diff options
| author | Joseph Myers <joseph@codesourcery.com> | 2016-10-07 23:28:33 +0000 |
|---|---|---|
| committer | Joseph Myers <joseph@codesourcery.com> | 2016-10-07 23:28:33 +0000 |
| commit | d0800ae0e7162631136a0042e87eb42c641455d9 (patch) | |
| tree | 2bd2e6a01da8827ad5e2536b8c3aa4ae258a481b /math/math.h | |
| parent | 12e5d361a4dcfe3579f1dd263f911ac193b0fa29 (diff) | |
| download | glibc-d0800ae0e7162631136a0042e87eb42c641455d9.zip glibc-d0800ae0e7162631136a0042e87eb42c641455d9.tar.gz glibc-d0800ae0e7162631136a0042e87eb42c641455d9.tar.bz2 | |
Make iseqsig handle excess precision.
iseqsig, like other type-generic comparison macros, should behave like
a comparison operator in not removing excess range and precision from
its arguments (see C11 F.10.11). This patch implements this by making
definitions of iseqsig appropriately conditional on
__FLT_EVAL_METHOD__ (including support for TS 18661-3 values of that
macro), with a corresponding testcase (that failed for 32-bit x86 in
the absence of the math.h changes) being added. (Of course the
definitions may need reworking when float128 support is added, just as
with other type-generic macros.)
Tested for x86_64 and x86.
* math/math.h [__GLIBC_USE (IEC_60559_BFP_EXT)] (iseqsig): Define
conditional on value of [__FLT_EVAL_METHOD__].
* math/test-iseqsig-excess-precision.c: New file.
* math/Makefile (tests): Add test-iseqsig-excess-precision.
Diffstat (limited to 'math/math.h')
| -rw-r--r-- | math/math.h | 19 |
1 files changed, 17 insertions, 2 deletions
diff --git a/math/math.h b/math/math.h index 394cefa..880b4a0 100644 --- a/math/math.h +++ b/math/math.h @@ -535,17 +535,32 @@ extern int matherr (struct exception *__exc); /* Return X == Y but raising "invalid" and setting errno if X or Y is a NaN. */ # ifdef __NO_LONG_DOUBLE_MATH -# define iseqsig(x, y) \ +# if (__FLT_EVAL_METHOD__ == 1 \ + || __FLT_EVAL_METHOD__ == 2 \ + || __FLT_EVAL_METHOD__ > 32) +# define iseqsig(x, y) __iseqsig ((x), (y)) +# else +# define iseqsig(x, y) \ (sizeof ((x) + (y)) == sizeof (float) \ ? __iseqsigf ((x), (y)) \ : __iseqsig ((x), (y))) +# endif # else -# define iseqsig(x, y) \ +# if __FLT_EVAL_METHOD__ == 2 || __FLT_EVAL_METHOD__ > 64 +# define iseqsig(x, y) __iseqsigl ((x), (y)) +# elif __FLT_EVAL_METHOD__ == 1 || __FLT_EVAL_METHOD__ > 32 +# define iseqsig(x, y) \ + (sizeof ((x) + (y)) <= sizeof (double) \ + ? __iseqsig ((x), (y)) \ + : __iseqsigl ((x), (y))) +# else +# define iseqsig(x, y) \ (sizeof ((x) + (y)) == sizeof (float) \ ? __iseqsigf ((x), (y)) \ : sizeof ((x) + (y)) == sizeof (double) \ ? __iseqsig ((x), (y)) \ : __iseqsigl ((x), (y))) +# endif # endif #endif |