This is the mail archive of the
ecos-discuss@sourceware.org
mailing list for the eCos project.
Re: Socket leak when accept() is aborted by TCP RST?
- From: Grant Edwards <grant dot b dot edwards at gmail dot com>
- To: ecos-discuss at sources dot redhat dot com
- Date: Tue, 27 Jul 2010 22:39:47 +0000 (UTC)
- Subject: [ECOS] Re: Socket leak when accept() is aborted by TCP RST?
- References: <i2nlp3$s1i$1@dough.gmane.org> <4C4F5F3A.1040005@systech.com>
On 2010-07-27, Jay Foster <jay@systech.com> wrote:
> I don't know about the socket leak, per se, but I have had some
> dealings with the original BSD network stack with regards to the
> accept() call. From my memory, I remember there being two queues on
> which the socket might be placed (so_comp, so_incomp). Depending on the
> timing, the socket might be on the incomplete or the completed queue
> when the RST arrives. Perhaps there is a bug in one of these code paths
> when receiving a RST (probably the so_incomp queue).
Thanks for the clue. I've been looking at BSD code for a little
while, and I'm afraid I'm completely lost. Can anybody point me to
any docs on how the stack works? (e.g. what the normal flow is and
where the new socket gets allocated.)
--
Grant Edwards grant.b.edwards Yow! ... the HIGHWAY is
at made out of LIME JELLO and
gmail.com my HONDA is a barbequeued
OYSTER! Yum!
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss