[FFmpeg-devel] [PATCH 3/3] rawdec: only exempt BIT0 with need_copy from buffer sanity check

Andreas Cadhalpun andreas.cadhalpun at googlemail.com
Sun Dec 20 12:20:14 CET 2015


On 20.12.2015 01:11, Michael Niedermayer wrote:
> On Sat, Dec 19, 2015 at 11:49:14PM +0100, Andreas Cadhalpun wrote:
>> Otherwise the too samll buffer is directly used in the frame, causing
>> segmentation faults, when trying to use the frame.
>>
>> Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun at googlemail.com>
>> ---
>>  libavcodec/rawdec.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/libavcodec/rawdec.c b/libavcodec/rawdec.c
>> index 611178f..50cee39 100644
>> --- a/libavcodec/rawdec.c
>> +++ b/libavcodec/rawdec.c
>> @@ -257,7 +257,7 @@ static int raw_decode(AVCodecContext *avctx, void *data, int *got_frame,
>>          buf += buf_size - context->frame_size;
>>  
>>      len = context->frame_size - (avctx->pix_fmt==AV_PIX_FMT_PAL8 ? AVPALETTE_SIZE : 0);
>> -    if (buf_size < len && (avctx->codec_tag & 0xFFFFFF) != MKTAG('B','I','T', 0)) {
>> +    if (buf_size < len && ((avctx->codec_tag & 0xFFFFFF) != MKTAG('B','I','T', 0) || !need_copy)) {
>>          av_log(avctx, AV_LOG_ERROR, "Invalid buffer size, packet size %d < expected frame_size %d\n", buf_size, len);
>>          av_buffer_unref(&frame->buf[0]);
>>          return AVERROR(EINVAL);
> 
> should be ok,

Pushed.

> alterantively you could also force need_copy in these cases

With valid samples it already uses need_copy, so it seems better to error here.

Best regards,
Andreas



More information about the ffmpeg-devel mailing list