Re: [ECOS] ecos and gettimeofday()

[Peter Korsgaard <jacmet at sunsite dot dk>]

>>>>> "Peter" == Peter Korsgaard <jacmet@sunsite.dk> writes:

 Peter> If the problem is that you don't know if the clock DSR has run or not,
 Peter> perhaps you could do something hackish like this to make sure
 Peter> it hasn't:

No, this also doesn't work.

We want to be able to do a HAL_CLOCK_READ() and cyg_current_time() and
want the two values to be in sync. When HAL_CLOCK_READ() reaches
CYGNUM_HAL_RTC_PERIOD an interrupt is generated and the clock is reset
to 0. The DSR will then at some later time be executed and
cyg_current_time() incremented.

If cyg_current_time() was updated in the ISR my routine would work.
What we need to detect is the case of a HAL_CLOCK_READ overflow
(interrupt) having occurred, but the DSR not executed yet. This can be
done by:

1) Updating the counter in the ISR and use that for cyg_current_time()
   (it still needs to be done in the DSR for alarms as well).

2) Set a flag in the ISR that the DSR is pending, clear it in the
   DSR. Increment cyg_current_time if it's set. Alternatively use the
   dsr_count from Cyg_interrupt.

Which of these would be preferred? Or is there a cleaner solution?

-- 
Bye, Peter Korsgaard

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