Teaching libffi to stdcall

Fri Nov 8 09:31:00 GMT 2002

Anthony Green wrote:
> If you just want to put a comment in the source about it, that would be
> fine. We don't need fastcall for JNI compatability, do we?

No, not at all. Besides, in the stock FSF GCC, fastcall isn't
even reliably implemented (as of 3.2). The patch to GCC 3.3
to finally support this was applied only the day before
(6 Nov).
>>4. GCC 2.95.x and 3.x seem to go to great lengths to ensure a
>> default stack alignment of 128 bits. At least the
>> ffi_call_SYSV in x86/win32.S does not seem to give an
>> aligned stack to the called function.
>>> We should probably fix this. Do you see how to change this?

Sort of. I'll give it a try.
>>5. On Win32, the ffitest program fails while testing closures.
>> Is this an expected failure?
>>> Do the same tests pass on Linux?

Yes.
Ranjit.
-- 
Ranjit Mathew Email: rmathew AT hotmail DOT com
Bangalore, INDIA. Web: http://ranjitmathew.tripod.com/