diff options
author | Chandler Carruth <chandlerc@gmail.com> | 2024-12-14 09:09:47 +0000 |
---|---|---|
committer | Chandler Carruth <chandlerc@gmail.com> | 2025-02-04 18:04:57 +0000 |
commit | cd269fee05a0f78fb53b65f701b4e06e9ddab424 (patch) | |
tree | 2a248752b2d85d3ae6bd73a7cf5091499f236006 /clang/lib/Sema/SemaChecking.cpp | |
parent | f308af757d72412d0d1429f43d93dedcc87c49f0 (diff) | |
download | llvm-cd269fee05a0f78fb53b65f701b4e06e9ddab424.zip llvm-cd269fee05a0f78fb53b65f701b4e06e9ddab424.tar.gz llvm-cd269fee05a0f78fb53b65f701b4e06e9ddab424.tar.bz2 |
[StrTable] Switch Clang builtins to use string tables
This both reapplies #118734, the initial attempt at this, and updates it
significantly.
First, it uses the newly added `StringTable` abstraction for string
tables, and simplifies the construction to build the string table and
info arrays separately. This should reduce any `constexpr` compile time
memory or CPU cost of the original PR while significantly improving the
APIs throughout.
It also restructures the builtins to support sharding across several
independent tables. This accomplishes two improvements from the
original PR:
1) It improves the APIs used significantly.
2) When builtins are defined from different sources (like SVE vs MVE in
AArch64), this allows each of them to build their own string table
independently rather than having to merge the string tables and info
structures.
3) It allows each shard to factor out a common prefix, often cutting the
size of the strings needed for the builtins by a factor two.
The second point is important both to allow different mechanisms of
construction (for example a `.def` file and a tablegen'ed `.inc` file,
or different tablegen'ed `.inc files), it also simply reduces the sizes
of these tables which is valuable given how large they are in some
cases. The third builds on that size reduction.
Initially, we use this new sharding rather than merging tables in
AArch64, LoongArch, RISCV, and X86. Mostly this helps ensure the system
works, as without further changes these still push scaling limits.
Subsequent commits will more deeply leverage the new structure,
including using the prefix capabilities which cannot be easily factored
out here and requires deep changes to the targets.
Diffstat (limited to 'clang/lib/Sema/SemaChecking.cpp')
-rw-r--r-- | clang/lib/Sema/SemaChecking.cpp | 16 |
1 files changed, 9 insertions, 7 deletions
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp index 61b2c8c..b345e40 100644 --- a/clang/lib/Sema/SemaChecking.cpp +++ b/clang/lib/Sema/SemaChecking.cpp @@ -1236,7 +1236,9 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD, bool IsChkVariant = false; auto GetFunctionName = [&]() { - StringRef FunctionName = getASTContext().BuiltinInfo.getName(BuiltinID); + std::string FunctionNameStr = + getASTContext().BuiltinInfo.getName(BuiltinID); + llvm::StringRef FunctionName = FunctionNameStr; // Skim off the details of whichever builtin was called to produce a better // diagnostic, as it's unlikely that the user wrote the __builtin // explicitly. @@ -1246,7 +1248,7 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD, } else { FunctionName.consume_front("__builtin_"); } - return FunctionName; + return FunctionName.str(); }; switch (BuiltinID) { @@ -1290,7 +1292,7 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD, unsigned SourceSize) { DiagID = diag::warn_fortify_scanf_overflow; unsigned Index = ArgIndex + DataIndex; - StringRef FunctionName = GetFunctionName(); + std::string FunctionName = GetFunctionName(); DiagRuntimeBehavior(TheCall->getArg(Index)->getBeginLoc(), TheCall, PDiag(DiagID) << FunctionName << (Index + 1) << DestSize << SourceSize); @@ -1439,7 +1441,7 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD, llvm::APSInt::compareValues(*SourceSize, *DestinationSize) <= 0) return; - StringRef FunctionName = GetFunctionName(); + std::string FunctionName = GetFunctionName(); SmallString<16> DestinationStr; SmallString<16> SourceStr; @@ -4584,7 +4586,7 @@ ExprResult Sema::BuiltinAtomicOverloaded(ExprResult TheCallResult) { // Get the decl for the concrete builtin from this, we can tell what the // concrete integer type we should convert to is. unsigned NewBuiltinID = BuiltinIndices[BuiltinIndex][SizeIndex]; - StringRef NewBuiltinName = Context.BuiltinInfo.getName(NewBuiltinID); + std::string NewBuiltinName = Context.BuiltinInfo.getName(NewBuiltinID); FunctionDecl *NewBuiltinDecl; if (NewBuiltinID == BuiltinID) NewBuiltinDecl = FDecl; @@ -8379,7 +8381,7 @@ static void emitReplacement(Sema &S, SourceLocation Loc, SourceRange Range, unsigned AbsKind, QualType ArgType) { bool EmitHeaderHint = true; const char *HeaderName = nullptr; - StringRef FunctionName; + std::string FunctionName; if (S.getLangOpts().CPlusPlus && !ArgType->isAnyComplexType()) { FunctionName = "std::abs"; if (ArgType->isIntegralOrEnumerationType()) { @@ -8545,7 +8547,7 @@ void Sema::CheckAbsoluteValueFunction(const CallExpr *Call, // Unsigned types cannot be negative. Suggest removing the absolute value // function call. if (ArgType->isUnsignedIntegerType()) { - StringRef FunctionName = + std::string FunctionName = IsStdAbs ? "std::abs" : Context.BuiltinInfo.getName(AbsKind); Diag(Call->getExprLoc(), diag::warn_unsigned_abs) << ArgType << ParamType; Diag(Call->getExprLoc(), diag::note_remove_abs) |