
This is a major change on how we represent nested name qualifications in the AST. * The nested name specifier itself and how it's stored is changed. The prefixes for types are handled within the type hierarchy, which makes canonicalization for them super cheap, no memory allocation required. Also translating a type into nested name specifier form becomes a no-op. An identifier is stored as a DependentNameType. The nested name specifier gains a lightweight handle class, to be used instead of passing around pointers, which is similar to what is implemented for TemplateName. There is still one free bit available, and this handle can be used within a PointerUnion and PointerIntPair, which should keep bit-packing aficionados happy. * The ElaboratedType node is removed, all type nodes in which it could previously apply to can now store the elaborated keyword and name qualifier, tail allocating when present. * TagTypes can now point to the exact declaration found when producing these, as opposed to the previous situation of there only existing one TagType per entity. This increases the amount of type sugar retained, and can have several applications, for example in tracking module ownership, and other tools which care about source file origins, such as IWYU. These TagTypes are lazily allocated, in order to limit the increase in AST size. This patch offers a great performance benefit. It greatly improves compilation time for [stdexec](https://github.com/NVIDIA/stdexec). For one datapoint, for `test_on2.cpp` in that project, which is the slowest compiling test, this patch improves `-c` compilation time by about 7.2%, with the `-fsyntax-only` improvement being at ~12%. This has great results on compile-time-tracker as well:  This patch also further enables other optimziations in the future, and will reduce the performance impact of template specialization resugaring when that lands. It has some other miscelaneous drive-by fixes. About the review: Yes the patch is huge, sorry about that. Part of the reason is that I started by the nested name specifier part, before the ElaboratedType part, but that had a huge performance downside, as ElaboratedType is a big performance hog. I didn't have the steam to go back and change the patch after the fact. There is also a lot of internal API changes, and it made sense to remove ElaboratedType in one go, versus removing it from one type at a time, as that would present much more churn to the users. Also, the nested name specifier having a different API avoids missing changes related to how prefixes work now, which could make existing code compile but not work. How to review: The important changes are all in `clang/include/clang/AST` and `clang/lib/AST`, with also important changes in `clang/lib/Sema/TreeTransform.h`. The rest and bulk of the changes are mostly consequences of the changes in API. PS: TagType::getDecl is renamed to `getOriginalDecl` in this patch, just for easier to rebasing. I plan to rename it back after this lands. Fixes #136624 Fixes https://github.com/llvm/llvm-project/issues/43179 Fixes https://github.com/llvm/llvm-project/issues/68670 Fixes https://github.com/llvm/llvm-project/issues/92757
1646 lines
60 KiB
C++
1646 lines
60 KiB
C++
//=== SemaFunctionEffects.cpp - Sema handling of function effects ---------===//
|
|
//
|
|
// 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 Sema handling of function effects.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "clang/AST/Decl.h"
|
|
#include "clang/AST/DeclCXX.h"
|
|
#include "clang/AST/DynamicRecursiveASTVisitor.h"
|
|
#include "clang/AST/ExprObjC.h"
|
|
#include "clang/AST/Stmt.h"
|
|
#include "clang/AST/StmtObjC.h"
|
|
#include "clang/AST/Type.h"
|
|
#include "clang/Basic/SourceManager.h"
|
|
#include "clang/Sema/SemaInternal.h"
|
|
|
|
#define DEBUG_TYPE "effectanalysis"
|
|
|
|
using namespace clang;
|
|
|
|
namespace {
|
|
|
|
enum class ViolationID : uint8_t {
|
|
None = 0, // Sentinel for an empty Violation.
|
|
// These first 5 map to a %select{} in one of several FunctionEffects
|
|
// diagnostics, e.g. warn_func_effect_violation.
|
|
BaseDiagnosticIndex,
|
|
AllocatesMemory = BaseDiagnosticIndex,
|
|
ThrowsOrCatchesExceptions,
|
|
HasStaticLocalVariable,
|
|
AccessesThreadLocalVariable,
|
|
AccessesObjCMethodOrProperty,
|
|
|
|
// These only apply to callees, where the analysis stops at the Decl.
|
|
DeclDisallowsInference,
|
|
|
|
// These both apply to indirect calls. The difference is that sometimes
|
|
// we have an actual Decl (generally a variable) which is the function
|
|
// pointer being called, and sometimes, typically due to a cast, we only
|
|
// have an expression.
|
|
CallsDeclWithoutEffect,
|
|
CallsExprWithoutEffect,
|
|
};
|
|
|
|
// Information about the AST context in which a violation was found, so
|
|
// that diagnostics can point to the correct source.
|
|
class ViolationSite {
|
|
public:
|
|
enum class Kind : uint8_t {
|
|
Default, // Function body.
|
|
MemberInitializer,
|
|
DefaultArgExpr
|
|
};
|
|
|
|
private:
|
|
llvm::PointerIntPair<CXXDefaultArgExpr *, 2, Kind> Impl;
|
|
|
|
public:
|
|
ViolationSite() = default;
|
|
|
|
explicit ViolationSite(CXXDefaultArgExpr *E)
|
|
: Impl(E, Kind::DefaultArgExpr) {}
|
|
|
|
Kind kind() const { return static_cast<Kind>(Impl.getInt()); }
|
|
CXXDefaultArgExpr *defaultArgExpr() const { return Impl.getPointer(); }
|
|
|
|
void setKind(Kind K) { Impl.setPointerAndInt(nullptr, K); }
|
|
};
|
|
|
|
// Represents a violation of the rules, potentially for the entire duration of
|
|
// the analysis phase, in order to refer to it when explaining why a caller has
|
|
// been made unsafe by a callee. Can be transformed into either a Diagnostic
|
|
// (warning or a note), depending on whether the violation pertains to a
|
|
// function failing to be verifed as holding an effect vs. a function failing to
|
|
// be inferred as holding that effect.
|
|
struct Violation {
|
|
FunctionEffect Effect;
|
|
std::optional<FunctionEffect>
|
|
CalleeEffectPreventingInference; // Only for certain IDs; can be nullopt.
|
|
ViolationID ID = ViolationID::None;
|
|
ViolationSite Site;
|
|
SourceLocation Loc;
|
|
const Decl *Callee =
|
|
nullptr; // Only valid for ViolationIDs Calls{Decl,Expr}WithoutEffect.
|
|
|
|
Violation(FunctionEffect Effect, ViolationID ID, ViolationSite VS,
|
|
SourceLocation Loc, const Decl *Callee = nullptr,
|
|
std::optional<FunctionEffect> CalleeEffect = std::nullopt)
|
|
: Effect(Effect), CalleeEffectPreventingInference(CalleeEffect), ID(ID),
|
|
Site(VS), Loc(Loc), Callee(Callee) {}
|
|
|
|
unsigned diagnosticSelectIndex() const {
|
|
return unsigned(ID) - unsigned(ViolationID::BaseDiagnosticIndex);
|
|
}
|
|
};
|
|
|
|
enum class SpecialFuncType : uint8_t { None, OperatorNew, OperatorDelete };
|
|
enum class CallableType : uint8_t {
|
|
// Unknown: probably function pointer.
|
|
Unknown,
|
|
Function,
|
|
Virtual,
|
|
Block
|
|
};
|
|
|
|
// Return whether a function's effects CAN be verified.
|
|
// The question of whether it SHOULD be verified is independent.
|
|
static bool functionIsVerifiable(const FunctionDecl *FD) {
|
|
if (FD->isTrivial()) {
|
|
// Otherwise `struct x { int a; };` would have an unverifiable default
|
|
// constructor.
|
|
return true;
|
|
}
|
|
return FD->hasBody();
|
|
}
|
|
|
|
static bool isNoexcept(const FunctionDecl *FD) {
|
|
const auto *FPT = FD->getType()->getAs<FunctionProtoType>();
|
|
return FPT && (FPT->isNothrow() || FD->hasAttr<NoThrowAttr>());
|
|
}
|
|
|
|
// This list is probably incomplete.
|
|
// FIXME: Investigate:
|
|
// __builtin_eh_return?
|
|
// __builtin_allow_runtime_check?
|
|
// __builtin_unwind_init and other similar things that sound exception-related.
|
|
// va_copy?
|
|
// coroutines?
|
|
static FunctionEffectKindSet getBuiltinFunctionEffects(unsigned BuiltinID) {
|
|
FunctionEffectKindSet Result;
|
|
|
|
switch (BuiltinID) {
|
|
case 0: // Not builtin.
|
|
default: // By default, builtins have no known effects.
|
|
break;
|
|
|
|
// These allocate/deallocate heap memory.
|
|
case Builtin::ID::BI__builtin_calloc:
|
|
case Builtin::ID::BI__builtin_malloc:
|
|
case Builtin::ID::BI__builtin_realloc:
|
|
case Builtin::ID::BI__builtin_free:
|
|
case Builtin::ID::BI__builtin_operator_delete:
|
|
case Builtin::ID::BI__builtin_operator_new:
|
|
case Builtin::ID::BIaligned_alloc:
|
|
case Builtin::ID::BIcalloc:
|
|
case Builtin::ID::BImalloc:
|
|
case Builtin::ID::BImemalign:
|
|
case Builtin::ID::BIrealloc:
|
|
case Builtin::ID::BIfree:
|
|
|
|
case Builtin::ID::BIfopen:
|
|
case Builtin::ID::BIpthread_create:
|
|
case Builtin::ID::BI_Block_object_dispose:
|
|
Result.insert(FunctionEffect(FunctionEffect::Kind::Allocating));
|
|
break;
|
|
|
|
// These block in some other way than allocating memory.
|
|
// longjmp() and friends are presumed unsafe because they are the moral
|
|
// equivalent of throwing a C++ exception, which is unsafe.
|
|
case Builtin::ID::BIlongjmp:
|
|
case Builtin::ID::BI_longjmp:
|
|
case Builtin::ID::BIsiglongjmp:
|
|
case Builtin::ID::BI__builtin_longjmp:
|
|
case Builtin::ID::BIobjc_exception_throw:
|
|
|
|
// Objective-C runtime.
|
|
case Builtin::ID::BIobjc_msgSend:
|
|
case Builtin::ID::BIobjc_msgSend_fpret:
|
|
case Builtin::ID::BIobjc_msgSend_fp2ret:
|
|
case Builtin::ID::BIobjc_msgSend_stret:
|
|
case Builtin::ID::BIobjc_msgSendSuper:
|
|
case Builtin::ID::BIobjc_getClass:
|
|
case Builtin::ID::BIobjc_getMetaClass:
|
|
case Builtin::ID::BIobjc_enumerationMutation:
|
|
case Builtin::ID::BIobjc_assign_ivar:
|
|
case Builtin::ID::BIobjc_assign_global:
|
|
case Builtin::ID::BIobjc_sync_enter:
|
|
case Builtin::ID::BIobjc_sync_exit:
|
|
case Builtin::ID::BINSLog:
|
|
case Builtin::ID::BINSLogv:
|
|
|
|
// stdio.h
|
|
case Builtin::ID::BIfread:
|
|
case Builtin::ID::BIfwrite:
|
|
|
|
// stdio.h: printf family.
|
|
case Builtin::ID::BIprintf:
|
|
case Builtin::ID::BI__builtin_printf:
|
|
case Builtin::ID::BIfprintf:
|
|
case Builtin::ID::BIsnprintf:
|
|
case Builtin::ID::BIsprintf:
|
|
case Builtin::ID::BIvprintf:
|
|
case Builtin::ID::BIvfprintf:
|
|
case Builtin::ID::BIvsnprintf:
|
|
case Builtin::ID::BIvsprintf:
|
|
|
|
// stdio.h: scanf family.
|
|
case Builtin::ID::BIscanf:
|
|
case Builtin::ID::BIfscanf:
|
|
case Builtin::ID::BIsscanf:
|
|
case Builtin::ID::BIvscanf:
|
|
case Builtin::ID::BIvfscanf:
|
|
case Builtin::ID::BIvsscanf:
|
|
Result.insert(FunctionEffect(FunctionEffect::Kind::Blocking));
|
|
break;
|
|
}
|
|
|
|
return Result;
|
|
}
|
|
|
|
// Transitory, more extended information about a callable, which can be a
|
|
// function, block, or function pointer.
|
|
struct CallableInfo {
|
|
// CDecl holds the function's definition, if any.
|
|
// FunctionDecl if CallableType::Function or Virtual
|
|
// BlockDecl if CallableType::Block
|
|
const Decl *CDecl;
|
|
|
|
// Remember whether the callable is a function, block, virtual method,
|
|
// or (presumed) function pointer.
|
|
CallableType CType = CallableType::Unknown;
|
|
|
|
// Remember whether the callable is an operator new or delete function,
|
|
// so that calls to them are reported more meaningfully, as memory
|
|
// allocations.
|
|
SpecialFuncType FuncType = SpecialFuncType::None;
|
|
|
|
// We inevitably want to know the callable's declared effects, so cache them.
|
|
FunctionEffectKindSet Effects;
|
|
|
|
CallableInfo(const Decl &CD, SpecialFuncType FT = SpecialFuncType::None)
|
|
: CDecl(&CD), FuncType(FT) {
|
|
FunctionEffectsRef DeclEffects;
|
|
if (auto *FD = dyn_cast<FunctionDecl>(CDecl)) {
|
|
// Use the function's definition, if any.
|
|
if (const FunctionDecl *Def = FD->getDefinition())
|
|
CDecl = FD = Def;
|
|
CType = CallableType::Function;
|
|
if (auto *Method = dyn_cast<CXXMethodDecl>(FD);
|
|
Method && Method->isVirtual())
|
|
CType = CallableType::Virtual;
|
|
DeclEffects = FD->getFunctionEffects();
|
|
} else if (auto *BD = dyn_cast<BlockDecl>(CDecl)) {
|
|
CType = CallableType::Block;
|
|
DeclEffects = BD->getFunctionEffects();
|
|
} else if (auto *VD = dyn_cast<ValueDecl>(CDecl)) {
|
|
// ValueDecl is function, enum, or variable, so just look at its type.
|
|
DeclEffects = FunctionEffectsRef::get(VD->getType());
|
|
}
|
|
Effects = FunctionEffectKindSet(DeclEffects);
|
|
}
|
|
|
|
CallableType type() const { return CType; }
|
|
|
|
bool isCalledDirectly() const {
|
|
return CType == CallableType::Function || CType == CallableType::Block;
|
|
}
|
|
|
|
bool isVerifiable() const {
|
|
switch (CType) {
|
|
case CallableType::Unknown:
|
|
case CallableType::Virtual:
|
|
return false;
|
|
case CallableType::Block:
|
|
return true;
|
|
case CallableType::Function:
|
|
return functionIsVerifiable(dyn_cast<FunctionDecl>(CDecl));
|
|
}
|
|
llvm_unreachable("undefined CallableType");
|
|
}
|
|
|
|
/// Generate a name for logging and diagnostics.
|
|
std::string getNameForDiagnostic(Sema &S) const {
|
|
std::string Name;
|
|
llvm::raw_string_ostream OS(Name);
|
|
|
|
if (auto *FD = dyn_cast<FunctionDecl>(CDecl))
|
|
FD->getNameForDiagnostic(OS, S.getPrintingPolicy(),
|
|
/*Qualified=*/true);
|
|
else if (auto *BD = dyn_cast<BlockDecl>(CDecl))
|
|
OS << "(block " << BD->getBlockManglingNumber() << ")";
|
|
else if (auto *VD = dyn_cast<NamedDecl>(CDecl))
|
|
VD->printQualifiedName(OS);
|
|
return Name;
|
|
}
|
|
};
|
|
|
|
// ----------
|
|
// Map effects to single Violations, to hold the first (of potentially many)
|
|
// violations pertaining to an effect, per function.
|
|
class EffectToViolationMap {
|
|
// Since we currently only have a tiny number of effects (typically no more
|
|
// than 1), use a SmallVector with an inline capacity of 1. Since it
|
|
// is often empty, use a unique_ptr to the SmallVector.
|
|
// Note that Violation itself contains a FunctionEffect which is the key.
|
|
// FIXME: Is there a way to simplify this using existing data structures?
|
|
using ImplVec = llvm::SmallVector<Violation, 1>;
|
|
std::unique_ptr<ImplVec> Impl;
|
|
|
|
public:
|
|
// Insert a new Violation if we do not already have one for its effect.
|
|
void maybeInsert(const Violation &Viol) {
|
|
if (Impl == nullptr)
|
|
Impl = std::make_unique<ImplVec>();
|
|
else if (lookup(Viol.Effect) != nullptr)
|
|
return;
|
|
|
|
Impl->push_back(Viol);
|
|
}
|
|
|
|
const Violation *lookup(FunctionEffect Key) {
|
|
if (Impl == nullptr)
|
|
return nullptr;
|
|
|
|
auto *Iter = llvm::find_if(
|
|
*Impl, [&](const auto &Item) { return Item.Effect == Key; });
|
|
return Iter != Impl->end() ? &*Iter : nullptr;
|
|
}
|
|
|
|
size_t size() const { return Impl ? Impl->size() : 0; }
|
|
};
|
|
|
|
// ----------
|
|
// State pertaining to a function whose AST is walked and whose effect analysis
|
|
// is dependent on a subsequent analysis of other functions.
|
|
class PendingFunctionAnalysis {
|
|
friend class CompleteFunctionAnalysis;
|
|
|
|
public:
|
|
struct DirectCall {
|
|
const Decl *Callee;
|
|
SourceLocation CallLoc;
|
|
// Not all recursive calls are detected, just enough
|
|
// to break cycles.
|
|
bool Recursed = false;
|
|
ViolationSite VSite;
|
|
|
|
DirectCall(const Decl *D, SourceLocation CallLoc, ViolationSite VSite)
|
|
: Callee(D), CallLoc(CallLoc), VSite(VSite) {}
|
|
};
|
|
|
|
// We always have two disjoint sets of effects to verify:
|
|
// 1. Effects declared explicitly by this function.
|
|
// 2. All other inferrable effects needing verification.
|
|
FunctionEffectKindSet DeclaredVerifiableEffects;
|
|
FunctionEffectKindSet EffectsToInfer;
|
|
|
|
private:
|
|
// Violations pertaining to the function's explicit effects.
|
|
SmallVector<Violation, 0> ViolationsForExplicitEffects;
|
|
|
|
// Violations pertaining to other, non-explicit, inferrable effects.
|
|
EffectToViolationMap InferrableEffectToFirstViolation;
|
|
|
|
// These unverified direct calls are what keeps the analysis "pending",
|
|
// until the callees can be verified.
|
|
SmallVector<DirectCall, 0> UnverifiedDirectCalls;
|
|
|
|
public:
|
|
PendingFunctionAnalysis(Sema &S, const CallableInfo &CInfo,
|
|
FunctionEffectKindSet AllInferrableEffectsToVerify)
|
|
: DeclaredVerifiableEffects(CInfo.Effects) {
|
|
// Check for effects we are not allowed to infer.
|
|
FunctionEffectKindSet InferrableEffects;
|
|
|
|
for (FunctionEffect effect : AllInferrableEffectsToVerify) {
|
|
std::optional<FunctionEffect> ProblemCalleeEffect =
|
|
effect.effectProhibitingInference(*CInfo.CDecl, CInfo.Effects);
|
|
if (!ProblemCalleeEffect)
|
|
InferrableEffects.insert(effect);
|
|
else {
|
|
// Add a Violation for this effect if a caller were to
|
|
// try to infer it.
|
|
InferrableEffectToFirstViolation.maybeInsert(Violation(
|
|
effect, ViolationID::DeclDisallowsInference, ViolationSite{},
|
|
CInfo.CDecl->getLocation(), nullptr, ProblemCalleeEffect));
|
|
}
|
|
}
|
|
// InferrableEffects is now the set of inferrable effects which are not
|
|
// prohibited.
|
|
EffectsToInfer = FunctionEffectKindSet::difference(
|
|
InferrableEffects, DeclaredVerifiableEffects);
|
|
}
|
|
|
|
// Hide the way that Violations for explicitly required effects vs. inferred
|
|
// ones are handled differently.
|
|
void checkAddViolation(bool Inferring, const Violation &NewViol) {
|
|
if (!Inferring)
|
|
ViolationsForExplicitEffects.push_back(NewViol);
|
|
else
|
|
InferrableEffectToFirstViolation.maybeInsert(NewViol);
|
|
}
|
|
|
|
void addUnverifiedDirectCall(const Decl *D, SourceLocation CallLoc,
|
|
ViolationSite VSite) {
|
|
UnverifiedDirectCalls.emplace_back(D, CallLoc, VSite);
|
|
}
|
|
|
|
// Analysis is complete when there are no unverified direct calls.
|
|
bool isComplete() const { return UnverifiedDirectCalls.empty(); }
|
|
|
|
const Violation *violationForInferrableEffect(FunctionEffect effect) {
|
|
return InferrableEffectToFirstViolation.lookup(effect);
|
|
}
|
|
|
|
// Mutable because caller may need to set a DirectCall's Recursing flag.
|
|
MutableArrayRef<DirectCall> unverifiedCalls() {
|
|
assert(!isComplete());
|
|
return UnverifiedDirectCalls;
|
|
}
|
|
|
|
ArrayRef<Violation> getSortedViolationsForExplicitEffects(SourceManager &SM) {
|
|
if (!ViolationsForExplicitEffects.empty())
|
|
llvm::sort(ViolationsForExplicitEffects,
|
|
[&SM](const Violation &LHS, const Violation &RHS) {
|
|
return SM.isBeforeInTranslationUnit(LHS.Loc, RHS.Loc);
|
|
});
|
|
return ViolationsForExplicitEffects;
|
|
}
|
|
|
|
void dump(Sema &SemaRef, llvm::raw_ostream &OS) const {
|
|
OS << "Pending: Declared ";
|
|
DeclaredVerifiableEffects.dump(OS);
|
|
OS << ", " << ViolationsForExplicitEffects.size() << " violations; ";
|
|
OS << " Infer ";
|
|
EffectsToInfer.dump(OS);
|
|
OS << ", " << InferrableEffectToFirstViolation.size() << " violations";
|
|
if (!UnverifiedDirectCalls.empty()) {
|
|
OS << "; Calls: ";
|
|
for (const DirectCall &Call : UnverifiedDirectCalls) {
|
|
CallableInfo CI(*Call.Callee);
|
|
OS << " " << CI.getNameForDiagnostic(SemaRef);
|
|
}
|
|
}
|
|
OS << "\n";
|
|
}
|
|
};
|
|
|
|
// ----------
|
|
class CompleteFunctionAnalysis {
|
|
// Current size: 2 pointers
|
|
public:
|
|
// Has effects which are both the declared ones -- not to be inferred -- plus
|
|
// ones which have been successfully inferred. These are all considered
|
|
// "verified" for the purposes of callers; any issue with verifying declared
|
|
// effects has already been reported and is not the problem of any caller.
|
|
FunctionEffectKindSet VerifiedEffects;
|
|
|
|
private:
|
|
// This is used to generate notes about failed inference.
|
|
EffectToViolationMap InferrableEffectToFirstViolation;
|
|
|
|
public:
|
|
// The incoming Pending analysis is consumed (member(s) are moved-from).
|
|
CompleteFunctionAnalysis(ASTContext &Ctx, PendingFunctionAnalysis &&Pending,
|
|
FunctionEffectKindSet DeclaredEffects,
|
|
FunctionEffectKindSet AllInferrableEffectsToVerify)
|
|
: VerifiedEffects(DeclaredEffects) {
|
|
for (FunctionEffect effect : AllInferrableEffectsToVerify)
|
|
if (Pending.violationForInferrableEffect(effect) == nullptr)
|
|
VerifiedEffects.insert(effect);
|
|
|
|
InferrableEffectToFirstViolation =
|
|
std::move(Pending.InferrableEffectToFirstViolation);
|
|
}
|
|
|
|
const Violation *firstViolationForEffect(FunctionEffect Effect) {
|
|
return InferrableEffectToFirstViolation.lookup(Effect);
|
|
}
|
|
|
|
void dump(llvm::raw_ostream &OS) const {
|
|
OS << "Complete: Verified ";
|
|
VerifiedEffects.dump(OS);
|
|
OS << "; Infer ";
|
|
OS << InferrableEffectToFirstViolation.size() << " violations\n";
|
|
}
|
|
};
|
|
|
|
// ==========
|
|
class Analyzer {
|
|
Sema &S;
|
|
|
|
// Subset of Sema.AllEffectsToVerify
|
|
FunctionEffectKindSet AllInferrableEffectsToVerify;
|
|
|
|
using FuncAnalysisPtr =
|
|
llvm::PointerUnion<PendingFunctionAnalysis *, CompleteFunctionAnalysis *>;
|
|
|
|
// Map all Decls analyzed to FuncAnalysisPtr. Pending state is larger
|
|
// than complete state, so use different objects to represent them.
|
|
// The state pointers are owned by the container.
|
|
class AnalysisMap : llvm::DenseMap<const Decl *, FuncAnalysisPtr> {
|
|
using Base = llvm::DenseMap<const Decl *, FuncAnalysisPtr>;
|
|
|
|
public:
|
|
~AnalysisMap();
|
|
|
|
// Use non-public inheritance in order to maintain the invariant
|
|
// that lookups and insertions are via the canonical Decls.
|
|
|
|
FuncAnalysisPtr lookup(const Decl *Key) const {
|
|
return Base::lookup(Key->getCanonicalDecl());
|
|
}
|
|
|
|
FuncAnalysisPtr &operator[](const Decl *Key) {
|
|
return Base::operator[](Key->getCanonicalDecl());
|
|
}
|
|
|
|
/// Shortcut for the case where we only care about completed analysis.
|
|
CompleteFunctionAnalysis *completedAnalysisForDecl(const Decl *D) const {
|
|
if (FuncAnalysisPtr AP = lookup(D);
|
|
isa_and_nonnull<CompleteFunctionAnalysis *>(AP))
|
|
return cast<CompleteFunctionAnalysis *>(AP);
|
|
return nullptr;
|
|
}
|
|
|
|
void dump(Sema &SemaRef, llvm::raw_ostream &OS) {
|
|
OS << "\nAnalysisMap:\n";
|
|
for (const auto &item : *this) {
|
|
CallableInfo CI(*item.first);
|
|
const auto AP = item.second;
|
|
OS << item.first << " " << CI.getNameForDiagnostic(SemaRef) << " : ";
|
|
if (AP.isNull()) {
|
|
OS << "null\n";
|
|
} else if (auto *CFA = dyn_cast<CompleteFunctionAnalysis *>(AP)) {
|
|
OS << CFA << " ";
|
|
CFA->dump(OS);
|
|
} else if (auto *PFA = dyn_cast<PendingFunctionAnalysis *>(AP)) {
|
|
OS << PFA << " ";
|
|
PFA->dump(SemaRef, OS);
|
|
} else
|
|
llvm_unreachable("never");
|
|
}
|
|
OS << "---\n";
|
|
}
|
|
};
|
|
AnalysisMap DeclAnalysis;
|
|
|
|
public:
|
|
Analyzer(Sema &S) : S(S) {}
|
|
|
|
void run(const TranslationUnitDecl &TU) {
|
|
// Gather all of the effects to be verified to see what operations need to
|
|
// be checked, and to see which ones are inferrable.
|
|
for (FunctionEffect Effect : S.AllEffectsToVerify) {
|
|
const FunctionEffect::Flags Flags = Effect.flags();
|
|
if (Flags & FunctionEffect::FE_InferrableOnCallees)
|
|
AllInferrableEffectsToVerify.insert(Effect);
|
|
}
|
|
LLVM_DEBUG(llvm::dbgs() << "AllInferrableEffectsToVerify: ";
|
|
AllInferrableEffectsToVerify.dump(llvm::dbgs());
|
|
llvm::dbgs() << "\n";);
|
|
|
|
// We can use DeclsWithEffectsToVerify as a stack for a
|
|
// depth-first traversal; there's no need for a second container. But first,
|
|
// reverse it, so when working from the end, Decls are verified in the order
|
|
// they are declared.
|
|
SmallVector<const Decl *> &VerificationQueue = S.DeclsWithEffectsToVerify;
|
|
std::reverse(VerificationQueue.begin(), VerificationQueue.end());
|
|
|
|
while (!VerificationQueue.empty()) {
|
|
const Decl *D = VerificationQueue.back();
|
|
if (FuncAnalysisPtr AP = DeclAnalysis.lookup(D)) {
|
|
if (auto *Pending = dyn_cast<PendingFunctionAnalysis *>(AP)) {
|
|
// All children have been traversed; finish analysis.
|
|
finishPendingAnalysis(D, Pending);
|
|
}
|
|
VerificationQueue.pop_back();
|
|
continue;
|
|
}
|
|
|
|
// Not previously visited; begin a new analysis for this Decl.
|
|
PendingFunctionAnalysis *Pending = verifyDecl(D);
|
|
if (Pending == nullptr) {
|
|
// Completed now.
|
|
VerificationQueue.pop_back();
|
|
continue;
|
|
}
|
|
|
|
// Analysis remains pending because there are direct callees to be
|
|
// verified first. Push them onto the queue.
|
|
for (PendingFunctionAnalysis::DirectCall &Call :
|
|
Pending->unverifiedCalls()) {
|
|
FuncAnalysisPtr AP = DeclAnalysis.lookup(Call.Callee);
|
|
if (AP.isNull()) {
|
|
VerificationQueue.push_back(Call.Callee);
|
|
continue;
|
|
}
|
|
|
|
// This indicates recursion (not necessarily direct). For the
|
|
// purposes of effect analysis, we can just ignore it since
|
|
// no effects forbid recursion.
|
|
assert(isa<PendingFunctionAnalysis *>(AP));
|
|
Call.Recursed = true;
|
|
}
|
|
}
|
|
}
|
|
|
|
private:
|
|
// Verify a single Decl. Return the pending structure if that was the result,
|
|
// else null. This method must not recurse.
|
|
PendingFunctionAnalysis *verifyDecl(const Decl *D) {
|
|
CallableInfo CInfo(*D);
|
|
bool isExternC = false;
|
|
|
|
if (const FunctionDecl *FD = dyn_cast<FunctionDecl>(D))
|
|
isExternC = FD->getCanonicalDecl()->isExternCContext();
|
|
|
|
// For C++, with non-extern "C" linkage only - if any of the Decl's declared
|
|
// effects forbid throwing (e.g. nonblocking) then the function should also
|
|
// be declared noexcept.
|
|
if (S.getLangOpts().CPlusPlus && !isExternC) {
|
|
for (FunctionEffect Effect : CInfo.Effects) {
|
|
if (!(Effect.flags() & FunctionEffect::FE_ExcludeThrow))
|
|
continue;
|
|
|
|
bool IsNoexcept = false;
|
|
if (auto *FD = D->getAsFunction()) {
|
|
IsNoexcept = isNoexcept(FD);
|
|
} else if (auto *BD = dyn_cast<BlockDecl>(D)) {
|
|
if (auto *TSI = BD->getSignatureAsWritten()) {
|
|
auto *FPT = TSI->getType()->castAs<FunctionProtoType>();
|
|
IsNoexcept = FPT->isNothrow() || BD->hasAttr<NoThrowAttr>();
|
|
}
|
|
}
|
|
if (!IsNoexcept)
|
|
S.Diag(D->getBeginLoc(), diag::warn_perf_constraint_implies_noexcept)
|
|
<< GetCallableDeclKind(D, nullptr) << Effect.name();
|
|
break;
|
|
}
|
|
}
|
|
|
|
// Build a PendingFunctionAnalysis on the stack. If it turns out to be
|
|
// complete, we'll have avoided a heap allocation; if it's incomplete, it's
|
|
// a fairly trivial move to a heap-allocated object.
|
|
PendingFunctionAnalysis FAnalysis(S, CInfo, AllInferrableEffectsToVerify);
|
|
|
|
LLVM_DEBUG(llvm::dbgs()
|
|
<< "\nVerifying " << CInfo.getNameForDiagnostic(S) << " ";
|
|
FAnalysis.dump(S, llvm::dbgs()););
|
|
|
|
FunctionBodyASTVisitor Visitor(*this, FAnalysis, CInfo);
|
|
|
|
Visitor.run();
|
|
if (FAnalysis.isComplete()) {
|
|
completeAnalysis(CInfo, std::move(FAnalysis));
|
|
return nullptr;
|
|
}
|
|
// Move the pending analysis to the heap and save it in the map.
|
|
PendingFunctionAnalysis *PendingPtr =
|
|
new PendingFunctionAnalysis(std::move(FAnalysis));
|
|
DeclAnalysis[D] = PendingPtr;
|
|
LLVM_DEBUG(llvm::dbgs() << "inserted pending " << PendingPtr << "\n";
|
|
DeclAnalysis.dump(S, llvm::dbgs()););
|
|
return PendingPtr;
|
|
}
|
|
|
|
// Consume PendingFunctionAnalysis, create with it a CompleteFunctionAnalysis,
|
|
// inserted in the container.
|
|
void completeAnalysis(const CallableInfo &CInfo,
|
|
PendingFunctionAnalysis &&Pending) {
|
|
if (ArrayRef<Violation> Viols =
|
|
Pending.getSortedViolationsForExplicitEffects(S.getSourceManager());
|
|
!Viols.empty())
|
|
emitDiagnostics(Viols, CInfo);
|
|
|
|
CompleteFunctionAnalysis *CompletePtr = new CompleteFunctionAnalysis(
|
|
S.getASTContext(), std::move(Pending), CInfo.Effects,
|
|
AllInferrableEffectsToVerify);
|
|
DeclAnalysis[CInfo.CDecl] = CompletePtr;
|
|
LLVM_DEBUG(llvm::dbgs() << "inserted complete " << CompletePtr << "\n";
|
|
DeclAnalysis.dump(S, llvm::dbgs()););
|
|
}
|
|
|
|
// Called after all direct calls requiring inference have been found -- or
|
|
// not. Repeats calls to FunctionBodyASTVisitor::followCall() but without
|
|
// the possibility of inference. Deletes Pending.
|
|
void finishPendingAnalysis(const Decl *D, PendingFunctionAnalysis *Pending) {
|
|
CallableInfo Caller(*D);
|
|
LLVM_DEBUG(llvm::dbgs() << "finishPendingAnalysis for "
|
|
<< Caller.getNameForDiagnostic(S) << " : ";
|
|
Pending->dump(S, llvm::dbgs()); llvm::dbgs() << "\n";);
|
|
for (const PendingFunctionAnalysis::DirectCall &Call :
|
|
Pending->unverifiedCalls()) {
|
|
if (Call.Recursed)
|
|
continue;
|
|
|
|
CallableInfo Callee(*Call.Callee);
|
|
followCall(Caller, *Pending, Callee, Call.CallLoc,
|
|
/*AssertNoFurtherInference=*/true, Call.VSite);
|
|
}
|
|
completeAnalysis(Caller, std::move(*Pending));
|
|
delete Pending;
|
|
}
|
|
|
|
// Here we have a call to a Decl, either explicitly via a CallExpr or some
|
|
// other AST construct. PFA pertains to the caller.
|
|
void followCall(const CallableInfo &Caller, PendingFunctionAnalysis &PFA,
|
|
const CallableInfo &Callee, SourceLocation CallLoc,
|
|
bool AssertNoFurtherInference, ViolationSite VSite) {
|
|
const bool DirectCall = Callee.isCalledDirectly();
|
|
|
|
// Initially, the declared effects; inferred effects will be added.
|
|
FunctionEffectKindSet CalleeEffects = Callee.Effects;
|
|
|
|
bool IsInferencePossible = DirectCall;
|
|
|
|
if (DirectCall)
|
|
if (CompleteFunctionAnalysis *CFA =
|
|
DeclAnalysis.completedAnalysisForDecl(Callee.CDecl)) {
|
|
// Combine declared effects with those which may have been inferred.
|
|
CalleeEffects.insert(CFA->VerifiedEffects);
|
|
IsInferencePossible = false; // We've already traversed it.
|
|
}
|
|
|
|
if (AssertNoFurtherInference) {
|
|
assert(!IsInferencePossible);
|
|
}
|
|
|
|
if (!Callee.isVerifiable())
|
|
IsInferencePossible = false;
|
|
|
|
LLVM_DEBUG(llvm::dbgs()
|
|
<< "followCall from " << Caller.getNameForDiagnostic(S)
|
|
<< " to " << Callee.getNameForDiagnostic(S)
|
|
<< "; verifiable: " << Callee.isVerifiable() << "; callee ";
|
|
CalleeEffects.dump(llvm::dbgs()); llvm::dbgs() << "\n";
|
|
llvm::dbgs() << " callee " << Callee.CDecl << " canonical "
|
|
<< Callee.CDecl->getCanonicalDecl() << "\n";);
|
|
|
|
auto Check1Effect = [&](FunctionEffect Effect, bool Inferring) {
|
|
if (!Effect.shouldDiagnoseFunctionCall(DirectCall, CalleeEffects))
|
|
return;
|
|
|
|
// If inference is not allowed, or the target is indirect (virtual
|
|
// method/function ptr?), generate a Violation now.
|
|
if (!IsInferencePossible ||
|
|
!(Effect.flags() & FunctionEffect::FE_InferrableOnCallees)) {
|
|
if (Callee.FuncType == SpecialFuncType::None)
|
|
PFA.checkAddViolation(Inferring,
|
|
{Effect, ViolationID::CallsDeclWithoutEffect,
|
|
VSite, CallLoc, Callee.CDecl});
|
|
else
|
|
PFA.checkAddViolation(
|
|
Inferring,
|
|
{Effect, ViolationID::AllocatesMemory, VSite, CallLoc});
|
|
} else {
|
|
// Inference is allowed and necessary; defer it.
|
|
PFA.addUnverifiedDirectCall(Callee.CDecl, CallLoc, VSite);
|
|
}
|
|
};
|
|
|
|
for (FunctionEffect Effect : PFA.DeclaredVerifiableEffects)
|
|
Check1Effect(Effect, false);
|
|
|
|
for (FunctionEffect Effect : PFA.EffectsToInfer)
|
|
Check1Effect(Effect, true);
|
|
}
|
|
|
|
// Describe a callable Decl for a diagnostic.
|
|
// (Not an enum class because the value is always converted to an integer for
|
|
// use in a diagnostic.)
|
|
enum CallableDeclKind {
|
|
CDK_Function,
|
|
CDK_Constructor,
|
|
CDK_Destructor,
|
|
CDK_Lambda,
|
|
CDK_Block,
|
|
CDK_MemberInitializer,
|
|
};
|
|
|
|
// Describe a call site or target using an enum mapping to a %select{}
|
|
// in a diagnostic, e.g. warn_func_effect_violation,
|
|
// warn_perf_constraint_implies_noexcept, and others.
|
|
static CallableDeclKind GetCallableDeclKind(const Decl *D,
|
|
const Violation *V) {
|
|
if (V != nullptr &&
|
|
V->Site.kind() == ViolationSite::Kind::MemberInitializer)
|
|
return CDK_MemberInitializer;
|
|
if (isa<BlockDecl>(D))
|
|
return CDK_Block;
|
|
if (auto *Method = dyn_cast<CXXMethodDecl>(D)) {
|
|
if (isa<CXXConstructorDecl>(D))
|
|
return CDK_Constructor;
|
|
if (isa<CXXDestructorDecl>(D))
|
|
return CDK_Destructor;
|
|
const CXXRecordDecl *Rec = Method->getParent();
|
|
if (Rec->isLambda())
|
|
return CDK_Lambda;
|
|
}
|
|
return CDK_Function;
|
|
};
|
|
|
|
// Should only be called when function's analysis is determined to be
|
|
// complete.
|
|
void emitDiagnostics(ArrayRef<Violation> Viols, const CallableInfo &CInfo) {
|
|
if (Viols.empty())
|
|
return;
|
|
|
|
auto MaybeAddTemplateNote = [&](const Decl *D) {
|
|
if (const FunctionDecl *FD = dyn_cast<FunctionDecl>(D)) {
|
|
while (FD != nullptr && FD->isTemplateInstantiation() &&
|
|
FD->getPointOfInstantiation().isValid()) {
|
|
S.Diag(FD->getPointOfInstantiation(),
|
|
diag::note_func_effect_from_template);
|
|
FD = FD->getTemplateInstantiationPattern();
|
|
}
|
|
}
|
|
};
|
|
|
|
// For note_func_effect_call_indirect.
|
|
enum { Indirect_VirtualMethod, Indirect_FunctionPtr };
|
|
|
|
auto MaybeAddSiteContext = [&](const Decl *D, const Violation &V) {
|
|
// If a violation site is a member initializer, add a note pointing to
|
|
// the constructor which invoked it.
|
|
if (V.Site.kind() == ViolationSite::Kind::MemberInitializer) {
|
|
unsigned ImplicitCtor = 0;
|
|
if (auto *Ctor = dyn_cast<CXXConstructorDecl>(D);
|
|
Ctor && Ctor->isImplicit())
|
|
ImplicitCtor = 1;
|
|
S.Diag(D->getLocation(), diag::note_func_effect_in_constructor)
|
|
<< ImplicitCtor;
|
|
}
|
|
|
|
// If a violation site is a default argument expression, add a note
|
|
// pointing to the call site using the default argument.
|
|
else if (V.Site.kind() == ViolationSite::Kind::DefaultArgExpr)
|
|
S.Diag(V.Site.defaultArgExpr()->getUsedLocation(),
|
|
diag::note_in_evaluating_default_argument);
|
|
};
|
|
|
|
// Top-level violations are warnings.
|
|
for (const Violation &Viol1 : Viols) {
|
|
StringRef effectName = Viol1.Effect.name();
|
|
switch (Viol1.ID) {
|
|
case ViolationID::None:
|
|
case ViolationID::DeclDisallowsInference: // Shouldn't happen
|
|
// here.
|
|
llvm_unreachable("Unexpected violation kind");
|
|
break;
|
|
case ViolationID::AllocatesMemory:
|
|
case ViolationID::ThrowsOrCatchesExceptions:
|
|
case ViolationID::HasStaticLocalVariable:
|
|
case ViolationID::AccessesThreadLocalVariable:
|
|
case ViolationID::AccessesObjCMethodOrProperty:
|
|
S.Diag(Viol1.Loc, diag::warn_func_effect_violation)
|
|
<< GetCallableDeclKind(CInfo.CDecl, &Viol1) << effectName
|
|
<< Viol1.diagnosticSelectIndex();
|
|
MaybeAddSiteContext(CInfo.CDecl, Viol1);
|
|
MaybeAddTemplateNote(CInfo.CDecl);
|
|
break;
|
|
case ViolationID::CallsExprWithoutEffect:
|
|
S.Diag(Viol1.Loc, diag::warn_func_effect_calls_expr_without_effect)
|
|
<< GetCallableDeclKind(CInfo.CDecl, &Viol1) << effectName;
|
|
MaybeAddSiteContext(CInfo.CDecl, Viol1);
|
|
MaybeAddTemplateNote(CInfo.CDecl);
|
|
break;
|
|
|
|
case ViolationID::CallsDeclWithoutEffect: {
|
|
CallableInfo CalleeInfo(*Viol1.Callee);
|
|
std::string CalleeName = CalleeInfo.getNameForDiagnostic(S);
|
|
|
|
S.Diag(Viol1.Loc, diag::warn_func_effect_calls_func_without_effect)
|
|
<< GetCallableDeclKind(CInfo.CDecl, &Viol1) << effectName
|
|
<< GetCallableDeclKind(CalleeInfo.CDecl, nullptr) << CalleeName;
|
|
MaybeAddSiteContext(CInfo.CDecl, Viol1);
|
|
MaybeAddTemplateNote(CInfo.CDecl);
|
|
|
|
// Emit notes explaining the transitive chain of inferences: Why isn't
|
|
// the callee safe?
|
|
for (const Decl *Callee = Viol1.Callee; Callee != nullptr;) {
|
|
std::optional<CallableInfo> MaybeNextCallee;
|
|
CompleteFunctionAnalysis *Completed =
|
|
DeclAnalysis.completedAnalysisForDecl(CalleeInfo.CDecl);
|
|
if (Completed == nullptr) {
|
|
// No result - could be
|
|
// - non-inline and extern
|
|
// - indirect (virtual or through function pointer)
|
|
// - effect has been explicitly disclaimed (e.g. "blocking")
|
|
|
|
CallableType CType = CalleeInfo.type();
|
|
if (CType == CallableType::Virtual)
|
|
S.Diag(Callee->getLocation(),
|
|
diag::note_func_effect_call_indirect)
|
|
<< Indirect_VirtualMethod << effectName;
|
|
else if (CType == CallableType::Unknown)
|
|
S.Diag(Callee->getLocation(),
|
|
diag::note_func_effect_call_indirect)
|
|
<< Indirect_FunctionPtr << effectName;
|
|
else if (CalleeInfo.Effects.contains(Viol1.Effect.oppositeKind()))
|
|
S.Diag(Callee->getLocation(),
|
|
diag::note_func_effect_call_disallows_inference)
|
|
<< GetCallableDeclKind(CInfo.CDecl, nullptr) << effectName
|
|
<< FunctionEffect(Viol1.Effect.oppositeKind()).name();
|
|
else if (const FunctionDecl *FD = dyn_cast<FunctionDecl>(Callee);
|
|
FD == nullptr || FD->getBuiltinID() == 0) {
|
|
// A builtin callee generally doesn't have a useful source
|
|
// location at which to insert a note.
|
|
S.Diag(Callee->getLocation(), diag::note_func_effect_call_extern)
|
|
<< effectName;
|
|
}
|
|
break;
|
|
}
|
|
const Violation *PtrViol2 =
|
|
Completed->firstViolationForEffect(Viol1.Effect);
|
|
if (PtrViol2 == nullptr)
|
|
break;
|
|
|
|
const Violation &Viol2 = *PtrViol2;
|
|
switch (Viol2.ID) {
|
|
case ViolationID::None:
|
|
llvm_unreachable("Unexpected violation kind");
|
|
break;
|
|
case ViolationID::DeclDisallowsInference:
|
|
S.Diag(Viol2.Loc, diag::note_func_effect_call_disallows_inference)
|
|
<< GetCallableDeclKind(CalleeInfo.CDecl, nullptr) << effectName
|
|
<< Viol2.CalleeEffectPreventingInference->name();
|
|
break;
|
|
case ViolationID::CallsExprWithoutEffect:
|
|
S.Diag(Viol2.Loc, diag::note_func_effect_call_indirect)
|
|
<< Indirect_FunctionPtr << effectName;
|
|
break;
|
|
case ViolationID::AllocatesMemory:
|
|
case ViolationID::ThrowsOrCatchesExceptions:
|
|
case ViolationID::HasStaticLocalVariable:
|
|
case ViolationID::AccessesThreadLocalVariable:
|
|
case ViolationID::AccessesObjCMethodOrProperty:
|
|
S.Diag(Viol2.Loc, diag::note_func_effect_violation)
|
|
<< GetCallableDeclKind(CalleeInfo.CDecl, &Viol2) << effectName
|
|
<< Viol2.diagnosticSelectIndex();
|
|
MaybeAddSiteContext(CalleeInfo.CDecl, Viol2);
|
|
break;
|
|
case ViolationID::CallsDeclWithoutEffect:
|
|
MaybeNextCallee.emplace(*Viol2.Callee);
|
|
S.Diag(Viol2.Loc, diag::note_func_effect_calls_func_without_effect)
|
|
<< GetCallableDeclKind(CalleeInfo.CDecl, &Viol2) << effectName
|
|
<< GetCallableDeclKind(Viol2.Callee, nullptr)
|
|
<< MaybeNextCallee->getNameForDiagnostic(S);
|
|
break;
|
|
}
|
|
MaybeAddTemplateNote(Callee);
|
|
Callee = Viol2.Callee;
|
|
if (MaybeNextCallee) {
|
|
CalleeInfo = *MaybeNextCallee;
|
|
CalleeName = CalleeInfo.getNameForDiagnostic(S);
|
|
}
|
|
}
|
|
} break;
|
|
}
|
|
}
|
|
}
|
|
|
|
// ----------
|
|
// This AST visitor is used to traverse the body of a function during effect
|
|
// verification. This happens in 2 situations:
|
|
// [1] The function has declared effects which need to be validated.
|
|
// [2] The function has not explicitly declared an effect in question, and is
|
|
// being checked for implicit conformance.
|
|
//
|
|
// Violations are always routed to a PendingFunctionAnalysis.
|
|
struct FunctionBodyASTVisitor : DynamicRecursiveASTVisitor {
|
|
Analyzer &Outer;
|
|
PendingFunctionAnalysis &CurrentFunction;
|
|
CallableInfo &CurrentCaller;
|
|
ViolationSite VSite;
|
|
const Expr *TrailingRequiresClause = nullptr;
|
|
const Expr *NoexceptExpr = nullptr;
|
|
|
|
FunctionBodyASTVisitor(Analyzer &Outer,
|
|
PendingFunctionAnalysis &CurrentFunction,
|
|
CallableInfo &CurrentCaller)
|
|
: Outer(Outer), CurrentFunction(CurrentFunction),
|
|
CurrentCaller(CurrentCaller) {
|
|
ShouldVisitImplicitCode = true;
|
|
ShouldWalkTypesOfTypeLocs = false;
|
|
}
|
|
|
|
// -- Entry point --
|
|
void run() {
|
|
// The target function may have implicit code paths beyond the
|
|
// body: member and base destructors. Visit these first.
|
|
if (auto *Dtor = dyn_cast<CXXDestructorDecl>(CurrentCaller.CDecl))
|
|
followDestructor(dyn_cast<CXXRecordDecl>(Dtor->getParent()), Dtor);
|
|
|
|
if (auto *FD = dyn_cast<FunctionDecl>(CurrentCaller.CDecl)) {
|
|
TrailingRequiresClause = FD->getTrailingRequiresClause().ConstraintExpr;
|
|
|
|
// Note that FD->getType->getAs<FunctionProtoType>() can yield a
|
|
// noexcept Expr which has been boiled down to a constant expression.
|
|
// Going through the TypeSourceInfo obtains the actual expression which
|
|
// will be traversed as part of the function -- unless we capture it
|
|
// here and have TraverseStmt skip it.
|
|
if (TypeSourceInfo *TSI = FD->getTypeSourceInfo()) {
|
|
if (FunctionProtoTypeLoc TL =
|
|
TSI->getTypeLoc().getAs<FunctionProtoTypeLoc>())
|
|
if (const FunctionProtoType *FPT = TL.getTypePtr())
|
|
NoexceptExpr = FPT->getNoexceptExpr();
|
|
}
|
|
}
|
|
|
|
// Do an AST traversal of the function/block body
|
|
TraverseDecl(const_cast<Decl *>(CurrentCaller.CDecl));
|
|
}
|
|
|
|
// -- Methods implementing common logic --
|
|
|
|
// Handle a language construct forbidden by some effects. Only effects whose
|
|
// flags include the specified flag receive a violation. \p Flag describes
|
|
// the construct.
|
|
void diagnoseLanguageConstruct(FunctionEffect::FlagBit Flag,
|
|
ViolationID VID, SourceLocation Loc,
|
|
const Decl *Callee = nullptr) {
|
|
// If there are any declared verifiable effects which forbid the construct
|
|
// represented by the flag, store just one violation.
|
|
for (FunctionEffect Effect : CurrentFunction.DeclaredVerifiableEffects) {
|
|
if (Effect.flags() & Flag) {
|
|
addViolation(/*inferring=*/false, Effect, VID, Loc, Callee);
|
|
break;
|
|
}
|
|
}
|
|
// For each inferred effect which forbids the construct, store a
|
|
// violation, if we don't already have a violation for that effect.
|
|
for (FunctionEffect Effect : CurrentFunction.EffectsToInfer)
|
|
if (Effect.flags() & Flag)
|
|
addViolation(/*inferring=*/true, Effect, VID, Loc, Callee);
|
|
}
|
|
|
|
void addViolation(bool Inferring, FunctionEffect Effect, ViolationID VID,
|
|
SourceLocation Loc, const Decl *Callee = nullptr) {
|
|
CurrentFunction.checkAddViolation(
|
|
Inferring, Violation(Effect, VID, VSite, Loc, Callee));
|
|
}
|
|
|
|
// Here we have a call to a Decl, either explicitly via a CallExpr or some
|
|
// other AST construct. CallableInfo pertains to the callee.
|
|
void followCall(CallableInfo &CI, SourceLocation CallLoc) {
|
|
// Check for a call to a builtin function, whose effects are
|
|
// handled specially.
|
|
if (const auto *FD = dyn_cast<FunctionDecl>(CI.CDecl)) {
|
|
if (unsigned BuiltinID = FD->getBuiltinID()) {
|
|
CI.Effects = getBuiltinFunctionEffects(BuiltinID);
|
|
if (CI.Effects.empty()) {
|
|
// A builtin with no known effects is assumed safe.
|
|
return;
|
|
}
|
|
// A builtin WITH effects doesn't get any special treatment for
|
|
// being noreturn/noexcept, e.g. longjmp(), so we skip the check
|
|
// below.
|
|
} else {
|
|
// If the callee is both `noreturn` and `noexcept`, it presumably
|
|
// terminates. Ignore it for the purposes of effect analysis.
|
|
// If not C++, `noreturn` alone is sufficient.
|
|
if (FD->isNoReturn() &&
|
|
(!Outer.S.getLangOpts().CPlusPlus || isNoexcept(FD)))
|
|
return;
|
|
}
|
|
}
|
|
|
|
Outer.followCall(CurrentCaller, CurrentFunction, CI, CallLoc,
|
|
/*AssertNoFurtherInference=*/false, VSite);
|
|
}
|
|
|
|
void checkIndirectCall(CallExpr *Call, QualType CalleeType) {
|
|
FunctionEffectKindSet CalleeEffects;
|
|
if (FunctionEffectsRef Effects = FunctionEffectsRef::get(CalleeType);
|
|
!Effects.empty())
|
|
CalleeEffects.insert(Effects);
|
|
|
|
auto Check1Effect = [&](FunctionEffect Effect, bool Inferring) {
|
|
if (Effect.shouldDiagnoseFunctionCall(
|
|
/*direct=*/false, CalleeEffects))
|
|
addViolation(Inferring, Effect, ViolationID::CallsExprWithoutEffect,
|
|
Call->getBeginLoc());
|
|
};
|
|
|
|
for (FunctionEffect Effect : CurrentFunction.DeclaredVerifiableEffects)
|
|
Check1Effect(Effect, false);
|
|
|
|
for (FunctionEffect Effect : CurrentFunction.EffectsToInfer)
|
|
Check1Effect(Effect, true);
|
|
}
|
|
|
|
// This destructor's body should be followed by the caller, but here we
|
|
// follow the field and base destructors.
|
|
void followDestructor(const CXXRecordDecl *Rec,
|
|
const CXXDestructorDecl *Dtor) {
|
|
SourceLocation DtorLoc = Dtor->getLocation();
|
|
for (const FieldDecl *Field : Rec->fields())
|
|
followTypeDtor(Field->getType(), DtorLoc);
|
|
|
|
if (const auto *Class = dyn_cast<CXXRecordDecl>(Rec))
|
|
for (const CXXBaseSpecifier &Base : Class->bases())
|
|
followTypeDtor(Base.getType(), DtorLoc);
|
|
}
|
|
|
|
void followTypeDtor(QualType QT, SourceLocation CallSite) {
|
|
const Type *Ty = QT.getTypePtr();
|
|
while (Ty->isArrayType()) {
|
|
const ArrayType *Arr = Ty->getAsArrayTypeUnsafe();
|
|
QT = Arr->getElementType();
|
|
Ty = QT.getTypePtr();
|
|
}
|
|
|
|
if (Ty->isRecordType()) {
|
|
if (const CXXRecordDecl *Class = Ty->getAsCXXRecordDecl()) {
|
|
if (CXXDestructorDecl *Dtor = Class->getDestructor();
|
|
Dtor && !Dtor->isDeleted()) {
|
|
CallableInfo CI(*Dtor);
|
|
followCall(CI, CallSite);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// -- Methods for use of RecursiveASTVisitor --
|
|
|
|
bool VisitCXXThrowExpr(CXXThrowExpr *Throw) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeThrow,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Throw->getThrowLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitCXXCatchStmt(CXXCatchStmt *Catch) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeCatch,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Catch->getCatchLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCAtThrowStmt(ObjCAtThrowStmt *Throw) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeThrow,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Throw->getThrowLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCAtCatchStmt(ObjCAtCatchStmt *Catch) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeCatch,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Catch->getAtCatchLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCAtFinallyStmt(ObjCAtFinallyStmt *Finally) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeCatch,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Finally->getAtFinallyLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCMessageExpr(ObjCMessageExpr *Msg) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeObjCMessageSend,
|
|
ViolationID::AccessesObjCMethodOrProperty,
|
|
Msg->getBeginLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCAutoreleasePoolStmt(ObjCAutoreleasePoolStmt *ARP) override {
|
|
// Under the hood, @autorelease (potentially?) allocates memory and
|
|
// invokes ObjC methods. We don't currently have memory allocation as
|
|
// a "language construct" but we do have ObjC messaging, so diagnose that.
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeObjCMessageSend,
|
|
ViolationID::AccessesObjCMethodOrProperty,
|
|
ARP->getBeginLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitObjCAtSynchronizedStmt(ObjCAtSynchronizedStmt *Sync) override {
|
|
// Under the hood, this calls objc_sync_enter and objc_sync_exit, wrapped
|
|
// in a @try/@finally block. Diagnose this generically as "ObjC
|
|
// messaging".
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeObjCMessageSend,
|
|
ViolationID::AccessesObjCMethodOrProperty,
|
|
Sync->getBeginLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitSEHExceptStmt(SEHExceptStmt *Exc) override {
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeCatch,
|
|
ViolationID::ThrowsOrCatchesExceptions,
|
|
Exc->getExceptLoc());
|
|
return true;
|
|
}
|
|
|
|
bool VisitCallExpr(CallExpr *Call) override {
|
|
LLVM_DEBUG(llvm::dbgs()
|
|
<< "VisitCallExpr : "
|
|
<< Call->getBeginLoc().printToString(Outer.S.SourceMgr)
|
|
<< "\n";);
|
|
|
|
Expr *CalleeExpr = Call->getCallee();
|
|
if (const Decl *Callee = CalleeExpr->getReferencedDeclOfCallee()) {
|
|
CallableInfo CI(*Callee);
|
|
followCall(CI, Call->getBeginLoc());
|
|
return true;
|
|
}
|
|
|
|
if (isa<CXXPseudoDestructorExpr>(CalleeExpr)) {
|
|
// Just destroying a scalar, fine.
|
|
return true;
|
|
}
|
|
|
|
// No Decl, just an Expr. Just check based on its type.
|
|
checkIndirectCall(Call, CalleeExpr->getType());
|
|
|
|
return true;
|
|
}
|
|
|
|
bool VisitVarDecl(VarDecl *Var) override {
|
|
LLVM_DEBUG(llvm::dbgs()
|
|
<< "VisitVarDecl : "
|
|
<< Var->getBeginLoc().printToString(Outer.S.SourceMgr)
|
|
<< "\n";);
|
|
|
|
if (Var->isStaticLocal())
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeStaticLocalVars,
|
|
ViolationID::HasStaticLocalVariable,
|
|
Var->getLocation());
|
|
|
|
const QualType::DestructionKind DK =
|
|
Var->needsDestruction(Outer.S.getASTContext());
|
|
if (DK == QualType::DK_cxx_destructor)
|
|
followTypeDtor(Var->getType(), Var->getLocation());
|
|
return true;
|
|
}
|
|
|
|
bool VisitCXXNewExpr(CXXNewExpr *New) override {
|
|
// RecursiveASTVisitor does not visit the implicit call to operator new.
|
|
if (FunctionDecl *FD = New->getOperatorNew()) {
|
|
CallableInfo CI(*FD, SpecialFuncType::OperatorNew);
|
|
followCall(CI, New->getBeginLoc());
|
|
}
|
|
|
|
// It's a bit excessive to check operator delete here, since it's
|
|
// just a fallback for operator new followed by a failed constructor.
|
|
// We could check it via New->getOperatorDelete().
|
|
|
|
// It DOES however visit the called constructor
|
|
return true;
|
|
}
|
|
|
|
bool VisitCXXDeleteExpr(CXXDeleteExpr *Delete) override {
|
|
// RecursiveASTVisitor does not visit the implicit call to operator
|
|
// delete.
|
|
if (FunctionDecl *FD = Delete->getOperatorDelete()) {
|
|
CallableInfo CI(*FD, SpecialFuncType::OperatorDelete);
|
|
followCall(CI, Delete->getBeginLoc());
|
|
}
|
|
|
|
// It DOES however visit the called destructor
|
|
|
|
return true;
|
|
}
|
|
|
|
bool VisitCXXConstructExpr(CXXConstructExpr *Construct) override {
|
|
LLVM_DEBUG(llvm::dbgs() << "VisitCXXConstructExpr : "
|
|
<< Construct->getBeginLoc().printToString(
|
|
Outer.S.SourceMgr)
|
|
<< "\n";);
|
|
|
|
// RecursiveASTVisitor does not visit the implicit call to the
|
|
// constructor.
|
|
const CXXConstructorDecl *Ctor = Construct->getConstructor();
|
|
CallableInfo CI(*Ctor);
|
|
followCall(CI, Construct->getLocation());
|
|
|
|
return true;
|
|
}
|
|
|
|
bool TraverseStmt(Stmt *Statement) override {
|
|
// If this statement is a `requires` clause from the top-level function
|
|
// being traversed, ignore it, since it's not generating runtime code.
|
|
// We skip the traversal of lambdas (beyond their captures, see
|
|
// TraverseLambdaExpr below), so just caching this from our constructor
|
|
// should suffice.
|
|
if (Statement != TrailingRequiresClause && Statement != NoexceptExpr)
|
|
return DynamicRecursiveASTVisitor::TraverseStmt(Statement);
|
|
return true;
|
|
}
|
|
|
|
bool TraverseConstructorInitializer(CXXCtorInitializer *Init) override {
|
|
ViolationSite PrevVS = VSite;
|
|
if (Init->isAnyMemberInitializer())
|
|
VSite.setKind(ViolationSite::Kind::MemberInitializer);
|
|
bool Result =
|
|
DynamicRecursiveASTVisitor::TraverseConstructorInitializer(Init);
|
|
VSite = PrevVS;
|
|
return Result;
|
|
}
|
|
|
|
bool TraverseCXXDefaultArgExpr(CXXDefaultArgExpr *E) override {
|
|
LLVM_DEBUG(llvm::dbgs()
|
|
<< "TraverseCXXDefaultArgExpr : "
|
|
<< E->getUsedLocation().printToString(Outer.S.SourceMgr)
|
|
<< "\n";);
|
|
|
|
ViolationSite PrevVS = VSite;
|
|
if (VSite.kind() == ViolationSite::Kind::Default)
|
|
VSite = ViolationSite{E};
|
|
|
|
bool Result = DynamicRecursiveASTVisitor::TraverseCXXDefaultArgExpr(E);
|
|
VSite = PrevVS;
|
|
return Result;
|
|
}
|
|
|
|
bool TraverseLambdaExpr(LambdaExpr *Lambda) override {
|
|
// We override this so as to be able to skip traversal of the lambda's
|
|
// body. We have to explicitly traverse the captures. Why not return
|
|
// false from shouldVisitLambdaBody()? Because we need to visit a lambda's
|
|
// body when we are verifying the lambda itself; we only want to skip it
|
|
// in the context of the outer function.
|
|
for (unsigned I = 0, N = Lambda->capture_size(); I < N; ++I)
|
|
TraverseLambdaCapture(Lambda, Lambda->capture_begin() + I,
|
|
Lambda->capture_init_begin()[I]);
|
|
|
|
return true;
|
|
}
|
|
|
|
bool TraverseBlockExpr(BlockExpr * /*unused*/) override {
|
|
// As with lambdas, don't traverse the block's body.
|
|
// TODO: are the capture expressions (ctor call?) safe?
|
|
return true;
|
|
}
|
|
|
|
bool VisitDeclRefExpr(DeclRefExpr *E) override {
|
|
const ValueDecl *Val = E->getDecl();
|
|
if (const auto *Var = dyn_cast<VarDecl>(Val)) {
|
|
if (Var->getTLSKind() != VarDecl::TLS_None) {
|
|
// At least on macOS, thread-local variables are initialized on
|
|
// first access, including a heap allocation.
|
|
diagnoseLanguageConstruct(FunctionEffect::FE_ExcludeThreadLocalVars,
|
|
ViolationID::AccessesThreadLocalVariable,
|
|
E->getLocation());
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
bool TraverseGenericSelectionExpr(GenericSelectionExpr *Node) override {
|
|
return TraverseStmt(Node->getResultExpr());
|
|
}
|
|
bool
|
|
TraverseUnaryExprOrTypeTraitExpr(UnaryExprOrTypeTraitExpr *Node) override {
|
|
return true;
|
|
}
|
|
|
|
bool TraverseTypeOfExprTypeLoc(TypeOfExprTypeLoc Node,
|
|
bool TraverseQualifier) override {
|
|
return true;
|
|
}
|
|
|
|
bool TraverseDecltypeTypeLoc(DecltypeTypeLoc Node,
|
|
bool TraverseQualifier) override {
|
|
return true;
|
|
}
|
|
|
|
bool TraverseCXXNoexceptExpr(CXXNoexceptExpr *Node) override {
|
|
return true;
|
|
}
|
|
|
|
bool TraverseCXXTypeidExpr(CXXTypeidExpr *Node) override { return true; }
|
|
|
|
// Skip concept requirements since they don't generate code.
|
|
bool TraverseConceptRequirement(concepts::Requirement *R) override {
|
|
return true;
|
|
}
|
|
};
|
|
};
|
|
|
|
Analyzer::AnalysisMap::~AnalysisMap() {
|
|
for (const auto &Item : *this) {
|
|
FuncAnalysisPtr AP = Item.second;
|
|
if (auto *PFA = dyn_cast<PendingFunctionAnalysis *>(AP))
|
|
delete PFA;
|
|
else
|
|
delete cast<CompleteFunctionAnalysis *>(AP);
|
|
}
|
|
}
|
|
|
|
} // anonymous namespace
|
|
|
|
namespace clang {
|
|
|
|
bool Sema::diagnoseConflictingFunctionEffect(
|
|
const FunctionEffectsRef &FX, const FunctionEffectWithCondition &NewEC,
|
|
SourceLocation NewAttrLoc) {
|
|
// If the new effect has a condition, we can't detect conflicts until the
|
|
// condition is resolved.
|
|
if (NewEC.Cond.getCondition() != nullptr)
|
|
return false;
|
|
|
|
// Diagnose the new attribute as incompatible with a previous one.
|
|
auto Incompatible = [&](const FunctionEffectWithCondition &PrevEC) {
|
|
Diag(NewAttrLoc, diag::err_attributes_are_not_compatible)
|
|
<< ("'" + NewEC.description() + "'")
|
|
<< ("'" + PrevEC.description() + "'") << false;
|
|
// We don't necessarily have the location of the previous attribute,
|
|
// so no note.
|
|
return true;
|
|
};
|
|
|
|
// Compare against previous attributes.
|
|
FunctionEffect::Kind NewKind = NewEC.Effect.kind();
|
|
|
|
for (const FunctionEffectWithCondition &PrevEC : FX) {
|
|
// Again, can't check yet when the effect is conditional.
|
|
if (PrevEC.Cond.getCondition() != nullptr)
|
|
continue;
|
|
|
|
FunctionEffect::Kind PrevKind = PrevEC.Effect.kind();
|
|
// Note that we allow PrevKind == NewKind; it's redundant and ignored.
|
|
|
|
if (PrevEC.Effect.oppositeKind() == NewKind)
|
|
return Incompatible(PrevEC);
|
|
|
|
// A new allocating is incompatible with a previous nonblocking.
|
|
if (PrevKind == FunctionEffect::Kind::NonBlocking &&
|
|
NewKind == FunctionEffect::Kind::Allocating)
|
|
return Incompatible(PrevEC);
|
|
|
|
// A new nonblocking is incompatible with a previous allocating.
|
|
if (PrevKind == FunctionEffect::Kind::Allocating &&
|
|
NewKind == FunctionEffect::Kind::NonBlocking)
|
|
return Incompatible(PrevEC);
|
|
}
|
|
|
|
return false;
|
|
}
|
|
|
|
void Sema::diagnoseFunctionEffectMergeConflicts(
|
|
const FunctionEffectSet::Conflicts &Errs, SourceLocation NewLoc,
|
|
SourceLocation OldLoc) {
|
|
for (const FunctionEffectSet::Conflict &Conflict : Errs) {
|
|
Diag(NewLoc, diag::warn_conflicting_func_effects)
|
|
<< Conflict.Kept.description() << Conflict.Rejected.description();
|
|
Diag(OldLoc, diag::note_previous_declaration);
|
|
}
|
|
}
|
|
|
|
// Decl should be a FunctionDecl or BlockDecl.
|
|
void Sema::maybeAddDeclWithEffects(const Decl *D,
|
|
const FunctionEffectsRef &FX) {
|
|
if (!D->hasBody()) {
|
|
if (const auto *FD = D->getAsFunction(); FD && !FD->willHaveBody())
|
|
return;
|
|
}
|
|
|
|
if (Diags.getIgnoreAllWarnings() ||
|
|
(Diags.getSuppressSystemWarnings() &&
|
|
SourceMgr.isInSystemHeader(D->getLocation())))
|
|
return;
|
|
|
|
if (hasUncompilableErrorOccurred())
|
|
return;
|
|
|
|
// For code in dependent contexts, we'll do this at instantiation time.
|
|
// Without this check, we would analyze the function based on placeholder
|
|
// template parameters, and potentially generate spurious diagnostics.
|
|
if (cast<DeclContext>(D)->isDependentContext())
|
|
return;
|
|
|
|
addDeclWithEffects(D, FX);
|
|
}
|
|
|
|
void Sema::addDeclWithEffects(const Decl *D, const FunctionEffectsRef &FX) {
|
|
// To avoid the possibility of conflict, don't add effects which are
|
|
// not FE_InferrableOnCallees and therefore not verified; this removes
|
|
// blocking/allocating but keeps nonblocking/nonallocating.
|
|
// Also, ignore any conditions when building the list of effects.
|
|
bool AnyVerifiable = false;
|
|
for (const FunctionEffectWithCondition &EC : FX)
|
|
if (EC.Effect.flags() & FunctionEffect::FE_InferrableOnCallees) {
|
|
AllEffectsToVerify.insert(EC.Effect);
|
|
AnyVerifiable = true;
|
|
}
|
|
|
|
// Record the declaration for later analysis.
|
|
if (AnyVerifiable)
|
|
DeclsWithEffectsToVerify.push_back(D);
|
|
}
|
|
|
|
void Sema::performFunctionEffectAnalysis(TranslationUnitDecl *TU) {
|
|
if (hasUncompilableErrorOccurred() || Diags.getIgnoreAllWarnings())
|
|
return;
|
|
if (TU == nullptr)
|
|
return;
|
|
Analyzer{*this}.run(*TU);
|
|
}
|
|
|
|
Sema::FunctionEffectDiffVector::FunctionEffectDiffVector(
|
|
const FunctionEffectsRef &Old, const FunctionEffectsRef &New) {
|
|
|
|
FunctionEffectsRef::iterator POld = Old.begin();
|
|
FunctionEffectsRef::iterator OldEnd = Old.end();
|
|
FunctionEffectsRef::iterator PNew = New.begin();
|
|
FunctionEffectsRef::iterator NewEnd = New.end();
|
|
|
|
while (true) {
|
|
int cmp = 0;
|
|
if (POld == OldEnd) {
|
|
if (PNew == NewEnd)
|
|
break;
|
|
cmp = 1;
|
|
} else if (PNew == NewEnd)
|
|
cmp = -1;
|
|
else {
|
|
FunctionEffectWithCondition Old = *POld;
|
|
FunctionEffectWithCondition New = *PNew;
|
|
if (Old.Effect.kind() < New.Effect.kind())
|
|
cmp = -1;
|
|
else if (New.Effect.kind() < Old.Effect.kind())
|
|
cmp = 1;
|
|
else {
|
|
cmp = 0;
|
|
if (Old.Cond.getCondition() != New.Cond.getCondition()) {
|
|
// FIXME: Cases where the expressions are equivalent but
|
|
// don't have the same identity.
|
|
push_back(FunctionEffectDiff{
|
|
Old.Effect.kind(), FunctionEffectDiff::Kind::ConditionMismatch,
|
|
Old, New});
|
|
}
|
|
}
|
|
}
|
|
|
|
if (cmp < 0) {
|
|
// removal
|
|
FunctionEffectWithCondition Old = *POld;
|
|
push_back(FunctionEffectDiff{Old.Effect.kind(),
|
|
FunctionEffectDiff::Kind::Removed, Old,
|
|
std::nullopt});
|
|
++POld;
|
|
} else if (cmp > 0) {
|
|
// addition
|
|
FunctionEffectWithCondition New = *PNew;
|
|
push_back(FunctionEffectDiff{New.Effect.kind(),
|
|
FunctionEffectDiff::Kind::Added,
|
|
std::nullopt, New});
|
|
++PNew;
|
|
} else {
|
|
++POld;
|
|
++PNew;
|
|
}
|
|
}
|
|
}
|
|
|
|
bool Sema::FunctionEffectDiff::shouldDiagnoseConversion(
|
|
QualType SrcType, const FunctionEffectsRef &SrcFX, QualType DstType,
|
|
const FunctionEffectsRef &DstFX) const {
|
|
|
|
switch (EffectKind) {
|
|
case FunctionEffect::Kind::NonAllocating:
|
|
// nonallocating can't be added (spoofed) during a conversion, unless we
|
|
// have nonblocking.
|
|
if (DiffKind == Kind::Added) {
|
|
for (const auto &CFE : SrcFX) {
|
|
if (CFE.Effect.kind() == FunctionEffect::Kind::NonBlocking)
|
|
return false;
|
|
}
|
|
}
|
|
[[fallthrough]];
|
|
case FunctionEffect::Kind::NonBlocking:
|
|
// nonblocking can't be added (spoofed) during a conversion.
|
|
switch (DiffKind) {
|
|
case Kind::Added:
|
|
return true;
|
|
case Kind::Removed:
|
|
return false;
|
|
case Kind::ConditionMismatch:
|
|
// FIXME: Condition mismatches are too coarse right now -- expressions
|
|
// which are equivalent but don't have the same identity are detected as
|
|
// mismatches. We're going to diagnose those anyhow until expression
|
|
// matching is better.
|
|
return true;
|
|
}
|
|
break;
|
|
case FunctionEffect::Kind::Blocking:
|
|
case FunctionEffect::Kind::Allocating:
|
|
return false;
|
|
}
|
|
llvm_unreachable("unknown effect kind");
|
|
}
|
|
|
|
bool Sema::FunctionEffectDiff::shouldDiagnoseRedeclaration(
|
|
const FunctionDecl &OldFunction, const FunctionEffectsRef &OldFX,
|
|
const FunctionDecl &NewFunction, const FunctionEffectsRef &NewFX) const {
|
|
switch (EffectKind) {
|
|
case FunctionEffect::Kind::NonAllocating:
|
|
case FunctionEffect::Kind::NonBlocking:
|
|
// nonblocking/nonallocating can't be removed in a redeclaration.
|
|
switch (DiffKind) {
|
|
case Kind::Added:
|
|
return false; // No diagnostic.
|
|
case Kind::Removed:
|
|
return true; // Issue diagnostic.
|
|
case Kind::ConditionMismatch:
|
|
// All these forms of mismatches are diagnosed.
|
|
return true;
|
|
}
|
|
break;
|
|
case FunctionEffect::Kind::Blocking:
|
|
case FunctionEffect::Kind::Allocating:
|
|
return false;
|
|
}
|
|
llvm_unreachable("unknown effect kind");
|
|
}
|
|
|
|
Sema::FunctionEffectDiff::OverrideResult
|
|
Sema::FunctionEffectDiff::shouldDiagnoseMethodOverride(
|
|
const CXXMethodDecl &OldMethod, const FunctionEffectsRef &OldFX,
|
|
const CXXMethodDecl &NewMethod, const FunctionEffectsRef &NewFX) const {
|
|
switch (EffectKind) {
|
|
case FunctionEffect::Kind::NonAllocating:
|
|
case FunctionEffect::Kind::NonBlocking:
|
|
switch (DiffKind) {
|
|
|
|
// If added on an override, that's fine and not diagnosed.
|
|
case Kind::Added:
|
|
return OverrideResult::NoAction;
|
|
|
|
// If missing from an override (removed), propagate from base to derived.
|
|
case Kind::Removed:
|
|
return OverrideResult::Merge;
|
|
|
|
// If there's a mismatch involving the effect's polarity or condition,
|
|
// issue a warning.
|
|
case Kind::ConditionMismatch:
|
|
return OverrideResult::Warn;
|
|
}
|
|
break;
|
|
case FunctionEffect::Kind::Blocking:
|
|
case FunctionEffect::Kind::Allocating:
|
|
return OverrideResult::NoAction;
|
|
}
|
|
llvm_unreachable("unknown effect kind");
|
|
}
|
|
|
|
} // namespace clang
|