Currently, it also tests whether extended_data points to something
different than the AVFrame's data array and frees extended_data
if it is different. Yet this is only necessary for one of its three
callers, namely av_frame_unref(); meanwhile the other two callers
took measures to avoid this (or rather, to make it to an av_free(NULL)).
This commit moves this chunk to av_frame_unref() (so that
get_frame_defaults() now treats its input as uninitialized)
and removes the now superfluous code in the other two callers.
Signed-off-by: Andreas Rheinhardt <andreas.rheinha...@outlook.com>
---
No code actually tests the scenario in which data != extended_data.
Of the 7984267 calls to get_frame_defaults() during FATE, 2784297
freed the extended_data, but in all these scenarios (which all came
from av_frame_alloc() and av_frame_move_ref()) extended_data was NULL.
libavutil/frame.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/libavutil/frame.c b/libavutil/frame.c
index d4d3ad6988..95bdc0be72 100644
--- a/libavutil/frame.c
+++ b/libavutil/frame.c
@@ -52,9 +52,6 @@ const char *av_get_colorspace_name(enum AVColorSpace val)
#endif
static void get_frame_defaults(AVFrame *frame)
{
- if (frame->extended_data != frame->data)
- av_freep(&frame->extended_data);
-
memset(frame, 0, sizeof(*frame));
frame->pts =
@@ -98,12 +95,11 @@ static void wipe_side_data(AVFrame *frame)
AVFrame *av_frame_alloc(void)
{
- AVFrame *frame = av_mallocz(sizeof(*frame));
+ AVFrame *frame = av_malloc(sizeof(*frame));
if (!frame)
return NULL;
- frame->extended_data = NULL;
get_frame_defaults(frame);
return frame;
@@ -455,6 +451,9 @@ void av_frame_unref(AVFrame *frame)
av_buffer_unref(&frame->opaque_ref);
av_buffer_unref(&frame->private_ref);
+ if (frame->extended_data != frame->data)
+ av_freep(&frame->extended_data);
+
get_frame_defaults(frame);
}
@@ -466,7 +465,6 @@ void av_frame_move_ref(AVFrame *dst, AVFrame *src)
*dst = *src;
if (src->extended_data == src->data)
dst->extended_data = dst->data;
- memset(src, 0, sizeof(*src));
get_frame_defaults(src);
}
--
2.32.0
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
