https://github.com/mleleszi commented:
More of a general question looking at the complexity of the solution: have you actually measured that this perfect hash approach does better than a simple identity hash? I haven't looked at the hash function it generates in detail, but I assume it has a pretty big constant overhead over a simple modulo. Is it worth it over a few collisions? The lower and upper conversion tables will have only ~1400 entries and even the library author says PtrHash is optimized for tables with [at least a million entries](https://github.com/RagnarGrootKoerkamp/PtrHash?tab=readme-ov-file#input). https://github.com/llvm/llvm-project/pull/174798 _______________________________________________ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
