https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114855
--- Comment #30 from rguenther at suse dot de <rguenther at suse dot de> --- On Wed, 4 Sep 2024, amacleod at redhat dot com wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114855 > > --- Comment #29 from Andrew Macleod <amacleod at redhat dot com> --- > Huh. Do we calculate *all* paths ahead of time? Yes, we first identify all profitable threading candidates in a function and then perform the threading. > I tried running valgrind, which died, but up to that point it showed 77% of > the > time spend in the body of > back_jt_path_registry::adjust_paths_after_duplication () Hmm, I can have a look there if you don't beat me to it (though the profiles didn't have this part in AFAIR).
