linux.git/fs/proc/array.c, branch v2.6.33

proc: partially revert "procfs: provide stack information for threads"

2010-01-11T17:34:06+00:00

Commit d899bf7b (procfs: provide stack information for threads) introduced
to show stack information in /proc/{pid}/status.  But it cause large
performance regression.  Unfortunately /proc/{pid}/status is used ps
command too and ps is one of most important component.  Because both to
take mmap_sem and page table walk are heavily operation.

If many process run, the ps performance is,

[before d899bf7b]

% perf stat ps >/dev/null

 Performance counter stats for 'ps':

     4090.435806  task-clock-msecs         #      0.032 CPUs
             229  context-switches         #      0.000 M/sec
               0  CPU-migrations           #      0.000 M/sec
             234  page-faults              #      0.000 M/sec
      8587565207  cycles                   #   2099.425 M/sec
      9866662403  instructions             #      1.149 IPC
      3789415411  cache-references         #    926.409 M/sec
        30419509  cache-misses             #      7.437 M/sec

   128.859521955  seconds time elapsed

[after d899bf7b]

% perf stat  ps  > /dev/null

 Performance counter stats for 'ps':

     4305.081146  task-clock-msecs         #      0.028 CPUs
             480  context-switches         #      0.000 M/sec
               2  CPU-migrations           #      0.000 M/sec
             237  page-faults              #      0.000 M/sec
      9021211334  cycles                   #   2095.480 M/sec
     10605887536  instructions             #      1.176 IPC
      3612650999  cache-references         #    839.160 M/sec
        23917502  cache-misses             #      5.556 M/sec

   152.277819582  seconds time elapsed

Thus, this patch revert it. Fortunately /proc/{pid}/task/{tid}/smaps
provide almost same information. we can use it.

Commit d899bf7b introduced two features:

 1) Add the annotattion of [thread stack: xxxx] mark to
    /proc/{pid}/task/{tid}/maps.
 2) Add StackUsage field to /proc/{pid}/status.

I only revert (2), because I haven't seen (1) cause regression.

Signed-off-by: KOSAKI Motohiro 
Cc: Stefani Seibold 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Alexey Dobriyan 
Cc: "Eric W. Biederman" 
Cc: Randy Dunlap 
Cc: Andrew Morton 
Cc: Andi Kleen 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

sched: Assert task state bits at build time

2009-12-17T12:22:45+00:00

Since everybody is lazy and prone to forgetting things, make the
compiler help us a bit.

Signed-off-by: Peter Zijlstra 
LKML-Reference: <20091217121830.060186433@chello.nl>
Signed-off-by: Ingo Molnar

sched: Update task_state_arraypwith new states

2009-12-17T12:22:45+00:00

Neglected because its hidden... (who reads comments anyway)

Signed-off-by: Peter Zijlstra 
LKML-Reference: <20091217121829.970166036@chello.nl>
Signed-off-by: Ingo Molnar

Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

2009-12-05T23:30:49+00:00

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (35 commits)
  sched, cputime: Introduce thread_group_times()
  sched, cputime: Cleanups related to task_times()
  Revert "sched, x86: Optimize branch hint in __switch_to()"
  sched: Fix isolcpus boot option
  sched: Revert 498657a478c60be092208422fefa9c7b248729c2
  sched, time: Define nsecs_to_jiffies()
  sched: Remove task_{u,s,g}time()
  sched: Introduce task_times() to replace task_{u,s}time() pair
  sched: Limit the number of scheduler debug messages
  sched.c: Call debug_show_all_locks() when dumping all tasks
  sched, x86: Optimize branch hint in __switch_to()
  sched: Optimize branch hint in context_switch()
  sched: Optimize branch hint in pick_next_task_fair()
  sched_feat_write(): Update ppos instead of file->f_pos
  sched: Sched_rt_periodic_timer vs cpu hotplug
  sched, kvm: Fix race condition involving sched_in_preempt_notifers
  sched: More generic WAKE_AFFINE vs select_idle_sibling()
  sched: Cleanup select_task_rq_fair()
  sched: Fix granularity of task_u/stime()
  sched: Fix/add missing update_rq_clock() calls
  ...

sched, cputime: Introduce thread_group_times()

2009-12-02T16:32:40+00:00

This is a real fix for problem of utime/stime values decreasing
described in the thread:

   http://lkml.org/lkml/2009/11/3/522

Now cputime is accounted in the following way:

 - {u,s}time in task_struct are increased every time when the thread
   is interrupted by a tick (timer interrupt).

 - When a thread exits, its {u,s}time are added to signal->{u,s}time,
   after adjusted by task_times().

 - When all threads in a thread_group exits, accumulated {u,s}time
   (and also c{u,s}time) in signal struct are added to c{u,s}time
   in signal struct of the group's parent.

So {u,s}time in task struct are "raw" tick count, while
{u,s}time and c{u,s}time in signal struct are "adjusted" values.

And accounted values are used by:

 - task_times(), to get cputime of a thread:
   This function returns adjusted values that originates from raw
   {u,s}time and scaled by sum_exec_runtime that accounted by CFS.

 - thread_group_cputime(), to get cputime of a thread group:
   This function returns sum of all {u,s}time of living threads in
   the group, plus {u,s}time in the signal struct that is sum of
   adjusted cputimes of all exited threads belonged to the group.

The problem is the return value of thread_group_cputime(),
because it is mixed sum of "raw" value and "adjusted" value:

  group's {u,s}time = foreach(thread){{u,s}time} + exited({u,s}time)

This misbehavior can break {u,s}time monotonicity.
Assume that if there is a thread that have raw values greater
than adjusted values (e.g. interrupted by 1000Hz ticks 50 times
but only runs 45ms) and if it exits, cputime will decrease (e.g.
-5ms).

To fix this, we could do:

  group's {u,s}time = foreach(t){task_times(t)} + exited({u,s}time)

But task_times() contains hard divisions, so applying it for
every thread should be avoided.

This patch fixes the above problem in the following way:

 - Modify thread's exit (= __exit_signal()) not to use task_times().
   It means {u,s}time in signal struct accumulates raw values instead
   of adjusted values.  As the result it makes thread_group_cputime()
   to return pure sum of "raw" values.

 - Introduce a new function thread_group_times(*task, *utime, *stime)
   that converts "raw" values of thread_group_cputime() to "adjusted"
   values, in same calculation procedure as task_times().

 - Modify group's exit (= wait_task_zombie()) to use this introduced
   thread_group_times().  It make c{u,s}time in signal struct to
   have adjusted values like before this patch.

 - Replace some thread_group_cputime() by thread_group_times().
   This replacements are only applied where conveys the "adjusted"
   cputime to users, and where already uses task_times() near by it.
   (i.e. sys_times(), getrusage(), and /proc//stat.)

This patch have a positive side effect:

 - Before this patch, if a group contains many short-life threads
   (e.g. runs 0.9ms and not interrupted by ticks), the group's
   cputime could be invisible since thread's cputime was accumulated
   after adjusted: imagine adjustment function as adj(ticks, runtime),
     {adj(0, 0.9) + adj(0, 0.9) + ....} = {0 + 0 + ....} = 0.
   After this patch it will not happen because the adjustment is
   applied after accumulated.

v2:
 - remove if()s, put new variables into signal_struct.

Signed-off-by: Hidetoshi Seto 
Acked-by: Peter Zijlstra 
Cc: Spencer Candland 
Cc: Americo Wang 
Cc: Oleg Nesterov 
Cc: Balbir Singh 
Cc: Stanislaw Gruszka 
LKML-Reference: <4B162517.8040909@jp.fujitsu.com>
Signed-off-by: Ingo Molnar

sched: Remove task_{u,s,g}time()

2009-11-26T11:59:20+00:00

Now all task_{u,s}time() pairs are replaced by task_times().
And task_gtime() is too simple to be an inline function.

Cleanup them all.

Signed-off-by: Hidetoshi Seto 
Acked-by: Peter Zijlstra 
Cc: Stanislaw Gruszka 
Cc: Spencer Candland 
Cc: Oleg Nesterov 
Cc: Balbir Singh 
Cc: Americo Wang 
LKML-Reference: <4B0E16D1.70902@jp.fujitsu.com>
Signed-off-by: Ingo Molnar

sched: Introduce task_times() to replace task_{u,s}time() pair

2009-11-26T11:59:19+00:00

Functions task_{u,s}time() are called in pair in almost all
cases.  However task_stime() is implemented to call task_utime()
from its inside, so such paired calls run task_utime() twice.

It means we do heavy divisions (div_u64 + do_div) twice to get
utime and stime which can be obtained at same time by one set
of divisions.

This patch introduces a function task_times(*tsk, *utime,
*stime) to retrieve utime and stime at once in better, optimized
way.

Signed-off-by: Hidetoshi Seto 
Acked-by: Peter Zijlstra 
Cc: Stanislaw Gruszka 
Cc: Spencer Candland 
Cc: Oleg Nesterov 
Cc: Balbir Singh 
Cc: Americo Wang 
LKML-Reference: <4B0E16AE.906@jp.fujitsu.com>
Signed-off-by: Ingo Molnar

procfs: fix /proc//stat stack pointer for kernel threads

2009-11-18T01:40:33+00:00

Fix a small issue for the stack pointer in /proc//stat.  In case of a
kernel thread the value of the printed stack pointer should be 0.

Signed-off-by: Stefani Seibold 
Cc: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Merge branch 'linus' into sched/core

2009-10-25T16:30:53+00:00

Conflicts:
	fs/proc/array.c

Merge reason: resolve conflict and queue up dependent patch.

Signed-off-by: Ingo Molnar

procfs: disable per-task stack usage on NOMMU

2009-09-25T00:11:24+00:00

It needs walk_page_range().

Reported-by: Michal Simek 
Tested-by: Michal Simek 
Cc: Stefani Seibold 
Cc: David Howells 
Cc: Paul Mundt 
Cc: Geert Uytterhoeven 
Cc: Greg Ungerer 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds