|
From: Paul G. <pga...@te...> - 2000-12-08 01:29:30
|
On 7 Dec 2000, at 22:23, the Illustrious Niels Diepeveen wrote: > > Hi Ben, > > Ben Allfree schreef: > > > > I'm still working on getting Python 2.0 to compile using > > MinGW32 2.95.2-1, > > So am I. Maybe we could work together on this? I have built the > complete 1.5.2 and 1.6 earlier and I'm now working on adapting > those changes to 2.0. It builds, except for some modules that > require extra libraries (_tkinter, pyexpat etc). Then I get some > nasty problems running regrtest. Has anyone ever used _alloca()? > I think it may be broken. Try using malloc() instead. Afaik, _alloca() is a Unix call, and is not part of the mingw runtime headers...nor is allocate(). > > > > I don't understand how the largeint.h functions ever worked > > (maybe they never did?), because gcc complains every time for > > me. > > I have been messing with largeint.h for a long time, and got it > nearly working several times, but ultimately I decided that it > was easier to rewrite the one function that uses it (os.clock()) > using LONGLONG arithmetic directly. I think getting largeint.h to > work would require rewriting most of it. largeint.h was originally started by Mumit as a first step to enable 64bit support for mingw. However, it was not ever fully implemented or tested. Peace, Paul G. > > > -- > Niels Diepeveen > Endea automatisering > > _______________________________________________ > MinGW-users mailing list > Min...@li... > > You may change your MinGW Account Options at: > http://lists.sourceforge.net/mailman/listinfo/mingw-users > Nothing real can be threatened. Nothing unreal exists. |