Libvirt provides a portable, long term stable C API for managing the virtualization technologies provided by many operating systems. It includes support for QEMU, KVM, Xen, LXC, bhyve, Virtuozzo, VMware vCenter and ESX, VMware Desktop, Hyper-V, VirtualBox and the POWER Hypervisor.
Go to file
Michal Privoznik 3ec6d586bc qemu: Start emulator thread with more generous cpuset.mems
Consider a domain with two guest NUMA nodes and the following
<numatune/> setting :

  <numatune>
    <memory mode="strict" nodeset="0"/>
    <memnode cellid="0" mode="strict" nodeset="1"/>
  </numatune>

What this means is the emulator thread is pinned onto host NUMA
node #0 (by setting corresponding cpuset.mems to "0"), and two
memory-backend-* objects are created:

  -object '{"qom-type":"memory-backend-ram","id":"ram-node0", .., "host-nodes":[1],"policy":"bind"}' \
  -numa node,nodeid=0,cpus=0-1,memdev=ram-node0 \
  -object '{"qom-type":"memory-backend-ram","id":"ram-node1", .., "host-nodes":[0],"policy":"bind"}' \
  -numa node,nodeid=1,cpus=2-3,memdev=ram-node1 \

Note, the emulator thread is pinned well before QEMU is even
exec()-ed.

Now, the way memory allocation works in QEMU is: the emulator
thread calls mmap() followed by mbind() (which is sane, that's
how everybody should do it). BUT, because the thread is already
restricted by CGroups to just NUMA node #0, calling:

  mbind(host-nodes:[1]); /* made up syntax (TM) */

fails. This is expected though. Kernel was instructed to place
the memory at NUMA node "0" and yet, process is trying to place
it elsewhere.

We used to solve this by not restricting emulator thread at all
initially, and only after it's done initializing (i.e. we got the
QMP greeting) we placed it onto desired nodes. But this had its
own problems (e.g. QEMU might have locked pieces of its memory
which were then unable to migrate onto different NUMA nodes).

Therefore, in v5.1.0-rc1~282 we've changed this and set cgroups
upfront (even before exec()-ing QEMU). And this used to work, but
something has changed (I can't really put my finger on it).

Therefore, for the initialization start the thread with union of
all configured host NUMA nodes ("0-1" in our example) and fix the
placement only after QEMU is started.

NB, the memory hotplug suffers the same problem, but that will
be fixed in the next commit.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2138150
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
2023-05-23 17:21:16 +02:00
.ctags.d maint: Add support for .ctags.d 2019-05-31 17:54:28 +02:00
.github/workflows Switch to new GitHub repo-lockdown configuration 2021-10-20 14:27:20 +01:00
.gitlab/issue_templates gitlab: issue_template: Remove labelling commands 2022-06-01 12:27:10 +02:00
build-aux tests: Introduce viracpitest 2023-04-06 12:48:22 +02:00
ci ci: integration: Flip QEMU upstream integration tests to Fedora 38 2023-05-10 16:41:21 +02:00
docs Revert "conf: Introduce MTE domain feature" 2023-05-22 15:13:19 +02:00
examples Use G_N_ELEMENTS() more 2023-03-13 13:29:07 +01:00
include Add public API for parallel compression method 2023-05-18 15:47:30 +02:00
po Translated using Weblate (Hungarian) 2023-05-12 20:21:07 +02:00
scripts docs: xsl: Simplify templating XSL 2023-05-18 10:06:51 +02:00
src qemu: Start emulator thread with more generous cpuset.mems 2023-05-23 17:21:16 +02:00
subprojects Move src/keycodemapdb -> subprojects/keycodemapdb 2023-04-17 15:02:38 +02:00
tests cpu-data.py: Filter out apic current logical processor 2023-05-23 16:25:12 +02:00
tools virsh: Add migrate options to set parallel compress level 2023-05-18 15:47:30 +02:00
.ctags ctags: Generate tags for headers, i.e. function prototypes 2018-09-18 14:21:33 +02:00
.dir-locals.el build: avoid tabs that failed syntax-check 2012-09-06 09:43:46 -06:00
.editorconfig Add .editorconfig 2019-09-06 12:47:46 +02:00
.gitattributes Add .gitattributes file 2022-03-17 14:33:12 +01:00
.gitignore Revert ".gitignore: Ignore cscope and other *tags files" 2023-02-08 17:24:31 +01:00
.gitlab-ci.yml .gitlab-ci.yml: Fix Leap 15.4 codestyle job 2023-05-09 08:42:09 +02:00
.gitmodules Move src/keycodemapdb -> subprojects/keycodemapdb 2023-04-17 15:02:38 +02:00
.gitpublish gitpublish: add a subject prefix 2020-01-16 13:04:11 +00:00
.mailmap mailmap: consolidate my email addresses 2020-10-06 12:05:09 +02:00
AUTHORS.rst.in AUTHORS: change my (Nikolay Shirokovskiy) email 2022-04-06 11:00:53 +03:00
config.h configure: bump min required CLang to 6.0 / XCode 10.0 2022-01-17 10:44:29 +00:00
configmake.h.in meson: generate configmake.h 2020-08-03 09:26:48 +02:00
CONTRIBUTING.rst meson: adjust our documentation to mention meson instead of autoconf 2020-08-03 09:27:09 +02:00
COPYING maint: follow recommended practice for using LGPL 2013-05-20 14:15:21 -06:00
COPYING.LESSER maint: Remove control characters from LGPL license file 2015-09-25 09:16:24 +02:00
gitdm.config gitdm: add 'ibm' file 2019-10-18 17:32:52 +02:00
libvirt-admin.pc.in Add libvirt-admin library 2015-06-16 13:46:20 +02:00
libvirt-lxc.pc.in Add pkg-config files for libvirt-qemu & libvirt-lxc 2014-06-23 16:17:27 +01:00
libvirt-qemu.pc.in Add pkg-config files for libvirt-qemu & libvirt-lxc 2014-06-23 16:17:27 +01:00
libvirt.pc.in Add pkg-config files for libvirt-qemu & libvirt-lxc 2014-06-23 16:17:27 +01:00
libvirt.spec.in rpm: Explain BuildRequires on qemu-img 2023-05-22 15:12:29 +02:00
meson_options.txt meson: Introduce initconfdir option 2023-05-05 15:08:19 +02:00
meson.build meson: Stop looking for dbus-daemon 2023-05-11 15:04:54 +02:00
NEWS.rst NEWS: Mention support for compressing parallel migration 2023-05-23 12:32:10 +02:00
README.rst README: drop Travis CI badge 2020-08-03 15:08:28 +02:00
run.in run: gracefully handle SIGHUP, SIGQUIT, SIGTERM 2022-03-10 08:06:12 +00:00

GitLab CI Build Status

CII Best Practices

Translation status

Libvirt API for virtualization

Libvirt provides a portable, long term stable C API for managing the virtualization technologies provided by many operating systems. It includes support for QEMU, KVM, Xen, LXC, bhyve, Virtuozzo, VMware vCenter and ESX, VMware Desktop, Hyper-V, VirtualBox and the POWER Hypervisor.

For some of these hypervisors, it provides a stateful management daemon which runs on the virtualization host allowing access to the API both by non-privileged local users and remote users.

Layered packages provide bindings of the libvirt C API into other languages including Python, Perl, PHP, Go, Java, OCaml, as well as mappings into object systems such as GObject, CIM and SNMP.

Further information about the libvirt project can be found on the website:

https://libvirt.org

License

The libvirt C API is distributed under the terms of GNU Lesser General Public License, version 2.1 (or later). Some parts of the code that are not part of the C library may have the more restrictive GNU General Public License, version 2.0 (or later). See the files COPYING.LESSER and COPYING for full license terms & conditions.

Installation

Instructions on building and installing libvirt can be found on the website:

https://libvirt.org/compiling.html

Contributing

The libvirt project welcomes contributions in many ways. For most components the best way to contribute is to send patches to the primary development mailing list. Further guidance on this can be found on the website:

https://libvirt.org/contribute.html

Contact

The libvirt project has two primary mailing lists:

Further details on contacting the project are available on the website:

https://libvirt.org/contact.html