[FFmpeg-devel] Update on lbos

[Michael Niedermayer]

On Thu, Feb 07, 2008 at 11:54:13AM -0500, Rich Felker wrote:
> On Thu, Feb 07, 2008 at 09:34:50AM +0100, Lucabe72 wrote:
> > Hi Dave,
> > [...]
> > > gcc -Ii:/usr/src/libbrokenos/include -Zomf -fomit-frame-pointer -g -Wall 
> > > -Wno-switch -Wdisabled-optimization -Wpointer-arith -Wredundant-decls 
> > > -Wcast-qual -Wwrite-strings -O3 -I"/usr/src/ffmpeg" -I"/usr/src/ffmpeg" 
> > > -I"/usr/src/ffmpeg"/libavutil -I"/usr/src/ffmpeg"/libavcodec 
> > > -I"/usr/src/ffmpeg"/libavformat -I"/usr/src/ffmpeg"/libswscale 
> > > -I"/usr/src/ffmpeg"/libavdevice -D_FILE_OFFSET_BITS=64 
> > > -D_LARGEFILE_SOURCE -D_ISOC9X_SOURCE -DHAVE_AV_CONFIG_H -c -o ffserver.o 
> > > ffserver.c
> > > In file included from i:/usr/src/libbrokenos/include/unistd.h:1,
> > >                   from ffserver.c:31:
> > > i:/usr/src/libbrokenos/include/unistd_orig.h:338: warning: redundant 
> > > redeclaration of `_exit' in same scope
> > > i:/usr/include/stdlib.h:431: warning: previous declaration of `_exit'
> > 
> > Ok, thanks; these are the information I needed.
> > It looks like _exit() is defined both in unistd.h and
> > stdlib.h?
> 
> This is incorrect; it should only be in unistd.h.
> 
> > I am wondering why you do not see this kind of warnings
> > when compiling programs that include both stdlib.h and
> > unistd.h...
> 
> It is normal to have 2 or more prototypes for the same function as
> long as they do not conflict with one another. I have never heard of a
> warning for such a thing. Who put -Wredundant-decls in the default
> warnings?? 

Maybe it was me. Theres no reason to have redundant declarations IN ffmpeg.


> This is sure to get all kinds of bogus warnings from
> system headers!!

Go report the bug to the gcc devels. Gcc does disable all warnings for
system headers due to the "high" quality of the code in them.
If you dont like the default theres -Wsystem-headers though ive not tried it.

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080207/6ec22297/attachment.pgp>