From: Costin Popescu <Costin.Popescu@enea.com>
To: 金健 <jin_j@neusoft.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] Can't make RSS work with 82598EB NIC
Date: Mon, 14 Oct 2013 06:37:10 +0000 [thread overview]
Message-ID: <88A25766747250469E73AEB2F1B9D6442882B857@sestoex09.enea.se> (raw)
In-Reply-To: <5258A202.2030103@neusoft.com>
I have changed the application to use multiple queues (one rx queue per core and per port).
I have also changed the port configuration by making .rxmode.mq_mode = ETH_MQ_RX_RSS and .rx_adv_conf.rss_conf.rss_hf = ETH_RSS_IPV4.
Furthermore my modified application works on the development machine but not on the test machine.
I was thinking that is a matter of either disabled or not supported feature on that hardware.
Costin
-----Original Message-----
From: 金健 [mailto:jin_j@neusoft.com]
Sent: 12 octombrie 2013 04:13
To: Costin Popescu
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] Can't make RSS work with 82598EB NIC
DPDK1.3.1 or newer obsoletes the l2fwd-vf example. Because l2fwd also works well on virtual function platform.
So l2fwd use only one rx queue for each port which attach to one core.
Then RSS will not work for only one rx queue.
On 2013年10月11日 20:48, Costin Popescu wrote:
> Hi everyone!
>
> I have modified the l2fwd example application provided with dpdk in order to benefit from RSS. I have a development machine with an Intel I350 NIC (1Gb) where RSS seem to work (traffic forwarded on all cores).
> On the test machine there is a 82598EB based network card and here the RSS doesn't work, only core 0 is forwarding traffic.
> Looking into ixgbe_dev_mq_rx_configure() function from dpdk lib I can see that for this type of network controller the configuring function exits immediately.
> i.e. if (hw->mac.type == ixgbe_mac_82598EB) return 0; However the
> datasheet claims that RSS is supported.
> The test machine is an Intel Xeon E5420 @ 2.50GHz.
>
> Does anyone have an idea about why RSS would not work in the 82598EB case?
>
> Thanks,
> Costin
>
>
---------------------------------------------------------------------------------------------------
Confidentiality Notice: The information contained in this e-mail and any accompanying attachment(s) is intended only for the use of the intended recipient and may be confidential and/or privileged of Neusoft Corporation, its subsidiaries and/or its affiliates. If any reader of this communication is not the intended recipient, unauthorized use, forwarding, printing, storing, disclosure or copying is strictly prohibited, and may be unlawful.If you have received this communication in error,please immediately notify the sender by return e-mail, and delete the original message and all copies from your system. Thank you.
---------------------------------------------------------------------------------------------------
prev parent reply other threads:[~2013-10-14 6:36 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-11 12:48 Costin Popescu
2013-10-12 1:12 ` 金健
2013-10-14 6:37 ` Costin Popescu [this message]
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=88A25766747250469E73AEB2F1B9D6442882B857@sestoex09.enea.se \
--to=costin.popescu@enea.com \
--cc=dev@dpdk.org \
--cc=jin_j@neusoft.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).