OK thanks, it makes sense.
On Fri, Jun 15, 2018 at 11:03 AM, Sebastian Huber <sebastian.hu...@embedded-brains.de> wrote: > ----- Am 15. Jun 2018 um 16:16 schrieb Gedare Bloom ged...@rtems.org: > >> Hi Sebastian, >> >> The only real question I have is about the default value of 1 GHz. >> (How) does this affect the accuracy/correctness of the BSPs that use >> this default implementation? > > If a BSP doesn't provide a CPU counter implementation, then the default value > will be almost always wrong. The value of 1GHz was chosen to get reasonable > safe busy wait functions, e.g. they return not too early. > >> >> As a side note, using Hz units in uint32_t means we limit to about 4.2 >> GHz. This seems a safe assumption, but worth documenting? > > Yes, sorry, the CPU counter API documentation is quite long on my TODO list: > > https://devel.rtems.org/ticket/2848 > > Faster CPUs can scale down the CPU counter frequency if necessary. > > -- > Sebastian Huber, embedded brains GmbH > > Address : Dornierstr. 4, D-82178 Puchheim, Germany > Phone : +49 89 189 47 41-16 > Fax : +49 89 189 47 41-09 > E-Mail : sebastian.hu...@embedded-brains.de > PGP : Public key available on request. > > Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
