- Devel - libvirt List Archives

[libvirt] [PATCH v5 00/15] CPU Model Baseline and Comparison for s390x
by Collin Walling 07 Oct '19

07 Oct '19

Note: since I've made some changes to a lot of these patches / split up some patches, I've decided to hold off on adding any r-b's in case there is a specific change that someone does not agree with. Changelog: - Properly refactored code from CPU model expansion function - Introduced a cleanup patch for CPU model expansion function - Introduced patches that modifies the refactored code to suit needs for baseline/comparison - CPU expansion function now accepts a virCPUDefPtr - Removed props parsing from CPU model comparison (they weren't used) - Cleaner error reporting when baselining/comparing with erroneous CPU models / features - Various cleanups based on feedback ___ To run these patches, execute the virsh hypervisor-cpu-compare or hypervisor-cpu-baseline commands and pass an XML file describing one or more CPU definition. You can use the definition from virsh domcapabilities or from a guest XML. There is no need extract it from the file and place it a new one, as the XML parser will look specifically for the CPU tags. ___ These patches hookup the virsh hypervisor-cpu-compare/baseline commands for the s390x architecture. They take an XML file describing some CPU definitions and passes the data to QEMU, where the actual CPU model comparison / baseline calculation is handled (available since QEMU 2.8.5). When baselining CPU models with the --features argument, s390x will report a full CPU model expansion. Thanks. Collin Walling (15): qemu_monitor: refactor cpu model expansion qemu_monitor: expansion cleanups qemu_monitor: use cpu def instead of char for expansion qemu_monitor: add features to CPU model for QMP command qemu_monitor: allow cpu props to be optional qemu_monitor: make qemuMonitorJSONParseCPUModelData command-agnostic qemu_monitor: implement query-cpu-model-baseline qemu_capabilities: introduce QEMU_CAPS_QUERY_CPU_MODEL_BASELINE qemu_driver: hook up query-cpu-model-baseline qemu_driver: expand cpu features after baseline qemu_monitor: implement query-cpu-model-comparison qemu_capabilities: introduce QEMU_CAPS_QUERY_CPU_MODEL_COMPARISON cpu_conf: xml to cpu definition parse helper qemu_driver: hook up query-cpu-model-comparison qemu_driver: improve comparison/baseline error reporting src/conf/cpu_conf.c | 29 +++ src/conf/cpu_conf.h | 5 + src/cpu/cpu.c | 14 +- src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 21 +- src/qemu/qemu_capabilities.h | 4 + src/qemu/qemu_driver.c | 215 +++++++++++++++++ src/qemu/qemu_monitor.c | 39 +++- src/qemu/qemu_monitor.h | 13 +- src/qemu/qemu_monitor_json.c | 279 +++++++++++++++++------ src/qemu/qemu_monitor_json.h | 17 +- tests/cputest.c | 11 +- tests/qemucapabilitiesdata/caps_2.10.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_2.11.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_2.12.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_2.8.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_2.9.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_3.0.0.s390x.xml | 2 + tests/qemucapabilitiesdata/caps_4.0.0.s390x.xml | 2 + 19 files changed, 573 insertions(+), 89 deletions(-) -- 2.7.4

3 33

[libvirt] [PATCH] maint: Post-release version bump to 5.9.0
by Peter Krempa 07 Oct '19

07 Oct '19

Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- configure.ac | 2 +- docs/news.xml | 8 ++++++++ 2 files changed, 9 insertions(+), 1 deletion(-) Pushed now. diff --git a/configure.ac b/configure.ac index af8cbcdfd8..5269db425c 100644 --- a/configure.ac +++ b/configure.ac @@ -16,7 +16,7 @@ dnl You should have received a copy of the GNU Lesser General Public dnl License along with this library. If not, see dnl <http://www.gnu.org/licenses/>. -AC_INIT([libvirt], [5.8.0], [libvir-list(a)redhat.com] [], [https://libvirt.org] +AC_INIT([libvirt], [5.9.0], [libvir-list(a)redhat.com] [], [https://libvirt.org] AC_CONFIG_SRCDIR([src/libvirt.c]) AC_CONFIG_AUX_DIR([build-aux]) AC_CONFIG_HEADERS([config.h]) diff --git a/docs/news.xml b/docs/news.xml index 70c325795e..9fd8f65515 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -39,6 +39,14 @@ --> <libvirt> + <release version="v5.9.0" date="unreleased"> + <section title="New features"> + </section> + <section title="Improvements"> + </section> + <section title="Bug fixes"> + </section> + </release> <release version="v5.8.0" date="2019-10-05"> <section title="Removed features"> <change> -- 2.21.0

1 0

[libvirt] libvirt? qemu change that mmaps ELF files breaks libvirt svirt handling for os.kernel
by Christian Borntraeger 04 Oct '19

04 Oct '19

Stefano, Paolo, I have an interesting fail in QEMU 2019-10-04T12:00:32.675188Z qemu-system-s390x: GLib: g_mapped_file_unref: assertion 'file != NULL' failed that bisected to commit 816b9fe450220e19acb91a0ce4a8ade7000648d1 (refs/bisect/bad) elf-ops.h: Map into memory the ELF to load strace tells that I can read the ELF file, but not mmap strace: 214365 openat(AT_FDCWD, "/var/lib/libvirt/images/test_cpu_timer.elf", O_RDONLY) = 36 214365 read(46, "\177ELF\2\2\1\0\0\0\0\0\0\0\0\0", 16) = 16 214365 lseek(46, 0, SEEK_SET) = 0 [...] 214365 fstat(46, {st_mode=S_IFREG|0755, st_size=168176, ...}) = 0 214365 mmap(NULL, 168176, PROT_READ|PROT_WRITE, MAP_PRIVATE, 46, 0) = -1 EACCES (Permission denied) So reading from /var/lib/libvirt/images/test_cpu_timer.elf does work, mmaping does not. setenforce 0 makes the problem go away. This might be more of an issue in libvirt, setting the svirt context too restrictive, but I am not too deep into the svirt part of libvirt. Reverting the qemu commit makes the problem go away.

3 5

[libvirt] [PATCH 00/11] Integrate usage of glib into libvirt
by Daniel P. Berrangé 04 Oct '19

04 Oct '19

This is a followup to a previous patch series: https://www.redhat.com/archives/libvir-list/2019-August/msg01374.html The first abort-on-oom parts of that series merged already. As well as fixing the issues mentioned last time, the glib parts now do a little more: - Demonstrate conversion of virObject to GObject - Convert to gbase64 APIs - Start to convert getopt to GOptionContext I have explicitly now confirmed we can freely mix g_malloc/malloc and g_free/free, given our min glib version. my intention with the glib code will be to focus on converting bits of code that allow us to eliminate gnulib modules. About 50% of the gnulib stuff is related to the Windows portability for sockets() and poll(). We can address this by integrate the event loop with GMainLoop, and using GIO for its GSocket APIs. The other gnulib stuff is a random bag of APIs. Some may be replaced by glib APIs, for others we can pull the gnulib fix straight into libvirt, for others the portability problems might be obsolete. Daniel P. Berrangé (11): build: probe for glib-2 library in configure build: link to glib, gobject, gio libraries remote: don't pull anonymous enums into rpc protocol structs util: use glib memory allocation functions util: use glib string allocation/formatting functions util: use glib base64 encoding/decoding APIs util: convert virIdentity class to use GObject src: convert over to use GRegex for regular exprssions virsh: convert command line parsing to use GOptionContext virt-admin: convert command line parsing to use GOptionContext virt-login-shell: convert command line parsing to use GOptionContext bootstrap.conf | 5 - configure.ac | 2 + libvirt.spec.in | 1 + m4/virt-glib.m4 | 36 ++++ mingw-libvirt.spec.in | 2 + src/Makefile.am | 5 +- src/access/viraccessdriverpolkit.c | 21 +- src/admin/admin_server.c | 3 +- src/admin_protocol-structs | 9 - src/conf/domain_event.c | 25 +-- src/conf/virsecretobj.c | 38 +--- src/internal.h | 1 + src/libvirt_private.syms | 1 - src/libxl/libxl_capabilities.c | 44 ++-- src/libxl/libxl_conf.c | 3 +- src/lxc/Makefile.inc.am | 2 + src/qemu/qemu_agent.c | 9 +- src/qemu/qemu_command.c | 5 +- src/qemu/qemu_domain.c | 8 +- src/qemu/qemu_process.c | 4 +- src/remote/Makefile.inc.am | 1 + src/remote/remote_daemon.c | 3 +- src/remote/remote_daemon_dispatch.c | 35 ++-- src/remote_protocol-structs | 9 - src/rpc/virnetserverclient.c | 57 +++--- src/rpc/virnetserverprogram.c | 13 +- src/secret/secret_driver.c | 1 - src/storage/storage_backend_rbd.c | 4 +- src/util/Makefile.inc.am | 1 + src/util/viralloc.c | 29 +-- src/util/viridentity.c | 87 ++++---- src/util/viridentity.h | 7 +- src/util/virstring.c | 40 +--- src/util/virstring.h | 2 - tests/Makefile.am | 3 +- tests/viridentitytest.c | 45 ++--- tests/virnetserverclienttest.c | 3 +- tools/Makefile.am | 1 + tools/virsh-secret.c | 17 +- tools/virsh.c | 303 +++++++++++++--------------- tools/virt-admin.c | 207 +++++++++---------- tools/virt-login-shell-helper.c | 66 ++---- 42 files changed, 495 insertions(+), 663 deletions(-) create mode 100644 m4/virt-glib.m4 -- 2.21.0

6 56

[libvirt] [jenkins-ci PATCH] Use all_machines whenever possible
by Andrea Bolognani 04 Oct '19

04 Oct '19

The reasons we had to prevent some targets from building libvirt-dbus and libvirt-ocaml in the past clearly no longer apply, since we're back to building them on all targets. Drop the explicit target list and use all_machines instead. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- Pushed as trivial. guests/playbooks/build/projects/libvirt-dbus.yml | 14 +------------- guests/playbooks/build/projects/libvirt-ocaml.yml | 14 +------------- jenkins/projects/libvirt-dbus.yaml | 10 +--------- jenkins/projects/libvirt-ocaml.yaml | 10 +--------- 4 files changed, 4 insertions(+), 44 deletions(-) diff --git a/guests/playbooks/build/projects/libvirt-dbus.yml b/guests/playbooks/build/projects/libvirt-dbus.yml index a6ff025..218bcc1 100644 --- a/guests/playbooks/build/projects/libvirt-dbus.yml +++ b/guests/playbooks/build/projects/libvirt-dbus.yml @@ -1,19 +1,7 @@ --- - set_fact: name: libvirt-dbus - machines: - - libvirt-centos-7 - - libvirt-debian-9 - - libvirt-debian-10 - - libvirt-debian-sid - - libvirt-fedora-29 - - libvirt-fedora-30 - - libvirt-fedora-rawhide - - libvirt-freebsd-11 - - libvirt-freebsd-12 - - libvirt-freebsd-current - - libvirt-ubuntu-16 - - libvirt-ubuntu-18 + machines: '{{ all_machines }}' archive_format: xz git_url: '{{ git_urls["libvirt-dbus"][git_remote] }}' diff --git a/guests/playbooks/build/projects/libvirt-ocaml.yml b/guests/playbooks/build/projects/libvirt-ocaml.yml index 58f017b..4202835 100644 --- a/guests/playbooks/build/projects/libvirt-ocaml.yml +++ b/guests/playbooks/build/projects/libvirt-ocaml.yml @@ -1,19 +1,7 @@ --- - set_fact: name: libvirt-ocaml - machines: - - libvirt-centos-7 - - libvirt-debian-9 - - libvirt-debian-10 - - libvirt-debian-sid - - libvirt-fedora-29 - - libvirt-fedora-30 - - libvirt-fedora-rawhide - - libvirt-freebsd-11 - - libvirt-freebsd-12 - - libvirt-freebsd-current - - libvirt-ubuntu-16 - - libvirt-ubuntu-18 + machines: '{{ all_machines }}' archive_format: xz git_url: '{{ git_urls["libvirt-ocaml"][git_remote] }}' diff --git a/jenkins/projects/libvirt-dbus.yaml b/jenkins/projects/libvirt-dbus.yaml index 2b72fb4..f7a52ef 100644 --- a/jenkins/projects/libvirt-dbus.yaml +++ b/jenkins/projects/libvirt-dbus.yaml @@ -1,15 +1,7 @@ --- - project: name: libvirt-dbus - machines: - - libvirt-centos-7 - - libvirt-debian-9 - - libvirt-debian-10 - - libvirt-fedora-29 - - libvirt-fedora-30 - - libvirt-fedora-rawhide - - libvirt-freebsd-11 - - libvirt-freebsd-12 + machines: '{all_machines}' title: Libvirt D-Bus archive_format: xz git_url: '{git_urls[libvirt-dbus][default]}' diff --git a/jenkins/projects/libvirt-ocaml.yaml b/jenkins/projects/libvirt-ocaml.yaml index 0c819cc..4a9bb65 100644 --- a/jenkins/projects/libvirt-ocaml.yaml +++ b/jenkins/projects/libvirt-ocaml.yaml @@ -1,15 +1,7 @@ --- - project: name: libvirt-ocaml - machines: - - libvirt-centos-7 - - libvirt-debian-9 - - libvirt-debian-10 - - libvirt-fedora-29 - - libvirt-fedora-30 - - libvirt-fedora-rawhide - - libvirt-freebsd-11 - - libvirt-freebsd-12 + machines: '{all_machines}' title: Libvirt OCaml archive_format: xz git_url: '{git_urls[libvirt-ocaml][default]}' -- 2.21.0

1 0

[libvirt] [PATCH] virsh: fixed handling of targetless disks (e.g. empty CDROM) in 'domblkinfo' cmd
by Pavel Mores 04 Oct '19

04 Oct '19

virDomainGetBlockInfo() returns error if called on a disk with no target (a targetless disk might be a removable media drive with no media in it, for instance an empty CDROM drive). So far this caused the virsh domblkinfo --all command to abort and ignore any remaining (not yet displayed) disk devices. This patch fixes it by ignoring virDomainGetBlockInfo() errors for CDROM and floppy drives, similar to how it's done for network drives. https://bugzilla.redhat.com/show_bug.cgi?id=1619625 Signed-off-by: Pavel Mores <pmores(a)redhat.com> --- tools/virsh-domain-monitor.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 0e2c4191d7..0f495c1a3f 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -473,6 +473,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) char *cap = NULL; char *alloc = NULL; char *phy = NULL; + char *device_type = NULL; vshTablePtr table = NULL; if (!(dom = virshCommandOptDomain(ctl, cmd, NULL))) @@ -510,6 +511,8 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) rc = virDomainGetBlockInfo(dom, target, &info, 0); if (rc < 0) { + device_type = virXPathString("string(./@device)", ctxt); + /* If protocol is present that's an indication of a networked * storage device which cannot provide statistics, so generate * 0 based data and get the next disk. */ @@ -518,9 +521,16 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) virGetLastErrorDomain() == VIR_FROM_STORAGE) { memset(&info, 0, sizeof(info)); vshResetLibvirtError(); + } else if (device_type != NULL && + (STRCASEEQ(device_type, "cdrom") || + STRCASEEQ(device_type, "floppy"))) { + memset(&info, 0, sizeof(info)); + vshResetLibvirtError(); } else { goto cleanup; } + + VIR_FREE(device_type); } if (!cmdDomblkinfoGet(ctl, &info, &cap, &alloc, &phy, human)) @@ -556,6 +566,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) VIR_FREE(target); VIR_FREE(protocol); VIR_FREE(disks); + VIR_FREE(device_type); xmlXPathFreeContext(ctxt); xmlFreeDoc(xmldoc); return ret; -- 2.21.0

3 10

[libvirt] Entering freeze for libvirt-5.8.0
by Daniel Veillard 03 Oct '19

03 Oct '19

per previous mail, apologies, I totally forgot. It's tagged in git, I have pushed RC1 signed tarball and source rpm to the usual place at: https://libvirt.org/sources/ I will reckon my testing was even more minimal than usual, I will just say it built for me ! https://ci.centos.org/view/libvirt/ is green with a small exception, so apparently it's not too broken, but please give it a try ! I will push rc2 on Thursday, and if everything is fine will do the release on Saturday, a bit late but hopefully not a big deal ! thanks, Daniel -- Daniel Veillard | Red Hat Developers Tools http://developer.redhat.com/ veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 1

[libvirt] [PATCH 00/11] build: eliminate many redundant/obsolete gnulib modules
by Daniel P. Berrangé 03 Oct '19

03 Oct '19

The modues in this series are either obsolete (problems now fixed), or redundant (fixing problems we never cared about - hello AIX portability), or trivially replaced. Daniel P. Berrangé (11): bhyve: stop using private gnulib _getopt_internal_r func build: drop the getopt-posix gnulib module build: drop the inet_pton gnulib module build: drop the mktempd gnulib module util: drop the stpcpy gnulib module Revert "build: use autobuild module to make build logs nicer" build: drop the usleep gnulib module build: drop the stdarg gnulib module build: drop the gitlog-to-changelog gnulib module build: drop the perror gnulib module build: drop the pthread gnulib module Makefile.am | 1 - bootstrap.conf | 10 ---- configure.ac | 6 --- po/POTFILES | 1 - src/bhyve/bhyve_parse_command.c | 88 +++++++++++++++++---------------- src/util/virerror.c | 5 +- tests/test-lib.sh | 3 +- 7 files changed, 50 insertions(+), 64 deletions(-) -- 2.21.0

2 24

[libvirt] [RFC] cpu_map: Remove pconfig from Icelake-Server CPU model
by Eduardo Habkost 03 Oct '19

03 Oct '19

The pconfig feature never worked, and adding "pconfig=off" to the QEMU command-line triggers a regression in QEMU 3.1.1 and 4.0.0. Signed-off-by: Eduardo Habkost <ehabkost(a)redhat.com> --- I'm sending this as an RFC because I couldn't test it properly, and because I don't know what are the consequences of changing cpu_map between libvirt versions. --- src/cpu_map/x86_Icelake-Server.xml | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/cpu_map/x86_Icelake-Server.xml b/src/cpu_map/x86_Icelake-Server.xml index fb15977a59..188a781282 100644 --- a/src/cpu_map/x86_Icelake-Server.xml +++ b/src/cpu_map/x86_Icelake-Server.xml @@ -56,7 +56,9 @@ <feature name='pat'/> <feature name='pcid'/> <feature name='pclmuldq'/> - <feature name='pconfig'/> +  <feature name='pdpe1gb'/> <feature name='pge'/> <feature name='pku'/> -- 2.21.0

5 8

[libvirt] [jenkins-ci PATCH 00/17] Meson, more meson!
by Fabiano Fidêncio 03 Oct '19

03 Oct '19

This patch series is fixing a few problems at the very same time. Would be better to split it and have a few series depending on each other? Maybe! Okay, the very first patch fixes teaches libvirt-jenkins-ci to deal with multilib path in systems where it matters as Meson uses it by default. The next 12 patches are basically adding the whole machinery we'll need to perform meson builds on OSes where we do not have Meson packaged at all or the Meson version is older than the minimum required by our projects which already adopted it (0.49.0). Next 3 patches build and test osinfo-db-tools, osinfo-db, and libosinfo on all supported machines as those now have the minimum requirements needed. Last patch in the series does the libvirt-dbus builds switch to using meson instead of autotools. Fabiano Fidêncio (17): guests: Deal with multilib path guests: CentOS7 provides python3 guests: Install EPEL on CentOS7 guests: Map python36-* CentOS7 packages mappings: Mark the OSes where meson is not supported guests: Add meson to the base project mappings: Add ninja guests: Add ninja to the base project Deal with ninja x ninja-build bin names guests: Keep base project's packages sorted mappings: Add python3-pip guests: Add pip to the base project guests: Install meson via pip osinfo-db-tools: Build again on all machines libosinfo: Build again on all machines osinfo-db: Build again on all machines Switch libvirt-dbus builds to meson guests/host_vars/libvirt-centos-7/main.yml | 2 ++ guests/host_vars/libvirt-debian-9/main.yml | 2 ++ guests/host_vars/libvirt-ubuntu-16/main.yml | 2 ++ guests/host_vars/libvirt-ubuntu-18/main.yml | 2 ++ guests/lcitool | 13 +++++++++ .../playbooks/build/jobs/meson-build-job.yml | 4 +-- .../playbooks/build/jobs/meson-check-job.yml | 2 +- guests/playbooks/build/jobs/meson-rpm-job.yml | 2 +- guests/playbooks/build/projects/libosinfo.yml | 12 +-------- .../playbooks/build/projects/libvirt-dbus.yml | 23 +++------------- .../build/projects/osinfo-db-tools.yml | 12 +-------- guests/playbooks/build/projects/osinfo-db.yml | 14 +--------- guests/playbooks/update/tasks/base.yml | 7 +++++ guests/playbooks/update/tasks/packages.yml | 8 ++++++ guests/playbooks/update/tasks/paths.yml | 15 +++++++++++ guests/playbooks/update/templates/bashrc.j2 | 18 +++++++++++++ guests/vars/mappings.yml | 27 ++++++++++++++----- guests/vars/projects/base.yml | 5 +++- guests/vars/projects/libosinfo+mingw32.yml | 1 - guests/vars/projects/libosinfo+mingw64.yml | 1 - guests/vars/projects/libosinfo.yml | 1 - .../vars/projects/osinfo-db-tools+mingw32.yml | 1 - .../vars/projects/osinfo-db-tools+mingw64.yml | 1 - guests/vars/projects/osinfo-db-tools.yml | 1 - jenkins/jobs/meson.yaml | 8 +++--- jenkins/projects/libosinfo.yaml | 8 +----- jenkins/projects/libvirt-dbus.yaml | 18 +++---------- jenkins/projects/osinfo-db-tools.yaml | 10 +------ jenkins/projects/osinfo-db.yaml | 10 +------ 29 files changed, 114 insertions(+), 116 deletions(-) -- 2.23.0

2 35