Modifying reassociate for improved CSE: fairly large perf gains
http://lists.llvm.org/pipermail/llvm-dev/2017-October/118476.html [lists.llvm.org]
2018-07-25 16:48
Wed Oct 25 11:36:54 PDT 2017
When playing around with reassociate I noticed a seemingly obvious optimization that was not getting done anywhere in llvm… nor in gcc or ICC.
source: danluu