1 STMicroelectronics 10/100/1000 Synopsys Ethernet driver
3 Copyright (C) 2007-2010 STMicroelectronics Ltd
4 Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
6 This is the driver for the MAC 10/100/1000 on-chip Ethernet controllers
9 Currently this network device driver is for all STM embedded MAC/GMAC
10 (i.e. 7xxx/5xxx SoCs), SPEAr (arm), Loongson1B (mips) and XLINX XC2V3000
11 FF1152AMT0221 D1215994A VIRTEX FPGA board.
13 DWC Ether MAC 10/100/1000 Universal version 3.60a (and older) and DWC Ether
14 MAC 10/100 Universal version 4.0 have been used for developing this driver.
16 This driver supports both the platform bus and PCI.
18 Please, for more information also visit: www.stlinux.com
20 1) Kernel Configuration
21 The kernel configuration option is STMMAC_ETH:
22 Device Drivers ---> Network device support ---> Ethernet (1000 Mbit) --->
23 STMicroelectronics 10/100/1000 Ethernet driver (STMMAC_ETH)
25 2) Driver parameters list:
26 debug: message level (0: no output, 16: all);
27 phyaddr: to manually provide the physical address to the PHY device;
28 dma_rxsize: DMA rx ring size;
29 dma_txsize: DMA tx ring size;
30 buf_sz: DMA buffer size;
31 tc: control the HW FIFO threshold;
32 tx_coe: Enable/Disable Tx Checksum Offload engine;
33 watchdog: transmit timeout (in milliseconds);
34 flow_ctrl: Flow control ability [on/off];
35 pause: Flow Control Pause Time;
36 tmrate: timer period (only if timer optimisation is configured).
38 3) Command line options
39 Driver parameters can be also passed in command line by using:
40 stmmaceth=dma_rxsize:128,dma_txsize:512
42 4) Driver information and notes
45 The xmit method is invoked when the kernel needs to transmit a packet; it sets
46 the descriptors in the ring and informs the DMA engine that there is a packet
47 ready to be transmitted.
48 Once the controller has finished transmitting the packet, an interrupt is
49 triggered; So the driver will be able to release the socket buffers.
50 By default, the driver sets the NETIF_F_SG bit in the features field of the
51 net_device structure enabling the scatter/gather feature.
54 When one or more packets are received, an interrupt happens. The interrupts
55 are not queued so the driver has to scan all the descriptors in the ring during
57 This is based on NAPI so the interrupt handler signals only if there is work
58 to be done, and it exits.
59 Then the poll method will be scheduled at some future point.
60 The incoming packets are stored, by the DMA, in a list of pre-allocated socket
61 buffers in order to avoid the memcpy (Zero-copy).
63 4.3) Timer-Driver Interrupt
64 Instead of having the device that asynchronously notifies the frame receptions,
65 the driver configures a timer to generate an interrupt at regular intervals.
66 Based on the granularity of the timer, the frames that are received by the
67 device will experience different levels of latency. Some NICs have dedicated
68 timer device to perform this task. STMMAC can use either the RTC device or the
69 TMU channel 2 on STLinux platforms.
70 The timers frequency can be passed to the driver as parameter; when change it,
71 take care of both hardware capability and network stability/performance impact.
72 Several performance tests on STM platforms showed this optimisation allows to
73 spare the CPU while having the maximum throughput.
76 Wake up on Lan feature through Magic and Unicast frames are supported for the
80 Driver handles both normal and enhanced descriptors. The latter has been only
81 tested on DWC Ether MAC 10/100/1000 Universal version 3.41a and later.
83 STMMAC supports DMA descriptor to operate both in dual buffer (RING)
84 and linked-list(CHAINED) mode. In RING each descriptor points to two
85 data buffer pointers whereas in CHAINED mode they point to only one data
86 buffer pointer. RING mode is the default.
88 In CHAINED mode each descriptor will have pointer to next descriptor in
89 the list, hence creating the explicit chaining in the descriptor itself,
90 whereas such explicit chaining is not possible in RING mode.
93 Ethtool is supported. Driver statistics and internal errors can be taken using:
94 ethtool -S ethX command. It is possible to dump registers etc.
96 4.7) Jumbo and Segmentation Offloading
97 Jumbo frames are supported and tested for the GMAC.
98 The GSO has been also added but it's performed in software.
102 The driver is compatible with PAL to work with PHY and GPHY devices.
104 4.9) Platform information
105 Several driver's information can be passed through the platform
106 These are included in the include/linux/stmmac.h header file
107 and detailed below as well:
109 struct plat_stmmacenet_data {
114 struct stmmac_mdio_bus_data *mdio_bus_data;
115 struct stmmac_dma_cfg *dma_cfg;
123 int force_sf_dma_mode;
124 void (*fix_mac_speed)(void *priv, unsigned int speed);
125 void (*bus_setup)(void __iomem *ioaddr);
126 int (*init)(struct platform_device *pdev);
127 void (*exit)(struct platform_device *pdev);
134 o phy_bus_name: phy bus name to attach to the stmmac.
135 o bus_id: bus identifier.
136 o phy_addr: the physical address can be passed from the platform.
137 If it is set to -1 the driver will automatically
138 detect it at run-time by probing all the 32 addresses.
139 o interface: PHY device's interface.
140 o mdio_bus_data: specific platform fields for the MDIO bus.
141 o dma_cfg: internal DMA parameters
142 o pbl: the Programmable Burst Length is maximum number of beats to
143 be transferred in one DMA transaction.
144 GMAC also enables the 4xPBL by default.
145 o fixed_burst/mixed_burst/burst_len
146 o clk_csr: fixed CSR Clock range selection.
147 o has_gmac: uses the GMAC core.
148 o enh_desc: if sets the MAC will use the enhanced descriptor structure.
149 o tx_coe: core is able to perform the tx csum in HW.
150 o rx_coe: the supports three check sum offloading engine types:
151 type_1, type_2 (full csum) and no RX coe.
152 o bugged_jumbo: some HWs are not able to perform the csum in HW for
153 over-sized frames due to limited buffer sizes.
154 Setting this flag the csum will be done in SW on
156 o pmt: core has the embedded power module (optional).
157 o force_sf_dma_mode: force DMA to use the Store and Forward mode
158 instead of the Threshold.
159 o fix_mac_speed: this callback is used for modifying some syscfg registers
160 (on ST SoCs) according to the link speed negotiated by the
162 o bus_setup: perform HW setup of the bus. For example, on some ST platforms
163 this field is used to configure the AMBA bridge to generate more
164 efficient STBus traffic.
165 o init/exit: callbacks used for calling a custom initialisation;
166 this is sometime necessary on some platforms (e.g. ST boxes)
167 where the HW needs to have set some PIO lines or system cfg
169 o custom_cfg/custom_data: this is a custom configuration that can be passed
170 while initialising the resources.
171 o bsp_priv: another private poiter.
173 For MDIO bus The we have:
175 struct stmmac_mdio_bus_data {
177 int (*phy_reset)(void *priv);
178 unsigned int phy_mask;
184 o bus_id: bus identifier;
185 o phy_reset: hook to reset the phy device attached to the bus.
186 o phy_mask: phy mask passed when register the MDIO bus within the driver.
187 o irqs: list of IRQs, one per PHY.
188 o probed_phy_irq: if irqs is NULL, use this for probed PHY.
190 For DMA engine we have the following internal fields that should be
191 tuned according to the HW capabilities.
193 struct stmmac_dma_cfg {
196 int burst_len_supported;
200 o pbl: Programmable Burst Length
201 o fixed_burst: program the DMA to use the fixed burst mode
202 o burst_len: this is the value we put in the register
203 supported values are provided as macros in
204 linux/stmmac.h header file.
208 Below an example how the structures above are using on ST platforms.
210 static struct plat_stmmacenet_data stxYYY_ethernet_platform_data = {
213 .fix_mac_speed = stxYYY_ethernet_fix_mac_speed,
215 |-> to write an internal syscfg
216 | on this platform when the
217 | link speed changes from 10 to
219 .init = &stmmac_claim_resource,
221 |-> On ST SoC this calls own "PAD"
222 | manager framework to claim
223 | all the resources necessary
224 | (GPIO ...). The .custom_cfg field
225 | is used to pass a custom config.
228 Below the usage of the stmmac_mdio_bus_data: on this SoC, in fact,
229 there are two MAC cores: one MAC is for MDIO Bus/PHY emulation
230 with fixed_link support.
232 static struct stmmac_mdio_bus_data stmmac1_mdio_bus = {
235 |-> phy device on the bus_id 1
236 .phy_reset = phy_reset;
238 |-> function to provide the phy_reset on this board
242 static struct fixed_phy_status stmmac0_fixed_phy_status = {
248 During the board's device_init we can configure the first
249 MAC for fixed_link by calling:
250 fixed_phy_add(PHY_POLL, 1, &stmmac0_fixed_phy_status));)
251 and the second one, with a real PHY device attached to the bus,
252 by using the stmmac_mdio_bus_data structure (to provide the id, the
253 reset procedure etc).
255 4.10) List of source files:
258 o stmmac_main.c: main network device driver;
259 o stmmac_mdio.c: mdio functions;
260 o stmmac_pci: PCI driver;
261 o stmmac_platform.c: platform driver
262 o stmmac_ethtool.c: ethtool support;
263 o stmmac_timer.[ch]: timer code used for mitigating the driver dma interrupts
264 (only tested on ST40 platforms based);
265 o stmmac.h: private driver structure;
266 o common.h: common definitions and VFTs;
267 o descs.h: descriptor structure definitions;
268 o dwmac1000_core.c: GMAC core functions;
269 o dwmac1000_dma.c: dma functions for the GMAC chip;
270 o dwmac1000.h: specific header file for the GMAC;
271 o dwmac100_core: MAC 100 core and dma code;
272 o dwmac100_dma.c: dma funtions for the MAC chip;
273 o dwmac1000.h: specific header file for the MAC;
274 o dwmac_lib.c: generic DMA functions shared among chips;
275 o enh_desc.c: functions for handling enhanced descriptors;
276 o norm_desc.c: functions for handling normal descriptors;
277 o chain_mode.c/ring_mode.c:: functions to manage RING/CHAINED modes;
278 o mmc_core.c/mmc.h: Management MAC Counters;
282 The driver exports many information i.e. internal statistics,
283 debug information, MAC and DMA registers etc.
285 These can be read in several ways depending on the
286 type of the information actually needed.
288 For example a user can be use the ethtool support
289 to get statistics: e.g. using: ethtool -S ethX
290 (that shows the Management counters (MMC) if supported)
291 or sees the MAC/DMA registers: e.g. using: ethtool -d ethX
293 Compiling the Kernel with CONFIG_DEBUG_FS and enabling the
294 STMMAC_DEBUG_FS option the driver will export the following
297 /sys/kernel/debug/stmmaceth/descriptors_status
298 To show the DMA TX/RX descriptor rings
300 Developer can also use the "debug" module parameter to get
301 further debug information.
303 In the end, there are other macros (that cannot be enabled
304 via menuconfig) to turn-on the RX/TX DMA debugging,
305 specific MAC core debug printk etc. Others to enable the
306 debug in the TX and RX processes.
307 All these are only useful during the developing stage
308 and should never enabled inside the code for general usage.
309 In fact, these can generate an huge amount of debug messages.
311 6) Energy Efficient Ethernet
313 Energy Efficient Ethernet(EEE) enables IEEE 802.3 MAC sublayer along
314 with a family of Physical layer to operate in the Low power Idle(LPI)
315 mode. The EEE mode supports the IEEE 802.3 MAC operation at 100Mbps,
318 The LPI mode allows power saving by switching off parts of the
319 communication device functionality when there is no data to be
320 transmitted & received. The system on both the side of the link can
321 disable some functionalities & save power during the period of low-link
322 utilization. The MAC controls whether the system should enter or exit
323 the LPI mode & communicate this to PHY.
325 As soon as the interface is opened, the driver verifies if the EEE can
326 be supported. This is done by looking at both the DMA HW capability
327 register and the PHY devices MCD registers.
328 To enter in Tx LPI mode the driver needs to have a software timer
329 that enable and disable the LPI mode when there is nothing to be
333 o XGMAC is not supported.
334 o Add the PTP - precision time protocol