Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1391363 > unrolled thread

[PATCH 0/7] sched/fair: fixes and cleanups

Back to article view | Back to linux.kernel

Contents

  [PATCH 0/7] sched/fair: fixes and cleanups Dietmar Eggemann <dietmar.eggemann@arm.com> - 2016-04-29 21:40 +0200
    [PATCH 6/7] sched/fair: Reorder code in update_sd_lb_stats() Dietmar Eggemann <dietmar.eggemann@arm.com> - 2016-04-29 21:40 +0200

#1391363 — [PATCH 0/7] sched/fair: fixes and cleanups

This is a collection of rather loosely coupled small changes to the cfs 
scheduler related to: 

Comments (patch 1/7, 2/7):

  Remove/fix some outdated comments related to 'power aware scheduling'
  and smpnice.

Load-balance (patch 3/7 - 5/7):

  Change unit of load_above_capacity in calculate_imbalance() from
  [load] to [load/capacity]

  Refactoring logic in fix_small_imbalance() and remove
  cpu_avg_load_per_task().

Misc (patch 6/7, 7/7):

  Reorder update_sd_lb_stats() and use group_cfs_rq(se) throughout
  fair.c.

Test-results:

IVB-EP (2*10*2):

  perf stat --null --repeat 10 -- perf bench sched messaging -g 50 -l 5000

   Before:                     After:
   4.784556741 ( +- 0.99% )    4.760953662 ( +- 0.47% )

Juno (ARM64) (2B/4l*1):

  perf stat --null --repeat 10 -- perf bench sched messaging -g 20 -l 500

  Before:                     After:
  7.227439638 ( +- 0.50% )    7.151685712 ( +- 0.64% )

Dietmar Eggemann (5):
  sched/fair: Remove remaining power aware scheduling comments
  sched/fair: Fix comment in calculate_imbalance()
  sched/fair: Clean up the logic in fix_small_imbalance()
  sched/fair: Reorder code in update_sd_lb_stats()
  sched/fair: Use group_cfs_rq(se) instead of se->my_q

Morten Rasmussen (2):
  sched/fair: Correct unit of load_above_capacity
  sched/fair: Remove cpu_avg_load_per_task()

 kernel/sched/fair.c | 71 ++++++++++++++++++++---------------------------------
 1 file changed, 26 insertions(+), 45 deletions(-)

-- 
1.9.1

[toc] | [next] | [standalone]

#1391365 — [PATCH 6/7] sched/fair: Reorder code in update_sd_lb_stats()

Do the update of total load and total capacity of sched_domain
statistics before detecting if it is the local group. This and the
inclusion of sg=sg->next into the condition of the do...while loop
makes the code easier to read.

Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
---
 kernel/sched/fair.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e68b9eb5cb97..58da724b6ca4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6801,8 +6801,12 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 		update_sg_lb_stats(env, sg, load_idx, local_group, sgs,
 						&overload);
 
+		/* Update sd_lb_stats */
+		sds->total_load += sgs->group_load;
+		sds->total_capacity += sgs->group_capacity;
+
 		if (local_group)
-			goto next_group;
+			continue;
 
 		/*
 		 * In case the child domain prefers tasks go to siblings
@@ -6826,13 +6830,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 			sds->busiest_stat = *sgs;
 		}
 
-next_group:
-		/* Now, start updating sd_lb_stats */
-		sds->total_load += sgs->group_load;
-		sds->total_capacity += sgs->group_capacity;
-
-		sg = sg->next;
-	} while (sg != env->sd->groups);
+	} while (sg = sg->next, sg != env->sd->groups);
 
 	if (env->sd->flags & SD_NUMA)
 		env->fbq_type = fbq_classify_group(&sds->busiest_stat);
-- 
1.9.1

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web