Allow top-app tasks to be upmigrated always

Signed-off-by: Tyler Nijmeh <tylernij@gmail.com>

README.md

@@ -86,8 +86,8 @@ Set the minimum task scheduling period to 2.5ms. With kernel.sched_latency_ns se
 Increase the time that a task is considered to be cache hot. According to RedHat, increasing this tunable reduces the number of task migrations. This should reduce time spent balancing tasks and increase per-task performance.
 See RedHat: https://www.redhat.com/files/summit/session-assets/2018/Performance-analysis-and-tuning-of-Red-Hat-Enterprise-Linux-Part-1.pdf
 
-### kernel.sched_min_task_util_for_colocation: 35 --> 20
-This value determines when top-app tasks (which are of greater priority than background tasks) can be sched_boosted. Lower this value a bit to use big clusters more for top-app tasks.
+### kernel.sched_min_task_util_for_colocation: 35 --> 0
+This value determines when top-app tasks (which are of greater priority than background tasks) can be sched_boosted. Set this value to zero to allow top-app tasks to always be upmigrated if the sched_{up,down}migrate values are met.
 
 ### kernel.sched_nr_migrate: 32 --> 128
 When migrating tasks between CPUs, allow the scheduler to migrate twice as many as usual. This should increase scheduling latency marginally, but increase the performance of SCHED_OTHER tasks. In testing, `cyclictest` reported a 2 microsecond increase in average latency, an a **decrease** in maximum latency of SCHED_FIFO tasks.

system/bin/ktweak

@@ -117,7 +117,7 @@ write /proc/sys/kernel/sched_tunable_scaling 0
 write /proc/sys/kernel/sched_latency_ns 10000000
 write /proc/sys/kernel/sched_min_granularity_ns 2500000
 write /proc/sys/kernel/sched_migration_cost_ns 1000000
-write /proc/sys/kernel/sched_min_task_util_for_colocation 20
+write /proc/sys/kernel/sched_min_task_util_for_colocation 0
 write /proc/sys/kernel/sched_nr_migrate 128
 write /proc/sys/kernel/sched_schedstats 0
 write /proc/sys/kernel/sched_wakeup_granularity_ns 10000000