Intel-Wired-Lan Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Tony Nguyen <anthony.l.nguyen@intel.com>
To: Paul Menzel <pmenzel@molgen.mpg.de>,
	Przemek Kitszel <przemyslaw.kitszel@intel.com>
Cc: <intel-wired-lan@lists.osuosl.org>, <it+intel-wired-lan@molgen.mpg.de>
Subject: Re: [Intel-wired-lan] ice: E810-XXV: Dell GBIC S28-10G-25G-SR-85C does not work
Date: Tue, 17 Feb 2026 10:16:02 -0800	[thread overview]
Message-ID: <3f6918da-0fcd-4a87-bca2-adbd1d88100b@intel.com> (raw)
In-Reply-To: <b235e4e1-3a1c-4b09-bf4f-bd23d308e3e7@molgen.mpg.de>



On 2/17/2026 9:15 AM, Paul Menzel wrote:

Hi Paul,

I spoke to one of our link people about this.


> It works with Broadcom network controller BCM57414:
> 
>      $ lspci -nn -s c4:00
>      c4:00.0 Ethernet controller [0200]: Broadcom Inc. and subsidiaries 
> BCM57414 NetXtreme-E 10Gb/25Gb RDMA Ethernet Controller [14e4:16d7] (rev 
> 01)
>      c4:00.1 Ethernet controller [0200]: Broadcom Inc. and subsidiaries 
> BCM57414 NetXtreme-E 10Gb/25Gb RDMA Ethernet Controller [14e4:16d7] (rev 
> 01)
> 
> The difference seems to be that the Broadcom device supports auto- 
> negotiation, and the Intel device does not:

Strictly speaking, optical links do not provide auto-negotiation.

> 
> Intel E810-XXV:
> 
>      Supported ports: [ FIBRE ]
>      Supported link modes:   1000baseT/Full
>                              10000baseT/Full
>                              25000baseCR/Full
>                              25000baseSR/Full
>                              1000baseX/Full
>                              10000baseCR/Full
>                              10000baseSR/Full
>                              10000baseLR/Full
>      Supported pause frame use: Symmetric
>      Supports auto-negotiation: No
>      Supported FEC modes: None
>      Advertised link modes:  25000baseSR/Full

The important part is here 10G is not an advertised link mode...

>      Advertised pause frame use: No
>      Advertised auto-negotiation: No
>      Advertised FEC modes: None
>      Speed: Unknown!
>      Duplex: Unknown! (255)
>      Auto-negotiation: off
>      Port: FIBRE
>      PHYAD: 0
>      Transceiver: internal
>      Supports Wake-on: d
>      Wake-on: d
>          Current message level: 0x00000007 (7)
>                                 drv probe link
>      Link detected: no
> 
> Broadcom BCM57414 NetXtreme-E:
> 
>      Supported ports: [ FIBRE ]
>      Supported link modes:   25000baseSR/Full
>                              10000baseSR/Full
>      Supported pause frame use: Symmetric Receive-only
>      Supports auto-negotiation: Yes
>      Supported FEC modes: RS     BASER
>      Advertised link modes:  25000baseSR/Full
>                              10000baseSR/Full

... where it is here.

>      Advertised pause frame use: No
>      Advertised auto-negotiation: Yes
>      Advertised FEC modes: Not reported
>      Speed: Unknown!
>      Duplex: Unknown! (255)
>      Auto-negotiation: on
>      Port: FIBRE
>      PHYAD: 1
>      Transceiver: internal
>      Supports Wake-on: g
>      Wake-on: d
>          Current message level: 0x00002081 (8321)
>                                 drv tx_err hw
>      Link detected: no
> 
> 
> Kind regards,
> 
> Paul
> 
> 
>> PS:
>>
>> ```
>> $ ip link show net04
>> 7: net04: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq 
>> switchid b48351ffff278d44 state DOWN mode DEFAULT group default qlen 1000
>>      link/ether b4:83:51:27:8d:44 brd ff:ff:ff:ff:ff:ff
>>      alias eth4
>> $ sudo ethtool -m net04
>>      Identifier                                : 0x03 (SFP)
>>      Extended identifier                       : 0x04 (GBIC/SFP 
>> defined by 2-wire interface ID)
>>      Connector                                 : 0x07 (LC)
>>      Transceiver codes                         : 0x00 0x00 0x00 0x00 
>> 0x00 0x00 0x00 0x00 0x02

0x10 would be here for advertised 10G support. It is not, which is why 
it's not being advertised. He mentioned it's very common for dual rates 
to claim it on paper but not advertise it properly.

Could you provide the output for 'ethool -m <INT> hex on'?

Thanks,
Tony


  reply	other threads:[~2026-02-17 18:16 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-12 14:06 [Intel-wired-lan] ice: E810-XXV: Dell GBIC S28-10G-25G-SR-85C does not work Paul Menzel
2026-02-17 17:15 ` Paul Menzel
2026-02-17 18:16   ` Tony Nguyen [this message]
2026-02-18 12:09     ` Paul Menzel
2026-02-18 15:00       ` Paul Menzel
2026-03-25  7:02         ` Paul Menzel
2026-03-26 17:52           ` Tony Nguyen

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=3f6918da-0fcd-4a87-bca2-adbd1d88100b@intel.com \
    --to=anthony.l.nguyen@intel.com \
    --cc=intel-wired-lan@lists.osuosl.org \
    --cc=it+intel-wired-lan@molgen.mpg.de \
    --cc=pmenzel@molgen.mpg.de \
    --cc=przemyslaw.kitszel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox