Re: [patch v3 0/8] sched: use runnable avg in load balance

[Alex Shi <alex.shi@intel.com>]

On 04/02/2013 11:23 AM, Alex Shi wrote:
> This version resolved the aim7 liked benchmark issue by patch 8th.
> Thanks for MikeG's avg_idle that is a good bursty wakeup indicator.
> 
> The first 3 patches were also include in my power aware scheduling patchset.
> 
> Morten, you can rebase your patch on this new version that bases on latest
> Linus tree. :)
> 
> a git tree for this patchset:
> 	https://github.com/alexshi/power-scheduling.git runnablelb

I removed the 3rd and 8th patches, left 1,2,4,5,6,7. and updated them in
above git tree.

I tested the kbuild, specjbb2005, aim9, fileio-cfq, hackbench and
dbench. on my NHM EP and 2 sockets SNB EP box. hackbench increased 3~5%
on both machines. kbuild has suspicious 2% increase. others has no clear
change.

Any more comments or concern about this patch?
> 
> Thanks
> Alex
> 
> [patch v3 1/8] Revert "sched: Introduce temporary FAIR_GROUP_SCHED
> [patch v3 2/8] sched: set initial value of runnable avg for new
> [patch v3 3/8] sched: only count runnable avg on cfs_rq's nr_running
> [patch v3 4/8] sched: update cpu load after task_tick.
> [patch v3 5/8] sched: compute runnable load avg in cpu_load and
> [patch v3 6/8] sched: consider runnable load average in move_tasks
> [patch v3 7/8] sched: consider runnable load average in
> [patch v3 8/8] sched: use instant load for burst wake up
> 


-- 
Thanks Alex