[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <65bbd2ce.050a0220.5ff09.69d5@mx.google.com>
Date: Thu, 1 Feb 2024 18:20:10 +0100
From: Christian Marangi <ansuelsmth@...il.com>
To: Antoine Tenart <atenart@...nel.org>
Cc: Andrew Lunn <andrew@...n.ch>, Bjorn Andersson <andersson@...nel.org>,
Conor Dooley <conor+dt@...nel.org>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Frank Rowand <frowand.list@...il.com>,
Heiner Kallweit <hkallweit1@...il.com>,
Jakub Kicinski <kuba@...nel.org>,
Konrad Dybcio <konrad.dybcio@...aro.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
Paolo Abeni <pabeni@...hat.com>, Rob Herring <robh+dt@...nel.org>,
Robert Marko <robert.marko@...tura.hr>,
Russell King <linux@...linux.org.uk>, devicetree@...r.kernel.org,
linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org
Subject: Re: [net-next PATCH v5 2/9] net: phy: add support for scanning PHY
in PHY packages nodes
On Thu, Feb 01, 2024 at 05:25:36PM +0100, Antoine Tenart wrote:
> Quoting Christian Marangi (2024-02-01 16:17:28)
> >
> > +static int __of_mdiobus_parse_phys(struct mii_bus *mdio, struct device_node *np,
> > + int base_addr, bool *scanphys)
> > +{
> > + struct device_node *child;
> > + int addr, rc = 0;
> > +
> > + /* Loop over the child nodes and register a phy_device for each phy */
> > + for_each_available_child_of_node(np, child) {
> > + if (of_node_name_eq(child, "ethernet-phy-package")) {
> > + rc = of_property_read_u32(child, "reg", &addr);
> > + if (rc)
> > + goto exit;
>
> This means a PHY package node w/o a reg property will prevent all other
> PHYs in the same parent node to be found?
>
Since this is something new, would it be a problem to make it mandatory
to define a reg? (And return error if we find something? Or print a
warn?)
> > +
> > + rc = __of_mdiobus_parse_phys(mdio, child, addr, scanphys);
>
> You might want to save passing scanphys down, PHYs w/o a reg property in
> a PHY package won't be "auto scanned" later.
>
I might be confused by this, but isn't this already done? (passing
scanphys in each recursive call so we can set it to true if needed?)
Also I think the scanphys should be skipped for the PHY package
(assuming we make reg mandatory, it would be an error condition and
should not be handled?)
> > diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
> > index afbad1ad8683..7737d0101d7b 100644
> > --- a/drivers/net/phy/mdio_bus.c
> > +++ b/drivers/net/phy/mdio_bus.c
> > @@ -459,20 +459,33 @@ EXPORT_SYMBOL(of_mdio_find_bus);
> > * found, set the of_node pointer for the mdio device. This allows
> > * auto-probed phy devices to be supplied with information passed in
> > * via DT.
> > + * If a PHY package is found, PHY is searched also there.
> > */
> > -static void of_mdiobus_link_mdiodev(struct mii_bus *bus,
> > - struct mdio_device *mdiodev)
> > +static int of_mdiobus_find_phy(struct device *dev, struct mdio_device *mdiodev,
> > + struct device_node *np, int base_addr)
> > {
> > - struct device *dev = &mdiodev->dev;
> > struct device_node *child;
> >
> > - if (dev->of_node || !bus->dev.of_node)
> > - return;
> > + for_each_available_child_of_node(np, child) {
> > + int addr, ret;
> >
> > - for_each_available_child_of_node(bus->dev.of_node, child) {
> > - int addr;
> > + if (of_node_name_eq(child, "ethernet-phy-package")) {
> > + ret = of_property_read_u32(child, "reg", &addr);
> > + if (ret)
> > + return ret;
>
> of_node_put
--
Ansuel
Powered by blists - more mailing lists