Re: listen(2) backlog

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>>> It looks to me that in reality the length of the queue of pending
>>> complete connections ends up being backlog _plus one_ before
>>> incoming SYNs are ignored.

>>     https://nxr.netbsd.org/xref/src/sys/kern/uipc_socket2.c#274

> Ah, interesting.  Thanks!  Looks like it's always been that way.

I don't know what the current code is; I thought I had picked up a copy
of a NetBSD 10 source tree for work, but I can't seem to find it now.

But 9.1, 5.2, and even 1.4T all have more or less the same code, and it
does not lead to backlog-plus-one except possibly for one backlog
setting (to be certain which, I'd have to think more than I feel
capable of right now - I'm getting over an illness).  This comes from
the 9.1 source:

	if (head->so_qlen + head->so_q0len > 3 * head->so_qlimit / 2) {

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B