1//-------------------------------------------------------------------------------------------------- 2// WHEN CREATING A NEW TEST, PLEASE JUST COPY & PASTE WITHOUT EDITS. 3// 4// Set-up that's shared across all tests in this directory. In principle, this 5// config could be moved to lit.local.cfg. However, there are downstream users that 6// do not use these LIT config files. Hence why this is kept inline. 7// 8// DEFINE: %{sparsifier_opts} = enable-runtime-library=true 9// DEFINE: %{sparsifier_opts_sve} = enable-arm-sve=true %{sparsifier_opts} 10// DEFINE: %{compile} = mlir-opt %s --sparsifier="%{sparsifier_opts}" 11// DEFINE: %{compile_sve} = mlir-opt %s --sparsifier="%{sparsifier_opts_sve}" 12// DEFINE: %{run_libs} = -shared-libs=%mlir_c_runner_utils,%mlir_runner_utils 13// DEFINE: %{run_libs_sve} = -shared-libs=%native_mlir_runner_utils,%native_mlir_c_runner_utils 14// DEFINE: %{run_opts} = -e main -entry-point-result=void 15// DEFINE: %{run} = mlir-runner %{run_opts} %{run_libs} 16// DEFINE: %{run_sve} = %mcr_aarch64_cmd --march=aarch64 --mattr="+sve" %{run_opts} %{run_libs_sve} 17// 18// DEFINE: %{env} = 19//-------------------------------------------------------------------------------------------------- 20 21// RUN: %{compile} | %{run} | FileCheck %s 22// 23// Do the same run, but now with direct IR generation. 24// REDEFINE: %{sparsifier_opts} = enable-runtime-library=false 25// RUN: %{compile} | %{run} | FileCheck %s 26// 27// Do the same run, but now with vectorization. 28// REDEFINE: %{sparsifier_opts} = enable-runtime-library=false vl=4 29// RUN: %{compile} | %{run} | FileCheck %s 30// 31// Do the same run, but now with VLA vectorization. 32// RUN: %if mlir_arm_sve_tests %{ %{compile_sve} | %{run_sve} | FileCheck %s %} 33 34#CSR = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : compressed) }> 35 36#trait_scale = { 37 indexing_maps = [ 38 affine_map<(i,j) -> (i,j)> // X (out) 39 ], 40 iterator_types = ["parallel", "parallel"], 41 doc = "X(i,j) = X(i,j) * 2" 42} 43 44// 45// Integration test that lowers a kernel annotated as sparse to actual sparse 46// code, initializes a matching sparse storage scheme from a dense tensor, 47// and runs the resulting code with the JIT compiler. 48// 49module { 50 // 51 // A kernel that scales a sparse matrix A by a factor of 2.0. 52 // 53 func.func @sparse_scale(%argx: tensor<8x8xf32, #CSR>) -> tensor<8x8xf32, #CSR> { 54 %c = arith.constant 2.0 : f32 55 %0 = linalg.generic #trait_scale 56 outs(%argx: tensor<8x8xf32, #CSR>) { 57 ^bb(%x: f32): 58 %1 = arith.mulf %x, %c : f32 59 linalg.yield %1 : f32 60 } -> tensor<8x8xf32, #CSR> 61 return %0 : tensor<8x8xf32, #CSR> 62 } 63 64 // 65 // Main driver that converts a dense tensor into a sparse tensor 66 // and then calls the sparse scaling kernel with the sparse tensor 67 // as input argument. 68 // 69 func.func @main() { 70 %c0 = arith.constant 0 : index 71 %f0 = arith.constant 0.0 : f32 72 73 // Initialize a dense tensor. 74 %0 = arith.constant dense<[ 75 [1.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 1.0], 76 [0.0, 2.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 77 [0.0, 0.0, 3.0, 0.0, 0.0, 0.0, 0.0, 0.0], 78 [0.0, 0.0, 0.0, 4.0, 0.0, 0.0, 0.0, 0.0], 79 [0.0, 1.0, 0.0, 0.0, 5.0, 0.0, 0.0, 0.0], 80 [0.0, 1.0, 1.0, 0.0, 0.0, 6.0, 0.0, 0.0], 81 [0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 7.0, 1.0], 82 [0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 8.0] 83 ]> : tensor<8x8xf32> 84 85 // Convert dense tensor to sparse tensor and call sparse kernel. 86 %1 = sparse_tensor.convert %0 : tensor<8x8xf32> to tensor<8x8xf32, #CSR> 87 %2 = call @sparse_scale(%1) 88 : (tensor<8x8xf32, #CSR>) -> tensor<8x8xf32, #CSR> 89 90 // Print the resulting compacted values for verification. 91 // 92 // CHECK: ---- Sparse Tensor ---- 93 // CHECK-NEXT: nse = 16 94 // CHECK-NEXT: dim = ( 8, 8 ) 95 // CHECK-NEXT: lvl = ( 8, 8 ) 96 // CHECK-NEXT: pos[1] : ( 0, 3, 4, 5, 6, 8, 11, 14, 16 ) 97 // CHECK-NEXT: crd[1] : ( 0, 2, 7, 1, 2, 3, 1, 4, 1, 2, 5, 2, 6, 7, 2, 7 ) 98 // CHECK-NEXT: values : ( 2, 2, 2, 4, 6, 8, 2, 10, 2, 2, 12, 2, 14, 2, 2, 16 ) 99 // CHECK-NEXT: ---- 100 // 101 sparse_tensor.print %2 : tensor<8x8xf32, #CSR> 102 103 // Release the resources. 104 bufferization.dealloc_tensor %1 : tensor<8x8xf32, #CSR> 105 106 return 107 } 108} 109