Why decrementer

Thomas Doerfler Thomas.Doerfler at embedded-brains.de
Wed Nov 5 10:32:23 UTC 2008


Leon,

I can see several good reasons:

- The decrementer has various design flaws, but is available in (almost)
EVERY PPC core, so the code is rather generic

- The decrementer's inaccuracy during reload can be overcome, if you use
the timebase as a reference for reload. This is implemented in the
current shared DEC driver code.

- The decrementer has a MUCH HIGHER resolution than the PIT in the mpc8xx.
- On the mpc8xx based MBX8xx boards, using the PIT really was a pain. On
some boards, the external clock was 32kHz, which only allows very
strange values for "microseconds per tick". The decrementer usually
overcomes this

Many reasons to do it this way.

wkr,
Thomas.


Leon Pollak wrote:
> Hello, all.
> 
> Is here somebody who can explain the reason why the RTEMS system clock for the 
> mpc8260 CPU is implemented with decrementer (with all its problems) and not 
> with PIT (as it was in mpc8xx)?
> 
> Many thanks ahead.
> 


-- 
--------------------------------------------
embedded brains GmbH
Thomas Doerfler           Obere Lagerstr. 30
D-82178 Puchheim          Germany
Tel. : +49-89-18 90 80 79-2
Fax  : +49-89-18 90 80 79-9
email: Thomas.Doerfler at embedded-brains.de
PGP public key available on request

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.



More information about the users mailing list