Index: cfe/trunk/lib/Driver/Driver.cpp
===================================================================
--- cfe/trunk/lib/Driver/Driver.cpp
+++ cfe/trunk/lib/Driver/Driver.cpp
@@ -2312,6 +2312,8 @@
     /// compilation.
     bool CompileHostOnly = false;
     bool CompileDeviceOnly = false;
+    bool EmitLLVM = false;
+    bool EmitAsm = false;
 
     /// List of GPU architectures to use in this compilation.
     SmallVector<CudaArch, 4> GpuArchList;
@@ -2478,6 +2480,8 @@
       CompileDeviceOnly = PartialCompilationArg &&
                           PartialCompilationArg->getOption().matches(
                               options::OPT_cuda_device_only);
+      EmitLLVM = Args.getLastArg(options::OPT_emit_llvm);
+      EmitAsm = Args.getLastArg(options::OPT_S);
 
       // Collect all cuda_gpu_arch parameters, removing duplicates.
       std::set<CudaArch> GpuArchs;
@@ -2664,7 +2668,8 @@
       assert(!CompileHostOnly &&
              "Not expecting CUDA actions in host-only compilation.");
 
-      if (!Relocatable && CurPhase == phases::Backend) {
+      if (!Relocatable && CurPhase == phases::Backend && !EmitLLVM &&
+          !EmitAsm) {
         // If we are in backend phase, we attempt to generate the fat binary.
         // We compile each arch to IR and use a link action to generate code
         // object containing ISA. Then we use a special "link" action to create
@@ -2732,7 +2737,8 @@
         A = C.getDriver().ConstructPhaseAction(C, Args, CurPhase, A,
                                                AssociatedOffloadKind);
 
-      return ABRT_Success;
+      return (CompileDeviceOnly && CurPhase == FinalPhase) ? ABRT_Ignore_Host
+                                                           : ABRT_Success;
     }
 
     void appendLinkDependences(OffloadAction::DeviceDependences &DA) override {
Index: cfe/trunk/test/Driver/hip-device-compile.hip
===================================================================
--- cfe/trunk/test/Driver/hip-device-compile.hip
+++ cfe/trunk/test/Driver/hip-device-compile.hip
@@ -0,0 +1,72 @@
+// REQUIRES: clang-driver
+// REQUIRES: x86-registered-target
+// REQUIRES: amdgpu-registered-target
+
+// If -emit-llvm and/or -S is used in device only compilation,
+// the output should not be bundled.
+
+// RUN: %clang -c -emit-llvm --cuda-device-only -### -target x86_64-linux-gnu \
+// RUN:   -o a.bc -x hip --cuda-gpu-arch=gfx900 \
+// RUN:   --hip-device-lib=lib1.bc \
+// RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
+// RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
+// RUN: 2>&1 | FileCheck -check-prefixes=CHECK,BC %s
+
+// RUN: %clang -c -S -emit-llvm --cuda-device-only -### -target x86_64-linux-gnu \
+// RUN:   -o a.ll -x hip --cuda-gpu-arch=gfx900 \
+// RUN:   --hip-device-lib=lib1.bc \
+// RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
+// RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
+// RUN: 2>&1 | FileCheck -check-prefixes=CHECK,LL %s
+
+// RUN: %clang -c -S --cuda-device-only -### -target x86_64-linux-gnu \
+// RUN:   -o a.s -x hip --cuda-gpu-arch=gfx900 \
+// RUN:   --hip-device-lib=lib1.bc \
+// RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
+// RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
+// RUN: 2>&1 | FileCheck -check-prefixes=CHECK,ASM %s
+
+// CHECK: {{".*clang.*"}} "-cc1" "-triple" "amdgcn-amd-amdhsa"
+// CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
+// BC-SAME: "-emit-llvm-bc"
+// LL-SAME: "-emit-llvm"
+// ASM-NOT: "-emit-llvm"
+// CHECK-SAME: "-main-file-name" "a.cu" {{.*}} "-target-cpu" "gfx900"
+// CHECK-SAME: "-fcuda-is-device"
+// CHECK-SAME: {{".*lib1.bc"}}
+// BC-SAME: "-o" "a.bc"
+// LL-SAME: "-o" "a.ll"
+// ASM-SAME: "-o" "a.s"
+// CHECK-SAME: {{".*a.cu"}}
+
+// CHECK-NOT: {{"*.llvm-link"}}
+// CHECK-NOT: {{".*opt"}}
+// CHECK-NOT: {{".*llc"}}
+// CHECK-NOT: {{".*lld"}}
+// CHECK-NOT: {{".*clang-offload-bundler"}}
+// CHECK-NOT: {{".*ld.*"}}
+
+// If neither -emit-llvm nor -S is used in device only compilation,
+// the output should be bundled.
+
+// RUN: %clang -c --cuda-device-only -### -target x86_64-linux-gnu \
+// RUN:   -o a.s -x hip --cuda-gpu-arch=gfx900 \
+// RUN:   --hip-device-lib=lib1.bc \
+// RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
+// RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
+// RUN: 2>&1 | FileCheck -check-prefixes=BUNDLE %s
+
+// RUN: %clang --cuda-device-only -### -target x86_64-linux-gnu \
+// RUN:   -o a.s -x hip --cuda-gpu-arch=gfx900 \
+// RUN:   --hip-device-lib=lib1.bc \
+// RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
+// RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
+// RUN: 2>&1 | FileCheck -check-prefixes=BUNDLE %s
+
+// BUNDLE: {{"*.clang.*"}}
+// BUNDLE: {{"*.llvm-link"}}
+// BUNDLE: {{".*opt"}}
+// BUNDLE: {{".*llc"}}
+// BUNDLE: {{".*lld"}}
+// BUNDLE: {{".*clang-offload-bundler"}}
+