[libc-commits] [libc] [libc][math][c23] Add ffma{, l, f128} and fdiv{, l, f128} C23 math functions #101089 (PR #101253)

[via libc-commits]

================
@@ -414,6 +414,14 @@ def StdC : StandardSpec<"stdc"> {
           FunctionSpec<"fdiml", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>]>,
           GuardedFunctionSpec<"fdimf16", RetValSpec<Float16Type>, [ArgSpec<Float16Type>, ArgSpec<Float16Type>], "LIBC_TYPES_HAS_FLOAT16">,
           GuardedFunctionSpec<"fdimf128", RetValSpec<Float128Type>, [ArgSpec<Float128Type>, ArgSpec<Float128Type>], "LIBC_TYPES_HAS_FLOAT128">,
+          
+          FunctionSpec<"fdiv", RetValSpec<FloatType>, [ArgSpec<DoubleType>, ArgSpec<DoubleType>]>,
+          FunctionSpec<"fdivl", RetValSpec<FloatType>, [ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>]>,
+          GuardedFunctionSpec<"fdivf128", RetValSpec<FloatType>, [ArgSpec<Float128Type>, ArgSpec<Float128Type>], "LIBC_TYPES_HAS_FLOAT128">,
+
+          FunctionSpec<"ffma", RetValSpec<FloatType>, [ArgSpec<DoubleType>, ArgSpec<DoubleType>, ArgSpec<DoubleType>]>,
+          FunctionSpec<"ffmal", RetValSpec<FloatType>, [ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>]>,
+          GuardedFunctionSpec<"ffmaf128", RetValSpec<FloatType>, [ArgSpec<Float128Type>, ArgSpec<Float128Type>, ArgSpec<Float128Type>], "LIBC_TYPES_HAS_FLOAT128">,
----------------
overmighty wrote:

`fdivf128` and `ffmaf128` should be moved to libc/spec/llvm_libc_ext.td.

https://github.com/llvm/llvm-project/pull/101253