Howdy,
I have an X15 I'm using to push with kernel builds - I want to make sure I'm reading the temperature correctly, because I'm concerned with what I'm seeing through thermal/thermal_zone*:
Hi,
I might be wrong but i think there is be a bug there.
According to http://www.ti.com/product/am5728, operating temp should be 0-90°C
And it seems to me that the DT thermal-zone used for AM5728 is this one : http://git.ti.com/ti-linux-kernel/ti-linux-kernel/blobs/ti-linux-4.1.y/arch/arm/boot/dts/omap4-cpu-thermal.dtsi
--> where alert temps = 100°C and crit temp = 125°C
That doesn't make sense to me.
I hope this message can help
N.
In the end I installed lm-sensors and just use watch sensors
Every 2.0s: sensors Mon Nov 30 15:39:24 2015
gpio_fan-isa-0000
Adapter: ISA adapter
fan1: 13000 RPM (min = 0 RPM, max = 13000 RPM)
tmp102-i2c-0-48
Adapter: OMAP I2C adapter
temp1: +55.9°C (high = +160.0°C, hyst = +150.0°C)
dspeve-virtual-0
Adapter: Virtual device
temp1: +67.4°C (crit = +110.0°C)
iva-virtual-0
Adapter: Virtual device
temp1: +69.0°C (crit = +110.0°C)
One can definitely heat this up,
[ 3.669098] thermal thermal_zone3: critical temperature reached(105 C),shutting down
[ 3.680808] reboot: Power down
Inane question - for working with this on a bench/no case, would a silly USB-powered (by the X15) fan blowing air across the surface be useful?
In the end I installed lm-sensors and just use watch sensors
Every 2.0s: sensors
Mon Nov 30 15:39:24 2015gpio_fan-isa-0000
Adapter: ISA adapter
fan1: 13000 RPM (min = 0 RPM, max = 13000 RPM)tmp102-i2c-0-48
Adapter: OMAP I2C adapter
temp1: +55.9°C (high = +160.0°C, hyst = +150.0°C)dspeve-virtual-0
Adapter: Virtual device
temp1: +67.4°C (crit = +110.0°C)iva-virtual-0
Adapter: Virtual device
temp1: +69.0°C (crit = +110.0°C)One can definitely heat this up,
[ 3.669098] thermal thermal_zone3: critical temperature reached(105
C),shutting down
this looks like the tmp102 kicking in..
[ 3.680808] reboot: Power down
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/hwmon/tmp102.c#n64
Comparing the register dumps with
http://www.ti.com/lit/ds/symlink/tmp102.pdf -> I think the
implementation might be flawed. trying to dig in.
bah.. actually, I take it back.. the trouble is in lower temp, which
is not the case here!
Hi
the dragonboard guys seem to have fixed a similar problem by patching their kernel device tree https://www.96boards.org/forums/topic/recommended-heatsinkfan-for-410c/#post-9838
after having a look at X15 device tree, i found out that cpu thermal zone is defined in file omap4-cpu-thermal.dtsi where alertTemp=100°C and critTemp=125°C
i don't understand why those values are used for am5728
as far as i understand the datasheet and device tree, crit temp should be set 105°C
hope this helps
N.
Indeed!
http://www.ti.com/product/AM5728/datasheet/specifications#POH
Hmm... Looks like reuse from DRA7 is a dual edged sword 
Folks,
Should we create an am57xx temp dtsi for usage for AM57xx platforms ->
alert at 90 and critical at 105C?