2 Regressions?

[Chad Gatesman]

Alexandre Oliva wrote:

> Chad Gatesman <chadg@redrose.net> writes:
>
> > First is that if you put inlines in the C or C++ module instead of in
> > the header file, it does not link correctly (example below):
>
> Which is correct, at least for C++.  The C++ Standard requires you to
> *define* an inline function in every translation unit that uses it.
> I don't know whether the C standard does too.

Ok.  It had linked correctly before with egcs-1.0.3.  I take it egcs
aggressively tryed to follow the standard in this case for 1.1, and that
is why my code now breaks.  I can accept that.  I wasn't aware that was
standard...looks like I will be fixing some code ;)

--
Chad Gatesman                 Home:  203.730.1389
Software Engineer             Work:  203.798.1007 x144
Bristol Technology            http://www.bristol.com/