It's been years since Adreno has been registered as a cooling device,
yet only so many platforms had the correct DT setup for it. This series
attempts to hook it up on most supported snapdragons.
Signed-off-by: Konrad Dybcio <[email protected]>
---
Konrad Dybcio (12):
arm64: dts: qcom: msm8916: Hook up GPU cooling device
arm64: dts: qcom: msm8939: Hook up GPU cooling device
arm64: dts: qcom: sc8180x: Hook up GPU cooling device
arm64: dts: qcom: sdm845: Hook up GPU cooling device
arm64: dts: qcom: sm6115: Hook up GPU cooling device
arm64: dts: qcom: sm6115: Mark GPU @ 125C critical
arm64: dts: qcom: sm8150: Hook up GPU cooling device
arm64: dts: qcom: sm8250: Hook up GPU cooling device
arm64: dts: qcom: sm8350: Hook up GPU cooling device
arm64: dts: qcom: sm8450: Hook up GPU cooling device
arm64: dts: qcom: sm8550: Hook up GPU cooling device
arm64: dts: qcom: sdm630: Hook up GPU cooling device
arch/arm64/boot/dts/qcom/msm8916.dtsi | 9 ++++++
arch/arm64/boot/dts/qcom/msm8939.dtsi | 9 ++++++
arch/arm64/boot/dts/qcom/sc8180x.dtsi | 20 ++++++++++--
arch/arm64/boot/dts/qcom/sdm630.dtsi | 9 ++++++
arch/arm64/boot/dts/qcom/sdm845.dtsi | 19 ++++++++++--
arch/arm64/boot/dts/qcom/sm6115.dtsi | 13 ++++++--
arch/arm64/boot/dts/qcom/sm8150.dtsi | 19 ++++++++++--
arch/arm64/boot/dts/qcom/sm8250.dtsi | 19 ++++++++++--
arch/arm64/boot/dts/qcom/sm8350.dtsi | 19 ++++++++++--
arch/arm64/boot/dts/qcom/sm8450.dtsi | 19 ++++++++++--
arch/arm64/boot/dts/qcom/sm8550.dtsi | 57 +++++++++++++++++++++++++++++++++++
11 files changed, 198 insertions(+), 14 deletions(-)
---
base-commit: ab0b3e6ef50d305278b1971891cf1d82ab050b35
change-id: 20240102-topic-gpu_cooling-155d4d4b051e
Best regards,
--
Konrad Dybcio <[email protected]>
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/msm8916.dtsi | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/msm8916.dtsi b/arch/arm64/boot/dts/qcom/msm8916.dtsi
index 7f8327b0dbdb..31ceb1df550a 100644
--- a/arch/arm64/boot/dts/qcom/msm8916.dtsi
+++ b/arch/arm64/boot/dts/qcom/msm8916.dtsi
@@ -1794,6 +1794,8 @@ gpu: gpu@1c00000 {
power-domains = <&gcc OXILI_GDSC>;
operating-points-v2 = <&gpu_opp_table>;
iommus = <&gpu_iommu 1>, <&gpu_iommu 2>;
+ #cooling-cells = <2>;
+
status = "disabled";
gpu_opp_table: opp-table {
@@ -2697,6 +2699,13 @@ gpu-thermal {
thermal-sensors = <&tsens 2>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
gpu_alert0: trip-point0 {
temperature = <75000>;
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/msm8939.dtsi | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/msm8939.dtsi b/arch/arm64/boot/dts/qcom/msm8939.dtsi
index 29f6bd9df2eb..e0521a060cf8 100644
--- a/arch/arm64/boot/dts/qcom/msm8939.dtsi
+++ b/arch/arm64/boot/dts/qcom/msm8939.dtsi
@@ -1439,6 +1439,8 @@ gpu: gpu@1c00000 {
power-domains = <&gcc OXILI_GDSC>;
operating-points-v2 = <&opp_table>;
iommus = <&gpu_iommu 1>, <&gpu_iommu 2>;
+ #cooling-cells = <2>;
+
status = "disabled";
opp_table: opp-table {
@@ -2468,6 +2470,13 @@ gpu-thermal {
thermal-sensors = <&tsens 3>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
gpu_alert0: trip-point0 {
temperature = <75000>;
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sc8180x.dtsi | 20 ++++++++++++++++++--
1 file changed, 18 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sc8180x.dtsi b/arch/arm64/boot/dts/qcom/sc8180x.dtsi
index 0430d99091e3..b1227e04429c 100644
--- a/arch/arm64/boot/dts/qcom/sc8180x.dtsi
+++ b/arch/arm64/boot/dts/qcom/sc8180x.dtsi
@@ -2173,6 +2173,8 @@ gpu: gpu@2c00000 {
interconnect-names = "gfx-mem";
qcom,gmu = <&gmu>;
+ #cooling-cells = <2>;
+
status = "disabled";
gpu_opp_table: opp-table {
@@ -3880,8 +3882,15 @@ gpu-top-thermal {
thermal-sensors = <&tsens0 15>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- trip-point0 {
+ gpu_top_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
@@ -4030,8 +4039,15 @@ gpu-bottom-thermal {
thermal-sensors = <&tsens1 11>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- trip-point0 {
+ gpu_bottom_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sdm845.dtsi | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi
index c2244824355a..1c2d01f0cf07 100644
--- a/arch/arm64/boot/dts/qcom/sdm845.dtsi
+++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi
@@ -4760,6 +4760,7 @@ gpu: gpu@5000000 {
operating-points-v2 = <&gpu_opp_table>;
qcom,gmu = <&gmu>;
+ #cooling-cells = <2>;
interconnects = <&mem_noc MASTER_GFX3D 0 &mem_noc SLAVE_EBI1 0>;
interconnect-names = "gfx-mem";
@@ -5602,8 +5603,15 @@ gpu-top-thermal {
thermal-sensors = <&tsens0 11>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu1_alert0: trip-point0 {
+ gpu_top_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
@@ -5617,8 +5625,15 @@ gpu-bottom-thermal {
thermal-sensors = <&tsens0 12>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu2_alert0: trip-point0 {
+ gpu_bottom_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm6115.dtsi | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi
index 160e098f1075..a8c819d53104 100644
--- a/arch/arm64/boot/dts/qcom/sm6115.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi
@@ -14,6 +14,7 @@
#include <dt-bindings/interconnect/qcom,sm6115.h>
#include <dt-bindings/interrupt-controller/arm-gic.h>
#include <dt-bindings/power/qcom-rpmpd.h>
+#include <dt-bindings/thermal/thermal.h>
/ {
interrupt-parent = <&intc>;
@@ -1646,6 +1647,7 @@ gpu: gpu@5900000 {
nvmem-cells = <&gpu_speed_bin>;
nvmem-cell-names = "speed_bin";
+ #cooling-cells = <2>;
status = "disabled";
@@ -3294,8 +3296,15 @@ gpu-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens0 15>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- trip-point0 {
+ gpu_alert0: trip-point0 {
temperature = <115000>;
hysteresis = <5000>;
type = "passive";
--
2.43.0
If the GPU ever reaches this temperature, the "critical" signal shuold
definitely be propagated. Fix the wrong type.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm6115.dtsi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi
index a8c819d53104..be51cbaeeb7e 100644
--- a/arch/arm64/boot/dts/qcom/sm6115.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi
@@ -3313,7 +3313,7 @@ gpu_alert0: trip-point0 {
trip-point1 {
temperature = <125000>;
hysteresis = <1000>;
- type = "passive";
+ type = "critical";
};
};
};
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones. Also, update the trip point label
to be more telling, while at it.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm8150.dtsi | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi
index 761a6757dc26..26dbda713e51 100644
--- a/arch/arm64/boot/dts/qcom/sm8150.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi
@@ -2198,6 +2198,7 @@ gpu: gpu@2c00000 {
nvmem-cells = <&gpu_speed_bin>;
nvmem-cell-names = "speed_bin";
+ #cooling-cells = <2>;
status = "disabled";
@@ -5101,8 +5102,15 @@ gpu-top-thermal {
thermal-sensors = <&tsens0 15>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu1_alert0: trip-point0 {
+ gpu_top_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
@@ -5281,8 +5289,15 @@ gpu-bottom-thermal {
thermal-sensors = <&tsens1 11>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu2_alert0: trip-point0 {
+ gpu_bottom_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones. Also, update the trip point label
to be more telling, while at it.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm8350.dtsi | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8350.dtsi b/arch/arm64/boot/dts/qcom/sm8350.dtsi
index e78c83a897c2..1c6f80b14d09 100644
--- a/arch/arm64/boot/dts/qcom/sm8350.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8350.dtsi
@@ -1847,6 +1847,7 @@ gpu: gpu@3d00000 {
operating-points-v2 = <&gpu_opp_table>;
qcom,gmu = <&gmu>;
+ #cooling-cells = <2>;
status = "disabled";
@@ -4214,8 +4215,15 @@ gpu-top-thermal {
thermal-sensors = <&tsens1 1>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu1_alert0: trip-point0 {
+ gpu_top_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <1000>;
type = "hot";
@@ -4229,8 +4237,15 @@ gpu-bottom-thermal {
thermal-sensors = <&tsens1 2>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu2_alert0: trip-point0 {
+ gpu_bottom_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <1000>;
type = "hot";
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones. Also, update the trip point label
to be more telling, while at it.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm8450.dtsi | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8450.dtsi b/arch/arm64/boot/dts/qcom/sm8450.dtsi
index 01e4dfc4babd..8f5a093c8828 100644
--- a/arch/arm64/boot/dts/qcom/sm8450.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8450.dtsi
@@ -2038,6 +2038,7 @@ gpu: gpu@3d00000 {
operating-points-v2 = <&gpu_opp_table>;
qcom,gmu = <&gmu>;
+ #cooling-cells = <2>;
status = "disabled";
@@ -4890,6 +4891,13 @@ gpu-top-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens0 14>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -4909,7 +4917,7 @@ reset-mon-cfg {
type = "passive";
};
- gpu0_tj_cfg: tj-cfg {
+ gpu_top_alert0: trip-point0 {
temperature = <95000>;
hysteresis = <5000>;
type = "passive";
@@ -4922,6 +4930,13 @@ gpu-bottom-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens0 15>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -4941,7 +4956,7 @@ reset-mon-cfg {
type = "passive";
};
- gpu1_tj_cfg: tj-cfg {
+ gpu_bottom_alert0: trip-point0 {
temperature = <95000>;
hysteresis = <5000>;
type = "passive";
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones. Also, unify the naming scheme of the
thermal zones across the tree while at it.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm8550.dtsi | 57 ++++++++++++++++++++++++++++++++++++
1 file changed, 57 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/sm8550.dtsi b/arch/arm64/boot/dts/qcom/sm8550.dtsi
index ee1ba5a8c8fc..692cd85f03fb 100644
--- a/arch/arm64/boot/dts/qcom/sm8550.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8550.dtsi
@@ -2012,6 +2012,7 @@ gpu: gpu@3d00000 {
operating-points-v2 = <&gpu_opp_table>;
qcom,gmu = <&gmu>;
+ #cooling-cells = <2>;
status = "disabled";
@@ -5304,6 +5305,13 @@ gpuss-0-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 1>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu0_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5336,6 +5344,13 @@ gpuss-1-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 2>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu1_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5368,6 +5383,13 @@ gpuss-2-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 3>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu2_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5400,6 +5422,13 @@ gpuss-3-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 4>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu3_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5432,6 +5461,13 @@ gpuss-4-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 5>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu4_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5464,6 +5500,13 @@ gpuss-5-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 6>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu5_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5496,6 +5539,13 @@ gpuss-6-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 7>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu6_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
@@ -5528,6 +5578,13 @@ gpuss-7-thermal {
polling-delay = <0>;
thermal-sensors = <&tsens2 8>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu7_junction_config>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
thermal-engine-config {
temperature = <125000>;
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sdm630.dtsi | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/sdm630.dtsi b/arch/arm64/boot/dts/qcom/sdm630.dtsi
index 775700f78e0f..fc06665861e2 100644
--- a/arch/arm64/boot/dts/qcom/sdm630.dtsi
+++ b/arch/arm64/boot/dts/qcom/sdm630.dtsi
@@ -13,6 +13,7 @@
#include <dt-bindings/power/qcom-rpmpd.h>
#include <dt-bindings/gpio/gpio.h>
#include <dt-bindings/interrupt-controller/arm-gic.h>
+#include <dt-bindings/thermal/thermal.h>
#include <dt-bindings/soc/qcom,apr.h>
/ {
@@ -1120,6 +1121,7 @@ adreno_gpu: gpu@5000000 {
interconnect-names = "gfx-mem";
operating-points-v2 = <&gpu_sdm630_opp_table>;
+ #cooling-cells = <2>;
status = "disabled";
@@ -2580,6 +2582,13 @@ gpu-thermal {
thermal-sensors = <&tsens 8>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_alert0>;
+ cooling-device = <&adreno_gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
gpu_alert0: trip-point0 {
temperature = <90000>;
--
2.43.0
In order to allow for throttling the GPU, hook up the cooling device
to the respective thermal zones. Also, update the trip point label
to be more telling, while at it.
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sm8250.dtsi | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi b/arch/arm64/boot/dts/qcom/sm8250.dtsi
index 760501c1301a..03bf49818798 100644
--- a/arch/arm64/boot/dts/qcom/sm8250.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi
@@ -2888,6 +2888,7 @@ gpu: gpu@3d00000 {
nvmem-cells = <&gpu_speed_bin>;
nvmem-cell-names = "speed_bin";
+ #cooling-cells = <2>;
status = "disabled";
@@ -6791,8 +6792,15 @@ gpu-top-thermal {
thermal-sensors = <&tsens0 15>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_top_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu1_alert0: trip-point0 {
+ gpu_top_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
@@ -6926,8 +6934,15 @@ gpu-bottom-thermal {
thermal-sensors = <&tsens1 8>;
+ cooling-maps {
+ map0 {
+ trip = <&gpu_bottom_alert0>;
+ cooling-device = <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+ };
+ };
+
trips {
- gpu2_alert0: trip-point0 {
+ gpu_bottom_alert0: trip-point0 {
temperature = <90000>;
hysteresis = <2000>;
type = "hot";
--
2.43.0
On Tue, 2 Jan 2024 at 15:35, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/msm8916.dtsi | 9 +++++++++
> 1 file changed, 9 insertions(+)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:35, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/msm8939.dtsi | 9 +++++++++
> 1 file changed, 9 insertions(+)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:35, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sc8180x.dtsi | 20 ++++++++++++++++++--
> 1 file changed, 18 insertions(+), 2 deletions(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:36, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sdm845.dtsi | 19 +++++++++++++++++--
> 1 file changed, 17 insertions(+), 2 deletions(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:36, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm6115.dtsi | 11 ++++++++++-
> 1 file changed, 10 insertions(+), 1 deletion(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:36, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones. Also, update the trip point label
> to be more telling, while at it.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm8150.dtsi | 19 +++++++++++++++++--
> 1 file changed, 17 insertions(+), 2 deletions(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:36, Konrad Dybcio <[email protected]> wrote:
>
> If the GPU ever reaches this temperature, the "critical" signal shuold
> definitely be propagated. Fix the wrong type.
I hope it will be shut down before reaching 125°C
Reviewed-by: Dmitry Baryshkov <[email protected]>
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm6115.dtsi | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi
> index a8c819d53104..be51cbaeeb7e 100644
> --- a/arch/arm64/boot/dts/qcom/sm6115.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi
> @@ -3313,7 +3313,7 @@ gpu_alert0: trip-point0 {
> trip-point1 {
> temperature = <125000>;
> hysteresis = <1000>;
> - type = "passive";
> + type = "critical";
> };
> };
> };
>
> --
> 2.43.0
>
>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:37, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones. Also, update the trip point label
> to be more telling, while at it.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm8250.dtsi | 19 +++++++++++++++++--
> 1 file changed, 17 insertions(+), 2 deletions(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:37, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones. Also, update the trip point label
> to be more telling, while at it.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm8350.dtsi | 19 +++++++++++++++++--
> 1 file changed, 17 insertions(+), 2 deletions(-)
>
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:38, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sdm630.dtsi | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:37, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones. Also, update the trip point label
> to be more telling, while at it.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm8450.dtsi | 19 +++++++++++++++++--
> 1 file changed, 17 insertions(+), 2 deletions(-)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 2 Jan 2024 at 15:38, Konrad Dybcio <[email protected]> wrote:
>
> In order to allow for throttling the GPU, hook up the cooling device
> to the respective thermal zones. Also, unify the naming scheme of the
> thermal zones across the tree while at it.
>
> Signed-off-by: Konrad Dybcio <[email protected]>
> ---
> arch/arm64/boot/dts/qcom/sm8550.dtsi | 57 ++++++++++++++++++++++++++++++++++++
> 1 file changed, 57 insertions(+)
Reviewed-by: Dmitry Baryshkov <[email protected]>
--
With best wishes
Dmitry
On Tue, 02 Jan 2024 14:34:04 +0100, Konrad Dybcio wrote:
> It's been years since Adreno has been registered as a cooling device,
> yet only so many platforms had the correct DT setup for it. This series
> attempts to hook it up on most supported snapdragons.
>
>
Applied, thanks!
[01/12] arm64: dts: qcom: msm8916: Hook up GPU cooling device
commit: 04ee83047e4f7ba3e12f96315c08e146d10e1b7b
[02/12] arm64: dts: qcom: msm8939: Hook up GPU cooling device
commit: 93c4e1fb9dd9e56b09adc4a85692de3a97209960
[03/12] arm64: dts: qcom: sc8180x: Hook up GPU cooling device
commit: f48cea4a6fffa3651ced6f8f348f62ff4f268d87
[04/12] arm64: dts: qcom: sdm845: Hook up GPU cooling device
commit: 08b1b831e1a91793a20a41e090f969f41195a03d
[05/12] arm64: dts: qcom: sm6115: Hook up GPU cooling device
commit: de5e4e88ac148eb3aa9759095c91cbf810901774
[06/12] arm64: dts: qcom: sm6115: Mark GPU @ 125C critical
commit: 834932689f2f545f05ee1bbaad5e5e7f4c9fe136
[07/12] arm64: dts: qcom: sm8150: Hook up GPU cooling device
commit: f18c63a8213913394e20b07790e1ddd60d42f44c
[08/12] arm64: dts: qcom: sm8250: Hook up GPU cooling device
commit: fb18c893816b37c907c1c0883b92e19dcad58c97
[09/12] arm64: dts: qcom: sm8350: Hook up GPU cooling device
commit: 43c925e4567390bbf7adb4778b4cae3c93b924ec
[10/12] arm64: dts: qcom: sm8450: Hook up GPU cooling device
commit: 36fd56ab4d1d13adb9ced8592dc51e009240442d
[11/12] arm64: dts: qcom: sm8550: Hook up GPU cooling device
commit: 6a464089284a6192aac8405918aeed92b3bba8b6
[12/12] arm64: dts: qcom: sdm630: Hook up GPU cooling device
commit: ac68c7d3912d09878b8b42866b47fdd6459f5703
Best regards,
--
Bjorn Andersson <[email protected]>