netdev - Re: [RFC PATCH net-next] net: phy: Don't suspend/resume device not in use

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: <c5238a4e-b4b1-484a-87f3-ea942b6aa04a@lunn.ch>
Date: Thu, 7 Mar 2024 17:22:57 +0100
From: Andrew Lunn <andrew@...n.ch>
To: "Jan Petrous (OSS)" <jan.petrous@....nxp.com>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	Heiner Kallweit <hkallweit1@...il.com>,
	Russell King <linux@...linux.org.uk>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>
Subject: Re: [RFC PATCH net-next] net: phy: Don't suspend/resume device not
 in use

> Because such device didn't go through attach process, internal
> parameters like phy_dev->interface is set to the default value, which
> is not correct for some drivers. Ie. Aquantia PHY AQR107 doesn't
> support PHY_INTERFACE_MODE_GMII and trying to use phy_init_hw()
> in mdio_bus_phy_resume() ends up with the following error caused
> by initial check of supported interfaces in aqr107_config_init():
> 
> [   63.927708] Aquantia AQR113C stmmac-0:08: PM: failed to resume: error -19']
> 
> Signed-off-by: Jan Petrous <jan.petrous@....nxp.com>
> ---
>  drivers/net/phy/phy_device.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
> index 3611ea64875e..30c03ac6b84c 100644
> --- a/drivers/net/phy/phy_device.c
> +++ b/drivers/net/phy/phy_device.c
> @@ -311,6 +311,10 @@ static __maybe_unused int mdio_bus_phy_suspend(struct device *dev)
>  {
>         struct phy_device *phydev = to_phy_device(dev);
> 
> +       /* Don't suspend device if not in use state */
> +       if (phydev->state <= PHY_READY)
> +               return 0;
> +
>         if (phydev->mac_managed_pm)
>                 return 0;

If nothing is using it, suspending it does however make sense. It is
consuming power, which could be saved by suspending it. It makes the
code asymmetrical, but i would throw this hunk away.

> 
> @@ -344,6 +348,10 @@ static __maybe_unused int mdio_bus_phy_resume(struct device *dev)
>         struct phy_device *phydev = to_phy_device(dev);
>         int ret;
> 
> +       /* Don't resume device which wasn't previously in use state */
> +       if (phydev->state <= PHY_READY)
> +               return 0;
> +

This is the real fix to your problem. phy_attach_direct() also does a
phy_resume(), so if the device does come along and claim the PHY after
the resume, it looks like this should work, without the matching
suspend.

	Andrew