This series patches supported the mail in devicetree and used the
thermal IPA by default.
Verified with kernel is based on Linus's master branch and Heiko's
v4.14-armsoc-tmp/dts64 branch. ( The Linux version 4.12.0 for now).
The most rockchip SoCs will be supported with IPA mode for thermal
in later.
---
History version:
1. The first version found on
https://www.spinics.net/lists/arm-kernel/msg593118.html
Tested on Kevin board with bringing up ChromeOS.
OS VERSION:
CHROMEOS_RELEASE_DESCRIPTION=9693.1.0 (Official Build) dev-channel kevin test
BIOS VERSION:
Google_Kevin.8785.211.2017_06_20_1043
EC VERSION:
Build info: kevin_v1.10.217-24514961d 2017-07-03 07:46:36 wxt@nb
With the ARM's lastest mali driver TX011-SW-99002-r18p0-01rel0 on
https://developer.arm.com/products/software/mali-drivers/midgard-kernel
>From the bootup log:
localhost devfreq0 # dmesg |grep mali
[ 0.933334] mali ff9a0000.gpu: GPU identified as 0x0860 r2p0 status 0
[ 0.940830] mali ff9a0000.gpu: Protected mode not available
[ 0.947334] mali ff9a0000.gpu: Using configured power model mali-simple-power-model, and fallback mali-simple-power-model
[ 0.960083] mali ff9a0000.gpu: Probed as mali0
localhost devfreq0 # pwd
/sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0
localhost devfreq0 # ls
available_frequencies device min_freq subsystem uevent
available_governors governor polling_interval target_freq userspace
cur_freq max_freq power trans_stat
localhost ff9a0000.gpu # ls
core_availability_policy gpuinfo modalias soft_job_timeout
core_mask js_scheduling_period of_node subsystem
devfreq js_timeouts pm_poweroff uevent
driver mem_pool_max_size power
driver_override mem_pool_size power_policy
dvfs_period misc reset_timeout
---
And for thermal with IPA.
Try to run 'md5sum /dev/zero &' and octane/benchmark scripts to go up the temperature.
>From the scripts to have a look at the actual control.
"
while true; do grep "" /sys/class/thermal/thermal_zone[0-1]/temp /sys/devices/system/cpu/cpu[0-5]/cpufreq/scaling_cur_freq /sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0/cur_freq;date;sleep .5; done &
"
-Caesar
Changes in v2:
As Heiko comments on https://patchwork.kernel.org/patch/9835939/
- interrupt-name use the lower case.
- use the correct compatible "arm,mali-t860"
- remove the clock name, since the mali only have one input clock.
- foo@ will produce warnings when used without reg property.
- update the commit to explain the two passive trip points changed.
Caesar Wang (5):
dt-bindings: gpu: add the RK3399 mali for rockchip specifics
dt-bindings: gpu: add a power_model optional properties for MALI
arm64: dts: rockchip: add ARM Mali GPU node for RK3399 SoCs
arm64: dts: rockchip: enable the GPU for RK3399-GRU
arm64: dts: rockchip: update the thermal zones for RK3399 SoCs
.../devicetree/bindings/gpu/arm,mali-midgard.txt | 13 ++++
arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi | 5 ++
arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi | 33 ++++++++++
arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi | 33 ++++++++++
arch/arm64/boot/dts/rockchip/rk3399.dtsi | 74 ++++++++++++----------
5 files changed, 125 insertions(+), 33 deletions(-)
--
2.7.4
This patch adds the MALI's power-model to set the IPA model to be used
for power management.
Signed-off-by: Caesar Wang <[email protected]>
---
Changes in v2: None
Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
index a461e47..b616e6b 100644
--- a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
+++ b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
@@ -37,6 +37,18 @@ Optional properties:
- operating-points-v2 : Refer to Documentation/devicetree/bindings/power/opp.txt
for details.
+- power_model : Sets power model parameters. Note that this model was designed for the Juno
+ platform, and may not be suitable for other platforms. A structure containing :
+ - compatible: Should be arm,mali-simple-power-model
+ - dynamic-coefficient: Coefficient, in pW/(Hz V^2), which is multiplied
+ by v^2*f to calculate the dynamic power consumption.
+ - static-coefficient: Coefficient, in uW/V^3, which is multiplied by
+ v^3 to calculate the static power consumption.
+ - ts: An array containing coefficients for the temperature scaling
+ factor. This is used to scale the static power by a factor of
+ tsf/1000000, where tsf = ts[3]*T^3 + ts[2]*T^2 + ts[1]*T + ts[0],
+ and T = temperature in degrees.
+ - thermal-zone: A string identifying the thermal zone used for the GPU
Example for a Mali-T760:
--
2.7.4
This patch enables the gpu and adds the mali-supply power for RK3399-GRU
devices.
Signed-off-by: Caesar Wang <[email protected]>
---
Changes in v2: None
arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi
index 90259cf..d48e98b 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi
@@ -595,6 +595,11 @@
status = "okay";
};
+&gpu {
+ mali-supply = <&ppvar_gpu>;
+ status = "okay";
+};
+
ap_i2c_mic: &i2c1 {
status = "okay";
--
2.7.4
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.
The Thermal power allocator governor works optimatly with two passive trip
points, for the better performance we will use the trip-point0 with 70
degree above which the governor control starts operating and trip-point1
with 85 degree is the target temperature by controlling.
Signed-off-by: Caesar Wang <[email protected]>
---
Changes in v2:
- foo@ will produce warnings when used without reg property.
- update the commit to explain the two passive trip points changed.
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 77d67cb..6d8a5eb 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>;
};
};
@@ -690,24 +690,25 @@
};
thermal_zones: thermal-zones {
- cpu_thermal: cpu {
+ soc_thermal: soc-thermal {
polling-delay-passive = <100>;
polling-delay = <1000>;
+ sustainable-power = <1000>;
thermal-sensors = <&tsadc 0>;
trips {
- cpu_alert0: cpu_alert0 {
+ threshold: trip-point0 {
temperature = <70000>;
hysteresis = <2000>;
type = "passive";
};
- cpu_alert1: cpu_alert1 {
- temperature = <75000>;
+ target: trip-point1 {
+ temperature = <85000>;
hysteresis = <2000>;
type = "passive";
};
- cpu_crit: cpu_crit {
+ soc_crit: soc-crit {
temperature = <95000>;
hysteresis = <2000>;
type = "critical";
@@ -716,45 +717,31 @@
cooling-maps {
map0 {
- trip = <&cpu_alert0>;
+ trip = <&target>;
cooling-device =
- <&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ <&cpu_l0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ contribution = <4096>;
};
map1 {
- trip = <&cpu_alert1>;
+ trip = <&target>;
cooling-device =
- <&cpu_l0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
<&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ contribution = <1024>;
+ };
+ map2 {
+ trip = <&target>;
+ cooling-device =
+ <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ contribution = <4096>;
};
};
};
- gpu_thermal: gpu {
+ gpu_thermal: gpu-thermal {
polling-delay-passive = <100>;
polling-delay = <1000>;
thermal-sensors = <&tsadc 1>;
-
- trips {
- gpu_alert0: gpu_alert0 {
- temperature = <75000>;
- hysteresis = <2000>;
- type = "passive";
- };
- gpu_crit: gpu_crit {
- temperature = <95000>;
- hysteresis = <2000>;
- type = "critical";
- };
- };
-
- cooling-maps {
- map0 {
- trip = <&gpu_alert0>;
- cooling-device =
- <&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
- };
- };
};
};
@@ -1451,8 +1438,17 @@
<GIC_SPI 21 IRQ_TYPE_LEVEL_HIGH 0>;
interrupt-names = "gpu", "job", "mmu";
clocks = <&cru ACLK_GPU>;
+ #cooling-cells = <2>;
power-domains = <&power RK3399_PD_GPU>;
status = "disabled";
+
+ gpu_power_model: power_model {
+ compatible = "arm,mali-simple-power-model";
+ static-coefficient = <1079403>;
+ dynamic-coefficient = <977>;
+ ts = <32000 4700 (-80) 2>;
+ thermal-zone = "gpu-thermal";
+ };
};
pinctrl: pinctrl {
--
2.7.4
Add Mali GPU device tree node for the RK3399 SoCs, with devfreq
opp table.
RK3399 and RK3399-OP1 SoCs have a different recommendation table with
gpu opp. Also, the ARM's mali driver found on
https://developer.arm.com/products/software/mali-drivers/midgard-kernel.
Signed-off-by: Caesar Wang <[email protected]>
---
Changes in v2:
As Heiko comments on https://patchwork.kernel.org/patch/9835939/
- interrupt-name use the lower case.
- use the correct compatible "arm,mali-t860"
- remove the clock name, since the mali only have one input clock.
arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi | 33 ++++++++++++++++++++++++
arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi | 33 ++++++++++++++++++++++++
arch/arm64/boot/dts/rockchip/rk3399.dtsi | 12 +++++++++
3 files changed, 78 insertions(+)
diff --git a/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
index be7fe63..d8a120f 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
@@ -118,6 +118,35 @@
opp-microvolt = <1250000>;
};
};
+
+ gpu_opp_table: opp-table2 {
+ compatible = "operating-points-v2";
+
+ opp00 {
+ opp-hz = /bits/ 64 <200000000>;
+ opp-microvolt = <800000>;
+ };
+ opp01 {
+ opp-hz = /bits/ 64 <297000000>;
+ opp-microvolt = <800000>;
+ };
+ opp02 {
+ opp-hz = /bits/ 64 <400000000>;
+ opp-microvolt = <825000>;
+ };
+ opp03 {
+ opp-hz = /bits/ 64 <500000000>;
+ opp-microvolt = <850000>;
+ };
+ opp04 {
+ opp-hz = /bits/ 64 <600000000>;
+ opp-microvolt = <925000>;
+ };
+ opp05 {
+ opp-hz = /bits/ 64 <800000000>;
+ opp-microvolt = <1075000>;
+ };
+ };
};
&cpu_l0 {
@@ -143,3 +172,7 @@
&cpu_b1 {
operating-points-v2 = <&cluster1_opp>;
};
+
+&gpu {
+ operating-points-v2 = <&gpu_opp_table>;
+};
diff --git a/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
index c83460d..81617bc 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
@@ -110,6 +110,35 @@
opp-microvolt = <1200000>;
};
};
+
+ gpu_opp_table: opp-table2 {
+ compatible = "operating-points-v2";
+
+ opp00 {
+ opp-hz = /bits/ 64 <200000000>;
+ opp-microvolt = <800000>;
+ };
+ opp01 {
+ opp-hz = /bits/ 64 <297000000>;
+ opp-microvolt = <800000>;
+ };
+ opp02 {
+ opp-hz = /bits/ 64 <400000000>;
+ opp-microvolt = <825000>;
+ };
+ opp03 {
+ opp-hz = /bits/ 64 <500000000>;
+ opp-microvolt = <875000>;
+ };
+ opp04 {
+ opp-hz = /bits/ 64 <600000000>;
+ opp-microvolt = <925000>;
+ };
+ opp05 {
+ opp-hz = /bits/ 64 <800000000>;
+ opp-microvolt = <1100000>;
+ };
+ };
};
&cpu_l0 {
@@ -135,3 +164,7 @@
&cpu_b1 {
operating-points-v2 = <&cluster1_opp>;
};
+
+&gpu {
+ operating-points-v2 = <&gpu_opp_table>;
+};
diff --git a/arch/arm64/boot/dts/rockchip/rk3399.dtsi b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
index 1cbd7a2..77d67cb 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
@@ -1443,6 +1443,18 @@
status = "disabled";
};
+ gpu: gpu@ff9a0000 {
+ compatible = "arm,rk3399-mali", "arm,mali-t860";
+ reg = <0x0 0xff9a0000 0x0 0x10000>;
+ interrupts = <GIC_SPI 19 IRQ_TYPE_LEVEL_HIGH 0>,
+ <GIC_SPI 20 IRQ_TYPE_LEVEL_HIGH 0>,
+ <GIC_SPI 21 IRQ_TYPE_LEVEL_HIGH 0>;
+ interrupt-names = "gpu", "job", "mmu";
+ clocks = <&cru ACLK_GPU>;
+ power-domains = <&power RK3399_PD_GPU>;
+ status = "disabled";
+ };
+
pinctrl: pinctrl {
compatible = "rockchip,rk3399-pinctrl";
rockchip,grf = <&grf>;
--
2.7.4
RK3399's GPU uses the quad-core Mali-T860, which is the new generation of
high-end graphics processors from ARM.
This patch added "rockchip,rk3399-mali" for dt-bindings, in order to
support IPA of gpu thermal in later.
Signed-off-by: Caesar Wang <[email protected]>
---
Changes in v2: None
Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 1 +
1 file changed, 1 insertion(+)
diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
index d3b6e1a4..a461e47 100644
--- a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
+++ b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
@@ -17,6 +17,7 @@ Required properties:
* which must be preceded by one of the following vendor specifics:
+ "amlogic,meson-gxm-mali"
+ "rockchip,rk3288-mali"
+ + "rockchip,rk3399-mali"
- reg : Physical base address of the device and length of the register area.
--
2.7.4
On Mon, Jul 17, 2017 at 04:14:27PM +0800, Caesar Wang wrote:
> RK3399's GPU uses the quad-core Mali-T860, which is the new generation of
> high-end graphics processors from ARM.
>
> This patch added "rockchip,rk3399-mali" for dt-bindings, in order to
> support IPA of gpu thermal in later.
>
> Signed-off-by: Caesar Wang <[email protected]>
> ---
>
> Changes in v2: None
>
> Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 1 +
> 1 file changed, 1 insertion(+)
Acked-by: Rob Herring <[email protected]>
On Mon, Jul 17, 2017 at 04:14:28PM +0800, Caesar Wang wrote:
> This patch adds the MALI's power-model to set the IPA model to be used
> for power management.
What's IPA? India Pale Ale or Intermediate Physical Address?
>
> Signed-off-by: Caesar Wang <[email protected]>
> ---
>
> Changes in v2: None
>
> Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 12 ++++++++++++
> 1 file changed, 12 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> index a461e47..b616e6b 100644
> --- a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> +++ b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> @@ -37,6 +37,18 @@ Optional properties:
> - operating-points-v2 : Refer to Documentation/devicetree/bindings/power/opp.txt
> for details.
>
> +- power_model : Sets power model parameters. Note that this model was designed for the Juno
> + platform, and may not be suitable for other platforms. A structure containing :
> + - compatible: Should be arm,mali-simple-power-model
> + - dynamic-coefficient: Coefficient, in pW/(Hz V^2), which is multiplied
> + by v^2*f to calculate the dynamic power consumption.
> + - static-coefficient: Coefficient, in uW/V^3, which is multiplied by
> + v^3 to calculate the static power consumption.
> + - ts: An array containing coefficients for the temperature scaling
> + factor. This is used to scale the static power by a factor of
> + tsf/1000000, where tsf = ts[3]*T^3 + ts[2]*T^2 + ts[1]*T + ts[0],
> + and T = temperature in degrees.
> + - thermal-zone: A string identifying the thermal zone used for the GPU
This can all easily be implied by the compatible string. I'm not
inclined to accept something Mali specific here.
This looks *very* precise, but I'd be surprised if these values are any
more than magic values (at least the dynamic coef) adjusted until the
desired power/performance requirements are achieved. To put it another
way, why don't we have similar values for CPUs?
Rob
Rob,
在 2017年07月18日 04:07, Rob Herring 写道:
> On Mon, Jul 17, 2017 at 04:14:28PM +0800, Caesar Wang wrote:
>> This patch adds the MALI's power-model to set the IPA model to be used
>> for power management.
> What's IPA? India Pale Ale or Intermediate Physical Address?
IPA is intelligent Power Allocator. (As the ARM introduced on
https://developer.arm.com/open-source/intelligent-power-allocation)
>
>> Signed-off-by: Caesar Wang <[email protected]>
>> ---
>>
>> Changes in v2: None
>>
>> Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 12 ++++++++++++
>> 1 file changed, 12 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
>> index a461e47..b616e6b 100644
>> --- a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
>> +++ b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
>> @@ -37,6 +37,18 @@ Optional properties:
>> - operating-points-v2 : Refer to Documentation/devicetree/bindings/power/opp.txt
>> for details.
>>
>> +- power_model : Sets power model parameters. Note that this model was designed for the Juno
>> + platform, and may not be suitable for other platforms. A structure containing :
>> + - compatible: Should be arm,mali-simple-power-model
>> + - dynamic-coefficient: Coefficient, in pW/(Hz V^2), which is multiplied
>> + by v^2*f to calculate the dynamic power consumption.
>> + - static-coefficient: Coefficient, in uW/V^3, which is multiplied by
>> + v^3 to calculate the static power consumption.
>> + - ts: An array containing coefficients for the temperature scaling
>> + factor. This is used to scale the static power by a factor of
>> + tsf/1000000, where tsf = ts[3]*T^3 + ts[2]*T^2 + ts[1]*T + ts[0],
>> + and T = temperature in degrees.
>> + - thermal-zone: A string identifying the thermal zone used for the GPU
> This can all easily be implied by the compatible string. I'm not
> inclined to accept something Mali specific here.
Isn't arm,mali-midgard.txt document suit for Mali specific? :-)
>
> This looks *very* precise, but I'd be surprised if these values are any
> more than magic values (at least the dynamic coef) adjusted until the
> desired power/performance requirements are achieved. To put it another
> way, why don't we have similar values for CPUs?
These value was calculated by running full GPU process.
CPU had the similar value for dtsi.
Say: arch/arm64/boot/dts/rockchip/rk3399.dtsi
cpu_b0: cpu@100 {
...
dynamic-power-coefficient = <436>;
...
};
-Caesar
>
> Rob
>
>
>
Hi Rob & Heiko,
Do we have the chance to merge these patches?
I'm try to bring up the display and run webgl for testing with my github on
https://github.com/Caesar-github/rockchip/commits/gru/next-stable-chromeos
I believe the Rocky@RK will post patches to support the other SoCs after.
-Caesar
在 2017年07月17日 16:14, Caesar Wang 写道:
> This series patches supported the mail in devicetree and used the
> thermal IPA by default.
>
> Verified with kernel is based on Linus's master branch and Heiko's
> v4.14-armsoc-tmp/dts64 branch. ( The Linux version 4.12.0 for now).
>
> The most rockchip SoCs will be supported with IPA mode for thermal
> in later.
>
> ---
>
> History version:
> 1. The first version found on
> https://www.spinics.net/lists/arm-kernel/msg593118.html
>
> Tested on Kevin board with bringing up ChromeOS.
> OS VERSION:
> CHROMEOS_RELEASE_DESCRIPTION=9693.1.0 (Official Build) dev-channel kevin test
>
> BIOS VERSION:
> Google_Kevin.8785.211.2017_06_20_1043
>
> EC VERSION:
> Build info: kevin_v1.10.217-24514961d 2017-07-03 07:46:36 wxt@nb
>
> With the ARM's lastest mali driver TX011-SW-99002-r18p0-01rel0 on
> https://developer.arm.com/products/software/mali-drivers/midgard-kernel
>
> From the bootup log:
> localhost devfreq0 # dmesg |grep mali
> [ 0.933334] mali ff9a0000.gpu: GPU identified as 0x0860 r2p0 status 0
> [ 0.940830] mali ff9a0000.gpu: Protected mode not available
> [ 0.947334] mali ff9a0000.gpu: Using configured power model mali-simple-power-model, and fallback mali-simple-power-model
> [ 0.960083] mali ff9a0000.gpu: Probed as mali0
>
> localhost devfreq0 # pwd
> /sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0
> localhost devfreq0 # ls
> available_frequencies device min_freq subsystem uevent
> available_governors governor polling_interval target_freq userspace
> cur_freq max_freq power trans_stat
>
> localhost ff9a0000.gpu # ls
> core_availability_policy gpuinfo modalias soft_job_timeout
> core_mask js_scheduling_period of_node subsystem
> devfreq js_timeouts pm_poweroff uevent
> driver mem_pool_max_size power
> driver_override mem_pool_size power_policy
> dvfs_period misc reset_timeout
> ---
>
> And for thermal with IPA.
> Try to run 'md5sum /dev/zero &' and octane/benchmark scripts to go up the temperature.
> From the scripts to have a look at the actual control.
> "
> while true; do grep "" /sys/class/thermal/thermal_zone[0-1]/temp /sys/devices/system/cpu/cpu[0-5]/cpufreq/scaling_cur_freq /sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0/cur_freq;date;sleep .5; done &
> "
>
> -Caesar
>
>
> Changes in v2:
> As Heiko comments on https://patchwork.kernel.org/patch/9835939/
> - interrupt-name use the lower case.
> - use the correct compatible "arm,mali-t860"
> - remove the clock name, since the mali only have one input clock.
> - foo@ will produce warnings when used without reg property.
> - update the commit to explain the two passive trip points changed.
>
> Caesar Wang (5):
> dt-bindings: gpu: add the RK3399 mali for rockchip specifics
> dt-bindings: gpu: add a power_model optional properties for MALI
> arm64: dts: rockchip: add ARM Mali GPU node for RK3399 SoCs
> arm64: dts: rockchip: enable the GPU for RK3399-GRU
> arm64: dts: rockchip: update the thermal zones for RK3399 SoCs
>
> .../devicetree/bindings/gpu/arm,mali-midgard.txt | 13 ++++
> arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi | 5 ++
> arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi | 33 ++++++++++
> arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi | 33 ++++++++++
> arch/arm64/boot/dts/rockchip/rk3399.dtsi | 74 ++++++++++++----------
> 5 files changed, 125 insertions(+), 33 deletions(-)
>
Am Mittwoch, 19. Juli 2017, 17:06:59 CEST schrieb Caesar Wang:
> Hi Rob & Heiko,
>
> Do we have the chance to merge these patches?
>From the mali-thermal thingy, it doesn't look like Rob likes that IPA stuff.
So I'll most likely pick the other patches that rely on approved bindings
after I return from my vacation and leave the thermal issue until it can
be resolved.
Heiko
>
> I'm try to bring up the display and run webgl for testing with my github on
> https://github.com/Caesar-github/rockchip/commits/gru/next-stable-chromeos
>
> I believe the Rocky@RK will post patches to support the other SoCs after.
>
> -Caesar
>
> 在 2017年07月17日 16:14, Caesar Wang 写道:
> > This series patches supported the mail in devicetree and used the
> > thermal IPA by default.
> >
> > Verified with kernel is based on Linus's master branch and Heiko's
> > v4.14-armsoc-tmp/dts64 branch. ( The Linux version 4.12.0 for now).
> >
> > The most rockchip SoCs will be supported with IPA mode for thermal
> > in later.
> >
> > ---
> >
> > History version:
> > 1. The first version found on
> > https://www.spinics.net/lists/arm-kernel/msg593118.html
> >
> > Tested on Kevin board with bringing up ChromeOS.
> > OS VERSION:
> > CHROMEOS_RELEASE_DESCRIPTION=9693.1.0 (Official Build) dev-channel kevin test
> >
> > BIOS VERSION:
> > Google_Kevin.8785.211.2017_06_20_1043
> >
> > EC VERSION:
> > Build info: kevin_v1.10.217-24514961d 2017-07-03 07:46:36 wxt@nb
> >
> > With the ARM's lastest mali driver TX011-SW-99002-r18p0-01rel0 on
> > https://developer.arm.com/products/software/mali-drivers/midgard-kernel
> >
> > From the bootup log:
> > localhost devfreq0 # dmesg |grep mali
> > [ 0.933334] mali ff9a0000.gpu: GPU identified as 0x0860 r2p0 status 0
> > [ 0.940830] mali ff9a0000.gpu: Protected mode not available
> > [ 0.947334] mali ff9a0000.gpu: Using configured power model mali-simple-power-model, and fallback mali-simple-power-model
> > [ 0.960083] mali ff9a0000.gpu: Probed as mali0
> >
> > localhost devfreq0 # pwd
> > /sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0
> > localhost devfreq0 # ls
> > available_frequencies device min_freq subsystem uevent
> > available_governors governor polling_interval target_freq userspace
> > cur_freq max_freq power trans_stat
> >
> > localhost ff9a0000.gpu # ls
> > core_availability_policy gpuinfo modalias soft_job_timeout
> > core_mask js_scheduling_period of_node subsystem
> > devfreq js_timeouts pm_poweroff uevent
> > driver mem_pool_max_size power
> > driver_override mem_pool_size power_policy
> > dvfs_period misc reset_timeout
> > ---
> >
> > And for thermal with IPA.
> > Try to run 'md5sum /dev/zero &' and octane/benchmark scripts to go up the temperature.
> > From the scripts to have a look at the actual control.
> > "
> > while true; do grep "" /sys/class/thermal/thermal_zone[0-1]/temp /sys/devices/system/cpu/cpu[0-5]/cpufreq/scaling_cur_freq /sys/devices/platform/ff9a0000.gpu/devfreq/devfreq0/cur_freq;date;sleep .5; done &
> > "
> >
> > -Caesar
> >
> >
> > Changes in v2:
> > As Heiko comments on https://patchwork.kernel.org/patch/9835939/
> > - interrupt-name use the lower case.
> > - use the correct compatible "arm,mali-t860"
> > - remove the clock name, since the mali only have one input clock.
> > - foo@ will produce warnings when used without reg property.
> > - update the commit to explain the two passive trip points changed.
> >
> > Caesar Wang (5):
> > dt-bindings: gpu: add the RK3399 mali for rockchip specifics
> > dt-bindings: gpu: add a power_model optional properties for MALI
> > arm64: dts: rockchip: add ARM Mali GPU node for RK3399 SoCs
> > arm64: dts: rockchip: enable the GPU for RK3399-GRU
> > arm64: dts: rockchip: update the thermal zones for RK3399 SoCs
> >
> > .../devicetree/bindings/gpu/arm,mali-midgard.txt | 13 ++++
> > arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi | 5 ++
> > arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi | 33 ++++++++++
> > arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi | 33 ++++++++++
> > arch/arm64/boot/dts/rockchip/rk3399.dtsi | 74 ++++++++++++----------
> > 5 files changed, 125 insertions(+), 33 deletions(-)
> >
>
>
>
>
Am Montag, 17. Juli 2017, 16:14:27 CEST schrieb Caesar Wang:
> RK3399's GPU uses the quad-core Mali-T860, which is the new generation of
> high-end graphics processors from ARM.
>
> This patch added "rockchip,rk3399-mali" for dt-bindings, in order to
> support IPA of gpu thermal in later.
>
> Signed-off-by: Caesar Wang <[email protected]>
applied for 4.14 with Rob's Ack
Heiko
Am Montag, 17. Juli 2017, 16:14:29 CEST schrieb Caesar Wang:
> Add Mali GPU device tree node for the RK3399 SoCs, with devfreq
> opp table.
>
> RK3399 and RK3399-OP1 SoCs have a different recommendation table with
> gpu opp. Also, the ARM's mali driver found on
> https://developer.arm.com/products/software/mali-drivers/midgard-kernel.
>
> Signed-off-by: Caesar Wang <[email protected]>
>
> ---
>
> Changes in v2:
> As Heiko comments on https://patchwork.kernel.org/patch/9835939/
> - interrupt-name use the lower case.
> - use the correct compatible "arm,mali-t860"
> - remove the clock name, since the mali only have one input clock.
>
> arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi | 33 ++++++++++++++++++++++++
> arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi | 33 ++++++++++++++++++++++++
> arch/arm64/boot/dts/rockchip/rk3399.dtsi | 12 +++++++++
> 3 files changed, 78 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
> index be7fe63..d8a120f 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3399-op1-opp.dtsi
> @@ -118,6 +118,35 @@
> opp-microvolt = <1250000>;
> };
> };
> +
> + gpu_opp_table: opp-table2 {
> + compatible = "operating-points-v2";
> +
> + opp00 {
> + opp-hz = /bits/ 64 <200000000>;
> + opp-microvolt = <800000>;
> + };
> + opp01 {
> + opp-hz = /bits/ 64 <297000000>;
> + opp-microvolt = <800000>;
> + };
> + opp02 {
> + opp-hz = /bits/ 64 <400000000>;
> + opp-microvolt = <825000>;
> + };
> + opp03 {
> + opp-hz = /bits/ 64 <500000000>;
> + opp-microvolt = <850000>;
> + };
> + opp04 {
> + opp-hz = /bits/ 64 <600000000>;
> + opp-microvolt = <925000>;
> + };
> + opp05 {
> + opp-hz = /bits/ 64 <800000000>;
> + opp-microvolt = <1075000>;
> + };
> + };
> };
>
> &cpu_l0 {
> @@ -143,3 +172,7 @@
> &cpu_b1 {
> operating-points-v2 = <&cluster1_opp>;
> };
> +
> +&gpu {
> + operating-points-v2 = <&gpu_opp_table>;
> +};
> diff --git a/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
> index c83460d..81617bc 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3399-opp.dtsi
> @@ -110,6 +110,35 @@
> opp-microvolt = <1200000>;
> };
> };
> +
> + gpu_opp_table: opp-table2 {
> + compatible = "operating-points-v2";
> +
> + opp00 {
> + opp-hz = /bits/ 64 <200000000>;
> + opp-microvolt = <800000>;
> + };
> + opp01 {
> + opp-hz = /bits/ 64 <297000000>;
> + opp-microvolt = <800000>;
> + };
> + opp02 {
> + opp-hz = /bits/ 64 <400000000>;
> + opp-microvolt = <825000>;
> + };
> + opp03 {
> + opp-hz = /bits/ 64 <500000000>;
> + opp-microvolt = <875000>;
> + };
> + opp04 {
> + opp-hz = /bits/ 64 <600000000>;
> + opp-microvolt = <925000>;
> + };
> + opp05 {
> + opp-hz = /bits/ 64 <800000000>;
> + opp-microvolt = <1100000>;
> + };
> + };
> };
>
> &cpu_l0 {
> @@ -135,3 +164,7 @@
> &cpu_b1 {
> operating-points-v2 = <&cluster1_opp>;
> };
> +
> +&gpu {
> + operating-points-v2 = <&gpu_opp_table>;
> +};
> diff --git a/arch/arm64/boot/dts/rockchip/rk3399.dtsi b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> index 1cbd7a2..77d67cb 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> @@ -1443,6 +1443,18 @@
> status = "disabled";
> };
>
> + gpu: gpu@ff9a0000 {
> + compatible = "arm,rk3399-mali", "arm,mali-t860";
applied for 4.14 after fixing the compatible to rockchip,rk3399-mali
Heiko
Am Montag, 17. Juli 2017, 16:14:30 CEST schrieb Caesar Wang:
> This patch enables the gpu and adds the mali-supply power for RK3399-GRU
> devices.
>
> Signed-off-by: Caesar Wang <[email protected]>
applied for 4.14
Thanks
Heiko
Hi Caesar,
Am Montag, 17. Juli 2017, 16:14:31 CEST schrieb Caesar Wang:
> 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.
>
> The Thermal power allocator governor works optimatly with two passive trip
> points, for the better performance we will use the trip-point0 with 70
> degree above which the governor control starts operating and trip-point1
> with 85 degree is the target temperature by controlling.
>
> Signed-off-by: Caesar Wang <[email protected]>
>
> ---
>
> Changes in v2:
> - foo@ will produce warnings when used without reg property.
> - update the commit to explain the two passive trip points changed.
>
> 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 77d67cb..6d8a5eb 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>;
Adjusting the coefficients should be a separate patch and the
commit message should explain how they were calculated and
why they are the exacter ones over the old values.
> };
> };
>
> @@ -690,24 +690,25 @@
> };
>
> thermal_zones: thermal-zones {
> - cpu_thermal: cpu {
> + soc_thermal: soc-thermal {
> polling-delay-passive = <100>;
> polling-delay = <1000>;
> + sustainable-power = <1000>;
>
> thermal-sensors = <&tsadc 0>;
>
> trips {
> - cpu_alert0: cpu_alert0 {
> + threshold: trip-point0 {
> temperature = <70000>;
> hysteresis = <2000>;
> type = "passive";
> };
> - cpu_alert1: cpu_alert1 {
> - temperature = <75000>;
> + target: trip-point1 {
> + temperature = <85000>;
> hysteresis = <2000>;
> type = "passive";
> };
> - cpu_crit: cpu_crit {
> + soc_crit: soc-crit {
> temperature = <95000>;
> hysteresis = <2000>;
> type = "critical";
> @@ -716,45 +717,31 @@
>
> cooling-maps {
> map0 {
> - trip = <&cpu_alert0>;
> + trip = <&target>;
still both maps use &target as trip point. Is that intentional
and if so, why is the &threshold trip point never referenced?
> cooling-device =
> - <&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> + <&cpu_l0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> + contribution = <4096>;
> };
> map1 {
> - trip = <&cpu_alert1>;
> + trip = <&target>;
> cooling-device =
> - <&cpu_l0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
> <&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> + contribution = <1024>;
> + };
> + map2 {
> + trip = <&target>;
> + cooling-device =
> + <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> + contribution = <4096>;
> };
> };
> };
>
> - gpu_thermal: gpu {
> + gpu_thermal: gpu-thermal {
> polling-delay-passive = <100>;
> polling-delay = <1000>;
>
> thermal-sensors = <&tsadc 1>;
> -
> - trips {
> - gpu_alert0: gpu_alert0 {
> - temperature = <75000>;
> - hysteresis = <2000>;
> - type = "passive";
> - };
> - gpu_crit: gpu_crit {
> - temperature = <95000>;
> - hysteresis = <2000>;
> - type = "critical";
> - };
> - };
> -
> - cooling-maps {
> - map0 {
> - trip = <&gpu_alert0>;
> - cooling-device =
> - <&cpu_b0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> - };
> - };
> };
> };
>
> @@ -1451,8 +1438,17 @@
> <GIC_SPI 21 IRQ_TYPE_LEVEL_HIGH 0>;
> interrupt-names = "gpu", "job", "mmu";
> clocks = <&cru ACLK_GPU>;
> + #cooling-cells = <2>;
> power-domains = <&power RK3399_PD_GPU>;
> status = "disabled";
> +
> + gpu_power_model: power_model {
> + compatible = "arm,mali-simple-power-model";
> + static-coefficient = <1079403>;
> + dynamic-coefficient = <977>;
> + ts = <32000 4700 (-80) 2>;
> + thermal-zone = "gpu-thermal";
> + };
You might want to have the gpu thermal work without the
power-model-thingy for now, so most likely just drop that
gpu-related change for now.
Heiko
On Tue, Jul 18, 2017 at 08:58:50AM +0800, Caesar Wang wrote:
> Rob,
>
> 在 2017年07月18日 04:07, Rob Herring 写道:
> > On Mon, Jul 17, 2017 at 04:14:28PM +0800, Caesar Wang wrote:
> > > This patch adds the MALI's power-model to set the IPA model to be used
> > > for power management.
> > What's IPA? India Pale Ale or Intermediate Physical Address?
>
> IPA is intelligent Power Allocator. (As the ARM introduced on
> https://developer.arm.com/open-source/intelligent-power-allocation)
>
> >
> > > Signed-off-by: Caesar Wang <[email protected]>
> > > ---
> > >
> > > Changes in v2: None
> > >
> > > Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt | 12 ++++++++++++
> > > 1 file changed, 12 insertions(+)
> > >
> > > diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> > > index a461e47..b616e6b 100644
> > > --- a/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> > > +++ b/Documentation/devicetree/bindings/gpu/arm,mali-midgard.txt
> > > @@ -37,6 +37,18 @@ Optional properties:
> > > - operating-points-v2 : Refer to Documentation/devicetree/bindings/power/opp.txt
> > > for details.
> > > +- power_model : Sets power model parameters. Note that this model was designed for the Juno
> > > + platform, and may not be suitable for other platforms. A structure containing :
> > > + - compatible: Should be arm,mali-simple-power-model
> > > + - dynamic-coefficient: Coefficient, in pW/(Hz V^2), which is multiplied
> > > + by v^2*f to calculate the dynamic power consumption.
> > > + - static-coefficient: Coefficient, in uW/V^3, which is multiplied by
> > > + v^3 to calculate the static power consumption.
> > > + - ts: An array containing coefficients for the temperature scaling
> > > + factor. This is used to scale the static power by a factor of
> > > + tsf/1000000, where tsf = ts[3]*T^3 + ts[2]*T^2 + ts[1]*T + ts[0],
> > > + and T = temperature in degrees.
> > > + - thermal-zone: A string identifying the thermal zone used for the GPU
> > This can all easily be implied by the compatible string. I'm not
> > inclined to accept something Mali specific here.
>
> Isn't arm,mali-midgard.txt document suit for Mali specific? :-)
It is, but I'm saying we shouldn't have something Mali specific here. It
should be something that works across different GPUs at least. IOW, get
some agreement with say adreno folks that these properties are useful and
I'll be more receptive.
> >
> > This looks *very* precise, but I'd be surprised if these values are any
> > more than magic values (at least the dynamic coef) adjusted until the
> > desired power/performance requirements are achieved. To put it another
> > way, why don't we have similar values for CPUs?
>
> These value was calculated by running full GPU process.
>
> CPU had the similar value for dtsi.
>
> Say: arch/arm64/boot/dts/rockchip/rk3399.dtsi
> cpu_b0: cpu@100 {
> ...
> dynamic-power-coefficient = <436>;
> ...
> };
Indeed. While it is documented for ARM CPUs, I don't see that it
is widely used as only the hi6220 dts defines it. So either support for
platforms is just missing, or upstream is not really using this property.
And if we are going to use this, then it needs to be documented in a
common location and moved out of arm/cpus.txt.
Rob
Hi Heiko,
Thanks your comments.
在 2017年07月23日 05:48, Heiko Stuebner 写道:
> Hi Caesar,
>
> Am Montag, 17. Juli 2017, 16:14:31 CEST schrieb Caesar Wang:
>> 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.
>>
>> The Thermal power allocator governor works optimatly with two passive trip
>> points, for the better performance we will use the trip-point0 with 70
>> degree above which the governor control starts operating and trip-point1
>> with 85 degree is the target temperature by controlling.
>>
>> Signed-off-by: Caesar Wang <[email protected]>
>>
>> ---
>>
>> Changes in v2:
>> - foo@ will produce warnings when used without reg property.
>> - update the commit to explain the two passive trip points changed.
>>
>> 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 77d67cb..6d8a5eb 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>;
> Adjusting the coefficients should be a separate patch and the
> commit message should explain how they were calculated and
> why they are the exacter ones over the old values.
Okay, i don't know why the dynamic-power-coefficient is 100 for
b-cluster before.:-)
>
>
>> };
>> };
>>
>> @@ -690,24 +690,25 @@
>> };
>>
>> thermal_zones: thermal-zones {
>> - cpu_thermal: cpu {
>> + soc_thermal: soc-thermal {
>> polling-delay-passive = <100>;
>> polling-delay = <1000>;
>> + sustainable-power = <1000>;
>>
>> thermal-sensors = <&tsadc 0>;
>>
>> trips {
>> - cpu_alert0: cpu_alert0 {
>> + threshold: trip-point0 {
>> temperature = <70000>;
>> hysteresis = <2000>;
>> type = "passive";
>> };
>> - cpu_alert1: cpu_alert1 {
>> - temperature = <75000>;
>> + target: trip-point1 {
>> + temperature = <85000>;
>> hysteresis = <2000>;
>> type = "passive";
>> };
>> - cpu_crit: cpu_crit {
>> + soc_crit: soc-crit {
>> temperature = <95000>;
>> hysteresis = <2000>;
>> type = "critical";
>> @@ -716,45 +717,31 @@
>>
>> cooling-maps {
>> map0 {
>> - trip = <&cpu_alert0>;
>> + trip = <&target>;
> still both maps use &target as trip point. Is that intentional
> and if so, why is the &threshold trip point never referenced?
For the power allocator governor, the &threshold trip point just control
starts operating, not need for map.
For other governor (e.g: step_wise) will need the first trip point.
Looks like we have to think about how to support them.
> };
> };
>
> @@ -1451,8 +1438,17 @@
> <GIC_SPI 21 IRQ_TYPE_LEVEL_HIGH 0>;
> interrupt-names = "gpu", "job", "mmu";
> clocks = <&cru ACLK_GPU>;
> + #cooling-cells = <2>;
> power-domains = <&power RK3399_PD_GPU>;
> status = "disabled";
> +
> + gpu_power_model: power_model {
> + compatible = "arm,mali-simple-power-model";
> + static-coefficient = <1079403>;
> + dynamic-coefficient = <977>;
> + ts = <32000 4700 (-80) 2>;
> + thermal-zone = "gpu-thermal";
> + };
> You might want to have the gpu thermal work without the
> power-model-thingy for now, so most likely just drop that
> gpu-related change for now.
Well, I guess the mediatek/hisilicon/rockchip ARM SoCs should support
the gpu's power mode with private kernel,
not in the upstream kernel.
-Caesar
>
>
> Heiko
>
> _______________________________________________
> Linux-rockchip mailing list
> [email protected]
> http://lists.infradead.org/mailman/listinfo/linux-rockchip