[mlir][vector] Relax the requirements on broadcast dims

mlir/include/mlir/Dialect/Vector/IR/VectorOps.td

@@ -1290,12 +1290,12 @@ def Vector_TransferReadOp :
     specifies if the transfer is guaranteed to be within the source bounds. If
     set to "false", accesses (including the starting point) may run
     out-of-bounds along the respective vector dimension as the index increases.
-    Non-vector and broadcast dimensions *must* always be in-bounds. The
-    `in_bounds` array length has to be equal to the vector rank. This attribute
-    has a default value: `false` (i.e. "out-of-bounds"). When skipped in the
-    textual IR, the default value is assumed. Similarly, the OP printer will
-    omit this attribute when all dimensions are out-of-bounds (i.e. the default
-    value is used).
+    Non-vector dimensions *must* always be in-bounds. The `in_bounds` array
+    length has to be equal to the vector rank. This attribute has a default
+    value: `false` (i.e. "out-of-bounds"). When skipped in the textual IR, the
+    default value is assumed. Similarly, the OP printer will omit this
+    attribute when all dimensions are out-of-bounds (i.e. the default value is
+    used).
 
     A `vector.transfer_read` can be lowered to a simple load if all dimensions
     are specified to be within bounds and no `mask` was specified.
@@ -1535,12 +1535,12 @@ def Vector_TransferWriteOp :
     specifies if the transfer is guaranteed to be within the source bounds. If
     set to "false", accesses (including the starting point) may run
     out-of-bounds along the respective vector dimension as the index increases.
-    Non-vector and broadcast dimensions *must* always be in-bounds. The
-    `in_bounds` array length has to be equal to the vector rank. This attribute
-    has a default value: `false` (i.e. "out-of-bounds"). When skipped in the
-    textual IR, the default value is assumed. Similarly, the OP printer will
-    omit this attribute when all dimensions are out-of-bounds (i.e. the default
-    value is used).
+    Non-vector dimensions *must* always be in-bounds. The `in_bounds` array
+    length has to be equal to the vector rank. This attribute has a default
+    value: `false` (i.e. "out-of-bounds"). When skipped in the textual IR, the
+    default value is assumed. Similarly, the OP printer will omit this
+    attribute when all dimensions are out-of-bounds (i.e. the default value is
+    used).
 
      A `vector.transfer_write` can be lowered to a simple store if all
      dimensions are specified to be within bounds and no `mask` was specified.

mlir/include/mlir/Interfaces/VectorInterfaces.td

@@ -234,12 +234,9 @@ def VectorTransferOpInterface : OpInterface<"VectorTransferOpInterface"> {
       return constExpr && constExpr.getValue() == 0;
     }
 
-    /// Return "true" if the vector transfer dimension `dim` is in-bounds. Also
-    /// return "true" if the dimension is a broadcast dimension. Return "false"
-    /// otherwise.
+    /// Return "true" if the vector transfer dimension `dim` is in-bounds.
+    /// Return "false" otherwise.
     bool isDimInBounds(unsigned dim) {
-      if ($_op.isBroadcastDim(dim))
-        return true;
       auto inBounds = $_op.getInBounds();
       return ::llvm::cast<::mlir::BoolAttr>(inBounds[dim]).getValue();
     }

mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp

@@ -1223,19 +1223,8 @@ static Operation *vectorizeAffineLoad(AffineLoadOp loadOp,
   LLVM_DEBUG(dbgs() << "\n[early-vect]+++++ permutationMap: ");
   LLVM_DEBUG(permutationMap.print(dbgs()));
 
-  // Make sure that the in_bounds attribute corresponding to a broadcast dim
-  // is set to `true` - that's required by the xfer Op.
-  // FIXME: We're not veryfying whether the corresponding access is in bounds.
-  // TODO: Use masking instead.
-  SmallVector<unsigned> broadcastedDims = permutationMap.getBroadcastDims();
-  SmallVector<bool> inBounds(vectorType.getRank(), false);
-
-  for (auto idx : broadcastedDims)
-    inBounds[idx] = true;
-
   auto transfer = state.builder.create<vector::TransferReadOp>(
-      loadOp.getLoc(), vectorType, loadOp.getMemRef(), indices, permutationMap,
-      inBounds);
+      loadOp.getLoc(), vectorType, loadOp.getMemRef(), indices, permutationMap);
 
   // Register replacement for future uses in the scope.
   state.registerOpVectorReplacement(loadOp, transfer);

mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp

@@ -1380,17 +1380,8 @@ vectorizeAsLinalgGeneric(RewriterBase &rewriter, VectorizationState &state,
 
     SmallVector<Value> indices(linalgOp.getShape(opOperand).size(), zero);
 
-    // Make sure that the in_bounds attribute corresponding to a broadcast dim
-    // is `true`
-    SmallVector<unsigned> broadcastedDims = readMap.getBroadcastDims();
-    SmallVector<bool> inBounds(readType.getRank(), false);
-
-    for (auto idx : broadcastedDims)
-      inBounds[idx] = true;
-
     Operation *read = rewriter.create<vector::TransferReadOp>(
-        loc, readType, opOperand->get(), indices, readMap,
-        ArrayRef<bool>(inBounds));
+        loc, readType, opOperand->get(), indices, readMap);
     read = state.maskOperation(rewriter, read, linalgOp, indexingMap);
     Value readValue = read->getResult(0);
 

mlir/lib/Dialect/Vector/IR/VectorOps.cpp

@@ -3947,10 +3947,6 @@ verifyTransferOp(VectorTransferOpInterface op, ShapedType shapedType,
                            "as permutation_map results: ")
            << AffineMapAttr::get(permutationMap)
            << " vs inBounds of size: " << inBounds.size();
-  for (unsigned int i = 0, e = permutationMap.getNumResults(); i < e; ++i)
-    if (isa<AffineConstantExpr>(permutationMap.getResult(i)) &&
-        !llvm::cast<BoolAttr>(inBounds.getValue()[i]).getValue())
-      return op->emitOpError("requires broadcast dimensions to be in-bounds");
 
   return success();
 }
@@ -4138,22 +4134,43 @@ static LogicalResult foldTransferInBoundsAttribute(TransferOp op) {
   bool changed = false;
   SmallVector<bool, 4> newInBounds;
   newInBounds.reserve(op.getTransferRank());
+  // Idxs of non-bcast dims - used when analysing bcast dims.
+  SmallVector<unsigned> nonBcastDims;
+
+  // 1. Process non-broadcast dims
   for (unsigned i = 0; i < op.getTransferRank(); ++i) {
-    // Already marked as in-bounds, nothing to see here.
+    // 1.1. Already marked as in-bounds, nothing to see here.
     if (op.isDimInBounds(i)) {
       newInBounds.push_back(true);
       continue;
     }
-    // Currently out-of-bounds, check whether we can statically determine it is
-    // inBounds.
+    // 1.2. Currently out-of-bounds, check whether we can statically determine
+    // it is inBounds.
+    bool inBounds = false;
     auto dimExpr = dyn_cast<AffineDimExpr>(permutationMap.getResult(i));
-    assert(dimExpr && "Broadcast dims must be in-bounds");
-    auto inBounds =
-        isInBounds(op, /*resultIdx=*/i, /*indicesIdx=*/dimExpr.getPosition());
+    if (dimExpr) {
+      inBounds = isInBounds(op, /*resultIdx=*/i,
+                            /*indicesIdx=*/dimExpr.getPosition());
+      nonBcastDims.push_back(i);
+    }
+
     newInBounds.push_back(inBounds);
     // We commit the pattern if it is "more inbounds".
     changed |= inBounds;
   }
+
+  // 2. Handle broadcast dims
+  // If all non-broadcast dims are "in bounds", then all bcast dims should be
+  // "in bounds" as well.
+  bool allNonBcastDimsInBounds = llvm::all_of(
+      nonBcastDims, [&newInBounds](unsigned idx) { return newInBounds[idx]; });
+  if (allNonBcastDimsInBounds) {
+    for (size_t idx : permutationMap.getBroadcastDims()) {
+      changed |= !newInBounds[idx];
+      newInBounds[idx] = true;
+    }
+  }
+
   if (!changed)
     return failure();
   // OpBuilder is only used as a helper to build an I64ArrayAttr.

mlir/test/Conversion/VectorToSCF/vector-to-scf.mlir

@@ -133,7 +133,7 @@ func.func @materialize_read(%M: index, %N: index, %O: index, %P: index) {
     affine.for %i1 = 0 to %N {
       affine.for %i2 = 0 to %O {
         affine.for %i3 = 0 to %P step 5 {
-          %f = vector.transfer_read %A[%i0, %i1, %i2, %i3], %f0 {in_bounds = [false, true, false], permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, 0, d0)>} : memref<?x?x?x?xf32>, vector<5x4x3xf32>
+          %f = vector.transfer_read %A[%i0, %i1, %i2, %i3], %f0 {permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, 0, d0)>} : memref<?x?x?x?xf32>, vector<5x4x3xf32>
           // Add a dummy use to prevent dead code elimination from removing
           // transfer read ops.
           "dummy_use"(%f) : (vector<5x4x3xf32>) -> ()
@@ -507,7 +507,7 @@ func.func @transfer_read_with_tensor(%arg: tensor<f32>) -> vector<1xf32> {
     // CHECK-NEXT: %[[RESULT:.*]] = vector.broadcast %[[EXTRACTED]] : f32 to vector<1xf32>
     // CHECK-NEXT: return %[[RESULT]] : vector<1xf32>
     %f0 = arith.constant 0.0 : f32
-    %0 = vector.transfer_read %arg[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
+    %0 = vector.transfer_read %arg[], %f0 {permutation_map = affine_map<()->(0)>} :
       tensor<f32>, vector<1xf32>
     return %0: vector<1xf32>
 }
@@ -746,7 +746,7 @@ func.func @cannot_lower_transfer_read_with_leading_scalable(%arg0: memref<?x4xf3
 func.func @does_not_crash_on_unpack_one_dim(%subview:  memref<1x1x1x1xi32>, %mask: vector<1x1xi1>) -> vector<1x1x1x1xi32> {
   %c0 = arith.constant 0 : index
   %c0_i32 = arith.constant 0 : i32
-  %3 = vector.transfer_read %subview[%c0, %c0, %c0, %c0], %c0_i32, %mask {in_bounds = [false, true, true, false], permutation_map = #map1}
+  %3 = vector.transfer_read %subview[%c0, %c0, %c0, %c0], %c0_i32, %mask {permutation_map = #map1}
           : memref<1x1x1x1xi32>, vector<1x1x1x1xi32>
   return %3 : vector<1x1x1x1xi32>
 }

mlir/test/Dialect/Affine/SuperVectorize/vectorize_1d.mlir

@@ -22,7 +22,7 @@ func.func @vec1d_1(%A : memref<?x?xf32>, %B : memref<?x?x?xf32>) {
 // CHECK-NEXT: %{{.*}} = affine.apply #[[$map_id1]](%[[C0]])
 // CHECK-NEXT: %{{.*}} = affine.apply #[[$map_id1]](%[[C0]])
 // CHECK-NEXT: %{{.*}} = arith.constant 0.0{{.*}}: f32
-// CHECK-NEXT: {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {in_bounds = [true], permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
+// CHECK-NEXT: {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
    affine.for %i0 = 0 to %M { // vectorized due to scalar -> vector
      %a0 = affine.load %A[%c0, %c0] : memref<?x?xf32>
    }
@@ -425,7 +425,7 @@ func.func @vec_rejected_8(%A : memref<?x?xf32>, %B : memref<?x?x?xf32>) {
 // CHECK:     %{{.*}} = affine.apply #[[$map_id1]](%{{.*}})
 // CHECK:     %{{.*}} = affine.apply #[[$map_id1]](%{{.*}})
 // CHECK:     %{{.*}} = arith.constant 0.0{{.*}}: f32
-// CHECK:     {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {in_bounds = [true], permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
+// CHECK:     {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
    affine.for %i17 = 0 to %M { // not vectorized, the 1-D pattern that matched %{{.*}} in DFS post-order prevents vectorizing %{{.*}}
      affine.for %i18 = 0 to %M { // vectorized due to scalar -> vector
        %a18 = affine.load %A[%c0, %c0] : memref<?x?xf32>
@@ -459,7 +459,7 @@ func.func @vec_rejected_9(%A : memref<?x?xf32>, %B : memref<?x?x?xf32>) {
 // CHECK:      %{{.*}} = affine.apply #[[$map_id1]](%{{.*}})
 // CHECK-NEXT: %{{.*}} = affine.apply #[[$map_id1]](%{{.*}})
 // CHECK-NEXT: %{{.*}} = arith.constant 0.0{{.*}}: f32
-// CHECK-NEXT: {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {in_bounds = [true], permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
+// CHECK-NEXT: {{.*}} = vector.transfer_read %{{.*}}[%{{.*}}, %{{.*}}], %{{.*}} {permutation_map = #[[$map_proj_d0d1_0]]} : memref<?x?xf32>, vector<128xf32>
    affine.for %i17 = 0 to %M { // not vectorized, the 1-D pattern that matched %i18 in DFS post-order prevents vectorizing %{{.*}}
      affine.for %i18 = 0 to %M { // vectorized due to scalar -> vector
        %a18 = affine.load %A[%c0, %c0] : memref<?x?xf32>

mlir/test/Dialect/Affine/SuperVectorize/vectorize_2d.mlir

@@ -123,8 +123,8 @@ func.func @vectorize_matmul(%arg0: memref<?x?xf32>, %arg1: memref<?x?xf32>, %arg
   //      VECT:  affine.for %[[I2:.*]] = #[[$map_id1]](%[[C0]]) to #[[$map_id1]](%[[M]]) step 4 {
   // VECT-NEXT:    affine.for %[[I3:.*]] = #[[$map_id1]](%[[C0]]) to #[[$map_id1]](%[[N]]) step 8 {
   // VECT-NEXT:      affine.for %[[I4:.*]] = #[[$map_id1]](%[[C0]]) to #[[$map_id1]](%[[K]]) {
-  //      VECT:        %[[A:.*]] = vector.transfer_read %{{.*}}[%[[I4]], %[[I3]]], %{{.*}} {in_bounds = [true, false], permutation_map = #[[$map_proj_d0d1_zerod1]]} : memref<?x?xf32>, vector<4x8xf32>
-  //      VECT:        %[[B:.*]] = vector.transfer_read %{{.*}}[%[[I2]], %[[I4]]], %{{.*}} {in_bounds = [false, true], permutation_map = #[[$map_proj_d0d1_d0zero]]} : memref<?x?xf32>, vector<4x8xf32>
+  //      VECT:        %[[A:.*]] = vector.transfer_read %{{.*}}[%[[I4]], %[[I3]]], %{{.*}} {permutation_map = #[[$map_proj_d0d1_zerod1]]} : memref<?x?xf32>, vector<4x8xf32>
+  //      VECT:        %[[B:.*]] = vector.transfer_read %{{.*}}[%[[I2]], %[[I4]]], %{{.*}} {permutation_map = #[[$map_proj_d0d1_d0zero]]} : memref<?x?xf32>, vector<4x8xf32>
   // VECT-NEXT:        %[[C:.*]] = arith.mulf %[[B]], %[[A]] : vector<4x8xf32>
   //      VECT:        %[[D:.*]] = vector.transfer_read %{{.*}}[%[[I2]], %[[I3]]], %{{.*}} : memref<?x?xf32>, vector<4x8xf32>
   // VECT-NEXT:        %[[E:.*]] = arith.addf %[[D]], %[[C]] : vector<4x8xf32>

mlir/test/Dialect/Affine/SuperVectorize/vectorize_affine_apply.mlir

@@ -141,7 +141,7 @@ func.func @affine_map_with_expr_2(%arg0: memref<8x12x16xf32>, %arg1: memref<8x24
 // CHECK-NEXT:       %[[S1:.*]] = affine.apply #[[$MAP_ID4]](%[[ARG3]], %[[ARG4]], %[[I0]])
 // CHECK-NEXT:       %[[S2:.*]] = affine.apply #[[$MAP_ID5]](%[[ARG3]], %[[ARG4]], %[[I0]])
 // CHECK-NEXT:       %[[CST:.*]] = arith.constant 0.000000e+00 : f32
-// CHECK-NEXT:       %[[S3:.*]] = vector.transfer_read %[[ARG0]][%[[S0]], %[[S1]], %[[S2]]], %[[CST]] {in_bounds = [true], permutation_map = #[[$MAP_ID6]]} : memref<8x12x16xf32>, vector<8xf32>
+// CHECK-NEXT:       %[[S3:.*]] = vector.transfer_read %[[ARG0]][%[[S0]], %[[S1]], %[[S2]]], %[[CST]] {permutation_map = #[[$MAP_ID6]]} : memref<8x12x16xf32>, vector<8xf32>
 // CHECK-NEXT:       vector.transfer_write %[[S3]], %[[ARG1]][%[[ARG3]], %[[ARG4]], %[[ARG5]]] : vector<8xf32>, memref<8x24x48xf32>
 // CHECK-NEXT:     }
 // CHECK-NEXT:   }

mlir/test/Dialect/Linalg/hoisting.mlir

@@ -200,7 +200,7 @@ func.func @hoist_vector_transfer_pairs_in_affine_loops(%memref0: memref<64x64xi3
   affine.for %arg3 = 0 to 64 {
     affine.for %arg4 = 0 to 64 step 16 {
       affine.for %arg5 = 0 to 64 {
-        %0 = vector.transfer_read %memref0[%arg3, %arg5], %c0_i32 {in_bounds = [true], permutation_map = affine_map<(d0, d1) -> (0)>} : memref<64x64xi32>, vector<16xi32>
+        %0 = vector.transfer_read %memref0[%arg3, %arg5], %c0_i32 {permutation_map = affine_map<(d0, d1) -> (0)>} : memref<64x64xi32>, vector<16xi32>
         %1 = vector.transfer_read %memref1[%arg5, %arg4], %c0_i32 : memref<64x64xi32>, vector<16xi32>
         %2 = vector.transfer_read %memref2[%arg3, %arg4], %c0_i32 : memref<64x64xi32>, vector<16xi32>
         %3 = arith.muli %0, %1 : vector<16xi32>

mlir/test/Dialect/Linalg/vectorization.mlir

@@ -130,7 +130,7 @@ func.func @vectorize_dynamic_1d_broadcast(%arg0: tensor<?xf32>,
 // CHECK-LABEL:   @vectorize_dynamic_1d_broadcast
 // CHECK:           %[[VAL_3:.*]] = arith.constant 0 : index
 // CHECK:           %[[VAL_4:.*]] = tensor.dim %{{.*}}, %[[VAL_3]] : tensor<?xf32>
-// CHECK:           %[[VAL_7:.*]] = vector.transfer_read %{{.*}} {in_bounds = {{.*}}, permutation_map = #{{.*}}} : tensor<?xf32>, vector<4xf32>
+// CHECK:           %[[VAL_7:.*]] = vector.transfer_read %{{.*}} {permutation_map = #{{.*}}} : tensor<?xf32>, vector<4xf32>
 // CHECK:           %[[VAL_9:.*]] = vector.create_mask %[[VAL_4]] : vector<4xi1>
 // CHECK:           %[[VAL_10:.*]] = vector.mask %[[VAL_9]] { vector.transfer_read %{{.*}} {in_bounds = [true]} : tensor<?xf32>, vector<4xf32> } : vector<4xi1> -> vector<4xf32>
 // CHECK:           %[[VAL_12:.*]] = vector.mask %[[VAL_9]] { vector.transfer_read %{{.*}} {in_bounds = [true]} : tensor<?xf32>, vector<4xf32> } : vector<4xi1> -> vector<4xf32>

mlir/test/Dialect/Vector/invalid.mlir

@@ -454,6 +454,15 @@ func.func @test_vector.transfer_read(%arg0: memref<?x?xf32>) {
 
 // -----
 
+func.func @test_vector.transfer_read(%arg0: memref<?x?xf32>) {
+  %c3 = arith.constant 3 : index
+  %cst = arith.constant 3.0 : f32
+  // expected-error@+1 {{requires a projected permutation_map (at most one dim or the zero constant can appear in each result)}}
+  %0 = vector.transfer_read %arg0[%c3, %c3], %cst {permutation_map = affine_map<(d0, d1)->(1)>} : memref<?x?xf32>, vector<128xf32>
+}
+
+// -----
+
 func.func @test_vector.transfer_read(%arg0: memref<?x?x?xf32>) {
   %c3 = arith.constant 3 : index
   %cst = arith.constant 3.0 : f32
@@ -505,16 +514,6 @@ func.func @test_vector.transfer_read(%arg0: memref<?x?xvector<2x3xf32>>) {
 
 // -----
 
-func.func @test_vector.transfer_read(%arg0: memref<?x?xvector<2x3xf32>>) {
-  %c3 = arith.constant 3 : index
-  %f0 = arith.constant 0.0 : f32
-  %vf0 = vector.splat %f0 : vector<2x3xf32>
-  // expected-error@+1 {{requires broadcast dimensions to be in-bounds}}
-  %0 = vector.transfer_read %arg0[%c3, %c3], %vf0 {in_bounds = [false, true], permutation_map = affine_map<(d0, d1)->(0, d1)>} : memref<?x?xvector<2x3xf32>>, vector<1x1x2x3xf32>
-}
-
-// -----
-
 func.func @test_vector.transfer_read(%arg0: memref<?x?xvector<2x3xf32>>) {
   %c3 = arith.constant 3 : index
   %f0 = arith.constant 0.0 : f32
@@ -618,6 +617,15 @@ func.func @test_vector.transfer_write(%arg0: memref<?x?xf32>) {
 
 // -----
 
+func.func @test_vector.transfer_write(%arg0: memref<?x?xf32>) {
+  %c3 = arith.constant 3 : index
+  %cst = arith.constant dense<3.0> : vector<128 x f32>
+  // expected-error@+1 {{requires a projected permutation_map (at most one dim or the zero constant can appear in each result)}}
+  vector.transfer_write %cst, %arg0[%c3, %c3] {permutation_map = affine_map<(d0, d1)->(1)>} : vector<128xf32>, memref<?x?xf32>
+}
+
+// -----
+
 func.func @test_vector.transfer_write(%arg0: memref<?x?x?xf32>) {
   %c3 = arith.constant 3 : index
   %cst = arith.constant dense<3.0> : vector<3 x 7 x f32>

mlir/test/Dialect/Vector/ops.mlir

@@ -70,7 +70,7 @@ func.func @vector_transfer_ops(%arg0: memref<?x?xf32>,
   // CHECK: vector.transfer_read %{{.*}}[%[[C3]], %[[C3]]], %{{.*}}, %{{.*}} : memref<?x?xf32>, vector<5xf32>
   %8 = vector.transfer_read %arg0[%c3, %c3], %f0, %m : memref<?x?xf32>, vector<5xf32>
   // CHECK: vector.transfer_read %{{.*}}[%[[C3]], %[[C3]], %[[C3]]], %{{.*}}, %{{.*}} : memref<?x?x?xf32>, vector<5x4x8xf32>
-  %9 = vector.transfer_read %arg4[%c3, %c3, %c3], %f0, %m2 {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2)->(d1, d0, 0)>} : memref<?x?x?xf32>, vector<5x4x8xf32>
+  %9 = vector.transfer_read %arg4[%c3, %c3, %c3], %f0, %m2 {permutation_map = affine_map<(d0, d1, d2)->(d1, d0, 0)>} : memref<?x?x?xf32>, vector<5x4x8xf32>
 
   // CHECK: vector.transfer_write
   vector.transfer_write %0, %arg0[%c3, %c3] {permutation_map = affine_map<(d0, d1)->(d0)>} : vector<128xf32>, memref<?x?xf32>

mlir/test/Dialect/Vector/vector-transfer-permutation-lowering.mlir

@@ -327,13 +327,11 @@ func.func @masked_permutation_xfer_read_fixed_width(
   %c0 = arith.constant 0 : index
   %3 = vector.mask %mask {
     vector.transfer_read %dest[%c0, %c0], %cst {
-      in_bounds = [false, true, false],
       permutation_map = affine_map<(d0, d1) -> (d1, 0, d0)>
     } : tensor<?x1xf32>, vector<1x4x4xf32>
   } : vector<4x1xi1> -> vector<1x4x4xf32>
 
   "test.some_use"(%3) : (vector<1x4x4xf32>) -> ()
-
   return
 }