jhuber6 created this revision.
jhuber6 added reviewers: jdoerfert, tianshilei1992, JonChesterfield, tra,
yaxunl.
Herald added a subscriber: guansong.
Herald added a project: All.
jhuber6 requested review of this revision.
Herald added subscribers: cfe-commits, sstefan1, MaskRay.
Herald added a project: clang.
This patch adds support for '--offload-arch=native' to OpenMP
offloading. This will automatically generate the toolchains required to
fulfil whatever GPUs the user has installed. Getting this to work
requires a bit of a hack. The problem is that we need the ToolChain to
launch its searching program. But we do not yet have that ToolChain
built. I had to temporarily make the ToolChain and also add some logic
to ignore regular warnings & errors.
Depends on D141078 <https://reviews.llvm.org/D141078>
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D141105
Files:
clang/include/clang/Driver/Driver.h
clang/lib/Driver/Driver.cpp
Index: clang/lib/Driver/Driver.cpp
===================================================================
--- clang/lib/Driver/Driver.cpp
+++ clang/lib/Driver/Driver.cpp
@@ -848,9 +848,29 @@
HostTC->getTriple());
// Attempt to deduce the offloading triple from the set of architectures.
- // We can only correctly deduce NVPTX / AMDGPU triples currently.
- llvm::DenseSet<StringRef> Archs =
- getOffloadArchs(C, C.getArgs(), Action::OFK_OpenMP, nullptr);
+ // We can only correctly deduce NVPTX / AMDGPU triples currently. We need
+ // to temporarily create these toolchains so that we can access tools for
+ // inferring architectures.
+ llvm::DenseSet<StringRef> Archs;
+ if (NVPTXTriple) {
+ auto TempTC = std::make_unique<toolchains::CudaToolChain>(
+ *this, *NVPTXTriple, *HostTC, C.getInputArgs());
+ for (StringRef Arch : getOffloadArchs(
+ C, C.getArgs(), Action::OFK_OpenMP, &*TempTC, true))
+ Archs.insert(Arch);
+ }
+ if (AMDTriple) {
+ auto TempTC = std::make_unique<toolchains::AMDGPUOpenMPToolChain>(
+ *this, *AMDTriple, *HostTC, C.getInputArgs());
+ for (StringRef Arch : getOffloadArchs(
+ C, C.getArgs(), Action::OFK_OpenMP, &*TempTC, true))
+ Archs.insert(Arch);
+ }
+ if (!AMDTriple && !NVPTXTriple) {
+ for (StringRef Arch :
+ getOffloadArchs(C, C.getArgs(), Action::OFK_OpenMP, nullptr, true))
+ Archs.insert(Arch);
+ }
for (StringRef Arch : Archs) {
if (NVPTXTriple && IsNVIDIAGpuArch(StringToCudaArch(
getProcessorFromTargetID(*NVPTXTriple, Arch)))) {
@@ -4182,16 +4202,17 @@
static StringRef getCanonicalArchString(Compilation &C,
const llvm::opt::DerivedArgList &Args,
StringRef ArchStr,
- const llvm::Triple &Triple) {
+ const llvm::Triple &Triple,
+ bool Query = false) {
// Lookup the CUDA / HIP architecture string. Only report an error if we were
// expecting the triple to be only NVPTX / AMDGPU.
CudaArch Arch = StringToCudaArch(getProcessorFromTargetID(Triple, ArchStr));
- if (Triple.isNVPTX() &&
+ if (!Query && Triple.isNVPTX() &&
(Arch == CudaArch::UNKNOWN || !IsNVIDIAGpuArch(Arch))) {
C.getDriver().Diag(clang::diag::err_drv_offload_bad_gpu_arch)
<< "CUDA" << ArchStr;
return StringRef();
- } else if (Triple.isAMDGPU() &&
+ } else if (!Query && Triple.isAMDGPU() &&
(Arch == CudaArch::UNKNOWN || !IsAMDGpuArch(Arch))) {
C.getDriver().Diag(clang::diag::err_drv_offload_bad_gpu_arch)
<< "HIP" << ArchStr;
@@ -4234,7 +4255,8 @@
llvm::DenseSet<StringRef>
Driver::getOffloadArchs(Compilation &C, const llvm::opt::DerivedArgList &Args,
- Action::OffloadKind Kind, const ToolChain *TC) const {
+ Action::OffloadKind Kind, const ToolChain *TC,
+ bool Query) const {
if (!TC)
TC = &C.getDefaultToolChain();
@@ -4271,18 +4293,22 @@
if (Arch == "native") {
auto GPUsOrErr = TC->getSystemGPUArchs(Args);
if (!GPUsOrErr) {
- TC->getDriver().Diag(diag::err_drv_undetermined_gpu_arch)
- << llvm::Triple::getArchTypeName(TC->getArch())
- << llvm::toString(GPUsOrErr.takeError()) << "--offload-arch";
+ if (Query)
+ llvm::consumeError(GPUsOrErr.takeError());
+ else
+ TC->getDriver().Diag(diag::err_drv_undetermined_gpu_arch)
+ << llvm::Triple::getArchTypeName(TC->getArch())
+ << llvm::toString(GPUsOrErr.takeError()) << "--offload-arch";
continue;
}
- for (auto ArchStr : *GPUsOrErr)
- Archs.insert(
- getCanonicalArchString(C, Args, ArchStr, TC->getTriple()));
+ for (auto ArchStr : *GPUsOrErr) {
+ Archs.insert(getCanonicalArchString(
+ C, Args, Args.MakeArgString(ArchStr), TC->getTriple(), Query));
+ }
} else {
StringRef ArchStr =
- getCanonicalArchString(C, Args, Arch, TC->getTriple());
+ getCanonicalArchString(C, Args, Arch, TC->getTriple(), Query);
if (ArchStr.empty())
return Archs;
Archs.insert(ArchStr);
@@ -4294,7 +4320,7 @@
Archs.clear();
} else {
StringRef ArchStr =
- getCanonicalArchString(C, Args, Arch, TC->getTriple());
+ getCanonicalArchString(C, Args, Arch, TC->getTriple(), Query);
if (ArchStr.empty())
return Archs;
Archs.erase(ArchStr);
@@ -4309,6 +4335,10 @@
C.setContainsError();
}
+ // Skip filling defaults if we're just querying what is availible.
+ if (Query)
+ return Archs;
+
if (Archs.empty()) {
if (Kind == Action::OFK_Cuda)
Archs.insert(CudaArchToString(CudaArch::CudaDefault));
Index: clang/include/clang/Driver/Driver.h
===================================================================
--- clang/include/clang/Driver/Driver.h
+++ clang/include/clang/Driver/Driver.h
@@ -481,10 +481,11 @@
/// Returns the set of bound architectures active for this offload kind.
/// If there are no bound architctures we return a set containing only the
- /// empty string.
+ /// empty string. The \p Query option is used to suppress errors on failure.
llvm::DenseSet<StringRef>
getOffloadArchs(Compilation &C, const llvm::opt::DerivedArgList &Args,
- Action::OffloadKind Kind, const ToolChain *TC) const;
+ Action::OffloadKind Kind, const ToolChain *TC,
+ bool Query = false) const;
/// Check that the file referenced by Value exists. If it doesn't,
/// issue a diagnostic and return false.
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
