[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20241107191617.00e0584a@fedora.home>
Date: Thu, 7 Nov 2024 19:16:17 +0100
From: Maxime Chevallier <maxime.chevallier@...tlin.com>
To: Andrew Lunn <andrew@...n.ch>
Cc: davem@...emloft.net, Jakub Kicinski <kuba@...nel.org>, Eric Dumazet
<edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>, Russell King
<linux@...linux.org.uk>, Christophe Leroy <christophe.leroy@...roup.eu>,
Heiner Kallweit <hkallweit1@...il.com>, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org, thomas.petazzoni@...tlin.com, Herve Codina
<herve.codina@...tlin.com>, Uwe Kleine-König
<u.kleine-koenig@...libre.com>, linuxppc-dev@...ts.ozlabs.org
Subject: Re: [PATCH net-next 4/7] net: freescale: ucc_geth: Fix WOL
configuration
Hello Andrew,
On Thu, 7 Nov 2024 18:49:00 +0100
Andrew Lunn <andrew@...n.ch> wrote:
> > + if (phydev)
> > + phy_ethtool_get_wol(phydev, wol);
> > +
> > if (qe_alive_during_sleep())
> > wol->supported |= WAKE_MAGIC;
>
> So get WoL will return whatever methods the PHY supports, plus
> WAKE_MAGIC is added because i assume the MAC can do that. So depending
> on the PHY, it could be the full list:
> > + if (wol->wolopts & WAKE_PHY)
> > + ret = phy_ethtool_set_wol(phydev, wol);
>
> Here, the code only calls into the PHY for WAKE_PHY, when in fact the
> PHY could be handling WAKE_UCAST, WAKE_MCAST, WAKE_ARP etc.
>
> So these conditions are wrong. It could we be that X years ago when
> this code was added only WAKE_PHY and WAKE_MAGIC existed?
Ah that's right indeed. So yeah my changes aren't enough, I'll go over
that patch again.
Thanks for spotting this,
Maxime
Powered by blists - more mailing lists