Numerous misuses of Packet::uniqueify about fastclick HOT 2 CLOSED

On Oct 31, 2018, at 1:24 AM, Tom Barbette ***@***.***> wrote: I the freeing behavior is actually true in FastClick. A change (quite undocumented but discussed recently in #101), in FastClick is that we have a pool of packet, and a pool of packet with buffer (while the second in vanilla is a pool of buffer only). As a result the code of uniqueify has a single allocation that do not depend on the number of references. Later, after the packet+buffer allocation, if we actually did not need the new packet because the use count of original was 1, the new packet is freed. But that being said, even in vanilla I see a return 0 if either the packet alloc is 0 or the buffer alloc is 0 ? However, there is still the problem of all code paths not checking if return of uniqueify is null. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.