[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d2f49dc7-7218-448c-aea3-d695445c80c7@intel.com>
Date: Mon, 13 May 2024 09:04:57 +0300
From: Sasha Neftin <sasha.neftin@...el.com>
To: Andrew Lunn <andrew@...n.ch>
CC: "Ruinskiy, Dima" <dima.ruinskiy@...el.com>, Ricky Wu
<en-wei.wu@...onical.com>, <netdev@...r.kernel.org>, <rickywu0421@...il.com>,
<linux-kernel@...r.kernel.org>, <edumazet@...gle.com>,
<intel-wired-lan@...ts.osuosl.org>, <kuba@...nel.org>,
<anthony.l.nguyen@...el.com>, <pabeni@...hat.com>, <davem@...emloft.net>,
"Lifshits, Vitaly" <vitaly.lifshits@...el.com>, naamax.meir
<naamax.meir@...ux.intel.com>, "Avivi, Amir" <amir.avivi@...el.com>, "Keller,
Jacob E" <jacob.e.keller@...el.com>, "Lifshits, Vitaly"
<vitaly.lifshits@...el.com>
Subject: Re: [Intel-wired-lan] [PATCH v2 2/2] e1000e: fix link fluctuations
problem
On 10/05/2024 15:32, Andrew Lunn wrote:
>>> It would be interesting to see what the link partner sees. What does
>>> it think the I219-LM is advertising? Is it advertising 1000BaseT_Half?
>>
>> i219 parts come with LSI PHY. 1000BASE-T half-duplex is not supported.
>> 1000BASET half-duplex not advertised in IEEE 1000BASE-T Control Register 9.
>
> That is the theory. But in practice? What does the link partner really
> see? I've come across systems which get advertisement wrong. However,
> in that case, i suspect it is the software above the PHY, not the PHY
> itself which was wrong.
>
> Andrew
Not only in the theory. On the link partner IEEE 1000BASE-T Status
Register - Address 10, Link Partner 1000BASE-T Half-Duplex Capability is
0. (1000BAS-T HD not advertised). I believe here is some false positive
indication during the auto-negotiation process.(this is not related to
the link fluctuation)
Sasha
Powered by blists - more mailing lists