Skip to content

[mlir][linalg] Allow fusing reshapes with nonparallel operands #130148

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Mar 7, 2025
Merged

[mlir][linalg] Allow fusing reshapes with nonparallel operands #130148

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f31cf5a
[mlir][linalg] Allow fusing reshapes with parallel operands
IanWood1 Mar 7, 2025
0201daa
Add test to check reduction reshape fusion
IanWood1 Mar 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 1 addition & 6 deletions mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -566,7 +566,6 @@ static bool isFusableWithReshapeByDimExpansion(LinalgOp linalgOp,
// - All the indexing maps for operands and results are projected
// permutations.
// - The fused tensor is not a scalar.
// - All the loops for the reshaped operand are parallel loops.
SmallVector<utils::IteratorType> iteratorTypes =
linalgOp.getIteratorTypesArray();
AffineMap operandMap = linalgOp.getMatchingIndexingMap(fusableOpOperand);
Expand All @@ -577,11 +576,7 @@ static bool isFusableWithReshapeByDimExpansion(LinalgOp linalgOp,
.getValue()
.isProjectedPermutation();
}) &&
operandMap.getNumResults() > 0 &&
llvm::all_of(operandMap.getResults(), [&](AffineExpr expr) {
return isParallelIterator(
iteratorTypes[cast<AffineDimExpr>(expr).getPosition()]);
});
operandMap.getNumResults() > 0;
}

namespace {
Expand Down
25 changes: 25 additions & 0 deletions mlir/test/Dialect/Linalg/reshape_fusion.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -482,6 +482,31 @@ func.func @generic_op_reshape_consumer_fusion_projected(%arg0 : tensor<?x?xf32>,

// -----

func.func @fuse_collapse_reduction(%arg0: tensor<10x10x20xf32>) -> tensor<100xf32> {
%c0 = arith.constant 0 : index
%c_0 = arith.constant 0.0 : f32
%0 = tensor.collapse_shape %arg0 [[0, 1], [2]] : tensor<10x10x20xf32> into tensor<100x20xf32>
%2 = tensor.empty() : tensor<100xf32>
%3 = linalg.fill ins(%c_0 : f32) outs(%2 : tensor<100xf32>) -> tensor<100xf32>
%4 = linalg.generic {
indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>, affine_map<(d0, d1) -> (d0)>],
iterator_types = ["parallel", "reduction"]}
ins(%0 : tensor<100x20xf32>) outs(%3 : tensor<100xf32>) {
^bb0(%arg1 : f32, %arg2: f32):
%4 = arith.addf %arg1, %arg2 : f32
linalg.yield %4 : f32
} -> tensor<100xf32>
return %4 : tensor<100xf32>
}

// CHECK: func @fuse_collapse_reduction
// CHECK-SAME: %[[ARG0:.+]]: tensor<10x10x20xf32>
// CHECK: %[[GENERIC:.+]] = linalg.generic
// CHECK-SAME: ins(%[[ARG0]] : tensor<10x10x20xf32>)
// CHECK: %[[COLLAPSE:.+]] = tensor.collapse_shape %[[GENERIC]]
// CHECK: return %[[COLLAPSE]]
// -----

func.func @no_fuse_dynamic_dims(%arg0: tensor<?x?xf32>) -> tensor<?xf32> {
%c0 = arith.constant 0 : index
%0 = tensor.collapse_shape %arg0 [[0, 1]] : tensor<?x?xf32> into tensor<?xf32>
Expand Down