- Devel - libvirt List Archives

[libvirt] [PATCH 0/2] Cleanup memoryBacking docs and move some hugepages checks
by John Ferlan 12 Sep '18

12 Sep '18

While reviewing the -memfd series: https://www.redhat.com/archives/libvir-list/2018-September/msg00381.html I noted that a couple of checks seemed out of place in the the domain_conf processing. They seem more "in place" in the hypvervisor processing. And while doing this I also noted the documentation is clearly not up to snuff, so I modified that too. That's a followup to another recent review comment: https://www.redhat.com/archives/libvir-list/2018-September/msg00176.html although I didn't delve into much detail, just the basics of what the attributes are and the acceptible values. John Ferlan (2): doc: Update the wording around the backingStore conf: Move hypervisor specific nhugepage checks docs/formatdomain.html.in | 13 ++++++++----- src/conf/domain_conf.c | 14 -------------- src/qemu/qemu_domain.c | 27 ++++++++++++++++++++++----- 3 files changed, 30 insertions(+), 24 deletions(-) -- 2.17.1

2 4

[libvirt] [jenkins-ci PATCH v2 0/2] Enable libosinfo tests that require network connectivity
by Fabiano Fidêncio 12 Sep '18

12 Sep '18

Let's enable both media and tree uris tests in order to early catch URL changes that need to be fixed in osinfo-db. Fabiano Fidêncio (2): guests: Enable {media,tree}uris tests for libosinfo projects: Enable {media,tree}uri tests for libosinfo guests/playbooks/build/projects/libosinfo.yml | 4 ++++ projects/libosinfo.yaml | 3 +++ 2 files changed, 7 insertions(+) -- 2.17.1

2 5

[libvirt] [PATCH 0/6] Introduce VIR_AUTOCLOSE macro for closing fd automatically
by Shi Lei 12 Sep '18

12 Sep '18

This series introduce VIR_AUTOCLOSE macro which can force close fd of the file automatically when the fd goes out of scope. It's used to eliminate VIR_FORCE_CLOSE in cleanup sections. The new macro consults VIR_AUTOFREE and VIR_AUTOPTR. Shi Lei (6): util: file: introduce VIR_AUTOCLOSE macro to close fd of the file automatically util: file: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup sections util: netdevbridge: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup sections util: netdev: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup sections phyp: driver: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup sections uml: conf: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup sections src/phyp/phyp_driver.c | 29 ++--- src/uml/uml_conf.c | 13 +- src/util/virfile.c | 21 +-- src/util/virfile.h | 20 ++- src/util/virnetdev.c | 253 +++++++++++++------------------------ src/util/virnetdevbridge.c | 120 ++++++------------ 6 files changed, 163 insertions(+), 293 deletions(-) -- 2.17.1

3 21

[libvirt] [PATCH 0/3] Allow inputvol when creating vol from inputvol to be encrypted
by John Ferlan 12 Sep '18

12 Sep '18

https://bugzilla.redhat.com/show_bug.cgi?id=1613737 Details in the patches (and even more in the bz). John Ferlan (3): storage: Remove secretPath from _virStorageBackendQemuImgInfo storage: Allow for inputvol to have any format for encryption storage: Allow inputvol to be encrypted src/storage/storage_util.c | 79 ++++++++++++++++--- src/storage/storage_util.h | 1 + .../luks-convert-encrypt.argv | 11 +++ .../luks-convert-encrypt2fileqcow2.argv | 7 ++ .../luks-convert-encrypt2fileraw.argv | 7 ++ .../luks-convert-qcow2.argv | 9 +++ tests/storagevolxml2argvtest.c | 19 ++++- tests/storagevolxml2xmlin/vol-encrypt1.xml | 21 +++++ tests/storagevolxml2xmlin/vol-encrypt2.xml | 21 +++++ tests/storagevolxml2xmlin/vol-file-qcow2.xml | 21 +++++ 10 files changed, 184 insertions(+), 12 deletions(-) create mode 100644 tests/storagevolxml2argvdata/luks-convert-encrypt.argv create mode 100644 tests/storagevolxml2argvdata/luks-convert-encrypt2fileqcow2.argv create mode 100644 tests/storagevolxml2argvdata/luks-convert-encrypt2fileraw.argv create mode 100644 tests/storagevolxml2argvdata/luks-convert-qcow2.argv create mode 100644 tests/storagevolxml2xmlin/vol-encrypt1.xml create mode 100644 tests/storagevolxml2xmlin/vol-encrypt2.xml create mode 100644 tests/storagevolxml2xmlin/vol-file-qcow2.xml -- 2.17.1

2 10

[libvirt] Fw: Re: [PATCH 4/6] util: netdev: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE
by Shi Lei 12 Sep '18

12 Sep '18

On 2018-09-12 at 16:25, Shi Lei wrote: On 2018-09-12 at 15:21, Erik Skultety wrote: >On Tue, Sep 11, 2018 at 11:37:45PM +0800, Shi Lei wrote: >> On 2018-09-11 at 20:44, Erik Skultety wrote: >> >On Mon, Sep 10, 2018 at 11:47:53AM +0800, Shi Lei wrote: >> >> By making use of GNU C's cleanup attribute handled by the VIR_AUTOCLOSE macro, >> >> many of the VIR_FORCE_CLOSE calls can be dropped, which in turn leads to >> >> getting rid of many of our cleanup sections. >> >> >> >> Signed-off-by: Shi Lei <shi_lei(a)massclouds.com> >> >> --- >> >> static int >> >> @@ -909,9 +867,9 @@ char *virNetDevGetName(int ifindex) >> >> #if defined(SIOCGIFINDEX) && defined(HAVE_STRUCT_IFREQ) >> >> int virNetDevGetIndex(const char *ifname, int *ifindex) >> >> { >> >> - int ret = -1; >> >> struct ifreq ifreq; >> >> - int fd = socket(VIR_NETDEV_FAMILY, SOCK_DGRAM, 0); >> >> + VIR_AUTOCLOSE(fd); >> >> + socket(VIR_NETDEV_FAMILY, SOCK_DGRAM, 0); >> > >> >^this could not potentially work... >> > >> >Erik >> >> Sorry! I'm too careless. >> >> I think that a new syntax-check rule might make sense. >> This rule checks below: >> >> type foo(a0, a1 ...); >> >> [type] var = foo(a0, a1, ...); /* It's OK */ >> ignore_value(foo(a0, a1, ...)); /* It's OK */ >> >> foo(a0, a1, ...); /* Report this usage */ > >but this would go away with the syntax-check rule you proposed in you response >to the first patch? If not, then would you be more specific to help me >understand the problem more? > >Thanks, >Erik Sorry! I was muddle-headed last night. It is not the syntax-check rule in my response to the first patch. This is another rule and I wanted to introduce it to avoid some faults then. Today I find that it is not easy as I thought. Please ignore it ... :-) Shi Lei > >> >> And it takes effect before compilation and even it's in the inactive >> side of the #if-#else conditons. >> >> It would not check macros since their name are upper case and >> we don't care for the function as condition in if-else statement. >> >> How about it? >> And I can try it if it's a bit helpful ... >> >> Shi Lei >> >> >> >> -- >> libvir-list mailing list >> libvir-list(a)redhat.com >> https://www.redhat.com/mailman/listinfo/libvir-list

1 0

[libvirt] [PATCHv3 0/3] Add virNetlinkNewLink helper for simplifying virNetDev*Create
by Shi Lei 12 Sep '18

12 Sep '18

v2 here: https://www.redhat.com/archives/libvir-list/2018-September/msg00131.html Diff from V2: This series remove all unrelated changes. Those changes would be standalone patches. Shi Lei (3): util: netlink: Introduce virNetlinkNewLink helper util: netlink: Add wrapper macros to make virNetlinkNewLink more readable util: netlink: Using virNetlinkNewLink to simplify virNetDev*Create src/libvirt_private.syms | 1 + src/util/virnetdevbridge.c | 73 +++--------------------- src/util/virnetdevmacvlan.c | 110 +++++------------------------------- src/util/virnetlink.c | 104 ++++++++++++++++++++++++++++++++++ src/util/virnetlink.h | 43 ++++++++++++++ 5 files changed, 172 insertions(+), 159 deletions(-) -- 2.17.1

3 17

[libvirt] [PATCH] numa: fix unsafe access to numa_nodes_ptr
by Wang Yechao 12 Sep '18

12 Sep '18

numa_nodes_ptr is a global variable in libnuma.so. It is been freed after main thread exits. If we have many running vms, restart the libvirtd service continuously at intervals of a few seconds, the main thread may exit before qemuProcessReconnect thread, and a segfault error occurs. Backstrace as follows: 0 0x00007f40e3d2dd72 in numa_bitmask_isbitset () from /lib64/libnuma.so.1 1 0x00007f40e4d14c55 in virNumaNodeIsAvailable (node=node@entry=0) at util/virnuma.c:396 2 0x00007f40e4d16010 in virNumaGetHostMemoryNodeset () at util/virnuma.c:1011 3 0x00007f40b94ced90 in qemuRestoreCgroupState (vm=0x7f407c39df00, vm=0x7f407c39df00) at qemu/qemu_cgroup.c:877 4 qemuConnectCgroup (driver=driver@entry=0x7f407c21fe80, vm=0x7f407c39df00) at qemu/qemu_cgroup.c:969 5 0x00007f40b94eef93 in qemuProcessReconnect (opaque=<optimized out>) at qemu/qemu_process.c:3531 6 0x00007f40e4d34bd2 in virThreadHelper (data=<optimized out>) at util/virthread.c:206 7 0x00007f40e214ee25 in start_thread () from /lib64/libpthread.so.0 8 0x00007f40e1e7c36d in clone () from /lib64/libc.so.6 Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> --- src/util/virnuma.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/src/util/virnuma.c b/src/util/virnuma.c index 67e6c86..502b1d5 100644 --- a/src/util/virnuma.c +++ b/src/util/virnuma.c @@ -381,7 +381,10 @@ virNumaGetMaxCPUs(void) bool virNumaNodeIsAvailable(int node) { - return numa_bitmask_isbitset(numa_nodes_ptr, node); + if (numa_nodes_ptr) + return numa_bitmask_isbitset(numa_nodes_ptr, node); + else + return false; } -- 1.8.3.1

1 0

[libvirt] [PATCH 0/2] Fix issues in qemuARPGetInterfaces and virArpTableGet
by Lin Ma 12 Sep '18

12 Sep '18

Lin Ma (2): qemu: Remove network type limitation for qemuARPGetInterfaces virarptable: Return a virArpTablePtr when the nlmsghdr for loop is over src/qemu/qemu_driver.c | 3 --- src/util/virarptable.c | 2 ++ 2 files changed, 2 insertions(+), 3 deletions(-) -- 2.16.2

2 6

Re: [libvirt] [PATCH v3 4/4] block/rbd: add deprecation documenation for filename keyvalue pairs
by Eric Blake 11 Sep '18

11 Sep '18

[MAINTAINERS says libvir-list should have been cc'd; not sure why that didn't happen] On 9/11/18 4:34 PM, Jeff Cody wrote: > Signed-off-by: Jeff Cody <jcody(a)redhat.com> In the subject: s/documenation/documentation/ > --- > qemu-deprecated.texi | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/qemu-deprecated.texi b/qemu-deprecated.texi > index 1b9c007f12..4df8ac442d 100644 > --- a/qemu-deprecated.texi > +++ b/qemu-deprecated.texi > @@ -190,6 +190,13 @@ used instead. > In order to prevent QEMU from automatically opening an image's backing > chain, use ``"backing": null'' instead. > > +@subsubsection "rbd keyvalue pair encoded filenames": "" (since 3.1.0) > + > +Options for ``rbd'' should be specified according to its runtime options, > +like other block drivers. Legacy parsing of keyvalue pair encoded > +filenames is useful to open images with the old format for backing files; > +These image files should be updated to use the current format. Can we give an example? Cribbing from patch 3, an example might look like changing: json:{"file.driver":"rbd", "file.filename":"rbd:rbd/name"} into: json:{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"} I'll let Peter or John comment on whether libvirt's RBD pool handler is impacted by this deprecation, but it seems reasonable to me. -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org

2 1

[libvirt] [PATCH] qemu_domain: Drop namespace check from top level functions
by Michal Privoznik 11 Sep '18

11 Sep '18

In some cases we are checking if the mount namespace is enabled at two places: one is at the beginning of exported function (e.g. qemuDomainNamespaceSetupDisk()) and the other is at the beginning of qemuDomainNamespaceMknodPaths() which is called from the former function anyway. Then we have some other functions which rely on the later check solely. In order to compensate for possibly needles function call, qemuDomainNamespaceMknodPaths() returns early if @npaths is zero. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_domain.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 5329899b13..15325aa4c1 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -12428,7 +12428,8 @@ qemuDomainNamespaceMknodPaths(virDomainObjPtr vm, int ret = -1; size_t i; - if (!qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT)) + if (!qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT) || + !npaths) return 0; cfg = virQEMUDriverGetConfig(driver); @@ -12521,9 +12522,6 @@ qemuDomainNamespaceSetupDisk(virDomainObjPtr vm, char *dmPath = NULL; int ret = -1; - if (!qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT)) - return 0; - for (next = src; virStorageSourceIsBacking(next); next = next->backingStore) { if (virStorageSourceIsEmpty(next) || !virStorageSourceIsLocalStorage(next)) { @@ -12574,9 +12572,6 @@ qemuDomainNamespaceSetupHostdev(virDomainObjPtr vm, char **paths = NULL; size_t i, npaths = 0; - if (!qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT)) - return 0; - if (qemuDomainGetHostdevPath(NULL, hostdev, false, &npaths, &paths, NULL) < 0) goto cleanup; @@ -12600,9 +12595,6 @@ qemuDomainNamespaceTeardownHostdev(virDomainObjPtr vm, char **paths = NULL; size_t i, npaths = 0; - if (!qemuDomainNamespaceEnabled(vm, QEMU_DOMAIN_NS_MOUNT)) - return 0; - if (qemuDomainGetHostdevPath(vm->def, hostdev, true, &npaths, &paths, NULL) < 0) goto cleanup; -- 2.16.4

2 1

[libvirt] Terminal input “udevadm trigger” command,then produce “nl_recv returned with error:No buffer space available”
by xiajidong＠cmss.chinamobile.com 11 Sep '18

11 Sep '18

Dear libvirt experts， I have add some patch as follow.this two patch add buffer size to avoid to produce “nl_recv returned with error:No buffer space available” 1. libvirt-nodedev-Increase-the-netlink-socket-buffer-size-to-t.patch 2. libvirt-util-increase-libnl-buffer-size.patch but as I set some configuration in /etc/libvirt/libvitd.conf as follows # Logging level: 4 errors, 3 warnings, 2 information, 1 debug log_level =2 # Multiple outputs can be defined, they just need to be separated by spaces. # e.g. to log all warnings and errors to syslog under the libvirtd ident: log_outputs="2:syslog:libvirtd" when I input “udevadm trigger” command [root@localhost ~]# vi /etc/libvirt/libvirtd.conf [root@localhost ~]# systemctl restart libvirtd [root@localhost ~]# udevadm trigger [root@localhost ~]# udevadm trigger [root@localhost ~]# udevadm trigger [root@localhost ~]# then will produce “nl_recv returned with error:No buffer space available” [root@localhost ~]# tail -f /var/log/messages | grep buffer Sep 2 22:44:00 localhost libvirtd: 4004: error : virNetlinkEventCallback:699 : nl_recv returned with error: No buffer space available Sep 2 23:11:33 localhost libvirtd: 4004: error : virNetlinkEventCallback:699 : nl_recv returned with error: No buffer space available Sep 2 23:11:45 localhost libvirtd: 4004: error : virNetlinkEventCallback:699 : nl_recv returned with error: No buffer space available So my question is: When input “udevadm trigger” in terminal, if setting log_level and log_outputs="2:syslog:libvirtd",how to avoid to produce “error virNetlinkEventCallback:699 : nl_recv returned with error: No buffer space available” log information? I am looking forward to your reply. _____ Best Regards, Jidong Xia China Mobile (Suzhou) Software Technology Company.cloud computing product department Tel: (+86) 18896726087 E-Mail：xiajidong(a)cmss.chinamobile.com <mailto:xiajidong@cmss.chinamobile.com>

2 1

[libvirt] [PATCH] test: Remove possible infinite loop in virnetsockettest
by John Ferlan 11 Sep '18

11 Sep '18

Commit 39015a6f3 modified the test to be more reliable/realistic, but without checking the return status of virEventRunDefaultImpl it's possible that the test could run infinitely. Found by Coverity Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- tests/virnetsockettest.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/tests/virnetsockettest.c b/tests/virnetsockettest.c index cccb90d0be..490f720fd0 100644 --- a/tests/virnetsockettest.c +++ b/tests/virnetsockettest.c @@ -243,8 +243,13 @@ testSocketAccept(const void *opaque) &cdata) < 0) goto cleanup; - while (rsock == NULL) - virEventRunDefaultImpl(); + while (rsock == NULL) { + if (virEventRunDefaultImpl() < 0) { + VIR_WARN("Failed to run event loop: %s\n", + virGetLastErrorMessage()); + break; + } + } for (i = 0; i < nlsock; i++) { if (lsock[i] == rsock) { -- 2.17.1

2 2

[libvirt] [PATCH] news: Mention Xen support for PM suspend and wakeup
by Jim Fehlig 11 Sep '18

11 Sep '18

Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- docs/news.xml | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index b8bee35df0..9d16b2930b 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -35,6 +35,15 @@ <libvirt> <release version="v4.8.0" date="unreleased"> <section title="New features"> + <change> + <summary> + Xen: Support PM Suspend and Wakeup + </summary> + <description> + The libxl driver now supports the virDomainPMSuspendForDuration + and virDomainPMWakeup APIs. + </description> + </change> </section> <section title="Improvements"> <change> -- 2.18.0

2 1

[libvirt] [PATCH 0/9] qemuxml2argvtest cleanups
by Ján Tomko 11 Sep '18

11 Sep '18

While looking at how slow this test has gotten, I found a lost test case, some redundancies. Thankfully, not running the HostModel functions twice helped, but there is still room for improvement: we spend a lot of time in the CPU driver comparing the same strings over and over again and with over 300 entries in virQEMUCaps enum, we can reduce some of these string comparisons in the future to cope with more and more QEMU versions being tested. Ján Tomko (9): tests: drop 'drive' from qemuxml2startup tests tests: add macro for dealing with exclusive flags tests: introduce macro for qemu XML->startup XML tests: only run startup XML tests if requested tests: report errors in QEMU XML->startup XML tests tests: drop redundant virQEMUCapsFilterByMachineType tests: do not mangle real qemu caps in xml2argvtest tests: turn skipLegacyCPUs into a flag qemu: remove unnecessary virQEMUCapsFreeHostCPUModel src/qemu/qemu_capabilities.c | 25 +------ src/qemu/qemu_capspriv.h | 5 -- tests/qemuxml2argvtest.c | 76 +++++++++++++--------- .../{disk-drive-shared.xml => disk-shared.xml} | 0 4 files changed, 49 insertions(+), 57 deletions(-) rename tests/qemuxml2startupxmloutdata/{disk-drive-shared.xml => disk-shared.xml} (100%) -- 2.16.4

2 18

[libvirt] [PATCH] qemu: free SEV caps in virQEMUCapsDispose
by Ján Tomko 11 Sep '18

11 Sep '18

Commit 77f51ab5 started parsing an copying the SEV capabilities, but omitted the free call. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- src/qemu/qemu_capabilities.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 08cf822b88..308f5602ca 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -1662,6 +1662,8 @@ void virQEMUCapsDispose(void *obj) VIR_FREE(qemuCaps->gicCapabilities); + virSEVCapabilitiesFree(qemuCaps->sevCapabilities); + virQEMUCapsHostCPUDataClear(&qemuCaps->kvmCPU); virQEMUCapsHostCPUDataClear(&qemuCaps->tcgCPU); } -- 2.16.4

2 1

[libvirt] [PATCH 0/2] Remove unnecessary/duplicated qemuAgentCheckError calls
by John Ferlan 11 Sep '18

11 Sep '18

Followup to review comments I made for recent agent series. John Ferlan (2): qemu: Remove duplicated qemuAgentCheckError qemu: Remove duplicated qemuAgentCheckError src/qemu/qemu_agent.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) -- 2.17.1

2 3

[libvirt] [PATCHv4 0/2] Add virNetlinkNewLink helper for simplifying virNetDev*Create
by Shi Lei 11 Sep '18

11 Sep '18

v3 here: https://www.redhat.com/archives/libvir-list/2018-September/msg00233.html Diff from v3: This series make changes according to the comments from Erik and John. Shi Lei (2): util: netlink: Introduce virNetlinkNewLink helper to simplify virNetDev*Create util: netlink: Add wrapper macros to make virNetlinkNewLink more readable src/libvirt_private.syms | 1 + src/util/virnetdevbridge.c | 73 +++--------------------- src/util/virnetdevmacvlan.c | 110 +++++------------------------------ src/util/virnetlink.c | 111 ++++++++++++++++++++++++++++++++++++ src/util/virnetlink.h | 36 ++++++++++++ 5 files changed, 172 insertions(+), 159 deletions(-) -- 2.17.1

2 4

[libvirt] [PATCH] add nodeset='all' and default for interleave mode
by Peng Hao 11 Sep '18

11 Sep '18

For interleave mode,sometimes we want to allocate mmeory regularly in all nodes on the host. But different hosts has different node number. So we add nodeset='all' for interleave mode and if nodeset=NULL default nodeset is 'all' for interleave mode. Signed-off-by: Peng Hao <peng.hao2(a)zte.com.cn> --- src/conf/numa_conf.c | 73 ++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 57 insertions(+), 16 deletions(-) diff --git a/src/conf/numa_conf.c b/src/conf/numa_conf.c index 97a3ca4..de5fb9c 100644 --- a/src/conf/numa_conf.c +++ b/src/conf/numa_conf.c @@ -29,6 +29,10 @@ #include "virnuma.h" #include "virstring.h" +#if WITH_NUMACTL +#include <numa.h> +#endif + /* * Distance definitions defined Conform ACPI 2.0 SLIT. * See include/linux/topology.h @@ -66,6 +70,7 @@ typedef virDomainNumaNode *virDomainNumaNodePtr; struct _virDomainNuma { struct { bool specified; + bool allnode; virBitmapPtr nodeset; virDomainNumatuneMemMode mode; virDomainNumatunePlacement placement; @@ -259,13 +264,17 @@ virDomainNumatuneParseXML(virDomainNumaPtr numa, tmp = virXMLPropString(node, "nodeset"); if (tmp) { - if (virBitmapParse(tmp, &nodeset, VIR_DOMAIN_CPUMASK_LEN) < 0) - goto cleanup; - - if (virBitmapIsAllClear(nodeset)) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("Invalid value of 'nodeset': %s"), tmp); - goto cleanup; + if (strcmp(tmp, "all") == 0) { + numa->memory.allnode = true; + } else { + if (virBitmapParse(tmp, &nodeset, VIR_DOMAIN_CPUMASK_LEN) < 0) + goto cleanup; + + if (virBitmapIsAllClear(nodeset)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("Invalid value of 'nodeset': %s"), tmp); + goto cleanup; + } } VIR_FREE(tmp); @@ -319,10 +328,14 @@ virDomainNumatuneFormatXML(virBufferPtr buf, virBufferAsprintf(buf, "<memory mode='%s' ", tmp); if (numatune->memory.placement == VIR_DOMAIN_NUMATUNE_PLACEMENT_STATIC) { - if (!(nodeset = virBitmapFormat(numatune->memory.nodeset))) - return -1; - virBufferAsprintf(buf, "nodeset='%s'/>\n", nodeset); - VIR_FREE(nodeset); + if (numatune->memory.allnode == true) { + virBufferAsprintf(buf, "nodeset='all'/>\n"); + } else { + if (!(nodeset = virBitmapFormat(numatune->memory.nodeset))) + return -1; + virBufferAsprintf(buf, "nodeset='%s'/>\n", nodeset); + VIR_FREE(nodeset); + } } else if (numatune->memory.placement) { tmp = virDomainNumatunePlacementTypeToString(numatune->memory.placement); virBufferAsprintf(buf, "placement='%s'/>\n", tmp); @@ -497,6 +510,9 @@ virDomainNumatuneSet(virDomainNumaPtr numa, virBitmapPtr nodeset) { int ret = -1; + int i = 0, maxnode = 0; + virBitmapPtr bitmap = NULL; + /* No need to do anything in this case */ if (mode == -1 && placement == -1 && !nodeset) @@ -538,7 +554,7 @@ virDomainNumatuneSet(virDomainNumaPtr numa, } if (placement == VIR_DOMAIN_NUMATUNE_PLACEMENT_DEFAULT) { - if (numa->memory.nodeset || placement_static) + if (numa->memory.nodeset || placement_static || numa->memory.allnode) placement = VIR_DOMAIN_NUMATUNE_PLACEMENT_STATIC; else placement = VIR_DOMAIN_NUMATUNE_PLACEMENT_AUTO; @@ -546,10 +562,35 @@ virDomainNumatuneSet(virDomainNumaPtr numa, if (placement == VIR_DOMAIN_NUMATUNE_PLACEMENT_STATIC && !numa->memory.nodeset) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("nodeset for NUMA memory tuning must be set " - "if 'placement' is 'static'")); - goto cleanup; + if (mode != VIR_DOMAIN_NUMATUNE_MEM_INTERLEAVE) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("nodeset for NUMA memory tuning must be set " + "if 'placement' is 'static'")); + goto cleanup; + } else { + /* default set all node if nodeset is NULL, when placement + * is static and mode is interleave. + */ + numa->memory.allnode = true; + } + } + + if (placement == VIR_DOMAIN_NUMATUNE_PLACEMENT_STATIC && + mode == VIR_DOMAIN_NUMATUNE_MEM_INTERLEAVE && + numa->memory.allnode) { + if ((bitmap = virBitmapNew(VIR_DOMAIN_CPUMASK_LEN)) == NULL) + goto cleanup; + virBitmapClearAll(bitmap); + maxnode = numa_max_node(); + for (i = 0; i <= maxnode; i++) { + if (virBitmapSetBit(bitmap, i) < 0) { + virBitmapFree(bitmap); + goto cleanup; + } + } + if (numa->memory.nodeset) + virBitmapFree(numa->memory.nodeset); + numa->memory.nodeset = bitmap; } /* setting nodeset when placement auto is invalid */ -- 1.8.3.1

2 1

[libvirt] [PATCH] docs: Fix missing timestamp inside backingStore.
by Julio Faracco 11 Sep '18

11 Sep '18

All backingStore XML definitions have a XML tag with the timestamp. This timestamp is not defined insinde RNG volume storage schema and it is causing some problems to validate and check volume XMLs. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1594266 Signed-off-by: Julio Faracco <jcfaracco(a)gmail.com> --- docs/schemas/storagevol.rng | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/schemas/storagevol.rng b/docs/schemas/storagevol.rng index 96d6b3e6d0..32aaa2784d 100644 --- a/docs/schemas/storagevol.rng +++ b/docs/schemas/storagevol.rng @@ -139,6 +139,7 @@ </element> <ref name='format'/> <ref name='permissions'/> + <ref name='timestamps'/> </interleave> </element> </define> -- 2.17.1

2 1

[libvirt] [PATCH] qemu: fix deadlock if create qemuProcessReconnect thread failed
by Wang Yechao 11 Sep '18

11 Sep '18

qemuProcessReconnectHelper has hold the doms lock, if create qemuProcessReconnect thread failed, it will get the doms lock again to remove the dom from doms list. add obj->inReconnetCtx flag to avoid deadlock. Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> --- src/conf/domain_conf.h | 1 + src/conf/virdomainobjlist.c | 6 ++++-- src/qemu/qemu_process.c | 1 + 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index e30a4b2..5bc5771 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2608,6 +2608,7 @@ struct _virDomainObj { virDomainSnapshotObjPtr current_snapshot; bool hasManagedSave; + bool inReconnectCtx; void *privateData; void (*privateDataFreeFunc)(void *); diff --git a/src/conf/virdomainobjlist.c b/src/conf/virdomainobjlist.c index 805fe94..30300b4 100644 --- a/src/conf/virdomainobjlist.c +++ b/src/conf/virdomainobjlist.c @@ -397,11 +397,13 @@ virDomainObjListRemove(virDomainObjListPtr doms, dom->removing = true; virObjectRef(dom); virObjectUnlock(dom); - virObjectRWLockWrite(doms); + if (!dom->inReconnectCtx) + virObjectRWLockWrite(doms); virObjectLock(dom); virDomainObjListRemoveLocked(doms, dom); virObjectUnref(dom); - virObjectRWUnlock(doms); + if (!dom->inReconnectCtx) + virObjectRWUnlock(doms); } diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index eb9904b..8c30850 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -8029,6 +8029,7 @@ qemuProcessReconnectHelper(virDomainObjPtr obj, */ qemuProcessStop(src->driver, obj, VIR_DOMAIN_SHUTOFF_FAILED, QEMU_ASYNC_JOB_NONE, 0); + obj->inReconnectCtx = true; qemuDomainRemoveInactiveJob(src->driver, obj); virDomainObjEndAPI(&obj); -- 1.8.3.1

1 0

[libvirt] [PATCH V2 0/2] libxl: drop support for Xen 4.4 and 4.5
by Jim Fehlig 11 Sep '18

11 Sep '18

V1: https://www.redhat.com/archives/libvir-list/2018-March/msg01704.html Jim Fehlig (2): libxl: drop support for Xen < 4.6 news: Announce dropping support for Xen 4.4 and 4.5 docs/drvxen.html.in | 2 +- docs/news.xml | 9 +++++++++ m4/virt-driver-libxl.m4 | 22 +--------------------- 3 files changed, 11 insertions(+), 22 deletions(-) -- 2.18.0

4 6

[libvirt] [PATCH] qemu: check for vhostusers bandwidth
by Roland Schulz 11 Sep '18

11 Sep '18

https://bugzilla.redhat.com/show_bug.cgi?id=1524230 Signed-off-by: Roland Schulz <schullzroll(a)gmail.com> --- src/qemu/qemu_command.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ff9589f593..284c2709fc 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -8244,6 +8244,8 @@ qemuBuildVhostuserCommandLine(virQEMUDriverPtr driver, virQEMUCapsPtr qemuCaps, unsigned int bootindex) { + virNetDevBandwidthPtr actualBandwidth = virDomainNetGetActualBandwidth(net); + virDomainNetType actualType = virDomainNetGetActualType(net); virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); char *chardev = NULL; char *netdev = NULL; @@ -8257,6 +8259,19 @@ qemuBuildVhostuserCommandLine(virQEMUDriverPtr driver, goto cleanup; } + /* Set bandwidth or warn if requested and not supported. */ + if (actualBandwidth) { + if (virNetDevSupportBandwidth(actualType)) { + if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false, + !virDomainNetTypeSharesHostView(net)) < 0) + goto cleanup; + } else { + VIR_WARN("setting bandwidth on interfaces of " + "type '%s' is not implemented yet", + virDomainNetTypeToString(actualType)); + } + } + switch ((virDomainChrType)net->data.vhostuser->type) { case VIR_DOMAIN_CHR_TYPE_UNIX: if (!(chardev = qemuBuildChrChardevStr(logManager, secManager, -- 2.17.1

3 3

[libvirt] [PATCH 0/5] libxl: various migration V3 improvements
by Jim Fehlig 11 Sep '18

11 Sep '18

Patch 5 fixes a long standing problem found by some very slow hosts in xen's osstest https://lists.xenproject.org/archives/html/xen-devel/2018-08/msg01945.html While working on the fix, I discovered other problems in libxl's V3 migration protocol. E.g. a modify job on the migrating VM was not handled properly across the phases on either src or dst host. Patches 1-4 fix this and other problems found along the way. Jim Fehlig (5): libxl: migration: defer removing VM until finish phase libxl: fix logic in P2P migration libxl: fix job handling across migration phases on src libxl: fix job handling across migration phases on dst libxl: join with thread receiving migration data src/libxl/libxl_domain.h | 1 + src/libxl/libxl_driver.c | 7 -- src/libxl/libxl_migration.c | 168 ++++++++++++++++++++++++------------ 3 files changed, 114 insertions(+), 62 deletions(-) -- 2.18.0

2 6

[libvirt] [PATCH] nwfilter: fix deadlock when nwfilter reload
by Wang Yechao 11 Sep '18

11 Sep '18

user run "firewalld-cmd --reload" nwfilterStateReload called in main thread step 1. virRWLockWrite(&updateLock) step 2. virNWFilterLoadAllConfigs step 3. virRWLockUnlock(&updateLock); lauch a vm: qemuDomainCreateXML runs in other thread step 1. virRWLockRead(&updateLock); step 2. qemuProcessStart step 3. qemuProcessWaitForMonitor step 4. ... step 5 virRWLockUnlock(&updateLock); if nwfilterStateReload called in the middle of step 1 and step 5 of qemuDomainCreateXML, it can't get the updateLock and then block the event_loop, so event_loop can't handle the qemu-monitor messages, cause deadlock move nwfilterStateReload into thread to fix this problem. Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> Reviewed-by: Wang Yi <wang.yi59(a)zte.com.cn> --- src/nwfilter/nwfilter_driver.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/src/nwfilter/nwfilter_driver.c b/src/nwfilter/nwfilter_driver.c index 1ee5162..8dcc40b 100644 --- a/src/nwfilter/nwfilter_driver.c +++ b/src/nwfilter/nwfilter_driver.c @@ -80,18 +80,27 @@ static void nwfilterDriverUnlock(void) } #if HAVE_FIREWALLD +static void nwfilterReloadThread(void *opaque ATTRIBUTE_UNUSED) +{ + nwfilterStateReload(); +} static DBusHandlerResult nwfilterFirewalldDBusFilter(DBusConnection *connection ATTRIBUTE_UNUSED, DBusMessage *message, void *user_data ATTRIBUTE_UNUSED) { + virThread thread; + if (dbus_message_is_signal(message, DBUS_INTERFACE_DBUS, "NameOwnerChanged") || dbus_message_is_signal(message, "org.fedoraproject.FirewallD1", "Reloaded")) { VIR_DEBUG("Reload in nwfilter_driver because of firewalld."); - nwfilterStateReload(); + + if (virThreadCreate(&thread, false, nwfilterReloadThread, NULL) < 0) { + VIR_ERROR("create nwfilterThread failed."); + } } return DBUS_HANDLER_RESULT_NOT_YET_HANDLED; -- 1.8.3.1

1 0

[libvirt] [PATCH] rpm: increase min required Fedora to 27
by Fabiano Fidêncio 11 Sep '18

11 Sep '18

Fedora 26 has reached its EOL on May 29th 2018. https://lists.fedoraproject.org/archives/list/announce@lists.fedoraproject.… Signed-off-by: Fabiano Fidêncio <fidencio(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 cc2e662e27..b31947b0f4 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -4,7 +4,7 @@ # that's still supported by the vendor. It may work on other distros # or versions, but no effort will be made to ensure that going forward. %define min_rhel 7 -%define min_fedora 26 +%define min_fedora 27 %if (0%{?fedora} && 0%{?fedora} >= %{min_fedora}) || (0%{?rhel} && 0%{?rhel} >= %{min_rhel}) %define supported_platform 1 -- 2.17.1

2 1