Messages in this thread | | | Date | Thu, 15 May 2014 07:05:28 -0700 | From | Sören Brinkmann <> | Subject | Re: [RFC PATCH 0/5] Frequency resolution in CCF vs. cpufreq |
| |
Hi Viresh,
On Thu, 2014-05-15 at 11:42AM +0530, Viresh Kumar wrote: > On 15 May 2014 04:00, Soren Brinkmann <soren.brinkmann@xilinx.com> wrote: > > I have one or two problems with cpufreq and the CCF, which are caused by > > rounding/different frequency resolutions. > > > > cpufreq works with kHz, while the CCF uses Hz. On Zynq our default frequency is > > 666666666 Hz which the CCF, due to rounding, reports as 666666660. And for > > cpufreq, which simply divides values it obtains through clk_round_rate() by > > 1000, 666666. > > Since passing 666666 to clk_round_rate() does not result in 666666660 > > (clk_round_rate() always rounds down!), we chose to put 666667 in the OPP. This > > causes issue 1: cpufreq stats are broken. > > I know it might a big exercise, but wouldn't it be worth to make cpufreq start > using frequencies in Hz ?
I haven't looked into this. As you say yourself, that might be a rather big project. I will take a look at it, but I can't promise that I have time to dedicate to this. Also, as said above, even though our CPU is supposed to run at 666666666 Hz, rounding lets the last 6 Hz disappear. I think we have to handle deviances either way.
Sören
-- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |