libffi & powerpc

[Joel Sherrill]

Mike Stump wrote:
>
> On May 14, 2007, at 11:53 AM, Eric Christopher wrote:
>
>>
>> On May 14, 2007, at 11:35 AM, Mike Stump wrote:
>>
>>> On May 14, 2007, at 8:46 AM, Patrick Olinet wrote:
>>>> Running with gdb, it looks like the problem comes from the
>>>> ppc_closure.S file of the libffi/src/powerpc directory, at line 32 :
>>>
>>> Maybe something like:
>>>
>>> #ifndef _SOFT_FLOAT
>>>> stfd %f1, 48(%r1)
>>> #endif
>>>
>>> but then, you might have to have something like:
>>>
>>> #define CPP_SPEC \
>>> "%{msoft-float: -D_SOFT_FLOAT} \
>>>
>>> (picked at random from lynx.h) in the .h file for your port.
>>>
>>> :-)  But I'm sure there would be more work to do.
>>
>> FWIW the backend already does this in rs6000-c.c :)
>
> Yeah, I thought it was in there someplace, but my grep

FWIW a few years ago, we tried to make sure all the RTEMS targets
defined _SOFT_FLOAT when msoft-float was specified.  Hopefully
that is still around and consistent.  RTEMS is built multilib so we
need a conditional to let us know which multilib it is so we can do
the right thing.  I imagine newlib depends on them too in many places.

--joel
>>
>>
>> -eric
>