[FFmpeg-devel] [Patch] fix ffprobe crash #3603

[anshul]

On 05/07/2014 05:01 PM, anshul wrote:
> On 05/07/2014 01:31 PM, Clément Boesch wrote:
>> On Wed, May 07, 2014 at 01:28:40PM +0530, anshul wrote:
>>> On 05/07/2014 12:48 PM, anshul wrote:
>>>> On 05/07/2014 11:56 AM, anshul wrote:
>>>>> ffprobe crashes since number of stream increased in the attched video
>>>>> at #3603 from 3 to 6 which were not taken in account in commit
>>>>> 29b9aee4.
>>>>>
>>>>> so this line need extra guard
>>>>> nb_streams_frames[pkt.stream_index]++;
>>>>>
>>>>> Thanks
>>>>> Anshul
>>>> Ignore this patch it have lot of flaws, will send another soon
>>>> -Anshul
>>> I have attached new patch for same problem
>>>
>>> -Anshul
>>>  From 08eaf8857ffab170baa5079ade50c5d2dafb4b1c Mon Sep 17 00:00:00 2001
>>> From: Anshul Maheshwari<er.anshul.maheshwari at gmail.com>
>>> Date: Wed, 7 May 2014 13:27:43 +0530
>>> Subject: [PATCH] Fix #3603 crashes in ffprobe
>>>
>> mark the fix in the description, and describe the fix in the title
>>
>>> ---
>>>   ffprobe.c | 15 +++++++++++++--
>>>   1 file changed, 13 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/ffprobe.c b/ffprobe.c
>>> index c6e0469..11a946c 100644
>>> --- a/ffprobe.c
>>> +++ b/ffprobe.c
>>> @@ -191,6 +191,7 @@ static const char unit_hertz_str[]          = "Hz"   ;
>>>   static const char unit_byte_str[]           = "byte" ;
>>>   static const char unit_bit_per_second_str[] = "bit/s";
>>>   
>>> +static int nb_streams;
>>>   static uint64_t *nb_streams_packets;
>>>   static uint64_t *nb_streams_frames;
>>>   static int *selected_streams;
>>> @@ -1920,7 +1921,17 @@ static int read_interval_packets(WriterContext *w, AVFormatContext *fmt_ctx,
>>>               if (do_read_packets) {
>>>                   if (do_show_packets)
>>>                       show_packet(w, fmt_ctx, &pkt, i++);
>>> -                nb_streams_packets[pkt.stream_index]++;
>>> +                if(nb_streams_packets)
>>> +                {
>>> +                    if(pkt.stream_index >= nb_streams)
>>> +                    {
>> please use a consistent style
>>
>>> +                        nb_streams_frames  = av_realloc(nb_streams_frames,fmt_ctx->nb_streams* sizeof(*nb_streams_frames));
>>> +                        nb_streams_packets = av_realloc(nb_streams_packets,fmt_ctx->nb_streams* sizeof(*nb_streams_packets));
>>> +                        selected_streams   = av_realloc(selected_streams,fmt_ctx->nb_streams* sizeof(*selected_streams));
>> av_realloc() is not compatible with av_calloc()
>>
>>> +                        nb_streams = fmt_ctx->nb_streams;
>> trailing whitespace
>>
>>> +                    }
>>> +                    nb_streams_packets[pkt.stream_index]++;
>>> +                }
>>>               }
>>>               if (do_read_frames) {
>>>                   pkt1 = pkt;
>>> @@ -2373,7 +2384,7 @@ static int probe_file(WriterContext *wctx, const char *filename)
>>>           return ret;
>>>   
>>>   #define CHECK_END if (ret < 0) goto end
>>> -
>>> +    nb_streams = fmt_ctx->nb_streams;
>>>       nb_streams_frames  = av_calloc(fmt_ctx->nb_streams, sizeof(*nb_streams_frames));
>>>       nb_streams_packets = av_calloc(fmt_ctx->nb_streams, sizeof(*nb_streams_packets));
>> so use it here too
>>
>>>       selected_streams   = av_calloc(fmt_ctx->nb_streams, sizeof(*selected_streams));
>>> -- 
>>> 1.8.1.4
>>>
>>> _______________________________________________
>>> ffmpeg-devel mailing list
>>> ffmpeg-devel at ffmpeg.org
>>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>>
>> _______________________________________________
>> ffmpeg-devel mailing list
>> ffmpeg-devel at ffmpeg.org
>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> I have attached new patch with some more indentation.
>
>
> Thanks
> Anshul
my previous patch was not considering  nb_streams_frames and 
selected_streams if number of stream increased after avformat_open_input,
this patch consider all three things.

Thanks
Anshul