Peter Zijlstra 3e71a462dd sched/core: Move task_rq_lock() out of line
Its a rather large function, inline doesn't seems to make much sense:

 $ size defconfig-build/kernel/sched/core.o{.orig,}
    text    data     bss     dec     hex filename
   56533   21037    2320   79890   13812 defconfig-build/kernel/sched/core.o.orig
   55733   21037    2320   79090   134f2 defconfig-build/kernel/sched/core.o

The 'perf bench sched messaging' micro-benchmark shows a visible improvement
of 4-5%:

  $ for i in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor ; do echo performance > $i ; done
  $ perf stat --null --repeat 25 -- perf bench sched messaging -g 40 -l 5000

  pre:
       4.582798193 seconds time elapsed          ( +-  1.41% )
       4.733374877 seconds time elapsed          ( +-  2.10% )
       4.560955136 seconds time elapsed          ( +-  1.43% )
       4.631062303 seconds time elapsed          ( +-  1.40% )

  post:
       4.364765213 seconds time elapsed          ( +-  0.91% )
       4.454442734 seconds time elapsed          ( +-  1.18% )
       4.448893817 seconds time elapsed          ( +-  1.41% )
       4.424346872 seconds time elapsed          ( +-  0.97% )

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-05 09:23:58 +02:00
..
2015-12-04 22:46:25 +01:00
2016-03-22 15:36:02 -07:00
2015-04-11 22:27:55 -04:00
2015-11-19 17:51:48 +01:00
2016-01-22 18:04:28 -05:00
2016-02-08 11:25:39 -05:00
2014-08-08 15:57:18 -07:00
2015-12-14 14:54:37 -05:00
2015-04-12 21:03:31 +02:00
2016-03-22 15:36:02 -07:00
2015-11-23 09:44:58 +01:00
2016-03-22 15:36:02 -07:00
2016-03-22 15:36:02 -07:00
2016-03-22 15:36:02 -07:00
2016-02-16 13:04:58 -05:00
2015-01-17 10:02:23 +13:00
2016-01-22 18:04:28 -05:00
2016-03-01 20:36:56 +01:00