Hi Brian,

�� 2017��07��13�� 01:32, Brian Norris �:
Hi Caesar,

On Wed, Jul 12, 2017 at 02:29:30PM +0800, Caesar Wang wrote:
As RK3399 had used the Power allocator thermal governor by default,
enabled this to manage thermals by dynamically allocating and limiting
power to devices.

Also, this patch supported the dynamic-power-coefficient/sustainable_power
and GPU's power model for needed parameters with thermal IPA.

Signed-off-by: Caesar Wang <wxt@rock-chips.com>

---

 arch/arm64/boot/dts/rockchip/rk3399.dtsi | 62 +++++++++++++++-----------------
 1 file changed, 29 insertions(+), 33 deletions(-)

diff --git a/arch/arm64/boot/dts/rockchip/rk3399.dtsi b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
index 8c6438b..139f58c 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
@@ -147,7 +147,7 @@
 			enable-method = "psci";
 			#cooling-cells = <2>; /* min followed by max */
 			clocks = <&cru ARMCLKB>;
-			dynamic-power-coefficient = <100>;
+			dynamic-power-coefficient = <436>;
 		};
 
 		cpu_b1: cpu@101 {
@@ -156,7 +156,7 @@
 			reg = <0x0 0x101>;
 			enable-method = "psci";
 			clocks = <&cru ARMCLKB>;
-			dynamic-power-coefficient = <100>;
+			dynamic-power-coefficient = <436>;
There are 6 of these properties (1 for each core now; not just 1 for
each cluster), and you're only changing 2 of them.

BTW, are these values determined from measurement this time? And heavily
tested? The previous values were suspiciously round, but they'd been
heavily tested so I didn't mind :)


The coefficient is used to calulate the dynamic power as below -
Pdyn= dynamic-power-coefficient * V^2 * f

The value was tested on rockchip inside, that should be resonable for me.
As the PLT tested the little cpu power comsumption Pl = 0.9mw, but big cpu power comsumption Pb= 0.28mw

-Caesar

 		};
 	};
 
...

Brian

_______________________________________________
Linux-rockchip mailing list
Linux-rockchip@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-rockchip