[LLVMdev] Autotuning parameters/heuristics within LLVM

Robert Stevenson 9golive9 at gmail.com
Thu Oct 2 07:19:49 PDT 2014


I am planning to begin a project to explore the space of tuning LLVM
internals in an effort to increase performance. I am wondering if
anyone can point to me any parameterizations, heuristics, or
priorities functions within LLVM that can be tuned/adjusted. So far,
I'm considering BranchProbabilityInfo and InlineCost. Does anyone have
any other suggestions?


More information about the llvm-dev mailing list