On October 17, 2015 11:26:43 AM GMT+02:00, Eric Botcazou <ebotca...@adacore.com> wrote: >> Well, it would (I think) ICE on assigning a packed variant to a >non-packed >> variant of a strict that happens to get a non-BLKmode when not >packed. > >Is "it" GIMPLE here? My sentence was not very clear, I meant that I >don't see >why GIMPLE would have to care about whether there is a VCE or not in >the IL.
Huh, I thought your question was about the mode check in useless_type_conversion_p. >And AFAIK nobody answered the question: what do we gain by making this >change? >So far I have only seen breakages, suspicious fixes and code >duplication... Honza wants the structural equality predicate (operand_equal_p) complete (optimistically) for GIMPLE. Richard.
