Author: LLVM - Bewertung: 0x - Views:1
2026 EuroLLVM Developers' Meeting
https://llvm.org/devmtg/2026-04/
------
Title: Self-Contained, Target-Specific GEMM Code Generation in MLIR
Speaker: Adam Siemieniuk, Renato Golin, Rolf Morel
------
Slides: https://llvm.org/devmtg/2026-04/slides/quick_talk/quick_talk_golin_morel_siemieniuk.pdf
-----
We present an MLIR-based approach for generating target-specific, highly optimized GEMM kernels that is fully self-contained within the LLVM/MLIR compiler infrastructure and does not rely on external libraries such as LIBXSMM, Intel MKL, or oneDNN. Building on prior work in the TPP-MLIR compiler, we upstream FP32, BF16, and INT8 code-generation techniques into MLIR and propose a transform schedule that combines existing and newly upstreamed passes to lower matmul, batch matmul, and batch-reduce matmul operations into optimized kernels, achieving performance competitive with the LIBXSMM library.
-----
Videos Edited by Bash Films: http://www.BashFilms.com
SOCIAL SHARE CARD GENERATOR