jhuber6 created this revision.
jhuber6 added reviewers: jdoerfert, tra, tianshilei1992, yaxunl.
Herald added a subscriber: asavonic.
Herald added a project: All.
jhuber6 requested review of this revision.
Herald added subscribers: cfe-commits, sstefan1.
Herald added a project: clang.
The NVPTX toolchain uses target features to determine the PTX version to
use. However this isn't exposed externally like most other toolchain
specific target features are. Add this functionaliy in preparation for
using it in for OpenMP offloading.
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D122089
Files:
clang/lib/Driver/ToolChains/Clang.cpp
clang/lib/Driver/ToolChains/Cuda.cpp
clang/lib/Driver/ToolChains/Cuda.h
Index: clang/lib/Driver/ToolChains/Cuda.h
===================================================================
--- clang/lib/Driver/ToolChains/Cuda.h
+++ clang/lib/Driver/ToolChains/Cuda.h
@@ -124,6 +124,11 @@
const char *LinkingOutput) const override;
};
+void getNVPTXTargetFeatures(const Driver &D, const llvm::Triple &Triple,
+ const llvm::opt::ArgList &Args,
+ std::vector<StringRef> &Features,
+ Optional<clang::CudaVersion> CudaVersion = None);
+
} // end namespace NVPTX
} // end namespace tools
Index: clang/lib/Driver/ToolChains/Cuda.cpp
===================================================================
--- clang/lib/Driver/ToolChains/Cuda.cpp
+++ clang/lib/Driver/ToolChains/Cuda.cpp
@@ -630,6 +630,43 @@
Exec, CmdArgs, Inputs, Output));
}
+void NVPTX::getNVPTXTargetFeatures(const Driver &D, const llvm::Triple &Triple,
+ const llvm::opt::ArgList &Args,
+ std::vector<StringRef> &Features,
+ Optional<clang::CudaVersion> CudaVersion) {
+ if (!CudaVersion) {
+ CudaInstallationDetector CudaInstallation(D, Triple, Args);
+ CudaVersion = CudaInstallation.version();
+ }
+
+ // New CUDA versions often introduce new instructions that are only supported
+ // by new PTX version, so we need to raise PTX level to enable them in NVPTX
+ // back-end.
+ const char *PtxFeature = nullptr;
+ switch (*CudaVersion) {
+#define CASE_CUDA_VERSION(CUDA_VER, PTX_VER) \
+ case CudaVersion::CUDA_##CUDA_VER: \
+ PtxFeature = "+ptx" #PTX_VER; \
+ break;
+ CASE_CUDA_VERSION(115, 75);
+ CASE_CUDA_VERSION(114, 74);
+ CASE_CUDA_VERSION(113, 73);
+ CASE_CUDA_VERSION(112, 72);
+ CASE_CUDA_VERSION(111, 71);
+ CASE_CUDA_VERSION(110, 70);
+ CASE_CUDA_VERSION(102, 65);
+ CASE_CUDA_VERSION(101, 64);
+ CASE_CUDA_VERSION(100, 63);
+ CASE_CUDA_VERSION(92, 61);
+ CASE_CUDA_VERSION(91, 61);
+ CASE_CUDA_VERSION(90, 60);
+#undef CASE_CUDA_VERSION
+ default:
+ PtxFeature = "+ptx42";
+ }
+ Features.push_back(PtxFeature);
+}
+
/// CUDA toolchain. Our assembler is ptxas, and our "linker" is fatbinary,
/// which isn't properly a linker but nonetheless performs the step of stitching
/// together object files from the assembler into a single blob.
@@ -701,32 +738,11 @@
clang::CudaVersion CudaInstallationVersion = CudaInstallation.version();
- // New CUDA versions often introduce new instructions that are only supported
- // by new PTX version, so we need to raise PTX level to enable them in NVPTX
- // back-end.
- const char *PtxFeature = nullptr;
- switch (CudaInstallationVersion) {
-#define CASE_CUDA_VERSION(CUDA_VER, PTX_VER) \
- case CudaVersion::CUDA_##CUDA_VER: \
- PtxFeature = "+ptx" #PTX_VER; \
- break;
- CASE_CUDA_VERSION(115, 75);
- CASE_CUDA_VERSION(114, 74);
- CASE_CUDA_VERSION(113, 73);
- CASE_CUDA_VERSION(112, 72);
- CASE_CUDA_VERSION(111, 71);
- CASE_CUDA_VERSION(110, 70);
- CASE_CUDA_VERSION(102, 65);
- CASE_CUDA_VERSION(101, 64);
- CASE_CUDA_VERSION(100, 63);
- CASE_CUDA_VERSION(92, 61);
- CASE_CUDA_VERSION(91, 61);
- CASE_CUDA_VERSION(90, 60);
-#undef CASE_CUDA_VERSION
- default:
- PtxFeature = "+ptx42";
- }
- CC1Args.append({"-target-feature", PtxFeature});
+ std::vector<StringRef> Features;
+ NVPTX::getNVPTXTargetFeatures(getDriver(), getTriple(), DriverArgs, Features,
+ CudaInstallationVersion);
+ for (StringRef PtxFeature : Features)
+ CC1Args.append({"-target-feature", DriverArgs.MakeArgString(PtxFeature)});
if (DriverArgs.hasFlag(options::OPT_fcuda_short_ptr,
options::OPT_fno_cuda_short_ptr, false))
CC1Args.append({"-mllvm", "--nvptx-short-ptr"});
Index: clang/lib/Driver/ToolChains/Clang.cpp
===================================================================
--- clang/lib/Driver/ToolChains/Clang.cpp
+++ clang/lib/Driver/ToolChains/Clang.cpp
@@ -369,6 +369,10 @@
case llvm::Triple::amdgcn:
amdgpu::getAMDGPUTargetFeatures(D, Triple, Args, Features);
break;
+ case llvm::Triple::nvptx:
+ case llvm::Triple::nvptx64:
+ NVPTX::getNVPTXTargetFeatures(D, Triple, Args, Features);
+ break;
case llvm::Triple::m68k:
m68k::getM68kTargetFeatures(D, Triple, Args, Features);
break;
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
