Skip to content

[mlir][linalg] Use ub.poison when vectorizing pack+unpack Ops #159536

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
Sep 23, 2025
Merged

[mlir][linalg] Use ub.poison when vectorizing pack+unpack Ops #159536

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
9006029
[mlir][linalg] Use ub.poison when vectorizing linalg.pack
banach-space Sep 18, 2025
470c798
Address suggestion from Diego, extend to linalg.unpack
banach-space Sep 22, 2025
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion mlir/include/mlir/Dialect/Vector/Utils/VectorUtils.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -227,7 +227,8 @@ bool isLinearizableVector(VectorType type);
///
/// Note: all read offsets are set to 0.
Value createReadOrMaskedRead(OpBuilder &builder, Location loc, Value source,
ArrayRef<int64_t> inputVectorSizes, Value padValue,
ArrayRef<int64_t> inputVectorSizes,
std::optional<Value> padValue = std::nullopt,
bool useInBoundsInsteadOfMasking = false,
ArrayRef<bool> inputScalableVecDims = {});

Expand Down
14 changes: 4 additions & 10 deletions mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1770,12 +1770,9 @@ vectorizeAsTensorPackOp(RewriterBase &rewriter, linalg::PackOp packOp,
rewriter.setInsertionPoint(packOp);

Location loc = packOp.getLoc();
auto padValue = packOp.getPaddingValue();
if (!padValue) {
padValue = arith::ConstantOp::create(
rewriter, loc,
rewriter.getZeroAttr(packOp.getSourceType().getElementType()));
}
std::optional<Value> padValue = packOp.getPaddingValue()
? std::optional(packOp.getPaddingValue())
: std::nullopt;

// If the input vector sizes are not provided, then the vector sizes are
// determined by the result tensor shape. In case the vector sizes aren't
Expand Down Expand Up @@ -1936,11 +1933,8 @@ vectorizeAsTensorUnpackOp(RewriterBase &rewriter, linalg::UnPackOp unpackOp,
}

// -- Generate the read operation --
auto padValue = arith::ConstantOp::create(
rewriter, loc,
rewriter.getZeroAttr(unpackOp.getSourceType().getElementType()));
Value readResult = vector::createReadOrMaskedRead(
rewriter, loc, unpackOp.getSource(), readVectorSizes, padValue,
rewriter, loc, unpackOp.getSource(), readVectorSizes, std::nullopt,
useInBoundsInsteadOfMasking, readScalableVectorFlags);

// -- Generate the transpose operation --
Expand Down
10 changes: 6 additions & 4 deletions mlir/lib/Dialect/Vector/Utils/VectorUtils.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -319,7 +319,7 @@ bool vector::isLinearizableVector(VectorType type) {
Value vector::createReadOrMaskedRead(OpBuilder &builder, Location loc,
Value source,
ArrayRef<int64_t> inputVectorSizes,
Value padValue,
std::optional<Value> padValue,
bool useInBoundsInsteadOfMasking,
ArrayRef<bool> inputScalableVecDims) {
assert(!llvm::is_contained(inputVectorSizes, ShapedType::kDynamic) &&
Expand All @@ -328,9 +328,11 @@ Value vector::createReadOrMaskedRead(OpBuilder &builder, Location loc,
auto sourceShape = sourceShapedType.getShape();
assert(sourceShape.size() == inputVectorSizes.size() &&
"expected same ranks.");
auto vectorType = VectorType::get(inputVectorSizes, padValue.getType(),
inputScalableVecDims);
assert(padValue.getType() == sourceShapedType.getElementType() &&
auto vectorType =
VectorType::get(inputVectorSizes, sourceShapedType.getElementType(),
inputScalableVecDims);
assert((!padValue.has_value() ||
padValue.value().getType() == sourceShapedType.getElementType()) &&
"expected same pad element type to match source element type");
int64_t readRank = inputVectorSizes.size();
auto zero = arith::ConstantIndexOp::create(builder, loc, 0);
Expand Down
4 changes: 2 additions & 2 deletions mlir/test/Dialect/Linalg/vectorization/linalg-ops-with-patterns.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -339,8 +339,8 @@ module attributes {transform.with_named_sequence} {
// CHECK-LABEL: func.func @test_vectorize_pack(
// CHECK-SAME: %[[VAL_0:.*]]: tensor<32x8x16xf32>,
// CHECK-SAME: %[[VAL_1:.*]]: tensor<4x1x32x16x2xf32>) -> tensor<4x1x32x16x2xf32> {
// CHECK: %[[VAL_2:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[VAL_3:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[VAL_2:.*]] = ub.poison : f32
// CHECK-DAG: %[[VAL_3:.*]] = arith.constant 0 : index
// CHECK: %[[VAL_4:.*]] = vector.transfer_read %[[VAL_0]]{{\[}}%[[VAL_3]], %[[VAL_3]], %[[VAL_3]]], %[[VAL_2]] {in_bounds = [true, true, true]} : tensor<32x8x16xf32>, vector<32x8x16xf32>
// CHECK: %[[VAL_5:.*]] = vector.shape_cast %[[VAL_4]] : vector<32x8x16xf32> to vector<32x4x2x1x16xf32>
// CHECK: %[[VAL_6:.*]] = vector.transpose %[[VAL_5]], [1, 3, 0, 4, 2] : vector<32x4x2x1x16xf32> to vector<4x1x32x16x2xf32>
Expand Down
66 changes: 33 additions & 33 deletions mlir/test/Dialect/Linalg/vectorization/linalg-ops.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1068,16 +1068,16 @@ module attributes {transform.with_named_sequence} {
// CHECK-SAME: %[[DEST:.*]]: tensor<?x?xf32>,
// CHECK-SAME: %[[SRC:.*]]: tensor<?x?x16x2xf32>
func.func @test_vectorize_dynamic_shapes_unpack_scalable_vec(%dest: tensor<?x?xf32>, %src: tensor<?x?x16x2xf32>) -> tensor<?x?xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00
// CHECK: %[[C01:.*]] = arith.constant 0
// CHECK: %[[C02:.*]] = arith.constant 0
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C01:.*]] = arith.constant 0
// CHECK-DAG: %[[C02:.*]] = arith.constant 0
// CHECK: %[[DIM4:.*]] = tensor.dim %[[SRC]], %[[C02]] : tensor<?x?x16x2xf32>
// CHECK: %[[CNST14:.*]] = arith.constant 1
// CHECK: %[[DIM6:.*]] = tensor.dim %[[SRC]], %[[CNST14]] : tensor<?x?x16x2xf32>
// CHECK: %[[CNST16:.*]] = arith.constant 16 : index
// CHECK: %[[CNST2:.*]] = arith.constant 2 : index
// CHECK: %[[MASK_READ:.*]] = vector.create_mask %[[DIM4]], %[[DIM6]], %[[CNST16]], %[[CNST2]] : vector<2x1x[16]x2xi1>
// CHECK: %[[READ:.*]] = vector.mask %[[MASK_READ]] {{.*}} vector.transfer_read %{{.*}} : tensor<?x?x16x2xf32>, vector<2x1x[16]x2xf32> } : vector<2x1x[16]x2xi1> -> vector<2x1x[16]x2xf32>
// CHECK: %[[READ:.*]] = vector.mask %[[MASK_READ]] {{.*}} vector.transfer_read %{{.*}} %[[PAD]] {{.*}}: tensor<?x?x16x2xf32>, vector<2x1x[16]x2xf32> } : vector<2x1x[16]x2xi1> -> vector<2x1x[16]x2xf32>
// CHECK: %[[TR:.*]] = vector.transpose %[[READ]], [0, 3, 1, 2] : vector<2x1x[16]x2xf32> to vector<2x2x1x[16]xf32>
// CHECK: %[[SC:.*]] = vector.shape_cast %[[TR]] : vector<2x2x1x[16]xf32> to vector<4x[16]xf32>
// CHECK: %[[MASK_WRITE:.*]] = vector.create_mask {{.*}} : vector<4x[16]xi1>
Expand All @@ -1100,17 +1100,17 @@ module attributes {transform.with_named_sequence} {
// CHECK-SAME: %[[DEST:.*]]: tensor<?x?xf32>,
// CHECK-SAME: %[[SRC:.*]]: tensor<?x?x?x2xf32>
func.func @test_vectorize_dynamic_shapes_unpack_scalable_vec_and_tile_size(%dest: tensor<?x?xf32>, %src: tensor<?x?x?x2xf32>) -> tensor<?x?xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00
// CHECK: %[[C01:.*]] = arith.constant 0
// CHECK: %[[C02:.*]] = arith.constant 0
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C01:.*]] = arith.constant 0
// CHECK-DAG: %[[C02:.*]] = arith.constant 0
// CHECK: %[[DIM4:.*]] = tensor.dim %[[SRC]], %[[C02]] : tensor<?x?x?x2xf32>
// CHECK: %[[C1_2:.*]] = arith.constant 1
// CHECK: %[[DIM6:.*]] = tensor.dim %[[SRC]], %[[C1_2]] : tensor<?x?x?x2xf32>
// CHECK: %[[C2:.*]] = arith.constant 2 : index
// CHECK: %[[DIM_2:.*]] = tensor.dim %[[SRC]], %[[C2]] : tensor<?x?x?x2xf32>
// CHECK: %[[C2_1:.*]] = arith.constant 2 : index
// CHECK: %[[MASK_READ:.*]] = vector.create_mask %[[DIM4]], %[[DIM6]], %[[DIM_2]], %[[C2_1]] : vector<2x1x[16]x2xi1>
// CHECK: %[[READ:.*]] = vector.mask %[[MASK_READ]] {{.*}} vector.transfer_read %{{.*}} : tensor<?x?x?x2xf32>, vector<2x1x[16]x2xf32> } : vector<2x1x[16]x2xi1> -> vector<2x1x[16]x2xf32>
// CHECK: %[[READ:.*]] = vector.mask %[[MASK_READ]] {{.*}} vector.transfer_read %{{.*}} %[[PAD]] {{.*}}: tensor<?x?x?x2xf32>, vector<2x1x[16]x2xf32> } : vector<2x1x[16]x2xi1> -> vector<2x1x[16]x2xf32>
// CHECK: %[[TR:.*]] = vector.transpose %[[READ]], [0, 3, 1, 2] : vector<2x1x[16]x2xf32> to vector<2x2x1x[16]xf32>
// CHECK: %[[SC:.*]] = vector.shape_cast %[[TR]] : vector<2x2x1x[16]xf32> to vector<4x[16]xf32>
// CHECK: %[[MASK_WRITE:.*]] = vector.create_mask {{.*}} : vector<4x[16]xi1>
Expand Down Expand Up @@ -1138,14 +1138,14 @@ module attributes {transform.with_named_sequence} {
// CHECK-SAME: %[[SRC:.*]]: tensor<8x8x32x16xf32>
// CHECK-SAME: %[[DEST:.*]]: tensor<256x128xf32>
func.func @test_vectorize_unpack(%source: tensor<8x8x32x16xf32>, %dest: tensor<256x128xf32>) -> tensor<256x128xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]]= arith.constant 0 : index
// CHECK: %[[C8:.*]] = arith.constant 8 : index
// CHECK: %[[C80:.*]] = arith.constant 8 : index
// CHECK: %[[C32:.*]] = arith.constant 32 : index
// CHECK: %[[C16:.*]] = arith.constant 16 : index
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]]= arith.constant 0 : index
// CHECK-DAG: %[[C8:.*]] = arith.constant 8 : index
// CHECK-DAG: %[[C80:.*]] = arith.constant 8 : index
// CHECK-DAG: %[[C32:.*]] = arith.constant 32 : index
// CHECK-DAG: %[[C16:.*]] = arith.constant 16 : index
// CHECK: %[[MSK0:.*]] = vector.create_mask %[[C8]], %[[C80]], %[[C32]], %[[C16]] : vector<16x8x32x16xi1>
// CHECK: %[[READ0:.*]] = vector.mask %[[MSK0]] { vector.transfer_read %[[SRC]]{{.*}}} : vector<16x8x32x16xi1> -> vector<16x8x32x16xf32>
// CHECK: %[[READ0:.*]] = vector.mask %[[MSK0]] { vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : vector<16x8x32x16xi1> -> vector<16x8x32x16xf32>
// CHECK: %[[TRANSP0:.*]] = vector.transpose %[[READ0]], [0, 2, 1, 3] : vector<16x8x32x16xf32> to vector<16x32x8x16xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP0]] : vector<16x32x8x16xf32> to vector<512x128xf32>
// CHECK: %[[C01:.*]] = arith.constant 0 : index
Expand All @@ -1171,9 +1171,9 @@ func.func @test_vectorize_unpack(%source: tensor<8x8x32x16xf32>, %dest: tensor<2
// CHECK-SAME: %[[SRC:.*]]: tensor<8x8x32x16xf32>
// CHECK-SAME: %[[DEST:.*]]: tensor<256x128xf32>
func.func @test_vectorize_unpack_no_masks(%source: tensor<8x8x32x16xf32>, %dest: tensor<256x128xf32>) -> tensor<256x128xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-AD: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK: %[[TRANSP:.*]] = vector.transpose %[[READ]], [0, 2, 1, 3] : vector<8x8x32x16xf32> to vector<8x32x8x16xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP]] : vector<8x32x8x16xf32> to vector<256x128xf32>
// CHECK: %[[C00:.*]] = arith.constant 0 : index
Expand All @@ -1196,9 +1196,9 @@ func.func @test_vectorize_unpack_no_masks(%source: tensor<8x8x32x16xf32>, %dest:
// CHECK-SAME: %[[SRC:.*]]: tensor<8x8x32x16xf32>
// CHECK-SAME: %[[DEST:.*]]: tensor<256x128xf32>
func.func @test_vectorize_unpack_with_outer_perm(%source: tensor<8x8x32x16xf32>, %dest: tensor<256x128xf32>) -> tensor<256x128xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK: %[[TRANSP:.*]] = vector.transpose %[[READ]], [1, 2, 0, 3] : vector<8x8x32x16xf32> to vector<8x32x8x16xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP]] : vector<8x32x8x16xf32> to vector<256x128xf32>
// CHECK: %[[C00:.*]] = arith.constant 0 : index
Expand All @@ -1221,9 +1221,9 @@ func.func @test_vectorize_unpack_no_masks(%source: tensor<8x8x32x16xf32>, %dest:
// CHECK-SAME: %[[SRC:.*]]: tensor<8x8x32x16xf32>
// CHECK-SAME: %[[DEST:.*]]: tensor<256x128xf32>
func.func @test_vectorize_unpack_no_vector_sizes(%source: tensor<8x8x32x16xf32>, %dest: tensor<256x128xf32>) -> tensor<256x128xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : tensor<8x8x32x16xf32>, vector<8x8x32x16xf32>
// CHECK: %[[TRANSP:.*]] = vector.transpose %[[READ]], [0, 2, 1, 3] : vector<8x8x32x16xf32> to vector<8x32x8x16xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP]] : vector<8x32x8x16xf32> to vector<256x128xf32>
// CHECK: %[[C00:.*]] = arith.constant 0 : index
Expand All @@ -1246,9 +1246,9 @@ func.func @test_vectorize_unpack_no_vector_sizes(%source: tensor<8x8x32x16xf32>,
// CHECK-SAME: %[[SRC:.*]]: tensor<8x4x16x16xf32>
// CHECK-SAME: %[[DEST:.*]]: tensor<64x127xf32>
func.func @test_vectorize_unpack_no_vector_sizes_slice_output(%source: tensor<8x4x16x16xf32>, %dest: tensor<64x127xf32>) -> tensor<64x127xf32> {
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}}} : tensor<8x4x16x16xf32>, vector<8x4x16x16xf32>
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : tensor<8x4x16x16xf32>, vector<8x4x16x16xf32>
// CHECK: %[[TRANSP:.*]] = vector.transpose %[[READ]], [1, 2, 0, 3] : vector<8x4x16x16xf32> to vector<4x16x8x16xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP]] : vector<4x16x8x16xf32> to vector<64x128xf32>
// CHECK: %[[C00:.*]] = arith.constant 0 : index
Expand All @@ -1275,9 +1275,9 @@ func.func @test_vectorize_unpack_no_vector_sizes_permute(%source: tensor<4x7x4xf
%0 = linalg.unpack %source outer_dims_perm=[1, 0] inner_dims_pos = [1] inner_tiles = [4] into %dest : tensor<4x7x4xf32> -> tensor<7x16xf32>
return %0 : tensor<7x16xf32>
}
// CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}}} : tensor<4x7x4xf32>, vector<4x7x4xf32>
// CHECK-DAG: %[[PAD:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %[[SRC]]{{.*}} %[[PAD]] {{.*}} : tensor<4x7x4xf32>, vector<4x7x4xf32>
// CHECK: %[[TRANSP:.*]] = vector.transpose %[[READ]], [1, 0, 2] : vector<4x7x4xf32> to vector<7x4x4xf32>
// CHECK: %[[SHAPC:.*]] = vector.shape_cast %[[TRANSP]] : vector<7x4x4xf32> to vector<7x16xf32>
// CHECK: %[[C00:.*]] = arith.constant 0 : index
Expand Down Expand Up @@ -1308,7 +1308,7 @@ func.func @test_vectorize_pack(%src: tensor<32x8x16xf32>, %dest: tensor<4x1x32x1
%pack = linalg.pack %src outer_dims_perm = [1, 2, 0] inner_dims_pos = [2, 1] inner_tiles = [16, 2] into %dest : tensor<32x8x16xf32> -> tensor<4x1x32x16x2xf32>
return %pack : tensor<4x1x32x16x2xf32>
}
// CHECK-DAG: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK-DAG: %[[CST:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %{{.*}}[%[[C0]], %[[C0]], %[[C0]]], %[[CST]]
// CHECK-SAME: {in_bounds = [true, true, true]} : tensor<32x8x16xf32>, vector<32x8x16xf32>
Expand Down Expand Up @@ -1376,7 +1376,7 @@ func.func @test_vectorize_dynamic_pack(%src: tensor<?x?xf32>, %dest: tensor<?x?x
return %pack : tensor<?x?x16x2xf32>
}

// CHECK-DAG: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK-DAG: %[[CST:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0_1:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[C0_0:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[C1_0:.*]] = arith.constant 1 : index
Expand Down Expand Up @@ -1417,7 +1417,7 @@ func.func @test_vectorize_pack_no_vector_sizes(%src: tensor<64x4xf32>, %dest: te
%pack = linalg.pack %src outer_dims_perm = [1, 0] inner_dims_pos = [0, 1] inner_tiles = [16, 2] into %dest : tensor<64x4xf32> -> tensor<2x4x16x2xf32>
return %pack : tensor<2x4x16x2xf32>
}
// CHECK-DAG: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
// CHECK-DAG: %[[CST:.*]] = ub.poison : f32
// CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
// CHECK: %[[READ:.*]] = vector.transfer_read %{{.*}}[%[[C0]], %[[C0]]], %[[CST]]
// CHECK-SAME: {in_bounds = [true, true]} : tensor<64x4xf32>, vector<64x4xf32>
Expand Down