- Devel - libvirt List Archives

[libvirt PATCH 0/3] xml: use g_auto for xmlXPathObject (glib chronicles)
by Ján Tomko 12 Aug '21

12 Aug '21

Ján Tomko (3): util: define cleanup func for xmlXPathObject util: xml: use g_auto for xmlXPathObject util: xml: clean up unnecessary ret variables src/util/virxml.c | 49 +++++++++++++---------------------------------- src/util/virxml.h | 1 + 2 files changed, 14 insertions(+), 36 deletions(-) -- 2.31.1

2 4

Plans for the next release
by Jiri Denemark 12 Aug '21

12 Aug '21

I'm sending this early as I'll be out until Aug 25, but don't worry I'll be back in time to handle the whole release process. To aim for the release on Sep 01 I suggest entering the freeze on Thursday Aug 26 and tagging RC2 on Monday Aug 30. I hope this works for everyone. Jirka

1 0

[PATCH 0/5] src: move validation to virXMLParseHelper()
by Kristina Hanicova 12 Aug '21

12 Aug '21

This series avoids duplication of the same code used for validation of XML document against rng schema and makes it easier to add validation of other objects in the future as well. Kristina Hanicova (5): src: add 'schema' and 'validate' variable to virXMLParseHelper() util: add xml validation against schema in virXMLParseHelper() domain_conf: replace validation with variables passed to virXMLParse() domain_conf: Remove redundant variable conf: replace validation with variables passed to virXMLParse() src/conf/backup_conf.c | 13 ++----------- src/conf/checkpoint_conf.c | 12 ++---------- src/conf/domain_conf.c | 27 +++++++-------------------- src/conf/interface_conf.c | 2 +- src/conf/network_conf.c | 2 +- src/conf/node_device_conf.c | 2 +- src/conf/nwfilter_conf.c | 2 +- src/conf/secret_conf.c | 2 +- src/conf/snapshot_conf.c | 15 ++------------- src/conf/storage_conf.c | 4 ++-- src/conf/virnetworkportdef.c | 2 +- src/conf/virnwfilterbindingdef.c | 2 +- src/conf/virnwfilterbindingobj.c | 2 +- src/test/test_driver.c | 2 +- src/util/virxml.c | 16 +++++++++++++++- src/util/virxml.h | 20 +++++++++++--------- src/vbox/vbox_snapshot_conf.c | 6 +++--- src/vz/vz_sdk.c | 2 +- tests/qemuxml2argvtest.c | 2 +- 19 files changed, 55 insertions(+), 80 deletions(-) -- 2.31.1

2 6

[libvirt PATCH] virsh: refactor cmdDomblkinfo
by Ján Tomko 12 Aug '21

12 Aug '21

Use automatic memory cleanup to get rid of the cleanup section, and of the memory leak that happens inside the loop, because cap, alloc and phy are only freed once per function. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- tools/virsh-domain-monitor.c | 62 +++++++++++++++--------------------- 1 file changed, 25 insertions(+), 37 deletions(-) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index a2bf5c05fc..55c8dcf118 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -452,22 +452,16 @@ static bool cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) { virDomainBlockInfo info; - virDomainPtr dom; - bool ret = false; + g_autoptr(virshDomain) dom = NULL; bool human = false; bool all = false; const char *device = NULL; - xmlDocPtr xmldoc = NULL; - xmlXPathContextPtr ctxt = NULL; + g_autoptr(xmlDoc) xmldoc = NULL; + g_autoptr(xmlXPathContext) ctxt = NULL; int ndisks; size_t i; - xmlNodePtr *disks = NULL; - char *target = NULL; - char *protocol = NULL; - char *cap = NULL; - char *alloc = NULL; - char *phy = NULL; - vshTable *table = NULL; + g_autofree xmlNodePtr *disks = NULL; + g_autoptr(vshTable) table = NULL; VSH_EXCLUSIVE_OPTIONS("all", "device"); @@ -477,7 +471,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) all = vshCommandOptBool(cmd, "all"); if (!all && vshCommandOptStringQuiet(ctl, cmd, "device", &device) <= 0) { vshError(ctl, "command 'domblkinfo' requires <device> option"); - goto cleanup; + return false; } human = vshCommandOptBool(cmd, "human"); @@ -487,18 +481,24 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) int rc; if (virshDomainGetXML(ctl, cmd, 0, &xmldoc, &ctxt) < 0) - goto cleanup; + return false; ndisks = virXPathNodeSet("./devices/disk", ctxt, &disks); if (ndisks < 0) - goto cleanup; + return false; /* title */ table = vshTableNew(_("Target"), _("Capacity"), _("Allocation"), _("Physical"), NULL); if (!table) - goto cleanup; + return false; for (i = 0; i < ndisks; i++) { + g_autofree char *target = NULL; + g_autofree char *protocol = NULL; + g_autofree char *cap = NULL; + g_autofree char *alloc = NULL; + g_autofree char *phy = NULL; + ctxt->node = disks[i]; protocol = virXPathString("string(./source/@protocol)", ctxt); target = virXPathString("string(./target/@dev)", ctxt); @@ -517,7 +517,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) memset(&info, 0, sizeof(info)); vshResetLibvirtError(); } else { - goto cleanup; + return false; } } } else { @@ -527,41 +527,29 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) } if (!cmdDomblkinfoGet(&info, &cap, &alloc, &phy, human)) - goto cleanup; + return false; if (vshTableRowAppend(table, target, cap, alloc, phy, NULL) < 0) - goto cleanup; - - VIR_FREE(target); - VIR_FREE(protocol); + return false; } vshTablePrintToStdout(table, ctl); } else { + g_autofree char *cap = NULL; + g_autofree char *alloc = NULL; + g_autofree char *phy = NULL; + if (virDomainGetBlockInfo(dom, device, &info, 0) < 0) - goto cleanup; + return false; if (!cmdDomblkinfoGet(&info, &cap, &alloc, &phy, human)) - goto cleanup; + return false; vshPrint(ctl, "%-15s %s\n", _("Capacity:"), cap); vshPrint(ctl, "%-15s %s\n", _("Allocation:"), alloc); vshPrint(ctl, "%-15s %s\n", _("Physical:"), phy); } - ret = true; - - cleanup: - vshTableFree(table); - VIR_FREE(cap); - VIR_FREE(alloc); - VIR_FREE(phy); - virshDomainFree(dom); - VIR_FREE(target); - VIR_FREE(protocol); - VIR_FREE(disks); - xmlXPathFreeContext(ctxt); - xmlFreeDoc(xmldoc); - return ret; + return true; } /* -- 2.31.1

2 1

[PATCH 0/2] versions of: save status xml after generating taint message
by Kristina Hanicova 11 Aug '21

11 Aug '21

Both of these patches solve the same bug and only one needs to be merged. The first one is a bit longer, but I think it is more effective (because of function qemuDomainObjCheckTaint, where qemuDomainObjTaint could be potentially called in multiple for cycles). The second one is short and probably more readable. I will leave it up to the reviewer to choose one. Kristina Hanicova (2): qemu: save status xml after generating taint message qemu: save status xml after generating taint message src/qemu/qemu_domain.c | 14 ++++++++++++++ src/qemu/qemu_domain.h | 5 +++++ src/qemu/qemu_driver.c | 4 ++-- 3 files changed, 21 insertions(+), 2 deletions(-) -- 2.31.1

2 3

[PATCH v2 00/11] Implement IOThreads related APIs for test driver
by Luke Yue 11 Aug '21

11 Aug '21

v2: - Move the extracted functions to hypervisor/domain_driver.c - Implement virDomainPinIOThread - Implement virDomainSetIOThreadParams - Implement virConnectGetAllDomainStats, currently only supports get state and IOThread info - Add more tests Luke Yue (11): domain_driver.c: Introduce and use virDomainDriverAddIOThreadCheck() test_driver: Introduce testIOThreadInfo and generate IOThread infos test_driver: Implement virDomainAddIOThread test_driver: Implement virDomainDelIOThread domain_driver.c: Introduce and use virDomainDriverGetIOThreadsConfig() test_driver: Implement virDomainGetIOThreadInfo test_driver: Implement virDomainPinIOThread test_driver: Implement testDomainSetIOThreadParams test_driver: Implement virConnectGetAllDomainStats test_driver: Introduce testDomainGetStatsIOThread tests: Test IOThread related functions for test driver examples/xml/test/testdomfc4.xml | 5 + src/hypervisor/domain_driver.c | 123 ++++++++ src/hypervisor/domain_driver.h | 9 + src/libvirt_private.syms | 3 + src/qemu/qemu_driver.c | 113 +------ src/test/meson.build | 1 + src/test/test_driver.c | 509 +++++++++++++++++++++++++++++++ tests/virshtest.c | 90 ++++++ 8 files changed, 745 insertions(+), 108 deletions(-) -- 2.32.0

2 13

[PATCH 0/4] Implement some job related APIs for test driver
by Luke Yue 11 Aug '21

11 Aug '21

Luke Yue (4): test_driver: Implement virDomainGetJobInfo test_driver: Implement virDomainGetJobStats test_driver: Implement virDomainAbortJob virshtest: add test for domjobinfo src/test/test_driver.c | 170 +++++++++++++++++++++++++++++++++++++++++ tests/virshtest.c | 11 +++ 2 files changed, 181 insertions(+) -- 2.32.0

2 8

[libvirt PATCH 0/3] Invalidate the cpu flags cache on changes of kernel command line
by Tim Wiederhake 11 Aug '21

11 Aug '21

The kernel command line can contain settings affecting the availability of cpu features, eg. "tsx=on". This series adds the kernel command line to the cpu flags cache and declares the cache invalid if the current kernel command line differs. Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1953389. Tim Wiederhake (3): virQEMUCaps: Add kernel command line virQEMUCapsCachePriv: Add kernel command line qemu: Invalidate capabilities cache on kernel command line mismatch src/qemu/qemu_capabilities.c | 34 ++++++++++++++++++++++++++++++++-- src/qemu/qemu_capspriv.h | 3 ++- tests/qemucapsprobe.c | 2 +- 3 files changed, 35 insertions(+), 4 deletions(-) -- 2.31.1

4 12

[libvirt PATCH] run: set abs_builddir variables
by Ján Tomko 11 Aug '21

11 Aug '21

test-lib.sh needs these to be set. Export them so that the virsh-* tests can be run using: builddir$ ./run srcdir/tests/virsh-snapshot Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- meson.build | 1 + run.in | 3 +++ 2 files changed, 4 insertions(+) diff --git a/meson.build b/meson.build index 32ad688c9c..5af09d319a 100644 --- a/meson.build +++ b/meson.build @@ -2233,6 +2233,7 @@ configure_file(output: 'meson-config.h', configuration: conf) # generate run helper run_conf = configuration_data() run_conf.set('abs_builddir', meson.build_root()) +run_conf.set('abs_top_builddir', meson.build_root()) configure_file( input: 'run.in', output: '@BASENAME@', diff --git a/run.in b/run.in index 9aae27a79e..64f99751fe 100644 --- a/run.in +++ b/run.in @@ -80,6 +80,9 @@ env["LIBVIRT_DIR_OVERRIDE"] = "1" # read problems when using glibc. env["MALLOC_PERTURB_"] = "%d" % random.randint(1, 255) +env["abs_builddir"] = "@abs_builddir@" +env["abs_top_builddir"] = "@abs_top_builddir@" + modular_daemons = [ "virtinterfaced", "virtlxcd", -- 2.31.1

2 1

[libvirt PATCH 00/10] virsh: moments: use g_auto more (glib chronicles)
by Ján Tomko 11 Aug '21

11 Aug '21

Ján Tomko (10): vsh: table: define cleanup function virsh: add cleanup function for virshDomainCheckpoint virsh: add cleanup function for virshDomainSnapshot virsh: checkpoint: move variables inside the loop virsh: snapshot: move variables inside the loop virsh: cmdCheckpointInfo: delete unused variables virsh: checkpoint: use g_auto where possible virsh: checkpoint: remove pointless cleanup labels virsh: snapshot: use g_auto where possible virsh: snapshot: remove pointless cleanup labels tools/virsh-checkpoint.c | 188 ++++++++----------------- tools/virsh-snapshot.c | 288 +++++++++++++-------------------------- tools/virsh-util.h | 4 + tools/vsh-table.h | 1 + 4 files changed, 156 insertions(+), 325 deletions(-) -- 2.31.1

2 11

[libvirt PATCH 0/5] vsh: use g_auto more (glib chronicles)
by Ján Tomko 11 Aug '21

11 Aug '21

Except for vshCommandParse. -EPARSE Ján Tomko (5): vsh: do not cast away const vsh: cmdEcho: use separate variable for argument vsh: use g_auto where possible vsh: remove pointless cleanup labels vsh: use g_clear_pointer tools/vsh.c | 91 ++++++++++++++++++----------------------------------- 1 file changed, 31 insertions(+), 60 deletions(-) -- 2.31.1

2 11

[libvirt PATCH] meson: Always use the / operator to join paths
by Andrea Bolognani 10 Aug '21

10 Aug '21

This is the preferred way to do it, but there were a few instances in which some of the path components had embedded slashes instead. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- docs/internals/meson.build | 2 +- docs/kbase/meson.build | 2 +- docs/manpages/meson.build | 2 +- src/security/meson.build | 2 +- src/storage/meson.build | 4 ++-- 5 files changed, 6 insertions(+), 6 deletions(-) diff --git a/docs/internals/meson.build b/docs/internals/meson.build index 5aa67ded5c..2d0ae6195c 100644 --- a/docs/internals/meson.build +++ b/docs/internals/meson.build @@ -12,7 +12,7 @@ html_xslt_gen = [] foreach name : internals_in_files html_xslt_gen += { 'name': name, - 'source': 'docs/internals' / name + '.html.in', + 'source': 'docs' / 'internals' / name + '.html.in', } endforeach diff --git a/docs/kbase/meson.build b/docs/kbase/meson.build index 8d8fd695ea..73dc11837f 100644 --- a/docs/kbase/meson.build +++ b/docs/kbase/meson.build @@ -31,7 +31,7 @@ foreach name : docs_kbase_files html_xslt_gen += { 'name': name, 'file': docs_rst2html5_gen.process(rst_file), - 'source': 'docs/kbase' / rst_file, + 'source': 'docs' / 'kbase' / rst_file, } endforeach diff --git a/docs/manpages/meson.build b/docs/manpages/meson.build index fe91a43614..cafe29a545 100644 --- a/docs/manpages/meson.build +++ b/docs/manpages/meson.build @@ -125,7 +125,7 @@ foreach data : docs_man_files html_xslt_gen += { 'name': data['name'], 'file': html_in, - 'source': 'docs/manpages' / rst_in_file, + 'source': 'docs' / 'manpages' / rst_in_file, } endforeach diff --git a/src/security/meson.build b/src/security/meson.build index 6f5e1dec1d..6230b34aa4 100644 --- a/src/security/meson.build +++ b/src/security/meson.build @@ -47,7 +47,7 @@ if conf.has('WITH_LIBVIRTD') and conf.has('WITH_APPARMOR') conf_inc_dir, hypervisor_inc_dir, include_directories('.'), - include_directories('../storage_file'), + include_directories('..' / 'storage_file'), ], } endif diff --git a/src/storage/meson.build b/src/storage/meson.build index 915ae46f61..d1aa66b147 100644 --- a/src/storage/meson.build +++ b/src/storage/meson.build @@ -79,7 +79,7 @@ if conf.has('WITH_STORAGE') ], include_directories: [ conf_inc_dir, - include_directories('../storage_file'), + include_directories('..' / 'storage_file'), ], ) @@ -163,7 +163,7 @@ if conf.has('WITH_STORAGE_GLUSTER') glusterfs_dep, ], 'include': [ - include_directories('../storage_file'), + include_directories('..' / 'storage_file'), ], 'install_dir': storage_backend_install_dir, } -- 2.31.1

2 1

[PATCH v2 0/4] interface define: add support for validation against schema
by Kristina Hanicova 10 Aug '21

10 Aug '21

This is v2 of: https://listman.redhat.com/archives/libvir-list/2021-August/msg00187.html Changes since v1 (suggested by Peter): * split of the patch containing api and virsh adjustments into separate patches * change of documentation (which I forgot to do before) Kristina Hanicova (4): conf: propagate xmlDocPtr and flags into virInterfaceDefParseXML() api: add virInterfaceDefineFlags virsh: add support for '--validate' option in define interface interface_conf: add validation against schema in define docs/manpages/virsh.rst | 5 ++-- include/libvirt/libvirt-interface.h | 4 +++ src/conf/interface_conf.c | 38 ++++++++++++++++++------- src/conf/interface_conf.h | 6 ++-- src/conf/virinterfaceobj.c | 2 +- src/interface/interface_backend_netcf.c | 6 ++-- src/libvirt-interface.c | 2 +- src/test/test_driver.c | 4 +-- tests/interfacexml2xmltest.c | 2 +- tools/virsh-interface.c | 10 ++++++- 10 files changed, 55 insertions(+), 24 deletions(-) -- 2.31.1

2 6

[PATCH 0/4] qemu: Open chardev logfile on behalf of QEMU
by Michal Privoznik 10 Aug '21

10 Aug '21

See 4/4 for explanation. Michal Prívozník (4): virDomainDefGetSecurityLabelDef: Fix const correctness qemuDomainOpenFile: Take virDomainDef instead of virDomainObj qemuDomainOpenFile: Take @cfg instead of driver qemu: Open chardev logfile on behalf of QEMU src/conf/domain_conf.c | 2 +- src/conf/domain_conf.h | 2 +- src/qemu/qemu_command.c | 68 +++++++++++++++++++++++++++++---------- src/qemu/qemu_domain.c | 13 ++++---- src/qemu/qemu_domain.h | 4 +-- src/qemu/qemu_driver.c | 4 +-- src/qemu/qemu_saveimage.c | 5 +-- 7 files changed, 66 insertions(+), 32 deletions(-) -- 2.31.1

3 6

[PATCH v2] domain_conf: Fix memory leak when free'ing virDomainObj
by Luke Yue 10 Aug '21

10 Aug '21

Free dom->deprecations when free'ing virDomainObj Signed-off-by: Luke Yue <lukedyue(a)gmail.com> --- src/conf/domain_conf.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 09da4ab952..7e1d30bdb2 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -3730,6 +3730,16 @@ void virDomainDefFree(virDomainDef *def) g_free(def); } +static void +virDomainObjDeprecationFree(virDomainObj *dom) +{ + size_t i = 0; + for (i = 0; i < dom->ndeprecations; i++) { + g_free(dom->deprecations[i]); + } + g_free(dom->deprecations); +} + static void virDomainObjDispose(void *obj) { virDomainObj *dom = obj; @@ -3742,6 +3752,7 @@ static void virDomainObjDispose(void *obj) if (dom->privateDataFreeFunc) (dom->privateDataFreeFunc)(dom->privateData); + virDomainObjDeprecationFree(dom); virDomainSnapshotObjListFree(dom->snapshots); virDomainCheckpointObjListFree(dom->checkpoints); } -- 2.32.0

2 1

[PATCH 0/2] virDomain: fix problems when setting qos on ovs managed interface
by zhangjl02 10 Aug '21

10 Aug '21

Two problems are found and fixed below: 1. Ingress rules is not clean on previous version of virNetDevOpenvswitchInterfaceClearQos. 2. If errors occurs when removing inbound qos on multi interfaces vm, some rules may not be delete as aspected. Fix by: Instead of cleaning all qos rules each time new qos is set, tx and rx's qos are set or cleaned respectively. Replace virReportError with VIR_WARN to let the cleaning process continue when error occurs. Add ifname into ovs querying statements, which will reduce failure of removing qos on the other interfaces of the same vm. Test virNetDevOpenvswitchInterfaceSetQos and virNetDevOpenvswitchInterfaceClearQos with dryrun method. Since commands in tests are not actually run, it is difficult to emulate some complex senario, such as set and then update qos. So basic tests are added in patches. Jinsheng Zhang (2): virDomain: interface: add virNetDevOpenvswitchInterfaceClearTxQos and virNetDevOpenvswitchInterfaceClearRxQos. tests: add test on virNetDevOpenvswitchInterfaceSetQos and virNetDevOpenvswitchInterfaceClearQos src/libvirt_private.syms | 2 + src/util/virnetdevopenvswitch.c | 69 +++++++++-- src/util/virnetdevopenvswitch.h | 7 ++ tests/virnetdevopenvswitchtest.c | 196 ++++++++++++++++++++++++++++++- 4 files changed, 262 insertions(+), 12 deletions(-) -- 2.30.2.windows.1

2 3

[PATCH] domain_conf: Fix memory leak when free'ing virDomainObj
by Luke Yue 10 Aug '21

10 Aug '21

Free dom->deprecations when free'ing virDomainObj Signed-off-by: Luke Yue <lukedyue(a)gmail.com> --- src/conf/domain_conf.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 09da4ab952..2527cd1d70 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -3733,6 +3733,7 @@ void virDomainDefFree(virDomainDef *def) static void virDomainObjDispose(void *obj) { virDomainObj *dom = obj; + size_t i = 0; VIR_DEBUG("obj=%p", dom); virCondDestroy(&dom->cond); @@ -3742,6 +3743,11 @@ static void virDomainObjDispose(void *obj) if (dom->privateDataFreeFunc) (dom->privateDataFreeFunc)(dom->privateData); + for (i = 0; i < dom->ndeprecations; i++) { + g_free(dom->deprecations[i]); + } + g_free(dom->deprecations); + virDomainSnapshotObjListFree(dom->snapshots); virDomainCheckpointObjListFree(dom->checkpoints); } -- 2.32.0

2 2

[PATCH 0/3] interface define: add support for validation against schema
by Kristina Hanicova 06 Aug '21

06 Aug '21

This is just the first series of patches of many more. I will send them soon, I just wanted to know if I am on the right path as they will follow the same pattern. Partially resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1972586 Kristina Hanicova (3): conf: propagate xmlDocPtr and flags into virInterfaceDefParseXML() virsh: add support for '--validate' option in define interface interface_conf: add validation against schema in define include/libvirt/libvirt-interface.h | 4 +++ src/conf/interface_conf.c | 38 ++++++++++++++++++------- src/conf/interface_conf.h | 6 ++-- src/conf/virinterfaceobj.c | 2 +- src/interface/interface_backend_netcf.c | 6 ++-- src/test/test_driver.c | 4 +-- tests/interfacexml2xmltest.c | 2 +- tools/virsh-interface.c | 10 ++++++- 8 files changed, 51 insertions(+), 21 deletions(-) -- 2.31.1

2 4

[PATCH] src: stop checking virIdentityNew return value
by Daniel P. Berrangé 06 Aug '21

06 Aug '21

This method will always succeed. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/remote/remote_daemon_dispatch.c | 5 +---- src/rpc/virnetserverclient.c | 5 +---- src/util/viridentity.c | 5 +---- 3 files changed, 3 insertions(+), 12 deletions(-) diff --git a/src/remote/remote_daemon_dispatch.c b/src/remote/remote_daemon_dispatch.c index 69d0a015ba..0e38b314a2 100644 --- a/src/remote/remote_daemon_dispatch.c +++ b/src/remote/remote_daemon_dispatch.c @@ -2192,7 +2192,7 @@ remoteDispatchConnectSetIdentity(virNetServer *server G_GNUC_UNUSED, int nparams = 0; int rv = -1; virConnectPtr conn = remoteGetHypervisorConn(client); - g_autoptr(virIdentity) ident = NULL; + g_autoptr(virIdentity) ident = virIdentityNew(); if (!conn) goto cleanup; @@ -2209,9 +2209,6 @@ remoteDispatchConnectSetIdentity(virNetServer *server G_GNUC_UNUSED, if (virConnectSetIdentityEnsureACL(conn) < 0) goto cleanup; - if (!(ident = virIdentityNew())) - goto cleanup; - if (virIdentitySetParameters(ident, params, nparams) < 0) goto cleanup; diff --git a/src/rpc/virnetserverclient.c b/src/rpc/virnetserverclient.c index 236702ced6..7050430e17 100644 --- a/src/rpc/virnetserverclient.c +++ b/src/rpc/virnetserverclient.c @@ -747,10 +747,7 @@ virNetServerClientCreateIdentity(virNetServerClient *client) g_autofree char *username = NULL; g_autofree char *groupname = NULL; g_autofree char *seccontext = NULL; - g_autoptr(virIdentity) ret = NULL; - - if (!(ret = virIdentityNew())) - return NULL; + g_autoptr(virIdentity) ret = virIdentityNew(); if (client->sock && virNetSocketIsLocal(client->sock)) { gid_t gid; diff --git a/src/util/viridentity.c b/src/util/viridentity.c index c18326c8cb..e36e54ae4b 100644 --- a/src/util/viridentity.c +++ b/src/util/viridentity.c @@ -313,15 +313,12 @@ virIdentity *virIdentityGetSystem(void) g_autofree char *username = NULL; g_autofree char *groupname = NULL; unsigned long long startTime; - g_autoptr(virIdentity) ret = NULL; + g_autoptr(virIdentity) ret = virIdentityNew(); #if WITH_SELINUX char *con; #endif g_autofree char *token = NULL; - if (!(ret = virIdentityNew())) - return NULL; - if (virIdentitySetProcessID(ret, getpid()) < 0) return NULL; -- 2.31.1

2 1

[PATCH 00/13] Clean up string list freeing
by Peter Krempa 06 Aug '21

06 Aug '21

Switch remaining users of virStringListFreeCount to g_auto(GStrv) and clean up some usage of string lists. Depends on the refactor of the virtual function code I've posted earlier. Pipeline: https://gitlab.com/pipo.sk/libvirt/-/pipelines/348627403 Peter Krempa (13): testQemuAgentSSHKeys: Refactor cleanup remote: dispatch: Don't use virStringListFreeCount for NULL terminated lists qemu: firmware: Store machine types as a NULL-terminated string list qemu: domain: Store passthrough environment variables in a struct qemu: domain: Store capability overrides in NULL-terminated string list qemu: domain: Store passthrough arguments in NULL-terminated string list network: bridge: Store dnsmasq passthrough options in NULL-terminated string list qemuNamespacePrepareOneItem: Restructure code to avoid temporary variables qemuNamespacePrepareOneItem: Don't pass count of elements qemuNamespaceMknodPaths: Remove 'ndevMountsPath' qemuDomainUnshareNamespace: Use automatic memory clearing for string lists lxcContainerSetReadOnly: Refactor cleanup handling util: virstring: Remove unused virStringListFreeCount src/libvirt_private.syms | 1 - src/lxc/lxc_container.c | 27 ++++----- src/network/bridge_driver.c | 23 ++++---- src/qemu/qemu_command.c | 9 +-- src/qemu/qemu_domain.c | 91 +++++++++++++++-------------- src/qemu/qemu_domain.h | 12 ++-- src/qemu/qemu_firmware.c | 4 +- src/qemu/qemu_namespace.c | 46 ++++++--------- src/qemu/qemu_process.c | 13 ++--- src/remote/remote_daemon_dispatch.c | 8 +-- src/util/virstring.c | 23 -------- src/util/virstring.h | 3 - tests/qemuagenttest.c | 54 ++++++++--------- 13 files changed, 135 insertions(+), 179 deletions(-) -- 2.31.1

2 15

[PATCH 0/6] util: Refactor fetching of virtual function list
by Peter Krempa 06 Aug '21

06 Aug '21

Note that this applies on top of the 'VIR_APPEND_ELEMENT' refactor I've sent earlier. Pipeline: https://gitlab.com/pipo.sk/libvirt/-/pipelines/348524810 Peter Krempa (6): virNetDevGetVirtualFunctions: Remove 'max_vfs' argument virPCIGetVirtualFunctions: Simplify cleanup of returned data virPCIGetVirtualFunctionIndex: Refactor cleanup virPCIGetNetName: Make 'physPortID' argument const virPCIGetVirtualFunctions: Fetch also network interface name if needed virNetDevGetVirtualFunctions: Directly return virPCIVirtualFunctionList src/conf/node_device_conf.c | 16 +++-- src/libvirt_private.syms | 2 + src/network/bridge_driver.c | 23 ++----- src/util/virnetdev.c | 65 +++--------------- src/util/virnetdev.h | 8 +-- src/util/virpci.c | 127 +++++++++++++++++++++--------------- src/util/virpci.h | 24 +++++-- 7 files changed, 122 insertions(+), 143 deletions(-) -- 2.31.1

2 7

[PATCH 0/7] vmx: Support super wide SCSI bus
by Michal Privoznik 06 Aug '21

06 Aug '21

As of vSphere 6.7 there can be up to 64 units on a SCSI bus. Honestly, I'm not a big fan of 5/7 but we want to support three different values and one of them doesn't depend on the driver that's assigning the disk address, but at a runtime information. Thus using xmlopt is not good enough. Any clever idea is very welcome. Michal Prívozník (7): vmx: Fill virtualHW.version to ESX version mapping vmx2xmltest: Add a test case vmx: Rework disk def allocation vmx: Drop needless check in virVMXParseDisk() conf: Store SCSI bus length in virDomainDef vmx: Support super wide SCSI bus DO NOT MERGE src/bhyve/bhyve_parse_command.c | 2 +- src/conf/domain_conf.c | 33 +++--- src/conf/domain_conf.h | 10 +- src/hyperv/hyperv_driver.c | 4 +- src/libxl/libxl_driver.c | 2 +- src/libxl/xen_xl.c | 2 +- src/libxl/xen_xm.c | 2 +- src/lxc/lxc_native.c | 2 +- src/openvz/openvz_conf.c | 2 +- src/qemu/qemu_process.c | 2 +- src/vbox/vbox_common.c | 8 +- src/vmx/vmx.c | 108 +++++++++--------- src/vz/vz_sdk.c | 2 +- tests/genericxml2xmltest.c | 2 +- tests/openvzutilstest.c | 2 +- tests/qemublocktest.c | 2 +- tests/qemumonitortestutils.c | 2 +- tests/securityselinuxtest.c | 2 +- .../vmx2xml-esx-in-the-wild-11.vmx | 97 ++++++++++++++++ .../vmx2xml-esx-in-the-wild-11.xml | 43 +++++++ tests/vmx2xmltest.c | 1 + 21 files changed, 239 insertions(+), 91 deletions(-) create mode 100644 tests/vmx2xmldata/vmx2xml-esx-in-the-wild-11.vmx create mode 100644 tests/vmx2xmldata/vmx2xml-esx-in-the-wild-11.xml -- 2.31.1

2 9

[libvirt PATCH 0/7] Fix some mutex related issues
by Tim Wiederhake 05 Aug '21

05 Aug '21

The calls to pthread_mutex_{lock,unlock,destroy} in "virthread.c" ignore the return value of these functions. This series fixes all "double unlock mutex" and "destroy locked mutex" issues I came across and adds checks to the calls to pthread_mutex_*, so no new issues of this type are introduced. Patches 1, 2, 3, and 4 fix the issues, patches 5, 6, and 7 add the checks. See https://gitlab.com/twiederh/libvirt/-/pipelines/348505233 and https://gitlab.com/twiederh/libvirt/-/pipelines/348505233 (only patches 5, 6, and 7). Regards, Tim Tim Wiederhake (7): tests: qemuMonitorTestFree: Unlock virDomainObj before freeing qemusecuritytest: Unlock virDomainObj before disposal qemuxml2argvtest: Unlock virDomainObj before disposal qemuhotplugtest: Lock reused virDomainObj Meson: Define "DYNAMIC_ANALYSIS" if built with sanitizers enabled virMutex*: Warn on error virMutex*: Fail loudly meson.build | 4 ++++ src/util/virthread.c | 24 +++++++++++++++++++++--- tests/qemuhotplugtest.c | 1 + tests/qemumonitortestutils.c | 2 +- tests/qemusecuritytest.c | 3 ++- tests/qemuxml2argvtest.c | 12 +++++++----- 6 files changed, 36 insertions(+), 10 deletions(-) -- 2.31.1

3 20

[PATCH 00/23] Remove return value from VIR_APPEND_ELEMENT and cleanups
by Peter Krempa 05 Aug '21

05 Aug '21

Since this touches all kind of conditionally compiled code you can find the pipeline build results here: https://gitlab.com/pipo.sk/libvirt/-/pipelines/347734973 Peter Krempa (23): virInsertElementsN: Rename 'add' argument virInsertElementsN: Split out actual insertion code util: alloc: Introduce virAppendElement helper util: alloc: Reimplement VIR_APPEND_ELEMENT_(COPY_)INPLACE using virAppendElement util: alloc: Reimplement VIR_APPEND_ELEMENT_COPY using virAppendElement util: alloc: Reimplement VIR_APPEND_ELEMENT_QUIET using virAppendElement util: alloc: Reimplement VIR_APPEND_ELEMENT using virAppendElement util: alloc: Completely replace VIR_APPEND_ELEMENT_QUIET by VIR_APPEND_ELEMENT prlsdkAddDomainVideoInfoCt: Remove pointless cleanup section xenParseXMDisk: Use automatic memory clearing and remove 'ret' variable qemuDomainUSBAddressAddHubs: Refactor cleanup virSecuritySELinuxContextListAppend: Remove unreachable cleanup qemuDomainAttachDeviceConfig: Remove pointless assignment virNWFilterRuleDefToRuleInst: Remove pointless assignment virObjectEventCallbackListAddID: Remove pointless cleanup of 'cb' virNWFilterRuleDefToRuleInst: Restructure code to avoid cleanup virNWFilterIncludeDefToRuleInst: Refactor cleanup qemuProcessSetupHotpluggableVcpus: Use automatic memory freeing virNetServerGetClients: Remove pointless cleanup lxcNetworkParseDataIPs: Automatically free string list virStorageBackendLogicalParseVolExtents: Declare one variable per line virStorageBackendLogicalParseVolExtents: Move 'extents' inside the loop virStorageBackendLogicalParseVolExtents: Remove 'cleanup' and 'ret' src/bhyve/bhyve_parse_command.c | 18 +-- src/conf/capabilities.c | 11 +- src/conf/domain_conf.c | 95 ++++++---------- src/conf/network_conf.c | 9 +- src/conf/node_device_conf.c | 24 ++-- src/conf/nwfilter_conf.c | 12 +- src/conf/object_event.c | 9 +- src/conf/storage_conf.c | 6 +- src/conf/virdomainobjlist.c | 6 +- src/conf/virnwfilterobj.c | 8 +- src/cpu/cpu_arm.c | 6 +- src/cpu/cpu_ppc64.c | 6 +- src/cpu/cpu_x86.c | 21 ++-- src/esx/esx_driver.c | 7 +- src/esx/esx_network_driver.c | 3 +- src/hyperv/hyperv_driver.c | 22 +--- src/hyperv/hyperv_network_driver.c | 3 +- src/libvirt_private.syms | 1 + src/libxl/xen_common.c | 17 +-- src/libxl/xen_xl.c | 21 +--- src/libxl/xen_xm.c | 21 +--- src/logging/log_handler.c | 6 +- src/lxc/lxc_container.c | 5 +- src/lxc/lxc_controller.c | 3 +- src/lxc/lxc_native.c | 12 +- src/network/bridge_driver.c | 4 +- src/node_device/node_device_driver.c | 3 +- src/nwfilter/nwfilter_ebiptables_driver.c | 3 +- src/nwfilter/nwfilter_gentech_driver.c | 57 ++++------ src/openvz/openvz_conf.c | 8 +- src/qemu/qemu_block.c | 9 +- src/qemu/qemu_command.c | 14 +-- src/qemu/qemu_domain.c | 6 +- src/qemu/qemu_domain_address.c | 9 +- src/qemu/qemu_driver.c | 17 +-- src/qemu/qemu_firmware.c | 3 +- src/qemu/qemu_hotplug.c | 6 +- src/qemu/qemu_monitor_json.c | 3 +- src/qemu/qemu_namespace.c | 5 +- src/qemu/qemu_process.c | 17 +-- src/remote/remote_daemon_dispatch.c | 56 ++++------ src/rpc/virnetdaemon.c | 4 +- src/rpc/virnetmessage.c | 3 +- src/rpc/virnetserver.c | 12 +- src/security/security_dac.c | 3 +- src/security/security_manager.c | 5 +- src/security/security_selinux.c | 15 +-- src/storage/storage_backend_logical.c | 38 +++---- src/storage/storage_backend_rbd.c | 3 +- src/util/viralloc.c | 129 +++++++++++++++++----- src/util/viralloc.h | 48 ++++---- src/util/virfirewall.c | 8 +- src/util/viriscsi.c | 3 +- src/util/virjson.c | 5 +- src/util/virlog.c | 10 +- src/util/virmdev.c | 7 +- src/util/virnetdev.c | 3 +- src/util/virnvme.c | 7 +- src/util/virpci.c | 13 +-- src/util/virprocess.c | 3 +- src/util/virresctrl.c | 6 +- src/util/virscsi.c | 7 +- src/util/virscsivhost.c | 4 +- src/util/virusb.c | 4 +- src/util/virutil.c | 9 +- src/vbox/vbox_common.c | 13 +-- src/vmx/vmx.c | 3 +- src/vz/vz_sdk.c | 32 ++---- tests/nwfilterxml2firewalltest.c | 13 +-- tests/qemumonitortestutils.c | 11 +- tests/testutils.c | 10 +- tests/virfilewrapper.c | 9 +- tests/virpcimock.c | 10 +- tools/virsh-domain.c | 3 +- tools/vsh-table.c | 9 +- 75 files changed, 404 insertions(+), 630 deletions(-) -- 2.31.1

2 25

[libvirt PATCH v2 0/5] add support for Fibre Channel VMID
by Pavel Hrdina 05 Aug '21

05 Aug '21

changes in v2: - refactor of resource parsing and formatting code - use <fibrechannel vmid=''/> element - use stat() directly - report only single system error Pavel Hrdina (5): vircgroup: introduce virCgroupGetInode function conf: refactor virDomainResourceDefParse conf: refactor virDomainResourceDefFormat conf: introduce support for Fibre Channel VMID qemu: implement support for Fibre Channel VMID docs/formatdomain.rst | 21 ++++++++++ docs/schemas/domaincommon.rng | 21 ++++++++-- src/conf/domain_conf.c | 42 +++++++++++-------- src/conf/domain_conf.h | 1 + src/conf/domain_validate.c | 19 +++++++++ src/libvirt_private.syms | 1 + src/qemu/qemu_cgroup.c | 27 ++++++++++++ src/util/vircgroup.c | 31 ++++++++++++++ src/util/vircgroup.h | 2 + .../fibrechannel-vmid.xml | 21 ++++++++++ tests/genericxml2xmltest.c | 2 + 11 files changed, 167 insertions(+), 21 deletions(-) create mode 100644 tests/genericxml2xmlindata/fibrechannel-vmid.xml -- 2.31.1

1 5

[PATCH] tests: Don't leak cpu defs
by Michal Privoznik 05 Aug '21

05 Aug '21

There are cpu definitions that are allocated in qemuTestDriverInit() but are missing corresponding virCPUDefFree() call in qemuTestDriverFree(). It's safe to call the free function because the definitions contain a refcounter and thus even if they were still in use the refcounter would be just decreased. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/testutilsqemu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tests/testutilsqemu.c b/tests/testutilsqemu.c index fab676c070..9a0666724a 100644 --- a/tests/testutilsqemu.c +++ b/tests/testutilsqemu.c @@ -332,6 +332,11 @@ void qemuTestDriverFree(virQEMUDriver *driver) virObjectUnref(driver->caps); virObjectUnref(driver->config); virObjectUnref(driver->securityManager); + + virCPUDefFree(cpuDefault); + virCPUDefFree(cpuHaswell); + virCPUDefFree(cpuPower8); + virCPUDefFree(cpuPower9); } int qemuTestCapsCacheInsert(virFileCache *cache, -- 2.31.1

2 1

[PATCH] rpm: add conditionals around post scripts
by Daniel P. Berrangé 05 Aug '21

05 Aug '21

The hypervisor drivers can be disabled in certain build scenarios, so their corresponding post scripts need to match. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- libvirt.spec.in | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index a162bc333d..011f5dd49e 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1458,6 +1458,7 @@ fi %libvirt_daemon_perform_restart virtstoraged +%if %{with_qemu} %post daemon-driver-qemu %if %{with_modular_daemons} %libvirt_daemon_systemd_post virtqemud @@ -1469,8 +1470,10 @@ fi %posttrans daemon-driver-qemu %libvirt_daemon_perform_restart virtqemud +%endif +%if %{with_lxc} %post daemon-driver-lxc %if %{with_modular_daemons} %libvirt_daemon_systemd_post virtlxcd @@ -1482,8 +1485,10 @@ fi %posttrans daemon-driver-lxc %libvirt_daemon_perform_restart virtlxcd +%endif +%if %{with_vbox} %post daemon-driver-vbox %if %{with_modular_daemons} %libvirt_daemon_systemd_post virtvboxd @@ -1495,8 +1500,10 @@ fi %posttrans daemon-driver-vbox %libvirt_daemon_perform_restart virtvboxd +%endif +%if %{with_libxl} %post daemon-driver-libxl %if %{with_modular_daemons} %libvirt_daemon_systemd_post virtxend @@ -1508,6 +1515,7 @@ fi %posttrans daemon-driver-libxl %libvirt_daemon_perform_restart virtxend +%endif %post daemon-config-network -- 2.31.1

2 1

[PATCH] util: directly query KVM for TSC scaling support
by Daniel P. Berrangé 05 Aug '21

05 Aug '21

We currently query the host MSRs to determine if TSC scaling is supported. This works OK when running privileged and can open the /dev/cpu/0/msr. When unprivileged we fallback to querying MSRs from /dev/kvm. This is incorrect because /dev/kvm only reports accurate info for MSRs that are valid to use from inside a guest. The TSC scaling support MSR is not, thus we always end up reporting lack of TSC scaling when unprivileged. The solution to this is easy, because KVM can directly report whether TSC scaling is available, which matches what QEMU will do at startup. https://gitlab.com/libvirt/libvirt/-/issues/188 Reported-by: Roman Mohr <rmohr(a)redhat.com> Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/util/virhostcpu.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/src/util/virhostcpu.c b/src/util/virhostcpu.c index 7aa92ad11d..f140610b47 100644 --- a/src/util/virhostcpu.c +++ b/src/util/virhostcpu.c @@ -1349,11 +1349,10 @@ virHostCPUGetMSR(unsigned long index, virHostCPUTscInfo * virHostCPUGetTscInfo(void) { - virHostCPUTscInfo *info; + g_autofree virHostCPUTscInfo *info = g_new0(virHostCPUTscInfo, 1); VIR_AUTOCLOSE kvmFd = -1; VIR_AUTOCLOSE vmFd = -1; VIR_AUTOCLOSE vcpuFd = -1; - uint64_t msr = 0; int rc; if ((kvmFd = open(KVM_DEVICE, O_RDONLY)) < 0) { @@ -1378,23 +1377,19 @@ virHostCPUGetTscInfo(void) _("Unable to probe TSC frequency")); return NULL; } - - info = g_new0(virHostCPUTscInfo, 1); - info->frequency = rc * 1000ULL; - if (virHostCPUGetMSR(VMX_PROCBASED_CTLS2_MSR, &msr) == 0) { - /* High 32 bits of the MSR value indicate whether specific control - * can be set to 1. */ - msr >>= 32; - - info->scaling = virTristateBoolFromBool(!!(msr & VMX_USE_TSC_SCALING)); + if ((rc = ioctl(kvmFd, KVM_CHECK_EXTENSION, KVM_CAP_TSC_CONTROL)) < 0) { + virReportSystemError(errno, "%s", + _("Unable to query TSC scaling support")); + return NULL; } + info->scaling = rc ? VIR_TRISTATE_BOOL_YES : VIR_TRISTATE_BOOL_NO; VIR_DEBUG("Detected TSC frequency %llu Hz, scaling %s", info->frequency, virTristateBoolTypeToString(info->scaling)); - return info; + return g_steal_pointer(&info); } #else -- 2.31.1

2 1

[PATCH] Revert "remote: remove probing logic from virtproxyd dispatcher"
by Daniel P. Berrangé 05 Aug '21

05 Aug '21

This reverts commit 05bd8db60b35b7706100d9cbbaeb13992965e0f2. It is true that the remote driver client now contains logic for probing the driver to connect to when using modular daemons. This logic, however, only runs when the remote driver is NOT running inside a daemon since we don't want it activated inside libvirtd. Since the same remote driver build is used in all daemons, we can't rely on it in virtproxyd either. Thus we need to keep the virtproxyd probing logic Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- src/remote/remote_daemon_dispatch.c | 77 +++++++++++++++++++++++++++++ 1 file changed, 77 insertions(+) diff --git a/src/remote/remote_daemon_dispatch.c b/src/remote/remote_daemon_dispatch.c index 65aa20f7d1..36d4d00b79 100644 --- a/src/remote/remote_daemon_dispatch.c +++ b/src/remote/remote_daemon_dispatch.c @@ -1948,6 +1948,73 @@ void *remoteClientNew(virNetServerClient *client, /*----- Functions. -----*/ +#ifdef VIRTPROXYD +/* + * When running in virtproxyd regular auto-probing of drivers + * does not work as we don't have any drivers present (except + * stateless ones inside libvirt.so). All the interesting + * drivers are in separate daemons. Thus when we get a NULL + * URI we need to simulate probing that virConnectOpen would + * previously do. We use the existence of the UNIX domain + * socket as our hook for probing. + * + * This assumes no stale sockets left over from a now dead + * daemon, but that's reasonable since libvirtd unlinks + * sockets it creates on shutdown, or uses systemd activation + * + * We only try to probe for primary hypervisor drivers, + * not the secondary drivers. + */ +static int +remoteDispatchProbeURI(bool readonly, + char **probeduri) +{ + g_autofree char *driver = NULL; + const char *suffix; + *probeduri = NULL; + VIR_DEBUG("Probing for driver daemon sockets"); + + /* + * If running root, either the daemon is running and the socket + * exists, or we're using socket activation so the socket exists + * too. + * + * If running non-root, the daemon may or may not already be + * running, and socket activation probably isn't relevant. + * So if no viable socket exists, we need to check which daemons + * are actually installed. This is not a big deal as only QEMU & + * VBox run as non-root, anyway. + */ + if (geteuid() != 0) { + if (remoteProbeSessionDriverFromSocket(false, &driver) < 0) + return -1; + + if (driver == NULL && + remoteProbeSessionDriverFromBinary(&driver) < 0) + return -1; + + suffix = "session"; + } else { + if (remoteProbeSystemDriverFromSocket(readonly, &driver) < 0) + return -1; + + suffix = "system"; + } + + /* Even if we didn't probe any socket, we won't + * return error. Just let virConnectOpen's normal + * logic run which will likely return an error anyway + */ + if (!driver) + return 0; + + *probeduri = g_strdup_printf("%s:///%s", driver, suffix); + VIR_DEBUG("Probed URI %s for driver %s", *probeduri, driver); + return 0; +} +#endif /* VIRTPROXYD */ + + static int remoteDispatchConnectOpen(virNetServer *server G_GNUC_UNUSED, virNetServerClient *client, @@ -1956,6 +2023,9 @@ remoteDispatchConnectOpen(virNetServer *server G_GNUC_UNUSED, struct remote_connect_open_args *args) { const char *name; +#ifdef VIRTPROXYD + g_autofree char *probeduri = NULL; +#endif unsigned int flags; struct daemonClientPrivate *priv = virNetServerClientGetPrivateData(client); int rv = -1; @@ -1984,6 +2054,13 @@ remoteDispatchConnectOpen(virNetServer *server G_GNUC_UNUSED, priv->readonly = flags & VIR_CONNECT_RO; #ifdef VIRTPROXYD + if (!name || STREQ(name, "")) { + if (remoteDispatchProbeURI(priv->readonly, &probeduri) < 0) + goto cleanup; + + name = probeduri; + } + preserveIdentity = true; #endif /* VIRTPROXYD */ -- 2.31.1

2 1

[libvirt PATCHv2 0/6] kill whole virtiofsd process group (virtio-fs epopee)
by Ján Tomko 05 Aug '21

05 Aug '21

v2 of: https://listman.redhat.com/archives/libvir-list/2021-June/msg00570.html Ján Tomko (6): virProcessKillPainfullyDelay: use 'rc' variable util: Introduce virProcessGroupKill util: introduce virProcessGroupGet util: virPidFileForceCleanupPath: add group argument qemu: virtiofs: kill the whole process group util: fix typo src/libvirt_private.syms | 1 + src/qemu/qemu_process.c | 3 ++- src/qemu/qemu_virtiofs.c | 2 +- src/util/virpidfile.c | 15 ++++++++++-- src/util/virpidfile.h | 2 ++ src/util/virprocess.c | 53 +++++++++++++++++++++++++++++++++++----- src/util/virprocess.h | 5 +++- 7 files changed, 70 insertions(+), 11 deletions(-) -- 2.31.1

3 13

[libvirt PATCH 0/3] add support for Fibre Channel VMID
by Pavel Hrdina 04 Aug '21

04 Aug '21

Pavel Hrdina (3): vircgroup: introduce virCgroupGetInode function conf: introduce support for Fibre Channel VMID qemu: implement support for Fibre Channel VMID docs/formatdomain.rst | 13 ++++++++++ docs/schemas/domaincommon.rng | 11 +++++++++ src/conf/domain_conf.c | 5 ++++ src/conf/domain_conf.h | 2 ++ src/conf/domain_validate.c | 19 +++++++++++++++ src/libvirt_private.syms | 1 + src/qemu/qemu_cgroup.c | 32 +++++++++++++++++++++++++ src/util/vircgroup.c | 37 +++++++++++++++++++++++++++++ src/util/vircgroup.h | 2 ++ tests/genericxml2xmlindata/vmid.xml | 19 +++++++++++++++ tests/genericxml2xmltest.c | 2 ++ 11 files changed, 143 insertions(+) create mode 100644 tests/genericxml2xmlindata/vmid.xml -- 2.31.1

3 16

Re: [PATCH for-6.1? v2 5/7] job: Add job_cancel_requested()
by Kevin Wolf 04 Aug '21

04 Aug '21

[ Peter, the question for you is at the end. ] Am 04.08.2021 um 10:07 hat Max Reitz geschrieben: > On 03.08.21 16:25, Kevin Wolf wrote: > > Am 26.07.2021 um 16:46 hat Max Reitz geschrieben: > > > Most callers of job_is_cancelled() actually want to know whether the job > > > is on its way to immediate termination. For example, we refuse to pause > > > jobs that are cancelled; but this only makes sense for jobs that are > > > really actually cancelled. > > > > > > A mirror job that is cancelled during READY with force=false should > > > absolutely be allowed to pause. This "cancellation" (which is actually > > > a kind of completion) may take an indefinite amount of time, and so > > > should behave like any job during normal operation. For example, with > > > on-target-error=stop, the job should stop on write errors. (In > > > contrast, force-cancelled jobs should not get write errors, as they > > > should just terminate and not do further I/O.) > > > > > > Therefore, redefine job_is_cancelled() to only return true for jobs that > > > are force-cancelled (which as of HEAD^ means any job that interprets the > > > cancellation request as a request for immediate termination), and add > > > job_cancel_request() as the general variant, which returns true for any > > > jobs which have been requested to be cancelled, whether it be > > > immediately or after an arbitrarily long completion phase. > > > > > > Buglink: https://gitlab.com/qemu-project/qemu/-/issues/462 > > > Signed-off-by: Max Reitz <mreitz(a)redhat.com> > > > --- > > > include/qemu/job.h | 8 +++++++- > > > block/mirror.c | 10 ++++------ > > > job.c | 7 ++++++- > > > 3 files changed, 17 insertions(+), 8 deletions(-) > > > > > > diff --git a/include/qemu/job.h b/include/qemu/job.h > > > index 8aa90f7395..032edf3c5f 100644 > > > --- a/include/qemu/job.h > > > +++ b/include/qemu/job.h > > > @@ -436,9 +436,15 @@ const char *job_type_str(const Job *job); > > > /** Returns true if the job should not be visible to the management layer. */ > > > bool job_is_internal(Job *job); > > > -/** Returns whether the job is scheduled for cancellation. */ > > > +/** Returns whether the job is being cancelled. */ > > > bool job_is_cancelled(Job *job); > > > +/** > > > + * Returns whether the job is scheduled for cancellation (at an > > > + * indefinite point). > > > + */ > > > +bool job_cancel_requested(Job *job); > > I don't think non-force blockdev-cancel for mirror should actually be > > considered cancellation, so what is the question that this function > > answers? > > > > "Is this a cancelled job, or a mirror block job that is supposed to > > complete soon, but only if it doesn't switch over the users to the > > target on completion"? > > Well, technically yes, but it was more intended as “Has the user ever > invoked (block-)job-cancel on this job?”. I understand this, but is this much more useful to know than "Has the user ever called HMP 'change'?", if you know what I mean? > > Is this ever a reasonable question to ask, except maybe inside the > > mirror implementation itself? > > I asked myself the same for v3, but found two places in job.c where I > would like to keep it: > > First, there’s an assertion in job_completed_txn_abort(). All jobs > other than @job have been force-cancelled, and so job_is_cancelled() > would be true for them. As for @job itself, the function is mostly > called when the job’s return value is not 0, but a soft-cancelled > mirror does have a return value of 0 and so would not end up in that > function. > But job_cancel() invokes job_completed_txn_abort() if the job has been > deferred to the main loop, which mostly correlates with the job having > been completed (in which case the assertion is skipped), but not 100 % > (there’s a small window between setting deferred_to_main_loop and the > job changing to a completed state). > So I’d prefer to keep the assertion as-is functionally, i.e. to only > check job->cancelled. Well, you don't. It's still job_is_cancelled() after this patch. So the scenario you're concerned about is a job that has just finished successfully (job->ret = 0) and then gets a cancel request? With force=false, I'm pretty sure the code is wrong anyway because calling job_completed_txn_abort() is not the right response. It should return an error because you're trying to complete twice, possibly with conflicting completion modes. Second best is just ignoring the cancel request because we obviously already fulfilled the request of completing the job (the completion mode might be different, though). With force=true, arguably still letting the job fail is correct. However, letting it fail involves more than just letting the transaction fail. We would have to call job_update_rc() as well so that instead of reporting success for the job, ECANCELED is returned and the job transitions to JOB_STATUS_ABORTING (after which job_is_completed() returns true). So, just bugs to be fixed. After this, I think we could even assert(job->ret != 0 || job->status == JOB_STATUS_PENDING) in job_completed_txn_abort(). ret == 0 can only happen when called from job_do_finalize(), when the job is only failing because other jobs in the same transaction have failed in their .prepare callback. > Second, job_complete() refuses to let a job complete that has been > cancelled. This function is basically only invoked by the user > (through qmp_block_job_complete()/qmp_job_complete(), or > job_complete_sync(), which comes from qemu-img), so I believe that it > should correspond to the external interface we have right now; i.e., > if the user has invoked (block-)job-cancel at one point, > job_complete() should generally return an error. True. But it should also return an error if the user has invoked job-complete at some point. The distinction between complete and non-force cancel doesn't make sense there. And cancelling with force=false should fail, too, when either job-cancel or job-complete was called for the job before. > > job_complete() is the only function outside of mirror that seems to use > > it. But even there, it feels wrong to make a difference. Either we > > accept redundant completion requests, or we don't. It doesn't really > > matter how the job reconfigures the graph on completion. (Also, I feel > > this should really have been part of the state machine, but I'm not sure > > if we want to touch it now...) > > Well, yes, I don’t think it makes a difference because I don’t think > anyone will first tell the job via block-job-cancel to complete > without pivoting, and then change their mind and call > block-job-complete after all. (Not least because that’s an error > pre-series.) Right, I'm just arguing that we shouldn't allow the opposite order either. Currently I think we do, and it's buggy, as explained above. > Also, I’m not even sure whether completing after a soft cancel request > works. I don’t think any of our code accounts for such a case, so I’d > rather avoid allowing it if there’s no need to allow it anyway. Yes, definitely avoid it. We should allow only one completion request (be it with job-complete or block-job-cancel) and return an error for all future completion requests for the same job. We could in theory keep allowing redundant completion requests when the completion mode doesn't conflict, but I don't see the point of that. Unless libvirt can actually issue multiple completion requests (again, this includes both (block-)job-complete and non-force block-job-cancel for mirror) for the same block job - Peter, I hope it doesn't? Kevin

3 2

[libvirt PATCH 0/6] rpm: enable modular daemons by default in Fedora 35 / RHEL 9
by Daniel P. Berrangé 04 Aug '21

04 Aug '21

Daniel P. Berrangé (6): rpm: macroize logic for restarting daemons post-transaction rpm: restart virtnwfilter/virnetworkd if configs change rpm: restart modular daemons on upgrade rpm: macroize logic for enabling/disabling daemons post/postun-install rpm: handle enabling/disabling modular daemons post/postun-install rpm: use direct remote connection for Fedora >= 35 / RHEL >= 9 libvirt.spec.in | 225 ++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 190 insertions(+), 35 deletions(-) -- 2.31.1

2 7

[PATCH] qemu_migration: check for interface type 'hostdev'
by Kristina Hanicova 03 Aug '21

03 Aug '21

When we try to migrate vm, we check if it contains only devices that are able to migrate. If a hostdev device is not able to migrate we raise an error with <hostdev/>, but it can actually be <interface/>, so we need to check if hostdev device was created by us from interface and show the right error message. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1942315 Signed-off-by: Kristina Hanicova <khanicov(a)redhat.com> --- src/qemu/qemu_migration.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 4d651aeb1a..34eee9c8b6 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1272,9 +1272,15 @@ qemuMigrationSrcIsAllowedHostdev(const virDomainDef *def) } /* all other PCI hostdevs can't be migrated */ - virReportError(VIR_ERR_OPERATION_UNSUPPORTED, - _("cannot migrate a domain with <hostdev mode='subsystem' type='%s'>"), - virDomainHostdevSubsysTypeToString(hostdev->source.subsys.type)); + if (hostdev->parentnet) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + _("cannot migrate a domain with <interface type='%s'>"), + virDomainNetTypeToString(hostdev->parentnet->type)); + } else { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + _("cannot migrate a domain with <hostdev mode='subsystem' type='%s'>"), + virDomainHostdevSubsysTypeToString(hostdev->source.subsys.type)); + } return false; case VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_LAST: -- 2.31.1

3 3

[PATCH v4 0/4] Implement virDomainGetMessages for test driver
by Luke Yue 03 Aug '21

03 Aug '21

v4: - Move testDomainObjCheckTaint to testParseDomains() - Add CPU tainted and deprecation check - Add a new xml with more tainted configs Luke Yue (4): conf: domain: Introduce and use virDomainObjGetMessages() test_driver: Implement virDomainGetMessages test_driver: Introduce testDomainObjCheckTaint examples: test: Add a new test xml with more tainted configs for testing examples/xml/test/testdomfc5.xml | 51 +++++++++++++++ examples/xml/test/testnode.xml | 1 + examples/xml/test/testnodeinline.xml | 51 +++++++++++++++ src/conf/domain_conf.c | 53 +++++++++++++++ src/conf/domain_conf.h | 5 ++ src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 34 +--------- src/test/test_driver.c | 98 ++++++++++++++++++++++++++++ tests/virshtest.c | 52 +++++++++++++-- 9 files changed, 306 insertions(+), 40 deletions(-) create mode 100644 examples/xml/test/testdomfc5.xml -- 2.32.0

2 7

[PATCH] vmx: Parse vm.genid
by Michal Privoznik 03 Aug '21

03 Aug '21

The VMware metadata file contains genid but we are not parsing and thus reporting it in domain XML. However, it's not as straightforward as one might think. The UUID reported by VMware is not in its usual string form, but split into two signed long longs. That means, we have to do a bit of trickery when parsing. But looking around it's the same magic that libguestfs does: https://github.com/libguestfs/virt-v2v/blob/master/v2v/input_vmx.ml#L421 It's also explained by Rich on qemu-devel: https://lists.nongnu.org/archive/html/qemu-devel/2018-07/msg02019.html Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1598348 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- I've successfully ran vmx2xmltest on an s390x machine which means that there shouldn't be any endiandness problem. src/vmx/vmx.c | 30 +++++++++++++++++++ .../vmx2xml-esx-in-the-wild-10.xml | 1 + 2 files changed, 31 insertions(+) diff --git a/src/vmx/vmx.c b/src/vmx/vmx.c index 1cd5a82227..04eabff18a 100644 --- a/src/vmx/vmx.c +++ b/src/vmx/vmx.c @@ -1337,6 +1337,32 @@ virVMXConfigScanResultsCollector(const char* name, } +static int +virVMXParseGenID(virConf *conf, + virDomainDef *def) +{ + long long vmid[2] = { 0 }; + g_autofree char *uuidstr = NULL; + + if (virVMXGetConfigLong(conf, "vm.genid", &vmid[0], 0, true) < 0 || + virVMXGetConfigLong(conf, "vm.genidX", &vmid[1], 0, true) < 0) + return -1; + + if (vmid[0] == 0 && vmid[1] == 0) + return 0; + + uuidstr = g_strdup_printf("%.16llx%.16llx", vmid[0], vmid[1]); + if (virUUIDParse(uuidstr, def->genid) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Could not parse UUID from string '%s'"), uuidstr); + return -1; + } + def->genidRequested = true; + + return 0; +} + + /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * VMX -> Domain XML @@ -1466,6 +1492,10 @@ virVMXParseConfig(virVMXContext *ctx, } } + /* vmx:vm.genid + vm.genidX -> def:genid */ + if (virVMXParseGenID(conf, def) < 0) + goto cleanup; + /* vmx:annotation -> def:description */ if (virVMXGetConfigString(conf, "annotation", &def->description, true) < 0) { diff --git a/tests/vmx2xmldata/vmx2xml-esx-in-the-wild-10.xml b/tests/vmx2xmldata/vmx2xml-esx-in-the-wild-10.xml index b8c522af1f..47ed637920 100644 --- a/tests/vmx2xmldata/vmx2xml-esx-in-the-wild-10.xml +++ b/tests/vmx2xmldata/vmx2xml-esx-in-the-wild-10.xml @@ -1,6 +1,7 @@ <domain type='vmware'> <name>w2019biosvmware</name> <uuid>421a6177-5aa9-abb7-5924-fc376c18a1b4</uuid> + <genid>13c67c91-9f47-526f-b0d6-e4dd2e4bb4f9</genid> <memory unit='KiB'>4194304</memory> <currentMemory unit='KiB'>4194304</currentMemory> <vcpu placement='static'>2</vcpu> -- 2.31.1

5 10

[libvirt PATCH 0/3] virHashNew refactorings - part VIII
by Tim Wiederhake 03 Aug '21

03 Aug '21

"virHashNew" cannot return NULL, yet we check for NULL in various places. See https://listman.redhat.com/archives/libvir-list/2021-July/msg00074.html. * Patch #1 is a fixed version of https://listman.redhat.com/archives/libvir-list/2021-July/msg00720.html. * Patch #2 is a fixed version of https://listman.redhat.com/archives/libvir-list/2021-July/msg00721.html. * For patch #3, see https://listman.redhat.com/archives/libvir-list/2021-July/msg00691.html. I mixed something up and was under the impression that the libvirt code style forbade `for` loops with anything else than `size_t` typed variables. Tim Wiederhake (3): virLockSpaceNewPostExecRestart: `virHashNew` cannot return NULL WIP virStoragePoolObjListNew: `virHashNew` cannot return NULL virNWFilterParseParamAttributes: Simplify loop src/conf/nwfilter_params.c | 7 ++----- src/conf/virstorageobj.c | 7 ++----- src/util/virlockspace.c | 3 +-- 3 files changed, 5 insertions(+), 12 deletions(-) -- 2.31.1

3 6

lxc container startup error and RFC patch
by Cole Robinson 03 Aug '21

03 Aug '21

Hi all, I'm seeing LXC container startup failures. This is with libvirt git, fedora 34 host with systemd-248.6-1.fc34.x86_64 (I didn't confirm with other versions). Reproducer: sudo virt-install --connect lxc:/// --name test-container --memory 128 --boot init=/bin/sh Starting install... ERROR error from service: GDBus.Error:org.freedesktop.machine1.NoMachineForPID: PID 2145047 does not belong to any known machine libvirt 7.0.0 works but 7.1.0+ does not. The root error seems to predate that, showing up in syslog, but commit 9c1693eff made it fatal: commit 9c1693eff427661616ce1bd2795688f87288a412 Author: Pavel Hrdina <phrdina(a)redhat.com> Date: Fri Feb 5 16:17:35 2021 +0100 vircgroup: use DBus call to systemd for some APIs The error comes from virSystemdGetMachineByPID. The PID that shows up in the above error message does not match the leader PID as reported by machinectl. This change fixes the error but I don't know if it's correct or if it has other implications: diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index 066e013ed4..54ecb1316b 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -866,12 +866,12 @@ static int virLXCControllerSetupCgroupLimits(virLXCController *ctrl) nodeset = virDomainNumatuneGetNodeset(ctrl->def->numa, auto_nodeset, -1); if (!(ctrl->cgroup = virLXCCgroupCreate(ctrl->def, - ctrl->initpid, + getpid(), ctrl->nnicindexes, ctrl->nicindexes))) goto cleanup; - if (virCgroupAddMachineProcess(ctrl->cgroup, getpid()) < 0) + if (virCgroupAddMachineProcess(ctrl->cgroup, ctrl->initpid) < 0) goto cleanup; /* Add all qemu-nbd tasks to the cgroup */ Maybe something else isn't working elsewhere. Clearly we try to add both pids to the systemd machine, but virSystemdGetMachineByPID is not working to match the non-leader pid, which is the one that the LXC driver knows about. Thoughts? Can anyone else reproduce? Thanks, Cole

3 2

[PATCH v4 13/13] docs/deprecated: deprecate passing plugin args through `arg=`
by Mahmoud Mandour 02 Aug '21

02 Aug '21

Signed-off-by: Mahmoud Mandour <ma.mandourr(a)gmail.com> --- docs/system/deprecated.rst | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index e2e0090878..7ae6f1f727 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -126,6 +126,13 @@ other options have been processed. This will either have no effect (if if they were not given. The property is therefore useless and should not be specified. +Plugin argument passing through ``arg=<string>`` (since 6.1) +'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' + +Passing TCG plugins arguments through ``arg=`` is redundant is makes the +command-line less readable, especially when the argument itself consist of a +name and a value, e.g. ``-plugin plugin_name,arg="arg_name=arg_value"``. +Therefore, the usage of ``arg`` is redundant. QEMU Machine Protocol (QMP) commands ------------------------------------ -- 2.25.1

2 2

[PATCH] rpm: fix mingw obsoletes lines
by Daniel P. Berrangé 02 Aug '21

02 Aug '21

The package is mistakenly obsoleting itself, when it should be obsoleting the -static packages we dropped. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- mingw-libvirt.spec.in | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mingw-libvirt.spec.in b/mingw-libvirt.spec.in index bcc2bd93e3..dbcb6d4b8f 100644 --- a/mingw-libvirt.spec.in +++ b/mingw-libvirt.spec.in @@ -69,7 +69,7 @@ MinGW Windows libvirt virtualization library. # Mingw32 %package -n mingw32-libvirt Summary: %{summary} -Obsoletes: mingw32-libvirt < 7.0.0 +Obsoletes: mingw32-libvirt-static < 7.0.0 %description -n mingw32-libvirt MinGW Windows libvirt virtualization library. @@ -77,7 +77,7 @@ MinGW Windows libvirt virtualization library. # Mingw64 %package -n mingw64-libvirt Summary: %{summary} -Obsoletes: mingw64-libvirt < 7.0.0 +Obsoletes: mingw64-libvirt-static < 7.0.0 %description -n mingw64-libvirt MinGW Windows libvirt virtualization library. -- 2.31.1

2 1

[PATCH] testutils: Don't leak @testBitmap and @failedTests
by Michal Privoznik 02 Aug '21

02 Aug '21

In virTestMain() the @failedTests bitmap is allocated and optionally @testBitmap too. But neither of them is freed. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/testutils.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/testutils.c b/tests/testutils.c index 7d87e30a5c..58d3300998 100644 --- a/tests/testutils.c +++ b/tests/testutils.c @@ -856,6 +856,9 @@ int virTestMain(int argc, fprintf(stderr, "Some tests failed. Run them using:\n"); fprintf(stderr, "VIR_TEST_DEBUG=1 VIR_TEST_RANGE=%s %s\n", failed, argv[0]); } + + virBitmapFree(testBitmap); + virBitmapFree(failedTests); virLogReset(); return ret; } -- 2.31.1

3 2

Release of libvirt-7.6.0
by Jiri Denemark 02 Aug '21

02 Aug '21

The 7.6.0 release of both libvirt and libvirt-python is tagged and signed tarballs and source RPMs are available at https://libvirt.org/sources/ https://libvirt.org/sources/python/ Thanks everybody who helped with this release by sending patches, reviewing, testing, or providing any other feedback. Your work is greatly appreciated. * Security * storage: Unlock pool objects on ACL check failures in ``storagePoolLookupByTargetPath`` (CVE-2021-3667) A logic bug in ``storagePoolLookupByTargetPath`` where the storage pool object was left locked after a failure of the ACL check could potentially deprive legitimate users access to a storage pool object by users who don't have access. * New features * qemu: Incremental backup support via ``virDomainBackupBegin`` libvirt-7.6 along with the unreleased qemu-6.1 will fully support the change block tracking features (block-dirty-bitmaps) to be able to do incremental backups and management of the checkpoint states via the appropriate APIs. * qemu: Add support for launch security type s390-pv Specifying s390-pv as launch security type in an s390 domain prepares for running the guest in protected virtualization secure mode, also known as IBM Secure Execution. This simplifies the definition and reduces the risk of an incorrect definition, e.g. by forgetting to specify ``iommu=on`` on all virtio devices. * domstats: Add haltpolling time statistic interface Domstats now provide the data of cpu haltpolling time. This feature relies on statistics available after kernel version 5.8. This will allow the user to get more accurate CPU usage information if needed. * Bug fixes * qemu: Fix migration with ``VIR_MIGRATE_NON_SHARED_INC`` libvirt 7.3.0 introduced a bug where ``VIR_MIGRATE_NON_SHARED_INC`` would not actually migrate the contents of the disk due to broken logic and at the same time could trigger migration of storage when ``VIR_MIGRATE_TUNNELLED`` is requested. This release fixes the bug. * qemu: Don't emit ``VIR_DOMAIN_EVENT_ID_BLOCK_THRESHOLD`` twice when registered with index When registering the threshold event with the index notation (e.g. ``vda[3]``) libvirt would emit the event also for ``vda`` if the image is in the top layer. The intention was to emit two events only when the original registration was done without the index. * qemu: Pass discard requests for disks with ``copy_on_read='on'`` When a disk using the ``copy_on_read='on'`` option is configured also with ``discard='unmap'`` the discard requests will now be passed to the underlying image freeing up the space. Enjoy. Jirka

1 0

[PATCH] gitlab: Add disclaimer to the 'feature' issue template
by Peter Krempa 30 Jul '21

30 Jul '21

Add a disclaimer that filing a feature request issue has no guarantees that anybody will actually implement the feature. Based on the disclaimer in the QEMU project. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- .gitlab/issue_templates/feature.md | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/.gitlab/issue_templates/feature.md b/.gitlab/issue_templates/feature.md index f32faae82b..50f5486701 100644 --- a/.gitlab/issue_templates/feature.md +++ b/.gitlab/issue_templates/feature.md @@ -1,3 +1,19 @@ + + ## Goal  -- 2.31.1

2 1

[PATCH v2 0/2] qemu: Ignore RESET/SHUTDOWN events during startup
by Peter Krempa 30 Jul '21

30 Jul '21

Before we resume the CPUs there was no guest code executed so the events emitted by qemu are purely based on libvirt actions. Ignore them since they are emitted when we are resetting the machine so that it picks up hotplugged disks. Peter Krempa (2): qemu: process: Ignore 'RESET' event during startup qemu: process: Ignore 'SHUTDOWN' event during startup src/qemu/qemu_process.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) -- 2.31.1

2 3

[libvirt PATCH] gitlab: use custom docker:dind image
by Daniel P. Berrangé 30 Jul '21

30 Jul '21

The current docker:dind container has broken default seccomp filter that results in clone3 being blocked, which in turn breaks Fedora 35 rawhide. This custom image has a workaround that causes the seccomp filter to return ENOSYS for clone3 instad of EPERM, thus triggering glibc to fallback to clone correctly. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- .gitlab-ci.yml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 344ecdf3ba..d1609c260d 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -21,7 +21,8 @@ stages: stage: containers needs: [] services: - - docker:dind + - name: registry.gitlab.com/libvirt/libvirt-ci/docker-dind:master + alias: docker rules: - if: "$TEMPORARILY_DISABLED" allow_failure: true -- 2.31.1

2 1

Entering freeze for libvirt-7.6.0
by Pavel Hrdina 29 Jul '21

29 Jul '21

I have just tagged v7.6.0-rc1 in the repository and pushed signed tarballs and source RPMs to https://libvirt.org/sources/ Please give the release candidate some testing and in case you find a serious issue which should have a fix in the upcoming release, feel free to reply to this thread to make sure the issue is more visible. If you have not done so yet, please update NEWS.rst to document any significant change you made since the last release. Thanks, Pavel

1 1

[PATCH] util: fix: duplicated index at nested loop in virNVMeDeviceListCreateReAttachList
by Yi Wang 29 Jul '21

29 Jul '21

From: Jia Zhou <zhou.jia2(a)zte.com.cn> When loop in function virNVMeDeviceListCreateReAttachList() there may be reused index @i, this patch fix this by using a new @j. Signed-off-by: Jia Zhou <zhou.jia2(a)zte.com.cn> Signed-off-by: Yi Wang <wang.yi59(a)zte.com.cn> --- src/util/virnvme.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/util/virnvme.c b/src/util/virnvme.c index 49102e3..b54a195 100644 --- a/src/util/virnvme.c +++ b/src/util/virnvme.c @@ -399,7 +399,7 @@ virNVMeDeviceListCreateReAttachList(virNVMeDeviceListPtr activeList, virNVMeDeviceListPtr toReAttachList) { g_autoptr(virPCIDeviceList) pciDevices = NULL; - size_t i; + size_t i, j; if (!(pciDevices = virPCIDeviceListNew())) return NULL; @@ -412,8 +412,8 @@ virNVMeDeviceListCreateReAttachList(virNVMeDeviceListPtr activeList, /* Check if there is any other NVMe device with the same PCI address as * @d. To simplify this, let's just count how many NVMe devices with * the same PCI address there are on the @activeList. */ - for (i = 0; i < activeList->count; i++) { - virNVMeDevicePtr other = activeList->devs[i]; + for (j = 0; j < activeList->count; j++) { + virNVMeDevicePtr other = activeList->devs[j]; if (!virPCIDeviceAddressEqual(&d->address, &other->address)) continue; -- 1.8.3.1

3 3

[PATCH] NEWS: Add haltpolling time statistic interface
by Yang Fei 29 Jul '21

29 Jul '21

Signed-off-by: Yang Fei <yangfei85(a)huawei.com> --- NEWS.rst | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 00f8788536..a5021b386d 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -36,6 +36,12 @@ v7.6.0 (unreleased) of an incorrect definition, e.g. by forgetting to specify ``iommu=on`` on all virtio devices. + * domstats: Add haltpolling time statistic interface + + Domstats now provide the data of cpu haltpolling time. This feature relies + on statistics available after kernel version 5.8. This will allow the user + to get more accurate CPU usage information if needed. + * **Improvements** * **Bug fixes** -- 2.23.0

2 1

[PATCH] NEWS: Fix false positive of sc_prohibit_doubled_word
by Michal Privoznik 28 Jul '21

28 Jul '21

In the previous commit we've added a sentence into NEWS.rst that supposedly contains doubled word. Well, it doesn't really. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Pushed as trivial. NEWS.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/NEWS.rst b/NEWS.rst index 4f46f92949..00f8788536 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -33,8 +33,8 @@ v7.6.0 (unreleased) Specifying s390-pv as launch security type in an s390 domain prepares for running the guest in protected virtualization secure mode, also known as IBM Secure Execution. This simplifies the definition and reduces the risk - of an incorrect definition, e.g. by forgetting to specify iommu=on on all - virtio devices. + of an incorrect definition, e.g. by forgetting to specify ``iommu=on`` on + all virtio devices. * **Improvements** -- 2.31.1

2 1

[PATCH] rpm: properly disble -Werror
by Daniel P. Berrangé 28 Jul '21

28 Jul '21

Since we use git to manage RPM applied patches, we need to disable both meson's -Werror config knob and libvirt's equivalent. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- libvirt.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index cb48dd0be0..b09336b441 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -192,7 +192,7 @@ %if 0%{?rhel} %define enable_werror -Dwerror=true %else - %define enable_werror -Dwerror=false + %define enable_werror -Dwerror=false -Wgit_werror=disabled %endif %define tls_priority "@LIBVIRT,SYSTEM" -- 2.31.1

3 3