4.12 on MVME6100 (beatnik)

[Joel Sherrill]

On Fri, Mar 11, 2016 at 5:21 AM, Heinz Junkes <junkes at fhi-berlin.mpg.de>
wrote:

> Hi,
>
> does anyone got the RTEMS 4.12 running on the beatnik (MVME6100)?
>
> This is unfortunately an example where we rely on the community to
periodically
test. If no one checks in a while, ... :(


> We can’t get one of the examples booting on that target. It stucks in the
> startup routine.
> It does not reach the bsp_start routine.
>
> Is the executable even well-formed? Does the format and address map look
correct per
the one that works? Entry point the same, etc.


> Examples are running with 4.12 on an older VMEbus-Board (MVME2100) with
> just some litte issues.
>
>
> With this version:
>
> |--------------------
> commit 9827d51e7405aad26d3eda2d3fb82504b7fa68c3
> Author: Joel Sherrill <joel.sherrill at oarcorp.com>
> Date:   Wed Oct 8 09:24:02 2014 -0500
>
>   gen5200/startup/cpuinit.c: Warning for function which should have been
> void
> …
> ——————————|
>
> On the MVME6100
>
> ticker.ralf works with RTEMS 4.10 compiler
> ticker.ralf fails with RTEMS 4.12 compiler
>
> any ideas?
>
> Does it work on the 4.11 branch?


> We tried the following to find the problem:
>
> - study the changes/diffs of the relevant startup files (including
> assembly)
> - try to find the point where it gets stuck by inserting a few instructions
>    which light LEDs or throw exceptions.
>
> (We couldn’t  get the LED (BDFAIL) on. Assuming
> cache/memory-corruption/???)
>
> What changes were in the startup code?

--joel


> Heinz
>
>
>
> _______________________________________________
> 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/20160314/1607ed32/attachment-0002.html>