[Ffmpeg-cvslog] r6732 - in trunk/libavcodec: cabac.h h264.c

Uoti Urpala uoti.urpala
Mon Nov 27 22:47:57 CET 2006


On Mon, 2006-11-27 at 16:04 -0500, Rich Felker wrote:
> On Mon, Nov 27, 2006 at 12:08:40PM +0100, Dominik 'Rathann' Mierzejewski wrote:
> > > maybe noone uses PIC lavc ...
> > 
> > Believe me, people do use it.
> 
> I'll remember to cite this thread next time uau flames about gcc 2.95
> support, saying it doesn't matter because the c++ declarations lasted
> for a few months with nobody noticing. :)) Apparently his precious PIC
> support is just as unnecessary. :)

"his precious" PIC support? That sounds like I would have promoted
ffmpeg PIC support as being particularly important, which I have not
done.

Is there any case where PIC with that asm could be compiled except
enabling shared libraries on AMD64 or manually forcing -pic on x86-32
(in which case reporting errors would be less likely)? There certainly
have been reports of problems with asm on AMD64 on #ffmpeg at least,
though I haven't paid much attention to the details.

Even if no one would have noticed PIC breakage (which does not actually
seem to be true) for as long as MPlayer not supporting gcc 2.95 was
unnoticed that alone of course wouldn't make PIC support "just as
unnecessary". There are other reasons why support for gcc 2.95 should be
dropped; that it already was unsupported for months without serious
problems just demonstrates that dropping support is a practical choice.





More information about the ffmpeg-cvslog mailing list