shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
454,683 Commits 1,713 Branches 298 Tags
Commit Graph

12 Commits

Author SHA1 Message Date
Jie Fu
7707ed9727 [mlir] Fix two build warnings (NFC) 
/data/llvm-project/mlir/lib/Dialect/Tensor/Utils/Utils.cpp:62:11: error: comparison of integers of different signs: 'int64_t' (aka 'long') and 'size_t' (aka 'unsigned long') [-Werror,-Wsign-compare]
  if (dim >= shape.size())
      ~~~ ^  ~~~~~~~~~~~~
1 error generated.

/data/llvm-project/mlir/lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp:484:8: error: unused variable 'appendIndex' [-Werror,-Wunused-variable]
  auto appendIndex = [&](Value val, SmallVector<Value> &dynIndices,
       ^
1 error generated.
 2023-03-15 12:07:25 +08:00
Mahesh Ravishankar
d3fa067e69 [mlir][Tensor] Use folded evaluators in tiling implementation of tensor.pad. 
Reviewed By: springerm

Differential Revision: https://reviews.llvm.org/D145135
 2023-03-15 03:19:55 +00:00
Nicolas Vasilache
2f07d627a1 [mlir][Linalg] Refactor HoistPadding and add support for hoisting in the absence of packing loops. 
This revision cleans up the implementation of hoist padding and extends it to also work in the
absence of packing loops.
This allows better composition when hoisting the padded result of a DPS operation.

A systematic usage of RewriterBase is applied to the implementation.

Depends on: D144856

Differential Revision: https://reviews.llvm.org/D144855
 2023-02-28 05:21:57 -08:00
Aliia Khasanova
399638f98c Merge kDynamicSize and kDynamicSentinel into one constant. 
resolve conflicts

Differential Revision: https://reviews.llvm.org/D138282
 2022-11-21 13:01:26 +00:00
Mahesh Ravishankar
188b041bf5 [mlir][Tensor] Change createDimValues to return a list of OpFoldResults. 
Reviewed By: nicolasvasilache, hanchung, ThomasRaoux

Differential Revision: https://reviews.llvm.org/D136733
 2022-10-27 03:13:56 +00:00
Mahesh Ravishankar
94b8469a88 [mlir][Tensor] Add a helper build method for pad operations with constant padding. 
Drop the `createPadScalarOp` from Utils.h since it is a duplicate of
the `build` method added here.

Differential Revision: https://reviews.llvm.org/D136493
 2022-10-24 18:11:53 +00:00
Jakub Kuderski
abc362a107 [mlir][arith] Change dialect name from Arithmetic to Arith 
Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22.

Tested with:
`ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples`

and `bazel build --config=generic_clang @llvm-project//mlir:all`.

Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini

Differential Revision: https://reviews.llvm.org/D134762
 2022-09-29 11:23:28 -04:00
Frederik Gossen
2c3ca3b684 [MLIR] Add utility function to create values for all dimensions of a tensor value 
This is a variant of the already provided `createDynamicDimValues` helper.

Differential Revision: https://reviews.llvm.org/D131798
 2022-08-12 14:42:27 -04:00
Jacques Pienaar
04235d07ad [mlir] Update flipped accessors (NFC) 
Follow up with memref flipped and flipping any intermediate changes
made.
 2022-06-28 13:11:26 -07:00
Frederik Gossen
ff6ce9e8fc Add createDynamicDimValues to tensor dialect utils 
The function creates dim ops for each dynamic dimension of the raked tensor
argument and returns these as values.

Differential Revision: https://reviews.llvm.org/D127533
 2022-06-13 08:26:28 -04:00
gysit
d26c42af57 [mlir][linalg] Control dimensions to pad. 
This revision supports padding only a subset of the iteration dimensions via an additional padding-dimensions parameter. This control allows us to pad an operation in multiple steps. For example, one may want to pad only the output dimensions of a producer matmul fused into a consumer loop nest, before tiling and padding its reduction dimension.

Depends On D122309

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D122560
 2022-03-28 14:39:57 +00:00
Alexander Belyaev
fd0c6f5391 [mlir] Move linalg::PadTensorOp to tensor::PadOp. 
RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785

Differential Revision: https://reviews.llvm.org/D117892
 2022-01-21 20:02:39 +01:00