Hi, recently when I experience the new version of bento4. I find an NPD bug in program "mp42aac".
./mp42acc inputs
The bug logic is that when the data size is not large enough and apply reallocation, the reallocation does not check whether the new buffer is successfully allocated.
This is the execution trace.
4c7a.png">
In SetDataSize, the function realloc buffer when new size is larger than the current one.
This means the two values of two size variable are not zero.
In reallocation, there is no null pointer check for the return value of the allocation and leads to the crash when apply AP4_CopyMemory which is an alias of memcpy function.
I have uploaded the report and related bug trace to help understand this problem. report_input.zip
The text was updated successfully, but these errors were encountered:
I look into the related bug and fix #342 , this is the same cause but does not fix the root cause since it only modifies the upper layer of mp42hls but this function is used in many programs and the bug still occurs.
Hi, recently when I experience the new version of bento4. I find an NPD bug in program "mp42aac".
./mp42acc inputsThe bug logic is that when the data size is not large enough and apply reallocation, the reallocation does not check whether the new buffer is successfully allocated.

This is the execution trace.
4c7a.png">
In SetDataSize, the function realloc buffer when new size is larger than the current one.


This means the two values of two size variable are not zero.
In reallocation, there is no null pointer check for the return value of the allocation and leads to the crash when apply AP4_CopyMemory which is an alias of memcpy function.
I have uploaded the report and related bug trace to help understand this problem.
report_input.zip
The text was updated successfully, but these errors were encountered: