- Devel - libvirt List Archives

[libvirt] Libvirt management daemon issue
by Bilal Arif 04 Nov '15

04 Nov '15

robo@robo:~$ sudo /etc/init.d/libvirt-bin start [sudo] password for robo: * Starting libvirt management daemon libvirtd/usr/sbin/libvirtd: /usr/local/lib/libvirt.so.0: version `LIBVIRT_PRIVATE_1.2.2' not found (required by /usr/sbin/libvirtd) Giving up waiting for /var/run/libvirt/libvirt-sock. After installing libvirt from source a lot of Ubuntu user face this issue, as i searched on internet about this, no solution works. this is mostly in case of ubuntu users as mentioned in many posted threads. so please help us as soon as possible. (this is only in case of installation libvirt from source)

2 1

[libvirt] [PATCH] qemu: add /usr/lib to AC_PATH_PROG for qemu-bridge-helper
by Michel Normand 04 Nov '15

04 Nov '15

For openSUSE the qemu-bridge-helper is installed in /usr/lib So libvirt has to search it in this directory. Signed-off-by: Michel Normand <normand(a)linux.vnet.ibm.com> --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index 75e95b7..77106ac 100644 --- a/configure.ac +++ b/configure.ac @@ -2520,7 +2520,7 @@ AC_DEFINE_UNQUOTED([QEMU_USER], ["$QEMU_USER"], [QEMU user account]) AC_DEFINE_UNQUOTED([QEMU_GROUP], ["$QEMU_GROUP"], [QEMU group account]) AC_PATH_PROG([QEMU_BRIDGE_HELPER], [qemu-bridge-helper], [/usr/libexec/qemu-bridge-helper], - [/usr/libexec:/usr/lib/qemu]) + [/usr/libexec:/usr/lib/qemu:/usr/lib]) AC_DEFINE_UNQUOTED([QEMU_BRIDGE_HELPER], ["$QEMU_BRIDGE_HELPER"], [QEMU bridge helper]) AC_ARG_WITH([macvtap], -- 1.7.9.5

2 1

[libvirt] Release of libvirt-1.2.21
by Daniel Veillard 04 Nov '15

04 Nov '15

So as planned I tagged libvirt-1.2.21 in git, we are out of freeze ! The signed tarballs and rpms for 1.2.21 are in the usual places for libvirt ftp://libvirt.org/libvirt/ and for python bindings ftp://libvirt.org/libvirt/python/ This is a moderately sized release with around 150 commits, a lot of refactoring and cleanups on the migration code, the python bindings and updates to the vz driver: Documentation: - event impl. registration before hypervisor connection (Dominik Perpeet) - libvirt-secret: Fix typo (Wei Jiangang) - NEWS: Fix XSLT stylesheet (Andrea Bolognani) - NEWS: Split releases by year (Andrea Bolognani) - NEWS: Unify date format (Andrea Bolognani) - NEWS: Unify section titles (Andrea Bolognani) - NEWS: Add empty lines (Andrea Bolognani) - NEWS: Fix indentation (Andrea Bolognani) - NEWS: Fix newlines (Andrea Bolognani) - NEWS: Fix whitespace (Andrea Bolognani) - NEWS: Organize old entries (Andrea Bolognani) - NEWS: Split old entries (2009) (Andrea Bolognani) - NEWS: Split old entries (2008) (Andrea Bolognani) - NEWS: Split old entries (2006-2007) (Andrea Bolognani) - NEWS: Properly escape > in HTML (Andrea Bolognani) - NEWS: Split off merged sections (Andrea Bolognani) - NEWS: Remove empty sections (Andrea Bolognani) - NEWS: Include description for release 0.7.3 (Andrea Bolognani) - domain: Show canonical pvspinlock XML (Cole Robinson) - Add Cuckoo Sandbox into apps.html (Martin Kletzander) Portability: - Fix virNetDevWaitDadFinish stub (Roman Bogorodskiy) - util: implement virProcessGetStartTime on GNU/kFreeBSD (Pino Toscano) - build: include xdr cflags for libvirt-admin (Eric Blake) - virt-host-validate: disable mntent code on Win32 (Daniel P. Berrange) - src: Remove $(builddir) usage (Andrea Bolognani) Bug Fixes: - wireshark: Install to generic plugin directory (Michal Privoznik) - network: wait for DAD to finish for bridge IPv6 addresses (Maxim Perevedentsev) - qemu: Use live autoNodeset when numatune placement is auto (Luyao Huang) - virsh-domain: use correct base for virStrToLong_ui (Pavel Hrdina) - libvirt-domain: fix the error reporting when use the localhost as target uri (Luyao Huang) - qemu: fix migration flags undefinesource cannot work (Luyao Huang) - virsh: Display an error when passing count <= 0 to setvcpus (Luyao Huang) - virtlockd: Don't SIGSEGV on SIGUSR1 (Michal Privoznik) - Close the source fd if the destination qemu exits during tunnelled migration (Shivaprasad G Bhat) - qemu: Fix qemu startup check for QEMU_CAPS_OBJECT_IOTHREAD (John Ferlan) - virsh: Use 'format' argument only when specified (Peter Krempa) - storage: Track successful creation of LV for removal (John Ferlan) - storage: Fix a resource leak in storageVolCreateXML (John Ferlan) - storage: Fix incorrect format for <disk> <auth> XML (John Ferlan) - migration: check dconnuri in p2p mode (Nikolay Shirokovskiy) - qemu: Resolve Coverity FORWARD_NULL (John Ferlan) - libxl: Resolve Coverity FORWARD_NULL (John Ferlan) - tests: eventtest: Fix coverity warning (Cole Robinson) - qemu: migration: Skip few checks while doing offline migration (Peter Krempa) - qemu: Wait until destination QEMU consumes all migration data (Jiri Denemark) - conf: Remove <metadata> elements with no namespace (Peter Krempa) - storage: Adjust calculation of alloc/capacity for disk (John Ferlan) - storage: Prior to creating a volume, refresh the pool (John Ferlan) - virfile: Fix error path for forked virFileRemove (John Ferlan) - rpc: libssh2: Fix regression in ssh host key verification (Peter Krempa) - interface: fail on OOM from virGetInterface() (Laine Stump) - libvirt.spec: fix accidental conditional inclusion of polkit docs (Daniel P. Berrange) - qemu: Use memory-backing-file only when needed (Martin Kletzander) Improvements: - util: set max wait for IPv6 DAD to 20 seconds (Laine Stump) - util: set error if DAD is not finished (Luyao Huang) - netlink: add support for multi-part netlink messages. (Maxim Perevedentsev) - rbd: Remove snapshots if the DELETE_WITH_SNAPSHOTS flag has been provided (Wido den Hollander) - bhyve: implement domainGetOSType (Roman Bogorodskiy) - libvirt-domain: fix no error report when p2p migrate fail (Luyao Huang) - conf: Add serial target type to ABI stability check (Luyao Huang) - lock_daemon: Switch to wrapper locking functions (Michal Privoznik) - qemu: hostdev: Introduce qemuHostdevUpdateActiveDomainDevices() (Andrea Bolognani) - qemu: hostdev: Unify naming for qemuHostdevUpdateActive*Devices() (Andrea Bolognani) - qemu: hostdev: Unify naming for qemuHostdevReAttach*Devices() (Andrea Bolognani) - qemu: hostdev: Unify naming for qemuHostdevPrepare*Devices() (Andrea Bolognani) - hostdev: Rename virHostdevUpdateDomainActiveDevices() (Andrea Bolognani) - conf: Fix error message to use correct parameter (John Ferlan) - util: Produce friendlier error message to user (Luyao Huang) - util: Adjust error paths for virNumaSetPagePoolSize (Luyao Huang) - util: split the virNumaGetHugePageInfoPath into separate function (Luyao Huang) - Avoid using !STREQ and !STRNEQ (Ishmanpreet Kaur Khera) - vz: implement some domain API calls (Maxim Nestratov) - vz: implement API calls of nodeGetxxx family (Maxim Nestratov) - vz: implement connectGetMaxVcpus API calls (Maxim Nestratov) - security_dac: Introduce remember/recall APIs (Michal Privoznik) - security_dac: Limit usage of virSecurityDACSetOwnershipInternal (Michal Privoznik) - virSecurityDACRestoreSecurityFileLabel: Pass virSecurityDACDataPtr (Michal Privoznik) - virSecurityDACSetOwnership: Pass virSecurityDACDataPtr (Michal Privoznik) - virSecurityDACSetOwnershipInternal: Don't chown so often (Michal Privoznik) - security_dac: Fix TODO marks (Michal Privoznik) - conf: Optimize the iothreadid initialization (John Ferlan) - qemu: Check for niothreads == 0 in qemuSetupCgroupForIOThreads (John Ferlan) - qemu: Use 'niothreadids' instead of 'iothreads' (John Ferlan) - Fix conficts with HACKING doc (Wei Jiangang) - conf: Refactor the iothreadid initialization (John Ferlan) - tests: make redirects happen in correct order (Wei Jiangang) - storage: Rework error paths for virStorageBackendCreateExecCommand (John Ferlan) - storage: On error rmdir created directory in virDirCreate[NoFork] (John Ferlan) - storage: On error unlink created file in virFileOpen{As|Forked} (John Ferlan) - storage: Remove duplicitous refreshVol in Sheepdog buildVol (John Ferlan) - storage: Remove duplicitous refreshVol in RBD buildVol (John Ferlan) - virt-host-validate: check for IOMMU support (Daniel P. Berrange) - libvirt-domain: Drop virDomainMigrateCheckNotLocal attribute (Michal Privoznik) - virt-host-validate: check for required cgroups (Daniel P. Berrange) - virt-host-validate: distinguish exists vs accessible for devices (Daniel P. Berrange) - virt-host-validate.c: check for kernel namespaces (Daniel P. Berrange) - migration: refactor: introduce parameter checking function (Nikolay Shirokovskiy) - migration: merge all proto branches into single function (Nikolay Shirokovskiy) - migration: refactor: refactor parameter compatibility checks (Nikolay Shirokovskiy) - migration: refactor: extract parameter adaption functions (Nikolay Shirokovskiy) - migration: refactor: introduce params version of unmanaged (Nikolay Shirokovskiy) - migration: refactor: merge direct and p2p into unmanaged (Nikolay Shirokovskiy) - migration: refactor: rename uri parameter to miguri (Nikolay Shirokovskiy) - migration: remove direct migration dependency on version1 of driver (Michal Privoznik) - migration: move implementation check to branches in p2p (Nikolay Shirokovskiy) - migration: refactor: reuse p2p url check (Nikolay Shirokovskiy) - migration: refactor: get rid of use_params p2p_full (Nikolay Shirokovskiy) - virJSONValueArraySize: return ssize_t (Michal Privoznik) - src: Include $(builddir)/util in the header search path (Andrea Bolognani) - storage: Perform some cleanup of calls (John Ferlan) - AUTHORS: Add myself to the list of committers (Andrea Bolognani) - virSecurityManagerNew: Turn array of booleans into flags (Michal Privoznik) - qemu: Fix indentation issue (Andrea Bolognani) - qemu: Remove explicit values from virQEMUCapsFlags (Andrea Bolognani) - qemu: migration: Use migration flags in qemuMigrationIsAllowed (Peter Krempa) - qemu: migration: Drop @def from qemuMigrationIsAllowed (Peter Krempa) - qemu: migration: Split source and destination migration checks (Peter Krempa) - spec: Delete .git after applying patches (Cole Robinson) - tests: qemu: Add aarch64 virtio pci tests (Cole Robinson) - qemu: Make updating stats in qemuMigrationCheckJobStatus optional (Jiri Denemark) - qemu: Introduce flags in qemuMigrationCompleted (Jiri Denemark) - qemu: Copy completed migration stats only on success (Jiri Denemark) - qemu: Always update migration times on destination (Jiri Denemark) - tests: Check GIC-related XMLs in qemuxml2xmltest (Martin Kletzander) - Do not distribute generated virkeymaps.h (Martin Kletzander) - apparmor: differentiate between error and unconfined profiles (Cédric Bosdonnat) - testutils: Drop virtTestResult (Cole Robinson) - tests: eventtest: Open code virtTestResult (Cole Robinson) - tests: sheepdog: Drop use of virtTestResult (Cole Robinson) - testutils: Add coloring to verbose PASS/FAILED output (Cole Robinson) - storage: Introduce virStorageBackendDiskStartPool (John Ferlan) - storage: Add additional errors/checks for disk label (John Ferlan) - storage: Add param to check whether we can write a disk label (John Ferlan) - storage: Refactor disk label checking (John Ferlan) - virfile: Add extra check for direct delete in virFileRemove (John Ferlan) - qemu: Add conditions for qemu-kvm use on ppc64 (Andrea Bolognani) - rpc: libssh2: Add more debugging info (Peter Krempa) - conf: Reuse virDomainDefCheckDuplicateDiskWWN to check disk serial too (Peter Krempa) - qemu: Perform the disk WWN check only on fresh starts (Peter Krempa) - interface: let netcf pre-filter for active vs. inactive (Laine Stump) - interface: re-use name and mac address rather than re-retrieving (Laine Stump) - interface: report correct interface count when not returning list (Laine Stump) - qemu: Add -mem-path even with numa (Martin Kletzander) - qemu: Extract -mem-path building into its own function (Martin Kletzander) - qemu: Move memory size detection to the top of the function (Martin Kletzander) - qemu: Move simplification variable to begining of the function (Martin Kletzander) - qemu: Add test cases for gic-version option (Pavel Fedin) - qemu: Add support for gic-version machine option (Pavel Fedin) - qemu: Introduce QEMU_CAPS_MACH_VIRT_GIC_VERSION capability (Pavel Fedin) - Post-release version bump to 1.2.21 (Martin Kletzander) Cleanups: - tests: Remove unused nodeinfo test data (Andrea Bolognani) - vz: cleanup (Maxim Nestratov) - vz: remove storage driver as never used (Maxim Nestratov) - vz: remove network driver as never used (Maxim Nestratov) Thanks everybody for your contributions to this release ! Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

3 5

[libvirt] [PATCH 0/5] Some qemuProcessStart cleanups
by Jiri Denemark 04 Nov '15

04 Nov '15

Jiri Denemark (5): qemu: Use correct type when calling qemuPrepareNVRAM qemu: Rename cleanup label in qemuProcessStart qemu: Rename ret variable in qemuProcessStart qemu: Introduce cleanup label in qemuProcessStart qemu: Fix memory leak in qemuProcessStart src/qemu/qemu_process.c | 215 ++++++++++++++++++++++++------------------------ 1 file changed, 106 insertions(+), 109 deletions(-) -- 2.6.2

2 7

[libvirt] [PATCH v2] Remove new lines from debug messages
by Jiri Denemark 04 Nov '15

04 Nov '15

VIR_DEBUG will automatically add a new line to the message, having "\n" at the end or at the beginning of the message results in empty lines. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/nwfilter/nwfilter_dhcpsnoop.c | 2 +- src/nwfilter/nwfilter_gentech_driver.c | 2 +- src/nwfilter/nwfilter_learnipaddr.c | 10 +++++----- src/rpc/virnetsocket.c | 2 +- src/util/virfile.c | 2 +- src/util/virhash.c | 2 +- src/util/virnetdevmacvlan.c | 26 ++++++++++++------------ src/util/virprocess.c | 2 +- src/xen/xend_internal.c | 2 +- tests/virhostdevtest.c | 36 +++++++++++++++++----------------- tests/virnetsockettest.c | 2 +- tests/virtimetest.c | 2 +- 12 files changed, 45 insertions(+), 45 deletions(-) diff --git a/src/nwfilter/nwfilter_dhcpsnoop.c b/src/nwfilter/nwfilter_dhcpsnoop.c index f05d4a8..bd6d25f 100644 --- a/src/nwfilter/nwfilter_dhcpsnoop.c +++ b/src/nwfilter/nwfilter_dhcpsnoop.c @@ -2017,7 +2017,7 @@ static void virNWFilterSnoopJoinThreads(void) { while (virAtomicIntGet(&virNWFilterSnoopState.nThreads) != 0) { - VIR_WARN("Waiting for snooping threads to terminate: %u\n", + VIR_WARN("Waiting for snooping threads to terminate: %u", virAtomicIntGet(&virNWFilterSnoopState.nThreads)); usleep(1000 * 1000); } diff --git a/src/nwfilter/nwfilter_gentech_driver.c b/src/nwfilter/nwfilter_gentech_driver.c index 701f8d8..5a4cff1 100644 --- a/src/nwfilter/nwfilter_gentech_driver.c +++ b/src/nwfilter/nwfilter_gentech_driver.c @@ -540,7 +540,7 @@ virNWFilterDetermineMissingVarsRec(virNWFilterDefPtr filter, if (rc) break; } else if (inc) { - VIR_DEBUG("Following filter %s\n", inc->filterref); + VIR_DEBUG("Following filter %s", inc->filterref); obj = virNWFilterObjFindByName(&driver->nwfilters, inc->filterref); if (obj) { diff --git a/src/nwfilter/nwfilter_learnipaddr.c b/src/nwfilter/nwfilter_learnipaddr.c index 5b55055..1adbadb 100644 --- a/src/nwfilter/nwfilter_learnipaddr.c +++ b/src/nwfilter/nwfilter_learnipaddr.c @@ -413,7 +413,7 @@ learnIPAddressThread(void *arg) handle = pcap_open_live(listen_if, BUFSIZ, 0, PKT_TIMEOUT_MS, errbuf); if (handle == NULL) { - VIR_DEBUG("Couldn't open device %s: %s\n", listen_if, errbuf); + VIR_DEBUG("Couldn't open device %s: %s", listen_if, errbuf); req->status = ENODEV; goto done; } @@ -448,13 +448,13 @@ learnIPAddressThread(void *arg) filter = virBufferContentAndReset(&buf); if (pcap_compile(handle, &fp, filter, 1, 0) != 0) { - VIR_DEBUG("Couldn't compile filter '%s'.\n", filter); + VIR_DEBUG("Couldn't compile filter '%s'", filter); req->status = EINVAL; goto done; } if (pcap_setfilter(handle, &fp) != 0) { - VIR_DEBUG("Couldn't set filter '%s'.\n", filter); + VIR_DEBUG("Couldn't set filter '%s'", filter); req->status = EINVAL; pcap_freecode(&fp); goto done; @@ -626,7 +626,7 @@ learnIPAddressThread(void *arg) req->filtername, req->filterparams); VIR_DEBUG("Result from applying firewall rules on " - "%s with IP addr %s : %d\n", req->ifname, inetaddr, ret); + "%s with IP addr %s : %d", req->ifname, inetaddr, ret); } } else { if (showError) @@ -638,7 +638,7 @@ learnIPAddressThread(void *arg) techdriver->applyDropAllRules(req->ifname); } - VIR_DEBUG("pcap thread terminating for interface %s\n", req->ifname); + VIR_DEBUG("pcap thread terminating for interface %s", req->ifname); virNWFilterUnlockIface(req->ifname); diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c index 5e5f1ab..526d291 100644 --- a/src/rpc/virnetsocket.c +++ b/src/rpc/virnetsocket.c @@ -196,7 +196,7 @@ int virNetSocketCheckProtocols(bool *hasIPv4, freeaddrinfo(ai); - VIR_DEBUG("Protocols: v4 %d v6 %d\n", *hasIPv4, *hasIPv6); + VIR_DEBUG("Protocols: v4 %d v6 %d", *hasIPv4, *hasIPv6); ret = 0; cleanup: diff --git a/src/util/virfile.c b/src/util/virfile.c index e5cf2c5..f45e18f 100644 --- a/src/util/virfile.c +++ b/src/util/virfile.c @@ -681,7 +681,7 @@ static int virFileLoopDeviceOpen(char **dev_name) if (virFileLoopDeviceOpenLoopCtl(dev_name, &loop_fd) < 0) return -1; - VIR_DEBUG("Return from loop-control got fd %d\n", loop_fd); + VIR_DEBUG("Return from loop-control got fd %d", loop_fd); if (loop_fd >= 0) return loop_fd; diff --git a/src/util/virhash.c b/src/util/virhash.c index bc90c44..fab621b 100644 --- a/src/util/virhash.c +++ b/src/util/virhash.c @@ -287,7 +287,7 @@ virHashGrow(virHashTablePtr table, size_t size) VIR_FREE(oldtable); #ifdef DEBUG_GROW - VIR_DEBUG("virHashGrow : from %d to %d, %ld elems\n", oldsize, + VIR_DEBUG("virHashGrow : from %d to %d, %ld elems", oldsize, size, nbElem); #endif diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index 89985b8..de345e6 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -473,16 +473,16 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, case RTM_DELLINK: case RTM_SETLINK: case RTM_GETLINK: - VIR_DEBUG(" IFINFOMSG\n"); - VIR_DEBUG(" ifi_family = 0x%02x\n", + VIR_DEBUG(" IFINFOMSG"); + VIR_DEBUG(" ifi_family = 0x%02x", ((struct ifinfomsg *)data)->ifi_family); - VIR_DEBUG(" ifi_type = 0x%x\n", + VIR_DEBUG(" ifi_type = 0x%x", ((struct ifinfomsg *)data)->ifi_type); - VIR_DEBUG(" ifi_index = %i\n", + VIR_DEBUG(" ifi_index = %i", ((struct ifinfomsg *)data)->ifi_index); - VIR_DEBUG(" ifi_flags = 0x%04x\n", + VIR_DEBUG(" ifi_flags = 0x%04x", ((struct ifinfomsg *)data)->ifi_flags); - VIR_DEBUG(" ifi_change = 0x%04x\n", + VIR_DEBUG(" ifi_change = 0x%04x", ((struct ifinfomsg *)data)->ifi_change); } /* DEBUG end */ @@ -544,23 +544,23 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, if (tb[IFLA_IFNAME]) { ifname = (char *)RTA_DATA(tb[IFLA_IFNAME]); - VIR_DEBUG("IFLA_IFNAME = %s\n", ifname); + VIR_DEBUG("IFLA_IFNAME = %s", ifname); } if (tb[IFLA_OPERSTATE]) { rem = *(unsigned short *)RTA_DATA(tb[IFLA_OPERSTATE]); - VIR_DEBUG("IFLA_OPERSTATE = %d\n", rem); + VIR_DEBUG("IFLA_OPERSTATE = %d", rem); } if (tb[IFLA_VF_PORTS]) { struct nlattr *tb_vf_ports; - VIR_DEBUG("found IFLA_VF_PORTS\n"); + VIR_DEBUG("found IFLA_VF_PORTS"); nla_for_each_nested(tb_vf_ports, tb[IFLA_VF_PORTS], rem) { - VIR_DEBUG("iterating\n"); + VIR_DEBUG("iterating"); if (nla_type(tb_vf_ports) != IFLA_VF_PORT) { - VIR_DEBUG("not a IFLA_VF_PORT. skipping\n"); + VIR_DEBUG("not a IFLA_VF_PORT. skipping"); continue; } if (nla_parse_nested(tb3, IFLA_PORT_MAX, tb_vf_ports, @@ -600,7 +600,7 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, uuid = (unsigned char *) RTA_DATA(tb3[IFLA_PORT_INSTANCE_UUID]); instance2str(uuid, instance, sizeof(instance)); - VIR_DEBUG("IFLA_PORT_INSTANCE_UUID = %s\n", + VIR_DEBUG("IFLA_PORT_INSTANCE_UUID = %s", instance); } @@ -615,7 +615,7 @@ virNetDevMacVLanVPortProfileCallback(struct nlmsghdr *hdr, } if (tb3[IFLA_PORT_RESPONSE]) { - VIR_DEBUG("IFLA_PORT_RESPONSE = %d\n", *(uint16_t *) + VIR_DEBUG("IFLA_PORT_RESPONSE = %d", *(uint16_t *) RTA_DATA(tb3[IFLA_PORT_RESPONSE])); } } diff --git a/src/util/virprocess.c b/src/util/virprocess.c index 43118f8..103c114 100644 --- a/src/util/virprocess.c +++ b/src/util/virprocess.c @@ -407,7 +407,7 @@ virProcessKillPainfully(pid_t pid, bool force) int virProcessSetAffinity(pid_t pid, virBitmapPtr map) { size_t i; - VIR_DEBUG("Set process affinity on %lld\n", (long long)pid); + VIR_DEBUG("Set process affinity on %lld", (long long)pid); # ifdef CPU_ALLOC /* New method dynamically allocates cpu mask, allowing unlimted cpus */ int numcpus = 1024; diff --git a/src/xen/xend_internal.c b/src/xen/xend_internal.c index 62ce930..21d99e3 100644 --- a/src/xen/xend_internal.c +++ b/src/xen/xend_internal.c @@ -507,7 +507,7 @@ xend_op_ext(virConnectPtr xend, const char *path, const char *key, va_list ap) return -1; content = virBufferContentAndReset(&buf); - VIR_DEBUG("xend op: %s\n", content); + VIR_DEBUG("xend op: %s", content); ret = http2unix(xend_post(xend, path, content)); VIR_FREE(content); diff --git a/tests/virhostdevtest.c b/tests/virhostdevtest.c index 065b825..faebdd4 100644 --- a/tests/virhostdevtest.c +++ b/tests/virhostdevtest.c @@ -169,14 +169,14 @@ testVirHostdevPreparePCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); /* Test normal functionality */ - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, NULL, 0, 0) < 0) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); /* Test unmanaged hostdevs */ - VIR_DEBUG("Test >=1 unmanaged hostdevs\n"); + VIR_DEBUG("Test >=1 unmanaged hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, hostdevs, nhostdevs, 0) < 0) goto cleanup; @@ -186,21 +186,21 @@ testVirHostdevPreparePCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) /* Test conflict */ count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); - VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, &hostdevs[0], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); CHECK_LIST_COUNT(mgr->inactivePCIHostdevs, count2); - VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, "test_domain1", uuid, &hostdevs[1], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); CHECK_LIST_COUNT(mgr->inactivePCIHostdevs, count2); - VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, "test_driver1", dom_name, uuid, &hostdevs[2], 1, 0)) goto cleanup; @@ -223,7 +223,7 @@ testVirHostdevReAttachPCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) for (i = 0; i < nhostdevs; i++) { if (hostdevs[i]->managed != false) { - VIR_DEBUG("invalid test\n"); + VIR_DEBUG("invalid test"); return -1; } } @@ -231,11 +231,11 @@ testVirHostdevReAttachPCIHostdevs_unmanaged(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); count2 = virPCIDeviceListCount(mgr->inactivePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, NULL, 0, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 unmanaged hostdevs\n"); + VIR_DEBUG("Test >=1 unmanaged hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, hostdevs, nhostdevs, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1 - 3); @@ -261,7 +261,7 @@ testVirHostdevPreparePCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); /* Test normal functionality */ - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); if (virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, hostdevs, nhostdevs, 0) < 0) goto cleanup; @@ -269,19 +269,19 @@ testVirHostdevPreparePCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) /* Test conflict */ count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, dom_name, uuid, &hostdevs[0], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for same driver, diff domain again"); if (!virHostdevPreparePCIDevices(mgr, drv_name, "test_domain1", uuid, &hostdevs[1], 1, 0)) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again\n"); + VIR_DEBUG("Test: prepare same hostdevs for diff driver/domain again"); if (!virHostdevPreparePCIDevices(mgr, "test_driver1", dom_name, uuid, &hostdevs[2], 1, 0)) goto cleanup; @@ -303,18 +303,18 @@ testVirHostdevReAttachPCIHostdevs_managed(const void *oaque ATTRIBUTE_UNUSED) for (i = 0; i < nhostdevs; i++) { if (hostdevs[i]->managed != true) { - VIR_DEBUG("invalid test\n"); + VIR_DEBUG("invalid test"); return -1; } } count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, NULL, 0, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); virHostdevReAttachPCIDevices(mgr, drv_name, dom_name, hostdevs, nhostdevs, NULL); CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1 - 3); @@ -392,13 +392,13 @@ testVirHostdevUpdateActivePCIHostdevs(const void *oaque ATTRIBUTE_UNUSED) count1 = virPCIDeviceListCount(mgr->activePCIHostdevs); - VIR_DEBUG("Test 0 hostdevs\n"); + VIR_DEBUG("Test 0 hostdevs"); if (virHostdevUpdateActivePCIDevices(mgr, NULL, 0, drv_name, dom_name) < 0) goto cleanup; CHECK_LIST_COUNT(mgr->activePCIHostdevs, count1); - VIR_DEBUG("Test >=1 hostdevs\n"); + VIR_DEBUG("Test >=1 hostdevs"); if (virHostdevUpdateActivePCIDevices(mgr, hostdevs, nhostdevs, drv_name, dom_name) < 0) goto cleanup; @@ -432,7 +432,7 @@ mymain(void) # define DO_TEST(fnc) \ do { \ - VIR_DEBUG("\nTesting: %s", #fnc); \ + VIR_DEBUG("Testing: %s", #fnc); \ if (virtTestRun(#fnc, fnc, NULL) < 0) \ ret = -1; \ } while (0) diff --git a/tests/virnetsockettest.c b/tests/virnetsockettest.c index ce9eeab..5786870 100644 --- a/tests/virnetsockettest.c +++ b/tests/virnetsockettest.c @@ -105,7 +105,7 @@ checkProtocols(bool *hasIPv4, bool *hasIPv6, break; } - VIR_DEBUG("Choose port %d\n", *freePort); + VIR_DEBUG("Choose port %d", *freePort); ret = 0; diff --git a/tests/virtimetest.c b/tests/virtimetest.c index 9ab38ae..49040d3 100644 --- a/tests/virtimetest.c +++ b/tests/virtimetest.c @@ -92,7 +92,7 @@ testTimeLocalOffset(const void *args) return -1; if (data->offset != actual) { - VIR_DEBUG("Expect Offset %ld got %ld\n", + VIR_DEBUG("Expect Offset %ld got %ld", data->offset, actual); return -1; } -- 2.6.2

2 2

[libvirt] [REPOST PATCH 0/6] Have 'buildVol' callers to clean up after themselves
by John Ferlan 04 Nov '15

04 Nov '15

Since some time and changes have occurred since the original posting, I figured it would be just easier to updated and repost the remainder of the series: http://www.redhat.com/archives/libvir-list/2015-October/msg00312.html Where the patches 1-3 are v2's of the original patches 8 & 9: http://www.redhat.com/archives/libvir-list/2015-October/msg00515.html http://www.redhat.com/archives/libvir-list/2015-October/msg00516.html http://www.redhat.com/archives/libvir-list/2015-October/msg00518.html Note that patches 4-6 have been ACK'd already as patches 10-12, it's just that without the previous patches it doesn't make much sense to push them alone. John Ferlan (6): storage: Refactor and fix virStorageBackendCreateExecCommand storage: Cleanup failures virStorageBackendCreateExecCommand storage: Cleanup failures in virStorageBackendCreateRaw storage: Pull volume removal from pool in storageVolDeleteInternal Revert "storage: Prior to creating a volume, refresh the pool" storage: On 'buildVol' failure don't delete the volume src/storage/storage_backend.c | 40 ++++++++++++++++++++++++++------------- src/storage/storage_backend.h | 10 ++++++++++ src/storage/storage_driver.c | 44 +++++++++++++++++++++---------------------- 3 files changed, 59 insertions(+), 35 deletions(-) -- 2.1.0

2 17

[libvirt] [PATCH] Revert "utils: Remove the logging of errors from virNetDevSendEthtoolIoctl"
by Daniel P. Berrange 03 Nov '15

03 Nov '15

This reverts commit 6f2a0198e913c91a2ef8b99db79b7d3cc5396957. This commit removed error reporting from virNetDevSendEthtoolIoctl pushing responsibility onto the callers. This is wrong, however, since virNetDevSendEthtoolIoctl calls virNetDevSetupControl which can still report errors. So as a result virNetDevSendEthtoolIoctl may or may not report errors depending on which bit of it fails, and as a result callers now overwrite some errors. It also introduced a regression causing unprivileged libvirtd to spew error messages to the console due to inability to query the NIC features, an error which was previously ignored. virNetDevSetupControlFull:148 : Cannot open network interface control socket: Operation not permitted virNetDevFeatureAvailable:3062 : Cannot get device wlp3s0 flags: Operation not permitted virNetDevSetupControlFull:148 : Cannot open network interface control socket: Operation not permitted virNetDevFeatureAvailable:3062 : Cannot get device wlp3s0 flags: Operation not permitted virNetDevSetupControlFull:148 : Cannot open network interface control socket: Operation not permitted virNetDevFeatureAvailable:3062 : Cannot get device wlp3s0 flags: Operation not permitted virNetDevSetupControlFull:148 : Cannot open network interface control socket: Operation not permitted virNetDevFeatureAvailable:3062 : Cannot get device wlp3s0 flags: Operation not permitted Looking back at the original posting I see no explanation of why thsi refactoring was needed, so reverting the clearly broken error reporting logic looks like the best option. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/util/virnetdev.c | 56 +++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 40 insertions(+), 16 deletions(-) diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index 9789e93..657df3a 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -3145,15 +3145,39 @@ static int virNetDevSendEthtoolIoctl(const char *ifname, void *cmd) { int ret = -1; - int fd; - struct ifreq ifr; + int sock = -1; + virIfreq ifr; - if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0) - return ret; + sock = socket(AF_LOCAL, SOCK_DGRAM, 0); + if (sock < 0) { + virReportSystemError(errno, "%s", _("Cannot open control socket")); + goto cleanup; + } + + memset(&ifr, 0, sizeof(ifr)); + strcpy(ifr.ifr_name, ifname); ifr.ifr_data = cmd; - ret = ioctl(fd, SIOCETHTOOL, &ifr); + ret = ioctl(sock, SIOCETHTOOL, &ifr); + if (ret != 0) { + switch (errno) { + case EPERM: + VIR_DEBUG("ethtool ioctl: permission denied"); + break; + case EINVAL: + VIR_DEBUG("ethtool ioctl: invalid request"); + break; + case EOPNOTSUPP: + VIR_DEBUG("ethtool ioctl: request not supported"); + break; + default: + virReportSystemError(errno, "%s", _("ethtool ioctl error")); + goto cleanup; + } + } - VIR_FORCE_CLOSE(fd); + cleanup: + if (sock) + VIR_FORCE_CLOSE(sock); return ret; } @@ -3170,12 +3194,12 @@ virNetDevSendEthtoolIoctl(const char *ifname, void *cmd) static int virNetDevFeatureAvailable(const char *ifname, struct ethtool_value *cmd) { + int ret = -1; + cmd = (void*)cmd; - if (virNetDevSendEthtoolIoctl(ifname, cmd) < 0) { - virReportSystemError(errno, _("Cannot get device %s flags"), ifname); - return -1; - } - return cmd->data > 0 ? 1 : 0; + if (!virNetDevSendEthtoolIoctl(ifname, cmd)) + ret = cmd->data > 0 ? 1 : 0; + return ret; } @@ -3192,12 +3216,12 @@ virNetDevFeatureAvailable(const char *ifname, struct ethtool_value *cmd) static int virNetDevGFeatureAvailable(const char *ifname, struct ethtool_gfeatures *cmd) { + int ret = -1; + cmd = (void*)cmd; - if (virNetDevSendEthtoolIoctl(ifname, cmd) < 0) { - virReportSystemError(errno, _("Cannot get device %s generic features"), ifname); - return -1; - } - return FEATURE_BIT_IS_SET(cmd->features, TX_UDP_TNL, active); + if (!virNetDevSendEthtoolIoctl(ifname, cmd)) + ret = FEATURE_BIT_IS_SET(cmd->features, TX_UDP_TNL, active); + return ret; } # endif -- 2.5.0

2 4

[libvirt] [PATCH] virkeymaps.h: Fix VPATH build with --disable-dependency-tracking
by Michal Privoznik 03 Nov '15

03 Nov '15

Funny things happen when you try to do a VPATH build and pass --disable-dependency-tracking argument to the configure script. Not only the directory structure is not created at the end of configure phase, but also contradictory to others, virkeymaps.h does not depend on src/utils dirstamp. I don't know how to force automake to generate that rule, but this is the diff that makes things work for me again. Yes, as you can see it more or less reverts a9fe620372144db3e432bd7506cb909cc3179ef8. More than less. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Steps to reproduce: 1) mkdir /tmp/test_libvirt && cd $_ 2) $OLDPWD/configure --disable-dependency-tracking 3) make src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Makefile.am b/src/Makefile.am index 99b4993..f98912f 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -181,7 +181,7 @@ MAINTAINERCLEANFILES += util/virkeymaps.h util/virkeymaps.h: $(srcdir)/util/keymaps.csv \ $(srcdir)/util/virkeycode-mapgen.py $(AM_V_GEN)$(PYTHON) $(srcdir)/util/virkeycode-mapgen.py \ - <$(srcdir)/util/keymaps.csv >util/virkeymaps.h + <$(srcdir)/util/keymaps.csv >$(srcdir)/util/virkeymaps.h # Internal generic driver infrastructure NODE_INFO_SOURCES = nodeinfo.h nodeinfo.c nodeinfopriv.h -- 2.4.10

3 5

[libvirt] [PATCH] bhyve: add bootrom support
by Roman Bogorodskiy 03 Nov '15

03 Nov '15

Recently, bhyve got UEFI support that eliminates necessity of having two-step process of starting a VM, i.e. loading OS using loader like bhyveloader or grub and then calling bhyve itself. >From user perspective usage looks the following: if a domain XML contains the '<bootloader>' element and it's value is a path to non-executable file, then it's treated as a path to bootrom file and external loader is not execute. All the existing behavior remains the same, e.g. if bootloader is not specified, bhyveloader is used. If loader is specified and it points to an executable file, it's executed before running bhyve. Implementation consists of two major parts: * Add probing for capability by checking if the current bhyve binary accepts '-l bootrom' argument meaning that UEFI is supported * virBhyveProcessBuildBhyveCmd is extended with 'external_loader' argument. If it decides (based on the schema described above) that external loader is not needed, it sets it to false. All the related code changed accordingly to act based on value of external_loader. --- src/bhyve/bhyve_capabilities.c | 11 ++++ src/bhyve/bhyve_capabilities.h | 1 + src/bhyve/bhyve_command.c | 20 +++++++- src/bhyve/bhyve_command.h | 1 + src/bhyve/bhyve_driver.c | 15 +++--- src/bhyve/bhyve_process.c | 58 ++++++++++++---------- .../bhyvexml2argvdata/bhyvexml2argv-bhyveload.args | 3 ++ .../bhyvexml2argv-bhyveload.ldargs | 1 + .../bhyvexml2argvdata/bhyvexml2argv-bhyveload.xml | 24 +++++++++ tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.args | 4 ++ tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.xml | 23 +++++++++ tests/bhyvexml2argvmock.c | 13 +++++ tests/bhyvexml2argvtest.c | 39 ++++++++------- 13 files changed, 162 insertions(+), 51 deletions(-) create mode 100644 tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.args create mode 100644 tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.ldargs create mode 100644 tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.xml create mode 100644 tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.args create mode 100644 tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.xml diff --git a/src/bhyve/bhyve_capabilities.c b/src/bhyve/bhyve_capabilities.c index d2970a2..fbdd63c 100644 --- a/src/bhyve/bhyve_capabilities.c +++ b/src/bhyve/bhyve_capabilities.c @@ -166,6 +166,17 @@ virBhyveProbeCaps(unsigned int *caps) if (strstr(help, "-u:") != NULL) *caps |= BHYVE_CAP_RTC_UTC; + virCommandFree(cmd); + + cmd = virCommandNew(binary); + virCommandAddArgList(cmd, "-l", "bootrom", NULL); + if (virCommandRun(cmd, &exit) < 0) { + ret = -1; + goto out; + } + if (exit == 1) + *caps |= BHYVE_CAP_BOOTROM; + out: VIR_FREE(help); virCommandFree(cmd); diff --git a/src/bhyve/bhyve_capabilities.h b/src/bhyve/bhyve_capabilities.h index 0eb22a4..25dfb2c 100644 --- a/src/bhyve/bhyve_capabilities.h +++ b/src/bhyve/bhyve_capabilities.h @@ -33,6 +33,7 @@ typedef enum { typedef enum { BHYVE_CAP_RTC_UTC = 1, + BHYVE_CAP_BOOTROM = 2, } virBhyveCapsFlags; int virBhyveProbeGrubCaps(virBhyveGrubCapsFlags *caps); diff --git a/src/bhyve/bhyve_command.c b/src/bhyve/bhyve_command.c index 6576029..10d504e 100644 --- a/src/bhyve/bhyve_command.c +++ b/src/bhyve/bhyve_command.c @@ -216,7 +216,9 @@ bhyveBuildDiskArgStr(const virDomainDef *def ATTRIBUTE_UNUSED, virCommandPtr virBhyveProcessBuildBhyveCmd(virConnectPtr conn, - virDomainDefPtr def, bool dryRun) + virDomainDefPtr def, + bool *external_loader, + bool dryRun) { /* * /usr/sbin/bhyve -c 2 -m 256 -AI -H -P \ @@ -230,6 +232,22 @@ virBhyveProcessBuildBhyveCmd(virConnectPtr conn, virCommandPtr cmd = virCommandNew(BHYVE); + /* Check what loader to use */ + if ((def->os.bootloader != NULL) && + (!virFileIsExecutable(def->os.bootloader))) { + *external_loader = false; + if ((bhyveDriverGetCaps(conn) & BHYVE_CAP_BOOTROM) == 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Installed bhyve binary does not support " + "bootrom")); + goto error; + } + virCommandAddArg(cmd, "-l"); + virCommandAddArgFormat(cmd, "bootrom,%s", def->os.bootloader); + } else { + *external_loader = true; + } + /* CPUs */ virCommandAddArg(cmd, "-c"); virCommandAddArgFormat(cmd, "%d", def->vcpus); diff --git a/src/bhyve/bhyve_command.h b/src/bhyve/bhyve_command.h index 22a959d..20a3bf0 100644 --- a/src/bhyve/bhyve_command.h +++ b/src/bhyve/bhyve_command.h @@ -32,6 +32,7 @@ virCommandPtr virBhyveProcessBuildBhyveCmd(virConnectPtr conn, virDomainDefPtr def, + bool *external_loader, bool dryRun); virCommandPtr diff --git a/src/bhyve/bhyve_driver.c b/src/bhyve/bhyve_driver.c index 4840a0e..7290251 100644 --- a/src/bhyve/bhyve_driver.c +++ b/src/bhyve/bhyve_driver.c @@ -698,6 +698,7 @@ bhyveConnectDomainXMLToNative(virConnectPtr conn, virDomainDefPtr def = NULL; virCommandPtr cmd = NULL, loadcmd = NULL; virCapsPtr caps = NULL; + bool external_loader = NULL; char *ret = NULL; virCheckFlags(0, NULL); @@ -721,15 +722,17 @@ bhyveConnectDomainXMLToNative(virConnectPtr conn, if (bhyveDomainAssignAddresses(def, NULL) < 0) goto cleanup; - if (!(loadcmd = virBhyveProcessBuildLoadCmd(conn, def, "<device.map>", - NULL))) + if (!(cmd = virBhyveProcessBuildBhyveCmd(conn, def, &external_loader, true))) goto cleanup; - if (!(cmd = virBhyveProcessBuildBhyveCmd(conn, def, true))) - goto cleanup; + if (external_loader) { + if (!(loadcmd = virBhyveProcessBuildLoadCmd(conn, def, "<device.map>", + NULL))) + goto cleanup; - virBufferAdd(&buf, virCommandToString(loadcmd), -1); - virBufferAddChar(&buf, '\n'); + virBufferAdd(&buf, virCommandToString(loadcmd), -1); + virBufferAddChar(&buf, '\n'); + } virBufferAdd(&buf, virCommandToString(cmd), -1); if (virBufferCheckError(&buf) < 0) diff --git a/src/bhyve/bhyve_process.c b/src/bhyve/bhyve_process.c index 284641a..d62cccd 100644 --- a/src/bhyve/bhyve_process.c +++ b/src/bhyve/bhyve_process.c @@ -114,6 +114,7 @@ virBhyveProcessStart(virConnectPtr conn, virCommandPtr load_cmd = NULL; bhyveConnPtr privconn = conn->privateData; int ret = -1, rc; + bool external_loader = NULL; if (virAsprintf(&logfile, "%s/%s.log", BHYVE_LOG_DIR, vm->def->name) < 0) @@ -150,6 +151,7 @@ virBhyveProcessStart(virConnectPtr conn, /* Call bhyve to start the VM */ if (!(cmd = virBhyveProcessBuildBhyveCmd(conn, vm->def, + &external_loader, false))) goto cleanup; @@ -159,38 +161,40 @@ virBhyveProcessStart(virConnectPtr conn, virCommandSetPidFile(cmd, privconn->pidfile); virCommandDaemonize(cmd); - /* Now bhyve command is constructed, meaning the - * domain is ready to be started, so we can build - * and execute bhyveload command */ - rc = virBhyveFormatDevMapFile(vm->def->name, &devmap_file); - if (rc < 0) - goto cleanup; - - if (!(load_cmd = virBhyveProcessBuildLoadCmd(conn, vm->def, devmap_file, - &devicemap))) - goto cleanup; - virCommandSetOutputFD(load_cmd, &logfd); - virCommandSetErrorFD(load_cmd, &logfd); + if (external_loader) { + /* Now bhyve command is constructed, meaning the + * domain is ready to be started, so we can build + * and execute bhyveload command */ + rc = virBhyveFormatDevMapFile(vm->def->name, &devmap_file); + if (rc < 0) + goto cleanup; - if (devicemap != NULL) { - rc = virFileWriteStr(devmap_file, devicemap, 0644); - if (rc) { - virReportSystemError(errno, - _("Cannot write device.map '%s'"), - devmap_file); + if (!(load_cmd = virBhyveProcessBuildLoadCmd(conn, vm->def, devmap_file, + &devicemap))) goto cleanup; + virCommandSetOutputFD(load_cmd, &logfd); + virCommandSetErrorFD(load_cmd, &logfd); + + if (devicemap != NULL) { + rc = virFileWriteStr(devmap_file, devicemap, 0644); + if (rc) { + virReportSystemError(errno, + _("Cannot write device.map '%s'"), + devmap_file); + goto cleanup; + } } - } - /* Log generated command line */ - virCommandWriteArgLog(load_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))); + /* Log generated command line */ + virCommandWriteArgLog(load_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_DEBUG("Loading domain '%s'", vm->def->name); - if (virCommandRun(load_cmd, NULL) < 0) - goto cleanup; + VIR_DEBUG("Loading domain '%s'", vm->def->name); + if (virCommandRun(load_cmd, NULL) < 0) + goto cleanup; + } /* Now we can start the domain */ VIR_DEBUG("Starting domain '%s'", vm->def->name); diff --git a/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.args b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.args new file mode 100644 index 0000000..118735e --- /dev/null +++ b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.args @@ -0,0 +1,3 @@ +/usr/sbin/bhyve -c 1 -m 214 -u -H -P -s 0:0,hostbridge \ +-s 3:0,virtio-net,faketapdev,mac=52:54:00:00:00:00 \ +-s 2:0,ahci-hd,/tmp/freebsd.img bhyve diff --git a/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.ldargs b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.ldargs new file mode 100644 index 0000000..56f7fb3 --- /dev/null +++ b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.ldargs @@ -0,0 +1 @@ +/usr/sbin/bhyveload -foo diff --git a/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.xml b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.xml new file mode 100644 index 0000000..9dac10c --- /dev/null +++ b/tests/bhyvexml2argvdata/bhyvexml2argv-bhyveload.xml @@ -0,0 +1,24 @@ +<domain type='bhyve'> + <name>bhyve</name> + <uuid>df3be7e7-a104-11e3-aeb0-50e5492bd3dc</uuid> + <memory>219136</memory> + <vcpu>1</vcpu> + <bootloader>/usr/sbin/bhyveload</bootloader> + <bootloader_args>-foo</bootloader_args> + <os> + <type>hvm</type> + </os> + <devices> + <disk type='file'> + <driver name='file' type='raw'/> + <source file='/tmp/freebsd.img'/> + <target dev='hda' bus='sata'/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> + </disk> + <interface type='bridge'> + <model type='virtio'/> + <source bridge="virbr0"/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> + </interface> + </devices> +</domain> diff --git a/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.args b/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.args new file mode 100644 index 0000000..d46ffab --- /dev/null +++ b/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.args @@ -0,0 +1,4 @@ +/usr/sbin/bhyve -l bootrom,/path/to/rom.fd \ +-c 1 -m 214 -u -H -P -s 0:0,hostbridge \ +-s 3:0,virtio-net,faketapdev,mac=52:54:00:00:00:00 \ +-s 2:0,ahci-hd,/tmp/freebsd.img bhyve diff --git a/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.xml b/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.xml new file mode 100644 index 0000000..2da80c5 --- /dev/null +++ b/tests/bhyvexml2argvdata/bhyvexml2argv-bootrom.xml @@ -0,0 +1,23 @@ +<domain type='bhyve'> + <name>bhyve</name> + <uuid>df3be7e7-a104-11e3-aeb0-50e5492bd3dc</uuid> + <memory>219136</memory> + <vcpu>1</vcpu> + <bootloader>/path/to/rom.fd</bootloader> + <os> + <type>hvm</type> + </os> + <devices> + <disk type='file'> + <driver name='file' type='raw'/> + <source file='/tmp/freebsd.img'/> + <target dev='hda' bus='sata'/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> + </disk> + <interface type='bridge'> + <model type='virtio'/> + <source bridge="virbr0"/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> + </interface> + </devices> +</domain> diff --git a/tests/bhyvexml2argvmock.c b/tests/bhyvexml2argvmock.c index 41058ca..797dc20 100644 --- a/tests/bhyvexml2argvmock.c +++ b/tests/bhyvexml2argvmock.c @@ -1,5 +1,6 @@ #include <config.h> +#include "virfile.h" #include "virstring.h" #include "virnetdev.h" #include "virnetdevtap.h" @@ -48,3 +49,15 @@ int virNetDevSetOnline(const char *ifname ATTRIBUTE_UNUSED, { return 0; } + +bool virFileIsExecutable(const char *file) +{ + if (STREQ(file, "/fizz_buzz_bazz")) + return true; + else if (STREQ(file, "/usr/local/sbin/grub-bhyve")) + return true; + else if (STREQ(file, "/usr/sbin/bhyveload")) + return true; + + return false; +} diff --git a/tests/bhyvexml2argvtest.c b/tests/bhyvexml2argvtest.c index 3e57a78..2ff1ae3 100644 --- a/tests/bhyvexml2argvtest.c +++ b/tests/bhyvexml2argvtest.c @@ -23,6 +23,7 @@ static int testCompareXMLToArgvFiles(const char *xml, virDomainDefPtr vmdef = NULL; virCommandPtr cmd = NULL, ldcmd = NULL; virConnectPtr conn; + bool external_loader = NULL; int ret = -1; if (!(conn = virGetConnect())) @@ -34,33 +35,35 @@ static int testCompareXMLToArgvFiles(const char *xml, conn->privateData = &driver; - if (!(cmd = virBhyveProcessBuildBhyveCmd(conn, vmdef, false))) + if (!(cmd = virBhyveProcessBuildBhyveCmd(conn, vmdef, &external_loader, false))) goto out; if (!(actualargv = virCommandToString(cmd))) goto out; - if (!(ldcmd = virBhyveProcessBuildLoadCmd(conn, vmdef, "<device.map>", - &actualdm))) - goto out; - - if (actualdm != NULL) - virTrimSpaces(actualdm, NULL); - - if (!(actualld = virCommandToString(ldcmd))) - goto out; + if (external_loader) { + if (!(ldcmd = virBhyveProcessBuildLoadCmd(conn, vmdef, "<device.map>", + &actualdm))) + goto out; - if (virtTestCompareToFile(actualargv, cmdline) < 0) - goto out; + if (actualdm != NULL) + virTrimSpaces(actualdm, NULL); - if (virtTestCompareToFile(actualld, ldcmdline) < 0) - goto out; + if (!(actualld = virCommandToString(ldcmd))) + goto out; - if (virFileExists(dmcmdline) || actualdm) { - if (virtTestCompareToFile(actualdm, dmcmdline) < 0) + if (virtTestCompareToFile(actualld, ldcmdline) < 0) goto out; + + if (virFileExists(dmcmdline) || actualdm) { + if (virtTestCompareToFile(actualdm, dmcmdline) < 0) + goto out; + } } + if (virtTestCompareToFile(actualargv, cmdline) < 0) + goto out; + ret = 0; out: @@ -118,7 +121,7 @@ mymain(void) } while (0) driver.grubcaps = BHYVE_GRUB_CAP_CONSDEV; - driver.bhyvecaps = BHYVE_CAP_RTC_UTC; + driver.bhyvecaps = BHYVE_CAP_RTC_UTC | BHYVE_CAP_BOOTROM; DO_TEST("base"); DO_TEST("acpiapic"); @@ -130,8 +133,10 @@ mymain(void) DO_TEST("grub-defaults"); DO_TEST("grub-bootorder"); DO_TEST("grub-bootorder2"); + DO_TEST("bhyveload"); DO_TEST("bhyveload-explicitargs"); DO_TEST("custom-loader"); + DO_TEST("bootrom"); DO_TEST("disk-cdrom-grub"); DO_TEST("serial-grub"); DO_TEST("localtime"); -- 2.6.1

2 3

[libvirt] Ten years of libvirt
by Michal Privoznik 03 Nov '15

03 Nov '15

Dear list, join me on this big day in congratulations to libvirt that has just turned 10 and is starting new decade of its life. At the same time big thanks to all who contributed in any way. Cheers! Michal

6 6

Re: [libvirt] [PATCH 7/8] Postpone reprobing till all the devices in iommu group are unbound from vfio
by Andrea Bolognani 02 Nov '15

02 Nov '15

On Mon, 2015-11-02 at 10:06 +0530, Shivaprasad bhat wrote: > > > +{ > > > + int ret = 0; > > > + virPCIDevicePtr pci = NULL; > > > + > > > + if (!(pci = virPCIDeviceNew(devAddr->domain, devAddr->bus, > > > + devAddr->slot, devAddr->function))) > > > + goto cleanup; > > > + > > > + if (STREQ_NULLABLE(pci->stubDriver, "vfio-pci")) > > > + ret = -1; > > > > As mentioned in the comment for Patch 1, I think this is > > fairly obscure: if I had not read throught the whole > > series, I'd assume this checks whether the device is > > configured to use vfio-pci as stub driver, not whether > > it is currently bound to it, and it would not be > > immediately clear to me how this check fits in a function > > called virPCIDeviceBoundToVFIODriver(). > > > > I think it would be much cleaner to query the driver > > explicitly using virPCIDeviceGetDriverPathAndName() and > > remove that call from virPCIDeviceNew(). > > > > The PCIDeviceNew does exactly the same of going through > the sysfs and populating it. Its only code which is as though > the stubDriver scan is happening inside than explicitly outside. > > The Iterator today takes virPCIDeviceAddressPtr and > virPCIDeviceGetDriverPathAndName for which we anyway need to > call virPCIDevNew. I am just avoiding one extra function call by > moving it inside virPCIDevNew The problem I have with this approach is that, as far as I understand, before your patch dev->stubDriver is a configuration item: you set it to whatever's appropriate, and the value is later inspected in virPCIDeviceBindToStub()[1] to decide what stub driver the device should be bound to. After your patch, depending on the context, it could be either a configuration item, as before, or state, eg. the current stub driver the device is bound to, with no clear way to tell one situation from the other. I think using a single attribute to store related but different information should be avoided, as it has the potential to make the code much harder to understand, especially later on for someone who's not been involved in the implementation. Moreover you only need that value for comparison purposes in a couple of places, and by moving the code to virPCIDeviceNew() you make it so the value is loaded (by performing filesystem operations) every single time a virDevice is created, whether the value is actually used or not. Cheers. [1] Rather virPCIDeviceDetach() in reality - I have a pending patch that moves it to virPCIDeviceBindToStub() where it IMHO belongs -- Andrea Bolognani Software Engineer - Virtualization Team

1 0

[libvirt] [PATCH v6 00/13] qemu: Add quorum support to libvirt
by Matthias Gatto 02 Nov '15

02 Nov '15

The purpose of these patches is to introduce quorum for libvirt I've try to follow this proposal: http://www.redhat.com/archives/libvir-list/2014-May/msg00533.html This feature ask for 6 task: 1) Allow a _virStorageSource to contain more than one backing store. Because all the actual libvirt code use the backingStore field as a pointer and we needs want to change that, I've decide to encapsulate the backingStore field to simplifie the array manipulation. 2) Add the missing field a quorum need in _virStorageSource and the VIR_STORAGE_TYPE_QUORUM and VIR_STORAGE_FILE_QUORUM in their respectives enums. 3) Parse and format the xml Because a quorum allows to have more than one backing store at the same level we need to change virDomainDiskDefFormat and virDomainDiskDefParseXML to call virDomainDiskBackingStoreFormat and virDomainDiskBackingStoreParse in a loop. virDomainDiskBackingStoreFormat and virDomainDiskBackingStoreParse can call themself recursively in a loop because a quorum can contain another quorum 4) Build qemu string As for the xml, we have to call the function which create quorum recursively. But this task have the problem explained here: http://www.redhat.com/archives/libvir-list/2014-October/msg00529.html The _virStorageSource missing some informations that can be passed to a child, and therefore this version of quorum is incomplet. 5) Allow to hotplug/change a disk in a quorum This part is not present in these patches because for this task we have to use blockdev-add, and currently libvirt use device_add for hotpluging that doesn't allow to hotplug quorum childs. There is 3 way to handle this problem: 1) create a virDomainBlockDevAdd function in libvirt witch call blockdev-add. 2) use blockdev-add instead of device_add in qemuMonitorJSONAddDevice 3) write a hack which uses blockdev-add only when attaching a quorum child V2: -Rebase on master -Add Documentation V3: -Transforme the backingStore field in virStorageSource into an array of pointer instead of a pointer -Modify virStorageSourceSetBackingStore to allow it to expand the backingStore size. V4: -Rebase on master V5: -Rebase on master -patch 1-4/9: use patchs from John Ferlan -patch 4/9: check return of virStorageSourceSetBackingStore -patch 5/9: report type of error on virStorageSourceSetBackingStore v6 note: First at all, I'm sorry for the time between v5 and v6, I had other projects to work on and was unable to work at full time on libvirt, moreover I've try at first to support all snapshot and block jobs operations for quorum, but encounter a lot a problems. At the end I had a versions which was only handling some few operations, so I've block all unsupported operations for quorum, I plan to continue working on the quorum unsupported operations, and send it when it will be ready, but in the meantime I hope to upstream this serie of patch which should provide a very basic(but stable) quorum suport. v6 modifications: -Rebase on master -Remove node-name patch -patch 06/13: Add virStorageSourceIsRAID -patch 10/13: Add virDomainDefHasRAID -patch 11-13/13: Block all unsupported operations Matthias Gatto (13): virstoragefile: Add virStorageSourceGetBackingStore virstoragefile: Always use virStorageSourceGetBackingStore to get backing store virstoragefile: Add virStorageSourceSetBackingStore virstoragefile: Always use virStorageSourceSetBackingStore to set backing store virstoragefile: change backingStore to backingStores. virstoragefile: Add virStorageSourceIsRAID virstoragefile: Add quorum in virstoragefile domain_conf: Read and Write quorum config qemu: Add quorum support in qemuBuildDriveDevStr domain: add virDomainDefHasRAID qemu: Block snapshot operation with RAID qemu: qemuDomainGetBlockInfo quorum support qemu: qemuDiskPathToAlias quorum support docs/formatdomain.html.in | 23 +++- docs/schemas/domaincommon.rng | 21 +++- docs/schemas/storagecommon.rng | 1 + docs/schemas/storagevol.rng | 1 + src/conf/domain_conf.c | 196 +++++++++++++++++++++++++--------- src/conf/domain_conf.h | 1 + src/conf/storage_conf.c | 23 ++-- src/libvirt_private.syms | 4 + src/qemu/qemu_cgroup.c | 4 +- src/qemu/qemu_command.c | 94 ++++++++++++++++ src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_driver.c | 50 ++++++--- src/qemu/qemu_migration.c | 1 + src/qemu/qemu_monitor_json.c | 4 +- src/security/security_dac.c | 2 +- src/security/security_selinux.c | 4 +- src/security/virt-aa-helper.c | 2 +- src/storage/storage_backend.c | 22 ++-- src/storage/storage_backend_fs.c | 38 ++++--- src/storage/storage_backend_gluster.c | 11 +- src/storage/storage_backend_logical.c | 15 ++- src/storage/storage_driver.c | 3 +- src/util/virstoragefile.c | 118 +++++++++++++++++--- src/util/virstoragefile.h | 15 ++- tests/virstoragetest.c | 18 ++-- 25 files changed, 523 insertions(+), 150 deletions(-) -- 2.6.1

2 23

[libvirt] [PATCH v3 0/9] Few VFIO related fixes
by Shivaprasad G Bhat 02 Nov '15

02 Nov '15

The following series implements... --- Addressed comments on V2 and V1 Fixed the while loop exit in P8 virPCIDeviceUnbindFromStub() Shivaprasad G Bhat (9): Implement virPCIIsKnownStub function Initialize the stubDriver of pci devices if bound to a valid one Add iommu group number info to virPCIDevice Refuse to reattach from vfio if the iommu group is in use by any domain Wait for vfio-pci device cleanups before reassinging the device to host driver Split reprobe action from the virPCIUnbindFromStub into a new function Pass activeDevs and inactiveDevs to virPCIDeviceUnbindFromStub and virPCIDeviceBindToStub Postpone reprobing till all the devices in iommu group are unbound from vfio Wait for iommmu device to go away before reprobing the host driver src/libvirt_private.syms | 1 src/util/virhostdev.c | 18 +++ src/util/virpci.c | 300 +++++++++++++++++++++++++++++++++++++++------- src/util/virpci.h | 2 tests/virpcimock.c | 187 ++++++++++++++++++++++++++--- tests/virpcitest.c | 152 +++++++++++++++++++---- 6 files changed, 566 insertions(+), 94 deletions(-) -- Signature

1 9

[libvirt] [PATCH 0/4] *** Don't include .c file for editing support ***
by rodinasophie＠gmail.com 02 Nov '15

02 Nov '15

From: sonya <rodinasophie(a)gmail.com> *** BLURB HERE *** sonya (4): General structures and interfaces were added Virsh-edit interface was added Comment in virsh-edit.h was changed Style-errors are fixed tools/Makefile.am | 1 + tools/virsh-domain.c | 187 ++++++++++++++++++++++++++++++++++++------------ tools/virsh-edit.c | 100 ++++++-------------------- tools/virsh-edit.h | 42 +++++++++++ tools/virsh-interface.c | 64 +++++++++++++---- tools/virsh-network.c | 62 ++++++++++++---- tools/virsh-nwfilter.c | 64 +++++++++++++---- tools/virsh-pool.c | 64 +++++++++++++---- tools/virsh-snapshot.c | 76 ++++++++++++++------ 9 files changed, 464 insertions(+), 196 deletions(-) create mode 100644 tools/virsh-edit.h -- 2.1.4

2 6

[libvirt] [PATCH] syntax-check: Add sc_prohibit_space_in_label rule
by Andrea Bolognani 02 Nov '15

02 Nov '15

This guards against code such as cleanup : which is happily accepted by the compiler but does not conform to our style guidelines. --- cfg.mk | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/cfg.mk b/cfg.mk index a9bba38..8462051 100644 --- a/cfg.mk +++ b/cfg.mk @@ -919,6 +919,12 @@ sc_require_space_before_label: halt="Top-level labels should be indented by one space" \ $(_sc_search_regexp) +sc_prohibit_space_in_label: + @prohibit='^[_a-zA-Z0-9]+ +:$$' \ + in_vc_files='\.[ch]$$' \ + halt="There should be no space between label name and colon" \ + $(_sc_search_regexp) + # Doesn't catch all cases of mismatched braces across if-else, but it helps sc_require_if_else_matching_braces: @prohibit='( else( if .*\))? {|} else( if .*\))?$$)' \ -- 2.4.3

2 2

[libvirt] [PATCH v2 0/9] Few VFIO related fixes
by Shivaprasad G Bhat 02 Nov '15

02 Nov '15

The following series fixes few VFIO related host crashes during hotplug/unplug or bind/unbind. --- Shivaprasad G Bhat (9): Implement virPCIIsAKnownStub function Initialize the stubDriver of pci devices if bound to a valid one Add iommu group number info to virPCIDevice Refuse to reattach from vfio if the iommu group is in use by any domain Wait for vfio-pci device cleanups before reassinging the device to host driver Split reprobe action from the virPCIUnbindFromStub into a new function Pass activeDevs and inactiveDevs to virPCIDeviceUnbindFromStub and virPCIDeviceBindToStub Postpone reprobing till all the devices in iommu group are unbound from vfio Wait for iommmu device to go away before reprobing the host driver src/libvirt_private.syms | 1 src/util/virhostdev.c | 18 +++ src/util/virpci.c | 289 ++++++++++++++++++++++++++++++++++++++-------- src/util/virpci.h | 2 tests/virpcimock.c | 187 +++++++++++++++++++++++++++--- tests/virpcitest.c | 152 ++++++++++++++++++++---- 6 files changed, 555 insertions(+), 94 deletions(-) -- Signature

3 14

[libvirt] Happy Birthday Libvirt
by Daniel Veillard 02 Nov '15

02 Nov '15

Hello everybody, as the first commits were done on Nov 2 2005, libvirt is more or less ten years old. This has been quite a bit of a journey, but a very successful one in my opinion, the initial goal was to provide a long term stable API for virtualization in the Red Hat ecosystem, and I'm very happy to see the goal has been reached, and we went way beyond that. In the end I think libvirt helped avoid the "balkanization" which is classic in highly competitive domains where you end up with one vertical set of tools per vendor. I also think that libvirt played a key role to keep free and open source software in the main virtualization offerings and even more so for cloud management systems. So it's time to celebrate, there is obviously lot to be done still on libvirt (as the steady stream of patches show !), but I would like to thank everybody who has contributed to the project so far, with a special mention to Dan Berrange who has been leading the effort technically for most of that time. I would also say that I have been very pelased with how the project has been working in that decade, the community is technical but friendly, and it's my hope libvirt, the project as well as the code, will remain as successful for the next decade ! So Thank You everybody, for making this project and community successful ! Enjoy that very special day :-) Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH 0/8] Few vfio related fixes
by Shivaprasad G Bhat 02 Nov '15

02 Nov '15

The series fixes 3 issues which lead to host crash. The actual fixes are in p3, p4, p7 and p8. I'll fill in more details / descriptions in v2. The make check fails on virpcitest now, which I am still debugging. I anticipate some rework from review on p7 which I continue to refine. Sending now, to get any feedback that might change things drastically. --- Shivaprasad G Bhat (8): Initialize the stubDriver of pci devices if bound to a valid one Add iommu group number info to virPCIDevice Refuse to reattach from vfio if the iommu group is in use by any domain Wait for vfio-pci device cleanups before reassinging the device to host driver Split reprobe action from the virPCIUnbindFromStub into a new function Pass activeDevs and inactiveDevs to virPCIDeviceUnbindFromStub and virPCIDeviceBindToStub Postpone reprobing till all the devices in iommu group are unbound from vfio Wait for iommmu device to go away before reprobing the host driver src/util/virhostdev.c | 18 ++++ src/util/virpci.c | 206 ++++++++++++++++++++++++++++++++++++++----------- 2 files changed, 177 insertions(+), 47 deletions(-) -- Signature

4 19

[libvirt] Entering freeze for libvirt-1.2.21
by Daniel Veillard 02 Nov '15

02 Nov '15

As pointed our on Tuesday it's time for a new release. I have tagged the release candidate 1 in git and pushed signed tarball and rpms to the usual place at: ftp://libvirt.org/libvirt/ Based on my limited testing this works just fine, but that's very limited and doesn't test portability at all, so please give it a try ! I will likely push rc2 on the week-end and the final version on Tuesday or Wednesday, please raise issues if you fine any, thanks, Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

2 5

[libvirt] Outreachy 2015 DecemberMarch- Enhancing libvirt-designer
by Riyafa Abdul Hameed 01 Nov '15

01 Nov '15

Dear sir, I am Computer Science and Engineering undergraduate at University of Moratuwa, Sri Lanka. I am a keen learner and am much interested in open source. I have worked with WSO2 <http://wso2.com/> on open source projects and am quite familiar with FOSS development. I have a good knowledge in C and in using XML. I would like to participate in "Outreachy 2015 DecemberMarch" with QEMU, but it was already quite late when I came to know of this program as its application deadline closes by tomorrow. I would like to apply for the project "Enhancing libvirt-designer" this time for Outreachy. I sent a mail to the mentor, Michal Privoznik and I couldn't get in touch with him. Because it is already too late can anyone please suggest what I can do before I apply for the program by tomorrow? Thank you. Yours sincerely, Riyafa -- Riyafa Abdul Hameed Undergraduate, University of Moratuwa Email: riyafa3(a)gmail.com Website: https://riyafa.wordpress.com/ <http://riyafa.wordpress.com/> <http://facebook.com/riyafa.ahf> <http://lk.linkedin.com/in/riyafa> <http://twitter.com/Riyafa1>

1 0

[libvirt] [PATCH 1/3] cpu_map.xml: add cmt feature to x86
by Qiaowei Ren 31 Oct '15

31 Oct '15

Some Intel processor families (e.g. the Intel Xeon processor E5 v3 family) introduced CMT (Cache Monitoring Technology) to measure the usage of cache by applications running on the platform. This patch add it into x86 part of cpu_map.xml. Signed-off-by: Qiaowei Ren <qiaowei.ren(a)intel.com> --- .gnulib | 2 +- src/cpu/cpu_map.xml | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index f39477d..106a386 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit f39477dba778e99392948dd3dd19ec0d46aee932 +Subproject commit 106a3866d01f9dd57ab4f10dbeb0d5a8db73a9f7 diff --git a/src/cpu/cpu_map.xml b/src/cpu/cpu_map.xml index b9e95cf..14ccbd8 100644 --- a/src/cpu/cpu_map.xml +++ b/src/cpu/cpu_map.xml @@ -317,6 +317,9 @@ <feature name='rtm'> <cpuid function='0x00000007' ebx='0x00000800'/> </feature> + <feature name='cmt'> + <cpuid function='0x00000007' ebx='0x00001000'/> + </feature> <feature name='rdseed'> <cpuid function='0x00000007' ebx='0x00040000'/> </feature> -- 1.9.1

5 10

[libvirt] [python PATCH] fix crash introduced by commit 1d39dbaf
by Pavel Hrdina 31 Oct '15

31 Oct '15

Some of the libvirt_*Wrap functions steals the reference and we need to set the item in array to NULL no not free it on success. Those three places was accidentally removed by commit 1d39dbaf. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1270977 Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- Pushed as fix for critical crasher. libvirt-override.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/libvirt-override.c b/libvirt-override.c index c6c1da8..5f1c98f 100644 --- a/libvirt-override.c +++ b/libvirt-override.c @@ -2448,6 +2448,7 @@ libvirt_virDomainListAllSnapshots(PyObject *self ATTRIBUTE_UNUSED, for (i = 0; i < c_retval; i++) { VIR_PY_LIST_SET_GOTO(py_retval, i, libvirt_virDomainSnapshotPtrWrap(snaps[i]), error); + snaps[i] = NULL; } cleanup: @@ -2544,9 +2545,11 @@ libvirt_virDomainSnapshotListAllChildren(PyObject *self ATTRIBUTE_UNUSED, if (!(py_retval = PyList_New(c_retval))) goto cleanup; - for (i = 0; i < c_retval; i++) + for (i = 0; i < c_retval; i++) { VIR_PY_LIST_SET_GOTO(py_retval, i, libvirt_virDomainSnapshotPtrWrap(snaps[i]), error); + snaps[i] = NULL; + } cleanup: for (i = 0; i < c_retval; i++) @@ -3117,9 +3120,11 @@ libvirt_virConnectListAllNetworks(PyObject *self ATTRIBUTE_UNUSED, if (!(py_retval = PyList_New(c_retval))) goto cleanup; - for (i = 0; i < c_retval; i++) + for (i = 0; i < c_retval; i++) { VIR_PY_LIST_SET_GOTO(py_retval, i, libvirt_virNetworkPtrWrap(nets[i]), error); + nets[i] = NULL; + } cleanup: for (i = 0; i < c_retval; i++) -- 2.6.2

1 0

[libvirt] RFC: Improve performance of macvtap device creation
by Tony Krowiak 30 Oct '15

30 Oct '15

For a guest domain defined with a large number of macvtap devices, it takes an exceedingly long time to boot the guest. In a test of a guest domain configured with 82 macvtap devices, it took over two minutes for the guest to boot. An strace of the ioctl calls during guest start up showed the SIOCGIFFLAGS ioctl literally being invoked 3,403 times. I was able to isolate the source of the ioctl calls to the*virNetDevMacVLanCreateWithVPortProfile* function in*virnetdevmacvlan.c*. The macvtap interface name is created by looping over a counter variable, starting with zero, and appending the counter value to 'macvtap'. With each iteration, a call is made to*virNetDevExists* (SIOCGIFFLAGS ioctl) to determine if a device with that name already exists, until a unique name is created. In the test case cited above, to create an interface name for the 82nd macvtap device, the*virNetDevExists* function will be called for interface names 'macvtap0' to 'macvtap80' before it is determined that 'mavtap81' can be used. So if N is the number of macvtap interfaces defined for a guest, the SIOCGIFFLAGS ioctl will be invoked (N x N + N)/2 times to find an unused macvtap device names. That's assuming only one guest is being started, who knows how many times the ioctl may have to be called in an installation running a large number of guests defined with macvtap devices. I was able to reduce the amount of time for starting a guest domain defined with 82 macvtap devices from over 2 minutes to about 14 seconds by keeping track of the interface name suffixes previously used. I defined two static bit maps (virBitmap), one each for macvtap and macvlan device name suffixes. When a macvtap/macvlan device is created, the index of the next clear bit (virBitmapNextClearBit) is retrieved to create the name. If an interface with that name does not exist, the device is created and the bit at the index used to create the interface name is set (virBitmapSetBit). When a macvtap/macvlan device is deleted, if the interface name has the pattern 'macvtap%d' or 'macvlan%d', the suffix is parsed into a bit index and used to clear the (virBitMapClearBit) bit in the respective bitmap. I am not sure that is the best design because there is no way to track interface names used to create macvtap devices outside of libvirt, for example using the ip command. There may also be other issues I've not contemplated. I included a couple of additional ideas below and am looking for comments or other suggestions that I have not considered. * Define a global counter variable initialized to 0, that gets incremented each time an interface name is created, to keep track of the last used interface name suffix. At some maximum value, the counter will be set back to 0. * Append a random number to 'macvlan' or 'macvtap' when creating the interface name. Of course, the number of digits would have to be limited so the interface name would not exceed the maximum allowed. * Create the interface name in code that has more knowledge of the environment and pass the name into the *virNetDevMacVLanCreateWithVPortProfile* function via the *tgifname* parameter. For example, the *qemuBuildCommandLine* function in *qemu_command.c* contains the loop that iterates over the network devices defined for the guest domain that ultimately get created via the *virNetDevMacVLanCreateWithVPortProfile* function. That function has access to the network device configuration and at the very least could ensure none of the names previously defined for the guest aren't used. I believe it would be matter of creating a macvtap interface name - e.g., maybe a call to some function in *virnetdevmacvlan.c* - and setting the name in the virDomainNetDef structure prior to invoking *qemuBuildInterfaceCommandLine*? There are shortcomings in all of these ideas, so if you have a better one, feel free to present it.

4 7

[libvirt] RAM backend and guest ABI (was Re: [Qemu-devel] [PATCH v2] pc: memhp: enforce minimal 128Mb) alignment for pc-dimm
by Eduardo Habkost 30 Oct '15

30 Oct '15

(CCing Michal and libvir-list, so libvirt team is aware of this restriction) On Thu, Oct 29, 2015 at 02:36:37PM +0100, Igor Mammedov wrote: > On Tue, 27 Oct 2015 14:36:35 -0200 > Eduardo Habkost <ehabkost(a)redhat.com> wrote: > > > On Tue, Oct 27, 2015 at 10:14:56AM +0100, Igor Mammedov wrote: > > > On Tue, 27 Oct 2015 10:53:08 +0200 > > > "Michael S. Tsirkin" <mst(a)redhat.com> wrote: > > > > > > > On Tue, Oct 27, 2015 at 09:48:37AM +0100, Igor Mammedov wrote: > > > > > On Tue, 27 Oct 2015 10:31:21 +0200 > > > > > "Michael S. Tsirkin" <mst(a)redhat.com> wrote: > > > > > > > > > > > On Mon, Oct 26, 2015 at 02:24:32PM +0100, Igor Mammedov wrote: > > > > > > > Yep it's workaround but it works around QEMU's broken virtio > > > > > > > implementation in a simple way without need for guest side changes. > > > > > > > > > > > > > > Without foreseeable virtio fix it makes memory hotplug unusable and even > > > > > > > more so if there were a virtio fix it won't fix old guests since you've > > > > > > > said that virtio fix would require changes of both QEMU and guest sides. > > > > > > > > > > > > What makes it not foreseeable? > > > > > > Apparently only the fact that we have a work-around in place so no one > > > > > > works on it. I can code it up pretty quickly, but I'm flat out of time > > > > > > for testing as I'm going on vacation soon, and hard freeze is pretty > > > > > > close. > > > > > I can lend a hand for testing part. > > > > > > > > > > > > > > > > > GPA space is kind of cheap, but wasting it in chunks of 512M > > > > > > seems way too aggressive. > > > > > hotplug region is sized with 1Gb alignment reserve per DIMM so we aren't > > > > > actually wasting anything here. > > > > > > > > > > > > > If I allocate two 1G DIMMs, what will be the gap size? 512M? 1G? > > > > It's too much either way. > > > minimum would be 512, and if backend is 1Gb-hugepage gap will be > > > backend's natural alignment (i.e. 1Gb). > > > > Is backend configuration even allowed to affect the machine ABI? We need > > to be able to change backend configuration when migrating the VM to > > another host. > for now, one has to use the same type of backend on both sides > i.e. if source uses 1Gb huge pages backend then target also > need to use it. > The page size of the backend don't even depend on QEMU arguments, but on the kernel command-line or hugetlbfs mount options. So it's possible to have exactly the same QEMU command-line on source and destination (with an explicit versioned machine-type), and get a VM that can't be migrated? That means we are breaking our guarantees about migration and guest ABI. > We could change this for the next machine type to always force > max alignment (1Gb), then it would be possible to change > between backends with different alignments. I'm not sure what's the best solution here. If always using 1GB is too aggressive, we could require management to ask for an explicit alignment as a -machine option if they know they will need a specific backend page size. BTW, are you talking about the behavior introduced by aa8580cddf011e8cedcf87f7a0fdea7549fc4704 ("pc: memhp: force gaps between DIMM's GPA") only, or the backend page size was already affecting GPA allocation before that commit? -- Eduardo

2 1

[libvirt] [PATCH v2 0/3] improve virsh attach-interface
by Pavel Hrdina 30 Oct '15

30 Oct '15

new in v2: - removed netdev name as an option for --source parameter - removed --driver parameter, the default is vfio for new qemu and that's good enough Pavel Hrdina (3): virsh-nodedev: makes struct and functions for NodeDevice list available virsh-domain: update attach-interface to support type=hostdev virsh.pod: update and improve a attach-interface section tools/virsh-domain.c | 34 +++++++++++++++++++-- tools/virsh-nodedev.c | 16 ++++------ tools/virsh-nodedev.h | 11 +++++++ tools/virsh.pod | 82 +++++++++++++++++++++++++++++++-------------------- 4 files changed, 98 insertions(+), 45 deletions(-) -- 2.6.2

2 13