https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116078
--- Comment #10 from Filip Kastl <pheeck at gcc dot gnu.org> --- If you say that there's no mistake in the patch that could cause the slowdown, I believe you. After all, you understand the patch and I don't. Sorry I didn't make that clear earlier :). However, I guess it is still possible that the changes triggered some bug in GCC that caused the slowdown and it would be nice to know what that is. Then of course you aren't obliged to investigate that. What I still see is that if I compile and run the benchmark with GCC before the commit in question, I get 361s runtime. When I do that with GCC after the commit, I get 398s runtime. As richi mentioned though, we don't know how much is this still relevant since we don't even know if the slowdown remains in the current trunk.
