jlebar updated this revision to Diff 55646.
jlebar added a comment.
Add more comprehensive test.
http://reviews.llvm.org/D19346
Files:
include/clang/Basic/TargetOptions.h
lib/Basic/Targets.cpp
lib/Frontend/CompilerInstance.cpp
lib/Frontend/CompilerInvocation.cpp
test/Preprocessor/cuda-types.cu
Index: test/Preprocessor/cuda-types.cu
===================================================================
--- /dev/null
+++ test/Preprocessor/cuda-types.cu
@@ -0,0 +1,21 @@
+// Check that types, widths, etc. match on the host and device sides of CUDA
+// compilations. Note that we filter out long double, as this is intentionally
+// different on host and device.
+
+// RUN: %clang --cuda-host-only -nocudainc -target i386-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/i386-host-defines
+// RUN: %clang --cuda-device-only -nocudainc -target i386-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/i386-device-defines
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/i386-host-defines | grep -v __LDBL > %T/i386-host-defines-filtered
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/i386-device-defines | grep -v __LDBL > %T/i386-device-defines-filtered
+// RUN: diff %T/i386-host-defines-filtered %T/i386-device-defines-filtered
+
+// RUN: %clang --cuda-host-only -nocudainc -target x86_64-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/x86_64-host-defines
+// RUN: %clang --cuda-device-only -nocudainc -target x86_64-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/x86_64-device-defines
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/x86_64-host-defines | grep -v __LDBL > %T/x86_64-host-defines-filtered
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/x86_64-device-defines | grep -v __LDBL > %T/x86_64-device-defines-filtered
+// RUN: diff %T/x86_64-host-defines-filtered %T/x86_64-device-defines-filtered
+
+// RUN: %clang --cuda-host-only -nocudainc -target powerpc64-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/powerpc64-host-defines
+// RUN: %clang --cuda-device-only -nocudainc -target powerpc64-unknown-linux-gnu -x cuda -E -dM -o - /dev/null > %T/powerpc64-device-defines
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/powerpc64-host-defines | grep -v __LDBL > %T/powerpc64-host-defines-filtered
+// RUN: grep 'define __[^ ]*\(TYPE\|MAX\)' %T/powerpc64-device-defines | grep -v __LDBL > %T/powerpc64-device-defines-filtered
+// RUN: diff %T/powerpc64-host-defines-filtered %T/powerpc64-device-defines-filtered
Index: lib/Frontend/CompilerInvocation.cpp
===================================================================
--- lib/Frontend/CompilerInvocation.cpp
+++ lib/Frontend/CompilerInvocation.cpp
@@ -2151,6 +2151,12 @@
LangOpts.ObjCExceptions = 1;
}
+ // During CUDA device-side compilation, the aux triple is the triple used for
+ // host compilation.
+ if (LangOpts.CUDA && LangOpts.CUDAIsDevice) {
+ Res.getTargetOpts().HostTriple = Res.getFrontendOpts().AuxTriple;
+ }
+
// FIXME: Override value name discarding when asan or msan is used because the
// backend passes depend on the name of the alloca in order to print out
// names.
Index: lib/Frontend/CompilerInstance.cpp
===================================================================
--- lib/Frontend/CompilerInstance.cpp
+++ lib/Frontend/CompilerInstance.cpp
@@ -836,8 +836,9 @@
// Create TargetInfo for the other side of CUDA compilation.
if (getLangOpts().CUDA && !getFrontendOpts().AuxTriple.empty()) {
- std::shared_ptr<TargetOptions> TO(new TargetOptions);
+ auto TO = std::make_shared<TargetOptions>();
TO->Triple = getFrontendOpts().AuxTriple;
+ TO->HostTriple = getTarget().getTriple().str();
setAuxTarget(TargetInfo::CreateTargetInfo(getDiagnostics(), TO));
}
Index: lib/Basic/Targets.cpp
===================================================================
--- lib/Basic/Targets.cpp
+++ lib/Basic/Targets.cpp
@@ -65,6 +65,9 @@
Builder.defineMacro("__tune_" + CPUName + "__");
}
+static TargetInfo *AllocateTarget(const llvm::Triple &Triple,
+ const TargetOptions &Opts);
+
//===----------------------------------------------------------------------===//
// Defines specific to certain operating systems.
//===----------------------------------------------------------------------===//
@@ -1623,7 +1626,7 @@
} GPU;
public:
- NVPTXTargetInfo(const llvm::Triple &Triple, const TargetOptions &)
+ NVPTXTargetInfo(const llvm::Triple &Triple, const TargetOptions &Opts)
: TargetInfo(Triple) {
BigEndian = false;
TLSSupported = false;
@@ -1635,6 +1638,61 @@
NoAsmVariants = true;
// Set the default GPU to sm20
GPU = GK_SM20;
+
+ std::unique_ptr<TargetInfo> HostTarget(
+ AllocateTarget(llvm::Triple(Opts.HostTriple), Opts));
+ if (!HostTarget) {
+ return;
+ }
+
+ // Match the host's types.
+ PointerWidth = HostTarget->getPointerWidth(/* AddrSpace = */ 0);
+ PointerAlign = HostTarget->getPointerAlign(/* AddrSpace = */ 0);
+ BoolWidth = HostTarget->getBoolWidth();
+ BoolAlign = HostTarget->getBoolAlign();
+ IntWidth = HostTarget->getIntWidth();
+ IntAlign = HostTarget->getIntAlign();
+ HalfWidth = HostTarget->getHalfWidth();
+ HalfAlign = HostTarget->getHalfAlign();
+ FloatWidth = HostTarget->getFloatWidth();
+ FloatAlign = HostTarget->getFloatAlign();
+ DoubleWidth = HostTarget->getDoubleWidth();
+ DoubleAlign = HostTarget->getDoubleAlign();
+ LongWidth = HostTarget->getLongWidth();
+ LongAlign = HostTarget->getLongAlign();
+ LongLongWidth = HostTarget->getLongLongWidth();
+ LongLongAlign = HostTarget->getLongLongAlign();
+ MinGlobalAlign = HostTarget->getMinGlobalAlign();
+ DefaultAlignForAttributeAligned =
+ HostTarget->getDefaultAlignForAttributeAligned();
+ SizeType = HostTarget->getSizeType();
+ IntMaxType = HostTarget->getIntMaxType();
+ PtrDiffType = HostTarget->getPtrDiffType(/* AddrSpace = */ 0);
+ IntPtrType = HostTarget->getIntPtrType();
+ WCharType = HostTarget->getWCharType();
+ WIntType = HostTarget->getWIntType();
+ Char16Type = HostTarget->getChar16Type();
+ Char32Type = HostTarget->getChar32Type();
+ Int64Type = HostTarget->getInt64Type();
+ SigAtomicType = HostTarget->getSigAtomicType();
+ ProcessIDType = HostTarget->getProcessIDType();
+
+ UseBitFieldTypeAlignment = HostTarget->useBitFieldTypeAlignment();
+ UseZeroLengthBitfieldAlignment =
+ HostTarget->useZeroLengthBitfieldAlignment();
+ UseExplicitBitFieldAlignment = HostTarget->useExplicitBitFieldAlignment();
+ ZeroLengthBitfieldBoundary = HostTarget->getZeroLengthBitfieldBoundary();
+
+ // Properties intentionally not copied from host:
+ // - LargeArrayMinWidth, LargeArrayAlign: Not visible across the
+ // host/device boundary.
+ // - SuitableAlign: Not visible across the host/device boundary, and may
+ // correctly be different on host/device, e.g. if host has wider vector
+ // types than device.
+ // - LongDoubleWidth, LongDoubleAlign: nvptx's long double type is the same
+ // as its double type, but that's not necessarily true on the host.
+ // TODO: nvcc emits a warning when using long double on device; we should
+ // do the same.
}
void getTargetDefines(const LangOptions &Opts,
MacroBuilder &Builder) const override {
Index: include/clang/Basic/TargetOptions.h
===================================================================
--- include/clang/Basic/TargetOptions.h
+++ include/clang/Basic/TargetOptions.h
@@ -27,6 +27,10 @@
/// target will be selected to match the host.
std::string Triple;
+ /// When compiling for the device side, contains the triple used to compile
+ /// for the host.
+ std::string HostTriple;
+
/// If given, the name of the target CPU to generate code for.
std::string CPU;
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
