Index: polly/trunk/lib/CodeGen/PPCGCodeGeneration.cpp
===================================================================
--- polly/trunk/lib/CodeGen/PPCGCodeGeneration.cpp
+++ polly/trunk/lib/CodeGen/PPCGCodeGeneration.cpp
@@ -1591,7 +1591,15 @@
   const int NumArgs = F->arg_size();
   std::vector<int> ArgSizes(NumArgs);
 
-  Type *ArrayTy = ArrayType::get(Builder.getInt8PtrTy(), 2 * NumArgs);
+  // If we are using the OpenCL Runtime, we need to add the kernel argument
+  // sizes to the end of the launch-parameter list, so OpenCL can determine
+  // how big the respective kernel arguments are.
+  // Here we need to reserve adequate space for that.
+  Type *ArrayTy;
+  if (Runtime == GPURuntime::OpenCL)
+    ArrayTy = ArrayType::get(Builder.getInt8PtrTy(), 2 * NumArgs);
+  else
+    ArrayTy = ArrayType::get(Builder.getInt8PtrTy(), NumArgs);
 
   BasicBlock *EntryBlock =
       &Builder.GetInsertBlock()->getParent()->getEntryBlock();
@@ -1608,7 +1616,8 @@
     isl_id *Id = isl_space_get_tuple_id(Prog->array[i].space, isl_dim_set);
     const ScopArrayInfo *SAI = ScopArrayInfo::getFromId(isl::manage(Id));
 
-    ArgSizes[Index] = SAI->getElemSizeInBytes();
+    if (Runtime == GPURuntime::OpenCL)
+      ArgSizes[Index] = SAI->getElemSizeInBytes();
 
     Value *DevArray = nullptr;
     if (PollyManagedMemory) {
@@ -1663,7 +1672,8 @@
     Value *Val = IDToValue[Id];
     isl_id_free(Id);
 
-    ArgSizes[Index] = computeSizeInBytes(Val->getType());
+    if (Runtime == GPURuntime::OpenCL)
+      ArgSizes[Index] = computeSizeInBytes(Val->getType());
 
     Instruction *Param =
         new AllocaInst(Val->getType(), AddressSpace,
@@ -1683,7 +1693,8 @@
       Val = ValueMap[Val];
     isl_id_free(Id);
 
-    ArgSizes[Index] = computeSizeInBytes(Val->getType());
+    if (Runtime == GPURuntime::OpenCL)
+      ArgSizes[Index] = computeSizeInBytes(Val->getType());
 
     Instruction *Param =
         new AllocaInst(Val->getType(), AddressSpace,
@@ -1695,7 +1706,8 @@
   }
 
   for (auto Val : SubtreeValues) {
-    ArgSizes[Index] = computeSizeInBytes(Val->getType());
+    if (Runtime == GPURuntime::OpenCL)
+      ArgSizes[Index] = computeSizeInBytes(Val->getType());
 
     Instruction *Param =
         new AllocaInst(Val->getType(), AddressSpace,
@@ -1706,15 +1718,17 @@
     Index++;
   }
 
-  for (int i = 0; i < NumArgs; i++) {
-    Value *Val = ConstantInt::get(Builder.getInt32Ty(), ArgSizes[i]);
-    Instruction *Param =
-        new AllocaInst(Builder.getInt32Ty(), AddressSpace,
-                       Launch + "_param_size_" + std::to_string(i),
-                       EntryBlock->getTerminator());
-    Builder.CreateStore(Val, Param);
-    insertStoreParameter(Parameters, Param, Index);
-    Index++;
+  if (Runtime == GPURuntime::OpenCL) {
+    for (int i = 0; i < NumArgs; i++) {
+      Value *Val = ConstantInt::get(Builder.getInt32Ty(), ArgSizes[i]);
+      Instruction *Param =
+          new AllocaInst(Builder.getInt32Ty(), AddressSpace,
+                         Launch + "_param_size_" + std::to_string(i),
+                         EntryBlock->getTerminator());
+      Builder.CreateStore(Val, Param);
+      insertStoreParameter(Parameters, Param, Index);
+      Index++;
+    }
   }
 
   auto Location = EntryBlock->getTerminator();
Index: polly/trunk/test/GPGPU/cuda-managed-memory-simple.ll
===================================================================
--- polly/trunk/test/GPGPU/cuda-managed-memory-simple.ll
+++ polly/trunk/test/GPGPU/cuda-managed-memory-simple.ll
@@ -38,25 +38,17 @@
 ; CHECK:       %13 = call i8* @polly_initContextCUDA()
 ; CHECK-NEXT:  %14 = bitcast i32* %A to i8*
 ; CHECK-NEXT:  %15 = bitcast i32* %R to i8*
-; CHECK-NEXT:  %16 = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 0
+; CHECK-NEXT:  %16 = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_0_params, i64 0, i64 0
 ; CHECK-NEXT:  store i8* %14, i8** %polly_launch_0_param_0
 ; CHECK-NEXT:  %17 = bitcast i8** %polly_launch_0_param_0 to i8*
 ; CHECK-NEXT:  store i8* %17, i8** %16
-; CHECK-NEXT:  %18 = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 1
+; CHECK-NEXT:  %18 = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_0_params, i64 0, i64 1
 ; CHECK-NEXT:  store i8* %15, i8** %polly_launch_0_param_1
 ; CHECK-NEXT:  %19 = bitcast i8** %polly_launch_0_param_1 to i8*
 ; CHECK-NEXT:  store i8* %19, i8** %18
-; CHECK-NEXT:  store i32 4, i32* %polly_launch_0_param_size_0
-; CHECK-NEXT:  %20 = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 2
-; CHECK-NEXT:  %21 = bitcast i32* %polly_launch_0_param_size_0 to i8*
-; CHECK-NEXT:  store i8* %21, i8** %20
-; CHECK-NEXT:  store i32 4, i32* %polly_launch_0_param_size_1
-; CHECK-NEXT:  %22 = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 3
-; CHECK-NEXT:  %23 = bitcast i32* %polly_launch_0_param_size_1 to i8*
-; CHECK-NEXT:  store i8* %23, i8** %22
-; CHECK-NEXT:  %24 = call i8* @polly_getKernel(i8* getelementptr inbounds ([852 x i8], [852 x i8]* @FUNC_copy_SCOP_0_KERNEL_0, i32 0, i32 0), i8* getelementptr inbounds ([26 x i8], [26 x i8]* @FUNC_copy_SCOP_0_KERNEL_0_name, i32 0, i32 0))
-; CHECK-NEXT:  call void @polly_launchKernel(i8* %24, i32 2, i32 1, i32 32, i32 1, i32 1, i8* %polly_launch_0_params_i8ptr)
-; CHECK-NEXT:  call void @polly_freeKernel(i8* %24)
+; CHECK-NEXT:  %20 = call i8* @polly_getKernel(i8* getelementptr inbounds ([852 x i8], [852 x i8]* @FUNC_copy_SCOP_0_KERNEL_0, i32 0, i32 0), i8* getelementptr inbounds ([26 x i8], [26 x i8]* @FUNC_copy_SCOP_0_KERNEL_0_name, i32 0, i32 0))
+; CHECK-NEXT:  call void @polly_launchKernel(i8* %20, i32 2, i32 1, i32 32, i32 1, i32 1, i8* %polly_launch_0_params_i8ptr)
+; CHECK-NEXT:  call void @polly_freeKernel(i8* %20)
 ; CHECK-NEXT:  call void @polly_synchronizeDevice()
 ; CHECK-NEXT:  call void @polly_freeContext(i8* %13)
 
Index: polly/trunk/test/GPGPU/host-control-flow.ll
===================================================================
--- polly/trunk/test/GPGPU/host-control-flow.ll
+++ polly/trunk/test/GPGPU/host-control-flow.ll
@@ -31,7 +31,7 @@
 ; IR-NEXT:   %polly.indvar = phi i64 [ 0, %polly.loop_preheader ], [ %polly.indvar_next, %polly.loop_header ]
 ; ...
 ; IR:  store i64 %polly.indvar, i64* %polly_launch_0_param_1
-; IR-NEXT:  [[REGA:%.+]] = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 1
+; IR-NEXT:  [[REGA:%.+]] = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_0_params, i64 0, i64 1
 ; IR-NEXT:  [[REGB:%.+]] = bitcast i64* %polly_launch_0_param_1 to i8*
 ; IR-NEXT:  store i8* [[REGB]], i8** [[REGA]]
 ; IR: call i8* @polly_getKernel
Index: polly/trunk/test/GPGPU/invariant-load-hoisting.ll
===================================================================
--- polly/trunk/test/GPGPU/invariant-load-hoisting.ll
+++ polly/trunk/test/GPGPU/invariant-load-hoisting.ll
@@ -17,8 +17,8 @@
 ; SCOP-NEXT:              [n, tmp12] -> { Stmt_for_body6[i0, i1, i2] -> MemRef_invariant[0] };
 ; SCOP-NEXT:          Execution Context: [n, tmp12] -> {  : n > 0 }
 ; SCOP-NEXT:  }
-; HOST-IR:      call void @polly_launchKernel(i8* %219, i32 %225, i32 1, i32 32, i32 1, i32 1, i8* %polly_launch_0_params_i8ptr)
-; HOST-IR-NEXT: call void @polly_freeKernel(i8* %219)
+; HOST-IR:      call void @polly_launchKernel(i8* %209, i32 %215, i32 1, i32 32, i32 1, i32 1, i8* %polly_launch_0_params_i8ptr)
+; HOST-IR-NEXT: call void @polly_freeKernel(i8* %209)
 
 ; KERNEL-IR: define ptx_kernel void @FUNC_f_SCOP_0_KERNEL_0(i8 addrspace(1)* %MemRef_B, i8 addrspace(1)* %MemRef_A, i32 %n, i32 %tmp12, i32 %polly.preload.tmp21.merge)
 
Index: polly/trunk/test/GPGPU/kernel-params-only-some-arrays.ll
===================================================================
--- polly/trunk/test/GPGPU/kernel-params-only-some-arrays.ll
+++ polly/trunk/test/GPGPU/kernel-params-only-some-arrays.ll
@@ -48,13 +48,13 @@
 
 
 ; IR:       [[DEVPTR:%.*]] = call i8* @polly_getDevicePtr(i8* %p_dev_array_MemRef_A)
-; IR-NEXT:  [[SLOT:%.*]] = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_0_params, i64 0, i64 0
+; IR-NEXT:  [[SLOT:%.*]] = getelementptr [1 x i8*], [1 x i8*]* %polly_launch_0_params, i64 0, i64 0
 ; IR-NEXT:  store i8* [[DEVPTR]], i8** %polly_launch_0_param_0
 ; IR-NEXT:  [[DATA:%.*]] = bitcast i8** %polly_launch_0_param_0 to i8*
 ; IR-NEXT:  store i8* [[DATA]], i8** [[SLOT]]
 
 ; IR:       [[DEVPTR:%.*]] = call i8* @polly_getDevicePtr(i8* %p_dev_array_MemRef_B)
-; IR-NEXT:  [[SLOT:%.*]] = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_1_params, i64 0, i64 0
+; IR-NEXT:  [[SLOT:%.*]] = getelementptr [1 x i8*], [1 x i8*]* %polly_launch_1_params, i64 0, i64 0
 ; IR-NEXT:  store i8* [[DEVPTR]], i8** %polly_launch_1_param_0
 ; IR-NEXT:  [[DATA:%.*]] = bitcast i8** %polly_launch_1_param_0 to i8*
 ; IR-NEXT:  store i8* [[DATA]], i8** [[SLOT]]
Index: polly/trunk/test/GPGPU/parametric-loop-bound.ll
===================================================================
--- polly/trunk/test/GPGPU/parametric-loop-bound.ll
+++ polly/trunk/test/GPGPU/parametric-loop-bound.ll
@@ -32,7 +32,7 @@
 ; CODE-NEXT:     Stmt_bb2(32 * b0 + t0 + 1048576 * c0);
 
 ; IR: store i64 %n, i64* %polly_launch_0_param_1
-; IR-NEXT: [[REGA:%.+]] = getelementptr [4 x i8*], [4 x i8*]* %polly_launch_0_params, i64 0, i64 1
+; IR-NEXT: [[REGA:%.+]] = getelementptr [2 x i8*], [2 x i8*]* %polly_launch_0_params, i64 0, i64 1
 ; IR-NEXT: [[REGB:%.+]] = bitcast i64* %polly_launch_0_param_1 to i8*
 ; IR-NEXT: store i8* [[REGB]], i8** [[REGA]]
 
Index: polly/trunk/test/GPGPU/phi-nodes-in-kernel.ll
===================================================================
--- polly/trunk/test/GPGPU/phi-nodes-in-kernel.ll
+++ polly/trunk/test/GPGPU/phi-nodes-in-kernel.ll
@@ -48,7 +48,7 @@
 ; CODE-NEXT: if (32 * b0 + t0 <= 48)
 ; CODE-NEXT:     Stmt_for_body17(0, 32 * b0 + t0);
 
-; IR: [[REGC:%.+]] =   bitcast i32* %27 to i8*
+; IR: [[REGC:%.+]] =   bitcast i32* %23 to i8*
 ; IR-NEXT:  call void @polly_copyFromDeviceToHost(i8* %p_dev_array_MemRef_c, i8* [[REGC]], i64 196)
 
 ; KERNEL-IR: define ptx_kernel void @FUNC_kernel_dynprog_SCOP_0_KERNEL_0(i8 addrspace(1)* %MemRef_c, i32) #0 {