net: ks8851_mll: add ethernet support
The device interface is 16 bits wide.
All the available packets are read from the incoming fifo.
Signed-off-by: Roberto Cerati <roberto.cerati@bticino.it>
Signed-off-by: Raffaele Recalcati <raffaele.recalcati@bticino.it>
[voice.shen@atmel.com: address comments from review results]
[voice.shen@atmel.com: clean up for submit]
Signed-off-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Raffaele Recalcati <raffaele.recalcati@bticino.it>
diff --git a/drivers/net/ks8851_mll.c b/drivers/net/ks8851_mll.c
new file mode 100644
index 0000000..b02d59a
--- /dev/null
+++ b/drivers/net/ks8851_mll.c
@@ -0,0 +1,645 @@
+/*
+ * Micrel KS8851_MLL 16bit Network driver
+ * Copyright (c) 2011 Roberto Cerati <roberto.cerati@bticino.it>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <asm/io.h>
+#include <common.h>
+#include <command.h>
+#include <malloc.h>
+#include <net.h>
+#include <miiphy.h>
+
+#include "ks8851_mll.h"
+
+#define DRIVERNAME "ks8851_mll"
+
+#define MAX_RECV_FRAMES 32
+#define MAX_BUF_SIZE 2048
+#define TX_BUF_SIZE 2000
+#define RX_BUF_SIZE 2000
+
+static const struct chip_id chip_ids[] = {
+ {CIDER_ID, "KSZ8851"},
+ {0, NULL},
+};
+
+/*
+ * union ks_tx_hdr - tx header data
+ * @txb: The header as bytes
+ * @txw: The header as 16bit, little-endian words
+ *
+ * A dual representation of the tx header data to allow
+ * access to individual bytes, and to allow 16bit accesses
+ * with 16bit alignment.
+ */
+union ks_tx_hdr {
+ u8 txb[4];
+ __le16 txw[2];
+};
+
+/*
+ * struct ks_net - KS8851 driver private data
+ * @net_device : The network device we're bound to
+ * @txh : temporaly buffer to save status/length.
+ * @frame_head_info : frame header information for multi-pkt rx.
+ * @statelock : Lock on this structure for tx list.
+ * @msg_enable : The message flags controlling driver output (see ethtool).
+ * @frame_cnt : number of frames received.
+ * @bus_width : i/o bus width.
+ * @irq : irq number assigned to this device.
+ * @rc_rxqcr : Cached copy of KS_RXQCR.
+ * @rc_txcr : Cached copy of KS_TXCR.
+ * @rc_ier : Cached copy of KS_IER.
+ * @sharedbus : Multipex(addr and data bus) mode indicator.
+ * @cmd_reg_cache : command register cached.
+ * @cmd_reg_cache_int : command register cached. Used in the irq handler.
+ * @promiscuous : promiscuous mode indicator.
+ * @all_mcast : mutlicast indicator.
+ * @mcast_lst_size : size of multicast list.
+ * @mcast_lst : multicast list.
+ * @mcast_bits : multicast enabed.
+ * @mac_addr : MAC address assigned to this device.
+ * @fid : frame id.
+ * @extra_byte : number of extra byte prepended rx pkt.
+ * @enabled : indicator this device works.
+ */
+
+/* Receive multiplex framer header info */
+struct type_frame_head {
+ u16 sts; /* Frame status */
+ u16 len; /* Byte count */
+} fr_h_i[MAX_RECV_FRAMES];
+
+struct ks_net {
+ struct net_device *netdev;
+ union ks_tx_hdr txh;
+ struct type_frame_head *frame_head_info;
+ u32 msg_enable;
+ u32 frame_cnt;
+ int bus_width;
+ int irq;
+ u16 rc_rxqcr;
+ u16 rc_txcr;
+ u16 rc_ier;
+ u16 sharedbus;
+ u16 cmd_reg_cache;
+ u16 cmd_reg_cache_int;
+ u16 promiscuous;
+ u16 all_mcast;
+ u16 mcast_lst_size;
+ u8 mcast_lst[MAX_MCAST_LST][MAC_ADDR_LEN];
+ u8 mcast_bits[HW_MCAST_SIZE];
+ u8 mac_addr[6];
+ u8 fid;
+ u8 extra_byte;
+ u8 enabled;
+} ks_str, *ks;
+
+#define BE3 0x8000 /* Byte Enable 3 */
+#define BE2 0x4000 /* Byte Enable 2 */
+#define BE1 0x2000 /* Byte Enable 1 */
+#define BE0 0x1000 /* Byte Enable 0 */
+
+static u8 ks_rdreg8(struct eth_device *dev, u16 offset)
+{
+ u8 shift_bit = offset & 0x03;
+ u8 shift_data = (offset & 1) << 3;
+
+ writew(offset | (BE0 << shift_bit), dev->iobase + 2);
+
+ return (u8)(readw(dev->iobase) >> shift_data);
+}
+
+static u16 ks_rdreg16(struct eth_device *dev, u16 offset)
+{
+ writew(offset | ((BE1 | BE0) << (offset & 0x02)), dev->iobase + 2);
+
+ return readw(dev->iobase);
+}
+
+static void ks_wrreg8(struct eth_device *dev, u16 offset, u8 val)
+{
+ u8 shift_bit = (offset & 0x03);
+ u16 value_write = (u16)(val << ((offset & 1) << 3));
+
+ writew(offset | (BE0 << shift_bit), dev->iobase + 2);
+ writew(value_write, dev->iobase);
+}
+
+static void ks_wrreg16(struct eth_device *dev, u16 offset, u16 val)
+{
+ writew(offset | ((BE1 | BE0) << (offset & 0x02)), dev->iobase + 2);
+ writew(val, dev->iobase);
+}
+
+/*
+ * ks_inblk - read a block of data from QMU. This is called after sudo DMA mode
+ * enabled.
+ * @ks: The chip state
+ * @wptr: buffer address to save data
+ * @len: length in byte to read
+ */
+static inline void ks_inblk(struct eth_device *dev, u16 *wptr, u32 len)
+{
+ len >>= 1;
+
+ while (len--)
+ *wptr++ = readw(dev->iobase);
+}
+
+/*
+ * ks_outblk - write data to QMU. This is called after sudo DMA mode enabled.
+ * @ks: The chip information
+ * @wptr: buffer address
+ * @len: length in byte to write
+ */
+static inline void ks_outblk(struct eth_device *dev, u16 *wptr, u32 len)
+{
+ len >>= 1;
+
+ while (len--)
+ writew(*wptr++, dev->iobase);
+}
+
+static void ks_enable_int(struct eth_device *dev)
+{
+ ks_wrreg16(dev, KS_IER, ks->rc_ier);
+}
+
+static void ks_set_powermode(struct eth_device *dev, unsigned pwrmode)
+{
+ unsigned pmecr;
+
+ ks_rdreg16(dev, KS_GRR);
+ pmecr = ks_rdreg16(dev, KS_PMECR);
+ pmecr &= ~PMECR_PM_MASK;
+ pmecr |= pwrmode;
+
+ ks_wrreg16(dev, KS_PMECR, pmecr);
+}
+
+/*
+ * ks_read_config - read chip configuration of bus width.
+ * @ks: The chip information
+ */
+static void ks_read_config(struct eth_device *dev)
+{
+ u16 reg_data = 0;
+
+ /* Regardless of bus width, 8 bit read should always work. */
+ reg_data = ks_rdreg8(dev, KS_CCR) & 0x00FF;
+ reg_data |= ks_rdreg8(dev, KS_CCR + 1) << 8;
+
+ /* addr/data bus are multiplexed */
+ ks->sharedbus = (reg_data & CCR_SHARED) == CCR_SHARED;
+
+ /*
+ * There are garbage data when reading data from QMU,
+ * depending on bus-width.
+ */
+ if (reg_data & CCR_8BIT) {
+ ks->bus_width = ENUM_BUS_8BIT;
+ ks->extra_byte = 1;
+ } else if (reg_data & CCR_16BIT) {
+ ks->bus_width = ENUM_BUS_16BIT;
+ ks->extra_byte = 2;
+ } else {
+ ks->bus_width = ENUM_BUS_32BIT;
+ ks->extra_byte = 4;
+ }
+}
+
+/*
+ * ks_soft_reset - issue one of the soft reset to the device
+ * @ks: The device state.
+ * @op: The bit(s) to set in the GRR
+ *
+ * Issue the relevant soft-reset command to the device's GRR register
+ * specified by @op.
+ *
+ * Note, the delays are in there as a caution to ensure that the reset
+ * has time to take effect and then complete. Since the datasheet does
+ * not currently specify the exact sequence, we have chosen something
+ * that seems to work with our device.
+ */
+static void ks_soft_reset(struct eth_device *dev, unsigned op)
+{
+ /* Disable interrupt first */
+ ks_wrreg16(dev, KS_IER, 0x0000);
+ ks_wrreg16(dev, KS_GRR, op);
+ mdelay(10); /* wait a short time to effect reset */
+ ks_wrreg16(dev, KS_GRR, 0);
+ mdelay(1); /* wait for condition to clear */
+}
+
+void ks_enable_qmu(struct eth_device *dev)
+{
+ u16 w;
+
+ w = ks_rdreg16(dev, KS_TXCR);
+
+ /* Enables QMU Transmit (TXCR). */
+ ks_wrreg16(dev, KS_TXCR, w | TXCR_TXE);
+
+ /* Enable RX Frame Count Threshold and Auto-Dequeue RXQ Frame */
+ w = ks_rdreg16(dev, KS_RXQCR);
+ ks_wrreg16(dev, KS_RXQCR, w | RXQCR_RXFCTE);
+
+ /* Enables QMU Receive (RXCR1). */
+ w = ks_rdreg16(dev, KS_RXCR1);
+ ks_wrreg16(dev, KS_RXCR1, w | RXCR1_RXE);
+}
+
+static void ks_disable_qmu(struct eth_device *dev)
+{
+ u16 w;
+
+ w = ks_rdreg16(dev, KS_TXCR);
+
+ /* Disables QMU Transmit (TXCR). */
+ w &= ~TXCR_TXE;
+ ks_wrreg16(dev, KS_TXCR, w);
+
+ /* Disables QMU Receive (RXCR1). */
+ w = ks_rdreg16(dev, KS_RXCR1);
+ w &= ~RXCR1_RXE;
+ ks_wrreg16(dev, KS_RXCR1, w);
+}
+
+static inline void ks_read_qmu(struct eth_device *dev, u16 *buf, u32 len)
+{
+ u32 r = ks->extra_byte & 0x1;
+ u32 w = ks->extra_byte - r;
+
+ /* 1. set sudo DMA mode */
+ ks_wrreg16(dev, KS_RXFDPR, RXFDPR_RXFPAI);
+ ks_wrreg8(dev, KS_RXQCR, (ks->rc_rxqcr | RXQCR_SDA) & 0xff);
+
+ /*
+ * 2. read prepend data
+ *
+ * read 4 + extra bytes and discard them.
+ * extra bytes for dummy, 2 for status, 2 for len
+ */
+
+ if (r)
+ ks_rdreg8(dev, 0);
+
+ ks_inblk(dev, buf, w + 2 + 2);
+
+ /* 3. read pkt data */
+ ks_inblk(dev, buf, ALIGN(len, 4));
+
+ /* 4. reset sudo DMA Mode */
+ ks_wrreg8(dev, KS_RXQCR, (ks->rc_rxqcr & ~RXQCR_SDA) & 0xff);
+}
+
+static void ks_rcv(struct eth_device *dev, uchar **pv_data)
+{
+ struct type_frame_head *frame_hdr = ks->frame_head_info;
+ int i;
+
+ ks->frame_cnt = ks_rdreg16(dev, KS_RXFCTR) >> 8;
+
+ /* read all header information */
+ for (i = 0; i < ks->frame_cnt; i++) {
+ /* Checking Received packet status */
+ frame_hdr->sts = ks_rdreg16(dev, KS_RXFHSR);
+ /* Get packet len from hardware */
+ frame_hdr->len = ks_rdreg16(dev, KS_RXFHBCR);
+ frame_hdr++;
+ }
+
+ frame_hdr = ks->frame_head_info;
+ while (ks->frame_cnt--) {
+ if ((frame_hdr->sts & RXFSHR_RXFV) &&
+ (frame_hdr->len < RX_BUF_SIZE) &&
+ frame_hdr->len) {
+ /* read data block including CRC 4 bytes */
+ ks_read_qmu(dev, (u16 *)(*pv_data), frame_hdr->len);
+
+ /* NetRxPackets buffer size is ok (*pv_data pointer) */
+ NetReceive(*pv_data, frame_hdr->len);
+ pv_data++;
+ } else {
+ ks_wrreg16(dev, KS_RXQCR, (ks->rc_rxqcr | RXQCR_RRXEF));
+ printf(DRIVERNAME ": bad packet\n");
+ }
+ frame_hdr++;
+ }
+}
+
+/*
+ * ks_read_selftest - read the selftest memory info.
+ * @ks: The device state
+ *
+ * Read and check the TX/RX memory selftest information.
+ */
+static int ks_read_selftest(struct eth_device *dev)
+{
+ u16 both_done = MBIR_TXMBF | MBIR_RXMBF;
+ u16 mbir;
+ int ret = 0;
+
+ mbir = ks_rdreg16(dev, KS_MBIR);
+
+ if ((mbir & both_done) != both_done) {
+ printf(DRIVERNAME ": Memory selftest not finished\n");
+ return 0;
+ }
+
+ if (mbir & MBIR_TXMBFA) {
+ printf(DRIVERNAME ": TX memory selftest fails\n");
+ ret |= 1;
+ }
+
+ if (mbir & MBIR_RXMBFA) {
+ printf(DRIVERNAME ": RX memory selftest fails\n");
+ ret |= 2;
+ }
+
+ debug(DRIVERNAME ": the selftest passes\n");
+
+ return ret;
+}
+
+static void ks_setup(struct eth_device *dev)
+{
+ u16 w;
+
+ /* Setup Transmit Frame Data Pointer Auto-Increment (TXFDPR) */
+ ks_wrreg16(dev, KS_TXFDPR, TXFDPR_TXFPAI);
+
+ /* Setup Receive Frame Data Pointer Auto-Increment */
+ ks_wrreg16(dev, KS_RXFDPR, RXFDPR_RXFPAI);
+
+ /* Setup Receive Frame Threshold - 1 frame (RXFCTFC) */
+ ks_wrreg16(dev, KS_RXFCTR, 1 & RXFCTR_THRESHOLD_MASK);
+
+ /* Setup RxQ Command Control (RXQCR) */
+ ks->rc_rxqcr = RXQCR_CMD_CNTL;
+ ks_wrreg16(dev, KS_RXQCR, ks->rc_rxqcr);
+
+ /*
+ * set the force mode to half duplex, default is full duplex
+ * because if the auto-negotiation fails, most switch uses
+ * half-duplex.
+ */
+ w = ks_rdreg16(dev, KS_P1MBCR);
+ w &= ~P1MBCR_FORCE_FDX;
+ ks_wrreg16(dev, KS_P1MBCR, w);
+
+ w = TXCR_TXFCE | TXCR_TXPE | TXCR_TXCRC | TXCR_TCGIP;
+ ks_wrreg16(dev, KS_TXCR, w);
+
+ w = RXCR1_RXFCE | RXCR1_RXBE | RXCR1_RXUE | RXCR1_RXME | RXCR1_RXIPFCC;
+
+ /* Normal mode */
+ w |= RXCR1_RXPAFMA;
+
+ ks_wrreg16(dev, KS_RXCR1, w);
+}
+
+static void ks_setup_int(struct eth_device *dev)
+{
+ ks->rc_ier = 0x00;
+
+ /* Clear the interrupts status of the hardware. */
+ ks_wrreg16(dev, KS_ISR, 0xffff);
+
+ /* Enables the interrupts of the hardware. */
+ ks->rc_ier = (IRQ_LCI | IRQ_TXI | IRQ_RXI);
+}
+
+static int ks8851_mll_detect_chip(struct eth_device *dev)
+{
+ unsigned short val, i;
+
+ ks_read_config(dev);
+
+ val = ks_rdreg16(dev, KS_CIDER);
+
+ if (val == 0xffff) {
+ /* Special case -- no chip present */
+ printf(DRIVERNAME ": is chip mounted ?\n");
+ return -1;
+ } else if ((val & 0xfff0) != CIDER_ID) {
+ printf(DRIVERNAME ": Invalid chip id 0x%04x\n", val);
+ return -1;
+ }
+
+ debug("Read back KS8851 id 0x%x\n", val);
+
+ /* only one entry in the table */
+ val &= 0xfff0;
+ for (i = 0; chip_ids[i].id != 0; i++) {
+ if (chip_ids[i].id == val)
+ break;
+ }
+ if (!chip_ids[i].id) {
+ printf(DRIVERNAME ": Unknown chip ID %04x\n", val);
+ return -1;
+ }
+
+ dev->priv = (void *)&chip_ids[i];
+
+ return 0;
+}
+
+static void ks8851_mll_reset(struct eth_device *dev)
+{
+ /* wake up powermode to normal mode */
+ ks_set_powermode(dev, PMECR_PM_NORMAL);
+ mdelay(1); /* wait for normal mode to take effect */
+
+ /* Disable interrupt and reset */
+ ks_soft_reset(dev, GRR_GSR);
+
+ /* turn off the IRQs and ack any outstanding */
+ ks_wrreg16(dev, KS_IER, 0x0000);
+ ks_wrreg16(dev, KS_ISR, 0xffff);
+
+ /* shutdown RX/TX QMU */
+ ks_disable_qmu(dev);
+}
+
+static void ks8851_mll_phy_configure(struct eth_device *dev)
+{
+ u16 data;
+
+ ks_setup(dev);
+ ks_setup_int(dev);
+
+ /* Probing the phy */
+ data = ks_rdreg16(dev, KS_OBCR);
+ ks_wrreg16(dev, KS_OBCR, data | OBCR_ODS_16MA);
+
+ debug(DRIVERNAME ": phy initialized\n");
+}
+
+static void ks8851_mll_enable(struct eth_device *dev)
+{
+ ks_wrreg16(dev, KS_ISR, 0xffff);
+ ks_enable_int(dev);
+ ks_enable_qmu(dev);
+}
+
+static int ks8851_mll_init(struct eth_device *dev, bd_t *bd)
+{
+ struct chip_id *id = dev->priv;
+
+ debug(DRIVERNAME ": detected %s controller\n", id->name);
+
+ if (ks_read_selftest(dev)) {
+ printf(DRIVERNAME ": Selftest failed\n");
+ return -1;
+ }
+
+ ks8851_mll_reset(dev);
+
+ /* Configure the PHY, initialize the link state */
+ ks8851_mll_phy_configure(dev);
+
+ /* static allocation of private informations */
+ ks->frame_head_info = fr_h_i;
+
+ /* Turn on Tx + Rx */
+ ks8851_mll_enable(dev);
+
+ return 0;
+}
+
+static void ks_write_qmu(struct eth_device *dev, u8 *pdata, u16 len)
+{
+ /* start header at txb[0] to align txw entries */
+ ks->txh.txw[0] = 0;
+ ks->txh.txw[1] = cpu_to_le16(len);
+
+ /* 1. set sudo-DMA mode */
+ ks_wrreg16(dev, KS_TXFDPR, TXFDPR_TXFPAI);
+ ks_wrreg8(dev, KS_RXQCR, (ks->rc_rxqcr | RXQCR_SDA) & 0xff);
+ /* 2. write status/lenth info */
+ ks_outblk(dev, ks->txh.txw, 4);
+ /* 3. write pkt data */
+ ks_outblk(dev, (u16 *)pdata, ALIGN(len, 4));
+ /* 4. reset sudo-DMA mode */
+ ks_wrreg8(dev, KS_RXQCR, (ks->rc_rxqcr & ~RXQCR_SDA) & 0xff);
+ /* 5. Enqueue Tx(move the pkt from TX buffer into TXQ) */
+ ks_wrreg16(dev, KS_TXQCR, TXQCR_METFE);
+ /* 6. wait until TXQCR_METFE is auto-cleared */
+ do { } while (ks_rdreg16(dev, KS_TXQCR) & TXQCR_METFE);
+}
+
+static int ks8851_mll_send(struct eth_device *dev, void *packet, int length)
+{
+ u8 *data = (u8 *)packet;
+ u16 tmplen = (u16)length;
+ u16 retv;
+
+ /*
+ * Extra space are required:
+ * 4 byte for alignment, 4 for status/length, 4 for CRC
+ */
+ retv = ks_rdreg16(dev, KS_TXMIR) & 0x1fff;
+ if (retv >= tmplen + 12) {
+ ks_write_qmu(dev, data, tmplen);
+ return 0;
+ } else {
+ printf(DRIVERNAME ": failed to send packet: No buffer\n");
+ return -1;
+ }
+}
+
+static void ks8851_mll_halt(struct eth_device *dev)
+{
+ ks8851_mll_reset(dev);
+}
+
+/*
+ * Maximum receive ring size; that is, the number of packets
+ * we can buffer before overflow happens. Basically, this just
+ * needs to be enough to prevent a packet being discarded while
+ * we are processing the previous one.
+ */
+static int ks8851_mll_recv(struct eth_device *dev)
+{
+ u16 status;
+
+ status = ks_rdreg16(dev, KS_ISR);
+
+ ks_wrreg16(dev, KS_ISR, status);
+
+ if ((status & IRQ_RXI))
+ ks_rcv(dev, (uchar **)NetRxPackets);
+
+ if ((status & IRQ_LDI)) {
+ u16 pmecr = ks_rdreg16(dev, KS_PMECR);
+ pmecr &= ~PMECR_WKEVT_MASK;
+ ks_wrreg16(dev, KS_PMECR, pmecr | PMECR_WKEVT_LINK);
+ }
+
+ return 0;
+}
+
+static int ks8851_mll_write_hwaddr(struct eth_device *dev)
+{
+ u16 addrl, addrm, addrh;
+
+ addrh = (dev->enetaddr[0] << 8) | dev->enetaddr[1];
+ addrm = (dev->enetaddr[2] << 8) | dev->enetaddr[3];
+ addrl = (dev->enetaddr[4] << 8) | dev->enetaddr[5];
+
+ ks_wrreg16(dev, KS_MARH, addrh);
+ ks_wrreg16(dev, KS_MARM, addrm);
+ ks_wrreg16(dev, KS_MARL, addrl);
+
+ return 0;
+}
+
+int ks8851_mll_initialize(u8 dev_num, int base_addr)
+{
+ struct eth_device *dev;
+
+ dev = malloc(sizeof(*dev));
+ if (!dev) {
+ printf("Error: Failed to allocate memory\n");
+ return -1;
+ }
+ memset(dev, 0, sizeof(*dev));
+
+ dev->iobase = base_addr;
+
+ ks = &ks_str;
+
+ /* Try to detect chip. Will fail if not present. */
+ if (ks8851_mll_detect_chip(dev)) {
+ free(dev);
+ return -1;
+ }
+
+ dev->init = ks8851_mll_init;
+ dev->halt = ks8851_mll_halt;
+ dev->send = ks8851_mll_send;
+ dev->recv = ks8851_mll_recv;
+ dev->write_hwaddr = ks8851_mll_write_hwaddr;
+ sprintf(dev->name, "%s-%hu", DRIVERNAME, dev_num);
+
+ eth_register(dev);
+
+ return 0;
+}