June 2019 - Devel - libvirt List Archives

[libvirt] [PATCH 0/2] A couple coverity patches
by John Ferlan 22 Jun '19

22 Jun '19

Coverity has been tripped up a few more times recently... The memory leak patch probably I what I noted from Jiri's recent CPU series. The return value check is older. It wasn't clear though if the intention of the module was to ignore errors so I just chose to error out like other similar callers. If there's a better way - then other patches can be posted. John Ferlan (2): cpu_x86: Fix memory leak - virCPUx86GetHost cpu_x86: Handle error in x86DataToCPU when calling x86DataAdd src/cpu/cpu_x86.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) -- 2.20.1

2 3

[libvirt] [PATCH] qemu: Adjust ATTRIBUTE_NONNULL
by John Ferlan 21 Jun '19

21 Jun '19

Commit 7bf679ae removed the @json argument from the qemuMonitorOpen prototype; however, it did not update the ATTRIBUTE_NONNULL value which causes a build failure for when checking is enabled such as when lv_cv_static_analysis is enabled. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Pushing under build breaker rule. src/qemu/qemu_monitor.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index 35e3d28c22..c41428b048 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -343,7 +343,7 @@ qemuMonitorPtr qemuMonitorOpen(virDomainObjPtr vm, unsigned long long timeout, qemuMonitorCallbacksPtr cb, void *opaque) - ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(6); + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(5); qemuMonitorPtr qemuMonitorOpenFD(virDomainObjPtr vm, int sockfd, qemuMonitorCallbacksPtr cb, -- 2.20.1

1 0

[libvirt] [PATCH] maint: Add missing 'global:' tag in libvirt_public.syms
by Eric Blake 21 Jun '19

21 Jun '19

Fixes: c08fc8d1 Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing as trivial src/libvirt_public.syms | 1 + 1 file changed, 1 insertion(+) diff --git a/src/libvirt_public.syms b/src/libvirt_public.syms index 75f7fdfb0c..18500ec8b2 100644 --- a/src/libvirt_public.syms +++ b/src/libvirt_public.syms @@ -820,6 +820,7 @@ LIBVIRT_5.2.0 { } LIBVIRT_4.10.0; LIBVIRT_5.5.0 { + global: virNetworkListAllPorts; virNetworkPortLookupByUUID; virNetworkPortLookupByUUIDString; -- 2.20.1

1 0

[libvirt] [PATCH v8.5 00/13] Checkpoint APIs (incremental backup saga)
by Eric Blake 21 Jun '19

21 Jun '19

This is a subset of v8 of incremental backup, focusing more on just the checkpoint manipulation code (as this had fewer review comments that still needed addressing). I think there is enough here that even if this lands in 5.5 while the backup APIs slide into 5.6 that we'd still have a reasonably usable set of APIs. I probably still need to double-check that creating checkpoints while an external snapshot exists gives a sane error, and that creating an external snapshot while checkoints exist give a sane error (in both cases, with plans to relax that error later once we finish our design); that can be done as a patch on top of these or if I have to respin this part of the series once more before 5.5. I do plan on tagging a v9 label of the full backup series, but that will be in a separate thread built on top of this one (hence naming this one v8.5). Since v8: - Completely drop the virDomainHasCurrentCheckpoint API (you get the same effect by calling virDomainCheckpointCurrent) - Defer the virDomainCheckpointHasMetadata API for a later day (if we ever want it); for now, all checkpoints require libvirt metadta - Expand the code to implement Checkpoint APIs for the test driver - rebase to master - Defer the Backup API work to a later series - Split out the handling of VIR_DOMAIN_CHECKPOINT_XML_SIZE into its own patch (that one is probably the most likely to need a respin, or even deferring it post-5.5 as adding flag support is different than adding new API) 001/13:[----] [--] 'backup: Document new XML for checkpoints' 002/13:[0121] [FC] 'backup: Introduce virDomainCheckpoint APIs' 003/13:[0004] [FC] 'backup: Document nuances between different state capture APIs' 004/13:[0127] [FC] 'backup: Parse and output checkpoint XML' 005/13:[0111] [FC] 'backup: Allow for lists of checkpoint objects' 006/13:[----] [--] 'backup: Add new domain:checkpoint access control' 007/13:[down] 'backup: Implement checkpoint APIs for remote driver' 008/13:[0065] [FC] 'backup: Implement virsh support for checkpoints' 009/13:[down] 'snapshot: Refactor test to utilize virDomainMoment more' 010/13:[down] 'backup: test: Implement metadata tracking for checkpoint APIs' 011/13:[0184] [FC] 'backup: qemu: Implement metadata tracking for checkpoint APIs' 012/13:[0016] [FC] 'backup: Wire up qemu checkpoint commands over QMP' 013/13:[down] 'backup: qemu: Implement VIR_DOMAIN_CHECKPOINT_XML_SIZE flag' Eric Blake (13): backup: Document new XML for checkpoints backup: Introduce virDomainCheckpoint APIs backup: Document nuances between different state capture APIs backup: Parse and output checkpoint XML backup: Allow for lists of checkpoint objects backup: Add new domain:checkpoint access control backup: Implement checkpoint APIs for remote driver backup: Implement virsh support for checkpoints snapshot: Refactor test to utilize virDomainMoment more backup: test: Implement metadata tracking for checkpoint APIs backup: qemu: Implement metadata tracking for checkpoint APIs backup: Wire up qemu checkpoint commands over QMP backup: qemu: Implement VIR_DOMAIN_CHECKPOINT_XML_SIZE flag include/libvirt/libvirt-domain-checkpoint.h | 149 ++ include/libvirt/libvirt-domain.h | 6 + include/libvirt/libvirt.h | 5 +- src/access/viraccessperm.h | 6 + src/conf/checkpoint_conf.h | 101 ++ src/conf/domain_conf.h | 2 + src/conf/virconftypes.h | 9 + src/conf/virdomaincheckpointobjlist.h | 66 + src/conf/virdomainmomentobjlist.h | 5 +- src/conf/virdomainobjlist.h | 7 +- src/driver-hypervisor.h | 48 + src/qemu/qemu_block.h | 3 + src/qemu/qemu_conf.h | 2 + src/qemu/qemu_domain.h | 15 + src/qemu/qemu_monitor.h | 4 + src/qemu/qemu_monitor_json.h | 3 + tools/virsh-checkpoint.h | 26 + tools/virsh-completer.h | 4 + tools/virsh-util.h | 3 + tools/virsh.h | 1 + docs/Makefile.am | 3 + docs/apibuild.py | 2 + docs/docs.html.in | 9 +- docs/domainstatecapture.html.in | 315 ++++ docs/format.html.in | 1 + docs/formatcheckpoint.html.in | 204 +++ docs/formatsnapshot.html.in | 2 + docs/index.html.in | 3 +- docs/schemas/domaincheckpoint.rng | 87 ++ libvirt.spec.in | 2 + mingw-libvirt.spec.in | 4 + po/POTFILES | 3 + src/Makefile.am | 2 + src/access/viraccessperm.c | 3 +- src/conf/Makefile.inc.am | 4 + src/conf/checkpoint_conf.c | 662 ++++++++ src/conf/domain_conf.c | 6 + src/conf/virdomaincheckpointobjlist.c | 219 +++ src/conf/virdomainmomentobjlist.c | 2 +- src/conf/virdomainobjlist.c | 11 + src/libvirt-domain-checkpoint.c | 670 ++++++++ src/libvirt-domain.c | 18 +- src/libvirt_private.syms | 26 + src/libvirt_public.syms | 25 +- src/qemu/qemu_block.c | 12 + src/qemu/qemu_conf.c | 5 + src/qemu/qemu_domain.c | 183 +++ src/qemu/qemu_driver.c | 868 +++++++++++ src/qemu/qemu_monitor.c | 11 + src/qemu/qemu_monitor_json.c | 76 + src/remote/remote_daemon_dispatch.c | 20 + src/remote/remote_driver.c | 26 +- src/remote/remote_protocol.x | 155 +- src/remote_protocol-structs | 85 ++ src/rpc/gendispatch.pl | 32 +- src/test/test_driver.c | 461 +++++- tests/Makefile.am | 11 +- tests/domaincheckpointxml2xmlin/empty.xml | 1 + tests/domaincheckpointxml2xmlin/sample.xml | 7 + tests/domaincheckpointxml2xmlin/size.xml | 4 + tests/domaincheckpointxml2xmlout/empty.xml | 7 + .../internal-active-invalid.xml | 53 + .../internal-inactive-invalid.xml | 53 + tests/domaincheckpointxml2xmlout/redefine.xml | 63 + tests/domaincheckpointxml2xmlout/sample.xml | 12 + tests/domaincheckpointxml2xmlout/size.xml | 11 + tests/domaincheckpointxml2xmltest.c | 223 +++ tests/virschematest.c | 2 + tools/Makefile.am | 1 + tools/virsh-checkpoint.c | 1342 +++++++++++++++++ tools/virsh-completer.c | 51 + tools/virsh-domain-monitor.c | 23 + tools/virsh-domain.c | 9 + tools/virsh-util.c | 11 + tools/virsh.c | 2 + tools/virsh.pod | 238 ++- 76 files changed, 6745 insertions(+), 61 deletions(-) create mode 100644 include/libvirt/libvirt-domain-checkpoint.h create mode 100644 src/conf/checkpoint_conf.h create mode 100644 src/conf/virdomaincheckpointobjlist.h create mode 100644 tools/virsh-checkpoint.h create mode 100644 docs/domainstatecapture.html.in create mode 100644 docs/formatcheckpoint.html.in create mode 100644 docs/schemas/domaincheckpoint.rng create mode 100644 src/conf/checkpoint_conf.c create mode 100644 src/conf/virdomaincheckpointobjlist.c create mode 100644 src/libvirt-domain-checkpoint.c create mode 100644 tests/domaincheckpointxml2xmlin/empty.xml create mode 100644 tests/domaincheckpointxml2xmlin/sample.xml create mode 100644 tests/domaincheckpointxml2xmlin/size.xml create mode 100644 tests/domaincheckpointxml2xmlout/empty.xml create mode 100644 tests/domaincheckpointxml2xmlout/internal-active-invalid.xml create mode 100644 tests/domaincheckpointxml2xmlout/internal-inactive-invalid.xml create mode 100644 tests/domaincheckpointxml2xmlout/redefine.xml create mode 100644 tests/domaincheckpointxml2xmlout/sample.xml create mode 100644 tests/domaincheckpointxml2xmlout/size.xml create mode 100644 tests/domaincheckpointxml2xmltest.c create mode 100644 tools/virsh-checkpoint.c -- 2.20.1

2 24

[libvirt] [PATCH] Don't parse/format vram attribute for cirrus video
by Jonathon Jongsma 21 Jun '19

21 Jun '19

Since the cirrus vga memory size isn't configurable, we can ignore any 'vram' attribute when parsing a domain definition. However, when no value is specified, it ends up getting set to a default value of 16MB. This 16MB value is not used anywhere (for example, it is not passed as an argument to qemu), but is displayed in the XML definition. So by changing this default value to 0, it will also be omitted from the XML definition of the domain. Fixes: rhbz#1447831 Signed-off-by: Jonathon Jongsma <jjongsma(a)redhat.com> --- This is an attempt to apply the fix suggested by Gerd at https://bugzilla.redhat.com/show_bug.cgi?id=1447831#c2. I'm not totally confident that this is the right approach, since I'm relatively new to the code. Another approach might be to simply close the bug as NOTABUG since it doesn't seem that having this unused attribute in the domain definition has any significant drawbacks. src/conf/domain_conf.c | 2 +- src/qemu/qemu_domain.c | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index c69d382d70..d06e094b11 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -15154,7 +15154,6 @@ virDomainVideoDefaultRAM(const virDomainDef *def, { switch (type) { case VIR_DOMAIN_VIDEO_TYPE_VGA: - case VIR_DOMAIN_VIDEO_TYPE_CIRRUS: case VIR_DOMAIN_VIDEO_TYPE_VMVGA: if (def->virtType == VIR_DOMAIN_VIRT_VBOX) return 8 * 1024; @@ -15172,6 +15171,7 @@ virDomainVideoDefaultRAM(const virDomainDef *def, /* QEMU use 64M as the minimal video memory for qxl device */ return 64 * 1024; + case VIR_DOMAIN_VIDEO_TYPE_CIRRUS: case VIR_DOMAIN_VIDEO_TYPE_DEFAULT: case VIR_DOMAIN_VIDEO_TYPE_VBOX: case VIR_DOMAIN_VIDEO_TYPE_PARALLELS: diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index e5c6ef3fda..7f0baf57ac 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -6835,6 +6835,9 @@ qemuDomainDeviceVideoDefPostParse(virDomainVideoDefPtr video, if (video->type == VIR_DOMAIN_VIDEO_TYPE_QXL && !video->vgamem) { video->vgamem = QEMU_QXL_VGAMEM_DEFAULT; + } else if (video->type == VIR_DOMAIN_VIDEO_TYPE_CIRRUS) { + /* cirrus vram is not configurable. Ignore it */ + video->vram = 0; } return 0; -- 2.20.1

2 4

[libvirt] [PATCH] virStorageSourceChainHasManagedPR: Check propely
by Michal Privoznik 21 Jun '19

21 Jun '19

In the virStorageSourceChainHasManagedPR() function we iterate over whole backing chain trying to determine if one of the layers has managed PR configured. But due to a typo we in fact check the top layer only. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Pushed under trivial rule. src/util/virstoragefile.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index 2b1c0a5b8e..f7495ab6da 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -2083,7 +2083,7 @@ virStorageSourceChainHasManagedPR(virStorageSourcePtr src) virStorageSourcePtr n; for (n = src; virStorageSourceIsBacking(n); n = n->backingStore) { - if (virStoragePRDefIsManaged(src->pr)) + if (virStoragePRDefIsManaged(n->pr)) return true; } -- 2.21.0

2 1

[libvirt] [PATCH v2 0/7] Finish support for IA32_ARCH_CAPABILITIES MSR features
by Jiri Denemark 21 Jun '19

21 Jun '19

As promised in v1 this series contains a few additional patches which limit usage of MSR features to QEMU that supports "unavailable-features" CPU property. This series intentionally enables MSR features for all QEMU versions before restricting the usage to make this restriction more visible in the test results. Version 2: - all but 2 patches from version 1 were pushed - 5 new patches Jiri Denemark (7): conf: Introduce virCPUDefCheckFeatures cpu_x86: Turn virCPUx86DataIteratorInit into a function cpu_x86: Introduce virCPUx86FeatureIsMSR cpu_x86: Read CPU features from IA32_ARCH_CAPABILITIES MSR cpu_map: Introduce IA32_ARCH_CAPABILITIES MSR features qemu: Forbid MSR features with old QEMU qemu: Drop MSR features from host-model with old QEMU src/conf/cpu_conf.c | 33 +++++++ src/conf/cpu_conf.h | 6 ++ src/cpu/cpu_x86.c | 96 +++++++++++++++++-- src/cpu/cpu_x86.h | 3 + src/cpu_map/x86_features.xml | 20 ++++ src/libvirt_private.syms | 2 + src/qemu/qemu_capabilities.c | 16 ++++ src/qemu/qemu_process.c | 29 +++++- .../x86_64-cpuid-Core-i7-7600U-enabled.xml | 1 + .../x86_64-cpuid-Core-i7-7600U-json.xml | 1 + ...86_64-cpuid-Xeon-Platinum-8268-enabled.xml | 1 + .../x86_64-cpuid-Xeon-Platinum-8268-guest.xml | 4 + .../x86_64-cpuid-Xeon-Platinum-8268-host.xml | 4 + .../x86_64-cpuid-Xeon-Platinum-8268-json.xml | 3 + .../qemu_4.1.0.x86_64.xml | 1 + 15 files changed, 207 insertions(+), 13 deletions(-) -- 2.22.0

3 17

[libvirt] [PATCH] rpc: add $(SASL_CFLAGS) to libvirt_net_rpc_server_la_CFLAGS
by Daniel P. Berrangé 21 Jun '19

21 Jun '19

The files for libvirt-net-rpc-server.la refernce the sasl/sasl.h system header but never used the $(SASL_CFLAGS) variable. This was never noticed previously because the $(AVAHI_CLFAGS) were set and these typically pulled in the same include directory. When mDNS/Avahi support was removed this exposed the bug which caused FreeBSD builds to break as /usr/local/include was no longer searched for headers. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed as a build fix src/rpc/Makefile.inc.am | 1 + 1 file changed, 1 insertion(+) diff --git a/src/rpc/Makefile.inc.am b/src/rpc/Makefile.inc.am index e5b830605a..b8ca53c69a 100644 --- a/src/rpc/Makefile.inc.am +++ b/src/rpc/Makefile.inc.am @@ -111,6 +111,7 @@ libvirt_net_rpc_server_la_SOURCES = \ libvirt_net_rpc_server_la_CFLAGS = \ $(DBUS_CFLAGS) \ $(XDR_CFLAGS) \ + $(SASL_CFLAGS) \ $(AM_CFLAGS) \ $(NULL) libvirt_net_rpc_server_la_LDFLAGS = \ -- 2.21.0

1 0

[libvirt] [PATCH 0/5] cgroups v2 fixes and improvements
by Pavel Hrdina 21 Jun '19

21 Jun '19

Pavel Hrdina (5): util: vircgroup: pass parent cgroup into virCgroupDetectControllersCB util: vircgroup: improve controller detection util: vircgroupv2: use any controller to create thread directory util: vircgroupv2: enable CPU controller only if it's available util: vircgroupv2: don't error out if enabling controller fails src/util/vircgroup.c | 32 ++++++++++++++++---------------- src/util/vircgroupbackend.h | 3 ++- src/util/vircgroupv1.c | 3 ++- src/util/vircgroupv2.c | 36 ++++++++++++++++++++++++------------ 4 files changed, 44 insertions(+), 30 deletions(-) -- 2.21.0

2 11

[libvirt] [PATCH v2] test_driver: properly handle DHCP ranges and IPv6 networks in testDomainInterfaceAddresses
by Ilias Stamatis 21 Jun '19

21 Jun '19

testDomainInterfaceAddresses always returns the same hard-coded addresses. Change the behavior such as if there is a DHCP range defined, addresses are returned from that pool. The specific address returned depends on both the domain id and the specific guest interface in an attempt to return unique addresses *most of the time*. Additionally, properly handle IPv6 networks which were previously ignored completely. Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- src/test/test_driver.c | 44 +++++++++++++++++++++++++++++++++++++----- 1 file changed, 39 insertions(+), 5 deletions(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 2a0ffbc6c5..21bd95941e 100755 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3414,6 +3414,10 @@ static int testDomainBlockStats(virDomainPtr domain, return ret; } + +static virNetworkObjPtr testNetworkObjFindByName(testDriverPtr privconn, const char *name); + + static int testDomainInterfaceAddresses(virDomainPtr dom, virDomainInterfacePtr **ifaces, @@ -3422,11 +3426,15 @@ testDomainInterfaceAddresses(virDomainPtr dom, { size_t i; size_t ifaces_count = 0; + size_t addr_offset; int ret = -1; char macaddr[VIR_MAC_STRING_BUFLEN]; virDomainObjPtr vm = NULL; virDomainInterfacePtr iface = NULL; virDomainInterfacePtr *ifaces_ret = NULL; + virSocketAddr addr; + virNetworkObjPtr net = NULL; + virNetworkDefPtr net_def = NULL; virCheckFlags(0, -1); @@ -3447,6 +3455,12 @@ testDomainInterfaceAddresses(virDomainPtr dom, goto cleanup; for (i = 0; i < vm->def->nnets; i++) { + if (!(net = testNetworkObjFindByName(dom->conn->privateData, + vm->def->nets[i]->data.network.name))) + goto cleanup; + + net_def = virNetworkObjGetDef(net); + if (VIR_ALLOC(iface) < 0) goto cleanup; @@ -3460,14 +3474,33 @@ testDomainInterfaceAddresses(virDomainPtr dom, if (VIR_ALLOC(iface->addrs) < 0) goto cleanup; - iface->addrs[0].type = VIR_IP_ADDR_TYPE_IPV4; - iface->addrs[0].prefix = 24; - if (virAsprintf(&iface->addrs[0].addr, "192.168.0.%zu", 1 + i) < 0) - goto cleanup; - iface->naddrs = 1; + iface->addrs[0].prefix = virSocketAddrGetIPPrefix(&net_def->ips->address, + &net_def->ips->netmask, + net_def->ips->prefix); + + if (net_def->ips->nranges > 0) + addr = net_def->ips->ranges[0].start; + else + addr = net_def->ips->address; + + /* try using different addresses per different inf and domain */ + addr_offset = 20 * (vm->def->id - 1) + i + 1; + + if (net_def->ips->family && STREQ(net_def->ips->family, "ipv6")) { + iface->addrs[0].type = VIR_IP_ADDR_TYPE_IPV6; + addr.data.inet6.sin6_addr.s6_addr[15] += addr_offset; + } else { + iface->addrs[0].type = VIR_IP_ADDR_TYPE_IPV4; + addr.data.inet4.sin_addr.s_addr = \ + htonl(ntohl(addr.data.inet4.sin_addr.s_addr) + addr_offset); + } + + if (!(iface->addrs[0].addr = virSocketAddrFormat(&addr))) + goto cleanup; VIR_APPEND_ELEMENT_INPLACE(ifaces_ret, ifaces_count, iface); + virNetworkObjEndAPI(&net); } VIR_STEAL_PTR(*ifaces, ifaces_ret); @@ -3475,6 +3508,7 @@ testDomainInterfaceAddresses(virDomainPtr dom, cleanup: virDomainObjEndAPI(&vm); + virNetworkObjEndAPI(&net); if (ifaces_ret) { for (i = 0; i < ifaces_count; i++) -- 2.22.0

2 5