diff options
| author | Kirill Tkhai <[email protected]> | 2014-08-20 13:48:29 +0400 |
|---|---|---|
| committer | Ingo Molnar <[email protected]> | 2014-08-20 14:53:05 +0200 |
| commit | 163122b7fcfa28c0e4a838fcc8043c616746802e (patch) | |
| tree | 6e23b7cafd125e42192cdfd52e385ad0ddbb1861 /tools/perf/scripts/python/bin/stackcollapse-report | |
| parent | e5673f280501298dbb56efa46e333cf64ee5080a (diff) | |
sched/fair: Remove double_lock_balance() from load_balance()
Avoid double_rq_lock() and use TASK_ON_RQ_MIGRATING for
load_balance(). The advantage is (obviously) not holding two
rq->lock's at the same time and thereby increasing parallelism.
Further note that if there was no task to migrate we will not
have acquired the second rq->lock at all.
The important point to note is that because we acquire dst->lock
immediately after releasing src->lock the potential wait time of
task_rq_lock() callers on TASK_ON_RQ_MIGRATING is not longer
than it would have been in the double rq lock scenario.
Signed-off-by: Kirill Tkhai <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Paul Turner <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Kirill Tkhai <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Nicolas Pitre <[email protected]>
Cc: Linus Torvalds <[email protected]>
Link: http://lkml.kernel.org/r/1408528109.23412.94.camel@tkhai
Signed-off-by: Ingo Molnar <[email protected]>
Diffstat (limited to 'tools/perf/scripts/python/bin/stackcollapse-report')
0 files changed, 0 insertions, 0 deletions