[Bug target/127302] [17 Regression] 505.mcf_r miscompiled since r17-3330 (allow backward threading to loop exit)

[law at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127302

--- Comment #4 from Jeffrey A. Law <law at gcc dot gnu.org> ---
So interestingly enough, mcf does not fail for me on our internal target.  So
this might be sensitive to the ascalon tuning.  Obviously if tuning changes
behavior, then there's a bug somewhere, but it may help with trying to narrow
down where within the benchmark things go wrong.