Mahesh Ravishankar f12639d0d6 [mlir][Linalg] Avoid collapsing dimensions of linalg op that arent foldable. ...

The collapsing dimensions transformation is limited to only those
cases where the sequence of dimensions are contiguous in all the
ranges of the indexing maps of the operation. Add this check before
applying the transformation.

Differential Revision: https://reviews.llvm.org/D150176

2023-05-10 05:45:02 +00:00

..

BubbleUpExtractSlice.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

BufferizableOpInterfaceImpl.cpp

[mlir][bufferization][NFC] Rename getAliasingOpOperand/getAliasingOpResult

2023-02-01 10:07:41 +01:00

Bufferize.cpp

[MLIR] Update pass declarations to new autogenerated files

2022-08-31 12:28:45 +02:00

CMakeLists.txt

[mlir][linalg] Replace getUpperBoundForIndex implementation

2023-04-11 10:32:34 +09:00

ConstantFold.cpp

[mlir] Use std::optional instead of llvm::Optional (NFC)

2023-01-14 01:25:58 -08:00

ConvertConv2DToImg2Col.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

ConvertToDestinationStyle.cpp

[mlir][NFC] reifyResultShapes: Add extra error checking

2023-03-10 11:37:54 +01:00

DataLayoutPropagation.cpp

[MLIR][Linalg] Rename packElementWiseOp to packGenericOp (NFC)

2023-05-08 09:58:19 +02:00

DecomposeLinalgOps.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

Detensorize.cpp

[mlir][linalg] check if a function has a body during detensorizing

2023-05-02 15:39:00 -07:00

DropUnitDims.cpp

[mlir][MemRef] Move transform related functions in Transforms.h

2023-03-28 15:20:19 +02:00

ElementwiseOpFusion.cpp

[mlir][Linalg] Avoid collapsing dimensions of linalg op that arent foldable.

2023-05-10 05:45:02 +00:00

ElementwiseToLinalg.cpp

[mlir][Linalg] NFC - Extract an IndexingUtils from Linalg/Utils

2023-04-13 01:12:55 -07:00

EraseUnusedOperandsAndResults.cpp

[mlir][linalg] Fold duplicate and unused inputs in linalg.generic

2022-12-09 15:18:26 +01:00

FusePadOpWithLinalgProducer.cpp

[mlir][NFC] reifyResultShapes: Add extra error checking

2023-03-10 11:37:54 +01:00

Fusion.cpp

[mlir] Prevent implicit downcasting to interfaces

2023-04-20 16:31:54 -04:00

FusionOnTensors.cpp

[mlir] Prevent implicit downcasting to interfaces

2023-04-20 16:31:54 -04:00

Generalization.cpp

[mlir][Linalg] Fix Generalize transform

2023-03-31 00:38:06 -07:00

Hoisting.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

HoistPadding.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

InlineScalarOperands.cpp

[mlir] Rename getInputs->getDpsInputs and getOutputs->getDpsInits in DPS interface.

2022-10-28 15:41:12 +02:00

Interchange.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

Loops.cpp

[mlir] Prevent implicit downcasting to interfaces

2023-04-20 16:31:54 -04:00

NamedOpConversions.cpp

[mlir] Rename getInputs->getDpsInputs and getOutputs->getDpsInits in DPS interface.

2022-10-28 15:41:12 +02:00

Promotion.cpp

[mlir][linalg] Replace getUpperBoundForIndex implementation

2023-04-11 10:32:34 +09:00

Split.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

SplitReduction.cpp

[mlir] Prevent implicit downcasting to interfaces

2023-04-20 16:31:54 -04:00

SubsetHoisting.cpp

[mlir][Linalg] Reimplement hoisting on tensors as a subset-based transformation

2023-02-27 08:15:44 -08:00

SwapExtractSliceWithFillPatterns.cpp

[mlir][linalg] Swap tensor.extract_slice(linalg.fill)

2022-09-20 17:31:22 -04:00

Tiling.cpp

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

2023-04-20 11:19:21 +09:00

TilingInterfaceImpl.cpp

[mlir] Prevent implicit downcasting to interfaces

2023-04-20 16:31:54 -04:00

Transforms.cpp

[Transform] Support more case for the transform pad operation

2023-05-08 15:40:38 +02:00

Vectorization.cpp

[mlir][linalg] Relax masked_vectorize(pad) to allow zero low Values.

2023-05-01 14:07:53 -07:00