nickdesaulniers added a comment. I appreciate you taking the time to review, and find additional optimizations! It's not populated yet, but IIUC, this should be the link with more test results: http://llvm-compile-time-tracker.com/compare.php?from=7cc5307c73caa72f22edd4208b175e3c36eec46e&to=dffc1420451f674731cb36799c8ae084104ff0b5&stat=instructions
Again, if future travelers find that this change tickles a pathological case in their builds, please feel empowered to revert. We can always try to thread through the current API that we're expanding tokens of a GNU C Statement Expression, and thus should cache macro expansions. This is just less invasive and I didn't measure any slowdowns locally (only speedups for my pathological case). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D80681/new/ https://reviews.llvm.org/D80681 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
