brintos

brintos / llvm-project-archived public Read only

0
0
Text · 7.9 KiB · c702270 Raw
155 lines · plain
1// NOTE: Assertions have been autogenerated by utils/generate-test-checks.py2// RUN: mlir-opt %s --sparse-reinterpret-map -sparsification | FileCheck %s3 4// Test to demonstrate the difference between non-annotated dense tensors5// and all-dense-annotated "sparse" tensors. The former class remains as6// two-dimensional tensors that are bufferized by subsequent passes. The7// latter class is linearized into one-dimensional buffers that are backed8// by the runtime support library.9 10#DenseMatrix = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : dense) }>11 12#trait_2d = {13  indexing_maps = [14    affine_map<(i,j) -> (i,j)>,  // A15    affine_map<(i,j) -> (i,j)>   // X (out)16  ],17  iterator_types = ["parallel", "parallel"],18  doc = "X(i,j) = A(i,j) + 1"19}20 21#trait_3d = {22  indexing_maps = [23    affine_map<(i,j,k) -> (i,j,k)>,  // A24    affine_map<(i,j,k) -> (i,j)>     // X (out)25  ],26  iterator_types = ["parallel", "parallel", "reduction"],27  doc = "X(i,j) += A(i,j,k)"28}29 30//31// Test with an all-dense-annotated "sparse" matrix as input and32// a non-annotated dense matrix as output.33//34// CHECK-LABEL:   func @dense1(35// CHECK-SAME:                 %[[VAL_0:.*]]: tensor<32x16xf32, #sparse{{[0-9]*}}>,36// CHECK-SAME:                 %[[VAL_1:.*]]: tensor<32x16xf32>) -> tensor<32x16xf32> {37// CHECK-DAG:       %[[VAL_2:.*]] = arith.constant 1.000000e+00 : f3238// CHECK-DAG:       %[[VAL_3:.*]] = arith.constant 32 : index39// CHECK-DAG:       %[[VAL_4:.*]] = arith.constant 16 : index40// CHECK-DAG:       %[[VAL_5:.*]] = arith.constant 0 : index41// CHECK-DAG:       %[[VAL_6:.*]] = arith.constant 1 : index42// CHECK:           %[[VAL_7:.*]] = sparse_tensor.values %[[VAL_0]] : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xf32>43// CHECK:           %[[VAL_8:.*]] = bufferization.to_buffer %[[VAL_1]] : tensor<32x16xf32> to memref<32x16xf32>44// CHECK:           scf.for %[[VAL_9:.*]] = %[[VAL_5]] to %[[VAL_3]] step %[[VAL_6]] {45// CHECK:             %[[VAL_11:.*]] = arith.muli %[[VAL_9]], %[[VAL_4]] : index46// CHECK:             scf.for %[[VAL_10:.*]] = %[[VAL_5]] to %[[VAL_4]] step %[[VAL_6]] {47// CHECK:               %[[VAL_12:.*]] = arith.addi %[[VAL_10]], %[[VAL_11]] : index48// CHECK:               %[[VAL_13:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_12]]] : memref<?xf32>49// CHECK:               %[[VAL_14:.*]] = arith.addf %[[VAL_13]], %[[VAL_2]] : f3250// CHECK:               memref.store %[[VAL_14]], %[[VAL_8]]{{\[}}%[[VAL_9]], %[[VAL_10]]] : memref<32x16xf32>51// CHECK:             }52// CHECK:           }53// CHECK:           %[[VAL_15:.*]] = bufferization.to_tensor %[[VAL_8]] : memref<32x16xf32>54// CHECK:           return %[[VAL_15]] : tensor<32x16xf32>55// CHECK:         }56func.func @dense1(%arga: tensor<32x16xf32, #DenseMatrix>,57                  %argx: tensor<32x16xf32>)58	     -> tensor<32x16xf32> {59  %c = arith.constant 1.0 : f3260  %0 = linalg.generic #trait_2d61     ins(%arga: tensor<32x16xf32, #DenseMatrix>)62    outs(%argx: tensor<32x16xf32>) {63      ^bb(%a: f32, %x: f32):64        %1 = arith.addf %a, %c : f3265        linalg.yield %1 : f3266  } -> tensor<32x16xf32>67  return %0 : tensor<32x16xf32>68}69 70//71// Test with a non-annotated dense matrix as input and72// an all-dense annotated "sparse" matrix as output.73//74// CHECK-LABEL:   func @dense2(75// CHECK-SAME:      %[[VAL_0:.*]]: tensor<32x16xf32>,76// CHECK-SAME:      %[[VAL_1:.*]]: tensor<32x16xf32, #sparse{{[0-9]*}}>) -> tensor<32x16xf32, #sparse{{[0-9]*}}> {77// CHECK-DAG:       %[[VAL_2:.*]] = arith.constant 1.000000e+00 : f3278// CHECK-DAG:       %[[VAL_3:.*]] = arith.constant 32 : index79// CHECK-DAG:       %[[VAL_4:.*]] = arith.constant 16 : index80// CHECK-DAG:       %[[VAL_5:.*]] = arith.constant 0 : index81// CHECK-DAG:       %[[VAL_6:.*]] = arith.constant 1 : index82// CHECK:           %[[VAL_7:.*]] = bufferization.to_buffer %[[VAL_0]] : tensor<32x16xf32> to memref<32x16xf32>83// CHECK:           %[[VAL_8:.*]] = sparse_tensor.values %[[VAL_1]] : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xf32>84// CHECK:           scf.for %[[VAL_9:.*]] = %[[VAL_5]] to %[[VAL_3]] step %[[VAL_6]] {85// CHECK:             %[[VAL_11:.*]] = arith.muli %[[VAL_9]], %[[VAL_4]] : index86// CHECK:             scf.for %[[VAL_10:.*]] = %[[VAL_5]] to %[[VAL_4]] step %[[VAL_6]] {87// CHECK:               %[[VAL_12:.*]] = arith.addi %[[VAL_10]], %[[VAL_11]] : index88// CHECK:               %[[VAL_13:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_9]], %[[VAL_10]]] : memref<32x16xf32>89// CHECK:               %[[VAL_14:.*]] = arith.addf %[[VAL_13]], %[[VAL_2]] : f3290// CHECK:               memref.store %[[VAL_14]], %[[VAL_8]]{{\[}}%[[VAL_12]]] : memref<?xf32>91// CHECK:             }92// CHECK:           }93// CHECK:           %[[VAL_15:.*]] = sparse_tensor.load %[[VAL_1]] : tensor<32x16xf32, #sparse{{[0-9]*}}>94// CHECK:           return %[[VAL_15]] : tensor<32x16xf32, #sparse{{[0-9]*}}>95// CHECK:         }96func.func @dense2(%arga: tensor<32x16xf32>,97                  %argx: tensor<32x16xf32, #DenseMatrix>)98	     -> tensor<32x16xf32, #DenseMatrix> {99  %c = arith.constant 1.0 : f32100  %0 = linalg.generic #trait_2d101     ins(%arga: tensor<32x16xf32>)102    outs(%argx: tensor<32x16xf32, #DenseMatrix>) {103      ^bb(%a: f32, %x: f32):104        %1 = arith.addf %a, %c : f32105        linalg.yield %1 : f32106  } -> tensor<32x16xf32, #DenseMatrix>107  return %0 : tensor<32x16xf32, #DenseMatrix>108}109 110 111//112// Test with a non-annotated dense matrix as input and113// an all-dense annotated "sparse" matrix as output.114// The missing innermost "k" index (due to a reduction) is accounted115// for by scalarizing the reduction operation for the output tensor.116//117// CHECK-LABEL:   func @dense3(118// CHECK-SAME:      %[[VAL_0:.*]]: tensor<32x16x8xf32>,119// CHECK-SAME:      %[[VAL_1:.*]]: tensor<32x16xf32, #sparse{{[0-9]*}}>) -> tensor<32x16xf32, #sparse{{[0-9]*}}> {120// CHECK-DAG:       %[[VAL_2:.*]] = arith.constant 8 : index121// CHECK-DAG:       %[[VAL_3:.*]] = arith.constant 32 : index122// CHECK-DAG:       %[[VAL_4:.*]] = arith.constant 16 : index123// CHECK-DAG:       %[[VAL_5:.*]] = arith.constant 0 : index124// CHECK-DAG:       %[[VAL_6:.*]] = arith.constant 1 : index125// CHECK:           %[[VAL_7:.*]] = bufferization.to_buffer %[[VAL_0]] : tensor<32x16x8xf32> to memref<32x16x8xf32>126// CHECK:           %[[VAL_8:.*]] = sparse_tensor.values %[[VAL_1]] : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xf32>127// CHECK:           scf.for %[[VAL_9:.*]] = %[[VAL_5]] to %[[VAL_3]] step %[[VAL_6]] {128// CHECK:             %[[VAL_11:.*]] = arith.muli %[[VAL_9]], %[[VAL_4]] : index129// CHECK:             scf.for %[[VAL_10:.*]] = %[[VAL_5]] to %[[VAL_4]] step %[[VAL_6]] {130// CHECK:               %[[VAL_12:.*]] = arith.addi %[[VAL_10]], %[[VAL_11]] : index131// CHECK:               %[[VAL_13:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_12]]] : memref<?xf32>132// CHECK:               %[[VAL_14:.*]] = scf.for %[[VAL_15:.*]] = %[[VAL_5]] to %[[VAL_2]] step %[[VAL_6]] iter_args(%[[VAL_16:.*]] = %[[VAL_13]]) -> (f32) {133// CHECK:                 %[[VAL_17:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_9]], %[[VAL_10]], %[[VAL_15]]] : memref<32x16x8xf32>134// CHECK:                 %[[VAL_18:.*]] = arith.addf %[[VAL_16]], %[[VAL_17]] : f32135// CHECK:                 scf.yield %[[VAL_18]] : f32136// CHECK:               }137// CHECK:               memref.store %[[VAL_19:.*]], %[[VAL_8]]{{\[}}%[[VAL_12]]] : memref<?xf32>138// CHECK:             }139// CHECK:           }140// CHECK:           %[[VAL_20:.*]] = sparse_tensor.load %[[VAL_1]] : tensor<32x16xf32, #sparse{{[0-9]*}}>141// CHECK:           return %[[VAL_20]] : tensor<32x16xf32, #sparse{{[0-9]*}}>142// CHECK:         }143func.func @dense3(%arga: tensor<32x16x8xf32>,144                  %argx: tensor<32x16xf32, #DenseMatrix>)145	     -> tensor<32x16xf32, #DenseMatrix> {146  %0 = linalg.generic #trait_3d147     ins(%arga: tensor<32x16x8xf32>)148    outs(%argx: tensor<32x16xf32, #DenseMatrix>) {149      ^bb(%a: f32, %x: f32):150        %1 = arith.addf %x, %a : f32151        linalg.yield %1 : f32152  } -> tensor<32x16xf32, #DenseMatrix>153  return %0 : tensor<32x16xf32, #DenseMatrix>154}155