and of course the Pro uses a different switch, the AR8216.
David
On Sunday 10 April 2011, Jonas Gorski wrote:
> On 10 April 2011 12:09, Peter Lebbing <pe...@digitalbrains.com> wrote:
> > So, do you have any ideas on the best way to properly handle this PHY
> > that looks like 32 PHYs?
>
> Take a look at ar71xx, they define a (machine dependent) phy mask for
> each eth that says at which phy addresses a device may attach.
>
> e.g. the rs pro setup code has:
>
> ar71xx_add_device_mdio(~(UBNT_RSPRO_WAN_PHYMASK |
> UBNT_RSPRO_LAN_PHYMASK));
>
> ar71xx_init_mac(ar71xx_eth0_data.mac_addr, ar71xx_mac_base, 0);
> ar71xx_eth0_data.phy_if_mode = PHY_INTERFACE_MODE_RGMII;
> ar71xx_eth0_data.phy_mask = UBNT_RSPRO_WAN_PHYMASK;
>
> ubnt_init_secondary_mac(ar71xx_mac_base);
> ar71xx_eth1_data.phy_if_mode = PHY_INTERFACE_MODE_RGMII;
> ar71xx_eth1_data.phy_mask = UBNT_RSPRO_LAN_PHYMASK;
> ar71xx_eth1_data.speed = SPEED_1000;
> ar71xx_eth1_data.duplex = DUPLEX_FULL;
>
> This results that only for one of the "allowed" phy masks a phy driver
> gets attached:
>
> static int ag71xx_phy_connect_multi(struct ag71xx *ag)
> {
> (...)
> for (phy_addr = 0; phy_addr < PHY_MAX_ADDR; phy_addr++) {
> if (!(pdata->phy_mask & (1 << phy_addr)))
> continue;
>
> if (ag->mii_bus->phy_map[phy_addr] == NULL)
> continue;
>
> if (phydev == NULL)
> phydev = ag->mii_bus->phy_map[phy_addr];
> }
> (...)
> ag->phy_dev = phy_connect(dev, dev_name(&phydev->dev),
> &ag71xx_phy_link_adjust, 0,
> pdata->phy_if_mode);
> (...)
> }
>
>
> Of course this assumes you already know which device you have.
>
> Regards
> Jonas
> _______________________________________________
> openwrt-devel mailing list
> openwrt-devel@lists.openwrt.org
> https://lists.openwrt.org/mailman/listinfo/openwrt-devel
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
