Fwd: RTEMS core: bad work with _Watchdog_* ?

[Joel Sherrill]

Ilya Alexeev wrote:

> > I have been working offline with Ilya and sent him the suggestion
> > to comment out the _ISR_Flash in src/watchdoginsert.c.  Then he should
> > have a finer grain view of the problem.
> >
> > --joel
>
> Yes, after commenting out _ISR_Flash in src/watchdoginsert.c
> my 'stress test' works fine.
>
> There were a some my incorrectness in the previous letter:
> the application worked good with _ISR_Disable/Enable around the
> _Watchdog_Insert_ticks only.
>

Just to double check... you now have removed the _ISR_Disable/Enable's
you added and commented out the _ISR_Flash in src/watchdoginsert.c
and it works.  Right?

If so, then something is wrong with the critical section logic.  I wonder
if we need to mark something as volatile.


>
> Ilya.