[Mlir-commits] [mlir] [mlir][gpu] Lower the remaining subgroup_mma_elementwise ops to NVVM (PR #221288)

Hung-Kuan Tseng llvmlistbot at llvm.org
Sun Nov 1 06:00:29 PST 2026

Previous message: [Mlir-commits] [mlir] [mlir][arith] Expand the scaling ops with the value of the scale (PR #217892)
Next message: [Mlir-commits] [mlir] [mlir][AMDGPU] Keep alignment when lowering masked loads and stores (PR #221308)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Tim096 wrote:

Ping @fabianmcg @grypp @kuhar. This fixes an abort in `convert-gpu-to-nvvm`, and it still merges cleanly with main.

https://github.com/llvm/llvm-project/pull/221288

Previous message: [Mlir-commits] [mlir] [mlir][arith] Expand the scaling ops with the value of the scale (PR #217892)
Next message: [Mlir-commits] [mlir] [mlir][AMDGPU] Keep alignment when lowering masked loads and stores (PR #221308)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the Mlir-commits mailing list