Ethernet Driver Guide

The networking stack in Das U-Boot is designed for multiple network devices to be easily added and controlled at runtime. This guide is meant for people who wish to review the net driver stack with an eye towards implementing your own ethernet device driver. Here we will describe a new pseudo ‘APE’ driver.

Most existing drivers do already - and new network driver MUST - use the U-Boot core driver model. Generic information about this can be found in doc/driver-model/design.rst, this document will thus focus on the network specific code parts. Some drivers are still using the old Ethernet interface, differences between the two and hints about porting will be handled at the end.

Driver framework

A network driver following the driver model must declare itself using the UCLASS_ETH .id field in the U-Boot driver struct:

U_BOOT_DRIVER(eth_ape) = {
        .name                   = "eth_ape",
        .id                     = UCLASS_ETH,
        .of_match               = eth_ape_ids,
        .of_to_plat     = eth_ape_of_to_plat,
        .probe                  = eth_ape_probe,
        .ops                    = &eth_ape_ops,
        .priv_auto      = sizeof(struct eth_ape_priv),
        .plat_auto = sizeof(struct eth_ape_pdata),
        .flags                  = DM_FLAG_ALLOC_PRIV_DMA,
};

struct eth_ape_priv contains runtime per-instance data, like buffers, pointers to current descriptors, current speed settings, pointers to PHY related data (like struct mii_dev) and so on. Declaring its size in .priv_auto will let the driver framework allocate it at the right time. It can be retrieved using a dev_get_priv(dev) call.

struct eth_ape_pdata contains static platform data, like the MMIO base address, a hardware variant, the MAC address. struct eth_pdata eth_pdata as the first member of this struct helps to avoid duplicated code. If you don’t need any more platform data beside the standard member, just use sizeof(struct eth_pdata) for the plat_auto.

PCI devices add a line pointing to supported vendor/device ID pairs:

static struct pci_device_id supported[] = {
        { PCI_DEVICE(PCI_VENDOR_ID_APE, 0x4223) },
        {}
};

U_BOOT_PCI_DEVICE(eth_ape, supported);

It is also possible to declare support for a whole class of PCI devices:

{ PCI_DEVICE_CLASS(PCI_CLASS_SYSTEM_SDHCI << 8, 0xffff00) },

Device probing and instantiation will be handled by the driver model framework, so follow the guidelines there. The probe() function would initialise the platform specific parts of the hardware, like clocks, resets, GPIOs, the MDIO bus. Also it would take care of any special PHY setup (power rails, enable bits for internal PHYs, etc.).

Driver methods

The real work will be done in the driver method functions the driver provides by defining the members of struct eth_ops:

struct eth_ops {
        int (*start)(struct udevice *dev);
        int (*send)(struct udevice *dev, void *packet, int length);
        int (*recv)(struct udevice *dev, int flags, uchar **packetp);
        int (*free_pkt)(struct udevice *dev, uchar *packet, int length);
        void (*stop)(struct udevice *dev);
        int (*mcast)(struct udevice *dev, const u8 *enetaddr, int join);
        int (*write_hwaddr)(struct udevice *dev);
        int (*read_rom_hwaddr)(struct udevice *dev);
};

An up-to-date version of this struct together with more information can be found in include/net.h.

Only start, stop, send and recv are required, the rest are optional and are handled by generic code or ignored if not provided.

The start function initialises the hardware and gets it ready for send/recv operations. You often do things here such as resetting the MAC and/or PHY, and waiting for the link to autonegotiate. You should also take the opportunity to program the device’s MAC address with the enetaddr member of the generic struct eth_pdata (which would be the first member of your own plat struct). This allows the rest of U-Boot to dynamically change the MAC address and have the new settings be respected.

The send function does what you think – transmit the specified packet whose size is specified by length (in bytes). The packet buffer can (and will!) be reused for subsequent calls to send(), so it must be no longer used when the send() function returns. The easiest way to achieve this is to wait until the transmission is complete. Alternatively, if supported by the hardware, just waiting for the buffer to be consumed (by some DMA engine) might be an option as well. Another way of consuming the buffer could be to copy the data to be send, then just queue the copied packet (for instance handing it over to a DMA engine), and return immediately afterwards. In any case you should leave the state such that the send function can be called multiple times in a row.

The recv function polls for availability of a new packet. If none is available, it must return with -EAGAIN. If a packet has been received, make sure it is accessible to the CPU (invalidate caches if needed), then write its address to the packetp pointer, and return the length. If there is an error (receive error, too short or too long packet), return 0 if you require the packet to be cleaned up normally, or a negative error code otherwise (cleanup not necessary or already done). The U-Boot network stack will then process the packet.

If free_pkt is defined, U-Boot will call it after a received packet has been processed, so the packet buffer can be freed or recycled. Typically you would hand it back to the hardware to acquire another packet. free_pkt() will be called after recv(), for the same packet, so you don’t necessarily need to infer the buffer to free from the packet pointer, but can rely on that being the last packet that recv() handled. The common code sets up packet buffers for you already in the .bss (net_rx_packets), so there should be no need to allocate your own. This doesn’t mean you must use the net_rx_packets array however; you’re free to use any buffer you wish.

The stop function should turn off / disable the hardware and place it back in its reset state. It can be called at any time (before any call to the related start() function), so make sure it can handle this sort of thing.

The (optional) write_hwaddr function should program the MAC address stored in pdata->enetaddr into the Ethernet controller.

So the call graph at this stage would look something like:

(some net operation (ping / tftp / whatever...))
eth_init()
        ops->start()
eth_send()
        ops->send()
eth_rx()
        ops->recv()
        (process packet)
        if (ops->free_pkt)
                ops->free_pkt()
eth_halt()
        ops->stop()

CONFIG_PHYLIB / CONFIG_CMD_MII

If your device supports banging arbitrary values on the MII bus (pretty much every device does), you should add support for the mii command. Doing so is fairly trivial and makes debugging mii issues a lot easier at runtime.

In your driver’s probe() function, add a call to mdio_alloc() and mdio_register() like so:

bus = mdio_alloc();
if (!bus) {
        ...
        return -ENOMEM;
}

bus->read = ape_mii_read;
bus->write = ape_mii_write;
mdio_register(bus);

And then define the mii_read and mii_write functions if you haven’t already. Their syntax is straightforward:

int mii_read(struct mii_dev *bus, int addr, int devad, int reg);
int mii_write(struct mii_dev *bus, int addr, int devad, int reg,
              u16 val);

The read function should read the register ‘reg’ from the phy at address ‘addr’ and return the result to its caller. The implementation for the write function should logically follow.


Legacy network drivers

!!! WARNING !!!

This section below describes the old way of doing things. No new Ethernet drivers should be implemented this way. All new drivers should be written against the U-Boot core driver model, as described above.

The actual callback functions are fairly similar, the differences are:

  • start() is called init()

  • stop() is called halt()

  • The recv() function must loop until all packets have been received, for each packet it must call the net_process_received_packet() function, handing it over the pointer and the length. Afterwards it should free the packet, before checking for new data.

For porting an old driver to the new driver model, split the existing recv() function into the actual new recv() function, just fetching one packet, remove the call to net_process_received_packet(), then move the packet cleanup into the free_pkt() function.

Registering the driver and probing a device is handled very differently, follow the recommendations in the driver model design documentation for instructions on how to port this over. For the records, the old way of initialising a network driver is as follows:

Old network driver registration

When U-Boot initializes, it will call the common function eth_initialize(). This will in turn call the board-specific board_eth_init() (or if that fails, the cpu-specific cpu_eth_init()). These board-specific functions can do random system handling, but ultimately they will call the driver-specific register function which in turn takes care of initializing that particular instance.

Keep in mind that you should code the driver to avoid storing state in global data as someone might want to hook up two of the same devices to one board. Any such information that is specific to an interface should be stored in a private, driver-defined data structure and pointed to by eth->priv (see below).

So the call graph at this stage would look something like:

board_init()
        eth_initialize()
                board_eth_init() / cpu_eth_init()
                        driver_register()
                                initialize eth_device
                                eth_register()

At this point in time, the only thing you need to worry about is the driver’s register function. The pseudo code would look something like:

int ape_register(struct bd_info *bis, int iobase)
{
        struct ape_priv *priv;
        struct eth_device *dev;
        struct mii_dev *bus;

        priv = malloc(sizeof(*priv));
        if (priv == NULL)
                return -ENOMEM;

        dev = malloc(sizeof(*dev));
        if (dev == NULL) {
                free(priv);
                return -ENOMEM;
        }

        /* setup whatever private state you need */

        memset(dev, 0, sizeof(*dev));
        sprintf(dev->name, "APE");

        /*
         * if your device has dedicated hardware storage for the
         * MAC, read it and initialize dev->enetaddr with it
         */
        ape_mac_read(dev->enetaddr);

        dev->iobase = iobase;
        dev->priv = priv;
        dev->init = ape_init;
        dev->halt = ape_halt;
        dev->send = ape_send;
        dev->recv = ape_recv;
        dev->write_hwaddr = ape_write_hwaddr;

        eth_register(dev);

#ifdef CONFIG_PHYLIB
        bus = mdio_alloc();
        if (!bus) {
                free(priv);
                free(dev);
                return -ENOMEM;
        }

        bus->read = ape_mii_read;
        bus->write = ape_mii_write;
        mdio_register(bus);
#endif

        return 1;
}

The exact arguments needed to initialize your device are up to you. If you need to pass more/less arguments, that’s fine. You should also add the prototype for your new register function to include/netdev.h.

The return value for this function should be as follows: < 0 - failure (hardware failure, not probe failure) >=0 - number of interfaces detected

You might notice that many drivers seem to use xxx_initialize() rather than xxx_register(). This is the old naming convention and should be avoided as it causes confusion with the driver-specific init function.

Other than locating the MAC address in dedicated hardware storage, you should not touch the hardware in anyway. That step is handled in the driver-specific init function. Remember that we are only registering the device here, we are not checking its state or doing random probing.