Re: [ECOS] AT91SAM7S RTC speed

[Tom Duffy <tduffy at tascam dot com>]

I can verify that the SPI clock is twice the expected rate in a real 
world device
using the SAM7X512.
I haven't had time to investigate where the extra factor of 2 clock rate is
being introduced in the HAL.
I just coded values half the expected in the SPI table.
Timer rates seem to be correct based on delays used to blink a LED though.

Tom.

Andrew Lunn wrote:
> On Mon, Oct 13, 2008 at 04:04:38AM +0400, Igor B. Poretsky wrote:
>   
> > Hello, all!
> >
> > I am using Ecos for the AT91SAM7S target in several real
> > projects. When started I noticed that the
> > CYGNUM_HAL_ARM_AT91_CLOCK_SPEED gets the value twice greater than
> > expected. So I patched the sources to achieve proper time
> > calculation. It was some time ago and now I don't remember whether I
> > informed developers about the issue or not. Anyway, I apply my patches
> > below.
> >     
>
> I've also been using the AT91SAM7S and AT91SAM7X in real projects and
> have not had problems with the timer. Please could you explain what
> you see wrong? Maybe you have a pointer to some documentation? 
>
> Looking at page 105 of doc6175.pdf, it shows that the PIT is driven by
> MCK/16.
>
> I moved house recently, so my AT91SAM7SEK development board is still
> packed away in a box somewhere. Could you run the clocktruth test
> program and see if that shows a problem. Without your patch you would
> expect it to "tick" at twice the expected rate.
>
>        Thanks
>           Andrew
>
>   


--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss