Workshop on MLIR ecosystem. The workshop format is similar to that of the LLVM meeting itself with a mix of technical talks and roundtables, with a focus on the MLIR ecosystem. We invite technical talks on MLIR components and use cases as well as downstream projects using or inspired by MLIR. Presentations of work-in-progress, radical design proposals and other wild ideas are encouraged! We are open to considering unconventional session types. This will be an in-person only workshop with focus on active discussions, the sessions will not be recorded.

Agenda:

• 9:00 AM: Session 1

  • ParaNova: An IR-Augmenting Parallel Compute DSL for Dataflow Accelerators - Bowen Yang and Raghu Prabhakar (SambaNova Systems)
  • An ML Compilation Stack for Autonomous Vehicles using MLIR - Raghavan Raman, Sambhav Jain, Aaron StGeorge, Kaz Sadeghi, Matthew Francis-Landau, Michael Feliz, Muhammad Abubakar, Srinath Avadhanula, Ze Zhang (Cruise)
  • MLIR-TensorRT - Christopher Bate Sagar Shelke, Zixin Huang, Bo Wang, Jinkyu Koo, Pavani Majety, and Jhalak Patel

• 10:15 AM - Break

• 10:45 AM: Session 2

  • Raising Named Ops from Linalg Generic Javed Absar (Qualcomm)
  • GPU Data-Tiling in IREE - Hanhan Wang, Benoit Jacob, Quinlan Dawkins, and Max Dawkins (AMD)
  • MLIR-based Data Tiling Design for AI Engines - Mahesh Ravishankar, Jorn Tuyls, Vivian Zhang, Abhishek Varma, James Newling, Maks Levanthal, and Nirvedh Meshram (AMD)

• 12:00 PM:  Lunch

• 1:35 PM: Session 3

  • ClangIR status update - Bruno Cardoso Lopes and Nathan Lanza (Meta)
  • The XLG framework: an MLIR replacement for ASTs - Fabian Mora-Cordero and Sunita Chandrasekaran
  • Model Explorer - Eric Yang, Jing Jin, and Na Li (Google)
  • RFC for an incubator project for frontend bindings - Maksim Levental (AMD)

• 3:15 PM:  Break

• 4:00 PM: Roundtables

• 5:00 PM: Closing