- Devel - libvirt List Archives

[libvirt] [PATCH RFC 00/16] Add support for LUKS encrypted devices
by John Ferlan 08 Jun '16

08 Jun '16

Patches 1-3 were posted separately: http://www.redhat.com/archives/libvir-list/2016-June/msg00256.html But perhaps seeing the final direction will make things more clear as to why a "real" flag system wasn't used and keeping the current paradigm of constant value returns still works just fine. Patches 4-5 were posted separately: http://www.redhat.com/archives/libvir-list/2016-June/msg00091.html (4) http://www.redhat.com/archives/libvir-list/2016-June/msg00094.html (5) Although at one point patch 4 had an ACK: http://www.redhat.com/archives/libvir-list/2016-May/msg02115.html It wasn't clear if the more recent review rescinded that, so it still remains "in the list". I understand the concern about adding secret to cfg.mk checking, but without a better idea of how to handle - I left things as they were. Patches 6-16 are all new. Some parts are separable, but rather than continue piecemeal I just figured going with an RFC will at least Patch 6 is only there to "prove" that using the current encryption paradigm XML still works, although if I've read the tea leaves correctly, the qemu support isn't working as desired/expected. Patch 7 adds "usage" as an XML attribute for encryption and the associated tests with that. I've chosen to "reuse" the <encryption> XML element rather than inventing something new. I'm not opposed to something new, but let's decide up a name quickly... Patch 8-9 adds the ability for the storage backend to create/recognize a luks volume Patches 10-13 adds support for luks encryption in the storage backend. The new "<secret>" format uses "luks" as the usage type and "<key>" as the 'name'. If those names cause angst, I'm fine with changing, but just give a better suggestion! Adding <cipher> and <ivgen> were a result of using qemu constructs from qemu commit id '3e308f20'. Since we are parsing something new, I figure failing in the domain parse code for this new type was acceptible as opposed to some post processing check. Patches 14-16 adds support for luks encryption to the domain using <encryption type='luks'... <secret format='key' usage/uuid='xxx'>> I've tested using a "good" and "bad" password and got the expected results for starting a domain. I did not add 'virsh vol-create-as' support just yet. I figured that would be less to go back and redo if the names of elements changes. I've also run the changes through Coverity with no new issues detected. The whole series is a result of the following bz: https://bugzilla.redhat.com/show_bug.cgi?id=1301021 John Ferlan (16): storage: Adjust qemu-img switches check storage: Create helper to set backing for CreateQemuImg code storage: Create helper to set options for CreateQemuImg code storage: Use virSecretGetSecretString secret: Move virStorageSecretType to secret_util and rename tests: Adjust tests for encrypted storage util: Add 'usage' for encryption util: Modify the FileTypeInfo for meta data checks util: Add 'luks' to the FileTypeInfo conf: Add new secret type "luks" encryption: Add luks parsing for storageencryption encryption: Add <cipher> and <ivgen> to encryption storage: Add support to create a luks volume qemu: Change protocol parameter for secret setup qemu: Remove authdef from secret setup qemu: Add luks support for domain disk cfg.mk | 2 +- docs/aclpolkit.html.in | 4 + docs/formatsecret.html.in | 60 ++- docs/formatstorageencryption.html.in | 115 ++++- docs/schemas/secret.rng | 10 + docs/schemas/storagecommon.rng | 58 ++- include/libvirt/libvirt-secret.h | 3 +- src/Makefile.am | 1 + src/access/viraccessdriverpolkit.c | 13 + src/conf/domain_conf.c | 11 + src/conf/secret_conf.c | 26 +- src/conf/secret_conf.h | 3 +- src/conf/virsecretobj.c | 5 + src/libvirt_private.syms | 1 + src/libxl/libxl_conf.c | 2 +- src/qemu/qemu_command.c | 8 +- src/qemu/qemu_domain.c | 154 ++++--- src/qemu/qemu_process.c | 18 +- src/secret/secret_util.c | 18 +- src/secret/secret_util.h | 22 +- src/storage/storage_backend.c | 480 +++++++++++++++------ src/storage/storage_backend.h | 3 +- src/storage/storage_backend_fs.c | 10 +- src/storage/storage_backend_gluster.c | 2 + src/storage/storage_backend_iscsi.c | 55 +-- src/storage/storage_backend_rbd.c | 49 +-- src/util/virendian.h | 24 ++ src/util/virqemu.c | 23 + src/util/virqemu.h | 6 + src/util/virstorageencryption.c | 166 ++++++- src/util/virstorageencryption.h | 18 +- src/util/virstoragefile.c | 125 ++++-- src/util/virstoragefile.h | 18 +- tests/qemuargv2xmltest.c | 4 +- .../qemuxml2argv-encrypted-disk-usage.args | 24 ++ .../qemuxml2argv-encrypted-disk-usage.xml | 32 ++ .../qemuxml2argv-encrypted-disk.args | 26 +- .../qemuxml2argv-encrypted-disk.xml | 4 +- .../qemuxml2argv-luks-disk-cipher.args | 36 ++ .../qemuxml2argv-luks-disk-cipher.xml | 41 ++ .../qemuxml2argvdata/qemuxml2argv-luks-disks.args | 36 ++ tests/qemuxml2argvdata/qemuxml2argv-luks-disks.xml | 41 ++ tests/qemuxml2argvtest.c | 14 +- .../qemuxml2xmlout-encrypted-disk-usage.xml | 36 ++ .../qemuxml2xmlout-encrypted-disk.xml | 4 +- .../qemuxml2xmlout-luks-disk-cipher.xml | 45 ++ .../qemuxml2xmlout-luks-disks.xml | 45 ++ tests/qemuxml2xmltest.c | 3 + tests/secretxml2xmlin/usage-luks.xml | 7 + tests/secretxml2xmltest.c | 1 + tests/storagevolxml2argvdata/qcow2-flag.argv | 2 - .../qcow2-nobacking-convert-flag.argv | 2 - .../qcow2-nobacking-convert-none.argv | 2 - .../qcow2-nobacking-flag.argv | 1 - .../qcow2-nobacking-none.argv | 1 - tests/storagevolxml2argvdata/qcow2-none.argv | 1 - tests/storagevolxml2argvtest.c | 25 +- tests/storagevolxml2xmlin/vol-luks-cipher.xml | 23 + tests/storagevolxml2xmlin/vol-luks.xml | 21 + tests/storagevolxml2xmlout/vol-luks-cipher.xml | 23 + tests/storagevolxml2xmlout/vol-luks.xml | 21 + tests/storagevolxml2xmltest.c | 2 + tests/virendiantest.c | 18 + 63 files changed, 1619 insertions(+), 435 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-encrypted-disk-usage.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-encrypted-disk-usage.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-luks-disk-cipher.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-luks-disk-cipher.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-luks-disks.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-luks-disks.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-encrypted-disk-usage.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-luks-disk-cipher.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-luks-disks.xml create mode 100644 tests/secretxml2xmlin/usage-luks.xml delete mode 100644 tests/storagevolxml2argvdata/qcow2-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-convert-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-convert-none.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-none.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-none.argv create mode 100644 tests/storagevolxml2xmlin/vol-luks-cipher.xml create mode 100644 tests/storagevolxml2xmlin/vol-luks.xml create mode 100644 tests/storagevolxml2xmlout/vol-luks-cipher.xml create mode 100644 tests/storagevolxml2xmlout/vol-luks.xml -- 2.5.5

2 19

[libvirt] [PATCH 0/3] qemu: fix startup policy checking
by Peter Krempa 08 Jun '16

08 Jun '16

One of the recent refactors broke disk startup policy checking. Fix it with a few cleanups. Peter Krempa (3): qemu: domain: Sanitize return value handling in disk presence checker qemu: process: Unexport qemuProcessStartValidate qemu: process: Call disk startup policy check after cloning domain def src/qemu/qemu_domain.c | 16 +++++----------- src/qemu/qemu_process.c | 14 +++++++++----- src/qemu/qemu_process.h | 7 ------- 3 files changed, 14 insertions(+), 23 deletions(-) -- 2.8.3

2 7

[libvirt] [PATCH] qemu: driver: Unset log file watcher after restoring a VM save file
by Peter Krempa 08 Jun '16

08 Jun '16

qemuProcessStart does not unset the infrastructure that retrieves errors from the qemu log file in case of migration. As this wasn't handled properly in qemuDomainSaveImageStartVM we kept the logging context/fd open for the lifetime of the VM rather than closing it after it's not needed. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1325080 --- src/qemu/qemu_driver.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 10d3e3d..4f09630 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6509,6 +6509,11 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, if (!restored) goto cleanup; + /* qemuProcessStart doesn't unsed the qemu error reporting infrastructure + * in case of migration (which is used in this case) so we need to reset it + * so that the handle to virtlogd is not held open unnecessarily */ + qemuMonitorSetDomainLog(qemuDomainGetMonitor(vm), NULL, NULL, NULL); + event = virDomainEventLifecycleNewFromObj(vm, VIR_DOMAIN_EVENT_STARTED, VIR_DOMAIN_EVENT_STARTED_RESTORED); -- 2.8.3

2 2

[libvirt] [PATCH] xen: Also add sub-type for driver=tap2 in xen-xm
by Philipp Hahn 07 Jun '16

07 Jun '16

tap2 only handles 'aio', but not 'raw', which must be explicitly given: | $ virsh domxml-to-native yyy.xml > yyy.xm | $ xm new yyy.xm | Error: tap:/srv/xen/xxx.img not a valid disk type | $ sed -i -e 's/tap2:/&aio:/' yyy.xm | $ xm new yyy.xm Fix reading and writing "xen-xm" format for "tap2" by handling it the same as "tap". --- src/xenconfig/xen_xm.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/xenconfig/xen_xm.c b/src/xenconfig/xen_xm.c index e09d97e..6556886 100644 --- a/src/xenconfig/xen_xm.c +++ b/src/xenconfig/xen_xm.c @@ -196,7 +196,8 @@ xenParseXMDisk(virConfPtr conf, virDomainDefPtr def) } /* And the sub-type for tap:XXX: type */ - if (STREQ_NULLABLE(virDomainDiskGetDriver(disk), "tap")) { + if (STREQ_NULLABLE(virDomainDiskGetDriver(disk), "tap") || + STREQ_NULLABLE(virDomainDiskGetDriver(disk), "tap2")) { char *driverType; if (!(tmp = strchr(src, ':'))) @@ -298,7 +299,7 @@ xenFormatXMDisk(virConfValuePtr list, else type = virStorageFileFormatTypeToString(format); virBufferAsprintf(&buf, "%s:", driver); - if (STREQ(driver, "tap")) + if (STREQ(driver, "tap") || STREQ(driver, "tap2")) virBufferAsprintf(&buf, "%s:", type); } else { switch (virDomainDiskGetType(disk)) { -- 2.1.4

2 6

Re: [libvirt] Channel paths
by Sascha Silbe 07 Jun '16

07 Jun '16

Dear Daniel, [CC += libvir-list, didn't notice before this thread is on qemu-devel only] "Daniel P. Berrange" <berrange(a)redhat.com> writes: > On Thu, Jun 02, 2016 at 09:27:45PM +0200, Sascha Silbe wrote: >> Since 71408079 [qemu: Don't bother user with libvirt-internal paths], >> the path chosen by libvirt isn't included in the domain XML anymore. So >> now I need to choose the path inside the application. The only safe way >> to do that is by using something in an application-managed namespace >> (e.g. /var/lib/myapp/foo or /tmp/myapp/foo); I certainly wouldn't want >> to second guess what paths would be safe inside the libvirt namespace >> (/var/lib/libvirt/qemu). Except now I hear that anything outside >> /var/lib/libvirt/qemu is not guaranteed to work due to e.g. the SELinux >> policy configured by libvirt... > > IIUC that change 71408079 ought to only apply to the persistent XML > configuration on disk. When the guest is running the live XML ought > to still report the path libvirt chose, so you should still be able > to see it while running. If that isn't the case then please raise a > bug, because we certainly expect apps to be able to discover the > path libvirt picked for exactly the reason you describe OK, this makes a lot more sense now (persistent vs. live). Just tried it with current libvirt master and it works, thanks. It even worked with 71408079 itself. I'm pretty sure things were broken some time in between, but a few quick probes didn't turn up a faulty version and I'm not curious enough to do a linear search. Cannot completely rule out other effects having played a role, either. Sascha -- Softwareentwicklung Sascha Silbe, Niederhofenstraße 5/1, 71229 Leonberg https://se-silbe.de/ USt-IdNr. DE281696641

1 0

[libvirt] [PATCH v3 0/4] Replace VIR_ERROR with standard vir*Error in state driver init
by Jovanka Gulicoska 07 Jun '16

07 Jun '16

Replace VIR_ERROR logging macros for error reporting in driver startup routines with vir*Error function. Don't use virStrerror and remove ebuf. Link to task: http://wiki.libvirt.org/page/BiteSizedTasks#Replace_VIR_ERROR_with_standard… Jovanka Gulicoska (4): uml: Replace VIR_ERROR with standard vir*Error in state driver init qemu: Replace VIR_ERROR with standard vir*Error in state driver init xen: Replace VIR_ERROR with standard vir*Error in state driver init node_device: Replace VIR_ERROR with standard vir*Error in state driver init src/node_device/node_device_hal.c | 26 +++++++++++------- src/node_device/node_device_udev.c | 49 ++++++++++++++++++++-------------- src/qemu/qemu_driver.c | 54 +++++++++++++++++++++++--------------- src/uml/uml_driver.c | 20 ++++++-------- src/xen/xen_driver.c | 5 ++-- 5 files changed, 90 insertions(+), 64 deletions(-) -- 2.5.5

2 5

[libvirt] [PATCH] Updated to deal with specifying user IDs to that do not map to usernames
by Roy Keene 07 Jun '16

07 Jun '16

Patch to libvirt master to avoid failing when a user ID is specified, e.g. for <seclabel type='dac'>, that does not map to a user name. This is useful if you want to run each VM as a separate user and not bother creating an /etc/passwd entry for each UID. It compiles but is as yet untested. --- src/util/virutil.c | 69 +++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 50 insertions(+), 19 deletions(-) diff --git a/src/util/virutil.c b/src/util/virutil.c index d80d994..ae95237 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -790,26 +790,57 @@ virGetUserEnt(uid_t uid, char **name, gid_t *group, char **dir) if (VIR_RESIZE_N(strbuf, strbuflen, strbuflen, strbuflen) < 0) goto cleanup; } - if (rc != 0) { - virReportSystemError(rc, - _("Failed to find user record for uid '%u'"), - (unsigned int) uid); - goto cleanup; - } else if (pw == NULL) { - virReportError(VIR_ERR_SYSTEM_ERROR, - _("Failed to find user record for uid '%u'"), - (unsigned int) uid); - goto cleanup; - } - if (name && VIR_STRDUP(*name, pw->pw_name) < 0) - goto cleanup; - if (group) - *group = pw->pw_gid; - if (dir && VIR_STRDUP(*dir, pw->pw_dir) < 0) { - if (name) - VIR_FREE(*name); - goto cleanup; + if (rc != 0 || pw == NULL) { + /* + * If the user does not exist or its data is not present, return + * a created username. + */ + VIR_FREE(strbuf); + + strbuflen = 128; + + if (VIR_ALLOC_N(strbuf, strbuflen) < 0) { + return(-1); + } + + /* + * Fake user home directory: / + */ + if (dir) { + if (VIR_STRDUP(*dir, "/") < 0) { + goto cleanup; + } + } + + /* + * Fake user GID: Same as UID + */ + if (group) { + *group = (gid_t) uid; + } + + /* + * Fake user name: Same as UID (in string) + */ + snprintf(strbuf, strbuflen, "%llu", (unsigned long long) uid); + + if (name && VIR_STRDUP(*name, strbuf) < 0) { + if (dir) { + VIR_FREE(*dir); + } + goto cleanup; + } + } else { + if (name && VIR_STRDUP(*name, pw->pw_name) < 0) + goto cleanup; + if (group) + *group = pw->pw_gid; + if (dir && VIR_STRDUP(*dir, pw->pw_dir) < 0) { + if (name) + VIR_FREE(*name); + goto cleanup; + } } ret = 0; -- 2.7.4

4 8

[libvirt] [PATCH 0/5] qemu: Improve VM file logging using virtlogd
by Peter Krempa 07 Jun '16

07 Jun '16

Patch 4/5 fixes a bug that was introduced with virtlogd. Rest of the patches add infrastructure for logging singular events to the vm log file. Peter Krempa (5): log: handler: Add new API to append to logging files log: daemon: Add remote protocol handling for the log appending API qemu: domain: Implement helper for one-shot log entries to the VM log file qemu: process: Append the "shutting down" message using the new APIs qemu: migration: Add VM log entry on start of migration src/libvirt_private.syms | 1 + src/logging/log_daemon_dispatch.c | 26 ++++++++++++++++ src/logging/log_handler.c | 50 +++++++++++++++++++++++++++++++ src/logging/log_handler.h | 8 +++++ src/logging/log_manager.c | 34 +++++++++++++++++++++ src/logging/log_manager.h | 8 +++++ src/logging/log_protocol.x | 20 ++++++++++++- src/qemu/qemu_domain.c | 62 +++++++++++++++++++++++++++++++++++++++ src/qemu/qemu_domain.h | 5 ++++ src/qemu/qemu_migration.c | 7 +++++ src/qemu/qemu_process.c | 11 ++----- 11 files changed, 223 insertions(+), 9 deletions(-) -- 2.8.3

2 10

[libvirt] [PATCH 0/3] Speed up schema testing
by Ján Tomko 07 Jun '16

07 Jun '16

Currently, we spawn a new xmllint process for every tested XML file. Fortunately, xmllint accepts multiple files on its command line. Use xargs on XML files we expect to be valid and only fall back to file-by-file check if that fails (or if VIR_TEST_EXPENSIVE was requested). This speeds up successfully passing domainschematest 32x (from ~13s to ~0.5s) and make check over 2x (over 16s to ~8s), making it even more fun to use with git rebase -x. Ján Tomko (3): schematestutils: split out file-by-file schema checking schematestutils: split out check_one_file schematestutils: Add check_schema_quick tests/schematestutils.sh | 79 ++++++++++++++++++++++++++++++++++++------------ 1 file changed, 59 insertions(+), 20 deletions(-) -- 2.7.3

2 4

[libvirt] [PATCH] util: Perform proper virRandomBytes return value checking
by John Ferlan 07 Jun '16

07 Jun '16

Document the return value of virRandomBytes as 0 or some errno value and then make sure all callers make the proper checks. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/util/vircrypto.c | 2 +- src/util/virrandom.c | 2 ++ tests/vircryptotest.c | 6 ++++-- tests/virrandomtest.c | 2 +- 4 files changed, 8 insertions(+), 4 deletions(-) diff --git a/src/util/vircrypto.c b/src/util/vircrypto.c index 4f288f0..8452936 100644 --- a/src/util/vircrypto.c +++ b/src/util/vircrypto.c @@ -301,7 +301,7 @@ virCryptoGenerateRandom(size_t nbytes) /* If we don't have gnutls_rnd(), we will generate a less cryptographically * strong master buf from /dev/urandom. */ - if ((ret = virRandomBytes(buf, nbytes)) < 0) { + if ((ret = virRandomBytes(buf, nbytes))) { virReportSystemError(ret, "%s", _("failed to generate byte stream")); VIR_FREE(buf); return NULL; diff --git a/src/util/virrandom.c b/src/util/virrandom.c index 62a0e31..41daa40 100644 --- a/src/util/virrandom.c +++ b/src/util/virrandom.c @@ -167,6 +167,8 @@ uint32_t virRandomInt(uint32_t max) * * Generate a stream of random bytes from /dev/urandom * into @buf of size @buflen + * + * Returns 0 on success or an errno on failure */ int virRandomBytes(unsigned char *buf, diff --git a/tests/vircryptotest.c b/tests/vircryptotest.c index 72265d9..e7b7de5 100644 --- a/tests/vircryptotest.c +++ b/tests/vircryptotest.c @@ -87,9 +87,11 @@ testCryptoEncrypt(const void *opaque) VIR_ALLOC_N(iv, ivlen) < 0) goto cleanup; - if (virRandomBytes(enckey, enckeylen) < 0 || - virRandomBytes(iv, ivlen) < 0) + if (virRandomBytes(enckey, enckeylen) || + virRandomBytes(iv, ivlen)) { + fprintf(stderr, "Failed to generate random bytes\n"); goto cleanup; + } if (virCryptoEncryptData(data->algorithm, enckey, enckeylen, iv, ivlen, data->input, data->inputlen, diff --git a/tests/virrandomtest.c b/tests/virrandomtest.c index 367bdc7..687ebd9 100644 --- a/tests/virrandomtest.c +++ b/tests/virrandomtest.c @@ -40,7 +40,7 @@ testRandomBytes(const void *unused ATTRIBUTE_UNUSED) if (VIR_ALLOC_N(data, datalen) < 0) return -1; - if (virRandomBytes(data, datalen) < 0) { + if (virRandomBytes(data, datalen)) { fprintf(stderr, "Failed to generate random bytes"); goto cleanup; } -- 2.5.5

2 1

[libvirt] [PATCH] Reindent virNetDevSendEthtoolIoctl
by Ján Tomko 07 Jun '16

07 Jun '16

--- src/util/virnetdev.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) Pushed as trivial. diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index e6c57ac..af122f5 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -3213,15 +3213,15 @@ virNetDevSendEthtoolIoctl(int fd, struct ifreq *ifr) ret = ioctl(fd, SIOCETHTOOL, ifr); if (ret != 0) { switch (errno) { - case EINVAL: /* kernel doesn't support SIOCETHTOOL */ - VIR_DEBUG("ethtool ioctl: invalid request"); - break; - case EOPNOTSUPP: /* kernel doesn't support specific feature */ - VIR_DEBUG("ethtool ioctl: request not supported"); - break; - default: - virReportSystemError(errno, "%s", _("ethtool ioctl error")); - break; + case EINVAL: /* kernel doesn't support SIOCETHTOOL */ + VIR_DEBUG("ethtool ioctl: invalid request"); + break; + case EOPNOTSUPP: /* kernel doesn't support specific feature */ + VIR_DEBUG("ethtool ioctl: request not supported"); + break; + default: + virReportSystemError(errno, "%s", _("ethtool ioctl error")); + break; } } -- 2.7.3

1 0

[libvirt] [PATCH] Reindent comment of virNetDevFeatureAvailable
by Ján Tomko 07 Jun '16

07 Jun '16

--- src/util/virnetdev.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) Pushed as trivial. diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index 7db4497..5ad61fe 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -3241,14 +3241,14 @@ virNetDevSendEthtoolIoctl(const char *ifname, void *cmd) /** -* virNetDevFeatureAvailable -* This function checks for the availability of a network device feature -* -* @ifname: name of the interface -* @cmd: reference to an ethtool command structure -* -* Returns 0 if not found, 1 on success, and -1 on failure. -*/ + * virNetDevFeatureAvailable + * This function checks for the availability of a network device feature + * + * @ifname: name of the interface + * @cmd: reference to an ethtool command structure + * + * Returns 0 if not found, 1 on success, and -1 on failure. + */ static int virNetDevFeatureAvailable(const char *ifname, struct ethtool_value *cmd) { -- 2.7.3

1 0

[libvirt] [PATCH] spec: Advertise nvram paths of official fedora edk2 builds
by Cole Robinson 07 Jun '16

07 Jun '16

Fedora now ships edk2 firmware in its official repos, so adapt the nvram path list to match. Eventually we can remove the nightly links as well once some integration kinks have been worked out, and documentation updated. Move the macro building into the %build target, which lets us build up a shell variable and make things a bit more readable https://bugzilla.redhat.com/show_bug.cgi?id=1335395 --- libvirt.spec.in | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index c7fcf85..8b88eef 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -192,12 +192,6 @@ %define qemu_group qemu -# Advertise OVMF and AAVMF from nightly firmware repo -%if 0%{?fedora} - %define arg_loader_nvram --with-loader-nvram="/usr/share/edk2.git/ovmf-x64/OVMF_CODE-pure-efi.fd:/usr/share/edk2.git/ovmf-x64/OVMF_VARS-pure-efi.fd:/usr/share/edk2.git/aarch64/QEMU_EFI-pflash.raw:/usr/share/edk2.git/aarch64/vars-template-pflash.raw" -%endif - - %if 0%{?fedora} || 0%{?rhel} >= 7 %define with_systemd_macros 1 %else @@ -1097,6 +1091,18 @@ rm -rf .git %define arg_selinux_mount --with-selinux-mount="/selinux" %endif +%if 0%{?fedora} + # Nightly firmware repo x86/OVMF + LOADERS="/usr/share/edk2.git/ovmf-x64/OVMF_CODE-pure-efi.fd:/usr/share/edk2.git/ovmf-x64/OVMF_VARS-pure-efi.fd" + # Nightly firmware repo aarch64/AAVMF + LOADERS="$LOADERS:/usr/share/edk2.git/aarch64/QEMU_EFI-pflash.raw:/usr/share/edk2.git/aarch64/vars-template-pflash.raw" + # Fedora official x86/OVMF + LOADERS="$LOADERS:/usr/share/edk2/ovmf/OVMF_CODE.fd:/usr/share/edk2/ovmf/OVMF_VARS.fd" + # Fedora official aarch64/AAVMF + LOADERS="$LOADERS:/usr/share/edk2/aarch64/QEMU_EFI-pflash.raw:/usr/share/edk2/aarch64/vars-template-pflash.raw" + %define arg_loader_nvram --with-loader-nvram="$LOADERS" +%endif + # place macros above and build commands below this comment %if 0%{?enable_autotools} -- 2.7.4

2 3

[libvirt] [PATCH] qemu: process: Allow VIR_QEMU_PROCESS_START_NEW in qemuProcessLaunch
by Peter Krempa 07 Jun '16

07 Jun '16

The new flag was not added to virCheckFlags in commit '0d1c17aa' causing a regression where VMs were not able to start. --- Pushed under the trivial rule. src/qemu/qemu_process.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 2f5829c..66bc4b1 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -5062,7 +5062,8 @@ qemuProcessLaunch(virConnectPtr conn, * but doesn't hurt to check */ virCheckFlags(VIR_QEMU_PROCESS_START_COLD | VIR_QEMU_PROCESS_START_PAUSED | - VIR_QEMU_PROCESS_START_AUTODESTROY, -1); + VIR_QEMU_PROCESS_START_AUTODESTROY | + VIR_QEMU_PROCESS_START_NEW, -1); cfg = virQEMUDriverGetConfig(driver); -- 2.8.3

1 0

[libvirt] [PATCH REPOST 0/4] Couple of random and crypto adjustments
by John Ferlan 07 Jun '16

07 Jun '16

Originally posted : http://www.redhat.com/archives/libvir-list/2016-May/msg01650.html Updated to current head and reposted. John Ferlan (4): util: Add range parameters to virRandomBytes storage: Use virRandomBytes for virStorageGenerateQcowPassphrase util: Alter virCryptoEncryptData for non GNUTLS builds util: Adjust virCryptoEncryptData code to use macros src/storage/storage_backend.c | 10 +++-- src/util/vircrypto.c | 88 +++++++++++++++++++++++++---------------- src/util/virrandom.c | 13 ++++-- src/util/virrandom.h | 3 +- src/util/virstorageencryption.c | 42 +++++++------------- src/util/virstorageencryption.h | 4 +- src/util/viruuid.c | 2 +- tests/qemuxml2argvmock.c | 2 +- tests/vircryptotest.c | 4 +- tests/virrandommock.c | 10 +++-- tests/virrandomtest.c | 32 +++++++++++---- 11 files changed, 124 insertions(+), 86 deletions(-) -- 2.5.5

2 9

[libvirt] [PATCH v6 0/2] qemu: Support for QXL heads
by Martin Kletzander 07 Jun '16

07 Jun '16

v6: - Change heads when migrating (revert v2) - https://www.redhat.com/archives/libvir-list/2016-June/msg00106.html v5: - Just a rebase, plus v4 had some hunk in that didn't belong there so this one's cleaner to look at - https://www.redhat.com/archives/libvir-list/2016-April/msg00721.html v4: - Added test case without "heads=" - Don't consolidate 'size_t i, j' declarations - https://www.redhat.com/archives/libvir-list/2016-March/msg01436.html v3: - rebase on top of current master in order to cleanly apply - https://www.redhat.com/archives/libvir-list/2016-March/msg01198.html v2: - don't change heads when migrating - https://www.redhat.com/archives/libvir-list/2016-March/msg00811.html v1: - https://www.redhat.com/archives/libvir-list/2016-March/msg00410.html Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1283207 *** BLURB HERE *** Martin Kletzander (2): qemu: Check for qxl's max_outputs parameter qemu: Add support to QXL's max_outputs parameter src/qemu/qemu_capabilities.c | 5 +++ src/qemu/qemu_capabilities.h | 4 ++ src/qemu/qemu_command.c | 8 ++++ src/qemu/qemu_migration.c | 1 - tests/qemucapabilitiesdata/caps_2.4.0.x86_64.xml | 2 + tests/qemucapabilitiesdata/caps_2.5.0.x86_64.xml | 2 + tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 2 + .../qemuxml2argv-video-qxl-heads.args | 28 +++++++++++++ .../qemuxml2argv-video-qxl-heads.xml | 47 ++++++++++++++++++++++ .../qemuxml2argv-video-qxl-noheads.args | 24 +++++++++++ .../qemuxml2argv-video-qxl-noheads.xml | 39 ++++++++++++++++++ tests/qemuxml2argvtest.c | 16 ++++++++ .../qemuxml2xmlout-video-qxl-heads.xml | 47 ++++++++++++++++++++++ .../qemuxml2xmlout-video-qxl-noheads.xml | 39 ++++++++++++++++++ tests/qemuxml2xmltest.c | 3 ++ 15 files changed, 266 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-video-qxl-heads.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-video-qxl-heads.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-video-qxl-noheads.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-video-qxl-noheads.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-video-qxl-heads.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-video-qxl-noheads.xml -- 2.8.3

2 5

[libvirt] [PATCH v2 0/3] Adjust qemu-img switches and checks
by John Ferlan 07 Jun '16

07 Jun '16

v1:http://www.redhat.com/archives/libvir-list/2016-June/msg00081.html (patches 4, 6, and 7) Patch 4 is now patch 1 Patch 6 was reviewed, but this patch alters it not need backingType Patch 7 reworked to accommodate patch 1 changes John Ferlan (3): storage: Adjust qemu-img switches check storage: Create helper to set backing for CreateQemuImg code storage: Create helper to set options for CreateQemuImg code src/storage/storage_backend.c | 223 ++++++++++----------- tests/storagevolxml2argvdata/qcow2-flag.argv | 2 - .../qcow2-nobacking-convert-flag.argv | 2 - .../qcow2-nobacking-convert-none.argv | 2 - .../qcow2-nobacking-flag.argv | 1 - .../qcow2-nobacking-none.argv | 1 - tests/storagevolxml2argvdata/qcow2-none.argv | 1 - tests/storagevolxml2argvtest.c | 22 +- 8 files changed, 107 insertions(+), 147 deletions(-) delete mode 100644 tests/storagevolxml2argvdata/qcow2-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-convert-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-convert-none.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-flag.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-nobacking-none.argv delete mode 100644 tests/storagevolxml2argvdata/qcow2-none.argv -- 2.5.5

2 5

[libvirt] [PATCH v2 00/10] Add domain config validation infrastructure
by Peter Krempa 07 Jun '16

07 Jun '16

Similarly to post parse callbacks let's add infrastructure that will allow us to introduce checks that will reject configrations that were previously valid. This is achieved by flagging all the entry points to the config parser that need to ignore invadid configurations and adding a callback infrastructure for driver specific checks. Peter Krempa (10): conf: Rename VIR_DOMAIN_DEF_PARSE_VALIDATE to VIR_DOMAIN_DEF_PARSE_VALIDATE_SCHEMA conf: Add infrastructure for adding configuration validation conf: drop 'def' from struct virDomainDefPostParseDeviceIteratorData conf: Add device def validation callback qemu: process: Unexport qemuProcessStartValidate qemu: process: Convert multiple boolean args to a single flag qemu: process: Call the domain config validator when starting a new VM conf: Move disk info validator to the domain conf validator conf: Move validation of disk LUN device to the appropriate place qemu: Move check that validates 'min_guarantee' to qemuDomainDefValidate src/bhyve/bhyve_driver.c | 4 +- src/conf/domain_conf.c | 225 ++++++++++++++++++++++++++++++++++-------- src/conf/domain_conf.h | 31 +++++- src/conf/snapshot_conf.c | 3 +- src/conf/virdomainobjlist.c | 6 +- src/esx/esx_driver.c | 2 +- src/libvirt_private.syms | 2 +- src/libxl/libxl_domain.c | 3 +- src/libxl/libxl_driver.c | 10 +- src/libxl/libxl_migration.c | 6 +- src/lxc/lxc_driver.c | 7 +- src/openvz/openvz_driver.c | 7 +- src/phyp/phyp_driver.c | 2 +- src/qemu/qemu_domain.c | 20 +++- src/qemu/qemu_driver.c | 15 +-- src/qemu/qemu_migration.c | 14 ++- src/qemu/qemu_process.c | 42 ++++---- src/qemu/qemu_process.h | 9 +- src/security/virt-aa-helper.c | 3 +- src/test/test_driver.c | 4 +- src/uml/uml_driver.c | 7 +- src/vbox/vbox_common.c | 5 +- src/vmware/vmware_driver.c | 4 +- src/vz/vz_driver.c | 6 +- src/xen/xen_driver.c | 4 +- src/xen/xend_internal.c | 3 +- src/xen/xm_internal.c | 3 +- src/xenapi/xenapi_driver.c | 4 +- tests/qemuxml2argvtest.c | 6 +- 29 files changed, 325 insertions(+), 132 deletions(-) -- 2.8.3

2 21

[libvirt] [PATCH v2] systemd: directly notify systemd instead of using sd_notify
by Daniel P. Berrange 07 Jun '16

07 Jun '16

The sd_notify method is used to tell systemd when libvirtd has finished starting up. All it does is send a datagram containing the string parameter to systemd on a UNIX socket named in the NOTIFY_SOCKET environment variable. Rather than pulling in the systemd libraries for this, just code the notification directly in libvirt as this is a stable ABI from systemd's POV which explicitly allows independant implementations: See "Reimplementable Independently" column in the "$NOTIFY_SOCKET Daemon Notifications" row: https://www.freedesktop.org/wiki/Software/systemd/InterfacePortabilityAndSt… Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- configure.ac | 2 -- libvirt.spec.in | 12 ----------- m4/virt-systemd-daemon.m4 | 34 ------------------------------ src/Makefile.am | 4 ++-- src/util/virsystemd.c | 53 ++++++++++++++++++++++++++++++++++++++++++----- 5 files changed, 50 insertions(+), 55 deletions(-) delete mode 100644 m4/virt-systemd-daemon.m4 diff --git a/configure.ac b/configure.ac index 74c33b3..a63b912 100644 --- a/configure.ac +++ b/configure.ac @@ -256,7 +256,6 @@ LIBVIRT_CHECK_SANLOCK LIBVIRT_CHECK_SASL LIBVIRT_CHECK_SELINUX LIBVIRT_CHECK_SSH2 -LIBVIRT_CHECK_SYSTEMD_DAEMON LIBVIRT_CHECK_UDEV LIBVIRT_CHECK_WIRESHARK LIBVIRT_CHECK_NSS @@ -2787,7 +2786,6 @@ LIBVIRT_RESULT_SANLOCK LIBVIRT_RESULT_SASL LIBVIRT_RESULT_SELINUX LIBVIRT_RESULT_SSH2 -LIBVIRT_RESULT_SYSTEMD_DAEMON LIBVIRT_RESULT_UDEV LIBVIRT_RESULT_WIRESHARK LIBVIRT_RESULT_NSS diff --git a/libvirt.spec.in b/libvirt.spec.in index 8b88eef..b93a53c 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -79,7 +79,6 @@ %define with_firewalld 0%{!?_without_firewalld:0} %define with_libssh2 0%{!?_without_libssh2:0} %define with_wireshark 0%{!?_without_wireshark:0} -%define with_systemd_daemon 0%{!?_without_systemd_daemon:0} %define with_pm_utils 1 # Finally set the OS / architecture specific special cases @@ -133,7 +132,6 @@ # Fedora has systemd, libvirt still used sysvinit there. %if 0%{?fedora} || 0%{?rhel} >= 7 %define with_systemd 1 - %define with_systemd_daemon 1 %define with_pm_utils 0 %endif @@ -268,9 +266,6 @@ BuildRequires: python %if %{with_systemd} BuildRequires: systemd-units %endif -%if %{with_systemd_daemon} -BuildRequires: systemd-devel -%endif %if %{with_xen} || %{with_libxl} BuildRequires: xen-devel %endif @@ -1061,12 +1056,6 @@ rm -rf .git %define arg_wireshark --without-wireshark-dissector %endif -%if %{with_systemd_daemon} - %define arg_systemd_daemon --with-systemd-daemon -%else - %define arg_systemd_daemon --without-systemd-daemon -%endif - %if %{with_pm_utils} %define arg_pm_utils --with-pm-utils %else @@ -1157,7 +1146,6 @@ rm -f po/stamp-po --with-driver-modules \ %{?arg_firewalld} \ %{?arg_wireshark} \ - %{?arg_systemd_daemon} \ %{?arg_pm_utils} \ --with-nss-plugin \ %{arg_packager} \ diff --git a/m4/virt-systemd-daemon.m4 b/m4/virt-systemd-daemon.m4 deleted file mode 100644 index 8516e41..0000000 --- a/m4/virt-systemd-daemon.m4 +++ /dev/null @@ -1,34 +0,0 @@ -dnl The libsystemd-daemon.so library -dnl -dnl Copyright (C) 2012-2013 Red Hat, Inc. -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_CHECK_SYSTEMD_DAEMON],[ - LIBVIRT_CHECK_PKG([SYSTEMD_DAEMON], [libsystemd-daemon], [0.27.1]) - - old_CFLAGS="$CFLAGS" - old_LIBS="$LIBS" - CFLAGS="$CFLAGS $SYSTEMD_DAEMON_CFLAGS" - LIBS="$LIBS $SYSTEMD_DAEMON_LIBS" - AC_CHECK_FUNCS([sd_notify]) - CFLAGS="$old_CFLAGS" - LIBS="$old_LIBS" -]) - -AC_DEFUN([LIBVIRT_RESULT_SYSTEMD_DAEMON],[ - LIBVIRT_RESULT_LIB([SYSTEMD_DAEMON]) -]) diff --git a/src/Makefile.am b/src/Makefile.am index 12b66c2..11c79df 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -1106,12 +1106,12 @@ libvirt_util_la_SOURCES = \ libvirt_util_la_CFLAGS = $(CAPNG_CFLAGS) $(YAJL_CFLAGS) $(LIBNL_CFLAGS) \ $(AM_CFLAGS) $(AUDIT_CFLAGS) $(DEVMAPPER_CFLAGS) \ $(DBUS_CFLAGS) $(LDEXP_LIBM) $(NUMACTL_CFLAGS) \ - $(SYSTEMD_DAEMON_CFLAGS) $(POLKIT_CFLAGS) $(GNUTLS_CFLAGS) \ + $(POLKIT_CFLAGS) $(GNUTLS_CFLAGS) \ -I$(srcdir)/conf libvirt_util_la_LIBADD = $(CAPNG_LIBS) $(YAJL_LIBS) $(LIBNL_LIBS) \ $(THREAD_LIBS) $(AUDIT_LIBS) $(DEVMAPPER_LIBS) \ $(LIB_CLOCK_GETTIME) $(DBUS_LIBS) $(MSCOM_LIBS) $(LIBXML_LIBS) \ - $(SECDRIVER_LIBS) $(NUMACTL_LIBS) $(SYSTEMD_DAEMON_LIBS) \ + $(SECDRIVER_LIBS) $(NUMACTL_LIBS) \ $(POLKIT_LIBS) diff --git a/src/util/virsystemd.c b/src/util/virsystemd.c index 4883f94..871db7e 100644 --- a/src/util/virsystemd.c +++ b/src/util/virsystemd.c @@ -21,8 +21,9 @@ #include <config.h> -#ifdef WITH_SYSTEMD_DAEMON -# include <systemd/sd-daemon.h> +#include <sys/socket.h> +#ifdef HAVE_SYS_UN_H +# include <sys/un.h> #endif #include "virsystemd.h" @@ -34,6 +35,7 @@ #include "virutil.h" #include "virlog.h" #include "virerror.h" +#include "virfile.h" #define VIR_FROM_THIS VIR_FROM_SYSTEMD @@ -480,9 +482,50 @@ int virSystemdTerminateMachine(const char *name) void virSystemdNotifyStartup(void) { -#ifdef WITH_SYSTEMD_DAEMON - sd_notify(0, "READY=1"); -#endif +#ifdef HAVE_SYS_UN_H + const char *path; + const char *msg = "READY=1"; + int fd; + struct sockaddr_un un = { + .sun_family = AF_UNIX, + }; + struct iovec iov = { + .iov_base = (char *)msg, + .iov_len = strlen(msg), + }; + struct msghdr mh = { + .msg_name = &un, + .msg_namelen = sizeof(un), + .msg_iov = &iov, + .msg_iovlen = 1, + }; + + if (!(path = virGetEnvBlockSUID("NOTIFY_SOCKET"))) { + VIR_DEBUG("Skipping systemd notify, not requested"); + return; + } + + /* NB sun_path field is *not* NUL-terminated, hence >, not >= */ + if (strlen(path) > sizeof(un.sun_path)) { + VIR_WARN("Systemd notify socket path '%s' too long", path); + return; + } + + memcpy(un.sun_path, path, strlen(path)); + if (un.sun_path[0] == '@') + un.sun_path[0] = '\0'; + + fd = socket(AF_UNIX, SOCK_DGRAM, 0); + if (fd < 0) { + VIR_WARN("Unable to create socket FD"); + return; + } + + if (sendmsg(fd, &mh, MSG_NOSIGNAL) < 0) + VIR_WARN("Failed to notify systemd"); + + VIR_FORCE_CLOSE(fd); +#endif /* HAVE_SYS_UN_H */ } static int -- 2.5.5

4 3

[libvirt] [PATCH 0/7] Cleanup with storageencryption/qemu-img
by John Ferlan 06 Jun '16

06 Jun '16

The first 4 patches were already posted: http://www.redhat.com/archives/libvir-list/2016-May/msg01984.htm http://www.redhat.com/archives/libvir-list/2016-May/msg02147.html There was only one small change in a comment to any of those patches. The last 3 patches are new - just trying to make it easier to use the qemu-img code for creating a luks volume. John Ferlan (7): util: Clean up code formatting in virstorageencryption storage: Split out setting default secret for encryption storage: Split out a helper for encryption checks storage: Adjust qemu-img switches check storage: Create helper to set input for CreateQemuImg code storage: Create helper to set backing for CreateQemuImg code storage: Create helper to set options for CreateQemuImg code src/storage/storage_backend.c | 367 ++++++++++++++++++++++----------------- src/storage/storage_backend_fs.c | 79 +++++---- src/util/virstorageencryption.c | 58 +++---- 3 files changed, 282 insertions(+), 222 deletions(-) -- 2.5.5

2 18

[libvirt] [PATCH] systemd: directly notify systemd instead of using sd_notify
by Daniel P. Berrange 06 Jun '16

06 Jun '16

The sd_notify method is used to tell systemd when libvirtd has finished starting up. All it does is send a datagram containing the string parameter to systemd on a UNIX socket named in the NOTIFY_SOCKET environment variable. Rather than pulling in the systemd libraries for this, just code the notification directly in libvirt as this is a stable ABI from systemd's POV which explicitly allows independant implementations: See "Reimplementable Independently" column in the "$NOTIFY_SOCKET Daemon Notifications" row: https://www.freedesktop.org/wiki/Software/systemd/InterfacePortabilityAndSt… Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- configure.ac | 2 -- libvirt.spec.in | 12 ----------- m4/virt-systemd-daemon.m4 | 34 ------------------------------- src/Makefile.am | 4 ++-- src/util/virsystemd.c | 52 ++++++++++++++++++++++++++++++++++++++++++----- 5 files changed, 49 insertions(+), 55 deletions(-) delete mode 100644 m4/virt-systemd-daemon.m4 diff --git a/configure.ac b/configure.ac index 74c33b3..a63b912 100644 --- a/configure.ac +++ b/configure.ac @@ -256,7 +256,6 @@ LIBVIRT_CHECK_SANLOCK LIBVIRT_CHECK_SASL LIBVIRT_CHECK_SELINUX LIBVIRT_CHECK_SSH2 -LIBVIRT_CHECK_SYSTEMD_DAEMON LIBVIRT_CHECK_UDEV LIBVIRT_CHECK_WIRESHARK LIBVIRT_CHECK_NSS @@ -2787,7 +2786,6 @@ LIBVIRT_RESULT_SANLOCK LIBVIRT_RESULT_SASL LIBVIRT_RESULT_SELINUX LIBVIRT_RESULT_SSH2 -LIBVIRT_RESULT_SYSTEMD_DAEMON LIBVIRT_RESULT_UDEV LIBVIRT_RESULT_WIRESHARK LIBVIRT_RESULT_NSS diff --git a/libvirt.spec.in b/libvirt.spec.in index 8b88eef..b93a53c 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -79,7 +79,6 @@ %define with_firewalld 0%{!?_without_firewalld:0} %define with_libssh2 0%{!?_without_libssh2:0} %define with_wireshark 0%{!?_without_wireshark:0} -%define with_systemd_daemon 0%{!?_without_systemd_daemon:0} %define with_pm_utils 1 # Finally set the OS / architecture specific special cases @@ -133,7 +132,6 @@ # Fedora has systemd, libvirt still used sysvinit there. %if 0%{?fedora} || 0%{?rhel} >= 7 %define with_systemd 1 - %define with_systemd_daemon 1 %define with_pm_utils 0 %endif @@ -268,9 +266,6 @@ BuildRequires: python %if %{with_systemd} BuildRequires: systemd-units %endif -%if %{with_systemd_daemon} -BuildRequires: systemd-devel -%endif %if %{with_xen} || %{with_libxl} BuildRequires: xen-devel %endif @@ -1061,12 +1056,6 @@ rm -rf .git %define arg_wireshark --without-wireshark-dissector %endif -%if %{with_systemd_daemon} - %define arg_systemd_daemon --with-systemd-daemon -%else - %define arg_systemd_daemon --without-systemd-daemon -%endif - %if %{with_pm_utils} %define arg_pm_utils --with-pm-utils %else @@ -1157,7 +1146,6 @@ rm -f po/stamp-po --with-driver-modules \ %{?arg_firewalld} \ %{?arg_wireshark} \ - %{?arg_systemd_daemon} \ %{?arg_pm_utils} \ --with-nss-plugin \ %{arg_packager} \ diff --git a/m4/virt-systemd-daemon.m4 b/m4/virt-systemd-daemon.m4 deleted file mode 100644 index 8516e41..0000000 --- a/m4/virt-systemd-daemon.m4 +++ /dev/null @@ -1,34 +0,0 @@ -dnl The libsystemd-daemon.so library -dnl -dnl Copyright (C) 2012-2013 Red Hat, Inc. -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_CHECK_SYSTEMD_DAEMON],[ - LIBVIRT_CHECK_PKG([SYSTEMD_DAEMON], [libsystemd-daemon], [0.27.1]) - - old_CFLAGS="$CFLAGS" - old_LIBS="$LIBS" - CFLAGS="$CFLAGS $SYSTEMD_DAEMON_CFLAGS" - LIBS="$LIBS $SYSTEMD_DAEMON_LIBS" - AC_CHECK_FUNCS([sd_notify]) - CFLAGS="$old_CFLAGS" - LIBS="$old_LIBS" -]) - -AC_DEFUN([LIBVIRT_RESULT_SYSTEMD_DAEMON],[ - LIBVIRT_RESULT_LIB([SYSTEMD_DAEMON]) -]) diff --git a/src/Makefile.am b/src/Makefile.am index 12b66c2..11c79df 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -1106,12 +1106,12 @@ libvirt_util_la_SOURCES = \ libvirt_util_la_CFLAGS = $(CAPNG_CFLAGS) $(YAJL_CFLAGS) $(LIBNL_CFLAGS) \ $(AM_CFLAGS) $(AUDIT_CFLAGS) $(DEVMAPPER_CFLAGS) \ $(DBUS_CFLAGS) $(LDEXP_LIBM) $(NUMACTL_CFLAGS) \ - $(SYSTEMD_DAEMON_CFLAGS) $(POLKIT_CFLAGS) $(GNUTLS_CFLAGS) \ + $(POLKIT_CFLAGS) $(GNUTLS_CFLAGS) \ -I$(srcdir)/conf libvirt_util_la_LIBADD = $(CAPNG_LIBS) $(YAJL_LIBS) $(LIBNL_LIBS) \ $(THREAD_LIBS) $(AUDIT_LIBS) $(DEVMAPPER_LIBS) \ $(LIB_CLOCK_GETTIME) $(DBUS_LIBS) $(MSCOM_LIBS) $(LIBXML_LIBS) \ - $(SECDRIVER_LIBS) $(NUMACTL_LIBS) $(SYSTEMD_DAEMON_LIBS) \ + $(SECDRIVER_LIBS) $(NUMACTL_LIBS) \ $(POLKIT_LIBS) diff --git a/src/util/virsystemd.c b/src/util/virsystemd.c index 4883f94..71b8f33 100644 --- a/src/util/virsystemd.c +++ b/src/util/virsystemd.c @@ -21,8 +21,9 @@ #include <config.h> -#ifdef WITH_SYSTEMD_DAEMON -# include <systemd/sd-daemon.h> +#include <sys/socket.h> +#ifdef HAVE_SYS_UN_H +# include <sys/un.h> #endif #include "virsystemd.h" @@ -34,6 +35,7 @@ #include "virutil.h" #include "virlog.h" #include "virerror.h" +#include "virfile.h" #define VIR_FROM_THIS VIR_FROM_SYSTEMD @@ -480,9 +482,49 @@ int virSystemdTerminateMachine(const char *name) void virSystemdNotifyStartup(void) { -#ifdef WITH_SYSTEMD_DAEMON - sd_notify(0, "READY=1"); -#endif +#ifdef HAVE_SYS_UN_H + const char *path; + const char *msg = "READY=1"; + int fd; + struct sockaddr_un un = { + .sun_family = AF_UNIX, + }; + struct iovec iov = { + .iov_base = (char *)msg, + .iov_len = strlen(msg), + }; + struct msghdr mh = { + .msg_name = &un, + .msg_namelen = sizeof(un), + .msg_iov = &iov, + .msg_iovlen = 1, + }; + + if (!(path = virGetEnvBlockSUID("NOTIFY_SOCKET"))) { + VIR_DEBUG("Skipping systemd notify, not requested"); + return; + } + + if (strlen(path) > sizeof(un.sun_path)) { + VIR_WARN("Systemd notify socket path '%s' too long", path); + return; + } + + fd = socket(AF_UNIX, SOCK_DGRAM, 0); + if (fd < 0) { + VIR_WARN("Unable to create socket FD"); + return; + } + + ignore_value(virStrcpy(un.sun_path, path, sizeof(un.sun_path))); + if (un.sun_path[0] == '@') + un.sun_path[0] = '\0'; + + if (sendmsg(fd, &mh, MSG_NOSIGNAL) < 0) + VIR_WARN("Failed to notify systemd"); + + VIR_FORCE_CLOSE(fd); +#endif /* HAVE_SYS_UN_H */ } static int -- 2.5.5

2 4

[libvirt] [PATCH] docs: Document our event loop
by Michal Privoznik 06 Jun '16

06 Jun '16

I was asked the other day what's event loop and how libvirt uses it. Well, I haven't found any good sources on the Internet so I thought of writing the documentation on my own. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- NB, when I will push this, the images will go to our libvirt-media repo among with the original SVGs that PNGs were rendered from. I'm sending them here for easier review. docs/internals.html.in | 1 + docs/internals/event_loop_simple.png | Bin 0 -> 16043 bytes docs/internals/event_loop_worker.png | Bin 0 -> 29696 bytes docs/internals/eventloop.html.in | 106 +++++++++++++++++++++++++++++++++++ docs/sitemap.html.in | 4 ++ 5 files changed, 111 insertions(+) create mode 100644 docs/internals/event_loop_simple.png create mode 100644 docs/internals/event_loop_worker.png create mode 100644 docs/internals/eventloop.html.in diff --git a/docs/internals.html.in b/docs/internals.html.in index c30f52f..fcc0758 100644 --- a/docs/internals.html.in +++ b/docs/internals.html.in @@ -14,6 +14,7 @@ <li>Introduction to basic rules and guidelines for <a href="hacking.html">hacking</a> on libvirt code</li> <li>Guide to adding <a href="api_extension.html">public APIs</a></li> + <li>Insight into libvirt <a href="internals/eventloop.html">event loop and worker pool</a></li> <li>Approach for <a href="internals/command.html">spawning commands</a> from libvirt driver code</li> <li>The libvirt <a href="internals/rpc.html">RPC infrastructure</a></li> diff --git a/docs/internals/event_loop_simple.png b/docs/internals/event_loop_simple.png new file mode 100644 index 0000000000000000000000000000000000000000..daf4338a16bc1051aca0ca1c7ea13baa4e7179f1 GIT binary patch literal 16043 zcmch;by!wk*YA4+0!ky2A}9(9NP|ctAW8_*NK2P=H=>eCmy}3}NOvP4-6|=KbR*$S z?6H3D-simMdEWh=zxMXJR9M`sm}|~4zT@*9Gx({(BSJhHJQNB=DDzlK8HK{2hQE;) zao~v3Ui(M*?}DR*jOs=B<8{$E2)@5$_xPD3e0~J^iy?6*$qf!tJ3Z8NQn594a((V# zf^v0rWw)@gax{8wXTomlV3xEgOoKvQN6AP%P<2b*_~ovn>O9fD)Bk~?8AZhIkH&nT zovog2u^AgDXZc>evbw?|QTx4xWma6=_={)bHc8bcHi_8>nJ)2hrWV<ENUmSM9@K)u zZ25D-#p(a?BWE9h&eq7<6|+Rg0-o-X>Eje0Gx69=1*T|=((v&9X8~Ugu%&PhsR(ca zezdpW9y8{Cfg?z9w#%|8Oz!3FEp`bHPgb|Vt6J*eLm3yBiaUpqDJg7`k&z|VV<pK= zO--u$`qSZlczr%b%-DW~g<ByuziOT-tEi-$K2u_Bvc4ymqmXj9`awoW=gc~$tiQs8 z2M>}YB_+MRyu6ZiLs@R*=H|Y7^-6PYZq5U<HbpmFh;?Zlb26pVSWD2XFFE_gM8%IL z_J_hEB0GC~c}1>UQ>)a})cS(8bgtR-VTQXqI|82j%Sz|t<yJzqSJavrZr-#QFS96V zGrVNy!nP?dnVkSn#>dLqouMP};!4~-^}JX-a+ZalpoXfp4>wf27#R8r9tZ?|H9(yv zUkME?#eXUrYeT~=%N#fqp(HC1^CnEWE1lOkMwuw<Mn;Uo9YX4?7)gtcJ6W`{lG#^; zTsLWVc6Jmde5|eS+S}X5>oVQG9avPvx&QLpkQ{tKyVwwW{@YuuCd1<GuekI&@f%FH zZVhD0ka--ecfQ2NXRrOzW|;oG<&Ckasb2i}_dhxDa+GP^KKa*KSVBt8`iJV=9Y1Dd zTp}W3x_Pr@cEbDo^l+@qf^2Q1uyL-$w3J#*OpKeG`;%tz#}=zvt-s|j=Z53LLdi&{ zJl5E72&kH3m{mq=T{*Vu59v_B!NGO=Lvq-0=Q9Ds_nnqWP=kYmw4z>Aew9-mq3i3` z*$OFL-*Z(2oL9d;oRF1~nLV8L<xsDOr(6i-XqZ{(jFwLkjMeu!eyx(Durn9RG1Tbm z`-bRR*BipC?8isbzUL`}8IOi@6zQek(zMw&o|>k)ZqGbO^;l)bA)@W@JxA+*QMW3# zR2zBZZbtcm-S%L8%zC^m_;_!LjFj|i#0|-dmo7cuUmm>0$hh-6$#xemF&=-unpe1e zaPS)r{H~n#G1{7{C!pqjb??Ra%<Sxg!_8k9+k1Pp=GtPX`((tTN7tA7Qm|&WXIreN z>W-tzMGi)d66CWnhgU}m6(2voxY!+^<9WFG{mZjx-?QVurY7;t>LogbG|}EF`}sQ> z1+<61T=o6iO^uE1Ms!_^4ZdQ&=d=1Ic77r$D;rW@@gM+)%XL!+PS$^NyeHsua^JT8 z&(Cs^;?<hL2>a&@&2w#`$7krR>OViVG44AobnunmV<$+IW83THpB`=J;o*@{P-s}# zeO-QQSK>FE;^4)P&uXDUn^gg=tx_*vzNETmO=b1-OZ6S&sOV@JMa3L%^Z~4p(L?H) z#dv+w-#v*0*95QlnfAUXy>g|s`8EE|{{G$Z=nu&flHXo4-ntbL6(tc$!L~lERc5ZV zF8N+AS2+t$x5lYBoZsqa*xx^E`VQ{CzUK`2hXd)7!YA7;s-IPZVH7T3zRYfkfpGyF zo9lV=1+7Y(CmGUjmxptR$(ZDuMn>XqK6<y^PVc)7FJnBuzpINuQ1CqtA&pL(-siO^ zXk(p(v^07;I=W;L&!3`f=hLaFsm3NI<Zkj)>3y#qgg6VkM)ZB3w7enY5Rj6Rx?&+R zoc(0J^LCo?`Pr%UWL4zX09^gum_KqH^-_k04358hW&`Pc40;l{P!}#-D6$^oR!9}L zp#P$thZA;n00vr5PtR%fd$37&97aT3+_w;N*7ZHB@1OjCST4?d4M3gzT`Q!$Z}(xq z8T-P8h}c-?u{G1?=4Q<@^GNR#SctcnnOCADmBeN&EG%L;Y^`{6GBdZ~k`y(QW2I)n zp`jUuZ6Pob1K_^Mu3p6#hud^N{WBu-<ViV)FZ_Vvg}y*M9eZ_EmR$T$t*h<Nd~LI> z!VaqkZXO=owo}4_&Z|<VfB)nvW#Cl)a?ws-%T>+2467y!?%3Gac)NvI?8<!ttottO zR6J&V#wSNRj5WmY^TqT)y18gY(HjB+iSMon?e>eE>y+t0bb(iRgOiiQ*47qc<<b7I zB1QtY30{t3S~g5cUhB~aSa!polx|ko{L=bZ{y8dV={&T5cBC`tB1CJ|JoUFvlI|75 zIJ&HlK2J-7Ng^wY2kYG4{90RNT%0sqq1<Yu!TQk)1qC85F0O@zg{K+Pgr4|!6irQ; zl$4ZI^R>!5yd~PwM>?;hRTxkDeKq(1u|7U|Yir9_SJbX;c4Gs1zlBxFfq}T?mOpA1 zyc8At8S)=h6KzD>^RQ`^8Nj&pu}kJM44WG4RIiWVsqU=ID(_gxNi#1#{r8v~(rLiC zu*`-ldpqw($XA07dzWjQ3J}oi>g#*q4&h7!;!aMyQ&UsSeFZJQEIO`4=j1G<=4)53 z3al%eul>xQTUn`YA9Jn}5;@oRx`u^?RY@!6Gub{eGBQ^F>9XOyB#5gOt?lj8SnJL% zBAfEs+--*$ZHCW=YhAa>PNqrJH@mIx4d-bTmhJEF$C@`?(H1bHd`bhc+D1WE)^fW~ zXzSI{-@odg^y+KAn)A-hrc-aSe^F<Ju=+)(S{W8wJOw%VpV%>LAIyPw`ish&yovHx z-%1;m1O`4Lx-MAN4iPNUh?gc>o@p58gZC^)q4KK7fP#&)TA4`Uzhs;LdXWCh#Pomu zqwy+4N5nJE!N+>ld)>N~Au$t!5lZn!*fglYru%1vR|_Y1mM4$Zb5wI;SjbQOd1r(p z2rk;;babR_Zq*INC}*{?`eAD~884kpM+xaNXNC0Aa~k2t*WDoJ6ezv3NLSXPK;Guw zbhV276Dkuc)Q_$2t}L@!&4O*OIV(IJ35StTbfz~Kr0H#3hCv?LRhDIE?p4b&7X5}g zv4O6!*LSjH)A~e3MdKj1gocNQV=u3-zl7_yr@+BnA(wR$(#T)6uAzhGWhmVg-2~;X zQ}T-m!Zf_HxlvEeo6HB`1kEN`(gD)UYP1jWs0eAYzL!^h_8a`5B&++@>YoFuN>^A7 zc+F#!<5d_iG9F~au*e7FgyFnZW)Ta#fc%d~OUnsojD0b}h<uJEslW<3vs%Ud|M>oD zGR;3fw_zvcbVoj*tHWZji+n<`%D8C(uGLjv?Sy>6d`i&AE-uiD-2Jr8XwlakD~*x! z_06%;z*G^>?5ynU0~{B@^jB#AZ}AC~L|<y&mB$GFjlPo=u8mC~9!PZU=wQOO{;l-9 zbn@^=E<^M~c`{BCxBJsog=DKrT_)9SH-@sRI?{)engWl%)f(cbV>EGoU>MkV*<OaS z{C)NCf(J%8R;9q_r&$I3l|%``>?FLDZ{%@`uDa1D5MIDYq!cz~+t%EiK5MdVqA{Sb zyWrMmzsTb39TVc$Ilu6M`L1zl2ft5NL?XUdgzXfeIG)K-j(54@x6^7}`kvS-ar)lZ zF1l?ct|&r3lj;_^>fx)gPD-1Q(&}St;0%TjqT{x<XxT+swYjFJ>6%0hoEx5_?3yF7 zSr#v5t&*%LC)+Knj?OA?c+D>sUt|B#eeDp#klC%O?BJ#Is@U(YjhJA+PREph-?BHb zDq22>`SBMy<DGXl-k8TfAwI%zqt?VrcG9VqjjiJ*=X5VH8sNow|NH6|o||q_81BGT zs@EIe|GdsA5^<pCr=hxVLGZI?maJAH0Sttlhjy-VzC5lx9ZV5gNh~IM_@8}4tB)Jv zg+zdKZu2Q_g3MsynbVjbr2e?0Be$HNA0A|UZ>#g;8D-z}c-<ZNwi_obsKctTp6RA^ z7)(qMeqE2~y;Tj)T&KNn!{16?^8xW1=QSpHlMaus!JN{yrme8z81h%eBfc6>L`51I z6m(I*0d#E;t&od6i$;Nt!5OFv6ZFBxWiqmxiejf_ytY%5&&!e7^;FMF+9A-%nk|XH z=vU)H4R+WM%h-pZ6rFw;SjHfqcYV(feGe!)b5(MTHz%vH<MeBsaz!+Kj%IP6)q9i& z1qtu<@F*7>K1}2`36W3aUi{ed>Cf7T9Ha=^8V(xX%^cEg%p&TEA0aX)==6bf<2UR0 z-}1$ck9QYUe)p$s4G<frc1GWpt998h9x2e>-K^b`eM`fasZprMz~Hbw^ELEkI>??9 zpHmM~c!~QfIcXt-<}n)+mE8v$6O4R(Kh6?5Iv(e06sCHe?6eb2ry;8xVrmV3U4@Cs z<K{TxBtMHQJ8~b&{R;i>8{DXvxQt_jrc{p&V=Xrpbo2B~y7nuov!^FUHkM^fn~8xz zQFy17@3s8%Ksomb^RbewyP8E!O0~}q=flN9!ow9=#-plJ<wtjCeGEt|CN<^0=B42$ zE7xO9R6aVy_~U=6$-{b@?um<6y~(%v(FWCc49%j~awMFj6MJ>H+1T>UK(ea1ZaPhF z*eIq5K6G)p_vAfqv7*)B$A`f-^UH%7P+f6Z3}s%{-Foe^HT9%FRfLTqG>f>e`jOqU z*f2KwpIGz7i{CJ;sfXBY)9WH~UvDVrwfMV_o;j;A_Twk`Z@6IC8pxfjjM`Ynq_qTO zJ1lmA`g&fhxjx*NDq?j$F+3do<;xe$Z~@S+>1*6nc#ZVlZ6W0DGyb@;AI9Z|{p^`n znrN-3$FwAELYX*>;}(df1+vOYjILKK2?RB_Fp34@JX!3Dtv)@NcvfLejWR7w>=#A% zh&H-xjAv$M+CdSf&E<Kp4JwbkrhrB5bHw}iTGumCb#--kWihhval*RCG3&5S_-!SO ztPfj0Eqn>-im<Al8+q={>wB^j_&rZUQLEgN@$FUqIGFn1r*7Y5Ky(z8ebE$bT94C2 zy%*xNZJ5gxrKJ$HbeW3k4pX$uJ9;s81_lV{5q^GtmkYI}&<2x!{1u|_Ns9j9aX48D ztjf|-@T%vjHMg|185VaStLnP3l(oizw6t{GkxQr~*0&dTak5|LTMUm>mw#fgJ9?JW zEb;A!c~e#FS(CIMr!~K*dO?5sgFO(xqZ~z!4{11UZYNB7UQNp)IM)fdTEjjvSVcof zRcYMp@|8|~U%GuWTORu@PT0h5QyfQ7G``YEw)_R;dor|dj5f=U2N@VDEan|Cm|5KB zyauv_go(-@RpA%pb3fit35f~$FzCnZ%xf%5T}ebkn4l~JRir~Va<Ci@GFRdu2cN(} z$57<pA2=uwk9?B*fn+v;T_<vohu664Vm$Im?uT&DJ{URZ2nVThaNt6TnKx8$1Q_A! zU3Uo+xmRocIhcJz<&ADU^2t5;<o!T67$zYpJ^KL;DOK|tUwPULx1O;JiQa|rGA|A1 zDkcs0XFo&aSy^7a39ISgL%yuJKuE3jDGx4A<xEY(y&4sA1^Fx<ukm*i-Ft9NRZ{bg zPpmAw|9olD@ri+%1o;3A$tQ-j^^J`bLEkg4or41fY(b&g>x&fdi)t(Dpp)Lgbp#2i zoqz0~ek^$7cE1)gOLZmt->UNeUi^I8g*+Nd6YF<CDKA94-q2$t>=)Z)*EIK`CMKG? z$eXflW?dp?qUt+3>xLcX5lu&5iVKe0F83Nse62MqLgytP%SgI@gsR4d@>cwt>b<bn zJ$tyDC@sv?<Lvlu?=a@+G7{_QU7O~%=jz1f_(!@Xy<P6eBh-#ZtYFgO{=^c)Z1)?l z-h-l$yeX}T#ekTT&9%mLbN;;J)R%W4BLSUWK#V+P!4+;H|BdPvPHS(zW9?gon4ym6 zee~Q+*8@Ve1yMZ3e||0G=de|L7$l=CP%N9*h%I|4?y)&NE6&7uW#UEz@l*WAYzy{F zes`?B1zG*`#U(cDQk!lVI5jL{`PEcUJY|=cl#VUay+RO4J-Uf~a+w5ol91reg1!0& z!fldm&LRRgnj@5UTHi5#oG*noCo}(tK)F>*ofm06hfOnH^R|X-N&d}3K6}C!PU6sq z@lADB%<2LubnMK56S(qCE}v@z>^MlYI~=}_{?)6%GPNG*ThKMyW2i7-NvmzJ92Jbg z6)dq+<K{J1l@Y*_Z~KWGJ$?5&&hoo_{Va|F7FRumO`{>l__0KxLfQ(RC#;U~_}Dhz zepU`+KfW;K&|`Dw(Lf8MGv7aRVd}9C^P=XzR}lZTIM!7Vm=W_EVf^j6*vH`-d@Ku- zu)744u;^E<3GMltw;G#nPMcnNDxXzsBss`5;^&DopV$77;I26Arncr!knK(>{GPR( zvZESV_i|(8!i9;-Tt?Ruax;po$(L`!ST6foSL#Q2NwU12#+A-Jf#-{??`h0W<Sir5 z2$7bQCVzJJGpxc;RS7Pex656@BQqYN^bqJ$L5LWMf2xx5!zGs+!<_))4uEc1aZ-J` zvL5vPL7=h_K)O6gMkSrb;{;Bv-a)k|6fW4n-wr|7`9ax=-d`V!{_x>?SL`tU6SXme zaxEY}vlF(Bv57op+0X9A{V**(=DOZJR<7&s`6>*dbpTNajsUFG0gy}9eW$Dt`~JOb zp<ewE5$*k=F6(>5t}Y@PGZZ(5neW^&cQiNu6pj<7eD6O&yWK6t_9}*v#>lEpV-FkP zIpZ&tSm-kbJJ!ca>z^<UJN6YU%IY*Q&d<*~9zdo$gDTU-g;};b1`_!DdoSKw7i(fB z>vo$|oy|slH7GJQF>yD)U8s!xEMJ{P^;5)6hJuq>yVD2^DXC=L0A+Tl3H?HSeRLZX zVVzI?Z@}EYhXD4w3=Ehk(%CfMGkY7Zk;jtKx(j)Q4RiY|!vOE(<p}_}*pDCRwn5EJ zCFF9~pti*IXnXbq>THFA)$yvoeCUq_g%J@XW3{e@!xfE=O9~c0za$P<Y_Wh|x+!*c zl;Q2w9mh5VkB8YkgOjFTdAvEqKkXF|A5R%bO#d6QN2Xeyx_p{w3Y&JtHjodbyf*Di zmI3^;N7o9pqXTGLcx5F&kd}paLYt`TqN1sgPE#7s&#vn0>;FC2=>G5^prfzPv@?ov zeX9Oyz*lj&WFnUl4honY<3z=an?MbI_a^ZG!I5W*_SWu>=P&~W_`N_k2|&hPH%H@k z4^KZTA~CU3+qS+9;DUTCi<;Ztl^lecrVCxYdNp*+I$J42y3J$gE(QMH@7{SJX->-n z{@G8G07yx~+*t4b`te~XLN55H+-^qSR{UnnUF@(Zi(o!ba-x<$zjS&Z{%TKgoqD!> z4<T!wTlJ5^=|nW^+_+Fi?eDUH>K>T*94}Ie{$0Kq6&1BSl!YHIc6J~5kr&*{(ed$M zrL8{72fe{xVLRPd+ArE2M0#_y!CMH%=7IstGSx4d#RD|=Y$hLrraVS(H68=JsXG3h zgp39Z#60koTX*gR!Ngb2*WznB(DDYMXg%GK0*a5zdXy7=#>m993zTRNhWI%z&C0J@ ze#tikUoFdbe`?$AfbPQ_0QL%)4ZG>HTHa`lGpme@%%s!s6Zz!(k$!%@jaN<q^OZWU zsayW|+$KpZ(h3^86>b$@TwR?4*ylZjM`B=Lgk2MggVSKxEf3uIyT2j{_-pDgx=1Vk z`zNIU5zhmX%a<{iUE^iHgS4d<opd29DmH2_vH2y4yl@zGDjrjUOP4MQpB>E~EVhSH zx2*oiqZaYt74g_Nm}&A~pKhf0n{5e<hIbl9&4b`WAmpez7z+fg29w?9_BnB@t2;YQ zIVQWUz-c!lp6a<N0JyK}{0zOZM{!rPX(&t1eJ=R6`9uZV`EkGR(b0VP#@<%rxepPY zaL2HsSZ{vilq?AK;IOc}PGZxZTOz4myE5n_n6Uj&wWu#g9DZf*Y4XPsu%EjS&Gdu; zAUG(km=qyb%!uUV9{BZ;#zrx?yp+`PcsVOTco-1?M!hd8e@O$ug+il`u`)6tp<}Yz z5#AY*WFS$N^V&~@z(FxIZbi>)PI67i#dv+R*b6Y6@cGFeYSLpZKa5r&3c=0`3&wPA zGZ-%e0&<_GV^&T&;7eyjZvLtruJ?59is$g{qyg^r6($j~eBdP3qecEOYtrBvT&6u# zFte%oEMvTn&C<N%I2w*#!V0hlX3lNgL4v-~@kuG8Nb@eh$N<ut(spMj?x-mr^b0_K z?MFNF)M7ruz@QHPyngeB+n@=BjIOe>@-j@QDD>+oD65Lyb`0r!PwdAlZTkVb9s~T_ z2=@l*Jz8l?59?tOYV#cJN}E@KfmqwWT&D*cygfaeP%*3;?<!tQl)!3{6QwWcT1?Vr zU@-a?7I{PR&1h2aRs*^K7PfXtY3W0$1qiK|Nl30UFaWBX+t^?;Egfq(fn{n8QH+6M z2posw_0s-Zw{Pc&5AWsXq(<FNb&~++0PjTupzg;~Qm>Nl+e?5eb4yfIPf7~bFtgan zc7S)a``&MvJoS7jDPt&Nlq+p?JyRGMz|x3@ghN^L7QBw*jmqiu`Dw4cw@?aZ4Lk=# z{kmWtM#hkr2C3CV^I`ELlaQ1&QOLDGc(>ywHk~8z@a_WLTB*$CKsgAI6vV{DGV=0e ziV%nsy};#wn@*R+Jq8D0eNQ1pkP*Pg!O|&^(Ma$XWFQPJz*CfPxbg+cA$Q}Cg*)C3 zGr=1StEewu?vj#TA|xb-ut`c9{Pt}utoly4;f-#NXC<v?I#3GhK<vo~ouWj38QwPe ziudf&(rs|7E|W%j{8{aU)wt1JfA)6`v0z9^$0}{pz%v-u%L(de4JI>zmCV4v?Xd7T z@!kuqbTk?g#YDS(K>t?Od9d9B_a{=Q=uQLxZ}5{5U%Yr-K!Dm0;&>9=nN7p7DSDS8 z^fl9y_b5&%ageyF=QfLtSVC}WV3MYjRrbRB1Cjva2lKW0Q{Cn+LGk}4XdA9l`n&~) zhTkTv(->K0qobN9CnrI{!NrAFXR?!QO3qIqo{`ei>(6Z8W@K#h8Ffep45`xS<BfKM zg~s7t6?Lv(eFEhWtU03NEvC^j3l(%@(W7^?h*HP`E9&4FoGh4&!_|&U5+m-+5W`e+ zmD@!1$}QE<Ye<F1<-Tj2t(az$mImb5%S)Jl(hdt&>|*`ll+kp9_ZPkTzn?3T^zH2I z{5`-av9&z~h;MK&sk5)|_h_*Zl%|95JeYqLM2L|Lv(rv*29pL+FPF%e1^{BG`uw$c zLqIKIU~v6cjWhCrrGnZ`2`MUAYru6g;Dsf~GXZHWE-s$5Yax`9mZosiUl5ZzTPtk5 z@A#Vl%E$^uIT@LBn0lzpoE%DzpykEIHV7>R$JGuvm3WExVybj>ujw}u<(cG}$Zjj9 z-gWYUdtU*^1DBW%K^0CbL$5)k(C}Hd7B-%Bfd$k8R?zZ9B?tO&>oz23SeW*6t(U-@ zEY`}WdMhpU3T%kEwKX{j3CtgP8n}1tOd&N6etE_Nlgaf7TG;czdJ1x(#L_VuJ&`nW ztd_483$Bk;>sfM&sQ0*DTBot$LPumzn%Mb!%dRcu!Ut0>Q@-b?ayTOp2=2S@-hy}G zbueaj(DGjHwV9Gt)g-#PzdsxT>y?_f=*`;jPoM7ioS(XS{sIJzm;yjiNU5ni!5&>1 zDWpF~PoHxp^&-J<zCC<?aq)G~%+gX|(9;4P0T=|xGBCx-s*b$v=il-p97e*5kbvM8 z7uU|mmE(~m_Q`C;-<Gg;lDlGAOPyAf(Peu}{S#slFvx8NoC7t^R!j=XkKyUZ#>RlO zCnqP<uKvvL24f8fxgdBIOl)j~u+Y&-V6sh@AI_upx2DrH^xR+f^vFX-!n3Xf3j`8m zk;O0*nzwXJ-pR=+y?pCeZG0@N#(2BF=LR3h1-xr`_*8yC1!lIk`eD)%Hi5VqtFkxp z_t5yFsWh_#F)%Yb8?z#!M%7y*!}t6QMl`Lgj2B&&@hBX#f@l*-PUNiWMV3EU5YIF6 z<|EW6u*ty61-4I7`{v|5uSGK<|6~(lS^o2sWE2!QqgqWk0siXwTEJ><0aXKlRMyC` z`aQP=a^w>LSV5torHvfBTEmb!QcneXCe8&Me<S&#FIm9IR|K_Lx2IreWb~}jr`}u3 z((?1EyDenZbPOjjN#PY~d^C^28%Lq&Xpvrk{?Lrrt_X1Vj5o&1y^dxBc`d&OfnFad zdM*jj1$$KRp`?Vw@@Vl*u!d&e@=tw`c#Vgo{qNu>f{eXPODo_0?ppO;pU}$B{Hx%| zFx<NJ3X)c-@3{{cjq4NIRdzR?J$r`e5F}ORX>g#^WkHTUD>Wqq*Y5*_{xX|i?~`n5 zf?EQKDYejWR`xLm=Fhu}-K8!YTB4`>Dxi(NfycZ&QkYhs&8}O+Y4!6iJW4wx_HR%4 ze@VgfZO})^L^E+YuY%K<t2`56Uk=It;nSxigfzS#bZcDz{e)BMxnbn}=mOQ&(-C=d zWvX7NH;La2*rV{#Oh9H<7P5#oYu4B;M+%;Zi;Mdc+t{mqdg{+-^>h1pDUIy%<z`4i z3IRBT5MFRmAmo_0YdAEEP#hc_aqPNUGv0&QPp;NDtw?}5Y%yAN1O7}pbhCkqPC7aK zm@4Kg3U3xxHz*-WM8l%347d|DYhBn7%MQi^`7p1=&@0F=*ZhFE?}GG+j)*YO*H=}& z0zt?LqLNf<_*&<p(7TYETwF>JU>AYEVrRFvw|9dGyCou0-?Z<z)F%T)6r^Fwgw*Kh zn_$g3Kqg_eafE-l4O@dkv)jLYlT}q!_3vv~%Yv9%0zM?>F!-LZDjr)~f03+AiHy8) zTOm0jCgw_@&5~s@OtF%!(6X{J%;8?s(pPWZ$SNxjJzNV33yZ0$s)AGh{PK(&6g-@| ztxd+*c;Vqz=u(=m9@GhD^nlfB>UO(WvE#lHiZ<Ffri8IVj5uqDa-|1eU+d0%I}l$! z=m5AofZ<hN4>1OA1SwryHYYoQ;f_HEfdBArS22XRc2HD`rKY|2K?SmE6fD5EIUs)? zdYnKeNq{jftlPz7XfCrDPT(?<_E%Znv`YzlBYz1ANn<6ifna}D&+mdHqCI2RMg8&P z$0K+Zk7Z>)g3q|WF>&}F(-g9GPpSy|;|kPSzaiGM2P+qvw+4}dxTmLl;`=+%j7rdH z**SMs3j#oiNjC+O=(NgNKrG+cQP3t#jGEmAVpLis*%8zAPCqMVZQx(pJQjCv7;#cL zz=)T2KkZu7%vUGCk$hLWR&_a^m>S$yM|t_Me|{qM=+QjTs=0N+g4RXJZ^Ofm{jj+H z&{il@LbZ?wKJX?umKt+JBb`nJ1O$@~Cr9@5AV|zcKE}wwBVWM9)t;N5-@-T$tWE8< z3tcnzDbu`b<#D)~qOPnTqtca<*lP+(BsMNJwc+vz5mgf*Q9P>0bZ;38m*@)~IG1;| zE99Z!Ww6gcEe`Gzz63Ozh@2O6d)1KjJ+ayKnvkxPze4{X-?3ucc_95eF%(+@(ivO` zXh4n6i?*K8XN5eT=2*3?dYA@HFK_>6lgq!?(*N6jnEy*3tx0xh@<kQO^&EDk=x^4j zWlp{W`bf(<fC|F-wqrZ$w)pPRQeUNDmf+&O)-mim@xO^PF_!#|t+(QN#HBZ#e3xRg zC;djxGrpVDVQLoDYm_*i+jgcDd|yNm<PL7n=oG8FB#VKAsZtjka>z5$SDVKIj7Rdf zz;{Yc@~L06efkLfK^luD)@D%%*O3LFK<;}TQzRoZq;EqaI~o|xIv|gq2ArDAqolb3 zL|HDJR0Qhu^l8S?on_R^L1+@JiWJWKwNb2}X4clw4TI*ioS+ZVAh=w<az(1z!6QAF z<tj)Vu)4j_8<pJ96TutWRx%E+6~|bhwu03zJYwh1A6<xSNf?Owt^}L2v|Fhw7uv5@ zK!1#bCRR`E{XN@Pu!J|sU@VA@js1d%knj)i+%6_QzF5E+aymLXiUyUXYt;ny$6M@) z^4UwK{i*qER9U-RG{{n$m@wuRSz6bEIj+qO8r%%(_*MN{mltO7FpZ%g%(NKJ?xYB1 zY#J-qd9y~(PECMgrLG<q=(5YA7V*O}3zCum!(10V(FifMxK!&H-<`!M0vxQWqot8W zH1t(%K9NM`4V^d<%{^IW+#OUOhEE^66tkM3yFqQ1A-Ngl#~79rVZTN{ivM{SC5GWD zSW5!%*BB+?S1;~uU8&*3=aCYo96!T8()kM0)6XnLCrlYkmFMMgn`!nTK|yvdmqyg4 za@TVFbNoD0G(%bDiLME@Vi61b1wLA7_-KffgcB9LV<tNM?uEaZXE0_&3<PI3%0g5V z`Ha31Fe#=cyIy>1DRMaJSa&oR+Q&X$Q@#D;%cZ(YPN;@fm$VSE;!{I|p+9WB1)AR6 z&@q5mFekg+9MFW~H(^uz7BZkxx&*e=D+s-*P{0*|e1gIhkuSX<KQwEcm;u&J+P7bW zBK{2|SgeSYl-}B{hHjV;+uPe86w^cr=!B{Kwzuu#?mm++HogUJn_=^-OORDB`4L{d zhlm+?^I(e6LDFL6<cxr%U25EU8Tv&Kaf1LBXrDqe9-2W?h7PMpsLUY=AnA~!e*ZFv z+9g0XU!m8Bn#c4hR6>~jU*YGZq&UQM@?c3}K+w9YQxyT>X9jvb;@NfG){EOwZb(ZS zpczmSlWy~Gkt!LgtzwA1KMVC0K@5KV`c*!G(;vVjx&p*g8_0u87cUxs>_Fx<5V321 zAQVMUwk08_1c!$c1LVmBZH-D6bPfTjKuSsZ2)rr`O^_?V0FYL+>%zjxNGF#0z;y}^ zy`HIhPnd;-KpsjO?}0Pb7D|cOHBa-@*`Uah>V40Lq0PDi3gI{KKBN~0ir<$2z+zv$ zx-jK7Pxy{b1pLKJ+vx@Z8s4D&eJ9WiU4U@j!2E@jEetJW2en*UcPaF||ELksi^_uy zYd(~DiJ=)d7p^!|K%tZzNM(ZzfPAWOBIskR>kNpnffr*3rLO+ORiSvH{Z1WAX--b^ z(TmAV6r2jRjTZI<SMSM#hGN2}WRE#VH=ak=Zq~@XA*Kgc3tTHy_53>_V~}LKr-##W zpd60S=x*pRGXY^YRPAU+ea||<+q>TLFwS$c`VrI!Ow-<nDRt0$#=vlyf`SPOvw%9V z>`FWjZKwqtGAE(>U|`sV#3>82#~6yQ-Ny4%MbV=fEGU;_p*qclR+hW^4LUxje>9%e zy5xI-Qjh<Y1|s7zD9DA^lzI_h!^*++7a=*v;nI6OMJ|hA37E>i8ctXX>lk2~)gte< z1Zq_%XXC&d4h`xs3KwA_7CR|AIoazku^M522BvaQ5Yi~4RbkBx5+ep4q$|m`ff@e5 zT>_lzhT7ExxC8f#ao*~kZ&w+4cvPgMM9)tgbG0iApl?j+p7rQ;c$6rBViQ&f!$BUg z1n}%JEI!6NcOD}Uk(BfpkXDw}NCEaiIm{HKq({gCf`BhW2PG7J;dOBOU8qWAK)@IS zU}a#4N=#&iuB8a5NjpFvFa8<PkB7P24Pxfj&6|>-Raoe~wh;iQ4-)GpNW=P|p#9Y! za;fgWsgI#q&lo`DE(nv*AVj&<K{!Be4Ax35$gu~&ek}^~&H7V!p&PRcxxfz%{aBa_ zU9T>YNl8J6nhcOdMtDcj(5TW4A{zn7Q7I`ed{3ZLa-n=guXf!^_ewmPr1qjlA@Lpm z2*l1^pNEwj1wF2h4Glkv81GKm{RggrAT!>ck&%(*b-a6-lJYiS;b=cU6jxF&<%=c2 z(J{aS#@qKW7XU$&nD)|CFUDyv^d#~Cs`=eRy|cR;1+^1}yDt>#H(?c)h1<gDLn{`z zszE^=Fea@AoI!1dk+6E8)h!D6!9=`%xn=C6^N2369N=)ziVXZU^M7JaBmf24UmNLf z=)*TLH3hXBlPv763~%r@RNhxeBcaIY01#)YZv)HzAmS1+F=fV}ra=l6oAJ<J3R4Lr zB%bgVLq)_hXifg7p)Upi`lx6FMAen)#>Svg%mNV4*mTp-i+a2et$2%_T>)-(5vUa9 zh8*-BE<(u`3k9mgAs-)~8Z;L&gMyESXcHXN0iw6foL3eKH1BFySM792?u~pz=bqW* z0bP|h{sCiN0L&4^E>Y>uXLm|m9OLSOJt7HjYr}U03Mv>{=r^o_K}1!pWGAifLDTa_ zAQ=iOmQ3)u0k*vaay|s`B}{t$;RJMwUW6*z1eW@N1x$Y!K1W!}J1~osA=x4oRRFPQ z2h>IoI%i-Net>p9q`-&J$74Syb#lsa9(ofZ?Li|JCgw#vyyl6CpKpHyivW-z;CXNt z&IxUY13;6|yG#9PmGsTM$pX<pdy4I5C6Er-Z*Q+Y0ylwL)ayQCBtZJ;OA+D$QvUaF zi?|8;GFuUf4?UrfjIJ{?2LpM6Wh>+6b|2|wuw^~p^MD?JzBDoYnR0-o2T-^=p`Q<B z9h5^hfS8-#)Q|?fp=T%bf=)ruFFd{kLFEnP1N#<2eog>N|Ec8^;6yuM4Gn$o*xPAd zQLsFROU)D!wqR$+2^gM$_I?ChA0VZSmKF_^U#M5FUn}J(T-%+_FNXDf&=^iD&<tuE z|2s_eM{tiRg3g#t-QA(k#}|4~$;(1^BRVR|@O;9u9J&Cx63<RCTI-BF5N7A#;NZWg zqO8oBC=bxe4yL<A<#X=D1W|9ne&Id&^u3WU&sqQ_B3;-JDw@GCXeCqR^#1FFfM29V z1|jNzWAQJjJ!cn2JLHyfF+~T19N2->KQ4Kv5tN4=V54_%i*Q5%mTOQ*2m=f_)WVSK z+5wGG0pADWkqw2<UA?*lsEkfdto2>320ma!O_-?zhyz-(z9t4&_18j5aHz!a^ZDQ$ zkl;5yu28yQ>VH-l1^5Kx*6?>VPhzS6Js(o`rNTwHz+jh<lth3qLaFOL{>(woEJXce zc;#f|<ZV%ma-Y=lD4?wZWdP9^k|wSn#3JDHK|!Ny0^Wal{r&w9w2&rm1pk|}p#U9W z|1FU%9r7;V9D$X*G5e3j1qm4VpZ;<Q)L%2;jg+3_A4B&o2?$(h#l9x!be)SU5=uGf zDAwMB<|B|n7XHLzumC$dJ8wV>JU|=hW$L*ZE-V273$Zi7AZTrEompAwgkaqZKQKHW zDY<n42WJQ>e0W<%#4n+f9Ucz@ai|Qe>;X!@4htgyX(Rvu4O9wzb{9gz5C9`XL&L{% zaxDg&qmnNG)*?Zxk{=j8yh3qHOBV1sLGTQS7DCgmo4b3l*Rcb-49<n-y9rEY9tLfs zt`HeM&}pnBs&IR^n3w=i8E!q@9L&H43{foB52+F^@DChQOqB;c1h)W0W9)nfsNr3} z(G64oj<PbmeW2qQiClO(#PCWUBEjecE?8zYvibgSBqb&|P98>N1@I3zcntUM#X)32 zI3CPjcC7=Xix}Sa?{8OaGvPG>aM(ja2`ErxI1vX4d?G~kF-X9Yg3vb)@$JUlyI~0I zfz=rm6Z0x3=Qb`8tpO-MVf1?G_d@+NSkeGH-T(a1092m|Y*DpP?^?#-WAIPuz4w)Z zH_>p-J}?X|V4}hlLE4(YT}kylbBFMP;Q1%85HAuFn{~x7BiISj#xA(eLx6lirJ7Gx zah;+MbCk1WCqY;tybD4lDHYY@P>OS(jTaDc5i<?p*;%sWbZ|w_?lxG|FoC$ej_fGd zb%MZUg7stw;0O_4NLM{14afCSE<om9KhW;LHFseG+JR<OsrPUO?*uxcXkj-+JDdk* zzas=AMBM7=(6Xw3K_LbkgiR8-1x-yVpumFA*N(_Rr;YJgh%ZLPn#(`(;=Q$DSAiK! z3B-<q*f9KAjTP?F%$5y6T`Z-(S0rd8#5;jW*xA)J2immNoHr^Wq6vmXccw^3K%-dj z8w@*BVg12-xFB3j)1U6F8`{ldJ32b#`ocp(qDDtYL$l3HOk^}Q6JVZEB=`VXyo8S* zosf`#Iqc}*a7$27i-CdR7AtGs<8s|lFwSJ*qX_*@NR5fPm7boCEy(pyYG`<vk(XB$ zY53RDaB?aGnY|h^p#y3R?6&!jE3lR?lak(mEWyF)y))Ma*3Hq!^*)n_($ZZZ67ZbN zwe!`3-@S{2{S!+cmju8tQh<Cm=m-`h^YiD=4|uHNhrz-+2DbuVzEmq8IU86C*xCI_ z{E9Fcj`M!yOU&Y2Yb@~c5mDT9bbkSYR{_+!+onML`QpV&*WCpfC`13T9f7aW`<`%t zEmZ94q9+5*=sc!92JnK}&(}U|)$I`hLj#`#JX0b_xJMw;@Hgrjo=!asbU)Q<n+5t( z<zVj~Le8ySVVw@CD89a__XncUArR{Fp2D=B9W&ECWHp87HG>2QPlD;|>s#Zxh2X?w z-8e~@#l<fs?g7J<1T$g<_&;nu7^x=$n56V2Lg_QFV_*a`nOS^{a$m;3bR%^E7=kAD zm;<3KW@k1i=5YdM=U8QR!@2{*Dl01mn3%qencP`)fGs{q$5ZugR3GrCd1zZVgEi?v zq@9az{i{X|7Tw|f+IOt>i+gMCf;c!hHn8!eD!j0;u+&Pai!q_s6pT+cpx1{6=yFnx zMT(Km6Z2xtpZlA?gxJn2HDg)wdQ5pBP|3<ED(VMY;S!(<PO&3(5)%`XAP>^i0VBTA z(9l5sS)<T~ljd#KS;2wqB7`SkT&3XbgLj~>rK<W}1NuiUHT@9sF%ziu%X7{F$YBo6 z>mtI!4t8_ifzAW79z_3&l>oqG$?{h&M=QxzbN%PtQ~y0Jx<Y*Y5dsG8#QK*ajXjSp zp`!Qz7rYy~eE%ad$}vPuihb-72q(a@7cVNsSCWlY;K0mF(S4=-*{Oy&2*>jNKagPV zZ9X;`;bba*aouni+(a$v1Kt2;6&V%g3H<<0(YfY(fI0;<;`G*Lmuaw^{bc>le4ebt zU$}+k`&=mWJKg=+?;*gg$&RQVqtsF1n3Whm2K<y{$yMa8<GEep6s)GmQe`%}(Cp{` zAAI6Tvx{3MbRO-9eJCvrh3#Z|!J`V=EU;~C0fl2mnI6MZzhC+GhuWaF$UnQt{$m_g z9l9{iG3wa4rdeHms@K#b9fY8H&f$wVmi6@KYweBC=VOKFQiP^gV^*Py{iRq0PICK! z$$gAI6m|9RX^O^s<<Dhq*hL2T3G0*R+73CaS*pdy&~-H%#h;1Eg!Co_xQyn0ydS<% zU{;Aq?&oAcTGW>KkU2{>{vVzNw=(w+%;JFiAvqJj;2|x;oU>xgop=v@bd<~o><Z5% zW_6|+HVhX1JvzFTB%j?Y-rlUQmzA9J5?YCF!_K~3=r@T63krNbibYWfu8iE=%5iMk zab#3f39CT4NGlX<vXc8pMqD3O+Wfi;`L*xt%csj=4Q2teh@1SN)RuXO&D`++zfEv> z%{T<mC0h!QnZhtcS4)F<28hkjI-&;q-E9Aag@th-+X;t>rMZsb0YHF!@`VVUq8q~U z96|;!RFq?e#TV|^Qn*6rqdZQ)*Y~lpi%JE__weMd=KYX;jYsYvC8a-UI1s@sN;)W{ z9H)*GfU)nWeAh$lKkWhh?`JV=7j#yGW`LmZ@Wwd@AbDUfVUm!Ozy9%AtXEb2pAN9W zNAz5v;DPM&(0e=W{33x6UbU18(5I-r9=fIbp|Ww)T~qzcYrp$Tv7iv*=cBWeJw<T1 z@$tzr2Hgg^p=&(Cb;`Y6lnx>-WW%_go}M*zGnub0OTuad<5A2l4QGF?AjtMX`vD?c zQ4Gyc`gTD=v((>&iu*YfT7DZzo!|mv%s{sW0}oF$gnDl9Y#=WK_W?`yJph+6dVR=Z zARpQR3>D5#F>9pXP?4^O(itknK!nJ+Z5se`X$30<a3`<LWC8{j4p<ef>5_p|0uHxe zqaiD*@$7G&NCaea$!dqiW$=2SzHK+>MIZ&TNl#q-YwgWR_oWmdavL>UISvg-&!(;~ zbd5l6McgY$<^)KW4eWE9^T(z8px@}TK3*PrezGJ6m1Gn)KE(i3h@n5D#f*0m9};r0 zewj50ykjtnu2rgZnOJnhJb+mMk#n>DP!C-O{-ft3wkA-s(m;zWe%0^+Jmc7{86F<a z)hwQSI9y*>_gGPpLvOZ=4f$QBXsAkwp-IlK(RWrXeIG=Oa-p6G0u|3r(bXFj6YN?{ zhkC5oAU2~C65Iy&0X2c?7NTG0AT)kk1xKwwClU0<UKeX&M07OPC{a2hS&^-pK!O$k z&>&4-0Q?eSVq!vnl97;zdw3u_Vxen%q~IE~5(qhbeOm{d|1nf+*Qzui%=C$NdprTx zBonqbDnl`3F;b8Mey$1>wT$qak&xy)pte}=2Hd3u`!Vk@Fu(vtL0#J_YP-A83E24- z(o#(;05cGp)MDVZOvEEy@~}NHI4Gz}!BfLwKxMQG`##$7DN<27K%RyziX6qxJ|G2l zp|5449d;ft2jCL37yr>$H1D{QbkG@N@8jbG3o~|EFpz+{8Cvj=s%>g&3Uoa5Kh$0R z^~QnTBVb{!`xI9(R@&p{&fG9!&HS%zw*R`s<hiH`-q6;sX|EWUYjIH#2>h?zU;l4L z#SMD*?%Tu!dTvYQQy#Qc%R^}Jgq?ApDZ64MpwosNcBY9SzqayFx^0AulhdvnY7TGM z@SK8dN(O{DGh-m<<#i&`q-QoV-&e+{;w7eaH;xYom4e)pC(bTxpD)1V`G4#g!URKP z<Faf3th`>RPzk4C{*pj%CLVxzeyX?MF!}wJ3wkZfFJHUX3v@6GrmO|n)cU~ZZ=*Jj z)Yumlz!w-a=VkJAb6YS#;}jCvaJs>l@VmF*Cm81XN8Nb^ZRc2K^52{>f{CMGV>wFZ Mp@LL_gu%=I0mPN)82|tP literal 0 HcmV?d00001 diff --git a/docs/internals/event_loop_worker.png b/docs/internals/event_loop_worker.png new file mode 100644 index 0000000000000000000000000000000000000000..1bec1ea598b2806fea24ab25dc825c9055c76ee6 GIT binary patch literal 29696 zcmb5W1yoht+b_BZ0RfSa5TrvIX=#vdq`N_B5$P5wkrt5d?(PzhmhLX;lI}CN|L;5Z z+%xXI;~vNGm3OVV=brQF_&o$GDM&s+B}9cF=!vwHm@)*x`$7<0Jqi+davXKA0scXB zcq6Tf0{(cQm<ECWfBZp8%K@yPg8hYiL!0OV9zOphuK7vD&is?J(ML1L+1Z)d!q&>c z#OQ+=v)#vci3j|I5JUz^i;1YZB<(FYYpT2>LOSAun-R^C!$pQyQ<``8!F8_IzOY>Q ziuSfh`@<0L4!v@5<AnAuUa@-XYD!&KaFF#3QIv=Gdu3JC@h?>C_j(%4gT1{xib#qF zP4C2f!aulcC}zTYda4Mhu%mgV*F65B-9Qm6wrdhAB1WaC4eyKMA2t7C{T&&%Af2Kz z^I+z}B%|4sn)YV}B#YfFs)U_g0`1*xA_CdI$7lV))t@>k%7tgDu_9L=SAU6uPp36d zBKzGt*!#PucSxp(yHW<tl34XJ{l=?IT8c{KCto0C@sm(`)e%;Dy9_HOkqi^l_`+P1 z2$m5xI>y{$>e!(UGKy=LNB$J)!rx2uexw#hC&y>S#Sydg{Faj+q6>0U9V?W`5XMxc zaaSEK^*fU*E=op!GQgmnpp824#OCj(>L-7nRGp&vgIn3y*aTNtQBia;Ffu+15_i>} z$#}8PZ@NpB+qF%TO)dQw6+5*h$QA23hpEaz<H5}$rpub40ej|OTuZ`uRaUmjyo3@P z&6Q*{3JPW1dW*-`&*Q^G*_27%nP>l6bULh}6rbVjl~q!T7_QRSHe+X>v&1l0>*A77 z8?CxDOA7jmiTm}7eSEll<|@hiW?HERp{suCNO<4nVwQle0O4T7&p{@l3`7YU_SIFT z(IxK;lQS=)f`ir>xCJK3IR7NrTlSssF$)R_;i9BXW}0;P1;#lzR@`RD`0sG34O+Gu zUQ%XUV1e(rSnVh_#wP7vbi?zux(hmcByQ);fGFd1HtzObYq4v5wAt9TF+h5LURG?U zC{2mV<9P3zql{(em`*Uuf!E0(DTUe8DcxtVKSPnZT5CmhV0xy%Y(>Q`ImOxWUYEeh zk1W3DaR=SHbcBfh;_srQ$Xra#RxE9&hRd65{KSxiDU_Wzs;WEjFyvPmrY9Ya$~a1& zKYK=Xd(^B7BhRb3-oI#I)%CM5Jmy|priD-|UiCb>NGVeUpHj{a7!?vV>(n`F@?um= zUO3a!WK^W<^4J8|%O{H!qLMh>C3L^0YfK5x)?q3BXx2|ygYF}Cns3ciKs`6BTBGGQ zP+RdSf$_=W>*=Y1qR&=qs0uc(;ibgPqJ^oxwcLcODeuKpIyUUft2yydW1a3AHh)#q z)8@*XXVw~6loLx8(WJ%J_i!2Km-%q$EmF+Af$AHV>&r@Ske-w6x&H|qoW7Du+gO(G zLQb?_hLbAlX)1**?v_rBb46tlvoS=KF+@E%i9I)?im$b<ql%<&gVEyfdUUoYD<d++ z7dIT9CWB3WDC>>7tBT}Fpro5NePis?$Gp$8Y|IT+%0Iu2;V_$eBW5Q0D3jizVWWut zQ8M}{9=uOUq$@6E2>o^+^S7Wp4g#IY9!vsWbqyzC(q}{}i}T}z{J7{?y&<tRv2{)P z*}^Y3%JYxTOHw3|B+O~9`LQ^#EKmdx3Sa1XvPbVqA`^fwy<p{}vr}6f<~5jc#C$tr zY-<^vn{8st#5g>~J;9mRM(G<Yl}SLLEux`hSg^;tD5<#ArYSk+Lmm;dE`yA8MRUCo z>hG2lfX6El`{EH5MX0f%ak9$qp3yfmj!og}Jn3FMOm|Q3QTVY&x?c^CvOE*kQOslE zcr`LMvdJw^F-kkjdwmu4#uLYE>R1pvSXe_tR)fiV<x7xioRX{t)o|?AV>D`b0UH4v zuQ9MY1<WQO`=(Uvx{yl4eVLu*w->_djjO4t0Twu2DK9(w(c^G_N{T^c1sz$&Nxb(A zjTjUZW>{~i)I)G`u(5iHZ%DB$URB?aJ5qo@*B<D;V&D&9&`n2P8M(<Hw}9kE9$Hqq z8A^8cekH~~ISh{E&rqo)BqZY1W6im>xk{>CI})C2#OUE*VZD8e?n_TAxZ>{=5@zzw z+-$S<F*^FV-BOEywT(@Bc6RpOXx<kwibygVnqQYak(Bc0g9#hk%B@4k55(F6rJt)G zg?GNBponS1%=@hU>BWg?myFun-a@0|(6Car_r%vl5m{NZu&^)!L65pLu<f;8!zYWm ziZ#vo#?v`qfF*Wvgl8UMzJ1>139mK~!RjDTS%Id_4#g6K^G>&Vx9Ur%r(6u%pX|^! zl7QR#kSlF`f!*)bs|dKVZ>(%=Zww977W2ork?9-p{%Zjh4Ym8_<>e6+!Pe8=Eg5l2 zQD0NBOGZHfPXMo`U~V3po}8Rq(NFfMthQECLLw*ff4&UUWW+n7U9OM@r{*7giiMSr z8Xyd#@{`k@XRQ2>;NEVpED>R0(UD5R2L}ZOabSv)VPRoKA{nDlfmNfz|IG^^TR}lm z2&{<M;I#!=fG~ae>FFmbDk@r=yYlk#ppX#vLw@6Of^bry05J->>;kZS97mKaW;&Kh z=atv33nZ+rj>ly^twg^oFDLh<u#jQ4)-Ie{Hnz!mv8aeqwOAYX>gp<685}0EVN~4- zE=#b%Y5yzuG(4oNto+J+fWYHo!5P)yu*LI6DO2JJV2Z}dF6@xcUOXOug&mAos8RZB z%D_7`UnO^SAf6rN@nblMikjN!d~c?7u+^voNw?LD7{Vkb4%Ml(g@e>f^+SLDe1nLH zNJd5m4W$ZsTj^O?SfD(8`cUNdaJ5OgH(ibpP9pF|SQui~t@GI&N?i*i^%A}~SYZ9I zEWEeW`jC9_^y$;?GNX>+Y*~DCV%}#OCAyIdPW!W*-$(=q1>Da&6@~5)Bc)EYssY{d zal#NSe`r;juPnFwX*RjAQczNk<SLS?mg*CC1UwP__>oP$L^qh!>r#a4;de5hk)a_3 z_Sdcd_~_m5B#~4>&*3x?1Qh=_F7ECKker;{qW08G<-3tW4VJ9jTm;B=p}}7?5Tm)W zW~C#rV_;yf+Co7=Ay2=Vx2n2&Wq<#PySqET=k+r#F0S`eCB&<1Ypj4V?=39SO}fKM z7r42(@ona6x}vCMl`75UVJiVSDrQ@*{tykM3Xy`nV23_fY7L(#)bO6gk<B8zjTE6! zeik77gd$R^%(x5ng^<^avz-Y9XdZAWT4Hw*-S;rz-BOEZ#V+8kl(e+#=51GKr?jY{ zzJ7|e9tH-+`_Zh&QBhHbR#qYHe#j<(xOV<+-@iXwUtdqN7|Q{tZ`fl+XiMU8z$73D z@<l*~hb$+H9_#4nFzVEx^YZdiyJN#-jX*RFNm@=$Ok109eqq6E4E#sPJ$k=UQ&Tg| zad)y}m(yzUOLX)Y-zGOFC)3EsQiIkoue)mz0|Qb(4=Fi04=Y402p9=gi%}{UH@BZN zGtYr;w`WO58(Ld`o2#{3`xQ#0S@$8``u_80cri)IOqm#Z(qmenHsd%KkH>*Up}|D; zmy(x{k!QH}CMG6+_N=MpSh|7kLLLMYnNtryHq$L93J#mZBqU<r^DKEk2uOqy`pd>K zueA9hY%Mfi@F_lKcR#m=3GHH&>ublIXTQEZM}S^&b4z~s@QOshjjAz7i~@K%ictb| zUqUfO#ivX<HDcf$M8?NQq*G%ZxnTS59de}97ptifIEdMx<r~lnyhcU`2Zv|iJQq#A zKtq51-RL))azESg2UK>}mmzGcsH{vczWa#D$Her6h})LYsbzdzwX>^BL_-5#P*4z@ zjoaA71P)qgbnYI_mW_;$r%QY|SZJiy76LX`t8K8I(0qTkdTYL(9cUZnoPh!_d~Rl@ z-y<}<G_(HL4<A2%PEID?nJP8d`*?bGhJ}ggGh1y%2_!C8A=%A_ogD%epX}8uOq`br z-*-(wcrFZtrrh<!JS!_JO*{n8<@$76=cMb`uMEJ<vhs3YUthS@)zvb`U3I8!JYNNi zlr)@Ptso81Hz+(ju(Fc<ji~5`Hp9p!mTzDHW~6ZtnG5!xLzQ~<I5D%hH!P3)xLb?1 z3>qV(sDi$J{Tvg6i$TI4w7hJX^xFQ7lRJ>2(a}+*a$|A-$zvWs9V{XuG)c+s3z8mk zybvrL9E5UBbUr@58I%lFFE7Cq0e2)>*?#Q9=t~|%csQzjqccJNmoWi7i8sYb2`q1C z)*nue$x^Rv1^0%;sGQ>tzG&N1DIq*YgjX$4-A=AeOh|~8A-jHogoKo*UA=?6iG%9@ z;ll@)&3%}FYh%D@+~En&nbg?G7x08AXO9hteBO7kM@M~KU0YX|_>37QSVT95gxXFV z9xQ|Nu&64TUV%nNp-gCtLG%A?$_oh6m1wbo#rl~2o(WT7i_ku!Kc|WSjG*qs%rpM> z4ZYIX*!a615}oj;PaNMuL)nD*UjpmA%8FP(z#v<BG(4;nh8H7}o}P})5b}5GyB;#| z4Zv}J)aJT31_7m#t%O8Cg)BKwUr&!P?0<Lhs<{~cr&qJqT=SmjDVq;2(Q8~cH~-jg zq!kev>C3FgyO|KbMXJ;K^RNLH+y3)%qpg8nNkllrSGARu4hkPWlu^;sqw;HtH3NMR z6Bj3c{aSN%Z!Zal=;ZXh=2pd3Ur7k?S6_vQazE2+CgLwPR@qMu+sj_CXz_b519`E0 zXKG3q+-2-rVU-z$h3b!OVv>mtZn2W=xnuhW1vtTc!T@31^M1wXQaIW%WNeha+ImLm zyIc_F0LMGZG5<miP%6up;Qz>Pro4;@lFG)VZG<T3Ghbs%{jXaL7Y?@1&kOf;UXTJ- zTC7R^-#sxu?~foU@!+;(>L4jrvE~1~mH2=0)6M*^zPOV_vmM__iixT2{BPfcIfNr6 zmep6+h=@gp@_*|1fh<ux>BN-ErV$WOoX&9<4h$Sx{q(i*Zl&^dGS>nEZ&vQ+v8Obk zjp1!swz7$F27`i>oc)L5hy6&T#TL&jqa)Z#*pIrq-tJW7tX=J$Vj4wz+Q5?b@!AL@ zrbMO4^fz>6I#*p=?xgi|lYDc?7czYf)lEKdsWRg2ia@=vw?}tU*^*g%@PcWWx9(&V zU9{^`WHBXwLKoPqR2`;iFi}95H%;J^r$oi(D@<1McF;)Y<d<!HUr8j844hYp%K2#e zJB_$L>l13l8V@q+Oa%3M($ui<<^>3E99bk0_MK7uf9gV*sE=GZ3uYhwO#GzO135HU z0ir17uTzo;k5<*tByCkJMF1lS1IQ!3zu_fB5&roWSm=z=ZcHf|oThQ-LR+5wFUzE! zwX`tVcXWREjo@iECA{Mg)|+>PnQ)RlBea>eFAAuzBBxM(=Rm_!#Md-;*#Y~Za8Fyx z^A%#0GWxmRTj!zn6Nb%!c&ob2VgDFEPmMBrEG*B@Jkj`*cT16#4DW_gY;Ntg>SNG9 zt16@Ws_q0R=xV(iN=IWjKbXtiJUk5V2z<JG=Xm+E&(|b&Lc_^MOl-$=@q4!8{S$#& zV}nSq>%$QThIfjZZC)~@FdkcOb;dY4?ejg1EjGKK%~e?-K<_7uo?{U67S*zxj($WE zp~#4e=QguQRtfA_+{q1mN<4|iXYa-1N9KtFqQ8Vy__Sw!WLR%|aq*#S0>u0*0{DE@ zCKgDB6$Km1bNc4r>lj2)GCE!9|12G8jG=<P9t{N7e&Z<zeK*FWqHvq*Miiz)5^*C1 zl>edl*hd-tvzXY!)pv-ti%jb9C#DqYF%O*5SCNRIt&@Kn;Y#YvJW`sW42Ni`pL{0f zbJ7(2Fi;V!Gbz_VmCF`9{>B%W$82rJ!2i7Gb<!kxZPecRoFS(?)$(;1Vw7$#J0xQ- z77@e|$&ZlsodYFBz~1XhHw?v+KR`H&Nc`V<C`PHiqQDMA)iu~oh5r@ZM|qw@PAow< z5>BaLD`%yizQ)4kY~Bz*I7>-s-iG9Ceod#!E=6~iGe!;F_jE>v@*gSQYN->#Daok$ zEw?XW$VEyiVL%@%913E`=;HE<i<2=RMaqntn)ge$KRz-r^6C2vAGpb-ns0ShXX8QH z`cFU0zO>b1*QX}L#~RtB9fWAJzlyv%J#V+mez^E}KBG}9y$}<^J$RG5#UJQQl5c2g ztg1Bnwml@5B|eKfce`qTHJR$rZQ{8k{B|@QI8ieRq6Qm^^7CIT+X;rp-_0Bqhy@F) zysix}llbg^ea)wzR1}9$F*$6FI>sc26ehWc-D9JZcMaE+R#KB&_r8X^{cY;?E#$k) zcBswUbFjS*k8Z5GndLj)O6@LIM=DN^PBnN>jfsNll~hB~ZRcC+J;!WAfYgp4x`jgr zsD9^M))TXV&|S3tdr&MZ_WN@v`<_p8ee`cV4qm>s&|v?dLxEc9pT#w-o7Nd8iuKK> zgRO3h*N}r%h*dY`x%b^gvu87t4I^`LVy%(2#drY*uz;lUxbSJ;D<lz<cfyhqQc|?? z>*fknD+D~ZFQbUFwVq<-<EsHymWHD+E4O2b2sb7<{k)AJy=UredR!j7F&Z_N$fH^# zv@7&Jha1}3+A{r>(|JQ%BXiasDSS;^@C?fS&O<SM)qjob*|cNx_qd=|nks5*dv}{g zrqtO+<51UP@{QPzcnmm1K{$$~l~r!6Ar~ApkJf6(7%5d16|Kx5=zTqZb~UzHQqNPL zQxnD&_)Irv7MS{&AM{AVbY`I__f^Y{bE6MBjTi@NElndQ`$CF!F2~mhe!i3%7`=3~ z{57tdy@CPP4ZU8JjR*Ix4qI^=ip+Bq&59jPDsGOu<CnTVcD1z7kIOrW+Sah7!-PZw z!<}_Q$Ft5wp2jDqj$PTunn8J|dB)Ip<v=VEIs}uBQxNW|Y@Z2n?b(B9X%-sJ@EkK` zvME*#52(JsxQmk3h|$R2Sp}yIqDK~{Q=64M>Kzli3kn=UerVe6`O;8ba68v@jQ_4= z$<x^JV&%kEe6tbfxJczPRPr2p)#)6w*&r_&tLjiuzgS+pS#qPen&4XX<FyD+xKGeC z@fe_cfL=15GtFmbNdqsYA`RsCSRKZb%yrc7{vgAd8YH#m?N|KiY>5`Zxkmb=$?xcD zv!s8Y+-f7_7QtqvP7U--WCs(<R;5FM$NWY(YIN0Nvy^&mcC}?in&5gEiPk|l<*eoG zF|^k~q<*P?XEu|U)~&ea26JSG-}AmH0TYppDpIDb)-9B5#Uu2{XQyPsm_eh|C2ecp zrm;%eaw>EG@^aIj#JObA>SUs=q>x_CaeiHCLXZfWysh6W<x1B1?L~4mPq}w^a))h< zQ`qR*;8jLyg=&kDbD^9Iw%_v;|B9*j{N!vc>UUC1jLybDwtMCovD=eqCSeM?+M6Sb zemi$}YT6tz3VNm;n5;EhxU5B1S1i1{4&A|;uu~1s<KA?w&ey1^zi5=co5E}M1X{;} zP*>E$JAo@nu$4T{3d!XE3`!G*e*Czt6-sSMQScR`VAlAQ@+NXx>Dqh$<LtJ9vn2=L zOSg9#+u~7|8lE82ILi$Kh6z^!QJNK)Zq1z+fR5xScoUO_wsImZJsz${UdRx^Yjd-^ z`FRh??fn?yA_WpYSDWQCo$%A_u|Lc++gabVgQdL2d0>_<RvvI8w&FkBsIgi}IZ*F7 z)_7<eoI__gzrWiOnsOuD+7*miZ1LXwW`G`wp0Qh<+%VLczf|gmA#hQ8yAf2%W<`@R zq=E;t@`f{%F4$bZb1P(b#mw;E6r?jgYRgba6$}j{=393<%P%}l{lRy9_#`@)HQ4QN zcHrhztLb-22YwOJWPapmmPlyPXo+I9*dm|B#`tZn5-4~vTZIiB*_SRR-D%0kWJ8+c z+he&S=_Y~mH2P<1*QczfUfd*KwCt%OO#3V$L{%%X;X;4+>znhF!gWn@@6{mY5m5Eq zDJ((~_M!Z2_E?k*j&=?ZFy=EER*at6e`xRo-8T#CKhElzG!E<@Y1bww;~x?LUkMO{ zrLZ{YV6-aW28ne}@P8JrOXDE?TWE|ju9N_{09#1H@o&R|mtbKhbV&-_@%0OIp9dpM zRUvd=OGdtzhnWUKNNu&i+rHI5=)O?-_j5s<hTS#1{`k>o{t|y`!k}khU~;2{YSdMr zf^K#N+$&B_+~D9ozLy8PA|JFo&01@kzEApZW#=y;U87s`8D9>f)rQKjwE}Hd+sZ}% zyyHvhYiWJght*9<(s^omrndaHmL^&P?0NC~59g=VArL_{W<@<ggC<K2!hL)oNEqZ- zTsE_(v-{m)#G;K(`%2Z8YM`nCOR+zeuC$U7w0JE4P_xz*BQ)JvecYW%BCyMuQ@b>B zuP-Xu%7R|-5s*GNl?|mpFhH17eQ<=gRe>oNQ;ICe<+RDWuoOQ089tJl00b{ws^7f4 zxrq#cq_Dd)7{{RT?)q%^=;ERquyy6npMd+j8-^?lK(|5DUBhUDyAVtA2w}{#*9{NT zi_!nROTmQVb3`>F(X@oQa{E=1bfpLeJlRj3#=%~lnVWaop13YmZU{@kFYlYploQ~T z81YQHzT$3;WFqG%riOVx+-aOY;h>>GLc?QpvAGBN>k26DKt4S^QZo-zMOuDINiY7Z z2o)S049*r<Tgwd!5i1)T{-6-zFYxbp9PSGj?qA**m54EpV1;yDzYWj43EkdzuF?o0 z$6-mq;HDddG-I#n{L*$tnL}e3b*oEP&xI7+ySrYxo0@N^G1D?~;01TmWS84Mp@2`L z1{Z)^UWvYJA$A|S2vB;@@$UD$^3^XwkeQF=Dw^EiUV#)-7}T7Wx3_~qWrmWR91heJ zx~eS3Oiw`Jli110>4Sqqhd&xV$UJ>PZ6XbnQ>N?eDVv&`;UJL7R^0HrpQGc@$hFPa zJK$bxCU9B>JZCrG@qD<eiqUcQ-=_U!h6oLhV&}w-P5yqxHtuQs&dMuuXlgJF`O+y= z*-5WY4MqI0vCeZL<!kpn?{@O{i1o<CWDTv(d-SHA+$fDZjz1Q1OZk4e?pEcgUQwWw zYxJkk%k8w(ds;ChJ)2r?YIO4VF#E8(v4xHOTEA^^_yy5~Ui)3CQaSmDwZoUi2<H1g zq7o9e^?Tdds-!(PTeUfyJXIUjV>n9jILC17ZRL-Tv_6Mf-Px!6mnPIG(e@zt(8m-x zKQw*n3|P8I$!B|azNZE_s`h@3o|d+4XQI&l>aST{eSNxGp*pA}N`CruD#=nP;)gUm zIT_1YoUyw#nvDjEI)30$Z{ex2zrVlH@1z+X&j#-JDznf-YX6dN5CoGXY_F5@?bgi$ z^=4eG{i}}KC2y^kbHmH1wia3BbKB&!z3aXNO79o7`i;v|bnzDP2b1TM=?7=?_XQJW ze=;pg)J6$8PKJB#Ul0+Oj^%b8DS<u?JoH)PZhUgYymslfoJfHfrI0Q0?liM`FW2E? z6K`!2he;RJ`IxaTzvsgG%0hPzH_Lp%^`+PMO%_@1hYAK$#mgQGzqp#i?c%cg+%nA9 zAOE^tgz_mcJ2^N_Umf1hjOi83{h_neODH<5?Qaa}!WIC!@>Ezgv*6CIT?k`ow5eQ{ z26Cy+dydg|QCOm<vA<BOXcW48b+Spz$OwwUh1Lm;jW;hji=FG;FPK4DC7?I@Wxq{) z*JCs^QD7@Rwpbbj!j(O9t<H--n8+}1So*;{1SK_GkU7Z@9mbv!@mk#dIf)~c(`)=x zcYC2fr6ikM>L_%2Q{IJJ$akPyYt{QH(9&sd2=XGRu4~j}oWwzSbOnzh7N)l6oyfc7 zEL$948xIYq#Y?`1mOx>mWVYK&n}X%WqC1IZ@$5OYlqtd<eu^~6KgwD(#}4IYkp3!) z7jzS8?HW_JYa#aX-dT6e#2tNjR-w53Ye`YtZr=W*{Y6-)xqqtmUj~FUV}6g=G>3ij z^;all`ELikcJrGnXP0;yt&2Gz8GlA1ti?hd(WTM`r@^buRPSfXvR8jM-YO~ifD-vk z6{yM0JUD*(q@;9oZC-qOx{6GM2rp7T=IH1+k}3I=NA<=Q-KU@<yTN90&^yk0Z&k(n zfCaE;_V10fmD!Xk-ixcXy0i49a3<T8F_+0d!)mQ;P|GtQAScV`9Y&bK%H>}>CMQ~1 ztXuax8>&YkT28Z(>7&WygexaAp*u39HpNY)L1w#?ay6pFS{sgw;ol}9VYl|W<adhX z<gW}T!{}^(WAJ%DeeSJ*)t?&|5dQSM<n4mfChU?6JeAf9i7TbF;>hiw{+Mah5rBe$ zfdrYFni|g4*i_7cifLqQY(S&)Av^?JIMbZ{l{_9EUeCaQNAWhCPfE(7)zY_(P&g&d z{l(YYa-o(^jd$<%mG&b112kbe0Ap}cwK@Hb16D}R?Ydrh!QF4pRu;=za2H3rfYyv> z#EAV0LT1pdZspXYq<qaTuZtlxF1oCTUI9dA&H7qpOH1?OB~hCFh>MDCN90<?C(XDu z)^|5Q4c>mm_)JbtuG#Fy$uA)Aw$;mXy!2to*YMXb9=VkA|HcCRLP19t%ajORURm)4 zjg2(W2k2t%o}Yiw($ez5(b2!Yo`-}VzzbYXx3J8IQkX6p;E;sXF5Hi@_UZEDtK&*$ zHQvrt=T+KRIjQ%_^%rs4opGFRax>0eif+ZWPL+5YTr$Xvh$!9N&>^HX-ImcCy1Qvh z?C<ZM_cT8|PWcjglD{`9XPb+6@qXjM#!21w3OP=lDSC=cx+Op!wWRr^6b*Go61XQy zP&=EPoSd$<!oR;gQGD_GgRqmkr6oNZ8{6{M7VXrT7Y35>nNoF1X=-`oV#tw-ELX2p zeB_0K4iD!8ovMC+bZtyGf`_GAoPjsVb@pLxeSQ5#?H$q2d&ZYS3VM1s-pGm!rlcZg z`?<spgsxMm9Dr-K*R=$Ha_#aiDAs!l<o{~_8AQ%4UKhl^<fo~xNet=0b!CZLmyD$R z{3abUhC#z?p94`mhQ&bt5fL6j8t;cfBkz4z>x#gNU8Q%@kEN>__%w{YV<}t3O*(Yw z--{^;z=%SjT43wV#Cq;Of0H^dYN+m{u))P-Dfs>3qjZbRnqg#v<1YQ}I6|r)oKka@ z<?1;ZJHwFljALHq{9aREio8|1P=7pyU)p!KQxzm(@%)9JH4$5<lK*Tbl0My8X+FUL zJEr1@NSS)EHmcC|o(d~ekUCpkkcImU&+=OOn~<!LQF?Htj^p{eM@pYCRqwp|pggG< zy~3>EXTuNcB=F)C<su8SY84yJy2{^139wW-(S3IE?L9X^+HZZj!n3Uhpc*2M;RSm` zsLF(tp9=!u%0AUiIlPs6Q0b&I{(^MPYy5~Y?wMHhXEKDeYqtrFOZUxG$-Khc{`(@~ zglCA)A=#nAR-O6_Aud(;wC5rba1lW=el&<8NNp9q<REVSh-nI%;iIateq;!SGwqmC zrF0?xR((%rM1rjXg{ej5Sg<`A7M+R~>SuHxv!|kDxP^>h4*()~WJZp{2K|fV6?R^L z?+ccS$#VdkGB#Ej+^djncJ;ZtW-*fYw>&`>YcX)#FNjzoRR8h-pVpnlWF?`$dQ)Ds zvMasQYd13e0cz1}bW-cFd%^$@4)Y77(rFQ99Y5V23v=-(KbIo6jaX}nWhswah=Km- zo(W18_S-<R9!N@TN0=B!0CPj|NM++ckHliYmD2uzN8uTu6a+3l9!U?bBs>TTM5OW{ z2nScnXIFz=DFId=if}{K;$nQ^qa03X4p9xZRdRoK^a_#qf?%9Y!g*@|74ccjq8g;A ziJ0|XIIc*Md9s;>fxBbB)mp86uj<mTApf0>;?fOzG2~SsLIA&ANlW>k*ID1diSlM( zk7CUK(YvCE100UbA`^3d<BOn9PlnM7CY=+hh!%7^YBCy&o5^zQnUlVL)qvk;uO;>o z?En3zx7HW92nf`oUidQ-<1H437)u};p!pR3gH0q-NCMhRe<hDtm%;jXS&m2w*bQ;J zFr_}e9QsF+1i2L0t3^=ELnROY_lPPL_D`1)3YGs=(!VFCsY8zeMNXMPUZRwW4dZCg z;yK;F4+aVmNuuIcRILlKc0*F9NGU%<z8)p&{2l9V!h_<zZCX_K!Sl2Rym*|Y{5A>; zCLbA3PZDodu-a?k$=<?p#_=W=7)`KZ=5Ju9i+Kj>f(5JpSTWMgzzp7vb_0$n<5zsr zbh$WpeEDgAmik?96!q40x~Q#lY6A`dIFDf1()`;M`Pqro;SDvR{bBHGjb&Hf5Xy=g zM)KzW@!fE&k6Z&9`EuBrT2o+9`)kcWfAc0l=CcrqoT_SDhP?*fYVPADZy_7C#1f`o z$c%)}02UPKwmj0F#gV9v(OhvAKE?}vFzcXjL-rESjXKidIOr=EQrx*@B6RjIwV$Cr zoQh5K`~TA>Ki)C~7=mAGe%mHTXH0YJH`gDj{`Us@f7j9f*A~G4XZKx(k%?Mw_u%X% zps&PFkkp&?3&&`U&~L4Yd^Nf?H2g+g7(6A55$)yu_|f0iDJ0xva|;vKt4&c^u9jUv zBD{m6*N$6qZTC7lt#3X2v!npC&-X0DQ&ZzKG+VZ+)Etm+hc9*|qz@a0rl0HJ)mZOT zm@WM}eo*o77ZWSk7B{X?)3^22t8{v@DVwXsr%BCFNkcCH+a*ew&8ktoa!zk-@ILla zLazG;y4fUX05E@fBXT}&)`B0RFjr%#5u<|q``XlOeR_9io66nMqQ<$cQ$PG`<HIOj z;jSK^b7OOzQc<a05$Tw>?$FZQC(8t?=KVkZjW(Ue<zbcX)hg^$m&web-D1=om{MkM z&MdjG<}^aTbC%=NZEHW_9oQizj@DHH4T9-8ZQZ5TUgx;x;x3)m7SE&UiN9VBPG^b= zy#)Y-^5eum#AJG@RT1Nn4`r@x{PkT~@lwcmQQt=BlT2^)TXWf576BF6lX8HIhyq}S zTIVrx_wBO=dCS&1&+V4V;Cc(|vjjC3q400^*ZZ5zyT$uVH#IlE7+SVGYfBf`PKWN< zPb|6bS0!bw`m=fhhYGs*U2WmXdRx7fa)XWn{j4UPwMvUsm>)(TN<yC!yzX{ciH(<( zxf4+?*Y|I$6~q^cZi<sMo7-$P$Vj2)xgMN7K!sxeNX!t=x=@*#Z^m9&9yA!p(mo@8 z?Pa2LQi?Y%)KX@6?0tcaK-*Ru?VW6!SDH9ptCMLErNaI#Mf~L)^r+5_vM=o>0#~O{ zz2HNcRU_}*kiOgaQDUUf!{6nBB*F8VKOWHxPg30|T>FL?K1`Rb*H(}2hVuy|RvY$z zQ-bQ;?^=vHtC5G+QeRF8RPCn5?zbGzs-3ok9|+phYAEozwtcbtUFY`t9b6rK^ZxIR zpmO5tnq>Yzfp&IqZ5h&q6+<a?>u3P)BL0CYqVG$hbEf-(C*~LyG$8kcBInBKhx_LE zUaH+v$ONC)pD=$hikZK2Eg=qYj9w=#Gg?>YwrVyVu}EH?wMjLxV!WQFZ8glleqCeL zDm*Io=G<-wKB<#f^K;^yCg%Sux6RK+LD4ljG3Huv_-C?3Ja_O+Z*u*okVY<O=oEeA z)wbZ_aSi*mtrcmv=zLiPhyP2j<zk{Ug*}`wkx;K#cj5T3o~Tvm)7>i5+RrzU8|6X@ z@_gr;bgYn6CwB7ZTligNH%(P|&|;zabUW&4F=FU+ckbYMs4u(u4&>TguZBY>u~Gk` zp^D)Fh@{9!8Zsof$T{@_{Yk2=uHznxl~pu*CUWg8-m4r>Rb6B4nDy^PZr2K#=hB;l zEST{E-YLdUDKgmh=8yZ?n;NQquv(1t6hgzL+j(TrXYS2P`?B9lj^v$(^W`Dt-uG^G zSIxUmTTc#&-XGm5;yiqYR(nGvLai3lk3&XFQe^@*A7`;y)II!Rh9XW(ZEEpr<8H!8 z|5T%PalC^5(d$<xC$VRsoBSv?Dw?-u)nVM}H;?p|FsFdbWY_n<qaF3K@!y_a<85Ln z%&E<P1Z_Ihaw%(5W1}CBpl+wsDg^}@i?{P0Mu9_Go#HRIIPNP%`bL*NpC@W1r1Z=6 zSCoVEa?{y)*mgt_`=OzUS1dN#^}lT#bwfWjg3)jT_v<MlK(s=rsN$a^Y>Ye?IS-ps zG1Qm=(o~DBQnGXW?S2j3SFs#cd%BI;nvZ<XiKA~fesW6E%WN*3ce@srom_;<tg&KZ z(S=0Ib{U*ozpg3?+>CIvGPlt+vNk5r2c=2p)W5fr?zia&DSlJVg#s=nJ0YhHddQBq z2vC6hz7UH5Q>7bsxQKhBJH}Q03BKpW;nkU8>!>HysUYa}nvHT8r22{evfDA%bR0Ou zeq6O@)8Behg@~p>9M9)`B|@pU@G*FX=juG=AV$4GwKSsEO}|yAmeN@Lw#;Se+LZx= zItb*$<(fq&YXd2niQFH5GY)o(ie;gw#D)nZJWKWXJGgCi!Ls;~=JaQKw=?Bn1G+6~ ze0&mhKiblx)8@q%yF_y6mGfFvE9qh=ujl=xs}7|}N7H%p+^?OT?I5`2ev)FGmc0?T z^(&RQ{-x&Y;R;)m3zv&uw)~!PsKDD-$32{GV%RC{PSoF{9h!5t&Rk8^quO%rv1@Y{ zN5UCqpPr8grM5)h3$3;9+~2$@O%U*kwVrNKD;A5}!Tp<6n>@eootCEb>1ZlBmw{k= z8~^;Sp`A}rSY9m4w#b?h4g3ueSZ~Q!>{!ncd=d@UHa;Lmm27KiPQk{>-W#P~`bV2J zH^Hi_`%qglQ~gYtsJftA#VQ{LCaYu|I2F%qwZTz|a&5eJ$$_Nk4r)HmYDcZyY5(f2 zehr)@^_t*+;R4D3+m$NkOAqzn*KF4pb#%N7=IvUzKhnxjNKEPH&3d<f;wep~Le?G9 z6xSV6!R-26>!%M~C2yy=&=lRlS!jq?3C=4+^}wcYjOi6t8{+!-J)W-r=~ov~f_}C0 z2OG9EDZlfR8>x<E*yc}WhYr_fO@Bpx>HLI@vlb^A3!3!DpF$lw>7*?%mfWt0kM{*L z6?H#-#(-c-aOspfvZ=$;N(p=2u$OyE0wld)l4-9|lreHkuLX<ecXsc$OlW!5RAMWJ zD6+*kSj=wWXA9qVJxZ)dUIK2b9*lCajEiOS^*VBWFG;v%u-Py&x54?CePuvd>Zm&_ z=wMqe_V83n;3nF@!~M^0H<vH8N~mUCbA0O#pd2US2yt?OeKm84@IjrW*ZERZTv#Y% zPI=ai`yoio;9UB!M)7C-izQUy>ziin()*aAgsFFpQ8EBcEsL}1rC}{ZO!WH4-FLwt z@(3Bd6N%sPoCPy`pebGtkv&8GYpI%ll&Ik_^$CC<#j+Gy-Y-X47<8#mDn5E6lHzGL zw`9qE_Zjx-mmxISDEiahR)r+8H!ThH`kbYQT-kVHS)au&4}S=|Mqd1ty7GAEUCWvF zFy8|3^_`Rm$aZVHx?H^{j<}%4oJBP{A`D@N!Gq2*Z>>hAFGO`7W8;0YOiLC6A}2ce zqo4OWj*iu9pYfkgXY6SMU{?I$-eHa9?)&bm-*;hm-(}=7xsjj7pBBEG?#lJUjvf_Y zW3NQ-G6v9iPKyk@g}5YTrzZ&#w9jM!`h~riQO>r?TGQerz)ZA+zL7@Rjv=SyqRIR0 zIPE)rK8e;|t_*wjVUkK;iO87LO@6+AZzy5P1ec;Fhpx?azC=LVWvA*s-h#&Ofj(Rq z40ZJXf}x(R@ZrS30ni*K4vvY(dQT)xrTHUQVCg+?y}iQ{HQEb2fLKzVMJ4^n5U5=G z8UM(2l}ltVH_LZ%FW1BqK5S@_*D<9&Ccf`NC3OD9e8$aJQhMHrqpv}Ar9#%zuU@}? zU1k(1F)}*(IWv<o;3+9{5O`y^sab8)U+SGuvezJWH?UQE6lket>9X)ni2BKyxlr+* zX7&D41iXB4A>4cC+xUX}nq&_aQJTA*_!tI@X6Hje0Z~y=SN=dT%fX%=l$#6E;r<@x zQ<9z9{IH4#3EA4#p!MTtmPNQKQ|@?wf^`l}gs9<<DpE<Ny@gp=8i(D@gr$hhrcTHS zxv>eHj<?g=+Q_&VGbp}&`v$OmS<X8@$9(%r{KvcR3d4~fm8W7i{{mkl;w$2IQNpi$ z<yBx${)&}up`*JykdG0kxg7?D{r+8DXT~w&TFN!D&r3!Xos^VRWEO4WFqaS?o*nRZ ze{T;Um?HbiVaQnXSQs+)EV2u%WCUPHQ*-kI521-VOj^8<NAni(TQ=I-+RGLKTP9_x z*tm*%*ZIHv0g{7);=fQfQeRPgSND=U0%TziacM5fa<Zuh8eBM|HUZKK=!?<Er@k9i z{?Dfl#1i}5>p^HQMkNHENd&lsD+VkSG9PZh@UxM(JpzccYOk8@FGQ019#m0b&uF6E zQC<u;uN?Pm_U@N1d6lpnn&Dts;BDO1xa-@-mYMcbW;m<tFDb;iqa^YF;_EAFGIDva zPrUANpAkEC1z1uz2!LP}Qw7y*S+hpXpXwa2H{6}BI5SJ+-IkJk4`^!Q2L<`P=)=_L zXus}o(#$lUm3k^*HukQ54-HAm$$g7rpig}Ekv=YatdS|$g<z_(J18ROtEOOnTr!_C zg`l9m#cp_Ta67>L#ngDI*{^m!{zlA)2a#c-yfW><OcHQc?^TN|Ago=UTN@(nK2Mmn zTN!z4;3cFz)?w)4@>;vrmZ`DDW{!*M<;%FOmXkk&q5vy;JjlI7#>J)nmO7ShcdE3z z!mQtP3KRryw8)db=0;$?F4M^TlI}^QxIpS=BEeKzzqHVTUzi(o7UA}T$AJbk9<r*= z0Lsa5vPertr?uen{^pR?aYuE=(c9a5Ac;pM7&nr>aM>`%U_V5L96}t4EWbmzX5r81 zH5%^fdX&g*7ps@bW`>%PnJN2eDO)yfb-g#b&f|)WUcD$U54-;xpxZz)AO8LQJp_Ro z<u?LW_>;{cjgMw#;WZg}O->?9M4^@%&pFH%4Wf5Fc)VLPo3-+E921XXpCpq`esq?y znVP58d%MdV?0Iu80w9~UwhKR7AMQhV9d`^tE8w-q<tr|`MHrkA5eez%=qPqu|KK}- z!t6X=>joHLpTm|L4EY2O0A{A5GTa``u5&&l0dTyLYAX$olR<7!Iqxh`E5yXZL-G}d z!7BR&KY8EmbHCtoLZ_pn%Tp`FqmfTQWM^leX@E80z*iFi1Un4V34=A;ul<5^;o)d; zJF`p@@+JWl*Q2vD|HMR{@^E%8F1+W@pMP+2x-Z&fU|>K49jvxuo!a3dEq2i35Mjj2 zRLNDS*fa!1YS1n62?;^Fy}L6qHxB^VR$^&sY0X+&;`76$aL|p-a6S3cRYEB!NCFC} zN5{wQUvU{A2mp`8LB-VcM7Qn(T<iTcLsWG1NQ0vljL4u42yoi|!NHFK5F8b%`?!I6 ze}7{q_u~~D*a9HVMFB7yg0X#gVuGnA009|8SVxBlAc#8v0vuL^jc0$iI$dr8Yf6EB z8WsTo3b+KsVo)fb$hC6V`Vi34B8c>qgwfptcGiYZbO-bG2$YnRpAP1qdtRUVN>Tj; z9i3ya5`c~;3jEX3$e=dR6u=}T{JOqw>T%qKe{^!<2dbY%Ee2j!j{p)9pf+L1Rr~Wj zE!aUJ0DJTS$m8X`nM#1-?FOfL3$WEt+u80E7CwF;Xj%2fo^8Y#7#}USyBsaU6=-b2 z-JRCZ3571Q&s*($05w4hI85T_b9R;w{&&_dik-rf1zh|pb{X=)X)dp?fByUl&uzEZ zdA>i#>UsT%fQV=foVm_^4GG-r8z7+3SdP5Q<q<g}6B8@&3>?<em{3GYN?7H)p$q`9 zo^Enw2X)`Fj~g<YT3YGg0~J?a>48Xnd(Iw_knk1svD!yQuz>V~LVPE{{@S1XkykCz zeeUJuRaRMv0Kri5pcnYs;Sc)nRH1Ni*(Uba_t(3nKx`!5zo)gGuY<F*vjayp2JPL~ zAC@6qUENm>f8+=V2q2f6N|X6Idr%#J1o3;_67<Khbc6OIOoZSNklN<vh~nepHEXP& z10*v#0s=yn^$a$Um(OAp?Vy(X;b5M})XXd$v=RuoZC?Ojv03<CH>r0gXiCB$>@KdZ zD*zzBwbUvEkx^12z;+KM;$B`}hJbCDo|{8}fOJ09(9oE1fQOphFC<|0Xf0f*kL0PR zpufNH@<z~dn|^4{ZQcEswMjkvrr!pkdmj9IDJq~EEJlT$Nc(Yt^zX%zx5-e701$-^ zz~7b4%|HO@ZeLs^G5Pha9kiQSjXP0pK|2TXVKw=+43zo@FgN;5F210@Wqi2UoM+G~ z7=Zr#GvJ%apMiK7g8a!vsct=X;8W6HKyb(-rC>Lmxh56#2m+tU@VNRrU15gSa=pWF zd+7au05NJ+AOeJXcev1f5I&2(VVe)!D>k+$|F+VWYZ<(I7-S!`+m~0nLb|xVMnqu3 zKGhM7^Adu9ALs_v_OG4_pBx>{d{^bc(83p>ax{u5-++V|gNXxzSB4NURbN0adc08M zX{Ka&Kx!%}3@#1Z6SQ*jv?^)P(b1JaTLf5z<_nz<%F5V(hEl^BG)fG|^WN4u?&5iP zcwjz%9t>8mT^_Gp9yA`78MecNFvSq)T+p{~of+aGN~QV&9CnK@V(8W0WXr@1gO(QT zVjF!iq6-a<_J9Qc1|O=bm+oqW>{k(;4(5ZUqiKwQ1j0J{;2MS|CVqg%hG3e4fcFzE z&<SMd$15Xv@Ev|Y*2$@<9|65?2P!<AC5;0c*Z_tQ=0yP5+cXo&YO<&u(D#+yB0uC} z6TRn${PgH-$^eW~pz}HH32**Rw!c2LKtY6OG-!E2#AV$Et`iA#1WYBok(87?PdM8e zK?n#4AmDRi1bzvTg@pxXQo-QFYqwL{1TGs?-{saN_Nmd)H$aF#fOAj1No%t`HG@7l z)aZHNj1QbaN8*J%9f$Mb#3g&%ql6pPx3(ttJ5QH;3cG&8W1j9jh(PdahS#nT!GhQI zQ6BP8AcAn$&iU>XNqp_SN)Cc>uB9JK+^2i34s~>|i9&Ixipn^c8R`_2-y(W#9gjaQ zrm0CVKR<s2ysqKR`MzegB`&NnuaL|OGZYmp%TG!8eE`h7<^I|VgfD5pLc>(say-xX z#}5KTWaN&eCDLlEserIB3_>pJAON~=Yj01VK12OCmtr!KA^v_O<I&#U9$BQ+5wL6- zz;eR0GmJ#wD^P}hBrKRmC@XsfvyVVSeA8DNz1>%U=n9L8A?C;@1^|S+F$jraG}{@^ zM=&rj0Mk6k0P(L`Vfq+K11N;Dii#*trz*<{)HoJ{01)BA<P&t%jer$}IR*&{BnURt z!=m5RcDg+l2?8EI=YyWfhrJngbV3dkfU4JQ@!&aFYz_gg0M5sn80-#ntJmlIApnDt z>HTm|OG_(UX+AhyW1}ON^co9H|40J+WoGaLj09cJ;;TbtH8lav%>uuZc>HQ?=Dhy8 z`|7`R`S|gR;szl}a36D470RB3L97%NOd5b&D>v!JfP}%Q#-G8&z?X_C!eAr|_$pCA z_2bnpG}u51m<E9jFGP;?Gm$f>7qtOfRc5~?0hG!Jj6ZD6)gD@rR=`N6$@mMn&#%O; zef|BSy1Fk=(a=_azXQ9r_4T2Gsha5)Pd@D$>&FlnSRn^}NPjQ|vI<NI*n<TonM_Pf z5Q|?B=tz29t_0Rj9<TS}fJuwCu`!&seml*>*{5=fii%tmYU4l$U<i=9y4PBwmSBud zS5L3pdWMZrvkY=Rm~Wpf)-k)t*K6cbFVge{EwOemHj~I@6BU}&(!W1fyK-<4_T>vA z7{Y;Ecw}n8gb=XF=^zv!5OCwTK6P~+cq=D|o|K#%85s!&q8BjDVhAQ*V6Fo|Xv)n8 z386NC0ARJ9e?8yideXd$tS=bH%E}6BT|zK5hbeDHMn+n3aVW5sKf(F8x3^&;0E`GG znAuPni`$em0DW6nr~^Wmp9>4b(KPZu(?tCA-3?$OsHLR^`nqSxOZX`%DRlsK07Q3r zef<kClUhISj=M==Qz*bkA;dB3y#b*cFk^IR_0caCVbKFH*M8}uff;Z7Q7e|Us_@D+ zzomLzKW^*yd_>J^wd;8ltc>luu+{&!h=pynAx33)S;>`2aWR(o;=;b<)_?xm0j95{ z9(DPhLk1q7GJbz3%}*a?*@sUjdkSSy`i>zp-AhtL_7w8_$8+0}fSY2zc<~MC2_bBz zsb^@Y8(1J%KvASsDKTkBjr|zR#6$pTZ)+2-Yixw6Dn7HWaI4qd(b0-1<O61Kz*r1m z%6<D88X|mWcX!9^ETNH+(bi1mOBfk5GXcPRwu%l3oxmrOQZmE&a1l6?pDeAnf7>^l zwua#VjehVsGed?zGhe%(GlK8~f?$3MsHpu$A1;{FLb|)T=mtP@vs7ZBFn(aP#(dxh z9LOAR?=F_W=G!{EFbsK)LNa}$!CM|49vHiT$C$0M@B?YL>AQCw<KxeOmPL6Uld>8k z!aOCIJ2C;*3IGOTZJxmSFzYp>NrsaUy#7F@UG?7AuJsNN$f=i&9)Nj^fpum#T!p+q zoWif%Zz96)Ztfy(B5RAE@nYxgd=2yZ5g*^)+bexe118=iWn@sHWdLpm>h2R57-)SD zLt4DL{`?5-mD}lixl}=7AjW~f;eil40Ql}qki?f&SNq4s;Y)<QcnnOu6csisxOuq$ zX1+`P-&laVytkQ10|NtOU|I->Ar1~sXJ=<m>=Veq<Jf=i0jQ41n>RnL5}n6@=SehM z0)tqHFJHc#akR5LbhZs+B2@kA_ILd;$U!b=xs|~|r4k>&LG0Iiu)wIKnv*-w`V@ZG z&p_pH8wZ?Oxw+#W2}5O7Ro|O2z~;WKtwMn6<c$JgEbQ^)$9YQW@JJXhf`Cx#Ho0JV z-yU~M7T4wh)7-2c9vvOcs2QJ_2qIqufJcJizC=WP84YE3BL$zrLjE8uO4#rnaHl4t zS<*l`14>I-fJH(eA|e9%Y}6Tq^?t5~i3<BMjY3ioh#=a46)-n9#|#z=3Jn$aN5#d& z#XXz7$=G84XEZ|aSzr^KGgVbp9AK^$a+%2sK}&LTb8`Wf@THK3Y7G-fZ4E%@jQm6A z2%8$roO3VL`_<HO=0)c3%?btp>R(?RhJ%m<mJ^+Bjld?)z=RqWP|mrHb<iDzS>3(; zeIVPu$WKYyW@oias`*`xAX{783J3*q+!=2Na%wh}-Dv;4Gcnb~?z|MhYakYPF%OSh zom66w@dgN!wSjCDHruJGqXSEIe(<}p(W~b9fW!h88C`DYrVf|r@qHr|A_0ogGdS4g z1BdY8Y)4JCQ2oiTU%x=e3}*e1z!+6pVd2eeEGb~5*6U`6O^lx%9R@c8QZh1cKxhNk zw5Q?bW=ARmR)qlbgDu2;YD7p3zl1LE#^|`a7bdV^oHH4j8?;7oVN;2~0hf8*xmDRN z5Wao;7I5U(Mt|J#>9)$=VlxqFsCR)pHX<Uz2YBxlFbq08G6IsB&ykT>fJY$EGXfKh zot<J@Cv<M6#qsga!9-;+V2anMNeZVG-kUdXzBUg69aIDoX4R0LO$`P#fj2iDia*sU zuF8SsqhR%YR>D`eDJwF)A((miZRol;&C2g~it^{rpXnMKq5uqH)a8Yhnk>$yLr)`P z<IliK;5CMURB#o<<NzBNHvzKgP@q<;AT9+%ct2CU?}Nb%>k5Di_Vo0$0cfWE!@cL$ z_O=K`WLsw^BB*YF`pheKc2Q&F4C~f-?j?L?X68DleeLq<+39IAI=b!RZSgosRaKZJ z?U+*x6oG<8&E{-c9FT3)YL8y3fdCUWTM1n5`-uWv5W0MspVz%TpR)r7DSi(v>XSF` z_S-gKy8@dec?SmuOxxw?si`-rSzHe1%r@jnPk7V|)tNx9nm#=z8TKL=nEq98TFv^8 z)NZF+z96XKu$?EkI9%!m1`Fo!VfhMR&~&XGDWIY09Sbinq4)jOOTY$@k3}XYhYk$L z%}iEu->3nzRqn9)+t><>Q;Qg{&LJ!VG-&DmOR3buiAM31m!zs^{oREineU5Z2?zY| z%h}l2=!G8x-$agy0(|siAt50u8k&r<GE8uSKutyo5Jba9uYvHw^6r_L88Dw(Av6i- z{2b)06=uM?6N9f52NR6gI5-A6cKxwTK!bvTE$z!OGc|2Gp{-_r_j?&6IB|LgRqwGt z7!%U`3?G7P18y`YC#PbtaH>SF1H`}jh0fr^r3(*$w><+MSFT1D7Fw&BTZ_5w7YAT$ z_}FR*5u`e3%Nua7{E(jf1asYGR#VIX_qq(s3Iw5{p#e=M6B2r;w4&{sn83xQSB<cd zd8FUMLL%gazPy1_AxtKm%<K5d^{*-5A-Rx{!FfxbW;un2r>D`ocYz>4nYjVeSH#%Z z6z&+=;ocyO=;`n80D@{-iVVT>3P9NDM(0v%ZFBRSJy$#6)63zGLA~VteBCjTO3WR% zcEU^9Z09SBSqZxv`Wilb8?^Ve2h+EXJA{yLTzUZ*L)9!ZgllTjKj@pzktbr*t;4jL zt%@9si;1B|GlcOOq)dEqSNWdBAlm>iuSjdvJrKUUb~z#g-*Yk2e@#%VUHx;dJKTQj zVSm9@SM}^(=x8eGxuK5{vRRCHd$qFeDC);{hJ^Q)^$ZDL%`Pmp6U=}8aB|poDy`j+ ziLbW`(HL_NVJozA^<Hpa818%YEP{B()O^~U=j6sOiJ`u~E#63fs5?`V4gKBXyI5`( z!Oat%&D4jg`J9pWN}8G|&HCYcd)BniP{DLBK%Vl%*6)M724Q&mFqTmZK63;mRKgD! zORzu|rMcn9)AxJz$H&L8G(qgGn3EHi)84ev)jjyl3C9x}1}gwk9Xr?8*T?9che7ZP z0>vWv9ei=DT*cHDk=Beq?;}+@t2lrE{tf8P0%}~}KniiXJM|Q}>CW0EFSIoK5iu}Z z3Y$cKKTLMJt?2zZIvNN3mH;0xTMD8Rn9T(du_SPIuml-A1Yt5Is7e9nFAjcV$;R=b z$B8@vIugiBU;t%G$*`&yU(Y}?n49YL74U1N<>o-t+M2P!`7jKK{Z&9U$e}NDD7~F! z(v=K>K*9oLczF1u)esirfgh}~2_UvYtkw<%)d$$Gci38N<^y#U>&LHHnvUROne|YG zgj#9ZH*E7KED;|)A_qV^1SE7=;u6bZfDzsMg-#_0R*{mDl4|q3SO^rlKMsc_pkUeu zhR}p%oWLW20-FtXQ9?&=Z!cRbH81ZhoAm7tMH0wmIb9dPjP%oN<zVL}&)VLb)twjF zK)W)49S;f%lK=-k?szIh#>hDE*JXFI*iVi|R7wg3ocT*PiTl7`Lf}ndw}REOU<L`q zHJKn?ik0B?xcrdB=UmcjFa_rDX_=X8Zk4GTCK(P{<h(CJrWPwVkDQCiMFY}`iU$6A zJ?u{juSG_I6R)KQlKcdrjP2D7Y)_E7AcJ(Q-#Ccxe>8U9@l^kR|36BpY$|0_M#@M= zcCtk_ADhgMRfmJ@h9aAc2t~%pCOae%A!Hm|ly!{kJ$}!3eZSXryRP5w`rWS2?eoXy zbDVSD>ouN_`{Vh1o-Ww;90!VzX-IEhgizoy)cdnT-2foYnrYu^THr;<Ge65`V!u6t zT>9SLYDacX4x(0o+&tad@@z|y8p2rH!IqCCO8fXoBCi9vVV9I-gb@Aa-Mg2NS3s_q zh??I6B7EI%?>bDQXUIJU?HuRV3p=yp;QNz+BoyHxFV-FU!RYG1+?+W{%gHJw#l#uY zQe1PmTX!f2V?2rRX6Do`)>2uN#OPe{kAGaBZeHf$dEJYqW@H?;zO7_vcwUH}95%ni zHk(d#DK{@~AVg7+&CXk%)NTva^H66!S>}2hQ95a9>^@7f-PZ(7OeG;S@-%oxnajq< zM*+MwLBuI>Qbg-?RB9?E9UUDY-LifwQ6Zt%uV15xh=`WfUbsb7D%jX?2+_BzkI~iF z*T?B_p~BMLI5;=}A(iB+ec+LiVS!0wKUTZX%i;?Bqxsu6XUwAD>*c}_vjQbxiZPpw zxA<#_L8G<eFmOUdFk&DAFkbAg5U_qs&QE4|uPt$XGDW5#kga_<Md}J!B|~p-Z+3Pz zjMdM)Sy@8yE!H<>Wr-d=dX%^p+IO#Hj7dR3!QS5f1`OSp#CPxBX=-Ub1R2pfKH{)h z-_VfWg3EQb{`%rloWcwI*N%>7CVWa378WtnyTM#3AE8Tv0BAH+Sp7X%R?o@!`IKzb z2;a9<X81Kdjkg~Eqz9KC=laF8wV~lCGcz-OgfFHgF?7Xb4#9P>8Fbcodw?A?5u+hp zFFNMqI<jvhXusbFT0AN#DFo{B5z>ty;1r*g7S2s-Vj>W{yk;TrPy4>`%-o!J1E*)n zW}s>YTJYhnk*K%%-e%olBj5xCBO!DdtdDx=z~DQw?|haSv>bYSZq!kv!CvUj_@_Y8 z5iWM6Uz3wIh6`iSzc3GS)>;`V2Y!bqwU<|m^B<Humr}w$)(Daap;)kf|J?mu=Ynb8 zdY80b4N}Z7lc~3-2Z1H{{cmrktWLa_mGWBq_@n+Ya{6X9UY3B!)hXz1jTe1kZiqMc z3;z05RV`5n8M?LOguEd!fWPDgqts{5K7}b{nBFtwMZjg<gPAqw44p}&p^{{jM|d~{ z;v<0HM4*~He>@~XmaWm=G>Sjt;G(tQc{Bh0^*El`<~KeN;ftD_qXT<UjC`O*Rav>h zneAk!=b7kgsok3n&tXwjS0{fS5)uVbA~}PIkQlwY1ngkQ28V*(J=uWtE*sfHk!E;a zP{I*e70L-x?@EqS(uGTxP6961&z>1|8vS7S{Y#wzPwdxxia;u{=b(?u`|Va0)Au(3 zzkJL*%+6FaId>B{(yxHOh4PG4lq9s&0|E}I?U8m$7tvB>2iuc>YuS%tz=%8XXnT-W zlSb=8Obf9r)nK2>ShfN<fg=euYdX(JAM(U{xA&Gr({>#q^HFYclrx_~Hm_kMdH=<X z;;B=oUTfnVYg0v@-oWvb3_d?Rw)MaoTB8j4h;yNx01Ql_WFlg;{S_a~TI`N5{BH46 zbvp_#+w#kF%<=a-TQNFcT<WT2D#P{S){XM?3EDx9PxW&GjgpOIX5K92mDh=CrO)(d zJK^TD<kf{Xbg;g7l60rqRB@9Ndlvg|X?aA)m!R&4fE7<-!EvDP8|9r%SLEp3^%VDA zA7$Z6iLPSa;T>X9-ESD2<$5*z>zNW<8<G8d7TnPFZ#-EADTY$pFL?d^h)BpDqL%hk zDsw_sN+=pX%j2PS(aY79m5(j$zm(NDX;)TOBIDw0!-IqKFY{GwO;WNPKE&IkRcDUY zT43D|6gG4)|I{sHi3JXr=MUEO^X=f|Qn(6z1LBv5$2l(B@elY~><=1b)m2#E*T~CP z<anN@Mi>k-ddzylmXb17<fB1sl+Z==WhK<C**38!gXr73;+2t%A&L)W?(1ui+`iT` z>o%YDOMqBbAV*Tyh~K;;beF_^bf=AQ*S%YVNs!pA^zO*p7?h*m1SfSAl;PTn@8Ui; zxh1+fjs<S+Dw~oFP(G9V(?j9c%)f`m-`+z<|GKT18>c>=ND)sBntvv|8{74K2+h{# z+~K(C#9%&F*8a7C|8L_0#((8Z{>P{Ob}XWRWCaki%|w4Y7KuTU3F6t3KpyOW8IAw# zkpKS=zm2Ryv9nj5zkqriTYh(8A<AidZm7&~bzS&xdS^5B|8O)v({rDK-^yF__ZI{O z<vz@TBvQw7_~YFpKakJwu1}6z{NHCcA8M9Er{CT$FoChGTn(%BuPhvX&>KpAVa-U& zzoT1yxgEL=Slhjlde!(&Rek1>C4oOqs?z;?uYf1=&Vd*QF5YvQ2WjuLX$kY=S5!?e z5AglD3lcAi?@aJ@IqKUUMvM9%2&~7W1pE`krFP>#$OPGz2--A$Hr2)}mxUp>u$clo z%aBJJd}5gByjq&d8p?u3J&ntnN{;=Aha3DQrb@Bc>CbI*=TVTlUri=(xRbGmKTJwB zI{0-Phi2b}1ammgtg5xJzGF(ni_s^Gy-<-Si=RdYZB%<p_9$GyI$7`j83ObDE{aDR zv*IGg*zU?&i~j4iUP$_|4L|aI!~*F_RqT3+5Il6|+w<tt>u;U3@i#Xf<DtPxn(ffm z_wfGQ&|57AUxg`qdU`5C(zs))IF;7EuF3dwVadZ4sfbIeznV@j{Cpr@rLonQWuPI& zhE?}~1nQumMMJ&qGmQ={zw-q|f*$wc(0Gu|7r41kU@G@EWYl_&C#Z26YH3A%p#y*u z>6IbA5%|UI&}+NT&TMsgQ$*EtXkqD|05$bU2|jx40f+dz)yTHhm-((!ZL;8j9c2|e zU1<^y`Vb&%vtWl^m13X=<BrkFSllgBd@^!jY2L-l8B9e@$5&r>Z1`uod3eNEF|xPd zba@yh?`hu67Ugs_DJqHYSs$iA#_UK8a;XGe;iIfw!n!<XdT?l)zM@;*8Tpg$$bFkh zF}Z}7j7&&qKE^6v7nP8}os*xtcnQl?4iSO8yhyu6$xsdscb}4iqHJ&@Xb^FQX555% zj1TsXFrGwC=i>9U(2yHJ`CRd>mXuj4|2Zj%=O_(ms7p^fY+Q~xQ?4LyZT`2t$bT;J z|BFc#rQXuo`k4DUDgYGt-H^Kte;#(T?JLO#G&-4T4_Sp2Um)@OYKKgG#k;cLKeO4f z32;oI=klFa{H?ev^{xDMS1c8r;%^d;z*ILH1nY(h$`?Td`~c|hxl~V5WZ%UOO6FXc z;)lJJR)pS+`t3SZR8`#pbbQ@+M@YnZ7{~*ng~ngnrPTL2a4?V^|FFAQy#?TxExz?C zv>lUN{!EJsAf@B`k?b)l3o#RLrlNb}H@Ra~Q6C}fT>=V}mYI1R5H{jeA`vRcmT~0G z5W*tjVy#;v9r7|nM^X#w4(P>`DPS>I*3f{I3$h-xu;1RY(%X!TdN4Zy`rm*UEN1!# zT=z596NpiWa1>N_etuJuOKmXlQN#_H>3rh};emxn((<VH@A0Ma1_Ao8nZeS_+B!Pp z^Yft~;3uG^6gNLVa*VUQMo&Ow8#6nA-vB-6WWc1q3I;5rxZ4R3?MG2TX+a?IS;l-f znIt78Bcr454B+9hFiLry1{PB8m;-i5r@atj$bdoO5LSM4|M>+-#_O44lJulPP<TWH zC1UtNOSrv*%{nLu4^zxFtv*|Kxc_^(dTgHp!ki``14x%9B!U4Sj>UJ48svoCz$nOz zdo!JEAO{%0`c-=HKn%qSxd!*|-$$U{*(?790Hd&+_#BK1@ITxgLmUmnB?7{g1~FMS z4ks%oca(;PrnAo9Klh)aBXG(}SnN9Q<}|2OXgMZ(|Nhk!-wi`XF$^){$da)51c<p# zp-|v)I#&8a-P5So9Islk_!$V_8Ug2kt#6J(rFbujnN+!*<gZzyfd+II0zzhq_n`!_ z|NUKB#KM8{JZ1=L;l`to6J)z|sYURPsN0NYf?8O-<uLfOW`5f?km~Z&qi&Z8jQ&87 zba?Gn0OiYI23vtAiSTQ{LQCBmGDIlosqzFWXn{=QS`MUFnlEgvwDSMnTF;bdh=RQf z1MTciCO{+@(DeE{PKPVkxSRqs0xAG+=1mUw3KamZisj0$mblmb@MpLAergzO;@?uF zLLgj4I&A^@RA;t=E?sa>>fw;&C}fIs)%tovjDCTOiv-3t^2+$npO1mP9Yv|$x}{}o z90FtoRQDK$c*%%TatQ?^7Ih=N%L1$qL?UV9<e|nIp`bJCUhpq+USL80=}C4Yr*{IT zK|PF51a$!rO9M?!MN9+)=m-ixKT?QBDZRxCrqc3K1En#uH>g2@b7b6?F4e<>03LQ5 zF;D@K!hB^Y1Um>x3_)p{nK|RG!5B&P-8iqRs(K|o9RgzITelj9T<Q>_2GG^4yZ|QT zb^n7sCMkZm=|4*p1%d-g_Nt*to+eJWxuvCOxDRwkYh6ikac7C0Mk^jf#nQ$`Q)37n zlEE>IJTbuPTVP`%MjUce5r+ZPgQkuSQnww-EJY92;R$l`Mqti}Qv`Df;pA`~2>J#s zUIdvH_Ckq6iyF}C?8@QB`nrO$auCNq+dDfqPb=69(IY_KtA-h#VO$k3m^%ZT;AA5U z)Cp431~7K6q+<XpDZ(g5Iwpc%VkDtpER-C*T=3yhb~ZOqi=#090u|GdfgaGLi3d2y zAYFj83^X-m0`lREn=J6(&&i0ZsuHiLs4yunP|cu$n$dFK+cS%hj33ne^t_{w_r|Y5 zxm?9`YFMA3<y+j)>F0<0s%y*3vU+;4lMavTQF;2LkX{E~2RKC&Pb0?X=0YF~rd9ce z3pf*$v9YmRUt|s*A{70jj)v4RpeWG#Z5$p+Bk1_Z*w{e0cf%O}{QQ)f0X%*&1W*qE zDA$GnVQ8laqv0n&n@@1_kX7Q-r<+fnJoy1COG!%$vQIYaE{KVp2a^8;tQcg~28fld z%Y%aDlDp0J8tNlhT{bp0aSQQraUwX5>(|f0ApBsN2M<S-aT(Hn<TN!qc~dEkt>;qB z!=a01l-$-#dc>7_XMn*$!NGTn?@bs1a0POmWFyZ9ZSY|9AbAPt5z^100Scp}XD8+` z=M7@od+sijCq4NLH=W2{a?=!UGB8La`DX3+7z?OqSgaLQR8B&g3+%r^Ky?fidvn>Y zTp@!+RtDgZ)k*a6$3$Svp-{JGs(_{!y3K0aml{FP^5at=DXb{qb2XK8NQWSKQ`y-k zN=r*4Tdd<cw_#ICFD&d_@dRV70dA_Mu`wy|y1Knd{<z=kv)v?V+;?C_j~gNo2=`{c zbD$s*<mf$E6y`R@A*&5CIRbi9XwjelE+{~*J_1c515@I~ix(D<wj|0Vb?f%+CjWy~ z7O<;WA^f2aI}XD6p$Q*ifk_{1S+%_qqlIlM0Pq3qCPj7~%#!OpI$(V$z^v$ml}8t< zt$^PG3jCAh0lPKyXOqq9t_NfcITJVF&LB<{vLT@X>Lm4U7<0y0!gnQrm0RXKf_|A^ z0lTZ@;ohp})JwZa(9gjAki`P<5-hl$)5#vlWJC<}iVE<N*FM4y@&k4~Sg2ZEX(2ZY zwirX^-?0W}_23!9yd-(=fTw_{qUebcT<hNcZ~rf_3KxNK;OXi48omTjnj9o3GW)J+ zLAT9DK^AHh&*jVa2|brW8O6S%J5CBmPx?q)xXRA+ZhpgqgFQO#%3iG1)UATif!uA0 z$fRT@Dlz+8Ri$NiPA+~m0!6;cPkMJ&U*wa9FGyC`qdUfs|1@PXEXKlF$ij_YMnVEO zFq61j;j6_-eKZ=aX=oT+SXjv52N)lrKkBz5jtGUnKLa|qwUrb~h~tLVy`&JQpFnty zfSH9f?lkxc&X|+Hkkl#Ejk{sp(A3ilg#0GXXHQ|wYr$6{|KrLuM<5{Z7n#1>+rk6- zmX>VRf-h}~TG}1^3yHv)aTd&?vNsXXhV>pm6=%*vz>pCT8NvE<9JMr+<bsStmk-NL zQoqYj18B(Ie^4^Dp-Sp{^zX_lhtnOsw?5mBxXvRZ`<Y{PhS%LNn5yxg*CiyzmfU5j zsAy;~U$7&WAr}nNc+5Sj3kqCf3cX@;avWlgapS8DYk<);KrRZ{_92i!WeI}=f|Wz; z=IZ8V-FOEyXl23Sa8wG^k62$<RnbE-S6#1jmlP71rir*feBfSQ_@i<<r2!~Sezh-T zR>tYz`_qDgPK446pF;K|B!OpbMhgVs@N!3WMa4ggL=T&KTYChN;L|bgWLaHpr!<%g zDyE;nLz4}Z(&C8)A2GMbrvC&4#)%csf)5lC^}aaGOoIP;nkcC|b)ZO&BkKt@^?%iv zoAMwxTCAh1$k>yt1YfeAc+{RMAM;0Q1XouZtAm#4v7s7x1#3&T?xHwuu7U{>XH{G& z2Px-)Nr=b{%JvL?@9u6lt7XQU6Av^Az)`xSW=}H|^XerV74)Jr_yU`4Vcp0ET9LdH z=n`8MHw1Ma5W7bgdwn`tseN4%y_W5j{B0`2hr0B7o_+|Z4==meZi5noa0w2qR5{eq z$;)p#&h};Ln}2wL<>ut9f~BwhReU@N-&r=LJAyAG<Kv%!OY5qp=4N_(ypccN$$KY4 z8;|Ak9Dt&XDm&=|vnJs~lZuu7Gb}8gPVG6}@mCCob+U4_f)V*8lGX=TZaDeoyVrKM zU{#ZnQ|UA6k2%Qw&K{V`rT3=^XxDh+(wgb!JKq1T#APk2ilUh1UtHkd$Vl_&xP@6n za?t#--80ltddQCTzf<-DfrDO;Q8!*xQAltR7*dtcR%?4col+LpSlO$W8kQwvZ}QE& z&2LCKiFq@1r-{P@H^(~HDjt<iv$g%YdTq|;r{@GOd$E&D>g9E9Zfp9D9B(deVyBJD zn+5s%5_dE519LA=`xM3BWV3&p|9GCXfJsf>eSE3T)k<=hQ8n`p3Po#({6@g5l8pTH zbwmEOswy$N)I$yZHLGjPJcWabEu63RHEcsel5U9{Dx>FecXw7mLk<M4(tY(lUcM>G zfo0lgllZvmR^M1`rj3{FGk9Uz*IiyiBE#Zl0QazYYCqaAvyH!xB1t6ki>NmKP6sn{ z@9Nyj#JllryIUs68D<HT=o)JI*!0V#do8cZdO&S=kmKm*W2x$^>lmr+z1-Su<a5!s zQi;=3lX1)6zDIDXn35WH81v7kWUgQ&#4xs!|2|%()bf0-(zj{-um@J#k)m(&Gk6oK zxbORFxevtn8&<4Z;vC*g?Y8bEN1UFFi+H&*anDbBCjY+W3F?8j=7Oy+Xz6@J7JQt0 z6JDrD#j7#(=BW`9jgqPLQ>PD=wSTu7X04j<x&JOIy`|)|wMOZ{xkwi5Z0<6m^d?R} zLGPW9L(Ae_h4`;u$e^q*`WdZNYX#?LqJLI}MFfdyqQbUT%m-3rl6f}|He}+(ldCl$ zLi$Idtz&5Y^HgaBhlDjsKE(P4bc>ygNDcoWbLBxt(k%a#q$T`OzlYMZ^-%}syL{RC z<=s!#es{Vdq#=CA>f2tz<V)B5W{J6fZfB*vDmFXAVel09Vfc2NnXt%%k5)xCZT{i) zU(ZLfMLM>IAGJF&I^riOU9Q=!W86(;HfBL46k?m9B2=wn%20YG(4>?5v|!lRb}@;! zHPf@gA-io8j347x@4<yra;=-KZ1??c@RRh!G&iP_M*AK7ZdC5#_Il3Qr9;@sz9h?M zPaOFAKGwBmgJS8rlyiNd;e}{1lE}noaO_WI8XBeEvmHuPv2l3G@1tY;huH9y?T&rZ z!3E}F{twr=`_tc;M|?EGIb1}CC|>+B9ibdP#*Sr@@7XzbL<MuJ&!2V`-~Q#0=FFn& zj~jy(T18|&uuAfVvqfISPE6C3mzOd6TP+1FiOP&g9ys8TuB@2Tk_u!h=x;0hP1_Hu zZ3Ljgo4OPLQ?zbh#=H8H5%NBqZ~%GvaA=EpQP03`USPb4j}9gZ%-t?%?f5p7v+LUU z8|9RkQ0QeFdzCo4$NKiE`YTxKe|_k`O>(p_eV*(&NvF*q(qnqxM0XAXAMs4rce_V) z+LQa3BL^2aD9+ITIXiyQjIZG2g7~j}{++N9ky-S)51N{j!!Gk>DyGjq95zm_Bx-6< za%mBehNFYxPj8_0{EfwY;_XE)N9^lL*0n}%))EKg%Z+Nt9Ox{w$d?#ym&Z1Es6bs7 zb<~Be_2T=-@8+;RqsA8`=es`X2w$z#(^2CrAwk7-UGzHjD7O|~{8r(&;N~NPoRwra zv3AZaV<|_8fSjzrsKXuqRb$*y4^>_GV^o9XiMRT6c>aIzVqfjzS(^z&H>j-1lP~zV zQ$9%G3WLVIXx>veQr0goHVB0aN3#}R?#;zdAfHz@qKZ-%eAIm*XvRNxfHD-I&0P~F za$<`qZoorY&_nrR%i9`p$Lf%h4;P)*3lUTX7~bCVdjHdQpQ}<E&uC*?Y|adlxK$Z> zXq~pv#`n-naQ!~GO`y+JhpkI>GM{uwC)rP^&ts0nfmTD=ls4+ZE-TjGaU(d|T0LXx zVPaz0kj4<ZELGbgk4JJ=1i?+$v9r(cGK-x|Db3Rx47x`81wpCB9t!G@1+EKYGv+`B zQlCCeTxsIB%fgvtA^6f7f@NmlX^nE14c=K*UtHYVIV_9t&hriU&W&}F5C5fFskq!U zZ+iLjaWeg^cLN#=ou&J_HQG$o6U$-dGIe)X97YdP=Cj&AH`I%8`+XfX5SPcYU=DmT z=a7Ek^z>h!+SvWVh=FqfP=j5<U%FfPdTwi{Qr3bgAJ$Ypd@5`}!hDvo7pu&<iyNsb zOSMhua9cRXd~Hpm#G@$Hh)C{r*vsuZUXE}H_J;=@FBTUUwd3XV!DhJk<D+awnSPvr z_gyc3m=KuLEN>Sh&M;U%mbQ{{x&}y&g86#3P+t<Qw5ETP-=G?pDO1Qj$KlW-xgv@g z+HAsld3%eVlWsT%`1AmmYIN^>mQ%QD#v^gPyf!hdw!a_B%)PmL<IWv+t}ErTAW(`| zPMK=JMuj<@s7k(jmmwC=UbilKKqD!bt+Q#yt#lEhLG5^1Ue=2jiH%IdR0-lij0)=T zR^u}?G)I_r@0MI89y|w6{MGwmk(No!()-i$dC&IWFWiwSrTBkRa>s<)QC1XHUXAJB zKQp+G|Kf;crE}{W<@<~J$Xna5EYqBR)HFC`gFssLVoSTjB-)>Zsc5H#1)kg*S$K*0 zanCWT0Rg3LpVQMXHrGxrUAI}-*Fz~@|JJQ+O}h1<gh>%bdfM5y1)?XusdDA;XWOec z8pg-!ar8DE=T}a!S#VmtC!q2UVn4|GtzFoE{`~oC{2UPq9BqM@YP_cO)KR9EGW)4* zp^UL!Z9|t_4lNDRdP;>12BRagD!nU{858%EYfH=9sg$F405>dyC4Ch2-iTLvZ>eE! zF$R<dR1(Ni(U0YWAZP%{V_bTho!$70-{83ev-7Kc-7anY(PpoVr7aW1XB}`3=<u+P z-+pg@X*VWtnV;KHyyckcdPGGVFp|Q}n@3QuKYU<>N+CpRBn{j3zSHyE7_fVzDO77} z?m2!8$-jUY6<VXC*ApT9jJmMd4QJjP9($Nzl&B*;Kz2evj(L>BUgt(off-?NQQ>}{ zjSX@su_o`%(QaD*=GbSM)9a&&fGAGC#AQsH@%lX!r~$w|Ux0jBiX<Tl2eN8j1A9aT zbV%=#fd!l(?JExE5vYM6aMph;l#e|%D?oa>6QMe!HW{I9{&EXfss*&ki~DjCe0fN1 zIN0v+p8)9@Z*3kNWJI$&o7{4c(eGqT7!6jS`)d1<ta8B7od^Y6lwBUfCPY!yI(Gga ze!TR~@8A7Y^x?LGSiHAhtS+VNCPw0_Rd4pMxKjZrIab4#)PprHBw;N_Kkw0iV-r#W zU~V44mEw5>vhSLph6SXn3cx8V5Y0$g7ZD2hTLYMkS1msgBY7m4%HfsxT-A(EpsXXH zpB}IYkp~|@2l}i-cjJNgL5mc0G-C7{-4f-}+cgnkUOL{WVh3lDV9u_elTrTr@w(U% z-alKbTDqH#tv5;b?FeH5`@YU!&-{{)_G6^rH&qhbtpzQ?gGcD6xW^(H3i#C4YU*Jq za-IlJghZAQNcS0sEVh5@{G>pZyN|g04ZTOPUb{vM6ahaXDkw;E2nfU2IV&jW;zvI( zkFzexVt&SihhyQyUyoJ3_FXT@q8;*Ba)_)_G;X3&ZjXZq9O8BA3yS|Zch9ZI1%E*m z!ovh40;O^rRbo8l+6D$Bbae4^&fva6UWJ3%I8gnk`)iBxSju;L{^B04wu<#%B=nhI z!C<-ygUPllch92bpoo9rWvMZN-s!v?t8fx<TVmOdPzmy5auVCHi-<u`GRO}PAnE!5 zx6M3QUaGl|H~~iinq0TIQCutti7&!+p6u)lpZvUB?sz>Vn2l_<HMNC8ljR<gtk_@r zYBg@P*6xOulV;wgjNj=!pPxgEfb1h<yiNsq=uf#tQewKT1=#@Lp_YJ5k-F4wa{>=| zRzB)Q3O@iK5A+s0v<#pO&<f~q-g5H4xkUm=?dpr(FeU)B)E(P=9+bUalq#a3)}c9) zvzxj-L~3kHM*4i|(7t+S^~uoTi=UtkbNZ{Wm!ZH3@$H{6dz}Enha``~*ItE8GT`oP zkZuA8{T#_ACjiWBu529A-NAZz135;JsM83!o=66$8zes;?%)rnHtETZBjg;WnM=Fr zMRxQV*nsG|ujyuuif-17)d{ni5{w*QUYQ=k1yftr@|Nu-EQaTHn@8+#6bN~zK^B&o zSvVv#fPIkT=nCm`Td)lMDGvCk#Nzt|l2WZUop+N6Y%whswE}e!Jy^~3;k~Z`&9V}X z8}Xm3)*b{)*Za+PY}$X@&*|6jPtRNL-qa|tkf~A9Q#eEPidI>F5|RNSB_PMy71DK) zBvphH0r@z6=FB+Mvy>C8sHg|%H+0Z<$sSs-S_>M+aFzmcIhn)8UTi*}<;MFB93Y7B z{`KoG|Ff5BZw3Buk{Tzj_4QfU&q3<#r@Rks{#KY;W;&nMS5IYcrMu`X4BI@iBOw_u z%n&N{V5!7Td2Dox<iHDyaQB+veMk265ZSvomr1Uwp&?4LRhbI9ja_MccCOxUP%$Gb zE2+ju58k{s*OEs5U0?QWM3a|`PL7VZ#;faHy$ua=1WdCkEap9go}IT3{fxICTtG98 z7cG=$6sMMnffJ8|KYXJk0hxT(`%BafhSmB(SuS1YiE&6z>Ya0AoOLFJk1KM3U!T3y z>z%_Q1rk|4{La+Zx3rjy?ImtFW4WAO_Dh#=qXVV~z3+RXkkEt5{&DS?^c}3$!GPS= zt6f<5&6Fu98NG)o@xA&|4N?&0jEoY+Kw0NkEDguY%rkE*;pW&eLZl+<roPMCQ*Z1y zbgHpVGGXDt`K^4-dzj^CIs+PcH{w<odOimUL<r(Y-^{g}?U)w)4mVKeFUX$7K_ZX5 z)(aI=+L7}ReK9jH4x-FALuDy<oF`AbcKX_mKQ2!(7{!NW5-s8LHA~YdWZ5k_<EX<j zAM2XtR+XuZ*ZJayHu#ryufehznmzi^SnP;B6Ec8K;X+;~4ZXMhp@Mp~r7GW}*4?WD z=?-=E4fPP1%m2q<xWr2LfP#Y^qvqaq=&8!ajz0e=``Jms^Gd|Q&2P@B3CapQSA;l) zJ^DGSByBXiVEPC(JBC}Z$jma}gBd~GStT-r^MXMUO%(QJXd2lglhZ#;44j0TlQW=H zJzKaJPs%4)X?6h&OZKmgUg_NNx8Ua`N_hn<P52TUz3Kr8YB>n-{C)E*xuOkLPA;$q zJN&%O;$3`Fn54FxygmmZ>{OtAJ}rHlSuZ)cEweK#`?K$z7Egxf^?2cX(?sb@(^@`X z5SF*rQ|$VoZIFe(#C)!-X6n;8F3~HwMGqH^S1~<Dt#vEMMWC`T;R%7{sZ&l=Rh)6u zoGBl}+75-F=+q5}Yqa>qTi?`OLET6;bk)ki=e*^El}Xs7tM7rn1G%9#{=S4)Y5*be z_(#VHG1K5?@ixvARbE}Wo5Anmhyo1r&K9#sA8HmBRIANS>0ZWyug04(%33u|XCV5S zY-rUSCE}m*5xQ8-P7r=G6DAh5Ik3F))`#?HdY^Z#4ySrlekaqSZe;3*2$L6^=g**} zEGEQ`24{!SSbsVb_RW+p;afe$5mwYp>v|=5m3U>Y>@OZmcJ$dmRK0F`{PtH<On?I) zgU4aBsh$D4@fwG?Q^3RN3_xnxx!ThO4g$9Bj61>V{3j~oyr$y3<dWG-GBf)$2u4B8 zwRgW_53r@=bW;(YIYe9SrH>(muhiAc>*z*@9ejVGB;O;*le0!tb@}P1WEr=1xOLU* znD0ZCig|C83$AXdnvzlvaQWud<GJado89)3*(5lr-+Zlhe6^!@VgM7CU!jWcq&PxJ zP8*d>YkhMdHqpJew^ZTF`zG@q$j`&7c--QAELPp@lc}H64IPfVM!q|c)shbyIl0Ge zhk_(i3#@bwhiK;2)m;y-;#?hzsDF%eegi5d3p8oEoIo2jf)fo|oYXUNajIc+{c<=L zy**#@d(ag1fv_bL9>Vm~JHDz?s3J)n()X)flkLkQ9)!tXBsYA_jlYXsAZyx*o)4Lr z-JGh$U+fscERMDKMtE0Cv`TpIR_u6Ih|Rs=JJNAMSze@|R@SXXJALze?@OLzCCi~w z?L;4#N(QbN``)|#BqoQwNnsL=fu=F7zY2%$-IDLK6Orx2i+T#r5Pm%sx0-5I`8{nW zam2INi7p_aJ<{gDDmryoo%!bknT9CJdGR*(RG^B>S(DJ)m}^Ot)GAFD(I*9qWVOav zD`sF~I33mGzouv45Y@|d*W=i{^_is+ZU5_XvyAhMmS<lzn4Wp0HpMw}=8?|hHbYAx zSj;PL+Q;mw%d5k*RO5_qZTCw5#7%F~R+{FTtR=)iSdZ7t@6hWQ*7h$+t;x~*m{t2~ zYLPu>1B1E0d`<Maw|M>A)5+Gi1+{*l@35n>wz5K`$4lx_4{Fph@w!P2Jij>1dMVY$ zoVyK<6mhPXqXS*_UK;Sy4?d{4B-tidN%N@fOj@vORWRd2RywxCIcr;;nzFuZ?Bj#t lQLX|bsTql%3w~>dM_2TfUfo&mB}KtMDhjvd^KV)_`agXE3giF) literal 0 HcmV?d00001 diff --git a/docs/internals/eventloop.html.in b/docs/internals/eventloop.html.in new file mode 100644 index 0000000..93d9e8d --- /dev/null +++ b/docs/internals/eventloop.html.in @@ -0,0 +1,106 @@ +<?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>Libvirt's event loop</h1> + + <ul id="toc"></ul> + + <p> + This page describes the event loop approach used in + libvirt. Both server and client. + </p> + + <h2><a name="event_loop">Event driven programming</a></h2> + + <p>Traditionally, a program simply ran once, then terminated. + This type of program was very common in the early days of + computing, and lacked any form of user interactivity. This is + still used frequently, particularly in small one purpose + programs.</p> + + <p>However, that approach is not suitable for all the types + of applications. For instance graphical applications spend + most of their run time waiting for an input from user. Only + after it happened (in our example a button was clicked, a key + pressed, etc.) an event is generated to which they respond + by executing desired function. If generalized, this is how + many long running programs (daemons) work. Even those who are + not waiting for direct user input and have no graphical + interface. Such as Libvirt.</p> + + <img alt="event loop" src="event_loop_simple.png"/> + + <p>In Libvirt this approach is used in combination with + <code>poll(2)</code> as all the communication with its + clients (and domains it manages too) happens through sockets. + Therefore whenever new client connects, it is given exclusive + file descriptor which is then watched for incoming events, + e.g. messages. </p> + + <h2><a name="api">The event loop API</a></h2> + + <p>To work with event loop from our code we have plenty of + APIs.</p> + + <ul> + <li><code>virEventAddHandle</code>: Registers a + callback for monitoring file handle events.</li> + <li><code>virEventUpdateHandle</code>: Change set of events + monitored file handle is being watched for.</li> + <li><code>virEventRemoveHandle</code>: Unregisters + previously registered file handle so that it is no + longer monitored for any events.</li> + <li><code>virEventAddTimeout</code>: Registers a + callback for timer event.</li> + <li><code>virEventUpdateTimeout</code>: Changes frequency + for a timer.</li> + <li><code>virEventRemoveTimeout</code>: Unregisters + a timer.</li> + </ul> + + <p>For more information on these APIs continue reading <a + href="../html/libvirt-libvirt-event.html">here</a>.</p> + + <h2><a name="worker_pool">Worker pool</a></h2> + + <p>Looking back at the image above we can see one big + limitation. While processing a message event loop is blocked + and for an outside observer unresponsive. This is not + acceptable for Libvirt. Therefore we have came up with the + following solution.</p> + + <img alt="event loop" src="event_loop_worker.png"/> + + <p>The event loop does only necessary minimum and hand over + message processing to another thread. In fact, there can be + as many processing threads as configured increasing + processing power.</p> + + <p>To break this high level description into smaller pieces, + here is what happens when user calls an API:</p> + <ol> + <li>User (or management application) calls a Libvirt API. + Depending on the connection URI, this may or may not + involve server. Well, for the sake of our + demonstration we assume the former.</li> + <li>Remote driver encodes the API among it's arguments + into an <a href="rpc.html">RPC message</a> and sends + it to the server.</li> + <li>Here, server is waiting in <code>poll(2)</code> for + an event, like incoming message.</li> + <li>As soon as the first bytes of message are received, + even loop wakes up and server starts reading the + whole message.</li> + <li>Once fully read, the event loop notifies threads + known as worker threads from which one picks the incoming + message, decodes and process it.</li> + <li>As soon as API execution is finished, a reply is sent + to the client.</li> + </ol> + + <p>In case that there's no free worker to process an incoming + message in step 5, message is placed at the end of a message + queue and is processed in next iteration.</p> + </body> +</html> diff --git a/docs/sitemap.html.in b/docs/sitemap.html.in index d5cf723..757d5aa 100644 --- a/docs/sitemap.html.in +++ b/docs/sitemap.html.in @@ -397,6 +397,10 @@ <span>Adding new public libvirt APIs</span> </li> <li> + <a href="internals/eventloop.html">Event loop and worker pool</a> + <span>Libvirt's event loop and worker pool mode</span> + </li> + <li> <a href="internals/command.html">Spawning commands</a> <span>Spawning commands from libvirt driver code</span> </li> -- 2.8.3

2 1

[libvirt] [PATCH 00/32] node_device_udev: cleanups
by Ján Tomko 06 Jun '16

06 Jun '16

Mostly style changes and refactors. Ján Tomko (32): Initialize ret to -1 in nodeStateInitialize Split out pciaccess (de)initialization node_device_udev: initialize libpciaccess after the driver lock Reformat nodeStateCleanup Do not call nodeStateCleanup on early initialization error Assign node device driver private data earlier udevGetDMIData: remove unused variable udevProcessSCSIHost: use STRSKIP Rewrite usage of StrToLong_ui in udevProcess{PCI,SCSI} Remove udevStrToLong_ull Remove udevStrToLong_ui Remove udevStrToLong_i Do not VIR_STRDUP the string in udevGetDeviceProperty Move udevHasDeviceProperty earlier udevProcessFloppy; remove unnecessary allocation Fix the return value in udevKludgeStorageType Rewrite disk type checking in udevProcessStorage Only return two values in udevGetStringProperty Only return two values in udevGetUintProperty Remove extra allocation in udevGetDeviceSysfsAttr Only return two values in udevGetStringSysfsAttr Only return two values in udevGetIntSysfsAttr Only return two values in udevGetUintSysfsAttr Remove PROPERTY_* constants node_device_udev: switch to using virReportError udevProcessStorage: trim all whitespace from model and vendor Reformat udevProcessRemoveableMedia udevSetupSystemDev: return if allocation fails node_device_udev: remove yoda condition node_device_udev: remove unnecessary ret variables node_device_udev: rename labels to cleanup Add nomatch filters when enumerating udev devices src/node_device/node_device_udev.c | 1156 ++++++++++++++---------------------- src/node_device/node_device_udev.h | 3 - 2 files changed, 437 insertions(+), 722 deletions(-) -- 2.7.3

2 65

[libvirt] [PATCH 0/3] Fix 'make rpm' failure with systemd 230
by Andrea Bolognani 06 Jun '16

06 Jun '16

Turns out the libsystemd-daemon library, which we've been using for the sd_notify() feature, has been deprecated for a long time and has finally been removed as of systemd 230. This means 'make rpm' no longer works on Fedora rawhide, and that other distributions such as Debian sid are silently disabling the sd_notify() feature. Switch from libsystemd-daemon to libsystemd. Andrea Bolognani (3): maint: Use libsystemd instead of libsystemd-daemon spec: Enable libsystemd on Fedora >= 21, RHEL >= 7 systemd: Guard the call to sd_notify() properly configure.ac | 4 ++-- libvirt.spec.in | 18 +++++++++++------- m4/{virt-systemd-daemon.m4 => virt-libsystemd.m4} | 14 +++++++------- src/Makefile.am | 4 ++-- src/util/virsystemd.c | 4 ++-- 5 files changed, 24 insertions(+), 20 deletions(-) rename m4/{virt-systemd-daemon.m4 => virt-libsystemd.m4} (73%) -- 2.5.5

3 5

[libvirt] [PATCH v3 0/5] Adjustments for secinfo command line build and secret changes
by John Ferlan 06 Jun '16

06 Jun '16

v2: http://www.redhat.com/archives/libvir-list/2016-May/msg02140.html Patch 1 is the same as v2 patch 1 Patch 2 is already ACK'd - it was the previous patch 5 Patch 3 & 4 replace patches 2-4 from v2 Patch 5 is the same as v2 patch 6. John Ferlan (5): qemu: Move and rename qemuBuildObjectCommandlineFromJSON storage: Use virSecretGetSecretString qemu: Use virJSONValueObjectCreate for master key qemu: Remove need for qemuBuildSecretInfoProps secret: Move virStorageSecretType to secret_util and rename cfg.mk | 2 +- po/POTFILES.in | 1 + src/Makefile.am | 2 + src/conf/secret_conf.h | 2 +- src/libvirt_private.syms | 4 + src/libxl/libxl_conf.c | 2 +- src/qemu/qemu_command.c | 202 +++++++----------------------------- src/qemu/qemu_command.h | 4 - src/qemu/qemu_domain.c | 4 +- src/secret/secret_util.c | 18 ++-- src/secret/secret_util.h | 22 +++- src/storage/storage_backend_iscsi.c | 55 ++-------- src/storage/storage_backend_rbd.c | 49 +-------- src/util/virqemu.c | 142 +++++++++++++++++++++++++ src/util/virqemu.h | 34 ++++++ src/util/virstoragefile.c | 33 +++--- src/util/virstoragefile.h | 17 +-- tests/qemuargv2xmltest.c | 4 +- tests/qemucommandutiltest.c | 9 +- 19 files changed, 294 insertions(+), 312 deletions(-) create mode 100644 src/util/virqemu.c create mode 100644 src/util/virqemu.h -- 2.5.5

2 17

[libvirt] [PATCH v5 0/6] [REPOST] introduce new listen types for graphics
by Pavel Hrdina 06 Jun '16

06 Jun '16

There is no change since v4 [1], I've only rebased patches that are still waiting for review and removed the ones that are already pushed. [1] <https://www.redhat.com/archives/libvir-list/2016-May/msg01438.html> Pavel Hrdina (6): graphics: introduce listen type socket and use it for VNC qemu_capabilites: add QEMU_CAPS_SPICE_UNIX spice: add support for listen type socket spice: introduce spice_auto_unix_socket config option spice: introduce listen type none vnc: add support for listen type none docs/formatdomain.html.in | 28 +++ docs/schemas/domaincommon.rng | 15 ++ src/conf/domain_conf.c | 245 ++++++++++++++++++--- src/conf/domain_conf.h | 8 +- src/libvirt_private.syms | 1 + src/qemu/libvirtd_qemu.aug | 1 + src/qemu/qemu.conf | 17 +- src/qemu/qemu_capabilities.c | 3 + src/qemu/qemu_capabilities.h | 3 + src/qemu/qemu_command.c | 130 ++++++----- src/qemu/qemu_conf.c | 1 + src/qemu/qemu_conf.h | 1 + src/qemu/qemu_domain.c | 28 ++- src/qemu/qemu_hotplug.c | 9 + src/qemu/qemu_migration.c | 47 +++- src/qemu/qemu_parse_command.c | 2 +- src/qemu/qemu_process.c | 46 +++- src/qemu/test_libvirtd_qemu.aug.in | 1 + src/security/virt-aa-helper.c | 15 +- ...ric-graphics-vnc-socket-attr-listen-address.xml | 30 +++ ...hics-vnc-socket-attr-listen-socket-mismatch.xml | 30 +++ ...eric-graphics-vnc-socket-attr-listen-socket.xml | 30 +++ ...ric-graphics-vnc-socket-attr-listen-address.xml | 30 +++ ...eric-graphics-vnc-socket-attr-listen-socket.xml | 30 +++ .../generic-graphics-vnc-socket-listen.xml | 4 +- .../generic-graphics-vnc-socket.xml | 4 +- tests/genericxml2xmltest.c | 4 + .../qemuargv2xml-graphics-vnc-socket.xml | 4 +- tests/qemucapabilitiesdata/caps_2.4.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.5.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 1 + ...emuxml2argv-graphics-spice-auto-socket-cfg.args | 20 ++ ...qemuxml2argv-graphics-spice-auto-socket-cfg.xml | 30 +++ .../qemuxml2argv-graphics-spice-auto-socket.args | 20 ++ .../qemuxml2argv-graphics-spice-auto-socket.xml | 30 +++ .../qemuxml2argv-graphics-spice-socket.args | 20 ++ .../qemuxml2argv-graphics-spice-socket.xml | 30 +++ .../qemuxml2argv-graphics-vnc-auto-socket.args | 20 ++ .../qemuxml2argv-graphics-vnc-auto-socket.xml | 30 +++ .../qemuxml2argv-graphics-vnc-none.args | 20 ++ .../qemuxml2argv-graphics-vnc-none.xml | 30 +++ .../qemuxml2argv-graphics-vnc-socket.args | 4 +- .../qemuxml2argv-graphics-vnc-socket.xml | 10 +- .../qemuxml2argv-video-virtio-gpu-spice-gl.args | 2 +- tests/qemuxml2argvtest.c | 14 ++ ...muxml2xmlout-graphics-spice-auto-socket-cfg.xml | 35 +++ .../qemuxml2xmlout-graphics-spice-auto-socket.xml | 35 +++ .../qemuxml2xmlout-graphics-spice-socket.xml | 35 +++ .../qemuxml2xmlout-graphics-vnc-auto-socket.xml | 35 +++ ...graphics-vnc-remove-generated-socket-active.xml | 4 +- .../qemuxml2xmlout-graphics-vnc-socket.xml | 35 +++ .../qemuxml2xmlout-video-virtio-gpu-spice-gl.xml | 4 +- tests/qemuxml2xmltest.c | 8 + 53 files changed, 1096 insertions(+), 145 deletions(-) create mode 100644 tests/genericxml2xmlindata/generic-graphics-vnc-socket-attr-listen-address.xml create mode 100644 tests/genericxml2xmlindata/generic-graphics-vnc-socket-attr-listen-socket-mismatch.xml create mode 100644 tests/genericxml2xmlindata/generic-graphics-vnc-socket-attr-listen-socket.xml create mode 100644 tests/genericxml2xmloutdata/generic-graphics-vnc-socket-attr-listen-address.xml create mode 100644 tests/genericxml2xmloutdata/generic-graphics-vnc-socket-attr-listen-socket.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-auto-socket-cfg.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-auto-socket-cfg.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-auto-socket.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-auto-socket.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-socket.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-socket.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-auto-socket.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-auto-socket.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-none.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-none.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-graphics-spice-auto-socket-cfg.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-graphics-spice-auto-socket.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-graphics-spice-socket.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-graphics-vnc-auto-socket.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-graphics-vnc-socket.xml -- 2.8.3

2 9

[libvirt] [PATCH] conf: always format os.bootloaderArgs if set
by Fabian Freyer 06 Jun '16

06 Jun '16

At the moment the bootloader arguments never get formatted if the bootloader is unset. However, in cases where the bootloader defaults to a default value when unset, specifying bootloader arguments does make sense. --- src/conf/domain_conf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 568c699..66bba6e 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -22608,6 +22608,8 @@ virDomainDefFormatInternal(virDomainDefPtr def, if (def->os.bootloader) { virBufferEscapeString(buf, "<bootloader>%s</bootloader>\n", def->os.bootloader); + } + if (def->os.bootloaderArgs) { virBufferEscapeString(buf, "<bootloader_args>%s</bootloader_args>\n", def->os.bootloaderArgs); -- 2.7.0

3 3

[libvirt] [RFC] extend USB controller 'model' to support xen pvusb
by Chun Yan Liu 06 Jun '16

06 Jun '16

Hi, List, Currently libxl supports pvusb type USB controller hotplug and in creation time, I'd like to add the support in libvirt libxl driver too. According to current USB controller schema, we need to extend 'model' to support xen PVUSB type. Since libxl support USB controller of version 1.1 and version 2.0, we may add two models: "pvusb1" (pvusb type, version 1.1) and "pvusb2" (pvusb type, version 2.0). Do you have any opinions on that? If it is agreed, I'll try to implement. Thanks, Chunyan

1 0

[libvirt] Release of libvirt-1.3.5
by Daniel Veillard 04 Jun '16

04 Jun '16

Things didn't go as planned, I got stuck in travels etc, so I didn't push rc2 as expected earlier and I could not do any release on Friday. So I broke the rule and pushed today without an rc2, that's exceptional, but there was no negative feedback on RC1 and I didn't want to freeze development any longer, so the tree is open. Signed tarball and rpms are available at the usual place: ftp://libvirt.org/libvirt/ I also made a python binding release but it is virtually identical to 1.3.4 This is a very large release with more than 500 commits ! There is clear user visible new features, and an lot of code (and spec!) cleanups. Nonetheless libvirt-1.3.5.tar.gz is 5 Megabytes bigger than libvirt-1.3.4.tar.gz, likely due to a large update to localizations ! As usual a fair amount of bug fixes are included: Features: - admin: Introduce virAdmClientGetInfo API (Erik Skultety) - Various improvement on bhyve driver (Roman Bogorodskiy, Fabian Freyer) - Various improvement on libxl driver (Jim Fehlig, Chunyan Liu) Documentation: - Refresh po files from zanata (Daniel P. Berrange) - cpu: Fix documentation of cpuGetModels (Jiri Denemark) - docs: fix <spice><gl enable> since version (Ján Tomko) - docs: formatdomain: document virtio-mmio device addresses (Cole Robinson) - schemas: Improve outdated comment (Christophe Fergeau) - docs: Fix disk "volume" description (John Ferlan) - docs: fix version number in vlan tagging documentation (Laine Stump) - tools: virt-host-validate: fix missing translation marker (Bjoern Walk) - docs: Reformat the Controllers description (John Ferlan) - docs: clarify disk iothread support (John Ferlan) - docs: align spelling of S390 (Boris Fiuczynski) Portability: - virDomainChrGetDomainPtrsInternal: Return an integer (Michal Privoznik) - virSocketAddrIsPrivate: Work on 32bits platforms (Michal Privoznik) - makefile: fix build on systems where gnutls is not in /usr/include (Pavel Hrdina) - util: fix build without GNUTLS (Mikhail Feoktistov) - tests: qemu: test <address type='pci'/> with aarch64 (Cole Robinson) - tests: Link virtestmock with probes (Michal Privoznik) - libvit.spec.in: Add missing BuildRequires against sheepdog (Daniel P. Berrange) - libvirt.spec.in: remove client only build option (Daniel P. Berrange) - uml: only build on Linux (Roman Bogorodskiy) - virNetServerClientNewPostExecRestart: Avoid align problems (Michal Privoznik) - build: fix 32-bit build of admin (Eric Blake) Bug Fixes: - Fix building with -Og (Martin Kletzander) - virPerfEventIsEnabled: Don't crash on shut off domains (Michal Privoznik) - virDomainFormatSchedDef: Avoid false positive NULL dereference (Michal Privoznik) - ppc64Compute: Avoid possible NULL dereference (Michal Privoznik) - network: restart dnsmasq after adding/removing txt and srv records (Laine Stump) - QXL: fix reloading of vram64 attribute (Pavel Hrdina) - esxStorageVolGetXMLDesc: Lookup SCSI lun properly (Michal Privoznik) - qemuMonitorTextGetAllBlockStatsInfo: Fix line validation (Michal Privoznik) - lxc: Fix virLXCDomainObjBeginJob position in lxcDomainSetMemoryParameters (Katerina Koukiou) - esx: do not store escaped password in esxVI_Context. (Dawid Zamirski) - qemu: Fix error message when PCI bridge has index <= bus (Andrea Bolognani) - daemon: cleanup state drivers in order reverse to init order (Nikolay Shirokovskiy) - esx: use newer virtualHW version for 5.1+ hosts (Dawid Zamirski) - esx: Add VMCI device for virtualHW >= 7 (Dawid Zamirski) - esx: use lsilogic adapter type in vol create. (Dawid Zamirski) - libxl: default to qemu driver for network disks (Jim Fehlig) - lxc: Fix lxcDomainDestroyFlags endjob processing (John Ferlan) - qemu: Remove unused persistentAddrs (John Ferlan) - xenconfig: xm: check for driver on disk format (Joao Martins) - nwfilter: fix lock order deadlock (Maxim Nestratov) - maint: fix syntax-check sc_prohibit_int_ijk exclude rule (Pavel Hrdina) - vz: add error code for case if vm is already stopped (Mikhail Feoktistov) - storage: do not clear vols before volume upload (Ján Tomko) - util: Fix error path for virPCIGetVirtualFunctions (John Ferlan) - tests: Add forgotten backslash (Michal Privoznik) - qemu: bulk stats: Don't access possibly blocked storage (Peter Krempa) - qemu_cgroup: allow access to /dev/dri for virtio-vga (Ján Tomko) - Revert "qemu_hotplug: fix checking graphics ports" (Pavel Hrdina) - util: Remove disabling of autologin for iscsi-targets (Fritz Elfert) - xenFormatNet: correct `type=netfront' to 'type=vif' to match libxl (Chunyan Liu) - libxl: Free migration cookie (John Ferlan) - qemu: command: Use -name guest= if available (Cole Robinson) - qemu: command: escape commas in chardev socket path (Cole Robinson) - qemu: command: escape commas in VNC socket path (Cole Robinson) - qemu: command: escape commas in secret master path (Cole Robinson) - qemu: command: escape commas in VM name (Cole Robinson) - admin: Fix passing an incorrect readonly attribute to virNetServerServiceNew (Erik Skultety) - cpuGetModels: Fix memory leak on error (Jiri Denemark) - libxl: don't attempt to probe a non-existent emulator (Jim Fehlig) - network: log error when <bandwidth> is requested for hostdev interfaces (Laine Stump) - Call per-device post-parse callback even on implicit video (Ján Tomko) - util: Fix virGetLastErrorMessage to return proper error when 'err' is NULL (Erik Skultety) - storage: Fix virStorageBackendDiskDeleteVol for device mapper (John Ferlan) - storage: Fix algorithm generating path names for devmapper (John Ferlan) - storage: Need to clear pool prior to calling the refreshPool (John Ferlan) - storage: Fix regression cloning volume into a logical pool (John Ferlan) - conf: log error when incorrect PCI root controller is added to domain (Laine Stump) - conf: don't redefine virDomainCapsDeviceHostdev (Roman Bogorodskiy) - conf: Fix error path in virNodeDevPCICapabilityParseXML (John Ferlan) - util: polkit: Fix polkit agent startup (Peter Krempa) - qemu: domain: Don't treat unknown storage type as not having backing chain (Peter Krempa) - qemu: Reject invalid block copy targets for <disk device='lun'> (Peter Krempa) - lxc: Fix wrong error message on disk hotplug (Peter Krempa) - domain_conf: fix migration/managedsave with usb keyboard (Pavel Hrdina) - qemu: Refresh RTC adjustment on qemuProcessReconnect (Michal Privoznik) - qemu: fix error message for default panic device (Boris Fiuczynski) - fdstream: don't raise error on SIGPIPE if abort requested (Cole Robinson) - daemon: stream: Don't force error when client aborts (Cole Robinson) - daemon: stream: set stream->closed on removal (Cole Robinson) - daemon: stream: don't update events if stream->closed (Cole Robinson) - daemon: stream: Close stream on send failure (Cole Robinson) - fdstream: Raise explicit error when iohelper gets SIGPIPE (Cole Robinson) - nwfilter: Save config to disk if we generated a UUID (Cole Robinson) - nwfilter: Fix potential locking problems on ObjLoad failure (Cole Robinson) - network: Fix segfault on daemon reload (Cole Robinson) - send default USB controller in xml to destination during migration (Shivaprasad G Bhat) - conf: storage: pool: reject name containing '/' (Cole Robinson) - conf: network: reject name containing '/' (Cole Robinson) - conf: domain: reject name containing '/' (Cole Robinson) - qemu: hotplug: Fix possible memory leak of props (John Ferlan) - qemu: process: Fix failure semantics for perf events (Peter Krempa) - qemu: process: Refresh ejectable media tray state on VM start (Peter Krempa) - tools: Fix connect command (Martin Kletzander) Improvements: - Drop virPerfGetEventFd (Michal Privoznik) - virNetDevBridgeGet: Don't require users to virNetDevSetupControl (Michal Privoznik) - apibuild: Substitute only pure number tokens (Michal Privoznik) - build: use gnulib's unsetenv (Michal Privoznik) - Turn 1<<31 into 1U<<31 (Michal Privoznik) - docs: Teach apibuild to deal with (1U << 31) too (Michal Privoznik) - maint: update to latest gnulib (Eric Blake) - spec: Advertise nvram paths of official fedora edk2 builds (Cole Robinson) - maint: update to latest gnulib (Eric Blake) - esx: add pciBridge devices when SCSI is used (Dawid Zamirski) - conf: permit auto-assignment of controller indexes (Laine Stump) - conf: make virDomainControllerFindUnusedIndex() more generally usable (Laine Stump) - conf/qemu: make IS_USB2_CONTROLLER globally available (Laine Stump) - libxl: add .domainInterfaceAddresses (Chunyan Liu) - security: label the slic_table (Ján Tomko) - qemu: format SLIC ACPI table command line (Ján Tomko) - conf: add <acpi><table> to <os> (Ján Tomko) - qemucapstest: replace caps-1.6.50 with updated caps-1.7.0 (Pavel Hrdina) - qemu: Remove virDomainLiveConfigHelperMethod from qemuDomainSetSchedulerParametersFlags (Peter Krempa) - qemu: Remove virDomainLiveConfigHelperMethod from qemuDomainSetBlockIoTune (Peter Krempa) - qemu: Refactor qemuDomainGetSchedulerParametersFlags (Peter Krempa) - conf: Change virDomainCputune member 'shares' to unsigned long long (Peter Krempa) - qemu: Remove virDomainLiveConfigHelperMethod from qemuDomainGetSchedulerParametersFlags (Peter Krempa) - qemu: Refactor qemuDomainGetBlkioParameters (Peter Krempa) - qemu: Remove virDomainLiveConfigHelperMethod from qemuDomainGetBlkioParameters (Peter Krempa) - qemu: Remove virDomainLiveConfigHelperMethod from qemuDomainSetMemoryParameters (Peter Krempa) - qemu: Refactor typed params assignment in qemuDomainGetBlockIoTune (Peter Krempa) - qemu: Replace virDomainLiveConfigHelperMethod in qemuDomainGetBlockIoTune (Peter Krempa) - qemu: monitor: Remove 'supportMaxOptions' argument from qemuMonitorGetBlockIoThrottle (Peter Krempa) - qemu: driver: Allow disk update of startupPolicy/snapshot for all disks (Peter Krempa) - qemu: driver: Move around code to avoid need to rollback (Peter Krempa) - Call qemuDomainObjEndJob when qemuCaps is null during hotplug (Shivaprasad G Bhat) - Unref the cfg in qemuDomainAttachHostPCIDevice() (Shivaprasad G Bhat) - qemu: Remove dead code (John Ferlan) - conf: nodedev: Set PCI_PHYSICAL_FUNCTION flag more carefully (Andrea Bolognani) - pci: Fix virPCIGetPhysicalFunction()'s callers (Andrea Bolognani) - pci: Document virPCIGetPhysicalFunction() (Andrea Bolognani) - pci: Initialize return location in virPCIGetPhysicalFunction() (Andrea Bolognani) - qemu: hotplug: wait for the tray to eject only for drives with a tray (Peter Krempa) - qemu: hotplug: Fix error reported when cdrom tray is locked (Peter Krempa) - qemu: hotplug: Extract code for waiting for tray eject (Peter Krempa) - qemu: process: Fix and improve disk data extraction (Peter Krempa) - qemu: Move and rename qemuDomainCheckEjectableMedia to qemuProcessRefreshDisks (Peter Krempa) - qemu: Extract more information about qemu drives (Peter Krempa) - qemu: Move struct qemuDomainDiskInfo to qemu_domain.h (Peter Krempa) - lxc: support <interface type='ethernet'> (Laine Stump) - lxc: completely rework reference counting (Katerina Koukiou) - lxc: use job functions in lxcDomainLxcOpenNamespace & lxcDomainSendProcessSignal (Katerina Koukiou) - qemucapstest: update caps for qemu-2.6.0 (Pavel Hrdina) - qemucapstest: update caps for qemu-2.5.0 (Pavel Hrdina) - qemucapstest: update caps for qemu-2.4.0 (Pavel Hrdina) - qemucapstest: update caps for qemu-2.1.1 (Pavel Hrdina) - qemucapstest: update caps for qemu-1.6.0 (Pavel Hrdina) - qemucapstest: update caps for qemu-1.5.3 (Pavel Hrdina) - qemucapstest: update caps for qemu-1.4.2 (Pavel Hrdina) - qemucapstest: update caps for qemu-1.3.1 (Pavel Hrdina) - qemucapstest: update caps for qemu-1.2.2 (Pavel Hrdina) - qemuxml2argvtest: skip test that depends on gnutls_cipher_encrypt() (Pavel Hrdina) - storage: Replace VIR_ERROR with standard vir*Error in state driver init (Jovanka Gulicoska) - nwfilter: Replace VIR_ERROR with standard vir*Error in state driver init (Jovanka Gulicoska) - libxl: Replace VIR_ERROR with standard vir*Error in state driver init (Jovanka Gulicoska) - bhyve: Replace VIR_ERROR with standard vir*Error in state driver init (Jovanka Gulicoska) - qemu: simplify addition of USB controller in qemuParseCommandLine (Laine Stump) - Deprecate QEMU_CAPS_PCIDEVICE (Ján Tomko) - Deprecate QEMU_CAPS_DEVICE (Ján Tomko) - Deprecate QEMU_CAPS_DRIVE_READONLY (Ján Tomko) - qemu: always assume QEMU_CAPS_DRIVE_READONLY (Ján Tomko) - tests: always assume QEMU_CAPS_DRIVE_READONLY (Ján Tomko) - tests: remove <readonly/> from IDE disks (Ján Tomko) - tests: remove disk-drive-fat test (Ján Tomko) - qemu: assume QEMU_CAPS_DEVICE almost everywhere (Ján Tomko) - qemu_command: assume QEMU_CAPS_DEVICE (Ján Tomko) - qemu: auto-assign addresses when <address type='pci'/> is specified (Laine Stump) - bhyve: auto-assign addresses when <address type='pci'/> is specified (Laine Stump) - conf: allow type='pci' addresses with no address attributes specified (Laine Stump) - conf: new functions to check if PCI address is wanted/present (Laine Stump) - conf: move virDomainDeviceInfo definition from domain_conf.h to device_conf.h (Laine Stump) - virtestmock: Mock stat() properly (Michal Privoznik) - tests: Drop VIR_MOCK_CALL_STAT (Michal Privoznik) - qemu: Utilize qemu secret objects for RBD auth/secret (John Ferlan) - tests: Allow multiple mock libraries (Peter Krempa) - qemu: Introduce qemuDomainSecretSetup (John Ferlan) - util: Introduce virCryptoGenerateRandom (John Ferlan) - util: Introduce encryption APIs (John Ferlan) - tests: Add mock for virRandomBytes (John Ferlan) - qemu: parse: Handle suffixes for -m memory (Nishith Shah) - qemu: parse: Use qemuParseCommandLineMem for -m memory (Nishith Shah) - qemu_command: refactor spice channel code (Pavel Hrdina) - qemu_process: separate graphics socket and address generation (Pavel Hrdina) - graphics: resolve address for listen type network in qemu_process (Pavel Hrdina) - qemu_command: move sasl parameter after port and addr definition (Pavel Hrdina) - domain_conf: introduce virDomainGraphicsListenDefFormatAddr (Pavel Hrdina) - graphics: rename gListen to glisten (Pavel Hrdina) - tests: cleanup vnc auto socket test (Pavel Hrdina) - qemu_domain: add a empty listen type address if we remove socket for VNC (Pavel Hrdina) - cpu: Rework CPU map loading (Jiri Denemark) - cpu_ppc64: Use array of models in CPU map (Jiri Denemark) - cpu_ppc64: Use array of vendors in CPU map (Jiri Denemark) - cpu_x86: Use array of features in CPU map (Jiri Denemark) - cpu_x86: Use array of vendors in CPU map (Jiri Denemark) - cpu_x86: Use array of models in CPU map (Jiri Denemark) - Do not mask QEMU_CAPS_DEVICE in qemuBuildDriveStr (Ján Tomko) - Introduce qemuDiskBusNeedsDeviceArg (Ján Tomko) - Assume QEMU_CAPS_DEVICE in qemuBuildDiskDriveCommandLine (Ján Tomko) - Remove DISK_BUS_XEN support from qemuBuildDiskDriveCommandLine (Ján Tomko) - qemu: always add -nodefaults (Ján Tomko) - qemu: process: Drop !QEMU_CAPS_DEVICE code (Cole Robinson) - Remove qemuProcessInitPCIAddresses with dependencies (Ján Tomko) - qemu: driver: Fix function header alignment of some functions (Peter Krempa) - conf: disk: Rename virDomainDiskDefValidate to virDomainDiskDefParseValidate (Peter Krempa) - util: Remove need for STATIC_ANALYSIS check (John Ferlan) - util: Adjust return for virPCIGetDeviceAddressFromSysfsLink (John Ferlan) - util: Remove need for ret in virPCIGetPhysicalFunction (John Ferlan) - tests: nodeinfotest: Remove virSaveLastError() usage (Cole Robinson) - More usage of virGetLastErrorMessage (Jovanka Gulicoska) - tests: More usage of virGetLastErrorMessage() (Jovanka Gulicoska) - lxc: use job functions in lxcDomain* functions that perform modify actions. (Katerina Koukiou) - lxc: use job functions in lxcDomain* functions that do query operations. (Katerina Koukiou) - lxc: add job functions in lxcDomainSetAutostart (Katerina Koukiou) - lxc: use job functions in lxcDomain{AttachDeviceFlags, DetachDeviceFlags, UpdateDeviceFlags} (Katerina Koukiou) - lxc: use job functions in lxcDomain{Suspend, Resume} (Katerina Koukiou) - lxc: use job functions in lxcDomainSetMemoryFlags (Katerina Koukiou) - lxc: use job functions in lxcDomain{CreateXMLWithFiles, CreateWithFiles} (Katerina Koukiou) - lxc: Add job support to lxc driver (Katerina Koukiou) - qemu: driver: Separate bulk stats worker for block devices (Peter Krempa) - qemu: driver: Remove unnecessary flag in qemuDomainGetStatsBlock (Peter Krempa) - perf: add support to perf event for MBM (Qiaowei Ren) - Separate virDomainDefParseBootOptions (Ján Tomko) - vz: cleanup: define vz format of uuids (Nikolay Shirokovskiy) - vz: implement p2p migration (Nikolay Shirokovskiy) - vz: implement managed migration (Nikolay Shirokovskiy) - vz: fix const correctness case (Nikolay Shirokovskiy) - vz: save session uuid on login (Nikolay Shirokovskiy) - virt-admin: Introduce commands srv-clients-info and srv-clients-set (Erik Skultety) - admin: Introduce virAdmServerSetClientLimits (Erik Skultety) - admin: Introduce virAdmServerGetClientLimits (Erik Skultety) - admin: Introduce some public constants related to server's client limits (Erik Skultety) - virnetserver: Introduce server's client-related limits getters (Erik Skultety) - qemu: address: Remove QEMU_CAPS_DEVICE usage (Cole Robinson) - tests: qemuargv2xmltest: Drop disk for s390 aes tests (Cole Robinson) - qemu: Call virDomainDefPostParse via CONFIG hotplug (Cole Robinson) - qemu: Assign device addresses in PostParse (Cole Robinson) - domain: Add virDomainDefAssignAddressesCallback (Cole Robinson) - vz: drop prlsdkDomainHasSnapshots (Nikolay Shirokovskiy) - vz: add domain snapshots functionality (Nikolay Shirokovskiy) - iscsi: Remove initiatoriqn from virISCSIScanTargets (John Ferlan) - iscsi: Add exit status checking for virISCSIGetSession (John Ferlan) - util: Add exitstatus parameter to virCommandRunRegex (John Ferlan) - tests: Try different usable GIC versions (Andrea Bolognani) - tests: Prepare to have different usable GIC versions (Andrea Bolognani) - qemu: Add virQEMUCapsSetGICCapabilities() (Andrea Bolognani) - qemu: Automatically choose usable GIC version (Andrea Bolognani) - qemu: Add virQEMUCapsSupportsGICVersion() (Andrea Bolognani) - Change return value of VIR_APPEND*INPLACE* to void (Jiri Denemark) - Remove virDomainRNGInsert (Jiri Denemark) - xlconfigtest: add test case for type=vif in xl format (Chunyan Liu) - extract XEN_CONFIG_FORMAT_XM/XL to xen_common.h (Chunyan Liu) - bhyve: implement virConnectIsSecure (Fabian Freyer) - bhyve: Implement virConnectIsEncrypted (Fabian Freyer) - spec: Remove %defattr usage (Cole Robinson) - Revert "vz: handle sourceless cdroms" (Maxim Nestratov) - qemu: hotplug: Report error if we hit tray status timeout (Cole Robinson) - admin: include: Rename argument dmn to conn in virAdmConnectListServers (Erik Skultety) - vz: make error path code idiomatic (Nikolay Shirokovskiy) - vz: fix template ct creation (Mikhail Feoktistov) - vz: fix error message for readonly fs (Nikolay Shirokovskiy) - vz: handle sourceless cdroms (Mikhail Feoktistov) - vz: fix vzCheckUnsupportedDisks format checks for cdroms (Nikolay Shirokovskiy) - vz: remove check for auto file format for disks (Nikolay Shirokovskiy) - domain_conf: cleanup virDomainGraphicsListenDefParseXML (Pavel Hrdina) - graphics: make address attribute for listen type='address' optional (Pavel Hrdina) - domain_conf: parse listen attribute while parsing listen elements (Pavel Hrdina) - graphics: don't parse listens if socket attribute is present (Pavel Hrdina) - conf: Allow all volume modes for disk type='lun' sources (Peter Krempa) - qemu: Drop QEMU_CAPS_VIRTIO_BLK_SG_IO (Andrea Bolognani) - qemu: Drop QEMU_CAPS_CPU_HOST (Andrea Bolognani) - qemu: Drop QEMU_CAPS_PCI_ROMBAR (Andrea Bolognani) - qemu: More qemu_monitor_json cleanups (John Ferlan) - bhyve: implement virConnectIsAlive (Fabian Freyer) - qemu: command: Add qemuBufferEscapeComma (Cole Robinson) - qemu: alias: Remove QEMU_CAPS_DEVICE (Cole Robinson) - cpu: Properly report errors when parsing CPU map XML (Jiri Denemark) - cpu_x86: Check vendor early (Jiri Denemark) - cpu_x86: Don't ignore parsing errors in x86ModelLoad (Jiri Denemark) - cpu_x86: Don't ignore parsing errors in x86FeatureLoad (Jiri Denemark) - cpu_x86: Don't ignore parsing errors in x86VendorLoad (Jiri Denemark) - cpu_x86: Simplify insertions into a linked list (Jiri Denemark) - cpu_x86: Remove comparisons to NULL (Jiri Denemark) - cpu_x86: Use for loop in x86Decode (Jiri Denemark) - cpu_x86: Rename cleanup labels (Jiri Denemark) - cpu_x86: Compare CPU candidates in a separate function (Jiri Denemark) - cpu_x86: Rename struct virCPUx86DataIterator (Jiri Denemark) - cpu_x86: Rename enum compare_result (Jiri Denemark) - cpu_x86: Rename struct x86_map (Jiri Denemark) - cpu_x86: Rename struct x86_model (Jiri Denemark) - cpu_x86: Rename struct x86_kvm_feature (Jiri Denemark) - cpu_x86: Rename struct x86_feature (Jiri Denemark) - cpu_x86: Rename struct x86_vendor (Jiri Denemark) - qemu: command: Ignore QEMU_CAPS_DEVICE when building drive alias (Cole Robinson) - cpu: Add support for clflushopt and tsc_adjust Intel features (Alexander Burluka) - secret: Alter virSecretGetSecretString (John Ferlan) - qemu: domain: Fix names for functions that clear security info (Peter Krempa) - util: string: Introduce helper to determine whether a byte buffer is printable (John Ferlan) - util: string: Introduce virStringEncodeBase64 (Peter Krempa) - secret: util: Refactor virSecretGetSecretString (Peter Krempa) - util: alloc: Introduce freeing helpers that clear the memory before freeing (Peter Krempa) - capabilities: Advertise cpuselection if -cpu host is usable (Jiri Denemark) - qemuaincapstest: Give better names to test data files (Jiri Denemark) - domaincapstest: Give better names to test data files (Jiri Denemark) - domaincapstest: Use arch strings (Jiri Denemark) - tests: Introduce check-file-access.pl (Michal Privoznik) - virtestmock: Print invalid file accesses into a file (Michal Privoznik) - tests: Introduce global mock library (Michal Privoznik) - virmock.h: Introduce VIR_MOCK_CALL_STAT (Michal Privoznik) - securityselinuxhelper: Adapt to virmock.h (Michal Privoznik) - nssmock: Adapt to virmock.h (Michal Privoznik) - vircgroupmock: Adapt to virmock.h (Michal Privoznik) - virpcimock: Adapt to virmock.h (Michal Privoznik) - qemu: remove ATTRIBUTE_UNUSED in connectGetType (Fabian Freyer) - bhyve: implement virConnectGetType (Fabian Freyer) - xlconfigtests: use qemu-xen in all test data files (Jim Fehlig) - virt-aa-helper: remove replace_string and use virStringReplace instead (Pavel Hrdina) - virfile: Introduce virFileRemoveLastComponent (Michal Privoznik) - leave out the default USB controller only on i440fx during migration (Shivaprasad G Bhat) - qemu: domain: Fix name of macro defining AES IV length (Peter Krempa) - qemu: Change from SecretIV or _IV to SecretAES or _AES (John Ferlan) - Fix tests to include video ram size (Ján Tomko) - Fill out default vram in DeviceDefPostParse (Ján Tomko) - Move virDomainDefPostParseInternal after virDomainDeviceDefPostParse (Ján Tomko) - spec: Fix broken indentation reported by syntax-check (Erik Skultety) - libvirt.spec.in: remove all changelog entries (Daniel P. Berrange) - libvirt.spec.in: use explicit --with-xxx for all features (Daniel P. Berrange) - libvirt.spec.in: remove nss plugin conditional (Daniel P. Berrange) - libvirt.spec.in: remove libnl conditional (Daniel P. Berrange) - libvirt.spec.in: remove macvtap conditional (Daniel P. Berrange) - libvirt.spec.in: remove libpcap conditional (Daniel P. Berrange) - libvirt.spec.in: remove nwfilter driver conditional (Daniel P. Berrange) - libvirt.spec.in: remove nodedev conditional (Daniel P. Berrange) - libvirt.spec.in: remove most storage conditionals (Daniel P. Berrange) - libvirt.spec.in: remove network driver conditiaonl (Daniel P. Berrange) - libvirt.spec.in: remove interface driver conditional (Daniel P. Berrange) - libvirt.spec.in: remove selinux conditional (Daniel P. Berrange) - libvirt.spec.in: remove yajl conditional (Daniel P. Berrange) - libvirt.spec.in: remove polkit conditional (Daniel P. Berrange) - libvirt.spec.in: remove sasl conditionals (Daniel P. Berrange) - libvirt.spec.in: remove dtrace conditional (Daniel P. Berrange) - libvirt.spec.in: remove audit conditional (Daniel P. Berrange) - libvirt.spec.in: remove netcf conditional (Daniel P. Berrange) - libvirt.spec.in: remove udev conditional (Daniel P. Berrange) - libvirt.spec.in: remove capng conditional (Daniel P. Berrange) - libvirt.spec.in: remove apparmor conditional (Daniel P. Berrange) - libvirt.spec.in: remove hal conditional (Daniel P. Berrange) - libvirt.spec.in: remove avahi conditional (Daniel P. Berrange) - libvirt.spec.in: remove libvirtd conditional (Daniel P. Berrange) - libvirt.spec.in: remove conditional for driver modules (Daniel P. Berrange) - libvirt.spec.in: drop Fedora < 20 and RHEL < 6 (Daniel P. Berrange) - libvirt.spec.in: explicitly disable xenapi & vz drivers (Daniel P. Berrange) - headers: Remove unnecessary keyword extern from function declaration (Erik Skultety) - conf: make virDomainDefAddController() public (Laine Stump) - libxl: support migration stream V2 in migration (Jim Fehlig) - libxl: support Xen migration stream V2 in save/restore (Jim Fehlig) - libxl: switch to using libxl_domain_create_restore from v4.4 API (Jim Fehlig) - qemu: Add extra checks for secret destroy API's (John Ferlan) - util: set vlan tag for macvtap passthrough mode on SRIOV VFs (Laine Stump) - seclabeltest: Update to use VIRT_TEST_MAIN (Michal Privoznik) - virt-admin: Introduce client-disconnect command (Erik Skultety) - admin: Introduce virAdmClientClose API (Erik Skultety) - admin: Remove flags checking from virAdmConnectOpen public API (Erik Skultety) - admin: Remove flags checking from public API entry points (Erik Skultety) - qemu_hotplug: fix checking graphics ports (Pavel Hrdina) - qemu_process: merge graphics code into qemuProcessSetupGraphics (Pavel Hrdina) - graphics: generate fake ports also for tests (Pavel Hrdina) - qemu_process: separate graphics port reservation (Pavel Hrdina) - configure: split out UML driver checks (Roman Bogorodskiy) - admin: Clean up error path in adminServerListClients (John Ferlan) - Use virGetLastErrorMessage to avoid Coverity message (John Ferlan) - qemu: command: unconditionally allow accel3d='no' (Cole Robinson) - domaincaps: Report video modelType (Cole Robinson) - domaincaps: Report graphics type enum (Cole Robinson) - qemu: command: Remove unnecessary label in qemuCheckDiskConfig (Peter Krempa) - conf: Kill now unused virDomainDiskSourceIsBlockType (Peter Krempa) - qemu: command: Use more appropriate checking function for block devices (Peter Krempa) - qemu: Support <disk device='lun'> for iSCSI direct mapped volumes (Peter Krempa) - util: Replace virDomainDiskSourceIsBlockType with a new helper (Peter Krempa) - qemu_monitor_json: Follow refactor (Michal Privoznik) - virsh: blkdeviotune: accept human readable values for bytes (Nishith Shah) - genericxml2xml: add several graphics tests (Pavel Hrdina) - qemu_hotplug: cleanup error messages in qemuDomainChangeGraphics (Pavel Hrdina) - qemu_process: handle port allocation for VNC the same way as for Spice (Pavel Hrdina) - qemu_process: move listen code out of qemuProcessSetupGraphics (Pavel Hrdina) - graphics: use enums instead of int (Pavel Hrdina) - virCgroupValidateMachineGroup: Reflect change in CGroup struct naming (Michal Privoznik) - domaincapstest: Add tests for QEMU 2.6 (Jiri Denemark) - qemucapabilitiestest: Add tests for aarch64 and ppc64le (Jiri Denemark) - domaincapstest: Use default machine type (Jiri Denemark) - qemucapabilitiestest: Rename *.caps to *.xml (Jiri Denemark) - qemucapabilitiestest: Test all capabilities (Jiri Denemark) - qemucapabilitiestest: Reorder flags in caps files (Jiri Denemark) - qemucapabilitiestest: Reindent *.caps files (Jiri Denemark) - qemucapabilitiestest: Uses consistent names (Jiri Denemark) - qemu: Export caps cache APIs for tests (Jiri Denemark) - qemu: Separate formatting from saving into caps cache (Jiri Denemark) - qemucapabilitiestest: Prepare for testing non-x86_64 archs (Jiri Denemark) - tests: Refactor domaincapstest (Jiri Denemark) - qemu: Introduce qemuDomainSecretIV (John Ferlan) - qemu: Separate network URI command building code (John Ferlan) - qemu: Move qemuDomainSecretDestroy to qemuProcessLaunch (John Ferlan) - qemu: Move qemuDomainSecretPrepare to qemuProcessPrepareDomain (John Ferlan) - qemu: Split out the master key create and write (John Ferlan) - qemu: Adjust names of qemuDomainSecretInfoType enums (John Ferlan) - rpc: use virNetMessageClearPayload in client (Cole Robinson) - rpc: Clear more in virNetMessageClearPayload (Cole Robinson) - rpc: Add virNetMessageClearPayload (Cole Robinson) - virNetServerClientNewPostExecRestart: Drop useless typecasts (Michal Privoznik) - Add qemucapsprobe in .gitignore (Jiri Denemark) - tests: Add qemucapsprobe helper (Jiri Denemark) - virjson: Make pretty format more compact (Jiri Denemark) - qemu: Make qemuMonitorJSONIOProcessLine available for tests (Jiri Denemark) - qemu: Make virQEMUCapsNewForBinary usable from tests (Jiri Denemark) - tests: Decouple preload code from main() (Jiri Denemark) - tests: Create a shared library with qemu driver (Jiri Denemark) - bhyve: implement domainShutdown (Roman Bogorodskiy) - bhyve: drop virProcessKillPainfully() from destroy (Roman Bogorodskiy) - virsh: Pass the correct live/config xml to virshDomainDetachInterface. (Nitesh Konkar) - virsh: Introduce virshDomainDetachInterface function (Nitesh Konkar) - admin: Add a check to reject negative argument for number of typed params (Erik Skultety) - tools: virt-host-validate: HW virt support on s390 (Bjoern Walk) - tools: virt-host-validate: improve error handling (Bjoern Walk) - tools: virt-host-validate: fix CPU flag detection (Bjoern Walk) - qemu: Add 'iothread' to command line for supported controller (John Ferlan) - qemu: Use switch for qemuCheckIOThreads (John Ferlan) - conf: Add support for virtio-scsi iothreads (John Ferlan) - Pull latest translations from zanata (Daniel P. Berrange) - libvirt.spec: remove duplicate files from -docs package (Daniel P. Berrange) - virsh: volume: Add --bytes to 'vol-info' (Peter Krempa) - daemon: Add VIR_ERR_NO_SERVER and VIR_ERR_NO_CLIENT to daemonErrorLogFilter (Erik Skultety) - conf: Move virDomainControllerModelTypeToString (John Ferlan) - qemu: Add capability for virtio-scsi iothreads (John Ferlan) - spec: Rename %{init_scripts} -> %{with_init_script} (Andrea Bolognani) - spec: Type --with-init-script correctly (Andrea Bolognani) - configure: Remove nested conditionals in LIBVIRT_CHECK_INIT_SCRIPT (Andrea Bolognani) - configure: Change RHEL default from systemd+redhat to systemd (Andrea Bolognani) - maint: Ignore all .init, .service and .socket files (Andrea Bolognani) - virt-admin: Introduce command client-info (Erik Skultety) - admin: Introduce virAdmClientGetInfo API (Erik Skultety) - virnetserverclient: Add an internal method to retrieve client's identity (Erik Skultety) - virneserverclient: Introduce virNetServerClientHasSASLSession (Erik Skultety) - virnetsocket: Provide socket address format in a more standard form (Erik Skultety) - admin: include: Introduce some client's identity related typed params macros (Erik Skultety) - admin: Introduce virAdmServerLookupClient (Erik Skultety) - qemu_monitor_json: Follow our coding style (Michal Privoznik) - maint: Don't ignore libvirtd.pod any longer (Andrea Bolognani) - gitignore: ignore virt(log|lock)d.(service|socket) (Cole Robinson) - daemon: add option to read host uuid from /etc/machine-id (Nikolay Shirokovskiy) - util: factor out reading file into preallocated buffer (Nikolay Shirokovskiy) - qemu: support configuring usb3 controller port count (Cole Robinson) - qemu: caps: introduce QEMU_CAPS_NEC_USB_XHCI_PORTS (Cole Robinson) - qemu: parse: Use virControllerDefNew (Cole Robinson) - qemu_monitor_json: Drop redundant checks (Michal Privoznik) - qemuMonitorJSONQueryRxFilter: Validate qemu reply prior parsing it (Michal Privoznik) - Remove useless os.machine NULL check (Ján Tomko) - Introduce qemuDomainMachineIsVirt (Ján Tomko) - Rewrite the condition in qemuDomainAssignARMVirtioMMIOAddresses (Ján Tomko) - Remove useless variable in qemuDomainAssignAddresses (Ján Tomko) - Return void in qemuDomainAssignARMVirtioMMIOAddresses (Ján Tomko) - Invert condition in qemuDomainDefAddDefaultDevices (Ján Tomko) - Use qemuDomainMachineIs helpers when adding default devices (Ján Tomko) - qemu: Introduce qemuMonitorGetRTCTime (Michal Privoznik) - virt-admin: Introduce srv-clients-list command (Erik Skultety) - admin: Introduce listing clients (Erik Skultety) - rpc: virnetserverclient: Implement client connection transport retrieval (Erik Skultety) - include: admin: export connection transport constants (Erik Skultety) - rpc: virnetserver: Support retrieval of a list of clients (Erik Skultety) - rpc: gendispatch: Tune it to support client structure (Erik Skultety) - admin: Introduce virAdmClient client-side object (Erik Skultety) - rpc: virnetserverclient: Introduce new attribute conn_time to client (Erik Skultety) - rpc: virnetserverclient: Identify clients by an integer ID (Erik Skultety) - configure: Introduce LIBVIRT_{CHECK,RESULT}_INIT_SCRIPT (Andrea Bolognani) - configure: Add systemd detection to --with-init-script=check (Andrea Bolognani) - configure: Improve --with-init-script=check (Andrea Bolognani) - tests: add tests for panic device model s390 (Boris Fiuczynski) - qemu: add default panic device to S390 guests (Boris Fiuczynski) - qemu: add panic device support for S390 (Boris Fiuczynski) - qemu: merge S390 and S390X default device creation (Boris Fiuczynski) - rpc: protocol: Clarify VIR_NET_ERROR usage with streams (Cole Robinson) - fdstream: Report error with virProcessTranslateStatus (Cole Robinson) - fdstream: separate out virCommandPtr cleanup (Cole Robinson) - nwfilter: Push configFile building into LoadConfig (Cole Robinson) - conf: format runtime DAC seclabel, unless MIGRATABLE (Cole Robinson) - conf: Parse more of our nodedev XML (Martin Kletzander) - schemas: Update nodedev schema to match reality (Martin Kletzander) - Move capability formatting together (Martin Kletzander) - Change virDevicePCIAddress to virPCIDeviceAddress (Martin Kletzander) - qemu: hotplug: Adjust error path for attach hostdev scsi disk (John Ferlan) - qemu: hotplug: Adjust error path for attach virtio disk (John Ferlan) - qemu: hotplug: Adjust error path for attach scsi disk (John Ferlan) - qemu: Use qemuDomainSecretInfoPtr in qemuBuildNetworkDriveURI (John Ferlan) - qemu: Introduce qemuDomainSecretHostdevPrepare and Destroy (John Ferlan) - qemu: Introduce qemuDomainHostdevPrivatePtr (John Ferlan) - qemu: Introduce qemuDomainSecretPrepare and Destroy (John Ferlan) - qemu: Introduce qemuDomainSecretInfo (John Ferlan) - Adapt augeas profile to handle negative int values. (Cédric Bosdonnat) - qemu: monitor: Kill legacy PCI hotplug code (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachControllerDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainDetachNetDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainDetachHostPCIDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainDetachControllerDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainDetachVirtioDiskDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachHostPCIDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachNetDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachVirtioDiskDevice (Peter Krempa) - qemu: monitor: Kill legacy USB monitor code (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachHostUSBDevice (Peter Krempa) - qemu: hotplug: Assume QEMU_CAPS_DEVICE in qemuDomainAttachUSBMassStorageDevice (Peter Krempa) - qemu: remove default case from few typecasted enums (Peter Krempa) - qemu: process: Don't needlesly clear the perf events in qemuDomainPerfRestart (Peter Krempa) - util: perf: Adhere to coding style of error checks in qemuDomainSetPerfEvents (Peter Krempa) - util: perf: Adhere to coding style of error checks in virPerfEventEnable (Peter Krempa) - util: perf: Use 'error' label in virPerfCmtEnable (Peter Krempa) - qemu: perf: Don't ignore perf setup if allocation fails (Peter Krempa) - qemu: hotplug: Allow update of disk default snapshot location (Peter Krempa) - qemu: domain: Check few more fields for when changing disk source (Peter Krempa) - qemu: domain: Fix error message in qemuDomainDiskChangeSupported (Peter Krempa) - qemu: hotplug: Skip waiting for tray opening if qemu doesn't notify us (Peter Krempa) - virStream{Recv,Send}All: Increase client buffer (Michal Privoznik) - virnetclientstream: Process stream messages later (Michal Privoznik) - Revert "rpc: Fix slow volume download (virsh vol-download)" (Michal Privoznik) - Post-release version bump to 1.3.5 (John Ferlan) Cleanups: - Revert "qemu_cgroup: allow access to /dev/dri for virtio-vga" (Ján Tomko) Thanks everybody who helped with this release, be it by providing ideas bugs, patches, reviews, doc or localizations, hopefully I didn't mess it by rushing that release out ... Enjoy ! 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

Re: [libvirt] [PATCH RFC 4/8] target-i386: cpu: consolidate calls of object_property_parse() in x86_cpu_parse_featurestr
by Eduardo Habkost 03 Jun '16

03 Jun '16

(CCing libvirt folks) BTW: On Thu, Jun 02, 2016 at 02:22:22PM +0200, Igor Mammedov wrote: [...] > > /* Special cases: */ > > if (!strcmp(name, "xlevel")) { > > numvalue = strtoul(val, &err, 0); > > if (!*val || *err) { > > error_setg(errp, "bad numerical value %s", val); > > return; > > } > > if (numvalue < 0x80000000) { > > error_report("xlevel value shall always be >= 0x80000000" > > ", fixup will be removed in future versions"); > > numvalue += 0x80000000; > > snprintf(num, sizeof(num), "%" PRIu32, numvalue); > > val = num; [...] > > } else if (!strcmp(name, "hv-spinlocks")) { > > const int min = 0xFFF; > > > > numvalue = strtoul(val, &err, 0); > > if (!*val || *err) { > > error_setg(errp, "bad numerical value %s", val); > > return; > > } > > if (numvalue < min) { > > error_report("hv-spinlocks value shall always be >= 0x%x" > > ", fixup will be removed in future versions", > > min); > > numvalue = min; > > } Those "fixup will be removed in future versions" warnings are present since QEMU 1.7. Assuming that libvirt never allowed those invalid values to be used in the configuration (did it?), I believe we can safely remove the hv-spinlocks and xlevel fixups in QEMU 2.7. The hv-spinlocks setter already rejects invalid values. We just need to make x86_cpu_realizefn() reject invalid xlevel values. -- Eduardo

4 8

[libvirt] [PATCH] Fix building with -Og
by Martin Kletzander 03 Jun '16

03 Jun '16

When building using -Og, gcc sees that some variables can be used uninitialized It can be debatable whether it is possible with our codeflow, but functions should be self-contained and initializations are always good. The return instead of goto is due to actualType being used in the cleanup. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/lxc/lxc_driver.c | 2 +- src/nwfilter/nwfilter_ebiptables_driver.c | 2 +- src/util/virbitmap.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 67f14fe766a5..f0948eae774e 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -4275,7 +4275,7 @@ lxcDomainAttachDeviceNetLive(virConnectPtr conn, if (!priv->initpid) { virReportError(VIR_ERR_OPERATION_INVALID, "%s", _("Cannot attach disk until init PID is known")); - goto cleanup; + return -1; } if (virLXCProcessValidateInterface(net) < 0) diff --git a/src/nwfilter/nwfilter_ebiptables_driver.c b/src/nwfilter/nwfilter_ebiptables_driver.c index 423d069e1b26..b7be2917e29e 100644 --- a/src/nwfilter/nwfilter_ebiptables_driver.c +++ b/src/nwfilter/nwfilter_ebiptables_driver.c @@ -1570,7 +1570,7 @@ iptablesCreateRuleInstanceStateCtrl(virFirewallPtr fw, const char *ifname, virNWFilterVarCombIterPtr vars) { - int rc; + int rc = 0; bool directionIn = false; char chainPrefix[2]; bool maySkipICMP, inout = false; diff --git a/src/util/virbitmap.c b/src/util/virbitmap.c index 9283aef1735b..4ca59f9d6227 100644 --- a/src/util/virbitmap.c +++ b/src/util/virbitmap.c @@ -817,7 +817,7 @@ virBitmapLastSetBit(virBitmapPtr bitmap) ssize_t i; int unusedBits; ssize_t sz; - unsigned long bits; + unsigned long bits = 0; unusedBits = bitmap->map_len * VIR_BITMAP_BITS_PER_UNIT - bitmap->max_bit; -- 2.8.3

3 4

[libvirt] [PATCH 0/9] Add runnability info to query-cpu-definitions
by Eduardo Habkost 03 Jun '16

03 Jun '16

This series extends query-cpu-definitions to include two extra fields: "runnable", and "unavailable-features". This will return information based on the current machine and accelerator only. In the future we may extend these mechanisms to allow querying other machines and other accelerators without restarting QEMU, but it will require some reorganization of QEMU's main code. This series is based on my 'x86-next' branch, at: git://github.com/ehabkost/qemu.git x86-next Cc: David Hildenbrand <dahi(a)linux.vnet.ibm.com> Cc: Michael Mueller <mimu(a)linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger(a)de.ibm.com> Cc: Cornelia Huck <cornelia.huck(a)de.ibm.com> Cc: Jiri Denemark <jdenemar(a)redhat.com> Cc: libvir-list(a)redhat.com Eduardo Habkost (9): target-i386: Move TCG initialization check to tcg_x86_init() target-i386: Move TCG initialization to realize time target-i386: Call cpu_exec_init() on realize target-i386: List CPU models using subclass list target-i386: Move warning code outside x86_cpu_filter_features() target-i386: Define CPUID filtering functions before x86_cpu_list() qmp: Add runnability information to query-cpu-definitions target-i386: Use "-" instead of "_" on all feature names target-i386: Return runnability information on query-cpu-definitions qapi-schema.json | 10 +- target-i386/cpu-qom.h | 4 + target-i386/cpu.c | 275 +++++++++++++++++++++++++++++++++--------------- target-i386/translate.c | 6 ++ 4 files changed, 207 insertions(+), 88 deletions(-) -- 2.5.5

6 46

[libvirt] [PATCH] storage: Adjust qemu-img switches check
by John Ferlan 03 Jun '16

03 Jun '16

Since we support QEMU 0.12 and later, checking for support of specific flags added prior to that isn't necessary. Thus start with the base of having the "-o options" available for the qemu-img create option and then determine whether we have the compat option for qcow2 files (which would be necessary up through qemu 2.0 where the default changes to compat 0.11). NOTE: Keeping old tests around since it's still possible to create in the old format. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/storage/storage_backend.c | 66 ++++++++++++++++--------------------------- 1 file changed, 24 insertions(+), 42 deletions(-) diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index 3a23cd7..eaa6f65 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -869,9 +869,19 @@ virStoragePloopResize(virStorageVolDefPtr vol, return ret; } +/* Flag values shared w/ storagevolxml2argvtest.c. + * Since it's still possible to provide the old format args, just + * keep them; however, prefix with an "X_" (similar to qemu_capabilities.c) + * to indicate the are older. + * + * QEMU_IMG_BACKING_FORMAT_OPTIONS (added in qemu 0.11) + * QEMU_IMG_BACKING_FORMAT_OPTIONS_COMPAT + * was made necessary due to 2.0 change to change the default + * qcow2 file format from 0.10 to 1.1. + */ enum { - QEMU_IMG_BACKING_FORMAT_NONE = 0, - QEMU_IMG_BACKING_FORMAT_FLAG, + X_QEMU_IMG_BACKING_FORMAT_NONE = 0, + X_QEMU_IMG_BACKING_FORMAT_FLAG, QEMU_IMG_BACKING_FORMAT_OPTIONS, QEMU_IMG_BACKING_FORMAT_OPTIONS_COMPAT, }; @@ -904,46 +914,18 @@ virStorageBackendQemuImgSupportsCompat(const char *qemuimg) static int virStorageBackendQEMUImgBackingFormat(const char *qemuimg) { - char *help = NULL; - char *start; - char *end; - char *tmp; - int ret = -1; - int exitstatus; - virCommandPtr cmd = virCommandNewArgList(qemuimg, "-h", NULL); - - virCommandAddEnvString(cmd, "LC_ALL=C"); - virCommandSetOutputBuffer(cmd, &help); - virCommandClearCaps(cmd); - - /* qemuimg doesn't return zero exit status on -h, - * therefore we need to provide pointer for storing - * exit status, although we don't parse it any later */ - if (virCommandRun(cmd, &exitstatus) < 0) - goto cleanup; + /* As of QEMU 0.11 the [-o options] support was added via qemu + * commit id '9ea2ea71', so we start with that base and figure + * out what else we have */ + int ret = QEMU_IMG_BACKING_FORMAT_OPTIONS; + + /* QEMU 2.0 changed to using a format that only QEMU 1.1 and newer + * understands. Since we still support QEMU 0.12 and newer, we need + * to be able to handle the previous format as can be set via a + * compat=0.10 option. */ + if (virStorageBackendQemuImgSupportsCompat(qemuimg)) + ret = QEMU_IMG_BACKING_FORMAT_OPTIONS_COMPAT; - if ((start = strstr(help, " create ")) == NULL || - (end = strstr(start, "\n")) == NULL) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("unable to parse qemu-img output '%s'"), - help); - goto cleanup; - } - if (((tmp = strstr(start, "-F fmt")) && tmp < end) || - ((tmp = strstr(start, "-F backing_fmt")) && tmp < end)) { - ret = QEMU_IMG_BACKING_FORMAT_FLAG; - } else if ((tmp = strstr(start, "[-o options]")) && tmp < end) { - if (virStorageBackendQemuImgSupportsCompat(qemuimg)) - ret = QEMU_IMG_BACKING_FORMAT_OPTIONS_COMPAT; - else - ret = QEMU_IMG_BACKING_FORMAT_OPTIONS; - } else { - ret = QEMU_IMG_BACKING_FORMAT_NONE; - } - - cleanup: - virCommandFree(cmd); - VIR_FREE(help); return ret; } @@ -1196,7 +1178,7 @@ virStorageBackendCreateQemuImgCmdFromVol(virConnectPtr conn, VIR_FREE(opts); } else { if (info.backingPath) { - if (imgformat == QEMU_IMG_BACKING_FORMAT_FLAG) + if (imgformat == X_QEMU_IMG_BACKING_FORMAT_FLAG) virCommandAddArgList(cmd, "-F", backingType, NULL); else VIR_DEBUG("Unable to set backing store format for %s with %s", -- 2.5.5

1 1

[libvirt] [PATCH 0/3] Clean up some virstorageencryption code
by John Ferlan 03 Jun '16

03 Jun '16

While working through adding luks support for libvirt, I have a few patches that aren't really germane to adding support and rather than drop a large patch series when I'm done - I figured I'd post a few adjustments. In the long run the encryption code probably doesn't work, but I'm trying to move it to the side at least. John Ferlan (3): util: Clean up code formatting in virstorageencryption storage: Split out setting default secret for encryption storage: Split out a helper for encryption checks src/storage/storage_backend.c | 80 ++++++++++++++++++++++++++-------------- src/storage/storage_backend_fs.c | 79 ++++++++++++++++++++++++--------------- src/util/virstorageencryption.c | 58 ++++++++++++++--------------- 3 files changed, 128 insertions(+), 89 deletions(-) -- 2.5.5

1 4

[libvirt] [PATCH 0/2] Fix a virperf related crasher
by Michal Privoznik 03 Jun '16

03 Jun '16

I've ran into a crasher. Worse crasher over RO connection. Michal Privoznik (2): Drop virPerfGetEventFd virPerfEventIsEnabled: Don't crash on shut off domains src/libvirt_private.syms | 1 - src/util/virperf.c | 22 +++------------------- src/util/virperf.h | 3 --- 3 files changed, 3 insertions(+), 23 deletions(-) -- 2.8.3

2 4

[libvirt] Plan for the next release
by Daniel Veillard 03 Jun '16

03 Jun '16

I will be travelling end of this week and most of next week, but I think we can still try to get a release for June 1st, I suggest: - enter freeze this Thrusday 26 - Push an rc2 over the week-end - try to push the 1.3.5 release on Wed June 1st I hope this works for everybody ! 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 1

[libvirt] [PATCH v3] virDomainChrGetDomainPtrsInternal: Return an integer
by Michal Privoznik 03 Jun '16

03 Jun '16

There's this problem on the recent gcc-6.1: In file included from conf/domain_conf.c:37:0: conf/domain_conf.c: In function 'virDomainChrPreAlloc': conf/domain_conf.c:14109:35: error: potential null pointer dereference [-Werror=null-dereference] return VIR_REALLOC_N(*arrPtr, *cntPtr + 1); ^~ ./util/viralloc.h:158:73: note: in definition of macro 'VIR_REALLOC_N' # define VIR_REALLOC_N(ptr, count) virReallocN(&(ptr), sizeof(*(ptr)), (count), \ ^~~~~ conf/domain_conf.c: In function 'virDomainChrRemove': conf/domain_conf.c:14133:21: error: potential null pointer dereference [-Werror=null-dereference] for (i = 0; i < *cntPtr; i++) { ^~~~~~~ GCC basically fails to see, that the virDomainChrGetDomainPtrsInternal will never actually return NULL because it's never called over a domain char device with _LAST type. But to make it shut up, lets turn this function into returning an integer and check in the callers if a zero value value was returned. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- diff to v2: - Yet another improvement as suggested in review of v2 src/conf/domain_conf.c | 39 +++++++++++++++++++++++---------------- 1 file changed, 23 insertions(+), 16 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 7d46d0b..9f9fdf2 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -14038,7 +14038,7 @@ virDomainChrFind(virDomainDefPtr def, /* Return the address within vmdef to be modified when working with a * chrdefptr of the given type. */ -static void +static int ATTRIBUTE_RETURN_CHECK virDomainChrGetDomainPtrsInternal(virDomainDefPtr vmdef, virDomainChrDeviceType type, virDomainChrDefPtr ***arrPtr, @@ -14048,28 +14048,30 @@ virDomainChrGetDomainPtrsInternal(virDomainDefPtr vmdef, case VIR_DOMAIN_CHR_DEVICE_TYPE_PARALLEL: *arrPtr = &vmdef->parallels; *cntPtr = &vmdef->nparallels; - break; + return 0; case VIR_DOMAIN_CHR_DEVICE_TYPE_SERIAL: *arrPtr = &vmdef->serials; *cntPtr = &vmdef->nserials; - break; + return 0; case VIR_DOMAIN_CHR_DEVICE_TYPE_CONSOLE: *arrPtr = &vmdef->consoles; *cntPtr = &vmdef->nconsoles; - break; + return 0; case VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL: *arrPtr = &vmdef->channels; *cntPtr = &vmdef->nchannels; - break; + return 0; case VIR_DOMAIN_CHR_DEVICE_TYPE_LAST: - *arrPtr = NULL; - *cntPtr = NULL; break; } + + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unknown char device type: %d"), type); + return -1; } @@ -14085,14 +14087,13 @@ virDomainChrGetDomainPtrs(const virDomainDef *vmdef, size_t *cntVar = NULL; /* Cast away const; we add it back in the final assignment. */ - virDomainChrGetDomainPtrsInternal((virDomainDefPtr) vmdef, type, - &arrVar, &cntVar); - if (arrVar) { + if (virDomainChrGetDomainPtrsInternal((virDomainDefPtr) vmdef, type, + &arrVar, &cntVar) < 0) { + *arrPtr = NULL; + *cntPtr = 0; + } else { *arrPtr = (const virDomainChrDef **) *arrVar; *cntPtr = *cntVar; - } else { - *arrPtr = NULL; - *cntPtr = 0; } } @@ -14104,7 +14105,9 @@ virDomainChrPreAlloc(virDomainDefPtr vmdef, virDomainChrDefPtr **arrPtr = NULL; size_t *cntPtr = NULL; - virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, &arrPtr, &cntPtr); + if (virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, + &arrPtr, &cntPtr) < 0) + return -1; return VIR_REALLOC_N(*arrPtr, *cntPtr + 1); } @@ -14116,7 +14119,9 @@ virDomainChrInsertPreAlloced(virDomainDefPtr vmdef, virDomainChrDefPtr **arrPtr = NULL; size_t *cntPtr = NULL; - virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, &arrPtr, &cntPtr); + if (virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, + &arrPtr, &cntPtr) < 0) + return; VIR_APPEND_ELEMENT_INPLACE(*arrPtr, *cntPtr, chr); } @@ -14128,7 +14133,9 @@ virDomainChrRemove(virDomainDefPtr vmdef, virDomainChrDefPtr ret = NULL, **arrPtr = NULL; size_t i, *cntPtr = NULL; - virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, &arrPtr, &cntPtr); + if (virDomainChrGetDomainPtrsInternal(vmdef, chr->deviceType, + &arrPtr, &cntPtr) < 0) + return NULL; for (i = 0; i < *cntPtr; i++) { ret = (*arrPtr)[i]; -- 2.8.3

2 1

[libvirt] [PATCH 00/10] vz: improve locking in getting domain statistics
by Nikolay Shirokovskiy 03 Jun '16

03 Jun '16

Current stats cache uses same lock as domain cache, namely lock of virDomainObjPtr. First this is not necessary, domain cache is used to translate id of object of interest from libvirt to vzsdk one. Besides this domain object is not used. Second this is risky game as lock is dropped on waiting for stat events from sdk deep down on stack so callers throughout the stack should be written very carefully. This patch series makes stats cache self locking and thus making code much more solid. This is done thru introducing extra lock for stats cache. Statistics API functions first use domain cache then drop domain lock, query stats cache and unreference domain object. Patches 1-4 conform API functions to this structure. Patches 5-7 makes subsidiary improvments. Patch 8 finally introduce new lock. Last patches are cleanup[9] and bugfix[10]. Nikolay Shirokovskiy (10): vz: pass string instead of disk definition to block stat function vz: prepare disks names before getting all disks stats vz: move getting stats in vzDomainGetVcpus to the end vz: move getting stats in vzDomainGetInfo to the end vz: use consistent naming for different domain object in vz_driver.c vz: simplify refcount on sdkdom in prlsdkLoadDomain vz: hold stats cache in a pointer in vzDomObj vz: introduce stats cache lock vz: extract on stats cache update into a function vz: fix many stat request issue src/vz/vz_driver.c | 397 +++++++++++++++++++++++++++++++++++------------------ src/vz/vz_sdk.c | 230 +++++++++++++++++++++---------- src/vz/vz_sdk.h | 15 +- src/vz/vz_utils.c | 9 -- src/vz/vz_utils.h | 6 +- 5 files changed, 438 insertions(+), 219 deletions(-) -- 1.8.3.1

1 11

[libvirt] libvirt-tck test failure after commit b3d06987
by Jim Fehlig 02 Jun '16

02 Jun '16

We've noticed libvirt-tck test 100-apply-verify-host.t failing recently on libvirt.git master and I finally got around to bisecting it to commit b3d06987. I haven't looked at the test in detail, but it appears to expect a broadcast address of 10.1.2.255, however finds an address of 0.0.0.0 after commit b3d06987. I'm not terribly familiar with this code, but the following hunk of b3d06987 looks suspect @@ -1039,21 +1039,28 @@ virNetDevCreateNetlinkAddressMessage(int messageType, const char *ifname, virSocketAddr *addr, unsigned int prefix, - virSocketAddr *broadcast) + virSocketAddr *broadcast, + virSocketAddr *peer) { struct nl_msg *nlmsg = NULL; struct ifaddrmsg ifa; unsigned int ifindex; void *addrData = NULL; + void *peerData = NULL; void *broadcastData = NULL; size_t addrDataLen; if (virNetDevGetIPAddressBinary(addr, &addrData, &addrDataLen) < 0) return NULL; - if (broadcast && virNetDevGetIPAddressBinary(broadcast, &broadcastData, - &addrDataLen) < 0) - return NULL; + if (peer && VIR_SOCKET_ADDR_VALID(peer)) { + if (virNetDevGetIPAddressBinary(peer, &peerData, &addrDataLen) < 0) + return NULL; + } else if (broadcast) { + if (virNetDevGetIPAddressBinary(broadcast, &broadcastData, + &addrDataLen) < 0) + return NULL; + } /* Get the interface index */ if ((ifindex = if_nametoindex(ifname)) == 0) Vasiliy, I can look in more detail tomorrow, but in the meantime any suggestions you have would be much appreciated. Regards, Jim

3 3

[libvirt] [PATCH v2 0/4] Final round of build fixes
by Michal Privoznik 02 Jun '16

02 Jun '16

diff to v1: - Some patches from the original patch set are pushed now - For the other ones, I've managed to work in review suggestions Michal Privoznik (4): virNetDevBridgeGet: Don't require users to virNetDevSetupControl ppc64Compute: Avoid possible NULL dereference virDomainChrGetDomainPtrsInternal: Return an integer virDomainFormatSchedDef: Avoid false positive NULL dereference src/conf/domain_conf.c | 44 ++++++++++++++++++++++++++++++-------------- src/cpu/cpu_ppc64.c | 12 ++++++------ src/util/virnetdevbridge.c | 35 ++++++++++++----------------------- 3 files changed, 48 insertions(+), 43 deletions(-) -- 2.8.3

2 10

[libvirt] Python listDefinedDomains broken?
by Marc Richter 02 Jun '16

02 Jun '16

Hi everyone, please, don't roast me if I'm asking stupid things or fail to provide info, since I'm a freshman, using libvirt. I have two servers, both running the following versions: Compiled against library: libvirt 1.3.1 Using library: libvirt 1.3.1 Using API: QEMU 1.3.1 Running hypervisor: QEMU 0.14.0 I'm trying to create a Python program, which gets a list of all defined domains for each host and give a list for domains only defined on one of these hosts. This Python script is running on a third host (my workstation), using these versions: Python: 3.5.1 libvirt-python: 1.3.1 When I'm running virsh on my workstation (version 1.3.1), like this: virsh -c "qemu+ssh://root@host1/system?keyfile=/home/user/.ssh/id_rsa_libvirt" --readonly list --all Then a domain named "winxp_ausgeschaltet" is within that list: Id Name State ---------------------------------------------------- ... - winxp_ausgeschaltet shut off When I do the same for the second host, it is also in that list: Id Name State ---------------------------------------------------- ... 7 winxp_ausgeschaltet running When I do the following in Python, this domain is listed for the first host (where it's state is "shut off"), but not for the second one: import libvirt import sys class kvmhost: def __init__(self, host, keyfile): self.conn = self.connect_kvm(host, keyfile) self.doms = sorted(self.get_doms()) def connect_kvm(self, host, keyfile): try: conn = libvirt.openReadOnly('qemu+ssh://root@' + host + '/system?keyfile=' + keyfile) except libvirt.libvirtError as lve: print('Error: ' + str(lve)) sys.exit(1) return conn def get_doms(self): try: alldoms = self.conn.listDefinedDomains() except libvirt.libvirtError as lve: print('Error: ' + str(lve)) self.conn.close() sys.exit(1) return alldoms host1 = kvmhost('host1', '/path/to/ssh_keyfile') host2 = kvmhost('host2', '/path/to/ssh_keyfile') print(host1.doms) print(host2.doms) What am I doing wrong here or: is there a missbehavior in libvirt or the python module? Using listAllDomains instead of listDefinedDomains gives: libvirt: Remote Driver error : unknown procedure: 273 What is the difference between listAllDomains and listDefinedDomains at all? Thanks for your help. Marc

3 3

[libvirt] [PATCH v2 0/2] qemu: expand domain memory statistics
by Derbyshev Dmitriy 01 Jun '16

01 Jun '16

From: Derbyshev Dmitry <dderbyshev(a)virtuozzo.com> QEMU reports timestamp and available along with other memory statistics. This information was not saved into domain statistics. Changes since v1: * Enum numeration fixed * Macro getting "usage" field fixed Derbyshev Dmitry (2): qemu: expand domain memory statistics with 'usable' qemu: expand domain memory statistics with 'last-update' timestamp include/libvirt/libvirt-domain.h | 11 ++++++++++- src/libvirt-domain.c | 5 +++++ src/qemu/qemu_monitor_json.c | 23 +++++++++++++---------- tools/virsh-domain-monitor.c | 4 ++++ 4 files changed, 32 insertions(+), 11 deletions(-) -- 1.9.5.msysgit.0

2 3

[libvirt] [PATCH 0/4] Move secret object command line building helpers
by John Ferlan 01 Jun '16

01 Jun '16

More "extractable" changes from my work on adding luks support. While working through changes in storage_backend.c I found that the building of the 'qemu-img' command to create a luks volume will need to create a secret object. In order to do that I found if I took the qemu_command code and moved things around a bit, then I would be able to access the API's that I need. Moving the command line building from qemu_command.c to virjson.c seemed logical since it was multipurpose/generic anyway. I'm not "tied" to the API name chosen - if there's suggestions, I'll adjust. Then splitting out the generation of the secret object props into their own API and then eventually into secret_util.c seemed to be the best avenue. I can then include secret_util into the storage driver. The last change just cleans up the secinfo command building processing. It was "complicated" when there was going to be hostdev and disk code trying to generate the same objects... Now that just seems less important. When/if iSCSI/hostdev support for the AES secret is added, I'll made the appropriate adjustments then. John Ferlan (4): qemu: Move and rename qemuBuildObjectCommandlineFromJSON qemu: Introduce qemuBuildSecretObjectProps qemu: Move and rename qemuBuildSecretObjectProps qemu: Rework secinfo command line building src/libvirt_private.syms | 4 +- src/qemu/qemu_command.c | 199 ++++++-------------------------------------- src/qemu/qemu_command.h | 4 - src/secret/secret_util.c | 59 +++++++++++++ src/secret/secret_util.h | 10 +++ src/util/virjson.c | 117 +++++++++++++++++++++++++- src/util/virjson.h | 12 ++- tests/qemucommandutiltest.c | 7 +- 8 files changed, 219 insertions(+), 193 deletions(-) -- 2.5.5

3 23

[libvirt] KVM Forum 2016: Call For Participation
by Paolo Bonzini 01 Jun '16

01 Jun '16

================================================================= KVM Forum 2016: Call For Participation August 24-26, 2016 - Westin Harbor Castle - Toronto, Canada (All submissions must be received before midnight May 1, 2016) ================================================================= KVM Forum is an annual event that presents a rare opportunity for developers and users to meet, discuss the state of Linux virtualization technology, and plan for the challenges ahead. We invite you to lead part of the discussion by submitting a speaking proposal for KVM Forum 2016. At this highly technical conference, developers driving innovation in the KVM virtualization stack (Linux, KVM, QEMU, libvirt) can meet users who depend on KVM as part of their offerings, or to power their data centers and clouds. KVM Forum will include sessions on the state of the KVM virtualization stack, planning for the future, and many opportunities for attendees to collaborate. As we celebrate ten years of KVM development in the Linux kernel, KVM continues to be a critical part of the FOSS cloud infrastructure. This year, KVM Forum is joining LinuxCon and ContainerCon in Toronto, Canada. Selected talks from KVM Forum will be presented on Wednesday August 24 to the full audience of LinuxCon and ContainerCon. Also, attendees of KVM Forum will have access to all of the LinuxCon and ContainerCon talks on Wednesday. http://events.linuxfoundation.org/cfp Suggested topics: KVM and Linux * Scaling and optimizations * Nested virtualization * Linux kernel performance improvements * Resource management (CPU, I/O, memory) * Hardening and security * VFIO: SR-IOV, GPU, platform device assignment * Architecture ports QEMU * Management interfaces: QOM and QMP * New devices, new boards, new architectures * Scaling and optimizations * Desktop virtualization and SPICE * Virtual GPU * virtio and vhost, including non-Linux or non-virtualized uses * Hardening and security * New storage features * Live migration and fault tolerance * High availability and continuous backup * Real-time guest support * Emulation and TCG * Firmware: ACPI, UEFI, coreboot, u-Boot, etc. * Testing Management and infrastructure * Managing KVM: Libvirt, OpenStack, oVirt, etc. * Storage: glusterfs, Ceph, etc. * Software defined networking: Open vSwitch, OpenDaylight, etc. * Network Function Virtualization * Security * Provisioning * Performance tuning =============== SUBMITTING YOUR PROPOSAL =============== Abstracts due: May 1, 2016 Please submit a short abstract (~150 words) describing your presentation proposal. Slots vary in length up to 45 minutes. Also include the proposal type -- one of: - technical talk - end-user talk Submit your proposal here: http://events.linuxfoundation.org/cfp Please only use the categories "presentation" and "panel discussion" You will receive a notification whether or not your presentation proposal was accepted by May 27, 2016. Speakers will receive a complimentary pass for the event. In the instance that your submission has multiple presenters, only the primary speaker for a proposal will receive a complementary event pass. For panel discussions, all panelists will receive a complimentary event pass. TECHNICAL TALKS A good technical talk should not just report on what has happened over the last year; it should present a concrete problem and how it impacts the user and/or developer community. Whenever applicable, focus on work that needs to be done, difficulties that haven't yet been solved, and on decisions that other developers should be aware of. Summarizing recent developments is okay but it should not be more than a small portion of the overall talk. END-USER TALKS One of the big challenges as developers is to know what, where and how people actually use our software. We will reserve a few slots for end users talking about their deployment challenges and achievements. If you are using KVM in production you are encouraged submit a speaking proposal. Simply mark it as an end-user talk. As an end user, this is a unique opportunity to get your input to developers. HANDS-ON / BOF SESSIONS We will reserve some time for people to get together and discuss strategic decisions as well as other topics that are best solved within smaller groups. These sessions will be announced during the event. If you are interested in organizing such a session, please add it to the list at http://www.linux-kvm.org/page/KVM_Forum_2016_BOF Let people you think might be interested know about it, and encourage them to add their names to the wiki page as well. Please try to add your ideas to the list before KVM Forum starts. PANEL DISCUSSIONS If you are proposing a panel discussion, please make sure that you list all of your potential panelists in your abstract. We will request full biographies if a panel is accepted. =============== HOTEL / TRAVEL =============== This year's event will take place at the Westin Harbour Castle Toronto. For information on discounted room rates for conference attendees and on other hotels close to the conference, please visit http://events.linuxfoundation.org/events/kvm-forum/attend/hotel-travel. As of March 15, 2016, non-US citizens need either a visa or an Electronic Travel Authorization (eTA) in order to enter Canada. Detailed information on the travel documentation required for your country of origin can be found at http://www.cic.gc.ca/english/visit/visas.asp and http://events.linuxfoundation.org/events/kvm-forum/attend/hotel-travel. ** We urge you to start this process as quickly as possible to ensure ** receipt of appropriate travel documentation in time for your conference ** travel to Canada. For processing times for visa applications, please visit ** http://www.cic.gc.ca/english/information/times/. =============== IMPORTANT DATES =============== Notification: May 27, 2015 Schedule announced: June 3, 2015 Event dates: August 24-26, 2016 Thank you for your interest in KVM. We're looking forward to your submissions and seeing you at the KVM Forum 2016 in August! -your KVM Forum 2016 Program Committee Please contact us with any questions or comments at kvm-forum-2016-pc(a)redhat.com

2 2

[libvirt] [PATCH v2 0/6] Move secret object command line building helpers
by John Ferlan 01 Jun '16

01 Jun '16

v1: http://www.redhat.com/archives/libvir-list/2016-May/msg02017.html Changes in v2: Patch 1 - Move function to a new src/util/virqemu.c with adjustments to other affected areas Patches 2-4 - Similarly adjust the flow/name of the patches with an adjustment so that patch 2 is just the new code in virqemu.c and patch 3 shows how the master secret code would use the same function to build it's command line. Patch 4 just follows suit... Patch 5 was reviewed/ACK'd - this patch just shows the changes to remove the "virSecretPtr secret = NULL;" and "virObjectUnref(secret);" Patch 6 is unchanged John Ferlan (6): qemu: Move and rename qemuBuildObjectCommandlineFromJSON util: Introduce virQEMUBuildSecretObjectProps qemu: Use virQEMUBuildSecretObjectProps to build secret objects qemu: Rework secinfo command line building storage: Use virSecretGetSecretString secret: Move virStorageSecretType to secret_util and rename cfg.mk | 2 +- po/POTFILES.in | 1 + src/Makefile.am | 2 + src/libvirt_private.syms | 5 + src/libxl/libxl_conf.c | 2 +- src/qemu/qemu_command.c | 219 +++++---------------- src/qemu/qemu_command.h | 4 - src/qemu/qemu_domain.c | 4 +- src/secret/secret_util.c | 18 +- src/secret/secret_util.h | 22 ++- src/storage/storage_backend_iscsi.c | 55 +----- src/storage/storage_backend_rbd.c | 49 +---- src/util/virqemu.c | 211 ++++++++++++++++++++ src/util/virqemu.h | 42 ++++ src/util/virstoragefile.c | 33 ++-- src/util/virstoragefile.h | 17 +- tests/qemuargv2xmltest.c | 4 +- tests/qemucommandutiltest.c | 9 +- ...muxml2argv-disk-drive-network-rbd-auth-AES.args | 3 +- .../qemuxml2argvdata/qemuxml2argv-master-key.args | 3 +- .../qemuxml2argvdata/qemuxml2argv-name-escape.args | 3 +- 21 files changed, 380 insertions(+), 328 deletions(-) create mode 100644 src/util/virqemu.c create mode 100644 src/util/virqemu.h -- 2.5.5

1 7

[libvirt] RFC: turn cpu 'host-model' into 'custom' for active config on domain start
by Nikolay Shirokovskiy 01 Jun '16

01 Jun '16

Hi, everyone. Cpu mode 'host-model' is not convinient for migration in current form. Consider migration from less capable host to more capable one - it is possible. But then migration backwards is impossible despite the fact that qemu process on destination is running with the same model and features as on source because migration does not update 'host-model' cpu upon migration on destination. The problem is that on migrating back instead of taking current cpu config 'host-model' is expanded again and gain extra features of destination and migration failed by source (new destination) side. What if we change cpu for active config from 'host-model' into 'custom' in the starting process? Then this issue will be fixed. It is quite reasonable from my POV - 'host-model' is effectively about taking active domain cpu from host on start, after that active domain cpu is on its own. Persistent config stays 'host-model' so migrating of active domain and restarting it gains extra features of new host as before. Nikolay

3 2

[libvirt] NFS over AF_VSOCK in <filesystem>
by Stefan Hajnoczi 01 Jun '16

01 Jun '16

virtio-vsock support has been added to the nfs-ganesha NFS server. I'm currently working on upstreaming virtio-vsock into Linux and QEMU. I also have patches for the Linux NFS client and server. Users wishing to share a file system with the guest will need to configure the NFS server. Perhaps libvirt could handle that given that it already has <filesystem> syntax. The basic task is setting up either the kernel nfsd or nfs-ganesha for the VM to access the NFS export(s). When the VM is destroy the NFS server can be shut down. Does this sound like a responsibility that libvirt should handle? Stefan

4 11

[libvirt] [PATCH 0/3] bhyve: virConnectDomainXMLFromNative
by Fabian Freyer 01 Jun '16

01 Jun '16

The following series of patches implement virConnectDomainXMLFromNative. This is mostly some boiler plate code generating argv-lists from the native command string, and two getopt-based argument parsers for the /usr/sbin/bhyve and /usr/sbin/bhyveload commands. Since there is quite some string handling involved especially in the first patch that I am not 100% sure about, I'd appreciate a thorough review. Fabian Freyer (3): bhyve: implement virConnectDomainXMLFromNative bhyve: implement bhyve argument parser bhyve: implement argument parser for loader po/POTFILES.in | 1 + src/Makefile.am | 2 + src/bhyve/bhyve_driver.c | 42 ++ src/bhyve/bhyve_parse_command.c | 869 ++++++++++++++++++++++++++++++++++++++++ src/bhyve/bhyve_parse_command.h | 30 ++ 5 files changed, 944 insertions(+) create mode 100644 src/bhyve/bhyve_parse_command.c create mode 100644 src/bhyve/bhyve_parse_command.h -- 2.7.0

1 4

Re: [libvirt] [ovs-dev] [PATCH v2 2/2] netdev-dpdk: Support user-defined socket attribs
by Ansis Atteka 01 Jun '16

01 Jun '16

On Mon, May 30, 2016 at 12:29 AM, Christian Ehrhardt <christian.ehrhardt(a)canonical.com> wrote: > On Tue, May 24, 2016 at 4:10 PM, Aaron Conole <aconole(a)redhat.com> wrote: > >> Daniele Di Proietto <diproiettod(a)vmware.com> writes: >> >> > Hi Aaron, >> > >> > I'm still a little bit nervous about calling chown on a (partially) >> > user controlled file name. >> >> I agree, that always seems scary. >> >> > Before moving forward I wanted to discuss a couple of other options: >> > >> > * Ansis (in CC) suggested using -runas parameter in qemu. This way >> > qemu can open the socket as root and drop privileges before starting >> > guest execution. >> >> I'm not sure how to do this with libvirt, or via the OpenStack Neutron >> plugin. I also don't know if it would be an acceptable workaround for >> users. Additionally, I recall there being something of a "don't even >> know if this works" around it. Maybe Christian or Ansis (both in CC) >> can expound on it. >> Cross-posting to libvirt mailing list to hear opinion from libvirt developers. In short - the problem is that libvirtd process starts qemu process under non-root user. Since qemu starts under non-root process, then qemu can't connect to DPDK unix domain sockets created by ovs-vswitcd process that runs under "root". There are two solutions to this problem: 1. let ovs-vswitchd process to chown its socket from "root" to "libvirt" group and/or user (this is what Aarons patch proposes) 2. make libvirtd process to start qemu process under "root" but then let qemu to downgrade via "-run-as" flag after qemu has opened the Unix Domain socket. Regarding solution #2. I think the necessary changes roughly would be to: 1. invoke virCommandAddArgPair(cmd, "-runas", "libvirt") before starting qemu process; AND 2. revert virCommandSetUID() that automatically downgrades user from "root" to "libvirt" even before qemu starts. I would like to hear feasibility of such solution from libvirt developers? Or maybe there is even a better solution that I am missing? Regarding solution #1. There have already been precedents that other daemons are chowning their sockets so that non-privileged daemons could connect to them. IMHO this is not elegant from security perspective but will at least get the job done if Linux Distribution maintainers (Aaron and Christian) need to ship something out ASAP . The thing I am concerned about this solutions is that ovs-vswitchd could be used to chown() arbitrary files and sockets on file system if there happens to be a bug. Also, this socket is not created by ovs-vswitchd code base itself but rather from Intel DPDK library where ovs-vswitchd calls into. > Hi, > IIRC we kind of agree that long term a proper MAC will be much better but > most involved people needed something to get it working like "now". > Since they are complementary (other than the fix removing a bit of the > urgency for more MAC) it was kind of the least bad option. > > You have to be aware that I brought up the discussion on dev(a)dpdk.org - see > [1] and [2]: > But this will take time and eventually still be the applications task to > "do something" - no matter if via API or via the chmod's right now. > So Aaron is trying to get something that works now until the long term > things are in place, which I appreciate. > > FYI - I was even more in a hurry as it was clear that OVS-2.5 won't get > this in time I run with [3] for now. > I never intended to suggest that, but with the discussion in place, one > could ask if you (Aaron) want to pick up that instead. > That would keep OVS free for now until DPDK made up the API (see [2]) for > socket ownership control and this then could be implemented in OVS? > > (I hope) In some months/years we will all be happy to drop this bunch of > interim solutions, never the less we need it for now. > > [1]: http://dpdk.org/dev/patchwork/patch/12222/ > [2]: http://dpdk.org/ml/archives/dev/2015-December/030326.html > [3]: > https://git.launchpad.net/~ubuntu-server/dpdk/commit/?h=ubuntu-xenial-to-dp… > > [...] > > >> I think originally we quickly discussed 4 possible solutions (and >> hopefully I captured them correctly): >> >> 1. OVS downgrades to the ovs user, and kvm runs under the ovs >> group. I don't actually like this solution because kvm could then >> pollute the ovs database. >> >> 2. OVS runs as some user and sets the user/group ownership of the socket >> via chown/chmod where permissions come from the database (the >> original context had ovs running as root - but as I described above >> it doesn't need to be root provided ovs+DPDK can start without root). >> >> 3. OVS runs as some user, kvm starts as root, opens the socket and >> downgrades. IIRC, this doesn't actually work, or it may have >> implications on other projects. I don't remember exactly what was >> not as great about this solution, TBH. >> >> 4. OVS and KVM run as whatever users; MAC is used to enforce the >> layering between them. >> >> I think solution 2 and solution 4 don't actually interfere with each >> other, and can be used to a complementary effect (if implemented >> properly) so that the MAC layer enforces access, but even without MAC, >> the DAC layer can provide appropriate whitelisting behavior. >> > > I also remember several complex changes needed for the #1 and #3 that > always would end up with huge effort and a high risk not being accepted. > Probably that is what you refer to with "implications on other projects". > > Also keep in mind the position of dpdk out of the last few discussions > which I'd like to summarize as "dpdk got this path from an app, so this app > OWNS that path". > _______________________________________________ > dev mailing list > dev(a)openvswitch.org > http://openvswitch.org/mailman/listinfo/dev

3 4

[libvirt] [PATCH] Removed function virDomainDefNewFull.
by Tomáš Ryšavý 31 May '16

31 May '16

The function virDomainDefNewFull() in src/conf/domain_conf.c was a thin wrapper around virDomainDefNew() that was only used in a few places in the code. The function was removed and the callers were re-implemented. --- src/conf/domain_conf.c | 22 ---------------------- src/conf/domain_conf.h | 3 --- src/libvirt_private.syms | 1 - src/vz/vz_utils.c | 8 +++++++- src/xen/xen_hypervisor.c | 26 ++++++++++++++++++-------- src/xen/xend_internal.c | 23 +++++++++++++++++++---- src/xen/xm_internal.c | 24 ++++++++++++++++++++++-- 7 files changed, 66 insertions(+), 41 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index fb05bf7..5721800 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2754,28 +2754,6 @@ virDomainDefNew(void) } -virDomainDefPtr -virDomainDefNewFull(const char *name, - const unsigned char *uuid, - int id) -{ - virDomainDefPtr def; - - if (!(def = virDomainDefNew())) - return NULL; - - if (VIR_STRDUP(def->name, name) < 0) { - VIR_FREE(def); - return NULL; - } - - memcpy(def->uuid, uuid, VIR_UUID_BUFLEN); - def->id = id; - - return def; -} - - void virDomainObjAssignDef(virDomainObjPtr domain, virDomainDefPtr def, bool live, diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 82e581b..afb722c 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2507,9 +2507,6 @@ void virDomainDefFree(virDomainDefPtr vm); virDomainChrDefPtr virDomainChrDefNew(void); virDomainDefPtr virDomainDefNew(void); -virDomainDefPtr virDomainDefNewFull(const char *name, - const unsigned char *uuid, - int id); void virDomainObjAssignDef(virDomainObjPtr domain, virDomainDefPtr def, diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index fb5b419..1a09a95 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -234,7 +234,6 @@ virDomainDefMaybeAddController; virDomainDefMaybeAddInput; virDomainDefNeedsPlacementAdvice; virDomainDefNew; -virDomainDefNewFull; virDomainDefParseFile; virDomainDefParseNode; virDomainDefParseString; diff --git a/src/vz/vz_utils.c b/src/vz/vz_utils.c index 5427314..0e98a39 100644 --- a/src/vz/vz_utils.c +++ b/src/vz/vz_utils.c @@ -166,9 +166,15 @@ vzNewDomain(vzDriverPtr driver, const char *name, const unsigned char *uuid) virDomainObjPtr dom = NULL; vzDomObjPtr pdom = NULL; - if (!(def = virDomainDefNewFull(name, uuid, -1))) + if (!(def = virDomainDefNew())) goto error; + if (VIR_STRDUP(def->name, name) < 0) + goto error; + + memcpy(def->uuid, uuid, VIR_UUID_BUFLEN); + def->id = -1; + if (VIR_ALLOC(pdom) < 0) goto error; diff --git a/src/xen/xen_hypervisor.c b/src/xen/xen_hypervisor.c index fc9e1c6..1b9d2e1 100644 --- a/src/xen/xen_hypervisor.c +++ b/src/xen/xen_hypervisor.c @@ -2634,10 +2634,15 @@ xenHypervisorLookupDomainByID(virConnectPtr conn, int id) if (!name) return NULL; - ret = virDomainDefNewFull(name, - XEN_GETDOMAININFO_UUID(dominfo), - id); - VIR_FREE(name); + if (!(ret = virDomainDefNew())) { + VIR_FREE(name); + return NULL; + } + + ret->name = name; + memcpy(ret->uuid, XEN_GETDOMAININFO_UUID(dominfo), VIR_UUID_BUFLEN); + ret->id = id; + return ret; } @@ -2699,10 +2704,15 @@ xenHypervisorLookupDomainByUUID(virConnectPtr conn, const unsigned char *uuid) if (!name) return NULL; - ret = virDomainDefNewFull(name, uuid, id); - if (ret) - ret->id = id; - VIR_FREE(name); + if (!(ret = virDomainDefNew())) { + VIR_FREE(name); + return NULL; + } + + ret->name = name; + memcpy(ret->uuid, uuid, VIR_UUID_BUFLEN); + ret->id = id; + return ret; } diff --git a/src/xen/xend_internal.c b/src/xen/xend_internal.c index 4dd6b41..2ba5459 100644 --- a/src/xen/xend_internal.c +++ b/src/xen/xend_internal.c @@ -1111,12 +1111,21 @@ sexpr_to_domain(virConnectPtr conn ATTRIBUTE_UNUSED, const struct sexpr *root) if (sexpr_node(root, "domain/domid")) id = sexpr_int(root, "domain/domid"); - return virDomainDefNewFull(name, uuid, id); + if (!(ret = virDomainDefNew())) + goto error; + + if (VIR_STRDUP(ret->name, name) < 0) + goto error; + + memcpy(def->uuid, uuid, VIR_UUID_BUFLEN); + def->id = id; + + return ret; error: virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("failed to parse Xend domain information")); - virObjectUnref(ret); + virDomainDevFree(ret); return NULL; } @@ -2003,9 +2012,15 @@ xenDaemonLookupByUUID(virConnectPtr conn, const unsigned char *uuid) if (name == NULL) return NULL; - ret = virDomainDefNewFull(name, uuid, id); + if (!(ret = virDomainDefNew())) { + VIR_FREE(name); + return NULL; + } + + ret->name = name; + memcpy(ret->uuid, uuid, VIR_UUID_BUFLEN); + ret->id = id; - VIR_FREE(name); return ret; } diff --git a/src/xen/xm_internal.c b/src/xen/xm_internal.c index ef1a460..ca4577f 100644 --- a/src/xen/xm_internal.c +++ b/src/xen/xm_internal.c @@ -848,7 +848,17 @@ xenXMDomainLookupByName(virConnectPtr conn, const char *domname) if (!(entry = virHashLookup(priv->configCache, filename))) goto cleanup; - ret = virDomainDefNewFull(domname, entry->def->uuid, -1); + if (!(ret = virDomainDefNew())) + goto cleanup; + + if (VIR_STRDUP(ret->name, domname) < 0) { + virDomainDevFree(ret); + ret = NULL; + goto cleanup; + } + + memcpy(def->uuid, entry->def->uuid, VIR_UUID_BUFLEN); + def->id = -1; cleanup: xenUnifiedUnlock(priv); @@ -891,7 +901,17 @@ xenXMDomainLookupByUUID(virConnectPtr conn, const unsigned char *uuid) if (!(entry = virHashSearch(priv->configCache, xenXMDomainSearchForUUID, (const void *)uuid))) goto cleanup; - ret = virDomainDefNewFull(entry->def->name, uuid, -1); + if (!(ret = virDomainDefNew())) + goto cleanup; + + if (VIR_STRDUP(ret->name, entry->def->name) < 0) { + virDomainDevFree(ret); + ret = NULL; + goto cleanup; + } + + memcpy(def->uuid, uuid, VIR_UUID_BUFLEN); + def->id = -1; cleanup: xenUnifiedUnlock(priv); -- 2.5.5

3 2