linux-kernel - Re: [PATCH net-next] net: phy: marvell: Fix 88e1510 downshift counter errata

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <566608c6-2816-4110-bdd3-cea384bb6e5b@altera.com>
Date: Wed, 3 Sep 2025 09:37:35 +0530
From: "G Thomas, Rohan" <rohan.g.thomas@...era.com>
To: Andrew Lunn <andrew@...n.ch>
Cc: "Russell King (Oracle)" <linux@...linux.org.uk>,
 Heiner Kallweit <hkallweit1@...il.com>, "David S. Miller"
 <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>,
 Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
 netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
 Matthew Gerlach <matthew.gerlach@...era.com>
Subject: Re: [PATCH net-next] net: phy: marvell: Fix 88e1510 downshift counter
 errata

Hi Andrew,

Thanks for the review comments.

On 9/2/2025 8:38 PM, Andrew Lunn wrote:
>>> Also, what is a "link power down/up" ? Are you referring to setting
>>> BMCR_PDOWN and then clearing it? (please update the commit description
>>> and repost after 24 hours, thanks.)
>>>
>>
>> Yes, I'm referring to setting and the clearing BMCR_PDOWN. Will update
>> the commit description in the next version.
> 
> So it could be the bootloader left the PHY in powered down state, when
> booting.
> 

It is not just during initial boot, but whenever the PHY is resumed by
clearing BMCR_PDOWN and goes for autoneg it is recommended to clear the
downshift counter by disabling and re-enabling the downshift feature.

> There is some not so nice logic in phy_attach_direct() which calls
> phy_resume() when the MAC attaches to the PHY. So this case is covered
> by that. But it would be good to comment in the commit message, and
> maybe the resume function, that you are relying on this behaviour.
> 

For m88e1510_resume(), description is added. Will update the commit
message and function description in the next version.

> I do wounder if it should be more explicit, m88e1510_config_init()
> also calls the workaround? You would then not need the comment.
> 
> 	Andrew

Best Regards,
Rohan