Krzysztof Drewniak fbed673f52

[mlir][amdgpu] Update TDM ops to use the new barrier type, improve docs (#180572) ...

Now that we have an AMDGPU dialect type for the in-LDS barriers that the
tensor data mover can automatically visit, update the definition of the
tensor descriptor operations to use said types and document the behavior
of the barrier.

2026-02-09 13:37:57 -08:00

..

amdgpu-emulate-atomics.mlir

[mlir][AMDGPU] Enable emulating vector buffer_atomic_fadd for bf16 on gfx942 (#129029)

2025-03-13 14:30:45 -05:00

amdgpu-fold-memrefs.mlir

[mlir][amdgpu] fold memref.subview/expand_shape/collapse_shape into amdgpu.gather_to_lds for DST operand (#152277)

2025-08-08 05:47:33 -07:00

amdgpu-make-dma-descriptor-fold.mlir

[mlir][amdgpu] Lower make_gather_dma_descriptor. (#172083)

2025-12-15 13:32:57 -05:00

amdgpu-resolve-strided-metadata.mlir

…

canonicalize.mlir

[mlir][amdgpu] Remove redundant barriers (#175436)

2026-01-12 14:47:58 +03:00

inlining.mlir

[mlir][amdgpu] Add Inliner interface (#162873)

2025-10-10 21:34:00 +03:00

invalid.mlir

[mlir][amdgpu] Update TDM ops to use the new barrier type, improve docs (#180572)

2026-02-09 13:37:57 -08:00

maskedload-to-load.mlir

[mlir][AMDGPU] Add better load/store lowering for full mask (#146748)

2025-07-10 16:11:19 +01:00

ops.mlir

[mlir][amdgpu] Update TDM ops to use the new barrier type, improve docs (#180572)

2026-02-09 13:37:57 -08:00

resolve-shaped-type-result-dims.mlir

[MLIR][AMDGPU] Implement reifyDimOfResult for FatRawBufferCastOp (#171839)

2025-12-12 11:39:00 -08:00