return0e has quit [Remote host closed the connection]
return0e has joined #linux-rockchip
vstehle has quit [Ping timeout: 250 seconds]
drrty has quit [Ping timeout: 246 seconds]
drrty has joined #linux-rockchip
vagrantc has joined #linux-rockchip
camus has joined #linux-rockchip
kaspter has quit [Ping timeout: 250 seconds]
camus is now known as kaspter
chewitt has quit [Quit: Adios!]
samueldr has quit [Ping timeout: 246 seconds]
samueldr has joined #linux-rockchip
drrty has quit [Ping timeout: 246 seconds]
vstehle has joined #linux-rockchip
return0e_ has joined #linux-rockchip
return0e has quit [Ping timeout: 245 seconds]
afaerber has joined #linux-rockchip
vagrantc has quit [Quit: leaving]
default__ is now known as ldevulder
<dlezcano>
mmind00, while playing with the thermal I noticed a misuse of the thermal cooling device and the thermal zones
<dlezcano>
mmind00, performances are considerably dropped with the Intelligent Power Allocator
<dlezcano>
mmind00, GPU and big CPUs have the same cooling device, hence the IPA allocates max power to CPU by capping the GPU power
<dlezcano>
mmind00, but actually it caps the big CPUs and we end up with the big CPUs running at a lowest freq
<mmind00>
dlezcano: hmm, found that in the dtsi ... looking at the chromeos kernel, the thermal_zones look quite a bit different ... any suggestions on how to fix that? I haven't done that much with the whole thermal handling so far
field^Zzz4 has joined #linux-rockchip
<dlezcano>
mmind00, I'll have a look
<mmind00>
dlezcano: thanks :-)
<dlezcano>
mmind00, however removing simply the GPU thermal zone breaks the driver
<dlezcano>
mmind00, because it expects all the sensor to be registered
<dlezcano>
mmind00, so we have to fix the thermal driver also
<dlezcano>
mmind00, by registering only the defined sensor find in the DT