[cfe-commits] r58685 - /cfe/trunk/lib/Headers/stddef.h
sebastian.redl at getdesigned.at
Tue Nov 4 06:37:51 PST 2008
Chris Lattner wrote:
> On Nov 4, 2008, at 6:03 AM, Sebastian Redl wrote:
>> Chris Lattner wrote:
>>> This is fine in the short term, but I don't think this will work in
>> It's the way every C++ compiler out there does it.
> 0 or 0L?
VC++ 2003 and GCC 4.1 in the non-__null version use 0, not 0L. Other
compilers might be different, but I don't have access to them.
> Avoiding varargs isn't really an option
It is for the programmers. Not for us, of course. But the problem is
inherent in the language feature, not something we can do very much about.
If we implement __null, though, we should definitely warn if it is
passed to a varargs function.
More information about the cfe-commits