s", driver) < 0) + goto cleanup; + + /* Although driver may have dash, 'modprobe -c' adjusts to use underscore */ + for (i = 0; i < drvblklst[i]; i++) + if (drvblklst[i] == '-') + drvblklst[i] = '_'; + + cmd = virCommandNewArgList(MODPROBE, "-c", NULL); + virCommandSetOutputBuffer(cmd, &outbuf); + if (virCommandRun(cmd, NULL) < 0) + goto cleanup; + + /* Find driver on blacklist? */ + if (strstr(outbuf, drvblklst)) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to load PCI stub module %s: " + "administratively prohibited"), + driver); + } + +cleanup: + virCommandFree(cmd); + VIR_FREE(drvblklst); + VIR_FREE(outbuf); return -1; } @@ -1313,9 +1348,10 @@ virPCIDeviceDetach(virPCIDevicePtr dev, virPCIDeviceList *inactiveDevs) { if (virPCIProbeStubDriver(dev->stubDriver) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("Failed to load PCI stub module %s"), - dev->stubDriver); + if (virGetLastError() == NULL) + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to load PCI stub module %s"), + dev->stubDriver); return -1; } -- 1.8.4.2

2 3

[libvirt] [PATCHv2 0/6] Add BlkIO and CPU/mem stat API implementations for lxc
by Thorsten Behrens 29 Jan '14

29 Jan '14

This patch set adds block io, memory and domain cpu statistics API slot implementations to the LXC driver, in order to get linux container monitoring and accounting a bit closer to qemu standards. The last patch is a tad quirky (happy to hear suggestions on alternative ways), in that it widens the permissible value set at the .domainBlockStats slot: for lxc guests, it is relatively likely to have zero disk devices, since host filesystems can be used via passthrough bind mounts. Therefore, passing the zero-length string as device path, is interpreted as 'return summary stats for the entire domains's block io'. Thorsten Behrens (6): Add util virCgroupGetBlkioIo*Serviced methods. Implement domainMemoryStats API slot for LXC driver. Make qemuGetDomainTotalCPUStats a virCgroup function. Implement domainGetCPUStats for lxc driver. Implemet lxcDomainBlockStats for lxc driver Widening API change - accept empty path for virDomainBlockStats src/libvirt.c | 8 +- src/libvirt_private.syms | 3 + src/lxc/lxc_driver.c | 239 +++++++++++++++++++++++++++++++++++ src/qemu/qemu_driver.c | 54 +------- src/util/vircgroup.c | 291 +++++++++++++++++++++++++++++++++++++++++++ src/util/vircgroup.h | 17 +++ tools/virsh-domain-monitor.c | 11 +- tools/virsh.pod | 5 +- 8 files changed, 569 insertions(+), 59 deletions(-) -- 1.8.4

2 16

[libvirt] [PATCH v3 0/2] Enhance virnetdevbandwidthtest
by Michal Privoznik 29 Jan '14

29 Jan '14

yet another version - this time with virBuffer approach. Michal Privoznik (2): virCommand: Introduce virCommandSetDryRun virnetdevbandwidthtest: Introduce testVirNetDevBandwidthSet src/libvirt_private.syms | 1 + src/util/vircommand.c | 57 +++++++++++++++++++++++++++++-- src/util/vircommand.h | 2 ++ tests/virnetdevbandwidthtest.c | 76 ++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 134 insertions(+), 2 deletions(-) -- 1.8.5.2

2 4