[ECOS] problems with RTC

[ECOS] problems with RTC

[Michele Paselli]

Hi all,

I'm developing a radio driver for an AT91 platform and I need to 
change the time granularity in order to make some precise latency 
measures. In order to have a frequency multiplied by 10 (1 KHz since 
the default is 100Hz) I multiplied by 10 the CYGNUM_HAL_RTC_DENOMINATOR 
parameter. I also modified the CYGNUM_KERNEL_TIMESLICE_TICKS parameter 
in order to have the same timeslice interval as I would have in the 
default conditions. The ticks are now indeed 1 every millisecond, but 
the radio driver does not seem to work properly anymore. In particular, 
the driver seems to miss come interrupts. The only call to a function 
related to timers in the driver is a timeout counter but I multiplied 
its value by 10 to have the same conditions as before. Can anyone 
suggest me a possible cause? I think that the conditions for the driver 
in terms of timing are the same as before, but maybe I'm missing 
something.

Thanks a lot in advance.

Michele


Naviga e telefona senza limiti con Tiscali     
Scopri le promozioni Tiscali adsl: navighi e telefoni senza canone Telecom

http://abbonati.tiscali.it/adsl/


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

Re: [ECOS] problems with RTC

[Nick Garnett]

Michele Paselli <miguelon@tiscali.it> writes:

> Hi all,
> 
> I'm developing a radio driver for an AT91 platform and I need to 
> change the time granularity in order to make some precise latency 
> measures. In order to have a frequency multiplied by 10 (1 KHz since 
> the default is 100Hz) I multiplied by 10 the CYGNUM_HAL_RTC_DENOMINATOR 
> parameter. I also modified the CYGNUM_KERNEL_TIMESLICE_TICKS parameter 
> in order to have the same timeslice interval as I would have in the 
> default conditions. The ticks are now indeed 1 every millisecond, but 
> the radio driver does not seem to work properly anymore. In particular, 
> the driver seems to miss come interrupts. The only call to a function 
> related to timers in the driver is a timeout counter but I multiplied 
> its value by 10 to have the same conditions as before. Can anyone 
> suggest me a possible cause? I think that the conditions for the driver 
> in terms of timing are the same as before, but maybe I'm missing 
> something.

I don't have any suggestions for fixing your driver, in theory
everything you have done is correct. I have run many eCos
configurations with a 1KHz clock with no problems.

Something you should check is that the clock ticks are arriving at the
expected rate. Try running the clocktruth kernel test against a
stopwatch to check this.

-- 
Nick Garnett                                     eCos Kernel Architect
eCosCentric Limited     http://www.eCosCentric.com/   The eCos experts
Barnwell House, Barnwell Drive, Cambridge, UK.    Tel: +44 1223 245571
Registered in England and Wales: Reg No 4422071.


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