On Thu, Feb 11, 2021 at 10:34:11PM +0000, Mark Thompson wrote:
> On 11/02/2021 21:48, Lluís Batlle i Rossel wrote:
> > From: Lluís Batlle i Rossell <vi...@viric.name>
> >
> > +
>
> Random whitespace change?
likely. I can fix that.
>
> > if (!ref)
> > return AVERROR(ENOMEM);
> > @@ -140,8 +141,19 @@ static int ffmmal_set_ref(AVFrame *frame, FFPoolRef
> > *pool,
> > atomic_fetch_add_explicit(&ref->pool->refcount, 1,
> > memory_order_relaxed);
> > mmal_buffer_header_acquire(buffer);
> > - frame->format = AV_PIX_FMT_MMAL;
> > - frame->data[3] = (uint8_t *)ref->buffer;
> > + frame->format = avctx->pix_fmt;
> > +
> > + if (avctx->pix_fmt == AV_PIX_FMT_YUV420P) {
> > + int w = FFALIGN(avctx->width, 32);
> > + int h = FFALIGN(avctx->height, 16);
> > +
> > + av_image_fill_arrays(frame->data, frame->linesize,
> > + buffer->data + buffer->type->video.offset[0],
> > + avctx->pix_fmt, w, h, 1);
> > + } else {
> > + frame->data[3] = (uint8_t *)ref->buffer;
> > + }
> > +
> > return 0;
> > }
> > @@ -633,30 +645,14 @@ static int ffmal_copy_frame(AVCodecContext *avctx,
> > AVFrame *frame,
>
> The function name is very misleading after this change.
Definitely. I can fix that.
> > + if ((ret = ffmmal_set_ref(avctx, frame, ctx->pool_out, buffer)) < 0)
> > + goto done;
> > frame->pts = buffer->pts == MMAL_TIME_UNKNOWN ? AV_NOPTS_VALUE :
> > buffer->pts;
> > #if FF_API_PKT_PTS
> >
>
> What happens if the user holds on to multiple of the output frames, perhaps
> because they are being queued into some other operation? The number of
> buffers appears fixed, so what happens if they run out?
I don't know. I don't know how to trigger that situation. That would
affect also the "mmal" pixfmt, already present.
--
(Escriu-me xifrat si saps PGP / Write ciphered if you know PGP)
PGP key 7CBD1DA5 - https://emailselfdefense.fsf.org/
_______________________________________________
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".
