[llvm-commits] [PATCH] RFC: fix some problems in GeneralHash

[Chandler Carruth]

I just wanted to point out that I'm not sure how much time it's worth
investing micro-optimizing the current code...

There are both performance and robustness problems with the hashing
algorithm (and to an extent the API) in this class. I've now run it through
SMHasher, a testsuite designed to evaluate the performance and quality of
hashes, and it finds several issues that would be relevant even for the use
cases likely to be seen in LLVM and/or Clang (even inputs as small as 2
pointers).

I'm working on gathering all of the data, and also working on an
alternative implementation based on more robust hashing algorithms. I'm
hoping to have something Real Soon Now.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20120225/794490fd/attachment.html>