diff options
| author | Joseph Myers <joseph@codesourcery.com> | 2015-12-01 01:01:36 +0000 |
|---|---|---|
| committer | Joseph Myers <joseph@codesourcery.com> | 2015-12-01 01:01:36 +0000 |
| commit | 60f435bb0c097ead2d4609aa7e45a203eb24e43c (patch) | |
| tree | 57707e61b9a43a675aada5dc4802ed11edd1ecbb /sysdeps/ieee754/dbl-64/e_sqrt.c | |
| parent | 9627da32ec76600244e7723e99b9d4e27691f1ff (diff) | |
| download | glibc-60f435bb0c097ead2d4609aa7e45a203eb24e43c.tar.gz | |
Use hex float constants in sysdeps/ieee754/dbl-64/e_sqrt.c.
Various sysdeps/ieee754/dbl-64 functions use double constants defined
using a union between a double and two ints, with separate big-endian
and little-endian definitions of the constants.
With modern C, this is unnecessary complication; hex float constants
(or __builtin_inf etc.) suffice to specify the exact value desired,
and so can avoid separate versions for each endianness. Having this
complication also complicates cleanups such as removing slow paths
from these library functions, as they need to make sure to remove both
copies of variables that are no longer used after such a cleanup (and
in at least one case, proper removal of a slow path will also involve
removing slow-path-only values from the middle of an array - an array
with both big-endian and little-endian copies - and adjusting other
references to that array).
So it makes sense to clean up the code to define these constants using
hex floats and so eliminate the endianness conditional. This patch
does so in the case of sqrt, where the two constants are such that it
makes sense just to put them directly in the code using them and
eliminate the names for them altogether.
Tested for arm (the code generated for sqrt does change, though not in
any significant way).
* sysdeps/ieee754/dbl-64/e_sqrt.c: Do not include uroot.h.
(__ieee754_sqrt): Use hex float constants instead of tm256.x and
t512.x.
* sysdeps/ieee754/dbl-64/uroot.h: Remove file.
Diffstat (limited to 'sysdeps/ieee754/dbl-64/e_sqrt.c')
| -rw-r--r-- | sysdeps/ieee754/dbl-64/e_sqrt.c | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/sysdeps/ieee754/dbl-64/e_sqrt.c b/sysdeps/ieee754/dbl-64/e_sqrt.c index fff6d148fe..c1fed7d97e 100644 --- a/sysdeps/ieee754/dbl-64/e_sqrt.c +++ b/sysdeps/ieee754/dbl-64/e_sqrt.c @@ -21,7 +21,7 @@ /* */ /* FUNCTION: usqrt */ /* */ -/* FILES NEEDED: dla.h endian.h mydefs.h uroot.h */ +/* FILES NEEDED: dla.h endian.h mydefs.h */ /* uroot.tbl */ /* */ /* An ultimate sqrt routine. Given an IEEE double machine number x */ @@ -47,7 +47,6 @@ double __ieee754_sqrt (double x) { -#include "uroot.h" static const double rt0 = 9.99999999859990725855365213134618E-01, rt1 = 4.99999999495955425917856814202739E-01, @@ -134,7 +133,7 @@ __ieee754_sqrt (double x) return x; /* sqrt(+0)=+0, sqrt(-0)=-0 */ if (k < 0) return (x - x) / (x - x); /* sqrt(-ve)=sNaN */ - return tm256.x * __ieee754_sqrt (x * t512.x); + return 0x1p-256 * __ieee754_sqrt (x * 0x1p512); } } strong_alias (__ieee754_sqrt, __sqrt_finite) |