mirror of
https://github.com/qemu/qemu.git
synced 2024-12-13 14:33:31 +08:00
4f67d30b5e
The following patch will need to handle properties registration during class_init time. Let's use a device_class_set_props() setter. spatch --macro-file scripts/cocci-macro-file.h --sp-file ./scripts/coccinelle/qdev-set-props.cocci --keep-comments --in-place --dir . @@ typedef DeviceClass; DeviceClass *d; expression val; @@ - d->props = val + device_class_set_props(d, val) Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20200110153039.1379601-20-marcandre.lureau@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
690 lines
18 KiB
C
690 lines
18 KiB
C
/*
|
|
* QEMU ETRAX Ethernet Controller.
|
|
*
|
|
* Copyright (c) 2008 Edgar E. Iglesias, Axis Communications AB.
|
|
*
|
|
* Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
* of this software and associated documentation files (the "Software"), to deal
|
|
* in the Software without restriction, including without limitation the rights
|
|
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
* copies of the Software, and to permit persons to whom the Software is
|
|
* furnished to do so, subject to the following conditions:
|
|
*
|
|
* The above copyright notice and this permission notice shall be included in
|
|
* all copies or substantial portions of the Software.
|
|
*
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
|
* THE SOFTWARE.
|
|
*/
|
|
|
|
#include "qemu/osdep.h"
|
|
#include "qapi/error.h"
|
|
#include "hw/sysbus.h"
|
|
#include "net/net.h"
|
|
#include "hw/cris/etraxfs.h"
|
|
#include "qemu/error-report.h"
|
|
#include "qemu/module.h"
|
|
#include "trace.h"
|
|
|
|
#define D(x)
|
|
|
|
/* Advertisement control register. */
|
|
#define ADVERTISE_10HALF 0x0020 /* Try for 10mbps half-duplex */
|
|
#define ADVERTISE_10FULL 0x0040 /* Try for 10mbps full-duplex */
|
|
#define ADVERTISE_100HALF 0x0080 /* Try for 100mbps half-duplex */
|
|
#define ADVERTISE_100FULL 0x0100 /* Try for 100mbps full-duplex */
|
|
|
|
/*
|
|
* The MDIO extensions in the TDK PHY model were reversed engineered from the
|
|
* linux driver (PHYID and Diagnostics reg).
|
|
* TODO: Add friendly names for the register nums.
|
|
*/
|
|
struct qemu_phy
|
|
{
|
|
uint32_t regs[32];
|
|
|
|
int link;
|
|
|
|
unsigned int (*read)(struct qemu_phy *phy, unsigned int req);
|
|
void (*write)(struct qemu_phy *phy, unsigned int req, unsigned int data);
|
|
};
|
|
|
|
static unsigned int tdk_read(struct qemu_phy *phy, unsigned int req)
|
|
{
|
|
int regnum;
|
|
unsigned r = 0;
|
|
|
|
regnum = req & 0x1f;
|
|
|
|
switch (regnum) {
|
|
case 1:
|
|
if (!phy->link) {
|
|
break;
|
|
}
|
|
/* MR1. */
|
|
/* Speeds and modes. */
|
|
r |= (1 << 13) | (1 << 14);
|
|
r |= (1 << 11) | (1 << 12);
|
|
r |= (1 << 5); /* Autoneg complete. */
|
|
r |= (1 << 3); /* Autoneg able. */
|
|
r |= (1 << 2); /* link. */
|
|
break;
|
|
case 5:
|
|
/* Link partner ability.
|
|
We are kind; always agree with whatever best mode
|
|
the guest advertises. */
|
|
r = 1 << 14; /* Success. */
|
|
/* Copy advertised modes. */
|
|
r |= phy->regs[4] & (15 << 5);
|
|
/* Autoneg support. */
|
|
r |= 1;
|
|
break;
|
|
case 18:
|
|
{
|
|
/* Diagnostics reg. */
|
|
int duplex = 0;
|
|
int speed_100 = 0;
|
|
|
|
if (!phy->link) {
|
|
break;
|
|
}
|
|
|
|
/* Are we advertising 100 half or 100 duplex ? */
|
|
speed_100 = !!(phy->regs[4] & ADVERTISE_100HALF);
|
|
speed_100 |= !!(phy->regs[4] & ADVERTISE_100FULL);
|
|
|
|
/* Are we advertising 10 duplex or 100 duplex ? */
|
|
duplex = !!(phy->regs[4] & ADVERTISE_100FULL);
|
|
duplex |= !!(phy->regs[4] & ADVERTISE_10FULL);
|
|
r = (speed_100 << 10) | (duplex << 11);
|
|
}
|
|
break;
|
|
|
|
default:
|
|
r = phy->regs[regnum];
|
|
break;
|
|
}
|
|
trace_mdio_phy_read(regnum, r);
|
|
return r;
|
|
}
|
|
|
|
static void
|
|
tdk_write(struct qemu_phy *phy, unsigned int req, unsigned int data)
|
|
{
|
|
int regnum;
|
|
|
|
regnum = req & 0x1f;
|
|
trace_mdio_phy_write(regnum, data);
|
|
switch (regnum) {
|
|
default:
|
|
phy->regs[regnum] = data;
|
|
break;
|
|
}
|
|
}
|
|
|
|
static void
|
|
tdk_reset(struct qemu_phy *phy)
|
|
{
|
|
phy->regs[0] = 0x3100;
|
|
/* PHY Id. */
|
|
phy->regs[2] = 0x0300;
|
|
phy->regs[3] = 0xe400;
|
|
/* Autonegotiation advertisement reg. */
|
|
phy->regs[4] = 0x01E1;
|
|
phy->link = 1;
|
|
}
|
|
|
|
struct qemu_mdio
|
|
{
|
|
/* bus. */
|
|
int mdc;
|
|
int mdio;
|
|
|
|
/* decoder. */
|
|
enum {
|
|
PREAMBLE,
|
|
SOF,
|
|
OPC,
|
|
ADDR,
|
|
REQ,
|
|
TURNAROUND,
|
|
DATA
|
|
} state;
|
|
unsigned int drive;
|
|
|
|
unsigned int cnt;
|
|
unsigned int addr;
|
|
unsigned int opc;
|
|
unsigned int req;
|
|
unsigned int data;
|
|
|
|
struct qemu_phy *devs[32];
|
|
};
|
|
|
|
static void
|
|
mdio_attach(struct qemu_mdio *bus, struct qemu_phy *phy, unsigned int addr)
|
|
{
|
|
bus->devs[addr & 0x1f] = phy;
|
|
}
|
|
|
|
#ifdef USE_THIS_DEAD_CODE
|
|
static void
|
|
mdio_detach(struct qemu_mdio *bus, struct qemu_phy *phy, unsigned int addr)
|
|
{
|
|
bus->devs[addr & 0x1f] = NULL;
|
|
}
|
|
#endif
|
|
|
|
static void mdio_read_req(struct qemu_mdio *bus)
|
|
{
|
|
struct qemu_phy *phy;
|
|
|
|
phy = bus->devs[bus->addr];
|
|
if (phy && phy->read) {
|
|
bus->data = phy->read(phy, bus->req);
|
|
} else {
|
|
bus->data = 0xffff;
|
|
}
|
|
}
|
|
|
|
static void mdio_write_req(struct qemu_mdio *bus)
|
|
{
|
|
struct qemu_phy *phy;
|
|
|
|
phy = bus->devs[bus->addr];
|
|
if (phy && phy->write) {
|
|
phy->write(phy, bus->req, bus->data);
|
|
}
|
|
}
|
|
|
|
static void mdio_cycle(struct qemu_mdio *bus)
|
|
{
|
|
bus->cnt++;
|
|
|
|
trace_mdio_bitbang(bus->mdc, bus->mdio, bus->state, bus->cnt, bus->drive);
|
|
#if 0
|
|
if (bus->mdc) {
|
|
printf("%d", bus->mdio);
|
|
}
|
|
#endif
|
|
switch (bus->state) {
|
|
case PREAMBLE:
|
|
if (bus->mdc) {
|
|
if (bus->cnt >= (32 * 2) && !bus->mdio) {
|
|
bus->cnt = 0;
|
|
bus->state = SOF;
|
|
bus->data = 0;
|
|
}
|
|
}
|
|
break;
|
|
case SOF:
|
|
if (bus->mdc) {
|
|
if (bus->mdio != 1) {
|
|
printf("WARNING: no SOF\n");
|
|
}
|
|
if (bus->cnt == 1*2) {
|
|
bus->cnt = 0;
|
|
bus->opc = 0;
|
|
bus->state = OPC;
|
|
}
|
|
}
|
|
break;
|
|
case OPC:
|
|
if (bus->mdc) {
|
|
bus->opc <<= 1;
|
|
bus->opc |= bus->mdio & 1;
|
|
if (bus->cnt == 2*2) {
|
|
bus->cnt = 0;
|
|
bus->addr = 0;
|
|
bus->state = ADDR;
|
|
}
|
|
}
|
|
break;
|
|
case ADDR:
|
|
if (bus->mdc) {
|
|
bus->addr <<= 1;
|
|
bus->addr |= bus->mdio & 1;
|
|
|
|
if (bus->cnt == 5*2) {
|
|
bus->cnt = 0;
|
|
bus->req = 0;
|
|
bus->state = REQ;
|
|
}
|
|
}
|
|
break;
|
|
case REQ:
|
|
if (bus->mdc) {
|
|
bus->req <<= 1;
|
|
bus->req |= bus->mdio & 1;
|
|
if (bus->cnt == 5*2) {
|
|
bus->cnt = 0;
|
|
bus->state = TURNAROUND;
|
|
}
|
|
}
|
|
break;
|
|
case TURNAROUND:
|
|
if (bus->mdc && bus->cnt == 2*2) {
|
|
bus->mdio = 0;
|
|
bus->cnt = 0;
|
|
|
|
if (bus->opc == 2) {
|
|
bus->drive = 1;
|
|
mdio_read_req(bus);
|
|
bus->mdio = bus->data & 1;
|
|
}
|
|
bus->state = DATA;
|
|
}
|
|
break;
|
|
case DATA:
|
|
if (!bus->mdc) {
|
|
if (bus->drive) {
|
|
bus->mdio = !!(bus->data & (1 << 15));
|
|
bus->data <<= 1;
|
|
}
|
|
} else {
|
|
if (!bus->drive) {
|
|
bus->data <<= 1;
|
|
bus->data |= bus->mdio;
|
|
}
|
|
if (bus->cnt == 16 * 2) {
|
|
bus->cnt = 0;
|
|
bus->state = PREAMBLE;
|
|
if (!bus->drive) {
|
|
mdio_write_req(bus);
|
|
}
|
|
bus->drive = 0;
|
|
}
|
|
}
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
}
|
|
|
|
/* ETRAX-FS Ethernet MAC block starts here. */
|
|
|
|
#define RW_MA0_LO 0x00
|
|
#define RW_MA0_HI 0x01
|
|
#define RW_MA1_LO 0x02
|
|
#define RW_MA1_HI 0x03
|
|
#define RW_GA_LO 0x04
|
|
#define RW_GA_HI 0x05
|
|
#define RW_GEN_CTRL 0x06
|
|
#define RW_REC_CTRL 0x07
|
|
#define RW_TR_CTRL 0x08
|
|
#define RW_CLR_ERR 0x09
|
|
#define RW_MGM_CTRL 0x0a
|
|
#define R_STAT 0x0b
|
|
#define FS_ETH_MAX_REGS 0x17
|
|
|
|
#define TYPE_ETRAX_FS_ETH "etraxfs-eth"
|
|
#define ETRAX_FS_ETH(obj) \
|
|
OBJECT_CHECK(ETRAXFSEthState, (obj), TYPE_ETRAX_FS_ETH)
|
|
|
|
typedef struct ETRAXFSEthState
|
|
{
|
|
SysBusDevice parent_obj;
|
|
|
|
MemoryRegion mmio;
|
|
NICState *nic;
|
|
NICConf conf;
|
|
|
|
/* Two addrs in the filter. */
|
|
uint8_t macaddr[2][6];
|
|
uint32_t regs[FS_ETH_MAX_REGS];
|
|
|
|
struct etraxfs_dma_client *dma_out;
|
|
struct etraxfs_dma_client *dma_in;
|
|
|
|
/* MDIO bus. */
|
|
struct qemu_mdio mdio_bus;
|
|
unsigned int phyaddr;
|
|
int duplex_mismatch;
|
|
|
|
/* PHY. */
|
|
struct qemu_phy phy;
|
|
} ETRAXFSEthState;
|
|
|
|
static void eth_validate_duplex(ETRAXFSEthState *eth)
|
|
{
|
|
struct qemu_phy *phy;
|
|
unsigned int phy_duplex;
|
|
unsigned int mac_duplex;
|
|
int new_mm = 0;
|
|
|
|
phy = eth->mdio_bus.devs[eth->phyaddr];
|
|
phy_duplex = !!(phy->read(phy, 18) & (1 << 11));
|
|
mac_duplex = !!(eth->regs[RW_REC_CTRL] & 128);
|
|
|
|
if (mac_duplex != phy_duplex) {
|
|
new_mm = 1;
|
|
}
|
|
|
|
if (eth->regs[RW_GEN_CTRL] & 1) {
|
|
if (new_mm != eth->duplex_mismatch) {
|
|
if (new_mm) {
|
|
printf("HW: WARNING ETH duplex mismatch MAC=%d PHY=%d\n",
|
|
mac_duplex, phy_duplex);
|
|
} else {
|
|
printf("HW: ETH duplex ok.\n");
|
|
}
|
|
}
|
|
eth->duplex_mismatch = new_mm;
|
|
}
|
|
}
|
|
|
|
static uint64_t
|
|
eth_read(void *opaque, hwaddr addr, unsigned int size)
|
|
{
|
|
ETRAXFSEthState *eth = opaque;
|
|
uint32_t r = 0;
|
|
|
|
addr >>= 2;
|
|
|
|
switch (addr) {
|
|
case R_STAT:
|
|
r = eth->mdio_bus.mdio & 1;
|
|
break;
|
|
default:
|
|
r = eth->regs[addr];
|
|
D(printf("%s %x\n", __func__, addr * 4));
|
|
break;
|
|
}
|
|
return r;
|
|
}
|
|
|
|
static void eth_update_ma(ETRAXFSEthState *eth, int ma)
|
|
{
|
|
int reg;
|
|
int i = 0;
|
|
|
|
ma &= 1;
|
|
|
|
reg = RW_MA0_LO;
|
|
if (ma) {
|
|
reg = RW_MA1_LO;
|
|
}
|
|
|
|
eth->macaddr[ma][i++] = eth->regs[reg];
|
|
eth->macaddr[ma][i++] = eth->regs[reg] >> 8;
|
|
eth->macaddr[ma][i++] = eth->regs[reg] >> 16;
|
|
eth->macaddr[ma][i++] = eth->regs[reg] >> 24;
|
|
eth->macaddr[ma][i++] = eth->regs[reg + 1];
|
|
eth->macaddr[ma][i] = eth->regs[reg + 1] >> 8;
|
|
|
|
D(printf("set mac%d=%x.%x.%x.%x.%x.%x\n", ma,
|
|
eth->macaddr[ma][0], eth->macaddr[ma][1],
|
|
eth->macaddr[ma][2], eth->macaddr[ma][3],
|
|
eth->macaddr[ma][4], eth->macaddr[ma][5]));
|
|
}
|
|
|
|
static void
|
|
eth_write(void *opaque, hwaddr addr,
|
|
uint64_t val64, unsigned int size)
|
|
{
|
|
ETRAXFSEthState *eth = opaque;
|
|
uint32_t value = val64;
|
|
|
|
addr >>= 2;
|
|
switch (addr) {
|
|
case RW_MA0_LO:
|
|
case RW_MA0_HI:
|
|
eth->regs[addr] = value;
|
|
eth_update_ma(eth, 0);
|
|
break;
|
|
case RW_MA1_LO:
|
|
case RW_MA1_HI:
|
|
eth->regs[addr] = value;
|
|
eth_update_ma(eth, 1);
|
|
break;
|
|
|
|
case RW_MGM_CTRL:
|
|
/* Attach an MDIO/PHY abstraction. */
|
|
if (value & 2) {
|
|
eth->mdio_bus.mdio = value & 1;
|
|
}
|
|
if (eth->mdio_bus.mdc != (value & 4)) {
|
|
mdio_cycle(ð->mdio_bus);
|
|
eth_validate_duplex(eth);
|
|
}
|
|
eth->mdio_bus.mdc = !!(value & 4);
|
|
eth->regs[addr] = value;
|
|
break;
|
|
|
|
case RW_REC_CTRL:
|
|
eth->regs[addr] = value;
|
|
eth_validate_duplex(eth);
|
|
break;
|
|
|
|
default:
|
|
eth->regs[addr] = value;
|
|
D(printf("%s %x %x\n", __func__, addr, value));
|
|
break;
|
|
}
|
|
}
|
|
|
|
/* The ETRAX FS has a groupt address table (GAT) which works like a k=1 bloom
|
|
filter dropping group addresses we have not joined. The filter has 64
|
|
bits (m). The has function is a simple nible xor of the group addr. */
|
|
static int eth_match_groupaddr(ETRAXFSEthState *eth, const unsigned char *sa)
|
|
{
|
|
unsigned int hsh;
|
|
int m_individual = eth->regs[RW_REC_CTRL] & 4;
|
|
int match;
|
|
|
|
/* First bit on the wire of a MAC address signals multicast or
|
|
physical address. */
|
|
if (!m_individual && !(sa[0] & 1)) {
|
|
return 0;
|
|
}
|
|
|
|
/* Calculate the hash index for the GA registers. */
|
|
hsh = 0;
|
|
hsh ^= (*sa) & 0x3f;
|
|
hsh ^= ((*sa) >> 6) & 0x03;
|
|
++sa;
|
|
hsh ^= ((*sa) << 2) & 0x03c;
|
|
hsh ^= ((*sa) >> 4) & 0xf;
|
|
++sa;
|
|
hsh ^= ((*sa) << 4) & 0x30;
|
|
hsh ^= ((*sa) >> 2) & 0x3f;
|
|
++sa;
|
|
hsh ^= (*sa) & 0x3f;
|
|
hsh ^= ((*sa) >> 6) & 0x03;
|
|
++sa;
|
|
hsh ^= ((*sa) << 2) & 0x03c;
|
|
hsh ^= ((*sa) >> 4) & 0xf;
|
|
++sa;
|
|
hsh ^= ((*sa) << 4) & 0x30;
|
|
hsh ^= ((*sa) >> 2) & 0x3f;
|
|
|
|
hsh &= 63;
|
|
if (hsh > 31) {
|
|
match = eth->regs[RW_GA_HI] & (1 << (hsh - 32));
|
|
} else {
|
|
match = eth->regs[RW_GA_LO] & (1 << hsh);
|
|
}
|
|
D(printf("hsh=%x ga=%x.%x mtch=%d\n", hsh,
|
|
eth->regs[RW_GA_HI], eth->regs[RW_GA_LO], match));
|
|
return match;
|
|
}
|
|
|
|
static ssize_t eth_receive(NetClientState *nc, const uint8_t *buf, size_t size)
|
|
{
|
|
unsigned char sa_bcast[6] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
|
|
ETRAXFSEthState *eth = qemu_get_nic_opaque(nc);
|
|
int use_ma0 = eth->regs[RW_REC_CTRL] & 1;
|
|
int use_ma1 = eth->regs[RW_REC_CTRL] & 2;
|
|
int r_bcast = eth->regs[RW_REC_CTRL] & 8;
|
|
|
|
if (size < 12) {
|
|
return -1;
|
|
}
|
|
|
|
D(printf("%x.%x.%x.%x.%x.%x ma=%d %d bc=%d\n",
|
|
buf[0], buf[1], buf[2], buf[3], buf[4], buf[5],
|
|
use_ma0, use_ma1, r_bcast));
|
|
|
|
/* Does the frame get through the address filters? */
|
|
if ((!use_ma0 || memcmp(buf, eth->macaddr[0], 6))
|
|
&& (!use_ma1 || memcmp(buf, eth->macaddr[1], 6))
|
|
&& (!r_bcast || memcmp(buf, sa_bcast, 6))
|
|
&& !eth_match_groupaddr(eth, buf)) {
|
|
return size;
|
|
}
|
|
|
|
/* FIXME: Find another way to pass on the fake csum. */
|
|
etraxfs_dmac_input(eth->dma_in, (void *)buf, size + 4, 1);
|
|
|
|
return size;
|
|
}
|
|
|
|
static int eth_tx_push(void *opaque, unsigned char *buf, int len, bool eop)
|
|
{
|
|
ETRAXFSEthState *eth = opaque;
|
|
|
|
D(printf("%s buf=%p len=%d\n", __func__, buf, len));
|
|
qemu_send_packet(qemu_get_queue(eth->nic), buf, len);
|
|
return len;
|
|
}
|
|
|
|
static void eth_set_link(NetClientState *nc)
|
|
{
|
|
ETRAXFSEthState *eth = qemu_get_nic_opaque(nc);
|
|
D(printf("%s %d\n", __func__, nc->link_down));
|
|
eth->phy.link = !nc->link_down;
|
|
}
|
|
|
|
static const MemoryRegionOps eth_ops = {
|
|
.read = eth_read,
|
|
.write = eth_write,
|
|
.endianness = DEVICE_LITTLE_ENDIAN,
|
|
.valid = {
|
|
.min_access_size = 4,
|
|
.max_access_size = 4
|
|
}
|
|
};
|
|
|
|
static NetClientInfo net_etraxfs_info = {
|
|
.type = NET_CLIENT_DRIVER_NIC,
|
|
.size = sizeof(NICState),
|
|
.receive = eth_receive,
|
|
.link_status_changed = eth_set_link,
|
|
};
|
|
|
|
static void etraxfs_eth_reset(DeviceState *dev)
|
|
{
|
|
ETRAXFSEthState *s = ETRAX_FS_ETH(dev);
|
|
|
|
memset(s->regs, 0, sizeof(s->regs));
|
|
memset(s->macaddr, 0, sizeof(s->macaddr));
|
|
s->duplex_mismatch = 0;
|
|
|
|
s->mdio_bus.mdc = 0;
|
|
s->mdio_bus.mdio = 0;
|
|
s->mdio_bus.state = 0;
|
|
s->mdio_bus.drive = 0;
|
|
s->mdio_bus.cnt = 0;
|
|
s->mdio_bus.addr = 0;
|
|
s->mdio_bus.opc = 0;
|
|
s->mdio_bus.req = 0;
|
|
s->mdio_bus.data = 0;
|
|
|
|
tdk_reset(&s->phy);
|
|
}
|
|
|
|
static void etraxfs_eth_realize(DeviceState *dev, Error **errp)
|
|
{
|
|
SysBusDevice *sbd = SYS_BUS_DEVICE(dev);
|
|
ETRAXFSEthState *s = ETRAX_FS_ETH(dev);
|
|
|
|
if (!s->dma_out || !s->dma_in) {
|
|
error_setg(errp, "Unconnected ETRAX-FS Ethernet MAC");
|
|
return;
|
|
}
|
|
|
|
s->dma_out->client.push = eth_tx_push;
|
|
s->dma_out->client.opaque = s;
|
|
s->dma_in->client.opaque = s;
|
|
s->dma_in->client.pull = NULL;
|
|
|
|
memory_region_init_io(&s->mmio, OBJECT(dev), ð_ops, s,
|
|
"etraxfs-eth", 0x5c);
|
|
sysbus_init_mmio(sbd, &s->mmio);
|
|
|
|
qemu_macaddr_default_if_unset(&s->conf.macaddr);
|
|
s->nic = qemu_new_nic(&net_etraxfs_info, &s->conf,
|
|
object_get_typename(OBJECT(s)), dev->id, s);
|
|
qemu_format_nic_info_str(qemu_get_queue(s->nic), s->conf.macaddr.a);
|
|
|
|
s->phy.read = tdk_read;
|
|
s->phy.write = tdk_write;
|
|
mdio_attach(&s->mdio_bus, &s->phy, s->phyaddr);
|
|
}
|
|
|
|
static Property etraxfs_eth_properties[] = {
|
|
DEFINE_PROP_UINT32("phyaddr", ETRAXFSEthState, phyaddr, 1),
|
|
DEFINE_NIC_PROPERTIES(ETRAXFSEthState, conf),
|
|
DEFINE_PROP_END_OF_LIST(),
|
|
};
|
|
|
|
static void etraxfs_eth_class_init(ObjectClass *klass, void *data)
|
|
{
|
|
DeviceClass *dc = DEVICE_CLASS(klass);
|
|
|
|
dc->realize = etraxfs_eth_realize;
|
|
dc->reset = etraxfs_eth_reset;
|
|
device_class_set_props(dc, etraxfs_eth_properties);
|
|
/* Reason: dma_out, dma_in are not user settable */
|
|
dc->user_creatable = false;
|
|
}
|
|
|
|
|
|
/* Instantiate an ETRAXFS Ethernet MAC. */
|
|
DeviceState *
|
|
etraxfs_eth_init(NICInfo *nd, hwaddr base, int phyaddr,
|
|
struct etraxfs_dma_client *dma_out,
|
|
struct etraxfs_dma_client *dma_in)
|
|
{
|
|
DeviceState *dev;
|
|
qemu_check_nic_model(nd, "fseth");
|
|
|
|
dev = qdev_create(NULL, "etraxfs-eth");
|
|
qdev_set_nic_properties(dev, nd);
|
|
qdev_prop_set_uint32(dev, "phyaddr", phyaddr);
|
|
|
|
/*
|
|
* TODO: QOM design, define a QOM interface for "I am an etraxfs
|
|
* DMA client" (which replaces the current 'struct
|
|
* etraxfs_dma_client' ad-hoc interface), implement it on the
|
|
* ethernet device, and then have QOM link properties on the DMA
|
|
* controller device so that you can pass the interface
|
|
* implementations to it.
|
|
*/
|
|
ETRAX_FS_ETH(dev)->dma_out = dma_out;
|
|
ETRAX_FS_ETH(dev)->dma_in = dma_in;
|
|
qdev_init_nofail(dev);
|
|
sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base);
|
|
|
|
return dev;
|
|
}
|
|
|
|
static const TypeInfo etraxfs_eth_info = {
|
|
.name = TYPE_ETRAX_FS_ETH,
|
|
.parent = TYPE_SYS_BUS_DEVICE,
|
|
.instance_size = sizeof(ETRAXFSEthState),
|
|
.class_init = etraxfs_eth_class_init,
|
|
};
|
|
|
|
static void etraxfs_eth_register_types(void)
|
|
{
|
|
type_register_static(&etraxfs_eth_info);
|
|
}
|
|
|
|
type_init(etraxfs_eth_register_types)
|