December 2020 - Devel - libvirt List Archives

[PATCH v2 00/27] Introduce virtio memory support
by Michal Privoznik 07 Dec '20

07 Dec '20

v2 of: https://www.redhat.com/archives/libvir-list/2020-November/msg01525.html diff to v1: - I've made change to 23/26 which now updates <currentMemory/> on MEMORY_DEVICE_SIZE_CHANGE event - virsh manpage was written in 25/26 - 26/26 is new, it's NEWS.rst addition Original patchset was: Reviewed-by: Daniel Henrique Barboza <danielhb413(a)gmail.com> Tested-by: Daniel Henrique Barboza <danielhb413(a)gmail.com> Tested-by: Han Han <hhan(a)redhat.com> and I will add those tags before pushing. However, I'd like have review for this new version, because of the changes I made. Michal Prívozník (27): viruuid: Rework virUUIDIsValid() internal.h: Introduce VIR_IS_POW2() docs: Fix nvdimm example wrt to <uuid/> domain_conf: Check NVDIMM UUID in ABI stability qemu_domain_address: Reformat qemuDomainAssignS390Addresses() conf: Require nvdimm path in validate step domain_conf: Fix virDomainMemoryModel type virDomainMemorySourceDefFormat: Utilize virXMLFormatElement() virDomainMemoryTargetDefFormat: Utilize virXMLFormatElement() qemu: Move mem validation into post parse validator conf: Move some of virDomainMemoryDef members into a union conf: Introduce virtio-pmem <memory/> model qemu_capabilities: Introduce QEMU_CAPS_DEVICE_VIRTIO_{P}MEM_PCI qemu_validate: Require virtio-mem device for mem model virtio security: Relabel virtio mem qemu: Allow virtio-pmem in CGroups qemu: Create virtio-pmem in domain namespace qemu_command: Move dimm into qemuBuildDeviceAddressStr() qemu: Build command line for virtio-pmem conf: Introduce virtio-mem <memory/> model qemu: Build command line for virtio-mem qemu: Wire up <memory/> live update qemu: Wire up MEMORY_DEVICE_SIZE_CHANGE event qemu: Refresh the actual size of virtio-mem on monitor reconnect virsh: Simplify @flags handing in cmdSetmem() and cmdSetmaxmem() virsh: Introduce --virtio to setmem news: document recent virtio memory addition NEWS.rst | 9 + docs/formatdomain.rst | 70 +++- docs/manpages/virsh.rst | 6 + docs/schemas/domaincommon.rng | 16 + src/conf/domain_conf.c | 372 ++++++++++++++---- src/conf/domain_conf.h | 38 +- src/internal.h | 10 + src/libvirt_private.syms | 2 + src/qemu/qemu_alias.c | 59 ++- src/qemu/qemu_capabilities.c | 4 + src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_cgroup.c | 43 +- src/qemu/qemu_command.c | 172 +++++--- src/qemu/qemu_command.h | 5 +- src/qemu/qemu_domain.c | 99 +++-- src/qemu/qemu_domain.h | 1 + src/qemu/qemu_domain_address.c | 98 +++-- src/qemu/qemu_domain_address.h | 3 +- src/qemu/qemu_driver.c | 201 +++++++++- src/qemu/qemu_hotplug.c | 22 +- src/qemu/qemu_hotplug.h | 5 + src/qemu/qemu_monitor.c | 37 ++ src/qemu/qemu_monitor.h | 27 ++ src/qemu/qemu_monitor_json.c | 94 +++-- src/qemu/qemu_monitor_json.h | 5 + src/qemu/qemu_namespace.c | 19 +- src/qemu/qemu_process.c | 107 ++++- src/qemu/qemu_validate.c | 78 ++-- src/security/security_apparmor.c | 35 +- src/security/security_dac.c | 48 ++- src/security/security_selinux.c | 48 ++- src/security/virt-aa-helper.c | 22 +- src/util/virrandom.c | 2 +- src/util/viruuid.c | 17 +- src/util/viruuid.h | 2 +- .../caps_4.1.0.x86_64.xml | 1 + .../caps_4.2.0.x86_64.xml | 1 + .../caps_5.0.0.x86_64.xml | 1 + .../caps_5.1.0.x86_64.xml | 2 + .../caps_5.2.0.x86_64.xml | 2 + ...mory-hotplug-virtio-mem.x86_64-latest.args | 53 +++ .../memory-hotplug-virtio-mem.xml | 78 ++++ ...ory-hotplug-virtio-pmem.x86_64-latest.args | 45 +++ .../memory-hotplug-virtio-pmem.xml | 54 +++ tests/qemuxml2argvtest.c | 2 + ...emory-hotplug-virtio-mem.x86_64-latest.xml | 1 + ...mory-hotplug-virtio-pmem.x86_64-latest.xml | 1 + tests/qemuxml2xmltest.c | 3 + tools/virsh-domain.c | 138 ++++++- 49 files changed, 1766 insertions(+), 394 deletions(-) create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-mem.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-mem.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-pmem.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-pmem.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-virtio-mem.x86_64-latest.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-virtio-pmem.x86_64-latest.xml -- 2.26.2

4 57

[PATCH] qemu_monitor_json: Don't leak "option" in qemuMonitorJSONGetCommandLineOptions()
by Michal Privoznik 07 Dec '20

07 Dec '20

In recent commit of bf8bd93df0 (and friends) we switched the way we process queried command line arguments: from string lists to virJSONValue stored in a hash table. To achieve this qemuMonitorJSONGetCommandLineOptions() helper was introduced which executes the "query-command-line-options" monitor command and then calls virJSONValueArrayForeachSteal() to process the output. The array process function is also given qemuMonitorJSONGetCommandLineOptionsWorker() as the callback which is called over each item of the returned array. This callback then steals "parameters" attribute of each array iteam storing it in the hash table, but it leaves behind "option" attribute (because it's g_strdup()-ed). After all of this, the callback returns 0 which is a signal to the array processing function that the callback took ownership of the array item. But this is not true. While it removed "parameters" it did not take the rest ("option" for instance). And therefore, it leads to a memory leak: 5,347 (1,656 direct, 3,691 indirect) bytes in 69 blocks are definitely lost in loss record 2,752 of 2,794 at 0x483BEC5: calloc (vg_replace_malloc.c:760) by 0x4E25A10: g_malloc0 (in /usr/lib64/libglib-2.0.so.0.6400.5) by 0x4943317: virJSONValueNewObject (virjson.c:569) by 0x4945692: virJSONParserHandleStartMap (virjson.c:1768) by 0x5825A86: yajl_do_parse (in /usr/lib64/libyajl.so.2.1.0) by 0x4945BFA: virJSONValueFromString (virjson.c:1896) by 0xAF5C115: qemuMonitorJSONIOProcessLine (qemu_monitor_json.c:224) by 0xAF5C45E: qemuMonitorJSONIOProcess (qemu_monitor_json.c:279) by 0xAF4BB6C: qemuMonitorIOProcess (qemu_monitor.c:342) by 0xAF4C444: qemuMonitorIO (qemu_monitor.c:574) by 0x4FEF846: socket_source_dispatch (in /usr/lib64/libgio-2.0.so.0.6400.5) by 0x4E1F727: g_main_context_dispatch (in /usr/lib64/libglib-2.0.so.0.6400.5) The callback must return 1 so that the array item is properly freed. Fixes: ebeff6cd57d07c89d42e191ed0085a9dd89835c5 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_monitor_json.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 4db00e284a..8480338f46 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -6357,7 +6357,7 @@ qemuMonitorJSONGetCommandLineOptionsWorker(size_t pos G_GNUC_UNUSED, g_hash_table_insert(options, g_strdup(name), parameters); parameters = NULL; - return 0; + return 1; } -- 2.26.2

2 1

Default issue template for libvirt.git
by Peter Krempa 07 Dec '20

07 Dec '20

Hi, we've recently added templates for issues for the libvirt repo [1]. Unfortunately gitlab made a weird design decision, where the default template is not actually picked from the templates comitted to the repository, but needs to be added in the repo config in the web-ui into a separate textarea. This unfortunately creates a gap where the repo-comitted stuff can actually diverge from the default. I've asked Pavel to copy the 'bug.md' template into the "default" one so even if users don't select any they will see the bug one by default. My reasoning despite being vulnerable to diverging from the in-repo state is that we've added templates so that we prevent too-little information being filled. Having to select a template requires additional steps and also asks for more than the reporter might have reported normally, so it may be ofputting for them to select it manually. Thus we rather want to force a default. Alternatively we can add another comment on top asking them to switch to a different template for enhancement such as:  Peter [1] https://www.redhat.com/archives/libvir-list/2020-December/msg00003.html

1 0

[PATCH] spec: keep existing nwfilters uuid on update
by Nikolay Shirokovskiy 06 Dec '20

06 Dec '20

Now on every nwfilter config package update we overwrite existing filters entirely. It is desired to bring new version of filters on update but we'd better keep their uuids I guess. Actually patch primarily address noise in logs on update. If both libvirtd and firewalld are running and libvirt is using firewalld backend then on firewalld restart we reload all nwfilters. So if node is updated and we have update for both firewalld and libvirt then in the process of update first new nwfilters of libvirt package are copied to /etc/libvirt/nwfilters then firewalld is restarted and then libvirtd is restarted. In this process firewalld restart cause log messages like [1]. The issue is libvirt brings nwfilters without <uuid> in definition and on handling firewalld restart libvirt generates missing uuid and then fail to update filter definition because it is already present in filters list with different uuid. [1] virNWFilterObjListAssignDef:337 : operation failed: filter 'no-ip-spoofing' already exists with uuid c302edf9-8a48-40d8-a652-f70b2c563ad1 Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy(a)virtuozzo.com> --- libvirt.spec.in | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 2a4324b..6a31440 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1438,7 +1438,18 @@ fi rm -rf %{_localstatedir}/lib/rpm-state/libvirt || : %post daemon-config-nwfilter -cp %{_datadir}/libvirt/nwfilter/*.xml %{_sysconfdir}/libvirt/nwfilter/ +# keep existing filters uuid on update +for dfile in %{_datadir}/libvirt/nwfilter/*.xml; do + sfile=%{_sysconfdir}/libvirt/nwfilter/`basename $dfile` + if [ -f "$sfile" ]; then + uuidstr=`sed -n '/<uuid>.*<\/uuid>/p' "$sfile"` + if [ ! -z "$uuidstr" ]; then + sed -e "s,<filter .*>,&\n$uuidstr," "$dfile" > "$sfile" + continue + fi + fi + cp "$dfile" "$sfile" +done # libvirt saves these files with mode 600 chmod 600 %{_sysconfdir}/libvirt/nwfilter/*.xml # Make sure libvirt picks up the new nwfilter defininitons -- 1.8.3.1

4 5

[PATCH v2 00/10] qemu: add option to process offloaded legacy blockjob event ealier
by Nikolay Shirokovskiy 04 Dec '20

04 Dec '20

This is successor to [1] but I changed the subject as in the review the patch 'qemu: sync backing chain update in virDomainGetBlockJobInfo' was not considered good one from design POV. However I think the basic patch is helpful to address similar issues. Look at [*] for example, there it allows to sync backing chains during query block stats. In discussion of [1] I stated that first patch will also allow to get rid of stale block job events on reconnection to qemu. But this will require additional work actually so with this patch series stale events are still present. And in the discussion it was also mentioned by Peter that stale events are not harmful for legacy blockjobs code. I also removed previous attempts to eliminate some of stale events. I still keep patch for virDomainGetBlockJobInfo. May be in comparsion to [*] the approach the patch takes will look not so bad :) [1] First version of the patch series https://www.redhat.com/archives/libvir-list/2020-October/msg01133.html Nikolay Shirokovskiy (10): qemu: add option to process offloaded legacy blockjob event ealier qemu: reconnect: precreate legacy blockjobs qemu: remove extra block job finalize on reconnect qemu: remove stale cleanup in qemuProcessRefreshLegacyBlockjob qemu: add note for outdated legacy block job events qemu: use autoptr in qemuProcessRefreshLegacyBlockjobs qemu: refresh backing chain after block job reconnection qemu: move code that depends on backing chain appropriately qemu: fix race on legacy block completion and quering stats [*] qemu: sync backing chain update in virDomainGetBlockJobInfo src/qemu/qemu_blockjob.c | 2 +- src/qemu/qemu_blockjob.h | 7 +++ src/qemu/qemu_driver.c | 50 +++++++++++----- src/qemu/qemu_process.c | 149 ++++++++++++++++++++++++++++------------------- 4 files changed, 132 insertions(+), 76 deletions(-) -- 1.8.3.1

2 20

[PATCH v3 0/5] add post parse pSeries mem align, when ABI_UPDATE
by Daniel Henrique Barboza 04 Dec '20

04 Dec '20

Hi, This is the v3 of https://www.redhat.com/archives/libvir-list/2020-November/msg01057.html Some patches were already pushed. This series contains the remaining ones that got feedback from Andrea. Michal, I added your Reviewed-by tag in all my patches because I considered that the changes made weren't worth making you review them again. Changes in v3: - former patches 1, 2 and 4: already pushed. - patches 1 and 2: these are Andrea's patches from [1]. The other patches were rebased upon them. They're being sent in this series because they're not pushed upstream in the moment of this posting. - patch 3: reverted an extra commit that was made deprecated after reverting the post parse code, as suggested by Andrea. - patch 4 and 5 (former 5 and 6): these were taken from Andrea's local branch in [2], with the test changes suggested by Andrea. [1] https://www.redhat.com/archives/libvir-list/2020-December/msg00244.html [2] https://gitlab.com/abologna/libvirt/-/tree/ppc64-memalign Andrea Bolognani (2): tests: Sync some ppc64 tests tests: Simplify some ppc64 tests Daniel Henrique Barboza (3): domain_conf.c: revert auto-alignment of pSeries NVDIMM in post parse qemu_domain.c: post parse pSeries NVDIMM align with PARSE_ABI_UPDATE qemu_domain.c: align all pSeries mem modules when PARSE_ABI_UPDATE src/conf/domain_conf.c | 59 +------------- src/conf/domain_conf.h | 3 - src/libvirt_private.syms | 1 - src/qemu/qemu_domain.c | 80 ++++++++++++++++++- ...mory-hotplug-nvdimm-ppc64-abi-update.args} | 20 ++--- ...memory-hotplug-nvdimm-ppc64-abi-update.xml | 1 + .../memory-hotplug-nvdimm-ppc64.args | 6 +- .../memory-hotplug-nvdimm-ppc64.xml | 8 +- ...emory-hotplug-ppc64-nonuma-abi-update.args | 29 +++++++ ...memory-hotplug-ppc64-nonuma-abi-update.xml | 1 + .../memory-hotplug-ppc64-nonuma.args | 7 +- .../memory-hotplug-ppc64-nonuma.xml | 14 +++- tests/qemuxml2argvtest.c | 18 ++++- ...memory-hotplug-nvdimm-ppc64-abi-update.xml | 46 +++++++++++ .../memory-hotplug-nvdimm-ppc64.xml | 10 +-- ...memory-hotplug-ppc64-nonuma-abi-update.xml | 38 +++++++++ .../memory-hotplug-ppc64-nonuma.xml | 1 + tests/qemuxml2xmltest.c | 19 +++++ 18 files changed, 254 insertions(+), 107 deletions(-) rename tests/qemuxml2argvdata/{memory-hotplug-nvdimm-ppc64.ppc64-latest.args => memory-hotplug-nvdimm-ppc64-abi-update.args} (62%) create mode 120000 tests/qemuxml2argvdata/memory-hotplug-nvdimm-ppc64-abi-update.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-ppc64-nonuma-abi-update.args create mode 120000 tests/qemuxml2argvdata/memory-hotplug-ppc64-nonuma-abi-update.xml create mode 100644 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-ppc64-abi-update.xml create mode 100644 tests/qemuxml2xmloutdata/memory-hotplug-ppc64-nonuma-abi-update.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-ppc64-nonuma.xml -- 2.26.2

2 12

[PATCH v2 RESEND] lxc: Cleanup after failed startup
by Michal Privoznik 04 Dec '20

04 Dec '20

If starting an container fails, the virLXCProcessStop() is called. But since vm->def->id is not set until libvirt_lxc is spawned (the domain's ID is PID of that process), virLXCProcessStop() returns early as virDomainObjIsActive() returns false. But doing so leaves behind resources reserved for the containers during the startup process. Most notably, hostdevs are not re-attached to the host, the domain's transient XML is not removed, etc. To resolve this, virLXCProcessCleanup() is called in this case. However, it is modified to accept @flags which allows caller to run only specific cleanups (depending how far in container creation the failure occurred). There is plenty of cleanups which don't need this guard because either they detect a NULL pointer or try to release an unique resource. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- This is just a resend of this patch: https://www.redhat.com/archives/libvir-list/2020-November/msg00525.html src/lxc/lxc_process.c | 52 ++++++++++++++++++++++++++++++++++--------- 1 file changed, 41 insertions(+), 11 deletions(-) diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index 7e07f49f9a..0f818e2ee0 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -145,18 +145,27 @@ lxcProcessRemoveDomainStatus(virLXCDriverConfigPtr cfg, } +typedef enum { + VIR_LXC_PROCESS_CLEANUP_RELEASE_SECLABEL = (1 << 0), + VIR_LXC_PROCESS_CLEANUP_RESTORE_SECLABEL = (1 << 1), + VIR_LXC_PROCESS_CLEANUP_REMOVE_TRANSIENT = (1 << 2), +} virLXCProcessCleanupFlags; + /** * virLXCProcessCleanup: * @driver: pointer to driver structure * @vm: pointer to VM to clean up * @reason: reason for switching the VM to shutoff state + * @flags: allows to run selective cleanups only * - * Cleanout resources associated with the now dead VM - * + * Clean out resources associated with the now dead VM. + * If @flags is zero then whole cleanup process is done, + * otherwise only selected sections are run. */ static void virLXCProcessCleanup(virLXCDriverPtr driver, virDomainObjPtr vm, - virDomainShutoffReason reason) + virDomainShutoffReason reason, + unsigned int flags) { size_t i; virLXCDomainObjPrivatePtr priv = vm->privateData; @@ -164,8 +173,11 @@ static void virLXCProcessCleanup(virLXCDriverPtr driver, virLXCDriverConfigPtr cfg = virLXCDriverGetConfig(driver); virConnectPtr conn = NULL; - VIR_DEBUG("Cleanup VM name=%s pid=%d reason=%d", - vm->def->name, (int)vm->pid, (int)reason); + VIR_DEBUG("Cleanup VM name=%s pid=%d reason=%d flags=0x%x", + vm->def->name, (int)vm->pid, (int)reason, flags); + + if (flags == 0) + flags = ~0; /* now that we know it's stopped call the hook if present */ if (virHookPresent(VIR_HOOK_DRIVER_LXC)) { @@ -177,9 +189,15 @@ static void virLXCProcessCleanup(virLXCDriverPtr driver, NULL, xml, NULL); } - virSecurityManagerRestoreAllLabel(driver->securityManager, - vm->def, false, false); - virSecurityManagerReleaseLabel(driver->securityManager, vm->def); + if (flags & VIR_LXC_PROCESS_CLEANUP_RESTORE_SECLABEL) { + virSecurityManagerRestoreAllLabel(driver->securityManager, + vm->def, false, false); + } + + if (flags & VIR_LXC_PROCESS_CLEANUP_RELEASE_SECLABEL) { + virSecurityManagerReleaseLabel(driver->securityManager, vm->def); + } + /* Clear out dynamically assigned labels */ if (vm->def->nseclabels && vm->def->seclabels[0]->type == VIR_DOMAIN_SECLABEL_DYNAMIC) { @@ -258,7 +276,9 @@ static void virLXCProcessCleanup(virLXCDriverPtr driver, NULL, xml, NULL); } - virDomainObjRemoveTransientDef(vm); + if (flags & VIR_LXC_PROCESS_CLEANUP_REMOVE_TRANSIENT) + virDomainObjRemoveTransientDef(vm); + virObjectUnref(cfg); virObjectUnref(conn); } @@ -904,7 +924,7 @@ int virLXCProcessStop(virLXCDriverPtr driver, } cleanup: - virLXCProcessCleanup(driver, vm, reason); + virLXCProcessCleanup(driver, vm, reason, 0); return 0; } @@ -1198,6 +1218,7 @@ int virLXCProcessStart(virConnectPtr conn, g_autoptr(virCgroup) selfcgroup = NULL; int status; g_autofree char *pidfile = NULL; + unsigned int stopFlags = 0; if (virCgroupNewSelf(&selfcgroup) < 0) return -1; @@ -1265,6 +1286,7 @@ int virLXCProcessStart(virConnectPtr conn, VIR_DEBUG("Setting current domain def as transient"); if (virDomainObjSetDefTransient(driver->xmlopt, vm, NULL) < 0) goto cleanup; + stopFlags |= VIR_LXC_PROCESS_CLEANUP_REMOVE_TRANSIENT; /* Run an early hook to set-up missing devices */ if (virHookPresent(VIR_HOOK_DRIVER_LXC)) { @@ -1312,11 +1334,13 @@ int virLXCProcessStart(virConnectPtr conn, goto cleanup; } virDomainAuditSecurityLabel(vm, true); + stopFlags |= VIR_LXC_PROCESS_CLEANUP_RELEASE_SECLABEL; VIR_DEBUG("Setting domain security labels"); if (virSecurityManagerSetAllLabel(driver->securityManager, vm->def, NULL, false, false) < 0) goto cleanup; + stopFlags |= VIR_LXC_PROCESS_CLEANUP_RESTORE_SECLABEL; VIR_DEBUG("Setting up consoles"); for (i = 0; i < vm->def->nconsoles; i++) { @@ -1525,7 +1549,13 @@ int virLXCProcessStart(virConnectPtr conn, } if (rc != 0) { virErrorPreserveLast(&err); - virLXCProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED); + if (virDomainObjIsActive(vm)) { + virLXCProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED); + } else { + /* virLXCProcessStop() is NOP if the container is not active. + * If there was a failure whilst creating it, cleanup manually. */ + virLXCProcessCleanup(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED, stopFlags); + } } virCommandFree(cmd); for (i = 0; i < nttyFDs; i++) -- 2.26.2

2 1

[PATCH] qemu: Drop @qemuCaps argument from qemuDomainDefValidateMemoryHotplug()
by Michal Privoznik 04 Dec '20

04 Dec '20

After previous cleanup the @qemuCaps argument in qemuDomainDefValidateMemoryHotplug() is unused and thus doesn't need to be passed. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_domain.c | 2 -- src/qemu/qemu_domain.h | 1 - src/qemu/qemu_hotplug.c | 2 +- src/qemu/qemu_validate.c | 2 +- 4 files changed, 2 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index e705e8d8d5..5b58ccbf0e 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -8532,7 +8532,6 @@ qemuDomainDefValidateMemoryHotplugDevice(const virDomainMemoryDef *mem, /** * qemuDomainDefValidateMemoryHotplug: * @def: domain definition - * @qemuCaps: qemu capabilities object * @mem: definition of memory device that is to be added to @def with hotplug, * NULL in case of regular VM startup * @@ -8544,7 +8543,6 @@ qemuDomainDefValidateMemoryHotplugDevice(const virDomainMemoryDef *mem, */ int qemuDomainDefValidateMemoryHotplug(const virDomainDef *def, - virQEMUCapsPtr qemuCaps G_GNUC_UNUSED, const virDomainMemoryDef *mem) { unsigned int nmems = def->nmems; diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 010bae285d..9e870ff1e2 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -795,7 +795,6 @@ int qemuDomainAdjustMaxMemLockHostdev(virDomainObjPtr vm, virDomainHostdevDefPtr hostdev); int qemuDomainDefValidateMemoryHotplug(const virDomainDef *def, - virQEMUCapsPtr qemuCaps, const virDomainMemoryDef *mem); bool qemuDomainSupportsNewVcpuHotplug(virDomainObjPtr vm); diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 29a30c6812..9b93f256e8 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -2399,7 +2399,7 @@ qemuDomainAttachMemory(virQEMUDriverPtr driver, if (qemuDomainMemoryDeviceAlignSize(vm->def, mem) < 0) goto cleanup; - if (qemuDomainDefValidateMemoryHotplug(vm->def, priv->qemuCaps, mem) < 0) + if (qemuDomainDefValidateMemoryHotplug(vm->def, mem) < 0) goto cleanup; if (qemuDomainAssignMemoryDeviceSlot(vm->def, mem) < 0) diff --git a/src/qemu/qemu_validate.c b/src/qemu/qemu_validate.c index 8ceea022d7..52d15defed 100644 --- a/src/qemu/qemu_validate.c +++ b/src/qemu/qemu_validate.c @@ -1156,7 +1156,7 @@ qemuValidateDomainDef(const virDomainDef *def, if (qemuValidateDomainDefCpu(driver, def, qemuCaps) < 0) return -1; - if (qemuDomainDefValidateMemoryHotplug(def, qemuCaps, NULL) < 0) + if (qemuDomainDefValidateMemoryHotplug(def, NULL) < 0) return -1; if (qemuValidateDomainDefClockTimers(def, qemuCaps) < 0) -- 2.26.2

2 1

[PATCH] apparmor: Drop needless check in AppArmorSetMemoryLabel()
by Michal Privoznik 04 Dec '20

04 Dec '20

The AppArmorSetMemoryLabel() is a callback that is called from qemuSecuritySetMemoryLabel() which never passes NULL as @mem. Therefore, there is no need to check whether @mem is NULL. Also, no other driver does that and just dereference it immediately. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/security/security_apparmor.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/security/security_apparmor.c b/src/security/security_apparmor.c index eed66e460f..1b035cce2f 100644 --- a/src/security/security_apparmor.c +++ b/src/security/security_apparmor.c @@ -681,9 +681,6 @@ AppArmorSetMemoryLabel(virSecurityManagerPtr mgr, virDomainDefPtr def, virDomainMemoryDefPtr mem) { - if (mem == NULL) - return 0; - switch (mem->model) { case VIR_DOMAIN_MEMORY_MODEL_NVDIMM: if (!virFileExists(mem->nvdimmPath)) { -- 2.26.2

2 1

[PATCH v2 0/6] add post parse pSeries mem align, when ABI_UPDATE
by Daniel Henrique Barboza 04 Dec '20

04 Dec '20

Hi, This is a follow up of [1] after really comprehending what I was messing with and what I couldn't do. This version has a shorter scope, only pSeries guests are being taken care of. I can send a follow up to handle x86 as well depending on the popularity of this work. Patches 1 and 2 moves the existing qemuDomainAlignMemorySizes() from qemu_command.c to qemuProcessPrepareDomain(). They are not related/tied to everything else done here and can be pushed independently if needed. Patches 3, 4 and 5 are reworking the existing code to be consistent to our prerrogative of not aligning memory when migrating or 'snapshotting'. No significant behavioral change is done. Patch 6 is where the bacon is. For new ppc64 guests, without ABI breakage, the mem alignment that are overshooting the intended initial memory is fixed. Test cases were added to help me diagnose and assert what I was changing and what would remain untouched. [1] https://www.redhat.com/archives/libvir-list/2020-November/msg00544.html Daniel Henrique Barboza (6): qemu_process.c: check migrateURI when setting VIR_QEMU_PROCESS_START_NEW qemu: move memory size align to qemuProcessPrepareDomain() Revert "domain_conf.c: auto-align pSeries NVDIMM in virDomainMemoryDefPostParse()" qemuxml2xmltest.c: honor ARG_PARSEFLAGS qemu_domain.c: post parse pSeries NVDIMM align with PARSE_ABI_UPDATE qemu_domain.c: align all pSeries mem modules when PARSE_ABI_UPDATE src/conf/domain_conf.c | 23 +-------- src/qemu/qemu_command.c | 3 -- src/qemu/qemu_domain.c | 39 ++++++++++++++- src/qemu/qemu_process.c | 11 +++- ...memory-hotplug-nvdimm-ppc64-abi-update.xml | 50 +++++++++++++++++++ ...emory-hotplug-ppc64-nonuma-abi-update.args | 34 +++++++++++++ ...memory-hotplug-ppc64-nonuma-abi-update.xml | 32 ++++++++++++ tests/qemuxml2argvtest.c | 7 +++ ...memory-hotplug-nvdimm-ppc64-abi-update.xml | 50 +++++++++++++++++++ .../memory-hotplug-nvdimm-ppc64.xml | 2 +- ...memory-hotplug-ppc64-nonuma-abi-update.xml | 45 +++++++++++++++++ tests/qemuxml2xmltest.c | 20 +++++++- 12 files changed, 286 insertions(+), 30 deletions(-) create mode 100644 tests/qemuxml2argvdata/memory-hotplug-nvdimm-ppc64-abi-update.xml create mode 100644 tests/qemuxml2argvdata/memory-hotplug-ppc64-nonuma-abi-update.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-ppc64-nonuma-abi-update.xml create mode 100644 tests/qemuxml2xmloutdata/memory-hotplug-nvdimm-ppc64-abi-update.xml create mode 100644 tests/qemuxml2xmloutdata/memory-hotplug-ppc64-nonuma-abi-update.xml -- 2.26.2

3 19