Greetings,
I'm an LLVM newcomer interested in participating in the next GSoC under the project "Improve parallelism-aware analyses and optimizations". I've already read "Compiler Optimizations for OpenMP" and "Compiler Optimizations for Parallel Programs" both by Doerfert and Finkel. Also, I've watched their two LLVM meeting developers conferences "Representing parallelism within LLVM" and "Optimizing indirections, using abstractions without ...", so I have some background in the problems they are trying to tackle. Could you tell me more about the project?, what variants do you have?, or what can I focus my application on?.
Thank you.