docs/scheduler: fix unit error

The unit mentioned in the documentation of scheduler statistics is
outdated which may mislead the readers.

The unit of statistics that is reported by /proc/schedstat is modified
to nanosecond, and the unit of statistics that is reported by
/proc/PID/schedstat is provided as well to make the context consistent.

The rq_cpu_time and the rq_sched_info.run_delay of a run queue, and the
sched_info.run_delay of a task are all updated based on the clock of the
run queue, while the se.sum_exec_runtime of a task is updated based on
the clock_task of the run queue of the task. Both the clock and
clock_task are relied on the return value of the function sched_clock()
which is in the unit of nanosecond.

Signed-off-by: Jui-Tse Huang <juitse.huang@gmail.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>

diff --git a/Documentation/scheduler/sched-stats.rst b/Documentation/scheduler/sched-stats.rst
index 9078a27a4137abcfab25fa16e329c449197361e2..03c06291599800ecb86f9c9456245f7d1ba5ded4 100644 (file)
--- a/Documentation/scheduler/sched-stats.rst
+++ b/Documentation/scheduler/sched-stats.rst
@@ -155,8 +155,8 @@ schedstats also adds a new /proc/<pid>/schedstat file to include some of
 the same information on a per-process level.  There are three fields in
 this file correlating for that process to:
 
-     1) time spent on the cpu
-     2) time spent waiting on a runqueue
+     1) time spent on the cpu (in nanoseconds)
+     2) time spent waiting on a runqueue (in nanoseconds)
      3) # of timeslices run on this cpu
 
 A program could be easily written to make use of these extra fields to