clang/lib/Driver/ToolChains/AMDGPU.cpp

//===--- AMDGPU.cpp - AMDGPU ToolChain Implementations ----------*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//

#include "AMDGPU.h"
#include "CommonArgs.h"
#include "InputInfo.h"
#include "clang/Driver/Compilation.h"
#include "clang/Driver/DriverDiagnostic.h"
#include "llvm/Option/ArgList.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/VirtualFileSystem.h"

using namespace clang::driver;
using namespace clang::driver::tools;
using namespace clang::driver::toolchains;
using namespace clang;
using namespace llvm::opt;

void RocmInstallationDetector::scanLibDevicePath() {
  assert(!LibDevicePath.empty());

  const StringRef Suffix(".bc");
  const StringRef Suffix2(".amdgcn.bc");

  std::error_code EC;
  for (llvm::vfs::directory_iterator
           LI = D.getVFS().dir_begin(LibDevicePath, EC),
           LE;
       !EC && LI != LE; LI = LI.increment(EC)) {
    StringRef FilePath = LI->path();
    StringRef FileName = llvm::sys::path::filename(FilePath);
    if (!FileName.endswith(Suffix))
      continue;

    StringRef BaseName;
    if (FileName.endswith(Suffix2))
      BaseName = FileName.drop_back(Suffix2.size());
    else if (FileName.endswith(Suffix))
      BaseName = FileName.drop_back(Suffix.size());

    if (BaseName == "ocml") {
      OCML = FilePath;
    } else if (BaseName == "ockl") {
      OCKL = FilePath;
    } else if (BaseName == "opencl") {
      OpenCL = FilePath;
    } else if (BaseName == "hip") {
      HIP = FilePath;
    } else if (BaseName == "oclc_finite_only_off") {
      FiniteOnly.Off = FilePath;
    } else if (BaseName == "oclc_finite_only_on") {
      FiniteOnly.On = FilePath;
    } else if (BaseName == "oclc_daz_opt_on") {
      DenormalsAreZero.On = FilePath;
    } else if (BaseName == "oclc_daz_opt_off") {
      DenormalsAreZero.Off = FilePath;
    } else if (BaseName == "oclc_correctly_rounded_sqrt_on") {
      CorrectlyRoundedSqrt.On = FilePath;
    } else if (BaseName == "oclc_correctly_rounded_sqrt_off") {
      CorrectlyRoundedSqrt.Off = FilePath;
    } else if (BaseName == "oclc_unsafe_math_on") {
      UnsafeMath.On = FilePath;
    } else if (BaseName == "oclc_unsafe_math_off") {
      UnsafeMath.Off = FilePath;
    } else if (BaseName == "oclc_wavefrontsize64_on") {
      WavefrontSize64.On = FilePath;
    } else if (BaseName == "oclc_wavefrontsize64_off") {
      WavefrontSize64.Off = FilePath;
    } else {
      // Process all bitcode filenames that look like
      // ocl_isa_version_XXX.amdgcn.bc
      const StringRef DeviceLibPrefix = "oclc_isa_version_";
      if (!BaseName.startswith(DeviceLibPrefix))
        continue;

      StringRef IsaVersionNumber =
        BaseName.drop_front(DeviceLibPrefix.size());

      llvm::Twine GfxName = Twine("gfx") + IsaVersionNumber;
      SmallString<8> Tmp;
      LibDeviceMap.insert(
        std::make_pair(GfxName.toStringRef(Tmp), FilePath.str()));
    }
  }
}

RocmInstallationDetector::RocmInstallationDetector(
    const Driver &D, const llvm::Triple &HostTriple,
    const llvm::opt::ArgList &Args)
    : D(D) {
  struct Candidate {
    std::string Path;
    bool StrictChecking;

    Candidate(std::string Path, bool StrictChecking = false)
        : Path(Path), StrictChecking(StrictChecking) {}
  };

  SmallVector<Candidate, 4> Candidates;

  if (Args.hasArg(clang::driver::options::OPT_rocm_path_EQ)) {
    Candidates.emplace_back(
        Args.getLastArgValue(clang::driver::options::OPT_rocm_path_EQ).str());
  } else {
    // Try to find relative to the compiler binary.
    const char *InstallDir = D.getInstalledDir();

    // Check both a normal Unix prefix position of the clang binary, as well as
    // the Windows-esque layout the ROCm packages use with the host architecture
    // subdirectory of bin.

    // Strip off directory (usually bin)
    StringRef ParentDir = llvm::sys::path::parent_path(InstallDir);
    StringRef ParentName = llvm::sys::path::filename(ParentDir);

    // Some builds use bin/{host arch}, so go up again.
    if (ParentName == "bin") {
      ParentDir = llvm::sys::path::parent_path(ParentDir);
      ParentName = llvm::sys::path::filename(ParentDir);
    }

    if (ParentName == "llvm") {
      // Some versions of the rocm llvm package install to /opt/rocm/llvm/bin
      Candidates.emplace_back(llvm::sys::path::parent_path(ParentDir).str(),
                              /*StrictChecking=*/true);
    }

    Candidates.emplace_back(D.SysRoot + "/opt/rocm");
  }

  bool NoBuiltinLibs = Args.hasArg(options::OPT_nogpulib);

  assert(LibDevicePath.empty());

  if (Args.hasArg(clang::driver::options::OPT_hip_device_lib_path_EQ)) {
    LibDevicePath
      = Args.getLastArgValue(clang::driver::options::OPT_hip_device_lib_path_EQ);
  } else if (const char *LibPathEnv = ::getenv("HIP_DEVICE_LIB_PATH")) {
    LibDevicePath = LibPathEnv;
  }

  auto &FS = D.getVFS();
  if (!LibDevicePath.empty()) {
    // Maintain compatability with HIP flag/envvar pointing directly at the
    // bitcode library directory. This points directly at the library path instead
    // of the rocm root installation.
    if (!FS.exists(LibDevicePath))
      return;

    scanLibDevicePath();
    IsValid = allGenericLibsValid() && !LibDeviceMap.empty();
    return;
  }

  for (const auto &Candidate : Candidates) {
    InstallPath = Candidate.Path;
    if (InstallPath.empty() || !FS.exists(InstallPath))
      continue;

    // The install path situation in old versions of ROCm is a real mess, and
    // use a different install layout. Multiple copies of the device libraries
    // exist for each frontend project, and differ depending on which build
    // system produced the packages. Standalone OpenCL builds also have a
    // different directory structure from the ROCm OpenCL package.
    //
    // The desired structure is (${ROCM_ROOT} or
    // ${OPENCL_ROOT})/amdgcn/bitcode/*, so try to detect this layout.

    // BinPath = InstallPath + "/bin";
    llvm::sys::path::append(IncludePath, InstallPath, "include");
    llvm::sys::path::append(LibDevicePath, InstallPath, "amdgcn", "bitcode");

    // We don't need the include path for OpenCL, since clang already ships with
    // the default header.

    bool CheckLibDevice = (!NoBuiltinLibs || Candidate.StrictChecking);
    if (CheckLibDevice && !FS.exists(LibDevicePath))
      continue;

    scanLibDevicePath();

    if (!NoBuiltinLibs) {
      // Check that the required non-target libraries are all available.
      if (!allGenericLibsValid())
        continue;

      // Check that we have found at least one libdevice that we can link in if
      // -nobuiltinlib hasn't been specified.
      if (LibDeviceMap.empty())
        continue;
    }

    IsValid = true;
    break;
  }
}

void RocmInstallationDetector::print(raw_ostream &OS) const {
  if (isValid())
    OS << "Found ROCm installation: " << InstallPath << '\n';
}

void RocmInstallationDetector::AddHIPIncludeArgs(const ArgList &DriverArgs,
                                                 ArgStringList &CC1Args) const {
  if (!DriverArgs.hasArg(options::OPT_nobuiltininc)) {
    // HIP header includes standard library wrapper headers under clang
    // cuda_wrappers directory. Since these wrapper headers include_next
    // standard C++ headers, whereas libc++ headers include_next other clang
    // headers. The include paths have to follow this order:
    // - wrapper include path
    // - standard C++ include path
    // - other clang include path
    // Since standard C++ and other clang include paths are added in other
    // places after this function, here we only need to make sure wrapper
    // include path is added.
    SmallString<128> P(D.ResourceDir);
    llvm::sys::path::append(P, "include");
    llvm::sys::path::append(P, "cuda_wrappers");
    CC1Args.push_back("-internal-isystem");
    CC1Args.push_back(DriverArgs.MakeArgString(P));
  }

  if (DriverArgs.hasArg(options::OPT_nogpuinc))
    return;

  if (!isValid()) {
    D.Diag(diag::err_drv_no_rocm_installation);
    return;
  }

  CC1Args.push_back("-internal-isystem");
  CC1Args.push_back(DriverArgs.MakeArgString(getIncludePath()));
  CC1Args.push_back("-include");
  CC1Args.push_back("__clang_hip_runtime_wrapper.h");
}

void amdgpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
                                  const InputInfo &Output,
                                  const InputInfoList &Inputs,
                                  const ArgList &Args,
                                  const char *LinkingOutput) const {

  std::string Linker = getToolChain().GetProgramPath(getShortName());
  ArgStringList CmdArgs;
  AddLinkerInputs(getToolChain(), Inputs, Args, CmdArgs, JA);
  CmdArgs.push_back("-shared");
  CmdArgs.push_back("-o");
  CmdArgs.push_back(Output.getFilename());
  C.addCommand(
      std::make_unique<Command>(JA, *this, ResponseFileSupport::AtFileCurCP(),
                                Args.MakeArgString(Linker), CmdArgs, Inputs));
}

void amdgpu::getAMDGPUTargetFeatures(const Driver &D,
                                     const llvm::opt::ArgList &Args,
                                     std::vector<StringRef> &Features) {
  if (const Arg *dAbi = Args.getLastArg(options::OPT_mamdgpu_debugger_abi))
    D.Diag(diag::err_drv_clang_unsupported) << dAbi->getAsString(Args);

  if (Args.getLastArg(options::OPT_mwavefrontsize64)) {
    Features.push_back("-wavefrontsize16");
    Features.push_back("-wavefrontsize32");
    Features.push_back("+wavefrontsize64");
  }
  if (Args.getLastArg(options::OPT_mno_wavefrontsize64)) {
    Features.push_back("-wavefrontsize16");
    Features.push_back("+wavefrontsize32");
    Features.push_back("-wavefrontsize64");
  }

  handleTargetFeaturesGroup(
    Args, Features, options::OPT_m_amdgpu_Features_Group);
}

/// AMDGPU Toolchain
AMDGPUToolChain::AMDGPUToolChain(const Driver &D, const llvm::Triple &Triple,
                                 const ArgList &Args)
    : Generic_ELF(D, Triple, Args),
      OptionsDefault({{options::OPT_O, "3"},
                      {options::OPT_cl_std_EQ, "CL1.2"}}) {}

Tool *AMDGPUToolChain::buildLinker() const {
  return new tools::amdgpu::Linker(*this);
}

DerivedArgList *
AMDGPUToolChain::TranslateArgs(const DerivedArgList &Args, StringRef BoundArch,
                               Action::OffloadKind DeviceOffloadKind) const {

  DerivedArgList *DAL =
      Generic_ELF::TranslateArgs(Args, BoundArch, DeviceOffloadKind);

  // Do nothing if not OpenCL (-x cl)
  if (!Args.getLastArgValue(options::OPT_x).equals("cl"))
    return DAL;

  if (!DAL)
    DAL = new DerivedArgList(Args.getBaseArgs());
  for (auto *A : Args)
    DAL->append(A);

  const OptTable &Opts = getDriver().getOpts();

  // Phase 1 (.cl -> .bc)
  if (Args.hasArg(options::OPT_c) && Args.hasArg(options::OPT_emit_llvm)) {
    DAL->AddFlagArg(nullptr, Opts.getOption(getTriple().isArch64Bit()
                                                ? options::OPT_m64
                                                : options::OPT_m32));

    // Have to check OPT_O4, OPT_O0 & OPT_Ofast separately
    // as they defined that way in Options.td
    if (!Args.hasArg(options::OPT_O, options::OPT_O0, options::OPT_O4,
                     options::OPT_Ofast))
      DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_O),
                        getOptionDefault(options::OPT_O));
  }

  return DAL;
}

bool AMDGPUToolChain::getDefaultDenormsAreZeroForTarget(
    llvm::AMDGPU::GPUKind Kind) {

  // Assume nothing without a specific target.
  if (Kind == llvm::AMDGPU::GK_NONE)
    return false;

  const unsigned ArchAttr = llvm::AMDGPU::getArchAttrAMDGCN(Kind);

  // Default to enabling f32 denormals by default on subtargets where fma is
  // fast with denormals
  const bool BothDenormAndFMAFast =
      (ArchAttr & llvm::AMDGPU::FEATURE_FAST_FMA_F32) &&
      (ArchAttr & llvm::AMDGPU::FEATURE_FAST_DENORMAL_F32);
  return !BothDenormAndFMAFast;
}

llvm::DenormalMode AMDGPUToolChain::getDefaultDenormalModeForType(
    const llvm::opt::ArgList &DriverArgs, const JobAction &JA,
    const llvm::fltSemantics *FPType) const {
  // Denormals should always be enabled for f16 and f64.
  if (!FPType || FPType != &llvm::APFloat::IEEEsingle())
    return llvm::DenormalMode::getIEEE();

  if (JA.getOffloadingDeviceKind() == Action::OFK_HIP ||
      JA.getOffloadingDeviceKind() == Action::OFK_Cuda) {
    auto Kind = llvm::AMDGPU::parseArchAMDGCN(JA.getOffloadingArch());
    if (FPType && FPType == &llvm::APFloat::IEEEsingle() &&
        DriverArgs.hasFlag(options::OPT_fcuda_flush_denormals_to_zero,
                           options::OPT_fno_cuda_flush_denormals_to_zero,
                           getDefaultDenormsAreZeroForTarget(Kind)))
      return llvm::DenormalMode::getPreserveSign();

    return llvm::DenormalMode::getIEEE();
  }

  const StringRef GpuArch = DriverArgs.getLastArgValue(options::OPT_mcpu_EQ);
  auto Kind = llvm::AMDGPU::parseArchAMDGCN(GpuArch);

  // TODO: There are way too many flags that change this. Do we need to check
  // them all?
  bool DAZ = DriverArgs.hasArg(options::OPT_cl_denorms_are_zero) ||
             getDefaultDenormsAreZeroForTarget(Kind);

  // Outputs are flushed to zero (FTZ), preserving sign. Denormal inputs are
  // also implicit treated as zero (DAZ).
  return DAZ ? llvm::DenormalMode::getPreserveSign() :
               llvm::DenormalMode::getIEEE();
}

bool AMDGPUToolChain::isWave64(const llvm::opt::ArgList &DriverArgs,
                               llvm::AMDGPU::GPUKind Kind) {
  const unsigned ArchAttr = llvm::AMDGPU::getArchAttrAMDGCN(Kind);
  static bool HasWave32 = (ArchAttr & llvm::AMDGPU::FEATURE_WAVE32);

  return !HasWave32 || DriverArgs.hasFlag(
    options::OPT_mwavefrontsize64, options::OPT_mno_wavefrontsize64, false);
}


/// ROCM Toolchain
ROCMToolChain::ROCMToolChain(const Driver &D, const llvm::Triple &Triple,
                             const ArgList &Args)
  : AMDGPUToolChain(D, Triple, Args),
    RocmInstallation(D, Triple, Args) { }

void AMDGPUToolChain::addClangTargetOptions(
    const llvm::opt::ArgList &DriverArgs,
    llvm::opt::ArgStringList &CC1Args,
    Action::OffloadKind DeviceOffloadingKind) const {
  // Default to "hidden" visibility, as object level linking will not be
  // supported for the foreseeable future.
  if (!DriverArgs.hasArg(options::OPT_fvisibility_EQ,
                         options::OPT_fvisibility_ms_compat)) {
    CC1Args.push_back("-fvisibility");
    CC1Args.push_back("hidden");
    CC1Args.push_back("-fapply-global-visibility-to-externs");
  }
}

void ROCMToolChain::addClangTargetOptions(
    const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
    Action::OffloadKind DeviceOffloadingKind) const {
  AMDGPUToolChain::addClangTargetOptions(DriverArgs, CC1Args,
                                         DeviceOffloadingKind);

  // For the OpenCL case where there is no offload target, accept -nostdlib to
  // disable bitcode linking.
  if (DeviceOffloadingKind == Action::OFK_None &&
      DriverArgs.hasArg(options::OPT_nostdlib))
    return;

  if (DriverArgs.hasArg(options::OPT_nogpulib))
    return;

  if (!RocmInstallation.isValid()) {
    getDriver().Diag(diag::err_drv_no_rocm_installation);
    return;
  }

  // Get the device name and canonicalize it
  const StringRef GpuArch = DriverArgs.getLastArgValue(options::OPT_mcpu_EQ);
  auto Kind = llvm::AMDGPU::parseArchAMDGCN(GpuArch);
  const StringRef CanonArch = llvm::AMDGPU::getArchNameAMDGCN(Kind);
  std::string LibDeviceFile = RocmInstallation.getLibDeviceFile(CanonArch);
  if (LibDeviceFile.empty()) {
    getDriver().Diag(diag::err_drv_no_rocm_device_lib) << GpuArch;
    return;
  }

  bool Wave64 = isWave64(DriverArgs, Kind);

  // TODO: There are way too many flags that change this. Do we need to check
  // them all?
  bool DAZ = DriverArgs.hasArg(options::OPT_cl_denorms_are_zero) ||
             getDefaultDenormsAreZeroForTarget(Kind);
  bool FiniteOnly = DriverArgs.hasArg(options::OPT_cl_finite_math_only);

  bool UnsafeMathOpt =
      DriverArgs.hasArg(options::OPT_cl_unsafe_math_optimizations);
  bool FastRelaxedMath = DriverArgs.hasArg(options::OPT_cl_fast_relaxed_math);
  bool CorrectSqrt =
      DriverArgs.hasArg(options::OPT_cl_fp32_correctly_rounded_divide_sqrt);

  // Add the OpenCL specific bitcode library.
  CC1Args.push_back("-mlink-builtin-bitcode");
  CC1Args.push_back(DriverArgs.MakeArgString(RocmInstallation.getOpenCLPath()));

  // Add the generic set of libraries.
  RocmInstallation.addCommonBitcodeLibCC1Args(
      DriverArgs, CC1Args, LibDeviceFile, Wave64, DAZ, FiniteOnly,
      UnsafeMathOpt, FastRelaxedMath, CorrectSqrt);
}

void RocmInstallationDetector::addCommonBitcodeLibCC1Args(
    const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
    StringRef LibDeviceFile, bool Wave64, bool DAZ, bool FiniteOnly,
    bool UnsafeMathOpt, bool FastRelaxedMath, bool CorrectSqrt) const {
  static const char LinkBitcodeFlag[] = "-mlink-builtin-bitcode";

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(getOCMLPath()));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(getOCKLPath()));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(getDenormalsAreZeroPath(DAZ)));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(
      getUnsafeMathPath(UnsafeMathOpt || FastRelaxedMath)));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(
      getFiniteOnlyPath(FiniteOnly || FastRelaxedMath)));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(
      DriverArgs.MakeArgString(getCorrectlyRoundedSqrtPath(CorrectSqrt)));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(getWavefrontSize64Path(Wave64)));

  CC1Args.push_back(LinkBitcodeFlag);
  CC1Args.push_back(DriverArgs.MakeArgString(LibDeviceFile));
}