blob: 59f5a14e54e819d1f078286539d305716ade076e [file] [log] [blame]
// SPDX-License-Identifier: GPL-2.0+
/*
* (C) Copyright 2009 Ilya Yanok, Emcraft Systems Ltd <yanok@emcraft.com>
* (C) Copyright 2008,2009 Eric Jarrige <eric.jarrige@armadeus.org>
* (C) Copyright 2008 Armadeus Systems nc
* (C) Copyright 2007 Pengutronix, Sascha Hauer <s.hauer@pengutronix.de>
* (C) Copyright 2007 Pengutronix, Juergen Beisert <j.beisert@pengutronix.de>
*/
#include <common.h>
#include <cpu_func.h>
#include <dm.h>
#include <env.h>
#include <log.h>
#include <malloc.h>
#include <memalign.h>
#include <miiphy.h>
#include <net.h>
#include <netdev.h>
#include <asm/cache.h>
#include <asm/global_data.h>
#include <linux/delay.h>
#include <power/regulator.h>
#include <asm/io.h>
#include <linux/errno.h>
#include <linux/compiler.h>
#include <asm/arch/clock.h>
#include <asm/arch/imx-regs.h>
#include <asm/mach-imx/sys_proto.h>
#include <asm-generic/gpio.h>
#include "fec_mxc.h"
#include <eth_phy.h>
DECLARE_GLOBAL_DATA_PTR;
/*
* Timeout the transfer after 5 mS. This is usually a bit more, since
* the code in the tightloops this timeout is used in adds some overhead.
*/
#define FEC_XFER_TIMEOUT 5000
/*
* The standard 32-byte DMA alignment does not work on mx6solox, which requires
* 64-byte alignment in the DMA RX FEC buffer.
* Introduce the FEC_DMA_RX_MINALIGN which can cover mx6solox needs and also
* satisfies the alignment on other SoCs (32-bytes)
*/
#define FEC_DMA_RX_MINALIGN 64
#ifndef CONFIG_MII
#error "CONFIG_MII has to be defined!"
#endif
#ifndef CONFIG_FEC_XCV_TYPE
#define CONFIG_FEC_XCV_TYPE MII100
#endif
/*
* The i.MX28 operates with packets in big endian. We need to swap them before
* sending and after receiving.
*/
#ifdef CONFIG_MX28
#define CONFIG_FEC_MXC_SWAP_PACKET
#endif
#define RXDESC_PER_CACHELINE (ARCH_DMA_MINALIGN/sizeof(struct fec_bd))
/* Check various alignment issues at compile time */
#if ((ARCH_DMA_MINALIGN < 16) || (ARCH_DMA_MINALIGN % 16 != 0))
#error "ARCH_DMA_MINALIGN must be multiple of 16!"
#endif
#if ((PKTALIGN < ARCH_DMA_MINALIGN) || \
(PKTALIGN % ARCH_DMA_MINALIGN != 0))
#error "PKTALIGN must be multiple of ARCH_DMA_MINALIGN!"
#endif
#undef DEBUG
#ifdef CONFIG_FEC_MXC_SWAP_PACKET
static void swap_packet(uint32_t *packet, int length)
{
int i;
for (i = 0; i < DIV_ROUND_UP(length, 4); i++)
packet[i] = __swab32(packet[i]);
}
#endif
/* MII-interface related functions */
static int fec_mdio_read(struct ethernet_regs *eth, uint8_t phyaddr,
uint8_t regaddr)
{
uint32_t reg; /* convenient holder for the PHY register */
uint32_t phy; /* convenient holder for the PHY */
uint32_t start;
int val;
/*
* reading from any PHY's register is done by properly
* programming the FEC's MII data register.
*/
writel(FEC_IEVENT_MII, &eth->ievent);
reg = regaddr << FEC_MII_DATA_RA_SHIFT;
phy = phyaddr << FEC_MII_DATA_PA_SHIFT;
writel(FEC_MII_DATA_ST | FEC_MII_DATA_OP_RD | FEC_MII_DATA_TA |
phy | reg, &eth->mii_data);
/* wait for the related interrupt */
start = get_timer(0);
while (!(readl(&eth->ievent) & FEC_IEVENT_MII)) {
if (get_timer(start) > (CONFIG_SYS_HZ / 1000)) {
printf("Read MDIO failed...\n");
return -1;
}
}
/* clear mii interrupt bit */
writel(FEC_IEVENT_MII, &eth->ievent);
/* it's now safe to read the PHY's register */
val = (unsigned short)readl(&eth->mii_data);
debug("%s: phy: %02x reg:%02x val:%#x\n", __func__, phyaddr,
regaddr, val);
return val;
}
#ifndef imx_get_fecclk
u32 __weak imx_get_fecclk(void)
{
return 0;
}
#endif
static int fec_get_clk_rate(void *udev, int idx)
{
struct fec_priv *fec;
struct udevice *dev;
int ret;
if (IS_ENABLED(CONFIG_IMX8) ||
CONFIG_IS_ENABLED(CLK_CCF)) {
dev = udev;
if (!dev) {
ret = uclass_get_device(UCLASS_ETH, idx, &dev);
if (ret < 0) {
debug("Can't get FEC udev: %d\n", ret);
return ret;
}
}
fec = dev_get_priv(dev);
if (fec)
return fec->clk_rate;
return -EINVAL;
} else {
return imx_get_fecclk();
}
}
static void fec_mii_setspeed(struct ethernet_regs *eth)
{
/*
* Set MII_SPEED = (1/(mii_speed * 2)) * System Clock
* and do not drop the Preamble.
*
* The i.MX28 and i.MX6 types have another field in the MSCR (aka
* MII_SPEED) register that defines the MDIO output hold time. Earlier
* versions are RAZ there, so just ignore the difference and write the
* register always.
* The minimal hold time according to IEE802.3 (clause 22) is 10 ns.
* HOLDTIME + 1 is the number of clk cycles the fec is holding the
* output.
* The HOLDTIME bitfield takes values between 0 and 7 (inclusive).
* Given that ceil(clkrate / 5000000) <= 64, the calculation for
* holdtime cannot result in a value greater than 3.
*/
u32 pclk;
u32 speed;
u32 hold;
int ret;
ret = fec_get_clk_rate(NULL, 0);
if (ret < 0) {
printf("Can't find FEC0 clk rate: %d\n", ret);
return;
}
pclk = ret;
speed = DIV_ROUND_UP(pclk, 5000000);
hold = DIV_ROUND_UP(pclk, 100000000) - 1;
#ifdef FEC_QUIRK_ENET_MAC
speed--;
#endif
writel(speed << 1 | hold << 8, &eth->mii_speed);
debug("%s: mii_speed %08x\n", __func__, readl(&eth->mii_speed));
}
static int fec_mdio_write(struct ethernet_regs *eth, uint8_t phyaddr,
uint8_t regaddr, uint16_t data)
{
uint32_t reg; /* convenient holder for the PHY register */
uint32_t phy; /* convenient holder for the PHY */
uint32_t start;
reg = regaddr << FEC_MII_DATA_RA_SHIFT;
phy = phyaddr << FEC_MII_DATA_PA_SHIFT;
writel(FEC_MII_DATA_ST | FEC_MII_DATA_OP_WR |
FEC_MII_DATA_TA | phy | reg | data, &eth->mii_data);
/* wait for the MII interrupt */
start = get_timer(0);
while (!(readl(&eth->ievent) & FEC_IEVENT_MII)) {
if (get_timer(start) > (CONFIG_SYS_HZ / 1000)) {
printf("Write MDIO failed...\n");
return -1;
}
}
/* clear MII interrupt bit */
writel(FEC_IEVENT_MII, &eth->ievent);
debug("%s: phy: %02x reg:%02x val:%#x\n", __func__, phyaddr,
regaddr, data);
return 0;
}
static int fec_phy_read(struct mii_dev *bus, int phyaddr, int dev_addr,
int regaddr)
{
return fec_mdio_read(bus->priv, phyaddr, regaddr);
}
static int fec_phy_write(struct mii_dev *bus, int phyaddr, int dev_addr,
int regaddr, u16 data)
{
return fec_mdio_write(bus->priv, phyaddr, regaddr, data);
}
#ifndef CONFIG_PHYLIB
static int miiphy_restart_aneg(struct eth_device *dev)
{
int ret = 0;
#if !defined(CONFIG_FEC_MXC_NO_ANEG)
struct fec_priv *fec = (struct fec_priv *)dev->priv;
struct ethernet_regs *eth = fec->bus->priv;
/*
* Wake up from sleep if necessary
* Reset PHY, then delay 300ns
*/
#ifdef CONFIG_MX27
fec_mdio_write(eth, fec->phy_id, MII_DCOUNTER, 0x00FF);
#endif
fec_mdio_write(eth, fec->phy_id, MII_BMCR, BMCR_RESET);
udelay(1000);
/* Set the auto-negotiation advertisement register bits */
fec_mdio_write(eth, fec->phy_id, MII_ADVERTISE,
LPA_100FULL | LPA_100HALF | LPA_10FULL |
LPA_10HALF | PHY_ANLPAR_PSB_802_3);
fec_mdio_write(eth, fec->phy_id, MII_BMCR,
BMCR_ANENABLE | BMCR_ANRESTART);
if (fec->mii_postcall)
ret = fec->mii_postcall(fec->phy_id);
#endif
return ret;
}
#ifndef CONFIG_FEC_FIXED_SPEED
static int miiphy_wait_aneg(struct eth_device *dev)
{
uint32_t start;
int status;
struct fec_priv *fec = (struct fec_priv *)dev->priv;
struct ethernet_regs *eth = fec->bus->priv;
/* Wait for AN completion */
start = get_timer(0);
do {
if (get_timer(start) > (CONFIG_SYS_HZ * 5)) {
printf("%s: Autonegotiation timeout\n", dev->name);
return -1;
}
status = fec_mdio_read(eth, fec->phy_id, MII_BMSR);
if (status < 0) {
printf("%s: Autonegotiation failed. status: %d\n",
dev->name, status);
return -1;
}
} while (!(status & BMSR_LSTATUS));
return 0;
}
#endif /* CONFIG_FEC_FIXED_SPEED */
#endif
static int fec_rx_task_enable(struct fec_priv *fec)
{
writel(FEC_R_DES_ACTIVE_RDAR, &fec->eth->r_des_active);
return 0;
}
static int fec_rx_task_disable(struct fec_priv *fec)
{
return 0;
}
static int fec_tx_task_enable(struct fec_priv *fec)
{
writel(FEC_X_DES_ACTIVE_TDAR, &fec->eth->x_des_active);
return 0;
}
static int fec_tx_task_disable(struct fec_priv *fec)
{
return 0;
}
/**
* Initialize receive task's buffer descriptors
* @param[in] fec all we know about the device yet
* @param[in] count receive buffer count to be allocated
* @param[in] dsize desired size of each receive buffer
* @return 0 on success
*
* Init all RX descriptors to default values.
*/
static void fec_rbd_init(struct fec_priv *fec, int count, int dsize)
{
uint32_t size;
ulong data;
int i;
/*
* Reload the RX descriptors with default values and wipe
* the RX buffers.
*/
size = roundup(dsize, ARCH_DMA_MINALIGN);
for (i = 0; i < count; i++) {
data = fec->rbd_base[i].data_pointer;
memset((void *)data, 0, dsize);
flush_dcache_range(data, data + size);
fec->rbd_base[i].status = FEC_RBD_EMPTY;
fec->rbd_base[i].data_length = 0;
}
/* Mark the last RBD to close the ring. */
fec->rbd_base[i - 1].status = FEC_RBD_WRAP | FEC_RBD_EMPTY;
fec->rbd_index = 0;
flush_dcache_range((ulong)fec->rbd_base,
(ulong)fec->rbd_base + size);
}
/**
* Initialize transmit task's buffer descriptors
* @param[in] fec all we know about the device yet
*
* Transmit buffers are created externally. We only have to init the BDs here.\n
* Note: There is a race condition in the hardware. When only one BD is in
* use it must be marked with the WRAP bit to use it for every transmitt.
* This bit in combination with the READY bit results into double transmit
* of each data buffer. It seems the state machine checks READY earlier then
* resetting it after the first transfer.
* Using two BDs solves this issue.
*/
static void fec_tbd_init(struct fec_priv *fec)
{
ulong addr = (ulong)fec->tbd_base;
unsigned size = roundup(2 * sizeof(struct fec_bd),
ARCH_DMA_MINALIGN);
memset(fec->tbd_base, 0, size);
fec->tbd_base[0].status = 0;
fec->tbd_base[1].status = FEC_TBD_WRAP;
fec->tbd_index = 0;
flush_dcache_range(addr, addr + size);
}
/**
* Mark the given read buffer descriptor as free
* @param[in] last 1 if this is the last buffer descriptor in the chain, else 0
* @param[in] prbd buffer descriptor to mark free again
*/
static void fec_rbd_clean(int last, struct fec_bd *prbd)
{
unsigned short flags = FEC_RBD_EMPTY;
if (last)
flags |= FEC_RBD_WRAP;
writew(flags, &prbd->status);
writew(0, &prbd->data_length);
}
static int fec_get_hwaddr(int dev_id, unsigned char *mac)
{
imx_get_mac_from_fuse(dev_id, mac);
return !is_valid_ethaddr(mac);
}
#ifdef CONFIG_DM_ETH
static int fecmxc_set_hwaddr(struct udevice *dev)
#else
static int fec_set_hwaddr(struct eth_device *dev)
#endif
{
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
struct eth_pdata *pdata = dev_get_plat(dev);
uchar *mac = pdata->enetaddr;
#else
uchar *mac = dev->enetaddr;
struct fec_priv *fec = (struct fec_priv *)dev->priv;
#endif
writel(0, &fec->eth->iaddr1);
writel(0, &fec->eth->iaddr2);
writel(0, &fec->eth->gaddr1);
writel(0, &fec->eth->gaddr2);
/* Set physical address */
writel((mac[0] << 24) + (mac[1] << 16) + (mac[2] << 8) + mac[3],
&fec->eth->paddr1);
writel((mac[4] << 24) + (mac[5] << 16) + 0x8808, &fec->eth->paddr2);
return 0;
}
/* Do initial configuration of the FEC registers */
static void fec_reg_setup(struct fec_priv *fec)
{
uint32_t rcntrl;
/* Set interrupt mask register */
writel(0x00000000, &fec->eth->imask);
/* Clear FEC-Lite interrupt event register(IEVENT) */
writel(0xffffffff, &fec->eth->ievent);
/* Set FEC-Lite receive control register(R_CNTRL): */
/* Start with frame length = 1518, common for all modes. */
rcntrl = PKTSIZE << FEC_RCNTRL_MAX_FL_SHIFT;
if (fec->xcv_type != SEVENWIRE) /* xMII modes */
rcntrl |= FEC_RCNTRL_FCE | FEC_RCNTRL_MII_MODE;
if (fec->xcv_type == RGMII)
rcntrl |= FEC_RCNTRL_RGMII;
else if (fec->xcv_type == RMII)
rcntrl |= FEC_RCNTRL_RMII;
writel(rcntrl, &fec->eth->r_cntrl);
}
/**
* Start the FEC engine
* @param[in] dev Our device to handle
*/
#ifdef CONFIG_DM_ETH
static int fec_open(struct udevice *dev)
#else
static int fec_open(struct eth_device *edev)
#endif
{
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
#else
struct fec_priv *fec = (struct fec_priv *)edev->priv;
#endif
int speed;
ulong addr, size;
int i;
debug("fec_open: fec_open(dev)\n");
/* full-duplex, heartbeat disabled */
writel(1 << 2, &fec->eth->x_cntrl);
fec->rbd_index = 0;
/* Invalidate all descriptors */
for (i = 0; i < FEC_RBD_NUM - 1; i++)
fec_rbd_clean(0, &fec->rbd_base[i]);
fec_rbd_clean(1, &fec->rbd_base[i]);
/* Flush the descriptors into RAM */
size = roundup(FEC_RBD_NUM * sizeof(struct fec_bd),
ARCH_DMA_MINALIGN);
addr = (ulong)fec->rbd_base;
flush_dcache_range(addr, addr + size);
#ifdef FEC_QUIRK_ENET_MAC
/* Enable ENET HW endian SWAP */
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_DBSWAP,
&fec->eth->ecntrl);
/* Enable ENET store and forward mode */
writel(readl(&fec->eth->x_wmrk) | FEC_X_WMRK_STRFWD,
&fec->eth->x_wmrk);
#endif
/* Enable FEC-Lite controller */
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_ETHER_EN,
&fec->eth->ecntrl);
#ifdef FEC_ENET_ENABLE_TXC_DELAY
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_TXC_DLY,
&fec->eth->ecntrl);
#endif
#ifdef FEC_ENET_ENABLE_RXC_DELAY
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_RXC_DLY,
&fec->eth->ecntrl);
#endif
#if defined(CONFIG_MX25) || defined(CONFIG_MX53) || defined(CONFIG_MX6SL)
udelay(100);
/* setup the MII gasket for RMII mode */
/* disable the gasket */
writew(0, &fec->eth->miigsk_enr);
/* wait for the gasket to be disabled */
while (readw(&fec->eth->miigsk_enr) & MIIGSK_ENR_READY)
udelay(2);
/* configure gasket for RMII, 50 MHz, no loopback, and no echo */
writew(MIIGSK_CFGR_IF_MODE_RMII, &fec->eth->miigsk_cfgr);
/* re-enable the gasket */
writew(MIIGSK_ENR_EN, &fec->eth->miigsk_enr);
/* wait until MII gasket is ready */
int max_loops = 10;
while ((readw(&fec->eth->miigsk_enr) & MIIGSK_ENR_READY) == 0) {
if (--max_loops <= 0) {
printf("WAIT for MII Gasket ready timed out\n");
break;
}
}
#endif
#ifdef CONFIG_PHYLIB
{
/* Start up the PHY */
int ret = phy_startup(fec->phydev);
if (ret) {
printf("Could not initialize PHY %s\n",
fec->phydev->dev->name);
return ret;
}
speed = fec->phydev->speed;
}
#elif CONFIG_FEC_FIXED_SPEED
speed = CONFIG_FEC_FIXED_SPEED;
#else
miiphy_wait_aneg(edev);
speed = miiphy_speed(edev->name, fec->phy_id);
miiphy_duplex(edev->name, fec->phy_id);
#endif
#ifdef FEC_QUIRK_ENET_MAC
{
u32 ecr = readl(&fec->eth->ecntrl) & ~FEC_ECNTRL_SPEED;
u32 rcr = readl(&fec->eth->r_cntrl) & ~FEC_RCNTRL_RMII_10T;
if (speed == _1000BASET)
ecr |= FEC_ECNTRL_SPEED;
else if (speed != _100BASET)
rcr |= FEC_RCNTRL_RMII_10T;
writel(ecr, &fec->eth->ecntrl);
writel(rcr, &fec->eth->r_cntrl);
}
#endif
debug("%s:Speed=%i\n", __func__, speed);
/* Enable SmartDMA receive task */
fec_rx_task_enable(fec);
udelay(100000);
return 0;
}
#ifdef CONFIG_DM_ETH
static int fecmxc_init(struct udevice *dev)
#else
static int fec_init(struct eth_device *dev, struct bd_info *bd)
#endif
{
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
#else
struct fec_priv *fec = (struct fec_priv *)dev->priv;
#endif
u8 *mib_ptr = (uint8_t *)&fec->eth->rmon_t_drop;
u8 *i;
ulong addr;
/* Initialize MAC address */
#ifdef CONFIG_DM_ETH
fecmxc_set_hwaddr(dev);
#else
fec_set_hwaddr(dev);
#endif
/* Setup transmit descriptors, there are two in total. */
fec_tbd_init(fec);
/* Setup receive descriptors. */
fec_rbd_init(fec, FEC_RBD_NUM, FEC_MAX_PKT_SIZE);
fec_reg_setup(fec);
if (fec->xcv_type != SEVENWIRE)
fec_mii_setspeed(fec->bus->priv);
/* Set Opcode/Pause Duration Register */
writel(0x00010020, &fec->eth->op_pause); /* FIXME 0xffff0020; */
writel(0x2, &fec->eth->x_wmrk);
/* Set multicast address filter */
writel(0x00000000, &fec->eth->gaddr1);
writel(0x00000000, &fec->eth->gaddr2);
/* Do not access reserved register */
if (!is_mx6ul() && !is_mx6ull() && !is_imx8() && !is_imx8m()) {
/* clear MIB RAM */
for (i = mib_ptr; i <= mib_ptr + 0xfc; i += 4)
writel(0, i);
/* FIFO receive start register */
writel(0x520, &fec->eth->r_fstart);
}
/* size and address of each buffer */
writel(FEC_MAX_PKT_SIZE, &fec->eth->emrbr);
addr = (ulong)fec->tbd_base;
writel((uint32_t)addr, &fec->eth->etdsr);
addr = (ulong)fec->rbd_base;
writel((uint32_t)addr, &fec->eth->erdsr);
#ifndef CONFIG_PHYLIB
if (fec->xcv_type != SEVENWIRE)
miiphy_restart_aneg(dev);
#endif
fec_open(dev);
return 0;
}
/**
* Halt the FEC engine
* @param[in] dev Our device to handle
*/
#ifdef CONFIG_DM_ETH
static void fecmxc_halt(struct udevice *dev)
#else
static void fec_halt(struct eth_device *dev)
#endif
{
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
#else
struct fec_priv *fec = (struct fec_priv *)dev->priv;
#endif
int counter = 0xffff;
/* issue graceful stop command to the FEC transmitter if necessary */
writel(FEC_TCNTRL_GTS | readl(&fec->eth->x_cntrl),
&fec->eth->x_cntrl);
debug("eth_halt: wait for stop regs\n");
/* wait for graceful stop to register */
while ((counter--) && (!(readl(&fec->eth->ievent) & FEC_IEVENT_GRA)))
udelay(1);
/* Disable SmartDMA tasks */
fec_tx_task_disable(fec);
fec_rx_task_disable(fec);
/*
* Disable the Ethernet Controller
* Note: this will also reset the BD index counter!
*/
writel(readl(&fec->eth->ecntrl) & ~FEC_ECNTRL_ETHER_EN,
&fec->eth->ecntrl);
fec->rbd_index = 0;
fec->tbd_index = 0;
debug("eth_halt: done\n");
}
/**
* Transmit one frame
* @param[in] dev Our ethernet device to handle
* @param[in] packet Pointer to the data to be transmitted
* @param[in] length Data count in bytes
* @return 0 on success
*/
#ifdef CONFIG_DM_ETH
static int fecmxc_send(struct udevice *dev, void *packet, int length)
#else
static int fec_send(struct eth_device *dev, void *packet, int length)
#endif
{
unsigned int status;
u32 size;
ulong addr, end;
int timeout = FEC_XFER_TIMEOUT;
int ret = 0;
/*
* This routine transmits one frame. This routine only accepts
* 6-byte Ethernet addresses.
*/
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
#else
struct fec_priv *fec = (struct fec_priv *)dev->priv;
#endif
/*
* Check for valid length of data.
*/
if ((length > 1500) || (length <= 0)) {
printf("Payload (%d) too large\n", length);
return -1;
}
/*
* Setup the transmit buffer. We are always using the first buffer for
* transmission, the second will be empty and only used to stop the DMA
* engine. We also flush the packet to RAM here to avoid cache trouble.
*/
#ifdef CONFIG_FEC_MXC_SWAP_PACKET
swap_packet((uint32_t *)packet, length);
#endif
addr = (ulong)packet;
end = roundup(addr + length, ARCH_DMA_MINALIGN);
addr &= ~(ARCH_DMA_MINALIGN - 1);
flush_dcache_range(addr, end);
writew(length, &fec->tbd_base[fec->tbd_index].data_length);
writel((uint32_t)addr, &fec->tbd_base[fec->tbd_index].data_pointer);
/*
* update BD's status now
* This block:
* - is always the last in a chain (means no chain)
* - should transmitt the CRC
* - might be the last BD in the list, so the address counter should
* wrap (-> keep the WRAP flag)
*/
status = readw(&fec->tbd_base[fec->tbd_index].status) & FEC_TBD_WRAP;
status |= FEC_TBD_LAST | FEC_TBD_TC | FEC_TBD_READY;
writew(status, &fec->tbd_base[fec->tbd_index].status);
/*
* Flush data cache. This code flushes both TX descriptors to RAM.
* After this code, the descriptors will be safely in RAM and we
* can start DMA.
*/
size = roundup(2 * sizeof(struct fec_bd), ARCH_DMA_MINALIGN);
addr = (ulong)fec->tbd_base;
flush_dcache_range(addr, addr + size);
/*
* Below we read the DMA descriptor's last four bytes back from the
* DRAM. This is important in order to make sure that all WRITE
* operations on the bus that were triggered by previous cache FLUSH
* have completed.
*
* Otherwise, on MX28, it is possible to observe a corruption of the
* DMA descriptors. Please refer to schematic "Figure 1-2" in MX28RM
* for the bus structure of MX28. The scenario is as follows:
*
* 1) ARM core triggers a series of WRITEs on the AHB_ARB2 bus going
* to DRAM due to flush_dcache_range()
* 2) ARM core writes the FEC registers via AHB_ARB2
* 3) FEC DMA starts reading/writing from/to DRAM via AHB_ARB3
*
* Note that 2) does sometimes finish before 1) due to reordering of
* WRITE accesses on the AHB bus, therefore triggering 3) before the
* DMA descriptor is fully written into DRAM. This results in occasional
* corruption of the DMA descriptor.
*/
readl(addr + size - 4);
/* Enable SmartDMA transmit task */
fec_tx_task_enable(fec);
/*
* Wait until frame is sent. On each turn of the wait cycle, we must
* invalidate data cache to see what's really in RAM. Also, we need
* barrier here.
*/
while (--timeout) {
if (!(readl(&fec->eth->x_des_active) & FEC_X_DES_ACTIVE_TDAR))
break;
}
if (!timeout) {
ret = -EINVAL;
goto out;
}
/*
* The TDAR bit is cleared when the descriptors are all out from TX
* but on mx6solox we noticed that the READY bit is still not cleared
* right after TDAR.
* These are two distinct signals, and in IC simulation, we found that
* TDAR always gets cleared prior than the READY bit of last BD becomes
* cleared.
* In mx6solox, we use a later version of FEC IP. It looks like that
* this intrinsic behaviour of TDAR bit has changed in this newer FEC
* version.
*
* Fix this by polling the READY bit of BD after the TDAR polling,
* which covers the mx6solox case and does not harm the other SoCs.
*/
timeout = FEC_XFER_TIMEOUT;
while (--timeout) {
invalidate_dcache_range(addr, addr + size);
if (!(readw(&fec->tbd_base[fec->tbd_index].status) &
FEC_TBD_READY))
break;
}
if (!timeout)
ret = -EINVAL;
out:
debug("fec_send: status 0x%x index %d ret %i\n",
readw(&fec->tbd_base[fec->tbd_index].status),
fec->tbd_index, ret);
/* for next transmission use the other buffer */
if (fec->tbd_index)
fec->tbd_index = 0;
else
fec->tbd_index = 1;
return ret;
}
/**
* Pull one frame from the card
* @param[in] dev Our ethernet device to handle
* @return Length of packet read
*/
#ifdef CONFIG_DM_ETH
static int fecmxc_recv(struct udevice *dev, int flags, uchar **packetp)
#else
static int fec_recv(struct eth_device *dev)
#endif
{
#ifdef CONFIG_DM_ETH
struct fec_priv *fec = dev_get_priv(dev);
#else
struct fec_priv *fec = (struct fec_priv *)dev->priv;
#endif
struct fec_bd *rbd = &fec->rbd_base[fec->rbd_index];
unsigned long ievent;
int frame_length, len = 0;
uint16_t bd_status;
ulong addr, size, end;
int i;
#ifdef CONFIG_DM_ETH
*packetp = memalign(ARCH_DMA_MINALIGN, FEC_MAX_PKT_SIZE);
if (*packetp == 0) {
printf("%s: error allocating packetp\n", __func__);
return -ENOMEM;
}
#else
ALLOC_CACHE_ALIGN_BUFFER(uchar, buff, FEC_MAX_PKT_SIZE);
#endif
/* Check if any critical events have happened */
ievent = readl(&fec->eth->ievent);
writel(ievent, &fec->eth->ievent);
debug("fec_recv: ievent 0x%lx\n", ievent);
if (ievent & FEC_IEVENT_BABR) {
#ifdef CONFIG_DM_ETH
fecmxc_halt(dev);
fecmxc_init(dev);
#else
fec_halt(dev);
fec_init(dev, fec->bd);
#endif
printf("some error: 0x%08lx\n", ievent);
return 0;
}
if (ievent & FEC_IEVENT_HBERR) {
/* Heartbeat error */
writel(0x00000001 | readl(&fec->eth->x_cntrl),
&fec->eth->x_cntrl);
}
if (ievent & FEC_IEVENT_GRA) {
/* Graceful stop complete */
if (readl(&fec->eth->x_cntrl) & 0x00000001) {
#ifdef CONFIG_DM_ETH
fecmxc_halt(dev);
#else
fec_halt(dev);
#endif
writel(~0x00000001 & readl(&fec->eth->x_cntrl),
&fec->eth->x_cntrl);
#ifdef CONFIG_DM_ETH
fecmxc_init(dev);
#else
fec_init(dev, fec->bd);
#endif
}
}
/*
* Read the buffer status. Before the status can be read, the data cache
* must be invalidated, because the data in RAM might have been changed
* by DMA. The descriptors are properly aligned to cachelines so there's
* no need to worry they'd overlap.
*
* WARNING: By invalidating the descriptor here, we also invalidate
* the descriptors surrounding this one. Therefore we can NOT change the
* contents of this descriptor nor the surrounding ones. The problem is
* that in order to mark the descriptor as processed, we need to change
* the descriptor. The solution is to mark the whole cache line when all
* descriptors in the cache line are processed.
*/
addr = (ulong)rbd;
addr &= ~(ARCH_DMA_MINALIGN - 1);
size = roundup(sizeof(struct fec_bd), ARCH_DMA_MINALIGN);
invalidate_dcache_range(addr, addr + size);
bd_status = readw(&rbd->status);
debug("fec_recv: status 0x%x\n", bd_status);
if (!(bd_status & FEC_RBD_EMPTY)) {
if ((bd_status & FEC_RBD_LAST) && !(bd_status & FEC_RBD_ERR) &&
((readw(&rbd->data_length) - 4) > 14)) {
/* Get buffer address and size */
addr = readl(&rbd->data_pointer);
frame_length = readw(&rbd->data_length) - 4;
/* Invalidate data cache over the buffer */
end = roundup(addr + frame_length, ARCH_DMA_MINALIGN);
addr &= ~(ARCH_DMA_MINALIGN - 1);
invalidate_dcache_range(addr, end);
/* Fill the buffer and pass it to upper layers */
#ifdef CONFIG_FEC_MXC_SWAP_PACKET
swap_packet((uint32_t *)addr, frame_length);
#endif
#ifdef CONFIG_DM_ETH
memcpy(*packetp, (char *)addr, frame_length);
#else
memcpy(buff, (char *)addr, frame_length);
net_process_received_packet(buff, frame_length);
#endif
len = frame_length;
} else {
if (bd_status & FEC_RBD_ERR)
debug("error frame: 0x%08lx 0x%08x\n",
addr, bd_status);
}
/*
* Free the current buffer, restart the engine and move forward
* to the next buffer. Here we check if the whole cacheline of
* descriptors was already processed and if so, we mark it free
* as whole.
*/
size = RXDESC_PER_CACHELINE - 1;
if ((fec->rbd_index & size) == size) {
i = fec->rbd_index - size;
addr = (ulong)&fec->rbd_base[i];
for (; i <= fec->rbd_index ; i++) {
fec_rbd_clean(i == (FEC_RBD_NUM - 1),
&fec->rbd_base[i]);
}
flush_dcache_range(addr,
addr + ARCH_DMA_MINALIGN);
}
fec_rx_task_enable(fec);
fec->rbd_index = (fec->rbd_index + 1) % FEC_RBD_NUM;
}
debug("fec_recv: stop\n");
return len;
}
static void fec_set_dev_name(char *dest, int dev_id)
{
sprintf(dest, (dev_id == -1) ? "FEC" : "FEC%i", dev_id);
}
static int fec_alloc_descs(struct fec_priv *fec)
{
unsigned int size;
int i;
uint8_t *data;
ulong addr;
/* Allocate TX descriptors. */
size = roundup(2 * sizeof(struct fec_bd), ARCH_DMA_MINALIGN);
fec->tbd_base = memalign(ARCH_DMA_MINALIGN, size);
if (!fec->tbd_base)
goto err_tx;
/* Allocate RX descriptors. */
size = roundup(FEC_RBD_NUM * sizeof(struct fec_bd), ARCH_DMA_MINALIGN);
fec->rbd_base = memalign(ARCH_DMA_MINALIGN, size);
if (!fec->rbd_base)
goto err_rx;
memset(fec->rbd_base, 0, size);
/* Allocate RX buffers. */
/* Maximum RX buffer size. */
size = roundup(FEC_MAX_PKT_SIZE, FEC_DMA_RX_MINALIGN);
for (i = 0; i < FEC_RBD_NUM; i++) {
data = memalign(FEC_DMA_RX_MINALIGN, size);
if (!data) {
printf("%s: error allocating rxbuf %d\n", __func__, i);
goto err_ring;
}
memset(data, 0, size);
addr = (ulong)data;
fec->rbd_base[i].data_pointer = (uint32_t)addr;
fec->rbd_base[i].status = FEC_RBD_EMPTY;
fec->rbd_base[i].data_length = 0;
/* Flush the buffer to memory. */
flush_dcache_range(addr, addr + size);
}
/* Mark the last RBD to close the ring. */
fec->rbd_base[i - 1].status = FEC_RBD_WRAP | FEC_RBD_EMPTY;
fec->rbd_index = 0;
fec->tbd_index = 0;
return 0;
err_ring:
for (; i >= 0; i--) {
addr = fec->rbd_base[i].data_pointer;
free((void *)addr);
}
free(fec->rbd_base);
err_rx:
free(fec->tbd_base);
err_tx:
return -ENOMEM;
}
static void fec_free_descs(struct fec_priv *fec)
{
int i;
ulong addr;
for (i = 0; i < FEC_RBD_NUM; i++) {
addr = fec->rbd_base[i].data_pointer;
free((void *)addr);
}
free(fec->rbd_base);
free(fec->tbd_base);
}
struct mii_dev *fec_get_miibus(ulong base_addr, int dev_id)
{
struct ethernet_regs *eth = (struct ethernet_regs *)base_addr;
struct mii_dev *bus;
int ret;
bus = mdio_alloc();
if (!bus) {
printf("mdio_alloc failed\n");
return NULL;
}
bus->read = fec_phy_read;
bus->write = fec_phy_write;
bus->priv = eth;
fec_set_dev_name(bus->name, dev_id);
ret = mdio_register(bus);
if (ret) {
printf("mdio_register failed\n");
free(bus);
return NULL;
}
fec_mii_setspeed(eth);
return bus;
}
#ifndef CONFIG_DM_ETH
#ifdef CONFIG_PHYLIB
int fec_probe(struct bd_info *bd, int dev_id, uint32_t base_addr,
struct mii_dev *bus, struct phy_device *phydev)
#else
static int fec_probe(struct bd_info *bd, int dev_id, uint32_t base_addr,
struct mii_dev *bus, int phy_id)
#endif
{
struct eth_device *edev;
struct fec_priv *fec;
unsigned char ethaddr[6];
char mac[16];
uint32_t start;
int ret = 0;
/* create and fill edev struct */
edev = (struct eth_device *)malloc(sizeof(struct eth_device));
if (!edev) {
puts("fec_mxc: not enough malloc memory for eth_device\n");
ret = -ENOMEM;
goto err1;
}
fec = (struct fec_priv *)malloc(sizeof(struct fec_priv));
if (!fec) {
puts("fec_mxc: not enough malloc memory for fec_priv\n");
ret = -ENOMEM;
goto err2;
}
memset(edev, 0, sizeof(*edev));
memset(fec, 0, sizeof(*fec));
ret = fec_alloc_descs(fec);
if (ret)
goto err3;
edev->priv = fec;
edev->init = fec_init;
edev->send = fec_send;
edev->recv = fec_recv;
edev->halt = fec_halt;
edev->write_hwaddr = fec_set_hwaddr;
fec->eth = (struct ethernet_regs *)(ulong)base_addr;
fec->bd = bd;
fec->xcv_type = CONFIG_FEC_XCV_TYPE;
/* Reset chip. */
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_RESET, &fec->eth->ecntrl);
start = get_timer(0);
while (readl(&fec->eth->ecntrl) & FEC_ECNTRL_RESET) {
if (get_timer(start) > (CONFIG_SYS_HZ * 5)) {
printf("FEC MXC: Timeout resetting chip\n");
goto err4;
}
udelay(10);
}
fec_reg_setup(fec);
fec_set_dev_name(edev->name, dev_id);
fec->dev_id = (dev_id == -1) ? 0 : dev_id;
fec->bus = bus;
fec_mii_setspeed(bus->priv);
#ifdef CONFIG_PHYLIB
fec->phydev = phydev;
phy_connect_dev(phydev, edev);
/* Configure phy */
phy_config(phydev);
#else
fec->phy_id = phy_id;
#endif
eth_register(edev);
/* only support one eth device, the index number pointed by dev_id */
edev->index = fec->dev_id;
if (fec_get_hwaddr(fec->dev_id, ethaddr) == 0) {
debug("got MAC%d address from fuse: %pM\n", fec->dev_id, ethaddr);
memcpy(edev->enetaddr, ethaddr, 6);
if (fec->dev_id)
sprintf(mac, "eth%daddr", fec->dev_id);
else
strcpy(mac, "ethaddr");
if (!env_get(mac))
eth_env_set_enetaddr(mac, ethaddr);
}
return ret;
err4:
fec_free_descs(fec);
err3:
free(fec);
err2:
free(edev);
err1:
return ret;
}
int fecmxc_initialize_multi(struct bd_info *bd, int dev_id, int phy_id,
uint32_t addr)
{
uint32_t base_mii;
struct mii_dev *bus = NULL;
#ifdef CONFIG_PHYLIB
struct phy_device *phydev = NULL;
#endif
int ret;
if (CONFIG_IS_ENABLED(IMX_MODULE_FUSE)) {
if (enet_fused((ulong)addr)) {
printf("SoC fuse indicates Ethernet@0x%x is unavailable.\n", addr);
return -ENODEV;
}
}
#ifdef CONFIG_FEC_MXC_MDIO_BASE
/*
* The i.MX28 has two ethernet interfaces, but they are not equal.
* Only the first one can access the MDIO bus.
*/
base_mii = CONFIG_FEC_MXC_MDIO_BASE;
#else
base_mii = addr;
#endif
debug("eth_init: fec_probe(bd, %i, %i) @ %08x\n", dev_id, phy_id, addr);
bus = fec_get_miibus(base_mii, dev_id);
if (!bus)
return -ENOMEM;
#ifdef CONFIG_PHYLIB
phydev = phy_find_by_mask(bus, 1 << phy_id, PHY_INTERFACE_MODE_RGMII);
if (!phydev) {
mdio_unregister(bus);
free(bus);
return -ENOMEM;
}
ret = fec_probe(bd, dev_id, addr, bus, phydev);
#else
ret = fec_probe(bd, dev_id, addr, bus, phy_id);
#endif
if (ret) {
#ifdef CONFIG_PHYLIB
free(phydev);
#endif
mdio_unregister(bus);
free(bus);
}
return ret;
}
#ifdef CONFIG_FEC_MXC_PHYADDR
int fecmxc_initialize(struct bd_info *bd)
{
return fecmxc_initialize_multi(bd, -1, CONFIG_FEC_MXC_PHYADDR,
IMX_FEC_BASE);
}
#endif
#ifndef CONFIG_PHYLIB
int fecmxc_register_mii_postcall(struct eth_device *dev, int (*cb)(int))
{
struct fec_priv *fec = (struct fec_priv *)dev->priv;
fec->mii_postcall = cb;
return 0;
}
#endif
#else
static int fecmxc_read_rom_hwaddr(struct udevice *dev)
{
struct fec_priv *priv = dev_get_priv(dev);
struct eth_pdata *pdata = dev_get_plat(dev);
return fec_get_hwaddr(priv->dev_id, pdata->enetaddr);
}
static int fecmxc_free_pkt(struct udevice *dev, uchar *packet, int length)
{
if (packet)
free(packet);
return 0;
}
static const struct eth_ops fecmxc_ops = {
.start = fecmxc_init,
.send = fecmxc_send,
.recv = fecmxc_recv,
.free_pkt = fecmxc_free_pkt,
.stop = fecmxc_halt,
.write_hwaddr = fecmxc_set_hwaddr,
.read_rom_hwaddr = fecmxc_read_rom_hwaddr,
};
static int device_get_phy_addr(struct fec_priv *priv, struct udevice *dev)
{
struct ofnode_phandle_args phandle_args;
int reg, ret;
ret = dev_read_phandle_with_args(dev, "phy-handle", NULL, 0, 0,
&phandle_args);
if (ret) {
debug("Failed to find phy-handle (err = %d\n)", ret);
return ret;
}
if (!ofnode_is_available(phandle_args.node))
return -ENOENT;
priv->phy_of_node = phandle_args.node;
reg = ofnode_read_u32_default(phandle_args.node, "reg", 0);
return reg;
}
static int fec_phy_init(struct fec_priv *priv, struct udevice *dev)
{
struct phy_device *phydev;
int addr;
addr = device_get_phy_addr(priv, dev);
#ifdef CONFIG_FEC_MXC_PHYADDR
addr = CONFIG_FEC_MXC_PHYADDR;
#endif
phydev = phy_connect(priv->bus, addr, dev, priv->interface);
if (!phydev)
return -ENODEV;
priv->phydev = phydev;
priv->phydev->node = priv->phy_of_node;
phy_config(phydev);
return 0;
}
#if CONFIG_IS_ENABLED(DM_GPIO)
/* FEC GPIO reset */
static void fec_gpio_reset(struct fec_priv *priv)
{
debug("fec_gpio_reset: fec_gpio_reset(dev)\n");
if (dm_gpio_is_valid(&priv->phy_reset_gpio)) {
dm_gpio_set_value(&priv->phy_reset_gpio, 1);
mdelay(priv->reset_delay);
dm_gpio_set_value(&priv->phy_reset_gpio, 0);
if (priv->reset_post_delay)
mdelay(priv->reset_post_delay);
}
}
#endif
static int fecmxc_probe(struct udevice *dev)
{
struct eth_pdata *pdata = dev_get_plat(dev);
struct fec_priv *priv = dev_get_priv(dev);
struct mii_dev *bus = NULL;
uint32_t start;
int ret;
if (CONFIG_IS_ENABLED(IMX_MODULE_FUSE)) {
if (enet_fused((ulong)priv->eth)) {
printf("SoC fuse indicates Ethernet@0x%lx is unavailable.\n", (ulong)priv->eth);
return -ENODEV;
}
}
if (IS_ENABLED(CONFIG_IMX8)) {
ret = clk_get_by_name(dev, "ipg", &priv->ipg_clk);
if (ret < 0) {
debug("Can't get FEC ipg clk: %d\n", ret);
return ret;
}
ret = clk_enable(&priv->ipg_clk);
if (ret < 0) {
debug("Can't enable FEC ipg clk: %d\n", ret);
return ret;
}
priv->clk_rate = clk_get_rate(&priv->ipg_clk);
} else if (CONFIG_IS_ENABLED(CLK_CCF)) {
ret = clk_get_by_name(dev, "ipg", &priv->ipg_clk);
if (ret < 0) {
debug("Can't get FEC ipg clk: %d\n", ret);
return ret;
}
ret = clk_enable(&priv->ipg_clk);
if(ret)
return ret;
ret = clk_get_by_name(dev, "ahb", &priv->ahb_clk);
if (ret < 0) {
debug("Can't get FEC ahb clk: %d\n", ret);
return ret;
}
ret = clk_enable(&priv->ahb_clk);
if (ret)
return ret;
ret = clk_get_by_name(dev, "enet_out", &priv->clk_enet_out);
if (!ret) {
ret = clk_enable(&priv->clk_enet_out);
if (ret)
return ret;
}
ret = clk_get_by_name(dev, "enet_clk_ref", &priv->clk_ref);
if (!ret) {
ret = clk_enable(&priv->clk_ref);
if (ret)
return ret;
}
ret = clk_get_by_name(dev, "ptp", &priv->clk_ptp);
if (!ret) {
ret = clk_enable(&priv->clk_ptp);
if (ret)
return ret;
}
priv->clk_rate = clk_get_rate(&priv->ipg_clk);
}
ret = fec_alloc_descs(priv);
if (ret)
return ret;
#ifdef CONFIG_DM_REGULATOR
if (priv->phy_supply) {
ret = regulator_set_enable(priv->phy_supply, true);
if (ret) {
printf("%s: Error enabling phy supply\n", dev->name);
return ret;
}
}
#endif
#if CONFIG_IS_ENABLED(DM_GPIO)
fec_gpio_reset(priv);
#endif
/* Reset chip. */
writel(readl(&priv->eth->ecntrl) | FEC_ECNTRL_RESET,
&priv->eth->ecntrl);
start = get_timer(0);
while (readl(&priv->eth->ecntrl) & FEC_ECNTRL_RESET) {
if (get_timer(start) > (CONFIG_SYS_HZ * 5)) {
printf("FEC MXC: Timeout reseting chip\n");
goto err_timeout;
}
udelay(10);
}
fec_reg_setup(priv);
priv->dev_id = dev_seq(dev);
#ifdef CONFIG_DM_ETH_PHY
bus = eth_phy_get_mdio_bus(dev);
#endif
if (!bus) {
#ifdef CONFIG_FEC_MXC_MDIO_BASE
bus = fec_get_miibus((ulong)CONFIG_FEC_MXC_MDIO_BASE,
dev_seq(dev));
#else
bus = fec_get_miibus((ulong)priv->eth, dev_seq(dev));
#endif
}
if (!bus) {
ret = -ENOMEM;
goto err_mii;
}
#ifdef CONFIG_DM_ETH_PHY
eth_phy_set_mdio_bus(dev, bus);
#endif
priv->bus = bus;
priv->interface = pdata->phy_interface;
switch (priv->interface) {
case PHY_INTERFACE_MODE_MII:
priv->xcv_type = MII100;
break;
case PHY_INTERFACE_MODE_RMII:
priv->xcv_type = RMII;
break;
case PHY_INTERFACE_MODE_RGMII:
case PHY_INTERFACE_MODE_RGMII_ID:
case PHY_INTERFACE_MODE_RGMII_RXID:
case PHY_INTERFACE_MODE_RGMII_TXID:
priv->xcv_type = RGMII;
break;
default:
priv->xcv_type = CONFIG_FEC_XCV_TYPE;
printf("Unsupported interface type %d defaulting to %d\n",
priv->interface, priv->xcv_type);
break;
}
ret = fec_phy_init(priv, dev);
if (ret)
goto err_phy;
return 0;
err_phy:
mdio_unregister(bus);
free(bus);
err_mii:
err_timeout:
fec_free_descs(priv);
return ret;
}
static int fecmxc_remove(struct udevice *dev)
{
struct fec_priv *priv = dev_get_priv(dev);
free(priv->phydev);
fec_free_descs(priv);
mdio_unregister(priv->bus);
mdio_free(priv->bus);
#ifdef CONFIG_DM_REGULATOR
if (priv->phy_supply)
regulator_set_enable(priv->phy_supply, false);
#endif
return 0;
}
static int fecmxc_of_to_plat(struct udevice *dev)
{
int ret = 0;
struct eth_pdata *pdata = dev_get_plat(dev);
struct fec_priv *priv = dev_get_priv(dev);
const char *phy_mode;
pdata->iobase = dev_read_addr(dev);
priv->eth = (struct ethernet_regs *)pdata->iobase;
pdata->phy_interface = -1;
phy_mode = fdt_getprop(gd->fdt_blob, dev_of_offset(dev), "phy-mode",
NULL);
if (phy_mode)
pdata->phy_interface = phy_get_interface_by_name(phy_mode);
if (pdata->phy_interface == -1) {
debug("%s: Invalid PHY interface '%s'\n", __func__, phy_mode);
return -EINVAL;
}
#ifdef CONFIG_DM_REGULATOR
device_get_supply_regulator(dev, "phy-supply", &priv->phy_supply);
#endif
#if CONFIG_IS_ENABLED(DM_GPIO)
ret = gpio_request_by_name(dev, "phy-reset-gpios", 0,
&priv->phy_reset_gpio, GPIOD_IS_OUT);
if (ret < 0)
return 0; /* property is optional, don't return error! */
priv->reset_delay = dev_read_u32_default(dev, "phy-reset-duration", 1);
if (priv->reset_delay > 1000) {
printf("FEC MXC: phy reset duration should be <= 1000ms\n");
/* property value wrong, use default value */
priv->reset_delay = 1;
}
priv->reset_post_delay = dev_read_u32_default(dev,
"phy-reset-post-delay",
0);
if (priv->reset_post_delay > 1000) {
printf("FEC MXC: phy reset post delay should be <= 1000ms\n");
/* property value wrong, use default value */
priv->reset_post_delay = 0;
}
#endif
return 0;
}
static const struct udevice_id fecmxc_ids[] = {
{ .compatible = "fsl,imx28-fec" },
{ .compatible = "fsl,imx6q-fec" },
{ .compatible = "fsl,imx6sl-fec" },
{ .compatible = "fsl,imx6sx-fec" },
{ .compatible = "fsl,imx6ul-fec" },
{ .compatible = "fsl,imx53-fec" },
{ .compatible = "fsl,imx7d-fec" },
{ .compatible = "fsl,mvf600-fec" },
{ }
};
U_BOOT_DRIVER(fecmxc_gem) = {
.name = "fecmxc",
.id = UCLASS_ETH,
.of_match = fecmxc_ids,
.of_to_plat = fecmxc_of_to_plat,
.probe = fecmxc_probe,
.remove = fecmxc_remove,
.ops = &fecmxc_ops,
.priv_auto = sizeof(struct fec_priv),
.plat_auto = sizeof(struct eth_pdata),
};
#endif