diff options
| author | Szabolcs Nagy <szabolcs.nagy@arm.com> | 2021-12-31 09:50:50 +0000 |
|---|---|---|
| committer | Szabolcs Nagy <szabolcs.nagy@arm.com> | 2022-01-10 14:27:17 +0000 |
| commit | 347a5b592c91f0230d52a5117fd83e1f75028207 (patch) | |
| tree | cef1eb7585b59b17118ad72f82856b39f94cfd14 /sysdeps/ieee754/flt-32/e_j0f.c | |
| parent | e72ef23ee88187284b4b1ca9b2e314e618429d35 (diff) | |
| download | glibc-347a5b592c91f0230d52a5117fd83e1f75028207.tar glibc-347a5b592c91f0230d52a5117fd83e1f75028207.tar.gz glibc-347a5b592c91f0230d52a5117fd83e1f75028207.tar.bz2 glibc-347a5b592c91f0230d52a5117fd83e1f75028207.zip | |
math: Fix float conversion regressions with gcc-12 [BZ #28713]
Converting double precision constants to float is now affected by the
runtime dynamic rounding mode instead of being evaluated at compile
time with default rounding mode (except static object initializers).
This can change the computed result and cause performance regression.
The known correctness issues (increased ulp errors) are already fixed,
this patch fixes remaining cases of unnecessary runtime conversions.
Add float M_* macros to math.h as new GNU extension API. To avoid
conversions the new M_* macros are used and instead of casting double
literals to float, use float literals (only required if the conversion
is inexact).
The patch was tested on aarch64 where the following symbols had new
spurious conversion instructions that got fixed:
__clog10f
__gammaf_r_finite@GLIBC_2.17
__j0f_finite@GLIBC_2.17
__j1f_finite@GLIBC_2.17
__jnf_finite@GLIBC_2.17
__kernel_casinhf
__lgamma_negf
__log1pf
__y0f_finite@GLIBC_2.17
__y1f_finite@GLIBC_2.17
cacosf
cacoshf
casinhf
catanf
catanhf
clogf
gammaf_positive
Fixes bug 28713.
Reviewed-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Diffstat (limited to 'sysdeps/ieee754/flt-32/e_j0f.c')
| -rw-r--r-- | sysdeps/ieee754/flt-32/e_j0f.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/sysdeps/ieee754/flt-32/e_j0f.c b/sysdeps/ieee754/flt-32/e_j0f.c index 0453a30109..9ae91a9dcc 100644 --- a/sysdeps/ieee754/flt-32/e_j0f.c +++ b/sysdeps/ieee754/flt-32/e_j0f.c @@ -233,7 +233,7 @@ j0f_near_root (float x, float z) float index_f; int index; - index_f = roundf ((x - FIRST_ZERO_J0) / (float) M_PI); + index_f = roundf ((x - FIRST_ZERO_J0) / M_PIf); /* j0f_asympt fails to give an error <= 9 ulps for x=0x1.324e92p+7 (index 48) thus we can't reduce SMALL_SIZE below 49. */ if (index_f >= SMALL_SIZE) @@ -514,7 +514,7 @@ y0f_near_root (float x, float z) float index_f; int index; - index_f = roundf ((x - FIRST_ZERO_Y0) / (float) M_PI); + index_f = roundf ((x - FIRST_ZERO_Y0) / M_PIf); if (index_f >= SMALL_SIZE) return y0f_asympt (x); index = (int) index_f; |