netdev - Re: r8169 : always copying the rx buffer to new skb

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: <20110418182100.GA18548@electric-eye.fr.zoreil.com>
Date:	Mon, 18 Apr 2011 20:21:00 +0200
From:	Francois Romieu <romieu@...zoreil.com>
To:	John Lumby <johnlumby@...mail.com>
Cc:	netdev@...r.kernel.org, nic_swsd@...ltek.com
Subject: Re: r8169 :  always copying the rx buffer to new skb

John Lumby <johnlumby@...mail.com> :
> Summary  -  current r8169 always memcpy's every received buffer to new skb,
>             and I'd like to propose that it should not,
>             which can improve throughput / reduce CPU utilization by
> around 10%

Disclaimer: I'll read your suggestion more carefully when I am back home.
I am late and I could be off-topic.

> I think (not sure) this was related to a bug

Short answer: it's mostly related to CVE-2009-4537 (see git log).

I may resurrect some alternate fix - i.e. detect corrupted Tx descriptors
and reset before things gets wrong - but it is not easy to prove it right
since it may be necessary to tailor it for each member of the 816x / 810x
family. Some input from Realtek may help though.

-- 
Ueimor
--
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