Re: [RFC PATCH net-next 08/22] net: dsa: mt7530: change p{5,6}_interface to p{5,6}_configured

[Daniel Golle]

On Fri, Apr 21, 2023 at 09:25:39PM +0300, Arınç ÜNAL wrote:
> 
> 
> On 21.04.2023 21:20, Arınç ÜNAL wrote:
> > On 21.04.2023 21:17, Arınç ÜNAL wrote:
> > > On 21.04.2023 20:28, Daniel Golle wrote:
> > > > On Fri, Apr 21, 2023 at 05:36:34PM +0300, arinc9.unal@xxxxxxxxx wrote:
> > > > > From: Arınç ÜNAL <arinc.unal@xxxxxxxxxx>
> > > > > 
> > > > > The idea of p5_interface and p6_interface pointers is to prevent
> > > > > mt753x_mac_config() from running twice for MT7531, as it's
> > > > > already run with
> > > > > mt753x_cpu_port_enable() from mt7531_setup_common(), if the
> > > > > port is used as
> > > > > a CPU port.
> > > > > 
> > > > > Change p5_interface and p6_interface to p5_configured and
> > > > > p6_configured.
> > > > > Make them boolean.
> > > > > 
> > > > > Do not set them for any other reason.
> > > > > 
> > > > > The priv->p5_intf_sel check is useless as in this code path,
> > > > > it will always
> > > > > be P5_INTF_SEL_GMAC5.
> > > > > 
> > > > > There was also no need to set priv->p5_interface and
> > > > > priv->p6_interface to
> > > > > PHY_INTERFACE_MODE_NA on mt7530_setup() and mt7531_setup()
> > > > > as they would
> > > > > already be set to that when "priv" is allocated. The
> > > > > pointers were of the
> > > > > phy_interface_t enumeration type, and the first element of the enum is
> > > > > PHY_INTERFACE_MODE_NA. There was nothing in between that
> > > > > would change this
> > > > > beforehand.
> > > > > 
> > > > > Tested-by: Arınç ÜNAL <arinc.unal@xxxxxxxxxx>
> > > > > Signed-off-by: Arınç ÜNAL <arinc.unal@xxxxxxxxxx>
> > > > 
> > > > NACK. This assumes that a user port is configured exactly once.
> > > > However, interface mode may change because of mode-changing PHYs (e.g.
> > > > often using Cisco SGMII for 10M/100M/1000M but using 2500Base-X for
> > > > 2500M, ie. depending on actual link speed).
> > > > 
> > > > Also when using SFP modules (which can be hotplugged) the interface
> > > > mode may change after initially setting up the driver, e.g. when SFP
> > > > driver is loaded or a module is plugged or replaced.
> > > 
> > > I'm not sure I understand. pX_configured would be set to true only
> > > when the port is used as a CPU port. mt753x_mac_config() should run
> > > for user or DSA ports more than once, if needed.
> > 
> > Looking at this again, once pX_interface is true, the check will prevent
> > even user or DSA ports to be configured again. What about setting
> > pX_interface to false after mt753x_mac_config() is run?
> 
> On a third thought, pX_interface will never be true for the port if it's a
> user or DSA port so this should not be a problem at all.

I also followed the individual codepaths and conclude that you are
right. I've also tested this by now on several boards with MT753x,
incl. BPi-R3 and SerDes interface mode switching is anyway handled in
the PCS driver (I should have rembered that...)

Hence
Acked-by: Daniel Golle <daniel@xxxxxxxxxxxxxx>

> 
> Arınç