[Ffmpeg-cvslog] r6338 - in trunk: ffmpeg.c libavcodec/mpegvideo.c libavcodec/ratecontrol.c libavcodec/snow.c

[Guillaume POIRIER]

Hi,

On 9/29/06, Michael Niedermayer <michaelni at gmx.at> wrote:
> On Thu, Sep 28, 2006 at 05:58:21PM +0100, M?ns Rullg?rd wrote:
> > Ismail Donmez said:
> > > 28 Eyl 2006 Per 19:19 tarihinde, Panagiotis Issaris ??unlar?? yazm????t??:
> > >> Hi,
> > >>
> > >> On wo, 2006-09-27 at 23:17 +0100, M?ns Rullg?rd wrote:
> > >> >[...]
> > >> >
> > >> > >> Log:
> > >> > >> Handle possible failure of ff_eval.
> > >> > >
> > >> > > That commit completely breaks mpeg encoding. Am I the only only
> > >> > > actually encoding ? :>
> > >> >
> > >> > Should be fixed.
> > >> >
> > >> > To takis: please pay attention to compiler warnings when you make
> > >> > changes.
> > >>
> > >> You are right. But, "make mpegvideo.o | grep warning|wc -l" returns 324 on
> > >> my system, which makes it difficult to see the new ones in the pages of
> > >> existing warnings...
> > >
> > > remove -Winline from configure and then you can see useful warnings.
> >
> > Why was that flag added in the first place?
>
> because i like to know when gcc decides to ignore me
> furthermore if gcc ignores some inline then this isnt optimal, either
> the inline should be changed to always_inline or the compiler parameters
> should be changed or the inline should be removed

Well, in the case of "inline" keyword, gcc doesn't ignore you by not
inlining... The C norm does clearly say that this keyword is just a
hint passed to the compiler to effect its inlining heuristic...

But I'm sure you knew that ;-)

Guillaume
-- 
With DADVSI (http://en.wikipedia.org/wiki/DADVSI), France finally has
a lead on USA on selling out individuals right to corporations!
Vive la France!