From mboxrd@z Thu Jan 1 00:00:00 1970 From: Richard Henderson To: law@cygnus.com Cc: Richard Henderson , Jim Wilson , egcs@cygnus.com Subject: Re: alpha failure on 920810-1 Date: Sat, 09 May 1998 02:05:00 -0000 Message-id: <19980509020730.41243@dot.cygnus.com> References: <19980430173828.15223@dot.cygnus.com> <16568.894658694@hurl.cygnus.com> X-SW-Source: 1998-05/msg00311.html On Fri, May 08, 1998 at 02:18:14PM -0600, Jeffrey A Law wrote: > > * loop.c (get_condition): Don't combine when either compare is MODE_CC. > > * machmode.h (COMPLEX_MODE_P): New macro. > > * alpha.c (alpha_emit_conditional_branch): New function. Taken from > > the body of beq, additionally set the mode of the branch to CCmode for > > FP compares and not fast_math. > > (alpha_emit_conditional_move): Always use a compare insn for FP > > when not fast_math, as well as setting CCmode on the cmov. > > * alpha.md (beq, bne, blt, et al): Call alpha_emit_conditional_branch. Applied. > I note that your patch doesn't actually use COMPLEX_MODE_P, but I do > think we ought to have it to be consistent with the other *_MODE_P > tests. Heh, didn't even notice that got left in; a previous version of the patch did use it. I checked that in separately. r~