This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: RFA: Saving fixed call-saved registers in unwind functions and setjmp callers
- From: Eric Botcazou <ebotcazou at libertysurf dot fr>
- To: Richard Sandiford <richard at codesourcery dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Roman Zippel <zippel at linux-m68k dot org>
- Date: Tue, 11 Sep 2007 01:18:38 +0200
- Subject: Re: RFA: Saving fixed call-saved registers in unwind functions and setjmp callers
- References: <87ir72yxx7.fsf@firetop.home> <200709102105.38422.ebotcazou@libertysurf.fr> <87sl5m45n8.fsf@firetop.home>
> Nah, it's OK, thanks. I think Roman's patch does all I need. (TBH,
> I was tempted to add a boolean to struct function too, but shied away
> from it because I didn't think it'd be accepted. It does seem cleaner
> than redefining what "fixedness" means for !call_really_used registers.)
OK, sounds like we've finally found the right meaning for the flag. But I
agree with you that the fixed_regs check in reload1.c is redundant (there is
an assert in init_reg_sets_1 and I think that the invariant is preserved) so
if you feel brave enough...
--
Eric Botcazou