On Sun, Apr 5, 2015 at 11:40 AM James Almer <jamr...@gmail.com> wrote:
> On 05/04/15 3:26 PM, James Almer wrote: > > Signed-off-by: James Almer <jamr...@gmail.com> > > --- > > libswresample/resample_template.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/libswresample/resample_template.c b/libswresample/resample_ > template.c > > index 7e80ef9..d71efd6 100644 > > --- a/libswresample/resample_template.c > > +++ b/libswresample/resample_template.c > > @@ -53,7 +53,7 @@ > > # define FELEM_MAX INT32_MAX > > # define FELEM_MIN INT32_MIN > > # define OUT(d, v) (v) = ((v) + (1<<(FILTER_SHIFT-1)))>> > FILTER_SHIFT;\ > > - (d) = (uint64_t)((v) + 0x80000000) > 0xFFFFFFFF ? > ((v)>>63) ^ 0x7FFFFFFF : (v) > > + (d) = av_clipl_int32(v) > > Unlike the s16 template, this code is not being tested by fate-swr for > some reason. You can change this > line to "(d) = 0" and fate will still pass. > Not sure why, but s32 functions are not tested at all: http://coverage.ffmpeg.org/ffmpeg/libswresample/resample_dsp.c.gcov.html Timothy _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
