[CUDA] Driver changes to pass flags needed to use detected CUDA installation.

[tra]

Summary

• added include paths to detected CUDA installation
• added include paths for -aux-triple
• added -nocudainc to prevent addition of CUDA include path.
• added test cases to verify new functionality

[tra]

Updated to reflect latest changes in D13144

[tra]

Moved libdevice detection and linking into a separate patch D14556.

[echristo]

This needs a very long commit message describing exactly what's going on and why since it's not documented anywhere. Possibly in the code as well at AddPreprocessingOptions and other places you add AuxToolChain. Some of this is going to need to get refactored in the medium term when we look at being able to preprocess and add multiple target argument passing, but this is an ok intermediate step.

Thanks!

-eric

include/clang/Driver/Driver.h

public:
   std::string GetTemporaryPath(StringRef Prefix, const char *Suffix) const;
 
   /// ShouldUseClangCompiler - Should the clang compiler be used to
   /// handle this action.
   bool ShouldUseClangCompiler(const JobAction &JA) const;
 
   bool IsUsingLTO(const llvm::opt::ArgList &Args) const;
 
-private:
   /// \brief Retrieves a ToolChain for a particular \p Target triple.
   ///
   /// Will cache ToolChains for the life of the driver object, and create them
   /// on-demand.
   const ToolChain &getToolChain(const llvm::opt::ArgList &Args,
                                 const llvm::Triple &Target) const;
 
   /// @}
 
+private:
   /// \brief Get bitmasks for which option flags to include and exclude based on
   /// the driver mode.
   std::pair<unsigned, unsigned> getIncludeExcludeOptionFlagMasks() const;
 
 public:
   /// GetReleaseVersion - Parse (([0-9]+)(.([0-9]+)(.([0-9]+)?))?)? and
   /// return the grouped values as integers. Numbers which are not
   /// provided are set to 0.

include/clang/Driver/Options.td

 def no_canonical_prefixes : Flag<["-"], "no-canonical-prefixes">, Flags<[HelpHidden]>,
   HelpText<"Use relative instead of canonical paths">;
 def no_cpp_precomp : Flag<["-"], "no-cpp-precomp">, Group<clang_ignored_f_Group>;
 def no_integrated_cpp : Flag<["-", "--"], "no-integrated-cpp">, Flags<[DriverOption]>;
 def no_pedantic : Flag<["-", "--"], "no-pedantic">, Group<pedantic_Group>;
 def no__dead__strip__inits__and__terms : Flag<["-"], "no_dead_strip_inits_and_terms">;
 def nobuiltininc : Flag<["-"], "nobuiltininc">, Flags<[CC1Option]>,
   HelpText<"Disable builtin #include directories">;
+def nocudainc : Flag<["-"], "nocudainc">;
+def nocudalib : Flag<["-"], "nocudalib">;
 def nodefaultlibs : Flag<["-"], "nodefaultlibs">;
 def nofixprebinding : Flag<["-"], "nofixprebinding">;
 def nolibc : Flag<["-"], "nolibc">;
 def nomultidefs : Flag<["-"], "nomultidefs">;
 def nopie : Flag<["-"], "nopie">;
 def noprebind : Flag<["-"], "noprebind">;
 def noseglinkedit : Flag<["-"], "noseglinkedit">;
 def nostartfiles : Flag<["-"], "nostartfiles">;

include/clang/Driver/ToolChain.h

public:
   /// AddFastMathRuntimeIfAvailable - If a runtime library exists that sets
   /// global flags for unsafe floating point math, add it and return true.
   ///
   /// This checks for presence of the -Ofast, -ffast-math or -funsafe-math flags.
   virtual bool
   AddFastMathRuntimeIfAvailable(const llvm::opt::ArgList &Args,
                                 llvm::opt::ArgStringList &CmdArgs) const;
 
+  /// \brief Add arguments to use system-specific CUDA includes.
+  virtual void AddCudaIncludeArgs(const llvm::opt::ArgList &DriverArgs,
+                                  llvm::opt::ArgStringList &CC1Args) const;
+
   /// \brief Return sanitizers which are available in this toolchain.
   virtual SanitizerMask getSupportedSanitizers() const;
 };
 
 } // end namespace driver
 } // end namespace clang
 
 #endif

lib/Driver/ToolChain.cpp

SanitizerMask ToolChain::getSupportedSanitizers() const {
   using namespace SanitizerKind;
   SanitizerMask Res = (Undefined & ~Vptr & ~Function) | (CFI & ~CFIICall) |
                       CFICastStrict | UnsignedIntegerOverflow | LocalBounds;
   if (getTriple().getArch() == llvm::Triple::x86 ||
       getTriple().getArch() == llvm::Triple::x86_64)
     Res |= CFIICall;
   return Res;
 }
+
+void ToolChain::AddCudaIncludeArgs(const ArgList &DriverArgs,
+                                   ArgStringList &CC1Args) const {}

lib/Driver/ToolChains.h

protected:
   // \brief A class to find a viable CUDA installation
 
   class CudaInstallationDetector {
     bool IsValid;
     std::string CudaInstallPath;
     std::string CudaLibPath;
     std::string CudaLibDevicePath;
     std::string CudaIncludePath;
+    llvm::StringMap<std::string> CudaLibDeviceMap;
 
   public:
     CudaInstallationDetector() : IsValid(false) {}
     void init(const Driver &D, const llvm::Triple &TargetTriple,
               const llvm::opt::ArgList &Args);
 
     /// \brief Check whether we detected a valid Cuda install.
     bool isValid() const { return IsValid; }
     /// \brief Print information about the detected CUDA installation.
     void print(raw_ostream &OS) const;
 
     /// \brief Get the detected Cuda installation path.
     StringRef getInstallPath() const { return CudaInstallPath; }
     /// \brief Get the detected Cuda Include path.
     StringRef getIncludePath() const { return CudaIncludePath; }
     /// \brief Get the detected Cuda library path.
     StringRef getLibPath() const { return CudaLibPath; }
     /// \brief Get the detected Cuda device library path.
     StringRef getLibDevicePath() const { return CudaLibDevicePath; }
     /// \brief Get libdevice file for given architecture
+    StringRef getLibDeviceFile(StringRef Gpu) const {
+      return CudaLibDeviceMap.lookup(Gpu);
+    }
   };
 
   CudaInstallationDetector CudaInstallation;
 
 public:
   Generic_GCC(const Driver &D, const llvm::Triple &Triple,
               const llvm::opt::ArgList &Args);
   ~Generic_GCC() override;
public:
   bool HasNativeLLVMSupport() const override;
 
   void
   AddClangSystemIncludeArgs(const llvm::opt::ArgList &DriverArgs,
                             llvm::opt::ArgStringList &CC1Args) const override;
   void AddClangCXXStdlibIncludeArgs(
       const llvm::opt::ArgList &DriverArgs,
       llvm::opt::ArgStringList &CC1Args) const override;
+  void AddCudaIncludeArgs(const llvm::opt::ArgList &DriverArgs,
+                          llvm::opt::ArgStringList &CC1Args) const override;
   bool isPIEDefault() const override;
   SanitizerMask getSupportedSanitizers() const override;
 
   std::string Linker;
   std::vector<std::string> ExtraOpts;
 
 protected:
   Tool *buildAssembler() const override;

lib/Driver/ToolChains.cpp

for (const auto &CudaPath : CudaPathCandidates) {
     CudaLibPath =
         CudaInstallPath + (TargetTriple.isArch64Bit() ? "/lib64" : "/lib");
 
     if (!(llvm::sys::fs::exists(CudaIncludePath) &&
           llvm::sys::fs::exists(CudaLibPath) &&
           llvm::sys::fs::exists(CudaLibDevicePath)))
       continue;
 
+    const StringRef LibDeviceName = "libdevice.";
+    std::error_code EC;
+    for (llvm::sys::fs::directory_iterator LI(CudaLibDevicePath, EC), LE;
+         !EC && LI != LE; LI = LI.increment(EC)) {
+      StringRef FilePath = LI->path();
+      StringRef FileName = llvm::sys::path::filename(FilePath);
+      // Process all bitcode filenames that look like libdevice.compute_XX.YY.bc
+      if (!(FileName.startswith(LibDeviceName) && FileName.endswith(".bc")))
+        continue;
+      StringRef GpuArch = FileName.slice(
+          LibDeviceName.size(), FileName.find('.', LibDeviceName.size()));
+      CudaLibDeviceMap[GpuArch] = FilePath.str();
+      // Insert map entries for specifc devices with this compute capability.
+      if (GpuArch == "compute_20") {
+        CudaLibDeviceMap["sm_20"] = FilePath;
+        CudaLibDeviceMap["sm_21"] = FilePath;
+      } else if (GpuArch == "compute_30") {
+        CudaLibDeviceMap["sm_30"] = FilePath;
+        CudaLibDeviceMap["sm_32"] = FilePath;
+      } else if (GpuArch == "compute_35") {
+        CudaLibDeviceMap["sm_35"] = FilePath;
+        CudaLibDeviceMap["sm_37"] = FilePath;
+      }
+    }
+
     IsValid = true;
     break;
   }
 }
 
 void Generic_GCC::CudaInstallationDetector::print(raw_ostream &OS) const {
   if (isValid())
     OS << "Found CUDA installation: " << CudaInstallPath << "\n";
for (const auto &IncludePath : LibStdCXXIncludePathCandidates) {
     if (addLibStdCXXIncludePaths(IncludePath, /*Suffix*/ "", TripleStr,
                                  /*GCCMultiarchTriple*/ "",
                                  /*TargetMultiarchTriple*/ "",
                                  Multilib.includeSuffix(), DriverArgs, CC1Args))
       break;
   }
 }
 
+void Linux::AddCudaIncludeArgs(const ArgList &DriverArgs,
+                               ArgStringList &CC1Args) const {
+  if (DriverArgs.hasArg(options::OPT_nocudainc))
+    return;
+
+  if (CudaInstallation.isValid())
+    addSystemInclude(DriverArgs, CC1Args, CudaInstallation.getIncludePath());
+}
+
 bool Linux::isPIEDefault() const { return getSanitizerArgs().requiresPIE(); }
 
 SanitizerMask Linux::getSupportedSanitizers() const {
   const bool IsX86 = getTriple().getArch() == llvm::Triple::x86;
   const bool IsX86_64 = getTriple().getArch() == llvm::Triple::x86_64;
   const bool IsMIPS64 = getTriple().getArch() == llvm::Triple::mips64 ||
                         getTriple().getArch() == llvm::Triple::mips64el;
   const bool IsPowerPC64 = getTriple().getArch() == llvm::Triple::ppc64 ||
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");
+
+  if (DriverArgs.hasArg(options::OPT_nocudalib))
+    return;
+
+  std::string LibDeviceFile = CudaInstallation.getLibDeviceFile(
+      DriverArgs.getLastArgValue(options::OPT_march_EQ));
+  if (!LibDeviceFile.empty()) {
+    CC1Args.push_back("-mlink-bitcode-file");
+    CC1Args.push_back(DriverArgs.MakeArgString(LibDeviceFile));
+    CC1Args.push_back("-fcuda-uses-libdevice");
+
+    // Libdevice in CUDA-7.0 requires PTX version that's more recent
+    // than LLVM defaults to. Use PTX4.2 which is the PTX version that
+    // came with CUDA-7.0.
+    CC1Args.push_back("-target-feature");
+    CC1Args.push_back("+ptx42");
+  }
 }
 
 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();
 

lib/Driver/Tools.h

public:
   static const char *getDependencyFileName(const llvm::opt::ArgList &Args,
                                            const InputInfoList &Inputs);
 
 private:
   void AddPreprocessingOptions(Compilation &C, const JobAction &JA,
                                const Driver &D, const llvm::opt::ArgList &Args,
                                llvm::opt::ArgStringList &CmdArgs,
                                const InputInfo &Output,
-                               const InputInfoList &Inputs) const;
+                               const InputInfoList &Inputs,
+                               const char *AuxTriple) const;
 
   void AddAArch64TargetArgs(const llvm::opt::ArgList &Args,
                             llvm::opt::ArgStringList &CmdArgs) const;
   void AddARMTargetArgs(const llvm::Triple &Triple,
                         const llvm::opt::ArgList &Args,
                         llvm::opt::ArgStringList &CmdArgs,
                         bool KernelOrKext) const;
   void AddARM64TargetArgs(const llvm::opt::ArgList &Args,

lib/Driver/Tools.cpp

static bool forwardToGCC(const Option &O) {
   return O.getKind() != Option::InputClass &&
          !O.hasFlag(options::DriverOption) && !O.hasFlag(options::LinkerInput);
 }
 
 void Clang::AddPreprocessingOptions(Compilation &C, const JobAction &JA,
                                     const Driver &D, const ArgList &Args,
                                     ArgStringList &CmdArgs,
                                     const InputInfo &Output,
-                                    const InputInfoList &Inputs) const {
+                                    const InputInfoList &Inputs,
+                                    const char *AuxTriple) const {
   Arg *A;
 
   CheckPreprocessingOptions(D, Args);
 
   Args.AddLastArg(CmdArgs, options::OPT_C);
   Args.AddLastArg(CmdArgs, options::OPT_CC);
 
   // Handle dependency file generation.
void Clang::AddPreprocessingOptions(Compilation &C, const JobAction &JA,
   addDirectoryList(Args, CmdArgs, "-objcxx-isystem", "OBJCPLUS_INCLUDE_PATH");
 
   // Add C++ include arguments, if needed.
   if (types::isCXX(Inputs[0].getType()))
     getToolChain().AddClangCXXStdlibIncludeArgs(Args, CmdArgs);
 
   // Add system include arguments.
   getToolChain().AddClangSystemIncludeArgs(Args, CmdArgs);
+
+  // Add CUDA include arguments
+  if (types::isCuda(Inputs[0].getType())) {
+    getToolChain().AddCudaIncludeArgs(Args, CmdArgs);
+    if (AuxTriple) {
+      const ToolChain &AuxTC = D.getToolChain(Args, llvm::Triple(AuxTriple));
+      AuxTC.AddClangCXXStdlibIncludeArgs(Args, CmdArgs);
+      AuxTC.AddClangSystemIncludeArgs(Args, CmdArgs);
+    }
+  }
 }
 
 // FIXME: Move to target hook.
 static bool isSignedCharDefault(const llvm::Triple &Triple) {
   switch (Triple.getArch()) {
   default:
     return true;
 
void Clang::ConstructJob(Compilation &C, const JobAction &JA,
   //
   // FIXME: Implement custom jobs for internal actions.
   CmdArgs.push_back("-cc1");
 
   // Add the "effective" target triple.
   CmdArgs.push_back("-triple");
   CmdArgs.push_back(Args.MakeArgString(TripleStr));
 
-  if (IsCuda && AuxTriple != nullptr) {
+  if (IsCuda) {
+    if (AuxTriple != nullptr) {
       CmdArgs.push_back("-aux-triple");
       CmdArgs.push_back(AuxTriple);
     }
+    CmdArgs.push_back("-fcuda-target-overloads");
+  }
 
   if (Triple.isOSWindows() && (Triple.getArch() == llvm::Triple::arm ||
                                Triple.getArch() == llvm::Triple::thumb)) {
     unsigned Offset = Triple.getArch() == llvm::Triple::arm ? 4 : 6;
     unsigned Version;
     Triple.getArchName().substr(Offset).getAsInteger(10, Version);
     if (Version < 7)
       D.Diag(diag::err_target_unsupported_arch) << Triple.getArchName()
if (const Arg *A = Args.getLastArg(options::OPT_ccc_objcmt_migrate)) {
     Args.AddLastArg(CmdArgs, options::OPT_objcmt_whitelist_dir_path);
   }
 
   // Add preprocessing options like -I, -D, etc. if we are using the
   // preprocessor.
   //
   // FIXME: Support -fpreprocessed
   if (types::getPreprocessedType(InputType) != types::TY_INVALID)
-    AddPreprocessingOptions(C, JA, D, Args, CmdArgs, Output, Inputs);
+    AddPreprocessingOptions(C, JA, D, Args, CmdArgs, Output, Inputs, AuxTriple);
 
   // Don't warn about "clang -c -DPIC -fPIC test.i" because libtool.m4 assumes
   // that "The compiler can only warn and ignore the option if not recognized".
   // When building with ccache, it will pass -D options to clang even on
   // preprocessed inputs and configure concludes that -fPIC is not supported.
   Args.ClaimAllArgs(options::OPT_D);
 
   // Manually translate -O4 to -O3; let clang reject others.

test/Driver/cuda-detect.cu

 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 //
+// # Check that we properly detect CUDA installation.
 // RUN: %clang -v --target=i386-unknown-linux \
 // RUN:   --sysroot=/tmp/no-cuda-there 2>&1 | FileCheck %s -check-prefix NOCUDA
 // RUN: %clang -v --target=i386-unknown-linux \
+// RUN:   --sysroot=%S/Inputs/CUDA 2>&1 | FileCheck %s
+// RUN: %clang -v --target=i386-unknown-linux \
 // RUN:   --cuda-path=%S/Inputs/CUDA/usr/local/cuda 2>&1 | FileCheck %s
 
+// Make sure we map libdevice bitcode files to proper GPUs.
+// RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_21 \
+// RUN:   --cuda-path=%S/Inputs/CUDA/usr/local/cuda %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix LIBDEVICE -check-prefix LIBDEVICE21
+// RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_35 \
+// RUN:   --cuda-path=%S/Inputs/CUDA/usr/local/cuda %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix LIBDEVICE -check-prefix LIBDEVICE35 \
+// RUN:   -check-prefix CUDAINC
+// Verify that -nocudainc prevents adding include path to CUDA headers.
+// RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_35 \
+// RUN:   -nocudainc --cuda-path=%S/Inputs/CUDA/usr/local/cuda %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix LIBDEVICE -check-prefix LIBDEVICE35 \
+// RUN:   -check-prefix NOCUDAINC
+
+// Verify that no options related to bitcode linking are passes if
+// there's no bitcode file.
+// RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_30 \
+// RUN:   --cuda-path=%S/Inputs/CUDA/usr/local/cuda %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix NOLIBDEVICE
+// .. or if we explicitly passed -nocudalib
+// RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_35 \
+// RUN:   -nocudalib --cuda-path=%S/Inputs/CUDA/usr/local/cuda %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix NOLIBDEVICE
+
 // CHECK: Found CUDA installation: {{.*}}/Inputs/CUDA/usr/local/cuda
 // NOCUDA-NOT: Found CUDA installation:
+
+// LIBDEVICE: "-triple" "nvptx-nvidia-cuda"
+// LIBDEVICE-SAME: "-fcuda-is-device"
+// LIBDEVICE-SAME: "-mlink-bitcode-file"
+// LIBDEVICE21-SAME: libdevice.compute_20.10.bc
+// LIBDEVICE35-SAME: libdevice.compute_35.10.bc
+// LIBDEVICE-SAME: "-fcuda-uses-libdevice"
+// LIBDEVICE-SAME: "-target-feature" "+ptx42"
+// CUDAINC-SAME: "-internal-isystem" "{{.*}}/Inputs/CUDA/usr/local/cuda/include"
+// NOCUDAINC-NOT: "-internal-isystem" "{{.*}}/Inputs/CUDA/usr/local/cuda/include"
+// LIBDEVICE-SAME: "-x" "cuda"
+
+// NOLIBDEVICE: "-triple" "nvptx-nvidia-cuda"
+// NOLIBDEVICE-SAME: "-fcuda-is-device"
+// NOLIBDEVICE-NOT: "-mlink-bitcode-file"
+// NOLIBDEVICE-NOT: libdevice.compute_{{.*}}.bc
+// NOLIBDEVICE-NOT: "-fcuda-uses-libdevice"
+// NOLIBDEVICE-NOT: "-target-feature"
+// NOLIBDEVICE-SAME: "-x" "cuda"