llvm-project/clang/lib/Basic/Builtins.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

253 lines
9.1 KiB
C++
Raw Normal View History

//===--- Builtins.cpp - Builtin function implementation -------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file implements various things for builtin functions.
//
//===----------------------------------------------------------------------===//
#include "clang/Basic/Builtins.h"
#include "BuiltinTargetFeatures.h"
#include "clang/Basic/IdentifierTable.h"
#include "clang/Basic/LangOptions.h"
#include "clang/Basic/TargetInfo.h"
#include "llvm/ADT/StringRef.h"
using namespace clang;
const char *HeaderDesc::getName() const {
switch (ID) {
#define HEADER(ID, NAME) \
case ID: \
return NAME;
#include "clang/Basic/BuiltinHeaders.def"
#undef HEADER
};
llvm_unreachable("Unknown HeaderDesc::HeaderID enum");
}
static constexpr Builtin::Info BuiltinInfo[] = {
{"not a builtin function", nullptr, nullptr, nullptr, HeaderDesc::NO_HEADER,
ALL_LANGUAGES},
#define BUILTIN(ID, TYPE, ATTRS) \
{#ID, TYPE, ATTRS, nullptr, HeaderDesc::NO_HEADER, ALL_LANGUAGES},
#define LANGBUILTIN(ID, TYPE, ATTRS, LANGS) \
{#ID, TYPE, ATTRS, nullptr, HeaderDesc::NO_HEADER, LANGS},
#define LIBBUILTIN(ID, TYPE, ATTRS, HEADER, LANGS) \
{#ID, TYPE, ATTRS, nullptr, HeaderDesc::HEADER, LANGS},
#include "clang/Basic/Builtins.def"
};
const Builtin::Info &Builtin::Context::getRecord(unsigned ID) const {
if (ID < Builtin::FirstTSBuiltin)
return BuiltinInfo[ID];
assert(((ID - Builtin::FirstTSBuiltin) <
(TSRecords.size() + AuxTSRecords.size())) &&
"Invalid builtin ID!");
if (isAuxBuiltinID(ID))
return AuxTSRecords[getAuxBuiltinID(ID) - Builtin::FirstTSBuiltin];
return TSRecords[ID - Builtin::FirstTSBuiltin];
}
void Builtin::Context::InitializeTarget(const TargetInfo &Target,
const TargetInfo *AuxTarget) {
assert(TSRecords.empty() && "Already initialized target?");
TSRecords = Target.getTargetBuiltins();
if (AuxTarget)
AuxTSRecords = AuxTarget->getTargetBuiltins();
}
bool Builtin::Context::isBuiltinFunc(llvm::StringRef FuncName) {
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
bool InStdNamespace = FuncName.consume_front("std-");
for (unsigned i = Builtin::NotBuiltin + 1; i != Builtin::FirstTSBuiltin;
++i) {
if (FuncName.equals(BuiltinInfo[i].Name) &&
(bool)strchr(BuiltinInfo[i].Attributes, 'z') == InStdNamespace)
return strchr(BuiltinInfo[i].Attributes, 'f') != nullptr;
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
}
return false;
}
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
/// Is this builtin supported according to the given language options?
static bool builtinIsSupported(const Builtin::Info &BuiltinInfo,
const LangOptions &LangOpts) {
/* Builtins Unsupported */
if (LangOpts.NoBuiltin && strchr(BuiltinInfo.Attributes, 'f') != nullptr)
return false;
/* CorBuiltins Unsupported */
if (!LangOpts.Coroutines && (BuiltinInfo.Langs & COR_LANG))
return false;
/* MathBuiltins Unsupported */
if (LangOpts.NoMathBuiltin && BuiltinInfo.Header.ID == HeaderDesc::MATH_H)
return false;
/* GnuMode Unsupported */
if (!LangOpts.GNUMode && (BuiltinInfo.Langs & GNU_LANG))
return false;
/* MSMode Unsupported */
if (!LangOpts.MicrosoftExt && (BuiltinInfo.Langs & MS_LANG))
return false;
/* ObjC Unsupported */
if (!LangOpts.ObjC && BuiltinInfo.Langs == OBJC_LANG)
return false;
/* OpenCLC Unsupported */
if (!LangOpts.OpenCL && (BuiltinInfo.Langs & ALL_OCL_LANGUAGES))
return false;
/* OopenCL GAS Unsupported */
if (!LangOpts.OpenCLGenericAddressSpace && (BuiltinInfo.Langs & OCL_GAS))
return false;
/* OpenCL Pipe Unsupported */
if (!LangOpts.OpenCLPipes && (BuiltinInfo.Langs & OCL_PIPE))
return false;
// Device side enqueue is not supported until OpenCL 2.0. In 2.0 and higher
// support is indicated with language option for blocks.
/* OpenCL DSE Unsupported */
if ((LangOpts.getOpenCLCompatibleVersion() < 200 || !LangOpts.Blocks) &&
(BuiltinInfo.Langs & OCL_DSE))
return false;
/* OpenMP Unsupported */
if (!LangOpts.OpenMP && BuiltinInfo.Langs == OMP_LANG)
return false;
/* CUDA Unsupported */
if (!LangOpts.CUDA && BuiltinInfo.Langs == CUDA_LANG)
return false;
/* CPlusPlus Unsupported */
if (!LangOpts.CPlusPlus && BuiltinInfo.Langs == CXX_LANG)
return false;
return true;
}
/// initializeBuiltins - Mark the identifiers for all the builtins with their
/// appropriate builtin ID # and mark any non-portable builtin identifiers as
/// such.
void Builtin::Context::initializeBuiltins(IdentifierTable &Table,
const LangOptions& LangOpts) {
// Step #1: mark all target-independent builtins with their ID's.
for (unsigned i = Builtin::NotBuiltin+1; i != Builtin::FirstTSBuiltin; ++i)
if (builtinIsSupported(BuiltinInfo[i], LangOpts)) {
Table.get(BuiltinInfo[i].Name).setBuiltinID(i);
}
// Step #2: Register target-specific builtins.
for (unsigned i = 0, e = TSRecords.size(); i != e; ++i)
if (builtinIsSupported(TSRecords[i], LangOpts))
Table.get(TSRecords[i].Name).setBuiltinID(i + Builtin::FirstTSBuiltin);
// Step #3: Register target-specific builtins for AuxTarget.
for (unsigned i = 0, e = AuxTSRecords.size(); i != e; ++i)
Table.get(AuxTSRecords[i].Name)
.setBuiltinID(i + Builtin::FirstTSBuiltin + TSRecords.size());
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
// Step #4: Unregister any builtins specified by -fno-builtin-foo.
for (llvm::StringRef Name : LangOpts.NoBuiltinFuncs) {
bool InStdNamespace = Name.consume_front("std-");
auto NameIt = Table.find(Name);
if (NameIt != Table.end()) {
unsigned ID = NameIt->second->getBuiltinID();
if (ID != Builtin::NotBuiltin && isPredefinedLibFunction(ID) &&
isInStdNamespace(ID) == InStdNamespace) {
Table.get(Name).setBuiltinID(Builtin::NotBuiltin);
}
}
}
}
[Builtins][Attributes][X86] Tag all X86 builtins with their required vector width. Add a min_vector_width function attribute and tag all x86 instrinsics with it This is part of an ongoing attempt at making 512 bit vectors illegal in the X86 backend type legalizer due to CPU frequency penalties associated with wide vectors on Skylake Server CPUs. We want the loop vectorizer to be able to emit IR containing wide vectors as intermediate operations in vectorized code and allow these wide vectors to be legalized to 256 bits by the X86 backend even though we are targetting a CPU that supports 512 bit vectors. This is similar to what happens with an AVX2 CPU, the vectorizer can emit wide vectors and the backend will split them. We want this splitting behavior, but still be able to use new Skylake instructions that work on 256-bit vectors and support things like masking and gather/scatter. Of course if the user uses explicit vector code in their source code we need to not split those operations. Especially if they have used any of the 512-bit vector intrinsics from immintrin.h. And we need to make it so that merely using the intrinsics produces the expected code in order to be backwards compatible. To support this goal, this patch adds a new IR function attribute "min-legal-vector-width" that can indicate the need for a minimum vector width to be legal in the backend. We need to ensure this attribute is set to the largest vector width needed by any intrinsics from immintrin.h that the function uses. The inliner will be reponsible for merging this attribute when a function is inlined. We may also need a way to limit inlining in the future as well, but we can discuss that in the future. To make things more complicated, there are two different ways intrinsics are implemented in immintrin.h. Either as an always_inline function containing calls to builtins(can be target specific or target independent) or vector extension code. Or as a macro wrapper around a taget specific builtin. I believe I've removed all cases where the macro was around a target independent builtin. To support the always_inline function case this patch adds attribute((min_vector_width(128))) that can be used to tag these functions with their vector width. All x86 intrinsic functions that operate on vectors have been tagged with this attribute. To support the macro case, all x86 specific builtins have also been tagged with the vector width that they require. Use of any builtin with this property will implicitly increase the min_vector_width of the function that calls it. I've done this as a new property in the attribute string for the builtin rather than basing it on the type string so that we can opt into it on a per builtin basis and avoid any impact to target independent builtins. There will be future work to support vectors passed as function arguments and supporting inline assembly. And whatever else we can find that isn't covered by this patch. Special thanks to Chandler who suggested this direction and reviewed a preview version of this patch. And thanks to Eric Christopher who has had many conversations with me about this issue. Differential Revision: https://reviews.llvm.org/D48617 llvm-svn: 336583
2018-07-09 19:00:16 +00:00
unsigned Builtin::Context::getRequiredVectorWidth(unsigned ID) const {
const char *WidthPos = ::strchr(getRecord(ID).Attributes, 'V');
if (!WidthPos)
return 0;
++WidthPos;
assert(*WidthPos == ':' &&
"Vector width specifier must be followed by a ':'");
++WidthPos;
char *EndPos;
unsigned Width = ::strtol(WidthPos, &EndPos, 10);
assert(*EndPos == ':' && "Vector width specific must end with a ':'");
return Width;
}
bool Builtin::Context::isLike(unsigned ID, unsigned &FormatIdx,
bool &HasVAListArg, const char *Fmt) const {
assert(Fmt && "Not passed a format string");
assert(::strlen(Fmt) == 2 &&
"Format string needs to be two characters long");
assert(::toupper(Fmt[0]) == Fmt[1] &&
"Format string is not in the form \"xX\"");
const char *Like = ::strpbrk(getRecord(ID).Attributes, Fmt);
if (!Like)
return false;
HasVAListArg = (*Like == Fmt[1]);
++Like;
assert(*Like == ':' && "Format specifier must be followed by a ':'");
++Like;
assert(::strchr(Like, ':') && "Format specifier must end with a ':'");
FormatIdx = ::strtol(Like, nullptr, 10);
return true;
}
bool Builtin::Context::isPrintfLike(unsigned ID, unsigned &FormatIdx,
bool &HasVAListArg) {
return isLike(ID, FormatIdx, HasVAListArg, "pP");
}
bool Builtin::Context::isScanfLike(unsigned ID, unsigned &FormatIdx,
bool &HasVAListArg) {
return isLike(ID, FormatIdx, HasVAListArg, "sS");
}
bool Builtin::Context::performsCallback(unsigned ID,
SmallVectorImpl<int> &Encoding) const {
const char *CalleePos = ::strchr(getRecord(ID).Attributes, 'C');
if (!CalleePos)
return false;
++CalleePos;
assert(*CalleePos == '<' &&
"Callback callee specifier must be followed by a '<'");
++CalleePos;
char *EndPos;
int CalleeIdx = ::strtol(CalleePos, &EndPos, 10);
assert(CalleeIdx >= 0 && "Callee index is supposed to be positive!");
Encoding.push_back(CalleeIdx);
while (*EndPos == ',') {
const char *PayloadPos = EndPos + 1;
int PayloadIdx = ::strtol(PayloadPos, &EndPos, 10);
Encoding.push_back(PayloadIdx);
}
assert(*EndPos == '>' && "Callback callee specifier must end with a '>'");
return true;
}
bool Builtin::Context::canBeRedeclared(unsigned ID) const {
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
return ID == Builtin::NotBuiltin || ID == Builtin::BI__va_start ||
ID == Builtin::BI__builtin_assume_aligned ||
Treat `std::move`, `forward`, etc. as builtins. This is extended to all `std::` functions that take a reference to a value and return a reference (or pointer) to that same value: `move`, `forward`, `move_if_noexcept`, `as_const`, `addressof`, and the libstdc++-specific function `__addressof`. We still require these functions to be declared before they can be used, but don't instantiate their definitions unless their addresses are taken. Instead, code generation, constant evaluation, and static analysis are given direct knowledge of their effect. This change aims to reduce various costs associated with these functions -- per-instantiation memory costs, compile time and memory costs due to creating out-of-line copies and inlining them, code size at -O0, and so on -- so that they are not substantially more expensive than a cast. Most of these improvements are very small, but I measured a 3% decrease in -O0 object file size for a simple C++ source file using the standard library after this change. We now automatically infer the `const` and `nothrow` attributes on these now-builtin functions, in particular meaning that we get a warning for an unused call to one of these functions. In C++20 onwards, we disallow taking the addresses of these functions, per the C++20 "addressable function" rule. In earlier language modes, a compatibility warning is produced but the address can still be taken. The same infrastructure is extended to the existing MSVC builtin `__GetExceptionInfo`, which is now only recognized in namespace `std` like it always should have been. This is a re-commit of fc3090109643af8d2da9822d0f99c84742b9c877, a571f82a50416b767fd3cce0fb5027bb5dfec58c, 64c045e25b8471bbb572bd29159c294a82a86a2, and de6ddaeef3aaa8a9ae3663c12cdb57d9afc0f906, and reverts aa643f455a5362de7189eac630050d2c8aefe8f2. This change also includes a workaround for users using libc++ 3.1 and earlier (!!), as apparently happens on AIX, where std::move sometimes returns by value. Reviewed By: aaron.ballman Differential Revision: https://reviews.llvm.org/D123345 Revert "Fixup D123950 to address revert of D123345" This reverts commit aa643f455a5362de7189eac630050d2c8aefe8f2.
2022-04-20 17:13:56 -07:00
(!hasReferenceArgsOrResult(ID) && !hasCustomTypechecking(ID)) ||
isInStdNamespace(ID);
}
bool Builtin::evaluateRequiredTargetFeatures(
StringRef RequiredFeatures, const llvm::StringMap<bool> &TargetFetureMap) {
// Return true if the builtin doesn't have any required features.
if (RequiredFeatures.empty())
return true;
assert(!RequiredFeatures.contains(' ') && "Space in feature list");
TargetFeatures TF(TargetFetureMap);
return TF.hasRequiredFeatures(RequiredFeatures);
}