p71924506
/
linux_old1
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Documentation: scheduler: fix outdated information on sched groups 

The documentation claims that two sched groups must not overlap. This is
no longer true, as overlapping sched groups are used on NUMA systems.
This change has been introduced by commit e3589f6c81 and was
documented by an in-code comment in commit 35a566e6e8.

Signed-off-by: Adrian Freund <adrian@freund.io>
Link: https://lore.kernel.org/r/20200407130525.76663-1-adrian@freund.io
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
This commit is contained in:
Adrian Freund 2020-04-07 15:05:25 +02:00 committed by Jonathan Corbet
parent 5af438d0dc
commit 7b9121040d
1 changed files with 6 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
Documentation/scheduler/sched-domains.rst
View File

@ -19,10 +19,12 @@ CPUs".
Each scheduling domain must have one or more CPU groups (struct sched_group) Each scheduling domain must have one or more CPU groups (struct sched_group)
which are organised as a circular one way linked list from the ->groups which are organised as a circular one way linked list from the ->groups
pointer. The union of cpumasks of these groups MUST be the same as the pointer. The union of cpumasks of these groups MUST be the same as the
domain's span. The intersection of cpumasks from any two of these groups domain's span. The group pointed to by the ->groups pointer MUST contain the CPU
MUST be the empty set. The group pointed to by the ->groups pointer MUST to which the domain belongs. Groups may be shared among CPUs as they contain
contain the CPU to which the domain belongs. Groups may be shared among read only data after they have been set up. The intersection of cpumasks from
CPUs as they contain read only data after they have been set up. any two of these groups may be non empty. If this is the case the SD_OVERLAP
flag is set on the corresponding scheduling domain and its groups may not be
shared between CPUs.
Balancing within a sched domain occurs between groups. That is, each group Balancing within a sched domain occurs between groups. That is, each group
is treated as one entity. The load of a group is defined as the sum of the is treated as one entity. The load of a group is defined as the sum of the