[cuda] Driver changes to compile and stitch together host and device-side CUDA code.

  NOTE: reverts r242077 to reinstate r242058, r242065, 242067
        and includes fix for OS X test failures.

  - Changed driver pipeline to compile host and device side of CUDA
    files and incorporate results of device-side compilation into host
    object file.

  - Added a test for cuda pipeline creation in clang driver.

  New clang options:
  --cuda-host-only   - Do host-side compilation only.
  --cuda-device-only - Do device-side compilation only.

  --cuda-gpu-arch=<ARCH> - specify GPU architecture for device-side
    compilation. E.g. sm_35, sm_30. Default is sm_20. May be used more
    than once in which case one device-compilation will be done per
    unique specified GPU architecture.

  Differential Revision: http://reviews.llvm.org/D9509

llvm-svn: 242085
diff --git a/clang/lib/Driver/Action.cpp b/clang/lib/Driver/Action.cpp
index 360dbee..3219dc1 100644
--- a/clang/lib/Driver/Action.cpp
+++ b/clang/lib/Driver/Action.cpp
@@ -24,6 +24,8 @@
   switch (AC) {
   case InputClass: return "input";
   case BindArchClass: return "bind-arch";
+  case CudaDeviceClass: return "cuda-device";
+  case CudaHostClass: return "cuda-host";
   case PreprocessJobClass: return "preprocessor";
   case PrecompileJobClass: return "precompiler";
   case AnalyzeJobClass: return "analyzer";
@@ -53,6 +55,25 @@
                                const char *_ArchName)
     : Action(BindArchClass, std::move(Input)), ArchName(_ArchName) {}
 
+void CudaDeviceAction::anchor() {}
+
+CudaDeviceAction::CudaDeviceAction(std::unique_ptr<Action> Input,
+                                   const char *ArchName, bool AtTopLevel)
+    : Action(CudaDeviceClass, std::move(Input)), GpuArchName(ArchName),
+      AtTopLevel(AtTopLevel) {}
+
+void CudaHostAction::anchor() {}
+
+CudaHostAction::CudaHostAction(std::unique_ptr<Action> Input,
+                               const ActionList &_DeviceActions)
+    : Action(CudaHostClass, std::move(Input)), DeviceActions(_DeviceActions) {}
+
+CudaHostAction::~CudaHostAction() {
+  for (iterator it = DeviceActions.begin(), ie = DeviceActions.end(); it != ie;
+       ++it)
+    delete *it;
+}
+
 void JobAction::anchor() {}
 
 JobAction::JobAction(ActionClass Kind, std::unique_ptr<Action> Input,
diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
index f3ec151..180c412 100644
--- a/clang/lib/Driver/Driver.cpp
+++ b/clang/lib/Driver/Driver.cpp
@@ -174,8 +174,10 @@
   } else if ((PhaseArg = DAL.getLastArg(options::OPT_S))) {
     FinalPhase = phases::Backend;
 
-    // -c only runs up to the assembler.
-  } else if ((PhaseArg = DAL.getLastArg(options::OPT_c))) {
+    // -c and partial CUDA compilations only run up to the assembler.
+  } else if ((PhaseArg = DAL.getLastArg(options::OPT_c)) ||
+             (PhaseArg = DAL.getLastArg(options::OPT_cuda_device_only)) ||
+             (PhaseArg = DAL.getLastArg(options::OPT_cuda_host_only))) {
     FinalPhase = phases::Assemble;
 
     // Otherwise do everything.
@@ -900,9 +902,20 @@
   } else if (BindArchAction *BIA = dyn_cast<BindArchAction>(A)) {
     os << '"' << BIA->getArchName() << '"' << ", {"
        << PrintActions1(C, *BIA->begin(), Ids) << "}";
+  } else if (CudaDeviceAction *CDA = dyn_cast<CudaDeviceAction>(A)) {
+    os << '"' << CDA->getGpuArchName() << '"' << ", {"
+       << PrintActions1(C, *CDA->begin(), Ids) << "}";
   } else {
+    ActionList *AL;
+    if (CudaHostAction *CHA = dyn_cast<CudaHostAction>(A)) {
+      os << "{" << PrintActions1(C, *CHA->begin(), Ids) << "}"
+         << ", gpu binaries ";
+      AL = &CHA->getDeviceActions();
+    } else
+      AL = &A->getInputs();
+
     const char *Prefix = "{";
-    for (Action *PreRequisite : *A) {
+    for (Action *PreRequisite : *AL) {
       os << Prefix << PrintActions1(C, PreRequisite, Ids);
       Prefix = ", ";
     }
@@ -1215,6 +1228,93 @@
   }
 }
 
+// For each unique --cuda-gpu-arch= argument creates a TY_CUDA_DEVICE input
+// action and then wraps each in CudaDeviceAction paired with appropriate GPU
+// arch name. If we're only building device-side code, each action remains
+// independent. Otherwise we pass device-side actions as inputs to a new
+// CudaHostAction which combines both host and device side actions.
+static std::unique_ptr<Action>
+buildCudaActions(const Driver &D, const ToolChain &TC, DerivedArgList &Args,
+                 const Arg *InputArg, const types::ID InputType,
+                 std::unique_ptr<Action> Current, ActionList &Actions) {
+
+  assert(InputType == types::TY_CUDA &&
+         "CUDA Actions only apply to CUDA inputs.");
+
+  // Collect all cuda_gpu_arch parameters, removing duplicates.
+  SmallVector<const char *, 4> GpuArchList;
+  llvm::StringSet<> GpuArchNames;
+  for (Arg *A : Args) {
+    if (A->getOption().matches(options::OPT_cuda_gpu_arch_EQ)) {
+      A->claim();
+      if (GpuArchNames.insert(A->getValue()).second)
+        GpuArchList.push_back(A->getValue());
+    }
+  }
+
+  // Default to sm_20 which is the lowest common denominator for supported GPUs.
+  // sm_20 code should work correctly, if suboptimally, on all newer GPUs.
+  if (GpuArchList.empty())
+    GpuArchList.push_back("sm_20");
+
+  // Replicate inputs for each GPU architecture.
+  Driver::InputList CudaDeviceInputs;
+  for (unsigned i = 0, e = GpuArchList.size(); i != e; ++i)
+    CudaDeviceInputs.push_back(std::make_pair(types::TY_CUDA_DEVICE, InputArg));
+
+  // Build actions for all device inputs.
+  ActionList CudaDeviceActions;
+  D.BuildActions(TC, Args, CudaDeviceInputs, CudaDeviceActions);
+  assert(GpuArchList.size() == CudaDeviceActions.size() &&
+         "Failed to create actions for all devices");
+
+  // Check whether any of device actions stopped before they could generate PTX.
+  bool PartialCompilation = false;
+  bool DeviceOnlyCompilation = Args.hasArg(options::OPT_cuda_device_only);
+  for (unsigned i = 0, e = GpuArchList.size(); i != e; ++i) {
+    if (CudaDeviceActions[i]->getKind() != Action::BackendJobClass) {
+      PartialCompilation = true;
+      break;
+    }
+  }
+
+  // Figure out what to do with device actions -- pass them as inputs to the
+  // host action or run each of them independently.
+  if (PartialCompilation || DeviceOnlyCompilation) {
+    // In case of partial or device-only compilation results of device actions
+    // are not consumed by the host action device actions have to be added to
+    // top-level actions list with AtTopLevel=true and run independently.
+
+    // -o is ambiguous if we have more than one top-level action.
+    if (Args.hasArg(options::OPT_o) &&
+        (!DeviceOnlyCompilation || GpuArchList.size() > 1)) {
+      D.Diag(clang::diag::err_drv_output_argument_with_multiple_files);
+      return nullptr;
+    }
+
+    for (unsigned i = 0, e = GpuArchList.size(); i != e; ++i)
+      Actions.push_back(
+          new CudaDeviceAction(std::unique_ptr<Action>(CudaDeviceActions[i]),
+                               GpuArchList[i], /* AtTopLevel */ true));
+    // Kill host action in case of device-only compilation.
+    if (DeviceOnlyCompilation)
+      Current.reset(nullptr);
+    return Current;
+  } else {
+    // Outputs of device actions during complete CUDA compilation get created
+    // with AtTopLevel=false and become inputs for the host action.
+    ActionList DeviceActions;
+    for (unsigned i = 0, e = GpuArchList.size(); i != e; ++i)
+      DeviceActions.push_back(
+          new CudaDeviceAction(std::unique_ptr<Action>(CudaDeviceActions[i]),
+                               GpuArchList[i], /* AtTopLevel */ false));
+    // Return a new host action that incorporates original host action and all
+    // device actions.
+    return std::unique_ptr<Action>(
+        new CudaHostAction(std::move(Current), DeviceActions));
+  }
+}
+
 void Driver::BuildActions(const ToolChain &TC, DerivedArgList &Args,
                           const InputList &Inputs, ActionList &Actions) const {
   llvm::PrettyStackTraceString CrashInfo("Building compilation actions");
@@ -1312,6 +1412,25 @@
       continue;
     }
 
+    phases::ID CudaInjectionPhase;
+    if (isSaveTempsEnabled()) {
+      // All phases are done independently, inject GPU blobs during compilation
+      // phase as that's where we generate glue code to init them.
+      CudaInjectionPhase = phases::Compile;
+    } else {
+      // Assumes that clang does everything up until linking phase, so we inject
+      // cuda device actions at the last step before linking. Otherwise CUDA
+      // host action forces preprocessor into a separate invocation.
+      if (FinalPhase == phases::Link) {
+        for (auto i = PL.begin(), e = PL.end(); i != e; ++i) {
+          auto next = i + 1;
+          if (next != e && *next == phases::Link)
+            CudaInjectionPhase = *i;
+        }
+      } else
+        CudaInjectionPhase = FinalPhase;
+    }
+
     // Build the pipeline for this file.
     std::unique_ptr<Action> Current(new InputAction(*InputArg, InputType));
     for (SmallVectorImpl<phases::ID>::iterator i = PL.begin(), e = PL.end();
@@ -1337,6 +1456,15 @@
 
       // Otherwise construct the appropriate action.
       Current = ConstructPhaseAction(TC, Args, Phase, std::move(Current));
+
+      if (InputType == types::TY_CUDA && Phase == CudaInjectionPhase &&
+          !Args.hasArg(options::OPT_cuda_host_only)) {
+        Current = buildCudaActions(*this, TC, Args, InputArg, InputType,
+                                   std::move(Current), Actions);
+        if (!Current)
+          break;
+      }
+
       if (Current->getType() == types::TY_Nothing)
         break;
     }
@@ -1576,7 +1704,13 @@
   if (isa<BackendJobAction>(JA)) {
     // Check if the compiler supports emitting LLVM IR.
     assert(Inputs->size() == 1);
-    JobAction *CompileJA = cast<CompileJobAction>(*Inputs->begin());
+    JobAction *CompileJA;
+    // Extract real host action, if it's a CudaHostAction.
+    if (CudaHostAction *CudaHA = dyn_cast<CudaHostAction>(*Inputs->begin()))
+      CompileJA = cast<CompileJobAction>(*CudaHA->begin());
+    else
+      CompileJA = cast<CompileJobAction>(*Inputs->begin());
+
     const Tool *Compiler = TC->SelectTool(*CompileJA);
     if (!Compiler)
       return nullptr;
@@ -1610,6 +1744,20 @@
                                 InputInfo &Result) const {
   llvm::PrettyStackTraceString CrashInfo("Building compilation jobs");
 
+  InputInfoList CudaDeviceInputInfos;
+  if (const CudaHostAction *CHA = dyn_cast<CudaHostAction>(A)) {
+    InputInfo II;
+    // Append outputs of device jobs to the input list.
+    for (const Action *DA : CHA->getDeviceActions()) {
+      BuildJobsForAction(C, DA, TC, "", AtTopLevel,
+                         /*MultipleArchs*/ false, LinkingOutput, II);
+      CudaDeviceInputInfos.push_back(II);
+    }
+    // Override current action with a real host compile action and continue
+    // processing it.
+    A = *CHA->begin();
+  }
+
   if (const InputAction *IA = dyn_cast<InputAction>(A)) {
     // FIXME: It would be nice to not claim this here; maybe the old scheme of
     // just using Args was better?
@@ -1635,11 +1783,24 @@
     else
       TC = &C.getDefaultToolChain();
 
-    BuildJobsForAction(C, *BAA->begin(), TC, BAA->getArchName(), AtTopLevel,
+    BuildJobsForAction(C, *BAA->begin(), TC, ArchName, AtTopLevel,
                        MultipleArchs, LinkingOutput, Result);
     return;
   }
 
+  if (const CudaDeviceAction *CDA = dyn_cast<CudaDeviceAction>(A)) {
+    // Figure out which NVPTX triple to use for device-side compilation based on
+    // whether host is 64-bit.
+    llvm::Triple DeviceTriple(C.getDefaultToolChain().getTriple().isArch64Bit()
+                                  ? "nvptx64-nvidia-cuda"
+                                  : "nvptx-nvidia-cuda");
+    BuildJobsForAction(C, *CDA->begin(),
+                       &getToolChain(C.getArgs(), DeviceTriple),
+                       CDA->getGpuArchName(), CDA->isAtTopLevel(),
+                       /*MultipleArchs*/ true, LinkingOutput, Result);
+    return;
+  }
+
   const ActionList *Inputs = &A->getInputs();
 
   const JobAction *JA = cast<JobAction>(A);
@@ -1671,6 +1832,10 @@
   if (JA->getType() == types::TY_dSYM)
     BaseInput = InputInfos[0].getFilename();
 
+  // Append outputs of cuda device jobs to the input list
+  if (CudaDeviceInputInfos.size())
+    InputInfos.append(CudaDeviceInputInfos.begin(), CudaDeviceInputInfos.end());
+
   // Determine the place to write output to, if any.
   if (JA->getType() == types::TY_Nothing)
     Result = InputInfo(A->getType(), BaseInput);
@@ -2052,6 +2217,9 @@
         break;
       }
       break;
+    case llvm::Triple::CUDA:
+      TC = new toolchains::CudaToolChain(*this, Target, Args);
+      break;
     default:
       // Of these targets, Hexagon is the only one that might have
       // an OS of Linux, in which case it got handled above already.
diff --git a/clang/lib/Driver/ToolChain.cpp b/clang/lib/Driver/ToolChain.cpp
index da020a2..e6a1bc9 100644
--- a/clang/lib/Driver/ToolChain.cpp
+++ b/clang/lib/Driver/ToolChain.cpp
@@ -151,6 +151,8 @@
 
   case Action::InputClass:
   case Action::BindArchClass:
+  case Action::CudaDeviceClass:
+  case Action::CudaHostClass:
   case Action::LipoJobClass:
   case Action::DsymutilJobClass:
   case Action::VerifyDebugInfoJobClass:
diff --git a/clang/lib/Driver/ToolChains.cpp b/clang/lib/Driver/ToolChains.cpp
index eafc72b..15e36a1 100644
--- a/clang/lib/Driver/ToolChains.cpp
+++ b/clang/lib/Driver/ToolChains.cpp
@@ -3652,6 +3652,65 @@
   return new tools::dragonfly::Linker(*this);
 }
 
+/// Stub for CUDA toolchain. At the moment we don't have assembler or
+/// linker and need toolchain mainly to propagate device-side options
+/// to CC1.
+
+CudaToolChain::CudaToolChain(const Driver &D, const llvm::Triple &Triple,
+                             const ArgList &Args)
+    : Linux(D, Triple, Args) {}
+
+void
+CudaToolChain::addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
+                                     llvm::opt::ArgStringList &CC1Args) const {
+  Linux::addClangTargetOptions(DriverArgs, CC1Args);
+  CC1Args.push_back("-fcuda-is-device");
+}
+
+llvm::opt::DerivedArgList *
+CudaToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
+                             const char *BoundArch) const {
+  DerivedArgList *DAL = new DerivedArgList(Args.getBaseArgs());
+  const OptTable &Opts = getDriver().getOpts();
+
+  for (Arg *A : Args) {
+    if (A->getOption().matches(options::OPT_Xarch__)) {
+      // Skip this argument unless the architecture matches BoundArch
+      if (A->getValue(0) != StringRef(BoundArch))
+        continue;
+
+      unsigned Index = Args.getBaseArgs().MakeIndex(A->getValue(1));
+      unsigned Prev = Index;
+      std::unique_ptr<Arg> XarchArg(Opts.ParseOneArg(Args, Index));
+
+      // If the argument parsing failed or more than one argument was
+      // consumed, the -Xarch_ argument's parameter tried to consume
+      // extra arguments. Emit an error and ignore.
+      //
+      // We also want to disallow any options which would alter the
+      // driver behavior; that isn't going to work in our model. We
+      // use isDriverOption() as an approximation, although things
+      // like -O4 are going to slip through.
+      if (!XarchArg || Index > Prev + 1) {
+        getDriver().Diag(diag::err_drv_invalid_Xarch_argument_with_args)
+            << A->getAsString(Args);
+        continue;
+      } else if (XarchArg->getOption().hasFlag(options::DriverOption)) {
+        getDriver().Diag(diag::err_drv_invalid_Xarch_argument_isdriver)
+            << A->getAsString(Args);
+        continue;
+      }
+      XarchArg->setBaseArg(A);
+      A = XarchArg.release();
+      DAL->AddSynthesizedArg(A);
+    }
+    DAL->append(A);
+  }
+
+  DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ), BoundArch);
+  return DAL;
+}
+
 /// XCore tool chain
 XCore::XCore(const Driver &D, const llvm::Triple &Triple, const ArgList &Args)
     : ToolChain(D, Triple, Args) {
diff --git a/clang/lib/Driver/ToolChains.h b/clang/lib/Driver/ToolChains.h
index 3689682..327ff9b 100644
--- a/clang/lib/Driver/ToolChains.h
+++ b/clang/lib/Driver/ToolChains.h
@@ -699,6 +699,18 @@
   std::string computeSysRoot() const;
 };
 
+class LLVM_LIBRARY_VISIBILITY CudaToolChain : public Linux {
+public:
+  CudaToolChain(const Driver &D, const llvm::Triple &Triple,
+                const llvm::opt::ArgList &Args);
+
+  llvm::opt::DerivedArgList *
+  TranslateArgs(const llvm::opt::DerivedArgList &Args,
+                const char *BoundArch) const override;
+  void addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
+                             llvm::opt::ArgStringList &CC1Args) const override;
+};
+
 class LLVM_LIBRARY_VISIBILITY Hexagon_TC : public Linux {
 protected:
   GCCVersion GCCLibAndIncVersion;
diff --git a/clang/lib/Driver/Tools.cpp b/clang/lib/Driver/Tools.cpp
index bf9b4ba..c6dc178 100644
--- a/clang/lib/Driver/Tools.cpp
+++ b/clang/lib/Driver/Tools.cpp
@@ -1488,6 +1488,12 @@
     return CPUName;
   }
 
+  case llvm::Triple::nvptx:
+  case llvm::Triple::nvptx64:
+    if (const Arg *A = Args.getLastArg(options::OPT_march_EQ))
+      return A->getValue();
+    return "";
+
   case llvm::Triple::ppc:
   case llvm::Triple::ppc64:
   case llvm::Triple::ppc64le: {
@@ -2826,8 +2832,14 @@
       getToolChain().getTriple().isWindowsCygwinEnvironment();
   bool IsWindowsMSVC = getToolChain().getTriple().isWindowsMSVCEnvironment();
 
-  assert(Inputs.size() == 1 && "Unable to handle multiple inputs.");
+  // Check number of inputs for sanity. We need at least one input.
+  assert(Inputs.size() >= 1 && "Must have at least one input.");
   const InputInfo &Input = Inputs[0];
+  // CUDA compilation may have multiple inputs (source file + results of
+  // device-side compilations). All other jobs are expected to have exactly one
+  // input.
+  bool IsCuda = types::isCuda(Input.getType());
+  assert((IsCuda || Inputs.size() == 1) && "Unable to handle multiple inputs.");
 
   // Invoke ourselves in -cc1 mode.
   //
@@ -4812,14 +4824,12 @@
     assert(Output.isNothing() && "Invalid output.");
   }
 
-  for (const auto &II : Inputs) {
-    addDashXForInput(Args, II, CmdArgs);
+  addDashXForInput(Args, Input, CmdArgs);
 
-    if (II.isFilename())
-      CmdArgs.push_back(II.getFilename());
-    else
-      II.getInputArg().renderAsInput(Args, CmdArgs);
-  }
+  if (Input.isFilename())
+    CmdArgs.push_back(Input.getFilename());
+  else
+    Input.getInputArg().renderAsInput(Args, CmdArgs);
 
   Args.AddAllArgs(CmdArgs, options::OPT_undef);
 
@@ -4857,6 +4867,16 @@
     CmdArgs.push_back(SplitDwarfOut);
   }
 
+  // Host-side cuda compilation receives device-side outputs as Inputs[1...].
+  // Include them with -fcuda-include-gpubinary.
+  if (IsCuda && Inputs.size() > 1)
+    for (InputInfoList::const_iterator it = std::next(Inputs.begin()),
+                                       ie = Inputs.end();
+         it != ie; ++it) {
+      CmdArgs.push_back("-fcuda-include-gpubinary");
+      CmdArgs.push_back(it->getFilename());
+    }
+
   // Finally add the compile command to the compilation.
   if (Args.hasArg(options::OPT__SLASH_fallback) &&
       Output.getType() == types::TY_Object &&
diff --git a/clang/lib/Driver/Types.cpp b/clang/lib/Driver/Types.cpp
index 7b28145..2085b01 100644
--- a/clang/lib/Driver/Types.cpp
+++ b/clang/lib/Driver/Types.cpp
@@ -86,6 +86,7 @@
   case TY_C: case TY_PP_C:
   case TY_CL:
   case TY_CUDA: case TY_PP_CUDA:
+  case TY_CUDA_DEVICE:
   case TY_ObjC: case TY_PP_ObjC: case TY_PP_ObjC_Alias:
   case TY_CXX: case TY_PP_CXX:
   case TY_ObjCXX: case TY_PP_ObjCXX: case TY_PP_ObjCXX_Alias:
@@ -122,7 +123,19 @@
   case TY_ObjCXX: case TY_PP_ObjCXX: case TY_PP_ObjCXX_Alias:
   case TY_CXXHeader: case TY_PP_CXXHeader:
   case TY_ObjCXXHeader: case TY_PP_ObjCXXHeader:
-  case TY_CUDA: case TY_PP_CUDA:
+  case TY_CUDA: case TY_PP_CUDA: case TY_CUDA_DEVICE:
+    return true;
+  }
+}
+
+bool types::isCuda(ID Id) {
+  switch (Id) {
+  default:
+    return false;
+
+  case TY_CUDA:
+  case TY_PP_CUDA:
+  case TY_CUDA_DEVICE:
     return true;
   }
 }
@@ -206,10 +219,12 @@
         P.push_back(phases::Compile);
         P.push_back(phases::Backend);
       }
-      P.push_back(phases::Assemble);
+      if (Id != TY_CUDA_DEVICE)
+        P.push_back(phases::Assemble);
     }
   }
-  if (!onlyPrecompileType(Id)) {
+
+  if (!onlyPrecompileType(Id) && Id != TY_CUDA_DEVICE) {
     P.push_back(phases::Link);
   }
   assert(0 < P.size() && "Not enough phases in list");