Hi Piotr,

Thank you very much for pointing out the problems.

And about��

�� float CDECL MSVCRT__copysignf( float num, float sign )
�� {
-�� /* FIXME: Behaviour for Nan/Inf? */
-�� if (sign < 0.0)
-�� return num < 0.0 ? num : -num;
-�� return num < 0.0 ? -num : num;
+�� /* FIXME: Behaviour for signbit(NAN) is different in Linux and
+�� *�� Windows, where Windows gives a zero for -NAN
+�� */
+�� if (signbit(sign))
+�� return signbit(num) ? num : -num;
+�� return signbit(num) ? -num : num;

It doesn't make sense to add this comment and remove it in next patch. I would also prefer if signbit fallback implementation is changed in this patch.

��

I removed the comments now because I thought this is already fixed... I've combined the changes and sent two patches.

Thanks again :)

Regards,

Zheng