On 02 March 2006 15:32, Christopher Faylor wrote:
> On Thu, Mar 02, 2006 at 03:24:59PM -0000, Dave Korn wrote:
>> On 02 March 2006 15:14, Christopher Faylor wrote:
>>
>>> On Thu, Mar 02, 2006 at 02:49:52PM -0000, Dave Korn wrote:
>>>> On 02 March 2006 14:34, Corinna Vinschen wrote:
>>>>> Well, I did this:
>>>>> $ nm libcygwin.a | grep isnan
>>>>> 00000000 T ___isnand
>>>>> 00000000 I __imp____isnand
>>>>> 00000000 T ___isnanf
>>>>> 00000000 I __imp____isnanf
>>>>> 00000000 I __imp___isnan
>>>>> 00000000 T __isnan
>>>>> 00000000 I __imp___isnanf
>>>>> 00000000 T __isnanf
>>>>> 00000000 I __imp__isnan
>>>>> 00000000 T _isnan
>>>>> 00000000 I __imp__isnanf
>>>>> 00000000 T _isnanf
>>>>> $ nm libm.a | grep isnan
>>>>> 00000000 T ___isnand
>KOFF< >KOFF< ;-)
>>>>> 00000000 I __imp____isnand
>>>>> 00000000 T ___isnanf
>>>>> 00000000 I __imp____isnanf
>>>>> 00000000 I __imp__isnan
>>>>> 00000000 T _isnan
>>>>> 00000000 I __imp__isnanf
>>>>> 00000000 T _isnanf
>>>>>
>>>>> I don't see what would be wrong here.
>>>>
>>>> It's not in libc.a, does that make a difference?
>>>
>>> It's in libm.a.
>>
>> That's orthogonal to the question I asked!
>
> Huh? I was providing some information which was missing from any other
> message I saw in this thread.
Actually it's right there in the output from nm, about ten lines above where
you mentioned it........
> Since all math functions are supposed to be in libm.a, then it really
> shouldn't be in libc.a and anything which relied on it being in libc.a
> would be in error.
>> BTW, I keep getting this "Error: Version info is older than DLL API!"
>> it. Is anyone else getting this? Could it be significant?
>
> No.
>
> cgf
Thanks for clarifying those points :)
cheers,
DaveK
--
Can't think of a witty .sigline today....
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
