Index: lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
===================================================================
--- lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -122,7 +122,7 @@
   cl::location(AMDGPUTargetMachine::EnableLateStructurizeCFG),
   cl::Hidden);
 
-static cl::opt<bool, true> EnableAMDGPUFunctionCalls(
+static cl::opt<bool, true> EnableAMDGPUFunctionCallsOpt(
   "amdgpu-function-calls",
   cl::desc("Enable AMDGPU function call support"),
   cl::location(AMDGPUTargetMachine::EnableFunctionCalls),
@@ -330,7 +330,7 @@
 }
 
 bool AMDGPUTargetMachine::EnableLateStructurizeCFG = false;
-bool AMDGPUTargetMachine::EnableFunctionCalls = false;
+bool AMDGPUTargetMachine::EnableFunctionCalls = true;
 
 AMDGPUTargetMachine::~AMDGPUTargetMachine() = default;
 
@@ -361,11 +361,11 @@
 
   bool EnableOpt = getOptLevel() > CodeGenOpt::None;
   bool Internalize = InternalizeSymbols;
-  bool EarlyInline = EarlyInlineAll && EnableOpt && !EnableAMDGPUFunctionCalls;
+  bool EarlyInline = EarlyInlineAll && EnableOpt && !EnableFunctionCalls;
   bool AMDGPUAA = EnableAMDGPUAliasAnalysis && EnableOpt;
   bool LibCallSimplify = EnableLibCallSimplify && EnableOpt;
 
-  if (EnableAMDGPUFunctionCalls) {
+  if (EnableFunctionCalls) {
     delete Builder.Inliner;
     Builder.Inliner = createAMDGPUFunctionInliningPass();
   }
Index: test/CodeGen/AMDGPU/inline-calls.ll
===================================================================
--- test/CodeGen/AMDGPU/inline-calls.ll
+++ test/CodeGen/AMDGPU/inline-calls.ll
@@ -1,16 +1,16 @@
-; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck  %s
-; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck  %s
-; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,ALL %s
+; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,ALL %s
+; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s | FileCheck -check-prefixes=R600,ALL %s
 
-; CHECK-NOT: {{^}}func:
+; ALL-NOT: {{^}}func:
 define internal fastcc i32 @func(i32 %a) {
 entry:
   %tmp0 = add i32 %a, 1
   ret i32 %tmp0
 }
 
-; CHECK: {{^}}kernel:
-; CHECK-NOT: call
+; ALL: {{^}}kernel:
+; GCN-NOT: s_swappc_b64
 define amdgpu_kernel void @kernel(i32 addrspace(1)* %out) {
 entry:
   %tmp0 = call i32 @func(i32 1)
@@ -18,19 +18,19 @@
   ret void
 }
 
-; CHECK: {{^}}kernel2:
-; CHECK-NOT: call
+; ALL: {{^}}kernel2:
+; GCN-NOT: s_swappc_b64
 define amdgpu_kernel void @kernel2(i32 addrspace(1)* %out) {
 entry:
   call void @kernel(i32 addrspace(1)* %out)
   ret void
 }
 
-; CHECK-NOT: func_alias
+; ALL-NOT: func_alias
 @func_alias = alias i32 (i32), i32 (i32)* @func
 
-; CHECK: {{^}}kernel3:
-; CHECK-NOT: call
+; ALL: {{^}}kernel3:
+; GCN-NOT: s_swappc_b64
 define amdgpu_kernel void @kernel3(i32 addrspace(1)* %out) {
 entry:
   %tmp0 = call i32 @func_alias(i32 1)
@@ -38,11 +38,11 @@
   ret void
 }
 
-; CHECK-NOT: kernel_alias
+; ALL-NOT: kernel_alias
 @kernel_alias = alias void (i32 addrspace(1)*), void (i32 addrspace(1)*)* @kernel
 
-; CHECK: {{^}}kernel4:
-; CHECK-NOT: call
+; ALL: {{^}}kernel4:
+; GCN-NOT: s_swappc_b64
 define amdgpu_kernel void @kernel4(i32 addrspace(1)* %out) {
 entry:
   call void @kernel_alias(i32 addrspace(1)* %out)