[PATCH] D46475: [HIP] Set proper triple and offload kind for the toolchain

[Yaxun Liu via Phabricator via cfe-commits]

yaxunl created this revision.
yaxunl added reviewers: rjmccall, tra.

Also introduce --hip-link option to indicate HIP for linking.


https://reviews.llvm.org/D46475

Files:
  include/clang/Driver/Options.td
  lib/Driver/Driver.cpp


Index: lib/Driver/Driver.cpp
===================================================================
--- lib/Driver/Driver.cpp
+++ lib/Driver/Driver.cpp
@@ -537,24 +537,42 @@
                                               InputList &Inputs) {
 
   //
-  // CUDA
+  // CUDA/HIP
   //
   // We need to generate a CUDA toolchain if any of the inputs has a CUDA type.
-  if (llvm::any_of(Inputs, [](std::pair<types::ID, const llvm::opt::Arg *> &I) {
-        return types::isCuda(I.first);
-      })) {
+  // ToDo: Handle mixed CUDA/HIP input files and -x hip option. Diagnose
+  //       CUDA on amdgcn and HIP on nvptx.
+  bool IsHIP =
+      (C.getInputArgs().hasArg(options::OPT_x) &&
+       StringRef(C.getInputArgs().getLastArg(options::OPT_x)->getValue()) ==
+           "hip") ||
+      C.getInputArgs().hasArg(options::OPT_hip_link);
+  if (llvm::any_of(Inputs,
+                   [](std::pair<types::ID, const llvm::opt::Arg *> &I) {
+                     return types::isCuda(I.first);
+                   }) ||
+      IsHIP) {
     const ToolChain *HostTC = C.getSingleOffloadToolChain<Action::OFK_Host>();
     const llvm::Triple &HostTriple = HostTC->getTriple();
-    llvm::Triple CudaTriple(HostTriple.isArch64Bit() ? "nvptx64-nvidia-cuda"
-                                                     : "nvptx-nvidia-cuda");
-    // Use the CUDA and host triples as the key into the ToolChains map, because
-    // the device toolchain we create depends on both.
+    StringRef DeviceTripleStr;
+    auto OFK = IsHIP ? Action::OFK_HIP : Action::OFK_Cuda;
+    if (IsHIP) {
+      // HIP is only supported on amdgcn.
+      DeviceTripleStr = "amdgcn-amd-amdhsa";
+    } else {
+      // CUDA is only supported on nvptx.
+      DeviceTripleStr = HostTriple.isArch64Bit() ? "nvptx64-nvidia-cuda"
+                                                 : "nvptx-nvidia-cuda";
+    }
+    llvm::Triple CudaTriple(DeviceTripleStr);
+    // Use the CUDA/HIP and host triples as the key into the ToolChains map,
+    // because the device toolchain we create depends on both.
     auto &CudaTC = ToolChains[CudaTriple.str() + "/" + HostTriple.str()];
     if (!CudaTC) {
       CudaTC = llvm::make_unique<toolchains::CudaToolChain>(
-          *this, CudaTriple, *HostTC, C.getInputArgs(), Action::OFK_Cuda);
+          *this, CudaTriple, *HostTC, C.getInputArgs(), OFK);
     }
-    C.addOffloadDeviceToolChain(CudaTC.get(), Action::OFK_Cuda);
+    C.addOffloadDeviceToolChain(CudaTC.get(), OFK);
   }
 
   //
Index: include/clang/Driver/Options.td
===================================================================
--- include/clang/Driver/Options.td
+++ include/clang/Driver/Options.td
@@ -555,6 +555,8 @@
   HelpText<"Do not include PTX for the follwing GPU architecture (e.g. sm_35) or 'all'. May be specified more than once.">;
 def cuda_gpu_arch_EQ : Joined<["--"], "cuda-gpu-arch=">, Flags<[DriverOption]>,
   HelpText<"CUDA GPU architecture (e.g. sm_35).  May be specified more than once.">;
+def hip_link : Flag<["--"], "hip-link">,
+  HelpText<"Link clang-offload-bundler bundles for HIP">;
 def no_cuda_gpu_arch_EQ : Joined<["--"], "no-cuda-gpu-arch=">, Flags<[DriverOption]>,
   HelpText<"Remove GPU architecture (e.g. sm_35) from the list of GPUs to compile for. "
            "'all' resets the list to its default value.">;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D46475.145293.patch
Type: text/x-patch
Size: 3329 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20180504/85a87d89/attachment.bin>