[uClibc]small bug in 0.9.16 and even cvs snapshot 20021209

Russ Dill Russ.Dill at asu.edu
Wed Dec 11 06:21:22 UTC 2002


On Tue, 2002-12-10 at 18:31, Manuel Novoa III wrote:
> On Wed, Dec 11, 2002 at 02:21:22AM +0100, Cristian Ionescu-Idbohrn wrote:
> > On Wed, 10 Dec 2002, Russ Dill wrote:
> > 
> > > On Tue, 2002-12-10 at 17:03, Cristian Ionescu-Idbohrn wrote:
> > > > Application compiled with -Werror reveals this:
> > > >
> > > > ,----
> > > > | In file included from
> > > > | uclibc/include/ctype.h:25,
> > > > |                  from nanotop.h:28,
> > > > |                  from nanotop.c:23:
> > > > | uclibc/include/bits/uClibc_ctype.h:62:
> > > > | warning: comma at end of enumerator list
> > >
> > > thats a gcc extension, and its commonly done..its not really an error or
> > > a bug, it just makes coding a little easier
> > 
> > Meaning side effect is noone can use -Werror gcc option on anything built
> > with uClibc which includes ctype.h. It does not seem gcc 2.95.4 knows
> > about that extension. I personally can live without the comfort you're
> > talking about. Catching potentional problems is more valuable to me.
> 
> It's easy enough to change, so I'll take care of it.
> 


I think the correct fix would be to change the cflags so that 'comma at
end of enumerator list' is no longer an error, since this will pop up in
a lot of patches or submitted code (don't know what that cflag is)




More information about the uClibc mailing list