| Message ID | 5bb39f3115df1a487d717d3ae87e523b03749379.1573908197.git.robin.murphy@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | arm64: dts: rockchip: Fix NanoPC-T4 cooling maps | expand |
Am Samstag, 16. November 2019, 13:47:19 CET schrieb Robin Murphy: > Although it appeared to follow logically from the bindings, apparently > the thermal framework can't properly cope with a single cooling device > being shared between multiple maps. The CPU zone is probably easier to > overheat, so remove the references to the (optional) fan from the GPU > cooling zone to avoid things getting confused. Hopefully GPU-intensive > tasks will leak enough heat across to the CPU zone to still hit the > fan trips before reaching critical GPU temperatures. > > Signed-off-by: Robin Murphy <robin.murphy@arm.com> applied, after little bit of editing. > &i2s0 { > status = "okay"; > }; although my nano-pc dts does not seem to have that &i2s0 node yet. Did I miss a patch somewhere? Heiko
On 18/11/2019 12:52 am, Heiko Stuebner wrote: > Am Samstag, 16. November 2019, 13:47:19 CET schrieb Robin Murphy: >> Although it appeared to follow logically from the bindings, apparently >> the thermal framework can't properly cope with a single cooling device >> being shared between multiple maps. The CPU zone is probably easier to >> overheat, so remove the references to the (optional) fan from the GPU >> cooling zone to avoid things getting confused. Hopefully GPU-intensive >> tasks will leak enough heat across to the CPU zone to still hit the >> fan trips before reaching critical GPU temperatures. >> >> Signed-off-by: Robin Murphy <robin.murphy@arm.com> > > applied, after little bit of editing. > >> &i2s0 { >> status = "okay"; >> }; > > although my nano-pc dts does not seem to have that &i2s0 node yet. > Did I miss a patch somewhere? Oops, no, my fault - there were more changes in the middle of my local branch which I thought were entirely orthogonal. I should have checked the diff context more closely... thanks for cleaning it up! Robin.
diff --git a/arch/arm64/boot/dts/rockchip/rk3399-nanopc-t4.dts b/arch/arm64/boot/dts/rockchip/rk3399-nanopc-t4.dts index f399853c635a..f81a8429f139 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-nanopc-t4.dts +++ b/arch/arm64/boot/dts/rockchip/rk3399-nanopc-t4.dts @@ -113,33 +113,6 @@ }; }; -&gpu_thermal { - trips { - gpu_warm: gpu_warm { - temperature = <55000>; - hysteresis = <2000>; - type = "active"; - }; - - gpu_hot: gpu_hot { - temperature = <65000>; - hysteresis = <2000>; - type = "active"; - }; - }; - cooling-maps { - map1 { - trip = <&gpu_warm>; - cooling-device = <&fan THERMAL_NO_LIMIT 1>; - }; - - map2 { - trip = <&gpu_hot>; - cooling-device = <&fan 2 THERMAL_NO_LIMIT>; - }; - }; -}; - &i2s0 { status = "okay"; };
Although it appeared to follow logically from the bindings, apparently the thermal framework can't properly cope with a single cooling device being shared between multiple maps. The CPU zone is probably easier to overheat, so remove the references to the (optional) fan from the GPU cooling zone to avoid things getting confused. Hopefully GPU-intensive tasks will leak enough heat across to the CPU zone to still hit the fan trips before reaching critical GPU temperatures. Signed-off-by: Robin Murphy <robin.murphy@arm.com> --- .../boot/dts/rockchip/rk3399-nanopc-t4.dts | 27 ------------------- 1 file changed, 27 deletions(-)