[U-Boot] [PATCH] Move ICS CLK chip frequenty calculation code into a common board library

Kumar Gala galak at kernel.crashing.org
Sun May 23 00:39:59 CEST 2010


On May 21, 2010, at 3:16 PM, Wolfgang Denk wrote:

> Dear Timur Tabi,
> 
> In message <4BF6E5DF.5020306 at freescale.com> you wrote:
>> 
>>>> So here's a better version of that function that rounds to the nearest
>>>> MHz and is of a proper coding style:
>>> 
>>> Why do we need that?
>> 
>> Um, because you complained about it?
> 
> I mean, why do we need that function? strmhz() already includes
> rounding. Are you really rounding, or cutting off precsion?
> 
>> Also, because this is silly:
>> 
>> Clock Configuration:
>>       CPU0:799.992 MHz, CPU1:799.992 MHz,
>>       CCB:399.996 MHz,
>>       DDR:299.997 MHz (599.994 MT/s data rate) (Asynchronous), LBC:25   MHz
>> 
>> Why display 799.992 MHZ when 800 MHz makes more sense?
> 
> Hm... does it really make more sense? Or is it just less precise, or
> do you hush up an error?
> 
> 799.992 MHz seems to be 24 * 33,333,000 Hz. Are you sure this is the
> exact quartz frequency on your board? If yes, then the number printed
> should probably remain 799.992 MHz. Or is it 33,333,333 Hz? Or
> 33,000,000 ?

Its a 33,333,000 Hz input crystal so the math is correct.  Rounding in the ICS code is wrong.

- k


More information about the U-Boot mailing list