linux/drivers/net/tulip/tulip_core.c

1853 lines
54 KiB
C
Raw Normal View History

/* tulip_core.c: A DEC 21x4x-family ethernet driver for Linux. */
/*
Maintained by Valerie Henson <val_henson@linux.intel.com>
Copyright 2000,2001 The Linux Kernel Team
Written/copyright 1994-2001 by Donald Becker.
This software may be used and distributed according to the terms
of the GNU General Public License, incorporated herein by reference.
Please refer to Documentation/DocBook/tulip-user.{pdf,ps,html}
for more information on this driver, or visit the project
Web page at http://sourceforge.net/projects/tulip/
*/
#define DRV_NAME "tulip"
#ifdef CONFIG_TULIP_NAPI
#define DRV_VERSION "1.1.15-NAPI" /* Keep at least for test */
#else
#define DRV_VERSION "1.1.15"
#endif
#define DRV_RELDATE "Feb 27, 2007"
#include <linux/module.h>
#include <linux/pci.h>
#include "tulip.h"
#include <linux/init.h>
#include <linux/etherdevice.h>
#include <linux/delay.h>
#include <linux/mii.h>
#include <linux/ethtool.h>
#include <linux/crc32.h>
#include <asm/unaligned.h>
#include <asm/uaccess.h>
#ifdef CONFIG_SPARC
#include <asm/prom.h>
#endif
static char version[] __devinitdata =
"Linux Tulip driver version " DRV_VERSION " (" DRV_RELDATE ")\n";
/* A few user-configurable values. */
/* Maximum events (Rx packets, etc.) to handle at each interrupt. */
static unsigned int max_interrupt_work = 25;
#define MAX_UNITS 8
/* Used to pass the full-duplex flag, etc. */
static int full_duplex[MAX_UNITS];
static int options[MAX_UNITS];
static int mtu[MAX_UNITS]; /* Jumbo MTU for interfaces. */
/* The possible media types that can be set in options[] are: */
const char * const medianame[32] = {
"10baseT", "10base2", "AUI", "100baseTx",
"10baseT-FDX", "100baseTx-FDX", "100baseT4", "100baseFx",
"100baseFx-FDX", "MII 10baseT", "MII 10baseT-FDX", "MII",
"10baseT(forced)", "MII 100baseTx", "MII 100baseTx-FDX", "MII 100baseT4",
"MII 100baseFx-HDX", "MII 100baseFx-FDX", "Home-PNA 1Mbps", "Invalid-19",
"","","","", "","","","", "","","","Transceiver reset",
};
/* Set the copy breakpoint for the copy-only-tiny-buffer Rx structure. */
#if defined(__alpha__) || defined(__arm__) || defined(__hppa__) \
|| defined(CONFIG_SPARC) || defined(__ia64__) \
|| defined(__sh__) || defined(__mips__)
static int rx_copybreak = 1518;
#else
static int rx_copybreak = 100;
#endif
/*
Set the bus performance register.
Typical: Set 16 longword cache alignment, no burst limit.
Cache alignment bits 15:14 Burst length 13:8
0000 No alignment 0x00000000 unlimited 0800 8 longwords
4000 8 longwords 0100 1 longword 1000 16 longwords
8000 16 longwords 0200 2 longwords 2000 32 longwords
C000 32 longwords 0400 4 longwords
Warning: many older 486 systems are broken and require setting 0x00A04800
8 longword cache alignment, 8 longword burst.
ToDo: Non-Intel setting could be better.
*/
#if defined(__alpha__) || defined(__ia64__)
static int csr0 = 0x01A00000 | 0xE000;
#elif defined(__i386__) || defined(__powerpc__) || defined(__x86_64__)
static int csr0 = 0x01A00000 | 0x8000;
#elif defined(CONFIG_SPARC) || defined(__hppa__)
/* The UltraSparc PCI controllers will disconnect at every 64-byte
* crossing anyways so it makes no sense to tell Tulip to burst
* any more than that.
*/
static int csr0 = 0x01A00000 | 0x9000;
#elif defined(__arm__) || defined(__sh__)
static int csr0 = 0x01A00000 | 0x4800;
#elif defined(__mips__)
static int csr0 = 0x00200000 | 0x4000;
#else
#warning Processor architecture undefined!
static int csr0 = 0x00A00000 | 0x4800;
#endif
/* Operational parameters that usually are not changed. */
/* Time in jiffies before concluding the transmitter is hung. */
#define TX_TIMEOUT (4*HZ)
MODULE_AUTHOR("The Linux Kernel Team");
MODULE_DESCRIPTION("Digital 21*4* Tulip ethernet driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_VERSION);
module_param(tulip_debug, int, 0);
module_param(max_interrupt_work, int, 0);
module_param(rx_copybreak, int, 0);
module_param(csr0, int, 0);
module_param_array(options, int, NULL, 0);
module_param_array(full_duplex, int, NULL, 0);
#define PFX DRV_NAME ": "
#ifdef TULIP_DEBUG
int tulip_debug = TULIP_DEBUG;
#else
int tulip_debug = 1;
#endif
static void tulip_timer(unsigned long data)
{
struct net_device *dev = (struct net_device *)data;
struct tulip_private *tp = netdev_priv(dev);
if (netif_running(dev))
schedule_work(&tp->media_work);
}
/*
* This table use during operation for capabilities and media timer.
*
* It is indexed via the values in 'enum chips'
*/
struct tulip_chip_table tulip_tbl[] = {
{ }, /* placeholder for array, slot unused currently */
{ }, /* placeholder for array, slot unused currently */
/* DC21140 */
{ "Digital DS21140 Tulip", 128, 0x0001ebef,
HAS_MII | HAS_MEDIA_TABLE | CSR12_IN_SROM | HAS_PCI_MWI, tulip_timer,
tulip_media_task },
/* DC21142, DC21143 */
{ "Digital DS21142/43 Tulip", 128, 0x0801fbff,
HAS_MII | HAS_MEDIA_TABLE | ALWAYS_CHECK_MII | HAS_ACPI | HAS_NWAY
| HAS_INTR_MITIGATION | HAS_PCI_MWI, tulip_timer, t21142_media_task },
/* LC82C168 */
{ "Lite-On 82c168 PNIC", 256, 0x0001fbef,
HAS_MII | HAS_PNICNWAY, pnic_timer, },
/* MX98713 */
{ "Macronix 98713 PMAC", 128, 0x0001ebef,
HAS_MII | HAS_MEDIA_TABLE | CSR12_IN_SROM, mxic_timer, },
/* MX98715 */
{ "Macronix 98715 PMAC", 256, 0x0001ebef,
HAS_MEDIA_TABLE, mxic_timer, },
/* MX98725 */
{ "Macronix 98725 PMAC", 256, 0x0001ebef,
HAS_MEDIA_TABLE, mxic_timer, },
/* AX88140 */
{ "ASIX AX88140", 128, 0x0001fbff,
HAS_MII | HAS_MEDIA_TABLE | CSR12_IN_SROM | MC_HASH_ONLY
| IS_ASIX, tulip_timer, tulip_media_task },
/* PNIC2 */
{ "Lite-On PNIC-II", 256, 0x0801fbff,
HAS_MII | HAS_NWAY | HAS_8023X | HAS_PCI_MWI, pnic2_timer, },
/* COMET */
{ "ADMtek Comet", 256, 0x0001abef,
HAS_MII | MC_HASH_ONLY | COMET_MAC_ADDR, comet_timer, },
/* COMPEX9881 */
{ "Compex 9881 PMAC", 128, 0x0001ebef,
HAS_MII | HAS_MEDIA_TABLE | CSR12_IN_SROM, mxic_timer, },
/* I21145 */
{ "Intel DS21145 Tulip", 128, 0x0801fbff,
HAS_MII | HAS_MEDIA_TABLE | ALWAYS_CHECK_MII | HAS_ACPI
| HAS_NWAY | HAS_PCI_MWI, tulip_timer, tulip_media_task },
/* DM910X */
{ "Davicom DM9102/DM9102A", 128, 0x0001ebef,
HAS_MII | HAS_MEDIA_TABLE | CSR12_IN_SROM | HAS_ACPI,
tulip_timer, tulip_media_task },
/* RS7112 */
{ "Conexant LANfinity", 256, 0x0001ebef,
HAS_MII | HAS_ACPI, tulip_timer, tulip_media_task },
};
static struct pci_device_id tulip_pci_tbl[] = {
{ 0x1011, 0x0009, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DC21140 },
{ 0x1011, 0x0019, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DC21143 },
{ 0x11AD, 0x0002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, LC82C168 },
{ 0x10d9, 0x0512, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MX98713 },
{ 0x10d9, 0x0531, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MX98715 },
/* { 0x10d9, 0x0531, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MX98725 },*/
{ 0x125B, 0x1400, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AX88140 },
{ 0x11AD, 0xc115, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PNIC2 },
{ 0x1317, 0x0981, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1317, 0x0985, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1317, 0x1985, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1317, 0x9511, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x13D1, 0xAB02, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x13D1, 0xAB03, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x13D1, 0xAB08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x104A, 0x0981, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x104A, 0x2774, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1259, 0xa120, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x11F6, 0x9881, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMPEX9881 },
{ 0x8086, 0x0039, PCI_ANY_ID, PCI_ANY_ID, 0, 0, I21145 },
{ 0x1282, 0x9100, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DM910X },
{ 0x1282, 0x9102, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DM910X },
{ 0x1113, 0x1216, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1113, 0x1217, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MX98715 },
{ 0x1113, 0x9511, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1186, 0x1541, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1186, 0x1561, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1186, 0x1591, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x14f1, 0x1803, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CONEXANT },
{ 0x1626, 0x8410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1737, 0xAB09, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x1737, 0xAB08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x17B3, 0xAB08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ 0x10b7, 0x9300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET }, /* 3Com 3CSOHO100B-TX */
2005-06-27 04:18:32 +00:00
{ 0x14ea, 0xab08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET }, /* Planex FNW-3602-TX */
{ 0x1414, 0x0002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, COMET },
{ } /* terminate list */
};
MODULE_DEVICE_TABLE(pci, tulip_pci_tbl);
/* A full-duplex map for media types. */
const char tulip_media_cap[32] =
{0,0,0,16, 3,19,16,24, 27,4,7,5, 0,20,23,20, 28,31,0,0, };
static void tulip_tx_timeout(struct net_device *dev);
static void tulip_init_ring(struct net_device *dev);
static int tulip_start_xmit(struct sk_buff *skb, struct net_device *dev);
static int tulip_open(struct net_device *dev);
static int tulip_close(struct net_device *dev);
static void tulip_up(struct net_device *dev);
static void tulip_down(struct net_device *dev);
static struct net_device_stats *tulip_get_stats(struct net_device *dev);
static int private_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
static void set_rx_mode(struct net_device *dev);
#ifdef CONFIG_NET_POLL_CONTROLLER
static void poll_tulip(struct net_device *dev);
#endif
static void tulip_set_power_state (struct tulip_private *tp,
int sleep, int snooze)
{
if (tp->flags & HAS_ACPI) {
u32 tmp, newtmp;
pci_read_config_dword (tp->pdev, CFDD, &tmp);
newtmp = tmp & ~(CFDD_Sleep | CFDD_Snooze);
if (sleep)
newtmp |= CFDD_Sleep;
else if (snooze)
newtmp |= CFDD_Snooze;
if (tmp != newtmp)
pci_write_config_dword (tp->pdev, CFDD, newtmp);
}
}
static void tulip_up(struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
int next_tick = 3*HZ;
u32 reg;
int i;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-03 23:41:36 +00:00
#ifdef CONFIG_TULIP_NAPI
napi_enable(&tp->napi);
#endif
/* Wake the chip from sleep/snooze mode. */
tulip_set_power_state (tp, 0, 0);
/* On some chip revs we must set the MII/SYM port before the reset!? */
if (tp->mii_cnt || (tp->mtable && tp->mtable->has_mii))
iowrite32(0x00040000, ioaddr + CSR6);
/* Reset the chip, holding bit 0 set at least 50 PCI cycles. */
iowrite32(0x00000001, ioaddr + CSR0);
pci_read_config_dword(tp->pdev, PCI_COMMAND, &reg); /* flush write */
udelay(100);
/* Deassert reset.
Wait the specified 50 PCI cycles after a reset by initializing
Tx and Rx queues and the address filter list. */
iowrite32(tp->csr0, ioaddr + CSR0);
pci_read_config_dword(tp->pdev, PCI_COMMAND, &reg); /* flush write */
udelay(100);
if (tulip_debug > 1)
printk(KERN_DEBUG "%s: tulip_up(), irq==%d.\n", dev->name, dev->irq);
iowrite32(tp->rx_ring_dma, ioaddr + CSR3);
iowrite32(tp->tx_ring_dma, ioaddr + CSR4);
tp->cur_rx = tp->cur_tx = 0;
tp->dirty_rx = tp->dirty_tx = 0;
if (tp->flags & MC_HASH_ONLY) {
u32 addr_low = get_unaligned_le32(dev->dev_addr);
u32 addr_high = get_unaligned_le16(dev->dev_addr + 4);
if (tp->chip_id == AX88140) {
iowrite32(0, ioaddr + CSR13);
iowrite32(addr_low, ioaddr + CSR14);
iowrite32(1, ioaddr + CSR13);
iowrite32(addr_high, ioaddr + CSR14);
} else if (tp->flags & COMET_MAC_ADDR) {
iowrite32(addr_low, ioaddr + 0xA4);
iowrite32(addr_high, ioaddr + 0xA8);
iowrite32(0, ioaddr + 0xAC);
iowrite32(0, ioaddr + 0xB0);
}
} else {
/* This is set_rx_mode(), but without starting the transmitter. */
u16 *eaddrs = (u16 *)dev->dev_addr;
u16 *setup_frm = &tp->setup_frame[15*6];
dma_addr_t mapping;
/* 21140 bug: you must add the broadcast address. */
memset(tp->setup_frame, 0xff, sizeof(tp->setup_frame));
/* Fill the final entry of the table with our physical address. */
*setup_frm++ = eaddrs[0]; *setup_frm++ = eaddrs[0];
*setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1];
*setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2];
mapping = pci_map_single(tp->pdev, tp->setup_frame,
sizeof(tp->setup_frame),
PCI_DMA_TODEVICE);
tp->tx_buffers[tp->cur_tx].skb = NULL;
tp->tx_buffers[tp->cur_tx].mapping = mapping;
/* Put the setup frame on the Tx list. */
tp->tx_ring[tp->cur_tx].length = cpu_to_le32(0x08000000 | 192);
tp->tx_ring[tp->cur_tx].buffer1 = cpu_to_le32(mapping);
tp->tx_ring[tp->cur_tx].status = cpu_to_le32(DescOwned);
tp->cur_tx++;
}
tp->saved_if_port = dev->if_port;
if (dev->if_port == 0)
dev->if_port = tp->default_port;
/* Allow selecting a default media. */
i = 0;
if (tp->mtable == NULL)
goto media_picked;
if (dev->if_port) {
int looking_for = tulip_media_cap[dev->if_port] & MediaIsMII ? 11 :
(dev->if_port == 12 ? 0 : dev->if_port);
for (i = 0; i < tp->mtable->leafcount; i++)
if (tp->mtable->mleaf[i].media == looking_for) {
printk(KERN_INFO "%s: Using user-specified media %s.\n",
dev->name, medianame[dev->if_port]);
goto media_picked;
}
}
if ((tp->mtable->defaultmedia & 0x0800) == 0) {
int looking_for = tp->mtable->defaultmedia & MEDIA_MASK;
for (i = 0; i < tp->mtable->leafcount; i++)
if (tp->mtable->mleaf[i].media == looking_for) {
printk(KERN_INFO "%s: Using EEPROM-set media %s.\n",
dev->name, medianame[looking_for]);
goto media_picked;
}
}
/* Start sensing first non-full-duplex media. */
for (i = tp->mtable->leafcount - 1;
(tulip_media_cap[tp->mtable->mleaf[i].media] & MediaAlwaysFD) && i > 0; i--)
;
media_picked:
tp->csr6 = 0;
tp->cur_index = i;
tp->nwayset = 0;
if (dev->if_port) {
if (tp->chip_id == DC21143 &&
(tulip_media_cap[dev->if_port] & MediaIsMII)) {
/* We must reset the media CSRs when we force-select MII mode. */
iowrite32(0x0000, ioaddr + CSR13);
iowrite32(0x0000, ioaddr + CSR14);
iowrite32(0x0008, ioaddr + CSR15);
}
tulip_select_media(dev, 1);
} else if (tp->chip_id == DC21142) {
if (tp->mii_cnt) {
tulip_select_media(dev, 1);
if (tulip_debug > 1)
printk(KERN_INFO "%s: Using MII transceiver %d, status "
"%4.4x.\n",
dev->name, tp->phys[0], tulip_mdio_read(dev, tp->phys[0], 1));
iowrite32(csr6_mask_defstate, ioaddr + CSR6);
tp->csr6 = csr6_mask_hdcap;
dev->if_port = 11;
iowrite32(0x0000, ioaddr + CSR13);
iowrite32(0x0000, ioaddr + CSR14);
} else
t21142_start_nway(dev);
} else if (tp->chip_id == PNIC2) {
/* for initial startup advertise 10/100 Full and Half */
tp->sym_advertise = 0x01E0;
/* enable autonegotiate end interrupt */
iowrite32(ioread32(ioaddr+CSR5)| 0x00008010, ioaddr + CSR5);
iowrite32(ioread32(ioaddr+CSR7)| 0x00008010, ioaddr + CSR7);
pnic2_start_nway(dev);
} else if (tp->chip_id == LC82C168 && ! tp->medialock) {
if (tp->mii_cnt) {
dev->if_port = 11;
tp->csr6 = 0x814C0000 | (tp->full_duplex ? 0x0200 : 0);
iowrite32(0x0001, ioaddr + CSR15);
} else if (ioread32(ioaddr + CSR5) & TPLnkPass)
pnic_do_nway(dev);
else {
/* Start with 10mbps to do autonegotiation. */
iowrite32(0x32, ioaddr + CSR12);
tp->csr6 = 0x00420000;
iowrite32(0x0001B078, ioaddr + 0xB8);
iowrite32(0x0201B078, ioaddr + 0xB8);
next_tick = 1*HZ;
}
} else if ((tp->chip_id == MX98713 || tp->chip_id == COMPEX9881)
&& ! tp->medialock) {
dev->if_port = 0;
tp->csr6 = 0x01880000 | (tp->full_duplex ? 0x0200 : 0);
iowrite32(0x0f370000 | ioread16(ioaddr + 0x80), ioaddr + 0x80);
} else if (tp->chip_id == MX98715 || tp->chip_id == MX98725) {
/* Provided by BOLO, Macronix - 12/10/1998. */
dev->if_port = 0;
tp->csr6 = 0x01a80200;
iowrite32(0x0f370000 | ioread16(ioaddr + 0x80), ioaddr + 0x80);
iowrite32(0x11000 | ioread16(ioaddr + 0xa0), ioaddr + 0xa0);
} else if (tp->chip_id == COMET || tp->chip_id == CONEXANT) {
/* Enable automatic Tx underrun recovery. */
iowrite32(ioread32(ioaddr + 0x88) | 1, ioaddr + 0x88);
dev->if_port = tp->mii_cnt ? 11 : 0;
tp->csr6 = 0x00040000;
} else if (tp->chip_id == AX88140) {
tp->csr6 = tp->mii_cnt ? 0x00040100 : 0x00000100;
} else
tulip_select_media(dev, 1);
/* Start the chip's Tx to process setup frame. */
tulip_stop_rxtx(tp);
barrier();
udelay(5);
iowrite32(tp->csr6 | TxOn, ioaddr + CSR6);
/* Enable interrupts by setting the interrupt mask. */
iowrite32(tulip_tbl[tp->chip_id].valid_intrs, ioaddr + CSR5);
iowrite32(tulip_tbl[tp->chip_id].valid_intrs, ioaddr + CSR7);
tulip_start_rxtx(tp);
iowrite32(0, ioaddr + CSR2); /* Rx poll demand */
if (tulip_debug > 2) {
printk(KERN_DEBUG "%s: Done tulip_up(), CSR0 %8.8x, CSR5 %8.8x CSR6 %8.8x.\n",
dev->name, ioread32(ioaddr + CSR0), ioread32(ioaddr + CSR5),
ioread32(ioaddr + CSR6));
}
/* Set the timer to switch to check for link beat and perhaps switch
to an alternate media type. */
tp->timer.expires = RUN_AT(next_tick);
add_timer(&tp->timer);
#ifdef CONFIG_TULIP_NAPI
init_timer(&tp->oom_timer);
tp->oom_timer.data = (unsigned long)dev;
tp->oom_timer.function = oom_timer;
#endif
}
static int
tulip_open(struct net_device *dev)
{
int retval;
if ((retval = request_irq(dev->irq, &tulip_interrupt, IRQF_SHARED, dev->name, dev)))
return retval;
tulip_init_ring (dev);
tulip_up (dev);
netif_start_queue (dev);
return 0;
}
static void tulip_tx_timeout(struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
unsigned long flags;
spin_lock_irqsave (&tp->lock, flags);
if (tulip_media_cap[dev->if_port] & MediaIsMII) {
/* Do nothing -- the media monitor should handle this. */
if (tulip_debug > 1)
printk(KERN_WARNING "%s: Transmit timeout using MII device.\n",
dev->name);
} else if (tp->chip_id == DC21140 || tp->chip_id == DC21142
|| tp->chip_id == MX98713 || tp->chip_id == COMPEX9881
|| tp->chip_id == DM910X) {
printk(KERN_WARNING "%s: 21140 transmit timed out, status %8.8x, "
"SIA %8.8x %8.8x %8.8x %8.8x, resetting...\n",
dev->name, ioread32(ioaddr + CSR5), ioread32(ioaddr + CSR12),
ioread32(ioaddr + CSR13), ioread32(ioaddr + CSR14), ioread32(ioaddr + CSR15));
tp->timeout_recovery = 1;
schedule_work(&tp->media_work);
goto out_unlock;
} else if (tp->chip_id == PNIC2) {
printk(KERN_WARNING "%s: PNIC2 transmit timed out, status %8.8x, "
"CSR6/7 %8.8x / %8.8x CSR12 %8.8x, resetting...\n",
dev->name, (int)ioread32(ioaddr + CSR5), (int)ioread32(ioaddr + CSR6),
(int)ioread32(ioaddr + CSR7), (int)ioread32(ioaddr + CSR12));
} else {
printk(KERN_WARNING "%s: Transmit timed out, status %8.8x, CSR12 "
"%8.8x, resetting...\n",
dev->name, ioread32(ioaddr + CSR5), ioread32(ioaddr + CSR12));
dev->if_port = 0;
}
#if defined(way_too_many_messages)
if (tulip_debug > 3) {
int i;
for (i = 0; i < RX_RING_SIZE; i++) {
u8 *buf = (u8 *)(tp->rx_ring[i].buffer1);
int j;
printk(KERN_DEBUG "%2d: %8.8x %8.8x %8.8x %8.8x "
"%2.2x %2.2x %2.2x.\n",
i, (unsigned int)tp->rx_ring[i].status,
(unsigned int)tp->rx_ring[i].length,
(unsigned int)tp->rx_ring[i].buffer1,
(unsigned int)tp->rx_ring[i].buffer2,
buf[0], buf[1], buf[2]);
for (j = 0; buf[j] != 0xee && j < 1600; j++)
if (j < 100) printk(" %2.2x", buf[j]);
printk(" j=%d.\n", j);
}
printk(KERN_DEBUG " Rx ring %8.8x: ", (int)tp->rx_ring);
for (i = 0; i < RX_RING_SIZE; i++)
printk(" %8.8x", (unsigned int)tp->rx_ring[i].status);
printk("\n" KERN_DEBUG " Tx ring %8.8x: ", (int)tp->tx_ring);
for (i = 0; i < TX_RING_SIZE; i++)
printk(" %8.8x", (unsigned int)tp->tx_ring[i].status);
printk("\n");
}
#endif
tulip_tx_timeout_complete(tp, ioaddr);
out_unlock:
spin_unlock_irqrestore (&tp->lock, flags);
dev->trans_start = jiffies;
netif_wake_queue (dev);
}
/* Initialize the Rx and Tx rings, along with various 'dev' bits. */
static void tulip_init_ring(struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
int i;
tp->susp_rx = 0;
tp->ttimer = 0;
tp->nir = 0;
for (i = 0; i < RX_RING_SIZE; i++) {
tp->rx_ring[i].status = 0x00000000;
tp->rx_ring[i].length = cpu_to_le32(PKT_BUF_SZ);
tp->rx_ring[i].buffer2 = cpu_to_le32(tp->rx_ring_dma + sizeof(struct tulip_rx_desc) * (i + 1));
tp->rx_buffers[i].skb = NULL;
tp->rx_buffers[i].mapping = 0;
}
/* Mark the last entry as wrapping the ring. */
tp->rx_ring[i-1].length = cpu_to_le32(PKT_BUF_SZ | DESC_RING_WRAP);
tp->rx_ring[i-1].buffer2 = cpu_to_le32(tp->rx_ring_dma);
for (i = 0; i < RX_RING_SIZE; i++) {
dma_addr_t mapping;
/* Note the receive buffer must be longword aligned.
dev_alloc_skb() provides 16 byte alignment. But do *not*
use skb_reserve() to align the IP header! */
struct sk_buff *skb = dev_alloc_skb(PKT_BUF_SZ);
tp->rx_buffers[i].skb = skb;
if (skb == NULL)
break;
mapping = pci_map_single(tp->pdev, skb->data,
PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
tp->rx_buffers[i].mapping = mapping;
skb->dev = dev; /* Mark as being used by this device. */
tp->rx_ring[i].status = cpu_to_le32(DescOwned); /* Owned by Tulip chip */
tp->rx_ring[i].buffer1 = cpu_to_le32(mapping);
}
tp->dirty_rx = (unsigned int)(i - RX_RING_SIZE);
/* The Tx buffer descriptor is filled in as needed, but we
do need to clear the ownership bit. */
for (i = 0; i < TX_RING_SIZE; i++) {
tp->tx_buffers[i].skb = NULL;
tp->tx_buffers[i].mapping = 0;
tp->tx_ring[i].status = 0x00000000;
tp->tx_ring[i].buffer2 = cpu_to_le32(tp->tx_ring_dma + sizeof(struct tulip_tx_desc) * (i + 1));
}
tp->tx_ring[i-1].buffer2 = cpu_to_le32(tp->tx_ring_dma);
}
static int
tulip_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
int entry;
u32 flag;
dma_addr_t mapping;
spin_lock_irq(&tp->lock);
/* Calculate the next Tx descriptor entry. */
entry = tp->cur_tx % TX_RING_SIZE;
tp->tx_buffers[entry].skb = skb;
mapping = pci_map_single(tp->pdev, skb->data,
skb->len, PCI_DMA_TODEVICE);
tp->tx_buffers[entry].mapping = mapping;
tp->tx_ring[entry].buffer1 = cpu_to_le32(mapping);
if (tp->cur_tx - tp->dirty_tx < TX_RING_SIZE/2) {/* Typical path */
flag = 0x60000000; /* No interrupt */
} else if (tp->cur_tx - tp->dirty_tx == TX_RING_SIZE/2) {
flag = 0xe0000000; /* Tx-done intr. */
} else if (tp->cur_tx - tp->dirty_tx < TX_RING_SIZE - 2) {
flag = 0x60000000; /* No Tx-done intr. */
} else { /* Leave room for set_rx_mode() to fill entries. */
flag = 0xe0000000; /* Tx-done intr. */
netif_stop_queue(dev);
}
if (entry == TX_RING_SIZE-1)
flag = 0xe0000000 | DESC_RING_WRAP;
tp->tx_ring[entry].length = cpu_to_le32(skb->len | flag);
/* if we were using Transmit Automatic Polling, we would need a
* wmb() here. */
tp->tx_ring[entry].status = cpu_to_le32(DescOwned);
wmb();
tp->cur_tx++;
/* Trigger an immediate transmit demand. */
iowrite32(0, tp->base_addr + CSR1);
spin_unlock_irq(&tp->lock);
dev->trans_start = jiffies;
return 0;
}
static void tulip_clean_tx_ring(struct tulip_private *tp)
{
unsigned int dirty_tx;
for (dirty_tx = tp->dirty_tx ; tp->cur_tx - dirty_tx > 0;
dirty_tx++) {
int entry = dirty_tx % TX_RING_SIZE;
int status = le32_to_cpu(tp->tx_ring[entry].status);
if (status < 0) {
tp->stats.tx_errors++; /* It wasn't Txed */
tp->tx_ring[entry].status = 0;
}
/* Check for Tx filter setup frames. */
if (tp->tx_buffers[entry].skb == NULL) {
/* test because dummy frames not mapped */
if (tp->tx_buffers[entry].mapping)
pci_unmap_single(tp->pdev,
tp->tx_buffers[entry].mapping,
sizeof(tp->setup_frame),
PCI_DMA_TODEVICE);
continue;
}
pci_unmap_single(tp->pdev, tp->tx_buffers[entry].mapping,
tp->tx_buffers[entry].skb->len,
PCI_DMA_TODEVICE);
/* Free the original skb. */
dev_kfree_skb_irq(tp->tx_buffers[entry].skb);
tp->tx_buffers[entry].skb = NULL;
tp->tx_buffers[entry].mapping = 0;
}
}
static void tulip_down (struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
unsigned long flags;
flush_scheduled_work();
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-03 23:41:36 +00:00
#ifdef CONFIG_TULIP_NAPI
napi_disable(&tp->napi);
#endif
del_timer_sync (&tp->timer);
#ifdef CONFIG_TULIP_NAPI
del_timer_sync (&tp->oom_timer);
#endif
spin_lock_irqsave (&tp->lock, flags);
/* Disable interrupts by clearing the interrupt mask. */
iowrite32 (0x00000000, ioaddr + CSR7);
/* Stop the Tx and Rx processes. */
tulip_stop_rxtx(tp);
/* prepare receive buffers */
tulip_refill_rx(dev);
/* release any unconsumed transmit buffers */
tulip_clean_tx_ring(tp);
if (ioread32 (ioaddr + CSR6) != 0xffffffff)
tp->stats.rx_missed_errors += ioread32 (ioaddr + CSR8) & 0xffff;
spin_unlock_irqrestore (&tp->lock, flags);
init_timer(&tp->timer);
tp->timer.data = (unsigned long)dev;
tp->timer.function = tulip_tbl[tp->chip_id].media_timer;
dev->if_port = tp->saved_if_port;
/* Leave the driver in snooze, not sleep, mode. */
tulip_set_power_state (tp, 0, 1);
}
static int tulip_close (struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
int i;
netif_stop_queue (dev);
tulip_down (dev);
if (tulip_debug > 1)
printk (KERN_DEBUG "%s: Shutting down ethercard, status was %2.2x.\n",
dev->name, ioread32 (ioaddr + CSR5));
free_irq (dev->irq, dev);
/* Free all the skbuffs in the Rx queue. */
for (i = 0; i < RX_RING_SIZE; i++) {
struct sk_buff *skb = tp->rx_buffers[i].skb;
dma_addr_t mapping = tp->rx_buffers[i].mapping;
tp->rx_buffers[i].skb = NULL;
tp->rx_buffers[i].mapping = 0;
tp->rx_ring[i].status = 0; /* Not owned by Tulip chip. */
tp->rx_ring[i].length = 0;
/* An invalid address. */
tp->rx_ring[i].buffer1 = cpu_to_le32(0xBADF00D0);
if (skb) {
pci_unmap_single(tp->pdev, mapping, PKT_BUF_SZ,
PCI_DMA_FROMDEVICE);
dev_kfree_skb (skb);
}
}
for (i = 0; i < TX_RING_SIZE; i++) {
struct sk_buff *skb = tp->tx_buffers[i].skb;
if (skb != NULL) {
pci_unmap_single(tp->pdev, tp->tx_buffers[i].mapping,
skb->len, PCI_DMA_TODEVICE);
dev_kfree_skb (skb);
}
tp->tx_buffers[i].skb = NULL;
tp->tx_buffers[i].mapping = 0;
}
return 0;
}
static struct net_device_stats *tulip_get_stats(struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
if (netif_running(dev)) {
unsigned long flags;
spin_lock_irqsave (&tp->lock, flags);
tp->stats.rx_missed_errors += ioread32(ioaddr + CSR8) & 0xffff;
spin_unlock_irqrestore(&tp->lock, flags);
}
return &tp->stats;
}
static void tulip_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
struct tulip_private *np = netdev_priv(dev);
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
strcpy(info->bus_info, pci_name(np->pdev));
}
static const struct ethtool_ops ops = {
.get_drvinfo = tulip_get_drvinfo
};
/* Provide ioctl() calls to examine the MII xcvr state. */
static int private_ioctl (struct net_device *dev, struct ifreq *rq, int cmd)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
struct mii_ioctl_data *data = if_mii(rq);
const unsigned int phy_idx = 0;
int phy = tp->phys[phy_idx] & 0x1f;
unsigned int regnum = data->reg_num;
switch (cmd) {
case SIOCGMIIPHY: /* Get address of MII PHY in use. */
if (tp->mii_cnt)
data->phy_id = phy;
else if (tp->flags & HAS_NWAY)
data->phy_id = 32;
else if (tp->chip_id == COMET)
data->phy_id = 1;
else
return -ENODEV;
case SIOCGMIIREG: /* Read MII PHY register. */
if (data->phy_id == 32 && (tp->flags & HAS_NWAY)) {
int csr12 = ioread32 (ioaddr + CSR12);
int csr14 = ioread32 (ioaddr + CSR14);
switch (regnum) {
case 0:
if (((csr14<<5) & 0x1000) ||
(dev->if_port == 5 && tp->nwayset))
data->val_out = 0x1000;
else
data->val_out = (tulip_media_cap[dev->if_port]&MediaIs100 ? 0x2000 : 0)
| (tulip_media_cap[dev->if_port]&MediaIsFD ? 0x0100 : 0);
break;
case 1:
data->val_out =
0x1848 +
((csr12&0x7000) == 0x5000 ? 0x20 : 0) +
((csr12&0x06) == 6 ? 0 : 4);
data->val_out |= 0x6048;
break;
case 4:
/* Advertised value, bogus 10baseTx-FD value from CSR6. */
data->val_out =
((ioread32(ioaddr + CSR6) >> 3) & 0x0040) +
((csr14 >> 1) & 0x20) + 1;
data->val_out |= ((csr14 >> 9) & 0x03C0);
break;
case 5: data->val_out = tp->lpar; break;
default: data->val_out = 0; break;
}
} else {
data->val_out = tulip_mdio_read (dev, data->phy_id & 0x1f, regnum);
}
return 0;
case SIOCSMIIREG: /* Write MII PHY register. */
if (!capable (CAP_NET_ADMIN))
return -EPERM;
if (regnum & ~0x1f)
return -EINVAL;
if (data->phy_id == phy) {
u16 value = data->val_in;
switch (regnum) {
case 0: /* Check for autonegotiation on or reset. */
tp->full_duplex_lock = (value & 0x9000) ? 0 : 1;
if (tp->full_duplex_lock)
tp->full_duplex = (value & 0x0100) ? 1 : 0;
break;
case 4:
tp->advertising[phy_idx] =
tp->mii_advertise = data->val_in;
break;
}
}
if (data->phy_id == 32 && (tp->flags & HAS_NWAY)) {
u16 value = data->val_in;
if (regnum == 0) {
if ((value & 0x1200) == 0x1200) {
if (tp->chip_id == PNIC2) {
pnic2_start_nway (dev);
} else {
t21142_start_nway (dev);
}
}
} else if (regnum == 4)
tp->sym_advertise = value;
} else {
tulip_mdio_write (dev, data->phy_id & 0x1f, regnum, data->val_in);
}
return 0;
default:
return -EOPNOTSUPP;
}
return -EOPNOTSUPP;
}
/* Set or clear the multicast filter for this adaptor.
Note that we only use exclusion around actually queueing the
new frame, not around filling tp->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 tulip_private *tp = netdev_priv(dev);
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 = &tp->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 tulip_private *tp = netdev_priv(dev);
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 = &tp->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 set_rx_mode(struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->base_addr;
int csr6;
csr6 = ioread32(ioaddr + CSR6) & ~0x00D5;
tp->csr6 &= ~0x00D5;
if (dev->flags & IFF_PROMISC) { /* Set promiscuous. */
tp->csr6 |= AcceptAllMulticast | AcceptAllPhys;
csr6 |= AcceptAllMulticast | AcceptAllPhys;
} else if ((dev->mc_count > 1000) || (dev->flags & IFF_ALLMULTI)) {
/* Too many to filter well -- accept all multicasts. */
tp->csr6 |= AcceptAllMulticast;
csr6 |= AcceptAllMulticast;
} else if (tp->flags & MC_HASH_ONLY) {
/* Some work-alikes have only a 64-entry hash filter table. */
/* Should verify correctness on big-endian/__powerpc__ */
struct dev_mc_list *mclist;
int i;
if (dev->mc_count > 64) { /* Arbitrary non-effective limit. */
tp->csr6 |= AcceptAllMulticast;
csr6 |= AcceptAllMulticast;
} else {
u32 mc_filter[2] = {0, 0}; /* Multicast hash filter */
int filterbit;
for (i = 0, mclist = dev->mc_list; mclist && i < dev->mc_count;
i++, mclist = mclist->next) {
if (tp->flags & COMET_MAC_ADDR)
filterbit = ether_crc_le(ETH_ALEN, mclist->dmi_addr);
else
filterbit = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26;
filterbit &= 0x3f;
mc_filter[filterbit >> 5] |= 1 << (filterbit & 31);
if (tulip_debug > 2) {
DECLARE_MAC_BUF(mac);
printk(KERN_INFO "%s: Added filter for %s"
" %8.8x bit %d.\n",
dev->name, print_mac(mac, mclist->dmi_addr),
ether_crc(ETH_ALEN, mclist->dmi_addr), filterbit);
}
}
if (mc_filter[0] == tp->mc_filter[0] &&
mc_filter[1] == tp->mc_filter[1])
; /* No change. */
else if (tp->flags & IS_ASIX) {
iowrite32(2, ioaddr + CSR13);
iowrite32(mc_filter[0], ioaddr + CSR14);
iowrite32(3, ioaddr + CSR13);
iowrite32(mc_filter[1], ioaddr + CSR14);
} else if (tp->flags & COMET_MAC_ADDR) {
iowrite32(mc_filter[0], ioaddr + 0xAC);
iowrite32(mc_filter[1], ioaddr + 0xB0);
}
tp->mc_filter[0] = mc_filter[0];
tp->mc_filter[1] = mc_filter[1];
}
} else {
unsigned long flags;
u32 tx_flags = 0x08000000 | 192;
/* 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(tp->setup_frame, dev);
tx_flags = 0x08400000 | 192;
} else {
build_setup_frame_perfect(tp->setup_frame, dev);
}
spin_lock_irqsave(&tp->lock, flags);
if (tp->cur_tx - tp->dirty_tx > TX_RING_SIZE - 2) {
/* Same setup recently queued, we need not add it. */
} else {
unsigned int entry;
int dummy = -1;
/* Now add this frame to the Tx list. */
entry = tp->cur_tx++ % TX_RING_SIZE;
if (entry != 0) {
/* Avoid a chip errata by prefixing a dummy entry. */
tp->tx_buffers[entry].skb = NULL;
tp->tx_buffers[entry].mapping = 0;
tp->tx_ring[entry].length =
(entry == TX_RING_SIZE-1) ? cpu_to_le32(DESC_RING_WRAP) : 0;
tp->tx_ring[entry].buffer1 = 0;
/* Must set DescOwned later to avoid race with chip */
dummy = entry;
entry = tp->cur_tx++ % TX_RING_SIZE;
}
tp->tx_buffers[entry].skb = NULL;
tp->tx_buffers[entry].mapping =
pci_map_single(tp->pdev, tp->setup_frame,
sizeof(tp->setup_frame),
PCI_DMA_TODEVICE);
/* Put the setup frame on the Tx list. */
if (entry == TX_RING_SIZE-1)
tx_flags |= DESC_RING_WRAP; /* Wrap ring. */
tp->tx_ring[entry].length = cpu_to_le32(tx_flags);
tp->tx_ring[entry].buffer1 =
cpu_to_le32(tp->tx_buffers[entry].mapping);
tp->tx_ring[entry].status = cpu_to_le32(DescOwned);
if (dummy >= 0)
tp->tx_ring[dummy].status = cpu_to_le32(DescOwned);
if (tp->cur_tx - tp->dirty_tx >= TX_RING_SIZE - 2)
netif_stop_queue(dev);
/* Trigger an immediate transmit demand. */
iowrite32(0, ioaddr + CSR1);
}
spin_unlock_irqrestore(&tp->lock, flags);
}
iowrite32(csr6, ioaddr + CSR6);
}
#ifdef CONFIG_TULIP_MWI
static void __devinit tulip_mwi_config (struct pci_dev *pdev,
struct net_device *dev)
{
struct tulip_private *tp = netdev_priv(dev);
u8 cache;
u16 pci_command;
u32 csr0;
if (tulip_debug > 3)
printk(KERN_DEBUG "%s: tulip_mwi_config()\n", pci_name(pdev));
tp->csr0 = csr0 = 0;
/* if we have any cache line size at all, we can do MRM and MWI */
csr0 |= MRM | MWI;
/* Enable MWI in the standard PCI command bit.
* Check for the case where MWI is desired but not available
*/
pci_try_set_mwi(pdev);
/* read result from hardware (in case bit refused to enable) */
pci_read_config_word(pdev, PCI_COMMAND, &pci_command);
if ((csr0 & MWI) && (!(pci_command & PCI_COMMAND_INVALIDATE)))
csr0 &= ~MWI;
/* if cache line size hardwired to zero, no MWI */
pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &cache);
if ((csr0 & MWI) && (cache == 0)) {
csr0 &= ~MWI;
pci_clear_mwi(pdev);
}
/* assign per-cacheline-size cache alignment and
* burst length values
*/
switch (cache) {
case 8:
csr0 |= MRL | (1 << CALShift) | (16 << BurstLenShift);
break;
case 16:
csr0 |= MRL | (2 << CALShift) | (16 << BurstLenShift);
break;
case 32:
csr0 |= MRL | (3 << CALShift) | (32 << BurstLenShift);
break;
default:
cache = 0;
break;
}
/* if we have a good cache line size, we by now have a good
* csr0, so save it and exit
*/
if (cache)
goto out;
/* we don't have a good csr0 or cache line size, disable MWI */
if (csr0 & MWI) {
pci_clear_mwi(pdev);
csr0 &= ~MWI;
}
/* sane defaults for burst length and cache alignment
* originally from de4x5 driver
*/
csr0 |= (8 << BurstLenShift) | (1 << CALShift);
out:
tp->csr0 = csr0;
if (tulip_debug > 2)
printk(KERN_DEBUG "%s: MWI config cacheline=%d, csr0=%08x\n",
pci_name(pdev), cache, csr0);
}
#endif
/*
* Chips that have the MRM/reserved bit quirk and the burst quirk. That
* is the DM910X and the on chip ULi devices
*/
static int tulip_uli_dm_quirk(struct pci_dev *pdev)
{
if (pdev->vendor == 0x1282 && pdev->device == 0x9102)
return 1;
return 0;
}
static int __devinit tulip_init_one (struct pci_dev *pdev,
const struct pci_device_id *ent)
{
struct tulip_private *tp;
/* See note below on the multiport cards. */
static unsigned char last_phys_addr[6] = {0x00, 'L', 'i', 'n', 'u', 'x'};
static struct pci_device_id early_486_chipsets[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82424) },
{ PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_496) },
{ },
};
static int last_irq;
static int multiport_cnt; /* For four-port boards w/one EEPROM */
int i, irq;
unsigned short sum;
unsigned char *ee_data;
struct net_device *dev;
void __iomem *ioaddr;
static int board_idx = -1;
int chip_idx = ent->driver_data;
const char *chip_name = tulip_tbl[chip_idx].chip_name;
unsigned int eeprom_missing = 0;
unsigned int force_csr0 = 0;
DECLARE_MAC_BUF(mac);
#ifndef MODULE
static int did_version; /* Already printed version info. */
if (tulip_debug > 0 && did_version++ == 0)
printk (KERN_INFO "%s", version);
#endif
board_idx++;
/*
* Lan media wire a tulip chip to a wan interface. Needs a very
* different driver (lmc driver)
*/
if (pdev->subsystem_vendor == PCI_VENDOR_ID_LMC) {
printk (KERN_ERR PFX "skipping LMC card.\n");
return -ENODEV;
}
/*
* Early DM9100's need software CRC and the DMFE driver
*/
if (pdev->vendor == 0x1282 && pdev->device == 0x9100)
{
/* Read Chip revision */
if (pdev->revision < 0x30)
{
printk(KERN_ERR PFX "skipping early DM9100 with Crc bug (use dmfe)\n");
return -ENODEV;
}
}
/*
* Looks for early PCI chipsets where people report hangs
* without the workarounds being on.
*/
/* 1. Intel Saturn. Switch to 8 long words burst, 8 long word cache
aligned. Aries might need this too. The Saturn errata are not
pretty reading but thankfully it's an old 486 chipset.
2. The dreaded SiS496 486 chipset. Same workaround as Intel
Saturn.
*/
if (pci_dev_present(early_486_chipsets)) {
csr0 = MRL | MRM | (8 << BurstLenShift) | (1 << CALShift);
force_csr0 = 1;
}
/* bugfix: the ASIX must have a burst limit or horrible things happen. */
if (chip_idx == AX88140) {
if ((csr0 & 0x3f00) == 0)
csr0 |= 0x2000;
}
/* PNIC doesn't have MWI/MRL/MRM... */
if (chip_idx == LC82C168)
csr0 &= ~0xfff10000; /* zero reserved bits 31:20, 16 */
/* DM9102A has troubles with MRM & clear reserved bits 24:22, 20, 16, 7:1 */
if (tulip_uli_dm_quirk(pdev)) {
csr0 &= ~0x01f100ff;
#if defined(CONFIG_SPARC)
csr0 = (csr0 & ~0xff00) | 0xe000;
#endif
}
/*
* And back to business
*/
i = pci_enable_device(pdev);
if (i) {
printk (KERN_ERR PFX
"Cannot enable tulip board #%d, aborting\n",
board_idx);
return i;
}
irq = pdev->irq;
/* alloc_etherdev ensures aligned and zeroed private structures */
dev = alloc_etherdev (sizeof (*tp));
if (!dev) {
printk (KERN_ERR PFX "ether device alloc failed, aborting\n");
return -ENOMEM;
}
SET_NETDEV_DEV(dev, &pdev->dev);
if (pci_resource_len (pdev, 0) < tulip_tbl[chip_idx].io_size) {
printk (KERN_ERR PFX "%s: I/O region (0x%llx@0x%llx) too small, "
"aborting\n", pci_name(pdev),
(unsigned long long)pci_resource_len (pdev, 0),
(unsigned long long)pci_resource_start (pdev, 0));
goto err_out_free_netdev;
}
/* grab all resources from both PIO and MMIO regions, as we
* don't want anyone else messing around with our hardware */
if (pci_request_regions (pdev, "tulip"))
goto err_out_free_netdev;
ioaddr = pci_iomap(pdev, TULIP_BAR, tulip_tbl[chip_idx].io_size);
if (!ioaddr)
goto err_out_free_res;
/*
* initialize private data structure 'tp'
* it is zeroed and aligned in alloc_etherdev
*/
tp = netdev_priv(dev);
tp->dev = dev;
tp->rx_ring = pci_alloc_consistent(pdev,
sizeof(struct tulip_rx_desc) * RX_RING_SIZE +
sizeof(struct tulip_tx_desc) * TX_RING_SIZE,
&tp->rx_ring_dma);
if (!tp->rx_ring)
goto err_out_mtable;
tp->tx_ring = (struct tulip_tx_desc *)(tp->rx_ring + RX_RING_SIZE);
tp->tx_ring_dma = tp->rx_ring_dma + sizeof(struct tulip_rx_desc) * RX_RING_SIZE;
tp->chip_id = chip_idx;
tp->flags = tulip_tbl[chip_idx].flags;
tp->pdev = pdev;
tp->base_addr = ioaddr;
tp->revision = pdev->revision;
tp->csr0 = csr0;
spin_lock_init(&tp->lock);
spin_lock_init(&tp->mii_lock);
init_timer(&tp->timer);
tp->timer.data = (unsigned long)dev;
tp->timer.function = tulip_tbl[tp->chip_id].media_timer;
INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task);
dev->base_addr = (unsigned long)ioaddr;
#ifdef CONFIG_TULIP_MWI
if (!force_csr0 && (tp->flags & HAS_PCI_MWI))
tulip_mwi_config (pdev, dev);
#endif
/* Stop the chip's Tx and Rx processes. */
tulip_stop_rxtx(tp);
pci_set_master(pdev);
#ifdef CONFIG_GSC
if (pdev->subsystem_vendor == PCI_VENDOR_ID_HP) {
switch (pdev->subsystem_device) {
default:
break;
case 0x1061:
case 0x1062:
case 0x1063:
case 0x1098:
case 0x1099:
case 0x10EE:
tp->flags |= HAS_SWAPPED_SEEPROM | NEEDS_FAKE_MEDIA_TABLE;
chip_name = "GSC DS21140 Tulip";
}
}
#endif
/* Clear the missed-packet counter. */
ioread32(ioaddr + CSR8);
/* The station address ROM is read byte serially. The register must
be polled, waiting for the value to be read bit serially from the
EEPROM.
*/
ee_data = tp->eeprom;
memset(ee_data, 0, sizeof(tp->eeprom));
sum = 0;
if (chip_idx == LC82C168) {
for (i = 0; i < 3; i++) {
int value, boguscnt = 100000;
iowrite32(0x600 | i, ioaddr + 0x98);
do
value = ioread32(ioaddr + CSR9);
while (value < 0 && --boguscnt > 0);
put_unaligned_le16(value, ((__le16 *)dev->dev_addr) + i);
sum += value & 0xffff;
}
} else if (chip_idx == COMET) {
/* No need to read the EEPROM. */
put_unaligned_le32(ioread32(ioaddr + 0xA4), dev->dev_addr);
put_unaligned_le16(ioread32(ioaddr + 0xA8), dev->dev_addr + 4);
for (i = 0; i < 6; i ++)
sum += dev->dev_addr[i];
} else {
/* A serial EEPROM interface, we read now and sort it out later. */
int sa_offset = 0;
int ee_addr_size = tulip_read_eeprom(dev, 0xff, 8) & 0x40000 ? 8 : 6;
int ee_max_addr = ((1 << ee_addr_size) - 1) * sizeof(u16);
if (ee_max_addr > sizeof(tp->eeprom))
ee_max_addr = sizeof(tp->eeprom);
for (i = 0; i < ee_max_addr ; i += sizeof(u16)) {
u16 data = tulip_read_eeprom(dev, i/2, ee_addr_size);
ee_data[i] = data & 0xff;
ee_data[i + 1] = data >> 8;
}
/* DEC now has a specification (see Notes) but early board makers
just put the address in the first EEPROM locations. */
/* This does memcmp(ee_data, ee_data+16, 8) */
for (i = 0; i < 8; i ++)
if (ee_data[i] != ee_data[16+i])
sa_offset = 20;
if (chip_idx == CONEXANT) {
/* Check that the tuple type and length is correct. */
if (ee_data[0x198] == 0x04 && ee_data[0x199] == 6)
sa_offset = 0x19A;
} else if (ee_data[0] == 0xff && ee_data[1] == 0xff &&
ee_data[2] == 0) {
sa_offset = 2; /* Grrr, damn Matrox boards. */
multiport_cnt = 4;
}
#ifdef CONFIG_MIPS_COBALT
if ((pdev->bus->number == 0) &&
((PCI_SLOT(pdev->devfn) == 7) ||
(PCI_SLOT(pdev->devfn) == 12))) {
/* Cobalt MAC address in first EEPROM locations. */
sa_offset = 0;
2005-06-26 21:45:52 +00:00
/* Ensure our media table fixup get's applied */
memcpy(ee_data + 16, ee_data, 8);
}
#endif
#ifdef CONFIG_GSC
/* Check to see if we have a broken srom */
if (ee_data[0] == 0x61 && ee_data[1] == 0x10) {
/* pci_vendor_id and subsystem_id are swapped */
ee_data[0] = ee_data[2];
ee_data[1] = ee_data[3];
ee_data[2] = 0x61;
ee_data[3] = 0x10;
/* HSC-PCI boards need to be byte-swaped and shifted
* up 1 word. This shift needs to happen at the end
* of the MAC first because of the 2 byte overlap.
*/
for (i = 4; i >= 0; i -= 2) {
ee_data[17 + i + 3] = ee_data[17 + i];
ee_data[16 + i + 5] = ee_data[16 + i];
}
}
#endif
for (i = 0; i < 6; i ++) {
dev->dev_addr[i] = ee_data[i + sa_offset];
sum += ee_data[i + sa_offset];
}
}
/* Lite-On boards have the address byte-swapped. */
if ((dev->dev_addr[0] == 0xA0 || dev->dev_addr[0] == 0xC0 || dev->dev_addr[0] == 0x02)
&& dev->dev_addr[1] == 0x00)
for (i = 0; i < 6; i+=2) {
char tmp = dev->dev_addr[i];
dev->dev_addr[i] = dev->dev_addr[i+1];
dev->dev_addr[i+1] = tmp;
}
/* On the Zynx 315 Etherarray and other multiport boards only the
first Tulip has an EEPROM.
On Sparc systems the mac address is held in the OBP property
"local-mac-address".
The addresses of the subsequent ports are derived from the first.
Many PCI BIOSes also incorrectly report the IRQ line, so we correct
that here as well. */
if (sum == 0 || sum == 6*0xff) {
#if defined(CONFIG_SPARC)
struct device_node *dp = pci_device_to_OF_node(pdev);
const unsigned char *addr;
int len;
#endif
eeprom_missing = 1;
for (i = 0; i < 5; i++)
dev->dev_addr[i] = last_phys_addr[i];
dev->dev_addr[i] = last_phys_addr[i] + 1;
#if defined(CONFIG_SPARC)
addr = of_get_property(dp, "local-mac-address", &len);
if (addr && len == 6)
memcpy(dev->dev_addr, addr, 6);
#endif
#if defined(__i386__) || defined(__x86_64__) /* Patch up x86 BIOS bug. */
if (last_irq)
irq = last_irq;
#endif
}
for (i = 0; i < 6; i++)
last_phys_addr[i] = dev->dev_addr[i];
last_irq = irq;
dev->irq = irq;
/* The lower four bits are the media type. */
if (board_idx >= 0 && board_idx < MAX_UNITS) {
if (options[board_idx] & MEDIA_MASK)
tp->default_port = options[board_idx] & MEDIA_MASK;
if ((options[board_idx] & FullDuplex) || full_duplex[board_idx] > 0)
tp->full_duplex = 1;
if (mtu[board_idx] > 0)
dev->mtu = mtu[board_idx];
}
if (dev->mem_start & MEDIA_MASK)
tp->default_port = dev->mem_start & MEDIA_MASK;
if (tp->default_port) {
printk(KERN_INFO "tulip%d: Transceiver selection forced to %s.\n",
board_idx, medianame[tp->default_port & MEDIA_MASK]);
tp->medialock = 1;
if (tulip_media_cap[tp->default_port] & MediaAlwaysFD)
tp->full_duplex = 1;
}
if (tp->full_duplex)
tp->full_duplex_lock = 1;
if (tulip_media_cap[tp->default_port] & MediaIsMII) {
u16 media2advert[] = { 0x20, 0x40, 0x03e0, 0x60, 0x80, 0x100, 0x200 };
tp->mii_advertise = media2advert[tp->default_port - 9];
tp->mii_advertise |= (tp->flags & HAS_8023X); /* Matching bits! */
}
if (tp->flags & HAS_MEDIA_TABLE) {
sprintf(dev->name, "tulip%d", board_idx); /* hack */
tulip_parse_eeprom(dev);
strcpy(dev->name, "eth%d"); /* un-hack */
}
if ((tp->flags & ALWAYS_CHECK_MII) ||
(tp->mtable && tp->mtable->has_mii) ||
( ! tp->mtable && (tp->flags & HAS_MII))) {
if (tp->mtable && tp->mtable->has_mii) {
for (i = 0; i < tp->mtable->leafcount; i++)
if (tp->mtable->mleaf[i].media == 11) {
tp->cur_index = i;
tp->saved_if_port = dev->if_port;
tulip_select_media(dev, 2);
dev->if_port = tp->saved_if_port;
break;
}
}
/* Find the connected MII xcvrs.
Doing this in open() would allow detecting external xcvrs
later, but takes much time. */
tulip_find_mii (dev, board_idx);
}
/* The Tulip-specific entries in the device structure. */
dev->open = tulip_open;
dev->hard_start_xmit = tulip_start_xmit;
dev->tx_timeout = tulip_tx_timeout;
dev->watchdog_timeo = TX_TIMEOUT;
#ifdef CONFIG_TULIP_NAPI
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-03 23:41:36 +00:00
netif_napi_add(dev, &tp->napi, tulip_poll, 16);
#endif
dev->stop = tulip_close;
dev->get_stats = tulip_get_stats;
dev->do_ioctl = private_ioctl;
dev->set_multicast_list = set_rx_mode;
#ifdef CONFIG_NET_POLL_CONTROLLER
dev->poll_controller = &poll_tulip;
#endif
SET_ETHTOOL_OPS(dev, &ops);
if (register_netdev(dev))
goto err_out_free_ring;
printk(KERN_INFO "%s: %s rev %d at "
#ifdef CONFIG_TULIP_MMIO
"MMIO"
#else
"Port"
#endif
" %#llx,", dev->name, chip_name, pdev->revision,
(unsigned long long) pci_resource_start(pdev, TULIP_BAR));
pci_set_drvdata(pdev, dev);
if (eeprom_missing)
printk(" EEPROM not present,");
printk(" %s", print_mac(mac, dev->dev_addr));
printk(", IRQ %d.\n", irq);
if (tp->chip_id == PNIC2)
tp->link_change = pnic2_lnk_change;
else if (tp->flags & HAS_NWAY)
tp->link_change = t21142_lnk_change;
else if (tp->flags & HAS_PNICNWAY)
tp->link_change = pnic_lnk_change;
/* Reset the xcvr interface and turn on heartbeat. */
switch (chip_idx) {
case DC21140:
case DM910X:
default:
if (tp->mtable)
iowrite32(tp->mtable->csr12dir | 0x100, ioaddr + CSR12);
break;
case DC21142:
if (tp->mii_cnt || tulip_media_cap[dev->if_port] & MediaIsMII) {
iowrite32(csr6_mask_defstate, ioaddr + CSR6);
iowrite32(0x0000, ioaddr + CSR13);
iowrite32(0x0000, ioaddr + CSR14);
iowrite32(csr6_mask_hdcap, ioaddr + CSR6);
} else
t21142_start_nway(dev);
break;
case PNIC2:
/* just do a reset for sanity sake */
iowrite32(0x0000, ioaddr + CSR13);
iowrite32(0x0000, ioaddr + CSR14);
break;
case LC82C168:
if ( ! tp->mii_cnt) {
tp->nway = 1;
tp->nwayset = 0;
iowrite32(csr6_ttm | csr6_ca, ioaddr + CSR6);
iowrite32(0x30, ioaddr + CSR12);
iowrite32(0x0001F078, ioaddr + CSR6);
iowrite32(0x0201F078, ioaddr + CSR6); /* Turn on autonegotiation. */
}
break;
case MX98713:
case COMPEX9881:
iowrite32(0x00000000, ioaddr + CSR6);
iowrite32(0x000711C0, ioaddr + CSR14); /* Turn on NWay. */
iowrite32(0x00000001, ioaddr + CSR13);
break;
case MX98715:
case MX98725:
iowrite32(0x01a80000, ioaddr + CSR6);
iowrite32(0xFFFFFFFF, ioaddr + CSR14);
iowrite32(0x00001000, ioaddr + CSR12);
break;
case COMET:
/* No initialization necessary. */
break;
}
/* put the chip in snooze mode until opened */
tulip_set_power_state (tp, 0, 1);
return 0;
err_out_free_ring:
pci_free_consistent (pdev,
sizeof (struct tulip_rx_desc) * RX_RING_SIZE +
sizeof (struct tulip_tx_desc) * TX_RING_SIZE,
tp->rx_ring, tp->rx_ring_dma);
err_out_mtable:
kfree (tp->mtable);
pci_iounmap(pdev, ioaddr);
err_out_free_res:
pci_release_regions (pdev);
err_out_free_netdev:
free_netdev (dev);
return -ENODEV;
}
#ifdef CONFIG_PM
static int tulip_suspend (struct pci_dev *pdev, pm_message_t state)
{
struct net_device *dev = pci_get_drvdata(pdev);
if (!dev)
return -EINVAL;
if (netif_running(dev))
tulip_down(dev);
netif_device_detach(dev);
free_irq(dev->irq, dev);
pci_save_state(pdev);
pci_disable_device(pdev);
pci_set_power_state(pdev, pci_choose_state(pdev, state));
return 0;
}
static int tulip_resume(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
int retval;
if (!dev)
return -EINVAL;
pci_set_power_state(pdev, PCI_D0);
pci_restore_state(pdev);
if ((retval = pci_enable_device(pdev))) {
printk (KERN_ERR "tulip: pci_enable_device failed in resume\n");
return retval;
}
if ((retval = request_irq(dev->irq, &tulip_interrupt, IRQF_SHARED, dev->name, dev))) {
printk (KERN_ERR "tulip: request_irq failed in resume\n");
return retval;
}
netif_device_attach(dev);
if (netif_running(dev))
tulip_up(dev);
return 0;
}
#endif /* CONFIG_PM */
static void __devexit tulip_remove_one (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata (pdev);
struct tulip_private *tp;
if (!dev)
return;
tp = netdev_priv(dev);
unregister_netdev(dev);
pci_free_consistent (pdev,
sizeof (struct tulip_rx_desc) * RX_RING_SIZE +
sizeof (struct tulip_tx_desc) * TX_RING_SIZE,
tp->rx_ring, tp->rx_ring_dma);
kfree (tp->mtable);
pci_iounmap(pdev, tp->base_addr);
free_netdev (dev);
pci_release_regions (pdev);
pci_set_drvdata (pdev, NULL);
/* pci_power_off (pdev, -1); */
}
#ifdef CONFIG_NET_POLL_CONTROLLER
/*
* Polling 'interrupt' - used by things like netconsole to send skbs
* without having to re-enable interrupts. It's not called while
* the interrupt routine is executing.
*/
static void poll_tulip (struct net_device *dev)
{
/* disable_irq here is not very nice, but with the lockless
interrupt handler we have no other choice. */
disable_irq(dev->irq);
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 13:55:46 +00:00
tulip_interrupt (dev->irq, dev);
enable_irq(dev->irq);
}
#endif
static struct pci_driver tulip_driver = {
.name = DRV_NAME,
.id_table = tulip_pci_tbl,
.probe = tulip_init_one,
.remove = __devexit_p(tulip_remove_one),
#ifdef CONFIG_PM
.suspend = tulip_suspend,
.resume = tulip_resume,
#endif /* CONFIG_PM */
};
static int __init tulip_init (void)
{
#ifdef MODULE
printk (KERN_INFO "%s", version);
#endif
/* copy module parms into globals */
tulip_rx_copybreak = rx_copybreak;
tulip_max_interrupt_work = max_interrupt_work;
/* probe for and init boards */
return pci_register_driver(&tulip_driver);
}
static void __exit tulip_cleanup (void)
{
pci_unregister_driver (&tulip_driver);
}
module_init(tulip_init);
module_exit(tulip_cleanup);