diff options
author | Ronald S. Bultje <rsbultje@gmail.com> | 2011-06-30 17:35:13 -0700 |
---|---|---|
committer | Ronald S. Bultje <rsbultje@gmail.com> | 2011-07-01 09:08:26 -0700 |
commit | 8a8d0ce208b77f506759185ff580fa61b5c41f70 (patch) | |
tree | 544655de68668abe121739fd51ab4b964aa6ca96 /libswscale/utils.c | |
parent | cdc2c1c57616956d975c57b4b69eb73865f513f5 (diff) | |
download | ffmpeg-8a8d0ce208b77f506759185ff580fa61b5c41f70.tar.gz |
swscale: for >8bit scaling, read in native bit-depth.
For 9/10bit, it means we don't have to upscale to 16bit before
actual scaling or pixel format conversion, and thus a performance
gain.
Diffstat (limited to 'libswscale/utils.c')
-rw-r--r-- | libswscale/utils.c | 1 |
1 files changed, 0 insertions, 1 deletions
diff --git a/libswscale/utils.c b/libswscale/utils.c index b7ccac94ba..eea32a130a 100644 --- a/libswscale/utils.c +++ b/libswscale/utils.c @@ -877,7 +877,6 @@ int sws_init_context(SwsContext *c, SwsFilter *srcFilter, SwsFilter *dstFilter) } } - // FIXME it's even nicer if bpp isn't 16, but max({src,dst}formatbpp) c->scalingBpp = FFMAX(av_pix_fmt_descriptors[srcFormat].comp[0].depth_minus1, av_pix_fmt_descriptors[dstFormat].comp[0].depth_minus1) >= 8 ? 16 : 8; if (c->scalingBpp == 16) |