Index: include/clang/Driver/Options.td =================================================================== --- include/clang/Driver/Options.td +++ include/clang/Driver/Options.td @@ -1595,6 +1595,7 @@ 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">; Index: lib/Driver/ToolChains.h =================================================================== --- lib/Driver/ToolChains.h +++ lib/Driver/ToolChains.h @@ -166,6 +166,7 @@ std::string CudaLibPath; std::string CudaLibDevicePath; std::string CudaIncludePath; + llvm::StringMap CudaLibDeviceMap; public: CudaInstallationDetector(const Driver &D) : IsValid(false), D(D) {} @@ -185,6 +186,9 @@ /// \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; Index: lib/Driver/ToolChains.cpp =================================================================== --- lib/Driver/ToolChains.cpp +++ lib/Driver/ToolChains.cpp @@ -1632,6 +1632,7 @@ Args.getLastArgValue(options::OPT_cuda_path_EQ)); else { CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda"); + CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda-7.5"); CudaPathCandidates.push_back(D.SysRoot + "/usr/local/cuda-7.0"); } @@ -1650,6 +1651,31 @@ D.getVFS().exists(CudaLibDevicePath))) continue; + 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 + const StringRef LibDeviceName = "libdevice."; + 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; } @@ -4041,6 +4067,22 @@ 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-cuda-bitcode"); + CC1Args.push_back(DriverArgs.MakeArgString(LibDeviceFile)); + + // 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 * Index: test/Driver/cuda-detect.cu =================================================================== --- test/Driver/cuda-detect.cu +++ test/Driver/cuda-detect.cu @@ -8,26 +8,51 @@ // RUN: %clang -v --target=i386-unknown-linux \ // RUN: --sysroot=%S/Inputs/CUDA 2>&1 | FileCheck %s // 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 -// Verify that CUDA include path gets added +// 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 COMMON \ +// RUN: -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 COMMON -check-prefix CUDAINC -// Verify that -nocudainc disables CUDA include paths. +// RUN: | FileCheck %s -check-prefix COMMON -check-prefix CUDAINC \ +// RUN: -check-prefix LIBDEVICE -check-prefix LIBDEVICE35 +// 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 COMMON -check-prefix NOCUDAINC +// RUN: | FileCheck %s -check-prefix COMMON -check-prefix NOCUDAINC \ +// RUN: -check-prefix LIBDEVICE -check-prefix LIBDEVICE35 // We should not add any CUDA include paths if there's no valid CUDA installation // RUN: %clang -### -v --target=i386-unknown-linux --cuda-gpu-arch=sm_35 \ // RUN: --cuda-path=%S/no-cuda-there %s 2>&1 \ // RUN: | FileCheck %s -check-prefix COMMON -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 COMMON -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 COMMON -check-prefix NOLIBDEVICE + // CHECK: Found CUDA installation: {{.*}}/Inputs/CUDA/usr/local/cuda // NOCUDA-NOT: Found CUDA installation: // COMMON: "-triple" "nvptx-nvidia-cuda" // COMMON-SAME: "-fcuda-is-device" +// LIBDEVICE-SAME: "-mlink-cuda-bitcode" +// NOLIBDEVICE-NOT: "-mlink-cuda-bitcode" +// LIBDEVICE21-SAME: libdevice.compute_20.10.bc +// LIBDEVICE35-SAME: libdevice.compute_35.10.bc +// NOLIBDEVICE-NOT: libdevice.compute_{{.*}}.bc +// LIBDEVICE-SAME: "-target-feature" "+ptx42" +// NOLIBDEVICE-NOT: "-target-feature" "+ptx42" // CUDAINC-SAME: "-internal-isystem" "{{.*}}/Inputs/CUDA/usr/local/cuda/include" // NOCUDAINC-NOT: "-internal-isystem" "{{.*}}/cuda/include" // COMMON-SAME: "-x" "cuda"