https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71437
--- Comment #5 from amker at gcc dot gnu.org --- (In reply to Richard Biener from comment #4) > With -fwhole-program there's no regression from GCC 6.2 to current trunk. > Without I still can see a small regression (here 0.86s vs 0.92s). > > From looking at the assembly it's hard to tell what the issue is. perf shows > hot spots at mispredicted branches it seems (for both good and bad case). > ... > at least this is the most notable difference in the innermost loops on GIMPLE > (plenty of differences in the outer loop stuff). > > Bin, any idea why IVO does the "bad" choice here? Thanks for the investigation, I will have a look.
