[ECOS] problem with the RealTimeClock ISR routine "return" instruction

[ECOS] problem with the RealTimeClock ISR routine "return" instruction

[Satish Kumar]

Hi,
    i m testing the clocktruth.cxx on my target board which has
arm926ej-s core.i m using gcc 3.2.1 for compiling my code.
i have noticed a strange behaviour in the ISR routine.
when the program calls Cyg_RealTimeClock() routine to initialize
the external timer..and when the timer generates an periodic interrupt 
the Cyg_RealTimeClock::isr() routine is called with the correct vector
number and when this instruction i.e
(  return Cyg_Interrupt::CALL_DSR|Cyg_Interrupt::HANDLED )
is excuted in the ISR routine, it is actually branching to spurious_IRQ
handler routine..
due to this my clocktruth program hangs in the while loop unable to 
read the rtc->counter_value_lo and returns 0 everytime..failin to print
the time value.

my concern is why "the spurious IRQ"  when the ISR routine was 
called with the correct timer interrupt no:...? is it somethin to do
with the HAL porting issues..?


some one help plz..
saty




 isr routine 
-- 
Satish Kumar <satish@bvt.sc.sanyo.co.jp>


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

RE: [ECOS] problem with the RealTimeClock ISR routine "return" instruction

[Robert Cragie]

It may be due to the timer interrupt not being negated or masked properly
(see hal_clock_reset()) and is possibly causing a re-entrant interrupt. This
action is very specific to the timers on your target platform so I would
check this carefully.

Alternatively, if you are getting a spurious interrupt, I would check *all*
your possible interrupt sources on your target hardware and make sure that
they are:

a) Catered for in your HAL port
b) If not used, correctly masked out in the initialisation
c) If used, correctly handled

I may be wrong, but I feel that your problem is very unlikely to be a
generic eCos one. Also, you are asking quite specific questions but you have
given very little detail of your actual target hardware with regard to what
sort of timers you are using and the interrupts they use. If you are having
real trouble with this, it might be an idea to use the services of an
organisation like eCosCentric or MIND, who are experts in HAL porting.

Robert Cragie, Design Engineer
_______________________________________________________________
Jennic Ltd, Furnival Street, Sheffield, S1 4QT,  UK
http://www.jennic.com  Tel: +44 (0) 114 281 2655
_______________________________________________________________

> -----Original Message-----
> From: ecos-discuss-owner@sources.redhat.com
> [mailto:ecos-discuss-owner@sources.redhat.com]On Behalf Of Satish Kumar
> Sent: 19 September 2003 08:23
> To: ecos-discuss@sources.redhat.com
> Subject: [ECOS] problem with the RealTimeClock ISR routine "return"
> instruction
>
>
> Hi,
>     i m testing the clocktruth.cxx on my target board which has
> arm926ej-s core.i m using gcc 3.2.1 for compiling my code.
> i have noticed a strange behaviour in the ISR routine.
> when the program calls Cyg_RealTimeClock() routine to initialize
> the external timer..and when the timer generates an periodic interrupt
> the Cyg_RealTimeClock::isr() routine is called with the correct vector
> number and when this instruction i.e
> (  return Cyg_Interrupt::CALL_DSR|Cyg_Interrupt::HANDLED )
> is excuted in the ISR routine, it is actually branching to spurious_IRQ
> handler routine..
> due to this my clocktruth program hangs in the while loop unable to
> read the rtc->counter_value_lo and returns 0 everytime..failin to print
> the time value.
>
> my concern is why "the spurious IRQ"  when the ISR routine was
> called with the correct timer interrupt no:...? is it somethin to do
> with the HAL porting issues..?
>
>
> some one help plz..
> saty
>
>
>
>
>  isr routine
> --
> Satish Kumar <satish@bvt.sc.sanyo.co.jp>
>
>
> --
> Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
> and search the list archive: http://sources.redhat.com/ml/ecos-discuss
>
>


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

Re: [ECOS] problem with the RealTimeClock ISR routine "return" instruction

[Satish Kumar]

Ok,
     Plz ignore my previous question on spurious IRQ being generated.
i m sorry for that..well, my timer interrupts are been serviced properly
, but its the problem with this "Return from exception"..from were i
jump into the clocktruth.cxx file to display the time, were in it hangs
up on this while loop. the rtc->current_value_lo() always returns 0.


       // First, print 100 lines as fast as you can, of distinct ticks.
     for ( i = 0; i < 100; i++ ) {
         now = rtc->current_value_lo();
         then = now;
         while ( then == now )
             now = rtc->current_value_lo();

         diag_printf( "INFO<time now %8d>\n", now );
     }

i followed up the patch posted as per 
http://sources.redhat.com/ml/ecos-patches/2003-04/msg00114.html

but still couldn`t over come this problem..i m not able to figure out
whatz the error like..?

satish


On Fri, 19 Sep 2003 09:28:04 +0100
"Robert Cragie" <rcc@jennic.com> wrote:

> It may be due to the timer interrupt not being negated or masked properly
> (see hal_clock_reset()) and is possibly causing a re-entrant interrupt. This
> action is very specific to the timers on your target platform so I would
> check this carefully.
> 
> Alternatively, if you are getting a spurious interrupt, I would check *all*
> your possible interrupt sources on your target hardware and make sure that
> they are:
> 
> a) Catered for in your HAL port
> b) If not used, correctly masked out in the initialisation
> c) If used, correctly handled
> 
> I may be wrong, but I feel that your problem is very unlikely to be a
> generic eCos one. Also, you are asking quite specific questions but you have
> given very little detail of your actual target hardware with regard to what
> sort of timers you are using and the interrupts they use. If you are having
> real trouble with this, it might be an idea to use the services of an
> organisation like eCosCentric or MIND, who are experts in HAL porting.
> 
> Robert Cragie, Design Engineer
> _______________________________________________________________
> Jennic Ltd, Furnival Street, Sheffield, S1 4QT,  UK
> http://www.jennic.com  Tel: +44 (0) 114 281 2655
> _______________________________________________________________
> 
> > -----Original Message-----
> > From: ecos-discuss-owner@sources.redhat.com
> > [mailto:ecos-discuss-owner@sources.redhat.com]On Behalf Of Satish Kumar
> > Sent: 19 September 2003 08:23
> > To: ecos-discuss@sources.redhat.com
> > Subject: [ECOS] problem with the RealTimeClock ISR routine "return"
> > instruction
> >
> >
> > Hi,
> >     i m testing the clocktruth.cxx on my target board which has
> > arm926ej-s core.i m using gcc 3.2.1 for compiling my code.
> > i have noticed a strange behaviour in the ISR routine.
> > when the program calls Cyg_RealTimeClock() routine to initialize
> > the external timer..and when the timer generates an periodic interrupt
> > the Cyg_RealTimeClock::isr() routine is called with the correct vector
> > number and when this instruction i.e
> > (  return Cyg_Interrupt::CALL_DSR|Cyg_Interrupt::HANDLED )
> > is excuted in the ISR routine, it is actually branching to spurious_IRQ
> > handler routine..
> > due to this my clocktruth program hangs in the while loop unable to
> > read the rtc->counter_value_lo and returns 0 everytime..failin to print
> > the time value.
> >
> > my concern is why "the spurious IRQ"  when the ISR routine was
> > called with the correct timer interrupt no:...? is it somethin to do
> > with the HAL porting issues..?
> >
> >
> > some one help plz..
> > saty
> >
> >
> >
> >
> >  isr routine
> > --
> > Satish Kumar <satish@bvt.sc.sanyo.co.jp>
> >
> >
> > --
> > Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
> > and search the list archive: http://sources.redhat.com/ml/ecos-discuss
> >
> >

-- 
Satish Kumar <satish@bvt.sc.sanyo.co.jp>


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