diff options
| author | Michael Niedermayer <[email protected]> | 2019-10-12 13:46:04 +0200 |
|---|---|---|
| committer | Michael Niedermayer <[email protected]> | 2019-11-15 12:25:46 +0100 |
| commit | 386b987f2a8c4abe507d2d58136acc0ef4ace9a3 (patch) | |
| tree | dd4e194889797930ea47a45261dde63efdbbfcce | |
| parent | 25a917e46f798350460df75e35cf87c07bff83b9 (diff) | |
avfilter/vf_geq: Use av_clipd() instead of av_clipf()
With floats we cannot represent all 32bit integer dimensions
Reviewed-by: Paul B Mahol <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
(cherry picked from commit c8813b1a984714f0027cabeea2394035df20cf38)
Signed-off-by: Michael Niedermayer <[email protected]>
| -rw-r--r-- | libavfilter/vf_geq.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/libavfilter/vf_geq.c b/libavfilter/vf_geq.c index 36dbd421ce..f441c591f9 100644 --- a/libavfilter/vf_geq.c +++ b/libavfilter/vf_geq.c @@ -82,8 +82,8 @@ static inline double getpix(void *priv, double x, double y, int plane) if (!src) return 0; - xi = x = av_clipf(x, 0, w - 2); - yi = y = av_clipf(y, 0, h - 2); + xi = x = av_clipd(x, 0, w - 2); + yi = y = av_clipd(y, 0, h - 2); x -= xi; y -= yi; |