Skip to content

[mlir][linalg] Add support for scalable vectorization of linalg.batch_mmt4d #152984

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
Aug 14, 2025
Merged

[mlir][linalg] Add support for scalable vectorization of linalg.batch_mmt4d #152984

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
8b0b165
[mlir][linalg] Add support for scalable vectorization of linalg.batch…
egebeysel Aug 8, 2025
e174db9
address comments & properly name vars
egebeysel Aug 13, 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
1 change: 1 addition & 0 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 @@ -2615,6 +2615,7 @@ vectorizeScalableVectorPrecondition(Operation *op,
isa<linalg::MatmulTransposeAOp>(op) ||
isa<linalg::DepthwiseConv1DNwcWcOp>(op) ||
isa<linalg::MatvecOp>(op) || isa<linalg::Mmt4DOp>(op) ||
isa<linalg::BatchMmt4DOp>(op) ||
hasReductionIterator(linalgOp));
}

Expand Down
124 changes: 109 additions & 15 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 @@ -880,22 +880,22 @@ func.func @mmt4d_scalable(%A: memref<16x16x8x1xf32>, %B: memref<16x16x?x1xf32>,
// CHECK-SAME: %[[A:.*]]: memref<16x16x8x1xf32>,
// CHECK-SAME: %[[B:.*]]: memref<16x16x?x1xf32>,
// CHECK-SAME: %[[C_IN:.*]]: memref<16x16x8x?xf32>) {
// CHECK: %[[VAL_0:.*]] = arith.constant 16 : index
// CHECK: %[[VAL_1:.*]] = arith.constant 16 : index
// CHECK: %[[VAL_2:.*]] = arith.constant 16 : index
// CHECK: %[[C16_M:.*]] = arith.constant 16 : index
// CHECK: %[[C16_N:.*]] = arith.constant 16 : index
// CHECK: %[[C16_K:.*]] = arith.constant 16 : index
// CHECK: %[[C8:.*]] = arith.constant 8 : index
// CHECK: %[[C2:.*]] = arith.constant 2 : index
// CHECK: %[[DIM_2:.*]] = memref.dim %[[B]], %[[C2]] : memref<16x16x?x1xf32>
// CHECK: %[[VAL_6:.*]] = arith.constant 1 : index
// CHECK: %[[C1:.*]] = arith.constant 1 : index
// CHECK: %[[VEC_A:.*]] = vector.transfer_read %[[A]]{{.*}} : memref<16x16x8x1xf32>, vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_1:.*]] = vector.create_mask %[[VAL_1]], %[[VAL_2]], %[[DIM_2]], %[[VAL_6]] : vector<16x16x[4]x1xi1>
// CHECK: %[[MASK_1:.*]] = vector.create_mask %[[C16_N]], %[[C16_K]], %[[DIM_2]], %[[C1]] : vector<16x16x[4]x1xi1>
// CHECK: %[[VEC_B:.*]] = vector.mask %[[MASK_1]] { vector.transfer_read %[[B]]{{.*}} : memref<16x16x?x1xf32>, vector<16x16x16x8x[4]x1xf32> } : vector<16x16x[4]x1xi1> -> vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_2:.*]] = vector.create_mask %[[VAL_0]], %[[VAL_1]], %[[C8]], %[[DIM_2]] : vector<16x16x8x[4]xi1>
// CHECK: %[[VAL_15:.*]] = vector.mask %[[MASK_2]] { vector.transfer_read %[[C_IN]]{{.*}} : memref<16x16x8x?xf32>, vector<16x16x8x[4]xf32> } : vector<16x16x8x[4]xi1> -> vector<16x16x8x[4]xf32>
// CHECK: %[[VAL_16:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_3:.*]] = vector.create_mask %[[VAL_0]], %[[VAL_1]], %[[VAL_2]], %[[C8]], %[[DIM_2]], %[[VAL_6]] : vector<16x16x16x8x[4]x1xi1>
// CHECK: %[[VAL_18:.*]] = vector.mask %[[MASK_3]] { vector.multi_reduction <add>, %[[VAL_16]], %[[VAL_15]] [2, 5] : vector<16x16x16x8x[4]x1xf32> to vector<16x16x8x[4]xf32> } : vector<16x16x16x8x[4]x1xi1> -> vector<16x16x8x[4]xf32>
// CHECK: vector.mask %[[MASK_2]] { vector.transfer_write %[[VAL_18]], %[[C_IN]]{{.*}} : vector<16x16x8x[4]xf32>, memref<16x16x8x?xf32> } : vector<16x16x8x[4]xi1>
// CHECK: %[[MASK_2:.*]] = vector.create_mask %[[C16_M]], %[[C16_N]], %[[C8]], %[[DIM_2]] : vector<16x16x8x[4]xi1>
// CHECK: %[[VEC_C:.*]] = vector.mask %[[MASK_2]] { vector.transfer_read %[[C_IN]]{{.*}} : memref<16x16x8x?xf32>, vector<16x16x8x[4]xf32> } : vector<16x16x8x[4]xi1> -> vector<16x16x8x[4]xf32>
// CHECK: %[[MUL:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_3:.*]] = vector.create_mask %[[C16_M]], %[[C16_N]], %[[C16_K]], %[[C8]], %[[DIM_2]], %[[C1]] : vector<16x16x16x8x[4]x1xi1>
// CHECK: %[[RED:.*]] = vector.mask %[[MASK_3]] { vector.multi_reduction <add>, %[[MUL]], %[[VEC_C]] [2, 5] : vector<16x16x16x8x[4]x1xf32> to vector<16x16x8x[4]xf32> } : vector<16x16x16x8x[4]x1xi1> -> vector<16x16x8x[4]xf32>
// CHECK: vector.mask %[[MASK_2]] { vector.transfer_write %[[RED]], %[[C_IN]]{{.*}} : vector<16x16x8x[4]xf32>, memref<16x16x8x?xf32> } : vector<16x16x8x[4]xi1>


module attributes {transform.with_named_sequence} {
Expand All @@ -920,10 +920,10 @@ func.func @mmt4d_scalable_with_assume(%A: memref<16x16x8x1xf32>, %B: memref<16x1
// CHECK-NOT: mask
// CHECK: %[[VEC_A:.*]] = vector.transfer_read %[[A]]{{.*}} : memref<16x16x8x1xf32>, vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[VEC_B:.*]] = vector.transfer_read %[[B]]{{.*}} : memref<16x16x?x1xf32>, vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[VAL_13:.*]] = vector.transfer_read %[[C_IN]]{{.*}} : memref<16x16x8x?xf32>, vector<16x16x8x[4]xf32>
// CHECK: %[[VAL_14:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[VAL_15:.*]] = vector.multi_reduction <add>, %[[VAL_14]], %[[VAL_13]] [2, 5] : vector<16x16x16x8x[4]x1xf32> to vector<16x16x8x[4]xf32>
// CHECK: vector.transfer_write %[[VAL_15]], %[[C_IN]]{{.*}} : vector<16x16x8x[4]xf32>, memref<16x16x8x?xf32>
// CHECK: %[[VEC_C:.*]] = vector.transfer_read %[[C_IN]]{{.*}} : memref<16x16x8x?xf32>, vector<16x16x8x[4]xf32>
// CHECK: %[[MUL:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<16x16x16x8x[4]x1xf32>
// CHECK: %[[RED:.*]] = vector.multi_reduction <add>, %[[MUL]], %[[VEC_C]] [2, 5] : vector<16x16x16x8x[4]x1xf32> to vector<16x16x8x[4]xf32>
// CHECK: vector.transfer_write %[[RED]], %[[C_IN]]{{.*}} : vector<16x16x8x[4]xf32>, memref<16x16x8x?xf32>

module attributes {transform.with_named_sequence} {
transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
Expand All @@ -933,6 +933,100 @@ module attributes {transform.with_named_sequence} {
}
}

// -----

///----------------------------------------------------------------------------------------
/// Tests for linalg.batch_mmt4d
///----------------------------------------------------------------------------------------

func.func @batch_mmt4d(%A: memref<2x16x16x8x1xf32>, %B: memref<2x16x16x8x1xf32>, %C_in: memref<2x16x16x8x8xf32>) {
linalg.batch_mmt4d ins(%A, %B: memref<2x16x16x8x1xf32>, memref<2x16x16x8x1xf32>)
outs(%C_in: memref<2x16x16x8x8xf32>)
return
}

// CHECK-LABEL: func.func @batch_mmt4d(
// CHECK-SAME: %[[A:.*]]: memref<2x16x16x8x1xf32>, %[[B:.*]]: memref<2x16x16x8x1xf32>, %[[C:.*]]: memref<2x16x16x8x8xf32>) {
// CHECK: %[[VEC_A:.*]] = vector.transfer_read %[[A]]{{.*}} : memref<2x16x16x8x1xf32>, vector<2x16x16x16x8x8x1xf32>
// CHECK: %[[VEC_B:.*]] = vector.transfer_read %[[B]]{{.*}} : memref<2x16x16x8x1xf32>, vector<2x16x16x16x8x8x1xf32>
// CHECK: %[[VEC_C:.*]] = vector.transfer_read %[[C]]{{.*}} : memref<2x16x16x8x8xf32>, vector<2x16x16x8x8xf32>
// CHECK: %[[MUL:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<2x16x16x16x8x8x1xf32>
// CHECK: %[[RED:.*]] = vector.multi_reduction <add>, %[[MUL]], %[[VEC_C]] [3, 6] : vector<2x16x16x16x8x8x1xf32> to vector<2x16x16x8x8xf32>
// CHECK: vector.transfer_write %[[RED]], %[[C]]{{.*}} : vector<2x16x16x8x8xf32>, memref<2x16x16x8x8xf32>

module attributes {transform.with_named_sequence} {
transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
%batch_mmt4d = transform.structured.match ops{["linalg.batch_mmt4d"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.vectorize %batch_mmt4d : !transform.any_op
transform.yield
}
}

// -----

func.func @batch_mmt4d_scalable(%A: memref<2x16x16x8x1xf32>, %B: memref<2x16x16x?x1xf32>, %C_in: memref<2x16x16x8x?xf32>) {
linalg.batch_mmt4d ins(%A, %B: memref<2x16x16x8x1xf32>, memref<2x16x16x?x1xf32>)
outs(%C_in: memref<2x16x16x8x?xf32>)
return
}
// CHECK-LABEL: func.func @batch_mmt4d_scalable(
// CHECK-SAME: %[[A:.*]]: memref<2x16x16x8x1xf32>,
// CHECK-SAME: %[[B:.*]]: memref<2x16x16x?x1xf32>,
// CHECK-SAME: %[[C_IN:.*]]: memref<2x16x16x8x?xf32>) {
// CHECK: %[[C2:.*]] = arith.constant 2 : index
// CHECK: %[[C16_M:.*]] = arith.constant 16 : index
// CHECK: %[[C16_N:.*]] = arith.constant 16 : index
// CHECK: %[[C16_K:.*]] = arith.constant 16 : index
// CHECK: %[[C8:.*]] = arith.constant 8 : index
// CHECK: %[[C3:.*]] = arith.constant 3 : index
// CHECK: %[[DIM_N_IN:.*]] = memref.dim %[[B]], %[[C3]] : memref<2x16x16x?x1xf32>
// CHECK: %[[C1:.*]] = arith.constant 1 : index
// CHECK: %[[VEC_A:.*]] = vector.transfer_read %[[A]]{{.*}} : memref<2x16x16x8x1xf32>, vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_1:.*]] = vector.create_mask %[[C2]], %[[C16_N]], %[[C16_K]], %[[DIM_N_IN]], %[[C1]] : vector<2x16x16x[4]x1xi1>
// CHECK: %[[VEC_B:.*]] = vector.mask %[[MASK_1]] { vector.transfer_read %[[B]]{{.*}} : memref<2x16x16x?x1xf32>, vector<2x16x16x16x8x[4]x1xf32> } : vector<2x16x16x[4]x1xi1> -> vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_2:.*]] = vector.create_mask %[[C2]], %[[C16_M]], %[[C16_N]], %[[C8]], %[[DIM_N_IN]] : vector<2x16x16x8x[4]xi1>
// CHECK: %[[VEC_C:.*]] = vector.mask %[[MASK_2]] { vector.transfer_read %[[C_IN]]{{.*}} : memref<2x16x16x8x?xf32>, vector<2x16x16x8x[4]xf32> } : vector<2x16x16x8x[4]xi1> -> vector<2x16x16x8x[4]xf32>
// CHECK: %[[MUL:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[MASK_3:.*]] = vector.create_mask %[[C2]], %[[C16_M]], %[[C16_N]], %[[C16_K]], %[[C8]], %[[DIM_N_IN]], %[[C1]] : vector<2x16x16x16x8x[4]x1xi1>
// CHECK: %[[RED:.*]] = vector.mask %[[MASK_3]] { vector.multi_reduction <add>, %[[MUL]], %[[VEC_C]] [3, 6] : vector<2x16x16x16x8x[4]x1xf32> to vector<2x16x16x8x[4]xf32> } : vector<2x16x16x16x8x[4]x1xi1> -> vector<2x16x16x8x[4]xf32>
// CHECK: vector.mask %[[MASK_2]] { vector.transfer_write %[[RED]], %[[C_IN]]{{.*}} : vector<2x16x16x8x[4]xf32>, memref<2x16x16x8x?xf32> } : vector<2x16x16x8x[4]xi1>

module attributes {transform.with_named_sequence} {
transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
%batch_mmt4d = transform.structured.match ops{["linalg.batch_mmt4d"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.vectorize %batch_mmt4d vector_sizes [2, 16, 16, 16, 8, [4], 1] : !transform.any_op
transform.yield
}
}

// -----

func.func @batch_mmt4d_scalable_with_assume(%A: memref<2x16x16x8x1xf32>, %B: memref<2x16x16x?x1xf32>, %C_in: memref<2x16x16x8x?xf32>) {
linalg.batch_mmt4d ins(%A, %B: memref<2x16x16x8x1xf32>, memref<2x16x16x?x1xf32>)
outs(%C_in: memref<2x16x16x8x?xf32>)
return
}
// CHECK-LABEL: func.func @batch_mmt4d_scalable_with_assume(
// CHECK-SAME: %[[A:.*]]: memref<2x16x16x8x1xf32>,
// CHECK-SAME: %[[B:.*]]: memref<2x16x16x?x1xf32>,
// CHECK-SAME: %[[C_IN:.*]]: memref<2x16x16x8x?xf32>) {
// CHECK-NOT: mask
// CHECK: %[[VEC_A:.*]] = vector.transfer_read %[[A]]{{.*}} : memref<2x16x16x8x1xf32>, vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[VEC_B:.*]] = vector.transfer_read %[[B]]{{.*}} : memref<2x16x16x?x1xf32>, vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[VEC_C:.*]] = vector.transfer_read %[[C_IN]]{{.*}} : memref<2x16x16x8x?xf32>, vector<2x16x16x8x[4]xf32>
// CHECK: %[[MUL:.*]] = arith.mulf %[[VEC_A]], %[[VEC_B]] : vector<2x16x16x16x8x[4]x1xf32>
// CHECK: %[[RED:.*]] = vector.multi_reduction <add>, %[[MUL]], %[[VEC_C]] [3, 6] : vector<2x16x16x16x8x[4]x1xf32> to vector<2x16x16x8x[4]xf32>
// CHECK: vector.transfer_write %[[RED]], %[[C_IN]]{{.*}} : vector<2x16x16x8x[4]xf32>, memref<2x16x16x8x?xf32>

module attributes {transform.with_named_sequence} {
transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
%batch_mmt4d = transform.structured.match ops{["linalg.batch_mmt4d"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.vectorize %batch_mmt4d vector_sizes [2, 16, 16, 16, 8, [4], 1] {assume_dynamic_dims_match_vec_sizes} : !transform.any_op
transform.yield
}
}


// -----

///----------------------------------------------------------------------------------------
Expand Down