On 20/09/13 14:23, Derek Buitenhuis wrote:
> On 9/19/2013 11:23 PM, Luca Barbato wrote:
>> Any assert that you might trigger is a DOS waiting for you...
> 
> How exactly can checking the value of what looks to be a
> define be a DoS?

As I said, "you might trigger".

Actually that "assert" should be a compile check to error out if you got
that macro evaluate to non-1 for whatever reason (the macro seems to be
either 1 or -1). (We have already some construct such as
foo[value-expected_value] that makes the compile error out iirc)

The current codebase has plenty of triggerable asserts and removing them
isn't so simple, I'd rather not add more lines.

lu
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to