Index: lib/CodeGen/IslNodeBuilder.cpp
===================================================================
--- lib/CodeGen/IslNodeBuilder.cpp
+++ lib/CodeGen/IslNodeBuilder.cpp
@@ -296,6 +296,7 @@
   for (const auto &I : IDToValue)
     Values.insert(I.second);
 
+  // NOTE: this is populated in IslNodeBuilder::addParameters
   for (const auto &I : OutsideLoopIterations)
     Values.insert(cast<SCEVUnknown>(I.second)->getValue());
 
Index: lib/CodeGen/PPCGCodeGeneration.cpp
===================================================================
--- lib/CodeGen/PPCGCodeGeneration.cpp
+++ lib/CodeGen/PPCGCodeGeneration.cpp
@@ -417,7 +417,7 @@
   ///          referenced by the kernel, and whose second element contains the
   ///          set of functions referenced by the kernel. All functions in the
   ///          second set satisfy isValidFunctionInKernel.
-  std::pair<SetVector<Value *>, SetVector<Function *>>
+  std::tuple<SetVector<Value *>, SetVector<Function *>, SetVector<const Loop *>>
   getReferencesInKernel(ppcg_kernel *Kernel);
 
   /// Compute the sizes of the execution grid for a given kernel.
@@ -1389,7 +1389,7 @@
   return SubtreeFunctions;
 }
 
-std::pair<SetVector<Value *>, SetVector<Function *>>
+std::tuple<SetVector<Value *>, SetVector<Function *>, SetVector<const Loop *>>
 GPUNodeBuilder::getReferencesInKernel(ppcg_kernel *Kernel) {
   SetVector<Value *> SubtreeValues;
   SetVector<const SCEV *> SCEVs;
@@ -1400,11 +1400,21 @@
   for (const auto &I : IDToValue)
     SubtreeValues.insert(I.second);
 
+  // NOTE: this is populated in IslNodeBuilder::addParameters
+  for (const auto &I : OutsideLoopIterations)
+    SubtreeValues.insert(cast<SCEVUnknown>(I.second)->getValue());
+
   isl_ast_node_foreach_descendant_top_down(
       Kernel->tree, collectReferencesInGPUStmt, &References);
 
-  for (const SCEV *Expr : SCEVs)
+  for (const SCEV *Expr : SCEVs) {
     findValues(Expr, SE, SubtreeValues);
+    findLoops(Expr, Loops);
+  }
+
+  Loops.remove_if([this](const Loop *L) {
+    return S.contains(L) || L->contains(S.getEntry());
+  });
 
   for (auto &SAI : S.arrays())
     SubtreeValues.remove(SAI->getBasePtr());
@@ -1451,7 +1461,7 @@
     else
       ReplacedValues.insert(It->second);
   }
-  return std::make_pair(ReplacedValues, ValidSubtreeFunctions);
+  return std::make_tuple(ReplacedValues, ValidSubtreeFunctions, Loops);
 }
 
 void GPUNodeBuilder::clearDominators(Function *F) {
@@ -1528,6 +1538,7 @@
 Value *
 GPUNodeBuilder::createLaunchParameters(ppcg_kernel *Kernel, Function *F,
                                        SetVector<Value *> SubtreeValues) {
+
   const int NumArgs = F->arg_size();
   std::vector<int> ArgSizes(NumArgs);
 
@@ -1695,7 +1706,9 @@
 
   SetVector<Value *> SubtreeValues;
   SetVector<Function *> SubtreeFunctions;
-  std::tie(SubtreeValues, SubtreeFunctions) = getReferencesInKernel(Kernel);
+  SetVector<const Loop *> Loops;
+  std::tie(SubtreeValues, SubtreeFunctions, Loops) =
+      getReferencesInKernel(Kernel);
 
   assert(Kernel->tree && "Device AST of kernel node is empty");
 
@@ -1705,8 +1718,6 @@
   BlockGenerator::AllocaMapTy HostScalarMap = ScalarMap;
   ScalarMap.clear();
 
-  SetVector<const Loop *> Loops;
-
   // Create for all loops we depend on values that contain the current loop
   // iteration. These values are necessary to generate code for SCEVs that
   // depend on such loops. As a result we need to pass them to the subfunction.
Index: test/GPGPU/test.ll
===================================================================
--- /dev/null
+++ test/GPGPU/test.ll
@@ -0,0 +1,61 @@
+; RUN: opt %loadPolly -polly-codegen-ppcg -polly-acc-fail-on-verify-module-failure \
+; RUN: -disable-output < %s
+
+; ModuleID = 'reduced.ll'
+source_filename = "bugpoint-output-cb5513b.bc"
+target datalayout = "e-p:64:64:64-S128-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f16:16:16-f32:32:32-f64:64:64-f128:128:128-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+target triple = "x86_64-unknown-linux-gnu"
+
+declare void @fn_to_fence(i32 *%val)
+
+
+; void f(int *arr, bool shouldcont) {
+;     for(int i = 0; ; i++) {
+;         for(int j = 0; j < 10; j++) {
+;             arr[j] = i;
+;         }
+;         fence(arr);
+;         if (!shouldcont) break;
+;     }
+; }
+
+
+; Function Attrs: nounwind uwtable
+define void @f(i32 *%arr, i1 %shouldcont) #1 {
+entry:
+  br label %for.init
+
+for.init:                                             ; preds = %for.end, %entry.split
+  %i = phi i32 [ %i.next, %for.end ], [ 0, %entry ]
+  br label %for2.body
+
+for2.body:                                             ; preds = %"65", %"64"
+  %j = phi i32 [ %j.next, %for2.body ], [ 0, %for.init ]
+  %j.sext = sext i32 %j to i64
+  %arr.slot = getelementptr i32, i32* %arr, i64 %j.sext
+  store i32 %i, i32* %arr.slot, align 4
+  %exitcond = icmp eq i32 %j, 10
+  %j.next = add i32 %j, 1
+  br i1 %exitcond, label %for2.body.fence, label %for2.body
+
+for2.body.fence:                                             ; preds = %"65"
+  call void @fn_to_fence(i32* %arr) #2
+  br i1 %shouldcont, label %for.end, label %exit
+for.end:                                             ; preds = %"69"
+  %i.next = add i32 %i, 1
+  br label %for.init
+
+exit:                                             ; preds = %"69"
+  ret void
+
+}
+
+; Function Attrs: nounwind
+declare noalias i8* @malloc(i64) #2
+
+declare void @used_fn(...)
+
+
+attributes #0 = { argmemonly nounwind }
+attributes #1 = { nounwind uwtable }
+attributes #2 = { nounwind }