diff --git a/mlir/test/Transforms/loop-fusion-4.mlir b/mlir/test/Transforms/loop-fusion-4.mlir
--- a/mlir/test/Transforms/loop-fusion-4.mlir
+++ b/mlir/test/Transforms/loop-fusion-4.mlir
@@ -1,5 +1,6 @@
 // RUN: mlir-opt -allow-unregistered-dialect %s -affine-loop-fusion="mode=producer" -split-input-file | FileCheck %s --check-prefix=PRODUCER-CONSUMER
 // RUN: mlir-opt -allow-unregistered-dialect %s -affine-loop-fusion="fusion-maximal mode=sibling" -split-input-file | FileCheck %s --check-prefix=SIBLING-MAXIMAL
+// RUN: mlir-opt -allow-unregistered-dialect %s -affine-loop-fusion="fusion-maximal mode=producer" -split-input-file | FileCheck %s --check-prefix=PRODUCER-CONSUMER-MAXIMAL
 
 // Part I of fusion tests in  mlir/test/Transforms/loop-fusion.mlir.
 // Part II of fusion tests in mlir/test/Transforms/loop-fusion-2.mlir
@@ -141,3 +142,73 @@
 // SIBLING-MAXIMAL-NEXT:             affine.for %[[idx_1:.*]] = 0 to 64 {
 // SIBLING-MAXIMAL-NEXT:               %[[result_1:.*]] = affine.for %[[idx_2:.*]] = 0 to 32 iter_args(%[[iter_0:.*]] = %[[cst_1]]) -> (f32) {
 // SIBLING-MAXIMAL-NEXT:                 %[[result_0:.*]] = affine.for %[[idx_3:.*]] = 0 to 64 iter_args(%[[iter_1:.*]] = %[[cst_0]]) -> (f32) {
+
+// -----
+
+// Source loop nest %i1 is a reduction but due to fusion of preceding
+// preceding loop %i0 generates a producer for %i3. Check if the 
+// producer fusion happens at depth 2 with original bounds.
+// PRODUCER-CONSUMER-MAXIMAL-LABEL:   func @reduction_producer_consumer(
+func.func @reduction_producer_consumer(%arg0: memref<1024xf32, 1>, %arg1: memref<1xf32, 1>, %arg2: memref<1xf32, 1>, %arg3: memref<1xf32, 1>) {
+  %cst = arith.constant 0.000000e+00 : f32
+  %0 = memref.alloc() : memref<f32, 1>
+  %1 = memref.alloc() : memref<f32, 1>
+  %2 = memref.alloc() : memref<1024xf32, 1>
+  affine.for %i0 = 0 to 1024 {
+    %4 = affine.load %arg0[%i0] : memref<1024xf32, 1>
+    %6 = arith.addf %4, %4 : f32
+    affine.store %6, %2[%i0] : memref<1024xf32, 1>
+  }
+  affine.for %i1 = 0 to 1 {
+    affine.store %cst, %1[] : memref<f32, 1>
+    affine.for %i2 = 0 to 1024 {
+      %5 = affine.load %1[] : memref<f32, 1>
+      %6 = affine.load %2[%i2] : memref<1024xf32, 1>
+      %7 = arith.addf %5, %6 : f32
+      affine.store %7, %1[] : memref<f32, 1>
+    }
+    %4 = affine.load %1[] : memref<f32, 1>
+    affine.store %4, %arg2[%i1] : memref<1xf32, 1>
+  }
+  affine.for %i3 = 0 to 1 {
+    affine.store %cst, %0[] : memref<f32, 1>
+    affine.for %i4 = 0 to 1024 {
+      %5 = affine.load %0[] : memref<f32, 1>
+      %6 = affine.load %2[%i4] : memref<1024xf32, 1>
+      %7 = arith.addf %5, %6 : f32
+      affine.store %7, %0[] : memref<f32, 1>
+    }
+    %4 = affine.load %0[] : memref<f32, 1>
+    affine.store %4, %arg3[%i3] : memref<1xf32, 1>
+  }
+  return
+}
+// PRODUCER-CONSUMER-MAXIMAL:           %[[VAL_4:.*]] = memref.alloc() : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:           %[[VAL_5:.*]] = arith.constant 0 : index
+// PRODUCER-CONSUMER-MAXIMAL:           %[[VAL_6:.*]] = arith.constant 0.000000e+00 : f32
+// PRODUCER-CONSUMER-MAXIMAL:           %[[VAL_7:.*]] = memref.alloc() : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:           %[[VAL_8:.*]] = memref.alloc() : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:           affine.for %[[VAL_9:.*]] = 0 to 1 {
+// PRODUCER-CONSUMER-MAXIMAL:             affine.store %[[VAL_6]], %[[VAL_7]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:             affine.for %[[VAL_10:.*]] = 0 to 1024 {
+// PRODUCER-CONSUMER-MAXIMAL:               affine.store %[[VAL_6]], %[[VAL_8]][] : memref<f32, 1>
+//  Incorrect bounds for the innermost loop.
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_11:.*]] = affine.load %{{.*}}{{\[}}%[[VAL_10]]] : memref<1024xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_12:.*]] = arith.addf %[[VAL_11]], %[[VAL_11]] : f32
+// PRODUCER-CONSUMER-MAXIMAL:               affine.store %[[VAL_12]], %[[VAL_4]][0] : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_13:.*]] = affine.load %[[VAL_8]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_14:.*]] = affine.load %[[VAL_4]][0] : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_15:.*]] = arith.addf %[[VAL_13]], %[[VAL_14]] : f32
+// PRODUCER-CONSUMER-MAXIMAL:               affine.store %[[VAL_15]], %[[VAL_8]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_16:.*]] = affine.load %[[VAL_8]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               affine.store %[[VAL_16]], %{{.*}}{{\[}}%[[VAL_5]]] : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_17:.*]] = affine.load %[[VAL_7]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_18:.*]] = affine.load %[[VAL_4]][0] : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:               %[[VAL_19:.*]] = arith.addf %[[VAL_17]], %[[VAL_18]] : f32
+// PRODUCER-CONSUMER-MAXIMAL:               affine.store %[[VAL_19]], %[[VAL_7]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:             }
+// PRODUCER-CONSUMER-MAXIMAL:             %[[VAL_20:.*]] = affine.load %[[VAL_7]][] : memref<f32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:             affine.store %[[VAL_20]], %{{.*}}{{\[}}%[[VAL_9]]] : memref<1xf32, 1>
+// PRODUCER-CONSUMER-MAXIMAL:           }
+// PRODUCER-CONSUMER-MAXIMAL:           return
+// PRODUCER-CONSUMER-MAXIMAL:         }