p71924506
/
linux_old1
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
linux_old1/drivers/net/tulip/de2104x.c

2191 lines
53 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* de2104x.c: A Linux PCI Ethernet driver for Intel/Digital 21040/1 chips. */
/*
Copyright 2001,2003 Jeff Garzik <jgarzik@pobox.com>
Copyright 1994, 1995 Digital Equipment Corporation. [de4x5.c]
Written/copyright 1994-2001 by Donald Becker. [tulip.c]
This software may be used and distributed according to the terms of
the GNU General Public License (GPL), incorporated herein by reference.
Drivers based on or derived from this code fall under the GPL and must
retain the authorship, copyright and license notice. This file is not
a complete program and may only be used when the entire operating
system is licensed under the GPL.
See the file COPYING in this distribution for more information.
TODO, in rough priority order:
* Support forcing media type with a module parameter,
like dl2k.c/sundance.c
* Constants (module parms?) for Rx work limit
* Complete reset on PciErr
* Jumbo frames / dev->change_mtu
* Adjust Rx FIFO threshold and Max Rx DMA burst on Rx FIFO error
* Adjust Tx FIFO threshold and Max Tx DMA burst on Tx FIFO error
* Implement Tx software interrupt mitigation via
Tx descriptor bit
*/
#define DRV_NAME "de2104x"
#define DRV_VERSION "0.7"
#define DRV_RELDATE "Mar 17, 2004"
#include <linux/config.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/delay.h>
#include <linux/ethtool.h>
#include <linux/compiler.h>
#include <linux/rtnetlink.h>
#include <linux/crc32.h>
#include <asm/io.h>
#include <asm/irq.h>
#include <asm/uaccess.h>
#include <asm/unaligned.h>
/* These identify the driver base version and may not be removed. */
static char version[] =
KERN_INFO DRV_NAME " PCI Ethernet driver v" DRV_VERSION " (" DRV_RELDATE ")\n";
MODULE_AUTHOR("Jeff Garzik <jgarzik@pobox.com>");
MODULE_DESCRIPTION("Intel/Digital 21040/1 series PCI Ethernet driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_VERSION);
static int debug = -1;
module_param (debug, int, 0);
MODULE_PARM_DESC (debug, "de2104x bitmapped message enable number");
/* Set the copy breakpoint for the copy-only-tiny-buffer Rx structure. */
#if defined(__alpha__) || defined(__arm__) || defined(__hppa__) \
|| defined(__sparc_) || defined(__ia64__) \
|| defined(__sh__) || defined(__mips__)
static int rx_copybreak = 1518;
#else
static int rx_copybreak = 100;
#endif
module_param (rx_copybreak, int, 0);
MODULE_PARM_DESC (rx_copybreak, "de2104x Breakpoint at which Rx packets are copied");
#define PFX DRV_NAME ": "
#define DE_DEF_MSG_ENABLE (NETIF_MSG_DRV | \
NETIF_MSG_PROBE | \
NETIF_MSG_LINK | \
NETIF_MSG_IFDOWN | \
NETIF_MSG_IFUP | \
NETIF_MSG_RX_ERR | \
NETIF_MSG_TX_ERR)
#define DE_RX_RING_SIZE 64
#define DE_TX_RING_SIZE 64
#define DE_RING_BYTES \
((sizeof(struct de_desc) * DE_RX_RING_SIZE) + \
(sizeof(struct de_desc) * DE_TX_RING_SIZE))
#define NEXT_TX(N) (((N) + 1) & (DE_TX_RING_SIZE - 1))
#define NEXT_RX(N) (((N) + 1) & (DE_RX_RING_SIZE - 1))
#define TX_BUFFS_AVAIL(CP) \
(((CP)->tx_tail <= (CP)->tx_head) ? \
(CP)->tx_tail + (DE_TX_RING_SIZE - 1) - (CP)->tx_head : \
(CP)->tx_tail - (CP)->tx_head - 1)
#define PKT_BUF_SZ 1536 /* Size of each temporary Rx buffer.*/
#define RX_OFFSET 2
#define DE_SETUP_SKB ((struct sk_buff *) 1)
#define DE_DUMMY_SKB ((struct sk_buff *) 2)
#define DE_SETUP_FRAME_WORDS 96
#define DE_EEPROM_WORDS 256
#define DE_EEPROM_SIZE (DE_EEPROM_WORDS * sizeof(u16))
#define DE_MAX_MEDIA 5
#define DE_MEDIA_TP_AUTO 0
#define DE_MEDIA_BNC 1
#define DE_MEDIA_AUI 2
#define DE_MEDIA_TP 3
#define DE_MEDIA_TP_FD 4
#define DE_MEDIA_INVALID DE_MAX_MEDIA
#define DE_MEDIA_FIRST 0
#define DE_MEDIA_LAST (DE_MAX_MEDIA - 1)
#define DE_AUI_BNC (SUPPORTED_AUI | SUPPORTED_BNC)
#define DE_TIMER_LINK (60 * HZ)
#define DE_TIMER_NO_LINK (5 * HZ)
#define DE_NUM_REGS 16
#define DE_REGS_SIZE (DE_NUM_REGS * sizeof(u32))
#define DE_REGS_VER 1
/* Time in jiffies before concluding the transmitter is hung. */
#define TX_TIMEOUT (6*HZ)
#define DE_UNALIGNED_16(a) (u16)(get_unaligned((u16 *)(a)))
/* This is a mysterious value that can be written to CSR11 in the 21040 (only)
to support a pre-NWay full-duplex signaling mechanism using short frames.
No one knows what it should be, but if left at its default value some
10base2(!) packets trigger a full-duplex-request interrupt. */
#define FULL_DUPLEX_MAGIC 0x6969
enum {
/* NIC registers */
BusMode = 0x00,
TxPoll = 0x08,
RxPoll = 0x10,
RxRingAddr = 0x18,
TxRingAddr = 0x20,
MacStatus = 0x28,
MacMode = 0x30,
IntrMask = 0x38,
RxMissed = 0x40,
ROMCmd = 0x48,
CSR11 = 0x58,
SIAStatus = 0x60,
CSR13 = 0x68,
CSR14 = 0x70,
CSR15 = 0x78,
PCIPM = 0x40,
/* BusMode bits */
CmdReset = (1 << 0),
CacheAlign16 = 0x00008000,
BurstLen4 = 0x00000400,
/* Rx/TxPoll bits */
NormalTxPoll = (1 << 0),
NormalRxPoll = (1 << 0),
/* Tx/Rx descriptor status bits */
DescOwn = (1 << 31),
RxError = (1 << 15),
RxErrLong = (1 << 7),
RxErrCRC = (1 << 1),
RxErrFIFO = (1 << 0),
RxErrRunt = (1 << 11),
RxErrFrame = (1 << 14),
RingEnd = (1 << 25),
FirstFrag = (1 << 29),
LastFrag = (1 << 30),
TxError = (1 << 15),
TxFIFOUnder = (1 << 1),
TxLinkFail = (1 << 2) | (1 << 10) | (1 << 11),
TxMaxCol = (1 << 8),
TxOWC = (1 << 9),
TxJabber = (1 << 14),
SetupFrame = (1 << 27),
TxSwInt = (1 << 31),
/* MacStatus bits */
IntrOK = (1 << 16),
IntrErr = (1 << 15),
RxIntr = (1 << 6),
RxEmpty = (1 << 7),
TxIntr = (1 << 0),
TxEmpty = (1 << 2),
PciErr = (1 << 13),
TxState = (1 << 22) | (1 << 21) | (1 << 20),
RxState = (1 << 19) | (1 << 18) | (1 << 17),
LinkFail = (1 << 12),
LinkPass = (1 << 4),
RxStopped = (1 << 8),
TxStopped = (1 << 1),
/* MacMode bits */
TxEnable = (1 << 13),
RxEnable = (1 << 1),
RxTx = TxEnable | RxEnable,
FullDuplex = (1 << 9),
AcceptAllMulticast = (1 << 7),
AcceptAllPhys = (1 << 6),
BOCnt = (1 << 5),
MacModeClear = (1<<12) | (1<<11) | (1<<10) | (1<<8) | (1<<3) |
RxTx | BOCnt | AcceptAllPhys | AcceptAllMulticast,
/* ROMCmd bits */
EE_SHIFT_CLK = 0x02, /* EEPROM shift clock. */
EE_CS = 0x01, /* EEPROM chip select. */
EE_DATA_WRITE = 0x04, /* Data from the Tulip to EEPROM. */
EE_WRITE_0 = 0x01,
EE_WRITE_1 = 0x05,
EE_DATA_READ = 0x08, /* Data from the EEPROM chip. */
EE_ENB = (0x4800 | EE_CS),
/* The EEPROM commands include the alway-set leading bit. */
EE_READ_CMD = 6,
/* RxMissed bits */
RxMissedOver = (1 << 16),
RxMissedMask = 0xffff,
/* SROM-related bits */
SROMC0InfoLeaf = 27,
MediaBlockMask = 0x3f,
MediaCustomCSRs = (1 << 6),
/* PCIPM bits */
PM_Sleep = (1 << 31),
PM_Snooze = (1 << 30),
PM_Mask = PM_Sleep | PM_Snooze,
/* SIAStatus bits */
NWayState = (1 << 14) | (1 << 13) | (1 << 12),
NWayRestart = (1 << 12),
NonselPortActive = (1 << 9),
LinkFailStatus = (1 << 2),
NetCxnErr = (1 << 1),
};
static const u32 de_intr_mask =
IntrOK | IntrErr | RxIntr | RxEmpty | TxIntr | TxEmpty |
LinkPass | LinkFail | PciErr;
/*
* Set the programmable burst length to 4 longwords for all:
* DMA errors result without these values. Cache align 16 long.
*/
static const u32 de_bus_mode = CacheAlign16 | BurstLen4;
struct de_srom_media_block {
u8 opts;
u16 csr13;
u16 csr14;
u16 csr15;
} __attribute__((packed));
struct de_srom_info_leaf {
u16 default_media;
u8 n_blocks;
u8 unused;
} __attribute__((packed));
struct de_desc {
u32 opts1;
u32 opts2;
u32 addr1;
u32 addr2;
};
struct media_info {
u16 type; /* DE_MEDIA_xxx */
u16 csr13;
u16 csr14;
u16 csr15;
};
struct ring_info {
struct sk_buff *skb;
dma_addr_t mapping;
};
struct de_private {
unsigned tx_head;
unsigned tx_tail;
unsigned rx_tail;
void __iomem *regs;
struct net_device *dev;
spinlock_t lock;
struct de_desc *rx_ring;
struct de_desc *tx_ring;
struct ring_info tx_skb[DE_TX_RING_SIZE];
struct ring_info rx_skb[DE_RX_RING_SIZE];
unsigned rx_buf_sz;
dma_addr_t ring_dma;
u32 msg_enable;
struct net_device_stats net_stats;
struct pci_dev *pdev;
u16 setup_frame[DE_SETUP_FRAME_WORDS];
u32 media_type;
u32 media_supported;
u32 media_advertise;
struct media_info media[DE_MAX_MEDIA];
struct timer_list media_timer;
u8 *ee_data;
unsigned board_idx;
unsigned de21040 : 1;
unsigned media_lock : 1;
};
static void de_set_rx_mode (struct net_device *dev);
static void de_tx (struct de_private *de);
static void de_clean_rings (struct de_private *de);
static void de_media_interrupt (struct de_private *de, u32 status);
static void de21040_media_timer (unsigned long data);
static void de21041_media_timer (unsigned long data);
static unsigned int de_ok_to_advertise (struct de_private *de, u32 new_media);
static struct pci_device_id de_pci_tbl[] = {
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_PLUS,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },
{ },
};
MODULE_DEVICE_TABLE(pci, de_pci_tbl);
static const char * const media_name[DE_MAX_MEDIA] = {
"10baseT auto",
"BNC",
"AUI",
"10baseT-HD",
"10baseT-FD"
};
/* 21040 transceiver register settings:
* TP AUTO(unused), BNC(unused), AUI, TP, TP FD*/
static u16 t21040_csr13[] = { 0, 0, 0x8F09, 0x8F01, 0x8F01, };
static u16 t21040_csr14[] = { 0, 0, 0x0705, 0xFFFF, 0xFFFD, };
static u16 t21040_csr15[] = { 0, 0, 0x0006, 0x0000, 0x0000, };
/* 21041 transceiver register settings: TP AUTO, BNC, AUI, TP, TP FD*/
static u16 t21041_csr13[] = { 0xEF01, 0xEF09, 0xEF09, 0xEF01, 0xEF09, };
static u16 t21041_csr14[] = { 0xFFFF, 0xF7FD, 0xF7FD, 0x6F3F, 0x6F3D, };
static u16 t21041_csr15[] = { 0x0008, 0x0006, 0x000E, 0x0008, 0x0008, };
#define dr32(reg) readl(de->regs + (reg))
#define dw32(reg,val) writel((val), de->regs + (reg))
static void de_rx_err_acct (struct de_private *de, unsigned rx_tail,
u32 status, u32 len)
{
if (netif_msg_rx_err (de))
printk (KERN_DEBUG
"%s: rx err, slot %d status 0x%x len %d\n",
de->dev->name, rx_tail, status, len);
if ((status & 0x38000300) != 0x0300) {
/* Ingore earlier buffers. */
if ((status & 0xffff) != 0x7fff) {
if (netif_msg_rx_err(de))
printk(KERN_WARNING "%s: Oversized Ethernet frame "
"spanned multiple buffers, status %8.8x!\n",
de->dev->name, status);
de->net_stats.rx_length_errors++;
}
} else if (status & RxError) {
/* There was a fatal error. */
de->net_stats.rx_errors++; /* end of a packet.*/
if (status & 0x0890) de->net_stats.rx_length_errors++;
if (status & RxErrCRC) de->net_stats.rx_crc_errors++;
if (status & RxErrFIFO) de->net_stats.rx_fifo_errors++;
}
}
static void de_rx (struct de_private *de)
{
unsigned rx_tail = de->rx_tail;
unsigned rx_work = DE_RX_RING_SIZE;
unsigned drop = 0;
int rc;
while (rx_work--) {
u32 status, len;
dma_addr_t mapping;
struct sk_buff *skb, *copy_skb;
unsigned copying_skb, buflen;
skb = de->rx_skb[rx_tail].skb;
if (!skb)
BUG();
rmb();
status = le32_to_cpu(de->rx_ring[rx_tail].opts1);
if (status & DescOwn)
break;
len = ((status >> 16) & 0x7ff) - 4;
mapping = de->rx_skb[rx_tail].mapping;
if (unlikely(drop)) {
de->net_stats.rx_dropped++;
goto rx_next;
}
if (unlikely((status & 0x38008300) != 0x0300)) {
de_rx_err_acct(de, rx_tail, status, len);
goto rx_next;
}
copying_skb = (len <= rx_copybreak);
if (unlikely(netif_msg_rx_status(de)))
printk(KERN_DEBUG "%s: rx slot %d status 0x%x len %d copying? %d\n",
de->dev->name, rx_tail, status, len,
copying_skb);
buflen = copying_skb ? (len + RX_OFFSET) : de->rx_buf_sz;
copy_skb = dev_alloc_skb (buflen);
if (unlikely(!copy_skb)) {
de->net_stats.rx_dropped++;
drop = 1;
rx_work = 100;
goto rx_next;
}
copy_skb->dev = de->dev;
if (!copying_skb) {
pci_unmap_single(de->pdev, mapping,
buflen, PCI_DMA_FROMDEVICE);
skb_put(skb, len);
mapping =
de->rx_skb[rx_tail].mapping =
[NET]: Remove gratuitous use of skb->tail in network drivers. Many drivers use skb->tail unnecessarily. In these situations, the code roughly looks like: dev = dev_alloc_skb(...); [optional] skb_reserve(skb, ...); ... skb->tail ... But even if the skb_reserve() happens, skb->data equals skb->tail. So it doesn't make any sense to use anything other than skb->data in these cases. Another case was the s2io.c driver directly mucking with the skb->data and skb->tail pointers. It really just wanted to do an skb_reserve(), so that's what the code was changed to do instead. Another reason I'm making this change as it allows some SKB cleanups I have planned simpler to merge. In those cleanups, skb->head, skb->tail, and skb->end pointers are removed, and replaced with skb->head_room and skb->tail_room integers. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Jeff Garzik <jgarzik@pobox.com>
2005-06-29 06:25:31 +08:00
pci_map_single(de->pdev, copy_skb->data,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
buflen, PCI_DMA_FROMDEVICE);
de->rx_skb[rx_tail].skb = copy_skb;
} else {
pci_dma_sync_single_for_cpu(de->pdev, mapping, len, PCI_DMA_FROMDEVICE);
skb_reserve(copy_skb, RX_OFFSET);
[NET]: Remove gratuitous use of skb->tail in network drivers. Many drivers use skb->tail unnecessarily. In these situations, the code roughly looks like: dev = dev_alloc_skb(...); [optional] skb_reserve(skb, ...); ... skb->tail ... But even if the skb_reserve() happens, skb->data equals skb->tail. So it doesn't make any sense to use anything other than skb->data in these cases. Another case was the s2io.c driver directly mucking with the skb->data and skb->tail pointers. It really just wanted to do an skb_reserve(), so that's what the code was changed to do instead. Another reason I'm making this change as it allows some SKB cleanups I have planned simpler to merge. In those cleanups, skb->head, skb->tail, and skb->end pointers are removed, and replaced with skb->head_room and skb->tail_room integers. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Jeff Garzik <jgarzik@pobox.com>
2005-06-29 06:25:31 +08:00
memcpy(skb_put(copy_skb, len), skb->data, len);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pci_dma_sync_single_for_device(de->pdev, mapping, len, PCI_DMA_FROMDEVICE);
/* We'll reuse the original ring buffer. */
skb = copy_skb;
}
skb->protocol = eth_type_trans (skb, de->dev);
de->net_stats.rx_packets++;
de->net_stats.rx_bytes += skb->len;
de->dev->last_rx = jiffies;
rc = netif_rx (skb);
if (rc == NET_RX_DROP)
drop = 1;
rx_next:
de->rx_ring[rx_tail].opts1 = cpu_to_le32(DescOwn);
if (rx_tail == (DE_RX_RING_SIZE - 1))
de->rx_ring[rx_tail].opts2 =
cpu_to_le32(RingEnd | de->rx_buf_sz);
else
de->rx_ring[rx_tail].opts2 = cpu_to_le32(de->rx_buf_sz);
de->rx_ring[rx_tail].addr1 = cpu_to_le32(mapping);
rx_tail = NEXT_RX(rx_tail);
}
if (!rx_work)
printk(KERN_WARNING "%s: rx work limit reached\n", de->dev->name);
de->rx_tail = rx_tail;
}
static irqreturn_t de_interrupt (int irq, void *dev_instance, struct pt_regs *regs)
{
struct net_device *dev = dev_instance;
struct de_private *de = dev->priv;
u32 status;
status = dr32(MacStatus);
if ((!(status & (IntrOK|IntrErr))) || (status == 0xFFFF))
return IRQ_NONE;
if (netif_msg_intr(de))
printk(KERN_DEBUG "%s: intr, status %08x mode %08x desc %u/%u/%u\n",
dev->name, status, dr32(MacMode), de->rx_tail, de->tx_head, de->tx_tail);
dw32(MacStatus, status);
if (status & (RxIntr | RxEmpty)) {
de_rx(de);
if (status & RxEmpty)
dw32(RxPoll, NormalRxPoll);
}
spin_lock(&de->lock);
if (status & (TxIntr | TxEmpty))
de_tx(de);
if (status & (LinkPass | LinkFail))
de_media_interrupt(de, status);
spin_unlock(&de->lock);
if (status & PciErr) {
u16 pci_status;
pci_read_config_word(de->pdev, PCI_STATUS, &pci_status);
pci_write_config_word(de->pdev, PCI_STATUS, pci_status);
printk(KERN_ERR "%s: PCI bus error, status=%08x, PCI status=%04x\n",
dev->name, status, pci_status);
}
return IRQ_HANDLED;
}
static void de_tx (struct de_private *de)
{
unsigned tx_head = de->tx_head;
unsigned tx_tail = de->tx_tail;
while (tx_tail != tx_head) {
struct sk_buff *skb;
u32 status;
rmb();
status = le32_to_cpu(de->tx_ring[tx_tail].opts1);
if (status & DescOwn)
break;
skb = de->tx_skb[tx_tail].skb;
if (!skb)
BUG();
if (unlikely(skb == DE_DUMMY_SKB))
goto next;
if (unlikely(skb == DE_SETUP_SKB)) {
pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping,
sizeof(de->setup_frame), PCI_DMA_TODEVICE);
goto next;
}
pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping,
skb->len, PCI_DMA_TODEVICE);
if (status & LastFrag) {
if (status & TxError) {
if (netif_msg_tx_err(de))
printk(KERN_DEBUG "%s: tx err, status 0x%x\n",
de->dev->name, status);
de->net_stats.tx_errors++;
if (status & TxOWC)
de->net_stats.tx_window_errors++;
if (status & TxMaxCol)
de->net_stats.tx_aborted_errors++;
if (status & TxLinkFail)
de->net_stats.tx_carrier_errors++;
if (status & TxFIFOUnder)
de->net_stats.tx_fifo_errors++;
} else {
de->net_stats.tx_packets++;
de->net_stats.tx_bytes += skb->len;
if (netif_msg_tx_done(de))
printk(KERN_DEBUG "%s: tx done, slot %d\n", de->dev->name, tx_tail);
}
dev_kfree_skb_irq(skb);
}
next:
de->tx_skb[tx_tail].skb = NULL;
tx_tail = NEXT_TX(tx_tail);
}
de->tx_tail = tx_tail;
if (netif_queue_stopped(de->dev) && (TX_BUFFS_AVAIL(de) > (DE_TX_RING_SIZE / 4)))
netif_wake_queue(de->dev);
}
static int de_start_xmit (struct sk_buff *skb, struct net_device *dev)
{
struct de_private *de = dev->priv;
unsigned int entry, tx_free;
u32 mapping, len, flags = FirstFrag | LastFrag;
struct de_desc *txd;
spin_lock_irq(&de->lock);
tx_free = TX_BUFFS_AVAIL(de);
if (tx_free == 0) {
netif_stop_queue(dev);
spin_unlock_irq(&de->lock);
return 1;
}
tx_free--;
entry = de->tx_head;
txd = &de->tx_ring[entry];
len = skb->len;
mapping = pci_map_single(de->pdev, skb->data, len, PCI_DMA_TODEVICE);
if (entry == (DE_TX_RING_SIZE - 1))
flags |= RingEnd;
if (!tx_free || (tx_free == (DE_TX_RING_SIZE / 2)))
flags |= TxSwInt;
flags |= len;
txd->opts2 = cpu_to_le32(flags);
txd->addr1 = cpu_to_le32(mapping);
de->tx_skb[entry].skb = skb;
de->tx_skb[entry].mapping = mapping;
wmb();
txd->opts1 = cpu_to_le32(DescOwn);
wmb();
de->tx_head = NEXT_TX(entry);
if (netif_msg_tx_queued(de))
printk(KERN_DEBUG "%s: tx queued, slot %d, skblen %d\n",
dev->name, entry, skb->len);
if (tx_free == 0)
netif_stop_queue(dev);
spin_unlock_irq(&de->lock);
/* Trigger an immediate transmit demand. */
dw32(TxPoll, NormalTxPoll);
dev->trans_start = jiffies;
return 0;
}
/* Set or clear the multicast filter for this adaptor.
Note that we only use exclusion around actually queueing the
new frame, not around filling de->setup_frame. This is non-deterministic
when re-entered but still correct. */
#undef set_bit_le
#define set_bit_le(i,p) do { ((char *)(p))[(i)/8] |= (1<<((i)%8)); } while(0)
static void build_setup_frame_hash(u16 *setup_frm, struct net_device *dev)
{
struct de_private *de = dev->priv;
u16 hash_table[32];
struct dev_mc_list *mclist;
int i;
u16 *eaddrs;
memset(hash_table, 0, sizeof(hash_table));
set_bit_le(255, hash_table); /* Broadcast entry */
/* This should work on big-endian machines as well. */
for (i = 0, mclist = dev->mc_list; mclist && i < dev->mc_count;
i++, mclist = mclist->next) {
int index = ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x1ff;
set_bit_le(index, hash_table);
for (i = 0; i < 32; i++) {
*setup_frm++ = hash_table[i];
*setup_frm++ = hash_table[i];
}
setup_frm = &de->setup_frame[13*6];
}
/* Fill the final entry with our physical address. */
eaddrs = (u16 *)dev->dev_addr;
*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
}
static void build_setup_frame_perfect(u16 *setup_frm, struct net_device *dev)
{
struct de_private *de = dev->priv;
struct dev_mc_list *mclist;
int i;
u16 *eaddrs;
/* We have <= 14 addresses so we can use the wonderful
16 address perfect filtering of the Tulip. */
for (i = 0, mclist = dev->mc_list; i < dev->mc_count;
i++, mclist = mclist->next) {
eaddrs = (u16 *)mclist->dmi_addr;
*setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++;
*setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++;
*setup_frm++ = *eaddrs; *setup_frm++ = *eaddrs++;
}
/* Fill the unused entries with the broadcast address. */
memset(setup_frm, 0xff, (15-i)*12);
setup_frm = &de->setup_frame[15*6];
/* Fill the final entry with our physical address. */
eaddrs = (u16 *)dev->dev_addr;
*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
}
static void __de_set_rx_mode (struct net_device *dev)
{
struct de_private *de = dev->priv;
u32 macmode;
unsigned int entry;
u32 mapping;
struct de_desc *txd;
struct de_desc *dummy_txd = NULL;
macmode = dr32(MacMode) & ~(AcceptAllMulticast | AcceptAllPhys);
if (dev->flags & IFF_PROMISC) { /* Set promiscuous. */
macmode |= AcceptAllMulticast | AcceptAllPhys;
goto out;
}
if ((dev->mc_count > 1000) || (dev->flags & IFF_ALLMULTI)) {
/* Too many to filter well -- accept all multicasts. */
macmode |= AcceptAllMulticast;
goto out;
}
/* Note that only the low-address shortword of setup_frame is valid!
The values are doubled for big-endian architectures. */
if (dev->mc_count > 14) /* Must use a multicast hash table. */
build_setup_frame_hash (de->setup_frame, dev);
else
build_setup_frame_perfect (de->setup_frame, dev);
/*
* Now add this frame to the Tx list.
*/
entry = de->tx_head;
/* Avoid a chip errata by prefixing a dummy entry. */
if (entry != 0) {
de->tx_skb[entry].skb = DE_DUMMY_SKB;
dummy_txd = &de->tx_ring[entry];
dummy_txd->opts2 = (entry == (DE_TX_RING_SIZE - 1)) ?
cpu_to_le32(RingEnd) : 0;
dummy_txd->addr1 = 0;
/* Must set DescOwned later to avoid race with chip */
entry = NEXT_TX(entry);
}
de->tx_skb[entry].skb = DE_SETUP_SKB;
de->tx_skb[entry].mapping = mapping =
pci_map_single (de->pdev, de->setup_frame,
sizeof (de->setup_frame), PCI_DMA_TODEVICE);
/* Put the setup frame on the Tx list. */
txd = &de->tx_ring[entry];
if (entry == (DE_TX_RING_SIZE - 1))
txd->opts2 = cpu_to_le32(SetupFrame | RingEnd | sizeof (de->setup_frame));
else
txd->opts2 = cpu_to_le32(SetupFrame | sizeof (de->setup_frame));
txd->addr1 = cpu_to_le32(mapping);
wmb();
txd->opts1 = cpu_to_le32(DescOwn);
wmb();
if (dummy_txd) {
dummy_txd->opts1 = cpu_to_le32(DescOwn);
wmb();
}
de->tx_head = NEXT_TX(entry);
if (TX_BUFFS_AVAIL(de) < 0)
BUG();
if (TX_BUFFS_AVAIL(de) == 0)
netif_stop_queue(dev);
/* Trigger an immediate transmit demand. */
dw32(TxPoll, NormalTxPoll);
out:
if (macmode != dr32(MacMode))
dw32(MacMode, macmode);
}
static void de_set_rx_mode (struct net_device *dev)
{
unsigned long flags;
struct de_private *de = dev->priv;
spin_lock_irqsave (&de->lock, flags);
__de_set_rx_mode(dev);
spin_unlock_irqrestore (&de->lock, flags);
}
static inline void de_rx_missed(struct de_private *de, u32 rx_missed)
{
if (unlikely(rx_missed & RxMissedOver))
de->net_stats.rx_missed_errors += RxMissedMask;
else
de->net_stats.rx_missed_errors += (rx_missed & RxMissedMask);
}
static void __de_get_stats(struct de_private *de)
{
u32 tmp = dr32(RxMissed); /* self-clearing */
de_rx_missed(de, tmp);
}
static struct net_device_stats *de_get_stats(struct net_device *dev)
{
struct de_private *de = dev->priv;
/* The chip only need report frame silently dropped. */
spin_lock_irq(&de->lock);
if (netif_running(dev) && netif_device_present(dev))
__de_get_stats(de);
spin_unlock_irq(&de->lock);
return &de->net_stats;
}
static inline int de_is_running (struct de_private *de)
{
return (dr32(MacStatus) & (RxState | TxState)) ? 1 : 0;
}
static void de_stop_rxtx (struct de_private *de)
{
u32 macmode;
unsigned int work = 1000;
macmode = dr32(MacMode);
if (macmode & RxTx) {
dw32(MacMode, macmode & ~RxTx);
dr32(MacMode);
}
while (--work > 0) {
if (!de_is_running(de))
return;
cpu_relax();
}
printk(KERN_WARNING "%s: timeout expired stopping DMA\n", de->dev->name);
}
static inline void de_start_rxtx (struct de_private *de)
{
u32 macmode;
macmode = dr32(MacMode);
if ((macmode & RxTx) != RxTx) {
dw32(MacMode, macmode | RxTx);
dr32(MacMode);
}
}
static void de_stop_hw (struct de_private *de)
{
udelay(5);
dw32(IntrMask, 0);
de_stop_rxtx(de);
dw32(MacStatus, dr32(MacStatus));
udelay(10);
de->rx_tail = 0;
de->tx_head = de->tx_tail = 0;
}
static void de_link_up(struct de_private *de)
{
if (!netif_carrier_ok(de->dev)) {
netif_carrier_on(de->dev);
if (netif_msg_link(de))
printk(KERN_INFO "%s: link up, media %s\n",
de->dev->name, media_name[de->media_type]);
}
}
static void de_link_down(struct de_private *de)
{
if (netif_carrier_ok(de->dev)) {
netif_carrier_off(de->dev);
if (netif_msg_link(de))
printk(KERN_INFO "%s: link down\n", de->dev->name);
}
}
static void de_set_media (struct de_private *de)
{
unsigned media = de->media_type;
u32 macmode = dr32(MacMode);
if (de_is_running(de))
BUG();
if (de->de21040)
dw32(CSR11, FULL_DUPLEX_MAGIC);
dw32(CSR13, 0); /* Reset phy */
dw32(CSR14, de->media[media].csr14);
dw32(CSR15, de->media[media].csr15);
dw32(CSR13, de->media[media].csr13);
/* must delay 10ms before writing to other registers,
* especially CSR6
*/
mdelay(10);
if (media == DE_MEDIA_TP_FD)
macmode |= FullDuplex;
else
macmode &= ~FullDuplex;
if (netif_msg_link(de)) {
printk(KERN_INFO "%s: set link %s\n"
KERN_INFO "%s: mode 0x%x, sia 0x%x,0x%x,0x%x,0x%x\n"
KERN_INFO "%s: set mode 0x%x, set sia 0x%x,0x%x,0x%x\n",
de->dev->name, media_name[media],
de->dev->name, dr32(MacMode), dr32(SIAStatus),
dr32(CSR13), dr32(CSR14), dr32(CSR15),
de->dev->name, macmode, de->media[media].csr13,
de->media[media].csr14, de->media[media].csr15);
}
if (macmode != dr32(MacMode))
dw32(MacMode, macmode);
}
static void de_next_media (struct de_private *de, u32 *media,
unsigned int n_media)
{
unsigned int i;
for (i = 0; i < n_media; i++) {
if (de_ok_to_advertise(de, media[i])) {
de->media_type = media[i];
return;
}
}
}
static void de21040_media_timer (unsigned long data)
{
struct de_private *de = (struct de_private *) data;
struct net_device *dev = de->dev;
u32 status = dr32(SIAStatus);
unsigned int carrier;
unsigned long flags;
carrier = (status & NetCxnErr) ? 0 : 1;
if (carrier) {
if (de->media_type != DE_MEDIA_AUI && (status & LinkFailStatus))
goto no_link_yet;
de->media_timer.expires = jiffies + DE_TIMER_LINK;
add_timer(&de->media_timer);
if (!netif_carrier_ok(dev))
de_link_up(de);
else
if (netif_msg_timer(de))
printk(KERN_INFO "%s: %s link ok, status %x\n",
dev->name, media_name[de->media_type],
status);
return;
}
de_link_down(de);
if (de->media_lock)
return;
if (de->media_type == DE_MEDIA_AUI) {
u32 next_state = DE_MEDIA_TP;
de_next_media(de, &next_state, 1);
} else {
u32 next_state = DE_MEDIA_AUI;
de_next_media(de, &next_state, 1);
}
spin_lock_irqsave(&de->lock, flags);
de_stop_rxtx(de);
spin_unlock_irqrestore(&de->lock, flags);
de_set_media(de);
de_start_rxtx(de);
no_link_yet:
de->media_timer.expires = jiffies + DE_TIMER_NO_LINK;
add_timer(&de->media_timer);
if (netif_msg_timer(de))
printk(KERN_INFO "%s: no link, trying media %s, status %x\n",
dev->name, media_name[de->media_type], status);
}
static unsigned int de_ok_to_advertise (struct de_private *de, u32 new_media)
{
switch (new_media) {
case DE_MEDIA_TP_AUTO:
if (!(de->media_advertise & ADVERTISED_Autoneg))
return 0;
if (!(de->media_advertise & (ADVERTISED_10baseT_Half | ADVERTISED_10baseT_Full)))
return 0;
break;
case DE_MEDIA_BNC:
if (!(de->media_advertise & ADVERTISED_BNC))
return 0;
break;
case DE_MEDIA_AUI:
if (!(de->media_advertise & ADVERTISED_AUI))
return 0;
break;
case DE_MEDIA_TP:
if (!(de->media_advertise & ADVERTISED_10baseT_Half))
return 0;
break;
case DE_MEDIA_TP_FD:
if (!(de->media_advertise & ADVERTISED_10baseT_Full))
return 0;
break;
}
return 1;
}
static void de21041_media_timer (unsigned long data)
{
struct de_private *de = (struct de_private *) data;
struct net_device *dev = de->dev;
u32 status = dr32(SIAStatus);
unsigned int carrier;
unsigned long flags;
carrier = (status & NetCxnErr) ? 0 : 1;
if (carrier) {
if ((de->media_type == DE_MEDIA_TP_AUTO ||
de->media_type == DE_MEDIA_TP ||
de->media_type == DE_MEDIA_TP_FD) &&
(status & LinkFailStatus))
goto no_link_yet;
de->media_timer.expires = jiffies + DE_TIMER_LINK;
add_timer(&de->media_timer);
if (!netif_carrier_ok(dev))
de_link_up(de);
else
if (netif_msg_timer(de))
printk(KERN_INFO "%s: %s link ok, mode %x status %x\n",
dev->name, media_name[de->media_type],
dr32(MacMode), status);
return;
}
de_link_down(de);
/* if media type locked, don't switch media */
if (de->media_lock)
goto set_media;
/* if activity detected, use that as hint for new media type */
if (status & NonselPortActive) {
unsigned int have_media = 1;
/* if AUI/BNC selected, then activity is on TP port */
if (de->media_type == DE_MEDIA_AUI ||
de->media_type == DE_MEDIA_BNC) {
if (de_ok_to_advertise(de, DE_MEDIA_TP_AUTO))
de->media_type = DE_MEDIA_TP_AUTO;
else
have_media = 0;
}
/* TP selected. If there is only TP and BNC, then it's BNC */
else if (((de->media_supported & DE_AUI_BNC) == SUPPORTED_BNC) &&
de_ok_to_advertise(de, DE_MEDIA_BNC))
de->media_type = DE_MEDIA_BNC;
/* TP selected. If there is only TP and AUI, then it's AUI */
else if (((de->media_supported & DE_AUI_BNC) == SUPPORTED_AUI) &&
de_ok_to_advertise(de, DE_MEDIA_AUI))
de->media_type = DE_MEDIA_AUI;
/* otherwise, ignore the hint */
else
have_media = 0;
if (have_media)
goto set_media;
}
/*
* Absent or ambiguous activity hint, move to next advertised
* media state. If de->media_type is left unchanged, this
* simply resets the PHY and reloads the current media settings.
*/
if (de->media_type == DE_MEDIA_AUI) {
u32 next_states[] = { DE_MEDIA_BNC, DE_MEDIA_TP_AUTO };
de_next_media(de, next_states, ARRAY_SIZE(next_states));
} else if (de->media_type == DE_MEDIA_BNC) {
u32 next_states[] = { DE_MEDIA_TP_AUTO, DE_MEDIA_AUI };
de_next_media(de, next_states, ARRAY_SIZE(next_states));
} else {
u32 next_states[] = { DE_MEDIA_AUI, DE_MEDIA_BNC, DE_MEDIA_TP_AUTO };
de_next_media(de, next_states, ARRAY_SIZE(next_states));
}
set_media:
spin_lock_irqsave(&de->lock, flags);
de_stop_rxtx(de);
spin_unlock_irqrestore(&de->lock, flags);
de_set_media(de);
de_start_rxtx(de);
no_link_yet:
de->media_timer.expires = jiffies + DE_TIMER_NO_LINK;
add_timer(&de->media_timer);
if (netif_msg_timer(de))
printk(KERN_INFO "%s: no link, trying media %s, status %x\n",
dev->name, media_name[de->media_type], status);
}
static void de_media_interrupt (struct de_private *de, u32 status)
{
if (status & LinkPass) {
de_link_up(de);
mod_timer(&de->media_timer, jiffies + DE_TIMER_LINK);
return;
}
if (!(status & LinkFail))
BUG();
if (netif_carrier_ok(de->dev)) {
de_link_down(de);
mod_timer(&de->media_timer, jiffies + DE_TIMER_NO_LINK);
}
}
static int de_reset_mac (struct de_private *de)
{
u32 status, tmp;
/*
* Reset MAC. de4x5.c and tulip.c examined for "advice"
* in this area.
*/
if (dr32(BusMode) == 0xffffffff)
return -EBUSY;
/* Reset the chip, holding bit 0 set at least 50 PCI cycles. */
dw32 (BusMode, CmdReset);
mdelay (1);
dw32 (BusMode, de_bus_mode);
mdelay (1);
for (tmp = 0; tmp < 5; tmp++) {
dr32 (BusMode);
mdelay (1);
}
mdelay (1);
status = dr32(MacStatus);
if (status & (RxState | TxState))
return -EBUSY;
if (status == 0xffffffff)
return -ENODEV;
return 0;
}
static void de_adapter_wake (struct de_private *de)
{
u32 pmctl;
if (de->de21040)
return;
pci_read_config_dword(de->pdev, PCIPM, &pmctl);
if (pmctl & PM_Mask) {
pmctl &= ~PM_Mask;
pci_write_config_dword(de->pdev, PCIPM, pmctl);
/* de4x5.c delays, so we do too */
msleep(10);
}
}
static void de_adapter_sleep (struct de_private *de)
{
u32 pmctl;
if (de->de21040)
return;
pci_read_config_dword(de->pdev, PCIPM, &pmctl);
pmctl |= PM_Sleep;
pci_write_config_dword(de->pdev, PCIPM, pmctl);
}
static int de_init_hw (struct de_private *de)
{
struct net_device *dev = de->dev;
u32 macmode;
int rc;
de_adapter_wake(de);
macmode = dr32(MacMode) & ~MacModeClear;
rc = de_reset_mac(de);
if (rc)
return rc;
de_set_media(de); /* reset phy */
dw32(RxRingAddr, de->ring_dma);
dw32(TxRingAddr, de->ring_dma + (sizeof(struct de_desc) * DE_RX_RING_SIZE));
dw32(MacMode, RxTx | macmode);
dr32(RxMissed); /* self-clearing */
dw32(IntrMask, de_intr_mask);
de_set_rx_mode(dev);
return 0;
}
static int de_refill_rx (struct de_private *de)
{
unsigned i;
for (i = 0; i < DE_RX_RING_SIZE; i++) {
struct sk_buff *skb;
skb = dev_alloc_skb(de->rx_buf_sz);
if (!skb)
goto err_out;
skb->dev = de->dev;
de->rx_skb[i].mapping = pci_map_single(de->pdev,
[NET]: Remove gratuitous use of skb->tail in network drivers. Many drivers use skb->tail unnecessarily. In these situations, the code roughly looks like: dev = dev_alloc_skb(...); [optional] skb_reserve(skb, ...); ... skb->tail ... But even if the skb_reserve() happens, skb->data equals skb->tail. So it doesn't make any sense to use anything other than skb->data in these cases. Another case was the s2io.c driver directly mucking with the skb->data and skb->tail pointers. It really just wanted to do an skb_reserve(), so that's what the code was changed to do instead. Another reason I'm making this change as it allows some SKB cleanups I have planned simpler to merge. In those cleanups, skb->head, skb->tail, and skb->end pointers are removed, and replaced with skb->head_room and skb->tail_room integers. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Jeff Garzik <jgarzik@pobox.com>
2005-06-29 06:25:31 +08:00
skb->data, de->rx_buf_sz, PCI_DMA_FROMDEVICE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
de->rx_skb[i].skb = skb;
de->rx_ring[i].opts1 = cpu_to_le32(DescOwn);
if (i == (DE_RX_RING_SIZE - 1))
de->rx_ring[i].opts2 =
cpu_to_le32(RingEnd | de->rx_buf_sz);
else
de->rx_ring[i].opts2 = cpu_to_le32(de->rx_buf_sz);
de->rx_ring[i].addr1 = cpu_to_le32(de->rx_skb[i].mapping);
de->rx_ring[i].addr2 = 0;
}
return 0;
err_out:
de_clean_rings(de);
return -ENOMEM;
}
static int de_init_rings (struct de_private *de)
{
memset(de->tx_ring, 0, sizeof(struct de_desc) * DE_TX_RING_SIZE);
de->tx_ring[DE_TX_RING_SIZE - 1].opts2 = cpu_to_le32(RingEnd);
de->rx_tail = 0;
de->tx_head = de->tx_tail = 0;
return de_refill_rx (de);
}
static int de_alloc_rings (struct de_private *de)
{
de->rx_ring = pci_alloc_consistent(de->pdev, DE_RING_BYTES, &de->ring_dma);
if (!de->rx_ring)
return -ENOMEM;
de->tx_ring = &de->rx_ring[DE_RX_RING_SIZE];
return de_init_rings(de);
}
static void de_clean_rings (struct de_private *de)
{
unsigned i;
memset(de->rx_ring, 0, sizeof(struct de_desc) * DE_RX_RING_SIZE);
de->rx_ring[DE_RX_RING_SIZE - 1].opts2 = cpu_to_le32(RingEnd);
wmb();
memset(de->tx_ring, 0, sizeof(struct de_desc) * DE_TX_RING_SIZE);
de->tx_ring[DE_TX_RING_SIZE - 1].opts2 = cpu_to_le32(RingEnd);
wmb();
for (i = 0; i < DE_RX_RING_SIZE; i++) {
if (de->rx_skb[i].skb) {
pci_unmap_single(de->pdev, de->rx_skb[i].mapping,
de->rx_buf_sz, PCI_DMA_FROMDEVICE);
dev_kfree_skb(de->rx_skb[i].skb);
}
}
for (i = 0; i < DE_TX_RING_SIZE; i++) {
struct sk_buff *skb = de->tx_skb[i].skb;
if ((skb) && (skb != DE_DUMMY_SKB)) {
if (skb != DE_SETUP_SKB) {
dev_kfree_skb(skb);
de->net_stats.tx_dropped++;
pci_unmap_single(de->pdev,
de->tx_skb[i].mapping,
skb->len, PCI_DMA_TODEVICE);
} else {
pci_unmap_single(de->pdev,
de->tx_skb[i].mapping,
sizeof(de->setup_frame),
PCI_DMA_TODEVICE);
}
}
}
memset(&de->rx_skb, 0, sizeof(struct ring_info) * DE_RX_RING_SIZE);
memset(&de->tx_skb, 0, sizeof(struct ring_info) * DE_TX_RING_SIZE);
}
static void de_free_rings (struct de_private *de)
{
de_clean_rings(de);
pci_free_consistent(de->pdev, DE_RING_BYTES, de->rx_ring, de->ring_dma);
de->rx_ring = NULL;
de->tx_ring = NULL;
}
static int de_open (struct net_device *dev)
{
struct de_private *de = dev->priv;
int rc;
unsigned long flags;
if (netif_msg_ifup(de))
printk(KERN_DEBUG "%s: enabling interface\n", dev->name);
de->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32);
rc = de_alloc_rings(de);
if (rc) {
printk(KERN_ERR "%s: ring allocation failure, err=%d\n",
dev->name, rc);
return rc;
}
rc = de_init_hw(de);
if (rc) {
printk(KERN_ERR "%s: h/w init failure, err=%d\n",
dev->name, rc);
goto err_out_free;
}
rc = request_irq(dev->irq, de_interrupt, SA_SHIRQ, dev->name, dev);
if (rc) {
printk(KERN_ERR "%s: IRQ %d request failure, err=%d\n",
dev->name, dev->irq, rc);
goto err_out_hw;
}
netif_start_queue(dev);
mod_timer(&de->media_timer, jiffies + DE_TIMER_NO_LINK);
return 0;
err_out_hw:
spin_lock_irqsave(&de->lock, flags);
de_stop_hw(de);
spin_unlock_irqrestore(&de->lock, flags);
err_out_free:
de_free_rings(de);
return rc;
}
static int de_close (struct net_device *dev)
{
struct de_private *de = dev->priv;
unsigned long flags;
if (netif_msg_ifdown(de))
printk(KERN_DEBUG "%s: disabling interface\n", dev->name);
del_timer_sync(&de->media_timer);
spin_lock_irqsave(&de->lock, flags);
de_stop_hw(de);
netif_stop_queue(dev);
netif_carrier_off(dev);
spin_unlock_irqrestore(&de->lock, flags);
free_irq(dev->irq, dev);
de_free_rings(de);
de_adapter_sleep(de);
pci_disable_device(de->pdev);
return 0;
}
static void de_tx_timeout (struct net_device *dev)
{
struct de_private *de = dev->priv;
printk(KERN_DEBUG "%s: NIC status %08x mode %08x sia %08x desc %u/%u/%u\n",
dev->name, dr32(MacStatus), dr32(MacMode), dr32(SIAStatus),
de->rx_tail, de->tx_head, de->tx_tail);
del_timer_sync(&de->media_timer);
disable_irq(dev->irq);
spin_lock_irq(&de->lock);
de_stop_hw(de);
netif_stop_queue(dev);
netif_carrier_off(dev);
spin_unlock_irq(&de->lock);
enable_irq(dev->irq);
/* Update the error counts. */
__de_get_stats(de);
synchronize_irq(dev->irq);
de_clean_rings(de);
de_init_hw(de);
netif_wake_queue(dev);
}
static void __de_get_regs(struct de_private *de, u8 *buf)
{
int i;
u32 *rbuf = (u32 *)buf;
/* read all CSRs */
for (i = 0; i < DE_NUM_REGS; i++)
rbuf[i] = dr32(i * 8);
/* handle self-clearing RxMissed counter, CSR8 */
de_rx_missed(de, rbuf[8]);
}
static int __de_get_settings(struct de_private *de, struct ethtool_cmd *ecmd)
{
ecmd->supported = de->media_supported;
ecmd->transceiver = XCVR_INTERNAL;
ecmd->phy_address = 0;
ecmd->advertising = de->media_advertise;
switch (de->media_type) {
case DE_MEDIA_AUI:
ecmd->port = PORT_AUI;
ecmd->speed = 5;
break;
case DE_MEDIA_BNC:
ecmd->port = PORT_BNC;
ecmd->speed = 2;
break;
default:
ecmd->port = PORT_TP;
ecmd->speed = SPEED_10;
break;
}
if (dr32(MacMode) & FullDuplex)
ecmd->duplex = DUPLEX_FULL;
else
ecmd->duplex = DUPLEX_HALF;
if (de->media_lock)
ecmd->autoneg = AUTONEG_DISABLE;
else
ecmd->autoneg = AUTONEG_ENABLE;
/* ignore maxtxpkt, maxrxpkt for now */
return 0;
}
static int __de_set_settings(struct de_private *de, struct ethtool_cmd *ecmd)
{
u32 new_media;
unsigned int media_lock;
if (ecmd->speed != SPEED_10 && ecmd->speed != 5 && ecmd->speed != 2)
return -EINVAL;
if (de->de21040 && ecmd->speed == 2)
return -EINVAL;
if (ecmd->duplex != DUPLEX_HALF && ecmd->duplex != DUPLEX_FULL)
return -EINVAL;
if (ecmd->port != PORT_TP && ecmd->port != PORT_AUI && ecmd->port != PORT_BNC)
return -EINVAL;
if (de->de21040 && ecmd->port == PORT_BNC)
return -EINVAL;
if (ecmd->transceiver != XCVR_INTERNAL)
return -EINVAL;
if (ecmd->autoneg != AUTONEG_DISABLE && ecmd->autoneg != AUTONEG_ENABLE)
return -EINVAL;
if (ecmd->advertising & ~de->media_supported)
return -EINVAL;
if (ecmd->autoneg == AUTONEG_ENABLE &&
(!(ecmd->advertising & ADVERTISED_Autoneg)))
return -EINVAL;
switch (ecmd->port) {
case PORT_AUI:
new_media = DE_MEDIA_AUI;
if (!(ecmd->advertising & ADVERTISED_AUI))
return -EINVAL;
break;
case PORT_BNC:
new_media = DE_MEDIA_BNC;
if (!(ecmd->advertising & ADVERTISED_BNC))
return -EINVAL;
break;
default:
if (ecmd->autoneg == AUTONEG_ENABLE)
new_media = DE_MEDIA_TP_AUTO;
else if (ecmd->duplex == DUPLEX_FULL)
new_media = DE_MEDIA_TP_FD;
else
new_media = DE_MEDIA_TP;
if (!(ecmd->advertising & ADVERTISED_TP))
return -EINVAL;
if (!(ecmd->advertising & (ADVERTISED_10baseT_Full | ADVERTISED_10baseT_Half)))
return -EINVAL;
break;
}
media_lock = (ecmd->autoneg == AUTONEG_ENABLE) ? 0 : 1;
if ((new_media == de->media_type) &&
(media_lock == de->media_lock) &&
(ecmd->advertising == de->media_advertise))
return 0; /* nothing to change */
de_link_down(de);
de_stop_rxtx(de);
de->media_type = new_media;
de->media_lock = media_lock;
de->media_advertise = ecmd->advertising;
de_set_media(de);
return 0;
}
static void de_get_drvinfo (struct net_device *dev,struct ethtool_drvinfo *info)
{
struct de_private *de = dev->priv;
strcpy (info->driver, DRV_NAME);
strcpy (info->version, DRV_VERSION);
strcpy (info->bus_info, pci_name(de->pdev));
info->eedump_len = DE_EEPROM_SIZE;
}
static int de_get_regs_len(struct net_device *dev)
{
return DE_REGS_SIZE;
}
static int de_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
{
struct de_private *de = dev->priv;
int rc;
spin_lock_irq(&de->lock);
rc = __de_get_settings(de, ecmd);
spin_unlock_irq(&de->lock);
return rc;
}
static int de_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
{
struct de_private *de = dev->priv;
int rc;
spin_lock_irq(&de->lock);
rc = __de_set_settings(de, ecmd);
spin_unlock_irq(&de->lock);
return rc;
}
static u32 de_get_msglevel(struct net_device *dev)
{
struct de_private *de = dev->priv;
return de->msg_enable;
}
static void de_set_msglevel(struct net_device *dev, u32 msglvl)
{
struct de_private *de = dev->priv;
de->msg_enable = msglvl;
}
static int de_get_eeprom(struct net_device *dev,
struct ethtool_eeprom *eeprom, u8 *data)
{
struct de_private *de = dev->priv;
if (!de->ee_data)
return -EOPNOTSUPP;
if ((eeprom->offset != 0) || (eeprom->magic != 0) ||
(eeprom->len != DE_EEPROM_SIZE))
return -EINVAL;
memcpy(data, de->ee_data, eeprom->len);
return 0;
}
static int de_nway_reset(struct net_device *dev)
{
struct de_private *de = dev->priv;
u32 status;
if (de->media_type != DE_MEDIA_TP_AUTO)
return -EINVAL;
if (netif_carrier_ok(de->dev))
de_link_down(de);
status = dr32(SIAStatus);
dw32(SIAStatus, (status & ~NWayState) | NWayRestart);
if (netif_msg_link(de))
printk(KERN_INFO "%s: link nway restart, status %x,%x\n",
de->dev->name, status, dr32(SIAStatus));
return 0;
}
static void de_get_regs(struct net_device *dev, struct ethtool_regs *regs,
void *data)
{
struct de_private *de = dev->priv;
regs->version = (DE_REGS_VER << 2) | de->de21040;
spin_lock_irq(&de->lock);
__de_get_regs(de, data);
spin_unlock_irq(&de->lock);
}
static struct ethtool_ops de_ethtool_ops = {
.get_link = ethtool_op_get_link,
.get_tx_csum = ethtool_op_get_tx_csum,
.get_sg = ethtool_op_get_sg,
.get_drvinfo = de_get_drvinfo,
.get_regs_len = de_get_regs_len,
.get_settings = de_get_settings,
.set_settings = de_set_settings,
.get_msglevel = de_get_msglevel,
.set_msglevel = de_set_msglevel,
.get_eeprom = de_get_eeprom,
.nway_reset = de_nway_reset,
.get_regs = de_get_regs,
};
static void __init de21040_get_mac_address (struct de_private *de)
{
unsigned i;
dw32 (ROMCmd, 0); /* Reset the pointer with a dummy write. */
for (i = 0; i < 6; i++) {
int value, boguscnt = 100000;
do
value = dr32(ROMCmd);
while (value < 0 && --boguscnt > 0);
de->dev->dev_addr[i] = value;
udelay(1);
if (boguscnt <= 0)
printk(KERN_WARNING PFX "timeout reading 21040 MAC address byte %u\n", i);
}
}
static void __init de21040_get_media_info(struct de_private *de)
{
unsigned int i;
de->media_type = DE_MEDIA_TP;
de->media_supported |= SUPPORTED_TP | SUPPORTED_10baseT_Full |
SUPPORTED_10baseT_Half | SUPPORTED_AUI;
de->media_advertise = de->media_supported;
for (i = 0; i < DE_MAX_MEDIA; i++) {
switch (i) {
case DE_MEDIA_AUI:
case DE_MEDIA_TP:
case DE_MEDIA_TP_FD:
de->media[i].type = i;
de->media[i].csr13 = t21040_csr13[i];
de->media[i].csr14 = t21040_csr14[i];
de->media[i].csr15 = t21040_csr15[i];
break;
default:
de->media[i].type = DE_MEDIA_INVALID;
break;
}
}
}
/* Note: this routine returns extra data bits for size detection. */
static unsigned __init tulip_read_eeprom(void __iomem *regs, int location, int addr_len)
{
int i;
unsigned retval = 0;
void __iomem *ee_addr = regs + ROMCmd;
int read_cmd = location | (EE_READ_CMD << addr_len);
writel(EE_ENB & ~EE_CS, ee_addr);
writel(EE_ENB, ee_addr);
/* Shift the read command bits out. */
for (i = 4 + addr_len; i >= 0; i--) {
short dataval = (read_cmd & (1 << i)) ? EE_DATA_WRITE : 0;
writel(EE_ENB | dataval, ee_addr);
readl(ee_addr);
writel(EE_ENB | dataval | EE_SHIFT_CLK, ee_addr);
readl(ee_addr);
retval = (retval << 1) | ((readl(ee_addr) & EE_DATA_READ) ? 1 : 0);
}
writel(EE_ENB, ee_addr);
readl(ee_addr);
for (i = 16; i > 0; i--) {
writel(EE_ENB | EE_SHIFT_CLK, ee_addr);
readl(ee_addr);
retval = (retval << 1) | ((readl(ee_addr) & EE_DATA_READ) ? 1 : 0);
writel(EE_ENB, ee_addr);
readl(ee_addr);
}
/* Terminate the EEPROM access. */
writel(EE_ENB & ~EE_CS, ee_addr);
return retval;
}
static void __init de21041_get_srom_info (struct de_private *de)
{
unsigned i, sa_offset = 0, ofs;
u8 ee_data[DE_EEPROM_SIZE + 6] = {};
unsigned ee_addr_size = tulip_read_eeprom(de->regs, 0xff, 8) & 0x40000 ? 8 : 6;
struct de_srom_info_leaf *il;
void *bufp;
/* download entire eeprom */
for (i = 0; i < DE_EEPROM_WORDS; i++)
((u16 *)ee_data)[i] =
le16_to_cpu(tulip_read_eeprom(de->regs, i, ee_addr_size));
/* DEC now has a specification but early board makers
just put the address in the first EEPROM locations. */
/* This does memcmp(eedata, eedata+16, 8) */
[PATCH] de2104x: Resurrect Cobalt support for 2.6. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/tulip/de2104x.c | 5 +++++ 1 files changed, 5 insertions(+) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-10-10 21:50:46 +08:00
#ifndef CONFIG_MIPS_COBALT
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
for (i = 0; i < 8; i ++)
if (ee_data[i] != ee_data[16+i])
sa_offset = 20;
[PATCH] de2104x: Resurrect Cobalt support for 2.6. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/tulip/de2104x.c | 5 +++++ 1 files changed, 5 insertions(+) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-10-10 21:50:46 +08:00
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* store MAC address */
for (i = 0; i < 6; i ++)
de->dev->dev_addr[i] = ee_data[i + sa_offset];
/* get offset of controller 0 info leaf. ignore 2nd byte. */
ofs = ee_data[SROMC0InfoLeaf];
if (ofs >= (sizeof(ee_data) - sizeof(struct de_srom_info_leaf) - sizeof(struct de_srom_media_block)))
goto bad_srom;
/* get pointer to info leaf */
il = (struct de_srom_info_leaf *) &ee_data[ofs];
/* paranoia checks */
if (il->n_blocks == 0)
goto bad_srom;
if ((sizeof(ee_data) - ofs) <
(sizeof(struct de_srom_info_leaf) + (sizeof(struct de_srom_media_block) * il->n_blocks)))
goto bad_srom;
/* get default media type */
switch (DE_UNALIGNED_16(&il->default_media)) {
case 0x0001: de->media_type = DE_MEDIA_BNC; break;
case 0x0002: de->media_type = DE_MEDIA_AUI; break;
case 0x0204: de->media_type = DE_MEDIA_TP_FD; break;
default: de->media_type = DE_MEDIA_TP_AUTO; break;
}
if (netif_msg_probe(de))
printk(KERN_INFO "de%d: SROM leaf offset %u, default media %s\n",
de->board_idx, ofs,
media_name[de->media_type]);
/* init SIA register values to defaults */
for (i = 0; i < DE_MAX_MEDIA; i++) {
de->media[i].type = DE_MEDIA_INVALID;
de->media[i].csr13 = 0xffff;
de->media[i].csr14 = 0xffff;
de->media[i].csr15 = 0xffff;
}
/* parse media blocks to see what medias are supported,
* and if any custom CSR values are provided
*/
bufp = ((void *)il) + sizeof(*il);
for (i = 0; i < il->n_blocks; i++) {
struct de_srom_media_block *ib = bufp;
unsigned idx;
/* index based on media type in media block */
switch(ib->opts & MediaBlockMask) {
case 0: /* 10baseT */
de->media_supported |= SUPPORTED_TP | SUPPORTED_10baseT_Half
| SUPPORTED_Autoneg;
idx = DE_MEDIA_TP;
de->media[DE_MEDIA_TP_AUTO].type = DE_MEDIA_TP_AUTO;
break;
case 1: /* BNC */
de->media_supported |= SUPPORTED_BNC;
idx = DE_MEDIA_BNC;
break;
case 2: /* AUI */
de->media_supported |= SUPPORTED_AUI;
idx = DE_MEDIA_AUI;
break;
case 4: /* 10baseT-FD */
de->media_supported |= SUPPORTED_TP | SUPPORTED_10baseT_Full
| SUPPORTED_Autoneg;
idx = DE_MEDIA_TP_FD;
de->media[DE_MEDIA_TP_AUTO].type = DE_MEDIA_TP_AUTO;
break;
default:
goto bad_srom;
}
de->media[idx].type = idx;
if (netif_msg_probe(de))
printk(KERN_INFO "de%d: media block #%u: %s",
de->board_idx, i,
media_name[de->media[idx].type]);
bufp += sizeof (ib->opts);
if (ib->opts & MediaCustomCSRs) {
de->media[idx].csr13 = DE_UNALIGNED_16(&ib->csr13);
de->media[idx].csr14 = DE_UNALIGNED_16(&ib->csr14);
de->media[idx].csr15 = DE_UNALIGNED_16(&ib->csr15);
bufp += sizeof(ib->csr13) + sizeof(ib->csr14) +
sizeof(ib->csr15);
if (netif_msg_probe(de))
printk(" (%x,%x,%x)\n",
de->media[idx].csr13,
de->media[idx].csr14,
de->media[idx].csr15);
} else if (netif_msg_probe(de))
printk("\n");
if (bufp > ((void *)&ee_data[DE_EEPROM_SIZE - 3]))
break;
}
de->media_advertise = de->media_supported;
fill_defaults:
/* fill in defaults, for cases where custom CSRs not used */
for (i = 0; i < DE_MAX_MEDIA; i++) {
if (de->media[i].csr13 == 0xffff)
de->media[i].csr13 = t21041_csr13[i];
if (de->media[i].csr14 == 0xffff)
de->media[i].csr14 = t21041_csr14[i];
if (de->media[i].csr15 == 0xffff)
de->media[i].csr15 = t21041_csr15[i];
}
de->ee_data = kmalloc(DE_EEPROM_SIZE, GFP_KERNEL);
if (de->ee_data)
memcpy(de->ee_data, &ee_data[0], DE_EEPROM_SIZE);
return;
bad_srom:
/* for error cases, it's ok to assume we support all these */
for (i = 0; i < DE_MAX_MEDIA; i++)
de->media[i].type = i;
de->media_supported =
SUPPORTED_10baseT_Half |
SUPPORTED_10baseT_Full |
SUPPORTED_Autoneg |
SUPPORTED_TP |
SUPPORTED_AUI |
SUPPORTED_BNC;
goto fill_defaults;
}
static int __init de_init_one (struct pci_dev *pdev,
const struct pci_device_id *ent)
{
struct net_device *dev;
struct de_private *de;
int rc;
void __iomem *regs;
[netdrvr de2104x] store PCI bus addresses in unsigned long BZ# 4475.
2005-08-31 18:11:16 +08:00
unsigned long pciaddr;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int board_idx = -1;
board_idx++;
#ifndef MODULE
if (board_idx == 0)
printk("%s", version);
#endif
/* allocate a new ethernet device structure, and fill in defaults */
dev = alloc_etherdev(sizeof(struct de_private));
if (!dev)
return -ENOMEM;
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &pdev->dev);
dev->open = de_open;
dev->stop = de_close;
dev->set_multicast_list = de_set_rx_mode;
dev->hard_start_xmit = de_start_xmit;
dev->get_stats = de_get_stats;
dev->ethtool_ops = &de_ethtool_ops;
dev->tx_timeout = de_tx_timeout;
dev->watchdog_timeo = TX_TIMEOUT;
de = dev->priv;
de->de21040 = ent->driver_data == 0 ? 1 : 0;
de->pdev = pdev;
de->dev = dev;
de->msg_enable = (debug < 0 ? DE_DEF_MSG_ENABLE : debug);
de->board_idx = board_idx;
spin_lock_init (&de->lock);
init_timer(&de->media_timer);
if (de->de21040)
de->media_timer.function = de21040_media_timer;
else
de->media_timer.function = de21041_media_timer;
de->media_timer.data = (unsigned long) de;
netif_carrier_off(dev);
netif_stop_queue(dev);
/* wake up device, assign resources */
rc = pci_enable_device(pdev);
if (rc)
goto err_out_free;
/* reserve PCI resources to ensure driver atomicity */
rc = pci_request_regions(pdev, DRV_NAME);
if (rc)
goto err_out_disable;
/* check for invalid IRQ value */
if (pdev->irq < 2) {
rc = -EIO;
printk(KERN_ERR PFX "invalid irq (%d) for pci dev %s\n",
pdev->irq, pci_name(pdev));
goto err_out_res;
}
dev->irq = pdev->irq;
/* obtain and check validity of PCI I/O address */
pciaddr = pci_resource_start(pdev, 1);
if (!pciaddr) {
rc = -EIO;
printk(KERN_ERR PFX "no MMIO resource for pci dev %s\n",
pci_name(pdev));
goto err_out_res;
}
if (pci_resource_len(pdev, 1) < DE_REGS_SIZE) {
rc = -EIO;
printk(KERN_ERR PFX "MMIO resource (%lx) too small on pci dev %s\n",
pci_resource_len(pdev, 1), pci_name(pdev));
goto err_out_res;
}
/* remap CSR registers */
regs = ioremap_nocache(pciaddr, DE_REGS_SIZE);
if (!regs) {
rc = -EIO;
printk(KERN_ERR PFX "Cannot map PCI MMIO (%lx@%lx) on pci dev %s\n",
pci_resource_len(pdev, 1), pciaddr, pci_name(pdev));
goto err_out_res;
}
dev->base_addr = (unsigned long) regs;
de->regs = regs;
de_adapter_wake(de);
/* make sure hardware is not running */
rc = de_reset_mac(de);
if (rc) {
printk(KERN_ERR PFX "Cannot reset MAC, pci dev %s\n",
pci_name(pdev));
goto err_out_iomap;
}
/* get MAC address, initialize default media type and
* get list of supported media
*/
if (de->de21040) {
de21040_get_mac_address(de);
de21040_get_media_info(de);
} else {
de21041_get_srom_info(de);
}
/* register new network interface with kernel */
rc = register_netdev(dev);
if (rc)
goto err_out_iomap;
/* print info about board and interface just registered */
printk (KERN_INFO "%s: %s at 0x%lx, "
"%02x:%02x:%02x:%02x:%02x:%02x, "
"IRQ %d\n",
dev->name,
de->de21040 ? "21040" : "21041",
dev->base_addr,
dev->dev_addr[0], dev->dev_addr[1],
dev->dev_addr[2], dev->dev_addr[3],
dev->dev_addr[4], dev->dev_addr[5],
dev->irq);
pci_set_drvdata(pdev, dev);
/* enable busmastering */
pci_set_master(pdev);
/* put adapter to sleep */
de_adapter_sleep(de);
return 0;
err_out_iomap:
drivers/net: Remove pointless checks for NULL prior to calling kfree()
2005-10-29 04:53:13 +08:00
kfree(de->ee_data);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
iounmap(regs);
err_out_res:
pci_release_regions(pdev);
err_out_disable:
pci_disable_device(pdev);
err_out_free:
free_netdev(dev);
return rc;
}
static void __exit de_remove_one (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
struct de_private *de = dev->priv;
if (!dev)
BUG();
unregister_netdev(dev);
drivers/net: Remove pointless checks for NULL prior to calling kfree()
2005-10-29 04:53:13 +08:00
kfree(de->ee_data);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
iounmap(de->regs);
pci_release_regions(pdev);
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
free_netdev(dev);
}
#ifdef CONFIG_PM
[PATCH] u32 vs. pm_message_t fixes for drivers/net This fixes remaining u32s in drivers/ net. Signed-off-by: Pavel Machek <pavel@suse.cz> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-04-17 06:25:25 +08:00
static int de_suspend (struct pci_dev *pdev, pm_message_t state)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
struct net_device *dev = pci_get_drvdata (pdev);
struct de_private *de = dev->priv;
rtnl_lock();
if (netif_running (dev)) {
del_timer_sync(&de->media_timer);
disable_irq(dev->irq);
spin_lock_irq(&de->lock);
de_stop_hw(de);
netif_stop_queue(dev);
netif_device_detach(dev);
netif_carrier_off(dev);
spin_unlock_irq(&de->lock);
enable_irq(dev->irq);
/* Update the error counts. */
__de_get_stats(de);
synchronize_irq(dev->irq);
de_clean_rings(de);
de_adapter_sleep(de);
pci_disable_device(pdev);
} else {
netif_device_detach(dev);
}
rtnl_unlock();
return 0;
}
static int de_resume (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata (pdev);
struct de_private *de = dev->priv;
rtnl_lock();
if (netif_device_present(dev))
goto out;
if (netif_running(dev)) {
pci_enable_device(pdev);
de_init_hw(de);
netif_device_attach(dev);
} else {
netif_device_attach(dev);
}
out:
rtnl_unlock();
return 0;
}
#endif /* CONFIG_PM */
static struct pci_driver de_driver = {
.name = DRV_NAME,
.id_table = de_pci_tbl,
.probe = de_init_one,
.remove = __exit_p(de_remove_one),
#ifdef CONFIG_PM
.suspend = de_suspend,
.resume = de_resume,
#endif
};
static int __init de_init (void)
{
#ifdef MODULE
printk("%s", version);
#endif
return pci_module_init (&de_driver);
}
static void __exit de_exit (void)
{
pci_unregister_driver (&de_driver);
}
module_init(de_init);
module_exit(de_exit);