From: "Neftin, Sasha" <sasha.neftin@intel.com>
To: Paul Menzel <pmenzel@molgen.mpg.de>
Cc: Dima Ruinskiy <dima.ruinskiy@intel.com>,
intel-wired-lan@lists.osuosl.org, "Avivi,
Amir" <amir.avivi@intel.com>
Subject: Re: [Intel-wired-lan] I218-LM uses > 1 W with low traffic
Date: Mon, 23 Jan 2023 09:15:08 +0200 [thread overview]
Message-ID: <31d7cee2-86cf-9a57-18bf-db57a9282d15@intel.com> (raw)
In-Reply-To: <53168857-3dd4-17fe-67cb-15ab0c005ca5@molgen.mpg.de>
On 1/19/2023 16:15, Paul Menzel wrote:
> Dear Sasha,
>
>
> Am 14.04.22 um 09:07 schrieb Neftin, Sasha:
>> On 4/7/2022 18:58, Paul Menzel wrote:
>
>>> Am 07.04.22 um 13:03 schrieb Neftin, Sasha:
>>>> On 4/6/2022 13:09, Paul Menzel wrote:
>>>
>>>>> On a Dell Latitude E7250 with Debian sid/unstable, PowerTOP 2.14
>>>>> reports the network device uses over 1.5 Watt, which is almost ten
>>>>> percent of the whole system.
>>>>>
>>>>> 1.62 W 435,9 pkts/s Device
>>>>> Netzwerkschnittstelle: eno1 (e1000e)
>>>>>
>>>>> $ lspci -nn | grep Ether
>>>>> 00:19.0 Ethernet controller [0200]: Intel Corporation Ethernet
>>>>> Connection (3) I218-LM [8086:15a2] (rev 03)
>>>>> $ uname -a
>>>>> Linux ersatz 5.17.0-trunk-amd64 #1 SMP PREEMPT Debian
>>>>> 5.17.1-1~exp1 (2022-03-29) x86_64 GNU/Linux
>>>>>
>>>>> From atop:
>>>>>
>>>>> NET | eno1 0% | pcki 2421 | pcko 2194 | sp 1000
>>>>> Mbps | si 1501 Kbps | so 140 Kbps | erri 0 | erro
>>>>> 0 | drpo 0 |
>>>>>
>>>>> So I receive 1.5 Mbps, and send 0.14 Mbps.
>>>>>
>>>>> Is such high energy usage expected of the network hardware?
>>>
>>>> It is not the power consumption we expected.How do you measure it?
>>>> (as I know powertop does not show it)
>>>
>>> On my system, PowerTOP shows it. Though maybe it’s a bug in
>>> PowerTOP’s power usage estimation algorithm. No idea. At least with
>>> no network cable connected, it shows zero Watts being used.
>>>
>>> 0 mW 0,0 pkts/s Netzwerkschnittstelle: eno1 (e1000e)
>>>
>>>
>>>> We will try to find out about power consumption.
>>>
>>> Thank you very much.
>>
>> The power consumption during normal operation mode expected to be ~600mW.
>
> With Linux 6.1.4 I still see high power usage:
>
> 4.01 W 26,0 pkts/s Device
> Netzwerkschnittstelle: eno1 (e1000e)
>
> Disconnecting the cable, it still shows high usage:
>
> 5.34 W 0,0 pkts/s Device
> Netzwerkschnittstelle: eno1 (e1000e)
>
> Were you able to reproduce it on your test systems?
I didn't reproduce.
I do not know if we can trust the powertop tool. (LAN power consumption
could be measured via a resistor on the controller connection lanes.)
Did you contact your vendors regards this problem?
>
>
> Kind regards,
>
> Paul
_______________________________________________
Intel-wired-lan mailing list
Intel-wired-lan@osuosl.org
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
next prev parent reply other threads:[~2023-01-23 7:15 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-06 10:09 [Intel-wired-lan] I218-LM uses > 1 W with low traffic Paul Menzel
2022-04-07 11:03 ` Neftin, Sasha
2022-04-07 15:58 ` Paul Menzel
2022-04-14 7:07 ` Neftin, Sasha
2023-01-19 14:15 ` Paul Menzel
2023-01-23 7:15 ` Neftin, Sasha [this message]
2023-01-23 8:38 ` Paul Menzel
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=31d7cee2-86cf-9a57-18bf-db57a9282d15@intel.com \
--to=sasha.neftin@intel.com \
--cc=amir.avivi@intel.com \
--cc=dima.ruinskiy@intel.com \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=pmenzel@molgen.mpg.de \
/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