[mlir][linalg, tosa] Fix memory leaks in integration tests (#85366)

Buffers are no longer deallocation by One-Shot Bufferize. This is now
done by a separate buffer deallocation pass.

Also fix a bug in the `vector.mask` folding, which was triggered by
`-buffer-deallocation-pipeline`, which runs the canonicalizer.

Author: matthias-springer

mlir/lib/Dialect/Vector/IR/VectorOps.cpp

@@ -6064,7 +6064,7 @@ LogicalResult MaskOp::fold(FoldAdaptor adaptor,
   maskableOp->dropAllUses();
   maskableOp->moveBefore(getOperation());
 
-  results.push_back(maskableOp->getResult(0));
+  llvm::append_range(results, maskableOp->getResults());
   return success();
 }
 

mlir/test/Dialect/Vector/canonicalize.mlir

@@ -2483,6 +2483,18 @@ func.func @all_true_vector_mask(%a : vector<3x4xf32>) -> vector<3x4xf32> {
 
 // -----
 
+// CHECK-LABEL: func @all_true_vector_mask_no_result(
+func.func @all_true_vector_mask_no_result(%a : vector<3x4xf32>, %m : memref<3x4xf32>) {
+//   CHECK-NOT:   vector.mask
+//       CHECK:   vector.transfer_write
+  %c0 = arith.constant 0 : index
+  %all_true = vector.constant_mask [3, 4] : vector<3x4xi1>
+  vector.mask %all_true { vector.transfer_write %a, %m[%c0, %c0] : vector<3x4xf32>, memref<3x4xf32> } : vector<3x4xi1>
+  return
+}
+
+// -----
+
 // CHECK-LABEL:   func.func @fold_shape_cast_with_mask(
 // CHECK-SAME:     %[[VAL_0:.*]]: tensor<1x?xf32>) -> vector<1x4xi1> {
 func.func @fold_shape_cast_with_mask(%arg0: tensor<1x?xf32>) -> vector<1x4xi1> {

mlir/test/Integration/Dialect/Linalg/CPU/mmt4d.mlir

@@ -1,6 +1,6 @@
 // DEFINE: %{compile} =  mlir-opt %s \
 // DEFINE:    -transform-interpreter -test-transform-dialect-erase-schedule \
-// DEFINE:    -one-shot-bufferize -func-bufferize -cse -canonicalize -convert-vector-to-scf -test-lower-to-llvm -o %t
+// DEFINE:    -one-shot-bufferize="bufferize-function-boundaries" -buffer-deallocation-pipeline -cse -canonicalize -convert-vector-to-scf -test-lower-to-llvm -o %t
 // DEFINE: %{entry_point} = mmt4d
 // DEFINE: %{run} = mlir-cpu-runner %t -e %{entry_point} -entry-point-result=void \
 // DEFINE:    -shared-libs=%mlir_runner_utils,%mlir_c_runner_utils

mlir/test/Integration/Dialect/Linalg/CPU/test-matmul-masked-vec.mlir

@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s -transform-interpreter -test-transform-dialect-erase-schedule -one-shot-bufferize -func-bufferize -lower-vector-mask --test-lower-to-llvm | \
+// RUN: mlir-opt %s -transform-interpreter -test-transform-dialect-erase-schedule -one-shot-bufferize="bufferize-function-boundaries" -buffer-deallocation-pipeline -lower-vector-mask --test-lower-to-llvm | \
 // RUN: mlir-cpu-runner -e main -entry-point-result=void --shared-libs=%mlir_c_runner_utils,%mlir_runner_utils | \
 // RUN: FileCheck %s
 

mlir/test/Integration/Dialect/Tosa/CPU/test-fully-connected.mlir

@@ -1,5 +1,5 @@
 // RUN: mlir-opt %s -pass-pipeline="builtin.module(func.func(tosa-to-linalg-named,tosa-to-linalg,tosa-to-arith))" | \
-// RUN: mlir-opt -one-shot-bufferize -func-bufferize -test-lower-to-llvm | \
+// RUN: mlir-opt -one-shot-bufferize="bufferize-function-boundaries" -buffer-deallocation-pipeline -test-lower-to-llvm | \
 // RUN: mlir-cpu-runner -O3 -e main -entry-point-result=void \
 // RUN:   -shared-libs=%mlir_runner_utils \
 // RUN: | FileCheck %s