On 01/02/17 15:22, Joel Sherrill wrote:
On Feb 1, 2017 8:02 AM, "Sebastian Huber" <sebastian.hu...@embedded-brains.de <mailto:sebastian.hu...@embedded-brains.de>> wrote:On 01/02/17 14:53, Joel Sherrill wrote: The inclusion of SMP categorizes the option. Many, if not most, configure options have a second word which is the category. All the POSIX options include POSIX in the name, gnat, Go, etc. This is the only CONFIGURE_SMP_* option left, a category should consist of more than one member. The current pattern for the maximum defines is CONFIGURE_MAXIMUM[_CATEGORY]_*. A CONFIGURE_MAXIMUM_SMP_PROCESSORS is not really that great. All of our services are rtems_CATEGORY. It isn't about the shortest name with RTEMS. It is about the clearest name with grouping. I don't want to see this pattern violated and am opposed to this change. Otherwise, this would be CONF_MAX_CPUS or similar and we don't want that. In general I agree, however, SMP or non-SMP shouldn't be a category for the configuration. I think its like CONFIGURE_TICKS_PER_TIMESLICE or CONFIGURE_MICROSECONDS_PER_TICK a configuration option that sets up your application environment. Are there any options without SMP which should have them? If not, I agree.
I update currently the SMP chapter in the user manual. I thought that this would simplify the things a bit.
The clustered scheduler configuration uses the SCHEDULER category. I don't think we added anything SMP-specific except CONFIGURE_SMP_MAXIMUM_PROCESSORS (to be renamed) and CONFIGURE_SMP_APPLICATION (to be removed).
-- 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
