netdev - Re: [PATCH net-next v2] net: phy: dp83826: disable WOL at init

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

Message-ID: <13a68356-235e-4945-8cf7-5b7a42b0bf46@lunn.ch>
Date: Mon, 26 Feb 2024 17:53:33 +0100
From: Andrew Lunn <andrew@...n.ch>
To: "Russell King (Oracle)" <linux@...linux.org.uk>
Cc: Catalin Popescu <catalin.popescu@...ca-geosystems.com>,
	hkallweit1@...il.com, davem@...emloft.net, edumazet@...gle.com,
	kuba@...nel.org, pabeni@...hat.com, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	bsp-development.geo@...ca-geosystems.com, m.felsch@...gutronix.de
Subject: Re: [PATCH net-next v2] net: phy: dp83826: disable WOL at init

On Mon, Feb 26, 2024 at 04:37:55PM +0000, Russell King (Oracle) wrote:
> On Mon, Feb 26, 2024 at 05:23:39PM +0100, Catalin Popescu wrote:
> > Commit d1d77120bc28 ("net: phy: dp83826: support TX data voltage tuning")
> > introduced a regression in that WOL is not disabled by default for DP83826.
> > WOL should normally be enabled through ethtool.
> > 
> > Fixes: d1d77120bc28 ("net: phy: dp83826: support TX data voltage tuning")
> > Signed-off-by: Catalin Popescu <catalin.popescu@...ca-geosystems.com>
> 
> It seems rather interesting that WoL is disabled in the config_init()
> method - because this will be called when the PHY is attached to its
> network driver (reasonable I guess) but also at resume time - which
> means one can't just set the WoL mode once and that status will be
> preserved.
> 
> Maybe Andrew can clarify, but I thought once WoL was configured, that
> configuration should remain until the system is rebooted.

Yes, i noticed this as well.

d1d77120bc28 does change the behaviour, clearing WoL was dropped. As
you say, WoL was probably broken before d1d77120bc28 and it will still
be broken after this patch, so no real change there :-)

Catalin as a followup, could you please look at suspend/resume via
WoL. As Russell pointed out, if the machine is woken from suspend, we
expect the WoL settings to be kept. So WoL should be cleared on boot,
but not resume.

    Andrew