netdev - Re: tcp: lost_cnt_hint is -1 due to tcp_shifted

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] [day] [month] [year] [list]

Message-ID: <CADVnQymZ-KDsvCW=G4V+GO2yAuU4_5oaZiHLno-ywv3+S0536g@mail.gmail.com>
Date:	Sun, 12 Feb 2012 22:47:32 -0500
From:	Neal Cardwell <ncardwell@...gle.com>
To:	Vijay Subramanian <subramanian.vijay@...il.com>
Cc:	netdev <netdev@...r.kernel.org>, Yuchung Cheng <ycheng@...gle.com>
Subject: Re: tcp: lost_cnt_hint is -1 due to tcp_shifted_skb() processing

On Sun, Feb 12, 2012 at 8:13 PM, Vijay Subramanian
<subramanian.vijay@...il.com> wrote:
> Hi,
>
> With the current net-next, I noticed tp->lost_cnt_hint is -1 in some cases which
> looks like a bug.

Hi Vijay,

I recently noticed the same issue, and our team has been working on a
fix as well. I think your analysis is mostly correct, except that the
suggested fix of moving the call to tcp_sacktag_one() earlier in
tcp_shifted_skb() also will not work, since the end sequence number of
the skb is not the correct value to use in tcp_sacktag_one() (in
addition to the beginning sequence number being incorrect, as you
noted). Tongith I'll send out the patch that we have been working
with.

neal
--
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