On 04/13/2011 01:45 AM, Michael Matz wrote: > And is there any chance to transform this: > > +#ifdef HAVE_cc0 > + if (!sets_cc0_p (insn)) > +#endif > + max_to = insn; > > into this: > > + if (!sets_cc0_p (insn)) > max_to = insn; > > ? Yes, that implies making sets_cc0_p be always there and return false, > and write the conditionals in the corrent way?
The correct fix is to eliminate cc0 targets. > I'll also note that the first hunk of your change is in a loop commented > with "Compute upper bound, bla ...", meaning to be a heuristic, and your > second change is this: It's not a heuristic. Both of these loops are necessary to compute a valid end point. !sets_cc0_p is just an additional condition that must be satisfied in both of them. > - if (!bitmap_intersect_p (test_set, local_merge_live)) > + if (!bitmap_intersect_p (test_set, local_merge_live) > +#ifdef HAVE_cc0 > + && !sets_cc0_p (insn) > +#endif > + ) > > It seems to me, that those insn then shouldn't have been in test_set from > the start, instead of fiddling with the users of test_set. Hence, is my > feeling of the patch being a hack-around of a deeper problem or it being > the wrong place to hack wrong? I don't understand this sentence. Bernd
