On 1 December 2013 00:43, Manish Badarkhe <badarkhe.manish@xxxxxxxxx> wrote: > I have some doubts in order to set "target" frequency in cpu-frequency > driver. I gone through code which set > "target" CPU frequency and got details as below: > > 1. Core CPU frequency driver "target" function calls > (__cpufreq_driver_target) a platform specific "target" > functions(for e.g. exynos_target, davinci_target). > 2. I observed some of problems in setting target frequency especially > when doing target freq > current frequency. > 3. Consider an example of davinci frequency driver if want to set a > frequency higher than current frequency then > 1. Firstly, set regulator voltage and if it fails > then return error. > 2. Set clock rate and if it fails return error. > 3. Consider a scenario that regulator works fine > while setting voltage but somehow "clk_set_rate" > fails then here is a bug, that CPU runs at > higher voltage even though that is not required. > > I have not seen anywhere in code that regulator is brought back to > original voltage if CPU set rate for frequency > is failed. Am I missing something in code flow? or there is no need > to care about regulator voltage in such case. Its probably a bug, but people doesn't get around it unless clk_set_rate() fails. And so nobody has cared to fix it up.. See cpufreq-cpu0 instead, which probably has all the error handling code in place. -- To unsubscribe from this list: send the line "unsubscribe cpufreq" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
