netdev - Re: [PATCH net-next 4/5] net: phy: add Lynx PCS MDIO module

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: <20200618220101.GJ1551@shell.armlinux.org.uk>
Date:   Thu, 18 Jun 2020 23:01:01 +0100
From:   Russell King - ARM Linux admin <linux@...linux.org.uk>
To:     Ioana Ciornei <ioana.ciornei@....com>
Cc:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "davem@...emloft.net" <davem@...emloft.net>,
        Vladimir Oltean <vladimir.oltean@....com>,
        Claudiu Manoil <claudiu.manoil@....com>,
        Alexandru Marginean <alexandru.marginean@....com>,
        "michael@...le.cc" <michael@...le.cc>,
        "andrew@...n.ch" <andrew@...n.ch>,
        "f.fainelli@...il.com" <f.fainelli@...il.com>
Subject: Re: [PATCH net-next 4/5] net: phy: add Lynx PCS MDIO module

On Thu, Jun 18, 2020 at 05:34:49PM +0000, Ioana Ciornei wrote:
> I am not sure how this would work with Felix and DSA drivers in general
> since the DSA core is hiding the phylink_pcs_ops from the actual switch
> driver.

Here's an idea to work around DSA (untested):

diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index 1f8e0023f4f4..f578a5bb8ad0 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -67,6 +67,7 @@ enum phylink_op_type {
 struct phylink_config {
 	struct device *dev;
 	enum phylink_op_type type;
+	void *pcs_private;
 	bool pcs_poll;
 	bool poll_fixed_state;
 	void (*get_fixed_state)(struct phylink_config *config,
diff --git a/include/net/dsa.h b/include/net/dsa.h
index a50d5007fd39..0fbb3a542b6e 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -742,6 +742,9 @@ int dsa_port_get_phy_strings(struct dsa_port *dp, uint8_t *data);
 int dsa_port_get_ethtool_phy_stats(struct dsa_port *dp, uint64_t *data);
 int dsa_port_get_phy_sset_count(struct dsa_port *dp);
 void dsa_port_phylink_mac_change(struct dsa_switch *ds, int port, bool up);
+void dsa_slave_attach_phylink_pcs(struct dsa_switch *ds, int port,
+				  const struct phylink_pcs_ops *ops,
+				  void *priv);
 
 struct dsa_tag_driver {
 	const struct dsa_device_ops *ops;
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 2ae17f95cb63..c80f62d88b63 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1619,6 +1619,17 @@ static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
 	return phylink_connect_phy(dp->pl, slave_dev->phydev);
 }
 
+void dsa_slave_attach_phylink_pcs(struct dsa_switch *ds, int port,
+				  const struct phylink_pcs_ops *ops,
+				  void *priv)
+{
+	const struct dsa_port *dp = dsa_to_port(ds, port);
+
+	dp->pl_config.pcs_private = priv;
+	phylink_add_pcs(dp->pl, ops);
+}
+EXPORT_SYMBOL_GPL(dsa_slave_attach_phylink_pcs);
+
 static int dsa_slave_phy_setup(struct net_device *slave_dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(slave_dev);


dsa_slave_attach_phylink_pcs() can be passed anything as it's priv
argument - that could be "dp", or it could be a lynx PCS private
data structure, but that's up to the driver code to decide.

At least this gives a way for us to have some standardised PCS
code that can be bolted into either DSA or a MAC driver by the
higher levels without the PCS code having to know which it's
connected to, and without having to have veneers to bridge into
the PCS code.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!