]> git.ipfire.org Git - people/ms/u-boot.git/blob - doc/README.drivers.eth
arc: cache - build invalidate_icache_all() and invalidate_dcache_all()
[people/ms/u-boot.git] / doc / README.drivers.eth
1 -----------------------
2 Ethernet Driver Guide
3 -----------------------
4
5 The networking stack in Das U-Boot is designed for multiple network devices
6 to be easily added and controlled at runtime. This guide is meant for people
7 who wish to review the net driver stack with an eye towards implementing your
8 own ethernet device driver. Here we will describe a new pseudo 'APE' driver.
9
10 ------------------
11 Driver Functions
12 ------------------
13
14 All functions you will be implementing in this document have the return value
15 meaning of 0 for success and non-zero for failure.
16
17 ----------
18 Register
19 ----------
20
21 When U-Boot initializes, it will call the common function eth_initialize().
22 This will in turn call the board-specific board_eth_init() (or if that fails,
23 the cpu-specific cpu_eth_init()). These board-specific functions can do random
24 system handling, but ultimately they will call the driver-specific register
25 function which in turn takes care of initializing that particular instance.
26
27 Keep in mind that you should code the driver to avoid storing state in global
28 data as someone might want to hook up two of the same devices to one board.
29 Any such information that is specific to an interface should be stored in a
30 private, driver-defined data structure and pointed to by eth->priv (see below).
31
32 So the call graph at this stage would look something like:
33 board_init()
34 eth_initialize()
35 board_eth_init() / cpu_eth_init()
36 driver_register()
37 initialize eth_device
38 eth_register()
39
40 At this point in time, the only thing you need to worry about is the driver's
41 register function. The pseudo code would look something like:
42 int ape_register(bd_t *bis, int iobase)
43 {
44 struct ape_priv *priv;
45 struct eth_device *dev;
46 struct mii_dev *bus;
47
48 priv = malloc(sizeof(*priv));
49 if (priv == NULL)
50 return -ENOMEM;
51
52 dev = malloc(sizeof(*dev));
53 if (dev == NULL) {
54 free(priv);
55 return -ENOMEM;
56 }
57
58 /* setup whatever private state you need */
59
60 memset(dev, 0, sizeof(*dev));
61 sprintf(dev->name, "APE");
62
63 /*
64 * if your device has dedicated hardware storage for the
65 * MAC, read it and initialize dev->enetaddr with it
66 */
67 ape_mac_read(dev->enetaddr);
68
69 dev->iobase = iobase;
70 dev->priv = priv;
71 dev->init = ape_init;
72 dev->halt = ape_halt;
73 dev->send = ape_send;
74 dev->recv = ape_recv;
75 dev->write_hwaddr = ape_write_hwaddr;
76
77 eth_register(dev);
78
79 #ifdef CONFIG_PHYLIB
80 bus = mdio_alloc();
81 if (!bus) {
82 free(priv);
83 free(dev);
84 return -ENOMEM;
85 }
86
87 bus->read = ape_mii_read;
88 bus->write = ape_mii_write;
89 mdio_register(bus);
90 #endif
91
92 return 1;
93 }
94
95 The exact arguments needed to initialize your device are up to you. If you
96 need to pass more/less arguments, that's fine. You should also add the
97 prototype for your new register function to include/netdev.h.
98
99 The return value for this function should be as follows:
100 < 0 - failure (hardware failure, not probe failure)
101 >=0 - number of interfaces detected
102
103 You might notice that many drivers seem to use xxx_initialize() rather than
104 xxx_register(). This is the old naming convention and should be avoided as it
105 causes confusion with the driver-specific init function.
106
107 Other than locating the MAC address in dedicated hardware storage, you should
108 not touch the hardware in anyway. That step is handled in the driver-specific
109 init function. Remember that we are only registering the device here, we are
110 not checking its state or doing random probing.
111
112 -----------
113 Callbacks
114 -----------
115
116 Now that we've registered with the ethernet layer, we can start getting some
117 real work done. You will need five functions:
118 int ape_init(struct eth_device *dev, bd_t *bis);
119 int ape_send(struct eth_device *dev, volatile void *packet, int length);
120 int ape_recv(struct eth_device *dev);
121 int ape_halt(struct eth_device *dev);
122 int ape_write_hwaddr(struct eth_device *dev);
123
124 The init function checks the hardware (probing/identifying) and gets it ready
125 for send/recv operations. You often do things here such as resetting the MAC
126 and/or PHY, and waiting for the link to autonegotiate. You should also take
127 the opportunity to program the device's MAC address with the dev->enetaddr
128 member. This allows the rest of U-Boot to dynamically change the MAC address
129 and have the new settings be respected.
130
131 The send function does what you think -- transmit the specified packet whose
132 size is specified by length (in bytes). You should not return until the
133 transmission is complete, and you should leave the state such that the send
134 function can be called multiple times in a row.
135
136 The recv function should process packets as long as the hardware has them
137 readily available before returning. i.e. you should drain the hardware fifo.
138 For each packet you receive, you should call the NetReceive() function on it
139 along with the packet length. The common code sets up packet buffers for you
140 already in the .bss (NetRxPackets), so there should be no need to allocate your
141 own. This doesn't mean you must use the NetRxPackets array however; you're
142 free to call the NetReceive() function with any buffer you wish. So the pseudo
143 code here would look something like:
144 int ape_recv(struct eth_device *dev)
145 {
146 int length, i = 0;
147 ...
148 while (packets_are_available()) {
149 ...
150 length = ape_get_packet(&NetRxPackets[i]);
151 ...
152 NetReceive(&NetRxPackets[i], length);
153 ...
154 if (++i >= PKTBUFSRX)
155 i = 0;
156 ...
157 }
158 ...
159 return 0;
160 }
161
162 The halt function should turn off / disable the hardware and place it back in
163 its reset state. It can be called at any time (before any call to the related
164 init function), so make sure it can handle this sort of thing.
165
166 The write_hwaddr function should program the MAC address stored in dev->enetaddr
167 into the Ethernet controller.
168
169 So the call graph at this stage would look something like:
170 some net operation (ping / tftp / whatever...)
171 eth_init()
172 dev->init()
173 eth_send()
174 dev->send()
175 eth_rx()
176 dev->recv()
177 eth_halt()
178 dev->halt()
179
180 --------------------------------
181 CONFIG_PHYLIB / CONFIG_CMD_MII
182 --------------------------------
183
184 If your device supports banging arbitrary values on the MII bus (pretty much
185 every device does), you should add support for the mii command. Doing so is
186 fairly trivial and makes debugging mii issues a lot easier at runtime.
187
188 After you have called eth_register() in your driver's register function, add
189 a call to mdio_alloc() and mdio_register() like so:
190 bus = mdio_alloc();
191 if (!bus) {
192 free(priv);
193 free(dev);
194 return -ENOMEM;
195 }
196
197 bus->read = ape_mii_read;
198 bus->write = ape_mii_write;
199 mdio_register(bus);
200
201 And then define the mii_read and mii_write functions if you haven't already.
202 Their syntax is straightforward:
203 int mii_read(struct mii_dev *bus, int addr, int devad, int reg);
204 int mii_write(struct mii_dev *bus, int addr, int devad, int reg,
205 u16 val);
206
207 The read function should read the register 'reg' from the phy at address 'addr'
208 and return the result to its caller. The implementation for the write function
209 should logically follow.