linux/drivers/net/tokenring/lanstreamer.c

1910 lines
63 KiB
C
Raw Normal View History

/*
* lanstreamer.c -- driver for the IBM Auto LANStreamer PCI Adapter
*
* Written By: Mike Sullivan, IBM Corporation
*
* Copyright (C) 1999 IBM Corporation
*
* Linux driver for IBM PCI tokenring cards based on the LanStreamer MPC
* chipset.
*
* This driver is based on the olympic driver for IBM PCI TokenRing cards (Pit/Pit-Phy/Olympic
* chipsets) written by:
* 1999 Peter De Schrijver All Rights Reserved
* 1999 Mike Phillips (phillim@amtrak.com)
*
* Base Driver Skeleton:
* Written 1993-94 by Donald Becker.
*
* Copyright 1993 United States Government as represented by the
* Director, National Security Agency.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* NO WARRANTY
* THE PROGRAM IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OR
* CONDITIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT
* LIMITATION, ANY WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT,
* MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is
* solely responsible for determining the appropriateness of using and
* distributing the Program and assumes all risks associated with its
* exercise of rights under this Agreement, including but not limited to
* the risks and costs of program errors, damage to or loss of data,
* programs or equipment, and unavailability or interruption of operations.
*
* DISCLAIMER OF LIABILITY
* NEITHER RECIPIENT NOR ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
* TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
* USE OR DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED
* HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES
*
* 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
*
*
* 12/10/99 - Alpha Release 0.1.0
* First release to the public
* 03/03/00 - Merged to kernel, indented -kr -i8 -bri0, fixed some missing
* malloc free checks, reviewed code. <alan@redhat.com>
* 03/13/00 - Added spinlocks for smp
* 03/08/01 - Added support for module_init() and module_exit()
* 08/15/01 - Added ioctl() functionality for debugging, changed netif_*_queue
* calls and other incorrectness - Kent Yoder <yoder1@us.ibm.com>
* 11/05/01 - Restructured the interrupt function, added delays, reduced the
* the number of TX descriptors to 1, which together can prevent
* the card from locking up the box - <yoder1@us.ibm.com>
* 09/27/02 - New PCI interface + bug fix. - <yoder1@us.ibm.com>
* 11/13/02 - Removed free_irq calls which could cause a hang, added
* netif_carrier_{on|off} - <yoder1@us.ibm.com>
*
* To Do:
*
*
* If Problems do Occur
* Most problems can be rectified by either closing and opening the interface
* (ifconfig down and up) or rmmod and insmod'ing the driver (a bit difficult
* if compiled into the kernel).
*/
/* Change STREAMER_DEBUG to 1 to get verbose, and I mean really verbose, messages */
#define STREAMER_DEBUG 0
#define STREAMER_DEBUG_PACKETS 0
/* Change STREAMER_NETWORK_MONITOR to receive mac frames through the arb channel.
* Will also create a /proc/net/streamer_tr entry if proc_fs is compiled into the
* kernel.
* Intended to be used to create a ring-error reporting network module
* i.e. it will give you the source address of beaconers on the ring
*/
#define STREAMER_NETWORK_MONITOR 0
/* #define CONFIG_PROC_FS */
/*
* Allow or disallow ioctl's for debugging
*/
#define STREAMER_IOCTL 0
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/in.h>
#include <linux/ioport.h>
#include <linux/string.h>
#include <linux/proc_fs.h>
#include <linux/ptrace.h>
#include <linux/skbuff.h>
#include <linux/interrupt.h>
#include <linux/delay.h>
#include <linux/netdevice.h>
#include <linux/trdevice.h>
#include <linux/stddef.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/spinlock.h>
#include <linux/bitops.h>
#include <linux/jiffies.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 09:04:11 +01:00
#include <linux/slab.h>
#include <net/net_namespace.h>
#include <net/checksum.h>
#include <asm/io.h>
#include "lanstreamer.h"
#if (BITS_PER_LONG == 64)
#error broken on 64-bit: stores pointer to rx_ring->buffer in 32-bit int
#endif
/* I've got to put some intelligence into the version number so that Peter and I know
* which version of the code somebody has got.
* Version Number = a.b.c.d where a.b.c is the level of code and d is the latest author.
* So 0.0.1.pds = Peter, 0.0.1.mlp = Mike
*
* Official releases will only have an a.b.c version number format.
*/
static char version[] = "LanStreamer.c v0.4.0 03/08/01 - Mike Sullivan\n"
" v0.5.3 11/13/02 - Kent Yoder";
static DEFINE_PCI_DEVICE_TABLE(streamer_pci_tbl) = {
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_TR, PCI_ANY_ID, PCI_ANY_ID,},
{} /* terminating entry */
};
MODULE_DEVICE_TABLE(pci,streamer_pci_tbl);
static char *open_maj_error[] = {
"No error", "Lobe Media Test", "Physical Insertion",
"Address Verification", "Neighbor Notification (Ring Poll)",
"Request Parameters", "FDX Registration Request",
"FDX Lobe Media Test", "FDX Duplicate Address Check",
"Unknown stage"
};
static char *open_min_error[] = {
"No error", "Function Failure", "Signal Lost", "Wire Fault",
"Ring Speed Mismatch", "Timeout", "Ring Failure", "Ring Beaconing",
"Duplicate Node Address", "Request Parameters", "Remove Received",
"Reserved", "Reserved", "No Monitor Detected for RPL",
"Monitor Contention failer for RPL", "FDX Protocol Error"
};
/* Module parameters */
/* Ring Speed 0,4,16
* 0 = Autosense
* 4,16 = Selected speed only, no autosense
* This allows the card to be the first on the ring
* and become the active monitor.
*
* WARNING: Some hubs will allow you to insert
* at the wrong speed
*/
static int ringspeed[STREAMER_MAX_ADAPTERS] = { 0, };
module_param_array(ringspeed, int, NULL, 0);
/* Packet buffer size */
static int pkt_buf_sz[STREAMER_MAX_ADAPTERS] = { 0, };
module_param_array(pkt_buf_sz, int, NULL, 0);
/* Message Level */
static int message_level[STREAMER_MAX_ADAPTERS] = { 1, };
module_param_array(message_level, int, NULL, 0);
#if STREAMER_IOCTL
static int streamer_ioctl(struct net_device *, struct ifreq *, int);
#endif
static int streamer_reset(struct net_device *dev);
static int streamer_open(struct net_device *dev);
static netdev_tx_t streamer_xmit(struct sk_buff *skb,
struct net_device *dev);
static int streamer_close(struct net_device *dev);
static void streamer_set_rx_mode(struct net_device *dev);
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 15:55:46 +02:00
static irqreturn_t streamer_interrupt(int irq, void *dev_id);
static int streamer_set_mac_address(struct net_device *dev, void *addr);
static void streamer_arb_cmd(struct net_device *dev);
static int streamer_change_mtu(struct net_device *dev, int mtu);
static void streamer_srb_bh(struct net_device *dev);
static void streamer_asb_bh(struct net_device *dev);
#if STREAMER_NETWORK_MONITOR
#ifdef CONFIG_PROC_FS
static int streamer_proc_info(char *buffer, char **start, off_t offset,
int length, int *eof, void *data);
static int sprintf_info(char *buffer, struct net_device *dev);
struct streamer_private *dev_streamer=NULL;
#endif
#endif
static const struct net_device_ops streamer_netdev_ops = {
.ndo_open = streamer_open,
.ndo_stop = streamer_close,
.ndo_start_xmit = streamer_xmit,
.ndo_change_mtu = streamer_change_mtu,
#if STREAMER_IOCTL
.ndo_do_ioctl = streamer_ioctl,
#endif
.ndo_set_rx_mode = streamer_set_rx_mode,
.ndo_set_mac_address = streamer_set_mac_address,
};
static int __devinit streamer_init_one(struct pci_dev *pdev,
const struct pci_device_id *ent)
{
struct net_device *dev;
struct streamer_private *streamer_priv;
unsigned long pio_start, pio_end, pio_flags, pio_len;
unsigned long mmio_start, mmio_end, mmio_flags, mmio_len;
int rc = 0;
static int card_no=-1;
u16 pcr;
#if STREAMER_DEBUG
printk("lanstreamer::streamer_init_one, entry pdev %p\n",pdev);
#endif
card_no++;
dev = alloc_trdev(sizeof(*streamer_priv));
if (dev==NULL) {
printk(KERN_ERR "lanstreamer: out of memory.\n");
return -ENOMEM;
}
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
streamer_priv = netdev_priv(dev);
#if STREAMER_NETWORK_MONITOR
#ifdef CONFIG_PROC_FS
if (!dev_streamer)
create_proc_read_entry("streamer_tr", 0, init_net.proc_net,
streamer_proc_info, NULL);
streamer_priv->next = dev_streamer;
dev_streamer = streamer_priv;
#endif
#endif
rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
if (rc) {
printk(KERN_ERR "%s: No suitable PCI mapping available.\n",
dev->name);
rc = -ENODEV;
goto err_out;
}
rc = pci_enable_device(pdev);
if (rc) {
printk(KERN_ERR "lanstreamer: unable to enable pci device\n");
rc=-EIO;
goto err_out;
}
pci_set_master(pdev);
rc = pci_set_mwi(pdev);
if (rc) {
printk(KERN_ERR "lanstreamer: unable to enable MWI on pci device\n");
goto err_out_disable;
}
pio_start = pci_resource_start(pdev, 0);
pio_end = pci_resource_end(pdev, 0);
pio_flags = pci_resource_flags(pdev, 0);
pio_len = pci_resource_len(pdev, 0);
mmio_start = pci_resource_start(pdev, 1);
mmio_end = pci_resource_end(pdev, 1);
mmio_flags = pci_resource_flags(pdev, 1);
mmio_len = pci_resource_len(pdev, 1);
#if STREAMER_DEBUG
printk("lanstreamer: pio_start %x pio_end %x pio_len %x pio_flags %x\n",
pio_start, pio_end, pio_len, pio_flags);
printk("lanstreamer: mmio_start %x mmio_end %x mmio_len %x mmio_flags %x\n",
mmio_start, mmio_end, mmio_flags, mmio_len);
#endif
if (!request_region(pio_start, pio_len, "lanstreamer")) {
printk(KERN_ERR "lanstreamer: unable to get pci io addr %lx\n",
pio_start);
rc= -EBUSY;
goto err_out_mwi;
}
if (!request_mem_region(mmio_start, mmio_len, "lanstreamer")) {
printk(KERN_ERR "lanstreamer: unable to get pci mmio addr %lx\n",
mmio_start);
rc= -EBUSY;
goto err_out_free_pio;
}
streamer_priv->streamer_mmio=ioremap(mmio_start, mmio_len);
if (streamer_priv->streamer_mmio == NULL) {
printk(KERN_ERR "lanstreamer: unable to remap MMIO %lx\n",
mmio_start);
rc= -EIO;
goto err_out_free_mmio;
}
init_waitqueue_head(&streamer_priv->srb_wait);
init_waitqueue_head(&streamer_priv->trb_wait);
dev->netdev_ops = &streamer_netdev_ops;
dev->irq = pdev->irq;
dev->base_addr=pio_start;
SET_NETDEV_DEV(dev, &pdev->dev);
streamer_priv->streamer_card_name = (char *)pdev->resource[0].name;
streamer_priv->pci_dev = pdev;
if ((pkt_buf_sz[card_no] < 100) || (pkt_buf_sz[card_no] > 18000))
streamer_priv->pkt_buf_sz = PKT_BUF_SZ;
else
streamer_priv->pkt_buf_sz = pkt_buf_sz[card_no];
streamer_priv->streamer_ring_speed = ringspeed[card_no];
streamer_priv->streamer_message_level = message_level[card_no];
pci_set_drvdata(pdev, dev);
spin_lock_init(&streamer_priv->streamer_lock);
pci_read_config_word (pdev, PCI_COMMAND, &pcr);
pcr |= PCI_COMMAND_SERR;
pci_write_config_word (pdev, PCI_COMMAND, pcr);
printk("%s\n", version);
printk("%s: %s. I/O at %hx, MMIO at %p, using irq %d\n",dev->name,
streamer_priv->streamer_card_name,
(unsigned int) dev->base_addr,
streamer_priv->streamer_mmio,
dev->irq);
if (streamer_reset(dev))
goto err_out_unmap;
rc = register_netdev(dev);
if (rc)
goto err_out_unmap;
return 0;
err_out_unmap:
iounmap(streamer_priv->streamer_mmio);
err_out_free_mmio:
release_mem_region(mmio_start, mmio_len);
err_out_free_pio:
release_region(pio_start, pio_len);
err_out_mwi:
pci_clear_mwi(pdev);
err_out_disable:
pci_disable_device(pdev);
err_out:
free_netdev(dev);
#if STREAMER_DEBUG
printk("lanstreamer: Exit error %x\n",rc);
#endif
return rc;
}
static void __devexit streamer_remove_one(struct pci_dev *pdev)
{
struct net_device *dev=pci_get_drvdata(pdev);
struct streamer_private *streamer_priv;
#if STREAMER_DEBUG
printk("lanstreamer::streamer_remove_one entry pdev %p\n",pdev);
#endif
if (dev == NULL) {
printk(KERN_ERR "lanstreamer::streamer_remove_one, ERROR dev is NULL\n");
return;
}
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
streamer_priv=netdev_priv(dev);
if (streamer_priv == NULL) {
printk(KERN_ERR "lanstreamer::streamer_remove_one, ERROR dev->priv is NULL\n");
return;
}
#if STREAMER_NETWORK_MONITOR
#ifdef CONFIG_PROC_FS
{
struct streamer_private **p, **next;
for (p = &dev_streamer; *p; p = next) {
next = &(*p)->next;
if (*p == streamer_priv) {
*p = *next;
break;
}
}
if (!dev_streamer)
remove_proc_entry("streamer_tr", init_net.proc_net);
}
#endif
#endif
unregister_netdev(dev);
iounmap(streamer_priv->streamer_mmio);
release_mem_region(pci_resource_start(pdev, 1), pci_resource_len(pdev,1));
release_region(pci_resource_start(pdev, 0), pci_resource_len(pdev,0));
pci_clear_mwi(pdev);
pci_disable_device(pdev);
free_netdev(dev);
pci_set_drvdata(pdev, NULL);
}
static int streamer_reset(struct net_device *dev)
{
struct streamer_private *streamer_priv;
__u8 __iomem *streamer_mmio;
unsigned long t;
unsigned int uaa_addr;
struct sk_buff *skb = NULL;
__u16 misr;
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
streamer_priv = netdev_priv(dev);
streamer_mmio = streamer_priv->streamer_mmio;
writew(readw(streamer_mmio + BCTL) | BCTL_SOFTRESET, streamer_mmio + BCTL);
t = jiffies;
/* Hold soft reset bit for a while */
ssleep(1);
writew(readw(streamer_mmio + BCTL) & ~BCTL_SOFTRESET,
streamer_mmio + BCTL);
#if STREAMER_DEBUG
printk("BCTL: %x\n", readw(streamer_mmio + BCTL));
printk("GPR: %x\n", readw(streamer_mmio + GPR));
printk("SISRMASK: %x\n", readw(streamer_mmio + SISR_MASK));
#endif
writew(readw(streamer_mmio + BCTL) | (BCTL_RX_FIFO_8 | BCTL_TX_FIFO_8), streamer_mmio + BCTL );
if (streamer_priv->streamer_ring_speed == 0) { /* Autosense */
writew(readw(streamer_mmio + GPR) | GPR_AUTOSENSE,
streamer_mmio + GPR);
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Ringspeed autosense mode on\n",
dev->name);
} else if (streamer_priv->streamer_ring_speed == 16) {
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Trying to open at 16 Mbps as requested\n",
dev->name);
writew(GPR_16MBPS, streamer_mmio + GPR);
} else if (streamer_priv->streamer_ring_speed == 4) {
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Trying to open at 4 Mbps as requested\n",
dev->name);
writew(0, streamer_mmio + GPR);
}
skb = dev_alloc_skb(streamer_priv->pkt_buf_sz);
if (!skb) {
printk(KERN_INFO "%s: skb allocation for diagnostics failed...proceeding\n",
dev->name);
} else {
struct streamer_rx_desc *rx_ring;
u8 *data;
rx_ring=(struct streamer_rx_desc *)skb->data;
data=((u8 *)skb->data)+sizeof(struct streamer_rx_desc);
rx_ring->forward=0;
rx_ring->status=0;
rx_ring->buffer=cpu_to_le32(pci_map_single(streamer_priv->pci_dev, data,
512, PCI_DMA_FROMDEVICE));
rx_ring->framelen_buflen=512;
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev, rx_ring, 512, PCI_DMA_FROMDEVICE)),
streamer_mmio+RXBDA);
}
#if STREAMER_DEBUG
printk("GPR = %x\n", readw(streamer_mmio + GPR));
#endif
/* start solo init */
writew(SISR_MI, streamer_mmio + SISR_MASK_SUM);
while (!((readw(streamer_mmio + SISR)) & SISR_SRB_REPLY)) {
msleep_interruptible(100);
if (time_after(jiffies, t + 40 * HZ)) {
printk(KERN_ERR
"IBM PCI tokenring card not responding\n");
release_region(dev->base_addr, STREAMER_IO_SPACE);
if (skb)
dev_kfree_skb(skb);
return -1;
}
}
writew(~SISR_SRB_REPLY, streamer_mmio + SISR_RUM);
misr = readw(streamer_mmio + MISR_RUM);
writew(~misr, streamer_mmio + MISR_RUM);
if (skb)
dev_kfree_skb(skb); /* release skb used for diagnostics */
#if STREAMER_DEBUG
printk("LAPWWO: %x, LAPA: %x LAPE: %x\n",
readw(streamer_mmio + LAPWWO), readw(streamer_mmio + LAPA),
readw(streamer_mmio + LAPE));
#endif
#if STREAMER_DEBUG
{
int i;
writew(readw(streamer_mmio + LAPWWO),
streamer_mmio + LAPA);
printk("initialization response srb dump: ");
for (i = 0; i < 10; i++)
printk("%x:",
ntohs(readw(streamer_mmio + LAPDINC)));
printk("\n");
}
#endif
writew(readw(streamer_mmio + LAPWWO) + 6, streamer_mmio + LAPA);
if (readw(streamer_mmio + LAPD)) {
printk(KERN_INFO "tokenring card initialization failed. errorcode : %x\n",
ntohs(readw(streamer_mmio + LAPD)));
release_region(dev->base_addr, STREAMER_IO_SPACE);
return -1;
}
writew(readw(streamer_mmio + LAPWWO) + 8, streamer_mmio + LAPA);
uaa_addr = ntohs(readw(streamer_mmio + LAPDINC));
readw(streamer_mmio + LAPDINC); /* skip over Level.Addr field */
streamer_priv->streamer_addr_table_addr = ntohs(readw(streamer_mmio + LAPDINC));
streamer_priv->streamer_parms_addr = ntohs(readw(streamer_mmio + LAPDINC));
#if STREAMER_DEBUG
printk("UAA resides at %x\n", uaa_addr);
#endif
/* setup uaa area for access with LAPD */
{
int i;
__u16 addr;
writew(uaa_addr, streamer_mmio + LAPA);
for (i = 0; i < 6; i += 2) {
addr=ntohs(readw(streamer_mmio+LAPDINC));
dev->dev_addr[i]= (addr >> 8) & 0xff;
dev->dev_addr[i+1]= addr & 0xff;
}
#if STREAMER_DEBUG
printk("Adapter address: %pM\n", dev->dev_addr);
#endif
}
return 0;
}
static int streamer_open(struct net_device *dev)
{
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
struct streamer_private *streamer_priv = netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags;
char open_error[255];
int i, open_finished = 1;
__u16 srb_word;
__u16 srb_open;
int rc;
if (readw(streamer_mmio+BMCTL_SUM) & BMCTL_RX_ENABLED) {
rc=streamer_reset(dev);
}
if (request_irq(dev->irq, streamer_interrupt, IRQF_SHARED, "lanstreamer", dev)) {
return -EAGAIN;
}
#if STREAMER_DEBUG
printk("BMCTL: %x\n", readw(streamer_mmio + BMCTL_SUM));
printk("pending ints: %x\n", readw(streamer_mmio + SISR));
#endif
writew(SISR_MI | SISR_SRB_REPLY, streamer_mmio + SISR_MASK); /* more ints later, doesn't stop arb cmd interrupt */
writew(LISR_LIE, streamer_mmio + LISR); /* more ints later */
/* adapter is closed, so SRB is pointed to by LAPWWO */
writew(readw(streamer_mmio + LAPWWO), streamer_mmio + LAPA);
#if STREAMER_DEBUG
printk("LAPWWO: %x, LAPA: %x\n", readw(streamer_mmio + LAPWWO),
readw(streamer_mmio + LAPA));
printk("LAPE: %x\n", readw(streamer_mmio + LAPE));
printk("SISR Mask = %04x\n", readw(streamer_mmio + SISR_MASK));
#endif
do {
for (i = 0; i < SRB_COMMAND_SIZE; i += 2) {
writew(0, streamer_mmio + LAPDINC);
}
writew(readw(streamer_mmio+LAPWWO),streamer_mmio+LAPA);
writew(htons(SRB_OPEN_ADAPTER<<8),streamer_mmio+LAPDINC) ; /* open */
writew(htons(STREAMER_CLEAR_RET_CODE<<8),streamer_mmio+LAPDINC);
writew(STREAMER_CLEAR_RET_CODE, streamer_mmio + LAPDINC);
writew(readw(streamer_mmio + LAPWWO) + 8, streamer_mmio + LAPA);
#if STREAMER_NETWORK_MONITOR
/* If Network Monitor, instruct card to copy MAC frames through the ARB */
writew(htons(OPEN_ADAPTER_ENABLE_FDX | OPEN_ADAPTER_PASS_ADC_MAC | OPEN_ADAPTER_PASS_ATT_MAC | OPEN_ADAPTER_PASS_BEACON), streamer_mmio + LAPDINC); /* offset 8 word contains open options */
#else
writew(htons(OPEN_ADAPTER_ENABLE_FDX), streamer_mmio + LAPDINC); /* Offset 8 word contains Open.Options */
#endif
if (streamer_priv->streamer_laa[0]) {
writew(readw(streamer_mmio + LAPWWO) + 12, streamer_mmio + LAPA);
writew(htons((streamer_priv->streamer_laa[0] << 8) |
streamer_priv->streamer_laa[1]),streamer_mmio+LAPDINC);
writew(htons((streamer_priv->streamer_laa[2] << 8) |
streamer_priv->streamer_laa[3]),streamer_mmio+LAPDINC);
writew(htons((streamer_priv->streamer_laa[4] << 8) |
streamer_priv->streamer_laa[5]),streamer_mmio+LAPDINC);
memcpy(dev->dev_addr, streamer_priv->streamer_laa, dev->addr_len);
}
/* save off srb open offset */
srb_open = readw(streamer_mmio + LAPWWO);
#if STREAMER_DEBUG
writew(readw(streamer_mmio + LAPWWO),
streamer_mmio + LAPA);
printk("srb open request:\n");
for (i = 0; i < 16; i++) {
printk("%x:", ntohs(readw(streamer_mmio + LAPDINC)));
}
printk("\n");
#endif
spin_lock_irqsave(&streamer_priv->streamer_lock, flags);
streamer_priv->srb_queued = 1;
/* signal solo that SRB command has been issued */
writew(LISR_SRB_CMD, streamer_mmio + LISR_SUM);
spin_unlock_irqrestore(&streamer_priv->streamer_lock, flags);
while (streamer_priv->srb_queued) {
interruptible_sleep_on_timeout(&streamer_priv->srb_wait, 5 * HZ);
if (signal_pending(current)) {
printk(KERN_WARNING "%s: SRB timed out.\n", dev->name);
printk(KERN_WARNING "SISR=%x MISR=%x, LISR=%x\n",
readw(streamer_mmio + SISR),
readw(streamer_mmio + MISR_RUM),
readw(streamer_mmio + LISR));
streamer_priv->srb_queued = 0;
break;
}
}
#if STREAMER_DEBUG
printk("SISR_MASK: %x\n", readw(streamer_mmio + SISR_MASK));
printk("srb open response:\n");
writew(srb_open, streamer_mmio + LAPA);
for (i = 0; i < 10; i++) {
printk("%x:",
ntohs(readw(streamer_mmio + LAPDINC)));
}
#endif
/* If we get the same return response as we set, the interrupt wasn't raised and the open
* timed out.
*/
writew(srb_open + 2, streamer_mmio + LAPA);
srb_word = ntohs(readw(streamer_mmio + LAPD)) >> 8;
if (srb_word == STREAMER_CLEAR_RET_CODE) {
printk(KERN_WARNING "%s: Adapter Open time out or error.\n",
dev->name);
return -EIO;
}
if (srb_word != 0) {
if (srb_word == 0x07) {
if (!streamer_priv->streamer_ring_speed && open_finished) { /* Autosense , first time around */
printk(KERN_WARNING "%s: Retrying at different ring speed\n",
dev->name);
open_finished = 0;
} else {
__u16 error_code;
writew(srb_open + 6, streamer_mmio + LAPA);
error_code = ntohs(readw(streamer_mmio + LAPD));
strcpy(open_error, open_maj_error[(error_code & 0xf0) >> 4]);
strcat(open_error, " - ");
strcat(open_error, open_min_error[(error_code & 0x0f)]);
if (!streamer_priv->streamer_ring_speed &&
((error_code & 0x0f) == 0x0d))
{
printk(KERN_WARNING "%s: Tried to autosense ring speed with no monitors present\n", dev->name);
printk(KERN_WARNING "%s: Please try again with a specified ring speed\n", dev->name);
free_irq(dev->irq, dev);
return -EIO;
}
printk(KERN_WARNING "%s: %s\n",
dev->name, open_error);
free_irq(dev->irq, dev);
return -EIO;
} /* if autosense && open_finished */
} else {
printk(KERN_WARNING "%s: Bad OPEN response: %x\n",
dev->name, srb_word);
free_irq(dev->irq, dev);
return -EIO;
}
} else
open_finished = 1;
} while (!(open_finished)); /* Will only loop if ring speed mismatch re-open attempted && autosense is on */
writew(srb_open + 18, streamer_mmio + LAPA);
srb_word=ntohs(readw(streamer_mmio+LAPD)) >> 8;
if (srb_word & (1 << 3))
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Opened in FDX Mode\n", dev->name);
if (srb_word & 1)
streamer_priv->streamer_ring_speed = 16;
else
streamer_priv->streamer_ring_speed = 4;
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Opened in %d Mbps mode\n",
dev->name,
streamer_priv->streamer_ring_speed);
writew(srb_open + 8, streamer_mmio + LAPA);
streamer_priv->asb = ntohs(readw(streamer_mmio + LAPDINC));
streamer_priv->srb = ntohs(readw(streamer_mmio + LAPDINC));
streamer_priv->arb = ntohs(readw(streamer_mmio + LAPDINC));
readw(streamer_mmio + LAPDINC); /* offset 14 word is rsvd */
streamer_priv->trb = ntohs(readw(streamer_mmio + LAPDINC));
streamer_priv->streamer_receive_options = 0x00;
streamer_priv->streamer_copy_all_options = 0;
/* setup rx ring */
/* enable rx channel */
writew(~BMCTL_RX_DIS, streamer_mmio + BMCTL_RUM);
/* setup rx descriptors */
streamer_priv->streamer_rx_ring=
kmalloc( sizeof(struct streamer_rx_desc)*
STREAMER_RX_RING_SIZE,GFP_KERNEL);
if (!streamer_priv->streamer_rx_ring) {
printk(KERN_WARNING "%s ALLOC of streamer rx ring FAILED!!\n",dev->name);
return -EIO;
}
for (i = 0; i < STREAMER_RX_RING_SIZE; i++) {
struct sk_buff *skb;
skb = dev_alloc_skb(streamer_priv->pkt_buf_sz);
if (skb == NULL)
break;
skb->dev = dev;
streamer_priv->streamer_rx_ring[i].forward =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, &streamer_priv->streamer_rx_ring[i + 1],
sizeof(struct streamer_rx_desc), PCI_DMA_FROMDEVICE));
streamer_priv->streamer_rx_ring[i].status = 0;
streamer_priv->streamer_rx_ring[i].buffer =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, skb->data,
streamer_priv->pkt_buf_sz, PCI_DMA_FROMDEVICE));
streamer_priv->streamer_rx_ring[i].framelen_buflen = streamer_priv->pkt_buf_sz;
streamer_priv->rx_ring_skb[i] = skb;
}
streamer_priv->streamer_rx_ring[STREAMER_RX_RING_SIZE - 1].forward =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, &streamer_priv->streamer_rx_ring[0],
sizeof(struct streamer_rx_desc), PCI_DMA_FROMDEVICE));
if (i == 0) {
printk(KERN_WARNING "%s: Not enough memory to allocate rx buffers. Adapter disabled\n", dev->name);
free_irq(dev->irq, dev);
return -EIO;
}
streamer_priv->rx_ring_last_received = STREAMER_RX_RING_SIZE - 1; /* last processed rx status */
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev, &streamer_priv->streamer_rx_ring[0],
sizeof(struct streamer_rx_desc), PCI_DMA_TODEVICE)),
streamer_mmio + RXBDA);
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev, &streamer_priv->streamer_rx_ring[STREAMER_RX_RING_SIZE - 1],
sizeof(struct streamer_rx_desc), PCI_DMA_TODEVICE)),
streamer_mmio + RXLBDA);
/* set bus master interrupt event mask */
writew(MISR_RX_NOBUF | MISR_RX_EOF, streamer_mmio + MISR_MASK);
/* setup tx ring */
streamer_priv->streamer_tx_ring=kmalloc(sizeof(struct streamer_tx_desc)*
STREAMER_TX_RING_SIZE,GFP_KERNEL);
if (!streamer_priv->streamer_tx_ring) {
printk(KERN_WARNING "%s ALLOC of streamer_tx_ring FAILED\n",dev->name);
return -EIO;
}
writew(~BMCTL_TX2_DIS, streamer_mmio + BMCTL_RUM); /* Enables TX channel 2 */
for (i = 0; i < STREAMER_TX_RING_SIZE; i++) {
streamer_priv->streamer_tx_ring[i].forward = cpu_to_le32(pci_map_single(streamer_priv->pci_dev,
&streamer_priv->streamer_tx_ring[i + 1],
sizeof(struct streamer_tx_desc),
PCI_DMA_TODEVICE));
streamer_priv->streamer_tx_ring[i].status = 0;
streamer_priv->streamer_tx_ring[i].bufcnt_framelen = 0;
streamer_priv->streamer_tx_ring[i].buffer = 0;
streamer_priv->streamer_tx_ring[i].buflen = 0;
streamer_priv->streamer_tx_ring[i].rsvd1 = 0;
streamer_priv->streamer_tx_ring[i].rsvd2 = 0;
streamer_priv->streamer_tx_ring[i].rsvd3 = 0;
}
streamer_priv->streamer_tx_ring[STREAMER_TX_RING_SIZE - 1].forward =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, &streamer_priv->streamer_tx_ring[0],
sizeof(struct streamer_tx_desc), PCI_DMA_TODEVICE));
streamer_priv->free_tx_ring_entries = STREAMER_TX_RING_SIZE;
streamer_priv->tx_ring_free = 0; /* next entry in tx ring to use */
streamer_priv->tx_ring_last_status = STREAMER_TX_RING_SIZE - 1;
/* set Busmaster interrupt event mask (handle receives on interrupt only */
writew(MISR_TX2_EOF | MISR_RX_NOBUF | MISR_RX_EOF, streamer_mmio + MISR_MASK);
/* set system event interrupt mask */
writew(SISR_ADAPTER_CHECK | SISR_ARB_CMD | SISR_TRB_REPLY | SISR_ASB_FREE, streamer_mmio + SISR_MASK_SUM);
#if STREAMER_DEBUG
printk("BMCTL: %x\n", readw(streamer_mmio + BMCTL_SUM));
printk("SISR MASK: %x\n", readw(streamer_mmio + SISR_MASK));
#endif
#if STREAMER_NETWORK_MONITOR
writew(streamer_priv->streamer_addr_table_addr, streamer_mmio + LAPA);
printk("%s: Node Address: %04x:%04x:%04x\n", dev->name,
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)));
readw(streamer_mmio + LAPDINC);
readw(streamer_mmio + LAPDINC);
printk("%s: Functional Address: %04x:%04x\n", dev->name,
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)));
writew(streamer_priv->streamer_parms_addr + 4,
streamer_mmio + LAPA);
printk("%s: NAUN Address: %04x:%04x:%04x\n", dev->name,
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)));
#endif
netif_start_queue(dev);
netif_carrier_on(dev);
return 0;
}
/*
* When we enter the rx routine we do not know how many frames have been
* queued on the rx channel. Therefore we start at the next rx status
* position and travel around the receive ring until we have completed
* all the frames.
*
* This means that we may process the frame before we receive the end
* of frame interrupt. This is why we always test the status instead
* of blindly processing the next frame.
*
*/
static void streamer_rx(struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
struct streamer_rx_desc *rx_desc;
int rx_ring_last_received, length, frame_length, buffer_cnt = 0;
struct sk_buff *skb, *skb2;
/* setup the next rx descriptor to be received */
rx_desc = &streamer_priv->streamer_rx_ring[(streamer_priv->rx_ring_last_received + 1) & (STREAMER_RX_RING_SIZE - 1)];
rx_ring_last_received = streamer_priv->rx_ring_last_received;
while (rx_desc->status & 0x01000000) { /* While processed descriptors are available */
if (rx_ring_last_received != streamer_priv->rx_ring_last_received)
{
printk(KERN_WARNING "RX Error 1 rx_ring_last_received not the same %x %x\n",
rx_ring_last_received, streamer_priv->rx_ring_last_received);
}
streamer_priv->rx_ring_last_received = (streamer_priv->rx_ring_last_received + 1) & (STREAMER_RX_RING_SIZE - 1);
rx_ring_last_received = streamer_priv->rx_ring_last_received;
length = rx_desc->framelen_buflen & 0xffff; /* buffer length */
frame_length = (rx_desc->framelen_buflen >> 16) & 0xffff;
if (rx_desc->status & 0x7E830000) { /* errors */
if (streamer_priv->streamer_message_level) {
printk(KERN_WARNING "%s: Rx Error %x\n",
dev->name, rx_desc->status);
}
} else { /* received without errors */
if (rx_desc->status & 0x80000000) { /* frame complete */
buffer_cnt = 1;
skb = dev_alloc_skb(streamer_priv->pkt_buf_sz);
} else {
skb = dev_alloc_skb(frame_length);
}
if (skb == NULL)
{
printk(KERN_WARNING "%s: Not enough memory to copy packet to upper layers.\n", dev->name);
dev->stats.rx_dropped++;
} else { /* we allocated an skb OK */
if (buffer_cnt == 1) {
/* release the DMA mapping */
pci_unmap_single(streamer_priv->pci_dev,
le32_to_cpu(streamer_priv->streamer_rx_ring[rx_ring_last_received].buffer),
streamer_priv->pkt_buf_sz,
PCI_DMA_FROMDEVICE);
skb2 = streamer_priv->rx_ring_skb[rx_ring_last_received];
#if STREAMER_DEBUG_PACKETS
{
int i;
printk("streamer_rx packet print: skb->data2 %p skb->head %p\n", skb2->data, skb2->head);
for (i = 0; i < frame_length; i++)
{
printk("%x:", skb2->data[i]);
if (((i + 1) % 16) == 0)
printk("\n");
}
printk("\n");
}
#endif
skb_put(skb2, length);
skb2->protocol = tr_type_trans(skb2, dev);
/* recycle this descriptor */
streamer_priv->streamer_rx_ring[rx_ring_last_received].status = 0;
streamer_priv->streamer_rx_ring[rx_ring_last_received].framelen_buflen = streamer_priv->pkt_buf_sz;
streamer_priv->streamer_rx_ring[rx_ring_last_received].buffer =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, skb->data, streamer_priv->pkt_buf_sz,
PCI_DMA_FROMDEVICE));
streamer_priv->rx_ring_skb[rx_ring_last_received] = skb;
/* place recycled descriptor back on the adapter */
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev,
&streamer_priv->streamer_rx_ring[rx_ring_last_received],
sizeof(struct streamer_rx_desc), PCI_DMA_FROMDEVICE)),
streamer_mmio + RXLBDA);
/* pass the received skb up to the protocol */
netif_rx(skb2);
} else {
do { /* Walk the buffers */
pci_unmap_single(streamer_priv->pci_dev, le32_to_cpu(rx_desc->buffer), length, PCI_DMA_FROMDEVICE),
memcpy(skb_put(skb, length), (void *)rx_desc->buffer, length); /* copy this fragment */
streamer_priv->streamer_rx_ring[rx_ring_last_received].status = 0;
streamer_priv->streamer_rx_ring[rx_ring_last_received].framelen_buflen = streamer_priv->pkt_buf_sz;
/* give descriptor back to the adapter */
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev,
&streamer_priv->streamer_rx_ring[rx_ring_last_received],
length, PCI_DMA_FROMDEVICE)),
streamer_mmio + RXLBDA);
if (rx_desc->status & 0x80000000)
break; /* this descriptor completes the frame */
/* else get the next pending descriptor */
if (rx_ring_last_received!= streamer_priv->rx_ring_last_received)
{
printk("RX Error rx_ring_last_received not the same %x %x\n",
rx_ring_last_received,
streamer_priv->rx_ring_last_received);
}
rx_desc = &streamer_priv->streamer_rx_ring[(streamer_priv->rx_ring_last_received+1) & (STREAMER_RX_RING_SIZE-1)];
length = rx_desc->framelen_buflen & 0xffff; /* buffer length */
streamer_priv->rx_ring_last_received = (streamer_priv->rx_ring_last_received+1) & (STREAMER_RX_RING_SIZE - 1);
rx_ring_last_received = streamer_priv->rx_ring_last_received;
} while (1);
skb->protocol = tr_type_trans(skb, dev);
/* send up to the protocol */
netif_rx(skb);
}
dev->stats.rx_packets++;
dev->stats.rx_bytes += length;
} /* if skb == null */
} /* end received without errors */
/* try the next one */
rx_desc = &streamer_priv->streamer_rx_ring[(rx_ring_last_received + 1) & (STREAMER_RX_RING_SIZE - 1)];
} /* end for all completed rx descriptors */
}
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 15:55:46 +02:00
static irqreturn_t streamer_interrupt(int irq, void *dev_id)
{
struct net_device *dev = (struct net_device *) dev_id;
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u16 sisr;
__u16 misr;
u8 max_intr = MAX_INTR;
spin_lock(&streamer_priv->streamer_lock);
sisr = readw(streamer_mmio + SISR);
while((sisr & (SISR_MI | SISR_SRB_REPLY | SISR_ADAPTER_CHECK | SISR_ASB_FREE |
SISR_ARB_CMD | SISR_TRB_REPLY | SISR_PAR_ERR | SISR_SERR_ERR)) &&
(max_intr > 0)) {
if(sisr & SISR_PAR_ERR) {
writew(~SISR_PAR_ERR, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
else if(sisr & SISR_SERR_ERR) {
writew(~SISR_SERR_ERR, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
else if(sisr & SISR_MI) {
misr = readw(streamer_mmio + MISR_RUM);
if (misr & MISR_TX2_EOF) {
while(streamer_priv->streamer_tx_ring[(streamer_priv->tx_ring_last_status + 1) & (STREAMER_TX_RING_SIZE - 1)].status) {
streamer_priv->tx_ring_last_status = (streamer_priv->tx_ring_last_status + 1) & (STREAMER_TX_RING_SIZE - 1);
streamer_priv->free_tx_ring_entries++;
dev->stats.tx_bytes += streamer_priv->tx_ring_skb[streamer_priv->tx_ring_last_status]->len;
dev->stats.tx_packets++;
dev_kfree_skb_irq(streamer_priv->tx_ring_skb[streamer_priv->tx_ring_last_status]);
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].buffer = 0xdeadbeef;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].status = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].bufcnt_framelen = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].buflen = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].rsvd1 = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].rsvd2 = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_last_status].rsvd3 = 0;
}
netif_wake_queue(dev);
}
if (misr & MISR_RX_EOF) {
streamer_rx(dev);
}
/* MISR_RX_EOF */
if (misr & MISR_RX_NOBUF) {
/* According to the documentation, we don't have to do anything,
* but trapping it keeps it out of /var/log/messages.
*/
} /* SISR_RX_NOBUF */
writew(~misr, streamer_mmio + MISR_RUM);
(void)readw(streamer_mmio + MISR_RUM);
}
else if (sisr & SISR_SRB_REPLY) {
if (streamer_priv->srb_queued == 1) {
wake_up_interruptible(&streamer_priv->srb_wait);
} else if (streamer_priv->srb_queued == 2) {
streamer_srb_bh(dev);
}
streamer_priv->srb_queued = 0;
writew(~SISR_SRB_REPLY, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
else if (sisr & SISR_ADAPTER_CHECK) {
printk(KERN_WARNING "%s: Adapter Check Interrupt Raised, 8 bytes of information follow:\n", dev->name);
writel(readl(streamer_mmio + LAPWWO), streamer_mmio + LAPA);
printk(KERN_WARNING "%s: Words %x:%x:%x:%x:\n",
dev->name, readw(streamer_mmio + LAPDINC),
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)),
ntohs(readw(streamer_mmio + LAPDINC)));
netif_stop_queue(dev);
netif_carrier_off(dev);
printk(KERN_WARNING "%s: Adapter must be manually reset.\n", dev->name);
}
/* SISR_ADAPTER_CHECK */
else if (sisr & SISR_ASB_FREE) {
/* Wake up anything that is waiting for the asb response */
if (streamer_priv->asb_queued) {
streamer_asb_bh(dev);
}
writew(~SISR_ASB_FREE, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
/* SISR_ASB_FREE */
else if (sisr & SISR_ARB_CMD) {
streamer_arb_cmd(dev);
writew(~SISR_ARB_CMD, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
/* SISR_ARB_CMD */
else if (sisr & SISR_TRB_REPLY) {
/* Wake up anything that is waiting for the trb response */
if (streamer_priv->trb_queued) {
wake_up_interruptible(&streamer_priv->
trb_wait);
}
streamer_priv->trb_queued = 0;
writew(~SISR_TRB_REPLY, streamer_mmio + SISR_RUM);
(void)readw(streamer_mmio + SISR_RUM);
}
/* SISR_TRB_REPLY */
sisr = readw(streamer_mmio + SISR);
max_intr--;
} /* while() */
spin_unlock(&streamer_priv->streamer_lock) ;
return IRQ_HANDLED;
}
static netdev_tx_t streamer_xmit(struct sk_buff *skb,
struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags ;
spin_lock_irqsave(&streamer_priv->streamer_lock, flags);
if (streamer_priv->free_tx_ring_entries) {
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].status = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].bufcnt_framelen = 0x00020000 | skb->len;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].buffer =
cpu_to_le32(pci_map_single(streamer_priv->pci_dev, skb->data, skb->len, PCI_DMA_TODEVICE));
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].rsvd1 = skb->len;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].rsvd2 = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].rsvd3 = 0;
streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free].buflen = skb->len;
streamer_priv->tx_ring_skb[streamer_priv->tx_ring_free] = skb;
streamer_priv->free_tx_ring_entries--;
#if STREAMER_DEBUG_PACKETS
{
int i;
printk("streamer_xmit packet print:\n");
for (i = 0; i < skb->len; i++) {
printk("%x:", skb->data[i]);
if (((i + 1) % 16) == 0)
printk("\n");
}
printk("\n");
}
#endif
writel(cpu_to_le32(pci_map_single(streamer_priv->pci_dev,
&streamer_priv->streamer_tx_ring[streamer_priv->tx_ring_free],
sizeof(struct streamer_tx_desc), PCI_DMA_TODEVICE)),
streamer_mmio + TX2LFDA);
(void)readl(streamer_mmio + TX2LFDA);
streamer_priv->tx_ring_free = (streamer_priv->tx_ring_free + 1) & (STREAMER_TX_RING_SIZE - 1);
spin_unlock_irqrestore(&streamer_priv->streamer_lock,flags);
return NETDEV_TX_OK;
} else {
netif_stop_queue(dev);
spin_unlock_irqrestore(&streamer_priv->streamer_lock,flags);
return NETDEV_TX_BUSY;
}
}
static int streamer_close(struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags;
int i;
netif_stop_queue(dev);
netif_carrier_off(dev);
writew(streamer_priv->srb, streamer_mmio + LAPA);
writew(htons(SRB_CLOSE_ADAPTER << 8),streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
spin_lock_irqsave(&streamer_priv->streamer_lock, flags);
streamer_priv->srb_queued = 1;
writew(LISR_SRB_CMD, streamer_mmio + LISR_SUM);
spin_unlock_irqrestore(&streamer_priv->streamer_lock, flags);
while (streamer_priv->srb_queued)
{
interruptible_sleep_on_timeout(&streamer_priv->srb_wait,
jiffies + 60 * HZ);
if (signal_pending(current))
{
printk(KERN_WARNING "%s: SRB timed out.\n", dev->name);
printk(KERN_WARNING "SISR=%x MISR=%x LISR=%x\n",
readw(streamer_mmio + SISR),
readw(streamer_mmio + MISR_RUM),
readw(streamer_mmio + LISR));
streamer_priv->srb_queued = 0;
break;
}
}
streamer_priv->rx_ring_last_received = (streamer_priv->rx_ring_last_received + 1) & (STREAMER_RX_RING_SIZE - 1);
for (i = 0; i < STREAMER_RX_RING_SIZE; i++) {
if (streamer_priv->rx_ring_skb[streamer_priv->rx_ring_last_received]) {
dev_kfree_skb(streamer_priv->rx_ring_skb[streamer_priv->rx_ring_last_received]);
}
streamer_priv->rx_ring_last_received = (streamer_priv->rx_ring_last_received + 1) & (STREAMER_RX_RING_SIZE - 1);
}
/* reset tx/rx fifo's and busmaster logic */
/* TBD. Add graceful way to reset the LLC channel without doing a soft reset.
writel(readl(streamer_mmio+BCTL)|(3<<13),streamer_mmio+BCTL);
udelay(1);
writel(readl(streamer_mmio+BCTL)&~(3<<13),streamer_mmio+BCTL);
*/
#if STREAMER_DEBUG
writew(streamer_priv->srb, streamer_mmio + LAPA);
printk("srb): ");
for (i = 0; i < 2; i++) {
printk("%x ", ntohs(readw(streamer_mmio + LAPDINC)));
}
printk("\n");
#endif
free_irq(dev->irq, dev);
return 0;
}
static void streamer_set_rx_mode(struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u8 options = 0;
struct netdev_hw_addr *ha;
unsigned char dev_mc_address[5];
writel(streamer_priv->srb, streamer_mmio + LAPA);
options = streamer_priv->streamer_copy_all_options;
if (dev->flags & IFF_PROMISC)
options |= (3 << 5); /* All LLC and MAC frames, all through the main rx channel */
else
options &= ~(3 << 5);
/* Only issue the srb if there is a change in options */
if ((options ^ streamer_priv->streamer_copy_all_options))
{
/* Now to issue the srb command to alter the copy.all.options */
writew(htons(SRB_MODIFY_RECEIVE_OPTIONS << 8), streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
writew(htons((streamer_priv->streamer_receive_options << 8) | options),streamer_mmio+LAPDINC);
writew(htons(0x4a41),streamer_mmio+LAPDINC);
writew(htons(0x4d45),streamer_mmio+LAPDINC);
writew(htons(0x5320),streamer_mmio+LAPDINC);
writew(0x2020, streamer_mmio + LAPDINC);
streamer_priv->srb_queued = 2; /* Can't sleep, use srb_bh */
writel(LISR_SRB_CMD, streamer_mmio + LISR_SUM);
streamer_priv->streamer_copy_all_options = options;
return;
}
/* Set the functional addresses we need for multicast */
writel(streamer_priv->srb,streamer_mmio+LAPA);
dev_mc_address[0] = dev_mc_address[1] = dev_mc_address[2] = dev_mc_address[3] = 0 ;
netdev_for_each_mc_addr(ha, dev) {
dev_mc_address[0] |= ha->addr[2];
dev_mc_address[1] |= ha->addr[3];
dev_mc_address[2] |= ha->addr[4];
dev_mc_address[3] |= ha->addr[5];
}
writew(htons(SRB_SET_FUNC_ADDRESS << 8),streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
writew(0,streamer_mmio+LAPDINC);
writew(htons( (dev_mc_address[0] << 8) | dev_mc_address[1]),streamer_mmio+LAPDINC);
writew(htons( (dev_mc_address[2] << 8) | dev_mc_address[3]),streamer_mmio+LAPDINC);
streamer_priv->srb_queued = 2 ;
writel(LISR_SRB_CMD,streamer_mmio+LISR_SUM);
}
static void streamer_srb_bh(struct net_device *dev)
{
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
struct streamer_private *streamer_priv = netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u16 srb_word;
writew(streamer_priv->srb, streamer_mmio + LAPA);
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
/* SRB_MODIFY_RECEIVE_OPTIONS i.e. set_multicast_list options (promiscuous)
* At some point we should do something if we get an error, such as
* resetting the IFF_PROMISC flag in dev
*/
case SRB_MODIFY_RECEIVE_OPTIONS:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n", dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
default:
if (streamer_priv->streamer_message_level)
printk(KERN_WARNING "%s: Receive Options Modified to %x,%x\n",
dev->name,
streamer_priv->streamer_copy_all_options,
streamer_priv->streamer_receive_options);
break;
} /* switch srb[2] */
break;
/* SRB_SET_GROUP_ADDRESS - Multicast group setting
*/
case SRB_SET_GROUP_ADDRESS:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x00:
break;
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n",dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
case 0x3c:
printk(KERN_WARNING "%s: Group/Functional address indicator bits not set correctly\n", dev->name);
break;
case 0x3e: /* If we ever implement individual multicast addresses, will need to deal with this */
printk(KERN_WARNING "%s: Group address registers full\n", dev->name);
break;
case 0x55:
printk(KERN_INFO "%s: Group Address already set.\n", dev->name);
break;
default:
break;
} /* switch srb[2] */
break;
/* SRB_RESET_GROUP_ADDRESS - Remove a multicast address from group list
*/
case SRB_RESET_GROUP_ADDRESS:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x00:
break;
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n", dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
case 0x39: /* Must deal with this if individual multicast addresses used */
printk(KERN_INFO "%s: Group address not found\n", dev->name);
break;
default:
break;
} /* switch srb[2] */
break;
/* SRB_SET_FUNC_ADDRESS - Called by the set_rx_mode
*/
case SRB_SET_FUNC_ADDRESS:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x00:
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Functional Address Mask Set\n", dev->name);
break;
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n", dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
default:
break;
} /* switch srb[2] */
break;
/* SRB_READ_LOG - Read and reset the adapter error counters
*/
case SRB_READ_LOG:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x00:
{
int i;
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Read Log command complete\n", dev->name);
printk("Read Log statistics: ");
writew(streamer_priv->srb + 6,
streamer_mmio + LAPA);
for (i = 0; i < 5; i++) {
printk("%x:", ntohs(readw(streamer_mmio + LAPDINC)));
}
printk("\n");
}
break;
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n", dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
} /* switch srb[2] */
break;
/* SRB_READ_SR_COUNTERS - Read and reset the source routing bridge related counters */
case SRB_READ_SR_COUNTERS:
srb_word=ntohs(readw(streamer_mmio+LAPDINC)) >> 8;
switch (srb_word) {
case 0x00:
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Read Source Routing Counters issued\n", dev->name);
break;
case 0x01:
printk(KERN_WARNING "%s: Unrecognized srb command\n", dev->name);
break;
case 0x04:
printk(KERN_WARNING "%s: Adapter must be open for this operation, doh!!\n", dev->name);
break;
default:
break;
} /* switch srb[2] */
break;
default:
printk(KERN_WARNING "%s: Unrecognized srb bh return value.\n", dev->name);
break;
} /* switch srb[0] */
}
static int streamer_set_mac_address(struct net_device *dev, void *addr)
{
struct sockaddr *saddr = addr;
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
struct streamer_private *streamer_priv = netdev_priv(dev);
if (netif_running(dev))
{
printk(KERN_WARNING "%s: Cannot set mac/laa address while card is open\n", dev->name);
return -EIO;
}
memcpy(streamer_priv->streamer_laa, saddr->sa_data, dev->addr_len);
if (streamer_priv->streamer_message_level) {
printk(KERN_INFO "%s: MAC/LAA Set to = %x.%x.%x.%x.%x.%x\n",
dev->name, streamer_priv->streamer_laa[0],
streamer_priv->streamer_laa[1],
streamer_priv->streamer_laa[2],
streamer_priv->streamer_laa[3],
streamer_priv->streamer_laa[4],
streamer_priv->streamer_laa[5]);
}
return 0;
}
static void streamer_arb_cmd(struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u8 header_len;
__u16 frame_len, buffer_len;
struct sk_buff *mac_frame;
__u8 frame_data[256];
__u16 buff_off;
__u16 lan_status = 0, lan_status_diff; /* Initialize to stop compiler warning */
__u8 fdx_prot_error;
__u16 next_ptr;
__u16 arb_word;
#if STREAMER_NETWORK_MONITOR
struct trh_hdr *mac_hdr;
#endif
writew(streamer_priv->arb, streamer_mmio + LAPA);
arb_word=ntohs(readw(streamer_mmio+LAPD)) >> 8;
if (arb_word == ARB_RECEIVE_DATA) { /* Receive.data, MAC frames */
writew(streamer_priv->arb + 6, streamer_mmio + LAPA);
streamer_priv->mac_rx_buffer = buff_off = ntohs(readw(streamer_mmio + LAPDINC));
header_len=ntohs(readw(streamer_mmio+LAPDINC)) >> 8; /* 802.5 Token-Ring Header Length */
frame_len = ntohs(readw(streamer_mmio + LAPDINC));
#if STREAMER_DEBUG
{
int i;
__u16 next;
__u8 status;
__u16 len;
writew(ntohs(buff_off), streamer_mmio + LAPA); /*setup window to frame data */
next = htons(readw(streamer_mmio + LAPDINC));
status =
ntohs(readw(streamer_mmio + LAPDINC)) & 0xff;
len = ntohs(readw(streamer_mmio + LAPDINC));
/* print out 1st 14 bytes of frame data */
for (i = 0; i < 7; i++) {
printk("Loc %d = %04x\n", i,
ntohs(readw
(streamer_mmio + LAPDINC)));
}
printk("next %04x, fs %02x, len %04x\n", next,
status, len);
}
#endif
if (!(mac_frame = dev_alloc_skb(frame_len))) {
printk(KERN_WARNING "%s: Memory squeeze, dropping frame.\n",
dev->name);
goto drop_frame;
}
/* Walk the buffer chain, creating the frame */
do {
int i;
__u16 rx_word;
writew(htons(buff_off), streamer_mmio + LAPA); /* setup window to frame data */
next_ptr = ntohs(readw(streamer_mmio + LAPDINC));
readw(streamer_mmio + LAPDINC); /* read thru status word */
buffer_len = ntohs(readw(streamer_mmio + LAPDINC));
if (buffer_len > 256)
break;
i = 0;
while (i < buffer_len) {
rx_word=ntohs(readw(streamer_mmio+LAPDINC));
frame_data[i]=rx_word >> 8;
frame_data[i+1]=rx_word & 0xff;
i += 2;
}
memcpy(skb_put(mac_frame, buffer_len),
frame_data, buffer_len);
} while (next_ptr && (buff_off = next_ptr));
mac_frame->protocol = tr_type_trans(mac_frame, dev);
#if STREAMER_NETWORK_MONITOR
printk(KERN_WARNING "%s: Received MAC Frame, details:\n",
dev->name);
mac_hdr = tr_hdr(mac_frame);
printk(KERN_WARNING
"%s: MAC Frame Dest. Addr: %pM\n",
dev->name, mac_hdr->daddr);
printk(KERN_WARNING
"%s: MAC Frame Srce. Addr: %pM\n",
dev->name, mac_hdr->saddr);
#endif
netif_rx(mac_frame);
/* Now tell the card we have dealt with the received frame */
drop_frame:
/* Set LISR Bit 1 */
writel(LISR_ARB_FREE, streamer_priv->streamer_mmio + LISR_SUM);
/* Is the ASB free ? */
if (!(readl(streamer_priv->streamer_mmio + SISR) & SISR_ASB_FREE))
{
streamer_priv->asb_queued = 1;
writel(LISR_ASB_FREE_REQ, streamer_priv->streamer_mmio + LISR_SUM);
return;
/* Drop out and wait for the bottom half to be run */
}
writew(streamer_priv->asb, streamer_mmio + LAPA);
writew(htons(ASB_RECEIVE_DATA << 8), streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
writew(0, streamer_mmio + LAPDINC);
writew(htons(streamer_priv->mac_rx_buffer), streamer_mmio + LAPD);
writel(LISR_ASB_REPLY | LISR_ASB_FREE_REQ, streamer_priv->streamer_mmio + LISR_SUM);
streamer_priv->asb_queued = 2;
return;
} else if (arb_word == ARB_LAN_CHANGE_STATUS) { /* Lan.change.status */
writew(streamer_priv->arb + 6, streamer_mmio + LAPA);
lan_status = ntohs(readw(streamer_mmio + LAPDINC));
fdx_prot_error = ntohs(readw(streamer_mmio+LAPD)) >> 8;
/* Issue ARB Free */
writew(LISR_ARB_FREE, streamer_priv->streamer_mmio + LISR_SUM);
lan_status_diff = (streamer_priv->streamer_lan_status ^ lan_status) &
lan_status;
if (lan_status_diff & (LSC_LWF | LSC_ARW | LSC_FPE | LSC_RR))
{
if (lan_status_diff & LSC_LWF)
printk(KERN_WARNING "%s: Short circuit detected on the lobe\n", dev->name);
if (lan_status_diff & LSC_ARW)
printk(KERN_WARNING "%s: Auto removal error\n", dev->name);
if (lan_status_diff & LSC_FPE)
printk(KERN_WARNING "%s: FDX Protocol Error\n", dev->name);
if (lan_status_diff & LSC_RR)
printk(KERN_WARNING "%s: Force remove MAC frame received\n", dev->name);
/* Adapter has been closed by the hardware */
/* reset tx/rx fifo's and busmaster logic */
/* @TBD. no llc reset on autostreamer writel(readl(streamer_mmio+BCTL)|(3<<13),streamer_mmio+BCTL);
udelay(1);
writel(readl(streamer_mmio+BCTL)&~(3<<13),streamer_mmio+BCTL); */
netif_stop_queue(dev);
netif_carrier_off(dev);
printk(KERN_WARNING "%s: Adapter must be manually reset.\n", dev->name);
}
/* If serious error */
if (streamer_priv->streamer_message_level) {
if (lan_status_diff & LSC_SIG_LOSS)
printk(KERN_WARNING "%s: No receive signal detected\n", dev->name);
if (lan_status_diff & LSC_HARD_ERR)
printk(KERN_INFO "%s: Beaconing\n", dev->name);
if (lan_status_diff & LSC_SOFT_ERR)
printk(KERN_WARNING "%s: Adapter transmitted Soft Error Report Mac Frame\n", dev->name);
if (lan_status_diff & LSC_TRAN_BCN)
printk(KERN_INFO "%s: We are transmitting the beacon, aaah\n", dev->name);
if (lan_status_diff & LSC_SS)
printk(KERN_INFO "%s: Single Station on the ring\n", dev->name);
if (lan_status_diff & LSC_RING_REC)
printk(KERN_INFO "%s: Ring recovery ongoing\n", dev->name);
if (lan_status_diff & LSC_FDX_MODE)
printk(KERN_INFO "%s: Operating in FDX mode\n", dev->name);
}
if (lan_status_diff & LSC_CO) {
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Counter Overflow\n", dev->name);
/* Issue READ.LOG command */
writew(streamer_priv->srb, streamer_mmio + LAPA);
writew(htons(SRB_READ_LOG << 8),streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
writew(0, streamer_mmio + LAPDINC);
streamer_priv->srb_queued = 2; /* Can't sleep, use srb_bh */
writew(LISR_SRB_CMD, streamer_mmio + LISR_SUM);
}
if (lan_status_diff & LSC_SR_CO) {
if (streamer_priv->streamer_message_level)
printk(KERN_INFO "%s: Source routing counters overflow\n", dev->name);
/* Issue a READ.SR.COUNTERS */
writew(streamer_priv->srb, streamer_mmio + LAPA);
writew(htons(SRB_READ_SR_COUNTERS << 8),
streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8),
streamer_mmio+LAPDINC);
streamer_priv->srb_queued = 2; /* Can't sleep, use srb_bh */
writew(LISR_SRB_CMD, streamer_mmio + LISR_SUM);
}
streamer_priv->streamer_lan_status = lan_status;
} /* Lan.change.status */
else
printk(KERN_WARNING "%s: Unknown arb command\n", dev->name);
}
static void streamer_asb_bh(struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
if (streamer_priv->asb_queued == 1)
{
/* Dropped through the first time */
writew(streamer_priv->asb, streamer_mmio + LAPA);
writew(htons(ASB_RECEIVE_DATA << 8),streamer_mmio+LAPDINC);
writew(htons(STREAMER_CLEAR_RET_CODE << 8), streamer_mmio+LAPDINC);
writew(0, streamer_mmio + LAPDINC);
writew(htons(streamer_priv->mac_rx_buffer), streamer_mmio + LAPD);
writel(LISR_ASB_REPLY | LISR_ASB_FREE_REQ, streamer_priv->streamer_mmio + LISR_SUM);
streamer_priv->asb_queued = 2;
return;
}
if (streamer_priv->asb_queued == 2) {
__u8 rc;
writew(streamer_priv->asb + 2, streamer_mmio + LAPA);
rc=ntohs(readw(streamer_mmio+LAPD)) >> 8;
switch (rc) {
case 0x01:
printk(KERN_WARNING "%s: Unrecognized command code\n", dev->name);
break;
case 0x26:
printk(KERN_WARNING "%s: Unrecognized buffer address\n", dev->name);
break;
case 0xFF:
/* Valid response, everything should be ok again */
break;
default:
printk(KERN_WARNING "%s: Invalid return code in asb\n", dev->name);
break;
}
}
streamer_priv->asb_queued = 0;
}
static int streamer_change_mtu(struct net_device *dev, int mtu)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u16 max_mtu;
if (streamer_priv->streamer_ring_speed == 4)
max_mtu = 4500;
else
max_mtu = 18000;
if (mtu > max_mtu)
return -EINVAL;
if (mtu < 100)
return -EINVAL;
dev->mtu = mtu;
streamer_priv->pkt_buf_sz = mtu + TR_HLEN;
return 0;
}
#if STREAMER_NETWORK_MONITOR
#ifdef CONFIG_PROC_FS
static int streamer_proc_info(char *buffer, char **start, off_t offset,
int length, int *eof, void *data)
{
struct streamer_private *sdev=NULL;
struct pci_dev *pci_device = NULL;
int len = 0;
off_t begin = 0;
off_t pos = 0;
int size;
struct net_device *dev;
size = sprintf(buffer, "IBM LanStreamer/MPC Chipset Token Ring Adapters\n");
pos += size;
len += size;
for(sdev=dev_streamer; sdev; sdev=sdev->next) {
pci_device=sdev->pci_dev;
dev=pci_get_drvdata(pci_device);
size = sprintf_info(buffer + len, dev);
len += size;
pos = begin + len;
if (pos < offset) {
len = 0;
begin = pos;
}
if (pos > offset + length)
break;
} /* for */
*start = buffer + (offset - begin); /* Start of wanted data */
len -= (offset - begin); /* Start slop */
if (len > length)
len = length; /* Ending slop */
return len;
}
static int sprintf_info(char *buffer, struct net_device *dev)
{
struct streamer_private *streamer_priv =
dev->priv to netdev_priv(dev), drivers/net/tokenring/ Replacing accesses to dev->priv to netdev_priv(dev). The replacment is safe when netdev_priv is used to access a private structure that is right next to the net_device structure in memory. Cf http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/de19321bcd94dbb8/0d74a4adcd6177bd This is the case when the net_device structure was allocated with a call to alloc_netdev or one of its derivative. Here is an excerpt of the semantic patch that performs the transformation @ rule1 @ type T; struct net_device *dev; @@ dev = ( alloc_netdev | alloc_etherdev | alloc_trdev ) (sizeof(T), ...) @ rule1bis @ struct net_device *dev; expression E; @@ dev->priv = E @ rule2 depends on rule1 && !rule1bis @ struct net_device *dev; type rule1.T; @@ - (T*) dev->priv + netdev_priv(dev) PS: I have performed the same transformation on the whole kernel and it affects around 70 files, most of them in drivers/net/. Should I split my patch for each subnet directories ? (wireless/, wan/, etc) Thanks to Thomas Surrel for helping me refining my semantic patch. Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> 3c359.c | 58 +++++++++++++++++++++++++++++----------------------------- ibmtr.c | 38 +++++++++++++++++++------------------- lanstreamer.c | 32 ++++++++++++++++---------------- madgemc.c | 4 ++-- olympic.c | 36 ++++++++++++++++++------------------ tmspci.c | 4 ++-- 6 files changed, 86 insertions(+), 86 deletions(-) Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-07-23 15:18:21 +02:00
netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
struct streamer_adapter_addr_table sat;
struct streamer_parameters_table spt;
int size = 0;
int i;
writew(streamer_priv->streamer_addr_table_addr, streamer_mmio + LAPA);
for (i = 0; i < 14; i += 2) {
__u16 io_word;
__u8 *datap = (__u8 *) & sat;
io_word=ntohs(readw(streamer_mmio+LAPDINC));
datap[size]=io_word >> 8;
datap[size+1]=io_word & 0xff;
}
writew(streamer_priv->streamer_parms_addr, streamer_mmio + LAPA);
for (i = 0; i < 68; i += 2) {
__u16 io_word;
__u8 *datap = (__u8 *) & spt;
io_word=ntohs(readw(streamer_mmio+LAPDINC));
datap[size]=io_word >> 8;
datap[size+1]=io_word & 0xff;
}
size = sprintf(buffer, "\n%6s: Adapter Address : Node Address : Functional Addr\n", dev->name);
size += sprintf(buffer + size,
"%6s: %pM : %pM : %02x:%02x:%02x:%02x\n",
dev->name, dev->dev_addr, sat.node_addr,
sat.func_addr[0], sat.func_addr[1],
sat.func_addr[2], sat.func_addr[3]);
size += sprintf(buffer + size, "\n%6s: Token Ring Parameters Table:\n", dev->name);
size += sprintf(buffer + size, "%6s: Physical Addr : Up Node Address : Poll Address : AccPri : Auth Src : Att Code :\n", dev->name);
size += sprintf(buffer + size,
"%6s: %02x:%02x:%02x:%02x : %pM : %pM : %04x : %04x : %04x :\n",
dev->name, spt.phys_addr[0], spt.phys_addr[1],
spt.phys_addr[2], spt.phys_addr[3],
spt.up_node_addr, spt.poll_addr,
ntohs(spt.acc_priority), ntohs(spt.auth_source_class),
ntohs(spt.att_code));
size += sprintf(buffer + size, "%6s: Source Address : Bcn T : Maj. V : Lan St : Lcl Rg : Mon Err : Frame Correl : \n", dev->name);
size += sprintf(buffer + size,
"%6s: %pM : %04x : %04x : %04x : %04x : %04x : %04x : \n",
dev->name, spt.source_addr,
ntohs(spt.beacon_type), ntohs(spt.major_vector),
ntohs(spt.lan_status), ntohs(spt.local_ring),
ntohs(spt.mon_error), ntohs(spt.frame_correl));
size += sprintf(buffer + size, "%6s: Beacon Details : Tx : Rx : NAUN Node Address : NAUN Node Phys : \n",
dev->name);
size += sprintf(buffer + size,
"%6s: : %02x : %02x : %pM : %02x:%02x:%02x:%02x : \n",
dev->name, ntohs(spt.beacon_transmit),
ntohs(spt.beacon_receive),
spt.beacon_naun,
spt.beacon_phys[0], spt.beacon_phys[1],
spt.beacon_phys[2], spt.beacon_phys[3]);
return size;
}
#endif
#endif
static struct pci_driver streamer_pci_driver = {
.name = "lanstreamer",
.id_table = streamer_pci_tbl,
.probe = streamer_init_one,
.remove = __devexit_p(streamer_remove_one),
};
module_pci_driver(streamer_pci_driver);
MODULE_LICENSE("GPL");