[PATCH] D38883: [CMake][OpenMP] Customize default offloading arch
Jonas Hahnfeld via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Fri Oct 13 12:55:59 PDT 2017
Hahnfeld updated this revision to Diff 118961.
Hahnfeld marked an inline comment as done.
Hahnfeld edited the summary of this revision.
Hahnfeld added a comment.
Check that the user didn't specify a value lower than `sm_30` and re-add some code as discussed.
https://reviews.llvm.org/D38883
Files:
CMakeLists.txt
include/clang/Config/config.h.cmake
lib/Driver/ToolChains/Cuda.cpp
lib/Driver/ToolChains/Cuda.h
Index: lib/Driver/ToolChains/Cuda.h
===================================================================
--- lib/Driver/ToolChains/Cuda.h
+++ lib/Driver/ToolChains/Cuda.h
@@ -76,17 +76,6 @@
std::string getLibDeviceFile(StringRef Gpu) const {
return LibDeviceMap.lookup(Gpu);
}
- /// \brief Get lowest available compute capability
- /// for which a libdevice library exists.
- std::string getLowestExistingArch() const {
- std::string LibDeviceFile;
- for (auto key : LibDeviceMap.keys()) {
- LibDeviceFile = LibDeviceMap.lookup(key);
- if (!LibDeviceFile.empty())
- return key;
- }
- return "sm_20";
- }
};
namespace tools {
Index: lib/Driver/ToolChains/Cuda.cpp
===================================================================
--- lib/Driver/ToolChains/Cuda.cpp
+++ lib/Driver/ToolChains/Cuda.cpp
@@ -551,9 +551,9 @@
// flags are not duplicated.
// Also append the compute capability.
if (DeviceOffloadKind == Action::OFK_OpenMP) {
- for (Arg *A : Args){
+ for (Arg *A : Args) {
bool IsDuplicate = false;
- for (Arg *DALArg : *DAL){
+ for (Arg *DALArg : *DAL) {
if (A == DALArg) {
IsDuplicate = true;
break;
@@ -564,14 +564,9 @@
}
StringRef Arch = DAL->getLastArgValue(options::OPT_march_EQ);
- if (Arch.empty()) {
- // Default compute capability for CUDA toolchain is the
- // lowest compute capability supported by the installed
- // CUDA version.
- DAL->AddJoinedArg(nullptr,
- Opts.getOption(options::OPT_march_EQ),
- CudaInstallation.getLowestExistingArch());
- }
+ if (Arch.empty())
+ DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ),
+ CLANG_OPENMP_NVPTX_DEFAULT_ARCH);
return DAL;
}
Index: include/clang/Config/config.h.cmake
===================================================================
--- include/clang/Config/config.h.cmake
+++ include/clang/Config/config.h.cmake
@@ -20,6 +20,9 @@
/* Default OpenMP runtime used by -fopenmp. */
#define CLANG_DEFAULT_OPENMP_RUNTIME "${CLANG_DEFAULT_OPENMP_RUNTIME}"
+/* Default architecture for OpenMP offloading to Nvidia GPUs. */
+#define CLANG_OPENMP_NVPTX_DEFAULT_ARCH "${CLANG_OPENMP_NVPTX_DEFAULT_ARCH}"
+
/* Multilib suffix for libdir. */
#define CLANG_LIBDIR_SUFFIX "${CLANG_LIBDIR_SUFFIX}"
Index: CMakeLists.txt
===================================================================
--- CMakeLists.txt
+++ CMakeLists.txt
@@ -235,6 +235,17 @@
set(CLANG_DEFAULT_OPENMP_RUNTIME "libomp" CACHE STRING
"Default OpenMP runtime used by -fopenmp.")
+# OpenMP offloading requires at least sm_30 because we use shuffle instructions
+# to generate efficient code for reductions.
+set(CLANG_OPENMP_NVPTX_DEFAULT_ARCH "sm_30" CACHE STRING
+ "Default architecture for OpenMP offloading to Nvidia GPUs.")
+string(REGEX MATCH "^sm_([0-9]+)$" MATCHED_ARCH "${CLANG_OPENMP_NVPTX_DEFAULT_ARCH}")
+if (NOT DEFINED MATCHED_ARCH OR "${CMAKE_MATCH_1}" LESS 30)
+ message(WARNING "Resetting default architecture for OpenMP offloading to Nvidia GPUs to sm_30")
+ set(CLANG_OPENMP_NVPTX_DEFAULT_ARCH "sm_30" CACHE STRING
+ "Default architecture for OpenMP offloading to Nvidia GPUs." FORCE)
+endif()
+
set(CLANG_VENDOR ${PACKAGE_VENDOR} CACHE STRING
"Vendor-specific text for showing with version information.")
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D38883.118961.patch
Type: text/x-patch
Size: 3397 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20171013/f53f99d3/attachment-0001.bin>
More information about the cfe-commits
mailing list