[mlir][Linalg] Deprecate `tileAndFuseLinalgOps` method and associated patterns.

The `tileAndFuseLinalgOps` is a legacy approach for tiling + fusion of
Linalg operations. Since it was also intended t

[mlir][Linalg] Deprecate `tileAndFuseLinalgOps` method and associated patterns.

The `tileAndFuseLinalgOps` is a legacy approach for tiling + fusion of
Linalg operations. Since it was also intended to work on operations
with buffer operands, this method had fairly complex logic to make
sure tile and fuse was correct even with side-effecting linalg ops.
While complex, it still wasnt robust enough. This patch deprecates
this method and thereby deprecating the tiling + fusion method for ops
with buffer semantics. Note that the core transformation to do fusion
of a producer with a tiled consumer still exists. The deprecation here
only removes methods that auto-magically tried to tile and fuse
correctly in presence of side-effects.

The `tileAndFuseLinalgOps` also works with operations with tensor
semantics. There are at least two other ways the same functionality
exists.
1) The `tileConsumerAndFuseProducers` method. This does a similar
transformation, but using a slightly different logic to
automatically figure out the legal tile + fuse code. Note that this
is also to be deprecated soon.
2) The prefered way uses the `TilingInterface` for tile + fuse, and
relies on the caller to set the tiling options correctly to ensure
that the generated code is correct.
As proof that (2) is equivalent to the functionality provided by
`tileAndFuseLinalgOps`, relevant tests have been moved to use the
interface, where the test driver sets the tile sizes appropriately to
generate the expected code.

Differential Revision: https://reviews.llvm.org/D129901

show more ...

[mlir] Use value instead of getValue (NFC)

[mlir] Handle linalg.index correctly in TilingInterface

The existing implementation of the TilingInterface for Linalg ops was not
modifying the `linalg.index` ops contained within other Linalg ops (

[mlir] Handle linalg.index correctly in TilingInterface

The existing implementation of the TilingInterface for Linalg ops was not
modifying the `linalg.index` ops contained within other Linalg ops (they need
to be summed up with the values of respective tile loop induction variables),
which led to the interface-based tiling being incorrect for any Linalg op with
index semantics.

In the process, fix the function performing the index offsetting to use the
pattern rewriter API instead of RAUW as it is being called from patterns and
may mess up the internal state of the rewriter. Also rename the function to
clearly catch all uses.

Depends On D129365

Reviewed By: mravishankar

Differential Revision: https://reviews.llvm.org/D129366

show more ...

[mlir] Update flipped accessors (NFC)

Follow up with memref flipped and flipping any intermediate changes
made.

Revert "Don't use Optional::hasValue (NFC)"

This reverts commit aa8feeefd3ac6c78ee8f67bf033976fc7d68bc6d.

Don't use Optional::hasValue (NFC)

[mlir] Don't use Optional::getValue (NFC)

[mlir][linalg][transform] Add TileOp to transform dialect

This commit adds a tiling op to the transform dialect as an external op.

Differential Revision: https://reviews.llvm.org/D124661

[mlir] Bubble up tensor.extract_slice above linalg operation

Bubble up extract_slice above Linalg operation.

A sequence of operations

%0 = linalg.<op> ... arg0, arg1, ...
%1 = tensor.extra

[mlir] Bubble up tensor.extract_slice above linalg operation

Bubble up extract_slice above Linalg operation.

A sequence of operations

%0 = linalg.<op> ... arg0, arg1, ...
%1 = tensor.extract_slice %0 ...

can be replaced with

%0 = tensor.extract_slice %arg0
%1 = tensor.extract_slice %arg1
%2 = linalg.<op> ... %0, %1, ...

This results in the reduce computation of the linalg operation.

The implementation uses the tiling utility functions. One difference
from the tiling process is that we don't need to insert the checking
code for the out-of-bound accesses. The use of the slice itself
represents that the code writer is sure about the boundary condition.
To avoid adding the boundary condtion check code, `omitPartialTileCheck`
is introduced for the tiling utility functions.

Differential Revision: https://reviews.llvm.org/D122437

show more ...

[mlir] Purge linalg.tiled_loop.

Differential Revision: https://reviews.llvm.org/D119415

[mlir] Use SmallBitVector instead of SmallDenseSet for AffineMap::compressSymbols

This is both more efficient and more ergonomic to use, as inverting a
bit vector is trivial while inverting a set is

[mlir] Use SmallBitVector instead of SmallDenseSet for AffineMap::compressSymbols

This is both more efficient and more ergonomic to use, as inverting a
bit vector is trivial while inverting a set is annoying.

Sadly this leaks into a bunch of APIs downstream, so adapt them as well.

This would be NFC, but there is an ordering dependency in MemRefOps's
computeMemRefRankReductionMask. This is now deterministic, previously it
was dependent on SmallDenseSet's unspecified iteration order.

Differential Revision: https://reviews.llvm.org/D119076

show more ...

Revert "Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead.""

This reverts commit 25bf6a2a9bc6ecb3792199490c70c4ce50a94aea.

Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead."

This reverts commit 016956b68081705ffee511c334e31e414fa1ddbf.
Reverting it to fix NVidia build without being in a hurry.

[mlir] Purge `linalg.copy` and use `memref.copy` instead.

Differential Revision: https://reviews.llvm.org/D118028

[mlir][Linalg] NFC - Modernize APIs and get rid of unnecessary tiling paterns.

Tiling patterns can be reduced to a single pattern by using interface-based patterns.

Differential Revision: https://r

[mlir][Linalg] NFC - Modernize APIs and get rid of unnecessary tiling paterns.

Tiling patterns can be reduced to a single pattern by using interface-based patterns.

Differential Revision: https://reviews.llvm.org/D116733

show more ...

Apply clang-tidy fixes for performance-for-range-copy to MLIR (NFC)

[mlir] Remove unused "using" (NFC)

Identified by misc-unused-using-decls.

[mlir][linalg] Replace LinalgOps.h and LinalgTypes.h by a single header.

After removing the range type, Linalg does not define any type. The revision thus consolidates the LinalgOps.h and LinalgType

[mlir][linalg] Replace LinalgOps.h and LinalgTypes.h by a single header.

After removing the range type, Linalg does not define any type. The revision thus consolidates the LinalgOps.h and LinalgTypes.h into a single Linalg.h header. Additionally, LinalgTypes.cpp is renamed to LinalgDialect.cpp to follow the convention adopted by other dialects such as the tensor dialect.

Depends On D115727

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D115728

show more ...

[mlir][Linalg] NFC - Drop Optional in favor of FailureOr

Differential revision: https://reviews.llvm.org/D112332

[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][linalg] Retire Linalg ConvOp.

The convolution op is one of the remaining hard coded Linalg operations that have no region attached. It got obsolete due to the OpDSL convolution operations. Re

[mlir][linalg] Retire Linalg ConvOp.

The convolution op is one of the remaining hard coded Linalg operations that have no region attached. It got obsolete due to the OpDSL convolution operations. Removing it allows us to delete specialized code and tests that are not needed for the OpDSL counterparts that rely on the standard code paths.

Test needed due to specialized implementations are removed. Tiling and fusion tests are replaced by variants using linalg.conv_2d.

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D111233

show more ...

[mlir][linalg] Add helper to update IndexOps after tiling (NFC).

Add the addTileLoopIvsToIndexOpResults method to shift the IndexOp results after tiling.

Reviewed By: nicolasvasilache

Differential

[mlir][linalg] Add helper to update IndexOps after tiling (NFC).

Add the addTileLoopIvsToIndexOpResults method to shift the IndexOp results after tiling.

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D109761

show more ...

[mlir][linalg] Tiling: Use loop ub in extract_slice size computation if possible

When tiling a LinalgOp, extract_slice/insert_slice pairs are inserted. To avoid going out-of-bounds when the tile siz

[mlir][linalg] Tiling: Use loop ub in extract_slice size computation if possible

When tiling a LinalgOp, extract_slice/insert_slice pairs are inserted. To avoid going out-of-bounds when the tile size does not divide the shape size evenly (at the boundary), AffineMin ops are inserted. Some ops have assumptions regarding the dimensions of inputs/outputs. E.g., in a `A * B` matmul, `dim(A, 1) == dim(B, 0)`. However, loop bounds use either `dim(A, 1)` or `dim(B, 0)`.

With this change, AffineMin ops are expressed in terms of loop bounds instead of tensor sizes. (Both have the same runtime value.) This simplifies canonicalizations.

Differential Revision: https://reviews.llvm.org/D109267

show more ...

[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][linalg] All StructuredOp parameters are inputs or outputs.

Adapt the StructuredOp verifier to ensure all operands are either in the input or the output group. The change is possible after add

[mlir][linalg] All StructuredOp parameters are inputs or outputs.

Adapt the StructuredOp verifier to ensure all operands are either in the input or the output group. The change is possible after adding support for scalar input operands (https://reviews.llvm.org/D104220).

Differential Revision: https://reviews.llvm.org/D104783

show more ...