EAS Check List and Known Issues Tracking

NO.

Description

Category

Solution Or Fixes

1

Reflect sched_entity movement into task_group's utilization

CFS

Vincent's patch; Dietmar's patch

2

Decay Idle CPU utilization signal

CFS

N/A

3

Fix first task of a task group is attached twice

CFS

Vincent's patch; Yuyang's patch; Peter Zijlstra's patch

4

fix hierarchical order in rq->leaf_cfs_rq_list

CFS

Vincent's patch

5

Change unit of load_above_capacity from [load] to [load/capacity]

CFS

Patch 0 1 2

6

Fix new task's load avg removed from source CPU in wake_up_new_task()

CFS

Yuyang's patch

7

schedTune PE filter

EAS

Patrick's patch

8

schedTune payoff threshold array

EAS

Patrick's patches 0 1 2

9

Optimize over-utilized as per sched domain flag

EAS

WIP Thara

10

Spread tasks within big cluster

EAS

echo 0 > /proc/sys/kernel/sched_migration_cost_ns

11

Fix load balance when big cluster is heavily load but little cluster is idle

EAS

WIP

12

Migrate single task from little cluster to big cluster in the case under tipping point

EAS

Leo's Patch


CategoryHowTo

eas-checklist-and-known-issues-tracking (last modified 2016-06-29 12:34:39)