[FFmpeg-devel] [PATCH] avfilter/af_adelay: make per channel delay argument an int64_t

[James Almer]

On 4/23/2021 10:09 AM, Gyan Doshi wrote:
> 
> 
> On 2021-04-23 18:22, James Almer wrote:
>> Should fix ticket #9196
>>
>> Signed-off-by: James Almer <jamrial at gmail.com>
>> ---
>>   libavfilter/af_adelay.c | 11 +++++++----
>>   1 file changed, 7 insertions(+), 4 deletions(-)
>>
>> diff --git a/libavfilter/af_adelay.c b/libavfilter/af_adelay.c
>> index 6ac81c2a3e..1229bf5bc7 100644
>> --- a/libavfilter/af_adelay.c
>> +++ b/libavfilter/af_adelay.c
>> @@ -28,9 +28,9 @@
>>   #include "internal.h"
>>   typedef struct ChanDelay {
>> -    int delay;
>> -    unsigned delay_index;
>> -    unsigned index;
>> +    int64_t delay;
>> +    size_t delay_index;
>> +    size_t index;
>>       uint8_t *samples;
>>   } ChanDelay;
>> @@ -152,7 +152,7 @@ static int config_input(AVFilterLink *inlink)
>>           p = NULL;
>> -        ret = av_sscanf(arg, "%d%c", &d->delay, &type);
>> +        ret = av_sscanf(arg, "%"SCNd64"%c", &d->delay, &type);
>>           if (ret != 2 || type != 'S') {
>>               div = type == 's' ? 1.0 : 1000.0;
>>               if (av_sscanf(arg, "%f", &delay) != 1) {
>> @@ -194,6 +194,9 @@ static int config_input(AVFilterLink *inlink)
>>           if (!d->delay)
>>               continue;
>> +        if (d->delay > SIZE_MAX)
>> +            return AVERROR(ENOMEM);
> 
> If this can occur due to user input, it should be reported back to the 
> user.

It could even before this patch, as you can see in the existing 
av_malloc_array() check below, but sure, i can print an av_log() error.

> 
> 
>> +
>>           d->samples = av_malloc_array(d->delay, s->block_align);
>>           if (!d->samples)
>>               return AVERROR(ENOMEM);
> 
> Regards,
> Gyan
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".