6 Commits

Author SHA1 Message Date
MaheshRavishankar
c077a4f305
[mlir][Tensor] Add pattern to fold concats of empty. (#98994)
A concatenation of empty tensors can be replaced by a single empty
tensor of the concatenated shape. Add this pattern to
`populateFoldTensorEmptyPatterns`.
2024-07-17 09:51:00 -07:00
Adam Siemieniuk
b586149475
[mlir][tensor] Fold pack and unpack of empty input tensor (#92247)
Extends `tensor.empty` folding patterns with pack and unpack consumers
to fold away the operations when their source is empty.
2024-05-22 18:01:14 +02:00
Matthias Springer
40052b08de [mlir][tensor] Add option to fold only tensor.empty with a single use
This is useful for transformations such as bufferization, which is looking for tensor.extract_slice/insert_slice pairs.

Also fix the documentation of the corresponding tranform op.

Differential Revision: https://reviews.llvm.org/D152455
2023-06-09 12:36:55 +02:00
Matthias Springer
758329dc7c [mlir][NFC] reifyResultShapes: Add extra error checking
This change adds a new helper function `mlir::reifyResultShapes` that calls the corresponding interface method and also checks the result produced by the implementation when running in debug mode. Bugs due to incorrect interface implementations can be difficult to debug.

This helper function also reduces the amount of code needed at call sites: the cast to `ReifyRankedShapedTypeOpInterface` is done in the helper function.

Differential Revision: https://reviews.llvm.org/D145777
2023-03-10 11:37:54 +01:00
Matthias Springer
2a5b13e722 [mlir][Interfaces] ReifyRankedShapedTypeOpInterface returns OpFoldResults
`reifyResultShapes` now returns `OpFoldResult`s instead of `Value`s. This is often more efficient because many transformations immediately attempt to extract a constant from the reified values.

Differential Revision: https://reviews.llvm.org/D145250
2023-03-06 08:41:28 +01:00
Alexander Belyaev
f6fb0a4f35 [mlir] Make patterns for folding tensor.empty optional.
At the moment, they are a part of EmptyOp::getCanonicalizationPatterns. When
extract_slice(tensor.empty) is rewritten as a new tensor.empty, it could
happen that we end up with two tensor.empty ops, since the original
tensor.empty can have two users. After bufferization such cases result in two
allocations.

Differential Revision: https://reviews.llvm.org/D139308
2022-12-07 23:01:34 +01:00