Hanhan Wang 9325b8da17 [mlir][Linalg] Add conv ops with TF definition. ...

The dimension order of a filter in tensorflow is
[filter_height, filter_width, in_channels, out_channels], which is different
from current definition. The current definition follows TOSA spec. Add TF
version conv ops to .tc, so we do not have to insert a transpose op around a
conv op.

Reviewed By: antiagainst

Differential Revision: https://reviews.llvm.org/D96038

2021-02-10 22:59:38 -08:00

..

Bufferize.cpp

[mlir] drop unused statics

2021-01-26 13:30:45 +01:00

CMakeLists.txt

[mlir][sparse] sparse tensor storage implementation

2021-02-10 11:57:24 -08:00

CodegenStrategy.cpp

[mlir] Mark LogicalResult as LLVM_NODISCARD

2021-02-04 15:10:10 -08:00

DropUnitDims.cpp

Revert "Revert "Reorder MLIRContext location in BuiltinAttributes.h""

2021-02-08 10:39:58 +01:00

ElementwiseToLinalg.cpp

[MLIR][Linalg] Refactor transforms to use linalg::getDynOperands helper

2021-01-11 16:24:59 -08:00

Fusion.cpp

[mlir][Linalg] Drop SliceOp

2021-02-04 11:22:01 +00:00

FusionOnTensors.cpp

NFC; fix typo in comment

2021-02-10 21:34:29 -08:00

Generalization.cpp

[mlir] Mark LogicalResult as LLVM_NODISCARD

2021-02-04 15:10:10 -08:00

Hoisting.cpp

[mlir][Linalg] Fix pad hoisting.

2021-02-10 16:49:38 +00:00

Interchange.cpp

Revert "Revert "Reorder MLIRContext location in BuiltinAttributes.h""

2021-02-08 10:39:58 +01:00

Loops.cpp

[mlir] Mark LogicalResult as LLVM_NODISCARD

2021-02-04 15:10:10 -08:00

PassDetail.h

[mlir][Linalg] Add TensorsToBuffers support for Constant ops.

2020-10-08 13:15:45 +00:00

Promotion.cpp

[mlir] Mark LogicalResult as LLVM_NODISCARD

2021-02-04 15:10:10 -08:00

SparseLowering.cpp

[mlir][sparse] sparse tensor storage implementation

2021-02-10 11:57:24 -08:00

Sparsification.cpp

[mlir][sparse] sparse tensor storage implementation

2021-02-10 11:57:24 -08:00

Tiling.cpp

[mlir][Linalg] Fix crash when tileSizeComputationFunction is left unspecified

2021-02-10 22:47:05 +00:00

Transforms.cpp

[mlir][Linalg] Fix padding related bugs.

2021-02-08 18:59:24 +00:00

Vectorization.cpp

[mlir][Linalg] Add conv ops with TF definition.

2021-02-10 22:59:38 -08:00