84 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#SparseMatrix = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : compressed, d1 : compressed) }>5 6// A contrived example that demonstrates the many different ways7// in which scalar values can be involved in a sparse kernel8// through the linalg generic op.9 10#trait = {11 indexing_maps = [12 affine_map<(i,j) -> (i,j)>, // A (sparse tensor)13 affine_map<(i,j) -> ()>, // p (scalar tensor)14 affine_map<(i,j) -> ()>, // q (true scalar)15 affine_map<(i,j) -> (i,j)> // X (dense tensor out)16 ],17 iterator_types = ["parallel", "parallel"],18 doc = "X(i,j) += A(i,j) * p * q * r * s * 2.2"19}20 21// CHECK-LABEL: func @mul(22// CHECK-SAME: %[[VAL_0:.*0]]: tensor<32x16xf32, #sparse{{[0-9]*}}>,23// CHECK-SAME: %[[VAL_1:.*1]]: tensor<f32>,24// CHECK-SAME: %[[VAL_2:.*2]]: f32,25// CHECK-SAME: %[[VAL_3:.*3]]: f32,26// CHECK-SAME: %[[VAL_4:.*4]]: tensor<32x16xf32>) -> tensor<32x16xf32> {27// CHECK-DAG: %[[VAL_5:.*]] = arith.constant 2.200000e+00 : f3228// CHECK-DAG: %[[VAL_6:.*]] = arith.constant 0 : index29// CHECK-DAG: %[[VAL_7:.*]] = arith.constant 1 : index30// CHECK-DAG: %[[VAL_8:.*]] = arith.addf %[[VAL_2]], %[[VAL_3]] : f3231// CHECK-DAG: %[[VAL_9:.*]] = sparse_tensor.positions %[[VAL_0]] {level = 0 : index} : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xindex>32// CHECK-DAG: %[[VAL_10:.*]] = sparse_tensor.coordinates %[[VAL_0]] {level = 0 : index} : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xindex>33// CHECK-DAG: %[[VAL_11:.*]] = sparse_tensor.positions %[[VAL_0]] {level = 1 : index} : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xindex>34// CHECK-DAG: %[[VAL_12:.*]] = sparse_tensor.coordinates %[[VAL_0]] {level = 1 : index} : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xindex>35// CHECK-DAG: %[[VAL_13:.*]] = sparse_tensor.values %[[VAL_0]] : tensor<32x16xf32, #sparse{{[0-9]*}}> to memref<?xf32>36// CHECK-DAG: %[[VAL_14:.*]] = bufferization.to_buffer %[[VAL_1]] : tensor<f32> to memref<f32>37// CHECK-DAG: %[[VAL_15:.*]] = bufferization.to_buffer %[[VAL_4]] : tensor<32x16xf32> to memref<32x16xf32>38// CHECK-DAG: %[[VAL_16:.*]] = memref.load %[[VAL_14]][] : memref<f32>39// CHECK-DAG: %[[VAL_17:.*]] = memref.load %[[VAL_9]]{{\[}}%[[VAL_6]]] : memref<?xindex>40// CHECK-DAG: %[[VAL_18:.*]] = memref.load %[[VAL_9]]{{\[}}%[[VAL_7]]] : memref<?xindex>41// CHECK: scf.for %[[VAL_19:.*]] = %[[VAL_17]] to %[[VAL_18]] step %[[VAL_7]] {42// CHECK: %[[VAL_20:.*]] = memref.load %[[VAL_10]]{{\[}}%[[VAL_19]]] : memref<?xindex>43// CHECK: %[[VAL_21:.*]] = memref.load %[[VAL_11]]{{\[}}%[[VAL_19]]] : memref<?xindex>44// CHECK: %[[VAL_22:.*]] = arith.addi %[[VAL_19]], %[[VAL_7]] : index45// CHECK: %[[VAL_23:.*]] = memref.load %[[VAL_11]]{{\[}}%[[VAL_22]]] : memref<?xindex>46// CHECK: scf.for %[[VAL_24:.*]] = %[[VAL_21]] to %[[VAL_23]] step %[[VAL_7]] {47// CHECK: %[[VAL_25:.*]] = memref.load %[[VAL_12]]{{\[}}%[[VAL_24]]] : memref<?xindex>48// CHECK: %[[VAL_26:.*]] = memref.load %[[VAL_13]]{{\[}}%[[VAL_24]]] : memref<?xf32>49// CHECK: %[[VAL_27:.*]] = arith.mulf %[[VAL_26]], %[[VAL_16]] : f3250// CHECK: %[[VAL_28:.*]] = arith.mulf %[[VAL_27]], %[[VAL_2]] : f3251// CHECK: %[[VAL_29:.*]] = arith.mulf %[[VAL_28]], %[[VAL_3]] : f3252// CHECK: %[[VAL_30:.*]] = arith.mulf %[[VAL_29]], %[[VAL_8]] : f3253// CHECK: %[[VAL_31:.*]] = arith.mulf %[[VAL_30]], %[[VAL_5]] : f3254// CHECK: %[[VAL_32:.*]] = memref.load %[[VAL_15]]{{\[}}%[[VAL_20]], %[[VAL_25]]] : memref<32x16xf32>55// CHECK: %[[VAL_33:.*]] = arith.addf %[[VAL_31]], %[[VAL_32]] : f3256// CHECK: memref.store %[[VAL_33]], %[[VAL_15]]{{\[}}%[[VAL_20]], %[[VAL_25]]] : memref<32x16xf32>57// CHECK: }58// CHECK: }59// CHECK: %[[VAL_34:.*]] = bufferization.to_tensor %[[VAL_15]] : memref<32x16xf32>60// CHECK: return %[[VAL_34]] : tensor<32x16xf32>61// CHECK: }62func.func @mul(%arga: tensor<32x16xf32, #SparseMatrix>,63 %argp: tensor<f32>,64 %argq: f32,65 %argr: f32,66 %argx: tensor<32x16xf32>) -> tensor<32x16xf32> {67 %s = arith.addf %argq, %argr : f3268 %c = arith.constant 2.2 : f3269 %0 = linalg.generic #trait70 ins(%arga, %argp, %argq: tensor<32x16xf32, #SparseMatrix>, tensor<f32>, f32)71 outs(%argx: tensor<32x16xf32>) {72 ^bb(%a: f32, %p: f32, %q: f32, %x: f32):73 %0 = arith.mulf %a, %p : f32 // scalar tensor argument74 %1 = arith.mulf %0, %q : f32 // scalar argument75 %2 = arith.mulf %1, %argr : f32 // scalar argument from outside block76 %3 = arith.mulf %2, %s : f32 // scalar value from outside block77 %4 = arith.mulf %3, %c : f32 // direct constant from outside block78 %5 = arith.addf %4, %x : f3279 linalg.yield %5 : f3280 } -> tensor<32x16xf32>81 82 return %0 : tensor<32x16xf32>83}84