diff options
| author | Marton Balint <cus@passwd.hu> | 2024-01-26 01:42:13 +0100 |
|---|---|---|
| committer | Marton Balint <cus@passwd.hu> | 2024-02-04 20:04:18 +0100 |
| commit | 8fccd6d510fe802c7d2dbe291676b0780cd4cf7f (patch) | |
| tree | 3a0e7d36cc08fd81416b5122fa2e909f061f39f9 | |
| parent | 976cf2343d8c624c243895c21585d3c7711ab8fe (diff) | |
| download | ffmpeg-8fccd6d510fe802c7d2dbe291676b0780cd4cf7f.tar.gz | |
avutil/rational: increase av_d2q precision
Fixes parsing small timebases from expressions (where the expression API
converts the result to double), like in this command line:
ffprobe -f lavfi -i testsrc=d=1,settb=1/2000000000 -show_streams -show_entries stream=time_base
Before the patch timebase was parsed as 1/1999999999.
Signed-off-by: Marton Balint <cus@passwd.hu>
| -rw-r--r-- | libavutil/rational.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/libavutil/rational.c b/libavutil/rational.c index eb148ddb12..329fbf3302 100644 --- a/libavutil/rational.c +++ b/libavutil/rational.c @@ -114,7 +114,7 @@ AVRational av_d2q(double d, int max) return (AVRational) { d < 0 ? -1 : 1, 0 }; frexp(d, &exponent); exponent = FFMAX(exponent-1, 0); - den = 1LL << (61 - exponent); + den = 1LL << (62 - exponent); // (int64_t)rint() and llrint() do not work with gcc on ia64 and sparc64, // see Ticket2713 for affected gcc/glibc versions av_reduce(&a.num, &a.den, floor(d * den + 0.5), den, max); |