Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 39026008 authored by Greg Kroah-Hartman's avatar Greg Kroah-Hartman Committed by Alistair Delva
Browse files

Revert "ANDROID: sched/fair: Attempt to improve throughput for asym cap systems" 



This reverts commit 1a1d22a3.

We want to back-out the eas-dev merge that happened in the tree after
5.3-rc1 as those patches "should" all be in Linus's tree now.

This is done to handle the merge conflicts with 5.4-rc1.

Cc: Todd Kjos <tkjos@google.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@google.com>
Change-Id: I4f3a3b047d2c81f2f6de6da633964f009717a1cf
parent 504e400e

kernel/sched/fair.c

+1 −16
Original line number Diff line number Diff line
@@ -8461,22 +8461,7 @@ void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds) 
	capa_move /= SCHED_CAPACITY_SCALE;



	/* Move if we gain throughput */

	if (capa_move > capa_now) {

		env->imbalance = busiest->load_per_task;

		return;

	}



	/* We can't see throughput improvement with the load-based

	 * method, but it is possible depending upon group size and

	 * capacity range that there might still be an underutilized

	 * cpu available in an asymmetric capacity system. Do one last

	 * check just in case.

	 */

	if (env->sd->flags & SD_ASYM_CPUCAPACITY &&

		busiest->group_type == group_overloaded &&

		busiest->sum_nr_running > busiest->group_weight &&

		local->sum_nr_running < local->group_weight &&

		local->group_capacity < busiest->group_capacity)

	if (capa_move > capa_now)

		env->imbalance = busiest->load_per_task;

}