> Or rather this one. Same hammer, different color. It turns out that > the rtlanal.c change caused problems, so I've got to use a home-brewn > equivalent of remove_reg_equal_equiv_notes_for_regno...
This does not fix pr55006 on x86_64-apple-darwin10;-(while the patch in http://gcc.gnu.org/ml/gcc-patches/2012-11/msg01607.html does). > Test case is unchanged so I'm omitting it here. This test passes on x86_64-apple-darwin10 for a clean trunk. Thanks for looking at the problem. Dominique
