[Sorry for the delay] > Jeff/Eric, maybe you know whether it's usual to fend off against > paradoxical subregs for these kind of transforms.
No, the failure mode is quite unexpected and I don't recall anything similar. > PR rtl-optimization/127606 > * combine.cc (distribute_and_simplify_rtx): Gate the > (and (xor B C) (not A)) to (xor (ior A B) (ior A C)) > distribution on A not containing a paradoxical subreg. IMO that's good enough a solution for this kind of artificial bug reports. -- Eric Botcazou
