Teach tryEvaluate that fabs, copysign, and unary +/- are constants for
floats.
 - With testcase, which also has some other things GCC folds but we
   don't commented out in it.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@57624 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/Sema/constant-builtins-2.c b/test/Sema/constant-builtins-2.c
new file mode 100644
index 0000000..d47cc35
--- /dev/null
+++ b/test/Sema/constant-builtins-2.c
@@ -0,0 +1,50 @@
+// RUN: clang -fsyntax-only %s
+
+// Math stuff
+
+double       g0  = __builtin_huge_val();
+float        g1  = __builtin_huge_valf();
+long double  g2  = __builtin_huge_vall();
+
+double       g3  = __builtin_inf();
+float        g4  = __builtin_inff();
+long double  g5  = __builtin_infl();
+
+double       g6  = __builtin_nan("");
+float        g7  = __builtin_nanf("");
+long double  g8  = __builtin_nanl("");
+
+// GCC constant folds these too (via native strtol):
+//double       g6_1  = __builtin_nan("1");
+//float        g7_1  = __builtin_nanf("1");
+//long double  g8_1  = __builtin_nanl("1");
+
+// APFloat doesn't have signalling NaN functions.
+//double       g9  = __builtin_nans("");
+//float        g10 = __builtin_nansf("");
+//long double  g11 = __builtin_nansl("");
+
+//int          g12 = __builtin_abs(-12);
+
+double       g13 = __builtin_fabs(-12.);
+double       g13_0 = __builtin_fabs(-0.);
+double       g13_1 = __builtin_fabs(-__builtin_inf());
+float        g14 = __builtin_fabsf(-12.f);
+// GCC doesn't eat this one.
+//long double  g15 = __builtin_fabsfl(-12.0L);
+
+float        g16 = __builtin_copysign(1.0, -1.0);
+double       g17 = __builtin_copysignf(1.0f, -1.0f);
+long double  g18 = __builtin_copysignl(1.0L, -1.0L);
+
+//double       g19 = __builtin_powi(2.0, 4);
+//float        g20 = __builtin_powif(2.0f, 4);
+//long double  g21 = __builtin_powil(2.0L, 4);
+
+// GCC misc stuff
+
+extern int f();
+
+int h0 = __builtin_types_compatible_p(int, float);
+//int h1 = __builtin_choose_expr(1, 10, f());
+//int h2 = __builtin_expect(0, 0);