[mlir] Enable decoupling two kinds of greedy behavior. (#104649)

The greedy rewriter is used in many different flows and it has a lot of
convenience (work list management, debugging actions, tracin

[mlir] Enable decoupling two kinds of greedy behavior. (#104649)

The greedy rewriter is used in many different flows and it has a lot of
convenience (work list management, debugging actions, tracing, etc). But
it combines two kinds of greedy behavior 1) how ops are matched, 2)
folding wherever it can.

These are independent forms of greedy and leads to inefficiency. E.g.,
cases where one need to create different phases in lowering and is
required to applying patterns in specific order split across different
passes. Using the driver one ends up needlessly retrying folding/having
multiple rounds of folding attempts, where one final run would have
sufficed.

Of course folks can locally avoid this behavior by just building their
own, but this is also a common requested feature that folks keep on
working around locally in suboptimal ways.

For downstream users, there should be no behavioral change. Updating
from the deprecated should just be a find and replace (e.g., `find ./
-type f -exec sed -i
's|applyPatternsAndFoldGreedily|applyPatternsGreedily|g' {} \;` variety)
as the API arguments hasn't changed between the two.

show more ...

[mlir][tensor] Apply `InsertSliceOfTransferWriteOpFolder` only when `transfer_write` overwrites all elements of `insert_slice` (#108803)

Resolves #101708

The updated logic now correctly checks if

[mlir][tensor] Apply `InsertSliceOfTransferWriteOpFolder` only when `transfer_write` overwrites all elements of `insert_slice` (#108803)

Resolves #101708

The updated logic now correctly checks if `transfer_write` completely
overwrites `insert_slice` and only then applies the rewrite for this
pattern.

This check currently covers static sizes, for dynamic sizes
value bounds analysis is needed (see `TODO:`).

show more ...

[MLIR][Vector] Implement TransferReadOfExtractSliceOp as MaskableOpRewritePattern (#91960)

Split of https://github.com/llvm/llvm-project/pull/90835
Adds support for `TransferReadOfExtractSliceOpFol

[MLIR][Vector] Implement TransferReadOfExtractSliceOp as MaskableOpRewritePattern (#91960)

Split of https://github.com/llvm/llvm-project/pull/90835
Adds support for `TransferReadOfExtractSliceOpFolder` when the
`TransferReadOp` is inside a `MaskOp`.

show more ...

[mlir][vector] Remove duplicate tensor subset <-> vector transfer patterns

Remove patterns that fold tensor subset ops into vector transfer ops from the vector dialect. These patterns already exist

[mlir][vector] Remove duplicate tensor subset <-> vector transfer patterns

Remove patterns that fold tensor subset ops into vector transfer ops from the vector dialect. These patterns already exist in the tensor dialect.

Differential Revision: https://reviews.llvm.org/D154932

show more ...

[mlir][Affine][NFC] Wrap dialect in "affine" namespace

This cleanup aligns the affine dialect with all the other dialects.

Differential Revision: https://reviews.llvm.org/D148687

[mlir][Tensor] Add support for insert_slice in FoldTensorSubsetOps

Differential Revision: https://reviews.llvm.org/D148334

[mlir][Tensor] Add a FoldTensorSubsetOps pass and patterns

These patterns follow FoldMemRefAliasOps which is further refactored for reuse.
In the process, fix FoldMemRefAliasOps handling of strides

[mlir][Tensor] Add a FoldTensorSubsetOps pass and patterns

These patterns follow FoldMemRefAliasOps which is further refactored for reuse.
In the process, fix FoldMemRefAliasOps handling of strides for vector.transfer ops which was previously incorrect.

These opt-in patterns generalize the existing canonicalizations on vector.transfer ops.
In the future the blanket canonicalizations will be retired.
They are kept for now to minimize porting disruptions.

Differential Revision: https://reviews.llvm.org/D146624

show more ...