diff options
| author | Mark Dickinson <dickinsm@gmail.com> | 2009-02-09 14:18:43 (GMT) |
|---|---|---|
| committer | Mark Dickinson <dickinsm@gmail.com> | 2009-02-09 14:18:43 (GMT) |
| commit | c97c9096ed05c89041b2aab9578489ca634c9375 (patch) | |
| tree | 39281e41387e90170eb0370e3ada481ca33f99fd /Python | |
| parent | b20af944d27e6277a9afa9f01282673eb89f7cf6 (diff) | |
| download | cpython-c97c9096ed05c89041b2aab9578489ca634c9375.zip cpython-c97c9096ed05c89041b2aab9578489ca634c9375.tar.gz cpython-c97c9096ed05c89041b2aab9578489ca634c9375.tar.bz2 | |
Issue #4575: fix Py_IS_INFINITY macro to work correctly on x87 FPUs.
It now forces its argument to double before testing for infinity.
Diffstat (limited to 'Python')
| -rw-r--r-- | Python/pymath.c | 13 |
1 files changed, 13 insertions, 0 deletions
diff --git a/Python/pymath.c b/Python/pymath.c index 5cf61ab..5d09b4c 100644 --- a/Python/pymath.c +++ b/Python/pymath.c @@ -1,5 +1,18 @@ #include "Python.h" +#ifdef X87_DOUBLE_ROUNDING +/* On x86 platforms using an x87 FPU, this function is called from the + Py_FORCE_DOUBLE macro (defined in pymath.h) to force a floating-point + number out of an 80-bit x87 FPU register and into a 64-bit memory location, + thus rounding from extended precision to double precision. */ +double _Py_force_double(double x) +{ + volatile double y; + y = x; + return y; +} +#endif + #ifndef HAVE_HYPOT double hypot(double x, double y) { |