[mlir][memref] Improve type inference for rank-reducing subviews

The result shape of a rank-reducing subview cannot be inferred in the general case. Just the result rank is not enough. The only thin

[mlir][memref] Improve type inference for rank-reducing subviews

The result shape of a rank-reducing subview cannot be inferred in the general case. Just the result rank is not enough. The only thing that we can infer is the layout map.

This change also improves the bufferization patterns of tensor.extract_slice and tensor.insert_slice to fully support rank-reducing operations.

Differential Revision: https://reviews.llvm.org/D129144

show more ...

[mlir][tensor][bufferize][NFC] Clean up test case

Insert -split-input-file flag to make the test cases more stable.

Differential Revision: https://reviews.llvm.org/D129143

[mlir][MemRef] Fix a crash when expanding a scalar shape

In this case the reassociation is empty, yielding no strides for the
result type.

Differential Revision: https://reviews.llvm.org/D127232

[mlir] Add translation from tensor.reshape to memref.reshape

This patch augments the `tensor-bufferize` pass by adding a conversion
rule to translate ReshapeOp from the `tensor` dialect to the `memr

[mlir] Add translation from tensor.reshape to memref.reshape

This patch augments the `tensor-bufferize` pass by adding a conversion
rule to translate ReshapeOp from the `tensor` dialect to the `memref`
dialect, in addition to adding a unit test to validate the translation.

Reviewed By: springerm

Differential Revision: https://reviews.llvm.org/D125031

show more ...

Fix CollapsedLayoutMap for dim size 1 case

This change fixes `CollapsedLayoutMap` for cases where the collapsed
dims are size 1. The cases where inner most dims are size 1 and
noncontiguous can be r

Fix CollapsedLayoutMap for dim size 1 case

This change fixes `CollapsedLayoutMap` for cases where the collapsed
dims are size 1. The cases where inner most dims are size 1 and
noncontiguous can be represented by the strided form and therefore can
be allowed. For such cases, the new stride should be of the next entry
in an association whose dimension is not size 1. If the next entry is
dynamic, it's not possible to decide which stride to use at compilation
time and the stride is set to dynamic.

Differential Revision: https://reviews.llvm.org/D124137

show more ...

[mlir][bufferize][NFC] Use custom walk instead of GreedyPatternRewriter

The bufferization driver was previously using a GreedyPatternRewriter. This was problematic because bufferization must travers

[mlir][bufferize][NFC] Use custom walk instead of GreedyPatternRewriter

The bufferization driver was previously using a GreedyPatternRewriter. This was problematic because bufferization must traverse ops top-to-bottom. The GreedyPatternRewriter was previously configured via `useTopDownTraversal`, but this was a hack; this API was just meant for performance improvements and should not affect the result of the rewrite.

BEGIN_PUBLIC
No public commit message needed.
END_PUBLIC

Differential Revision: https://reviews.llvm.org/D123618

show more ...

[mlir][NFC] Update textual references of `func` to `func.func` in Tensor/Tosa/Vector tests

The special case parsing of `func` operations is being removed.

[mlir][tensor] Fix verifier and bufferization of collapse_shape

Insert a buffer copy unless the dims are guaranteed to be collapsible. In the verifier, accept collapses unless they are guaranteed to

[mlir][tensor] Fix verifier and bufferization of collapse_shape

Insert a buffer copy unless the dims are guaranteed to be collapsible. In the verifier, accept collapses unless they are guaranteed to be non-collapsible.

Differential Revision: https://reviews.llvm.org/D123316

show more ...

Reland [GreedPatternRewriter] Preprocess constants while building worklist when not processing top down

Reland Note: Adds a fix to properly mark a commutative operation as folded if we change the or

Reland [GreedPatternRewriter] Preprocess constants while building worklist when not processing top down

Reland Note: Adds a fix to properly mark a commutative operation as folded if we change the order
of its operands. This was uncovered by the fact that we no longer re-process constants.

This avoids accidentally reversing the order of constants during successive
application, e.g. when running the canonicalizer. This helps reduce the number
of iterations, and also avoids unnecessary changes to input IR.

Fixes #51892

Differential Revision: https://reviews.llvm.org/D122692

show more ...

[mlir][MemRef] Allow transposed layouts in ExpandShapeOp.

https://reviews.llvm.org/D122641 introduced fixes to the ExpandShapeOp verifier
but also introduced an artificial layout limitation that pre

[mlir][MemRef] Allow transposed layouts in ExpandShapeOp.

https://reviews.llvm.org/D122641 introduced fixes to the ExpandShapeOp verifier
but also introduced an artificial layout limitation that prevents the consideration of transposed layouts.

This revision fixes the omissions and reimplements the logic using saturated arithmetic which is more
idiomatic and avoids leaking internal implementation details.

Tests cases are added for transposed layouts.

Reviewed By: springerm

Differential Revision: https://reviews.llvm.org/D122845

show more ...

[mlir][tensor][bufferize] Support 0-d collapse_shape with offset

Differential Revision: https://reviews.llvm.org/D122901

Revert "[GreedPatternRewriter] Preprocess constants while building worklist when not processing top down"

This reverts commit 59bbc7a0851b6e0054bb3ed47df0958822f08880.

This exposes an issue breakin

Revert "[GreedPatternRewriter] Preprocess constants while building worklist when not processing top down"

This reverts commit 59bbc7a0851b6e0054bb3ed47df0958822f08880.

This exposes an issue breaking the contract of
`applyPatternsAndFoldGreedily` where we "converge" without applying
remaining patterns.

show more ...

[GreedPatternRewriter] Preprocess constants while building worklist when not processing top down

This avoids accidentally reversing the order of constants during successive
application, e.g. when ru

[GreedPatternRewriter] Preprocess constants while building worklist when not processing top down

This avoids accidentally reversing the order of constants during successive
application, e.g. when running the canonicalizer. This helps reduce the number
of iterations, and also avoids unnecessary changes to input IR.

Fixes #51892

Differential Revision: https://reviews.llvm.org/D122692

show more ...

[mlir][tensor] Fix bufferization of CollapseShapeOp / ExpandShapeOp

Infer a tighter MemRef type instead of always falling back to the most dynamic MemRef type. This is inefficient and caused op veri

[mlir][tensor] Fix bufferization of CollapseShapeOp / ExpandShapeOp

Infer a tighter MemRef type instead of always falling back to the most dynamic MemRef type. This is inefficient and caused op verification errors.

Differential Revision: https://reviews.llvm.org/D122649

show more ...

[mlir][bufferize] Extract buffer hoisting into separate function

This improves the modularity of the bufferization.

From now on, all ops that do not implement BufferizableOpInterface are considered

[mlir][bufferize] Extract buffer hoisting into separate function

This improves the modularity of the bufferization.

From now on, all ops that do not implement BufferizableOpInterface are considered hoisting barriers. Previously, all ops that do not implement the interface were not considered barriers and such ops had to be marked as barriers explicitly. This was unsafe because we could've hoisted across unknown ops where it was not safe to hoist.

As a side effect, this allows for cleaning up AffineBufferizableOpInterfaceImpl. This build unit no longer needed and can be deleted.

Differential Revision: https://reviews.llvm.org/D121519

show more ...

[mlir][bufferize] Support tensor.expand_shape and tensor.collapse_shape

Differential Revision: https://reviews.llvm.org/D112512

[mlir][tensor] Replace tensor-bufferize with BufferizableOpInterface impl

This commit switches the `tensor-bufferize` pass over to BufferizableOpInterface-based bufferization.

Differential Revision

[mlir][tensor] Replace tensor-bufferize with BufferizableOpInterface impl

This commit switches the `tensor-bufferize` pass over to BufferizableOpInterface-based bufferization.

Differential Revision: https://reviews.llvm.org/D118246

show more ...

[mlir] Extend `tensor.from_elements` to support N-D case.

RFC: https://llvm.discourse.group/t/rfc-extend-tensor-fromelementsop-to-n-d/4715

Differential Revision: https://reviews.llvm.org/D115821

[mlir] Add a missing pattern to bufferize tensor.rank.

Differential Revision: https://reviews.llvm.org/D115745

[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect.

https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712

Differential Revision: https://reviews.l

[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect.

https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712

Differential Revision: https://reviews.llvm.org/D114552

show more ...

[mlir:DialectConversion] Restructure how argument/target materializations get invoked

The current implementation invokes materializations
whenever an input operand does not have a mapping for the
de

[mlir:DialectConversion] Restructure how argument/target materializations get invoked

The current implementation invokes materializations
whenever an input operand does not have a mapping for the
desired type, i.e. it requires materialization at the earliest possible
point. This conflicts with goal of dialect conversion (and also the
current documentation) which states that a materialization is only
required if the materialization is supposed to persist after the
conversion process has finished.

This revision refactors this such that whenever a target
materialization "might" be necessary, we insert an
unrealized_conversion_cast to act as a temporary materialization.
This allows for deferring the invocation of the user
materialization hooks until the end of the conversion process,
where we actually have a better sense if it's actually
necessary. This has several benefits:

* In some cases a target materialization hook is no longer
necessary
When performing a full conversion, there are some situations
where a temporary materialization is necessary. Moving forward,
these users won't need to provide any target materializations,
as the temporary materializations do not require the user to
provide materialization hooks.

* getRemappedValue can now handle values that haven't been
converted yet
Before this commit, it wasn't well supported to get the remapped
value of a value that hadn't been converted yet (making it
difficult/impossible to convert multiple operations in many
situations). This commit updates getRemappedValue to properly
handle this case by inserting temporary materializations when
necessary.

Another code-health related benefit is that with this change we
can move a majority of the complexity related to materializations
to the end of the conversion process, instead of handling adhoc
while conversion is happening.

Differential Revision: https://reviews.llvm.org/D111620

show more ...

[MLIR] Replace std ops with arith dialect ops

Precursor: https://reviews.llvm.org/D110200

Removed redundant ops from the standard dialect that were moved to the
`arith` or `math` dialects.

Renamed

[MLIR] Replace std ops with arith dialect ops

Precursor: https://reviews.llvm.org/D110200

Removed redundant ops from the standard dialect that were moved to the
`arith` or `math` dialects.

Renamed all instances of operations in the codebase and in tests.

Reviewed By: rriddle, jpienaar

Differential Revision: https://reviews.llvm.org/D110797

show more ...

[mlir] Move BufferizeDimOp to Tensor/Transforms/Bufferize.cpp

Differential Revision: https://reviews.llvm.org/D105256

[mlir][tensor] Add tensor.dim operation

* Split memref.dim into two operations: memref.dim and tensor.dim. Both ops have the same builder interface and op argument names, so that they can be used wi

[mlir][tensor] Add tensor.dim operation

* Split memref.dim into two operations: memref.dim and tensor.dim. Both ops have the same builder interface and op argument names, so that they can be used with templates in patterns that apply to both tensors and memrefs (e.g., some patterns in Linalg).
* Add constant materializer to TensorDialect (needed for folding in affine.apply etc.).
* Remove some MemRefDialect dependencies, make some explicit.

Differential Revision: https://reviews.llvm.org/D105165

show more ...

[MLIR] Create memref dialect and move dialect-specific ops from std.

Create the memref dialect and move dialect-specific ops
from std dialect to this dialect.

Moved ops:
AllocOp -> MemRef_AllocOp
A

[MLIR] Create memref dialect and move dialect-specific ops from std.

Create the memref dialect and move dialect-specific ops
from std dialect to this dialect.

Moved ops:
AllocOp -> MemRef_AllocOp
AllocaOp -> MemRef_AllocaOp
AssumeAlignmentOp -> MemRef_AssumeAlignmentOp
DeallocOp -> MemRef_DeallocOp
DimOp -> MemRef_DimOp
MemRefCastOp -> MemRef_CastOp
MemRefReinterpretCastOp -> MemRef_ReinterpretCastOp
GetGlobalMemRefOp -> MemRef_GetGlobalOp
GlobalMemRefOp -> MemRef_GlobalOp
LoadOp -> MemRef_LoadOp
PrefetchOp -> MemRef_PrefetchOp
ReshapeOp -> MemRef_ReshapeOp
StoreOp -> MemRef_StoreOp
SubViewOp -> MemRef_SubViewOp
TransposeOp -> MemRef_TransposeOp
TensorLoadOp -> MemRef_TensorLoadOp
TensorStoreOp -> MemRef_TensorStoreOp
TensorToMemRefOp -> MemRef_BufferCastOp
ViewOp -> MemRef_ViewOp

The roadmap to split the memref dialect from std is discussed here:
https://llvm.discourse.group/t/rfc-split-the-memref-dialect-from-std/2667

Differential Revision: https://reviews.llvm.org/D98041

show more ...