January 2020 - Devel - libvirt List Archives

[PATCH] libxl: support getting and setting parameters for the Credit2
by Dario Faggioli 30 Jan '20

30 Jan '20

With Credit2 being Xen default scheduler, it's definitely the case to allow Credit2's scheduling parameters to be get and set via libvirt. This is easy, as Credit and Credit2 have (at least as of now) the very same parameters ('weight' and 'cap'). So we can just let credit2 pass the scheduler-type check and the same code will work for both. Signed-off-by: Dario Faggioli <dfaggioli(a)suse.com> --- Cc: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libxl_driver.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index e0a8ec5c24..78ca77d0f6 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -75,6 +75,7 @@ VIR_LOG_INIT("libxl.libxl_driver"); /* Number of Xen scheduler parameters */ #define XEN_SCHED_CREDIT_NPARAM 2 +#define XEN_SCHED_CREDIT2_NPARAM 2 #define LIBXL_CHECK_DOM0_GOTO(name, label) \ do { \ @@ -4586,6 +4587,8 @@ libxlDomainGetSchedulerType(virDomainPtr dom, int *nparams) break; case LIBXL_SCHEDULER_CREDIT2: name = "credit2"; + if (nparams) + *nparams = XEN_SCHED_CREDIT2_NPARAM; break; case LIBXL_SCHEDULER_ARINC653: name = "arinc653"; @@ -4632,11 +4635,11 @@ libxlDomainGetSchedulerParametersFlags(virDomainPtr dom, if (virDomainObjCheckActive(vm) < 0) goto cleanup; + /* Only credit and credit2 are supported for now. */ sched_id = libxl_get_scheduler(cfg->ctx); - - if (sched_id != LIBXL_SCHEDULER_CREDIT) { + if (sched_id != LIBXL_SCHEDULER_CREDIT && sched_id != LIBXL_SCHEDULER_CREDIT2) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Only 'credit' scheduler is supported")); + _("Only 'credit' and 'credit2' schedulers are supported")); goto cleanup; } @@ -4657,6 +4660,9 @@ libxlDomainGetSchedulerParametersFlags(virDomainPtr dom, goto cleanup; } + /* Credit and Credit2 have the same number (two) of parameters, + * so this is ok for both, at least as long as that stays true. */ + G_STATIC_ASSERT(XEN_SCHED_CREDIT_NPARAM == XEN_SCHED2_CREDIT_NPARAM); if (*nparams > XEN_SCHED_CREDIT_NPARAM) *nparams = XEN_SCHED_CREDIT_NPARAM; ret = 0; @@ -4711,9 +4717,11 @@ libxlDomainSetSchedulerParametersFlags(virDomainPtr dom, sched_id = libxl_get_scheduler(cfg->ctx); - if (sched_id != LIBXL_SCHEDULER_CREDIT) { + /* Only credit and credit2 are supported for now. */ + sched_id = libxl_get_scheduler(cfg->ctx); + if (sched_id != LIBXL_SCHEDULER_CREDIT && sched_id != LIBXL_SCHEDULER_CREDIT2) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Only 'credit' scheduler is supported")); + _("Only 'credit' and 'credit2' schedulers are supported")); goto endjob; } -- 2.24.1 -- Dario Faggioli, Ph.D http://about.me/dario.faggioli Virtualization Software Engineer SUSE Labs, SUSE https://www.suse.com/ ------------------------------------------------------------------- <<This happens because _I_ choose it to happen!>> (Raistlin Majere)

2 1

RHEL 7.7/libvirt 5.10 Question
by bart.m.bailey＠verizon.com 29 Jan '20

29 Jan '20

Hello, RHEL 7.7 installation attempts of libvirt-5.10.0-1.fc30.src.rpm are encountering the following - make[4]: Entering directory `/root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/gnulib/tests' CC test-accept.o CC test-alloca-opt.o CC test-arpa_inet.o CC test-binary-io.o ============================================================================ Testsuite summary for libvirt 5.10.0 ============================================================================ # TOTAL: 197 # PASS: 187 # SKIP: 10 # XFAIL: 0 # FAIL: 0 # XPASS: 0 # ERROR: 0 ============================================================================ With subsequent testsuite execcuiton registering the following failure - ============================================================================ Testsuite summary for libvirt 5.10.0 ============================================================================ # TOTAL: 131 # PASS: 130 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 ============================================================================ See tests/test-suite.log Please report to libvir-list(a)redhat.com<mailto:libvir-list@redhat.com> ============================================================================ make[3]: *** [test-suite.log] Error 1 make[3]: Leaving directory `/root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/tests' make[2]: *** [check-TESTS] Error 2 make[2]: Leaving directory `/root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/tests' make[1]: *** [check-am] Error 2 make[1]: Leaving directory `/root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/tests' make: *** [check-recursive] Error 1 + cat test-suite.log cat: test-suite.log: No such file or directory + true + exit 1 error: Bad exit status from /var/tmp/rpm-tmp.eMOkrN (%check) RPM build errors: Bad exit status from /var/tmp/rpm-tmp.eMOkrN (%check) The gnulib-test-suite.log file attachment registers the following event ... 35) Checking luks-disk-invalid.xml against domain.rng ... libvirt: XML Util error : XML document failed to validate against schema: Unable to validate doc against /root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/../docs/schemas/domain.rng Extra element devices in interleave Element domain failed to validate content OK 36) Checking disk-serial.xml against domain.rng ... OK 37) Checking pseries-no-parallel.xml against domain.rng ... OK ...with subsequent xml errors following similar to below. 2115) Checking dir-missing-target-path-invalid.xml against storagepool.rng ... libvirt: XML Util error : XML document failed to validate against schema: Unable to validate doc against /root/rpm/BUILD/libvirt-5.10.0/x86_64-redhat-linux-gnu/../docs/schemas/storagepool.rng Expecting an element path, got nothing Invalid sequence in interleave Element target failed to validate content Invalid sequence in interleave Element pool failed to validate content OK File attachments x86_64-redhat-linux-gnu/gnulib/tests/test-suite.log, x86_64-redhat-linux-gnu/tests/test-suite.log are provided and not sure if this helps but libvirt 5.10 candidate release libvirt-5.10.0-0rc2.fc30.src.rpm<https://libvirt.org/sources/libvirt-5.10.0-0rc2.fc30.src.rpm> completed RPM builds successfully. Libvirt 5.10 release libvirt-5.10.0-1.fc30.src.rpm is required for this application. Thanks and let us know if anything else is needed from this end. [cid:image002.png@01D5D5B6.9ADA0040] Bart Bailey (216) 462-2173 Verizon ME Lab Richardson 400 International Parkway Richardson, TX 75081

2 2

[PATCH v1 0/2] DAC,SELinux: do not remember/recall labels for
by Daniel Henrique Barboza 29 Jan '20

29 Jan '20

Libvirt is trying to do multiple open() calls in /dev/vfio files, which results in errors inside the security drivers and QEMU returning error 125 when attempting to hotplug a hostdev which belongs to the same IOMMU group as an existing domain hostdev. See patch 2 commit msg for more details. I found this problem when testing a PCI multifunction hotplug implementation. Since this is a problem that can happen in other currently supported scenarios, I decided it was worth sending the fix right away. Michal, I'm CCing you because I mentioned commit 4e95cdcbb3 ("security: Don't remember labels for TPM") in patch 2, which seems to fix a very similar problem. Figured you might want to take a look. Daniel Henrique Barboza (2): security: Allow 'remember' to be set for HostdevLabelHelper security: do not remember/recall labels for VFIO src/security/security_dac.c | 20 ++++++++++++-------- src/security/security_selinux.c | 20 ++++++++++++-------- 2 files changed, 24 insertions(+), 16 deletions(-) -- 2.24.1

2 3

[libvirt PATCH] docs: add warning about virt-qemu-run being experimental
by Daniel P. Berrangé 29 Jan '20

29 Jan '20

We currently note that the QEMU embedded driver in experimental in the drvqemu.html file, but we should do the same in the virt-qemu-run man page. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- docs/manpages/virt-qemu-run.rst | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/docs/manpages/virt-qemu-run.rst b/docs/manpages/virt-qemu-run.rst index 5074a928b7..b06c311b1d 100644 --- a/docs/manpages/virt-qemu-run.rst +++ b/docs/manpages/virt-qemu-run.rst @@ -33,6 +33,11 @@ If the virt-qemu-run program is interrupted (eg Ctrl-C) it will immediately terminate the virtual machine without giving the guest OS any opportunity to gracefully shutdown. +**NOTE: this tool is currently considered experimental.** Its +usage and behaviour is still subject to change in future libvirt +releases. For further information on its usage consult the +`QEMU driver documentation <https://libvirt.org/drvqemu.html#uriembedded>`_. + OPTIONS ======= -- 2.24.1

2 1

[libvirt PATCH 0/2] docs: make rst docs generation more robust
by Daniel P. Berrangé 29 Jan '20

29 Jan '20

This will avoid the various issues with generated docs containing build warnings. Daniel P. Berrangé (2): docs: fix various duplicate link targets in virsh.rst docs: use --strict arg when processing rst docs docs/Makefile.am | 10 +++++----- docs/manpages/virsh.rst | 36 ++++++++++++++++++------------------ 2 files changed, 23 insertions(+), 23 deletions(-) -- 2.24.1

2 3

[libvirt-rust] change in coding style to follow rustfmt
by Sahid Orentino Ferdjaoui 29 Jan '20

29 Jan '20

I'm proposing to merge a change [0] so the CI would check whether the source code is well-formatted using rustfmt [1]. The aim is to add an easy convention to follow regarding the coding style. For contributors the natural thing to do before to submit any patches is to verify the format using: `cargo fmt -v -- --check` The support of rustfmt via cargo can be added using: `rustup component add rustfmt` I did not want to pollute the mailing list with such big change which doesn't bring anything. Please let me know if you have any objections. commit 851296786626d61ebec985a9caeb515514955dd8 Author: Sahid Orentino Ferdjaoui <sahid.ferdjaoui(a)canonical.com> Date: Tue Jan 28 15:05:44 2020 +0100 ensure that the code is well formatted using rustfmt. In this commit we add CI rule so travis would check whether the code is well formatted using rustfmt. Signed-off-by: Sahid Orentino Ferdjaoui <sahid.ferdjaoui(a)canonical.com> .travis.yml | 4 + README.md | 6 + examples/auth.rs | 19 +- examples/hello.rs | 76 +-- examples/migrate.rs | 35 +- examples/suspend.rs | 31 +- src/connect.rs | 534 ++++++++++-------- src/domain.rs | 1323 +++++++++++++++++++++++++-------------------- src/domain_snapshot.rs | 88 +-- src/error.rs | 13 +- src/interface.rs | 38 +- src/lib.rs | 30 +- src/network.rs | 75 +-- src/nodedev.rs | 122 +++-- src/nwfilter.rs | 14 +- src/secret.rs | 79 +-- src/storage_pool.rs | 99 ++-- src/storage_vol.rs | 218 ++++---- src/stream.rs | 27 +- tests/common/mod.rs | 115 ++-- tests/connect.rs | 74 +-- tests/domain.rs | 14 +- tests/integration_qemu.rs | 39 +- tests/interface.rs | 2 - tests/network.rs | 1 - tests/storage_pool.rs | 18 +- 26 files changed, 1717 insertions(+), 1377 deletions(-) Thanks. s. [0] https://github.com/sahid/libvirt-rust/commit/851296786626d61ebec985a9caeb51… [1] https://travis-ci.org/sahid/libvirt-rust/builds/642931039

3 2

[PATCH v2 0/7] qemu: block iotune group fixes/improvements
by Michal Privoznik 29 Jan '20

29 Jan '20

This series was originally posted by Nikolay here: https://www.redhat.com/archives/libvir-list/2020-January/msg00283.html I've delayed review too much. Sorry for that. To make it up to you, I've fixed some patches and resending the series. Basically, I want to know whether you agree with my adjustments. I will add Dan's Reviewed-by when merging. Nikolay Shirokovskiy (7): qemu: Move qemuDiskConfigBlkdeviotuneHas* to conf conf: expand iotune params if only group name is given qemu: propagate iotune settings to all disks in the group qemu: check iotune params same for all disk in group qemu: fix using defaults when setting persistent iotune params qemu: get defaults from iotune group we move disk into qemu: when leaving iotune group update xml properly src/conf/domain_conf.c | 108 +++++++++++++++++++++++++++++++++++++++ src/conf/domain_conf.h | 22 ++++++++ src/libvirt_private.syms | 6 +++ src/qemu/qemu_command.c | 91 ++++++++++++++------------------- src/qemu/qemu_command.h | 3 ++ src/qemu/qemu_driver.c | 94 ++++++++++++++++++++++++++++++++-- src/qemu/qemu_hotplug.c | 2 +- tests/qemublocktest.c | 8 ++- 8 files changed, 275 insertions(+), 59 deletions(-) -- 2.24.1

2 9

[libvirt] [PATCH 0/7] qemu: block iotune group fixes/improvements
by Nikolay Shirokovskiy 29 Jan '20

29 Jan '20

Currently iotune group impl has several bugs/oddities this patchset aims to fix. Also patches (2) and (3) add/change functionality so that group iotune became easier/saner to use. TODO: add docs to API/virsh in respect to using iotune group Nikolay Shirokovskiy (7): conf: refactor virDomainBlockIoTuneInfoHas* conf: expand iotune params if only group name is given (2) qemu: check iotune params same for all disk in group qemu: fix using defaults when setting persistent iotune params qemu: propagate iotune settings to all disks in the group qemu: get defaults from iotune group we move disk into (3) qemu: when leaving iotune group update xml properly src/conf/domain_conf.c | 104 +++++++++++++++++++++++++++++++++++++++ src/conf/domain_conf.h | 16 ++++++ src/libvirt_private.syms | 5 ++ src/qemu/qemu_command.c | 90 ++++++++++++++------------------- src/qemu/qemu_command.h | 3 ++ src/qemu/qemu_driver.c | 98 ++++++++++++++++++++++++++++++++++-- src/qemu/qemu_hotplug.c | 2 +- tests/qemublocktest.c | 2 +- 8 files changed, 262 insertions(+), 58 deletions(-) -- 2.23.0

3 17

[libvirt PATCH 00/10] Various cleanups (virtio-fs epopee)
by Ján Tomko 29 Jan '20

29 Jan '20

2 22

[PATCH v2 0/4] lxc: Add suport to network indexes for LXC 3.X
by Julio Faracco 28 Jan '20

28 Jan '20

This series implement support for network syntax settings for LXC 3.X. Now, indexes are part of the settings to define network interfaces. Each interface has its own index. The old style uses 'type' tag to differentiate each interface and with this, it could use one single structure. Now, network interfaces are a single array with interface pointers. Old: lxc.network.type = veth lxc.network.flags = up lxc.network.link = virbr0 New: lxc.net.0.type = veth lxc.net.0.flags = up lxc.net.0.link = virbr0 v1-v2: Moving sscanf to virStrToLong_ull according Daniel's suggestion. Julio Faracco (4): lxc: refactor lxcNetworkParseData pointers to use new structures lxc: add LXC version 3 network parser lxc: remove domain definition from lxc network struct tests: update LXC config dataset to support V3 indexes src/lxc/lxc_native.c | 189 ++++++++++++------ .../lxcconf2xml-ethernet-v3.config | 16 +- .../lxcconf2xml-fstab-v3.config | 10 +- .../lxcconf2xml-macvlannetwork-v3.config | 10 +- .../lxcconf2xml-simple-v3.config | 18 +- 5 files changed, 151 insertions(+), 92 deletions(-) -- 2.20.1

1 4