[HIP] Move HIP Linking Logic into HIP ToolChain

[ashi1]

Summary

This patch is a follow up on https://reviews.llvm.org/D78759.

Extract the HIP Linker script from generic GNU linker,
and move it into HIP ToolChain. Update OffloadActionBuilder
Link actions feature to apply device linking and host linking
actions separately.

[JonChesterfield]

This looks great to me. A bunch of HIP specific code moving out of generic code and into HIP specific files. The host / device split in the driver looks reasonable and similar to existing code. I especially like that all the changes to tests are under HIP.

One request - I'd like this to land before D78759 instead of after. Cleanup then extend as opposed to extend then cleanup. Primarily so that this patch doesn't get lost if D78759 stalls further or ends up getting reverted.

Thanks!

[ashi1]

In D81963#2096658, @JonChesterfield wrote:

This looks great to me. A bunch of HIP specific code moving out of generic code and into HIP specific files. The host / device split in the driver looks reasonable and similar to existing code. I especially like that all the changes to tests are under HIP.

One request - I'd like this to land before D78759 instead of after. Cleanup then extend as opposed to extend then cleanup. Primarily so that this patch doesn't get lost if D78759 stalls further or ends up getting reverted.

Thanks!

Sounds good, I am working on that now and will update this patch to be standalone.

[yaxunl]

The only thing that concerns me is that you called clang-offload-unbundler twice to unbundle a fat binary, one for host, one for device. One call of clang-offload-unbundler is supposed to unbundle everything.

[ashi1]

I've updated this patch so it doesn't have dependence on the static-lib patch.

This patch can be submitted standalone.

[ashi1]

In D81963#2096814, @yaxunl wrote:

The only thing that concerns me is that you called clang-offload-unbundler twice to unbundle a fat binary, one for host, one for device. One call of clang-offload-unbundler is supposed to unbundle everything.

You are right, I am looking into this.

[ashi1]

@yaxunl - I am looking into the bundler issue. Is it okay to commit these current patches first, while we look into that bug? This patch should only affect HIP.

[yaxunl]

LGTM. Thanks.

[MaskRay]

Hi, your git commit contains extra Phabricator tags. You can drop Reviewers: Subscribers: Tags: and the text Summary: from the git commit with the following script:

arcfilter () {
        arc amend
        git log -1 --pretty=%B | awk '/Reviewers:|Subscribers:/{p=1} /Reviewed By:|Differential Revision:/{p=0} !p && !/^Summary:$/ {sub(/^Summary: /,"");print}' | git commit --amend --date=now -F -
}

Reviewed By: is considered important by some people. Please keep the tag. (I have updated my script to use --date=now (setting author date to committer date))

https://reviews.llvm.org/D80978 contains a git pre-push hook to automate this.

clang/lib/Driver/Driver.cpp

public:
     /// dependency of the current device action. By default it is inactive.
     virtual ActionBuilderReturnCode addDeviceDepences(Action *HostAction) {
       return ABRT_Inactive;
     }
 
     /// Append top level actions generated by the builder.
     virtual void appendTopLevelActions(ActionList &AL) {}
 
-    /// Append linker actions generated by the builder.
-    virtual void appendLinkActions(ActionList &AL) {}
+    /// Append linker device actions generated by the builder.
+    virtual void appendLinkDeviceActions(ActionList &AL) {}
+
+    /// Append linker host action generated by the builder.
+    virtual Action* appendLinkHostActions(ActionList &AL) { return nullptr; }
 
     /// Append linker actions generated by the builder.
     virtual void appendLinkDependences(OffloadAction::DeviceDependences &DA) {}
 
     /// Initialize the builder. Return true if any initialization errors are
     /// found.
     virtual bool initialize() { return false; }
 
getDeviceDependences(OffloadAction::DeviceDependences &DA,
       for (Action *&A : CudaDeviceActions)
         A = C.getDriver().ConstructPhaseAction(C, Args, CurPhase, A,
                                                AssociatedOffloadKind);
 
       return (CompileDeviceOnly && CurPhase == FinalPhase) ? ABRT_Ignore_Host
                                                            : ABRT_Success;
     }
 
-    void appendLinkDependences(OffloadAction::DeviceDependences &DA) override {
+    void appendLinkDeviceActions(ActionList &AL) override {
+      if (DeviceLinkerInputs.size() == 0)
+        return;
+
+      assert(DeviceLinkerInputs.size() == GpuArchList.size() &&
+             "Linker inputs and GPU arch list sizes do not match.");
+
       // Append a new link action for each device.
       unsigned I = 0;
       for (auto &LI : DeviceLinkerInputs) {
+        // Each entry in DeviceLinkerInputs corresponds to a GPU arch.
         auto *DeviceLinkAction =
             C.MakeAction<LinkJobAction>(LI, types::TY_Image);
-        DA.add(*DeviceLinkAction, *ToolChains[0],
-               CudaArchToString(GpuArchList[I]), AssociatedOffloadKind);
+        // Linking all inputs for the current GPU arch.
+        // LI contains all the inputs for the linker.
+        OffloadAction::DeviceDependences DeviceLinkDeps;
+        DeviceLinkDeps.add(*DeviceLinkAction, *ToolChains[0],
+        AL.push_back(C.MakeAction<OffloadAction>(DeviceLinkDeps,
+            DeviceLinkAction->getType()));
         ++I;
       }
+      DeviceLinkerInputs.clear();
+
+      // Create a host object from all the device images by embedding them
+      // in a fat binary.
+      OffloadAction::DeviceDependences DDeps;
+      auto *TopDeviceLinkAction =
+          C.MakeAction<LinkJobAction>(AL, types::TY_Object);
+      DDeps.add(*TopDeviceLinkAction, *ToolChains[0],
+          nullptr, AssociatedOffloadKind);
+
+      // Offload the host object to the host linker.
+      AL.push_back(C.MakeAction<OffloadAction>(DDeps, TopDeviceLinkAction->getType()));
     }
+
+    Action* appendLinkHostActions(ActionList &AL) override { return AL.back(); }
+
+    void appendLinkDependences(OffloadAction::DeviceDependences &DA) override {}
   };
 
   /// OpenMP action builder. The host bitcode is passed to the device frontend
   /// and all the device linked images are passed to the host link phase.
   class OpenMPActionBuilder final : public DeviceActionBuilder {
     /// The OpenMP actions for the current input.
     ActionList OpenMPDeviceActions;
 
void appendTopLevelActions(ActionList &AL) override {
         Dep.add(*A, **TI, /*BoundArch=*/nullptr, Action::OFK_OpenMP);
         AL.push_back(C.MakeAction<OffloadAction>(Dep, A->getType()));
         ++TI;
       }
       // We no longer need the action stored in this builder.
       OpenMPDeviceActions.clear();
     }
 
-    void appendLinkActions(ActionList &AL) override {
+    void appendLinkDeviceActions(ActionList &AL) override {
       assert(ToolChains.size() == DeviceLinkerInputs.size() &&
              "Toolchains and linker inputs sizes do not match.");
 
       // Append a new link action for each device.
       auto TC = ToolChains.begin();
       for (auto &LI : DeviceLinkerInputs) {
         auto *DeviceLinkAction =
             C.MakeAction<LinkJobAction>(LI, types::TY_Image);
         OffloadAction::DeviceDependences DeviceLinkDeps;
         DeviceLinkDeps.add(*DeviceLinkAction, **TC, /*BoundArch=*/nullptr,
             Action::OFK_OpenMP);
         AL.push_back(C.MakeAction<OffloadAction>(DeviceLinkDeps,
             DeviceLinkAction->getType()));
         ++TC;
       }
       DeviceLinkerInputs.clear();
     }
 
+    Action* appendLinkHostActions(ActionList &AL) override {
+      // Create wrapper bitcode from the result of device link actions and compile
+      // it to an object which will be added to the host link command.
+      auto *BC = C.MakeAction<OffloadWrapperJobAction>(AL, types::TY_LLVM_BC);
+      auto *ASM = C.MakeAction<BackendJobAction>(BC, types::TY_PP_Asm);
+      return C.MakeAction<AssembleJobAction>(ASM, types::TY_Object);
+    }
+
     void appendLinkDependences(OffloadAction::DeviceDependences &DA) override {}
 
     bool initialize() override {
       // Get the OpenMP toolchains. If we don't get any, the action builder will
       // know there is nothing to do related to OpenMP offloading.
       auto OpenMPTCRange = C.getOffloadToolChains<Action::OFK_OpenMP>();
       for (auto TI = OpenMPTCRange.first, TE = OpenMPTCRange.second; TI != TE;
            ++TI)
public:
   }
 
   Action* makeHostLinkAction() {
     // Build a list of device linking actions.
     ActionList DeviceAL;
     for (DeviceActionBuilder *SB : SpecializedBuilders) {
       if (!SB->isValid())
         continue;
-      SB->appendLinkActions(DeviceAL);
+      SB->appendLinkDeviceActions(DeviceAL);
     }
 
     if (DeviceAL.empty())
       return nullptr;
 
-    // Create wrapper bitcode from the result of device link actions and compile
-    // it to an object which will be added to the host link command.
-    auto *BC = C.MakeAction<OffloadWrapperJobAction>(DeviceAL, types::TY_LLVM_BC);
-    auto *ASM = C.MakeAction<BackendJobAction>(BC, types::TY_PP_Asm);
-    return C.MakeAction<AssembleJobAction>(ASM, types::TY_Object);
+    // Let builders add host linking actions.
+    Action* HA;
+    for (DeviceActionBuilder *SB : SpecializedBuilders) {
+      if (!SB->isValid())
+        continue;
+      HA = SB->appendLinkHostActions(DeviceAL);
+    }
+    return HA;
   }
 
   /// Processes the host linker action. This currently consists of replacing it
   /// with an offload action if there are device link objects and propagate to
   /// the host action all the offload kinds used in the current compilation. The
   /// resulting action is returned.
   Action *processHostLinkAction(Action *HostAction) {
     // Add all the dependences from the device linking actions.

clang/lib/Driver/ToolChains/CommonArgs.h

 
 void linkXRayRuntimeDeps(const ToolChain &TC,
                          llvm::opt::ArgStringList &CmdArgs);
 
 void AddRunTimeLibs(const ToolChain &TC, const Driver &D,
                     llvm::opt::ArgStringList &CmdArgs,
                     const llvm::opt::ArgList &Args);
 
-void AddHIPLinkerScript(const ToolChain &TC, Compilation &C,
-                        const InputInfo &Output, const InputInfoList &Inputs,
-                        const llvm::opt::ArgList &Args,
-                        llvm::opt::ArgStringList &CmdArgs, const JobAction &JA,
-                        const Tool &T);
-
 const char *SplitDebugName(const llvm::opt::ArgList &Args,
                            const InputInfo &Input, const InputInfo &Output);
 
 void SplitDebugInfo(const ToolChain &TC, Compilation &C, const Tool &T,
                     const JobAction &JA, const llvm::opt::ArgList &Args,
                     const InputInfo &Output, const char *OutFile);
 
 void addLTOOptions(const ToolChain &ToolChain, const llvm::opt::ArgList &Args,

clang/lib/Driver/ToolChains/CommonArgs.cpp

void tools::AddLinkerInputs(const ToolChain &TC, const InputInfoList &Inputs,
   Args.AddAllArgValues(CmdArgs, options::OPT_Zlinker_input);
 
   // LIBRARY_PATH are included before user inputs and only supported on native
   // toolchains.
   if (!TC.isCrossCompiling())
     addDirectoryList(Args, CmdArgs, "-L", "LIBRARY_PATH");
 
   for (const auto &II : Inputs) {
-    // If the current tool chain refers to an OpenMP or HIP offloading host, we
-    // should ignore inputs that refer to OpenMP or HIP offloading devices -
+    // If the current tool chain refers to an OpenMP offloading host, we
+    // should ignore inputs that refer to OpenMP offloading devices -
     // they will be embedded according to a proper linker script.
     if (auto *IA = II.getAction())
       if ((JA.isHostOffloading(Action::OFK_OpenMP) &&
-           IA->isDeviceOffloading(Action::OFK_OpenMP)) ||
-          (JA.isHostOffloading(Action::OFK_HIP) &&
-           IA->isDeviceOffloading(Action::OFK_HIP)))
+           IA->isDeviceOffloading(Action::OFK_OpenMP)))
         continue;
 
     if (!TC.HasNativeLLVMSupport() && types::isLLVMIR(II.getType()))
       // Don't try to pass LLVM inputs unless we have native support.
       D.Diag(diag::err_drv_no_linker_llvm_support) << TC.getTripleString();
 
     // Add filenames immediately.
     if (II.isFilename()) {
if (TC.getTriple().isKnownWindowsMSVCEnvironment()) {
             << Args.getLastArg(options::OPT_rtlib_EQ)->getValue() << "MSVC";
       }
     } else
       AddLibgcc(TC, D, CmdArgs, Args);
     break;
   }
 }
 
-/// Add HIP linker script arguments at the end of the argument list so that
-/// the fat binary is built by embedding the device images into the host. The
-/// linker script also defines a symbol required by the code generation so that
-/// the image can be retrieved at runtime. This should be used only in tool
-/// chains that support linker scripts.
-void tools::AddHIPLinkerScript(const ToolChain &TC, Compilation &C,
-                               const InputInfo &Output,
-                               const InputInfoList &Inputs, const ArgList &Args,
-                               ArgStringList &CmdArgs, const JobAction &JA,
-                               const Tool &T) {
-
-  // If this is not a HIP host toolchain, we don't need to do anything.
-  if (!JA.isHostOffloading(Action::OFK_HIP))
-    return;
-
-  InputInfoList DeviceInputs;
-  for (const auto &II : Inputs) {
-    const Action *A = II.getAction();
-    // Is this a device linking action?
-    if (A && isa<LinkJobAction>(A) && A->isDeviceOffloading(Action::OFK_HIP)) {
-      DeviceInputs.push_back(II);
-    }
-  }
-
-  if (DeviceInputs.empty())
-    return;
-
-  // Create temporary linker script. Keep it if save-temps is enabled.
-  const char *LKS;
-  std::string Name =
-      std::string(llvm::sys::path::filename(Output.getFilename()));
-  if (C.getDriver().isSaveTempsEnabled()) {
-    LKS = C.getArgs().MakeArgString(Name + ".lk");
-  } else {
-    auto TmpName = C.getDriver().GetTemporaryPath(Name, "lk");
-    LKS = C.addTempFile(C.getArgs().MakeArgString(TmpName));
-  }
-
-  // Add linker script option to the command.
-  CmdArgs.push_back("-T");
-  CmdArgs.push_back(LKS);
-
-  // Create a buffer to write the contents of the linker script.
-  std::string LksBuffer;
-  llvm::raw_string_ostream LksStream(LksBuffer);
-
-  // Get the HIP offload tool chain.
-  auto *HIPTC = static_cast<const toolchains::HIPToolChain *>(
-      C.getSingleOffloadToolChain<Action::OFK_HIP>());
-  assert(HIPTC->getTriple().getArch() == llvm::Triple::amdgcn &&
-         "Wrong platform");
-  (void)HIPTC;
-
-  const char *BundleFile;
-  if (C.getDriver().isSaveTempsEnabled()) {
-    BundleFile = C.getArgs().MakeArgString(Name + ".hipfb");
-  } else {
-    auto TmpName = C.getDriver().GetTemporaryPath(Name, "hipfb");
-    BundleFile = C.addTempFile(C.getArgs().MakeArgString(TmpName));
-  }
-  AMDGCN::constructHIPFatbinCommand(C, JA, BundleFile, DeviceInputs, Args, T);
-
-  // Add commands to embed target binaries. We ensure that each section and
-  // image is 16-byte aligned. This is not mandatory, but increases the
-  // likelihood of data to be aligned with a cache block in several main host
-  // machines.
-  LksStream << "/*\n";
-  LksStream << "       HIP Offload Linker Script\n";
-  LksStream << " *** Automatically generated by Clang ***\n";
-  LksStream << "*/\n";
-  LksStream << "TARGET(binary)\n";
-  LksStream << "INPUT(" << BundleFile << ")\n";
-  LksStream << "SECTIONS\n";
-  LksStream << "{\n";
-  LksStream << "  .hip_fatbin :\n";
-  LksStream << "  ALIGN(0x10)\n";
-  LksStream << "  {\n";
-  LksStream << "    PROVIDE_HIDDEN(__hip_fatbin = .);\n";
-  LksStream << "    " << BundleFile << "\n";
-  LksStream << "  }\n";
-  LksStream << "  /DISCARD/ :\n";
-  LksStream << "  {\n";
-  LksStream << "    * ( __CLANG_OFFLOAD_BUNDLE__* )\n";
-  LksStream << "  }\n";
-  LksStream << "}\n";
-  LksStream << "INSERT BEFORE .data\n";
-  LksStream.flush();
-
-  // Dump the contents of the linker script if the user requested that. We
-  // support this option to enable testing of behavior with -###.
-  if (C.getArgs().hasArg(options::OPT_fhip_dump_offload_linker_script))
-    llvm::errs() << LksBuffer;
-
-  // If this is a dry run, do not create the linker script file.
-  if (C.getArgs().hasArg(options::OPT__HASH_HASH_HASH))
-    return;
-
-  // Open script file and write the contents.
-  std::error_code EC;
-  llvm::raw_fd_ostream Lksf(LKS, EC, llvm::sys::fs::OF_None);
-
-  if (EC) {
-    C.getDriver().Diag(clang::diag::err_unable_to_make_temp) << EC.message();
-    return;
-  }
-
-  Lksf << LksBuffer;
-}
-
 SmallString<128> tools::getStatsFileName(const llvm::opt::ArgList &Args,
                                          const InputInfo &Output,
                                          const InputInfo &Input,
                                          const Driver &D) {
   const Arg *A = Args.getLastArg(options::OPT_save_stats_EQ);
   if (!A)
     return {};
 

clang/lib/Driver/ToolChains/Gnu.cpp

if (!Args.hasArg(options::OPT_nostartfiles) && !IsIAMCU) {
         }
         CmdArgs.push_back(Args.MakeArgString(P));
       }
       if (!isAndroid)
         CmdArgs.push_back(Args.MakeArgString(ToolChain.GetFilePath("crtn.o")));
     }
   }
 
-  // Add HIP offloading linker script args if required.
-  AddHIPLinkerScript(getToolChain(), C, Output, Inputs, Args, CmdArgs, JA,
-                     *this);
-
   Args.AddAllArgs(CmdArgs, options::OPT_T);
 
   const char *Exec = Args.MakeArgString(ToolChain.GetLinkerPath());
   C.addCommand(std::make_unique<Command>(JA, *this, Exec, CmdArgs, Inputs));
 }
 
 void tools::gnutools::Assembler::ConstructJob(Compilation &C,
                                               const JobAction &JA,

clang/lib/Driver/ToolChains/HIP.h

void ConstructJob(Compilation &C, const JobAction &JA,
                     const llvm::opt::ArgList &TCArgs,
                     const char *LinkingOutput) const override;
 
 private:
 
   void constructLldCommand(Compilation &C, const JobAction &JA,
                            const InputInfoList &Inputs, const InputInfo &Output,
                            const llvm::opt::ArgList &Args) const;
+
+  // Construct command for creating Object from HIP fatbin.
+  void constructGenerateObjFileFromHIPFatBinary(Compilation &C,
+                                                const InputInfo &Output,
+                                                const InputInfoList &Inputs,
+                                                const llvm::opt::ArgList &Args,
+                                                const JobAction &JA) const;
 };
 
 } // end namespace AMDGCN
 } // end namespace tools
 
 namespace toolchains {
 
 class LLVM_LIBRARY_VISIBILITY HIPToolChain final : public ROCMToolChain {

clang/lib/Driver/ToolChains/HIP.cpp

auto BundlerOutputArg = Args.MakeArgString(
       std::string("-outputs=").append(std::string(OutputFileName)));
   BundlerArgs.push_back(BundlerOutputArg);
 
   const char *Bundler = Args.MakeArgString(
       T.getToolChain().GetProgramPath("clang-offload-bundler"));
   C.addCommand(std::make_unique<Command>(JA, T, Bundler, BundlerArgs, Inputs));
 }
 
+/// Add Generated HIP Object File which has device images embedded into the
+/// host to the argument list for linking. Using MC directives, embed the
+/// device code and also define symbols required by the code generation so that
+/// the image can be retrieved at runtime.
+void AMDGCN::Linker::constructGenerateObjFileFromHIPFatBinary(
+    Compilation &C, const InputInfo &Output,
+    const InputInfoList &Inputs, const ArgList &Args,
+    const JobAction &JA) const {
+  const ToolChain &TC = getToolChain();
+  std::string Name =
+      std::string(llvm::sys::path::stem(Output.getFilename()));
+
+  // Create Temp Object File Generator,
+  // Offload Bundled file and Bundled Object file.
+  // Keep them if save-temps is enabled.
+  const char *McinFile;
+  const char *BundleFile;
+  if (C.getDriver().isSaveTempsEnabled()) {
+    McinFile = C.getArgs().MakeArgString(Name + ".mcin");
+    BundleFile = C.getArgs().MakeArgString(Name + ".hipfb");
+  } else {
+    auto TmpNameMcin = C.getDriver().GetTemporaryPath(Name, "mcin");
+    McinFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameMcin));
+    auto TmpNameFb = C.getDriver().GetTemporaryPath(Name, "hipfb");
+    BundleFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameFb));
+  }
+  constructHIPFatbinCommand(C, JA, BundleFile, Inputs, Args, *this);
+
+  // Create a buffer to write the contents of the temp obj generator.
+  std::string ObjBuffer;
+  llvm::raw_string_ostream ObjStream(ObjBuffer);
+
+  // Add MC directives to embed target binaries. We ensure that each
+  // section and image is 16-byte aligned. This is not mandatory, but
+  // increases the likelihood of data to be aligned with a cache block
+  // in several main host machines.
+  ObjStream << "#       HIP Object Generator\n";
+  ObjStream << "# *** Automatically generated by Clang ***\n";
+  ObjStream << "  .type __hip_fatbin,@object\n";
+  ObjStream << "  .section .hip_fatbin,\"aMS\",@progbits,1\n";
+  ObjStream << "  .data\n";
+  ObjStream << "  .globl __hip_fatbin\n";
+  ObjStream << "  .p2align 3\n";
+  ObjStream << "__hip_fatbin:\n";
+  ObjStream << "  .incbin \"" << BundleFile << "\"\n";
+  ObjStream.flush();
+
+  // Dump the contents of the temp object file gen if the user requested that.
+  // We support this option to enable testing of behavior with -###.
+  if (C.getArgs().hasArg(options::OPT_fhip_dump_offload_linker_script))
+    llvm::errs() << ObjBuffer;
+
+  // Open script file and write the contents.
+  std::error_code EC;
+  llvm::raw_fd_ostream Objf(McinFile, EC, llvm::sys::fs::OF_None);
+
+  if (EC) {
+    C.getDriver().Diag(clang::diag::err_unable_to_make_temp) << EC.message();
+    return;
+  }
+
+  Objf << ObjBuffer;
+
+  ArgStringList McArgs{"-triple", Args.MakeArgString(TC.getTripleString()),
+                       "-o",      Output.getFilename(),
+                       McinFile,  "--filetype=obj"};
+  const char *Mc = Args.MakeArgString(TC.GetProgramPath("llvm-mc"));
+  C.addCommand(std::make_unique<Command>(JA, *this, Mc, McArgs, Inputs));
+}
+
 // For amdgcn the inputs of the linker job are device bitcode and output is
 // object file. It calls llvm-link, opt, llc, then lld steps.
 void AMDGCN::Linker::ConstructJob(Compilation &C, const JobAction &JA,
                                    const InputInfo &Output,
                                    const InputInfoList &Inputs,
                                    const ArgList &Args,
                                    const char *LinkingOutput) const {
+  if (Inputs.size() > 0 &&
+      Inputs[0].getType() == types::TY_Image &&
+      JA.getType() == types::TY_Object)
+    return constructGenerateObjFileFromHIPFatBinary(C, Output, Inputs, Args, JA);
 
   if (JA.getType() == types::TY_HIP_FATBIN)
     return constructHIPFatbinCommand(C, JA, Output.getFilename(), Inputs, Args, *this);
 
   return constructLldCommand(C, JA, Inputs, Output, Args);
 }
 
 HIPToolChain::HIPToolChain(const Driver &D, const llvm::Triple &Triple,

clang/test/Driver/hip-binding.hip

 // RDCS: # "x86_64-unknown-linux-gnu" - "clang", inputs: ["[[IN]]"], output: "[[HOSTOBJ:.*o]]"
 // RDCS: # "x86_64-unknown-linux-gnu" - "offload bundler", inputs: ["[[BC1]]", "[[BC2]]", "[[HOSTOBJ]]"], output: "{{.*}}"
 
 // RUN: touch %t.o
 // RUN: %clang --hip-link -ccc-print-bindings -target x86_64-linux-gnu \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 -fgpu-rdc %t.o\
 // RUN: 2>&1 | FileCheck %s
 
-// CHECK: # "amdgcn-amd-amdhsa" - "offload bundler", inputs: ["[[IN:.*o]]"], outputs: ["[[OBJ1:.*o]]", "[[OBJ2:.*o]]", "[[OBJ3:.*o]]"] 
-// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[OBJ2]]"], output: "[[IMG2:.*out]]"
+// CHECK: # "x86_64-unknown-linux-gnu" - "offload bundler", inputs: ["[[IN:.*o]]"], outputs: ["[[HOSTOBJ:.*o]]", "{{.*o}}", "{{.*o}}"]
+// CHECK: # "amdgcn-amd-amdhsa" - "offload bundler", inputs: ["[[IN]]"], outputs: ["{{.*o}}", "[[DOBJ1:.*o]]", "[[DOBJ2:.*o]]"]
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[DOBJ1]]"], output: "[[IMG1:.*out]]"
 // CHECK-NOT: offload bundler
-// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[OBJ3]]"], output: "[[IMG3:.*out]]"
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[DOBJ2]]"], output: "[[IMG2:.*out]]"
 // CHECK-NOT: offload bundler
-// CHECK: # "x86_64-unknown-linux-gnu" - "GNU::Linker", inputs: ["[[OBJ1]]", "[[IMG2]]", "[[IMG3]]"], output: "a.out"
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[IMG1]]", "[[IMG2]]"], output: "[[FATBINOBJ:.*o]]"
+// CHECK-NOT: offload bundler
+// CHECK: # "x86_64-unknown-linux-gnu" - "GNU::Linker", inputs: ["[[HOSTOBJ]]", "[[FATBINOBJ]]"], output: "a.out"
 
 // RUN: %clang --hip-link -ccc-print-bindings -target x86_64-linux-gnu \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %t.o\
 // RUN: 2>&1 | FileCheck -check-prefix=NORDC %s
 
 // NORDC-NOT: offload bundler
 // NORDC: # "x86_64-unknown-linux-gnu" - "GNU::Linker", inputs: ["{{.*o}}"], output: "a.out"

clang/test/Driver/hip-link-save-temps.hip

 // CHECK: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=obj2-host-x86_64-unknown-linux-gnu.o,obj2-hip-amdgcn-amd-amdhsa-gfx900.o,obj2-hip-amdgcn-amd-amdhsa-gfx906.o" "-unbundle"
 // CHECK-NOT: llvm-link
 // CHECK-NOT: opt
 // CHECK-NOT: llc
 // CHECK: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx900" "obj1-hip-amdgcn-amd-amdhsa-gfx900.o" "obj2-hip-amdgcn-amd-amdhsa-gfx900.o"
 // CHECK: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx906" "obj1-hip-amdgcn-amd-amdhsa-gfx906.o" "obj2-hip-amdgcn-amd-amdhsa-gfx906.o"
-// OUT: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=executable.hipfb"
-// OUT: "{{.*ld.*}}" {{.*}} "-o" "executable" {{.*}} "-T" "executable.lk"
-// NOUT: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=a.out.hipfb"
-// NOUT: "{{.*ld.*}}" {{.*}} "-o" "a.out" {{.*}} "-T" "a.out.lk"
+// CHECK: {{".*llvm-mc.*"}} "-triple" "amdgcn-amd-amdhsa" "-o"
+// CHECK-SAME: "[[OBJBUNDLE:.*.o]]" "{{.*}}.mcin" "--filetype=obj"
+// OUT: "{{.*ld.*}}" {{.*}} "-o" "executable" {{.*}} "[[OBJBUNDLE]]"
+// NOUT: "{{.*ld.*}}" {{.*}} "-o" "a.out" {{.*}} "[[OBJBUNDLE]]"

clang/test/Driver/hip-link-shared-library.hip

 // RUN: touch %t.o
 // RUN: %clang --hip-link -ccc-print-bindings -target x86_64-linux-gnu \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %t.o %S/Inputs/in.so \
 // RUN:   -fgpu-rdc 2>&1 | FileCheck %s
 
-// CHECK: # "amdgcn-amd-amdhsa" - "offload bundler", inputs: ["[[IN:.*o]]"], outputs: ["[[OBJ1:.*o]]", "[[OBJ2:.*o]]", "[[OBJ3:.*o]]"]
-// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[OBJ2]]"], output: "[[IMG2:.*out]]"
+// CHECK: # "x86_64-unknown-linux-gnu" - "offload bundler", inputs: ["[[IN:.*o]]"], outputs: ["[[HOSTOBJ:.*o]]", "{{.*o}}", "{{.*o}}"]
+// CHECK: # "amdgcn-amd-amdhsa" - "offload bundler", inputs: ["[[IN]]"], outputs: ["{{.*o}}", "[[DOBJ1:.*o]]", "[[DOBJ2:.*o]]"]
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[DOBJ1]]"], output: "[[IMG1:.*out]]"
 // CHECK-NOT: offload bundler
-// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[OBJ3]]"], output: "[[IMG3:.*out]]"
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[DOBJ2]]"], output: "[[IMG2:.*out]]"
 // CHECK-NOT: offload bundler
-// CHECK: # "x86_64-unknown-linux-gnu" - "GNU::Linker", inputs: ["[[OBJ1]]", "{{.*}}/Inputs/in.so", "[[IMG2]]", "[[IMG3]]"], output: "a.out"
+// CHECK: # "amdgcn-amd-amdhsa" - "AMDGCN::Linker", inputs: ["[[IMG1]]", "[[IMG2]]"], output: "[[FATBINOBJ:.*o]]"
+// CHECK-NOT: offload bundler
+// CHECK: # "x86_64-unknown-linux-gnu" - "GNU::Linker", inputs: ["[[HOSTOBJ]]", "{{.*}}/Inputs/in.so", "[[FATBINOBJ]]"], output: "a.out"
 

clang/test/Driver/hip-phases.hip

 //
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 -fgpu-rdc %s 2>&1 \
 // RUN: | FileCheck -check-prefixes=BIN,RDC %s
 //
 // BIN-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (host-[[T]])
 // BIN-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]])
 // BIN-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]])
+// RDC-DAG: [[P12:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]])
+// RDC-DAG: [[P13:[0-9]+]]: assembler, {[[P12]]}, object, (host-[[T]])
 
 // BIN-DAG: [[P3:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH:gfx803]])
 // BIN-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]])
 // BIN-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH]])
 // NRD-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, assembler, (device-[[T]], [[ARCH]])
 // NRD-DAG: [[P7:[0-9]+]]: assembler, {[[P6]]}, object, (device-[[T]], [[ARCH]])
-// NRD-DAG: [[P8:[0-9]+]]: linker, {[[P7]]}, image, (device-[[T]], [[ARCH]])
-// NRD-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH]])" {[[P8]]}, image
+// RDC-DAG: [[P7:[0-9]+]]: backend, {[[P5]]}, ir, (device-[[T]], [[ARCH]])
+// BIN-DAG: [[P8:[0-9]+]]: linker, {[[P7]]}, image, (device-[[T]], [[ARCH]])
+// BIN-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH]])" {[[P8]]}, image
 // NRD-DAG: [[P10:[0-9]+]]: linker, {[[P9]]}, hip-fatbin, (device-[[T]])
-// RDC-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, ir, (device-[[T]], [[ARCH]])
-// RDC-DAG: [[P10:[0-9]+]]: linker, {[[P6]]}, image, (device-[[T]], [[ARCH]])
+// RDC-DAG: [[P10:[0-9]+]]: linker, {[[P9]]}, object, (device-[[T]])
 
-// NRD-DAG: [[P12:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P2]]}, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P10]]}, ir
+// NRD-DAG: [[P11:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P2]]}, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P10]]}, ir
+// RDC-DAG: [[P11:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P10]]}, object
+// NRD-DAG: [[P12:[0-9]+]]: backend, {[[P11]]}, assembler, (host-[[T]])
+// NRD-DAG: [[P13:[0-9]+]]: assembler, {[[P12]]}, object, (host-[[T]])
+// NRD-DAG: [[P14:[0-9]+]]: linker, {[[P13]]}, image, (host-[[T]])
+// RDC-DAG: [[P14:[0-9]+]]: linker, {[[P13]], [[P11]]}, image, (host-[[T]])
 
-// NRD-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, assembler, (host-[[T]])
-// RDC-DAG: [[P13:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]])
-// BIN-DAG: [[P14:[0-9]+]]: assembler, {[[P13]]}, object, (host-[[T]])
-// BIN-DAG: [[P15:[0-9]+]]: linker, {[[P14]]}, image, (host-[[T]])
-// RDC-DAG: [[P16:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P15]]}, "device-[[T]] (amdgcn-amd-amdhsa:gfx803)" {[[P10]]}, image
 //
 // Test single gpu architecture up to the assemble phase.
 //
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 %s -S 2>&1 \
 // RUN: | FileCheck -check-prefixes=ASM %s
 // ASM-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (device-[[T]], [[ARCH:gfx803]])
 // ASM-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]])
 // ASM-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]])
 
 // ASM-DAG: [[P5:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (host-[[T]])
 // ASM-DAG: [[P6:[0-9]+]]: preprocessor, {[[P5]]}, [[T]]-cpp-output, (host-[[T]])
 // ASM-DAG: [[P7:[0-9]+]]: compiler, {[[P6]]}, ir, (host-[[T]])
 // ASM-DAG: [[P8:[0-9]+]]: backend, {[[P7]]}, assembler, (host-[[T]])
 
 //
-// Test two gpu architectures with complete compilation.
+// Test two gpu architectures with complete compilation with -fno-gpu-rdc.
 //
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s 2>&1 \
-// RUN: | FileCheck -check-prefixes=BIN2,NRD2,CL2 %s
+// RUN: | FileCheck -check-prefixes=NRD2,NCL2 %s
 
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s -c 2>&1 \
-// RUN: | FileCheck -check-prefixes=BIN2,NRD2 %s
+// RUN: | FileCheck -check-prefixes=NRD2 %s
 
+// NRD2-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (host-[[T]])
+// NRD2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]])
+// NRD2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]])
+
+// NRD2-DAG: [[P3:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH1:gfx803]])
+// NRD2-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH1]])
+// NRD2-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH1]])
+// NRD2-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, assembler, (device-[[T]], [[ARCH1]])
+// NRD2-DAG: [[P7:[0-9]+]]: assembler, {[[P6]]}, object, (device-[[T]], [[ARCH1]])
+// NRD2-DAG: [[P8:[0-9]+]]: linker, {[[P7]]}, image, (device-[[T]], [[ARCH1]])
+// NRD2-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P8]]}, image
+
+// NRD2-DAG: [[P10:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH2:gfx900]])
+// NRD2-DAG: [[P11:[0-9]+]]: preprocessor, {[[P10]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]])
+// NRD2-DAG: [[P12:[0-9]+]]: compiler, {[[P11]]}, ir, (device-[[T]], [[ARCH2]])
+// NRD2-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, assembler, (device-[[T]], [[ARCH2]])
+// NRD2-DAG: [[P14:[0-9]+]]: assembler, {[[P13]]}, object, (device-[[T]], [[ARCH2]])
+// NRD2-DAG: [[P15:[0-9]+]]: linker, {[[P14]]}, image, (device-[[T]], [[ARCH2]])
+// NRD2-DAG: [[P16:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P15]]}, image
+// NRD2-DAG: [[P17:[0-9]+]]: linker, {[[P9]], [[P16]]}, hip-fatbin, (device-[[T]])
+// NRD2-DAG: [[P18:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P2]]}, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P17]]}, ir
+// NRD2-DAG: [[P19:[0-9]+]]: backend, {[[P18]]}, assembler, (host-[[T]])
+// NRD2-DAG: [[P20:[0-9]+]]: assembler, {[[P19]]}, object, (host-[[T]])
+// NCL2-DAG: [[P21:[0-9]+]]: linker, {[[P20]]}, image, (host-[[T]])
+
+//
+// Test two gpu architectures with complete compilation with -fgpu-rdc.
+//
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s -fgpu-rdc 2>&1 \
-// RUN: | FileCheck -check-prefixes=BIN2,RDC2,CL2,RCL2 %s
+// RUN: | FileCheck -check-prefixes=RDC2,CL2,RCL2 %s
 
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s -fgpu-rdc -c 2>&1 \
-// RUN: | FileCheck -check-prefixes=BIN2,RDC2,RC2 %s
+// RUN: | FileCheck -check-prefixes=RDC2,RC2 %s
 
-// BIN2-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (host-[[T]])
-// BIN2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]])
-// BIN2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]])
-
-// BIN2-DAG: [[P3:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH1:gfx803]])
-// BIN2-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH1]])
-// BIN2-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH1]])
-// NRD2-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, assembler, (device-[[T]], [[ARCH1]])
+// RCL2-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (host-[[T]])
+// RCL2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]])
+// RCL2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]])
+// RCL2-DAG: [[P19:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]])
+// RCL2-DAG: [[P20:[0-9]+]]: assembler, {[[P19]]}, object, (host-[[T]])
+
+// RDC2-DAG: [[P3:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (device-[[T]], [[ARCH1:gfx803]])
+// RDC2-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH1]])
+// RDC2-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH1]])
 // RDC2-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, ir, (device-[[T]], [[ARCH1]])
-// NRD2-DAG: [[P7:[0-9]+]]: assembler, {[[P6]]}, object, (device-[[T]], [[ARCH1]])
-// NRD2-DAG: [[P8:[0-9]+]]: linker, {[[P7]]}, image, (device-[[T]], [[ARCH1]])
 // RCL2-DAG: [[P8:[0-9]+]]: linker, {[[P6]]}, image, (device-[[T]], [[ARCH1]])
-// NRD2-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P8]]}, image
+// RCL2-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P8]]}, image
 // RC2-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P6]]}, ir
 
-// BIN2-DAG: [[P10:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH2:gfx900]])
-// BIN2-DAG: [[P11:[0-9]+]]: preprocessor, {[[P10]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]])
-// BIN2-DAG: [[P12:[0-9]+]]: compiler, {[[P11]]}, ir, (device-[[T]], [[ARCH2]])
-// NRD2-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, assembler, (device-[[T]], [[ARCH2]])
+// RDC2-DAG: [[P10:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T]], (device-[[T]], [[ARCH2:gfx900]])
+// RDC2-DAG: [[P11:[0-9]+]]: preprocessor, {[[P10]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]])
+// RDC2-DAG: [[P12:[0-9]+]]: compiler, {[[P11]]}, ir, (device-[[T]], [[ARCH2]])
 // RDC2-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, ir, (device-[[T]], [[ARCH2]])
-// NRD2-DAG: [[P14:[0-9]+]]: assembler, {[[P13]]}, object, (device-[[T]], [[ARCH2]])
-// NRD2-DAG: [[P15:[0-9]+]]: linker, {[[P14]]}, image, (device-[[T]], [[ARCH2]])
 // RCL2-DAG: [[P15:[0-9]+]]: linker, {[[P13]]}, image, (device-[[T]], [[ARCH2]])
-// NRD2-DAG: [[P16:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P15]]}, image
+// RCL2-DAG: [[P16:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P15]]}, image
 // RC2-DAG: [[P16:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P13]]}, ir
 
-// NRD2-DAG: [[P17:[0-9]+]]: linker, {[[P9]], [[P16]]}, hip-fatbin, (device-[[T]])
-
-// NRD2-DAG: [[P18:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P2]]}, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P17]]}, ir
-// NRD2-DAG: [[P19:[0-9]+]]: backend, {[[P18]]}, assembler, (host-[[T]])
-// RDC2-DAG: [[P19:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]])
-// BIN2-DAG: [[P20:[0-9]+]]: assembler, {[[P19]]}, object, (host-[[T]])
-// CL2-DAG: [[P21:[0-9]+]]: linker, {[[P20]]}, image, (host-[[T]])
-// RCL2-DAG: [[P22:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P21]]}, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P8]]}, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P15]]}, image
-// RC2-DAG: [[P22:[0-9]+]]: clang-offload-bundler, {[[P9]], [[P16]], [[P20]]}, object, (host-[[T]])
+// RC2-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (host-[[T]])
+// RC2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]])
+// RC2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]])
+// RC2-DAG: [[P19:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]])
+// RC2-DAG: [[P20:[0-9]+]]: assembler, {[[P19]]}, object, (host-[[T]])
+
+// RCL2-DAG: [[P17:[0-9]+]]: linker, {[[P9]], [[P16]]}, object, (device-[[T]])
+// RCL2-DAG: [[P22:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P17]]}, object
+// RCL2-DAG: [[P23:[0-9]+]]: linker, {[[P20]], [[P22]]}, image, (host-[[T]])
+// RC2-DAG: [[P23:[0-9]+]]: clang-offload-bundler, {[[P9]], [[P16]], [[P20]]}, object, (host-[[T]])
 
 //
 // Test two gpu architecturess up to the assemble phase.
 //
 // RUN: %clang -x hip -target x86_64-unknown-linux-gnu -ccc-print-phases \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s -S 2>&1 \
 // RUN: | FileCheck -check-prefixes=ASM2 %s
 // ASM2-DAG: [[P0:[0-9]+]]: input, "{{.*}}hip-phases.hip", [[T:hip]], (device-[[T]], [[ARCH1:gfx803]])

 // RUN: %clang -target x86_64-unknown-linux-gnu -ccc-print-phases --hip-link \
 // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %T/obj1.o %T/obj2.o \
 // RUN: -fgpu-rdc 2>&1 | FileCheck -check-prefixes=L2,RL2 %s
 //
 // L2-DAG: [[P0:[0-9]+]]: input, "{{.*}}obj1.o", object, (host-[[T:hip]])
 // RL2-DAG: [[P1:[0-9]+]]: clang-offload-unbundler, {[[P0]]}, object, (host-[[T]])
 // L2-DAG: [[P2:[0-9]+]]: input, "{{.*}}obj2.o", object, (host-[[T]])
 // RL2-DAG: [[P3:[0-9]+]]: clang-offload-unbundler, {[[P2]]}, object, (host-[[T]])
-// NL2-DAG: [[P4:[0-9]+]]: linker, {[[P0]], [[P2]]}, image, (host-[[T]])
-// RL2-DAG: [[P4:[0-9]+]]: linker, {[[P1]], [[P3]]}, image, (host-[[T]])
-// RL2-DAG: [[P5:[0-9]+]]: linker, {[[P1]], [[P3]]}, image, (device-[[T]], [[ARCH1:gfx803]])
+
+// RL2-DAG: [[P4:[0-9]+]]: linker, {[[P1]], [[P3]]}, image, (device-[[T]], [[ARCH1:gfx803]])
+// RL2-DAG: [[P5:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P4]]}, image
 // RL2-DAG: [[P6:[0-9]+]]: linker, {[[P1]], [[P3]]}, image, (device-[[T]], [[ARCH2:gfx900]])
-// RL2-DAG: [[P7:[0-9]+]]: offload, "host-[[T]] (x86_64-unknown-linux-gnu)" {[[P4]]}, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH1]])" {[[P5]]}, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P6]]}, image
+// RL2-DAG: [[P7:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa:[[ARCH2]])" {[[P6]]}, image
+// RL2-DAG: [[P8:[0-9]+]]: linker, {[[P5]], [[P7]]}, object, (device-[[T]])
+// RL2-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] (amdgcn-amd-amdhsa)" {[[P8]]}, object
+
+// NL2-DAG: [[P4:[0-9]+]]: linker, {[[P0]], [[P2]]}, image, (host-[[T]])
+// RL2-DAG: [[P4:[0-9]+]]: linker, {[[P1]], [[P3]], [[P9]]}, image, (host-[[T]])

clang/test/Driver/hip-save-temps.hip

 // -fgpu-rdc without -o with -c
 // RUN: %clang -### -target x86_64-linux-gnu -nogpulib -save-temps \
 // RUN:   -fgpu-rdc --cuda-gpu-arch=gfx900 -c %s 2>&1 | \
 // RUN:   FileCheck -check-prefixes=CHECK,RDC,RDCC %s
 
 // -fgpu-rdc without -o
 // RUN: %clang -### -target x86_64-linux-gnu -nogpulib -save-temps \
 // RUN:   -fgpu-rdc --cuda-gpu-arch=gfx900 %s 2>&1 | \
-// RUN:   FileCheck -check-prefixes=CHECK,RDC,RDCL,RDC-NOUT,NOUT %s
+// RUN:   FileCheck -check-prefixes=CHECK,RDC,RDCL,NOUT %s
 
 // -fgpu-rdc with -o
-// RUN: %clang -### -target x86_64-linux-gnu -nogpulib -save-temps \
-// RUN:   -o executable -fgpu-rdc --cuda-gpu-arch=gfx900 %s 2>&1 | \
-// RUN:   FileCheck -check-prefixes=CHECK,RDC,RDCL,RDC-WOUT,WOUT %s
+// UN: %clang -### -target x86_64-linux-gnu -nogpulib -save-temps \
+// UN:   -o executable -fgpu-rdc --cuda-gpu-arch=gfx900 %s 2>&1 | \
+// UN:   FileCheck -check-prefixes=CHECK,RDC,RDCL,WOUT %s
+
+// -fgpu-rdc host object path
+// RDCL: "{{.*clang.*}}" "-cc1" {{.*}} "-E" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.cui"
+// RDCL: "{{.*clang.*}}" "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.bc"
+// RDCL: "{{.*clang.*}}" "-cc1" {{.*}} "-S" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.s"
+// RDCL: "{{.*clang.*}}" "-cc1as" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.o"
 
+// device object paths
 // CHECK: {{".*clang.*"}} "-cc1" {{.*}} "-E" {{.*}} [[CPU:"-target-cpu" "gfx900"]] {{.*}}  "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.cui"
 // NORDC: {{".*clang.*"}} "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} [[CPU]] {{.*}} "-disable-llvm-passes" {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.bc"
 // RDC: {{".*clang.*"}} "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} [[CPU]] {{.*}} "-disable-llvm-passes" {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.tmp.bc"
 
 // NORDC: {{".*clang.*"}} "-cc1" {{.*}} "-S"
 // NORDC-SAME: "-mllvm" "-amdgpu-internalize-symbols"
 // NORDC-SAME: [[CPU]]
 // NORDC-SAME: "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.s"
 
 // RDC: {{".*clang.*"}} "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} [[CPU]] {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.bc"
 // NORDC: {{".*clang.*"}} "-cc1as" {{.*}} "-filetype" "obj" {{.*}} [[CPU]] {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.o"
-// CHECK-NOT: llvm-link
-// CHECK-NOT: opt
-// CHECK-NOT: llc
+// CHECK-NOT: "{{.*}}llvm-link"
+// CHECK-NOT: "{{.*}}opt"
+// CHECK-NOT: "{{.*}}llc"
 // NORDC: {{.*lld.*}}"-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.out"
-
 // RDCL: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // RDCL-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx900"
+// RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-E" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.cui"
+// RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.bc"
+// RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-S" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.s"
+// RDCC: "{{.*clang.*}}" "-cc1as" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.o"
+// RDCC: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps.o"
+// RDCL: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps-hip-amdgcn-amd-amdhsa.hipfb"
+// RDCL: {{.*}}llvm-mc{{.*}}"-o" "hip-save-temps-hip-amdgcn-amd-amdhsa.o" "hip-save-temps-hip-amdgcn-amd-amdhsa.mcin" "--filetype=obj"
 
-// NORDC: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps.hip-hip-amdgcn-amd-amdhsa.hipfb"
-// CHECK: "{{.*clang.*}}" "-cc1" {{.*}} "-E" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.cui"
+// -fno-gpu-rdc host object path
+// NORDC: "{{.*clang.*}}" "-cc1" {{.*}} "-E" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.cui"
 // NORDC: "{{.*clang.*}}" "-cc1" {{.*}} "-emit-llvm-bc" {{.*}}  "-fcuda-include-gpubinary" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.bc"
-// RDC: "{{.*clang.*}}" "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.bc"
-// CHECK: "{{.*clang.*}}" "-cc1" {{.*}} "-S" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.s"
-// CHECK: "{{.*clang.*}}" "-cc1as" {{.*}} "-o" "hip-save-temps{{.*}}.o"
-// RDCC: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps.o"
-// RDC-NOUT: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=a.out.hipfb"
-// RDC-WOUT: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=executable.hipfb"
-// NOUT: "{{.*ld.*}}" {{.*}} "-o" "a.out"
-// WOUT: "{{.*ld.*}}" {{.*}} "-o" "executable"
+// NORDC: "{{.*clang.*}}" "-cc1" {{.*}} "-S" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.s"
+// NORDC: "{{.*clang.*}}" "-cc1as" {{.*}} "-o" "hip-save-temps{{.*}}.o"
+
+// output to default a.out or -o specified file name
+// NOUT: {{.*}}ld{{.*}}"-o" "a.out"
+// WOUT: {{.*}}ld{{.*}}"-o" "executable"
 

clang/test/Driver/hip-toolchain-rdc-separate.hip

 // RUN: %clang --hip-link -### -target x86_64-linux-gnu \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 \
 // RUN:   -fuse-ld=lld -fgpu-rdc -nogpuinc \
 // RUN:   %T/a.o %T/b.o \
 // RUN: 2>&1 | FileCheck -check-prefix=LINK %s
 
 // LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // LINK-SAME: "-targets=host-x86_64-unknown-linux-gnu,hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
-// LINK-SAME: "-inputs=[[A_O:.*a.o]]" "-outputs=[[A_OBJ_HOST:.*o]],[[A_BC1:.*o]],[[A_BC2:.*o]]"
+// LINK-SAME: "-inputs=[[A_O:.*a.o]]" "-outputs=[[A_OBJ_HOST:.*o]],{{.*o}},{{.*o}}"
 // LINK: "-unbundle"
 
 // LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // LINK-SAME: "-targets=host-x86_64-unknown-linux-gnu,hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
-// LINK-SAME: "-inputs=[[B_O:.*b.o]]" "-outputs=[[B_OBJ_HOST:.*o]],[[B_BC1:.*o]],[[B_BC2:.*o]]"
+// LINK-SAME: "-inputs=[[B_O:.*b.o]]" "-outputs=[[B_OBJ_HOST:.*o]],{{.*o}},{{.*o}}"
+// LINK: "-unbundle"
+
+// LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
+// LINK-SAME: "-targets=host-x86_64-unknown-linux-gnu,hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
+// LINK-SAME: "-inputs=[[A_O]]" "-outputs={{.*o}},[[A_BC1:.*o]],[[A_BC2:.*o]]"
+// LINK: "-unbundle"
+
+// LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
+// LINK-SAME: "-targets=host-x86_64-unknown-linux-gnu,hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
+// LINK-SAME: "-inputs=[[B_O]]" "-outputs={{.*o}},[[B_BC1:.*o]],[[B_BC2:.*o]]"
 // LINK: "-unbundle"
 
 // LINK-NOT: "*.llvm-link"
 // LINK-NOT: ".*opt"
 // LINK-NOT: ".*llc"
 // LINK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // LINK-SAME: "-o" "[[IMG_DEV1:.*.out]]" "[[A_BC1]]" "[[B_BC1]]"
 
 // LINK-NOT: "*.llvm-link"
 // LINK-NOT: ".*opt"
 // LINK-NOT: ".*llc"
 // LINK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // LINK-SAME: "-o" "[[IMG_DEV2:.*.out]]" "[[A_BC2]]" "[[B_BC2]]"
 
 // LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // LINK-SAME: "-targets={{.*}},hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
 // LINK-SAME: "-inputs={{.*}},[[IMG_DEV1]],[[IMG_DEV2]]" "-outputs=[[BUNDLE:.*hipfb]]"
 
-// LINK: [[LD:".*ld.*"]] {{.*}} "[[A_OBJ_HOST]]" "[[B_OBJ_HOST]]"
-// LINK-SAME: {{.*}} "-T" "{{.*}}.lk"
+// LINK: {{".*llvm-mc.*"}} "-triple" "amdgcn-amd-amdhsa" "-o"
+// LINK-SAME: "[[OBJBUNDLE:.*o]]" "{{.*}}.mcin" "--filetype=obj"
+
+// LINK: [[LD:".*ld.*"]] {{.*}} "-o" "a.out" {{.*}} "[[A_OBJ_HOST]]"
+// LINK-SAME: "[[B_OBJ_HOST]]" "[[OBJBUNDLE]]"

clang/test/Driver/hip-toolchain-rdc.hip

 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 // REQUIRES: amdgpu-registered-target
 
 // RUN: %clang -### -target x86_64-linux-gnu \
 // RUN:   -x hip --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 \
 // RUN:   --hip-device-lib=lib1.bc --hip-device-lib=lib2.bc \
 // RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib1 \
 // RUN:   --hip-device-lib-path=%S/Inputs/hip_multiple_inputs/lib2 \
 // RUN:   -fuse-ld=lld -fgpu-rdc -nogpuinc \
 // RUN:   %S/Inputs/hip_multiple_inputs/a.cu \
 // RUN:   %S/Inputs/hip_multiple_inputs/b.hip \
 // RUN: 2>&1 | FileCheck %s
 
-// CHECK: [[CLANG:".*clang.*"]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
+// emit objects for host side path
+// CHECK: [[CLANG:".*clang.*"]] "-cc1" "-triple" "x86_64-unknown-linux-gnu"
+// CHECK-SAME: "-aux-triple" "amdgcn-amd-amdhsa"
+// CHECK-SAME: "-emit-obj"
+// CHECK-SAME: {{.*}} "-main-file-name" "a.cu"
+// CHECK-SAME: {{.*}} "-o" [[A_OBJ_HOST:".*o"]] "-x" "hip"
+// CHECK-SAME: {{.*}} [[A_SRC:".*a.cu"]]
+
+// CHECK: [[CLANG]] "-cc1" "-triple" "x86_64-unknown-linux-gnu"
+// CHECK-SAME: "-aux-triple" "amdgcn-amd-amdhsa"
+// CHECK-SAME: "-emit-obj"
+// CHECK-SAME: {{.*}} "-main-file-name" "b.hip"
+// CHECK-SAME: {{.*}} "-o" [[B_OBJ_HOST:".*o"]] "-x" "hip"
+// CHECK-SAME: {{.*}} [[B_SRC:".*b.hip"]]
+
+// generate image for device side path on gfx803
+// CHECK: [[CLANG]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
 // CHECK-SAME: "-emit-llvm-bc"
 // CHECK-SAME: {{.*}} "-main-file-name" "a.cu"
 // CHECK-SAME: "-fcuda-is-device" "-fgpu-rdc" "-fcuda-allow-variadic-functions" "-fvisibility" "hidden"
 // CHECK-SAME: "-fapply-global-visibility-to-externs"
 // CHECK-SAME: "{{.*}}lib1.bc" "{{.*}}lib2.bc"
 // CHECK-SAME: "-target-cpu" "gfx803"
 // CHECK-SAME: {{.*}} "-o" [[A_BC1:".*bc"]] "-x" "hip"
-// CHECK-SAME: {{.*}} [[A_SRC:".*a.cu"]]
+// CHECK-SAME: {{.*}} [[A_SRC]]
 
 // CHECK: [[CLANG]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
 // CHECK-SAME: "-emit-llvm-bc"
 // CHECK-SAME: {{.*}} "-main-file-name" "b.hip"
 // CHECK-SAME: "-fcuda-is-device" "-fgpu-rdc" "-fcuda-allow-variadic-functions" "-fvisibility" "hidden"
 // CHECK-SAME: "-fapply-global-visibility-to-externs"
 // CHECK-SAME: "{{.*}}lib1.bc" "{{.*}}lib2.bc"
 // CHECK-SAME: "-target-cpu" "gfx803"
 // CHECK-SAME: {{.*}} "-o" [[B_BC1:".*bc"]] "-x" "hip"
-// CHECK-SAME: {{.*}} [[B_SRC:".*b.hip"]]
+// CHECK-SAME: {{.*}} [[B_SRC]]
 
 // CHECK-NOT: "*.llvm-link"
 // CHECK-NOT: ".*opt"
 // CHECK-NOT: ".*llc"
 // CHECK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "[[IMG_DEV1:.*.out]]" [[A_BC1]] [[B_BC1]]
 
+// generate image for device side path on gfx900
 // CHECK: [[CLANG]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
 // CHECK-SAME: "-emit-llvm-bc"
 // CHECK-SAME: {{.*}} "-main-file-name" "a.cu"
 // CHECK-SAME: "-fcuda-is-device" "-fgpu-rdc"
 // CHECK-SAME: "{{.*}}lib1.bc" "{{.*}}lib2.bc"
 // CHECK-SAME: "-target-cpu" "gfx900"
 // CHECK-SAME: {{.*}} "-o" [[A_BC2:".*bc"]] "-x" "hip"

 // CHECK-SAME: {{.*}} [[B_SRC]]
 
 // CHECK-NOT: "*.llvm-link"
 // CHECK-NOT: ".*opt"
 // CHECK-NOT: ".*llc"
 // CHECK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "[[IMG_DEV2:.*.out]]" [[A_BC2]] [[B_BC2]]
 
-// CHECK: [[CLANG]] "-cc1" "-triple" "x86_64-unknown-linux-gnu"
-// CHECK-SAME: "-aux-triple" "amdgcn-amd-amdhsa"
-// CHECK-SAME: "-emit-obj"
-// CHECK-SAME: {{.*}} "-main-file-name" "a.cu"
-// CHECK-SAME: {{.*}} "-o" [[A_OBJ_HOST:".*o"]] "-x" "hip"
-// CHECK-SAME: {{.*}} [[A_SRC]]
-
-// CHECK: [[CLANG]] "-cc1" "-triple" "x86_64-unknown-linux-gnu"
-// CHECK-SAME: "-aux-triple" "amdgcn-amd-amdhsa"
-// CHECK-SAME: "-emit-obj"
-// CHECK-SAME: {{.*}} "-main-file-name" "b.hip"
-// CHECK-SAME: {{.*}} "-o" [[B_OBJ_HOST:".*o"]] "-x" "hip"
-// CHECK-SAME: {{.*}} [[B_SRC]]
-
+// combine images generated into hip fat binary object
 // CHECK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // CHECK-SAME: "-targets={{.*}},hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
 // CHECK-SAME: "-inputs={{.*}},[[IMG_DEV1]],[[IMG_DEV2]]" "-outputs=[[BUNDLE:.*hipfb]]"
 
-// CHECK: [[LD:".*ld.*"]] {{.*}} [[A_OBJ_HOST]] [[B_OBJ_HOST]]
-// CHECK-SAME: {{.*}} "-T" "{{.*}}.lk"
+// CHECK: [[MC:".*llvm-mc"]] "-triple" "amdgcn-amd-amdhsa"
+// CHECK-SAME: "-o" [[OBJBUNDLE:".*o"]] "{{.*}}.mcin" "--filetype=obj"
+
+// output the executable
+// CHECK: [[LD:".*ld.*"]] {{.*}}"-o" "a.out" {{.*}} [[A_OBJ_HOST]] [[B_OBJ_HOST]] [[OBJBUNDLE]]