Fatal source exception in loopback sample on powerpc/qoriq_e6500_64

[Joel Sherrill]

On Wed, Sep 16, 2020 at 7:51 AM Clukey (US), Steven A <
steven.a.clukey at boeing.com> wrote:

> Thanks! I had missed that. I've gotten libbsd compiled/linked and now I
> can run a loopback test
>
> A question though: if RTEMS_NETWORKING doesn't work on the 64-bit targets,
> why does it compile? Could the `--enable-networking` configure option be
> removed or marked as an error on the targets that it doesn't support?
>

PowerPC has 32 and 64 bit support and it does work on the 32-bit ones.

The other issue is that some BSPs have no network drivers, some have only
drivers for the legacy in-tree stack, some have only libbsd, and a few
support both stacks.

Checking this would be a per-BSP configure.ac change which would be OK if
we were sticking with the current build system but with the switch to waf
on the master and (unfunded) plan to move the legacy stack to its own build
tree, I'm not sure adding any error checks will get much attention. If
patches showed up, I don't think there would be opposition to merging them
though. I just don't think a core developer will invest their volunteer
time in it.

--joel



> _______________________________________________
> users mailing list
> users at rtems.org
> http://lists.rtems.org/mailman/listinfo/users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20200916/654dd3f2/attachment.html>