https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125855
--- Comment #3 from Josef Melcr <jmelcr at gcc dot gnu.org> --- I can only answer your second question, our only Intel machine (3rd generation Xeon) isn't many notable changes, apart from two speedups (don't know which patch did that): https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=803.70.0&plot.1=841.70.0&plot.2=822.70.0&plot.3=902.70.0&plot.4=1082.70.0&plot.5=1188.70.0&plot.6=1478.70.0&plot.7=784.70.0&; https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=804.70.0&plot.1=846.70.0&plot.2=831.70.0&plot.3=912.70.0&plot.4=1083.70.0&plot.5=1189.70.0&plot.6=1482.70.0&plot.7=788.70.0&; As for your comment, providing net changes for individual patches is impossible from a time standpoint, sadly. We run benchmarks on a schedule (some are run daily, some only once a week...) with the current trunk. A single run can include tens of patches. Running the entire suite for a single patch would take a few days on the low end. I understand your frustration and agree with you that it is hard to grasp the impact of a change with only regressions to look at, but that's the way it is unfortunately.
