All of lore.kernel.org
 help / color / mirror / Atom feed
From: Klaus Degner <kd@allegro-packets.com>
To: "Lu, Wenzhuo" <wenzhuo.lu@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: jumbo frame support for 82583V
Date: Mon, 20 Jul 2015 11:21:06 +0200	[thread overview]
Message-ID: <55ACBD82.4000200@allegro-packets.com> (raw)
In-Reply-To: <6A0DE07E22DDAD4C9103DF62FEBC0909CFB378@shsmsx102.ccr.corp.intel.com>

Hi Wenzhuo,
> Hi Klaus,
>
>> -----Original Message-----
>> From: Klaus Degner [mailto:kd@allegro-packets.com]
>> Sent: Friday, July 17, 2015 7:56 PM
>> To: Lu, Wenzhuo
>> Cc: dev@dpdk.org
>> Subject: jumbo frame support for 82583V
>>
>> Hi Wenzhuo,
>>
>> We are testing different Intel NICs for DPDK. We have tested the master branch
>> with the support for 82583V Intel chip.
>> It works very well except that we can only use up to 1518 bytes for maximum
>> packet capture.
>> We have debugged this and it is restricted in the initialization:
>>
>> http://dpdk.org/browse/dpdk/tree/drivers/net/e1000/em_ethdev.c#n855
>>
>> Without DPDK, the linux driver support and mtu up to 9k and ark.intel reports
>> that this chip is jumbo frame capable:
>>
>> http://ark.intel.com/de/products/41676/Intel-82583V-Gigabit-Ethernet-
>> Controller
>>
>> Is there any specific reason why DPDK cannot use jumbo frames for this NIC ?
>>
>> Thanks for help !
>>
>> Klaus
> Sorry, I'm not sure about the history.
> According to the datasheet's update log, 82583v doesn't support jumbo frame at first.
> I think the reason is the e1000 code is not updated after the datasheet said the jumbo frame is supported.
> As I don’t have the NIC on hand, would you like to change the code and have a try. I mean just let
> " case e1000_82583" return 9K. Thanks.

The DPDK jumbo frame send and receive works for 82583V.
We have tested it with 4x 85283V and the setup:

Linux send --> DPDK receive --> DPDK send --> Linux receive

We could send a 9k packet via this data path and have verified the
content of it.
I can generate a patch for DPDK if this is enough testing for you.

Thanks,

Klaus

  reply	other threads:[~2015-07-20  9:21 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-17 11:56 jumbo frame support for 82583V Klaus Degner
2015-07-20  3:14 ` Lu, Wenzhuo
2015-07-20  9:21   ` Klaus Degner [this message]
2015-07-21  0:40     ` Lu, Wenzhuo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55ACBD82.4000200@allegro-packets.com \
    --to=kd@allegro-packets.com \
    --cc=dev@dpdk.org \
    --cc=wenzhuo.lu@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.