[AArch64] Define the macro __ARM_FP16_ARGS

[keith.walker.arm]

Summary

The ACLE (ARM C Language Extensions) 2.0 defines that the predefined macro
ARM_FP16_ARGS should be defined if fp16 can be used as an argument and
result.

The support for __fp16 to be used as an argument and result is already
implemented for AArch64 so this change is just adding the missing macro.

[ab]

LGTM, thanks!

lib/Basic/Targets.cpp

void getTargetDefines(const LangOptions &Opts,
     Builder.defineMacro("__ARM_ALIGN_MAX_STACK_PWR", "4");
 
     // 0xe implies support for half, single and double precision operations.
     Builder.defineMacro("__ARM_FP", "0xe");
 
     // PCS specifies this for SysV variants, which is all we support. Other ABIs
     // may choose __ARM_FP16_FORMAT_ALTERNATIVE.
     Builder.defineMacro("__ARM_FP16_FORMAT_IEEE");
+    Builder.defineMacro("__ARM_FP16_ARGS");
 
     if (Opts.FastMath || Opts.FiniteMathOnly)
       Builder.defineMacro("__ARM_FP_FAST");
 
     if (Opts.C99 && !Opts.Freestanding)
       Builder.defineMacro("__ARM_FP_FENV_ROUNDING");
 
     Builder.defineMacro("__ARM_SIZEOF_WCHAR_T", Opts.ShortWChar ? "2" : "4");

test/Preprocessor/aarch64-target-features.c

 // CHECK-NOT: __ARM_FEATURE_CRYPTO 1
 // CHECK: __ARM_FEATURE_DIRECTED_ROUNDING 1
 // CHECK: __ARM_FEATURE_DIV 1
 // CHECK: __ARM_FEATURE_FMA 1
 // CHECK: __ARM_FEATURE_IDIV 1
 // CHECK: __ARM_FEATURE_NUMERIC_MAXMIN 1
 // CHECK: __ARM_FEATURE_UNALIGNED 1
 // CHECK: __ARM_FP 0xe
+// CHECK: __ARM_FP16_ARGS 1
 // CHECK: __ARM_FP16_FORMAT_IEEE 1
 // CHECK-NOT: __ARM_FP_FAST 1
 // CHECK: __ARM_FP_FENV_ROUNDING 1
 // CHECK: __ARM_NEON 1
 // CHECK: __ARM_NEON_FP 0xe
 // CHECK: __ARM_PCS_AAPCS64 1
 // CHECK-NOT: __ARM_SIZEOF_MINIMAL_ENUM 1
 // CHECK-NOT: __ARM_SIZEOF_WCHAR_T 2