[Mlir-commits] [mlir] [mlir][sparse_tensor] Fix memory leak in `sparse_index_dense.mlir` (PR #137454)

[Matthias Springer]

https://github.com/matthias-springer created https://github.com/llvm/llvm-project/pull/137454

None

>From 40176253d93bd46178677cfc690cc99a607b7c3f Mon Sep 17 00:00:00 2001
From: Matthias Springer <mspringer at nvidia.com>
Date: Sat, 26 Apr 2025 12:10:22 +0200
Subject: [PATCH] [mlir][sparse_tensor] Fix memory leak in
 `sparse_index_dense.mlir`

---
 .../SparseTensor/CPU/sparse_index_dense.mlir  | 41 +++++++++----------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_index_dense.mlir b/mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_index_dense.mlir
index 371b3f359f3bf..407c06077da9f 100644
--- a/mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_index_dense.mlir
+++ b/mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_index_dense.mlir
@@ -65,8 +65,9 @@ module {
   //
   // Kernel that uses index in the index notation (conjunction).
   //
-  func.func @sparse_index_1d_conj(%arga: tensor<8xi64, #SparseVector>,
-                                  %out: tensor<8xi64>) -> tensor<8xi64> {
+  func.func @sparse_index_1d_conj(%arga: tensor<8xi64, #SparseVector>)
+      -> tensor<8xi64> {
+    %out = tensor.empty() : tensor<8xi64>
     %r = linalg.generic #trait_1d
         ins(%arga: tensor<8xi64, #SparseVector>)
        outs(%out: tensor<8xi64>) {
@@ -82,8 +83,9 @@ module {
   //
   // Kernel that uses index in the index notation (disjunction).
   //
-  func.func @sparse_index_1d_disj(%arga: tensor<8xi64, #SparseVector>,
-                                  %out: tensor<8xi64>) -> tensor<8xi64> {
+  func.func @sparse_index_1d_disj(%arga: tensor<8xi64, #SparseVector>)
+      -> tensor<8xi64> {
+    %out = tensor.empty() : tensor<8xi64>
     %r = linalg.generic #trait_1d
         ins(%arga: tensor<8xi64, #SparseVector>)
        outs(%out: tensor<8xi64>) {
@@ -99,8 +101,9 @@ module {
   //
   // Kernel that uses indices in the index notation (conjunction).
   //
-  func.func @sparse_index_2d_conj(%arga: tensor<3x4xi64, #SparseMatrix>,
-                                  %out: tensor<3x4xi64>) -> tensor<3x4xi64> {
+  func.func @sparse_index_2d_conj(%arga: tensor<3x4xi64, #SparseMatrix>)
+      -> tensor<3x4xi64> {
+    %out = tensor.empty() : tensor<3x4xi64>
     %r = linalg.generic #trait_2d
         ins(%arga: tensor<3x4xi64, #SparseMatrix>)
        outs(%out: tensor<3x4xi64>) {
@@ -119,8 +122,9 @@ module {
   //
   // Kernel that uses indices in the index notation (disjunction).
   //
-  func.func @sparse_index_2d_disj(%arga: tensor<3x4xi64, #SparseMatrix>,
-                                  %out: tensor<3x4xi64>) -> tensor<3x4xi64> {
+  func.func @sparse_index_2d_disj(%arga: tensor<3x4xi64, #SparseMatrix>)
+      -> tensor<3x4xi64> {
+    %out = tensor.empty() : tensor<3x4xi64>
     %r = linalg.generic #trait_2d
         ins(%arga: tensor<3x4xi64, #SparseMatrix>)
        outs(%out: tensor<3x4xi64>) {
@@ -161,20 +165,15 @@ module {
                                   [ 1,  1,  3,  4 ] ]> : tensor<3x4xi64>
     %dm = sparse_tensor.convert %m2 : tensor<3x4xi64> to tensor<3x4xi64, #SparseMatrix>
 
-    // Setup out tensors.
-    // Note: Constants bufferize to read-only buffers.
-    %init_8 = tensor.empty() : tensor<8xi64>
-    %init_3_4 = tensor.empty() : tensor<3x4xi64>
-
     // Call the kernels.
-    %0 = call @sparse_index_1d_conj(%sv, %init_8) : (tensor<8xi64, #SparseVector>, tensor<8xi64>) -> tensor<8xi64>
-    %1 = call @sparse_index_1d_disj(%sv, %init_8) : (tensor<8xi64, #SparseVector>, tensor<8xi64>) -> tensor<8xi64>
-    %2 = call @sparse_index_1d_conj(%dv, %init_8) : (tensor<8xi64, #SparseVector>, tensor<8xi64>) -> tensor<8xi64>
-    %3 = call @sparse_index_1d_disj(%dv, %init_8) : (tensor<8xi64, #SparseVector>, tensor<8xi64>) -> tensor<8xi64>
-    %4 = call @sparse_index_2d_conj(%sm, %init_3_4) : (tensor<3x4xi64, #SparseMatrix>, tensor<3x4xi64>) -> tensor<3x4xi64>
-    %5 = call @sparse_index_2d_disj(%sm, %init_3_4) : (tensor<3x4xi64, #SparseMatrix>, tensor<3x4xi64>) -> tensor<3x4xi64>
-    %6 = call @sparse_index_2d_conj(%dm, %init_3_4) : (tensor<3x4xi64, #SparseMatrix>, tensor<3x4xi64>) -> tensor<3x4xi64>
-    %7 = call @sparse_index_2d_disj(%dm, %init_3_4) : (tensor<3x4xi64, #SparseMatrix>, tensor<3x4xi64>) -> tensor<3x4xi64>
+    %0 = call @sparse_index_1d_conj(%sv) : (tensor<8xi64, #SparseVector>) -> tensor<8xi64>
+    %1 = call @sparse_index_1d_disj(%sv) : (tensor<8xi64, #SparseVector>) -> tensor<8xi64>
+    %2 = call @sparse_index_1d_conj(%dv) : (tensor<8xi64, #SparseVector>) -> tensor<8xi64>
+    %3 = call @sparse_index_1d_disj(%dv) : (tensor<8xi64, #SparseVector>) -> tensor<8xi64>
+    %4 = call @sparse_index_2d_conj(%sm) : (tensor<3x4xi64, #SparseMatrix>) -> tensor<3x4xi64>
+    %5 = call @sparse_index_2d_disj(%sm) : (tensor<3x4xi64, #SparseMatrix>) -> tensor<3x4xi64>
+    %6 = call @sparse_index_2d_conj(%dm) : (tensor<3x4xi64, #SparseMatrix>) -> tensor<3x4xi64>
+    %7 = call @sparse_index_2d_disj(%dm) : (tensor<3x4xi64, #SparseMatrix>) -> tensor<3x4xi64>
 
     //
     // Verify result.