diff options
| author | Suren Baghdasaryan <surenb@google.com> | 2021-05-24 12:53:39 -0700 |
|---|---|---|
| committer | Tejun Heo <tj@kernel.org> | 2021-06-08 14:59:02 -0400 |
| commit | 3958e2d0c34e18c41b60dc01832bd670a59ef70f (patch) | |
| tree | 429f3434bd8e89fb688e9366c9087208b31fcda3 /Documentation/admin-guide/kernel-parameters.txt | |
| parent | 2ca11b0e043be6f5c2b188897e9a32275eaab046 (diff) | |
cgroup: make per-cgroup pressure stall tracking configurable
PSI accounts stalls for each cgroup separately and aggregates it at each
level of the hierarchy. This causes additional overhead with psi_avgs_work
being called for each cgroup in the hierarchy. psi_avgs_work has been
highly optimized, however on systems with large number of cgroups the
overhead becomes noticeable.
Systems which use PSI only at the system level could avoid this overhead
if PSI can be configured to skip per-cgroup stall accounting.
Add "cgroup_disable=pressure" kernel command-line option to allow
requesting system-wide only pressure stall accounting. When set, it
keeps system-wide accounting under /proc/pressure/ but skips accounting
for individual cgroups and does not expose PSI nodes in cgroup hierarchy.
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Diffstat (limited to 'Documentation/admin-guide/kernel-parameters.txt')
| -rw-r--r-- | Documentation/admin-guide/kernel-parameters.txt | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index cb89dbdedc46..653c62142f07 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -497,16 +497,21 @@ ccw_timeout_log [S390] See Documentation/s390/common_io.rst for details. - cgroup_disable= [KNL] Disable a particular controller - Format: {name of the controller(s) to disable} + cgroup_disable= [KNL] Disable a particular controller or optional feature + Format: {name of the controller(s) or feature(s) to disable} The effects of cgroup_disable=foo are: - foo isn't auto-mounted if you mount all cgroups in a single hierarchy - foo isn't visible as an individually mountable subsystem + - if foo is an optional feature then the feature is + disabled and corresponding cgroup files are not + created {Currently only "memory" controller deal with this and cut the overhead, others just disable the usage. So only cgroup_disable=memory is actually worthy} + Specifying "pressure" disables per-cgroup pressure + stall information accounting feature cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 Format: { { controller | "all" | "named" } |