- Devel - libvirt List Archives

[libvirt] [PATCH 0/3] conf: Fix some issues
by Andrea Bolognani 17 Nov '18

17 Nov '18

Fallout from merging the zPCI patches. Andrea Bolognani (3): conf: Fix error flow in virDomainPCIAddressEnsureAddr() conf: Add several cleanup paths conf: Perform error checking in virDomainDeviceInfoFormat() src/conf/domain_addr.c | 16 ++- src/conf/domain_conf.c | 234 +++++++++++++++++++++++++++++------------ 2 files changed, 177 insertions(+), 73 deletions(-) -- 2.19.1

3 7

[libvirt] [RFC 0/3] update NVDIMM support
by Luyao Zhong 17 Nov '18

17 Nov '18

Hi libvirt experts, This is the RFC for updating NVDIMM support in libvirt. QEMU has supported four more properties which libvirt has not introduced yet, including 'align', 'pmem', 'nvdimm-persistences' and 'unarmed'. The 'align' property allows users to specify the proper alignment. The previous alignment can only be 4K because QEMU use pagesize as alignment. But some backends may require alignments different from the pagesize. The 'pmem' property allows users to specify whether the backend storage of memory-backend-file is a real persistent memory. Then QEMU will know if it needs to guarrantee the write persistence to the vNVDIMM backend. The 'nvdimm-persistence' property allows users to set platform-supported features about NVDIMM data persistence of a guest. The 'unarmed' property allows users to mark vNVDIMM read-only. Only the device DAX on the real NVDIMM can guarantee the guest write persistence, so it's suggested to set 'unarmed' option to 'on' and then vNVDIMM device will be marked as read-only. Libvirt introduces 'alignsize', 'pmem', 'persistence' and 'unarmed' config elements into xml corresponding to 'align', 'pmem', 'nvdimm-persistence' and 'unarmed' properties in QEMU, and update xml parsing, formating and qemu command-line generating process for NVDIMM. Thanks, Zhong, Luyao Luyao Zhong (3): xml: introduce more config elements for NVDIMM memory xml: update xml parsing and formating about NVDIMM memory qemu: update qemu command-line generating for NVDIMM memory docs/formatdomain.html.in | 98 +++++++++++++++--- docs/schemas/domaincommon.rng | 31 +++++- src/conf/domain_conf.c | 115 +++++++++++++++++++-- src/conf/domain_conf.h | 14 +++ src/libvirt_private.syms | 2 + src/qemu/qemu_command.c | 25 +++++ .../memory-hotplug-nvdimm-align.args | 31 ++++++ .../memory-hotplug-nvdimm-align.xml | 58 +++++++++++ .../memory-hotplug-nvdimm-persistence.args | 31 ++++++ .../memory-hotplug-nvdimm-persistence.xml | 58 +++++++++++ .../memory-hotplug-nvdimm-pmem.args | 31 ++++++ .../memory-hotplug-nvdimm-pmem.xml | 58 +++++++++++ .../memory-hotplug-nvdimm-unarmed.args | 31 ++++++ .../memory-hotplug-nvdimm-unarmed.xml | 58 +++++++++++ tests/qemuxml2argvtest.c | 12 +++ .../memory-hotplug-nvdimm-align.xml | 1 + .../memory-hotplug-nvdimm-persistence.xml | 1 + .../memory-hotplug-nvdimm-pmem.xml | 1 + .../memory-hotplug-nvdimm-unarmed.xml | 1 + tests/qemuxml2xmltest.c | 4 + 20 files changed, 636 insertions(+), 25 deletions(-) create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-align.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-align.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-persistence.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-persistence.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-pmem.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-pmem.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-unarmed.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-unarmed.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-align.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-persistence.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-pmem.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-unarmed.xml -- 2.7.4

4 15

[libvirt] [PATCH 0/2] rpc: fixing compilation error due to deprecated ssh_is_server_known()
by Julio Faracco 16 Nov '18

16 Nov '18

After 0.8.0 release, libssh deprecated ssh_is_server_known() method to introduce ssh_session_is_known_server(). This commit check if ssh_session_is_known_server() is available to use. If it is not, it creates an alias to ssh_is_server_known() during the configuration. The commits are similar to c9da6cbe and 463fa9c7. Julio Faracco (2): rpc: replacing ssh_is_server_known() by ssh_session_is_known_server(). m4: checking if ssh_session_is_known_server() exists. m4/virt-libssh.m4 | 4 ++++ src/rpc/virnetlibsshsession.c | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) -- 2.19.1

1 3

[libvirt] [PATCH v4 00/15] Implement alternative metadata locking
by Michal Privoznik 16 Nov '18

16 Nov '18

v4 of: https://www.redhat.com/archives/libvir-list/2018-October/msg00861.html diff to v3: - Introduced a config knob to enable/disable metadata locking (except not really). We want to have a knob that enables/disables remembering of original owner. This knob in turn enables metadata locking. The reason is that metadata locking on its own doesn't make any sense. Anyway, the qemu.conf change is not done (it'll be done in upcoming patch set that implements original owner remembering), so if you want to see these patches in action you'll need to apply the following patch: diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 32da9a7351..0080b0d021 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -347,6 +347,8 @@ virQEMUDriverConfigPtr virQEMUDriverConfigNew(bool privileged) if (!(cfg->namespaces = virBitmapNew(QEMU_DOMAIN_NS_LAST))) goto error; + cfg->rememberOwner = true; + if (privileged && qemuDomainNamespaceAvailable(QEMU_DOMAIN_NS_MOUNT) && virBitmapSetBit(cfg->namespaces, QEMU_DOMAIN_NS_MOUNT) < 0) - I've fixed small issues raised in review of v3. Note that patches 01 and 02 are ACKed already but I'm sending them for completeness (probably doesn't make much sense to merge them while this is still under review, does it?). Michal Prívozník (15): virprocess: Introduce virProcessRunInFork virprocess: Make virProcessRunInMountNamespace use virProcessRunInFork qemu_tpm: Pass virDomainObjPtr instead of virDomainDefPtr qemu_domain: Track if domain remembers original owner virSecurityManagerTransactionCommit: Do metadata locking iff enabled in config security_manager: Rework metadata locking Revert "security_manager: Load lock plugin on init" Revert "qemu_conf: Introduce metadata_lock_manager" Revert "lock_manager: Allow disabling configFile for virLockManagerPluginNew" Revert "lock_driver: Introduce VIR_LOCK_MANAGER_ACQUIRE_ROLLBACK" Revert "lock_driver: Introduce VIR_LOCK_MANAGER_RESOURCE_TYPE_METADATA" Revert "_virLockManagerLockDaemonPrivate: Move @hasRWDisks into dom union" Revert "lock_driver: Introduce new VIR_LOCK_MANAGER_OBJECT_TYPE_DAEMON" Revert "lock_driver_lockd: Introduce VIR_LOCK_SPACE_PROTOCOL_ACQUIRE_RESOURCE_METADATA flag" Revert "virlockspace: Allow caller to specify start and length offset in virLockSpaceAcquireResource" cfg.mk | 4 +- src/libvirt_private.syms | 1 + src/locking/lock_daemon_dispatch.c | 11 +- src/locking/lock_driver.h | 12 - src/locking/lock_driver_lockd.c | 421 ++++++++++------------------- src/locking/lock_driver_lockd.h | 1 - src/locking/lock_driver_sanlock.c | 44 +-- src/locking/lock_manager.c | 10 +- src/lxc/lxc_controller.c | 3 +- src/lxc/lxc_driver.c | 2 +- src/qemu/qemu_conf.c | 1 - src/qemu/qemu_conf.h | 2 +- src/qemu/qemu_domain.c | 7 + src/qemu/qemu_domain.h | 3 + src/qemu/qemu_driver.c | 3 - src/qemu/qemu_extdevice.c | 16 +- src/qemu/qemu_extdevice.h | 4 +- src/qemu/qemu_process.c | 9 +- src/qemu/qemu_security.c | 87 ++++-- src/qemu/qemu_security.h | 4 +- src/qemu/qemu_tpm.c | 24 +- src/qemu/qemu_tpm.h | 4 +- src/security/security_dac.c | 54 ++-- src/security/security_driver.h | 3 +- src/security/security_manager.c | 259 +++++++++--------- src/security/security_manager.h | 22 +- src/security/security_selinux.c | 53 ++-- src/security/security_stack.c | 5 +- src/util/virlockspace.c | 15 +- src/util/virlockspace.h | 4 - src/util/virprocess.c | 82 ++++-- src/util/virprocess.h | 16 ++ tests/seclabeltest.c | 2 +- tests/securityselinuxlabeltest.c | 2 +- tests/securityselinuxtest.c | 2 +- tests/testutilsqemu.c | 2 +- tests/virlockspacetest.c | 29 +- 37 files changed, 573 insertions(+), 650 deletions(-) -- 2.18.1

2 22

[libvirt] [PATCH v3 0/3] Add "memfd" memory backing type (resend)
by marcandre.lureau＠redhat.com 16 Nov '18

16 Nov '18

From: Marc-André Lureau <marcandre.lureau(a)redhat.com> Hi, This is an alternative series from "[PATCH 0/5] Use memfd if possible". Instead of automatically using memfd for anonymous memory when available (as suggested by Daniel), it introduces the "memfd" memory backing type. Although using memfd transparently when possible is a good idea, it is a source of various complications for migration & save/restore. This could eventually be challenged in a different series. *please*: The first two patches have been modified and reviewed by John Ferlan. Hopefully they can be merged early, regardless of the last patch outcome, to avoid the painful rebase conflicts due to capabilities checks introduction. Thanks :) v3: - rebased, to fix capabilities check and ping the series Marc-André Lureau (3): qemu: add memory-backend-memfd capability check qemu: check memory-backend-memfd.hugetlb capability qemu: add memfd source type docs/formatdomain.html.in | 9 +- docs/schemas/domaincommon.rng | 1 + src/conf/domain_conf.c | 3 +- src/conf/domain_conf.h | 1 + src/qemu/qemu_capabilities.c | 10 ++ src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_command.c | 69 +++++++---- src/qemu/qemu_domain.c | 12 +- .../caps_2.12.0.aarch64.replies | 94 ++++++++++++--- .../caps_2.12.0.aarch64.xml | 4 +- .../caps_2.12.0.ppc64.replies | 90 +++++++++++--- .../caps_2.12.0.ppc64.xml | 4 +- .../caps_2.12.0.s390x.replies | 98 ++++++++++++---- .../caps_2.12.0.s390x.xml | 4 +- .../caps_2.12.0.x86_64.replies | 110 +++++++++++++----- .../caps_2.12.0.x86_64.xml | 4 +- .../caps_3.0.0.ppc64.replies | 90 +++++++++++--- .../qemucapabilitiesdata/caps_3.0.0.ppc64.xml | 4 +- .../caps_3.0.0.riscv32.replies | 86 +++++++++++--- .../caps_3.0.0.riscv32.xml | 2 + .../caps_3.0.0.riscv64.replies | 86 +++++++++++--- .../caps_3.0.0.riscv64.xml | 2 + .../caps_3.0.0.s390x.replies | 98 ++++++++++++---- .../qemucapabilitiesdata/caps_3.0.0.s390x.xml | 4 +- .../caps_3.0.0.x86_64.replies | 110 +++++++++++++----- .../caps_3.0.0.x86_64.xml | 4 +- .../memfd-memory-numa.x86_64-latest.args | 34 ++++++ tests/qemuxml2argvdata/memfd-memory-numa.xml | 36 ++++++ tests/qemuxml2argvtest.c | 2 + 29 files changed, 869 insertions(+), 204 deletions(-) create mode 100644 tests/qemuxml2argvdata/memfd-memory-numa.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/memfd-memory-numa.xml -- 2.19.1.708.g4ede3d42df

4 10

[libvirt] [PATCH] docs: Add notes for VIR_DOMAIN_SNAPSHOT_LIST_NO_METADATA
by Han Han 16 Nov '18

16 Nov '18

When listing snapshot with VIR_DOMAIN_SNAPSHOT_LIST_NO_METADATA, it always returns 0 or no snapshot. Because we never implement funtions to list no-metadata snapshot in virDomainSnapshotObjListGetNames(): if ((data.flags & VIR_DOMAIN_SNAPSHOT_FILTERS_METADATA) == VIR_DOMAIN_SNAPSHOT_LIST_NO_METADATA) return 0; Add notes for that flag. Please update the comment and man page of that flag when no-metadata snapshot list is implemented in the future. Signed-off-by: Han Han <hhan(a)redhat.com> --- include/libvirt/libvirt-domain-snapshot.h | 5 ++++- tools/virsh.pod | 3 +++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/include/libvirt/libvirt-domain-snapshot.h b/include/libvirt/libvirt-domain-snapshot.h index 20771f9b1e..2e19a52a5c 100644 --- a/include/libvirt/libvirt-domain-snapshot.h +++ b/include/libvirt/libvirt-domain-snapshot.h @@ -93,7 +93,10 @@ char *virDomainSnapshotGetXMLDesc(virDomainSnapshotPtr snapshot, * of flag (1<<0) depends on which function it is passed to; but serves * to toggle the per-call default of whether the listing is shallow or * recursive. Remaining bits come in groups; if all bits from a group are - * 0, then that group is not used to filter results. */ + * 0, then that group is not used to filter results. Internal functions + * for listing no-metadata snapshots aren't implemented. Functions above + * will return 0 when VIR_DOMAIN_SNAPSHOT_LIST_NO_METADATA is used. + * */ typedef enum { VIR_DOMAIN_SNAPSHOT_LIST_ROOTS = (1 << 0), /* Filter by snapshots with no parents, when diff --git a/tools/virsh.pod b/tools/virsh.pod index 86c041d575..b3d3840c2b 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -4689,6 +4689,9 @@ B<undefine> of a persistent domain, or be lost on B<destroy> of a transient domain. Likewise, if I<--no-metadata> is specified, the list will be filtered to just snapshots that exist without the need for libvirt metadata. +Note that - It will return no snapshot when I<--no-metadata> is +used since internal functions for listing no-metadata snapshot +are not implemented. If I<--inactive> is specified, the list will be filtered to snapshots that were taken when the domain was shut off. If I<--active> is -- 2.19.1

3 3

[libvirt] [PATCH] [patch] docs: remove redundant words and blank lines
by luzhipeng＠uniudc.com 16 Nov '18

16 Nov '18

From: ZhiPeng Lu <luzhipeng(a)uniudc.com> Signed-off-by: ZhiPeng Lu <luzhipeng(a)uniudc.com> --- docs/formatdomain.html.in | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 8a23b78..de25ad3 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -2835,7 +2835,6 @@ </source> <target dev='sdb' bus='scsi'/> </disk> - </disk> <disk type='network' device='lun'> <driver name='qemu' type='raw'/> <source protocol='iscsi' name='iqn.2013-07.com.example:iscsi-nopool/0'> @@ -5216,7 +5215,6 @@ <virtualport> <parameters instanceid='09b11c53-8b5c-4eeb-8f00-d84eaa0aaa4f'/> </virtualport> - </interface> </devices> ...</pre> @@ -5709,8 +5707,7 @@ qemu-kvm -net nic,model=? /dev/null <host csum='off' gso='off' tso4='off' tso6='off' ecn='off' ufo='off' mrg_rxbuf='off'/> <guest csum='off' tso4='off' tso6='off' ecn='off' ufo='off'/> </driver> - - </interface> + </interface> </devices> ...</pre> @@ -6212,7 +6209,7 @@ qemu-kvm -net nic,model=? /dev/null <route family='ipv4' address='192.168.122.0' prefix='24' gateway='192.168.122.1'/> <route family='ipv4' address='192.168.122.8' gateway='192.168.122.1'/> </hostdev> - + ... </devices> ... </pre> -- 1.8.3.1

1 0

[libvirt] [PATCH v2 0/5] qemu_domain, qemu_process: maxCpus check to non-x86 guests
by Daniel Henrique Barboza 15 Nov '18

15 Nov '18

v2: - original patch was split in 3 patches as John Ferlan requested; - patches 4 and 5 are cleanups; - first patch link: https://www.redhat.com/archives/libvir-list/2018-October/msg00251.html Daniel Henrique Barboza (5): qemu_process.c: adding maxCpus value to error message qemu_process.c: make qemuValidateCpuCount public qemu_domain.c: moving maxCpu validation to qemuDomainDefValidate qemu_process.c: moving qemuValidateCpuCount to qemu_domain.c qemu_process.c: removing qemuProcessStartValidateXML src/qemu/qemu_domain.c | 27 +++++++++++++++++++ src/qemu/qemu_process.c | 57 ++++------------------------------------- 2 files changed, 32 insertions(+), 52 deletions(-) -- 2.19.1

2 9

[libvirt] [REPOST PATCH v2 00/12] Allow modification of IOThread polling values (redux)
by John Ferlan 15 Nov '18

15 Nov '18

v2: https://www.redhat.com/archives/libvir-list/2018-October/msg01065.html NB: Minor mods for this are change using 4.10.0 instead of 4.9.0, merge of qemu_capabilities conflict, and updated news.xml ...... v2 Cover Letter: v1: https://www.redhat.com/archives/libvir-list/2018-October/msg00456.html Changes since v1 (from code review) Patch 2: Move the job back into qemuDomainGetIOThreadsLive Patch 3: Add check for ActiveJob before allowing, use true for *StatsWorker, and print 'iothread' in output not 'block' Patch 5: Use virCheckPositiveArgGoto(nparams, error) instead of using virCheckNonNegativeArgGoto(nparams, error). And then remove the if (nparams) before the virCheckNonNullArgGoto(params, error); Patch 6: Add ability to determine which parameter was set via bool set_poll_{max_ns|grow|shrink} values. Then use those in the macro that sets the value to determine whether or not the value will be set based on whether it was changed. Patch 10: Use bool's to set_ when the value is found in the incoming params list. Remove the check that says poll_max_ns needs to be set. Testing proves that if it's set to 0, then the grow and shrink values can be changed (although they do nothing) Patch 12: (NEW) - News article ...... v1 Cover Letter: This series attempts to resurrect the concept of being able to modify the IOThread polling parameters; however, in a slightly different manner than the previous series posted by posted by Pavel Hrdina <phrdina redhat com>: https://www.redhat.com/archives/libvir-list/2017-February/msg01047.html The work is prompted by continued pleas found in the bz: https://bugzilla.redhat.com/show_bug.cgi?id=1545732 to provide some way to modify the paremters without needing to supply QEMU command line pass through values. It's accepted that the values being changed are fairly or extremely low level type knobs; however, it's also shown that by being able to turn the knob it is possible for certain, specific appliances to be able to gain a performance benefit for the thread at the expense of other competing threads. Unlike the previous series, this series does not attempt to save the polling values in the guest XML. Rather, it only modifies the live guest's IOThread with new values. It also doesn't provide the polling values in a virsh iothread* command, rather it uses the domstats in order to fetch and display the values. The theory being this leaves the onus on the higher level appliance/application to provide the "proper guidance" and "concerns" related to changing the values to the consumer. Not saving the values means whatever values that are chosen do not "live" in perpetuity. Once the guest is shut down or the IOThread removed from guest, the hypervisor default values take over again. Perhaps not a perfect situation in terms of what the bz requests; however, storage of default values that could cause performance issues is not an optimal situation. So this I figured is a "comprimise" of sorts. If it's still felt that no we don't want to do this, then fine, but please in doing so own the bz, state your case, and close it. I'm 50/50 on it, but figured at least I'd present this option and see what the concensus was. John Ferlan (12): qemu: Check for and return IOThread polling values if available qemu: Split qemuDomainGetIOThreadsLive qemu: Implement the ability to return IOThread stats virsh: Add ability to display IOThread stats lib: Introduce virDomainSetIOThreadParams qemu: Add monitor functions to set IOThread params qemu: Alter qemuDomainChgIOThread to take enum instead of bool qemu: Alter qemuDomainChgIOThread to take qemuMonitorIOThreadInfo qemu: Detect whether iothread polling is supported qemu: Introduce qemuDomainSetIOThreadParams tools: Add virsh iothreadset command docs: Add news article for IOThread polling docs/news.xml | 13 + include/libvirt/libvirt-domain.h | 45 ++ src/driver-hypervisor.h | 8 + src/libvirt-domain.c | 108 +++++ src/libvirt_public.syms | 5 + src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_driver.c | 384 ++++++++++++++++-- src/qemu/qemu_monitor.c | 19 + src/qemu/qemu_monitor.h | 9 + src/qemu/qemu_monitor_json.c | 50 +++ src/qemu/qemu_monitor_json.h | 4 + src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 21 +- src/remote_protocol-structs | 10 + .../caps_2.10.0.aarch64.xml | 1 + .../caps_2.10.0.ppc64.xml | 1 + .../caps_2.10.0.s390x.xml | 1 + .../caps_2.10.0.x86_64.xml | 1 + .../caps_2.11.0.s390x.xml | 1 + .../caps_2.11.0.x86_64.xml | 1 + .../caps_2.12.0.aarch64.xml | 1 + .../caps_2.12.0.ppc64.xml | 1 + .../caps_2.12.0.s390x.xml | 1 + .../caps_2.12.0.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_2.9.0.ppc64.xml | 1 + .../qemucapabilitiesdata/caps_2.9.0.s390x.xml | 1 + .../caps_2.9.0.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_3.0.0.ppc64.xml | 1 + .../caps_3.0.0.riscv32.xml | 1 + .../caps_3.0.0.riscv64.xml | 1 + .../qemucapabilitiesdata/caps_3.0.0.s390x.xml | 1 + .../caps_3.0.0.x86_64.xml | 1 + tools/virsh-domain-monitor.c | 7 + tools/virsh-domain.c | 110 +++++ tools/virsh.pod | 47 ++- 36 files changed, 825 insertions(+), 37 deletions(-) -- 2.17.2

3 19

[libvirt] [PATCH v3] qemu: agent: Avoid agentError when closing the QEMU agent
by Wang Yechao 15 Nov '18

15 Nov '18

The commit 89563efc0209b854d2b2e554423423d7602acdbd fix the monitor error when closing the QEMU monitor. The QEMU agent has a problem similar to QEMU monitor. So fix the QEMU agent with the same method. Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> --- Changes in v3: - change the commit messages --- src/qemu/qemu_agent.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_agent.c b/src/qemu/qemu_agent.c index 97ad0e7..d842b0e 100644 --- a/src/qemu/qemu_agent.c +++ b/src/qemu/qemu_agent.c @@ -530,6 +530,9 @@ static void qemuAgentUpdateWatch(qemuAgentPtr mon) VIR_EVENT_HANDLE_HANGUP | VIR_EVENT_HANDLE_ERROR; + if (!mon->watch) + return; + if (mon->lastError.code == VIR_ERR_OK) { events |= VIR_EVENT_HANDLE_READABLE; @@ -555,6 +558,12 @@ qemuAgentIO(int watch, int fd, int events, void *opaque) VIR_DEBUG("Agent %p I/O on watch %d fd %d events %d", mon, watch, fd, events); #endif + if (mon->fd == -1 || mon->watch == 0) { + virObjectUnlock(mon); + virObjectUnref(mon); + return; + } + if (mon->fd != fd || mon->watch != watch) { if (events & (VIR_EVENT_HANDLE_HANGUP | VIR_EVENT_HANDLE_ERROR)) eof = true; @@ -788,8 +797,10 @@ void qemuAgentClose(qemuAgentPtr mon) virObjectLock(mon); if (mon->fd >= 0) { - if (mon->watch) + if (mon->watch) { virEventRemoveHandle(mon->watch); + mon->watch = 0; + } VIR_FORCE_CLOSE(mon->fd); } -- 1.8.3.1

2 1