DPDK usage discussions
 help / color / mirror / Atom feed
From: Alex Kiselev <kiselev99@gmail.com>
To: users <users@dpdk.org>, declan.doherty@intel.com
Subject: [dpdk-users] TX descriptor is not done, bonding driver LACP mode issues
Date: Fri, 14 Jul 2017 11:23:51 +0300	[thread overview]
Message-ID: <CAMKNYbz2FK_XZyFj4JQsU4u+FLbgEgJOLGH2VCVunAXgGnBj5w@mail.gmail.com> (raw)

Hi.

I will try to describe my problem shortly again so you don't need to
start reading the first posts in my thread "bonding driver LACP mode
issues".

I am having problems with bonding driver. The problems appeared right
after I switched from using two i40e ports to a single bonding port
build on top of the two i40e ports. Before changing the configuration
my application had been working for weeks without any problems.

I found out that after some time of normal operations tx queue 0 stop
transmitting with the error "TX descriptor is not done"

I google a little bit about "TX descriptor is not done" error and
found a discussion "NIC Stops Transmitting" where a guy was having the
same issues that I do.
http://dpdk.org/ml/archives/dev/2013-July/000357.html

One of the proposals to fix that problem was to check
pthresh/hthresh/wthresh values.

So, I start checking and found out that rte_eth_dev_info_get return 0
for for pthresh/hthresh/wthresh field, but recommended/default values
of this fields from i40e driver are completly different.

So far, it looks like this is what causing the bond port to stop
transmitting packets.

Is this a bug or a feature?

--
Alex Kiselev

                 reply	other threads:[~2017-07-14  8:23 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=CAMKNYbz2FK_XZyFj4JQsU4u+FLbgEgJOLGH2VCVunAXgGnBj5w@mail.gmail.com \
    --to=kiselev99@gmail.com \
    --cc=declan.doherty@intel.com \
    --cc=users@dpdk.org \
    /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).