aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation
diff options
context:
space:
mode:
authorJavi Merino <cibervicho@gmail.com>2011-01-31 23:11:36 +0000
committerIngo Molnar <mingo@elte.hu>2011-02-01 09:07:13 +0100
commit1e1dbb259c79b38a542c1c4c00fd8dfe936b183b (patch)
treeae240f978095a095253f93848cce0d3c9733706c /Documentation
parentebf53826e105f488f4f628703a108e98940d1dc5 (diff)
downloadkernel_samsung_crespo-1e1dbb259c79b38a542c1c4c00fd8dfe936b183b.zip
kernel_samsung_crespo-1e1dbb259c79b38a542c1c4c00fd8dfe936b183b.tar.gz
kernel_samsung_crespo-1e1dbb259c79b38a542c1c4c00fd8dfe936b183b.tar.bz2
sched, docs: Update schedstats documentation to version 15
Version 15 of schedstats was introduced in: 67aa0f767af4: sched: remove unused fields from struct rq and removed three unused counters in sched_yield(). Update the documentation. Signed-off-by: Javi Merino <cibervicho@gmail.com> Cc: henrix@sapo.pt Cc: rdunlap@xenotime.net Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1296515496-8229-1-git-send-email-cibervicho@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'Documentation')
-rw-r--r--Documentation/scheduler/sched-stats.txt33
1 files changed, 15 insertions, 18 deletions
diff --git a/Documentation/scheduler/sched-stats.txt b/Documentation/scheduler/sched-stats.txt
index 01e6940..1cd5d51 100644
--- a/Documentation/scheduler/sched-stats.txt
+++ b/Documentation/scheduler/sched-stats.txt
@@ -1,3 +1,7 @@
+Version 15 of schedstats dropped counters for some sched_yield:
+yld_exp_empty, yld_act_empty and yld_both_empty. Otherwise, it is
+identical to version 14.
+
Version 14 of schedstats includes support for sched_domains, which hit the
mainline kernel in 2.6.20 although it is identical to the stats from version
12 which was in the kernel from 2.6.13-2.6.19 (version 13 never saw a kernel
@@ -28,32 +32,25 @@ to write their own scripts, the fields are described here.
CPU statistics
--------------
-cpu<N> 1 2 3 4 5 6 7 8 9 10 11 12
-
-NOTE: In the sched_yield() statistics, the active queue is considered empty
- if it has only one process in it, since obviously the process calling
- sched_yield() is that process.
+cpu<N> 1 2 3 4 5 6 7 8 9
-First four fields are sched_yield() statistics:
- 1) # of times both the active and the expired queue were empty
- 2) # of times just the active queue was empty
- 3) # of times just the expired queue was empty
- 4) # of times sched_yield() was called
+First field is a sched_yield() statistic:
+ 1) # of times sched_yield() was called
Next three are schedule() statistics:
- 5) # of times we switched to the expired queue and reused it
- 6) # of times schedule() was called
- 7) # of times schedule() left the processor idle
+ 2) # of times we switched to the expired queue and reused it
+ 3) # of times schedule() was called
+ 4) # of times schedule() left the processor idle
Next two are try_to_wake_up() statistics:
- 8) # of times try_to_wake_up() was called
- 9) # of times try_to_wake_up() was called to wake up the local cpu
+ 5) # of times try_to_wake_up() was called
+ 6) # of times try_to_wake_up() was called to wake up the local cpu
Next three are statistics describing scheduling latency:
- 10) sum of all time spent running by tasks on this processor (in jiffies)
- 11) sum of all time spent waiting to run by tasks on this processor (in
+ 7) sum of all time spent running by tasks on this processor (in jiffies)
+ 8) sum of all time spent waiting to run by tasks on this processor (in
jiffies)
- 12) # of timeslices run on this cpu
+ 9) # of timeslices run on this cpu
Domain statistics