https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113255
--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> --- (In reply to Richard Biener from comment #4) > I'm unsure the above parallel is valid, isn't parallel executing stmts > in "parallel" (unspecified order)? I don't see anything invalid on it. In addition to the memory copying, it describes the other effects at the end of the pattern (that the counter goes to 0 and the di/si pointers are incremented by the original counter times 8). All the uses of pseudos in the pattern are the values of those pseudos before the instruction, then all the sets at the end set the updated pseudos.
