ARM: tegra: power: Use runnable threads average for hotplug
authorWen Yi <wyi@nvidia.com>
Tue, 12 Jun 2012 19:01:11 +0000 (12:01 -0700)
committerSimone Willett <swillett@nvidia.com>
Sun, 1 Jul 2012 16:15:23 +0000 (09:15 -0700)
commit69f4a0f9adb0d8f82b920ceffcc40f7bd770c2e2
tree282ae29b0961e6b99b02e1235f32bac2a25039e6
parent1802afb2ad9ee1b6c1e11207f3fcdd3a56a1e0f0
ARM: tegra: power: Use runnable threads average for hotplug

Sample scheduler runnable threads average in auto-hotplug work
function and use it to determine the auto-hotplug target for number
of on-line cores. Use cpu up delay as sampling period, and enforce
down delay by checking last cpu configuration change time stamp.

Bug 958978

Change-Id: I4280a11d39914687e6ffaa6f38df594d10aedaa9
Signed-off-by: Alex Frid <afrid@nvidia.com>
(cherry picked from commit 507e2ef5e4f09b23de2e924003dba259d3c8bc3c)
Reviewed-on: http://git-master/r/111638
Reviewed-by: Sai Gurrappadi <sgurrappadi@nvidia.com>
Tested-by: Sai Gurrappadi <sgurrappadi@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Peter Boonstoppel <pboonstoppel@nvidia.com>
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
arch/arm/mach-tegra/cpu-tegra3.c