diff --git a/llvm/lib/CodeGen/MachineScheduler.cpp b/llvm/lib/CodeGen/MachineScheduler.cpp
--- a/llvm/lib/CodeGen/MachineScheduler.cpp
+++ b/llvm/lib/CodeGen/MachineScheduler.cpp
@@ -1609,7 +1609,7 @@
 
     unsigned ChainPredID = DAG->SUnits.size();
     for (const SDep &Pred : SU.Preds) {
-      if (Pred.isCtrl()) {
+      if (Pred.isCtrl() && !Pred.isArtificial()) {
         ChainPredID = Pred.getSUnit()->NodeNum;
         break;
       }
diff --git a/llvm/test/CodeGen/AMDGPU/cluster_stores.ll b/llvm/test/CodeGen/AMDGPU/cluster_stores.ll
new file mode 100644
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/cluster_stores.ll
@@ -0,0 +1,27 @@
+; RUN: llc -march=amdgcn -mcpu=gfx900 -debug-only=machine-scheduler -o /dev/null %s 2>&1 | FileCheck %s
+; REQUIRES: asserts
+
+define amdgpu_kernel void @f(i32* noalias %lb, i32* noalias %sb) {
+bb:
+; CHECK-COUNT-3: Cluster ld/st
+  %la0 = getelementptr inbounds i32, i32* %lb, i32 0
+  %ld0 = load i32, i32* %la0
+  %la1 = getelementptr inbounds i32, i32* %lb, i32 2
+  %ld1 = load i32, i32* %la1
+  %la2 = getelementptr inbounds i32, i32* %lb, i32 4
+  %ld2 = load i32, i32* %la2
+  %la3 = getelementptr inbounds i32, i32* %lb, i32 6
+  %ld3 = load i32, i32* %la3
+
+; CHECK-COUNT-3: Cluster ld/st
+  %sa0 = getelementptr inbounds i32, i32* %sb, i32 0
+  store i32 %ld0, i32* %sa0
+  %sa1 = getelementptr inbounds i32, i32* %sb, i32 2
+  store i32 %ld1, i32* %sa1
+  %sa2 = getelementptr inbounds i32, i32* %sb, i32 4
+  store i32 %ld2, i32* %sa2
+  %sa3 = getelementptr inbounds i32, i32* %sb, i32 6
+  store i32 %ld3, i32* %sa3
+
+  ret void
+}