Skip to content
Snippets Groups Projects
Select Git revision
  • 3e3589c842115e94e291164f038cad8f44a37efc
  • master default protected
  • debian protected
  • pristine-tar protected
  • upstream protected
  • backupninja.conf.d
  • when-override
  • maethor-master-patch-46063
  • maethor-master-patch-70558
  • expand_pruning_options
  • systemd_integration
  • borg-sftp-support
  • nap-initial
  • mariaback_full-intial
  • borg-ssh-keygen
  • borg-custom-init-options
  • stretch-backports
  • backupninja_debian/1.2.2-1
  • backupninja_upstream/1.2.2
  • backupninja-1.2.2
  • backupninja_debian/1.2.1-1
  • backupninja_upstream/1.2.1
  • backupninja-1.2.1
  • backupninja_debian/1.2.0-1
  • backupninja_upstream/1.2.0
  • backupninja-1.2.0
  • backupninja-1.2.0-rc1
  • backupninja_debian/1.1.0-1
  • backupninja_upstream/1.1.0
  • backupninja-1.1.0
  • backupninja_debian/1.0.2-1
  • backupninja_upstream/1.0.2
  • backupninja-1.0.2
  • backupninja_debian/1.0.1-2
  • backupninja_debian/1.0.1-1
  • backupninja_upstream/1.0.1
  • backupninja-1.0.1
37 results

example.makecd

user avatar
maxigas authored and Daniel Kahn Gillmor committed
The capabilities are extensions to the CPU architecture's base instruction set.

New versions of software we care about such as the CentOS installer
requires "x86-64-v2" capabilities because glibc relies on them.

Also, Signal CLI (see #11)
apparently relies on them.

This is not the default setting in kvm in order to avoid issues with
live migration on some virtualization management systems, but
kvm-manager does not support live migration anyway, so this is
unlikely to become a problem.

see
https://www.qemu.org/docs/master/system/qemu-cpu-models.html#two-ways-to-configure-cpu-models-with-qemu-kvm
and
https://developers.redhat.com/blog/2021/01/05/building-red-hat-enterprise-linux-9-for-the-x86-64-v2-microarchitecture-level
for more details.

(dkg fleshed out these details, maxigas and jamie proposed the
concrete change)
466f58b6
History