On Wed, Jul 30, 2008 at 11:03 AM, Paul Martz <[EMAIL PROTECTED]> wrote:

>  Thanks, Philip. I checked in the <stdlib.h> change on the 2.6 branch.
>
> I'm not sure the WIN32_WINNT change is correct. Are you aware that you can
> set this as a /D compile option in cmake? Given that, do you still think
> adding this #define directly to the source is the right thing to do? (A
> search for WIN32_WINNT at msdn turns up a variety of info, none of it
> particularly clear to me...)
>

Paul,

_WIN32_WINNT is already being defined to 0x400 at Win32MutexPrivateData.h:25
and Win32ThreadPrivateData.h:22.  Given that, I don't think there's any
problem with defining it wherever I ended up needing to put it in my
submission to remove the compile warning.

Also, I just looked it up and as I suspected, defining _WIN32_WINNT to 0x400
allows your application or library to use functionality introduced in
Windows NT 4.0 (but not necessarily available in Windows 95).  I'm assuming
this isn't a problem these days since nearly everyone runs Windows 2000 or
better.

If in the future someone wanted to take advantage of functionality present
only in Windows XP or Windows Vista within OpenThreads or the OSG then CMake
could indeed come in handy there in manipulating _WIN32_WINNT.or the "new"
definition of choice, NTDDI_VERSION

http://msdn.microsoft.com/en-us/library/aa383745(VS.85).aspx

-- 
Philip Lowman
_______________________________________________
osg-submissions mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org

Reply via email to