Reported by Nikolaus Schaller, May 24, 2014
* cpufreq is max. 800 MHz root@gta04:~# cat /proc/cmdline console=ttyO2,115200n8 mux=GTA04A3+ mpurate=1000 vram=12M cma=12M omapfb.vram=0:8M,1:4M omapfb.mode=dvi:1024x768MR-16@60 omapfb.debug=y omapdss.def_disp=lt root@gta04:~# cpufreq-info cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009 Report errors and bugs to email@example.com, please. analyzing CPU 0: driver: generic_cpu0 CPUs which run at the same hardware frequency: 0 CPUs which need to have their frequency coordinated by software: 0 maximum transition latency: 300 us. hardware limits: 300 MHz - 800 MHz available frequency steps: 300 MHz, 600 MHz, 800 MHz available cpufreq governors: conservative, userspace, powersave, ondemand, performance current policy: frequency should be within 300 MHz and 800 MHz. The governor "ondemand" may decide which speed to use within this range. current CPU frequency is 300 MHz (asserted by call to hardware). cpufreq stats: 300 MHz:13.67%, 600 MHz:30.66%, 800 MHz:55.66% (41) root@gta04:~# If possible, the "Speed Binned Flag" should be dynamically queried and update the "available frequency steps" list. * USB not working (?)
Comment 1 by Nikolaus Schaller, Apr 10, 2020
We have submitted code upstream that switches to OPPv2 format which allows to make the 1GHz OPP depend on the speed binned bit. This works well. Unfortunately the dm3730 (GTA04A5) is a little unreliable at 1GHz. It can be demonstrated by running kmscube on Bullseye without SGX and switching cpufreq-set -f between 800M and 1G. At 1G there are drawing artefacts. The process may also break with SIGBUS or SIGILL. This indicates that the DM3730 isn't running at the correct OPP voltage. Maybe, 1GHz also needs Smart Reflex in operation while the standard ti-cpufreq code just switches voltages to predefined values.
Comment 2 by Nikolaus Schaller, Apr 10, 2020
Comment 3 by Nikolaus Schaller, Apr 10, 2020