On Mon, Apr 30, 2018 at 05:48:29PM +0200, Bartlomiej Zolnierkiewicz wrote: > On Monday, April 30, 2018 05:34:31 PM Daniel Lezcano wrote: > > On Thu, Apr 26, 2018 at 01:51:21PM +0200, Bartlomiej Zolnierkiewicz wrote: > > > of_thermal_get_ntrips() may return value bigger than supported > > > by a given SoC (i.e. on Exynos5422/5800) > > > > Can you elaborate a bit ? > > Odroid-XU3 DTS file [1] define 6 thermal trip points (2 passive ones) > while data->ntrip is 4, the current code works fine by accident as > the threshold values for trip points 5 & 6 don't fit into 32-bits > threshold value (however since they are passive ones this is okay).
Ah, I see. data->ntrip is the SoC specific tmu max value capping what is defined in the DT, right ? > Of course the code for handling passive trip points still needs to > be fixed to properly handle all odd cases (which are not present in > current DTS files). > > [1] arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi Yeah, it is definitively a good idea of cleaning up this driver :/ Reviewed-by: Daniel Lezcano <[email protected]> -- <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook | <http://twitter.com/#!/linaroorg> Twitter | <http://www.linaro.org/linaro-blog/> Blog
