MLIR-to-RTL simulation flow: от linalg.matmul до systolic array
Привет! Хотел бы рассказать о своем MVP проекта hw-mlir-lab , где я использую MLIR для lowering операции умножения матриц ( matmul ) на systolic array, который я симулирую в Verilator.
https://habr.com/ru/articles/1045754/
#MLIR #verilog #verilator #rtl #asic #asic_design #compiler #hardware_acceleration #system_on_chip
This project developed the LAPIS compiler framework, built on the Multilevel Intermediate Representation (MLIR), to optimize sparse linear algebra operations and support performance portability across diverse architectures. The main innovation of LAPIS is the Kokkos dialect, which allows for lowering codes from a high productivity language to different architectures in an elegant way. The dialect also allows the conversion of lower-level MLIR code to C++ Kokkos code, facilitating the integration of scientific machine learning (SciML) models into applications. To extend LAPIS for distributed memory architectures, a new partition dialect was created to manage the distribution of sparse tensors and express communication patterns for sparse linear algebra operations. This dialect also supports the distributed execution of operators and includes algorithmic optimizations to minimize communication to improve performance. The project also demonstrates that MLIR can enable effective linear algebra-level optimizations, improving performance on different GPUs for both sparse and dense linear algebra kernels. Key applications of LAPIS include sparse linear algebra and graph kernels, TenSQL, a relational database management solution built on GraphBLAS, and the development of subgraph isomorphism and monomorphism kernels, showcasing performance portability. In summary, the LAPIS framework supports productivity, performance, portability, and distributed memory execution, while also enabling linear algebra-level optimizations that are challenging in traditional programming languages, with successful applications ranging from simple sparse linear algebra to complex graph kernels. | OSTI.GOV
CUDA Tile IR is an MLIR-based intermediate representation and compiler infrastructure for CUDA kernel optimization, focusing on tile-based computation patterns and optimizations targeting NVIDIA te...
At FOSDEM 2026, LLVM will again participate with a dedicated devroom, on Saturday afternoon January 31st, in Brussels. As possibly the largest European Open Source Conference, FOSDEM attracts more than 600 lectures and over 8000 hackers - many core contributors of the world’s leading open source projects. Complementing the LLVM developer meetings, the devroom at FOSDEM provides a great opportunity for LLVM developers and the wider open source community to get together, connect and discuss. We ...
Doka Studio with MLiR - Mario Marini @ Doka - 30 Aug feat. MLiR, Mario Marini
mattpd
Habr
GripNews
Houston Haynes
kbeyls
HGPU group
SESH.sx