[PATCH RTEMS] mvme3100: Add BSP fatal extension

[Vijay Kumar Banerjee]

On Tue, Aug 1, 2023, 15:30 Joel Sherrill <joel at rtems.org> wrote:

>
>
> On Tue, Aug 1, 2023 at 2:52 PM Vijay Kumar Banerjee <vijay at rtems.org>
> wrote:
>
>>
>>
>> On Tue, Aug 1, 2023, 12:20 Joel Sherrill <joel at rtems.org> wrote:
>>
>>>
>>>
>>> On Tue, Aug 1, 2023, 11:41 AM Vijay Kumar Banerjee <vijay at rtems.org>
>>> wrote:
>>>
>>>> ---
>>>>  bsps/powerpc/mvme3100/start/bspclean.c        | 25 +++++++++++++++++++
>>>>  .../bsps/powerpc/mvme3100/bspmvme3100.yml     |  1 +
>>>>  2 files changed, 26 insertions(+)
>>>>  create mode 100644 bsps/powerpc/mvme3100/start/bspclean.c
>>>>
>>>> diff --git a/bsps/powerpc/mvme3100/start/bspclean.c
>>>> b/bsps/powerpc/mvme3100/start/bspclean.c
>>>> new file mode 100644
>>>> index 0000000000..251d47a46d
>>>> --- /dev/null
>>>> +++ b/bsps/powerpc/mvme3100/start/bspclean.c
>>>> @@ -0,0 +1,25 @@
>>>>
>>>
>>> There is no copyright, licence, or Doxygen file header block.
>>>
>>> +#include <bsp.h>
>>>> +#include <bsp/bootcard.h>
>>>> +#include <rtems/bspIo.h>
>>>> +
>>>> +void bsp_fatal_extension(
>>>> +  rtems_fatal_source source,
>>>> +  bool always_set_to_false,
>>>> +  rtems_fatal_code error
>>>> +)
>>>> +{
>>>> +  printk("fatal source: %s\n", rtems_fatal_source_text(source));
>>>> +
>>>> +  if (source == RTEMS_FATAL_SOURCE_EXCEPTION) {
>>>> +    rtems_exception_frame_print((const rtems_exception_frame *) error);
>>>> +  }
>>>> +
>>>> +  /* We can't go back to MotLoad since we blew it's memory area
>>>> +   * and vectors. Just pull the reset line...
>>>> +   */
>>>> +  printk(
>>>> +    "bsp_fatal_extension(): RTEMS terminated -- no way back to MotLoad
>>>> "
>>>> +      "so I reset the card\n"
>>>> +  );
>>>> +  bsp_reset();
>>>> +}
>>>>
>>>
>>> What does Motorola PowerPC or the other similar bsps do?
>>>
>>> Can the behaviour be unified and shared?
>>>
>>
>> Thanks for the suggestion!
>> Motorola powerpc uses bspfatal-default.c from bsps/shared. It was not
>> working with mvme3100 because it added the bspreset-empty in the build.
>> Just removing the empty reset makes it work. This patch is no longer
>> needed. I'll send a smaller patch with this much neater solution using the
>> shared defaults.
>>
>
> Wow! Good to hear. Probably should check beatnik and the other mvme* ppc
> BSPs  as well,
>

At least two bsps (mvme5500 and beatnik) has this issue. I will test on the
hardware and post patches for other boards too.

>
>>
>> Thanks!
>>
>
> No problem.
>
> --joel
>
>>
>>
>>> diff --git a/spec/build/bsps/powerpc/mvme3100/bspmvme3100.yml
>>>> b/spec/build/bsps/powerpc/mvme3100/bspmvme3100.yml
>>>> index 1667c1617a..dc04e4dd36 100644
>>>> --- a/spec/build/bsps/powerpc/mvme3100/bspmvme3100.yml
>>>> +++ b/spec/build/bsps/powerpc/mvme3100/bspmvme3100.yml
>>>> @@ -64,6 +64,7 @@ source:
>>>>  - bsps/powerpc/mvme3100/pci/detect_host_bridge.c
>>>>  - bsps/powerpc/mvme3100/rtc/todcfg.c
>>>>  - bsps/powerpc/mvme3100/start/bspstart.c
>>>> +- bsps/powerpc/mvme3100/start/bspclean.c
>>>>  - bsps/powerpc/mvme3100/start/misc.c
>>>>  - bsps/powerpc/shared/btimer/btimer-ppc-dec.c
>>>>  - bsps/powerpc/shared/cache/cache.c
>>>> --
>>>> 2.39.3
>>>>
>>>> _______________________________________________
>>>> devel mailing list
>>>> devel at rtems.org
>>>> http://lists.rtems.org/mailman/listinfo/devel
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20230801/6ac82a08/attachment.htm>