[FFmpeg-devel] Broken compile with latest libavutil/common.h

[Aurelien Jacobs]

On Sat, 17 Jan 2009 16:19:17 +0000
M?ns Rullg?rd <mans at mansr.com> wrote:

> Michael Niedermayer <michaelni at gmx.at> writes:
> 
> > On Sat, Jan 17, 2009 at 02:08:08PM +0100, Aurelien Jacobs wrote:
> >> On Sat, 17 Jan 2009 03:06:06 +0100
> >> Michael Niedermayer <michaelni at gmx.at> wrote:
> >> 
> >> > On Fri, Jan 16, 2009 at 12:53:24AM +0100, Aurelien Jacobs wrote:
> >> > > Michael Niedermayer wrote:
> >> > [...]
> >> > > > [...]
> >> > > > > 
> >> > > > > > +/* math */
> >> > > > > > +int64_t av_const ff_gcd(int64_t a, int64_t b);
> >> > > > > 
> >> > > > > The function is in mathematics.c.  Maybe mathematics.h would be a good
> >> > > > > place for this prototype.  
> >> > > > 
> >> > > > > Maybe we should even make that function
> >> > > > > public.
> >> > > > 
> >> > > > I think so too
> >> > > 
> >> > > Done in attached patch.
> >> > 
> >> > ok
> >> 
> >> Applied.
> >> 
> >> > > > > > +/**
> >> > > > > > + * converts fourcc string to int
> >> > > > > > + */
> >> > > > > > +static inline av_pure int ff_get_fourcc(const char *s){
> >> > > > > > +    assert( strlen(s)==4 );
> >> > > > > > +    return (s[0]) + (s[1]<<8) + (s[2]<<16) + (s[3]<<24);
> >> > > > > > +}
> >> > > > > 
> >> > > > > This looks a lot like AV_RL32().  
> >> > > > 
> >> > > > indeed and i wouldnt mind seeing all ff_get_fourcc replaced by it
> >> > > 
> >> > > See attached patch.
> >> > 
> >> > ok
> >> 
> >> Applied.
> >> 
> >> Now the only non-public part left in common.h is mid_pred().
> >> Is moving it to libavcodec/internal.h OK ?
> >> If not, where should it be moved ?
> >
> > i dont mind it being moved to libavcodec/internal.h, but IIRC there was
> > some uglyness in the patch ...
> 
> If it's going to libavcodec, I'd prefer it in mathops.h.  That file
> has per-arch overrides already, and separating the assembler in that
> function as well would be nice.

That sounds like a good idea. Attached patch moves mid_pred() in mathops.h
and the x86 specific implementation in x86/mathops.h.

Aurel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mathops_mid_pred.diff
Type: text/x-diff
Size: 8619 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090117/6b5373e3/attachment.diff>