Re: [ECOS]Strongarm rtc and oscr

[Andrea Acquaviva <aacquaviva@deis.unibo.it>]

Jonathan Larmour wrote:

> Andrea Acquaviva wrote:
> >
> > I want to measure eCos multi-threading energy costs. I use it on the
> > assabet board.
>
> Have you looked at the tm_basic kernel test?
>
> > In order to measure context switch overhead, I need some infos about
> > timeslicing:
> >
> > - the option CYGNUM_KERNEL_SCHED_TIMESLICE_TICKS refers to 3.6864MHz
> >    clock ticks?
>
> Yes.
>
> > - How does eCos use real time clock (RTC) and operating system timer
> > (OSCR) of
> >     StrongARM 1110?
>
> An interesting question because we are considering changing the way it
> operates right now! The interesting part is in
> hal/arm/sa11x0/var/current/src/sa11x0_misc.c and hal_clock_rest() and
> hal_clock_read(). Currently it resets it to 0 each interrupt, but we may be
> about to change it to make it a freerunning counter.
>
> Jifl
> --
> Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
> Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine

Hi,

OK. I see that eCos uses oscr timer0 as a real time clock instead of RTC.

Now I'm looked at tm_basic test, and I have some additional questions:

- Can you explain me how the alarm_cb2 function is called? In the ref man
   I found that it should be the DSR routine defined in clock.cxx
   that calls the function which responds to an alarm. It's that right?

- When I want to install an rtc based alarm, what's the steps I must follow?

Thanks a lot for your help.
Andrea.

--
Ing. Andrea Acquaviva
D.E.I.S. - Universita' di Bologna
V.le Risorgimento, 2    40136 BOLOGNA (ITALY)
Tel: (+39) 051 20 93787 Fax: (+39) 051 2093786