On 16/10/2018 15:16, Joel Sherrill wrote:
On Oct 16, 2018 8:07 AM, "Sebastian Huber"
<sebastian.hu...@embedded-brains.de
<mailto:sebastian.hu...@embedded-brains.de>> wrote:
On 15/10/2018 14:52, Joel Sherrill wrote:
> Does doing this mean that autoconf probes now need to pass an extra
> library to link against?
I think Autoconf is not an issue. By default, the feature probes will
use the dummy crt0.o provided by Newlib.
What if the probe references a method with configurable parameter? Max
file descriptors, etc? Are you sure there is no path from any single
POSIX method to a missing configuration?
You only link against the RTEMS libraries if you use the -specs
bsp_specs etc. I don't think this is used by Autoconf probes unless you
do something special.
The stuff in rtems_waf makes some problems. It uses the RTEMS libraries
for the probes which contain simply a main() function and no RTEMS
configuration. There are two options to address this issue.
1. Use test stub with an RTEMS configuration.
2. Use the librtemsdefaultconfig.a during the probes, but not for the
application.
--
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
