| /* Verify that overloaded built-ins for vec_sub with float | |
| inputs produce the right results. */ | |
| /* { dg-do compile } */ | |
| /* { dg-options "-maltivec -mno-vsx" } */ | |
| /* { dg-require-effective-target powerpc_altivec } */ | |
| #include <altivec.h> | |
| vector float | |
| test1 (vector float x, vector float y) | |
| { | |
| return vec_sub (x, y); | |
| } | |
| /* { dg-final { scan-assembler-times "vsubfp" 1 } } */ | |