Bug#565404: linux-image-2.6.26-2-amd64: atl1e: TSO is broken
- To: Eric Dumazet <eric.dumazet@gmail.com>
- Cc: "Huang, Xiong" <xiong@qca.qualcomm.com>, Ben Hutchings <ben@decadent.org.uk>, Anders Boström <anders@netinsight.net>, "netdev@vger.kernel.org" <netdev@vger.kernel.org>, "565404@bugs.debian.org" <565404@bugs.debian.org>
- Subject: Bug#565404: linux-image-2.6.26-2-amd64: atl1e: TSO is broken
- From: Hannes Frederic Sowa <hannes@stressinduktion.org>
- Date: Wed, 3 Apr 2013 00:15:20 +0200
- Message-id: <[🔎] 20130402221520.GF4924@order.stressinduktion.org>
- Mail-followup-to: Eric Dumazet <eric.dumazet@gmail.com>, "Huang, Xiong" <xiong@qca.qualcomm.com>, Ben Hutchings <ben@decadent.org.uk>, Anders Boström <anders@netinsight.net>, "netdev@vger.kernel.org" <netdev@vger.kernel.org>, "565404@bugs.debian.org" <565404@bugs.debian.org>
- Reply-to: Hannes Frederic Sowa <hannes@stressinduktion.org>, 565404@bugs.debian.org
- In-reply-to: <[🔎] 1364940038.5113.187.camel@edumazet-glaptop>
- References: <CDAFEDABF718A54BABD0DA5476695307475CCD42@SHEXMB-01.global.atheros.com> <20100125.163612.582489483209420341.anders@netinsight.net> <CDAFEDABF718A54BABD0DA5476695307475CCFF8@SHEXMB-01.global.atheros.com> <20100126.093439.651115405791675606.anders@netinsight.net> <1364689558.3557.22.camel@deadeye.wl.decadent.org.uk> <157393863283F442885425D2C45428564F202477@nasanexd02f.na.qualcomm.com> <1364695805.3557.41.camel@deadeye.wl.decadent.org.uk> <[🔎] 157393863283F442885425D2C45428564F20261B@nasanexd02f.na.qualcomm.com> <[🔎] 20130402211524.GE4924@order.stressinduktion.org> <[🔎] 1364940038.5113.187.camel@edumazet-glaptop>
On Tue, Apr 02, 2013 at 03:00:38PM -0700, Eric Dumazet wrote:
> On Tue, 2013-04-02 at 23:15 +0200, Hannes Frederic Sowa wrote:
>
> > The error vanishes as soon as I put a gso size limit of MAX_TX_BUF_LEN
> > in the driver. MAX_TX_BUF_LEN seems to be arbitrary set to 0x2000. I
> > can even raise it to 0x3000 and don't see any tcp retransmits. Do you
> > have an advice on how to size this value (e.g. should we switch to the
> > windows values)?
>
> This looks like an overflow error...
Thanks for your input, Eric.
I am limited in my time to work on this today but nontheless just tested
your patch without any of my changes and count a lot of TcpRetransSegs
again. Either there is really some hardware limitation or another
overflow.
Reply to: