steakhal wrote: > Benchmarking is taking a while for me, and I can only really run benchmarks > with low run-time variance overnight. I will post more over time as I get > more benchmark projects added. > > Here is the result from benchmarking one C project, libsodium, so far. These > results are for 13 runs after some warmup runs. This is with a debug build > and default checkers enabled using CodeChecker: > > base/main branch of clang: > > * Mean: 681.3 seconds > > * Stdev: 12.8 seconds > > > This branch: > > * Mean: 678.1 seconds > > * Stdev 15.4 seconds > > > edit: This is with the change that would not be merged: that there is a > lookup in the trait for each memregion rather than only for those with > UnknownSpaceRegion.
I see no risk in this change to regress analysis time. Consequently, if I were you, I wouldn't run benchmarks. Unless my primary objective would be to dust off the SATest.py and get on hands experience using that tool. https://github.com/llvm/llvm-project/pull/123003 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
