Re: [gnugo-devel] Even more warnings :)

[Dave Denholm]

Daniel Bump <address@hidden> writes:

> > Unlikely that these cause any real trouble.
> > It looks like I need to back out my changes to the TRACE family
> > of macros, since we seem unable to avoid these warnings.
> 
> We discussed this issue around the time the patch went in,
> 
> http://mail.gnu.org/pipermail/gnugo-devel/2002-February/001427.html
> http://mail.gnu.org/pipermail/gnugo-devel/2002-February/001433.html
> http://mail.gnu.org/pipermail/gnugo-devel/2002-February/001436.html
> http://mail.gnu.org/pipermail/gnugo-devel/2002-February/001437.html
> 
> Perhaps these warnings might be worth tolerating if we can avoid 
> pointless function calls at every trace when !verbose.

If we are still talking about the version

#define TRACE  (verbose) && gprintf

then perhaps a compromise might be to make a global dummy variable, and then use

  dummy = (verbose) && gprintf


It adds a little overhead, but much less than a function call.


But I notice from the published build :

checking for gcc... no
checking for cc... cc
...
checking whether we are using the GNU C compiler... yes
...
checking for variadic cpp define... no



Is this a particularly old gcc, or is the variadic cpp test broken ?

Hmm...

cc -DHAVE_CONFIG_H -I. ... -Wp,-lang-c89 ...
                           ^^^^^^^^^^^^^

if this is disabling gcc cpp extensions, then variadic macros might be 
disabled..?

(IIRC, forcing ansi cpp may have been put in to force // comments to break..?)




dd
-- 
address@hidden          http://www.insignia.com