Re: [ECOS] RedBoot/Basic test working... context not?

["Stefan Sommerfeld" <sommerfeld@mikrom.de>]

Hi,
>> >>
>> >> I'm using a  XScale SO-DIMM module (Ka-Ro Triton270) and moving from 
>> >> the
>> >> development board to the destination plattform. If I run 'basic' or
>> >> 'cache'
>> >> test on the destination plattform, it works, but context does not. I
>> >> guess
>> >> context switching doesn't work, but i wonder what is the cause of 
>> >> this.
>> >> What is needed for context switching? I would think it's only 
>> >> timer/cpu
>> >> related stuff, so there couldn't be any influence from GPIO's.
>> >>
>> >> Any hints why context switch doesn't work?
>> >
>> > What do you mean by doesn't work? Does it give an error message? Does
>> > an assert fail? Send out smoke signals as the SO-DIMM modules melts?
>>
>> I added some debug output and found the position where it stops:
>>
>>  HAL_THREAD_LOAD_CONTEXT( &sp[0] );
>>
>> 'context' executes this macro and stopped working (no output anymore).
>> There's also no output from entry0. I sure it's some hardware influence,
>> but why does the threading doesn't work?
>
> Because this is much more complicated than the tests that you've
> already run!  The biggest difference is that interrupts are normally
> enabled as part of the change of context.  If you've got some I/O
> misconfigured (based on your query about GPIO), this could be the
> problem.
>
I agress with you, it sounds like an IRQ problem due to incorrect GPIO 
config, but why does the program start work and the new threads not. Event 
a program with a main() function doesn't work. Some test are working, like 
thread0 or intr0, others not like thread1. Is there a test where i can come 
a little closer to the problem?

Bye... 


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