RE: [PATCH net v2 1/2] net: stmmac: fix PHY handle parsing

[Sit, Michael Wei Hong]

> -----Original Message-----
> From: Russell King <linux@xxxxxxxxxxxxxxx>
> Sent: Saturday, March 18, 2023 4:58 AM
> To: Andrew Lunn <andrew@xxxxxxx>
> Cc: Sit, Michael Wei Hong <michael.wei.hong.sit@xxxxxxxxx>;
> Giuseppe Cavallaro <peppe.cavallaro@xxxxxx>; Alexandre
> Torgue <alexandre.torgue@xxxxxxxxxxx>; Jose Abreu
> <joabreu@xxxxxxxxxxxx>; David S . Miller
> <davem@xxxxxxxxxxxxx>; Eric Dumazet
> <edumazet@xxxxxxxxxx>; Jakub Kicinski <kuba@xxxxxxxxxx>;
> Paolo Abeni <pabeni@xxxxxxxxxx>; Maxime Coquelin
> <mcoquelin.stm32@xxxxxxxxx>; Ong, Boon Leong
> <boon.leong.ong@xxxxxxxxx>; netdev@xxxxxxxxxxxxxxx; linux-
> stm32@xxxxxxxxxxxxxxxxxxxxxxxxxxxx; linux-arm-
> kernel@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Looi,
> Hong Aun <hong.aun.looi@xxxxxxxxx>; Voon, Weifeng
> <weifeng.voon@xxxxxxxxx>; Lai, Peter Jun Ann
> <peter.jun.ann.lai@xxxxxxxxx>
> Subject: Re: [PATCH net v2 1/2] net: stmmac: fix PHY handle
> parsing
> 
> On Fri, Mar 17, 2023 at 08:56:19PM +0100, Andrew Lunn wrote:
> > On Tue, Mar 14, 2023 at 03:02:07PM +0800, Michael Sit Wei
> Hong wrote:
> > > phylink_fwnode_phy_connect returns 0 when set to
> MLO_AN_INBAND.
> > > This causes the PHY handle parsing to skip and the PHY will not
> be
> > > attached to the MAC.
> >
> > Please could you expand the commit message because i'm
> having trouble
> > following this.
> >
> > phylink_fwnode_phy_connect() says:
> >
> > 	/* Fixed links and 802.3z are handled without needing a
> PHY */
> > 	if (pl->cfg_link_an_mode == MLO_AN_FIXED ||
> > 	    (pl->cfg_link_an_mode == MLO_AN_INBAND &&
> > 	     phy_interface_mode_is_8023z(pl->link_interface)))
> > 		return 0;
> >
> > So your first statement is not true. It should be
> MLO_AN_INBAND and
> > phy_interface_mode_is_8023z.
> >
> > > Add additional check for PHY handle parsing when set to
> MLO_AN_INBAND.
> >
> > Looking at the patch, there is no reference to
> MLO_AN_INBAND, or
> > managed = "in-band-status";
> 
> That's the pesky "xpcs_an_inband" which ends up as phylink's
> "ovr_an_inband"... I'm sure these are random renames of stuff
> to make sure that people struggle to follow the code.
> 
It is as mentioned above, the "xpcs_an_inband" will end up as
"ovr_an_inband" which will then
set pl->cfg_link_an_mode = MLO_AN_INBAND in the
phylink_parse_mode() in phylink.c

The phylink_fwnode_phy_connect() checks if both
MLO_AN_INBAND && phy_interface_mode_is_8023z() true
before returning 0.

But in our case, we only have MLO_AN_INBAND is true, which
then goes to the next part of the code.

	phy_fwnode = fwnode_get_phy_node(fwnode);
	if (IS_ERR(phy_fwnode)) {
		if (pl->cfg_link_an_mode == MLO_AN_PHY)
			return -ENODEV;
		return 0;
	}

Where here the IS_ERR(phy_fwnode) returns true, then it
Checks for MLO_AN_PHY, which in our case is not, so it returns
a 0.

When returned 0, our driver will then skip the manual phy parsing
due to if (!fwnode || ret)
> --
> RMK's Patch system:
> https://www.armlinux.org.uk/developer/patches/
> FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at
> last!