Re: [RFC PATCH 2/3] net: sparx5: Add Sparx5 switchdev driver - Russell King

From: Russell King - ARM Linux admin <linux@armlinux.org.uk>
To: Steen Hegelund <steen.hegelund@microchip.com>
Cc: Andrew Lunn <andrew@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Masahiro Yamada <masahiroy@kernel.org>,
	Lars Povlsen <lars.povlsen@microchip.com>,
	Bjarni Jonasson <bjarni.jonasson@microchip.com>,
	Microchip Linux Driver Support <UNGLinuxDriver@microchip.com>,
	Alexandre Belloni <alexandre.belloni@bootlin.com>,
	Microsemi List <microsemi@lists.bootlin.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH 2/3] net: sparx5: Add Sparx5 switchdev driver
Date: Mon, 30 Nov 2020 14:52:14 +0000	[thread overview]
Message-ID: <20201130145214.GX1551@shell.armlinux.org.uk> (raw)
In-Reply-To: <20201130141556.o4vg32lr4uykwxmu@mchp-dev-shegelun>

On Mon, Nov 30, 2020 at 03:15:56PM +0100, Steen Hegelund wrote:
> On 29.11.2020 10:52, Russell King - ARM Linux admin wrote:
> > EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
> > 
> > On Sat, Nov 28, 2020 at 10:28:28PM +0000, Russell King - ARM Linux admin wrote:
> > > On Sat, Nov 28, 2020 at 08:06:16PM +0100, Andrew Lunn wrote:
> > > > > +static void sparx5_phylink_mac_config(struct phylink_config *config,
> > > > > +                               unsigned int mode,
> > > > > +                               const struct phylink_link_state *state)
> > > > > +{
> > > > > + struct sparx5_port *port = netdev_priv(to_net_dev(config->dev));
> > > > > + struct sparx5_port_config conf;
> > > > > + int err = 0;
> > > > > +
> > > > > + conf = port->conf;
> > > > > + conf.autoneg = state->an_enabled;
> > > > > + conf.pause = state->pause;
> > > > > + conf.duplex = state->duplex;
> > > > > + conf.power_down = false;
> > > > > + conf.portmode = state->interface;
> > > > > +
> > > > > + if (state->speed == SPEED_UNKNOWN) {
> > > > > +         /* When a SFP is plugged in we use capabilities to
> > > > > +          * default to the highest supported speed
> > > > > +          */
> > > >
> > > > This looks suspicious.
> > > 
> > > Yes, it looks highly suspicious. The fact that
> > > sparx5_phylink_mac_link_up() is empty, and sparx5_phylink_mac_config()
> > > does all the work suggests that this was developed before the phylink
> > > re-organisation, and this code hasn't been updated for it.
> > > 
> > > Any new code for the kernel really ought to be updated for the new
> > > phylink methodology before it is accepted.
> > > 
> > > Looking at sparx5_port_config(), it also seems to use
> > > PHY_INTERFACE_MODE_1000BASEX for both 1000BASE-X and 2500BASE-X. All
> > > very well for the driver to do that internally, but it's confusing
> > > when it comes to reviewing this stuff, especially when people outside
> > > of the driver (such as myself) reviewing it need to understand what's
> > > going on with the configuration.
> > 
> 
> Hi Russell,
> 
> > There are other issues too.
> > 
> > Looking at sparx5_get_1000basex_status(), we have:
> > 
> > +       status->link = DEV2G5_PCS1G_LINK_STATUS_LINK_STATUS_GET(value) |
> > +                      DEV2G5_PCS1G_LINK_STATUS_SYNC_STATUS_GET(value);
> > 
> 
> > Why is the link status the logical OR of these?
> 
> Oops: It should have been AND. Well spotted.

Do you need to check the sync status? Isn't it impossible to have "link
up" on a link that is unsynchronised?

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