On 11/28/2012 08:37 PM, Måns Rullgård wrote: > Jordi Ortiz <[email protected]> writes: > >> --- >> Cosmetic changes to last version. >> >> libavcodec/dsputil.c | 8 ++++++++ >> libavcodec/dsputil.h | 27 +++++++++++++++++++++++++++ >> 2 files changed, 35 insertions(+) >> >> diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c >> index d4471dc..b0522c9 100644 >> --- a/libavcodec/dsputil.c >> +++ b/libavcodec/dsputil.c >> @@ -2912,6 +2912,14 @@ av_cold void ff_dsputil_init(DSPContext* c, >> AVCodecContext *avctx) >> c->add_pixels4 = FUNCC(add_pixels4 ## dct , >> depth);\ >> c->put_no_rnd_pixels_l2[0] = FUNCC(put_no_rnd_pixels16_l2, >> depth);\ >> c->put_no_rnd_pixels_l2[1] = FUNCC(put_no_rnd_pixels8_l2 , >> depth);\ >> + c->put_pixels_l2[0] = put_pixels16_l2_8; \ >> + c->put_pixels_l2[1] = put_pixels8_l2_8; \ >> + c->put_pixels_l4[0] = put_pixels16_l4_8; \ >> + c->put_pixels_l4[1] = put_pixels8_l4_8; \ >> + c->avg_pixels_l2[0] = avg_pixels16_l2_8; \ >> + c->avg_pixels_l2[1] = avg_pixels8_l2_8; \ >> + c->avg_pixels_l4[0] = avg_pixels16_l4_8; \ >> + c->avg_pixels_l4[1] = avg_pixels8_l4_8; \ > > I still want to know why this is needed.
Because sadly those functions are generated by a template that is brought in dsputil.c Pending cutting it out of dsputil (pretty much slicing it in half) that's the sanest solution to get those functions w/out building them twice. lu _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
