[Mlir-commits] [mlir] [mlir][vector] Refactor vector-transfer-to-vector-load-store.mlir (NFC) (PR #105509)

Pablo Antonio Martinez llvmlistbot at llvm.org
Tue Sep 3 01:32:39 PDT 2024


https://github.com/pabloantoniom updated https://github.com/llvm/llvm-project/pull/105509

>From 5d7a0f9cab6ed3394031e488b5166edc5d9e19f5 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Wed, 21 Aug 2024 20:23:07 +0800
Subject: [PATCH 1/8] [mlir][vector] Refactor
 vector-transfer-to-vector-load-store.mlir (NFC)

Overview of changes:

- All memref input arguments are re-named to %mem.
- All vector input arguments are re-named to %vec.
- All index input arguments are re-named to %idx.
- All vector input arguments are re-named to %arg.
- LIT variables are update to be consistent with input arguments.
- Renamed all output arguments as %res.
- Remove unused argument in `transfer_write_broadcast_unit_dim`.
- Unified identation of `FileCheck` commands.
---
 .../vector-transfer-to-vector-load-store.mlir | 207 +++++++++---------
 1 file changed, 103 insertions(+), 104 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 2b7906d4dd40cb..d376c61b611569 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -2,33 +2,33 @@
 
 // CHECK-LABEL: func @vector_transfer_ops_0d_memref(
 //  CHECK-SAME:   %[[MEM:.*]]: memref<f32>
-//  CHECK-SAME:   %[[VV:.*]]: vector<1x1x1xf32>
-func.func @vector_transfer_ops_0d_memref(%M: memref<f32>, %v: vector<1x1x1xf32>) {
+//  CHECK-SAME:   %[[VEC:.*]]: vector<1x1x1xf32>
+func.func @vector_transfer_ops_0d_memref(%mem: memref<f32>, %vec: vector<1x1x1xf32>) {
     %f0 = arith.constant 0.0 : f32
 
 //  CHECK-NEXT:   %[[s:.*]] = memref.load %[[MEM]][] : memref<f32>
 //  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[s]] : f32 to vector<f32>
-    %0 = vector.transfer_read %M[], %f0 : memref<f32>, vector<f32>
+    %0 = vector.transfer_read %mem[], %f0 : memref<f32>, vector<f32>
 
 //  CHECK-NEXT:   %[[ss:.*]] = vector.extractelement %[[V]][] : vector<f32>
 //  CHECK-NEXT:   memref.store %[[ss]], %[[MEM]][] : memref<f32>
-    vector.transfer_write %0, %M[] : vector<f32>, memref<f32>
+    vector.transfer_write %0, %mem[] : vector<f32>, memref<f32>
 
 //  CHECK-NEXT:   %[[VV:.*]] = vector.extract %arg1[0, 0, 0] : f32 from vector<1x1x1xf32>
 //  CHECK-NEXT:   memref.store %[[VV]], %[[MEM]][] : memref<f32>
-    vector.store %v, %M[] : memref<f32>, vector<1x1x1xf32>
+    vector.store %vec, %mem[] : memref<f32>, vector<1x1x1xf32>
 
     return
 }
 
 // CHECK-LABEL: func @vector_transfer_ops_0d_tensor(
-//  CHECK-SAME:   %[[SOURCE:.*]]: tensor<f32>
-func.func @vector_transfer_ops_0d_tensor(%M: tensor<f32>) -> vector<1xf32> {
+//  CHECK-SAME:   %[[ARG0:.*]]: tensor<f32>
+func.func @vector_transfer_ops_0d_tensor(%arg0: tensor<f32>) -> vector<1xf32> {
     %f0 = arith.constant 0.0 : f32
 
-//  CHECK-NEXT:   %[[S:.*]] = tensor.extract %[[SOURCE]][] : tensor<f32>
+//  CHECK-NEXT:   %[[S:.*]] = tensor.extract %[[ARG0]][] : tensor<f32>
 //  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[S]] : f32 to vector<1xf32>
-    %0 = vector.transfer_read %M[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
+    %0 = vector.transfer_read %arg0[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
       tensor<f32>, vector<1xf32>
 
 //  CHECK-NEXT:   return %[[V]]
@@ -37,200 +37,200 @@ func.func @vector_transfer_ops_0d_tensor(%M: tensor<f32>) -> vector<1xf32> {
 
 // transfer_read/write are lowered to vector.load/store
 // CHECK-LABEL:   func @transfer_to_load(
-// CHECK-SAME:                                %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                                %[[IDX:.*]]: index) -> vector<4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4xf32> {
 // CHECK-NEXT:      %[[RES:.*]] = vector.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>, vector<4xf32>
 // CHECK-NEXT:      vector.store  %[[RES:.*]], %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>, vector<4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_to_load(%mem : memref<8x8xf32>, %i : index) -> vector<4xf32> {
+func.func @transfer_to_load(%mem : memref<8x8xf32>, %idx : index) -> vector<4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true]} : memref<8x8xf32>, vector<4xf32>
-  vector.transfer_write %res, %mem[%i, %i] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true]} : memref<8x8xf32>, vector<4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32>
   return %res : vector<4xf32>
 }
 
 // Masked transfer_read/write inside are NOT lowered to vector.load/store
 // CHECK-LABEL:   func @masked_transfer_to_load(
-//  CHECK-SAME:                                %[[MEM:.*]]: memref<8x8xf32>,
-//  CHECK-SAME:                                %[[IDX:.*]]: index,
-//  CHECK-SAME:                                %[[MASK:.*]]: vector<4xi1>) -> memref<8x8xf32>
+//  CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+//  CHECK-SAME:      %[[IDX:.*]]: index,
+//  CHECK-SAME:      %[[MASK:.*]]: vector<4xi1>) -> memref<8x8xf32>
 //   CHECK-NOT:      vector.load 
 //   CHECK-NOT:      vector.store
 //       CHECK:      %[[READ:.*]] = vector.mask %[[MASK]] { vector.transfer_read %arg0[%[[IDX]], %[[IDX]]]{{.*}} : memref<8x8xf32>, vector<4xf32> } : vector<4xi1> -> vector<4xf32>
 //       CHECK:      vector.mask %[[MASK]] { vector.transfer_write %[[READ]], %[[MEM]][%[[IDX]], %[[IDX]]]{{.*}} : vector<4xf32>, memref<8x8xf32> } : vector<4xi1>
 
-func.func @masked_transfer_to_load(%mem : memref<8x8xf32>, %i : index, %mask : vector<4xi1>) -> memref<8x8xf32> {
+func.func @masked_transfer_to_load(%mem : memref<8x8xf32>, %idx : index, %mask : vector<4xi1>) -> memref<8x8xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %read = vector.mask %mask {vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true]} : memref<8x8xf32>, vector<4xf32>} : vector<4xi1> -> vector<4xf32>
-  vector.mask %mask {vector.transfer_write %read, %mem[%i, %i] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32> } : vector<4xi1> 
+  %read = vector.mask %mask {vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true]} : memref<8x8xf32>, vector<4xf32>} : vector<4xi1> -> vector<4xf32>
+  vector.mask %mask {vector.transfer_write %read, %mem[%idx, %idx] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32> } : vector<4xi1> 
   return %mem : memref<8x8xf32>
 }
 
 // n-D results are also supported.
 // CHECK-LABEL:   func @transfer_2D(
-// CHECK-SAME:                           %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                           %[[IDX:.*]]: index) -> vector<2x4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<2x4xf32> {
 // CHECK-NEXT:      %[[RES:.*]] = vector.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>, vector<2x4xf32>
 // CHECK-NEXT:      vector.store %[[RES:.*]], %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>, vector<2x4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<2x4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_2D(%mem : memref<8x8xf32>, %i : index) -> vector<2x4xf32> {
+func.func @transfer_2D(%mem : memref<8x8xf32>, %idx : index) -> vector<2x4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true, true]} : memref<8x8xf32>, vector<2x4xf32>
-  vector.transfer_write %res, %mem[%i, %i] {in_bounds = [true, true]} : vector<2x4xf32>, memref<8x8xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true, true]} : memref<8x8xf32>, vector<2x4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] {in_bounds = [true, true]} : vector<2x4xf32>, memref<8x8xf32>
   return %res : vector<2x4xf32>
 }
 
 // Vector element types are supported when the result has the same type.
 // CHECK-LABEL:   func @transfer_vector_element(
-// CHECK-SAME:                           %[[MEM:.*]]: memref<8x8xvector<2x4xf32>>,
-// CHECK-SAME:                           %[[IDX:.*]]: index) -> vector<2x4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xvector<2x4xf32>>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<2x4xf32> {
 // CHECK-NEXT:      %[[RES:.*]] = vector.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xvector<2x4xf32>>, vector<2x4xf32>
 // CHECK-NEXT:      vector.store %[[RES:.*]], %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xvector<2x4xf32>>, vector<2x4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<2x4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_vector_element(%mem : memref<8x8xvector<2x4xf32>>, %i : index) -> vector<2x4xf32> {
+func.func @transfer_vector_element(%mem : memref<8x8xvector<2x4xf32>>, %idx : index) -> vector<2x4xf32> {
   %cf0 = arith.constant dense<0.0> : vector<2x4xf32>
-  %res = vector.transfer_read %mem[%i, %i], %cf0 : memref<8x8xvector<2x4xf32>>, vector<2x4xf32>
-  vector.transfer_write %res, %mem[%i, %i] : vector<2x4xf32>, memref<8x8xvector<2x4xf32>>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 : memref<8x8xvector<2x4xf32>>, vector<2x4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] : vector<2x4xf32>, memref<8x8xvector<2x4xf32>>
   return %res : vector<2x4xf32>
 }
 
 // TODO: Vector element types are not supported yet when the result has a
 // different type.
 // CHECK-LABEL:   func @transfer_vector_element_different_types(
-// CHECK-SAME:                           %[[MEM:.*]]: memref<8x8xvector<2x4xf32>>,
-// CHECK-SAME:                           %[[IDX:.*]]: index) -> vector<1x2x4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xvector<2x4xf32>>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<1x2x4xf32> {
 // CHECK-NEXT:      %[[CF0:.*]] = arith.constant dense<0.000000e+00> : vector<2x4xf32>
 // CHECK-NEXT:      %[[RES:.*]] = vector.transfer_read %[[MEM]][%[[IDX]], %[[IDX]]], %[[CF0]] {in_bounds = [true]} : memref<8x8xvector<2x4xf32>>, vector<1x2x4xf32>
 // CHECK-NEXT:      vector.transfer_write %[[RES:.*]], %[[MEM]][%[[IDX]], %[[IDX]]] {in_bounds = [true]} : vector<1x2x4xf32>, memref<8x8xvector<2x4xf32>>
 // CHECK-NEXT:      return %[[RES]] : vector<1x2x4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_vector_element_different_types(%mem : memref<8x8xvector<2x4xf32>>, %i : index) -> vector<1x2x4xf32> {
+func.func @transfer_vector_element_different_types(%mem : memref<8x8xvector<2x4xf32>>, %idx : index) -> vector<1x2x4xf32> {
   %cf0 = arith.constant dense<0.0> : vector<2x4xf32>
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true]} : memref<8x8xvector<2x4xf32>>, vector<1x2x4xf32>
-  vector.transfer_write %res, %mem[%i, %i] {in_bounds = [true]} : vector<1x2x4xf32>, memref<8x8xvector<2x4xf32>>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true]} : memref<8x8xvector<2x4xf32>>, vector<1x2x4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] {in_bounds = [true]} : vector<1x2x4xf32>, memref<8x8xvector<2x4xf32>>
   return %res : vector<1x2x4xf32>
 }
 
 // TODO: transfer_read/write cannot be lowered because there is a dimension
 // that is not guaranteed to be in-bounds.
 // CHECK-LABEL:   func @transfer_2D_not_inbounds(
-// CHECK-SAME:                                  %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                                  %[[IDX:.*]]: index) -> vector<2x4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<2x4xf32> {
 // CHECK-NEXT:      %[[CF0:.*]] = arith.constant 0.000000e+00 : f32
 // CHECK-NEXT:      %[[RES:.*]] = vector.transfer_read %[[MEM]][%[[IDX]], %[[IDX]]], %[[CF0]] {in_bounds = [true, false]} : memref<8x8xf32>, vector<2x4xf32>
 // CHECK-NEXT:      vector.transfer_write %[[RES]], %[[MEM]][%[[IDX]], %[[IDX]]] {in_bounds = [false, true]} : vector<2x4xf32>, memref<8x8xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<2x4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_2D_not_inbounds(%mem : memref<8x8xf32>, %i : index) -> vector<2x4xf32> {
+func.func @transfer_2D_not_inbounds(%mem : memref<8x8xf32>, %idx : index) -> vector<2x4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true, false]} : memref<8x8xf32>, vector<2x4xf32>
-  vector.transfer_write %res, %mem[%i, %i] {in_bounds = [false, true]} : vector<2x4xf32>, memref<8x8xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true, false]} : memref<8x8xf32>, vector<2x4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] {in_bounds = [false, true]} : vector<2x4xf32>, memref<8x8xf32>
   return %res : vector<2x4xf32>
 }
 
 // TODO: transfer_read/write cannot be lowered because they are not guaranteed
 // to be in-bounds.
 // CHECK-LABEL:   func @transfer_not_inbounds(
-// CHECK-SAME:                               %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                               %[[IDX:.*]]: index) -> vector<4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4xf32> {
 // CHECK-NEXT:      %[[CF0:.*]] = arith.constant 0.000000e+00 : f32
 // CHECK-NEXT:      %[[RES:.*]] = vector.transfer_read %[[MEM]][%[[IDX]], %[[IDX]]], %[[CF0]] : memref<8x8xf32>, vector<4xf32>
 // CHECK-NEXT:      vector.transfer_write %[[RES]], %[[MEM]][%[[IDX]], %[[IDX]]] : vector<4xf32>, memref<8x8xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_not_inbounds(%mem : memref<8x8xf32>, %i : index) -> vector<4xf32> {
+func.func @transfer_not_inbounds(%mem : memref<8x8xf32>, %idx : index) -> vector<4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 : memref<8x8xf32>, vector<4xf32>
-  vector.transfer_write %res, %mem[%i, %i] : vector<4xf32>, memref<8x8xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 : memref<8x8xf32>, vector<4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] : vector<4xf32>, memref<8x8xf32>
   return %res : vector<4xf32>
 }
 
 // CHECK-LABEL:   func @transfer_nondefault_layout(
-// CHECK-SAME:                                          %[[MEM:.*]]: memref<8x8xf32, #{{.*}}>,
-// CHECK-SAME:                                          %[[IDX:.*]]: index) -> vector<4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32, #{{.*}}>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4xf32> {
 // CHECK-NEXT:      %[[RES:.*]] = vector.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32, #{{.*}}>, vector<4xf32>
 // CHECK-NEXT:      vector.store %[[RES]], %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32, #{{.*}}>,  vector<4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4xf32>
 // CHECK-NEXT:    }
 
 #layout = affine_map<(d0, d1) -> (d0*16 + d1)>
-func.func @transfer_nondefault_layout(%mem : memref<8x8xf32, #layout>, %i : index) -> vector<4xf32> {
+func.func @transfer_nondefault_layout(%mem : memref<8x8xf32, #layout>, %idx : index) -> vector<4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true]} : memref<8x8xf32, #layout>, vector<4xf32>
-  vector.transfer_write %res, %mem[%i, %i] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32, #layout>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true]} : memref<8x8xf32, #layout>, vector<4xf32>
+  vector.transfer_write %res, %mem[%idx, %idx] {in_bounds = [true]} : vector<4xf32>, memref<8x8xf32, #layout>
   return %res : vector<4xf32>
 }
 
 // TODO: transfer_read/write cannot be lowered to vector.load/store yet when the
 // permutation map is not the minor identity map (up to broadcasting).
 // CHECK-LABEL:   func @transfer_perm_map(
-// CHECK-SAME:                                 %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                                 %[[IDX:.*]]: index) -> vector<4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4xf32> {
 // CHECK-NEXT:      %[[CF0:.*]] = arith.constant 0.000000e+00 : f32
 // CHECK-NEXT:      %[[RES:.*]] = vector.transfer_read %[[MEM]][%[[IDX]], %[[IDX]]], %[[CF0]] {in_bounds = [true], permutation_map = #{{.*}}} : memref<8x8xf32>, vector<4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4xf32>
 // CHECK-NEXT:    }
 
-func.func @transfer_perm_map(%mem : memref<8x8xf32>, %i : index) -> vector<4xf32> {
+func.func @transfer_perm_map(%mem : memref<8x8xf32>, %idx : index) -> vector<4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true], permutation_map = affine_map<(d0, d1) -> (d0)>} : memref<8x8xf32>, vector<4xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true], permutation_map = affine_map<(d0, d1) -> (d0)>} : memref<8x8xf32>, vector<4xf32>
   return %res : vector<4xf32>
 }
 
 // Lowering of transfer_read with broadcasting is supported (note that a `load`
 // is generated instead of a `vector.load`).
 // CHECK-LABEL:   func @transfer_broadcasting(
-// CHECK-SAME:                                %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                                %[[IDX:.*]]: index) -> vector<4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4xf32> {
 // CHECK-NEXT:      %[[LOAD:.*]] = memref.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>
 // CHECK-NEXT:      %[[RES:.*]] = vector.broadcast %[[LOAD]] : f32 to vector<4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4xf32>
 // CHECK-NEXT:    }
 
 #broadcast_1d = affine_map<(d0, d1) -> (0)>
-func.func @transfer_broadcasting(%mem : memref<8x8xf32>, %i : index) -> vector<4xf32> {
+func.func @transfer_broadcasting(%mem : memref<8x8xf32>, %idx : index) -> vector<4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0
     {in_bounds = [true], permutation_map = #broadcast_1d}
       : memref<8x8xf32>, vector<4xf32>
   return %res : vector<4xf32>
 }
 
 // CHECK-LABEL:   func @transfer_scalar(
-// CHECK-SAME:                          %[[MEM:.*]]: memref<?x?xf32>,
-// CHECK-SAME:                          %[[IDX:.*]]: index) -> vector<1xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<?x?xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<1xf32> {
 // CHECK-NEXT:      %[[LOAD:.*]] = memref.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<?x?xf32>
 // CHECK-NEXT:      %[[RES:.*]] = vector.broadcast %[[LOAD]] : f32 to vector<1xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<1xf32>
 // CHECK-NEXT:    }
-func.func @transfer_scalar(%mem : memref<?x?xf32>, %i : index) -> vector<1xf32> {
+func.func @transfer_scalar(%mem : memref<?x?xf32>, %idx : index) -> vector<1xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0 {in_bounds = [true]} : memref<?x?xf32>, vector<1xf32>
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0 {in_bounds = [true]} : memref<?x?xf32>, vector<1xf32>
   return %res : vector<1xf32>
 }
 
 // An example with two broadcasted dimensions.
 // CHECK-LABEL:   func @transfer_broadcasting_2D(
-// CHECK-SAME:                                %[[MEM:.*]]: memref<8x8xf32>,
-// CHECK-SAME:                                %[[IDX:.*]]: index) -> vector<4x4xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<4x4xf32> {
 // CHECK-NEXT:      %[[LOAD:.*]] = memref.load %[[MEM]][%[[IDX]], %[[IDX]]] : memref<8x8xf32>
 // CHECK-NEXT:      %[[RES:.*]] = vector.broadcast %[[LOAD]] : f32 to vector<4x4xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<4x4xf32>
 // CHECK-NEXT:    }
 
 #broadcast_2d = affine_map<(d0, d1) -> (0, 0)>
-func.func @transfer_broadcasting_2D(%mem : memref<8x8xf32>, %i : index) -> vector<4x4xf32> {
+func.func @transfer_broadcasting_2D(%mem : memref<8x8xf32>, %idx : index) -> vector<4x4xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i], %cf0
+  %res = vector.transfer_read %mem[%idx, %idx], %cf0
     {in_bounds = [true, true], permutation_map = #broadcast_2d}
       : memref<8x8xf32>, vector<4x4xf32>
   return %res : vector<4x4xf32>
@@ -238,17 +238,17 @@ func.func @transfer_broadcasting_2D(%mem : memref<8x8xf32>, %i : index) -> vecto
 
 // More complex broadcasting case (here a `vector.load` is generated).
 // CHECK-LABEL:   func @transfer_broadcasting_complex(
-// CHECK-SAME:                                %[[MEM:.*]]: memref<10x20x30x8x8xf32>,
-// CHECK-SAME:                                %[[IDX:.*]]: index) -> vector<3x2x4x5xf32> {
+// CHECK-SAME:      %[[MEM:.*]]: memref<10x20x30x8x8xf32>,
+// CHECK-SAME:      %[[IDX:.*]]: index) -> vector<3x2x4x5xf32> {
 // CHECK-NEXT:      %[[LOAD:.*]] = vector.load %[[MEM]][%[[IDX]], %[[IDX]], %[[IDX]], %[[IDX]], %[[IDX]]] : memref<10x20x30x8x8xf32>, vector<3x1x1x5xf32>
 // CHECK-NEXT:      %[[RES:.*]] = vector.broadcast %[[LOAD]] : vector<3x1x1x5xf32> to vector<3x2x4x5xf32>
 // CHECK-NEXT:      return %[[RES]] : vector<3x2x4x5xf32>
 // CHECK-NEXT:    }
 
 #broadcast_2d_in_4d = affine_map<(d0, d1, d2, d3, d4) -> (d1, 0, 0, d4)>
-func.func @transfer_broadcasting_complex(%mem : memref<10x20x30x8x8xf32>, %i : index) -> vector<3x2x4x5xf32> {
+func.func @transfer_broadcasting_complex(%mem : memref<10x20x30x8x8xf32>, %idx : index) -> vector<3x2x4x5xf32> {
   %cf0 = arith.constant 0.0 : f32
-  %res = vector.transfer_read %mem[%i, %i, %i, %i, %i], %cf0
+  %res = vector.transfer_read %mem[%idx, %idx, %idx, %idx, %idx], %cf0
     {in_bounds = [true, true, true, true], permutation_map = #broadcast_2d_in_4d}
       : memref<10x20x30x8x8xf32>, vector<3x2x4x5xf32>
   return %res : vector<3x2x4x5xf32>
@@ -280,7 +280,7 @@ module attributes {transform.with_named_sequence} {
 // CHECK-DAG: #[[$MAP1:.*]] = affine_map<(d0, d1, d2, d3) -> (d1, 0, d3)>
 
 // CHECK-LABEL: func @transfer_read_permutations
-func.func @transfer_read_permutations(%arg0 : memref<?x?xf32>, %arg1 : memref<?x?x?x?xf32>, %m: i1)
+func.func @transfer_read_permutations(%mem_0 : memref<?x?xf32>, %mem_1 : memref<?x?x?x?xf32>, %m: i1)
     -> (vector<7x14x8x16xf32>, vector<7x14x8x16xf32>, vector<7x14x8x16xf32>,
        vector<7x14x8x16xf32>, vector<7x14x8x16xf32>, vector<7x14x8x16xf32>, vector<8xf32>) {
 // CHECK-DAG: %[[CF0:.*]] = arith.constant 0.000000e+00 : f32
@@ -290,38 +290,38 @@ func.func @transfer_read_permutations(%arg0 : memref<?x?xf32>, %arg1 : memref<?x
 
 // CHECK: %[[MASK0:.*]] = vector.splat %{{.*}} : vector<14x7xi1>
   %mask0 = vector.splat %m : vector<14x7xi1>
-  %0 = vector.transfer_read %arg1[%c0, %c0, %c0, %c0], %cst, %mask0 {in_bounds = [true, false, true, true], permutation_map = #map0} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
+  %0 = vector.transfer_read %mem_1[%c0, %c0, %c0, %c0], %cst, %mask0 {in_bounds = [true, false, true, true], permutation_map = #map0} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read {{.*}} %[[MASK0]] {in_bounds = [false, true, true, true], permutation_map = #[[$MAP0]]} : memref<?x?x?x?xf32>, vector<14x7x8x16xf32>
 // CHECK: vector.transpose %{{.*}}, [1, 0, 2, 3] : vector<14x7x8x16xf32> to vector<7x14x8x16xf32>
 
 // CHECK: %[[MASK1:.*]] = vector.splat %{{.*}} : vector<16x14xi1>
   %mask1 = vector.splat %m : vector<16x14xi1>
-  %1 = vector.transfer_read %arg1[%c0, %c0, %c0, %c0], %cst, %mask1 {in_bounds = [true, false, true, false], permutation_map = #map1} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
+  %1 = vector.transfer_read %mem_1[%c0, %c0, %c0, %c0], %cst, %mask1 {in_bounds = [true, false, true, false], permutation_map = #map1} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read {{.*}} %[[MASK1]] {in_bounds = [false, false, true, true], permutation_map = #[[$MAP0]]} : memref<?x?x?x?xf32>, vector<16x14x7x8xf32>
 // CHECK: vector.transpose %{{.*}}, [2, 1, 3, 0] : vector<16x14x7x8xf32> to vector<7x14x8x16xf32>
 
 // CHECK: %[[MASK3:.*]] = vector.splat %{{.*}} : vector<14x7xi1>
   %mask2 = vector.splat %m : vector<14x7xi1>
-  %2 = vector.transfer_read %arg1[%c0, %c0, %c0, %c0], %cst, %mask2 {in_bounds = [true, false, true, true], permutation_map = #map2} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
+  %2 = vector.transfer_read %mem_1[%c0, %c0, %c0, %c0], %cst, %mask2 {in_bounds = [true, false, true, true], permutation_map = #map2} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read {{.*}} %[[MASK3]] {in_bounds = [false, true, true], permutation_map = #[[$MAP1]]} : memref<?x?x?x?xf32>, vector<14x16x7xf32>
 // CHECK: vector.broadcast %{{.*}} : vector<14x16x7xf32> to vector<8x14x16x7xf32>
 // CHECK: vector.transpose %{{.*}}, [3, 1, 0, 2] : vector<8x14x16x7xf32> to vector<7x14x8x16xf32>
 
-  %3 = vector.transfer_read %arg0[%c0, %c0], %cst {in_bounds = [false, false, true, true], permutation_map = #map3} : memref<?x?xf32>, vector<7x14x8x16xf32>
+  %3 = vector.transfer_read %mem_0[%c0, %c0], %cst {in_bounds = [false, false, true, true], permutation_map = #map3} : memref<?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read %{{.*}}[%[[C0]], %[[C0]]], %[[CF0]] : memref<?x?xf32>, vector<14x7xf32>
 // CHECK: vector.broadcast %{{.*}} : vector<14x7xf32> to vector<8x16x14x7xf32>
 // CHECK: vector.transpose %{{.*}}, [3, 2, 0, 1] : vector<8x16x14x7xf32> to vector<7x14x8x16xf32>
 
-  %4 = vector.transfer_read %arg0[%c0, %c0], %cst {in_bounds = [true, false, true, false], permutation_map = #map4} : memref<?x?xf32>, vector<7x14x8x16xf32>
+  %4 = vector.transfer_read %mem_0[%c0, %c0], %cst {in_bounds = [true, false, true, false], permutation_map = #map4} : memref<?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read %{{.*}}[%[[C0]], %[[C0]]], %[[CF0]] : memref<?x?xf32>, vector<16x14xf32>
 // CHECK: vector.broadcast %{{.*}} : vector<16x14xf32> to vector<7x8x16x14xf32>
 // CHECK: vector.transpose %{{.*}}, [0, 3, 1, 2] : vector<7x8x16x14xf32> to vector<7x14x8x16xf32>
 
-  %5 = vector.transfer_read %arg1[%c0, %c0, %c0, %c0], %cst {permutation_map = #map5} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
+  %5 = vector.transfer_read %mem_1[%c0, %c0, %c0, %c0], %cst {permutation_map = #map5} : memref<?x?x?x?xf32>, vector<7x14x8x16xf32>
 // CHECK: vector.transfer_read %{{.*}}[%[[C0]], %[[C0]], %[[C0]], %[[C0]]], %[[CF0]] : memref<?x?x?x?xf32>, vector<16x14x7x8xf32>
 // CHECK: vector.transpose %{{.*}}, [2, 1, 3, 0] : vector<16x14x7x8xf32> to vector<7x14x8x16xf32>
 
-  %6 = vector.transfer_read %arg0[%c0, %c0], %cst {in_bounds = [true], permutation_map = #map6} : memref<?x?xf32>, vector<8xf32>
+  %6 = vector.transfer_read %mem_0[%c0, %c0], %cst {in_bounds = [true], permutation_map = #map6} : memref<?x?xf32>, vector<8xf32>
 // CHECK: memref.load %{{.*}}[%[[C0]], %[[C0]]] : memref<?x?xf32>
 // CHECK: vector.broadcast %{{.*}} : f32 to vector<8xf32>
 
@@ -331,53 +331,52 @@ func.func @transfer_read_permutations(%arg0 : memref<?x?xf32>, %arg1 : memref<?x
 }
 
 // CHECK-LABEL: func @transfer_write_permutations
-// CHECK-SAME:      %[[ARG0:.*]]: memref<?x?x?x?xf32>
-// CHECK-SAME:      %[[ARG1:.*]]: tensor<?x?x?x?xf32>
-// CHECK-SAME:      %[[ARG2:.*]]: vector<7x14x8x16xf32>
-// CHECK-SAME:      %[[ARG3:.*]]: vector<8x16xf32>
-// CHECK-SAME:      %[[M:.*]]: i1
+// CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
+// CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[VEC0:.*]]: vector<7x14x8x16xf32>
+// CHECK-SAME:    %[[VEC1:.*]]: vector<8x16xf32>
+// CHECK-SAME:    %[[M:.*]]: i1
 func.func @transfer_write_permutations(
-    %arg0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
-    %v1 : vector<7x14x8x16xf32>, %v2 : vector<8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
+    %mem_0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
+    %vec_0 : vector<7x14x8x16xf32>, %vec_1 : vector<8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
   // CHECK: %[[MASK:.*]] = vector.splat %[[M]] : vector<16x14x7x8xi1>
   %mask0 = vector.splat %m : vector<16x14x7x8xi1>
-  %0 = vector.transfer_write %v1, %arg1[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec_0, %arg1[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.transpose %{{.*}} [3, 1, 0, 2] : vector<7x14x8x16xf32> to vector<16x14x7x8xf32>
   // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[ARG1]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
 
-  vector.transfer_write %v2, %arg0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
+  vector.transfer_write %vec_1, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %{{.*}} [1, 0] : vector<8x16xf32> to vector<16x8xf32>
-  // CHECK: vector.transfer_write %[[NEW_VEC1]], %[[ARG0]][%c0, %c0, %c0, %c0] : vector<16x8xf32>, memref<?x?x?x?xf32>
+  // CHECK: vector.transfer_write %[[NEW_VEC1]], %[[MEM0]][%c0, %c0, %c0, %c0] : vector<16x8xf32>, memref<?x?x?x?xf32>
 
-  return %0 : tensor<?x?x?x?xf32>
+  return %res : tensor<?x?x?x?xf32>
 }
 
 // CHECK-LABEL: func @transfer_write_broadcast_unit_dim
-// CHECK-SAME:      %[[ARG0:.*]]: memref<?x?x?x?xf32>
-// CHECK-SAME:      %[[ARG1:.*]]: tensor<?x?x?x?xf32>
-// CHECK-SAME:      %[[ARG2:.*]]: vector<14x8x16xf32>
-// CHECK-SAME:      %[[ARG3:.*]]: vector<8x16xf32>
-// CHECK-SAME:      %[[M:.*]]: i1
+// CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
+// CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[VEC0:.*]]: vector<14x8x16xf32>
+// CHECK-SAME:    %[[VEC1:.*]]: vector<8x16xf32>
 func.func @transfer_write_broadcast_unit_dim(
-    %arg0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
-    %v1 : vector<14x8x16xf32>, %v2 : vector<8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
+    %mem_0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
+    %vec_0 : vector<14x8x16xf32>, %vec_1 : vector<8x16xf32>) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
-  %0 = vector.transfer_write %v1, %arg1[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec_0, %arg1[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.broadcast %{{.*}} : vector<14x8x16xf32> to vector<1x14x8x16xf32>
   // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %[[NEW_VEC0]], [1, 2, 0, 3] : vector<1x14x8x16xf32> to vector<14x8x1x16xf32>
   // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[ARG1]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
 
-  vector.transfer_write %v2, %arg0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d1, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
+  vector.transfer_write %vec_1, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d1, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC2:.*]] = vector.broadcast %{{.*}} : vector<8x16xf32> to vector<1x8x16xf32>
   // CHECK: %[[NEW_VEC3:.*]] = vector.transpose %[[NEW_VEC2]], [1, 2, 0] : vector<1x8x16xf32> to vector<8x16x1xf32>
-  // CHECK: vector.transfer_write %[[NEW_VEC3]], %[[ARG0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true]} : vector<8x16x1xf32>, memref<?x?x?x?xf32>
+  // CHECK: vector.transfer_write %[[NEW_VEC3]], %[[MEM0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true]} : vector<8x16x1xf32>, memref<?x?x?x?xf32>
 
-  return %0 : tensor<?x?x?x?xf32>
+  return %res : tensor<?x?x?x?xf32>
 }
 
 module attributes {transform.with_named_sequence} {
@@ -397,10 +396,10 @@ module attributes {transform.with_named_sequence} {
 /// vectors aren't generated for rank > 1 transfer ops.
 
 // CHECK-LABEL: @transfer_2D_masked
-// CHECK-NOT: vector.maskedload
-// CHECK-NOT: vector.maskedstore
-// CHECK: vector.transfer_read
-// CHECK: vector.transfer_write
+// CHECK-NOT:     vector.maskedload
+// CHECK-NOT:     vector.maskedstore
+// CHECK:         vector.transfer_read
+// CHECK:         vector.transfer_write
 func.func @transfer_2D_masked(%mem : memref<?x?xf32>, %mask : vector<2x4xi1>) -> vector<2x4xf32> {
   %c0 = arith.constant 0 : index
   %pad = arith.constant 0.0 : f32

>From b2f1de46433da4abda770679cea25cb8eb700c42 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Thu, 22 Aug 2024 16:01:12 +0800
Subject: [PATCH 2/8] Use uppercase for var names (suggested by nujaa)

---
 .../Vector/vector-transfer-to-vector-load-store.mlir      | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index d376c61b611569..5796d58e30e801 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -6,12 +6,12 @@
 func.func @vector_transfer_ops_0d_memref(%mem: memref<f32>, %vec: vector<1x1x1xf32>) {
     %f0 = arith.constant 0.0 : f32
 
-//  CHECK-NEXT:   %[[s:.*]] = memref.load %[[MEM]][] : memref<f32>
-//  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[s]] : f32 to vector<f32>
+//  CHECK-NEXT:   %[[S:.*]] = memref.load %[[MEM]][] : memref<f32>
+//  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[S]] : f32 to vector<f32>
     %0 = vector.transfer_read %mem[], %f0 : memref<f32>, vector<f32>
 
-//  CHECK-NEXT:   %[[ss:.*]] = vector.extractelement %[[V]][] : vector<f32>
-//  CHECK-NEXT:   memref.store %[[ss]], %[[MEM]][] : memref<f32>
+//  CHECK-NEXT:   %[[SS:.*]] = vector.extractelement %[[V]][] : vector<f32>
+//  CHECK-NEXT:   memref.store %[[SS]], %[[MEM]][] : memref<f32>
     vector.transfer_write %0, %mem[] : vector<f32>, memref<f32>
 
 //  CHECK-NEXT:   %[[VV:.*]] = vector.extract %arg1[0, 0, 0] : f32 from vector<1x1x1xf32>

>From a9d1f1d973b3ce337418b5f7815a459a4dea125c Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Thu, 22 Aug 2024 16:02:14 +0800
Subject: [PATCH 3/8] Renamed forgotten %0 to %res

---
 .../Dialect/Vector/vector-transfer-to-vector-load-store.mlir  | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 5796d58e30e801..355fb61d4f5742 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -28,11 +28,11 @@ func.func @vector_transfer_ops_0d_tensor(%arg0: tensor<f32>) -> vector<1xf32> {
 
 //  CHECK-NEXT:   %[[S:.*]] = tensor.extract %[[ARG0]][] : tensor<f32>
 //  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[S]] : f32 to vector<1xf32>
-    %0 = vector.transfer_read %arg0[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
+    %res = vector.transfer_read %arg0[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
       tensor<f32>, vector<1xf32>
 
 //  CHECK-NEXT:   return %[[V]]
-    return %0: vector<1xf32>
+    return %res: vector<1xf32>
 }
 
 // transfer_read/write are lowered to vector.load/store

>From 35a3d23eea781525f4a38cf3c6d93a1a40d9a444 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Thu, 22 Aug 2024 16:03:02 +0800
Subject: [PATCH 4/8] Split transfer_write_permutations into
 transfer_write_permutations_tensor and transfer_write_permutations_memref
 (suggested by banach-space)

---
 .../vector-transfer-to-vector-load-store.mlir | 33 ++++++++++++-------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 355fb61d4f5742..dc11b48ef0f9ba 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -330,31 +330,40 @@ func.func @transfer_read_permutations(%mem_0 : memref<?x?xf32>, %mem_1 : memref<
          vector<7x14x8x16xf32>, vector<8xf32>
 }
 
-// CHECK-LABEL: func @transfer_write_permutations
-// CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
+// CHECK-LABEL: func @transfer_write_permutations_tensor
 // CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>
-// CHECK-SAME:    %[[VEC0:.*]]: vector<7x14x8x16xf32>
-// CHECK-SAME:    %[[VEC1:.*]]: vector<8x16xf32>
+// CHECK-SAME:    %[[VEC:.*]]: vector<7x14x8x16xf32>
 // CHECK-SAME:    %[[M:.*]]: i1
-func.func @transfer_write_permutations(
-    %mem_0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
-    %vec_0 : vector<7x14x8x16xf32>, %vec_1 : vector<8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
+func.func @transfer_write_permutations_tensor(
+    %arg1 : tensor<?x?x?x?xf32>,
+    %vec : vector<7x14x8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
   // CHECK: %[[MASK:.*]] = vector.splat %[[M]] : vector<16x14x7x8xi1>
   %mask0 = vector.splat %m : vector<16x14x7x8xi1>
-  %res = vector.transfer_write %vec_0, %arg1[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec, %arg1[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.transpose %{{.*}} [3, 1, 0, 2] : vector<7x14x8x16xf32> to vector<16x14x7x8xf32>
   // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[ARG1]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
 
-  vector.transfer_write %vec_1, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
-  // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %{{.*}} [1, 0] : vector<8x16xf32> to vector<16x8xf32>
-  // CHECK: vector.transfer_write %[[NEW_VEC1]], %[[MEM0]][%c0, %c0, %c0, %c0] : vector<16x8xf32>, memref<?x?x?x?xf32>
-
   return %res : tensor<?x?x?x?xf32>
 }
 
+// CHECK-LABEL: func @transfer_write_permutations_memref
+// CHECK-SAME:    %[[MEM:.*]]: memref<?x?x?x?xf32>
+// CHECK-SAME:    %[[VEC:.*]]: vector<8x16xf32>
+func.func @transfer_write_permutations_memref(
+    %mem : memref<?x?x?x?xf32>, %vec : vector<8x16xf32>) {
+  // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
+  %c0 = arith.constant 0 : index
+
+  vector.transfer_write %vec, %mem[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d3, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
+  // CHECK: %[[NEW_VEC0:.*]] = vector.transpose %{{.*}} [1, 0] : vector<8x16xf32> to vector<16x8xf32>
+  // CHECK: vector.transfer_write %[[NEW_VEC0]], %[[MEM]][%c0, %c0, %c0, %c0] : vector<16x8xf32>, memref<?x?x?x?xf32>
+
+  return
+}
+
 // CHECK-LABEL: func @transfer_write_broadcast_unit_dim
 // CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
 // CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>

>From 5f31855355fae228e9a69d2f566d8403ca7df0d4 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Thu, 22 Aug 2024 16:07:14 +0800
Subject: [PATCH 5/8] Rename vector %arg as %src (suggested by nujaa and
 banach-space)

---
 .../vector-transfer-to-vector-load-store.mlir | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index dc11b48ef0f9ba..13e18085e4c8a8 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -22,13 +22,13 @@ func.func @vector_transfer_ops_0d_memref(%mem: memref<f32>, %vec: vector<1x1x1xf
 }
 
 // CHECK-LABEL: func @vector_transfer_ops_0d_tensor(
-//  CHECK-SAME:   %[[ARG0:.*]]: tensor<f32>
-func.func @vector_transfer_ops_0d_tensor(%arg0: tensor<f32>) -> vector<1xf32> {
+//  CHECK-SAME:   %[[SRC:.*]]: tensor<f32>
+func.func @vector_transfer_ops_0d_tensor(%src: tensor<f32>) -> vector<1xf32> {
     %f0 = arith.constant 0.0 : f32
 
-//  CHECK-NEXT:   %[[S:.*]] = tensor.extract %[[ARG0]][] : tensor<f32>
+//  CHECK-NEXT:   %[[S:.*]] = tensor.extract %[[SRC]][] : tensor<f32>
 //  CHECK-NEXT:   %[[V:.*]] = vector.broadcast %[[S]] : f32 to vector<1xf32>
-    %res = vector.transfer_read %arg0[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
+    %res = vector.transfer_read %src[], %f0 {in_bounds = [true], permutation_map = affine_map<()->(0)>} :
       tensor<f32>, vector<1xf32>
 
 //  CHECK-NEXT:   return %[[V]]
@@ -331,20 +331,20 @@ func.func @transfer_read_permutations(%mem_0 : memref<?x?xf32>, %mem_1 : memref<
 }
 
 // CHECK-LABEL: func @transfer_write_permutations_tensor
-// CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[SRC:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC:.*]]: vector<7x14x8x16xf32>
 // CHECK-SAME:    %[[M:.*]]: i1
 func.func @transfer_write_permutations_tensor(
-    %arg1 : tensor<?x?x?x?xf32>,
+    %src : tensor<?x?x?x?xf32>,
     %vec : vector<7x14x8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
   // CHECK: %[[MASK:.*]] = vector.splat %[[M]] : vector<16x14x7x8xi1>
   %mask0 = vector.splat %m : vector<16x14x7x8xi1>
-  %res = vector.transfer_write %vec, %arg1[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec, %src[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.transpose %{{.*}} [3, 1, 0, 2] : vector<7x14x8x16xf32> to vector<16x14x7x8xf32>
-  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[ARG1]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
+  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[SRC]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
 
   return %res : tensor<?x?x?x?xf32>
 }
@@ -366,19 +366,19 @@ func.func @transfer_write_permutations_memref(
 
 // CHECK-LABEL: func @transfer_write_broadcast_unit_dim
 // CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
-// CHECK-SAME:    %[[ARG1:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[SRC0:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC0:.*]]: vector<14x8x16xf32>
 // CHECK-SAME:    %[[VEC1:.*]]: vector<8x16xf32>
 func.func @transfer_write_broadcast_unit_dim(
-    %mem_0 : memref<?x?x?x?xf32>, %arg1 : tensor<?x?x?x?xf32>,
+    %mem_0 : memref<?x?x?x?xf32>, %src_0 : tensor<?x?x?x?xf32>,
     %vec_0 : vector<14x8x16xf32>, %vec_1 : vector<8x16xf32>) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
-  %res = vector.transfer_write %vec_0, %arg1[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec_0, %src_0[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.broadcast %{{.*}} : vector<14x8x16xf32> to vector<1x14x8x16xf32>
   // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %[[NEW_VEC0]], [1, 2, 0, 3] : vector<1x14x8x16xf32> to vector<14x8x1x16xf32>
-  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[ARG1]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
+  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[SRC0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
 
   vector.transfer_write %vec_1, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d1, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC2:.*]] = vector.broadcast %{{.*}} : vector<8x16xf32> to vector<1x8x16xf32>

>From 4871c153efed43b5d0b6f2ecf516c1c63bac1034 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Tue, 3 Sep 2024 16:25:58 +0800
Subject: [PATCH 6/8] Split transfer_write_broadcast_unit_dim into tensor and
 memref (as suggested by reviewers)

---
 .../vector-transfer-to-vector-load-store.mlir | 30 ++++++++++++-------
 1 file changed, 19 insertions(+), 11 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 13e18085e4c8a8..4f8b2c9ae0752b 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -364,14 +364,11 @@ func.func @transfer_write_permutations_memref(
   return
 }
 
-// CHECK-LABEL: func @transfer_write_broadcast_unit_dim
-// CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
+// CHECK-LABEL: func @transfer_write_broadcast_unit_dim_tensor
 // CHECK-SAME:    %[[SRC0:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC0:.*]]: vector<14x8x16xf32>
-// CHECK-SAME:    %[[VEC1:.*]]: vector<8x16xf32>
-func.func @transfer_write_broadcast_unit_dim(
-    %mem_0 : memref<?x?x?x?xf32>, %src_0 : tensor<?x?x?x?xf32>,
-    %vec_0 : vector<14x8x16xf32>, %vec_1 : vector<8x16xf32>) -> tensor<?x?x?x?xf32> {
+func.func @transfer_write_broadcast_unit_dim_tensor(
+    %src_0 : tensor<?x?x?x?xf32>, %vec_0 : vector<14x8x16xf32>) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
@@ -380,14 +377,25 @@ func.func @transfer_write_broadcast_unit_dim(
   // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %[[NEW_VEC0]], [1, 2, 0, 3] : vector<1x14x8x16xf32> to vector<14x8x1x16xf32>
   // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[SRC0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
 
-  vector.transfer_write %vec_1, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d1, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
-  // CHECK: %[[NEW_VEC2:.*]] = vector.broadcast %{{.*}} : vector<8x16xf32> to vector<1x8x16xf32>
-  // CHECK: %[[NEW_VEC3:.*]] = vector.transpose %[[NEW_VEC2]], [1, 2, 0] : vector<1x8x16xf32> to vector<8x16x1xf32>
-  // CHECK: vector.transfer_write %[[NEW_VEC3]], %[[MEM0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true]} : vector<8x16x1xf32>, memref<?x?x?x?xf32>
-
   return %res : tensor<?x?x?x?xf32>
 }
 
+// CHECK-LABEL: func @transfer_write_broadcast_unit_dim_memref
+// CHECK-SAME:    %[[MEM0:.*]]: memref<?x?x?x?xf32>
+// CHECK-SAME:    %[[VEC0:.*]]: vector<8x16xf32>
+func.func @transfer_write_broadcast_unit_dim_memref(
+    %mem_0 : memref<?x?x?x?xf32>, %vec_0 : vector<8x16xf32>) {
+  // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
+  %c0 = arith.constant 0 : index
+
+  vector.transfer_write %vec_0, %mem_0[%c0, %c0, %c0, %c0] {permutation_map = affine_map<(d0, d1, d2, d3) -> (d1, d2)>} : vector<8x16xf32>, memref<?x?x?x?xf32>
+  // CHECK: %[[NEW_VEC0:.*]] = vector.broadcast %{{.*}} : vector<8x16xf32> to vector<1x8x16xf32>
+  // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %[[NEW_VEC0]], [1, 2, 0] : vector<1x8x16xf32> to vector<8x16x1xf32>
+  // CHECK: vector.transfer_write %[[NEW_VEC1]], %[[MEM0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true]} : vector<8x16x1xf32>, memref<?x?x?x?xf32>
+
+  return
+}
+
 module attributes {transform.with_named_sequence} {
   transform.named_sequence @__transform_main(%root : !transform.any_op {transform.readonly}) {
     %func_op = transform.structured.match ops{["func.func"]} in %root : (!transform.any_op) -> !transform.op<"func.func">

>From c51424800cbecd97562792b413ac2b05817f9d41 Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Tue, 3 Sep 2024 16:29:22 +0800
Subject: [PATCH 7/8] Renamed transfer_write_permutations_tensor to masked to
 make it easier to differenciate with transfer_write_permutations_memref

---
 .../Dialect/Vector/vector-transfer-to-vector-load-store.mlir  | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 4f8b2c9ae0752b..30e6544f76741d 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -330,11 +330,11 @@ func.func @transfer_read_permutations(%mem_0 : memref<?x?xf32>, %mem_1 : memref<
          vector<7x14x8x16xf32>, vector<8xf32>
 }
 
-// CHECK-LABEL: func @transfer_write_permutations_tensor
+// CHECK-LABEL: func @transfer_write_permutations_tensor_masked
 // CHECK-SAME:    %[[SRC:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC:.*]]: vector<7x14x8x16xf32>
 // CHECK-SAME:    %[[M:.*]]: i1
-func.func @transfer_write_permutations_tensor(
+func.func @transfer_write_permutations_tensor_masked(
     %src : tensor<?x?x?x?xf32>,
     %vec : vector<7x14x8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index

>From ff7b93367aaee8208fbba86646ec37abc0508bed Mon Sep 17 00:00:00 2001
From: Pablo Antonio Martinez <pablo.antonio.martinez at huawei.com>
Date: Tue, 3 Sep 2024 16:31:29 +0800
Subject: [PATCH 8/8] Use dst rather than src when tensor is the destionation
 (suggested by banach-space)

---
 .../vector-transfer-to-vector-load-store.mlir    | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
index 30e6544f76741d..4d8e4a8296fb5a 100644
--- a/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
+++ b/mlir/test/Dialect/Vector/vector-transfer-to-vector-load-store.mlir
@@ -331,20 +331,20 @@ func.func @transfer_read_permutations(%mem_0 : memref<?x?xf32>, %mem_1 : memref<
 }
 
 // CHECK-LABEL: func @transfer_write_permutations_tensor_masked
-// CHECK-SAME:    %[[SRC:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[DST:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC:.*]]: vector<7x14x8x16xf32>
 // CHECK-SAME:    %[[M:.*]]: i1
 func.func @transfer_write_permutations_tensor_masked(
-    %src : tensor<?x?x?x?xf32>,
+    %dst : tensor<?x?x?x?xf32>,
     %vec : vector<7x14x8x16xf32>, %m: i1) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
   // CHECK: %[[MASK:.*]] = vector.splat %[[M]] : vector<16x14x7x8xi1>
   %mask0 = vector.splat %m : vector<16x14x7x8xi1>
-  %res = vector.transfer_write %vec, %src[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec, %dst[%c0, %c0, %c0, %c0], %mask0 {in_bounds = [true, false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d2, d1, d3, d0)>} : vector<7x14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.transpose %{{.*}} [3, 1, 0, 2] : vector<7x14x8x16xf32> to vector<16x14x7x8xf32>
-  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[SRC]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
+  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC0]], %[[DST]][%c0, %c0, %c0, %c0], %[[MASK]] {in_bounds = [true, false, true, false]} : vector<16x14x7x8xf32>, tensor<?x?x?x?xf32>
 
   return %res : tensor<?x?x?x?xf32>
 }
@@ -365,17 +365,17 @@ func.func @transfer_write_permutations_memref(
 }
 
 // CHECK-LABEL: func @transfer_write_broadcast_unit_dim_tensor
-// CHECK-SAME:    %[[SRC0:.*]]: tensor<?x?x?x?xf32>
+// CHECK-SAME:    %[[DST0:.*]]: tensor<?x?x?x?xf32>
 // CHECK-SAME:    %[[VEC0:.*]]: vector<14x8x16xf32>
 func.func @transfer_write_broadcast_unit_dim_tensor(
-    %src_0 : tensor<?x?x?x?xf32>, %vec_0 : vector<14x8x16xf32>) -> tensor<?x?x?x?xf32> {
+    %dst_0 : tensor<?x?x?x?xf32>, %vec_0 : vector<14x8x16xf32>) -> tensor<?x?x?x?xf32> {
   // CHECK-DAG: %[[C0:.*]] = arith.constant 0 : index
   %c0 = arith.constant 0 : index
 
-  %res = vector.transfer_write %vec_0, %src_0[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
+  %res = vector.transfer_write %vec_0, %dst_0[%c0, %c0, %c0, %c0] {in_bounds = [false, false, true], permutation_map = affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>} : vector<14x8x16xf32>, tensor<?x?x?x?xf32>
   // CHECK: %[[NEW_VEC0:.*]] = vector.broadcast %{{.*}} : vector<14x8x16xf32> to vector<1x14x8x16xf32>
   // CHECK: %[[NEW_VEC1:.*]] = vector.transpose %[[NEW_VEC0]], [1, 2, 0, 3] : vector<1x14x8x16xf32> to vector<14x8x1x16xf32>
-  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[SRC0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
+  // CHECK: %[[NEW_RES0:.*]] = vector.transfer_write %[[NEW_VEC1]], %[[DST0]][%[[C0]], %[[C0]], %[[C0]], %[[C0]]] {in_bounds = [false, false, true, true]} : vector<14x8x1x16xf32>, tensor<?x?x?x?xf32>
 
   return %res : tensor<?x?x?x?xf32>
 }



More information about the Mlir-commits mailing list