- Devel - libvirt List Archives

[libvirt] [PATCH] conf: add cpu check attribute to ABI check
by Nikolay Shirokovskiy 24 Apr '19

24 Apr '19

Different check values is not ABI compatible. For example if on migration we change 'full' to 'partial' then guest cpu on destination can be different. Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy(a)virtuozzo.com> --- src/conf/cpu_conf.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/src/conf/cpu_conf.c b/src/conf/cpu_conf.c index 33c8b99..bd2beab 100644 --- a/src/conf/cpu_conf.c +++ b/src/conf/cpu_conf.c @@ -894,6 +894,13 @@ virCPUDefIsEqual(virCPUDefPtr src, goto cleanup; } + if (src->check != dst->check) { + MISMATCH(_("Target CPU check %s does not match source %s"), + virCPUCheckTypeToString(dst->check), + virCPUCheckTypeToString(src->check)); + goto cleanup; + } + if (src->arch != dst->arch) { MISMATCH(_("Target CPU arch %s does not match source %s"), virArchToString(dst->arch), -- 1.8.3.1

2 1

[libvirt] JVM crashes during GC
by Sachin Soman 23 Apr '19

23 Apr '19

Hi, Could you tell me if the following is some known issue? While performing the following simple test, I see my JVM crashing (consistently): 1. Open a connection to an ESXi driver/host (passing ConnectAuthDefault instance). 2. Close the connection. 3. Invoke GC When GC is triggered, at some point, some unallocated native memory is being tried to release. That's failing. The error thrown is: java(78745,0x70000241e000) malloc: *** error for object 0x7fd5df561390: pointer being freed was not allocated *** set a breakpoint in malloc_error_break to debug Frames from core dump: frame #0: 0x00007fff5b274b66 libsystem_kernel.dylib`__pthread_kill + 10 frame #1: 0x00007fff5b43f080 libsystem_pthread.dylib`pthread_kill + 333 frame #2: 0x00007fff5b1d01ae libsystem_c.dylib`abort + 127 frame #3: 0x00007fff5b2ce8a6 libsystem_malloc.dylib`free + 521 frame #4: 0x00000001127f43a7 frame #5: 0x00000001127e3ffd frame #6: 0x00000001127e3ffd frame #7: 0x00000001127e3ffd frame #8: 0x00000001127e3ffd frame #9: 0x00000001127e4042 frame #10: 0x00000001127e3ffd frame #11: 0x00000001127e3ffd frame #12: 0x00000001127dc4e7 frame #13: 0x000000010c0e235e libjvm.dylib`JavaCalls::call_helper(JavaValue*, methodHandle*, JavaCallArguments*, Thread*) + 1710 frame #14: 0x000000010c0e2b02 libjvm.dylib`JavaCalls::call_virtual(JavaValue*, KlassHandle, Symbol*, Symbol*, JavaCallArguments*, Thread*) + 356 frame #15: 0x000000010c0e2cae libjvm.dylib`JavaCalls::call_virtual(JavaValue*, Handle, KlassHandle, Symbol*, Symbol*, Thread*) + 74 frame #16: 0x000000010c1208ee libjvm.dylib`thread_entry(JavaThread*, Thread*) + 124 frame #17: 0x000000010c33e84d libjvm.dylib`JavaThread::thread_main_inner() + 155 frame #18: 0x000000010c33ff12 libjvm.dylib`JavaThread::run() + 448 frame #19: 0x000000010c26058a libjvm.dylib`java_start(Thread*) + 246 frame #20: 0x00007fff5b43c661 libsystem_pthread.dylib`_pthread_body + 340 frame #21: 0x00007fff5b43c50d libsystem_pthread.dylib`_pthread_start + 377 frame #22: 0x00007fff5b43bbf9 libsystem_pthread.dylib`thread_start + 13 I have installed Libvirt 5.2.0. Java bindings libvirt-java 0.5.1 JNA 4.0.0 Tested Java environments: Oracle Java 8 and OpenJDK 8 on MAC, OpenJDK 11 on Ubuntu 16 Thanks & Regards Sachin Soman

3 11

[libvirt] [PATCH] networkStartNetworkVirtual: Don't overwrite error in 'err5'
by Michal Privoznik 23 Apr '19

23 Apr '19

If there's an error when setting up QoS on a bridge the control jumps over to 'err5' label. Here, the virNetDevBandwidthClear() is called to clear out any partially set QoS. This function can also report an error which would overwrite the actual error that caused us jumping here. 🤦 Use virErrorPreserveLast() to preserve the original error. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/network/bridge_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index ce4f4890f1..77206b4584 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2493,6 +2493,7 @@ networkStartNetworkVirtual(virNetworkDriverStatePtr driver, return 0; err5: + virErrorPreserveLast(&save_err); if (def->bandwidth) virNetDevBandwidthClear(def->bridge); -- 2.21.0

2 3

[libvirt] [PATCH 0/3] Emit event on lease attach/detach
by Michal Privoznik 23 Apr '19

23 Apr '19

Unfortunately, we can't emit VIR_DOMAIN_EVENT_ID_DEVICE_ADDED or VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED event because that carries device alias within itself and leases don't have one. Michal Prívozník (3): qemuDomainAttachDeviceLive: Rework event emitting Introduce VIR_DOMAIN_EVENT_ID_LEASE_CHANGE qemu: Emit event on lease attach/detach examples/object-events/event-test.c | 34 +++++++++++ include/libvirt/libvirt-domain.h | 33 +++++++++++ src/conf/domain_event.c | 89 +++++++++++++++++++++++++++++ src/conf/domain_event.h | 12 ++++ src/libvirt_private.syms | 2 + src/qemu/qemu_driver.c | 43 +++++++------- src/qemu/qemu_hotplug.c | 7 +++ src/remote/remote_daemon_dispatch.c | 40 +++++++++++++ src/remote/remote_driver.c | 32 +++++++++++ src/remote/remote_protocol.x | 16 +++++- src/remote_protocol-structs | 8 +++ tools/virsh-domain.c | 31 ++++++++++ 12 files changed, 325 insertions(+), 22 deletions(-) -- 2.21.0

3 7

[libvirt] [PATCH v1] network: Convert to virErrorRestore/virErrorPreserveLast
by Syed Humaid 23 Apr '19

23 Apr '19

Replaced usage of virSaveLastError and virSetError/virFreeError with virErrorPreserveLast and virErrorRestore respectively. Signed-off-by: Syed Humaid <syedhumaidbinharoon(a)gmail.com> --- src/network/bridge_driver.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index f824b8b4f3..dc21ac642c 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2495,7 +2495,7 @@ networkStartNetworkVirtual(virNetworkDriverStatePtr driver, err4: if (!save_err) - save_err = virSaveLastError(); + virErrorPreserveLast(&save_err); dnsmasqPid = virNetworkObjGetDnsmasqPid(obj); if (dnsmasqPid > 0) { @@ -2505,18 +2505,18 @@ networkStartNetworkVirtual(virNetworkDriverStatePtr driver, err3: if (!save_err) - save_err = virSaveLastError(); + virErrorPreserveLast(&save_err); ignore_value(virNetDevSetOnline(def->bridge, 0)); err2: if (!save_err) - save_err = virSaveLastError(); + virErrorPreserveLast(&save_err); if (def->forward.type != VIR_NETWORK_FORWARD_OPEN) networkRemoveFirewallRules(def); err1: if (!save_err) - save_err = virSaveLastError(); + virErrorPreserveLast(&save_err); if (macTapIfName) { VIR_FORCE_CLOSE(tapfd); @@ -2528,13 +2528,10 @@ networkStartNetworkVirtual(virNetworkDriverStatePtr driver, err0: if (!save_err) - save_err = virSaveLastError(); + virErrorPreserveLast(&save_err); ignore_value(virNetDevBridgeDelete(def->bridge)); - if (save_err) { - virSetError(save_err); - virFreeError(save_err); - } + virErrorRestore(&save_err); /* coverity[leaked_handle] - 'tapfd' is not leaked */ return -1; } -- 2.20.1

2 1

[libvirt] [PATCH 0/3] Couple of mem leak fixes
by Michal Privoznik 23 Apr '19

23 Apr '19

*** BLURB HERE *** Michal Prívozník (3): qemucaps2xmltest: Don't leak @binary libxlDriverConfigDispose: Free @configBaseDir too vmx: Free @firmware in virVMXParseConfig src/libxl/libxl_conf.c | 1 + src/vmx/vmx.c | 1 + tests/qemucaps2xmltest.c | 3 +-- 3 files changed, 3 insertions(+), 2 deletions(-) -- 2.21.0

3 6

[libvirt] [PATCH 0/3] qemu: monitor: detect more snapshot errors
by Cole Robinson 23 Apr '19

23 Apr '19

qemu in Fedora 30 rejects migration if the VM has nested VMX configured. This means managedsave and snapshots are rejected too. Unfortunately the scraping we need to resort to with the text monitor snapshot commands is not detecting that 'savevm' actually failed, so the snapshot appears to succeed. https://bugzilla.redhat.com/show_bug.cgi?id=1697997 qemu 4.0.0 will add an 'Error: ' prefix to most hmp command errors (qemu commit 66363e9a43f), let's use that to detect error as well Cole Robinson (3): qemu: monitor: cleanup loadvm error handling qemu: monitor cleanup delvm error handling qemu: monitor: check for common 'Error: ' string src/qemu/qemu_monitor_text.c | 39 ++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 19 deletions(-) -- 2.21.0

2 4

[libvirt] New Feature: Intel MKTME Support
by Carvalho, Larkins L 22 Apr '19

22 Apr '19

Hello Team, Greetings. We want to add Intel MKTME support to the Libvirt. Intel MKTME is a capability to encrypt entirety of physical memory of a system similar to AMD SEV. Please let us know what are the expectations from us to initiate the design and development of the feature. Regards, Larkins Carvalho Intel Corporation

5 15

[libvirt] [PATCH] news: add note about optional "Removed features" section
by Daniel P. Berrangé 18 Apr '19

18 Apr '19

Most releases don't need a "Removed features" section so don't include it in the template by default, but leave a reminder in case it is relevant. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- docs/news.xml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index b4926dd906..5346a25132 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -30,6 +30,12 @@ </section> </release> + If relevant for a given release, add a section for removed + features too + + <section title="Removed features"> + </section> + --> <libvirt> -- 2.20.1

2 1

[libvirt] [PATCH v3 0/2] qemu: Honor memory mode='strict'
by Michal Privoznik 18 Apr '19

18 Apr '19

v3 of: diff to v2: - Not calling qemuSetupCpusetMems() in qemuProcessLaunch - Dropping qemuSetupCpusetMems() Michal Prívozník (2): qemu: Set up EMULATOR thread and cpuset.mems before exec()-ing qemu qemu_cgroup: Remove unused qemuSetupCpusetMems src/qemu/qemu_cgroup.c | 35 ----------------------------------- src/qemu/qemu_cgroup.h | 1 - src/qemu/qemu_process.c | 8 ++++---- 3 files changed, 4 insertions(+), 40 deletions(-) -- 2.21.0

2 3