AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-05 08:26:49 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
3342cf0e59
linux_dsm_epyc7002/drivers/net/saa9730.c

1147 lines
30 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 05:20:36 +07:00
/*
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
* Copyright (C) 2000, 2005 MIPS Technologies, Inc. All rights reserved.
* Authors: Carsten Langgaard <carstenl@mips.com>
* Maciej W. Rozycki <macro@mips.com>
* Copyright (C) 2004 Ralf Baechle <ralf@linux-mips.org>
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 05:20:36 +07:00
*
* This program is free software; you can distribute it and/or modify it
* under the terms of the GNU General Public License (Version 2) as
* published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
* for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
*
* SAA9730 ethernet driver.
*
* Changes:
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
* Angelo Dell'Aera <buffer@antifork.org> : Conversion to the new PCI API
* (pci_driver).
* Conversion to spinlocks.
* Error handling fixes.
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 05:20:36 +07:00
*/
#include <linux/init.h>
#include <linux/netdevice.h>
#include <linux/delay.h>
#include <linux/etherdevice.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/pci.h>
#include <linux/spinlock.h>
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
#include <linux/types.h>
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 05:20:36 +07:00
#include <asm/addrspace.h>
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
#include <asm/io.h>
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 05:20:36 +07:00
#include <asm/mips-boards/prom.h>
#include "saa9730.h"
#ifdef LAN_SAA9730_DEBUG
int lan_saa9730_debug = LAN_SAA9730_DEBUG;
#else
int lan_saa9730_debug;
#endif
#define DRV_MODULE_NAME "saa9730"
static struct pci_device_id saa9730_pci_tbl[] = {
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
{ PCI_VENDOR_ID_PHILIPS, PCI_DEVICE_ID_PHILIPS_SAA9730,
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
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 05:20:36 +07:00
{ 0, }
};
MODULE_DEVICE_TABLE(pci, saa9730_pci_tbl);
/* Non-zero only if the current card is a PCI with BIOS-set IRQ. */
static unsigned int pci_irq_line;
static void evm_saa9730_enable_lan_int(struct lan_saa9730_private *lp)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptBlock1) | EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptBlock1);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptStatus1) | EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptStatus1);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptEnable1) | EVM_LAN_INT |
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 05:20:36 +07:00
EVM_MASTER_EN, &lp->evm_saa9730_regs->InterruptEnable1);
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
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 05:20:36 +07:00
static void evm_saa9730_disable_lan_int(struct lan_saa9730_private *lp)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptBlock1) & ~EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptBlock1);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptEnable1) & ~EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptEnable1);
}
static void evm_saa9730_clear_lan_int(struct lan_saa9730_private *lp)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(EVM_LAN_INT, &lp->evm_saa9730_regs->InterruptStatus1);
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 05:20:36 +07:00
}
static void evm_saa9730_block_lan_int(struct lan_saa9730_private *lp)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptBlock1) & ~EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptBlock1);
}
static void evm_saa9730_unblock_lan_int(struct lan_saa9730_private *lp)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->evm_saa9730_regs->InterruptBlock1) | EVM_LAN_INT,
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 05:20:36 +07:00
&lp->evm_saa9730_regs->InterruptBlock1);
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
static void __attribute_used__ show_saa9730_regs(struct lan_saa9730_private *lp)
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 05:20:36 +07:00
{
int i, j;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("TxmBufferA = %p\n", lp->TxmBuffer[0][0]);
printk("TxmBufferB = %p\n", lp->TxmBuffer[1][0]);
printk("RcvBufferA = %p\n", lp->RcvBuffer[0][0]);
printk("RcvBufferB = %p\n", lp->RcvBuffer[1][0]);
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 05:20:36 +07:00
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_TXM_Q_SIZE; j++) {
printk("TxmBuffer[%d][%d] = %x\n", i, j,
le32_to_cpu(*(unsigned int *)
lp->TxmBuffer[i][j]));
}
}
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_RCV_Q_SIZE; j++) {
printk("RcvBuffer[%d][%d] = %x\n", i, j,
le32_to_cpu(*(unsigned int *)
lp->RcvBuffer[i][j]));
}
}
printk("lp->evm_saa9730_regs->InterruptBlock1 = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->evm_saa9730_regs->InterruptBlock1));
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 05:20:36 +07:00
printk("lp->evm_saa9730_regs->InterruptStatus1 = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->evm_saa9730_regs->InterruptStatus1));
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 05:20:36 +07:00
printk("lp->evm_saa9730_regs->InterruptEnable1 = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->evm_saa9730_regs->InterruptEnable1));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->Ok2Use = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->Ok2Use));
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 05:20:36 +07:00
printk("lp->NextTxmBufferIndex = %x\n", lp->NextTxmBufferIndex);
printk("lp->NextTxmPacketIndex = %x\n", lp->NextTxmPacketIndex);
printk("lp->PendingTxmBufferIndex = %x\n",
lp->PendingTxmBufferIndex);
printk("lp->PendingTxmPacketIndex = %x\n",
lp->PendingTxmPacketIndex);
printk("lp->lan_saa9730_regs->LanDmaCtl = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->LanDmaCtl));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DmaStatus = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DmaStatus));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->CamCtl = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->CamCtl));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->TxCtl = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->TxCtl));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->TxStatus = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->TxStatus));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->RxCtl = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->RxCtl));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->RxStatus = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->RxStatus));
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 05:20:36 +07:00
for (i = 0; i < LAN_SAA9730_CAM_DWORDS; i++) {
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(i, &lp->lan_saa9730_regs->CamAddress);
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->CamData = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->CamData));
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 05:20:36 +07:00
}
printk("lp->stats.tx_packets = %lx\n", lp->stats.tx_packets);
printk("lp->stats.tx_errors = %lx\n", lp->stats.tx_errors);
printk("lp->stats.tx_aborted_errors = %lx\n",
lp->stats.tx_aborted_errors);
printk("lp->stats.tx_window_errors = %lx\n",
lp->stats.tx_window_errors);
printk("lp->stats.tx_carrier_errors = %lx\n",
lp->stats.tx_carrier_errors);
printk("lp->stats.tx_fifo_errors = %lx\n",
lp->stats.tx_fifo_errors);
printk("lp->stats.tx_heartbeat_errors = %lx\n",
lp->stats.tx_heartbeat_errors);
printk("lp->stats.collisions = %lx\n", lp->stats.collisions);
printk("lp->stats.rx_packets = %lx\n", lp->stats.rx_packets);
printk("lp->stats.rx_errors = %lx\n", lp->stats.rx_errors);
printk("lp->stats.rx_dropped = %lx\n", lp->stats.rx_dropped);
printk("lp->stats.rx_crc_errors = %lx\n", lp->stats.rx_crc_errors);
printk("lp->stats.rx_frame_errors = %lx\n",
lp->stats.rx_frame_errors);
printk("lp->stats.rx_fifo_errors = %lx\n",
lp->stats.rx_fifo_errors);
printk("lp->stats.rx_length_errors = %lx\n",
lp->stats.rx_length_errors);
printk("lp->lan_saa9730_regs->DebugPCIMasterAddr = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugPCIMasterAddr));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DebugLanTxStateMachine = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugLanTxStateMachine));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DebugLanRxStateMachine = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugLanRxStateMachine));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DebugLanTxFifoPointers = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugLanTxFifoPointers));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DebugLanRxFifoPointers = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugLanRxFifoPointers));
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 05:20:36 +07:00
printk("lp->lan_saa9730_regs->DebugLanCtlStateMachine = %x\n",
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
readl(&lp->lan_saa9730_regs->DebugLanCtlStateMachine));
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 05:20:36 +07:00
}
static void lan_saa9730_buffer_init(struct lan_saa9730_private *lp)
{
int i, j;
/* Init RX buffers */
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_RCV_Q_SIZE; j++) {
*(unsigned int *) lp->RcvBuffer[i][j] =
cpu_to_le32(RXSF_READY <<
RX_STAT_CTL_OWNER_SHF);
}
}
/* Init TX buffers */
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_TXM_Q_SIZE; j++) {
*(unsigned int *) lp->TxmBuffer[i][j] =
cpu_to_le32(TXSF_EMPTY <<
TX_STAT_CTL_OWNER_SHF);
}
}
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
static void lan_saa9730_free_buffers(struct pci_dev *pdev,
struct lan_saa9730_private *lp)
{
pci_free_consistent(pdev, lp->buffer_size, lp->buffer_start,
lp->dma_addr);
}
static int lan_saa9730_allocate_buffers(struct pci_dev *pdev,
struct lan_saa9730_private *lp)
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 05:20:36 +07:00
{
void *Pa;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
unsigned int i, j, rxoffset, txoffset;
int ret;
/* Initialize buffer space */
lp->DmaRcvPackets = LAN_SAA9730_RCV_Q_SIZE;
lp->DmaTxmPackets = LAN_SAA9730_TXM_Q_SIZE;
/* Initialize Rx Buffer Index */
lp->NextRcvPacketIndex = 0;
lp->NextRcvBufferIndex = 0;
/* Set current buffer index & next available packet index */
lp->NextTxmPacketIndex = 0;
lp->NextTxmBufferIndex = 0;
lp->PendingTxmPacketIndex = 0;
lp->PendingTxmBufferIndex = 0;
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 05:20:36 +07:00
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
* Allocate all RX and TX packets in one chunk.
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 05:20:36 +07:00
* The Rx and Tx packets must be PACKET_SIZE aligned.
*/
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->buffer_size = ((LAN_SAA9730_RCV_Q_SIZE + LAN_SAA9730_TXM_Q_SIZE) *
LAN_SAA9730_PACKET_SIZE * LAN_SAA9730_BUFFERS) +
LAN_SAA9730_PACKET_SIZE;
lp->buffer_start = pci_alloc_consistent(pdev, lp->buffer_size,
&lp->dma_addr);
if (!lp->buffer_start) {
ret = -ENOMEM;
goto out;
}
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 05:20:36 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
Pa = (void *)ALIGN((unsigned long)lp->buffer_start,
LAN_SAA9730_PACKET_SIZE);
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 05:20:36 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
rxoffset = Pa - lp->buffer_start;
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 05:20:36 +07:00
/* Init RX buffers */
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_RCV_Q_SIZE; j++) {
*(unsigned int *) Pa =
cpu_to_le32(RXSF_READY <<
RX_STAT_CTL_OWNER_SHF);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->RcvBuffer[i][j] = Pa;
Pa += LAN_SAA9730_PACKET_SIZE;
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 05:20:36 +07:00
}
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
txoffset = Pa - lp->buffer_start;
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 05:20:36 +07:00
/* Init TX buffers */
for (i = 0; i < LAN_SAA9730_BUFFERS; i++) {
for (j = 0; j < LAN_SAA9730_TXM_Q_SIZE; j++) {
*(unsigned int *) Pa =
cpu_to_le32(TXSF_EMPTY <<
TX_STAT_CTL_OWNER_SHF);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->TxmBuffer[i][j] = Pa;
Pa += LAN_SAA9730_PACKET_SIZE;
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 05:20:36 +07:00
}
}
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
* Set rx buffer A and rx buffer B to point to the first two buffer
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 05:20:36 +07:00
* spaces.
*/
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(lp->dma_addr + rxoffset,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->RxBuffA);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(lp->dma_addr + rxoffset +
LAN_SAA9730_PACKET_SIZE * LAN_SAA9730_RCV_Q_SIZE,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->RxBuffB);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
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 05:20:36 +07:00
* Set txm_buf_a and txm_buf_b to point to the first two buffer
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
* space
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 05:20:36 +07:00
*/
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(lp->dma_addr + txoffset,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->TxBuffA);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(lp->dma_addr + txoffset +
LAN_SAA9730_PACKET_SIZE * LAN_SAA9730_TXM_Q_SIZE,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->TxBuffB);
/* Set packet number */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl((lp->DmaRcvPackets << PK_COUNT_RX_A_SHF) |
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 05:20:36 +07:00
(lp->DmaRcvPackets << PK_COUNT_RX_B_SHF) |
(lp->DmaTxmPackets << PK_COUNT_TX_A_SHF) |
(lp->DmaTxmPackets << PK_COUNT_TX_B_SHF),
&lp->lan_saa9730_regs->PacketCount);
return 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
out:
return ret;
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 05:20:36 +07:00
}
static int lan_saa9730_cam_load(struct lan_saa9730_private *lp)
{
unsigned int i;
unsigned char *NetworkAddress;
NetworkAddress = (unsigned char *) &lp->PhysicalAddress[0][0];
for (i = 0; i < LAN_SAA9730_CAM_DWORDS; i++) {
/* First set address to where data is written */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(i, &lp->lan_saa9730_regs->CamAddress);
outl((NetworkAddress[0] << 24) | (NetworkAddress[1] << 16)
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 05:20:36 +07:00
| (NetworkAddress[2] << 8) | NetworkAddress[3],
&lp->lan_saa9730_regs->CamData);
NetworkAddress += 4;
}
return 0;
}
static int lan_saa9730_cam_init(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
unsigned int i;
/* Copy MAC-address into all entries. */
for (i = 0; i < LAN_SAA9730_CAM_ENTRIES; i++) {
memcpy((unsigned char *) lp->PhysicalAddress[i],
(unsigned char *) dev->dev_addr, 6);
}
return 0;
}
static int lan_saa9730_mii_init(struct lan_saa9730_private *lp)
{
int i, l;
/* Check link status, spin here till station is not busy. */
i = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
while (readl(&lp->lan_saa9730_regs->StationMgmtCtl) & MD_CA_BUSY) {
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 05:20:36 +07:00
i++;
if (i > 100) {
printk("Error: lan_saa9730_mii_init: timeout\n");
return -1;
}
mdelay(1); /* wait 1 ms. */
}
/* Now set the control and address register. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(MD_CA_BUSY | PHY_STATUS | PHY_ADDRESS << MD_CA_PHY_SHF,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->StationMgmtCtl);
/* check link status, spin here till station is not busy */
i = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
while (readl(&lp->lan_saa9730_regs->StationMgmtCtl) & MD_CA_BUSY) {
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 05:20:36 +07:00
i++;
if (i > 100) {
printk("Error: lan_saa9730_mii_init: timeout\n");
return -1;
}
mdelay(1); /* wait 1 ms. */
}
/* Wait for 1 ms. */
mdelay(1);
/* Check the link status. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if (readl(&lp->lan_saa9730_regs->StationMgmtData) &
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 05:20:36 +07:00
PHY_STATUS_LINK_UP) {
/* Link is up. */
return 0;
} else {
/* Link is down, reset the PHY first. */
/* set PHY address = 'CONTROL' */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(PHY_ADDRESS << MD_CA_PHY_SHF | MD_CA_WR | PHY_CONTROL,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->StationMgmtCtl);
/* Wait for 1 ms. */
mdelay(1);
/* set 'CONTROL' = force reset and renegotiate */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(PHY_CONTROL_RESET | PHY_CONTROL_AUTO_NEG |
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 05:20:36 +07:00
PHY_CONTROL_RESTART_AUTO_NEG,
&lp->lan_saa9730_regs->StationMgmtData);
/* Wait for 50 ms. */
mdelay(50);
/* set 'BUSY' to start operation */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(MD_CA_BUSY | PHY_ADDRESS << MD_CA_PHY_SHF | MD_CA_WR |
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 05:20:36 +07:00
PHY_CONTROL, &lp->lan_saa9730_regs->StationMgmtCtl);
/* await completion */
i = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
while (readl(&lp->lan_saa9730_regs->StationMgmtCtl) &
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 05:20:36 +07:00
MD_CA_BUSY) {
i++;
if (i > 100) {
printk
("Error: lan_saa9730_mii_init: timeout\n");
return -1;
}
mdelay(1); /* wait 1 ms. */
}
/* Wait for 1 ms. */
mdelay(1);
for (l = 0; l < 2; l++) {
/* set PHY address = 'STATUS' */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(MD_CA_BUSY | PHY_ADDRESS << MD_CA_PHY_SHF |
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 05:20:36 +07:00
PHY_STATUS,
&lp->lan_saa9730_regs->StationMgmtCtl);
/* await completion */
i = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
while (readl(&lp->lan_saa9730_regs->StationMgmtCtl) &
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 05:20:36 +07:00
MD_CA_BUSY) {
i++;
if (i > 100) {
printk
("Error: lan_saa9730_mii_init: timeout\n");
return -1;
}
mdelay(1); /* wait 1 ms. */
}
/* wait for 3 sec. */
mdelay(3000);
/* check the link status */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if (readl(&lp->lan_saa9730_regs->StationMgmtData) &
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 05:20:36 +07:00
PHY_STATUS_LINK_UP) {
/* link is up */
break;
}
}
}
return 0;
}
static int lan_saa9730_control_init(struct lan_saa9730_private *lp)
{
/* Initialize DMA control register. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl((LANMB_ANY << DMA_CTL_MAX_XFER_SHF) |
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 05:20:36 +07:00
(LANEND_LITTLE << DMA_CTL_ENDIAN_SHF) |
(LAN_SAA9730_RCV_Q_INT_THRESHOLD << DMA_CTL_RX_INT_COUNT_SHF)
| DMA_CTL_RX_INT_TO_EN | DMA_CTL_RX_INT_EN |
DMA_CTL_MAC_RX_INT_EN | DMA_CTL_MAC_TX_INT_EN,
&lp->lan_saa9730_regs->LanDmaCtl);
/* Initial MAC control register. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl((MACCM_MII << MAC_CONTROL_CONN_SHF) | MAC_CONTROL_FULL_DUP,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->MacCtl);
/* Initialize CAM control register. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(CAM_CONTROL_COMP_EN | CAM_CONTROL_BROAD_ACC,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->CamCtl);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
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 05:20:36 +07:00
* Initialize CAM enable register, only turn on first entry, should
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
* contain own addr.
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 05:20:36 +07:00
*/
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(0x0001, &lp->lan_saa9730_regs->CamEnable);
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 05:20:36 +07:00
/* Initialize Tx control register */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(TX_CTL_EN_COMP, &lp->lan_saa9730_regs->TxCtl);
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 05:20:36 +07:00
/* Initialize Rcv control register */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(RX_CTL_STRIP_CRC, &lp->lan_saa9730_regs->RxCtl);
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 05:20:36 +07:00
/* Reset DMA engine */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(DMA_TEST_SW_RESET, &lp->lan_saa9730_regs->DmaTest);
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 05:20:36 +07:00
return 0;
}
static int lan_saa9730_stop(struct lan_saa9730_private *lp)
{
int i;
/* Stop DMA first */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->lan_saa9730_regs->LanDmaCtl) &
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 05:20:36 +07:00
~(DMA_CTL_EN_TX_DMA | DMA_CTL_EN_RX_DMA),
&lp->lan_saa9730_regs->LanDmaCtl);
/* Set the SW Reset bits in DMA and MAC control registers */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(DMA_TEST_SW_RESET, &lp->lan_saa9730_regs->DmaTest);
outl(readl(&lp->lan_saa9730_regs->MacCtl) | MAC_CONTROL_RESET,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->MacCtl);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
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 05:20:36 +07:00
* Wait for MAC reset to have finished. The reset bit is auto cleared
* when the reset is done.
*/
i = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
while (readl(&lp->lan_saa9730_regs->MacCtl) & MAC_CONTROL_RESET) {
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 05:20:36 +07:00
i++;
if (i > 100) {
printk
("Error: lan_sa9730_stop: MAC reset timeout\n");
return -1;
}
mdelay(1); /* wait 1 ms. */
}
return 0;
}
static int lan_saa9730_dma_init(struct lan_saa9730_private *lp)
{
/* Stop lan controller. */
lan_saa9730_stop(lp);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(LAN_SAA9730_DEFAULT_TIME_OUT_CNT,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->Timeout);
return 0;
}
static int lan_saa9730_start(struct lan_saa9730_private *lp)
{
lan_saa9730_buffer_init(lp);
/* Initialize Rx Buffer Index */
lp->NextRcvPacketIndex = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->NextRcvBufferIndex = 0;
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 05:20:36 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
/* Set current buffer index & next available packet index */
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 05:20:36 +07:00
lp->NextTxmPacketIndex = 0;
lp->NextTxmBufferIndex = 0;
lp->PendingTxmPacketIndex = 0;
lp->PendingTxmBufferIndex = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->lan_saa9730_regs->LanDmaCtl) | DMA_CTL_EN_TX_DMA |
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 05:20:36 +07:00
DMA_CTL_EN_RX_DMA, &lp->lan_saa9730_regs->LanDmaCtl);
/* For Tx, turn on MAC then DMA */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->lan_saa9730_regs->TxCtl) | TX_CTL_TX_EN,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->TxCtl);
/* For Rx, turn on DMA then MAC */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(readl(&lp->lan_saa9730_regs->RxCtl) | RX_CTL_RX_EN,
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 05:20:36 +07:00
&lp->lan_saa9730_regs->RxCtl);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
/* Set Ok2Use to let hardware own the buffers. */
outl(OK2USE_RX_A | OK2USE_RX_B, &lp->lan_saa9730_regs->Ok2Use);
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 05:20:36 +07:00
return 0;
}
static int lan_saa9730_restart(struct lan_saa9730_private *lp)
{
lan_saa9730_stop(lp);
lan_saa9730_start(lp);
return 0;
}
static int lan_saa9730_tx(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
unsigned int *pPacket;
unsigned int tx_status;
if (lan_saa9730_debug > 5)
printk("lan_saa9730_tx interrupt\n");
/* Clear interrupt. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(DMA_STATUS_MAC_TX_INT, &lp->lan_saa9730_regs->DmaStatus);
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 05:20:36 +07:00
while (1) {
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
pPacket = lp->TxmBuffer[lp->PendingTxmBufferIndex]
[lp->PendingTxmPacketIndex];
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 05:20:36 +07:00
/* Get status of first packet transmitted. */
tx_status = le32_to_cpu(*pPacket);
/* Check ownership. */
if ((tx_status & TX_STAT_CTL_OWNER_MSK) !=
(TXSF_HWDONE << TX_STAT_CTL_OWNER_SHF)) break;
/* Check for error. */
if (tx_status & TX_STAT_CTL_ERROR_MSK) {
if (lan_saa9730_debug > 1)
printk("lan_saa9730_tx: tx error = %x\n",
tx_status);
lp->stats.tx_errors++;
if (tx_status &
(TX_STATUS_EX_COLL << TX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.tx_aborted_errors++;
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 05:20:36 +07:00
if (tx_status &
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
(TX_STATUS_LATE_COLL << TX_STAT_CTL_STATUS_SHF))
lp->stats.tx_window_errors++;
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 05:20:36 +07:00
if (tx_status &
(TX_STATUS_L_CARR << TX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.tx_carrier_errors++;
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 05:20:36 +07:00
if (tx_status &
(TX_STATUS_UNDER << TX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.tx_fifo_errors++;
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 05:20:36 +07:00
if (tx_status &
(TX_STATUS_SQ_ERR << TX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.tx_heartbeat_errors++;
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 05:20:36 +07:00
lp->stats.collisions +=
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
tx_status & TX_STATUS_TX_COLL_MSK;
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 05:20:36 +07:00
}
/* Free buffer. */
*pPacket =
cpu_to_le32(TXSF_EMPTY << TX_STAT_CTL_OWNER_SHF);
/* Update pending index pointer. */
lp->PendingTxmPacketIndex++;
if (lp->PendingTxmPacketIndex >= LAN_SAA9730_TXM_Q_SIZE) {
lp->PendingTxmPacketIndex = 0;
lp->PendingTxmBufferIndex ^= 1;
}
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
/* The tx buffer is no longer full. */
netif_wake_queue(dev);
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 05:20:36 +07:00
return 0;
}
static int lan_saa9730_rx(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
int len = 0;
struct sk_buff *skb = 0;
unsigned int rx_status;
int BufferIndex;
int PacketIndex;
unsigned int *pPacket;
unsigned char *pData;
if (lan_saa9730_debug > 5)
printk("lan_saa9730_rx interrupt\n");
/* Clear receive interrupts. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(DMA_STATUS_MAC_RX_INT | DMA_STATUS_RX_INT |
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 05:20:36 +07:00
DMA_STATUS_RX_TO_INT, &lp->lan_saa9730_regs->DmaStatus);
/* Address next packet */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
BufferIndex = lp->NextRcvBufferIndex;
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 05:20:36 +07:00
PacketIndex = lp->NextRcvPacketIndex;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
pPacket = lp->RcvBuffer[BufferIndex][PacketIndex];
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 05:20:36 +07:00
rx_status = le32_to_cpu(*pPacket);
/* Process each packet. */
while ((rx_status & RX_STAT_CTL_OWNER_MSK) ==
(RXSF_HWDONE << RX_STAT_CTL_OWNER_SHF)) {
/* Check the rx status. */
if (rx_status & (RX_STATUS_GOOD << RX_STAT_CTL_STATUS_SHF)) {
/* Received packet is good. */
len = (rx_status & RX_STAT_CTL_LENGTH_MSK) >>
RX_STAT_CTL_LENGTH_SHF;
pData = (unsigned char *) pPacket;
pData += 4;
skb = dev_alloc_skb(len + 2);
if (skb == 0) {
printk
("%s: Memory squeeze, deferring packet.\n",
dev->name);
lp->stats.rx_dropped++;
} else {
lp->stats.rx_bytes += len;
lp->stats.rx_packets++;
skb->dev = dev;
skb_reserve(skb, 2); /* 16 byte align */
skb_put(skb, len); /* make room */
eth_copy_and_sum(skb,
(unsigned char *) pData,
len, 0);
skb->protocol = eth_type_trans(skb, dev);
netif_rx(skb);
dev->last_rx = jiffies;
}
} else {
/* We got an error packet. */
if (lan_saa9730_debug > 2)
printk
("lan_saa9730_rx: We got an error packet = %x\n",
rx_status);
lp->stats.rx_errors++;
if (rx_status &
(RX_STATUS_CRC_ERR << RX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.rx_crc_errors++;
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 05:20:36 +07:00
if (rx_status &
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
(RX_STATUS_ALIGN_ERR << RX_STAT_CTL_STATUS_SHF))
lp->stats.rx_frame_errors++;
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 05:20:36 +07:00
if (rx_status &
(RX_STATUS_OVERFLOW << RX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.rx_fifo_errors++;
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 05:20:36 +07:00
if (rx_status &
(RX_STATUS_LONG_ERR << RX_STAT_CTL_STATUS_SHF))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->stats.rx_length_errors++;
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 05:20:36 +07:00
}
/* Indicate we have processed the buffer. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
*pPacket = cpu_to_le32(RXSF_READY << RX_STAT_CTL_OWNER_SHF);
/* Make sure A or B is available to hardware as appropriate. */
outl(BufferIndex ? OK2USE_RX_B : OK2USE_RX_A,
&lp->lan_saa9730_regs->Ok2Use);
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 05:20:36 +07:00
/* Go to next packet in sequence. */
lp->NextRcvPacketIndex++;
if (lp->NextRcvPacketIndex >= LAN_SAA9730_RCV_Q_SIZE) {
lp->NextRcvPacketIndex = 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->NextRcvBufferIndex ^= 1;
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 05:20:36 +07:00
}
/* Address next packet */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
BufferIndex = lp->NextRcvBufferIndex;
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 05:20:36 +07:00
PacketIndex = lp->NextRcvPacketIndex;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
pPacket = lp->RcvBuffer[BufferIndex][PacketIndex];
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 05:20:36 +07:00
rx_status = le32_to_cpu(*pPacket);
}
return 0;
}
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 20:55:46 +07:00
static irqreturn_t lan_saa9730_interrupt(const int irq, void *dev_id)
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 05:20:36 +07:00
{
drivers/net: eliminate irq handler impossible checks, needless casts - Eliminate check for irq handler 'dev_id==NULL' where the condition never occurs. - Eliminate needless casts to/from void* Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-10-07 01:56:04 +07:00
struct net_device *dev = dev_id;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
if (lan_saa9730_debug > 5)
printk("lan_saa9730_interrupt\n");
/* Disable the EVM LAN interrupt. */
evm_saa9730_block_lan_int(lp);
/* Clear the EVM LAN interrupt. */
evm_saa9730_clear_lan_int(lp);
/* Service pending transmit interrupts. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if (readl(&lp->lan_saa9730_regs->DmaStatus) & DMA_STATUS_MAC_TX_INT)
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 05:20:36 +07:00
lan_saa9730_tx(dev);
/* Service pending receive interrupts. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if (readl(&lp->lan_saa9730_regs->DmaStatus) &
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 05:20:36 +07:00
(DMA_STATUS_MAC_RX_INT | DMA_STATUS_RX_INT |
DMA_STATUS_RX_TO_INT)) lan_saa9730_rx(dev);
/* Enable the EVM LAN interrupt. */
evm_saa9730_unblock_lan_int(lp);
return IRQ_HANDLED;
}
static int lan_saa9730_open(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
/* Associate IRQ with lan_saa9730_interrupt */
if (request_irq(dev->irq, &lan_saa9730_interrupt, 0, "SAA9730 Eth",
dev)) {
printk("lan_saa9730_open: Can't get irq %d\n", dev->irq);
return -EAGAIN;
}
/* Enable the Lan interrupt in the event manager. */
evm_saa9730_enable_lan_int(lp);
/* Start the LAN controller */
if (lan_saa9730_start(lp))
return -1;
netif_start_queue(dev);
return 0;
}
static int lan_saa9730_write(struct lan_saa9730_private *lp,
struct sk_buff *skb, int skblen)
{
unsigned char *pbData = skb->data;
unsigned int len = skblen;
unsigned char *pbPacketData;
unsigned int tx_status;
int BufferIndex;
int PacketIndex;
if (lan_saa9730_debug > 5)
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("lan_saa9730_write: skb=%p\n", skb);
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 05:20:36 +07:00
BufferIndex = lp->NextTxmBufferIndex;
PacketIndex = lp->NextTxmPacketIndex;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
tx_status = le32_to_cpu(*(unsigned int *)lp->TxmBuffer[BufferIndex]
[PacketIndex]);
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 05:20:36 +07:00
if ((tx_status & TX_STAT_CTL_OWNER_MSK) !=
(TXSF_EMPTY << TX_STAT_CTL_OWNER_SHF)) {
if (lan_saa9730_debug > 4)
printk
("lan_saa9730_write: Tx buffer not available: tx_status = %x\n",
tx_status);
return -1;
}
lp->NextTxmPacketIndex++;
if (lp->NextTxmPacketIndex >= LAN_SAA9730_TXM_Q_SIZE) {
lp->NextTxmPacketIndex = 0;
lp->NextTxmBufferIndex ^= 1;
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
pbPacketData = lp->TxmBuffer[BufferIndex][PacketIndex];
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 05:20:36 +07:00
pbPacketData += 4;
/* copy the bits */
memcpy(pbPacketData, pbData, len);
/* Set transmit status for hardware */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
*(unsigned int *)lp->TxmBuffer[BufferIndex][PacketIndex] =
cpu_to_le32((TXSF_READY << TX_STAT_CTL_OWNER_SHF) |
(TX_STAT_CTL_INT_AFTER_TX <<
TX_STAT_CTL_FRAME_SHF) |
(len << TX_STAT_CTL_LENGTH_SHF));
/* Make sure A or B is available to hardware as appropriate. */
outl(BufferIndex ? OK2USE_TX_B : OK2USE_TX_A,
&lp->lan_saa9730_regs->Ok2Use);
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 05:20:36 +07:00
return 0;
}
static void lan_saa9730_tx_timeout(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
/* Transmitter timeout, serious problems */
lp->stats.tx_errors++;
printk("%s: transmit timed out, reset\n", dev->name);
/*show_saa9730_regs(lp); */
lan_saa9730_restart(lp);
dev->trans_start = jiffies;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
netif_wake_queue(dev);
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 05:20:36 +07:00
}
static int lan_saa9730_start_xmit(struct sk_buff *skb,
struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
unsigned long flags;
int skblen;
int len;
if (lan_saa9730_debug > 4)
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("Send packet: skb=%p\n", skb);
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 05:20:36 +07:00
skblen = skb->len;
spin_lock_irqsave(&lp->lock, flags);
len = (skblen <= ETH_ZLEN) ? ETH_ZLEN : skblen;
if (lan_saa9730_write(lp, skb, skblen)) {
spin_unlock_irqrestore(&lp->lock, flags);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("Error when writing packet to controller: skb=%p\n", skb);
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 05:20:36 +07:00
netif_stop_queue(dev);
return -1;
}
lp->stats.tx_bytes += len;
lp->stats.tx_packets++;
dev->trans_start = jiffies;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
netif_wake_queue(dev);
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 05:20:36 +07:00
dev_kfree_skb(skb);
spin_unlock_irqrestore(&lp->lock, flags);
return 0;
}
static int lan_saa9730_close(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
if (lan_saa9730_debug > 1)
printk("lan_saa9730_close:\n");
netif_stop_queue(dev);
/* Disable the Lan interrupt in the event manager. */
evm_saa9730_disable_lan_int(lp);
/* Stop the controller */
if (lan_saa9730_stop(lp))
return -1;
free_irq(dev->irq, (void *) dev);
return 0;
}
static struct net_device_stats *lan_saa9730_get_stats(struct net_device
*dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
return &lp->stats;
}
static void lan_saa9730_set_multicast(struct net_device *dev)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
/* Stop the controller */
lan_saa9730_stop(lp);
if (dev->flags & IFF_PROMISC) {
/* accept all packets */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(CAM_CONTROL_COMP_EN | CAM_CONTROL_STATION_ACC |
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 05:20:36 +07:00
CAM_CONTROL_GROUP_ACC | CAM_CONTROL_BROAD_ACC,
&lp->lan_saa9730_regs->CamCtl);
} else {
if (dev->flags & IFF_ALLMULTI) {
/* accept all multicast packets */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
outl(CAM_CONTROL_COMP_EN | CAM_CONTROL_GROUP_ACC |
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 05:20:36 +07:00
CAM_CONTROL_BROAD_ACC,
&lp->lan_saa9730_regs->CamCtl);
} else {
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
/*
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 05:20:36 +07:00
* Will handle the multicast stuff later. -carstenl
*/
}
}
lan_saa9730_restart(lp);
}
static void __devexit saa9730_remove_one(struct pci_dev *pdev)
{
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
struct net_device *dev = pci_get_drvdata(pdev);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
if (dev) {
unregister_netdev(dev);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lan_saa9730_free_buffers(pdev, lp);
iounmap(lp->lan_saa9730_regs);
iounmap(lp->evm_saa9730_regs);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
free_netdev(dev);
pci_release_regions(pdev);
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
}
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 05:20:36 +07:00
}
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
static int lan_saa9730_init(struct net_device *dev, struct pci_dev *pdev,
unsigned long ioaddr, int irq)
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 05:20:36 +07:00
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct lan_saa9730_private *lp = netdev_priv(dev);
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 05:20:36 +07:00
unsigned char ethernet_addr[6];
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
int ret;
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 05:20:36 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if (get_ethernet_addr(ethernet_addr)) {
ret = -ENODEV;
goto out;
}
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
memcpy(dev->dev_addr, ethernet_addr, 6);
dev->base_addr = ioaddr;
dev->irq = irq;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->pci_dev = pdev;
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 05:20:36 +07:00
/* Set SAA9730 LAN base address. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->lan_saa9730_regs = ioremap(ioaddr + SAA9730_LAN_REGS_ADDR,
SAA9730_LAN_REGS_SIZE);
if (!lp->lan_saa9730_regs) {
ret = -ENOMEM;
goto out;
}
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 05:20:36 +07:00
/* Set SAA9730 EVM base address. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
lp->evm_saa9730_regs = ioremap(ioaddr + SAA9730_EVM_REGS_ADDR,
SAA9730_EVM_REGS_SIZE);
if (!lp->evm_saa9730_regs) {
ret = -ENOMEM;
goto out_iounmap_lan;
}
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 05:20:36 +07:00
/* Allocate LAN RX/TX frame buffer space. */
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
if ((ret = lan_saa9730_allocate_buffers(pdev, lp)))
goto out_iounmap;
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 05:20:36 +07:00
/* Stop LAN controller. */
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
if ((ret = lan_saa9730_stop(lp)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
/* Initialize CAM registers. */
if ((ret = lan_saa9730_cam_init(dev)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
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 05:20:36 +07:00
/* Initialize MII registers. */
if ((ret = lan_saa9730_mii_init(lp)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
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 05:20:36 +07:00
/* Initialize control registers. */
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
if ((ret = lan_saa9730_control_init(lp)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
/* Load CAM registers. */
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
if ((ret = lan_saa9730_cam_load(lp)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
/* Initialize DMA context registers. */
if ((ret = lan_saa9730_dma_init(lp)))
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
spin_lock_init(&lp->lock);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
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 05:20:36 +07:00
dev->open = lan_saa9730_open;
dev->hard_start_xmit = lan_saa9730_start_xmit;
dev->stop = lan_saa9730_close;
dev->get_stats = lan_saa9730_get_stats;
dev->set_multicast_list = lan_saa9730_set_multicast;
dev->tx_timeout = lan_saa9730_tx_timeout;
dev->watchdog_timeo = (HZ >> 1);
dev->dma = 0;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
ret = register_netdev (dev);
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 05:20:36 +07:00
if (ret)
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_free_consistent;
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 05:20:36 +07:00
return 0;
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
out_free_consistent:
lan_saa9730_free_buffers(pdev, lp);
out_iounmap:
iounmap(lp->evm_saa9730_regs);
out_iounmap_lan:
iounmap(lp->lan_saa9730_regs);
out:
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 05:20:36 +07:00
return ret;
}
static int __devinit saa9730_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
{
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
struct net_device *dev = NULL;
unsigned long pci_ioaddr;
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 05:20:36 +07:00
int err;
if (lan_saa9730_debug > 1)
printk("saa9730.c: PCI bios is present, checking for devices...\n");
err = pci_enable_device(pdev);
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
if (err) {
printk(KERN_ERR "Cannot enable PCI device, aborting.\n");
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
}
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 05:20:36 +07:00
err = pci_request_regions(pdev, DRV_MODULE_NAME);
if (err) {
printk(KERN_ERR "Cannot obtain PCI resources, aborting.\n");
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
goto out_disable_pdev;
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 05:20:36 +07:00
}
pci_irq_line = pdev->irq;
/* LAN base address in located at BAR 1. */
pci_ioaddr = pci_resource_start(pdev, 1);
pci_set_master(pdev);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("Found SAA9730 (PCI) at %lx, irq %d.\n",
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 05:20:36 +07:00
pci_ioaddr, pci_irq_line);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
dev = alloc_etherdev(sizeof(struct lan_saa9730_private));
if (!dev)
goto out_disable_pdev;
err = lan_saa9730_init(dev, pdev, pci_ioaddr, pci_irq_line);
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 05:20:36 +07:00
if (err) {
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
printk("LAN init failed");
goto out_free_netdev;
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 05:20:36 +07:00
}
pci_set_drvdata(pdev, dev);
SET_NETDEV_DEV(dev, &pdev->dev);
return 0;
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
out_free_netdev:
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 05:20:36 +07:00
free_netdev(dev);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
out_disable_pdev:
pci_disable_device(pdev);
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 05:20:36 +07:00
out:
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
pci_set_drvdata(pdev, NULL);
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 05:20:36 +07:00
return err;
}
static struct pci_driver saa9730_driver = {
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
.name = DRV_MODULE_NAME,
.id_table = saa9730_pci_tbl,
.probe = saa9730_init_one,
.remove = __devexit_p(saa9730_remove_one),
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 05:20:36 +07:00
};
static int __init saa9730_init(void)
{
drivers/net: Remove deprecated use of pci_module_init() From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-08-20 04:48:59 +07:00
return pci_register_driver(&saa9730_driver);
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 05:20:36 +07:00
}
static void __exit saa9730_cleanup(void)
{
[PATCH] SAA9730: Whitespace cleanup. Kill trailing whitespace, replace leading whitespace with tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 98 +++++++++++++++++++++++++------------------------- 1 files changed, 49 insertions(+), 49 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-09 23:44:02 +07:00
pci_unregister_driver(&saa9730_driver);
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 05:20:36 +07:00
}
module_init(saa9730_init);
module_exit(saa9730_cleanup);
[PATCH] SAA9730: Driver overhaul o Try to work around some of the undocumented "features" of the SAA9730 o Use netdev_priv() instead of the previous broken mechanism to allocate the private data structure. o Try to make sure we don't leak resources on exit. o No more need to call SET_MODULE_OWNER in 2.6. o Use pci_free_consistent instead of homegrown architecture-specific allocation. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> drivers/net/saa9730.c | 531 +++++++++++++++++++++++--------------------------- 1 files changed, 249 insertions(+), 282 deletions(-) Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
2005-11-10 00:10:05 +07:00
MODULE_AUTHOR("Ralf Baechle <ralf@linux-mips.org>");
MODULE_DESCRIPTION("Philips SAA9730 ethernet driver");
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 05:20:36 +07:00
MODULE_LICENSE("GPL");
Reference in New Issue Copy Permalink