Markus Böck 6c9be27b52

[mlir][tensor] Fold identity reshape of 0d-tensors (#146375) ...

Just like 1d-tensors, reshapes of 0d-tensors (aka scalars) are always
no-folds as they only have one possible layout. This PR adds logic to
the `fold` implementation to optimize these away as is currently
implemented for 1d tensors.

2025-07-02 09:09:03 +02:00

..

bubble-reshapes.mlir

[mlir][Tensor] Generalize the pattern to swap tensor.collapse_shape -> tensor.expand_shape. (#133819)

2025-04-15 14:10:18 -07:00

bubble-up-extract-slice-op.mlir

[MLIR] Bubble up tensor.extract_slice through tensor.collapse_shape (#131982)

2025-04-02 21:06:43 +01:00

bufferize.mlir

[mlir][tensor] Relax input type requirement on tensor.splat (#145893)

2025-06-30 09:49:19 +02:00

canonicalize.mlir

[mlir][tensor] Fold identity reshape of 0d-tensors (#146375)

2025-07-02 09:09:03 +02:00

decompose-concat.mlir

[mlir][Tensor] Move concat operation decomposition as a method of the concat operation. (#116004)

2024-11-13 13:29:04 -08:00

drop-redundant-insert-slice-rank-expansion.mlir

[mlir][Tensor] Check for out-of-bounds slice in insert/extract_slice verifier (#130487)

2025-03-12 08:34:21 +01:00

extract-from-collapse-shape.mlir

[mlir][tensor] add tensor insert/extract op folders (#142458)

2025-06-03 09:16:03 -07:00

extract-slice-from-collapse-shape.mlir

[mlir][affine] Add static basis support to affine.delinearize (#113846)

2024-11-04 14:59:13 -06:00

fold-consecutive-insert-extract-slice.mlir

…

fold-constant-extract-slice.mlir

…

fold-empty-op.mlir

[mlir][tensor][linalg] Move Pack/UnPack Ops to Linalg (#123902)

2025-02-17 10:44:27 +00:00

fold-into-pack-and-unpack.mlir

feat(linalg): add a way to pass controlFn to foldIntoPackUnpackPatterns (#143685)

2025-07-01 07:22:38 -07:00

fold-reassociative-reshapes.mlir

[mlir][tensor] Fold unpadding collapse_shape into extract_slice (#93554)

2024-05-31 13:29:40 +02:00

fold-tensor-subset-ops-into-vector-transfers.mlir

[mlir] Drop outermost dims in slice rank reduction inference (#95020)

2024-06-25 12:33:02 -04:00

fold-tensor-subset-ops.mlir

[mlir][Tensor] Check for out-of-bounds slice in insert/extract_slice verifier (#130487)

2025-03-12 08:34:21 +01:00

int-range-inference.mlir

[mlir][IntRangeInference] Infer values for {memref,tensor}.dim (#122945)

2025-01-29 18:43:53 -06:00

invalid.mlir

[mlir][tensor] Relax input type requirement on tensor.splat (#145893)

2025-06-30 09:49:19 +02:00

mesh-spmdization.mlir

[mlir][vector] Support complete folding in single pass for vector.insert/vector.extract (#142124)

2025-06-18 09:26:04 -07:00

one-shot-bufferize-encodings.mlir

[mlir][bufferization] Fix OneShotBufferize when defaultMemorySpaceFn is used (#91524)

2024-11-26 09:45:57 -07:00

one-shot-bufferize-tensor-copy-insertion.mlir

[mlir][bufferization] Remove allow-return-allocs and create-deallocs pass options, remove bufferization.escape attribute (#66619)

2023-09-18 16:44:48 +02:00

one-shot-bufferize.mlir

[mlir][bufferization] Let bufferization.tensor_layout be any layout attr (#138567)

2025-05-06 11:21:52 -05:00

ops.mlir

[mlir][tensor] Relax input type requirement on tensor.splat (#145893)

2025-06-30 09:49:19 +02:00

reify-shapes.mlir

[mlir][memref] Add a new ReifyResultShapes pass (#145927)

2025-07-01 15:39:21 +02:00

resolve-shaped-type-result-dims.mlir

[mlir][tensor] Fix ReifyRankedShapedTypeOpInterface impl. of reshape ops

2023-08-24 12:23:10 +02:00

rewrite-as-constant.mlir

[mlir][tensor] Implement constant folder for tensor.pad (#92691)

2024-06-06 10:22:16 -04:00

tiling.mlir

[mlir][tensor][linalg] Move Pack/UnPack Ops to Linalg (#123902)

2025-02-17 10:44:27 +00:00

tracking-listener.mlir

[mlir][tensor] Fix bug in utility tensor::isCastLikeExtractSliceOp

2023-08-28 11:17:11 -06:00

transform-op-casting.mlir

[mlir][transform] Add an op for replacing values with function calls (#78398)

2024-01-19 13:21:52 -05:00

transform-op-make-loop-independent.mlir

[mlir] use transform-interpreter in test passes (#70040)

2023-10-24 16:12:34 +02:00

value-bounds-op-interface-impl.mlir

[mlir][tensor] Make tensor::PadOp a ReifyRankedShapedTypeOpInterface (#145867)

2025-06-26 14:40:57 +02:00