netdev - Re: TCP_DEFER_ACCEPT is missing counter update

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Pine.LNX.4.58.0910151045220.2830@u.domain.uli>
Date:	Thu, 15 Oct 2009 10:59:19 +0300 (EEST)
From:	Julian Anastasov <ja@....bg>
To:	David Miller <davem@...emloft.net>
cc:	w@....eu, netdev@...r.kernel.org, eric.dumazet@...il.com
Subject: Re: TCP_DEFER_ACCEPT is missing counter update


	Hello,

On Wed, 14 Oct 2009, David Miller wrote:

> For now I'm pushing Willy's change into Linus's tree.

	Better do not apply this change because it's reasoning is
wrong - counter is increased at the same time when threshold
decreases - what this change does is to half the TCP_DEFER_ACCEPT
period (assuming first ACK comes immediately) with the side
effect to switch off the TCP_DEFER_ACCEPT flag during
SYN-ACK next retransmits. When one uses TCP_DEFER_ACCEPT=1 as
flag the listener will see the new socket on the 2nd ACK which
contradicts with the TCP_DEFER_ACCEPT semantic to wakeup only
when data comes.

> After more discussion we can revert if necessary.
> 
> I won't submit this to -stable until the discussion is fully resolved.
> 
> Thanks!

Regards

--
Julian Anastasov <ja@....bg>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html