https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115825
--- Comment #18 from Segher Boessenkool <segher at gcc dot gnu.org> --- So, you are saying the change made us realise some insns can never be optimised away? It should be obvious in much more fundamental ways that these insns can never be optimised away (simply because they are always executed), but heh. > If you now no longer get "trivial" stuff unrolled then you run into > the size limit. A single insn (always 4 bytes) is too much?!? There is no way to make non-empty code smaller, are you saying we will never get *anything* unrolled now?!?
