[OpenMP] Make `-fopenmp-target=` use the `nvptx-arch` tool

[jhuber6]

Summary

Previously, if the user did not provide an architecture when using
-fopenmp-targets=nvptx64 we used the value from
CLANG_OPENMP_DEFAULT_NVPTX_ARCH which is defined at compile time. This
isn't ideal because it means that the default is set when the LLVM
compiler it built. Instead this patch uses the nvptx-arch tool to
query it at runtime. This matches the existing behaviour of the AMDGPU
toolchain with its amdgpu-arch tool.

[jdoerfert]

LG

clang/lib/Driver/ToolChains/Cuda.cpp

 #include "clang/Driver/Distro.h"
 #include "clang/Driver/Driver.h"
 #include "clang/Driver/DriverDiagnostic.h"
 #include "clang/Driver/InputInfo.h"
 #include "clang/Driver/Options.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/Option/ArgList.h"
 #include "llvm/Support/FileSystem.h"
+#include "llvm/Support/FormatAdapters.h"
+#include "llvm/Support/FormatVariadic.h"
 #include "llvm/Support/Host.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/Process.h"
 #include "llvm/Support/Program.h"
 #include "llvm/Support/TargetParser.h"
 #include "llvm/Support/VirtualFileSystem.h"
 #include <system_error>
 
CudaToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
   // For OpenMP device offloading, append derived arguments. Make sure
   // flags are not duplicated.
   // Also append the compute capability.
   if (DeviceOffloadKind == Action::OFK_OpenMP) {
     for (Arg *A : Args)
       if (!llvm::is_contained(*DAL, A))
         DAL->append(A);
 
-    if (!DAL->hasArg(options::OPT_march_EQ))
-      DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ),
-                        !BoundArch.empty() ? BoundArch
-                                           : CLANG_OPENMP_NVPTX_DEFAULT_ARCH);
+    if (!DAL->hasArg(options::OPT_march_EQ)) {
+      StringRef Arch = BoundArch;
+      if (Arch.empty()) {
+        auto ArchsOrErr = getSystemGPUArchs(Args);
+        if (!ArchsOrErr) {
+          std::string ErrMsg =
+              llvm::formatv("{0}", llvm::fmt_consume(ArchsOrErr.takeError()));
+          getDriver().Diag(diag::err_drv_undetermined_gpu_arch)
+              << llvm::Triple::getArchTypeName(getArch()) << ErrMsg << "-march";
+          Arch = CudaArchToString(CudaArch::CudaDefault);
+        } else {
+          Arch = Args.MakeArgString(ArchsOrErr->front());
+        }
+      }
+      DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ), Arch);
+    }
 
     return DAL;
   }
 
   for (Arg *A : Args) {
     DAL->append(A);
   }
 

clang/test/Driver/openmp-system-arch.c

 // RUN:     --nvptx-arch-tool=%t/nvptx_arch_empty --amdgpu-arch-tool=%t/amdgpu_arch_empty %s 2>&1 \
 // RUN:   | FileCheck %s --check-prefix=NO-OUTPUT-ERROR
 // NO-OUTPUT-ERROR: error: failed to deduce triple for target architecture 'native'; specify the triple using '-fopenmp-targets' and '-Xopenmp-target' instead.
 
 // case when amdgpu-arch succeeds.
 // RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp --offload-arch=native \
 // RUN:     --nvptx-arch-tool=%t/nvptx_arch_fail --amdgpu-arch-tool=%t/amdgpu_arch_gfx906 %s 2>&1 \
 // RUN:   | FileCheck %s --check-prefix=ARCH-GFX906
+// RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp -fopenmp-targets=amdgcn-amd-amdhsa \
+// RUN:     --nvptx-arch-tool=%t/nvptx_arch_fail --amdgpu-arch-tool=%t/amdgpu_arch_gfx906 %s 2>&1 \
+// RUN:   | FileCheck %s --check-prefix=ARCH-GFX906
 // ARCH-GFX906: "-cc1" "-triple" "amdgcn-amd-amdhsa"{{.*}}"-target-cpu" "gfx906"
 
 // case when nvptx-arch succeeds.
 // RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp --offload-arch=native \
 // RUN:     --nvptx-arch-tool=%t/nvptx_arch_sm_70 --amdgpu-arch-tool=%t/amdgpu_arch_fail %s 2>&1 \
 // RUN:   | FileCheck %s --check-prefix=ARCH-SM_70
+// RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda \
+// RUN:     --nvptx-arch-tool=%t/nvptx_arch_sm_70 --amdgpu-arch-tool=%t/amdgpu_arch_fail %s 2>&1 \
+// RUN:   | FileCheck %s --check-prefix=ARCH-SM_70
 // ARCH-SM_70: "-cc1" "-triple" "nvptx64-nvidia-cuda"{{.*}}"-target-cpu" "sm_70"
 
 // case when both nvptx-arch and amdgpu-arch succeed.
 // RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp --offload-arch=native \
 // RUN:     --nvptx-arch-tool=%t/nvptx_arch_sm_70 --amdgpu-arch-tool=%t/amdgpu_arch_gfx906 %s 2>&1 \
 // RUN:   | FileCheck %s --check-prefix=ARCH-SM_70-GFX906
 // ARCH-SM_70-GFX906: "-cc1" "-triple" "amdgcn-amd-amdhsa"{{.*}}"-target-cpu" "gfx906"
 // ARCH-SM_70-GFX906: "-cc1" "-triple" "nvptx64-nvidia-cuda"{{.*}}"-target-cpu" "sm_70"
 
 // case when both nvptx-arch and amdgpu-arch succeed with other archs.
 // RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp --offload-arch=native,sm_75,gfx1030 \
 // RUN:     --nvptx-arch-tool=%t/nvptx_arch_sm_70 --amdgpu-arch-tool=%t/amdgpu_arch_gfx906 %s 2>&1 \
 // RUN:   | FileCheck %s --check-prefix=ARCH-MULTIPLE
 // ARCH-MULTIPLE: "-cc1" "-triple" "amdgcn-amd-amdhsa"{{.*}}"-target-cpu" "gfx1030"
 // ARCH-MULTIPLE: "-cc1" "-triple" "amdgcn-amd-amdhsa"{{.*}}"-target-cpu" "gfx906"
 // ARCH-MULTIPLE: "-cc1" "-triple" "nvptx64-nvidia-cuda"{{.*}}"-target-cpu" "sm_70"
 // ARCH-MULTIPLE: "-cc1" "-triple" "nvptx64-nvidia-cuda"{{.*}}"-target-cpu" "sm_75"
+
+// case when 'nvptx-arch' returns nothing using `-fopenmp-targets=`.
+// RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp \
+// RUN:     -fopenmp-targets=nvptx64-nvidia-cuda --nvptx-arch-tool=%t/nvptx_arch_empty %s 2>&1 \
+// RUN:   | FileCheck %s --check-prefix=NVPTX
+// NVPTX: error: cannot determine nvptx64 architecture: No NVIDIA GPU detected in the system; consider passing it via '-march'
+
+// case when 'amdgpu-arch' returns nothing using `-fopenmp-targets=`.
+// RUN:   %clang -### --target=x86_64-unknown-linux-gnu -nogpulib -fopenmp=libomp \
+// RUN:     -fopenmp-targets=amdgcn-amd-amdhsa --amdgpu-arch-tool=%t/amdgpu_arch_empty %s 2>&1 \
+// RUN:   | FileCheck %s --check-prefix=AMDGPU
+// AMDGPU: error: cannot determine amdgcn architecture: No AMD GPU detected in the system; consider passing it via '-march'