[libc-commits] [libc] [libc][math][c23] Add f16{add, sub}{, l, f128} C23 math functions (PR #97072)

[via libc-commits]

================
@@ -729,6 +729,16 @@ def StdC : StandardSpec<"stdc"> {
 
           GuardedFunctionSpec<"setpayloadsigf16", RetValSpec<IntType>, [ArgSpec<Float16Ptr>, ArgSpec<Float16Type>], "LIBC_TYPES_HAS_FLOAT16">,
 
+          GuardedFunctionSpec<"f16add", RetValSpec<Float16Type>, [ArgSpec<DoubleType>, ArgSpec<DoubleType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16addf", RetValSpec<Float16Type>, [ArgSpec<FloatType>, ArgSpec<FloatType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16addl", RetValSpec<Float16Type>, [ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16addf128", RetValSpec<Float16Type>, [ArgSpec<Float128Type>, ArgSpec<Float128Type>], "LIBC_TYPES_HAS_FLOAT16_AND_FLOAT128">,
+
+          GuardedFunctionSpec<"f16sub", RetValSpec<Float16Type>, [ArgSpec<DoubleType>, ArgSpec<DoubleType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16subf", RetValSpec<Float16Type>, [ArgSpec<FloatType>, ArgSpec<FloatType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16subl", RetValSpec<Float16Type>, [ArgSpec<LongDoubleType>, ArgSpec<LongDoubleType>], "LIBC_TYPES_HAS_FLOAT16">,
+          GuardedFunctionSpec<"f16subf128", RetValSpec<Float16Type>, [ArgSpec<Float128Type>, ArgSpec<Float128Type>], "LIBC_TYPES_HAS_FLOAT16_AND_FLOAT128">,
+
----------------
overmighty wrote:

Should `f16{add,sub}{,l}` be moved to llvm_libc_ext.td since they're LLVM libc extensions? We should probably also document these extensions somewhere.

https://github.com/llvm/llvm-project/pull/97072