Diego Caballero 8ba018d72a [mlir][Vector] Add support for sub-byte transpose emulation (#80110) ...

This PR adds patterns to convert a sub-byte vector transpose into a
sequence of instructions that perform the transpose on i8 vector
elements. Whereas this rewrite may not lead to the absolute peak
performance, it should ensure correctness when dealing with sub-byte
transposes.

2024-01-31 17:26:50 -08:00

..

BufferizableOpInterfaceImpl.cpp

[mlir][IR] Rename "update root" to "modify op" in rewriter API (#78260)

2024-01-17 11:08:59 +01:00

Bufferize.cpp

…

CMakeLists.txt

[mlir][vector] Add emulation patterns for vector masked load/store (#74834)

2023-12-15 11:35:48 +00:00

LowerVectorBroadcast.cpp

[mlir][vector] Fix vector.broadcast lowering for scalable vectors (#66344)

2023-09-15 16:35:47 +01:00

LowerVectorContract.cpp

[mlir][vector] Fix invalid IR in ContractionOpLowering (#78130)

2024-01-16 09:40:24 +01:00

LowerVectorGather.cpp

[mlir][Vector] Add a rewrite pattern for gather over a strided memref (#72991)

2023-11-30 16:33:20 +00:00

LowerVectorMask.cpp

[mlir][vector] Handle empty MaskOp in LowerVectorMask, MaskOpRewritePattern (#72031)

2023-11-12 08:12:28 +01:00

LowerVectorMultiReduction.cpp

[mlir][Vector] Add support for scalable vectors in multi_reduction

2023-08-08 17:01:59 +00:00

LowerVectorScan.cpp

[mlir][vector][gpu] Align minf/maxf reduction kind names with arith (#75901)

2023-12-20 00:14:43 -05:00

LowerVectorShapeCast.cpp

[mlir][vector] add result type to vector.extract assembly format (#66499)

2023-09-28 11:11:16 +01:00

LowerVectorTransfer.cpp

[mlir][affine][nfc] cleanup deprecated T.cast style functions (#71269)

2023-11-14 13:01:19 +08:00

LowerVectorTranspose.cpp

[mlir][vector] Disable transpose -> shuffle lowering for scalable vectors (#79979)

2024-01-31 09:21:44 +00:00

SubsetOpInterfaceImpl.cpp

[mlir][vector] LISH: Implement SubsetOpInterface for transfer_read/write (#70629)

2023-11-01 12:19:30 +09:00

VectorDistribute.cpp

[mlir][IR] Rename "update root" to "modify op" in rewriter API (#78260)

2024-01-17 11:08:59 +01:00

VectorDropLeadUnitDim.cpp

[mlir] Exclude masked ops in VectorDropLeadUnitDim (#76468)

2024-01-20 19:37:46 -05:00

VectorEmulateMaskedLoadStore.cpp

[mlir][vector] Add emulation patterns for vector masked load/store (#74834)

2023-12-15 11:35:48 +00:00

VectorEmulateNarrowType.cpp

[mlir][Vector] Add support for sub-byte transpose emulation (#80110)

2024-01-31 17:26:50 -08:00

VectorInsertExtractStridedSliceRewritePatterns.cpp

[mlir] Move casting calls from methods to function calls

2023-05-12 11:21:25 +02:00

VectorTransferOpTransforms.cpp

[mlir][vector] Add pattern to drop unit dim from elementwise(a, b)) (#74817)

2023-12-13 20:29:12 +00:00

VectorTransferSplitRewritePatterns.cpp

[mlir][IR] Rename "update root" to "modify op" in rewriter API (#78260)

2024-01-17 11:08:59 +01:00

VectorTransforms.cpp

[mlir][vector] Drop inner unit dims for transfer ops on dynamic shapes. (#79752)

2024-01-29 00:30:19 -08:00

VectorUnroll.cpp

[mlir][vector] Modernize vector.transpose op (#72594)

2023-11-20 11:25:35 +01:00