DPDK patches and discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: "Morten Brørup" <mb@smartsharesystems.com>
Cc: "Bly, Mike" <mbly@ciena.com>, <dev@dpdk.org>
Subject: Re: e1000 forced 1G support?
Date: Fri, 11 Feb 2022 06:44:41 -0800	[thread overview]
Message-ID: <20220211064441.7cd12699@hermes.local> (raw)
In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35D86E87@smartserver.smartshare.dk>

On Fri, 11 Feb 2022 09:57:31 +0100
Morten Brørup <mb@smartsharesystems.com> wrote:

> > From: Bly, Mike [mailto:mbly@ciena.com] 
> > Sent: Friday, 11 February 2022 02.30
> > 
> > Hello,
> >
> > This is in regards to the DPDK E1000 driver used for the i350 [8086:1521] NIC.
> >
> > I am looking to see if we can get forced speed == 1000Mb (1Gb) support working on this NIC. The current DPDK driver does not appear to have support for forcing the NIC to 1G (1000M) speed. It only supports setting 100M and 10M. Is there a reason for this? Refer to: e1000_phy_force_speed_duplex_setup() in drivers/net/e1000/base/e1000_phy.c.
> >
> > Based on my reading of ethernet-controller-i350-datasheet.pdf it would seem we should be able to force the speed to 1G. However, even after "updating" the above mentioned function to try and support a 1G forced speed, the only way we can get two of these NICs to link up to each other at 1G, is to set the port to auto-neg. We can certainly force speed on one link to 100M or 10M and the other NIC will link up, but no luck for 1G, regardless of whether we have one or both sides in a forced speed vs. auto-neg mode.
> >
> > Is there a limitation I missed in the PDF perhaps?  
> 
> Hi Mike,
> 
> Auto-negotiation is a *requirement* for 1 Gbps according to the IEEE 802.3 Ethernet standard, so the way to force 1 Gbps is: Enable auto-negotiation and only advertise 1 Gbps.
> 
> In other words: You cannot establish a 1 Gbps link without auto-neg.
> 
> 
> Med venlig hilsen / Kind regards,
> -Morten Brørup
> 

Also, check your cable. 1G requires all 4 wire pairs to be connected. But 100M can get by with 2 pairs

  reply	other threads:[~2022-02-11 14:44 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-11  1:30 Bly, Mike
2022-02-11  8:57 ` Morten Brørup
2022-02-11 14:44   ` Stephen Hemminger [this message]
2022-02-11 17:23     ` [**EXTERNAL**] " Bly, Mike

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=20220211064441.7cd12699@hermes.local \
    --to=stephen@networkplumber.org \
    --cc=dev@dpdk.org \
    --cc=mb@smartsharesystems.com \
    --cc=mbly@ciena.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).