GSoC Linux UIO driver for PRU

[Nils Hölscher]

On Thu, 18 Jul 2019 at 19:18, Chris Johns <chrisj at rtems.org> wrote:

> On 18/7/19 7:11 pm, Nils Hölscher wrote:
> > @chrisj at rtems.org <mailto:chrisj at rtems.org>  but where do we place the
> shell
> > part, which exists between BSD and RTEMS?
>
> LibBSD has a number of shell commands so this should not be a problem.
>
> > However this doesn't have priority right now, does it?
>
> I think the shell command support is good to have. A user wanting to
> develop and
> run a PRU piece of code with an RTEMS app will need all the support we can
> provide.
>

Yes you are definitely right and I think it is pretty easy to port the
pructrl repo I posted before.

>
> Chris
>

I am currently adding more and more libbsd deps to my build, that are not
in rtems or rtems-libbsd.
Now I have been touching these to files of the BSD kernel:
https://github.com/freebsd/freebsd/blob/0966ce5cd477cd7c8b63b0f9de75396aa5939032/sys/arm/include/cpu-v4.h
https://github.com/freebsd/freebsd/blob/0966ce5cd477cd7c8b63b0f9de75396aa5939032/sys/arm/include/atomic-v6.h
Throwing a good dozen of warning when compiling and I only need one call
from these files.
This is needed in the pru driver to distinguish between prussv1 and
prussv2, a feature I definitely want.
FYI: prussv1 is on the TI Sitara AM18xx SoCs and v2 on the AM35xx.

So the question is should I try to resolve those warnings and try to port
the whole files to rtems or should I try to just pick  what I need and be
done?

Best,
Nils
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20190719/a661e85c/attachment-0002.html>