Hi, everyone,
The paper submission deadline for this year’s LLVM in HPC workshop has been further extended to September 14th (AoE). We’re looking for a few additional submissions, so if you have anything that could be submitted as a paper by the beginning of next week, please take advantage of this opportunity. If you have any questions, please let me know.
SC20 is now a virtual event. Please see for more information.
-Hal
Hi, everyone,
A quick note: The LLVM in HPC workshop at SC20 is happening now. If you’re interested in joining the online event, you need to register for the SC20 workshops () which costs $50 ($20 for students), and that provides access to all SC20 workshops (live streams, videos, etc.). See the SC20 page for more information.
Time (EST) Speaker Title 10:00 Hal Finkel Welcome 10:05 Mehdi Amini Keynote: MLIR: an Agile Infrastructure for Building a Compiler Ecosystem 11:15 Coffee Break 11:30 Rahim Mammadli Static Neural Compiler Optimization via Deep Reinforcement Learning 12:10 Michael Kruse Autotuning Search Space for Loop Transformations 12:50 Dibyendu Das Deep Learning-based Approximate Graph-Coloring Algorithm for Register Allocation 1:30 Lunch Break 2:30 Alok Mishra Extending the LLVM/Clang Framework for OpenMP Metadirective Support 3:10 Andrew Lamzed-Short Towards Automated Kernel Fusion for the Optimisation of Scientific Applications 3:50 Alexis Engelke Robust Practical Binary Optimization at Run-time using LLVM 4:30 Coffee Break 4:45 Hal Finkel Really Embedding Domain-Specific Languages into C++ 5:25 All Presenters Panel Discussion 6:25 Final Discussion 6:30 The End
-Hal
Hi, everyone,
Slides from this year’s LLVM in HPC workshop presentations have been posted on and presentation videos have been posted on the LLVM YouTube channel.
We thank the presenters for making these materials available for public posting.
Happy holidays!
-Hal