Message ID | 20210311124203.310-1-jamrial@gmail.com |
---|---|
State | Accepted |
Commit | 99042c2bf6cc79006036502a6abbec5e51f73673 |
Headers | show |
Series | [FFmpeg-devel,v2] avcodec/h264_slice: don't copy frame data during error concealment | expand |
Context | Check | Description |
---|---|---|
andriy/x86_make | success | Make finished |
andriy/x86_make_fate | success | Make fate finished |
andriy/PPC64_make | success | Make finished |
andriy/PPC64_make_fate | success | Make fate finished |
On Thu, Mar 11, 2021 at 09:42:03AM -0300, James Almer wrote: > In addition to the fact that av_image_copy() cannot handle hardware pixel formats, > h->short_ref[0]->f may not be writable at this point. > > Based on a patch by Hendrik Leppkes. > > Signed-off-by: James Almer <jamrial@gmail.com> > --- > This version fixes the fuzzed sample Michael talked about. as said privatly already, this works fine with everything i threw at it thx [...]
On 3/11/2021 4:54 PM, Michael Niedermayer wrote: > On Thu, Mar 11, 2021 at 09:42:03AM -0300, James Almer wrote: >> In addition to the fact that av_image_copy() cannot handle hardware pixel formats, >> h->short_ref[0]->f may not be writable at this point. >> >> Based on a patch by Hendrik Leppkes. >> >> Signed-off-by: James Almer <jamrial@gmail.com> >> --- >> This version fixes the fuzzed sample Michael talked about. > > as said privatly already, this works fine with everything i threw at it > > thx Pushed, thanks.
diff --git a/libavcodec/h264_slice.c b/libavcodec/h264_slice.c index fa7a639053..14b945756b 100644 --- a/libavcodec/h264_slice.c +++ b/libavcodec/h264_slice.c @@ -1599,14 +1599,15 @@ static int h264_field_start(H264Context *h, const H264SliceContext *sl, ff_thread_await_progress(&prev->tf, INT_MAX, 0); if (prev->field_picture) ff_thread_await_progress(&prev->tf, INT_MAX, 1); - av_image_copy(h->short_ref[0]->f->data, - h->short_ref[0]->f->linesize, - (const uint8_t **)prev->f->data, - prev->f->linesize, - prev->f->format, - prev->f->width, - prev->f->height); + ff_thread_release_buffer(h->avctx, &h->short_ref[0]->tf); + h->short_ref[0]->tf.f = h->short_ref[0]->f; + ret = ff_thread_ref_frame(&h->short_ref[0]->tf, &prev->tf); + if (ret < 0) + return ret; h->short_ref[0]->poc = prev->poc + 2U; + ff_thread_report_progress(&h->short_ref[0]->tf, INT_MAX, 0); + if (h->short_ref[0]->field_picture) + ff_thread_report_progress(&h->short_ref[0]->tf, INT_MAX, 1); } else if (!h->frame_recovered && !h->avctx->hwaccel) ff_color_frame(h->short_ref[0]->f, c); h->short_ref[0]->frame_num = h->poc.prev_frame_num;
In addition to the fact that av_image_copy() cannot handle hardware pixel formats, h->short_ref[0]->f may not be writable at this point. Based on a patch by Hendrik Leppkes. Signed-off-by: James Almer <jamrial@gmail.com> --- This version fixes the fuzzed sample Michael talked about. libavcodec/h264_slice.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-)