netdev - Re: Recurring trace from tcp

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: <CADVnQy=rwKG_ddduixcee1D4HaaSBBk=X=2Htmdow0k6Cbxhzg@mail.gmail.com>
Date:	Sat, 30 May 2015 13:29:14 -0400
From:	Neal Cardwell <ncardwell@...gle.com>
To:	Grant Zhang <gzhang@...tly.com>
Cc:	Netdev <netdev@...r.kernel.org>, Yuchung Cheng <ycheng@...gle.com>,
	Eric Dumazet <edumazet@...gle.com>
Subject: Re: Recurring trace from tcp_fragment()

On Fri, May 29, 2015 at 3:53 PM, Grant Zhang <gzhang@...tly.com> wrote:
> Hi Neal,
>
> I will be more happy to test the patch. Please send it my way.

Great. Thank you so much for being willing to do this. Attached is a
patch for testing. I generated it and tested it relative to Linux
v3.14.39, since your stack trace seemed to suggest that you were
seeing this on some variant of v3.14.39. (Newer kernels would need a
slightly different patch, since the reneging code path has changed a
little since 3.14.)

Can you please try it out and see if it makes that warning go away?

Also, I would be interested in seeing the value of your
TcpExtTCPSACKReneging counter, and some sense of how fast that value
is increasing, on a machine that's seeing this issue:
  nstat -z -a | grep Reneg

Thanks!

neal

Download attachment "0001-RFC-for-tests-on-v3.14.39-tcp-resegment-skbs-that-we.patch" of type "application/octet-stream" (4569 bytes)