https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116564
--- Comment #7 from Richard Sandiford <rsandifo at gcc dot gnu.org> --- (In reply to Alex Coplan from comment #6) > So I'm testing the following to do this (which so far survives bootstrap on > aarch64): > > diff --git a/gcc/df-problems.cc b/gcc/df-problems.cc > index f32185b3eac..a7f9eae10d3 100644 > --- a/gcc/df-problems.cc > +++ b/gcc/df-problems.cc > @@ -3895,7 +3895,9 @@ df_simulate_defs (rtx_insn *insn, bitmap live) > > /* If the def is to only part of the reg, it does > not kill the other defs that reach here. */ > - if (!(DF_REF_FLAGS (def) & (DF_REF_PARTIAL | DF_REF_CONDITIONAL))) > + if (DF_REF_FLAGS (def) & (DF_REF_PARTIAL | DF_REF_CONDITIONAL)) > + bitmap_set_bit (live, dregno); > + else > bitmap_clear_bit (live, dregno); > } > } Thanks for the clear analysis. Based on that, I agree that this looks like the right fix FWIW. I see that the commit message in g:f60226fd72331398dd5fd239e9d1b4feccc91988 only covered definitions in SETs, not CLOBBERs.
