netdev - Re: e1000e max frame calculation

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: <alpine.NEB.2.11.1504081019220.3388@chris.i8u.org>
Date:	Wed, 8 Apr 2015 10:25:45 -0700 (PDT)
From:	Hisashi T Fujinaka <htodd@...fifty.com>
To:	Alexander Duyck <alexander.duyck@...il.com>
cc:	Alexander Duyck <alexander.h.duyck@...hat.com>,
	Michael Cronenworth <mike@...tml.com>, netdev@...r.kernel.org
Subject: Re: e1000e max frame calculation

On Wed, 8 Apr 2015, Alexander Duyck wrote:

> On 04/06/2015 08:47 PM, Hisashi T Fujinaka wrote:
>> On Mon, 6 Apr 2015, Alexander Duyck wrote:
>>
>>> The e1000e pch parts seem very limited in terms of their jumbo frames
>>> support.  It looks like the first pch part only supported 4K, and in
>>> the case of the later generations I only see support for a 9018 jumbo
>>> frame listed in the datasheet.  I suspect this is why that is set as
>>> the limit in the driver. The question at this point is if 9018 is the
>>> limit, why is it the limit and is there any wiggle room to it.
>>
>> It's a limit that was told to us by the HW group for the 82579.
>
> The question I have is if the limit of 9018 includes the VLAN header or
> not.  The reason why I ask is that prior to 3.15 you could configure an
> MTU of 9000, and then still enable VLANs which would push the upper size
> to a theoretical limit of 9022.
>
> For example, what is the upper limit on the Windows driver for this
> part?  Does it not support a jumbo frame size of 9014 (Excluding the CRC
> and VLAN header)?  If so we could probably increase the value to 9022 in
> these drivers, not suffer any ill effects, and resolve the issue that
> Michael has been having after the change to include the VLAN_HLEN on the
> max frame size calculation.

I've been searching the documentation for the max MTU. This is something
coming from a completely different division and I haven't seen it
written down anywhere. In fact, I can't find much documentation for
client networking parts at all. It doesn't mean that it doesn't exist,
it just means that I'm very unfamiliar with the documentation and
processes from the Intel client groups.

I'll keep trying to track it down, but David Ertman (the old e1000e
driver maintainer) was very clear that he was told by the hardware
engineers not to exceed 9018.
--
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