[clang] Lex: add support for `i128` and `ui128` suffixes (PR #130993)
Saleem Abdulrasool via cfe-commits
cfe-commits at lists.llvm.org
Wed Mar 12 10:08:45 PDT 2025
https://github.com/compnerd created https://github.com/llvm/llvm-project/pull/130993
Microsoft's compiler supports an extension for 128-bit literals. This is referenced in `intsafe.h` which is included transitievly. When building with modules, the literal parsing causes a failure due to the missing support for the extension. To alleviate this issue, support parsing this literal, especially now that there is the BitInt extension.
Take the opportunity to tighten up the code slightly by ensuring that we do not access out-of-bounds characters when lexing the token.
>From 79ad6e94847e187e10a78670cdf79b91ebec4da2 Mon Sep 17 00:00:00 2001
From: Saleem Abdulrasool <compnerd at compnerd.org>
Date: Wed, 12 Mar 2025 09:52:58 -0700
Subject: [PATCH] Lex: add support for `i128` and `ui128` suffixes
Microsoft's compiler supports an extension for 128-bit literals. This is
referenced in `intsafe.h` which is included transitievly. When building
with modules, the literal parsing causes a failure due to the missing
support for the extension. To alleviate this issue, support parsing this
literal, especially now that there is the BitInt extension.
Take the opportunity to tighten up the code slightly by ensuring that we
do not access out-of-bounds characters when lexing the token.
---
clang/include/clang/Lex/LiteralSupport.h | 4 ++--
clang/lib/Lex/LiteralSupport.cpp | 14 ++++++++++----
clang/lib/Sema/SemaExpr.cpp | 16 ++++++++++++----
clang/test/Lexer/ms-extensions.c | 10 +++++++++-
4 files changed, 33 insertions(+), 11 deletions(-)
diff --git a/clang/include/clang/Lex/LiteralSupport.h b/clang/include/clang/Lex/LiteralSupport.h
index 705021fcfa5b1..ea5f63bc20399 100644
--- a/clang/include/clang/Lex/LiteralSupport.h
+++ b/clang/include/clang/Lex/LiteralSupport.h
@@ -82,8 +82,8 @@ class NumericLiteralParser {
bool isAccum : 1; // 1.0hk/k/lk/uhk/uk/ulk
bool isBitInt : 1; // 1wb, 1uwb (C23) or 1__wb, 1__uwb (Clang extension in C++
// mode)
- uint8_t MicrosoftInteger; // Microsoft suffix extension i8, i16, i32, or i64.
-
+ uint8_t MicrosoftInteger; // Microsoft suffix extension i8, i16, i32, i64, or
+ // i128.
bool isFixedPointLiteral() const {
return (saw_period || saw_exponent) && saw_fixed_point_suffix;
diff --git a/clang/lib/Lex/LiteralSupport.cpp b/clang/lib/Lex/LiteralSupport.cpp
index 225a6c2d15baa..218b9e2886f54 100644
--- a/clang/lib/Lex/LiteralSupport.cpp
+++ b/clang/lib/Lex/LiteralSupport.cpp
@@ -1074,29 +1074,35 @@ NumericLiteralParser::NumericLiteralParser(StringRef TokSpelling,
case 'i':
case 'I':
if (LangOpts.MicrosoftExt && !isFPConstant) {
- // Allow i8, i16, i32, and i64. First, look ahead and check if
+ // Allow i8, i16, i32, i64, and i128. First, look ahead and check if
// suffixes are Microsoft integers and not the imaginary unit.
uint8_t Bits = 0;
size_t ToSkip = 0;
+ if (s + 1 >= ThisTokEnd)
+ break;
switch (s[1]) {
case '8': // i8 suffix
Bits = 8;
ToSkip = 2;
break;
case '1':
- if (s[2] == '6') { // i16 suffix
+ if (s + 2 < ThisTokEnd && s[2] == '6') { // i16 suffix
Bits = 16;
ToSkip = 3;
+ } else if (s + 3 < ThisTokEnd && s[2] == '2' &&
+ s[3] == '8') { // i128 suffix
+ Bits = 128;
+ ToSkip = 4;
}
break;
case '3':
- if (s[2] == '2') { // i32 suffix
+ if (s + 2 < ThisTokEnd && s[2] == '2') { // i32 suffix
Bits = 32;
ToSkip = 3;
}
break;
case '6':
- if (s[2] == '4') { // i64 suffix
+ if (s + 2 < ThisTokEnd && s[2] == '4') { // i64 suffix
Bits = 64;
ToSkip = 3;
}
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index a36a2f563739e..86a492dd465a0 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -3924,10 +3924,18 @@ ExprResult Sema::ActOnNumericConstant(const Token &Tok, Scope *UDLScope) {
// to get the integer value from an overly-wide APInt is *extremely*
// expensive, so the naive approach of assuming
// llvm::IntegerType::MAX_INT_BITS is a big performance hit.
- unsigned BitsNeeded =
- Literal.isBitInt ? llvm::APInt::getSufficientBitsNeeded(
- Literal.getLiteralDigits(), Literal.getRadix())
- : Context.getTargetInfo().getIntMaxTWidth();
+ unsigned BitsNeeded = Context.getTargetInfo().getIntMaxTWidth();
+ if (Literal.isBitInt)
+ BitsNeeded = llvm::APInt::getSufficientBitsNeeded(
+ Literal.getLiteralDigits(), Literal.getRadix());
+ if (Literal.MicrosoftInteger) {
+ if (Literal.MicrosoftInteger == 128 &&
+ !Context.getTargetInfo().hasInt128Type())
+ PP.Diag(Tok.getLocation(), diag::err_integer_literal_too_large)
+ << Literal.isUnsigned;
+ BitsNeeded = Literal.MicrosoftInteger;
+ }
+
llvm::APInt ResultVal(BitsNeeded, 0);
if (Literal.GetIntegerValue(ResultVal)) {
diff --git a/clang/test/Lexer/ms-extensions.c b/clang/test/Lexer/ms-extensions.c
index f1eed337b8737..41a5c2c182cb6 100644
--- a/clang/test/Lexer/ms-extensions.c
+++ b/clang/test/Lexer/ms-extensions.c
@@ -13,16 +13,24 @@ __int64 w = 0x43ui64;
__int64 z = 9Li64; // expected-error {{invalid suffix}}
__int64 q = 10lli64; // expected-error {{invalid suffix}}
-__complex double c1 = 1i;
+__complex double c1 = 1i; // expected-error {{invalid suffix}}
__complex double c2 = 1.0i;
__complex float c3 = 1.0if;
+#define UINT128_MAX 0xffffffffffffffffffffffffffffffffui128
#define ULLONG_MAX 0xffffffffffffffffui64
#define UINT 0xffffffffui32
#define USHORT 0xffffui16
#define UCHAR 0xffui8
void a(void) {
+#if __SIZEOF_INT128__
+ __int128 j = UINT128_MAX;
+#else
+ int j = UINT128_MAX;
+ // expected-warning at -1{{implicit conversion from 'unsigned __int128' to 'int' changes value from 340282366920938463463374607431768211455 to -1}}
+ // expected-error at -2{{integer literal is too large to be represented in any integer type}}
+#endif
unsigned long long m = ULLONG_MAX;
unsigned int n = UINT;
unsigned short s = USHORT;
More information about the cfe-commits
mailing list