This is the mail archive of the
mailing list for the GCC project.
Re: gimplify.c fix for rs6000 and mips
- From: Andrew MacLeod <amacleod at redhat dot com>
- To: Richard Biener <richard dot guenther at gmail dot com>
- Cc: gcc-patches <gcc-patches at gcc dot gnu dot org>, Jan-Benedict Glaw <jbglaw at lug-owl dot de>
- Date: Thu, 10 Oct 2013 13:12:04 -0400
- Subject: Re: gimplify.c fix for rs6000 and mips
- Authentication-results: sourceware.org; auth=none
- References: <5256A73B dot 8010404 at redhat dot com> <CAFiYyc11CX1g4feELmMC6SoVjMx7fXyX0dX20nQaVz36Q8n76A at mail dot gmail dot com> <CAFiYyc0W75xogBY1E0eqjmrT+jKm17SopuT-OL1gHuPzF2FgWw at mail dot gmail dot com>
On 10/10/2013 10:35 AM, Richard Biener wrote:
yeah, gimplify.c already was calling it, so it seemed more appropriate
for a gimplification routine in gimplify.c. If we didn't move it,
gimpliify.c would need builtins.h for no reason other than to get the
prototype (once it was moved out of tree.h where it happened to be
declared) I didn't realize it would drag crud with it :-P
On Thu, Oct 10, 2013 at 4:30 PM, Richard Biener
On Thu, Oct 10, 2013 at 3:10 PM, Andrew MacLeod <firstname.lastname@example.org> wrote:
More fun target specific stuff. gimplify.c needs the back end supplied
macros/function for the va-arg padding functions to bootstrap...
Err ... don't we have the va_arg gimplify target hooks to hide this
kind of target dependency?
Ah - it's common code for the targets... which makes me believe
it belongs in targhooks.c, kind of.
Btw, you didn't move gimplify_va_arg_expr to gimplify.c for some reason?
seems like there should be a better solution for PAD_VARARGS_DOWN
(defined in config/targ.h) and ARGS_GROW_DOWNWARD.. and somewhere in
there something gets expanded to ARGS_SIZE_RTX which required expr.h
:-P . Its a bit ugly for sure.
At least it compiles now. I wont get to anything until next week as Im
almost out for the rest of the week now, but then I'll look to see if it
can be cleaned up a bit.