llvm-project/mlir/test/Dialect/SparseTensor/sparse_index.mlir

129 lines
7.7 KiB
MLIR

// NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
// RUN: mlir-opt %s -sparsification | FileCheck %s
#DenseMatrix = #sparse_tensor.encoding<{
dimLevelType = ["dense", "dense"]
}>
#SparseMatrix = #sparse_tensor.encoding<{
dimLevelType = ["compressed", "compressed"]
}>
#trait = {
indexing_maps = [
affine_map<(i,j) -> (i,j)>, // A
affine_map<(i,j) -> (i,j)> // X (out)
],
iterator_types = ["parallel", "parallel"],
doc = "X(i,j) = A(i,j) * i * j"
}
// CHECK-LABEL: func @dense_index(
// CHECK-SAME: %[[VAL_0:.*]]: tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_1:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[VAL_2:.*]] = arith.constant 1 : index
// CHECK-DAG: %[[VAL_3:.*]] = tensor.dim %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_4:.*]] = tensor.dim %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_5:.*]] = sparse_tensor.init{{\[}}%[[VAL_3]], %[[VAL_4]]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_6:.*]] = sparse_tensor.values %[[VAL_0]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_7:.*]] = tensor.dim %[[VAL_5]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_8:.*]] = tensor.dim %[[VAL_5]], %[[VAL_2]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_9:.*]] = sparse_tensor.values %[[VAL_5]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: scf.for %[[VAL_10:.*]] = %[[VAL_1]] to %[[VAL_7]] step %[[VAL_2]] {
// CHECK: scf.for %[[VAL_11:.*]] = %[[VAL_1]] to %[[VAL_8]] step %[[VAL_2]] {
// CHECK: %[[VAL_12:.*]] = arith.muli %[[VAL_8]], %[[VAL_10]] : index
// CHECK: %[[VAL_13:.*]] = arith.addi %[[VAL_12]], %[[VAL_11]] : index
// CHECK: %[[VAL_14:.*]] = arith.muli %[[VAL_8]], %[[VAL_10]] : index
// CHECK: %[[VAL_15:.*]] = arith.addi %[[VAL_14]], %[[VAL_11]] : index
// CHECK: %[[VAL_16:.*]] = arith.index_cast %[[VAL_11]] : index to i64
// CHECK: %[[VAL_17:.*]] = arith.index_cast %[[VAL_10]] : index to i64
// CHECK: %[[VAL_18:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_13]]] : memref<?xi64>
// CHECK: %[[VAL_19:.*]] = arith.muli %[[VAL_17]], %[[VAL_18]] : i64
// CHECK: %[[VAL_20:.*]] = arith.muli %[[VAL_16]], %[[VAL_19]] : i64
// CHECK: memref.store %[[VAL_20]], %[[VAL_9]]{{\[}}%[[VAL_15]]] : memref<?xi64>
// CHECK: }
// CHECK: }
// CHECK: %[[VAL_21:.*]] = sparse_tensor.load %[[VAL_5]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: return %[[VAL_21]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: }
func.func @dense_index(%arga: tensor<?x?xi64, #DenseMatrix>)
-> tensor<?x?xi64, #DenseMatrix> {
%c0 = arith.constant 0 : index
%c1 = arith.constant 0 : index
%0 = tensor.dim %arga, %c0 : tensor<?x?xi64, #DenseMatrix>
%1 = tensor.dim %arga, %c1 : tensor<?x?xi64, #DenseMatrix>
%init = sparse_tensor.init [%0, %1] : tensor<?x?xi64, #DenseMatrix>
%r = linalg.generic #trait
ins(%arga: tensor<?x?xi64, #DenseMatrix>)
outs(%init: tensor<?x?xi64, #DenseMatrix>) {
^bb(%a: i64, %x: i64):
%i = linalg.index 0 : index
%j = linalg.index 1 : index
%ii = arith.index_cast %i : index to i64
%jj = arith.index_cast %j : index to i64
%m1 = arith.muli %ii, %a : i64
%m2 = arith.muli %jj, %m1 : i64
linalg.yield %m2 : i64
} -> tensor<?x?xi64, #DenseMatrix>
return %r : tensor<?x?xi64, #DenseMatrix>
}
// CHECK-LABEL: func @sparse_index(
// CHECK-SAME: %[[VAL_0:.*]]: tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_1:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[VAL_2:.*]] = arith.constant 1 : index
// CHECK-DAG: %[[VAL_3:.*]] = arith.constant 2 : index
// CHECK-DAG: %[[VAL_4:.*]] = tensor.dim %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_5:.*]] = tensor.dim %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_6:.*]] = sparse_tensor.init{{\[}}%[[VAL_4]], %[[VAL_5]]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_7:.*]] = sparse_tensor.pointers %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_8:.*]] = sparse_tensor.indices %[[VAL_0]], %[[VAL_1]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_9:.*]] = sparse_tensor.pointers %[[VAL_0]], %[[VAL_2]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_10:.*]] = sparse_tensor.indices %[[VAL_0]], %[[VAL_2]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK-DAG: %[[VAL_11:.*]] = sparse_tensor.values %[[VAL_0]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: %[[VAL_12:.*]] = memref.alloca(%[[VAL_3]]) : memref<?xindex>
// CHECK: %[[VAL_13:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_1]]] : memref<?xindex>
// CHECK: %[[VAL_14:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_2]]] : memref<?xindex>
// CHECK: scf.for %[[VAL_15:.*]] = %[[VAL_13]] to %[[VAL_14]] step %[[VAL_2]] {
// CHECK: %[[VAL_16:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_15]]] : memref<?xindex>
// CHECK: memref.store %[[VAL_16]], %[[VAL_12]]{{\[}}%[[VAL_1]]] : memref<?xindex>
// CHECK: %[[VAL_17:.*]] = memref.load %[[VAL_9]]{{\[}}%[[VAL_15]]] : memref<?xindex>
// CHECK: %[[VAL_18:.*]] = arith.addi %[[VAL_15]], %[[VAL_2]] : index
// CHECK: %[[VAL_19:.*]] = memref.load %[[VAL_9]]{{\[}}%[[VAL_18]]] : memref<?xindex>
// CHECK: scf.for %[[VAL_20:.*]] = %[[VAL_17]] to %[[VAL_19]] step %[[VAL_2]] {
// CHECK: %[[VAL_21:.*]] = memref.load %[[VAL_10]]{{\[}}%[[VAL_20]]] : memref<?xindex>
// CHECK: memref.store %[[VAL_21]], %[[VAL_12]]{{\[}}%[[VAL_2]]] : memref<?xindex>
// CHECK: %[[VAL_22:.*]] = arith.index_cast %[[VAL_21]] : index to i64
// CHECK: %[[VAL_23:.*]] = arith.index_cast %[[VAL_16]] : index to i64
// CHECK: %[[VAL_24:.*]] = memref.load %[[VAL_11]]{{\[}}%[[VAL_20]]] : memref<?xi64>
// CHECK: %[[VAL_25:.*]] = arith.muli %[[VAL_23]], %[[VAL_24]] : i64
// CHECK: %[[VAL_26:.*]] = arith.muli %[[VAL_22]], %[[VAL_25]] : i64
// CHECK: sparse_tensor.lex_insert %[[VAL_6]], %[[VAL_12]], %[[VAL_26]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: }
// CHECK: }
// CHECK: %[[VAL_27:.*]] = sparse_tensor.load %[[VAL_6]] hasInserts : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: return %[[VAL_27]] : tensor<?x?xi64, #sparse_tensor.encoding
// CHECK: }
func.func @sparse_index(%arga: tensor<?x?xi64, #SparseMatrix>)
-> tensor<?x?xi64, #SparseMatrix> {
%c0 = arith.constant 0 : index
%c1 = arith.constant 0 : index
%0 = tensor.dim %arga, %c0 : tensor<?x?xi64, #SparseMatrix>
%1 = tensor.dim %arga, %c1 : tensor<?x?xi64, #SparseMatrix>
%init = sparse_tensor.init [%0, %1] : tensor<?x?xi64, #SparseMatrix>
%r = linalg.generic #trait
ins(%arga: tensor<?x?xi64, #SparseMatrix>)
outs(%init: tensor<?x?xi64, #SparseMatrix>) {
^bb(%a: i64, %x: i64):
%i = linalg.index 0 : index
%j = linalg.index 1 : index
%ii = arith.index_cast %i : index to i64
%jj = arith.index_cast %j : index to i64
%m1 = arith.muli %ii, %a : i64
%m2 = arith.muli %jj, %m1 : i64
linalg.yield %m2 : i64
} -> tensor<?x?xi64, #SparseMatrix>
return %r : tensor<?x?xi64, #SparseMatrix>
}