From mboxrd@z Thu Jan 1 00:00:00 1970 From: icenowy@aosc.io (icenowy at aosc.io) Date: Wed, 27 Sep 2017 19:51:30 +0800 Subject: [PATCH 0/3] Simple DVFS support for Allwinner A64 SoC In-Reply-To: <20170925102744.qixfwlheeimemhcf@flea.home> References: <20170923001531.14285-1-icenowy@aosc.io> <20170925101027.lghnnll4h6inreqm@flea.home> <27EF78BD-6285-4D8D-AA65-8294D797E2FB@aosc.io> <20170925102744.qixfwlheeimemhcf@flea.home> Message-ID: <9b3aeb6cb155bb2f9a7cee438de82ccb@aosc.io> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org ? 2017-09-25 18:27?Maxime Ripard ??? > On Mon, Sep 25, 2017 at 10:12:09AM +0000, Icenowy Zheng wrote: >> ? 2017?9?25? GMT+08:00 ??6:10:27, Maxime Ripard >> ??: >> >Hi, >> > >> >On Sat, Sep 23, 2017 at 12:15:28AM +0000, Icenowy Zheng wrote: >> >> This patchset imports simple DVFS support for Allwinner A64 SoC. >> >> >> >> As the thermal sensor driver is not yet implemented and some boards >> >> have still no AXP PMIC support, now only two OPPs are present -- >> >> 648MHz at 1.04V and 816MHz at 1.1V to prevent overheat or undervoltage. >> >> >> >> PATCH 1 is a fix to the CCU driver of A64, and the remaining patches >> >> set up the device tree bits of the DVFS on Pine64. >> > >> >How has this been tested? >> > >> >What tasks did you run, with what governor, etc... >> >> I only tested manual frequency switching between 648MHz and >> 816MHz, and tested the PLL stuck issue by change the OPPs to >> some random value. > > Ideally, we should test that it's actually reliable. Poorly chosen > OPPs might lead to corrupt data that you might not get before a while. > > Please test using: > https://linux-sunxi.org/Hardware_Reliability_Tests#Reliability_of_cpufreq_voltage.2Ffrequency_settings > > And post the report. ``` root at p64 [ cpuburn-arm at master ] # ./cpuburn-a53 & [1] 2543 root at p64 [ cpuburn-arm at master ] # ./cpufreq-ljt-stress-test Creating './whitenoise-1920x1080.jpg' ... done CPU stress test, which is doing JPEG decoding by libjpeg-turbo at different cpufreq operating points. Testing CPU 0 816 MHz ............................................................ OK 648 MHz ............................................................ OK Testing CPU 1 816 MHz ............................................................ OK 648 MHz ............................................................ OK Testing CPU 2 816 MHz ............................................................ OK 648 MHz ............................................................ OK Testing CPU 3 816 MHz ............................................................ OK 648 MHz ............................................................ OK Overall result : PASSED ``` > > Maxime