On 23/04/2019 14:56, Amit Kucheria wrote: > On Tue, Apr 23, 2019 at 3:01 PM Marc Gonzalez wrote: > >> On 02/04/2019 20:46, Amit Kucheria wrote: >> >>> On Tue, Apr 2, 2019 at 8:58 PM Marc Gonzalez wrote: >>> >>>> On 29/03/2019 11:12, Amit Kucheria wrote: >>>> >>>>> Changes since v1: >>>>> - Devices that are not associated with cooling-maps use the 'hot' type >>>>> - Remove critical trip from such devices too and restrict that type of CPUs >>>>> - For 8916, add the sensor on the modem >>>>> >>>>> Description: >>>>> - Expose all temperature sensors on msm8916, msm996, msm8998, sdm845 >>>>> - split up the register address map for msm8998 >>>>> - standardize names of the various thermal-zones across boards to make it >>>>> easy for test scripts to parse >>>> >>>> This is what I see on msm8998 with your patch series applied on top of >>>> yesterday's next: >>>> >>>> # dmesg | grep of_parse_thermal_zones >>>> of_parse_thermal_zones: build cpu0-thermal zone 0 >>>> of_parse_thermal_zones: build cpu1-thermal zone 0 >>>> of_parse_thermal_zones: build cpu2-thermal zone 0 >>>> of_parse_thermal_zones: build cpu3-thermal zone 0 >>>> of_parse_thermal_zones: build cpu4-thermal zone 0 >>>> of_parse_thermal_zones: build cpu5-thermal zone 0 >>>> of_parse_thermal_zones: build cpu6-thermal zone 0 >>>> of_parse_thermal_zones: build cpu7-thermal zone 0 >>>> of_parse_thermal_zones: build gpu-thermal-bottom zone 0 >>>> of_parse_thermal_zones: build gpu-thermal-top zone 0 >>>> of_parse_thermal_zones: build cluster0-mhm-thermal zone -22 >>>> of_parse_thermal_zones: build cluster1-mhm-thermal zone -22 >>>> of_parse_thermal_zones: build cluster1-l2-thermal zone 0 >>>> of_parse_thermal_zones: build modem-thermal zone 0 >>>> of_parse_thermal_zones: build mem-thermal zone 0 >>>> of_parse_thermal_zones: build wlan-thermal zone 0 >>>> of_parse_thermal_zones: build q6-dsp-thermal zone 0 >>>> of_parse_thermal_zones: build camera-thermal zone 0 >>>> of_parse_thermal_zones: build multimedia-thermal zone 0 >>>> of_parse_thermal_zones: build pm8998 zone 0 >>>> of_parse_thermal_zones: build battery-thermal zone 0 >>>> of_parse_thermal_zones: build skin-thermal zone 0 >>>> >>>> NB: -EINVAL for cluster0-mhm-thermal and cluster1-mhm-thermal >>>> >>>> >>>> # cat /sys/devices/virtual/thermal/thermal_zone*/temp >>>> 35800 >>>> 35200 >>>> 34500 >>>> 34800 >>>> 34500 >>>> 34100 >>>> 34100 >>>> 33500 >>>> 33800 >>>> cat: /sys/devices/virtual/thermal/thermal_zone17/temp: Invalid argument >>>> 36100 >>>> cat: /sys/devices/virtual/thermal/thermal_zone19/temp: Invalid argument >>>> 35500 >>>> 36100 >>>> 35800 >>>> 35200 >>>> 35500 >>>> 36100 >>>> cat: /sys/devices/virtual/thermal/thermal_zone8/temp: Invalid argument >>>> cat: /sys/devices/virtual/thermal/thermal_zone9/temp: Invalid argument >>>> >>>> >>>> Are the failures expected? >>>> (cluster*-mhm-thermal and thermal_zone*/temp) >>> >>> No they are not. Unfortunately, I didn't have HW to test, so >>> appreciate the help with testing. I'll check the 8, 9, 17 and 19 again >>> before resubmitting. >> >> Looks like Andy merged this series? > > Argh! my bad, I completely forgot about the need to fix this up. I > need to get me one of these boards. > >> https://git.kernel.org/pub/scm/linux/kernel/git/agross/linux.git/log/?h=for-next >> >> I was expecting a respin, or a discussion of the failures. >> >> Slightly confused. What next? > > We can either revert that one patch for the msm8998, but I think it > might be better if I send a fixup patch that could fixup those 4 zones > instead. > > Does that work for you? Sure. I can give you a Tested-by once everything works on the MEDIABOX I have. You mention the 4 thermal_zone*/temp failures. What about the cluster0-mhm-thermal and cluster1-mhm-thermal failures? Regards.
