[clang] Predefined macros for float128 support

[pranavk]

Expose some predefined macros for Float128 support.

[llvmbot]

@llvm/pr-subscribers-clang

Changes

Expose some predefined macros for Float128 support.

---

Full diff: https://github.com/llvm/llvm-project/pull/67196.diff

1 Files Affected:

• (modified) clang/lib/Frontend/InitPreprocessor.cpp (+2)

diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
index e5db8a654e6707d..9e4d4d398a21da5 100644
--- a/clang/lib/Frontend/InitPreprocessor.cpp
+++ b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -1076,6 +1076,8 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
   DefineFloatMacros(Builder, "FLT", &TI.getFloatFormat(), "F");
   DefineFloatMacros(Builder, "DBL", &TI.getDoubleFormat(), "");
   DefineFloatMacros(Builder, "LDBL", &TI.getLongDoubleFormat(), "L");
+  if (TI.hasFloat128Type())
+    DefineFloatMacros(Builder, "FLT128", &TI.getFloat128Format(), "Q");
 
   // Define a __POINTER_WIDTH__ macro for stdint.h.
   Builder.defineMacro("__POINTER_WIDTH__",

[alexander-shaposhnikov]

This needs tests (see e.g. https://reviews.llvm.org/D111382, https://github.com/llvm/llvm-project/blob/main/clang/test/Preprocessor)

[pranavk]

I am not adding any new float type or any literal suffix (unlike other revision you pointed by Ray). This is only adding float macros for which I modified the test that was expecting it.

[echristo]

In general this looks good. Since we don't have full compiler-rt support for this yet we may want to wait until that to merge. Alex is already looking at this and coordinating with his needs is probably the best solution.

[alexander-shaposhnikov]

LG