diff options
| author | Wilco Dijkstra <wdijkstr@arm.com> | 2015-09-18 21:19:48 +0100 |
|---|---|---|
| committer | Wilco Dijkstra <wdijkstr@arm.com> | 2015-09-18 21:19:48 +0100 |
| commit | 89faa0340ad399ead8104f8fb6b7fed81f7d939c (patch) | |
| tree | cc15f087d86807d128a1fa592d00890e6c0e5982 /math/w_tgammaf.c | |
| parent | 3cc652e951c71785032019fec82e3b8543d85305 (diff) | |
| download | glibc-89faa0340ad399ead8104f8fb6b7fed81f7d939c.tar.gz glibc-89faa0340ad399ead8104f8fb6b7fed81f7d939c.tar.xz glibc-89faa0340ad399ead8104f8fb6b7fed81f7d939c.zip | |
Cleanup a few cases where isinf is used to get the signbit to improve the readability and maintainability and allow inlining.
2015-09-18 Wilco Dijkstra <wdijkstr@arm.com>
* math/w_tgamma.c (__ieee754_gamma_r): Use explicit sign check.
* math/w_tgammaf.c (__ieee754_gammaf_r): Likewise.
* math/w_tgammal.c (__ieee754_gammal_r): Likewise.
* stdio-common/printf_fp.c (___printf_fp):
Use signbit to get the sign. Use isinf macro to allow inlining.
* stdio-common/printf_fphex.c (__printf_fphex): Likewise.
* stdio-common/printf_size.c (__printf_size): Likewise.
Diffstat (limited to 'math/w_tgammaf.c')
| -rw-r--r-- | math/w_tgammaf.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/math/w_tgammaf.c b/math/w_tgammaf.c index 9ed7660d73..dfce894cc8 100644 --- a/math/w_tgammaf.c +++ b/math/w_tgammaf.c @@ -24,7 +24,7 @@ __tgammaf(float x) float y = __ieee754_gammaf_r(x,&local_signgam); if(__glibc_unlikely (!isfinite (y) || y == 0) - && (isfinite (x) || isinf (x) < 0) + && (isfinite (x) || (isinf (x) && x < 0.0)) && _LIB_VERSION != _IEEE_) { if (x == (float)0.0) /* tgammaf pole */ |