summaryrefslogtreecommitdiffstats
path: root/clang/lib/Driver
diff options
context:
space:
mode:
Diffstat (limited to 'clang/lib/Driver')
-rw-r--r--clang/lib/Driver/Action.cpp36
-rw-r--r--clang/lib/Driver/Driver.cpp29
-rw-r--r--clang/lib/Driver/Tools.cpp7
3 files changed, 22 insertions, 50 deletions
diff --git a/clang/lib/Driver/Action.cpp b/clang/lib/Driver/Action.cpp
index f9e1024f920..425f315ccd6 100644
--- a/clang/lib/Driver/Action.cpp
+++ b/clang/lib/Driver/Action.cpp
@@ -51,43 +51,11 @@ void BindArchAction::anchor() {}
BindArchAction::BindArchAction(Action *Input, const char *_ArchName)
: Action(BindArchClass, Input), ArchName(_ArchName) {}
-// Converts CUDA GPU architecture, e.g. "sm_21", to its corresponding virtual
-// compute arch, e.g. "compute_20". Returns null if the input arch is null or
-// doesn't match an existing arch.
-static const char* GpuArchToComputeName(const char *ArchName) {
- if (!ArchName)
- return nullptr;
- return llvm::StringSwitch<const char *>(ArchName)
- .Cases("sm_20", "sm_21", "compute_20")
- .Case("sm_30", "compute_30")
- .Case("sm_32", "compute_32")
- .Case("sm_35", "compute_35")
- .Case("sm_37", "compute_37")
- .Case("sm_50", "compute_50")
- .Case("sm_52", "compute_52")
- .Case("sm_53", "compute_53")
- .Case("sm_60", "compute_60")
- .Case("sm_61", "compute_61")
- .Case("sm_62", "compute_62")
- .Default(nullptr);
-}
-
void CudaDeviceAction::anchor() {}
-CudaDeviceAction::CudaDeviceAction(Action *Input, const char *ArchName,
+CudaDeviceAction::CudaDeviceAction(Action *Input, CudaArch Arch,
bool AtTopLevel)
- : Action(CudaDeviceClass, Input), GpuArchName(ArchName),
- AtTopLevel(AtTopLevel) {
- assert(!GpuArchName || IsValidGpuArchName(GpuArchName));
-}
-
-const char *CudaDeviceAction::getComputeArchName() const {
- return GpuArchToComputeName(GpuArchName);
-}
-
-bool CudaDeviceAction::IsValidGpuArchName(llvm::StringRef ArchName) {
- return GpuArchToComputeName(ArchName.data()) != nullptr;
-}
+ : Action(CudaDeviceClass, Input), GpuArch(Arch), AtTopLevel(AtTopLevel) {}
void CudaHostAction::anchor() {}
diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
index 7235bd0efc8..78c3125cdb6 100644
--- a/clang/lib/Driver/Driver.cpp
+++ b/clang/lib/Driver/Driver.cpp
@@ -23,6 +23,7 @@
#include "clang/Driver/ToolChain.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringSet.h"
#include "llvm/ADT/StringSwitch.h"
@@ -1022,9 +1023,10 @@ static unsigned PrintActions1(const Compilation &C, Action *A,
os << '"' << BIA->getArchName() << '"' << ", {"
<< PrintActions1(C, *BIA->input_begin(), Ids) << "}";
} else if (CudaDeviceAction *CDA = dyn_cast<CudaDeviceAction>(A)) {
- os << '"'
- << (CDA->getGpuArchName() ? CDA->getGpuArchName() : "(multiple archs)")
- << '"' << ", {" << PrintActions1(C, *CDA->input_begin(), Ids) << "}";
+ CudaArch Arch = CDA->getGpuArch();
+ if (Arch != CudaArch::UNKNOWN)
+ os << "'" << CudaArchToString(Arch) << "', ";
+ os << "{" << PrintActions1(C, *CDA->input_begin(), Ids) << "}";
} else {
const ActionList *AL;
if (CudaHostAction *CHA = dyn_cast<CudaHostAction>(A)) {
@@ -1380,24 +1382,25 @@ static Action *buildCudaActions(Compilation &C, DerivedArgList &Args,
return C.MakeAction<CudaHostAction>(HostAction, ActionList());
// Collect all cuda_gpu_arch parameters, removing duplicates.
- SmallVector<const char *, 4> GpuArchList;
- llvm::StringSet<> GpuArchNames;
+ SmallVector<CudaArch, 4> GpuArchList;
+ llvm::SmallSet<CudaArch, 4> GpuArchs;
for (Arg *A : Args) {
if (!A->getOption().matches(options::OPT_cuda_gpu_arch_EQ))
continue;
A->claim();
- const auto& Arch = A->getValue();
- if (!CudaDeviceAction::IsValidGpuArchName(Arch))
- C.getDriver().Diag(clang::diag::err_drv_cuda_bad_gpu_arch) << Arch;
- else if (GpuArchNames.insert(Arch).second)
+ const auto &ArchStr = A->getValue();
+ CudaArch Arch = StringToCudaArch(ArchStr);
+ if (Arch == CudaArch::UNKNOWN)
+ C.getDriver().Diag(clang::diag::err_drv_cuda_bad_gpu_arch) << ArchStr;
+ else if (GpuArchs.insert(Arch).second)
GpuArchList.push_back(Arch);
}
// Default to sm_20 which is the lowest common denominator for supported GPUs.
// sm_20 code should work correctly, if suboptimally, on all newer GPUs.
if (GpuArchList.empty())
- GpuArchList.push_back("sm_20");
+ GpuArchList.push_back(CudaArch::SM_20);
// Replicate inputs for each GPU architecture.
Driver::InputList CudaDeviceInputs;
@@ -1463,7 +1466,7 @@ static Action *buildCudaActions(Compilation &C, DerivedArgList &Args,
}
auto FatbinAction = C.MakeAction<CudaDeviceAction>(
C.MakeAction<LinkJobAction>(DeviceActions, types::TY_CUDA_FATBIN),
- /* GpuArchName = */ nullptr,
+ CudaArch::UNKNOWN,
/* AtTopLevel = */ false);
// Return a new host action that incorporates original host action and all
// device actions.
@@ -2047,8 +2050,8 @@ InputInfo Driver::BuildJobsForActionNoCache(
// Call BuildJobsForAction() again, now with correct device parameters.
InputInfo II = BuildJobsForAction(
C, *CDA->input_begin(), C.getSingleOffloadToolChain<Action::OFK_Cuda>(),
- CDA->getGpuArchName(), CDA->isAtTopLevel(), /*MultipleArchs=*/true,
- LinkingOutput, CachedResults);
+ CudaArchToString(CDA->getGpuArch()), CDA->isAtTopLevel(),
+ /*MultipleArchs=*/true, LinkingOutput, CachedResults);
// Currently II's Action is *CDA->input_begin(). Set it to CDA instead, so
// that one can retrieve II's GPU arch.
II.setAction(A);
diff --git a/clang/lib/Driver/Tools.cpp b/clang/lib/Driver/Tools.cpp
index 2425619d4c9..14ed0d4e3ed 100644
--- a/clang/lib/Driver/Tools.cpp
+++ b/clang/lib/Driver/Tools.cpp
@@ -11222,9 +11222,10 @@ void NVPTX::Linker::ConstructJob(Compilation &C, const JobAction &JA,
auto* A = cast<const CudaDeviceAction>(II.getAction());
// We need to pass an Arch of the form "sm_XX" for cubin files and
// "compute_XX" for ptx.
- const char *Arch = (II.getType() == types::TY_PP_Asm)
- ? A->getComputeArchName()
- : A->getGpuArchName();
+ const char *Arch =
+ (II.getType() == types::TY_PP_Asm)
+ ? CudaVirtualArchToString(VirtualArchForCudaArch(A->getGpuArch()))
+ : CudaArchToString(A->getGpuArch());
CmdArgs.push_back(Args.MakeArgString(llvm::Twine("--image=profile=") +
Arch + ",file=" + II.getFilename()));
}
OpenPOWER on IntegriCloud