News for hackers. Stuff that matters.

A comment on AMD EPYC Performance Testing with systemd.

Re:AMD EPYC Performance Testing with systemd by viraptor on Thursday July 12, 2018 @04:55PM (#17518518)
Most likely the change is from cgroups, and not systemd itself. This could be verified by booting with and without cgroup_disable=cpu.
[ Parent | on HN ]
Re:AMD EPYC Performance Testing with systemd by zlynx on Thursday July 12, 2018 @05:14PM (#17518666)
And running systemd-cgtop might help see what's going on.
[ Parent | on HN ]
Re:AMD EPYC Performance Testing with systemd by v4tk on Sunday July 15, 2018 @05:18PM (#17537136)
cgroup_disable=cpu and as matter of fact cgroup_disable= all cgroups improves things (adds extra 12% for bash), but not to the full extent.
[ Parent | on HN ]