netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Otto Sabart <osabart@redhat.com>
To: Alexander Duyck <alexander.duyck@gmail.com>
Cc: Netdev <netdev@vger.kernel.org>,
	Jirka Hladky <jhladky@redhat.com>,
	Adam Okuliar <aokuliar@redhat.com>
Subject: Re: ixgbe: cannot enable LRO
Date: Fri, 6 May 2016 12:18:58 +0200	[thread overview]
Message-ID: <20160506101858.GC4106@redhat.com> (raw)
In-Reply-To: <CAKgT0UceTxomp+Zf5ks_B0bFomdx2jyhWLwSXh4716voJ5YFAg@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 1958 bytes --]

> On Wed, Apr 27, 2016 at 2:36 AM, Otto Sabart <osabart@redhat.com> wrote:
> >
> > Hello everyone,
> > does anybody have a problem with LRO on ixge (on latest 4.6-rc5)?
> > I cannot find a way to enable it.
> >
> > On stable RHEL7.2 kernel everything works fine.
> >
> > I opened a bug report [0].
> >
> > [0] https://bugzilla.kernel.org/show_bug.cgi?id=117291
> >
> >
> > Thanks!
> >
> > Ota
> 

Hello Alex,

> So I am able to turn on LRO without any issues.

Yes, I badly desciribed the problem. The LRO was not possible to turn on
immediately _after the boot_ (I was enabling it in /etc/rc.local).

When I reloaded the ixgbe driver, the LRO was possible to turn on
without problem.

I found out that the problem was caused by network manager. When I
disabled NM, the LRO started to work.

> 
> Do you know if you have done anything that might disable LRO such as
> modified the rx-usecs to a value less than 10 or enabled routing or
> bridging on the device?  Also I think a stacked device might be able
> to block you from enabling LRO unless all the devices stacked on the
> interface can support it.

I did not modify rx-usecs (at least not intentionally).  Its value is
always (with disabled or enabled LRO) equal to 1 (with disabled NM).

$ ethtool -c ixgbe | grep rx-usecs
rx-usecs: 1
rx-usecs-irq: 0
rx-usecs-low: 0
rx-usecs-high: 0

There is no linux bridge or routing enabled on this device. There is
only a VLAN configured. Does it matter?

$ ip l show ixgbe
9: ixgbe: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT qlen 1000
    link/ether 00:1b:21:90:c3:86 brd ff:ff:ff:ff:ff:ff

$ ip l show ixgbe.40
19: ixgbe.40@ixgbe: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT qlen 1000
    link/ether 00:1b:21:90:c3:86 brd ff:ff:ff:ff:ff:ff


I updated the bugzilla [0] and I think we can close this as a NOTABUG.

Thanks!

Ota

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 520 bytes --]

      reply	other threads:[~2016-05-06 10:19 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-27  9:36 ixgbe: cannot enable LRO Otto Sabart
2016-04-27 15:37 ` Alexander Duyck
2016-05-06 10:18   ` Otto Sabart [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=20160506101858.GC4106@redhat.com \
    --to=osabart@redhat.com \
    --cc=alexander.duyck@gmail.com \
    --cc=aokuliar@redhat.com \
    --cc=jhladky@redhat.com \
    --cc=netdev@vger.kernel.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).