Hsiangkai Wang 46ad5408fc
[mlir][gpu][vector] Lower Vector dialect to GPU for element-wise ops only (#159091)
Current convertVectorToMMAOps starts from vector.contract and finds its
dependencies as the targets to convert. In GPU dialect, we have
gpu.subgroup_mma_elementwise operation. We should be able to lower
element-wise operations to GPU MMA operations without vector.contract.
This patch adds this case to the pattern.
2025-09-17 09:10:52 +01:00
..