This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: va_list: mn10300
- To: Richard Henderson <rth at cygnus dot com>
- Subject: Re: va_list: mn10300
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Wed, 28 Jul 1999 03:19:51 -0600
- cc: gcc-patches at gcc dot gnu dot org
- Reply-To: law at cygnus dot com
In message <19990728014156.A13166@cygnus.com>you write:
> I'm a bit curious why you're using __builtin_saveregs instead
> of setup_incoming_varargs. I'm guessing that would make
> current_function_internal_arg_pointer go away and things
> generally cleaner. I'm also guessing it is the way it is
> because pa does it this way. ;-)
Lost in the annals of history.
Probably because I couldn't get setup_incoming_varargs to actually
do what I wanted at the time, and knowing the PA port I just copied it's
general scheme.
> I didn't get to test this on the sim because there's something
> broken 64-bit-wise. I got prologue insns like `add 4294967296,sp'.
> I could have just fallen back to an x86 box but didn't.
Yup, that's a 64x32 problem of some kind. Presumably in the mn103 backend.
Once I've got a PA64 compiler functional I'll have the joys of having to
find all these problems so that I can still use my HPs for development :-)
I'll spin this one during the next egcs -> cygnus merge.
jeff