netdev - Re: [PATCH net] e1000e: Change wthresh to 1 to avoid possible Tx stalls.

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: <5074A9CA.50308@tomt.net>
Date:	Wed, 10 Oct 2012 00:48:42 +0200
From:	Andre Tomt <andre@...t.net>
To:	Jesse Brandeburg <jesse.brandeburg@...el.com>
CC:	Eric Dumazet <eric.dumazet@...il.com>,
	Frank Reppin <frank@...ermydesk.org>, netdev@...r.kernel.org,
	Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
	e1000-devel@...ts.sourceforge.net
Subject: Re: [PATCH net] e1000e: Change wthresh to 1 to avoid possible Tx
 stalls.

On 09. okt. 2012 19:36, Jesse Brandeburg wrote:
> I'm not sure what went wrong internally here that this hasn't been
> fixed, and I'm personally embarrassed.  I am working on it until I have
> a patch/solution.
>
> currently am trying to reproduce the issue, am in some weird how to
> use BQL limbo, the lack of documentation on user usage of BQL is slowing
> me down.
>
> Hints or clues (I'm trying to follow the repro steps mentioned in
> some related threads) are appreciated.

I found it simplest to reproduce when doing forwarding, and *not* 
saturating the interface doing the TX. 100Mbps forwarding on gigabit 
link triggered it in seconds. Doing gigabit forwarding speeds (~980Mbps) 
did not trigger anything.

Setup looked somewhat like this, GE beeing gigabit link, FE 100Mbps;

reciever PC (iperf -s)
     |
     GE
     |
    eth0   <- TX lockups
router with 2*e1000e
    eth1
     |
     GE
     |
switch
     |
     FE
     |
source PC (iperf -c recieverPC)

I don't recall all the details anymore, but I'm fairly certain I didnt 
use any non-default qdiscs to reproduce - eg just pfifo_fast (usually 
doing fq_codel though).

For the bug to manifest itself was somewhat dependent on GRO and TSO in 
kernel 3.5, but with 3.6 it didnt matter anymore (at least the rc's).
--
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