From a0220ba7218f0331986f0595662342ecfd173d92 Mon Sep 17 00:00:00 2001 From: Quentin Colombet Date: Thu, 8 Dec 2022 11:45:19 +0000 Subject: [PATCH] [mlir][SparseTensor] Add a few tests for sparse vectorization These tests covers mulf, ori, and subi. NFC Differential Revision: https://reviews.llvm.org/D139625 --- .../Dialect/SparseTensor/vectorize_reduction.mlir | 436 +++++++++++++++++++++ 1 file changed, 436 insertions(+) create mode 100644 mlir/test/Dialect/SparseTensor/vectorize_reduction.mlir diff --git a/mlir/test/Dialect/SparseTensor/vectorize_reduction.mlir b/mlir/test/Dialect/SparseTensor/vectorize_reduction.mlir new file mode 100644 index 0000000..e292cec --- /dev/null +++ b/mlir/test/Dialect/SparseTensor/vectorize_reduction.mlir @@ -0,0 +1,436 @@ +// RUN: mlir-opt %s -sparsification -cse -sparse-vectorization="vl=8" -cse -split-input-file | \ +// RUN: FileCheck %s --check-prefix=CHECK-ON +// RUN: mlir-opt %s -sparsification -cse -split-input-file | \ +// RUN: FileCheck %s --check-prefix=CHECK-OFF + +// ----- + +// Check that we recognize a reduction with a mul operator. +// We use two dimensions here to check that the vectorization +// is not affected by how the outer loop is layed out. +// In other words, we should be able to vectorize the sparse inner loop +// regardless of whether the outer loop is dense or sparse. +// +// For this particular test, we expect: +// With vectorization on: +// dense scf.for +// init vector_accumulator = {scalar_accumulator, 1.0, 1.0, ...} +// sparse scf.for +// vectorized mul in vector_accumulator, vector_input +// horizontal reduction of the vector_accumulator to scalar_accumulator +// final store of scalar_accumulaor +// +// With vectorization off: +// dense scf.for +// sparse scf.for +// mul in accumulator +// final store +// +// CHECK-ON-LABEL: func.func @sparse_product_reduction_dense_sparse( +// CHECK-ON-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-ON-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-ON-DAG: %[[VAL_2:.*]] = arith.constant 8 : index +// CHECK-ON-DAG: %[[VAL_3:.*]] = arith.constant dense<1.000000e+00> : vector<8xf64> +// CHECK-ON-DAG: %[[VAL_4:.*]] = arith.constant dense<0.000000e+00> : vector<8xf64> +// CHECK-ON-DAG: %[[VAL_5:.*]] = arith.constant 0 : index +// CHECK-ON-DAG: %[[VAL_6:.*]] = arith.constant 1 : index +// CHECK-ON-DAG: %[[VAL_7:.*]] = tensor.dim %[[VAL_1]], %[[VAL_5]] : tensor> +// CHECK-ON: %[[VAL_8:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 1 : index} : tensor> to memref +// CHECK-ON: %[[VAL_9:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-ON: %[[VAL_10:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-ON: %[[VAL_11:.*]] = memref.load %[[VAL_10]][] : memref +// CHECK-ON: %[[VAL_12:.*]] = scf.for %[[VAL_13:.*]] = %[[VAL_5]] to %[[VAL_7]] step %[[VAL_6]] iter_args(%[[VAL_14:.*]] = %[[VAL_11]]) -> (f64) { +// CHECK-ON: %[[VAL_15:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_13]]] : memref +// CHECK-ON: %[[VAL_16:.*]] = arith.addi %[[VAL_13]], %[[VAL_6]] : index +// CHECK-ON: %[[VAL_17:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_16]]] : memref +// CHECK-ON: %[[VAL_18:.*]] = vector.insertelement %[[VAL_14]], %[[VAL_3]]{{\[}}%[[VAL_5]] : index] : vector<8xf64> +// CHECK-ON: %[[VAL_19:.*]] = scf.for %[[VAL_20:.*]] = %[[VAL_15]] to %[[VAL_17]] step %[[VAL_2]] iter_args(%[[VAL_21:.*]] = %[[VAL_18]]) -> (vector<8xf64>) { +// CHECK-ON: %[[VAL_22:.*]] = affine.min #map(%[[VAL_17]], %[[VAL_20]]){{\[}}%[[VAL_2]]] +// CHECK-ON: %[[VAL_23:.*]] = vector.create_mask %[[VAL_22]] : vector<8xi1> +// CHECK-ON: %[[VAL_24:.*]] = vector.maskedload %[[VAL_9]]{{\[}}%[[VAL_20]]], %[[VAL_23]], %[[VAL_4]] : memref, vector<8xi1>, vector<8xf64> into vector<8xf64> +// CHECK-ON: %[[VAL_25:.*]] = arith.mulf %[[VAL_21]], %[[VAL_24]] : vector<8xf64> +// CHECK-ON: %[[VAL_26:.*]] = arith.select %[[VAL_23]], %[[VAL_25]], %[[VAL_21]] : vector<8xi1>, vector<8xf64> +// CHECK-ON: scf.yield %[[VAL_26]] : vector<8xf64> +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: %[[VAL_27:.*]] = vector.reduction , %[[VAL_28:.*]] : vector<8xf64> into f64 +// CHECK-ON: scf.yield %[[VAL_27]] : f64 +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: memref.store %[[VAL_29:.*]], %[[VAL_10]][] : memref +// CHECK-ON: %[[VAL_30:.*]] = bufferization.to_tensor %[[VAL_10]] : memref +// CHECK-ON: return %[[VAL_30]] : tensor +// CHECK-ON: } +// +// CHECK-OFF-LABEL: func.func @sparse_product_reduction_dense_sparse( +// CHECK-OFF-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-OFF-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-OFF-DAG: %[[VAL_2:.*]] = arith.constant 0 : index +// CHECK-OFF-DAG: %[[VAL_3:.*]] = arith.constant 1 : index +// CHECK-OFF: %[[VAL_4:.*]] = tensor.dim %[[VAL_1]], %[[VAL_2]] : tensor> +// CHECK-OFF: %[[VAL_5:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 1 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_6:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-OFF: %[[VAL_7:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-OFF: %[[VAL_8:.*]] = memref.load %[[VAL_7]][] : memref +// CHECK-OFF: %[[VAL_9:.*]] = scf.for %[[VAL_10:.*]] = %[[VAL_2]] to %[[VAL_4]] step %[[VAL_3]] iter_args(%[[VAL_11:.*]] = %[[VAL_8]]) -> (f64) { +// CHECK-OFF: %[[VAL_12:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_10]]] : memref +// CHECK-OFF: %[[VAL_13:.*]] = arith.addi %[[VAL_10]], %[[VAL_3]] : index +// CHECK-OFF: %[[VAL_14:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_13]]] : memref +// CHECK-OFF: %[[VAL_15:.*]] = scf.for %[[VAL_16:.*]] = %[[VAL_12]] to %[[VAL_14]] step %[[VAL_3]] iter_args(%[[VAL_17:.*]] = %[[VAL_11]]) -> (f64) { +// CHECK-OFF: %[[VAL_18:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_16]]] : memref +// CHECK-OFF: %[[VAL_19:.*]] = arith.mulf %[[VAL_17]], %[[VAL_18]] : f64 +// CHECK-OFF: scf.yield %[[VAL_19]] : f64 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: scf.yield %[[VAL_20:.*]] : f64 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: memref.store %[[VAL_21:.*]], %[[VAL_7]][] : memref +// CHECK-OFF: %[[VAL_22:.*]] = bufferization.to_tensor %[[VAL_7]] : memref +// CHECK-OFF: return %[[VAL_22]] : tensor +// CHECK-OFF: } + +#SparseVector = #sparse_tensor.encoding<{dimLevelType = ["dense","compressed"]}> + +#trait = { + indexing_maps = [ + affine_map<(i,j) -> (i,j)>, // a (in) + affine_map<(i,j) -> ()> // x (out) + ], + iterator_types = ["reduction", "reduction"] +} + +func.func @sparse_product_reduction_dense_sparse(%argx: tensor, + %arga: tensor) + -> tensor { + %0 = linalg.generic #trait + ins(%arga: tensor) + outs(%argx: tensor) { + ^bb(%a: f64, %x: f64): + %t = arith.mulf %x, %a: f64 + linalg.yield %t : f64 + } -> tensor + return %0 : tensor +} + +// ----- + +// Same as sparse_product_reduction_dense_sparse but with the outer loop being sparse. +// +// CHECK-ON-LABEL: func.func @sparse_product_reduction_sparse_sparse( +// CHECK-ON-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-ON-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-ON-DAG: %[[VAL_2:.*]] = arith.constant 8 : index +// CHECK-ON-DAG: %[[VAL_3:.*]] = arith.constant dense<1.000000e+00> : vector<8xf64> +// CHECK-ON-DAG: %[[VAL_4:.*]] = arith.constant dense<0.000000e+00> : vector<8xf64> +// CHECK-ON-DAG: %[[VAL_5:.*]] = arith.constant 0 : index +// CHECK-ON-DAG: %[[VAL_6:.*]] = arith.constant 1 : index +// CHECK-ON: %[[VAL_7:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-ON: %[[VAL_8:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 1 : index} : tensor> to memref +// CHECK-ON: %[[VAL_9:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-ON: %[[VAL_10:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-ON: %[[VAL_11:.*]] = memref.load %[[VAL_10]][] : memref +// CHECK-ON: %[[VAL_12:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_5]]] : memref +// CHECK-ON: %[[VAL_13:.*]] = memref.load %[[VAL_7]]{{\[}}%[[VAL_6]]] : memref +// CHECK-ON: %[[VAL_14:.*]] = scf.for %[[VAL_15:.*]] = %[[VAL_12]] to %[[VAL_13]] step %[[VAL_6]] iter_args(%[[VAL_16:.*]] = %[[VAL_11]]) -> (f64) { +// CHECK-ON: %[[VAL_17:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_15]]] : memref +// CHECK-ON: %[[VAL_18:.*]] = arith.addi %[[VAL_15]], %[[VAL_6]] : index +// CHECK-ON: %[[VAL_19:.*]] = memref.load %[[VAL_8]]{{\[}}%[[VAL_18]]] : memref +// CHECK-ON: %[[VAL_20:.*]] = vector.insertelement %[[VAL_16]], %[[VAL_3]]{{\[}}%[[VAL_5]] : index] : vector<8xf64> +// CHECK-ON: %[[VAL_21:.*]] = scf.for %[[VAL_22:.*]] = %[[VAL_17]] to %[[VAL_19]] step %[[VAL_2]] iter_args(%[[VAL_23:.*]] = %[[VAL_20]]) -> (vector<8xf64>) { +// CHECK-ON: %[[VAL_24:.*]] = affine.min #map(%[[VAL_19]], %[[VAL_22]]){{\[}}%[[VAL_2]]] +// CHECK-ON: %[[VAL_25:.*]] = vector.create_mask %[[VAL_24]] : vector<8xi1> +// CHECK-ON: %[[VAL_26:.*]] = vector.maskedload %[[VAL_9]]{{\[}}%[[VAL_22]]], %[[VAL_25]], %[[VAL_4]] : memref, vector<8xi1>, vector<8xf64> into vector<8xf64> +// CHECK-ON: %[[VAL_27:.*]] = arith.mulf %[[VAL_23]], %[[VAL_26]] : vector<8xf64> +// CHECK-ON: %[[VAL_28:.*]] = arith.select %[[VAL_25]], %[[VAL_27]], %[[VAL_23]] : vector<8xi1>, vector<8xf64> +// CHECK-ON: scf.yield %[[VAL_28]] : vector<8xf64> +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: %[[VAL_29:.*]] = vector.reduction , %[[VAL_30:.*]] : vector<8xf64> into f64 +// CHECK-ON: scf.yield %[[VAL_29]] : f64 +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: memref.store %[[VAL_31:.*]], %[[VAL_10]][] : memref +// CHECK-ON: %[[VAL_32:.*]] = bufferization.to_tensor %[[VAL_10]] : memref +// CHECK-ON: return %[[VAL_32]] : tensor +// CHECK-ON: } +// +// CHECK-OFF-LABEL: func.func @sparse_product_reduction_sparse_sparse( +// CHECK-OFF-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-OFF-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-OFF-DAG: %[[VAL_2:.*]] = arith.constant 0 : index +// CHECK-OFF-DAG: %[[VAL_3:.*]] = arith.constant 1 : index +// CHECK-OFF: %[[VAL_4:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_5:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 1 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_6:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-OFF: %[[VAL_7:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-OFF: %[[VAL_8:.*]] = memref.load %[[VAL_7]][] : memref +// CHECK-OFF: %[[VAL_9:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_2]]] : memref +// CHECK-OFF: %[[VAL_10:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_3]]] : memref +// CHECK-OFF: %[[VAL_11:.*]] = scf.for %[[VAL_12:.*]] = %[[VAL_9]] to %[[VAL_10]] step %[[VAL_3]] iter_args(%[[VAL_13:.*]] = %[[VAL_8]]) -> (f64) { +// CHECK-OFF: %[[VAL_14:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_12]]] : memref +// CHECK-OFF: %[[VAL_15:.*]] = arith.addi %[[VAL_12]], %[[VAL_3]] : index +// CHECK-OFF: %[[VAL_16:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_15]]] : memref +// CHECK-OFF: %[[VAL_17:.*]] = scf.for %[[VAL_18:.*]] = %[[VAL_14]] to %[[VAL_16]] step %[[VAL_3]] iter_args(%[[VAL_19:.*]] = %[[VAL_13]]) -> (f64) { +// CHECK-OFF: %[[VAL_20:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_18]]] : memref +// CHECK-OFF: %[[VAL_21:.*]] = arith.mulf %[[VAL_19]], %[[VAL_20]] : f64 +// CHECK-OFF: scf.yield %[[VAL_21]] : f64 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: scf.yield %[[VAL_22:.*]] : f64 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: memref.store %[[VAL_23:.*]], %[[VAL_7]][] : memref +// CHECK-OFF: %[[VAL_24:.*]] = bufferization.to_tensor %[[VAL_7]] : memref +// CHECK-OFF: return %[[VAL_24]] : tensor +// CHECK-OFF: } +#SparseVector = #sparse_tensor.encoding<{dimLevelType = ["compressed","compressed"]}> + +#trait = { + indexing_maps = [ + affine_map<(i,j) -> (i,j)>, // a (in) + affine_map<(i,j) -> ()> // x (out) + ], + iterator_types = ["reduction", "reduction"] +} + +func.func @sparse_product_reduction_sparse_sparse(%argx: tensor, + %arga: tensor) + -> tensor { + %0 = linalg.generic #trait + ins(%arga: tensor) + outs(%argx: tensor) { + ^bb(%a: f64, %x: f64): + %t = arith.mulf %x, %a: f64 + linalg.yield %t : f64 + } -> tensor + return %0 : tensor +} + +// ----- + +// sparse_product_reduction_dense_sparse and +// sparse_product_reduction_sparse_sparse established that the outer loop +// doesn't matter for vectorization. +// As a result from this point forward, use tensors with fewer dimensions. + +// Check that we vectorize reductions with ori. +// Note: The weird element type here is to check that we create the right +// constant type for the pass-through value. +// CHECK-ON-LABEL: func.func @sparse_reduction_ori( +// CHECK-ON-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-ON-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-ON-DAG: %[[VAL_2:.*]] = arith.constant 8 : index +// CHECK-ON-DAG: %[[VAL_3:.*]] = arith.constant dense<0> : vector<8xi13> +// CHECK-ON-DAG: %[[VAL_4:.*]] = arith.constant 0 : index +// CHECK-ON-DAG: %[[VAL_5:.*]] = arith.constant 1 : index +// CHECK-ON: %[[VAL_6:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-ON: %[[VAL_7:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-ON: %[[VAL_8:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-ON: %[[VAL_9:.*]] = memref.load %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_10:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_4]]] : memref +// CHECK-ON: %[[VAL_11:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_5]]] : memref +// CHECK-ON: %[[VAL_12:.*]] = vector.broadcast %[[VAL_9]] : i13 to vector<8xi13> +// CHECK-ON: %[[VAL_13:.*]] = scf.for %[[VAL_14:.*]] = %[[VAL_10]] to %[[VAL_11]] step %[[VAL_2]] iter_args(%[[VAL_15:.*]] = %[[VAL_12]]) -> (vector<8xi13>) { +// CHECK-ON: %[[VAL_16:.*]] = affine.min #map(%[[VAL_11]], %[[VAL_14]]){{\[}}%[[VAL_2]]] +// CHECK-ON: %[[VAL_17:.*]] = vector.create_mask %[[VAL_16]] : vector<8xi1> +// CHECK-ON: %[[VAL_18:.*]] = vector.maskedload %[[VAL_7]]{{\[}}%[[VAL_14]]], %[[VAL_17]], %[[VAL_3]] : memref, vector<8xi1>, vector<8xi13> into vector<8xi13> +// CHECK-ON: %[[VAL_19:.*]] = arith.ori %[[VAL_15]], %[[VAL_18]] : vector<8xi13> +// CHECK-ON: %[[VAL_20:.*]] = arith.select %[[VAL_17]], %[[VAL_19]], %[[VAL_15]] : vector<8xi1>, vector<8xi13> +// CHECK-ON: scf.yield %[[VAL_20]] : vector<8xi13> +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: %[[VAL_21:.*]] = vector.reduction , %[[VAL_22:.*]] : vector<8xi13> into i13 +// CHECK-ON: memref.store %[[VAL_21]], %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_23:.*]] = bufferization.to_tensor %[[VAL_8]] : memref +// CHECK-ON: return %[[VAL_23]] : tensor +// CHECK-ON: } +// +// CHECK-OFF-LABEL: func.func @sparse_reduction_ori( +// CHECK-OFF-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-OFF-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-OFF-DAG: %[[VAL_2:.*]] = arith.constant 0 : index +// CHECK-OFF-DAG: %[[VAL_3:.*]] = arith.constant 1 : index +// CHECK-OFF: %[[VAL_4:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_5:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-OFF: %[[VAL_6:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-OFF: %[[VAL_7:.*]] = memref.load %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_8:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_2]]] : memref +// CHECK-OFF: %[[VAL_9:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_3]]] : memref +// CHECK-OFF: %[[VAL_10:.*]] = scf.for %[[VAL_11:.*]] = %[[VAL_8]] to %[[VAL_9]] step %[[VAL_3]] iter_args(%[[VAL_12:.*]] = %[[VAL_7]]) -> (i13) { +// CHECK-OFF: %[[VAL_13:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_11]]] : memref +// CHECK-OFF: %[[VAL_14:.*]] = arith.ori %[[VAL_12]], %[[VAL_13]] : i13 +// CHECK-OFF: scf.yield %[[VAL_14]] : i13 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: memref.store %[[VAL_15:.*]], %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_16:.*]] = bufferization.to_tensor %[[VAL_6]] : memref +// CHECK-OFF: return %[[VAL_16]] : tensor +// CHECK-OFF: } +#SparseVector = #sparse_tensor.encoding<{dimLevelType = ["compressed"]}> + +#trait = { + indexing_maps = [ + affine_map<(i) -> (i)>, // a (in) + affine_map<(i) -> ()> // x (out) + ], + iterator_types = ["reduction"] +} + +func.func @sparse_reduction_ori(%argx: tensor, + %arga: tensor) + -> tensor { + %0 = linalg.generic #trait + ins(%arga: tensor) + outs(%argx: tensor) { + ^bb(%a: i13, %x: i13): + %t = arith.ori %x, %a: i13 + linalg.yield %t : i13 + } -> tensor + return %0 : tensor +} + +// ----- + +// Same test as sparse_reduction_ori except that the accumulator is on the +// rhs of the operation. +// This checks that we can recognize a reduction irrespective to where the +// accumalator appears on commutative operations. + +// CHECK-ON-LABEL: func.func @sparse_reduction_ori_accumulator_on_rhs( +// CHECK-ON-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-ON-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-ON-DAG: %[[VAL_2:.*]] = arith.constant 8 : index +// CHECK-ON-DAG: %[[VAL_3:.*]] = arith.constant dense<0> : vector<8xi13> +// CHECK-ON-DAG: %[[VAL_4:.*]] = arith.constant 0 : index +// CHECK-ON-DAG: %[[VAL_5:.*]] = arith.constant 1 : index +// CHECK-ON: %[[VAL_6:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-ON: %[[VAL_7:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-ON: %[[VAL_8:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-ON: %[[VAL_9:.*]] = memref.load %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_10:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_4]]] : memref +// CHECK-ON: %[[VAL_11:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_5]]] : memref +// CHECK-ON: %[[VAL_12:.*]] = vector.broadcast %[[VAL_9]] : i13 to vector<8xi13> +// CHECK-ON: %[[VAL_13:.*]] = scf.for %[[VAL_14:.*]] = %[[VAL_10]] to %[[VAL_11]] step %[[VAL_2]] iter_args(%[[VAL_15:.*]] = %[[VAL_12]]) -> (vector<8xi13>) { +// CHECK-ON: %[[VAL_16:.*]] = affine.min #map(%[[VAL_11]], %[[VAL_14]]){{\[}}%[[VAL_2]]] +// CHECK-ON: %[[VAL_17:.*]] = vector.create_mask %[[VAL_16]] : vector<8xi1> +// CHECK-ON: %[[VAL_18:.*]] = vector.maskedload %[[VAL_7]]{{\[}}%[[VAL_14]]], %[[VAL_17]], %[[VAL_3]] : memref, vector<8xi1>, vector<8xi13> into vector<8xi13> +// CHECK-ON: %[[VAL_19:.*]] = arith.ori %[[VAL_18]], %[[VAL_15]] : vector<8xi13> +// CHECK-ON: %[[VAL_20:.*]] = arith.select %[[VAL_17]], %[[VAL_19]], %[[VAL_15]] : vector<8xi1>, vector<8xi13> +// CHECK-ON: scf.yield %[[VAL_20]] : vector<8xi13> +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: %[[VAL_21:.*]] = vector.reduction , %[[VAL_22:.*]] : vector<8xi13> into i13 +// CHECK-ON: memref.store %[[VAL_21]], %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_23:.*]] = bufferization.to_tensor %[[VAL_8]] : memref +// CHECK-ON: return %[[VAL_23]] : tensor +// CHECK-ON: } +// +// CHECK-OFF-LABEL: func.func @sparse_reduction_ori_accumulator_on_rhs( +// CHECK-OFF-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-OFF-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-OFF-DAG: %[[VAL_2:.*]] = arith.constant 0 : index +// CHECK-OFF-DAG: %[[VAL_3:.*]] = arith.constant 1 : index +// CHECK-OFF: %[[VAL_4:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_5:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-OFF: %[[VAL_6:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-OFF: %[[VAL_7:.*]] = memref.load %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_8:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_2]]] : memref +// CHECK-OFF: %[[VAL_9:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_3]]] : memref +// CHECK-OFF: %[[VAL_10:.*]] = scf.for %[[VAL_11:.*]] = %[[VAL_8]] to %[[VAL_9]] step %[[VAL_3]] iter_args(%[[VAL_12:.*]] = %[[VAL_7]]) -> (i13) { +// CHECK-OFF: %[[VAL_13:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_11]]] : memref +// CHECK-OFF: %[[VAL_14:.*]] = arith.ori %[[VAL_13]], %[[VAL_12]] : i13 +// CHECK-OFF: scf.yield %[[VAL_14]] : i13 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: memref.store %[[VAL_15:.*]], %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_16:.*]] = bufferization.to_tensor %[[VAL_6]] : memref +// CHECK-OFF: return %[[VAL_16]] : tensor +// CHECK-OFF: } +#SparseVector = #sparse_tensor.encoding<{dimLevelType = ["compressed"]}> + +#trait = { + indexing_maps = [ + affine_map<(i) -> (i)>, // a (in) + affine_map<(i) -> ()> // x (out) + ], + iterator_types = ["reduction"] +} + +func.func @sparse_reduction_ori_accumulator_on_rhs(%argx: tensor, + %arga: tensor) + -> tensor { + %0 = linalg.generic #trait + ins(%arga: tensor) + outs(%argx: tensor) { + ^bb(%a: i13, %x: i13): + %t = arith.ori %a, %x: i13 + linalg.yield %t : i13 + } -> tensor + return %0 : tensor +} + +// ----- + +// Check that we vectorize reduction with subi. +// +// CHECK-ON-LABEL: func.func @sparse_reduction_subi( +// CHECK-ON-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-ON-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-ON-DAG: %[[VAL_2:.*]] = arith.constant 8 : index +// CHECK-ON-DAG: %[[VAL_3:.*]] = arith.constant 0 : index +// CHECK-ON-DAG: %[[VAL_4:.*]] = arith.constant dense<0> : vector<8xi32> +// CHECK-ON-DAG: %[[VAL_5:.*]] = arith.constant 1 : index +// CHECK-ON: %[[VAL_6:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-ON: %[[VAL_7:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-ON: %[[VAL_8:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-ON: %[[VAL_9:.*]] = memref.load %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_10:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_3]]] : memref +// CHECK-ON: %[[VAL_11:.*]] = memref.load %[[VAL_6]]{{\[}}%[[VAL_5]]] : memref +// CHECK-ON: %[[VAL_12:.*]] = vector.insertelement %[[VAL_9]], %[[VAL_4]]{{\[}}%[[VAL_3]] : index] : vector<8xi32> +// CHECK-ON: %[[VAL_13:.*]] = scf.for %[[VAL_14:.*]] = %[[VAL_10]] to %[[VAL_11]] step %[[VAL_2]] iter_args(%[[VAL_15:.*]] = %[[VAL_12]]) -> (vector<8xi32>) { +// CHECK-ON: %[[VAL_16:.*]] = affine.min #map(%[[VAL_11]], %[[VAL_14]]){{\[}}%[[VAL_2]]] +// CHECK-ON: %[[VAL_17:.*]] = vector.create_mask %[[VAL_16]] : vector<8xi1> +// CHECK-ON: %[[VAL_18:.*]] = vector.maskedload %[[VAL_7]]{{\[}}%[[VAL_14]]], %[[VAL_17]], %[[VAL_4]] : memref, vector<8xi1>, vector<8xi32> into vector<8xi32> +// CHECK-ON: %[[VAL_19:.*]] = arith.subi %[[VAL_15]], %[[VAL_18]] : vector<8xi32> +// CHECK-ON: %[[VAL_20:.*]] = arith.select %[[VAL_17]], %[[VAL_19]], %[[VAL_15]] : vector<8xi1>, vector<8xi32> +// CHECK-ON: scf.yield %[[VAL_20]] : vector<8xi32> +// CHECK-ON: } {"Emitted from" = "linalg.generic"} +// CHECK-ON: %[[VAL_21:.*]] = vector.reduction , %[[VAL_22:.*]] : vector<8xi32> into i32 +// CHECK-ON: memref.store %[[VAL_21]], %[[VAL_8]][] : memref +// CHECK-ON: %[[VAL_23:.*]] = bufferization.to_tensor %[[VAL_8]] : memref +// CHECK-ON: return %[[VAL_23]] : tensor +// CHECK-ON: } +// +// CHECK-OFF-LABEL: func.func @sparse_reduction_subi( +// CHECK-OFF-SAME: %[[VAL_0:.*]]: tensor, +// CHECK-OFF-SAME: %[[VAL_1:.*]]: tensor>) -> tensor { +// CHECK-OFF-DAG: %[[VAL_2:.*]] = arith.constant 0 : index +// CHECK-OFF-DAG: %[[VAL_3:.*]] = arith.constant 1 : index +// CHECK-OFF: %[[VAL_4:.*]] = sparse_tensor.pointers %[[VAL_1]] {dimension = 0 : index} : tensor> to memref +// CHECK-OFF: %[[VAL_5:.*]] = sparse_tensor.values %[[VAL_1]] : tensor> to memref +// CHECK-OFF: %[[VAL_6:.*]] = bufferization.to_memref %[[VAL_0]] : memref +// CHECK-OFF: %[[VAL_7:.*]] = memref.load %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_8:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_2]]] : memref +// CHECK-OFF: %[[VAL_9:.*]] = memref.load %[[VAL_4]]{{\[}}%[[VAL_3]]] : memref +// CHECK-OFF: %[[VAL_10:.*]] = scf.for %[[VAL_11:.*]] = %[[VAL_8]] to %[[VAL_9]] step %[[VAL_3]] iter_args(%[[VAL_12:.*]] = %[[VAL_7]]) -> (i32) { +// CHECK-OFF: %[[VAL_13:.*]] = memref.load %[[VAL_5]]{{\[}}%[[VAL_11]]] : memref +// CHECK-OFF: %[[VAL_14:.*]] = arith.subi %[[VAL_12]], %[[VAL_13]] : i32 +// CHECK-OFF: scf.yield %[[VAL_14]] : i32 +// CHECK-OFF: } {"Emitted from" = "linalg.generic"} +// CHECK-OFF: memref.store %[[VAL_15:.*]], %[[VAL_6]][] : memref +// CHECK-OFF: %[[VAL_16:.*]] = bufferization.to_tensor %[[VAL_6]] : memref +// CHECK-OFF: return %[[VAL_16]] : tensor +// CHECK-OFF: } +#SparseVector = #sparse_tensor.encoding<{dimLevelType = ["compressed"]}> + +#trait = { + indexing_maps = [ + affine_map<(i) -> (i)>, // a (in) + affine_map<(i) -> ()> // x (out) + ], + iterator_types = ["reduction"] +} + +func.func @sparse_reduction_subi(%argx: tensor, + %arga: tensor) + -> tensor { + %0 = linalg.generic #trait + ins(%arga: tensor) + outs(%argx: tensor) { + ^bb(%a: i32, %x: i32): + %t = arith.subi %x, %a: i32 + linalg.yield %t : i32 + } -> tensor + return %0 : tensor +} + +// ----- -- 2.7.4