| /* { dg-do compile } */ | |
| /* { dg-require-effective-target hard_float } */ | |
| /* { dg-options "-fno-finite-math-only -fsigned-zeros -fno-signaling-nans -dp" } */ | |
| float | |
| fmaxf (float x, float y) | |
| { | |
| return __builtin_fmaxf (x, y); | |
| } | |
| /* { dg-final { scan-assembler-not "\ttail\tfmaxf\t" } } */ | |
| /* { dg-final { scan-assembler-not "\tfge\\.s\t" } } */ | |
| /* { dg-final { scan-assembler "\tfmax\\.s\t\[^\n\]* fmaxsf3\n" } } */ |