* Re: [PATCH] lro: myri10ge example how to use LRO
From: Andrew Gallatin @ 2007-08-03 21:00 UTC (permalink / raw)
To: Kok, Auke
Cc: netdev, raisch, themann, linux-kernel, linuxppc-dev, meder,
tklein, stefan.roscher, jeff, davem
In-Reply-To: <46B39063.6070301@intel.com>
Kok, Auke wrote:
> Andrew Gallatin wrote:
>> To follow up on Jan-Bernd Themann's LRO patch earlier today,
>> this patch shows how the generic LRO interface can be used for
>> page based drivers.
>>
>> Again, many thanks to Jan-Bernd Themann for leading this effort.
>>
>> Drew
>>
>> Singed off by: Andrew Gallatin <gallatin@myri.com>
>>
>
>
> please take a look at my lro patch for ethtool and see if it works for
> you, instead of adding another generic module parameter that doesn't
> need to be there.
That looks very nice, and will indeed work for me.
Thanks,
Drew
^ permalink raw reply
* Re: [PATCH] lro: myri10ge example how to use LRO
From: Kok, Auke @ 2007-08-03 20:30 UTC (permalink / raw)
To: Andrew Gallatin
Cc: netdev, raisch, themann, linux-kernel, linuxppc-dev, meder,
tklein, stefan.roscher, jeff, davem
In-Reply-To: <46B363A8.4010407@myri.com>
Andrew Gallatin wrote:
> To follow up on Jan-Bernd Themann's LRO patch earlier today,
> this patch shows how the generic LRO interface can be used for
> page based drivers.
>
> Again, many thanks to Jan-Bernd Themann for leading this effort.
>
> Drew
>
> Singed off by: Andrew Gallatin <gallatin@myri.com>
>
please take a look at my lro patch for ethtool and see if it works for you,
instead of adding another generic module parameter that doesn't need to be there.
Thanks.
Auke
^ permalink raw reply
* Re: [PATCH] lro: eHEA example how to use LRO
From: Kok, Auke @ 2007-08-03 20:29 UTC (permalink / raw)
To: Jan-Bernd Themann
Cc: netdev, Christoph Raisch, Jan-Bernd Themann, linux-kernel,
linux-ppc, Marcus Eder, Thomas Klein, Stefan Roscher,
Andrew Gallatin, Jeff Garzik, David Miller
In-Reply-To: <200708031441.26841.ossthema@de.ibm.com>
Jan-Bernd Themann wrote:
> This patch shows how the generic LRO interface is used for SKB mode
>
> Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
>
> ---
> drivers/net/Kconfig | 1 +
> drivers/net/ehea/ehea.h | 9 ++++-
> drivers/net/ehea/ehea_ethtool.c | 15 +++++++
> drivers/net/ehea/ehea_main.c | 84 +++++++++++++++++++++++++++++++++++---
> 4 files changed, 101 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
> index f8a602c..fec4004 100644
> --- a/drivers/net/Kconfig
> +++ b/drivers/net/Kconfig
<snip>
> +module_param(use_lro, int, 0);
Have you looked at my generic lro get/set patch that I posted this week? this
adds a useless module parameter while ethtool has all the structure already to
accomodate setting lro on/off.
Auke
^ permalink raw reply
* Re: strange tcp behavior
From: David Miller @ 2007-08-03 20:04 UTC (permalink / raw)
To: johnpol; +Cc: simon, john, netdev
In-Reply-To: <20070803082242.GA25582@2ka.mipt.ru>
From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Date: Fri, 3 Aug 2007 12:22:42 +0400
> On Thu, Aug 02, 2007 at 07:21:34PM -0700, David Miller (davem@davemloft.net) wrote:
> > What in the world are we doing allowing stream sockets to autobind?
> > That is totally bogus. Even if we autobind, that won't make a connect
> > happen.
>
> For accepted socket it is perfectly valid assumption - we could autobind
> it during the first send. Or may bind it during accept. Its a matter of
> taste I think. Autobinding during first sending can end up being a
> protection against DoS in some obscure rare case...
accept()ed socket is by definition fully bound and already in
established state.
^ permalink raw reply
* Re: Distributed storage.
From: Daniel Phillips @ 2007-08-03 19:48 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Evgeniy Polyakov, netdev, linux-kernel, linux-fsdevel,
Arnaldo Carvalho de Melo
In-Reply-To: <1186152817.12034.128.camel@twins>
On Friday 03 August 2007 07:53, Peter Zijlstra wrote:
> On Fri, 2007-08-03 at 17:49 +0400, Evgeniy Polyakov wrote:
> > On Fri, Aug 03, 2007 at 02:27:52PM +0200, Peter Zijlstra wrote:
> > ...my main position is to
> > allocate per socket reserve from socket's queue, and copy data
> > there from main reserve, all of which are allocated either in
> > advance (global one) or per sockoption, so that there would be no
> > fairness issues what to mark as special and what to not.
> >
> > Say we have a page per socket, each socket can assign a reserve for
> > itself from own memory, this accounts both tx and rx side. Tx is
> > not interesting, it is simple, rx has global reserve (always
> > allocated on startup or sometime way before reclaim/oom)where data
> > is originally received (including skb, shared info and whatever is
> > needed, page is just an exmaple), then it is copied into per-socket
> > reserve and reused for the next packet. Having per-socket reserve
> > allows to have progress in any situation not only in cases where
> > single action must be received/processed, and allows to be
> > completely fair for all users, but not only special sockets, thus
> > admin for example would be allowed to login, ipsec would work and
> > so on...
>
> Ah, I think I understand now. Yes this is indeed a good idea!
>
> It would be quite doable to implement this on top of that I already
> have. We would need to extend the socket with a sock_opt that would
> reserve a specified amount of data for that specific socket. And then
> on socket demux check if the socket has a non zero reserve and has
> not yet exceeded said reserve. If so, process the packet.
>
> This would also quite neatly work for -rt where we would not want
> incomming packet processing to be delayed by memory allocations.
At this point we need "anything that works" in mainline as a starting
point. By erring on the side of simplicity we can make this
understandable for folks who haven't spent the last two years wallowing
in it. The page per socket approach is about as simple as it gets. I
therefore propose we save our premature optimizations for later.
It will also help our cause if we keep any new internal APIs to strictly
what is needed to make deadlock go away. Not a whole lot more than
just the flag to mark a socket as part of the vm writeout path when you
get right down to essentials.
Regards,
Daniel
^ permalink raw reply
* Re: Distributed storage.
From: Daniel Phillips @ 2007-08-03 19:41 UTC (permalink / raw)
To: Evgeniy Polyakov
Cc: Peter Zijlstra, netdev, linux-kernel, linux-fsdevel,
Arnaldo Carvalho de Melo
In-Reply-To: <20070803134943.GA21221@2ka.mipt.ru>
On Friday 03 August 2007 06:49, Evgeniy Polyakov wrote:
> ...rx has global reserve (always allocated on
> startup or sometime way before reclaim/oom)where data is originally
> received (including skb, shared info and whatever is needed, page is
> just an exmaple), then it is copied into per-socket reserve and
> reused for the next packet. Having per-socket reserve allows to have
> progress in any situation not only in cases where single action must
> be received/processed, and allows to be completely fair for all
> users, but not only special sockets, thus admin for example would be
> allowed to login, ipsec would work and so on...
And when the global reserve is entirely used up your system goes back to
dropping vm writeout acknowledgements, not so good. I like your
approach, and specifically the copying idea cuts out considerable
complexity. But I believe the per-socket flag to mark a socket as part
of the vm writeout path is not optional, and in this case it will be a
better world if it is a slightly unfair world in favor of vm writeout
traffic.
Ssh will still work fine even with vm getting priority access to the
pool. During memory crunches, non-vm ssh traffic may get bumped till
after the crunch, but vm writeout is never supposed to hog the whole
machine. If vm writeout hogs your machine long enough to delay an ssh
login then that is a vm bug and should be fixed at that level.
Regards,
Daniel
^ permalink raw reply
* Re: strange tcp behavior
From: Simon Arlott @ 2007-08-03 18:29 UTC (permalink / raw)
To: Evgeniy Polyakov; +Cc: john, netdev, David Miller
In-Reply-To: <20070803173906.GA30688@2ka.mipt.ru>
On 03/08/07 18:39, Evgeniy Polyakov wrote:
> On Fri, Aug 03, 2007 at 05:51:42PM +0100, Simon Arlott (simon@fire.lp0.eu) wrote:
>
>> 17:38:03.533589 IP 192.168.7.4.50550 > 192.168.7.8.2500: R 82517592:82517592(0) win 1500 (raw)
>> vs
>> 17:37:38.383085 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 4259643274:4259643274(0) ack 1171836829 win 14360
>> What happened there ?
Erm... you seem to have removed parts of my message in a way that doesn't
make sense...
On Fri, Aug 03, 2007 at 05:51:42PM +0100, Simon Arlott wrote:
> 17:38:04.536277 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 1:1(0) ack 17 win 14360
> vs
> 17:37:38.383085 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 4259643274:4259643274(0) ack 1171836829 win 14360
> What happened there ?
The first one is the RST sent when the connection is close()d without
reading, and the second one is the same RST but after other connection
has been made on the same ports using a different socket.
> It is the same situation, which would happen if you will spam remote
> side with RST packets with arbitrary sequence number in hope that it
> will reset some connection.
Isn't it still possible that the connection that got reset is left open
(possibly for days) until another connection using the same ports is
using roughly the same sequence numbers?
--
Simon Arlott
^ permalink raw reply
* Re: [patch 0/5][RFC] Update network drivers to use devres
From: Brandon Philips @ 2007-08-03 18:23 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: Tejun Heo, netdev, teheo, bphilips
In-Reply-To: <20070803144409.3f95c91d@oldman.hamilton.local>
On 14:44 Fri 03 Aug 2007, Stephen Hemminger wrote:
> On Fri, 03 Aug 2007 20:33:04 +0900 Tejun Heo <htejun@gmail.com> wrote:
> > >> Devres makes low level drivers simpler, easier to get right and
> > >> maintain. Writing new drivers becomes easier too. So, why not?
> > >>
> > >>> Network devices seem to work fine thanks, and the resource requirements
> > >>> are different. If ain't broke, don't fix it.
> > >> Care to enlighten me on how the resource requirments are different
> > >> from ATA drivers?
> > >
> > > I was thinking of the hot remove (no mod ref counts) and lingering
> > > /sys open issues. ATA drivers use ref counts.
> >
> > I guess the hot removing is done by severing netdev from the actual
> > device, right? I don't see how that affects usage of devres on network
> > drivers. Am I missing something?
>
> The issue is that device may be removed at any time. So you can't rely
> on module ref counts to save you. And netdevice structure must still
> linger after module is removed, till dev ref count goes to zero.
These patches allow the net_device to linger. The code calls
free_netdev on device removal just as before.
This is how the net_device is handled on device removal by these
patches:
+static void devm_free_netdev(struct device *gendev, void *res)
+{
+ struct net_device *dev = dev_get_drvdata(gendev);
+ free_netdev(dev);
+}
> > On a separate note, can you explain lingering /sys open issue to me a
> > bit? With recent sysfs changes, sysfs nodes are disconnected
> > immediately on deletion. Would that make any difference to netdevs?
>
> Examples are in Documentation/networking/netdevices.txt
Isn't this the same problem as above? The net_device structure must
stay around if there are still references to it and it does.
Or am I missing something?
Thanks,
Brandon
^ permalink raw reply
* Re: strange tcp behavior
From: Evgeniy Polyakov @ 2007-08-03 17:39 UTC (permalink / raw)
To: Simon Arlott; +Cc: john, netdev, David Miller
In-Reply-To: <46B35D1E.9050501@simon.arlott.org.uk>
On Fri, Aug 03, 2007 at 05:51:42PM +0100, Simon Arlott (simon@fire.lp0.eu) wrote:
> 17:38:03.533589 IP 192.168.7.4.50550 > 192.168.7.8.2500: R 82517592:82517592(0) win 1500 (raw)
> vs
> 17:37:38.383085 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 4259643274:4259643274(0) ack 1171836829 win 14360
> What happened there ?
You mean what will happend if second rst (4259643274) is close enough to
first (82517592) to reset the connection? If this will be session hijiking
attack first (known) implemented by Kevin Mitnik. So far things moved
forward and sequence number generation algorithm changed a lot.
It is the same situation, which would happen if you will spam remote
side with RST packets with arbitrary sequence number in hope that it
will reset some connection.
--
Evgeniy Polyakov
^ permalink raw reply
* [PATCH 12/13] dev->priv to netdev_priv(dev), for drivers/net/wan
From: Yoann Padioleau @ 2007-08-03 17:37 UTC (permalink / raw)
To: kernel-janitors; +Cc: jeff, netdev, akpm, linux-kernel
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)
Signed-off-by: Yoann Padioleau <padator@wanadoo.fr>
Cc: jeff@garzik.org
Cc: netdev@vger.kernel.org
Cc: akpm@linux-foundation.org
---
drivers/net/wan/dlci.c | 30 +++++++++++++-------------
drivers/net/wan/lmc/lmc_main.c | 26 +++++++++++------------
drivers/net/wan/sdla.c | 46 ++++++++++++++++++++---------------------
drivers/net/wan/sealevel.c | 12 +++++-----
drivers/net/wan/x25_asy.c | 28 ++++++++++++------------
5 files changed, 71 insertions(+), 71 deletions(-)
diff --git a/drivers/net/wan/dlci.c b/drivers/net/wan/dlci.c
index 66be20c..d988161 100644
--- a/drivers/net/wan/dlci.c
+++ b/drivers/net/wan/dlci.c
@@ -74,7 +74,7 @@ static int dlci_header(struct sk_buff *s
unsigned int hlen;
char *dest;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
hdr.control = FRAD_I_UI;
switch(type)
@@ -110,7 +110,7 @@ static void dlci_receive(struct sk_buff
struct frhdr *hdr;
int process, header;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
if (!pskb_may_pull(skb, sizeof(*hdr))) {
printk(KERN_NOTICE "%s: invalid data no header\n",
dev->name);
@@ -197,7 +197,7 @@ static int dlci_transmit(struct sk_buff
if (!skb || !dev)
return(0);
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
netif_stop_queue(dev);
@@ -235,7 +235,7 @@ static int dlci_config(struct net_device
struct frad_local *flp;
int err;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
flp = dlp->slave->priv;
@@ -269,7 +269,7 @@ static int dlci_dev_ioctl(struct net_dev
if (!capable(CAP_NET_ADMIN))
return(-EPERM);
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
switch(cmd)
{
@@ -298,7 +298,7 @@ static int dlci_change_mtu(struct net_de
{
struct dlci_local *dlp;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
return((*dlp->slave->change_mtu)(dlp->slave, new_mtu));
}
@@ -309,7 +309,7 @@ static int dlci_open(struct net_device *
struct frad_local *flp;
int err;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
if (!*(short *)(dev->dev_addr))
return(-EINVAL);
@@ -335,7 +335,7 @@ static int dlci_close(struct net_device
netif_stop_queue(dev);
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
flp = dlp->slave->priv;
err = (*flp->deactivate)(dlp->slave, dev);
@@ -347,7 +347,7 @@ static struct net_device_stats *dlci_get
{
struct dlci_local *dlp;
- dlp = dev->priv;
+ dlp = netdev_priv(dev);
return(&dlp->stats);
}
@@ -365,7 +365,7 @@ static int dlci_add(struct dlci_add *dlc
if (!slave)
return -ENODEV;
- if (slave->type != ARPHRD_FRAD || slave->priv == NULL)
+ if (slave->type != ARPHRD_FRAD || netdev_priv(slave) == NULL)
goto err1;
/* create device name */
@@ -391,11 +391,11 @@ static int dlci_add(struct dlci_add *dlc
*(short *)(master->dev_addr) = dlci->dlci;
- dlp = (struct dlci_local *) master->priv;
+ dlp = netdev_priv(master);
dlp->slave = slave;
dlp->master = master;
- flp = slave->priv;
+ flp = netdev_priv(slave);
err = (*flp->assoc)(slave, master);
if (err < 0)
goto err2;
@@ -435,9 +435,9 @@ static int dlci_del(struct dlci_add *dlc
return(-EBUSY);
}
- dlp = master->priv;
+ dlp = netdev_priv(master);
slave = dlp->slave;
- flp = slave->priv;
+ flp = netdev_priv(slave);
rtnl_lock();
err = (*flp->deassoc)(slave, master);
@@ -487,7 +487,7 @@ static int dlci_ioctl(unsigned int cmd,
static void dlci_setup(struct net_device *dev)
{
- struct dlci_local *dlp = dev->priv;
+ struct dlci_local *dlp = netdev_priv(dev);
dev->flags = 0;
dev->open = dlci_open;
diff --git a/drivers/net/wan/lmc/lmc_main.c b/drivers/net/wan/lmc/lmc_main.c
index ae132c1..9aa04a6 100644
--- a/drivers/net/wan/lmc/lmc_main.c
+++ b/drivers/net/wan/lmc/lmc_main.c
@@ -126,7 +126,7 @@ int lmc_ioctl (struct net_device *dev, s
ret = -EOPNOTSUPP;
- sc = dev->priv;
+ sc = netdev_priv(dev);
lmc_trace(dev, "lmc_ioctl in");
@@ -634,7 +634,7 @@ static void lmc_watchdog (unsigned long
u_int32_t ticks;
unsigned long flags;
- sc = dev->priv;
+ sc = netdev_priv(dev);
lmc_trace(dev, "lmc_watchdog in");
@@ -872,7 +872,7 @@ #endif
lmc_first_load = 1;
}
- sc = dev->priv;
+ sc = netdev_priv(dev);
sc->lmc_device = dev;
sc->name = dev->name;
@@ -1018,7 +1018,7 @@ static void __devexit lmc_remove_one (st
struct net_device *dev = pci_get_drvdata(pdev);
if (dev) {
- lmc_softc_t *sc = dev->priv;
+ lmc_softc_t *sc = netdev_priv(dev);
printk("%s: removing...\n", dev->name);
lmc_proto_detach(sc);
@@ -1035,7 +1035,7 @@ static void __devexit lmc_remove_one (st
*/
static int lmc_open (struct net_device *dev) /*fold00*/
{
- lmc_softc_t *sc = dev->priv;
+ lmc_softc_t *sc = netdev_priv(dev);
lmc_trace(dev, "lmc_open in");
@@ -1153,7 +1153,7 @@ static int lmc_open (struct net_device *
static void lmc_running_reset (struct net_device *dev) /*fold00*/
{
- lmc_softc_t *sc = (lmc_softc_t *) dev->priv;
+ lmc_softc_t *sc = netdev_priv(dev);
lmc_trace(dev, "lmc_runnig_reset in");
@@ -1194,7 +1194,7 @@ static int lmc_close (struct net_device
lmc_trace(dev, "lmc_close in");
- sc = dev->priv;
+ sc = netdev_priv(dev);
sc->lmc_ok = 0;
sc->lmc_media->set_link_status (sc, 0);
del_timer (&sc->timer);
@@ -1210,7 +1210,7 @@ static int lmc_close (struct net_device
/* When the interface goes down, this is called */
static int lmc_ifdown (struct net_device *dev) /*fold00*/
{
- lmc_softc_t *sc = dev->priv;
+ lmc_softc_t *sc = netdev_priv(dev);
u32 csr6;
int i;
@@ -1287,7 +1287,7 @@ static irqreturn_t lmc_interrupt (int ir
lmc_trace(dev, "lmc_interrupt in");
- sc = dev->priv;
+ sc = netdev_priv(dev);
spin_lock(&sc->lmc_lock);
@@ -1473,7 +1473,7 @@ static int lmc_start_xmit (struct sk_buf
lmc_trace(dev, "lmc_start_xmit in");
- sc = dev->priv;
+ sc = netdev_priv(dev);
spin_lock_irqsave(&sc->lmc_lock, flags);
@@ -1570,7 +1570,7 @@ static int lmc_rx (struct net_device *de
lmc_trace(dev, "lmc_rx in");
- sc = dev->priv;
+ sc = netdev_priv(dev);
lmc_led_on(sc, LMC_DS3_LED3);
@@ -1764,7 +1764,7 @@ skip_out_of_mem:
static struct net_device_stats *lmc_get_stats (struct net_device *dev) /*fold00*/
{
- lmc_softc_t *sc = dev->priv;
+ lmc_softc_t *sc = netdev_priv(dev);
unsigned long flags;
lmc_trace(dev, "lmc_get_stats in");
@@ -2145,7 +2145,7 @@ static void lmc_driver_timeout(struct ne
lmc_trace(dev, "lmc_driver_timeout in");
- sc = dev->priv;
+ sc = netdev_priv(dev);
spin_lock_irqsave(&sc->lmc_lock, flags);
diff --git a/drivers/net/wan/sdla.c b/drivers/net/wan/sdla.c
index 792e588..a2f1849 100644
--- a/drivers/net/wan/sdla.c
+++ b/drivers/net/wan/sdla.c
@@ -185,7 +185,7 @@ static void sdla_stop(struct net_device
{
struct frad_local *flp;
- flp = dev->priv;
+ flp = netdev_priv(dev);
switch(flp->type)
{
case SDLA_S502A:
@@ -212,7 +212,7 @@ static void sdla_start(struct net_device
{
struct frad_local *flp;
- flp = dev->priv;
+ flp = netdev_priv(dev);
switch(flp->type)
{
case SDLA_S502A:
@@ -432,7 +432,7 @@ static int sdla_cmd(struct net_device *d
int ret, waiting, len;
long window;
- flp = dev->priv;
+ flp = netdev_priv(dev);
window = flp->type == SDLA_S508 ? SDLA_508_CMD_BUF : SDLA_502_CMD_BUF;
cmd_buf = (struct sdla_cmd *)(dev->mem_start + (window & SDLA_ADDR_MASK));
ret = 0;
@@ -509,7 +509,7 @@ static int sdla_activate(struct net_devi
struct frad_local *flp;
int i;
- flp = slave->priv;
+ flp = netdev_priv(slave);
for(i=0;i<CONFIG_DLCI_MAX;i++)
if (flp->master[i] == master)
@@ -531,7 +531,7 @@ static int sdla_deactivate(struct net_de
struct frad_local *flp;
int i;
- flp = slave->priv;
+ flp = netdev_priv(slave);
for(i=0;i<CONFIG_DLCI_MAX;i++)
if (flp->master[i] == master)
@@ -556,7 +556,7 @@ static int sdla_assoc(struct net_device
if (master->type != ARPHRD_DLCI)
return(-EINVAL);
- flp = slave->priv;
+ flp = netdev_priv(slave);
for(i=0;i<CONFIG_DLCI_MAX;i++)
{
@@ -589,7 +589,7 @@ static int sdla_deassoc(struct net_devic
struct frad_local *flp;
int i;
- flp = slave->priv;
+ flp = netdev_priv(slave);
for(i=0;i<CONFIG_DLCI_MAX;i++)
if (flp->master[i] == master)
@@ -619,7 +619,7 @@ static int sdla_dlci_conf(struct net_dev
int i;
short len, ret;
- flp = slave->priv;
+ flp = netdev_priv(slave);
for(i=0;i<CONFIG_DLCI_MAX;i++)
if (flp->master[i] == master)
@@ -628,7 +628,7 @@ static int sdla_dlci_conf(struct net_dev
if (i == CONFIG_DLCI_MAX)
return(-ENODEV);
- dlp = master->priv;
+ dlp = netdev_priv(master);
ret = SDLA_RET_OK;
len = sizeof(struct dlci_conf);
@@ -659,7 +659,7 @@ static int sdla_transmit(struct sk_buff
unsigned long flags;
struct buf_entry *pbuf;
- flp = dev->priv;
+ flp = netdev_priv(dev);
ret = 0;
accept = 1;
@@ -755,7 +755,7 @@ static void sdla_receive(struct net_devi
int i=0, received, success, addr, buf_base, buf_top;
short dlci, len, len2, split;
- flp = dev->priv;
+ flp = netdev_priv(dev);
success = 1;
received = addr = buf_top = buf_base = 0;
len = dlci = 0;
@@ -860,7 +860,7 @@ static void sdla_receive(struct net_devi
if (success)
{
flp->stats.rx_packets++;
- dlp = master->priv;
+ dlp = netdev_priv(master);
(*dlp->receive)(skb, master);
}
@@ -924,7 +924,7 @@ static void sdla_poll(unsigned long devi
struct frad_local *flp;
dev = (struct net_device *) device;
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (sdla_byte(dev, SDLA_502_RCV_BUF))
sdla_receive(dev);
@@ -940,7 +940,7 @@ static int sdla_close(struct net_device
int len, i;
short dlcis[CONFIG_DLCI_MAX];
- flp = dev->priv;
+ flp = netdev_priv(dev);
len = 0;
for(i=0;i<CONFIG_DLCI_MAX;i++)
@@ -1001,7 +1001,7 @@ static int sdla_open(struct net_device *
int len, i;
char byte;
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (!flp->initialized)
return(-EPERM);
@@ -1098,7 +1098,7 @@ static int sdla_config(struct net_device
if (dev->type == 0xFFFF)
return(-EUNATCH);
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (!get)
{
@@ -1229,7 +1229,7 @@ static int sdla_reconfig(struct net_devi
struct conf_data data;
int i, len;
- flp = dev->priv;
+ flp = netdev_priv(dev);
len = 0;
for(i=0;i<CONFIG_DLCI_MAX;i++)
@@ -1254,7 +1254,7 @@ static int sdla_ioctl(struct net_device
if(!capable(CAP_NET_ADMIN))
return -EPERM;
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (!flp->initialized)
return(-EINVAL);
@@ -1320,7 +1320,7 @@ static int sdla_change_mtu(struct net_de
{
struct frad_local *flp;
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (netif_running(dev))
return(-EBUSY);
@@ -1337,7 +1337,7 @@ static int sdla_set_config(struct net_de
unsigned base;
int err = -EINVAL;
- flp = dev->priv;
+ flp = netdev_priv(dev);
if (flp->initialized)
return(-EINVAL);
@@ -1592,14 +1592,14 @@ fail:
static struct net_device_stats *sdla_stats(struct net_device *dev)
{
struct frad_local *flp;
- flp = dev->priv;
+ flp = netdev_priv(dev);
return(&flp->stats);
}
static void setup_sdla(struct net_device *dev)
{
- struct frad_local *flp = dev->priv;
+ struct frad_local *flp = netdev_priv(dev);
netdev_boot_setup_check(dev);
@@ -1651,7 +1651,7 @@ static int __init init_sdla(void)
static void __exit exit_sdla(void)
{
- struct frad_local *flp = sdla->priv;
+ struct frad_local *flp = netdev_priv(sdla);
unregister_netdev(sdla);
if (flp->initialized) {
diff --git a/drivers/net/wan/sealevel.c b/drivers/net/wan/sealevel.c
index 11276bf..4414901 100644
--- a/drivers/net/wan/sealevel.c
+++ b/drivers/net/wan/sealevel.c
@@ -77,7 +77,7 @@ static void sealevel_input(struct z8530_
static int sealevel_open(struct net_device *d)
{
- struct slvl_device *slvl=d->priv;
+ struct slvl_device *slvl=netdev_priv(d);
int err = -1;
int unit = slvl->channel;
@@ -126,7 +126,7 @@ static int sealevel_open(struct net_devi
static int sealevel_close(struct net_device *d)
{
- struct slvl_device *slvl=d->priv;
+ struct slvl_device *slvl=netdev_priv(d);
int unit = slvl->channel;
/*
@@ -166,7 +166,7 @@ static int sealevel_ioctl(struct net_dev
static struct net_device_stats *sealevel_get_stats(struct net_device *d)
{
- struct slvl_device *slvl=d->priv;
+ struct slvl_device *slvl=netdev_priv(d);
if(slvl)
return z8530_get_stats(slvl->chan);
else
@@ -179,7 +179,7 @@ static struct net_device_stats *sealevel
static int sealevel_queue_xmit(struct sk_buff *skb, struct net_device *d)
{
- struct slvl_device *slvl=d->priv;
+ struct slvl_device *slvl=netdev_priv(d);
return z8530_queue_xmit(slvl->chan, skb);
}
@@ -204,7 +204,7 @@ static int sealevel_neigh_setup_dev(stru
static int sealevel_attach(struct net_device *dev)
{
- struct slvl_device *sv = dev->priv;
+ struct slvl_device *sv = netdev_priv(dev);
sppp_attach(&sv->pppdev);
return 0;
}
@@ -240,7 +240,7 @@ static inline struct slvl_device *slvl_a
if (!d)
return NULL;
- sv = d->priv;
+ sv = netdev_priv(d);
sv->if_ptr = &sv->pppdev;
sv->pppdev.dev = d;
d->base_addr = iobase;
diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c
index c48b1cc..d62821d 100644
--- a/drivers/net/wan/x25_asy.c
+++ b/drivers/net/wan/x25_asy.c
@@ -63,7 +63,7 @@ static struct x25_asy *x25_asy_alloc(voi
if (dev == NULL)
break;
- sl = dev->priv;
+ sl = netdev_priv(dev);
/* Not in use ? */
if (!test_and_set_bit(SLF_INUSE, &sl->flags))
return sl;
@@ -85,7 +85,7 @@ static struct x25_asy *x25_asy_alloc(voi
return NULL;
/* Initialize channel control data */
- sl = dev->priv;
+ sl = netdev_priv(dev);
dev->base_addr = i;
/* register device so that it can be ifconfig'ed */
@@ -119,7 +119,7 @@ static void x25_asy_free(struct x25_asy
static int x25_asy_change_mtu(struct net_device *dev, int newmtu)
{
- struct x25_asy *sl = dev->priv;
+ struct x25_asy *sl = netdev_priv(dev);
unsigned char *xbuff, *rbuff;
int len = 2* newmtu;
@@ -282,7 +282,7 @@ static void x25_asy_write_wakeup(struct
static void x25_asy_timeout(struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
spin_lock(&sl->lock);
if (netif_queue_stopped(dev)) {
@@ -303,7 +303,7 @@ static void x25_asy_timeout(struct net_d
static int x25_asy_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
int err;
if (!netif_running(sl->dev)) {
@@ -372,7 +372,7 @@ static int x25_asy_data_indication(struc
static void x25_asy_data_transmit(struct net_device *dev, struct sk_buff *skb)
{
- struct x25_asy *sl=dev->priv;
+ struct x25_asy *sl=netdev_priv(dev);
spin_lock(&sl->lock);
if (netif_queue_stopped(sl->dev) || sl->tty == NULL)
@@ -399,7 +399,7 @@ static void x25_asy_data_transmit(struct
static void x25_asy_connected(struct net_device *dev, int reason)
{
- struct x25_asy *sl = dev->priv;
+ struct x25_asy *sl = netdev_priv(dev);
struct sk_buff *skb;
unsigned char *ptr;
@@ -418,7 +418,7 @@ static void x25_asy_connected(struct net
static void x25_asy_disconnected(struct net_device *dev, int reason)
{
- struct x25_asy *sl = dev->priv;
+ struct x25_asy *sl = netdev_priv(dev);
struct sk_buff *skb;
unsigned char *ptr;
@@ -449,7 +449,7 @@ static struct lapb_register_struct x25_a
/* Open the low-level part of the X.25 channel. Easy! */
static int x25_asy_open(struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
unsigned long len;
int err;
@@ -499,7 +499,7 @@ norbuff:
/* Close the low-level part of the X.25 channel. Easy! */
static int x25_asy_close(struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
int err;
spin_lock(&sl->lock);
@@ -615,7 +615,7 @@ static void x25_asy_close_tty(struct tty
static struct net_device_stats *x25_asy_get_stats(struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
return &sl->stats;
}
@@ -730,7 +730,7 @@ static int x25_asy_ioctl(struct tty_stru
static int x25_asy_open_dev(struct net_device *dev)
{
- struct x25_asy *sl = (struct x25_asy*)(dev->priv);
+ struct x25_asy *sl = (struct x25_asy*)(netdev_priv(dev));
if(sl->tty==NULL)
return -ENODEV;
return 0;
@@ -739,7 +739,7 @@ static int x25_asy_open_dev(struct net_d
/* Initialise the X.25 driver. Called by the device init code */
static void x25_asy_setup(struct net_device *dev)
{
- struct x25_asy *sl = dev->priv;
+ struct x25_asy *sl = netdev_priv(dev);
sl->magic = X25_ASY_MAGIC;
sl->dev = dev;
@@ -805,7 +805,7 @@ static void __exit exit_x25_asy(void)
for (i = 0; i < x25_asy_maxdev; i++) {
dev = x25_asy_devs[i];
if (dev) {
- struct x25_asy *sl = dev->priv;
+ struct x25_asy *sl = netdev_priv(dev);
spin_lock_bh(&sl->lock);
if (sl->tty)
^ permalink raw reply related
* [PATCH 09/13] dev->priv to netdev_priv(dev), for drivers/net/tokenring
From: Yoann Padioleau @ 2007-08-03 17:36 UTC (permalink / raw)
To: kernel-janitors; +Cc: jeff, netdev, akpm, linux-kernel
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)
Signed-off-by: Yoann Padioleau <padator@wanadoo.fr>
Cc: jeff@garzik.org
Cc: netdev@vger.kernel.org
Cc: akpm@linux-foundation.org
---
drivers/net/tokenring/3c359.c | 58 ++++++++++++++++++------------------
drivers/net/tokenring/ibmtr.c | 38 +++++++++++------------
drivers/net/tokenring/lanstreamer.c | 32 +++++++++----------
drivers/net/tokenring/madgemc.c | 4 +-
drivers/net/tokenring/olympic.c | 36 +++++++++++-----------
drivers/net/tokenring/tmspci.c | 4 +-
6 files changed, 86 insertions(+), 86 deletions(-)
diff --git a/drivers/net/tokenring/3c359.c b/drivers/net/tokenring/3c359.c
index 9f1b6ab..a8573da 100644
--- a/drivers/net/tokenring/3c359.c
+++ b/drivers/net/tokenring/3c359.c
@@ -156,7 +156,7 @@ static void print_rx_state(struct net_de
static void print_tx_state(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
struct xl_tx_desc *txd ;
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
int i ;
@@ -179,7 +179,7 @@ static void print_tx_state(struct net_de
static void print_rx_state(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
struct xl_rx_desc *rxd ;
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
int i ;
@@ -213,7 +213,7 @@ #endif
static u16 xl_ee_read(struct net_device *dev, int ee_addr)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
/* Wait for EEProm to not be busy */
@@ -245,7 +245,7 @@ static u16 xl_ee_read(struct net_device
static void xl_ee_write(struct net_device *dev, int ee_addr, u16 ee_value)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
/* Wait for EEProm to not be busy */
@@ -305,11 +305,11 @@ static int __devinit xl_probe(struct pci
pci_release_regions(pdev) ;
return -ENOMEM ;
}
- xl_priv = dev->priv ;
+ xl_priv = netdev_priv(dev) ;
#if XL_DEBUG
printk("pci_device: %p, dev:%p, dev->priv: %p, ba[0]: %10x, ba[1]:%10x\n",
- pdev, dev, dev->priv, (unsigned int)pdev->resource[0].start, (unsigned int)pdev->resource[1].start) ;
+ pdev, dev, netdev_priv(dev), (unsigned int)pdev->resource[0].start, (unsigned int)pdev->resource[1].start) ;
#endif
dev->irq=pdev->irq;
@@ -365,7 +365,7 @@ #endif
static int __devinit xl_init(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
printk(KERN_INFO "%s \n", version);
printk(KERN_INFO "%s: I/O at %hx, MMIO at %p, using irq %d\n",
@@ -385,7 +385,7 @@ static int __devinit xl_init(struct net_
static int xl_hw_reset(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
unsigned long t ;
u16 i ;
@@ -568,7 +568,7 @@ #endif
static int xl_open(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
u8 i ;
u16 hwaddr[3] ; /* Should be u8[6] but we get word return values */
@@ -726,7 +726,7 @@ static int xl_open(struct net_device *de
static int xl_open_hw(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
u8 __iomem *xl_mmio = xl_priv->xl_mmio ;
u16 vsoff ;
char ver_str[33];
@@ -875,7 +875,7 @@ static int xl_open_hw(struct net_device
static void adv_rx_ring(struct net_device *dev) /* Advance rx_ring, cut down on bloat in xl_rx */
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
int prev_ring_loc ;
prev_ring_loc = (xl_priv->rx_ring_tail + XL_RX_RING_SIZE - 1) & (XL_RX_RING_SIZE - 1);
@@ -890,7 +890,7 @@ static void adv_rx_ring(struct net_devic
static void xl_rx(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
struct sk_buff *skb, *skb2 ;
int frame_length = 0, copy_len = 0 ;
@@ -997,7 +997,7 @@ static void xl_rx(struct net_device *dev
static void xl_reset(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
unsigned long t;
@@ -1020,7 +1020,7 @@ static void xl_reset(struct net_device *
static void xl_freemem(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv=netdev_priv(dev) ;
int i ;
for (i=0;i<XL_RX_RING_SIZE;i++) {
@@ -1044,7 +1044,7 @@ static void xl_freemem(struct net_device
static irqreturn_t xl_interrupt(int irq, void *dev_id)
{
struct net_device *dev = (struct net_device *)dev_id;
- struct xl_private *xl_priv =(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv =netdev_priv(dev);
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
u16 intstatus, macstatus ;
@@ -1171,7 +1171,7 @@ static irqreturn_t xl_interrupt(int irq,
static int xl_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
struct xl_tx_desc *txd ;
int tx_head, tx_tail, tx_prev ;
unsigned long flags ;
@@ -1232,7 +1232,7 @@ static int xl_xmit(struct sk_buff *skb,
static void xl_dn_comp(struct net_device *dev)
{
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
struct xl_tx_desc *txd ;
@@ -1268,7 +1268,7 @@ static void xl_dn_comp(struct net_device
static int xl_close(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
unsigned long t ;
@@ -1366,7 +1366,7 @@ static int xl_close(struct net_device *d
static void xl_set_rx_mode(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
struct dev_mc_list *dmi ;
unsigned char dev_mc_address[4] ;
u16 options ;
@@ -1407,7 +1407,7 @@ static void xl_set_rx_mode(struct net_de
static void xl_srb_bh(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
u8 srb_cmd, ret_code ;
int i ;
@@ -1476,14 +1476,14 @@ static void xl_srb_bh(struct net_device
static struct net_device_stats * xl_get_stats(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
return (struct net_device_stats *) &xl_priv->xl_stats;
}
static int xl_set_mac_address (struct net_device *dev, void *addr)
{
struct sockaddr *saddr = addr ;
- struct xl_private *xl_priv = (struct xl_private *)dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
if (netif_running(dev)) {
printk(KERN_WARNING "%s: Cannot set mac/laa address while card is open\n", dev->name) ;
@@ -1504,7 +1504,7 @@ static int xl_set_mac_address (struct ne
static void xl_arb_cmd(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv;
+ struct xl_private *xl_priv = netdev_priv(dev);
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
u8 arb_cmd ;
u16 lan_status, lan_status_diff ;
@@ -1632,7 +1632,7 @@ #endif
static void xl_asb_cmd(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
if (xl_priv->asb_queued == 1)
@@ -1663,7 +1663,7 @@ static void xl_asb_cmd(struct net_device
*/
static void xl_asb_bh(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
u8 ret_code ;
@@ -1691,7 +1691,7 @@ static void xl_asb_bh(struct net_device
static void xl_srb_cmd(struct net_device *dev, int srb_cmd)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
switch (srb_cmd) {
@@ -1748,7 +1748,7 @@ static void xl_srb_cmd(struct net_device
static void xl_wait_misr_flags(struct net_device *dev)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv ;
+ struct xl_private *xl_priv = netdev_priv(dev) ;
u8 __iomem * xl_mmio = xl_priv->xl_mmio ;
int i ;
@@ -1773,7 +1773,7 @@ static void xl_wait_misr_flags(struct ne
static int xl_change_mtu(struct net_device *dev, int mtu)
{
- struct xl_private *xl_priv = (struct xl_private *) dev->priv;
+ struct xl_private *xl_priv = netdev_priv(dev);
u16 max_mtu ;
if (xl_priv->xl_ring_speed == 4)
@@ -1795,7 +1795,7 @@ static int xl_change_mtu(struct net_devi
static void __devexit xl_remove_one (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct xl_private *xl_priv=(struct xl_private *)dev->priv;
+ struct xl_private *xl_priv=netdev_priv(dev);
unregister_netdev(dev);
iounmap(xl_priv->xl_mmio) ;
diff --git a/drivers/net/tokenring/ibmtr.c b/drivers/net/tokenring/ibmtr.c
index 1e8958e..5140969 100644
--- a/drivers/net/tokenring/ibmtr.c
+++ b/drivers/net/tokenring/ibmtr.c
@@ -327,7 +327,7 @@ #ifndef PCMCIA
release_region(dev->base_addr, IBMTR_IO_EXTENT);
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
iounmap(ti->mmio);
iounmap(ti->sram_virt);
}
@@ -384,7 +384,7 @@ static int __devinit ibmtr_probe1(struct
unsigned char segment, intr=0, irq=0, i, j, cardpresent=NOTOK, temp=0;
void __iomem * t_mmio = NULL;
- struct tok_info *ti = dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
void __iomem *cd_chanid;
unsigned char *tchanid, ctemp;
#ifndef PCMCIA
@@ -823,7 +823,7 @@ static unsigned char __devinit get_sram_
static int __devinit trdev_init(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
SET_PAGE(ti->srb_page);
ti->open_failure = NO ;
@@ -846,7 +846,7 @@ static int tok_init_card(struct net_devi
unsigned long i;
PIOaddr = dev->base_addr;
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
/* Special processing for first interrupt after reset */
ti->do_tok_int = FIRST_INT;
/* Reset adapter */
@@ -868,7 +868,7 @@ #endif
/*****************************************************************************/
static int tok_open(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
int i;
/*the case we were left in a failure state during a previous open */
@@ -927,7 +927,7 @@ static void tok_open_adapter(unsigned lo
struct tok_info *ti;
int i;
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
SET_PAGE(ti->init_srb_page);
writeb(~SRB_RESP_INT, ti->mmio + ACA_OFFSET + ACA_RESET + ISRP_ODD);
for (i = 0; i < sizeof(struct dir_open_adapter); i++)
@@ -962,7 +962,7 @@ static void tok_open_adapter(unsigned lo
static void open_sap(unsigned char type, struct net_device *dev)
{
int i;
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
SET_PAGE(ti->srb_page);
for (i = 0; i < sizeof(struct dlc_open_sap); i++)
@@ -986,7 +986,7 @@ #define STATION_COUNT_OFST 18
static void tok_set_multicast_list(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
struct dev_mc_list *mclist;
unsigned char address[4];
@@ -1029,7 +1029,7 @@ static int tok_send_packet(struct sk_buf
{
struct tok_info *ti;
unsigned long flags;
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
netif_stop_queue(dev);
@@ -1051,7 +1051,7 @@ static int tok_send_packet(struct sk_buf
static int tok_close(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
/* Important for PCMCIA hot unplug, otherwise, we'll pull the card, */
/* unloading the module from memory, and then if a timer pops, ouch */
@@ -1094,7 +1094,7 @@ static void __iomem *map_address(struct
static void dir_open_adapter (struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
unsigned char ret_code;
__u16 err;
@@ -1179,7 +1179,7 @@ #endif
#if TR_VERBOSE
DPRINTK("Int from tok_driver, dev : %p irq%d\n", dev,irq);
#endif
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
if (ti->sram_phys & 1)
return IRQ_NONE; /* PCMCIA card extraction flag */
spin_lock(&(ti->lock));
@@ -1498,7 +1498,7 @@ static void initial_tok_int(struct net_d
struct tok_info *ti;
unsigned char init_status; /*BMS 12/2000*/
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
ti->do_tok_int = NOT_FIRST;
@@ -1560,7 +1560,7 @@ #define RSAP_VALUE_OFST 9
static void tr_tx(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
struct trh_hdr *trhdr = (struct trh_hdr *) ti->current_skb->data;
unsigned int hdr_len;
__u32 dhb=0,dhb_base;
@@ -1674,7 +1674,7 @@ #define ETHERTYPE_OFST 6
static void tr_rx(struct net_device *dev)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
__u32 rbuffer;
void __iomem *rbuf, *rbufdata, *llc;
__u8 rbuffer_page = 0;
@@ -1846,7 +1846,7 @@ static void ibmtr_reset_timer(struct tim
void tok_rerun(unsigned long dev_addr){
struct net_device *dev = (struct net_device *)dev_addr;
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
if ( ti->open_action == RESTART){
ti->do_tok_int = FIRST_INT;
@@ -1868,7 +1868,7 @@ static void ibmtr_readlog(struct net_dev
{
struct tok_info *ti;
- ti = (struct tok_info *) dev->priv;
+ ti = netdev_priv(dev);
ti->readlog_pending = 0;
SET_PAGE(ti->srb_page);
@@ -1891,7 +1891,7 @@ static struct net_device_stats *tok_get_
{
struct tok_info *toki;
- toki = (struct tok_info *) dev->priv;
+ toki = netdev_priv(dev);
return (struct net_device_stats *) &toki->tr_stats;
}
@@ -1899,7 +1899,7 @@ static struct net_device_stats *tok_get_
static int ibmtr_change_mtu(struct net_device *dev, int mtu)
{
- struct tok_info *ti = (struct tok_info *) dev->priv;
+ struct tok_info *ti = netdev_priv(dev);
if (ti->ring_speed == 16 && mtu > ti->maxmtu16)
return -EINVAL;
diff --git a/drivers/net/tokenring/lanstreamer.c b/drivers/net/tokenring/lanstreamer.c
index 5d849c0..3448634 100644
--- a/drivers/net/tokenring/lanstreamer.c
+++ b/drivers/net/tokenring/lanstreamer.c
@@ -245,7 +245,7 @@ #endif
}
SET_MODULE_OWNER(dev);
- streamer_priv = dev->priv;
+ streamer_priv = netdev_priv(dev);
#if STREAMER_NETWORK_MONITOR
#ifdef CONFIG_PROC_FS
@@ -404,7 +404,7 @@ #endif
return;
}
- streamer_priv=dev->priv;
+ streamer_priv=netdev_priv(dev);
if (streamer_priv == NULL) {
printk(KERN_ERR "lanstreamer::streamer_remove_one, ERROR dev->priv is NULL\n");
return;
@@ -448,7 +448,7 @@ static int streamer_reset(struct net_dev
struct sk_buff *skb = NULL;
__u16 misr;
- streamer_priv = (struct streamer_private *) dev->priv;
+ streamer_priv = netdev_priv(dev);
streamer_mmio = streamer_priv->streamer_mmio;
writew(readw(streamer_mmio + BCTL) | BCTL_SOFTRESET, streamer_mmio + BCTL);
@@ -587,7 +587,7 @@ #endif
static int streamer_open(struct net_device *dev)
{
- struct streamer_private *streamer_priv = (struct streamer_private *) dev->priv;
+ struct streamer_private *streamer_priv = netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags;
char open_error[255];
@@ -904,7 +904,7 @@ #endif
static void streamer_rx(struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ 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;
@@ -1029,7 +1029,7 @@ static irqreturn_t streamer_interrupt(in
{
struct net_device *dev = (struct net_device *) dev_id;
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u16 sisr;
__u16 misr;
@@ -1152,7 +1152,7 @@ static irqreturn_t streamer_interrupt(in
static int streamer_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags ;
@@ -1203,7 +1203,7 @@ #endif
static int streamer_close(struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
unsigned long flags;
int i;
@@ -1269,7 +1269,7 @@ #endif
static void streamer_set_rx_mode(struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u8 options = 0;
struct dev_mc_list *dmi;
@@ -1328,7 +1328,7 @@ static void streamer_set_rx_mode(struct
static void streamer_srb_bh(struct net_device *dev)
{
- struct streamer_private *streamer_priv = (struct streamer_private *) dev->priv;
+ struct streamer_private *streamer_priv = netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u16 srb_word;
@@ -1493,14 +1493,14 @@ static void streamer_srb_bh(struct net_d
static struct net_device_stats *streamer_get_stats(struct net_device *dev)
{
struct streamer_private *streamer_priv;
- streamer_priv = (struct streamer_private *) dev->priv;
+ streamer_priv = netdev_priv(dev);
return (struct net_device_stats *) &streamer_priv->streamer_stats;
}
static int streamer_set_mac_address(struct net_device *dev, void *addr)
{
struct sockaddr *saddr = addr;
- struct streamer_private *streamer_priv = (struct streamer_private *) dev->priv;
+ struct streamer_private *streamer_priv = netdev_priv(dev);
if (netif_running(dev))
{
@@ -1525,7 +1525,7 @@ static int streamer_set_mac_address(stru
static void streamer_arb_cmd(struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
__u8 header_len;
__u16 frame_len, buffer_len;
@@ -1740,7 +1740,7 @@ drop_frame:
static void streamer_asb_bh(struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
if (streamer_priv->asb_queued == 1)
@@ -1784,7 +1784,7 @@ static void streamer_asb_bh(struct net_d
static int streamer_change_mtu(struct net_device *dev, int mtu)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u16 max_mtu;
if (streamer_priv->streamer_ring_speed == 4)
@@ -1848,7 +1848,7 @@ static int streamer_proc_info(char *buff
static int sprintf_info(char *buffer, struct net_device *dev)
{
struct streamer_private *streamer_priv =
- (struct streamer_private *) dev->priv;
+ netdev_priv(dev);
__u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
struct streamer_adapter_addr_table sat;
struct streamer_parameters_table spt;
diff --git a/drivers/net/tokenring/madgemc.c b/drivers/net/tokenring/madgemc.c
index f8f4d74..9eafc2e 100644
--- a/drivers/net/tokenring/madgemc.c
+++ b/drivers/net/tokenring/madgemc.c
@@ -690,7 +690,7 @@ static int madgemc_close(struct net_devi
static int madgemc_mcaproc(char *buf, int slot, void *d)
{
struct net_device *dev = (struct net_device *)d;
- struct net_local *tp = dev->priv;
+ struct net_local *tp = netdev_priv(dev);
struct card_info *curcard = tp->tmspriv;
int len = 0;
@@ -736,7 +736,7 @@ static int __devexit madgemc_remove(stru
BUG_ON(!dev);
- tp = dev->priv;
+ tp = netdev_priv(dev);
card = tp->tmspriv;
kfree(card);
tp->tmspriv = NULL;
diff --git a/drivers/net/tokenring/olympic.c b/drivers/net/tokenring/olympic.c
index 09b3cfb..a96784e 100644
--- a/drivers/net/tokenring/olympic.c
+++ b/drivers/net/tokenring/olympic.c
@@ -219,14 +219,14 @@ static int __devinit olympic_probe(struc
goto op_release_dev;
}
- olympic_priv = dev->priv ;
+ olympic_priv = netdev_priv(dev) ;
spin_lock_init(&olympic_priv->olympic_lock) ;
init_waitqueue_head(&olympic_priv->srb_wait);
init_waitqueue_head(&olympic_priv->trb_wait);
#if OLYMPIC_DEBUG
- printk(KERN_INFO "pci_device: %p, dev:%p, dev->priv: %p\n", pdev, dev, dev->priv);
+ printk(KERN_INFO "pci_device: %p, dev:%p, dev->priv: %p\n", pdev, dev, netdev_priv(dev));
#endif
dev->irq=pdev->irq;
dev->base_addr=pci_resource_start(pdev, 0);
@@ -297,7 +297,7 @@ static int __devinit olympic_init(struct
unsigned long t;
unsigned int uaa_addr;
- olympic_priv=(struct olympic_private *)dev->priv;
+ olympic_priv=netdev_priv(dev);
olympic_mmio=olympic_priv->olympic_mmio;
printk("%s \n", version);
@@ -435,7 +435,7 @@ #endif
static int olympic_open(struct net_device *dev)
{
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio,*init_srb;
unsigned long flags, t;
int i, open_finished = 1 ;
@@ -755,7 +755,7 @@ out:
*/
static void olympic_rx(struct net_device *dev)
{
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio;
struct olympic_rx_status *rx_status;
struct olympic_rx_desc *rx_desc ;
@@ -897,7 +897,7 @@ #endif
static void olympic_freemem(struct net_device *dev)
{
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
int i;
for(i=0;i<OLYMPIC_RX_RING_SIZE;i++) {
@@ -930,7 +930,7 @@ static void olympic_freemem(struct net_d
static irqreturn_t olympic_interrupt(int irq, void *dev_id)
{
struct net_device *dev= (struct net_device *)dev_id;
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio;
u32 sisr;
u8 __iomem *adapter_check_area ;
@@ -1046,7 +1046,7 @@ static irqreturn_t olympic_interrupt(int
static int olympic_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio;
unsigned long flags ;
@@ -1077,7 +1077,7 @@ static int olympic_xmit(struct sk_buff *
static int olympic_close(struct net_device *dev)
{
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio,*srb;
unsigned long t,flags;
@@ -1147,7 +1147,7 @@ #endif
static void olympic_set_rx_mode(struct net_device *dev)
{
- struct olympic_private *olympic_priv = (struct olympic_private *) dev->priv ;
+ struct olympic_private *olympic_priv = netdev_priv(dev) ;
u8 __iomem *olympic_mmio = olympic_priv->olympic_mmio ;
u8 options = 0;
u8 __iomem *srb;
@@ -1215,7 +1215,7 @@ static void olympic_set_rx_mode(struct n
static void olympic_srb_bh(struct net_device *dev)
{
- struct olympic_private *olympic_priv = (struct olympic_private *) dev->priv ;
+ struct olympic_private *olympic_priv = netdev_priv(dev) ;
u8 __iomem *olympic_mmio = olympic_priv->olympic_mmio ;
u8 __iomem *srb;
@@ -1361,14 +1361,14 @@ static void olympic_srb_bh(struct net_de
static struct net_device_stats * olympic_get_stats(struct net_device *dev)
{
struct olympic_private *olympic_priv ;
- olympic_priv=(struct olympic_private *) dev->priv;
+ olympic_priv=netdev_priv(dev);
return (struct net_device_stats *) &olympic_priv->olympic_stats;
}
static int olympic_set_mac_address (struct net_device *dev, void *addr)
{
struct sockaddr *saddr = addr ;
- struct olympic_private *olympic_priv = (struct olympic_private *)dev->priv ;
+ struct olympic_private *olympic_priv = netdev_priv(dev) ;
if (netif_running(dev)) {
printk(KERN_WARNING "%s: Cannot set mac/laa address while card is open\n", dev->name) ;
@@ -1389,7 +1389,7 @@ static int olympic_set_mac_address (stru
static void olympic_arb_cmd(struct net_device *dev)
{
- struct olympic_private *olympic_priv = (struct olympic_private *) dev->priv;
+ struct olympic_private *olympic_priv = netdev_priv(dev);
u8 __iomem *olympic_mmio=olympic_priv->olympic_mmio;
u8 __iomem *arb_block, *asb_block, *srb ;
u8 header_len ;
@@ -1575,7 +1575,7 @@ drop_frame:
static void olympic_asb_bh(struct net_device *dev)
{
- struct olympic_private *olympic_priv = (struct olympic_private *) dev->priv ;
+ struct olympic_private *olympic_priv = netdev_priv(dev) ;
u8 __iomem *arb_block, *asb_block ;
arb_block = (olympic_priv->olympic_lap + olympic_priv->arb) ;
@@ -1615,7 +1615,7 @@ static void olympic_asb_bh(struct net_de
static int olympic_change_mtu(struct net_device *dev, int mtu)
{
- struct olympic_private *olympic_priv = (struct olympic_private *) dev->priv;
+ struct olympic_private *olympic_priv = netdev_priv(dev);
u16 max_mtu ;
if (olympic_priv->olympic_ring_speed == 4)
@@ -1637,7 +1637,7 @@ static int olympic_change_mtu(struct net
static int olympic_proc_info(char *buffer, char **start, off_t offset, int length, int *eof, void *data)
{
struct net_device *dev = (struct net_device *)data ;
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
u8 __iomem *oat = (olympic_priv->olympic_lap + olympic_priv->olympic_addr_table_addr) ;
u8 __iomem *opt = (olympic_priv->olympic_lap + olympic_priv->olympic_parms_addr) ;
int size = 0 ;
@@ -1748,7 +1748,7 @@ static int olympic_proc_info(char *buffe
static void __devexit olympic_remove_one(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev) ;
- struct olympic_private *olympic_priv=(struct olympic_private *)dev->priv;
+ struct olympic_private *olympic_priv=netdev_priv(dev);
if (olympic_priv->olympic_network_monitor) {
char proc_name[20] ;
diff --git a/drivers/net/tokenring/tmspci.c b/drivers/net/tokenring/tmspci.c
index 3b2f00b..aec75c9 100644
--- a/drivers/net/tokenring/tmspci.c
+++ b/drivers/net/tokenring/tmspci.c
@@ -149,7 +149,7 @@ static int __devinit tms_pci_attach(stru
goto err_out_irq;
}
- tp = dev->priv;
+ tp = netdev_priv(dev);
tp->setnselout = tms_pci_setnselout_pins;
tp->sifreadb = tms_pci_sifreadb;
@@ -210,7 +210,7 @@ static void tms_pci_read_eeprom(struct n
static unsigned short tms_pci_setnselout_pins(struct net_device *dev)
{
unsigned short val = 0;
- struct net_local *tp = dev->priv;
+ struct net_local *tp = netdev_priv(dev);
struct card_info *cardinfo = tp->tmspriv;
if(tp->DataRate == SPEED_4)
^ permalink raw reply related
* [PATCH 03/13] dev->priv to netdev_priv(dev), for drivers/net/arcnet
From: Yoann Padioleau @ 2007-08-03 17:33 UTC (permalink / raw)
To: kernel-janitors; +Cc: jeff, netdev, akpm, linux-kernel
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)
Signed-off-by: Yoann Padioleau <padator@wanadoo.fr>
Cc: jeff@garzik.org
Cc: netdev@vger.kernel.org
Cc: akpm@linux-foundation.org
---
drivers/net/arcnet/arc-rawmode.c | 4 ++--
drivers/net/arcnet/arc-rimi.c | 16 ++++++++--------
drivers/net/arcnet/arcnet.c | 32 ++++++++++++++++----------------
drivers/net/arcnet/capmode.c | 6 +++---
drivers/net/arcnet/com20020-isa.c | 4 ++--
drivers/net/arcnet/com20020-pci.c | 2 +-
drivers/net/arcnet/com20020.c | 10 +++++-----
drivers/net/arcnet/com90io.c | 4 ++--
drivers/net/arcnet/com90xx.c | 10 +++++-----
drivers/net/arcnet/rfc1051.c | 8 ++++----
drivers/net/arcnet/rfc1201.c | 12 ++++++------
11 files changed, 54 insertions(+), 54 deletions(-)
diff --git a/drivers/net/arcnet/arc-rawmode.c b/drivers/net/arcnet/arc-rawmode.c
index e0a18e7..909ba77 100644
--- a/drivers/net/arcnet/arc-rawmode.c
+++ b/drivers/net/arcnet/arc-rawmode.c
@@ -87,7 +87,7 @@ MODULE_LICENSE("GPL");
static void rx(struct net_device *dev, int bufnum,
struct archdr *pkthdr, int length)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct sk_buff *skb;
struct archdr *pkt = pkthdr;
int ofs;
@@ -168,7 +168,7 @@ static int build_header(struct sk_buff *
static int prepare_tx(struct net_device *dev, struct archdr *pkt, int length,
int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct arc_hardware *hard = &pkt->hard;
int ofs;
diff --git a/drivers/net/arcnet/arc-rimi.c b/drivers/net/arcnet/arc-rimi.c
index 8c8d6c4..e3082a9 100644
--- a/drivers/net/arcnet/arc-rimi.c
+++ b/drivers/net/arcnet/arc-rimi.c
@@ -194,7 +194,7 @@ static int __init arcrimi_found(struct n
/* initialize the rest of the device structure. */
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->card_name = "RIM I";
lp->hw.command = arcrimi_command;
lp->hw.status = arcrimi_status;
@@ -260,7 +260,7 @@ err_free_irq:
*/
static int arcrimi_reset(struct net_device *dev, int really_reset)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->mem_start + 0x800;
BUGMSG(D_INIT, "Resetting %s (status=%02Xh)\n", dev->name, ASTATUS());
@@ -281,7 +281,7 @@ static int arcrimi_reset(struct net_devi
static void arcrimi_setmask(struct net_device *dev, int mask)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->mem_start + 0x800;
AINTMASK(mask);
@@ -289,7 +289,7 @@ static void arcrimi_setmask(struct net_d
static int arcrimi_status(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->mem_start + 0x800;
return ASTATUS();
@@ -297,7 +297,7 @@ static int arcrimi_status(struct net_dev
static void arcrimi_command(struct net_device *dev, int cmd)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->mem_start + 0x800;
ACOMMAND(cmd);
@@ -306,7 +306,7 @@ static void arcrimi_command(struct net_d
static void arcrimi_copy_to_card(struct net_device *dev, int bufnum, int offset,
void *buf, int count)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *memaddr = lp->mem_start + 0x800 + bufnum * 512 + offset;
TIME("memcpy_toio", count, memcpy_toio(memaddr, buf, count));
}
@@ -315,7 +315,7 @@ static void arcrimi_copy_to_card(struct
static void arcrimi_copy_from_card(struct net_device *dev, int bufnum, int offset,
void *buf, int count)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *memaddr = lp->mem_start + 0x800 + bufnum * 512 + offset;
TIME("memcpy_fromio", count, memcpy_fromio(buf, memaddr, count));
}
@@ -361,7 +361,7 @@ static int __init arc_rimi_init(void)
static void __exit arc_rimi_exit(void)
{
struct net_device *dev = my_dev;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
unregister_netdev(dev);
iounmap(lp->mem_start);
diff --git a/drivers/net/arcnet/arcnet.c b/drivers/net/arcnet/arcnet.c
index 681e20b..230b545 100644
--- a/drivers/net/arcnet/arcnet.c
+++ b/drivers/net/arcnet/arcnet.c
@@ -181,7 +181,7 @@ #if (ARCNET_DEBUG_MAX & (D_RX | D_TX))
static void arcnet_dump_packet(struct net_device *dev, int bufnum,
char *desc, int take_arcnet_lock)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int i, length;
unsigned long flags = 0;
static uint8_t buf[512];
@@ -247,7 +247,7 @@ void arcnet_unregister_proto(struct ArcP
*/
static void release_arcbuf(struct net_device *dev, int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int i;
lp->buf_queue[lp->first_free_buf++] = bufnum;
@@ -269,7 +269,7 @@ static void release_arcbuf(struct net_de
*/
static int get_arcbuf(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int buf = -1, i;
if (!atomic_dec_and_test(&lp->buf_lock)) {
@@ -353,7 +353,7 @@ struct net_device *alloc_arcdev(char *na
dev = alloc_netdev(sizeof(struct arcnet_local),
name && *name ? name : "arc%d", arcdev_setup);
if(dev) {
- struct arcnet_local *lp = (struct arcnet_local *) dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
spin_lock_init(&lp->lock);
}
@@ -370,7 +370,7 @@ struct net_device *alloc_arcdev(char *na
*/
static int arcnet_open(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int count, newmtu, error;
BUGMSG(D_INIT,"opened.");
@@ -470,7 +470,7 @@ static int arcnet_open(struct net_device
/* The inverse routine to arcnet_open - shuts down the card. */
static int arcnet_close(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
netif_stop_queue(dev);
@@ -491,7 +491,7 @@ static int arcnet_header(struct sk_buff
unsigned short type, void *daddr, void *saddr,
unsigned len)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
uint8_t _daddr, proto_num;
struct ArcProto *proto;
@@ -552,7 +552,7 @@ static int arcnet_header(struct sk_buff
static int arcnet_rebuild_header(struct sk_buff *skb)
{
struct net_device *dev = skb->dev;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int status = 0; /* default is failure */
unsigned short type;
uint8_t daddr=0;
@@ -599,7 +599,7 @@ #endif
/* Called by the kernel in order to transmit a packet. */
static int arcnet_send_packet(struct sk_buff *skb, struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct archdr *pkt;
struct arc_rfc1201 *soft;
struct ArcProto *proto;
@@ -689,7 +689,7 @@ static int arcnet_send_packet(struct sk_
*/
static int go_tx(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
BUGMSG(D_DURING, "go_tx: status=%Xh, intmask=%Xh, next_tx=%d, cur_tx=%d\n",
ASTATUS(), lp->intmask, lp->next_tx, lp->cur_tx);
@@ -719,7 +719,7 @@ static int go_tx(struct net_device *dev)
static void arcnet_timeout(struct net_device *dev)
{
unsigned long flags;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int status = ASTATUS();
char *msg;
@@ -768,7 +768,7 @@ irqreturn_t arcnet_interrupt(int irq, vo
BUGMSG(D_DURING, "in arcnet_interrupt\n");
- lp = dev->priv;
+ lp = netdev_priv(dev);
BUG_ON(!lp);
spin_lock(&lp->lock);
@@ -1005,7 +1005,7 @@ irqreturn_t arcnet_interrupt(int irq, vo
*/
static void arcnet_rx(struct net_device *dev, int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct archdr pkt;
struct arc_rfc1201 *soft;
int length, ofs;
@@ -1069,7 +1069,7 @@ static void arcnet_rx(struct net_device
*/
static struct net_device_stats *arcnet_get_stats(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
return &lp->stats;
}
@@ -1086,7 +1086,7 @@ static void null_rx(struct net_device *d
static int null_build_header(struct sk_buff *skb, struct net_device *dev,
unsigned short type, uint8_t daddr)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
BUGMSG(D_PROTO,
"tx: can't build header for encap %02Xh; load a protocol driver.\n",
@@ -1101,7 +1101,7 @@ static int null_build_header(struct sk_b
static int null_prepare_tx(struct net_device *dev, struct archdr *pkt,
int length, int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct arc_hardware newpkt;
BUGMSG(D_PROTO, "tx: no encap for this host; load a protocol driver.\n");
diff --git a/drivers/net/arcnet/capmode.c b/drivers/net/arcnet/capmode.c
index cc4610d..bd04be9 100644
--- a/drivers/net/arcnet/capmode.c
+++ b/drivers/net/arcnet/capmode.c
@@ -101,7 +101,7 @@ #endif /* MODULE */
static void rx(struct net_device *dev, int bufnum,
struct archdr *pkthdr, int length)
{
- struct arcnet_local *lp = (struct arcnet_local *) dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct sk_buff *skb;
struct archdr *pkt = pkthdr;
char *pktbuf, *pkthdrbuf;
@@ -196,7 +196,7 @@ static int build_header(struct sk_buff *
static int prepare_tx(struct net_device *dev, struct archdr *pkt, int length,
int bufnum)
{
- struct arcnet_local *lp = (struct arcnet_local *) dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct arc_hardware *hard = &pkt->hard;
int ofs;
@@ -248,7 +248,7 @@ static int prepare_tx(struct net_device
static int ack_tx(struct net_device *dev, int acked)
{
- struct arcnet_local *lp = (struct arcnet_local *) dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct sk_buff *ackskb;
struct archdr *ackpkt;
int length=sizeof(struct arc_cap);
diff --git a/drivers/net/arcnet/com20020-isa.c b/drivers/net/arcnet/com20020-isa.c
index 9289e61..ea53a94 100644
--- a/drivers/net/arcnet/com20020-isa.c
+++ b/drivers/net/arcnet/com20020-isa.c
@@ -52,7 +52,7 @@ static int __init com20020isa_probe(stru
{
int ioaddr;
unsigned long airqmask;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int err;
BUGLVL(D_NORMAL) printk(VERSION);
@@ -151,7 +151,7 @@ static int __init com20020_init(void)
if (node && node != 0xff)
dev->dev_addr[0] = node;
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->backplane = backplane;
lp->clockp = clockp & 7;
lp->clockm = clockm & 3;
diff --git a/drivers/net/arcnet/com20020-pci.c b/drivers/net/arcnet/com20020-pci.c
index b8c0fa6..8b51f63 100644
--- a/drivers/net/arcnet/com20020-pci.c
+++ b/drivers/net/arcnet/com20020-pci.c
@@ -72,7 +72,7 @@ static int __devinit com20020pci_probe(s
dev = alloc_arcdev(device);
if (!dev)
return -ENOMEM;
- lp = dev->priv;
+ lp = netdev_priv(dev);
pci_set_drvdata(pdev, dev);
diff --git a/drivers/net/arcnet/com20020.c b/drivers/net/arcnet/com20020.c
index 7cf0a25..dd02463 100644
--- a/drivers/net/arcnet/com20020.c
+++ b/drivers/net/arcnet/com20020.c
@@ -89,7 +89,7 @@ static void com20020_copy_to_card(struct
int com20020_check(struct net_device *dev)
{
int ioaddr = dev->base_addr, status;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
ARCRESET0;
mdelay(RESETtime);
@@ -159,7 +159,7 @@ int com20020_found(struct net_device *de
/* Initialize the rest of the device structure. */
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->hw.owner = THIS_MODULE;
lp->hw.command = com20020_command;
@@ -233,7 +233,7 @@ int com20020_found(struct net_device *de
*/
static int com20020_reset(struct net_device *dev, int really_reset)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
u_int ioaddr = dev->base_addr;
u_char inbyte;
@@ -300,7 +300,7 @@ static int com20020_status(struct net_de
static void com20020_close(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int ioaddr = dev->base_addr;
/* disable transmitter */
@@ -317,7 +317,7 @@ static void com20020_close(struct net_de
*/
static void com20020_set_mc_list(struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int ioaddr = dev->base_addr;
if ((dev->flags & IFF_PROMISC) && (dev->flags & IFF_UP)) { /* Enable promiscuous mode */
diff --git a/drivers/net/arcnet/com90io.c b/drivers/net/arcnet/com90io.c
index 1f03027..c6e373a 100644
--- a/drivers/net/arcnet/com90io.c
+++ b/drivers/net/arcnet/com90io.c
@@ -248,7 +248,7 @@ static int __init com90io_found(struct n
return -EBUSY;
}
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->card_name = "COM90xx I/O";
lp->hw.command = com90io_command;
lp->hw.status = com90io_status;
@@ -290,7 +290,7 @@ static int __init com90io_found(struct n
*/
static int com90io_reset(struct net_device *dev, int really_reset)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
short ioaddr = dev->base_addr;
BUGMSG(D_INIT, "Resetting %s (status=%02Xh)\n", dev->name, ASTATUS());
diff --git a/drivers/net/arcnet/com90xx.c b/drivers/net/arcnet/com90xx.c
index 0d45553..f4113d2 100644
--- a/drivers/net/arcnet/com90xx.c
+++ b/drivers/net/arcnet/com90xx.c
@@ -468,7 +468,7 @@ static int __init com90xx_found(int ioad
release_mem_region(shmem, MIRROR_SIZE);
return -ENOMEM;
}
- lp = dev->priv;
+ lp = netdev_priv(dev);
/* find the real shared memory start/end points, including mirrors */
/* guess the actual size of one "memory mirror" - the number of
@@ -585,7 +585,7 @@ static void com90xx_setmask(struct net_d
*/
int com90xx_reset(struct net_device *dev, int really_reset)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
short ioaddr = dev->base_addr;
BUGMSG(D_INIT, "Resetting (status=%02Xh)\n", ASTATUS());
@@ -621,7 +621,7 @@ int com90xx_reset(struct net_device *dev
static void com90xx_copy_to_card(struct net_device *dev, int bufnum, int offset,
void *buf, int count)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *memaddr = lp->mem_start + bufnum * 512 + offset;
TIME("memcpy_toio", count, memcpy_toio(memaddr, buf, count));
}
@@ -630,7 +630,7 @@ static void com90xx_copy_to_card(struct
static void com90xx_copy_from_card(struct net_device *dev, int bufnum, int offset,
void *buf, int count)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
void __iomem *memaddr = lp->mem_start + bufnum * 512 + offset;
TIME("memcpy_fromio", count, memcpy_fromio(buf, memaddr, count));
}
@@ -656,7 +656,7 @@ static void __exit com90xx_exit(void)
for (count = 0; count < numcards; count++) {
dev = cards[count];
- lp = dev->priv;
+ lp = netdev_priv(dev);
unregister_netdev(dev);
free_irq(dev->irq, dev);
diff --git a/drivers/net/arcnet/rfc1051.c b/drivers/net/arcnet/rfc1051.c
index 2de8877..ba58089 100644
--- a/drivers/net/arcnet/rfc1051.c
+++ b/drivers/net/arcnet/rfc1051.c
@@ -88,7 +88,7 @@ MODULE_LICENSE("GPL");
*/
static unsigned short type_trans(struct sk_buff *skb, struct net_device *dev)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct archdr *pkt = (struct archdr *) skb->data;
struct arc_rfc1051 *soft = &pkt->soft.rfc1051;
int hdr_size = ARC_HDR_SIZE + RFC1051_HDR_SIZE;
@@ -125,7 +125,7 @@ static unsigned short type_trans(struct
static void rx(struct net_device *dev, int bufnum,
struct archdr *pkthdr, int length)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct sk_buff *skb;
struct archdr *pkt = pkthdr;
int ofs;
@@ -169,7 +169,7 @@ static void rx(struct net_device *dev, i
static int build_header(struct sk_buff *skb, struct net_device *dev,
unsigned short type, uint8_t daddr)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int hdr_size = ARC_HDR_SIZE + RFC1051_HDR_SIZE;
struct archdr *pkt = (struct archdr *) skb_push(skb, hdr_size);
struct arc_rfc1051 *soft = &pkt->soft.rfc1051;
@@ -220,7 +220,7 @@ static int build_header(struct sk_buff *
static int prepare_tx(struct net_device *dev, struct archdr *pkt, int length,
int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct arc_hardware *hard = &pkt->hard;
int ofs;
diff --git a/drivers/net/arcnet/rfc1201.c b/drivers/net/arcnet/rfc1201.c
index 460a095..75998f4 100644
--- a/drivers/net/arcnet/rfc1201.c
+++ b/drivers/net/arcnet/rfc1201.c
@@ -92,7 +92,7 @@ static unsigned short type_trans(struct
{
struct archdr *pkt = (struct archdr *) skb->data;
struct arc_rfc1201 *soft = &pkt->soft.rfc1201;
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int hdr_size = ARC_HDR_SIZE + RFC1201_HDR_SIZE;
/* Pull off the arcnet header. */
@@ -134,7 +134,7 @@ static unsigned short type_trans(struct
static void rx(struct net_device *dev, int bufnum,
struct archdr *pkthdr, int length)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct sk_buff *skb;
struct archdr *pkt = pkthdr;
struct arc_rfc1201 *soft = &pkthdr->soft.rfc1201;
@@ -376,7 +376,7 @@ static void rx(struct net_device *dev, i
static int build_header(struct sk_buff *skb, struct net_device *dev,
unsigned short type, uint8_t daddr)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int hdr_size = ARC_HDR_SIZE + RFC1201_HDR_SIZE;
struct archdr *pkt = (struct archdr *) skb_push(skb, hdr_size);
struct arc_rfc1201 *soft = &pkt->soft.rfc1201;
@@ -443,7 +443,7 @@ static int build_header(struct sk_buff *
static void load_pkt(struct net_device *dev, struct arc_hardware *hard,
struct arc_rfc1201 *soft, int softlen, int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
int ofs;
/* assume length <= XMTU: someone should have handled that by now. */
@@ -476,7 +476,7 @@ static void load_pkt(struct net_device *
static int prepare_tx(struct net_device *dev, struct archdr *pkt, int length,
int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
const int maxsegsize = XMTU - RFC1201_HDR_SIZE;
struct Outgoing *out;
@@ -511,7 +511,7 @@ static int prepare_tx(struct net_device
static int continue_tx(struct net_device *dev, int bufnum)
{
- struct arcnet_local *lp = dev->priv;
+ struct arcnet_local *lp = netdev_priv(dev);
struct Outgoing *out = &lp->outgoing;
struct arc_hardware *hard = &out->pkt->hard;
struct arc_rfc1201 *soft = &out->pkt->soft.rfc1201, *newsoft;
^ permalink raw reply related
* [PATCH 02/13] dev->priv to netdev_priv(dev), for drivers/net/appletalk
From: Yoann Padioleau @ 2007-08-03 17:33 UTC (permalink / raw)
To: kernel-janitors; +Cc: jeff, netdev, akpm, linux-kernel
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)
Signed-off-by: Yoann Padioleau <padator@wanadoo.fr>
Cc: jeff@garzik.org
Cc: netdev@vger.kernel.org
Cc: akpm@linux-foundation.org
---
drivers/net/appletalk/ipddp.c | 6 +++---
drivers/net/appletalk/ltpc.c | 8 ++++----
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c
index f22e46d..61add0e 100644
--- a/drivers/net/appletalk/ipddp.c
+++ b/drivers/net/appletalk/ipddp.c
@@ -109,7 +109,7 @@ static struct net_device * __init ipddp_
*/
static struct net_device_stats *ipddp_get_stats(struct net_device *dev)
{
- return dev->priv;
+ return netdev_priv(dev);
}
/*
@@ -171,8 +171,8 @@ static int ipddp_xmit(struct sk_buff *sk
skb->protocol = htons(ETH_P_ATALK); /* Protocol has changed */
- ((struct net_device_stats *) dev->priv)->tx_packets++;
- ((struct net_device_stats *) dev->priv)->tx_bytes+=skb->len;
+ ((struct net_device_stats *)netdev_priv(dev))->tx_packets++;
+ ((struct net_device_stats *)netdev_priv(dev))->tx_bytes+=skb->len;
if(aarp_send_ddp(rt->dev, skb, &rt->at, NULL) < 0)
dev_kfree_skb(skb);
diff --git a/drivers/net/appletalk/ltpc.c b/drivers/net/appletalk/ltpc.c
index 6a6cbd3..be12c6b 100644
--- a/drivers/net/appletalk/ltpc.c
+++ b/drivers/net/appletalk/ltpc.c
@@ -726,7 +726,7 @@ static int sendup_buffer (struct net_dev
int dnode, snode, llaptype, len;
int sklen;
struct sk_buff *skb;
- struct net_device_stats *stats = &((struct ltpc_private *)dev->priv)->stats;
+ struct net_device_stats *stats = &((struct ltpc_private *)netdev_priv(dev))->stats;
struct lt_rcvlap *ltc = (struct lt_rcvlap *) ltdmacbuf;
if (ltc->command != LT_RCVLAP) {
@@ -823,7 +823,7 @@ static int ltpc_ioctl(struct net_device
{
struct sockaddr_at *sa = (struct sockaddr_at *) &ifr->ifr_addr;
/* we'll keep the localtalk node address in dev->pa_addr */
- struct atalk_addr *aa = &((struct ltpc_private *)dev->priv)->my_addr;
+ struct atalk_addr *aa = &((struct ltpc_private *)netdev_priv(dev))->my_addr;
struct lt_init c;
int ltflags;
@@ -913,7 +913,7 @@ static int ltpc_xmit(struct sk_buff *skb
* and skb->len is the length of the ddp data + ddp header
*/
- struct net_device_stats *stats = &((struct ltpc_private *)dev->priv)->stats;
+ struct net_device_stats *stats = &((struct ltpc_private *)netdev_priv(dev))->stats;
int i;
struct lt_sendlap cbuf;
@@ -952,7 +952,7 @@ static int ltpc_xmit(struct sk_buff *skb
static struct net_device_stats *ltpc_get_stats(struct net_device *dev)
{
- struct net_device_stats *stats = &((struct ltpc_private *) dev->priv)->stats;
+ struct net_device_stats *stats = &((struct ltpc_private *)netdev_priv(dev))->stats;
return stats;
}
^ permalink raw reply related
* [PATCH 01/13] dev->priv to netdev_priv(dev), for drivers/net
From: Yoann Padioleau @ 2007-08-03 17:33 UTC (permalink / raw)
To: kernel-janitors; +Cc: jeff, netdev, akpm, linux-kernel
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)
Signed-off-by: Yoann Padioleau <padator@wanadoo.fr>
Cc: jeff@garzik.org
Cc: netdev@vger.kernel.org
Cc: akpm@linux-foundation.org
---
drivers/net/3c501.c | 2 -
drivers/net/3c505.c | 36 +++++++++---------
drivers/net/3c523.c | 28 +++++++-------
drivers/net/3c59x.c | 4 +-
drivers/net/acenic.c | 2 -
drivers/net/atarilance.c | 22 +++++------
drivers/net/atp.c | 2 -
drivers/net/au1000_eth.c | 50 +++++++++++++-------------
drivers/net/de600.c | 8 ++--
drivers/net/de620.c | 14 +++----
drivers/net/declance.c | 2 -
drivers/net/depca.c | 42 ++++++++++-----------
drivers/net/dgrs.c | 32 ++++++++--------
drivers/net/dm9000.c | 32 ++++++++--------
drivers/net/eepro.c | 4 +-
drivers/net/epic100.c | 44 +++++++++++------------
drivers/net/eth16i.c | 2 -
drivers/net/fec.c | 2 -
drivers/net/ibmveth.c | 24 ++++++------
drivers/net/iseries_veth.c | 20 +++++-----
drivers/net/lp486e.c | 30 +++++++--------
drivers/net/mv643xx_eth.c | 2 -
drivers/net/mvme147.c | 4 +-
drivers/net/myri_sbus.c | 14 +++----
drivers/net/ne2k-pci.c | 2 -
drivers/net/ni5010.c | 2 -
drivers/net/ni52.c | 32 ++++++++--------
drivers/net/pci-skeleton.c | 38 +++++++++----------
drivers/net/plip.c | 8 ++--
drivers/net/qla3xxx.c | 2 -
drivers/net/rionet.c | 28 +++++++-------
drivers/net/s2io.c | 74 +++++++++++++++++++-------------------
drivers/net/sgiseeq.c | 2 -
drivers/net/shaper.c | 16 ++++----
drivers/net/sis900.c | 66 +++++++++++++++++-----------------
drivers/net/smc9194.c | 6 +--
drivers/net/sonic.h | 20 +++++-----
drivers/net/sun3_82586.c | 32 ++++++++--------
drivers/net/sunbmac.c | 18 ++++-----
drivers/net/sungem.c | 54 ++++++++++++++--------------
drivers/net/sunhme.c | 26 ++++++-------
drivers/net/sunqe.c | 16 ++++----
drivers/net/tc35815.c | 86 ++++++++++++++++++++++-----------------------
drivers/net/typhoon.c | 2 -
drivers/net/znet.c | 24 ++++++------
45 files changed, 488 insertions(+), 488 deletions(-)
diff --git a/drivers/net/3c501.c b/drivers/net/3c501.c
index 4bee99b..2461688 100644
--- a/drivers/net/3c501.c
+++ b/drivers/net/3c501.c
@@ -304,7 +304,7 @@ #endif
if (el_debug)
printk(KERN_DEBUG "%s", version);
- memset(dev->priv, 0, sizeof(struct net_local));
+ memset(netdev_priv(dev), 0, sizeof(struct net_local));
lp = netdev_priv(dev);
spin_lock_init(&lp->lock);
diff --git a/drivers/net/3c505.c b/drivers/net/3c505.c
index e985a85..4cf3dde 100644
--- a/drivers/net/3c505.c
+++ b/drivers/net/3c505.c
@@ -218,7 +218,7 @@ static inline int inb_command(unsigned i
static inline void outb_control(unsigned char val, struct net_device *dev)
{
outb(val, dev->base_addr + PORT_CONTROL);
- ((elp_device *)(dev->priv))->hcr_val = val;
+ ((elp_device *)(netdev_priv(dev)))->hcr_val = val;
}
#define HCR_VAL(x) (((elp_device *)((x)->priv))->hcr_val)
@@ -262,7 +262,7 @@ static inline int get_status(unsigned in
static inline void set_hsf(struct net_device *dev, int hsf)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
unsigned long flags;
spin_lock_irqsave(&adapter->lock, flags);
@@ -275,7 +275,7 @@ static int start_receive(struct net_devi
static inline void adapter_reset(struct net_device *dev)
{
unsigned long timeout;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
unsigned char orig_hcr = adapter->hcr_val;
outb_control(0, dev);
@@ -308,7 +308,7 @@ static inline void adapter_reset(struct
*/
static inline void check_3c505_dma(struct net_device *dev)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
if (adapter->dmaing && time_after(jiffies, adapter->current_dma.start_time + 10)) {
unsigned long flags, f;
printk(KERN_ERR "%s: DMA %s timed out, %d bytes left\n", dev->name, adapter->current_dma.direction ? "download" : "upload", get_dma_residue(dev->dma));
@@ -355,7 +355,7 @@ static inline unsigned int send_pcb_fast
/* Check to see if the receiver needs restarting, and kick it if so */
static inline void prime_rx(struct net_device *dev)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
while (adapter->rx_active < ELP_RX_PCBS && netif_running(dev)) {
if (!start_receive(dev, &adapter->itx_pcb))
break;
@@ -390,7 +390,7 @@ static int send_pcb(struct net_device *d
{
int i;
unsigned long timeout;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
unsigned long flags;
check_3c505_dma(dev);
@@ -478,7 +478,7 @@ static int receive_pcb(struct net_device
unsigned long timeout;
unsigned long flags;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
set_hsf(dev, 0);
@@ -558,7 +558,7 @@ static int receive_pcb(struct net_device
static int start_receive(struct net_device *dev, pcb_struct * tx_pcb)
{
int status;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
if (elp_debug >= 3)
printk(KERN_DEBUG "%s: restarting receiver\n", dev->name);
@@ -586,7 +586,7 @@ static int start_receive(struct net_devi
static void receive_packet(struct net_device *dev, int len)
{
int rlen;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
void *target;
struct sk_buff *skb;
unsigned long flags;
@@ -658,7 +658,7 @@ static irqreturn_t elp_interrupt(int irq
unsigned long timeout;
dev = dev_id;
- adapter = (elp_device *) dev->priv;
+ adapter = netdev_priv(dev);
spin_lock(&adapter->lock);
@@ -856,7 +856,7 @@ static int elp_open(struct net_device *d
elp_device *adapter;
int retval;
- adapter = dev->priv;
+ adapter = netdev_priv(dev);
if (elp_debug >= 3)
printk(KERN_DEBUG "%s: request to open device\n", dev->name);
@@ -986,7 +986,7 @@ static int elp_open(struct net_device *d
static int send_packet(struct net_device *dev, struct sk_buff *skb)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
unsigned long target;
unsigned long flags;
@@ -1056,7 +1056,7 @@ static int send_packet(struct net_device
static void elp_timeout(struct net_device *dev)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
int stat;
stat = inb_status(dev->base_addr);
@@ -1078,7 +1078,7 @@ static void elp_timeout(struct net_devic
static int elp_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
unsigned long flags;
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
spin_lock_irqsave(&adapter->lock, flags);
check_3c505_dma(dev);
@@ -1120,7 +1120,7 @@ static int elp_start_xmit(struct sk_buff
static struct net_device_stats *elp_get_stats(struct net_device *dev)
{
- elp_device *adapter = (elp_device *) dev->priv;
+ elp_device *adapter = netdev_priv(dev);
if (elp_debug >= 3)
printk(KERN_DEBUG "%s: request for stats\n", dev->name);
@@ -1184,7 +1184,7 @@ static int elp_close(struct net_device *
{
elp_device *adapter;
- adapter = dev->priv;
+ adapter = netdev_priv(dev);
if (elp_debug >= 3)
printk(KERN_DEBUG "%s: request to close device\n", dev->name);
@@ -1225,7 +1225,7 @@ static int elp_close(struct net_device *
static void elp_set_mc_list(struct net_device *dev)
{
- elp_device *adapter = (elp_device *) dev->priv;
+ elp_device *adapter = netdev_priv(dev);
struct dev_mc_list *dmi = dev->mc_list;
int i;
unsigned long flags;
@@ -1396,7 +1396,7 @@ static int __init elp_autodetect(struct
static int __init elplus_setup(struct net_device *dev)
{
- elp_device *adapter = dev->priv;
+ elp_device *adapter = netdev_priv(dev);
int i, tries, tries1, okay;
unsigned long timeout;
unsigned long cookie = 0;
diff --git a/drivers/net/3c523.c b/drivers/net/3c523.c
index ab18343..98fef26 100644
--- a/drivers/net/3c523.c
+++ b/drivers/net/3c523.c
@@ -309,7 +309,7 @@ static int elmc_open(struct net_device *
static int __init check586(struct net_device *dev, unsigned long where, unsigned size)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
char *iscp_addrs[2];
int i = 0;
@@ -350,7 +350,7 @@ static int __init check586(struct net_de
void alloc586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
elmc_id_reset586();
DELAY(2);
@@ -421,7 +421,7 @@ static int __init do_elmc_probe(struct n
int i = 0;
unsigned int size = 0;
int retval;
- struct priv *pr = dev->priv;
+ struct priv *pr = netdev_priv(dev);
SET_MODULE_OWNER(dev);
if (MCA_bus == 0) {
@@ -586,7 +586,7 @@ err_out:
static void cleanup_card(struct net_device *dev)
{
- mca_set_adapter_procfn(((struct priv *) (dev->priv))->slot, NULL, NULL);
+ mca_set_adapter_procfn(((struct priv *) (netdev_priv(dev)))->slot, NULL, NULL);
release_region(dev->base_addr, ELMC_IO_EXTENT);
}
@@ -622,7 +622,7 @@ static int init586(struct net_device *de
void *ptr;
unsigned long s;
int i, result = 0;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
volatile struct configure_cmd_struct *cfg_cmd;
volatile struct iasetup_cmd_struct *ias_cmd;
volatile struct tdr_cmd_struct *tdr_cmd;
@@ -860,7 +860,7 @@ static void *alloc_rfa(struct net_device
volatile struct rfd_struct *rfd = (struct rfd_struct *) ptr;
volatile struct rbd_struct *rbd;
int i;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
memset((char *) rfd, 0, sizeof(struct rfd_struct) * p->num_recv_buffs);
p->rfd_first = rfd;
@@ -921,7 +921,7 @@ elmc_interrupt(int irq, void *dev_id)
}
/* reading ELMC_CTRL also clears the INT bit. */
- p = (struct priv *) dev->priv;
+ p = netdev_priv(dev);
while ((stat = p->scb->status & STAT_MASK))
{
@@ -977,7 +977,7 @@ static void elmc_rcv_int(struct net_devi
unsigned short totlen;
struct sk_buff *skb;
struct rbd_struct *rbd;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
for (; (status = p->rfd_top->status) & STAT_COMPL;) {
rbd = (struct rbd_struct *) make32(p->rfd_top->rbd_offset);
@@ -1021,7 +1021,7 @@ static void elmc_rcv_int(struct net_devi
static void elmc_rnr_int(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
p->stats.rx_errors++;
@@ -1044,7 +1044,7 @@ static void elmc_rnr_int(struct net_devi
static void elmc_xmt_int(struct net_device *dev)
{
int status;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
status = p->xmit_cmds[p->xmit_last]->cmd_status;
if (!(status & STAT_COMPL)) {
@@ -1087,7 +1087,7 @@ #endif
static void startrecv586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
p->scb->rfa_offset = make16(p->rfd_first);
p->scb->cmd = RUC_START;
@@ -1101,7 +1101,7 @@ static void startrecv586(struct net_devi
static void elmc_timeout(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
/* COMMAND-UNIT active? */
if (p->scb->status & CU_ACTIVE) {
#ifdef DEBUG
@@ -1137,7 +1137,7 @@ static int elmc_send_packet(struct sk_bu
#ifndef NO_NOPCOMMANDS
int next_nop;
#endif
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
netif_stop_queue(dev);
@@ -1208,7 +1208,7 @@ #endif
static struct net_device_stats *elmc_get_stats(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
unsigned short crc, aln, rsc, ovrn;
crc = p->scb->crc_errs; /* get error-statistic from the ni82586 */
diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c
index 001c66d..a70fe56 100644
--- a/drivers/net/3c59x.c
+++ b/drivers/net/3c59x.c
@@ -804,7 +804,7 @@ static int vortex_suspend(struct pci_dev
{
struct net_device *dev = pci_get_drvdata(pdev);
- if (dev && dev->priv) {
+ if (dev && netdev_priv(dev)) {
if (netif_running(dev)) {
netif_device_detach(dev);
vortex_down(dev, 1);
@@ -3193,7 +3193,7 @@ #ifdef CONFIG_EISA
#endif
if (compaq_net_device) {
- vp = compaq_net_device->priv;
+ vp = netdev_priv(compaq_net_device);
ioaddr = ioport_map(compaq_net_device->base_addr,
VORTEX_TOTAL_SIZE);
diff --git a/drivers/net/acenic.c b/drivers/net/acenic.c
index 62e660a..dcfa951 100644
--- a/drivers/net/acenic.c
+++ b/drivers/net/acenic.c
@@ -468,7 +468,7 @@ static int __devinit acenic_probe_one(st
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &pdev->dev);
- ap = dev->priv;
+ ap = netdev_priv(dev);
ap->pdev = pdev;
ap->name = pci_name(pdev);
diff --git a/drivers/net/atarilance.c b/drivers/net/atarilance.c
index dfa8b9b..223df93 100644
--- a/drivers/net/atarilance.c
+++ b/drivers/net/atarilance.c
@@ -524,7 +524,7 @@ static unsigned long __init lance_probe1
return( 0 );
probe_ok:
- lp = (struct lance_private *)dev->priv;
+ lp = netdev_priv(dev);
MEM = (struct lance_memory *)memaddr;
IO = lp->iobase = (struct lance_ioreg *)ioaddr;
dev->base_addr = (unsigned long)ioaddr; /* informational only */
@@ -653,7 +653,7 @@ #endif
static int lance_open( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
struct lance_ioreg *IO = lp->iobase;
int i;
@@ -694,7 +694,7 @@ static int lance_open( struct net_device
static void lance_init_ring( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
int i;
unsigned offset;
@@ -742,7 +742,7 @@ #define CHECK_OFFSET(o) \
static void lance_tx_timeout (struct net_device *dev)
{
- struct lance_private *lp = (struct lance_private *) dev->priv;
+ struct lance_private *lp = netdev_priv(dev);
struct lance_ioreg *IO = lp->iobase;
AREG = CSR0;
@@ -785,7 +785,7 @@ #endif
static int lance_start_xmit( struct sk_buff *skb, struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
struct lance_ioreg *IO = lp->iobase;
int entry, len;
struct lance_tx_head *head;
@@ -879,7 +879,7 @@ static irqreturn_t lance_interrupt( int
return IRQ_NONE;
}
- lp = (struct lance_private *)dev->priv;
+ lp = netdev_priv(dev);
IO = lp->iobase;
spin_lock (&lp->devlock);
@@ -980,7 +980,7 @@ #endif
static int lance_rx( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
int entry = lp->cur_rx & RX_RING_MOD_MASK;
int i;
@@ -1073,7 +1073,7 @@ static int lance_rx( struct net_device *
static int lance_close( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
struct lance_ioreg *IO = lp->iobase;
netif_stop_queue (dev);
@@ -1093,7 +1093,7 @@ static int lance_close( struct net_devic
static struct net_device_stats *lance_get_stats( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
return &lp->stats;
}
@@ -1108,7 +1108,7 @@ static struct net_device_stats *lance_ge
static void set_multicast_list( struct net_device *dev )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
struct lance_ioreg *IO = lp->iobase;
if (netif_running(dev))
@@ -1150,7 +1150,7 @@ static void set_multicast_list( struct n
static int lance_set_mac_address( struct net_device *dev, void *addr )
-{ struct lance_private *lp = (struct lance_private *)dev->priv;
+{ struct lance_private *lp = netdev_priv(dev);
struct sockaddr *saddr = addr;
int i;
diff --git a/drivers/net/atp.c b/drivers/net/atp.c
index 82d78ff..96a9855 100644
--- a/drivers/net/atp.c
+++ b/drivers/net/atp.c
@@ -932,7 +932,7 @@ static void __exit atp_cleanup_module(vo
struct net_device *next_dev;
while (root_atp_dev) {
- next_dev = ((struct net_local *)root_atp_dev->priv)->next_module;
+ next_dev = ((struct net_local *)netdev_priv(root_atp_dev))->next_module;
unregister_netdev(root_atp_dev);
/* No need to release_region(), since we never snarf it. */
free_netdev(root_atp_dev);
diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c
index e86b369..21713fe 100644
--- a/drivers/net/au1000_eth.c
+++ b/drivers/net/au1000_eth.c
@@ -196,7 +196,7 @@ #endif
*/
static int mdio_read(struct net_device *dev, int phy_addr, int reg)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
volatile u32 *const mii_control_reg = &aup->mac->mii_control;
volatile u32 *const mii_data_reg = &aup->mac->mii_data;
u32 timedout = 20;
@@ -230,7 +230,7 @@ static int mdio_read(struct net_device *
static void mdio_write(struct net_device *dev, int phy_addr, int reg, u16 value)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
volatile u32 *const mii_control_reg = &aup->mac->mii_control;
volatile u32 *const mii_data_reg = &aup->mac->mii_data;
u32 timedout = 20;
@@ -285,7 +285,7 @@ static int mdiobus_reset(struct mii_bus
static int mii_probe (struct net_device *dev)
{
- struct au1000_private *const aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *const aup = netdev_priv(dev);
struct phy_device *phydev = NULL;
#if defined(AU1XXX_PHY_STATIC_CONFIG)
@@ -417,7 +417,7 @@ void ReleaseDB(struct au1000_private *au
static void enable_rx_tx(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (au1000_debug > 4)
printk(KERN_INFO "%s: enable_rx_tx\n", dev->name);
@@ -428,7 +428,7 @@ static void enable_rx_tx(struct net_devi
static void hard_stop(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (au1000_debug > 4)
printk(KERN_INFO "%s: hard stop\n", dev->name);
@@ -440,7 +440,7 @@ static void hard_stop(struct net_device
static void enable_mac(struct net_device *dev, int force_reset)
{
unsigned long flags;
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
spin_lock_irqsave(&aup->lock, flags);
@@ -459,7 +459,7 @@ static void enable_mac(struct net_device
static void reset_mac_unlocked(struct net_device *dev)
{
- struct au1000_private *const aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *const aup = netdev_priv(dev);
int i;
hard_stop(dev);
@@ -485,7 +485,7 @@ static void reset_mac_unlocked(struct ne
static void reset_mac(struct net_device *dev)
{
- struct au1000_private *const aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *const aup = netdev_priv(dev);
unsigned long flags;
if (au1000_debug > 4)
@@ -574,7 +574,7 @@ static int __init au1000_init_module(voi
static int au1000_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct au1000_private *aup = (struct au1000_private *)dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (aup->phy_dev)
return phy_ethtool_gset(aup->phy_dev, cmd);
@@ -584,7 +584,7 @@ static int au1000_get_settings(struct ne
static int au1000_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct au1000_private *aup = (struct au1000_private *)dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (!capable(CAP_NET_ADMIN))
return -EPERM;
@@ -598,7 +598,7 @@ static int au1000_set_settings(struct ne
static void
au1000_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct au1000_private *aup = (struct au1000_private *)dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
@@ -655,7 +655,7 @@ static struct net_device * au1000_probe(
printk("%s: Au1xx0 Ethernet found at 0x%x, irq %d\n",
dev->name, base, irq);
- aup = dev->priv;
+ aup = netdev_priv(dev);
/* Allocate the data buffers */
/* Snooping works fine with eth on all au1xxx */
@@ -820,7 +820,7 @@ err_out:
*/
static int au1000_init(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
u32 flags;
int i;
u32 control;
@@ -871,7 +871,7 @@ #endif
static void
au1000_adjust_link(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
struct phy_device *phydev = aup->phy_dev;
unsigned long flags;
@@ -951,7 +951,7 @@ au1000_adjust_link(struct net_device *de
static int au1000_open(struct net_device *dev)
{
int retval;
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (au1000_debug > 4)
printk("%s: open: dev=%p\n", dev->name, dev);
@@ -986,7 +986,7 @@ static int au1000_open(struct net_device
static int au1000_close(struct net_device *dev)
{
unsigned long flags;
- struct au1000_private *const aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *const aup = netdev_priv(dev);
if (au1000_debug > 4)
printk("%s: close: dev=%p\n", dev->name, dev);
@@ -1017,7 +1017,7 @@ static void __exit au1000_cleanup_module
for (i = 0; i < num_ifs; i++) {
dev = iflist[i].dev;
if (dev) {
- aup = (struct au1000_private *) dev->priv;
+ aup = netdev_priv(dev);
unregister_netdev(dev);
for (j = 0; j < NUM_RX_DMA; j++)
if (aup->rx_db_inuse[j])
@@ -1037,7 +1037,7 @@ static void __exit au1000_cleanup_module
static void update_tx_stats(struct net_device *dev, u32 status)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
struct net_device_stats *ps = &aup->stats;
if (status & TX_FRAME_ABORTED) {
@@ -1066,7 +1066,7 @@ static void update_tx_stats(struct net_d
*/
static void au1000_tx_ack(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
volatile tx_dma_t *ptxd;
ptxd = aup->tx_dma_ring[aup->tx_tail];
@@ -1093,7 +1093,7 @@ static void au1000_tx_ack(struct net_dev
*/
static int au1000_tx(struct sk_buff *skb, struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
struct net_device_stats *ps = &aup->stats;
volatile tx_dma_t *ptxd;
u32 buff_stat;
@@ -1147,7 +1147,7 @@ static int au1000_tx(struct sk_buff *skb
static inline void update_rx_stats(struct net_device *dev, u32 status)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
struct net_device_stats *ps = &aup->stats;
ps->rx_packets++;
@@ -1175,7 +1175,7 @@ static inline void update_rx_stats(struc
*/
static int au1000_rx(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
struct sk_buff *skb;
volatile rx_dma_t *prxd;
u32 buff_stat, status;
@@ -1283,7 +1283,7 @@ static void au1000_tx_timeout(struct net
static void set_rx_mode(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (au1000_debug > 4)
printk("%s: set_rx_mode: flags=%x\n", dev->name, dev->flags);
@@ -1315,7 +1315,7 @@ static void set_rx_mode(struct net_devic
static int au1000_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
- struct au1000_private *aup = (struct au1000_private *)dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (!netif_running(dev)) return -EINVAL;
@@ -1326,7 +1326,7 @@ static int au1000_ioctl(struct net_devic
static struct net_device_stats *au1000_get_stats(struct net_device *dev)
{
- struct au1000_private *aup = (struct au1000_private *) dev->priv;
+ struct au1000_private *aup = netdev_priv(dev);
if (au1000_debug > 4)
printk("%s: au1000_get_stats: dev=%p\n", dev->name, dev);
diff --git a/drivers/net/de600.c b/drivers/net/de600.c
index dae97b8..27cf8c7 100644
--- a/drivers/net/de600.c
+++ b/drivers/net/de600.c
@@ -156,7 +156,7 @@ static int de600_close(struct net_device
static struct net_device_stats *get_stats(struct net_device *dev)
{
- return (struct net_device_stats *)(dev->priv);
+ return (struct net_device_stats *)(netdev_priv(dev));
}
static inline void trigger_interrupt(struct net_device *dev)
@@ -308,7 +308,7 @@ static int de600_tx_intr(struct net_devi
if (!(irq_status & TX_FAILED16)) {
tx_fifo_out = (tx_fifo_out + 1) % TX_PAGES;
++free_tx_pages;
- ((struct net_device_stats *)(dev->priv))->tx_packets++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->tx_packets++;
netif_wake_queue(dev);
}
@@ -375,8 +375,8 @@ static void de600_rx_intr(struct net_dev
/* update stats */
dev->last_rx = jiffies;
- ((struct net_device_stats *)(dev->priv))->rx_packets++; /* count all receives */
- ((struct net_device_stats *)(dev->priv))->rx_bytes += size; /* count all received bytes */
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_packets++; /* count all receives */
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_bytes += size; /* count all received bytes */
/*
* If any worth-while packets have been received, netif_rx()
diff --git a/drivers/net/de620.c b/drivers/net/de620.c
index dc48924..e57b0f9 100644
--- a/drivers/net/de620.c
+++ b/drivers/net/de620.c
@@ -485,7 +485,7 @@ static int de620_close(struct net_device
*/
static struct net_device_stats *get_stats(struct net_device *dev)
{
- return (struct net_device_stats *)(dev->priv);
+ return (struct net_device_stats *)(netdev_priv(dev));
}
/*********************************************
@@ -579,7 +579,7 @@ static int de620_start_xmit(struct sk_bu
if(!(using_txbuf == (TXBF0 | TXBF1)))
netif_wake_queue(dev);
- ((struct net_device_stats *)(dev->priv))->tx_packets++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->tx_packets++;
spin_unlock_irqrestore(&de620_lock, flags);
dev_kfree_skb (skb);
return 0;
@@ -660,7 +660,7 @@ static int de620_rx_intr(struct net_devi
/* You win some, you lose some. And sometimes plenty... */
adapter_init(dev);
netif_wake_queue(dev);
- ((struct net_device_stats *)(dev->priv))->rx_over_errors++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_over_errors++;
return 0;
}
@@ -680,7 +680,7 @@ static int de620_rx_intr(struct net_devi
next_rx_page = header_buf.Rx_NextPage; /* at least a try... */
de620_send_command(dev, W_DUMMY);
de620_set_register(dev, W_NPRF, next_rx_page);
- ((struct net_device_stats *)(dev->priv))->rx_over_errors++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_over_errors++;
return 0;
}
next_rx_page = pagelink;
@@ -693,7 +693,7 @@ static int de620_rx_intr(struct net_devi
skb = dev_alloc_skb(size+2);
if (skb == NULL) { /* Yeah, but no place to put it... */
printk(KERN_WARNING "%s: Couldn't allocate a sk_buff of size %d.\n", dev->name, size);
- ((struct net_device_stats *)(dev->priv))->rx_dropped++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_dropped++;
}
else { /* Yep! Go get it! */
skb_reserve(skb,2); /* Align */
@@ -706,8 +706,8 @@ static int de620_rx_intr(struct net_devi
netif_rx(skb); /* deliver it "upstairs" */
dev->last_rx = jiffies;
/* count all receives */
- ((struct net_device_stats *)(dev->priv))->rx_packets++;
- ((struct net_device_stats *)(dev->priv))->rx_bytes += size;
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_packets++;
+ ((struct net_device_stats *)(netdev_priv(dev)))->rx_bytes += size;
}
}
diff --git a/drivers/net/declance.c b/drivers/net/declance.c
index b2577f4..d8cb974 100644
--- a/drivers/net/declance.c
+++ b/drivers/net/declance.c
@@ -1047,7 +1047,7 @@ static int __init dec_lance_probe(struct
dev = root_lance_dev;
while (dev) {
i++;
- lp = (struct lance_private *)dev->priv;
+ lp = netdev_priv(dev);
dev = lp->next;
}
snprintf(name, sizeof(name), fmt, i);
diff --git a/drivers/net/depca.c b/drivers/net/depca.c
index 1834970..ba32f83 100644
--- a/drivers/net/depca.c
+++ b/drivers/net/depca.c
@@ -602,7 +602,7 @@ static int __init depca_hw_init (struct
return -ENXIO;
}
- lp = (struct depca_private *) dev->priv;
+ lp = netdev_priv(dev);
mem_start = lp->mem_start;
if (!mem_start || lp->adapter < DEPCA || lp->adapter >=unknown)
@@ -826,7 +826,7 @@ out_priv:
static int depca_open(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
s16 nicsr;
int status = 0;
@@ -871,7 +871,7 @@ static int depca_open(struct net_device
/* Initialize the lance Rx and Tx descriptor rings. */
static void depca_init_ring(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_int i;
u_long offset;
@@ -929,7 +929,7 @@ static void depca_tx_timeout(struct net_
*/
static int depca_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
int status = 0;
@@ -977,7 +977,7 @@ static irqreturn_t depca_interrupt(int i
return IRQ_NONE;
}
- lp = (struct depca_private *) dev->priv;
+ lp = netdev_priv(dev);
ioaddr = dev->base_addr;
spin_lock(&lp->lock);
@@ -1015,7 +1015,7 @@ static irqreturn_t depca_interrupt(int i
/* Called with lp->lock held */
static int depca_rx(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
int i, entry;
s32 status;
@@ -1113,7 +1113,7 @@ static int depca_rx(struct net_device *d
*/
static int depca_tx(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
int entry;
s32 status;
u_long ioaddr = dev->base_addr;
@@ -1154,7 +1154,7 @@ static int depca_tx(struct net_device *d
static int depca_close(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
s16 nicsr;
u_long ioaddr = dev->base_addr;
@@ -1190,7 +1190,7 @@ static int depca_close(struct net_device
static void LoadCSRs(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
outw(CSR1, DEPCA_ADDR); /* initialisation block address LSW */
@@ -1207,7 +1207,7 @@ static void LoadCSRs(struct net_device *
static int InitRestartDepca(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
int i, status = 0;
@@ -1236,7 +1236,7 @@ static int InitRestartDepca(struct net_d
static struct net_device_stats *depca_get_stats(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
/* Null body since there is no framing error counter */
@@ -1248,7 +1248,7 @@ static struct net_device_stats *depca_ge
*/
static void set_multicast_list(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
netif_stop_queue(dev);
@@ -1277,7 +1277,7 @@ static void set_multicast_list(struct ne
*/
static void SetMulticastFilter(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
struct dev_mc_list *dmi = dev->mc_list;
char *addrs;
int i, j, bit, byte;
@@ -1445,7 +1445,7 @@ static int __init depca_mca_probe(struct
dev->irq = irq;
dev->base_addr = iobase;
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->depca_bus = DEPCA_BUS_MCA;
lp->adapter = depca_mca_adapter_type[mdev->index];
lp->mem_start = mem_start;
@@ -1548,7 +1548,7 @@ static int __init depca_isa_probe (struc
dev->base_addr = ioaddr;
dev->irq = irq; /* Use whatever value the user gave
* us, and 0 if he didn't. */
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->depca_bus = DEPCA_BUS_ISA;
lp->adapter = adapter;
lp->mem_start = mem_start;
@@ -1594,7 +1594,7 @@ static int __init depca_eisa_probe (stru
dev->base_addr = ioaddr;
dev->irq = irq;
- lp = dev->priv;
+ lp = netdev_priv(dev);
lp->depca_bus = DEPCA_BUS_EISA;
lp->adapter = edev->id.driver_data;
lp->mem_start = mem_start;
@@ -1619,7 +1619,7 @@ static int __devexit depca_device_remove
int bus;
dev = device->driver_data;
- lp = dev->priv;
+ lp = netdev_priv(dev);
unregister_netdev (dev);
iounmap (lp->sh_mem);
@@ -1761,7 +1761,7 @@ static int __init DevicePresent(u_long i
static int __init get_hw_addr(struct net_device *dev)
{
u_long ioaddr = dev->base_addr;
- struct depca_private *lp = dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
int i, k, tmp, status = 0;
u_short j, x, chksum;
@@ -1796,7 +1796,7 @@ static int __init get_hw_addr(struct net
*/
static int load_packet(struct net_device *dev, struct sk_buff *skb)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
int i, entry, end, len, status = 0;
entry = lp->tx_new; /* Ring around buffer number. */
@@ -1851,7 +1851,7 @@ static int load_packet(struct net_device
static void depca_dbg_open(struct net_device *dev)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
u_long ioaddr = dev->base_addr;
struct depca_init *p = &lp->init_block;
int i;
@@ -1926,7 +1926,7 @@ static void depca_dbg_open(struct net_de
*/
static int depca_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
- struct depca_private *lp = (struct depca_private *) dev->priv;
+ struct depca_private *lp = netdev_priv(dev);
struct depca_ioctl *ioc = (struct depca_ioctl *) &rq->ifr_ifru;
int i, status = 0;
u_long ioaddr = dev->base_addr;
diff --git a/drivers/net/dgrs.c b/drivers/net/dgrs.c
index df62c02..4a2651c 100644
--- a/drivers/net/dgrs.c
+++ b/drivers/net/dgrs.c
@@ -251,7 +251,7 @@ typedef struct
static void
proc_reset(struct net_device *dev0, int reset)
{
- DGRS_PRIV *priv0 = (DGRS_PRIV *) dev0->priv;
+ DGRS_PRIV *priv0 = netdev_priv(dev0);
if (priv0->plxreg)
{
@@ -275,7 +275,7 @@ proc_reset(struct net_device *dev0, int
static int
check_board_dma(struct net_device *dev0)
{
- DGRS_PRIV *priv0 = (DGRS_PRIV *) dev0->priv;
+ DGRS_PRIV *priv0 = netdev_priv(dev0);
ulong x;
/*
@@ -358,7 +358,7 @@ do_plx_dma(
{
int i;
ulong csr = 0;
- DGRS_PRIV *priv = (DGRS_PRIV *) dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
if (pciaddr)
{
@@ -482,7 +482,7 @@ dgrs_rcv_frame(
*/
if (devN == NULL)
goto out;
- privN = (DGRS_PRIV *) devN->priv;
+ privN = netdev_priv(devN);
}
else
{ /* Switch mode */
@@ -694,7 +694,7 @@ out:
static int dgrs_start_xmit(struct sk_buff *skb, struct net_device *devN)
{
- DGRS_PRIV *privN = (DGRS_PRIV *) devN->priv;
+ DGRS_PRIV *privN = netdev_priv(devN);
struct net_device *dev0;
DGRS_PRIV *priv0;
I596_RBD *rbdp;
@@ -707,7 +707,7 @@ static int dgrs_start_xmit(struct sk_buf
if (dgrs_nicmode)
{
dev0 = privN->devtbl[0];
- priv0 = (DGRS_PRIV *) dev0->priv;
+ priv0 = netdev_priv(dev0);
}
else
{
@@ -810,7 +810,7 @@ static int dgrs_close( struct net_device
*/
static struct net_device_stats *dgrs_get_stats( struct net_device *dev )
{
- DGRS_PRIV *priv = (DGRS_PRIV *) dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
return (&priv->stats);
}
@@ -821,7 +821,7 @@ static struct net_device_stats *dgrs_get
static void dgrs_set_multicast_list( struct net_device *dev)
{
- DGRS_PRIV *priv = (DGRS_PRIV *) dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
priv->port->is_promisc = (dev->flags & IFF_PROMISC) ? 1 : 0;
}
@@ -831,7 +831,7 @@ static void dgrs_set_multicast_list( str
*/
static int dgrs_ioctl(struct net_device *devN, struct ifreq *ifr, int cmd)
{
- DGRS_PRIV *privN = (DGRS_PRIV *) devN->priv;
+ DGRS_PRIV *privN = netdev_priv(devN);
DGRS_IOCTL ioc;
int i;
@@ -897,7 +897,7 @@ static int dgrs_ioctl(struct net_device
static irqreturn_t dgrs_intr(int irq, void *dev_id)
{
struct net_device *dev0 = dev_id;
- DGRS_PRIV *priv0 = dev0->priv;
+ DGRS_PRIV *priv0 = netdev_priv(dev0);
I596_CB *cbp;
int cmd;
int i;
@@ -987,7 +987,7 @@ ack_intr:
static int __init
dgrs_download(struct net_device *dev0)
{
- DGRS_PRIV *priv0 = (DGRS_PRIV *) dev0->priv;
+ DGRS_PRIV *priv0 = netdev_priv(dev0);
int is;
unsigned long i;
@@ -1151,7 +1151,7 @@ dgrs_download(struct net_device *dev0)
static int __init
dgrs_probe1(struct net_device *dev)
{
- DGRS_PRIV *priv = (DGRS_PRIV *) dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
unsigned long i;
int rc;
@@ -1229,7 +1229,7 @@ err_out:
static int __init
dgrs_initclone(struct net_device *dev)
{
- DGRS_PRIV *priv = (DGRS_PRIV *) dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
int i;
printk("%s: Digi RightSwitch port %d ",
@@ -1259,7 +1259,7 @@ dgrs_found_device(
if (!dev)
goto err0;
- priv = (DGRS_PRIV *)dev->priv;
+ priv = netdev_priv(dev);
dev->base_addr = io;
dev->mem_start = mem;
@@ -1305,7 +1305,7 @@ dgrs_found_device(
/* Don't copy the network device structure! */
/* copy the priv structure of dev[0] */
- privN = (DGRS_PRIV *)devN->priv;
+ privN = netdev_priv(devN);
*privN = *priv;
/* ... and zero out VM areas */
@@ -1350,7 +1350,7 @@ dgrs_found_device(
static void __devexit dgrs_remove(struct net_device *dev)
{
- DGRS_PRIV *priv = dev->priv;
+ DGRS_PRIV *priv = netdev_priv(dev);
int i;
unregister_netdev(dev);
diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c
index c3de81b..38064df 100644
--- a/drivers/net/dm9000.c
+++ b/drivers/net/dm9000.c
@@ -328,7 +328,7 @@ static void dm9000_set_io(struct board_i
/* Our watchdog timed out. Called by the networking layer */
static void dm9000_timeout(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
u8 reg_save;
unsigned long flags;
@@ -422,7 +422,7 @@ dm9000_probe(struct platform_device *pde
PRINTK2("dm9000_probe()");
/* setup board info structure */
- db = (struct board_info *) ndev->priv;
+ db = netdev_priv(ndev);
memset(db, 0, sizeof (*db));
spin_lock_init(&db->lock);
@@ -625,7 +625,7 @@ #endif
static int
dm9000_open(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
PRINTK2("entering dm9000_open\n");
@@ -658,7 +658,7 @@ dm9000_open(struct net_device *dev)
static void
dm9000_init_dm9000(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
PRINTK1("entering %s\n",__FUNCTION__);
@@ -700,7 +700,7 @@ dm9000_init_dm9000(struct net_device *de
static int
dm9000_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
PRINTK3("dm9000_start_xmit\n");
@@ -755,7 +755,7 @@ dm9000_start_xmit(struct sk_buff *skb, s
static void
dm9000_shutdown(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
/* RESET device */
dm9000_phy_write(dev, 0, MII_BMCR, BMCR_RESET); /* PHY RESET */
@@ -771,7 +771,7 @@ dm9000_shutdown(struct net_device *dev)
static int
dm9000_stop(struct net_device *ndev)
{
- board_info_t *db = (board_info_t *) ndev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(ndev);
PRINTK1("entering %s\n",__FUNCTION__);
@@ -831,7 +831,7 @@ dm9000_interrupt(int irq, void *dev_id)
}
/* A real interrupt coming */
- db = (board_info_t *) dev->priv;
+ db = (board_info_t *) netdev_priv(dev);
spin_lock(&db->lock);
/* Save previous register address */
@@ -869,7 +869,7 @@ dm9000_interrupt(int irq, void *dev_id)
static struct net_device_stats *
dm9000_get_stats(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
return &db->stats;
}
@@ -882,7 +882,7 @@ static void
dm9000_timer(unsigned long data)
{
struct net_device *dev = (struct net_device *) data;
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
PRINTK3("dm9000_timer()\n");
@@ -904,7 +904,7 @@ struct dm9000_rxhdr {
static void
dm9000_rx(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
struct dm9000_rxhdr rxhdr;
struct sk_buff *skb;
u8 rxbyte, *rdptr;
@@ -1061,7 +1061,7 @@ cal_CRC(unsigned char *Data, unsigned in
static void
dm9000_hash_table(struct net_device *dev)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
struct dev_mc_list *mcptr = dev->mc_list;
int mc_cnt = dev->mc_count;
u32 hash_val;
@@ -1104,7 +1104,7 @@ dm9000_hash_table(struct net_device *dev
static int
dm9000_phy_read(struct net_device *dev, int phy_reg_unused, int reg)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
unsigned long flags;
unsigned int reg_save;
int ret;
@@ -1138,7 +1138,7 @@ dm9000_phy_read(struct net_device *dev,
static void
dm9000_phy_write(struct net_device *dev, int phyaddr_unused, int reg, int value)
{
- board_info_t *db = (board_info_t *) dev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(dev);
unsigned long flags;
unsigned long reg_save;
@@ -1182,7 +1182,7 @@ static int
dm9000_drv_resume(struct platform_device *dev)
{
struct net_device *ndev = platform_get_drvdata(dev);
- board_info_t *db = (board_info_t *) ndev->priv;
+ board_info_t *db = (board_info_t *) netdev_priv(ndev);
if (ndev) {
@@ -1204,7 +1204,7 @@ dm9000_drv_remove(struct platform_device
platform_set_drvdata(pdev, NULL);
unregister_netdev(ndev);
- dm9000_release_board(pdev, (board_info_t *) ndev->priv);
+ dm9000_release_board(pdev, (board_info_t *) netdev_priv(ndev));
free_netdev(ndev); /* free device structure */
PRINTK1("clean_module() exit\n");
diff --git a/drivers/net/eepro.c b/drivers/net/eepro.c
index 4768023..52b29b7 100644
--- a/drivers/net/eepro.c
+++ b/drivers/net/eepro.c
@@ -611,7 +611,7 @@ #endif
static void __init printEEPROMInfo(struct net_device *dev)
{
- struct eepro_local *lp = (struct eepro_local *)dev->priv;
+ struct eepro_local *lp = netdev_priv(dev);
int ioaddr = dev->base_addr;
unsigned short Word;
int i,j;
@@ -1700,7 +1700,7 @@ eepro_transmit_interrupt(struct net_devi
static int eepro_ethtool_get_settings(struct net_device *dev,
struct ethtool_cmd *cmd)
{
- struct eepro_local *lp = (struct eepro_local *)dev->priv;
+ struct eepro_local *lp = netdev_priv(dev);
cmd->supported = SUPPORTED_10baseT_Half |
SUPPORTED_10baseT_Full |
diff --git a/drivers/net/epic100.c b/drivers/net/epic100.c
index 1197784..3413a88 100644
--- a/drivers/net/epic100.c
+++ b/drivers/net/epic100.c
@@ -366,7 +366,7 @@ #else
#endif
pci_set_drvdata(pdev, dev);
- ep = dev->priv;
+ ep = netdev_priv(dev);
ep->mii.dev = dev;
ep->mii.mdio_read = mdio_read;
ep->mii.mdio_write = mdio_write;
@@ -652,7 +652,7 @@ static void mdio_write(struct net_device
static int epic_open(struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
int i;
int retval;
@@ -760,7 +760,7 @@ #endif
static void epic_pause(struct net_device *dev)
{
long ioaddr = dev->base_addr;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
netif_stop_queue (dev);
@@ -783,7 +783,7 @@ static void epic_pause(struct net_device
static void epic_restart(struct net_device *dev)
{
long ioaddr = dev->base_addr;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
int i;
/* Soft reset the chip. */
@@ -835,7 +835,7 @@ #endif
static void check_media(struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
int mii_lpa = ep->mii_phy_cnt ? mdio_read(dev, ep->phys[0], MII_LPA) : 0;
int negotiated = mii_lpa & ep->mii.advertising;
@@ -857,7 +857,7 @@ static void check_media(struct net_devic
static void epic_timer(unsigned long data)
{
struct net_device *dev = (struct net_device *)data;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
int next_tick = 5*HZ;
@@ -878,7 +878,7 @@ static void epic_timer(unsigned long dat
static void epic_tx_timeout(struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
if (debug > 0) {
@@ -907,7 +907,7 @@ static void epic_tx_timeout(struct net_d
/* Initialize the Rx and Tx rings, along with various 'dev' bits. */
static void epic_init_ring(struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
int i;
ep->tx_full = 0;
@@ -953,7 +953,7 @@ static void epic_init_ring(struct net_de
static int epic_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
int entry, free_count;
u32 ctrl_word;
unsigned long flags;
@@ -1081,7 +1081,7 @@ #endif
static irqreturn_t epic_interrupt(int irq, void *dev_instance)
{
struct net_device *dev = dev_instance;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
unsigned int handled = 0;
int status;
@@ -1149,7 +1149,7 @@ out:
static int epic_rx(struct net_device *dev, int budget)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
int entry = ep->cur_rx % RX_RING_SIZE;
int rx_work_limit = ep->dirty_rx + RX_RING_SIZE - ep->cur_rx;
int work_done = 0;
@@ -1259,7 +1259,7 @@ static void epic_rx_err(struct net_devic
static int epic_poll(struct net_device *dev, int *budget)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
int work_done = 0, orig_budget;
long ioaddr = dev->base_addr;
@@ -1304,7 +1304,7 @@ rx_action:
static int epic_close(struct net_device *dev)
{
long ioaddr = dev->base_addr;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
struct sk_buff *skb;
int i;
@@ -1353,7 +1353,7 @@ static int epic_close(struct net_device
static struct net_device_stats *epic_get_stats(struct net_device *dev)
{
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
long ioaddr = dev->base_addr;
if (netif_running(dev)) {
@@ -1374,7 +1374,7 @@ static struct net_device_stats *epic_get
static void set_rx_mode(struct net_device *dev)
{
long ioaddr = dev->base_addr;
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
unsigned char mc_filter[8]; /* Multicast hash filter */
int i;
@@ -1413,7 +1413,7 @@ static void set_rx_mode(struct net_devic
static void netdev_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
strcpy (info->driver, DRV_NAME);
strcpy (info->version, DRV_VERSION);
@@ -1422,7 +1422,7 @@ static void netdev_get_drvinfo (struct n
static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
int rc;
spin_lock_irq(&np->lock);
@@ -1434,7 +1434,7 @@ static int netdev_get_settings(struct ne
static int netdev_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
int rc;
spin_lock_irq(&np->lock);
@@ -1446,13 +1446,13 @@ static int netdev_set_settings(struct ne
static int netdev_nway_reset(struct net_device *dev)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
return mii_nway_restart(&np->mii);
}
static u32 netdev_get_link(struct net_device *dev)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
return mii_link_ok(&np->mii);
}
@@ -1503,7 +1503,7 @@ static const struct ethtool_ops netdev_e
static int netdev_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
- struct epic_private *np = dev->priv;
+ struct epic_private *np = netdev_priv(dev);
long ioaddr = dev->base_addr;
struct mii_ioctl_data *data = if_mii(rq);
int rc;
@@ -1531,7 +1531,7 @@ static int netdev_ioctl(struct net_devic
static void __devexit epic_remove_one (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct epic_private *ep = dev->priv;
+ struct epic_private *ep = netdev_priv(dev);
pci_free_consistent(pdev, TX_TOTAL_SIZE, ep->tx_ring, ep->tx_ring_dma);
pci_free_consistent(pdev, RX_TOTAL_SIZE, ep->rx_ring, ep->rx_ring_dma);
diff --git a/drivers/net/eth16i.c b/drivers/net/eth16i.c
index 04abf59..3fe913c 100644
--- a/drivers/net/eth16i.c
+++ b/drivers/net/eth16i.c
@@ -1481,7 +1481,7 @@ void __exit cleanup_module(void)
for(this_dev = 0; this_dev < MAX_ETH16I_CARDS; this_dev++) {
struct net_device *dev = dev_eth16i[this_dev];
- if(dev->priv) {
+ if(netdev_priv(dev)) {
unregister_netdev(dev);
free_irq(dev->irq, dev);
release_region(dev->base_addr, ETH16I_IO_EXTENT);
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 4e8df91..fdad6f8 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -1165,7 +1165,7 @@ #define MII_DP8384X_PHYSTST 16 /* PH
static void mii_parse_dp8384x_sr2(uint mii_reg, struct net_device *dev)
{
- struct fec_enet_private *fep = dev->priv;
+ struct fec_enet_private *fep = netdev_priv(dev);
volatile uint *s = &(fep->phy_status);
*s &= ~(PHY_STAT_SPMASK | PHY_STAT_LINK | PHY_STAT_ANC);
diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index d96eb72..65fe9c5 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -470,7 +470,7 @@ retry:
static int ibmveth_open(struct net_device *netdev)
{
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
u64 mac_address = 0;
int rxq_entries = 1;
unsigned long lpar_rc;
@@ -582,7 +582,7 @@ static int ibmveth_open(struct net_devic
static int ibmveth_close(struct net_device *netdev)
{
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
long lpar_rc;
ibmveth_debug_printk("close starting\n");
@@ -651,7 +651,7 @@ #define page_offset(v) ((unsigned long)(
static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
{
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
union ibmveth_buf_desc desc[IbmVethMaxSendFrags];
unsigned long lpar_rc;
int nfrags = 0, curfrag;
@@ -769,7 +769,7 @@ out: spin_lock_irqsave(&adapter->stats_l
static int ibmveth_poll(struct net_device *netdev, int *budget)
{
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
int max_frames_to_process = netdev->quota;
int frames_processed = 0;
int more_work = 1;
@@ -846,7 +846,7 @@ static int ibmveth_poll(struct net_devic
static irqreturn_t ibmveth_interrupt(int irq, void *dev_instance)
{
struct net_device *netdev = dev_instance;
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
unsigned long lpar_rc;
if(netif_rx_schedule_prep(netdev)) {
@@ -859,13 +859,13 @@ static irqreturn_t ibmveth_interrupt(int
static struct net_device_stats *ibmveth_get_stats(struct net_device *dev)
{
- struct ibmveth_adapter *adapter = dev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(dev);
return &adapter->stats;
}
static void ibmveth_set_multicast_list(struct net_device *netdev)
{
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
unsigned long lpar_rc;
if((netdev->flags & IFF_PROMISC) || (netdev->mc_count > adapter->mcastFilterSize)) {
@@ -913,7 +913,7 @@ static void ibmveth_set_multicast_list(s
static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
{
- struct ibmveth_adapter *adapter = dev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(dev);
int new_mtu_oh = new_mtu + IBMVETH_BUFF_OH;
int reinit = 0;
int i, rc;
@@ -954,7 +954,7 @@ static int ibmveth_change_mtu(struct net
#ifdef CONFIG_NET_POLL_CONTROLLER
static void ibmveth_poll_controller(struct net_device *dev)
{
- ibmveth_replenish_task(dev->priv);
+ ibmveth_replenish_task(netdev_priv(dev));
ibmveth_interrupt(dev->irq, dev);
}
#endif
@@ -996,7 +996,7 @@ static int __devinit ibmveth_probe(struc
SET_MODULE_OWNER(netdev);
- adapter = netdev->priv;
+ adapter = netdev_priv(netdev);
memset(adapter, 0, sizeof(adapter));
dev->dev.driver_data = netdev;
@@ -1076,7 +1076,7 @@ #endif
static int __devexit ibmveth_remove(struct vio_dev *dev)
{
struct net_device *netdev = dev->dev.driver_data;
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
int i;
for(i = 0; i<IbmVethNumBufferPools; i++)
@@ -1256,7 +1256,7 @@ const char * buf, size_t count)
kobj);
struct net_device *netdev =
container_of(kobj->parent, struct device, kobj)->driver_data;
- struct ibmveth_adapter *adapter = netdev->priv;
+ struct ibmveth_adapter *adapter = netdev_priv(netdev);
long value = simple_strtol(buf, NULL, 10);
long rc;
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index 0433c41..ed0ffe3 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -936,7 +936,7 @@ static void veth_release_connection(stru
static int veth_open(struct net_device *dev)
{
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
memset(&port->stats, 0, sizeof (port->stats));
netif_start_queue(dev);
@@ -951,7 +951,7 @@ static int veth_close(struct net_device
static struct net_device_stats *veth_get_stats(struct net_device *dev)
{
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
return &port->stats;
}
@@ -966,7 +966,7 @@ static int veth_change_mtu(struct net_de
static void veth_set_multicast_list(struct net_device *dev)
{
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
unsigned long flags;
write_lock_irqsave(&port->mcast_gate, flags);
@@ -1058,7 +1058,7 @@ static struct net_device * __init veth_p
return NULL;
}
- port = (struct veth_port *) dev->priv;
+ port = netdev_priv(dev);
spin_lock_init(&port->queue_lock);
rwlock_init(&port->mcast_gate);
@@ -1120,7 +1120,7 @@ static int veth_transmit_to_one(struct s
struct net_device *dev)
{
struct veth_lpar_connection *cnx = veth_cnx[rlp];
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
HvLpEvent_Rc rc;
struct veth_msg *msg = NULL;
unsigned long flags;
@@ -1183,7 +1183,7 @@ static void veth_transmit_to_many(struct
HvLpIndexMap lpmask,
struct net_device *dev)
{
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
int i, success, error;
success = error = 0;
@@ -1210,7 +1210,7 @@ static void veth_transmit_to_many(struct
static int veth_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
unsigned char *frame = skb->data;
- struct veth_port *port = (struct veth_port *) dev->priv;
+ struct veth_port *port = netdev_priv(dev);
HvLpIndexMap lpmask;
if (! (frame[0] & 0x01)) {
@@ -1274,7 +1274,7 @@ static void veth_wake_queues(struct veth
if (! dev)
continue;
- port = (struct veth_port *)dev->priv;
+ port = netdev_priv(dev);
if (! (port->lpar_map & (1<<cnx->remote_lp)))
continue;
@@ -1303,7 +1303,7 @@ static void veth_stop_queues(struct veth
if (! dev)
continue;
- port = (struct veth_port *)dev->priv;
+ port = netdev_priv(dev);
/* If this cnx is not on the vlan for this port, continue */
if (! (port->lpar_map & (1 << cnx->remote_lp)))
@@ -1525,7 +1525,7 @@ static void veth_receive(struct veth_lpa
continue;
}
- port = (struct veth_port *)dev->priv;
+ port = netdev_priv(dev);
dest = *((u64 *) skb->data) & 0xFFFFFFFFFFFF0000;
if ((vlan > HVMAXARCHITECTEDVIRTUALLANS) || !port) {
diff --git a/drivers/net/lp486e.c b/drivers/net/lp486e.c
index 5fc18da..673a780 100644
--- a/drivers/net/lp486e.c
+++ b/drivers/net/lp486e.c
@@ -392,7 +392,7 @@ i596_timeout(struct net_device *dev, cha
struct i596_private *lp;
int boguscnt = ct;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
while (lp->scb.command) {
if (--boguscnt == 0) {
printk("%s: %s timed out - stat %4.4x, cmd %4.4x\n",
@@ -413,7 +413,7 @@ init_rx_bufs(struct net_device *dev, int
int i;
// struct i596_rbd *rbd;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
lp->scb.pa_rfd = I596_NULL;
for (i = 0; i < num; i++) {
@@ -470,7 +470,7 @@ remove_rx_bufs(struct net_device *dev) {
struct i596_private *lp;
struct i596_rfd *rfd;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
lp->rx_tail->pa_next = I596_NULL;
do {
@@ -545,7 +545,7 @@ #endif
static int
i596_scp_setup(struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
int boguscnt;
/* Setup SCP, ISCP, SCB */
@@ -626,7 +626,7 @@ init_i596(struct net_device *dev) {
if (i596_scp_setup(dev))
return 1;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
lp->scb.command = 0;
memcpy ((void *)lp->i596_config, init_setup, 14);
@@ -709,7 +709,7 @@ #endif
static int
i596_rx(struct net_device *dev) {
- struct i596_private *lp = (struct i596_private *) dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
struct i596_rfd *rfd;
int frames = 0;
@@ -742,7 +742,7 @@ i596_cleanup_cmd(struct net_device *dev)
struct i596_private *lp;
struct i596_cmd *cmd;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
while (lp->cmd_head) {
cmd = (struct i596_cmd *)lp->cmd_head;
@@ -810,7 +810,7 @@ static void i596_reset(struct net_device
}
static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
int ioaddr = dev->base_addr;
unsigned long flags;
@@ -869,7 +869,7 @@ static int i596_open(struct net_device *
}
static int i596_start_xmit (struct sk_buff *skb, struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
struct tx_cmd *tx_cmd;
short length;
@@ -917,7 +917,7 @@ static int i596_start_xmit (struct sk_bu
static void
i596_tx_timeout (struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
int ioaddr = dev->base_addr;
/* Transmitter timeout, serious problems. */
@@ -975,7 +975,7 @@ static int __init lp486e_probe(struct ne
return -EBUSY;
}
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
spin_lock_init(&lp->cmd_lock);
/*
@@ -1156,7 +1156,7 @@ i596_interrupt (int irq, void *dev_insta
unsigned short status, ack_cmd = 0;
int frames_in = 0;
- lp = (struct i596_private *) dev->priv;
+ lp = netdev_priv(dev);
/*
* The 82596 examines the command, performs the required action,
@@ -1222,7 +1222,7 @@ #endif
}
static int i596_close(struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
netif_stop_queue(dev);
@@ -1245,7 +1245,7 @@ static int i596_close(struct net_device
}
static struct net_device_stats * i596_get_stats(struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
return &lp->stats;
}
@@ -1255,7 +1255,7 @@ static struct net_device_stats * i596_ge
*/
static void set_multicast_list(struct net_device *dev) {
- struct i596_private *lp = dev->priv;
+ struct i596_private *lp = netdev_priv(dev);
struct i596_cmd *cmd;
if (i596_debug > 1)
diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 1799eee..85fca52 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -2709,7 +2709,7 @@ static int mv643xx_get_stats_count(struc
static void mv643xx_get_ethtool_stats(struct net_device *netdev,
struct ethtool_stats *stats, uint64_t *data)
{
- struct mv643xx_private *mp = netdev->priv;
+ struct mv643xx_private *mp = netdev_priv(netdev);
int i;
eth_update_mib_counters(mp);
diff --git a/drivers/net/mvme147.c b/drivers/net/mvme147.c
index e246d00..052ce2d 100644
--- a/drivers/net/mvme147.c
+++ b/drivers/net/mvme147.c
@@ -110,7 +110,7 @@ struct net_device * __init mvme147lance_
dev->dev_addr[3], dev->dev_addr[4],
dev->dev_addr[5]);
- lp = (struct m147lance_private *)dev->priv;
+ lp = netdev_priv(dev);
lp->ram = __get_dma_pages(GFP_ATOMIC, 3); /* 16K */
if (!lp->ram)
{
@@ -194,7 +194,7 @@ int __init init_module(void)
void __exit cleanup_module(void)
{
- struct m147lance_private *lp = dev_mvme147_lance->priv;
+ struct m147lance_private *lp = netdev_priv(dev_mvme147_lance);
unregister_netdev(dev_mvme147_lance);
free_pages(lp->ram, 3);
free_netdev(dev_mvme147_lance);
diff --git a/drivers/net/myri_sbus.c b/drivers/net/myri_sbus.c
index 13444da..4f7919d 100644
--- a/drivers/net/myri_sbus.c
+++ b/drivers/net/myri_sbus.c
@@ -538,7 +538,7 @@ static void myri_rx(struct myri_eth *mp,
static irqreturn_t myri_interrupt(int irq, void *dev_id)
{
struct net_device *dev = (struct net_device *) dev_id;
- struct myri_eth *mp = (struct myri_eth *) dev->priv;
+ struct myri_eth *mp = netdev_priv(dev);
void __iomem *lregs = mp->lregs;
struct myri_channel __iomem *chan = &mp->shmem->channel;
unsigned long flags;
@@ -577,14 +577,14 @@ static irqreturn_t myri_interrupt(int ir
static int myri_open(struct net_device *dev)
{
- struct myri_eth *mp = (struct myri_eth *) dev->priv;
+ struct myri_eth *mp = netdev_priv(dev);
return myri_init(mp, in_interrupt());
}
static int myri_close(struct net_device *dev)
{
- struct myri_eth *mp = (struct myri_eth *) dev->priv;
+ struct myri_eth *mp = netdev_priv(dev);
myri_clean_rings(mp);
return 0;
@@ -592,7 +592,7 @@ static int myri_close(struct net_device
static void myri_tx_timeout(struct net_device *dev)
{
- struct myri_eth *mp = (struct myri_eth *) dev->priv;
+ struct myri_eth *mp = netdev_priv(dev);
printk(KERN_ERR "%s: transmit timed out, resetting\n", dev->name);
@@ -603,7 +603,7 @@ static void myri_tx_timeout(struct net_d
static int myri_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct myri_eth *mp = (struct myri_eth *) dev->priv;
+ struct myri_eth *mp = netdev_priv(dev);
struct sendq __iomem *sq = mp->sq;
struct myri_txd __iomem *txd;
unsigned long flags;
@@ -807,7 +807,7 @@ static int myri_change_mtu(struct net_de
}
static struct net_device_stats *myri_get_stats(struct net_device *dev)
-{ return &(((struct myri_eth *)dev->priv)->enet_stats); }
+{ return &(((struct myri_eth *)netdev_priv(dev))->enet_stats); }
static void myri_set_multicast(struct net_device *dev)
{
@@ -911,7 +911,7 @@ static int __devinit myri_ether_init(str
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &sdev->ofdev.dev);
- mp = (struct myri_eth *) dev->priv;
+ mp = netdev_priv(dev);
spin_lock_init(&mp->irq_lock);
mp->myri_sdev = sdev;
diff --git a/drivers/net/ne2k-pci.c b/drivers/net/ne2k-pci.c
index f81d939..40039d4 100644
--- a/drivers/net/ne2k-pci.c
+++ b/drivers/net/ne2k-pci.c
@@ -626,7 +626,7 @@ #endif
static void ne2k_pci_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
- struct ei_device *ei = dev->priv;
+ struct ei_device *ei = netdev_priv(dev);
struct pci_dev *pci_dev = (struct pci_dev *) ei->priv;
strcpy(info->driver, DRV_NAME);
diff --git a/drivers/net/ni5010.c b/drivers/net/ni5010.c
index 22a3b3d..a11e1a1 100644
--- a/drivers/net/ni5010.c
+++ b/drivers/net/ni5010.c
@@ -331,7 +331,7 @@ #endif /* JUMPERED_INTERRUPTS */
outb(0, IE_RBUF); /* set buffer byte 0 to 0 again */
}
printk("-> bufsize rcv/xmt=%d/%d\n", bufsize_rcv, NI5010_BUFSIZE);
- memset(dev->priv, 0, sizeof(struct ni5010_local));
+ memset(netdev_priv(dev), 0, sizeof(struct ni5010_local));
dev->open = ni5010_open;
dev->stop = ni5010_close;
diff --git a/drivers/net/ni52.c b/drivers/net/ni52.c
index 5e7999d..242ea31 100644
--- a/drivers/net/ni52.c
+++ b/drivers/net/ni52.c
@@ -324,7 +324,7 @@ static int check586(struct net_device *d
*/
static void alloc586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
ni_reset586();
DELAY(1);
@@ -495,17 +495,17 @@ #else
dev->mem_end = dev->mem_start + size; /* set mem_end showed by 'ifconfig' */
#endif
- memset((char *) dev->priv,0,sizeof(struct priv));
+ memset((char *) netdev_priv(dev),0,sizeof(struct priv));
- ((struct priv *) (dev->priv))->memtop = isa_bus_to_virt(dev->mem_start) + size;
- ((struct priv *) (dev->priv))->base = (unsigned long) isa_bus_to_virt(dev->mem_start) + size - 0x01000000;
+ ((struct priv *) (netdev_priv(dev)))->memtop = isa_bus_to_virt(dev->mem_start) + size;
+ ((struct priv *) (netdev_priv(dev)))->base = (unsigned long) isa_bus_to_virt(dev->mem_start) + size - 0x01000000;
alloc586(dev);
/* set number of receive-buffs according to memsize */
if(size == 0x2000)
- ((struct priv *) dev->priv)->num_recv_buffs = NUM_RECV_BUFFS_8;
+ ((struct priv *)netdev_priv(dev))->num_recv_buffs = NUM_RECV_BUFFS_8;
else
- ((struct priv *) dev->priv)->num_recv_buffs = NUM_RECV_BUFFS_16;
+ ((struct priv *)netdev_priv(dev))->num_recv_buffs = NUM_RECV_BUFFS_16;
printk("Memaddr: 0x%lx, Memsize: %d, ",dev->mem_start,size);
@@ -558,7 +558,7 @@ static int init586(struct net_device *de
{
void *ptr;
int i,result=0;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
volatile struct configure_cmd_struct *cfg_cmd;
volatile struct iasetup_cmd_struct *ias_cmd;
volatile struct tdr_cmd_struct *tdr_cmd;
@@ -794,7 +794,7 @@ static void *alloc_rfa(struct net_device
volatile struct rfd_struct *rfd = (struct rfd_struct *)ptr;
volatile struct rbd_struct *rbd;
int i;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
memset((char *) rfd,0,sizeof(struct rfd_struct)*(p->num_recv_buffs+rfdadd));
p->rfd_first = rfd;
@@ -846,7 +846,7 @@ static irqreturn_t ni52_interrupt(int ir
printk ("ni5210-interrupt: irq %d for unknown device.\n",irq);
return IRQ_NONE;
}
- p = (struct priv *) dev->priv;
+ p = netdev_priv(dev);
if(debuglevel > 1)
printk("I");
@@ -916,7 +916,7 @@ static void ni52_rcv_int(struct net_devi
unsigned short totlen;
struct sk_buff *skb;
struct rbd_struct *rbd;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(debuglevel > 0)
printk("R");
@@ -1036,7 +1036,7 @@ #endif
static void ni52_rnr_int(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
p->stats.rx_errors++;
@@ -1060,7 +1060,7 @@ static void ni52_rnr_int(struct net_devi
static void ni52_xmt_int(struct net_device *dev)
{
int status;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(debuglevel > 0)
printk("X");
@@ -1110,7 +1110,7 @@ #endif
static void startrecv586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
WAIT_4_SCB_CMD();
WAIT_4_SCB_CMD_RUC();
@@ -1122,7 +1122,7 @@ static void startrecv586(struct net_devi
static void ni52_timeout(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
#ifndef NO_NOPCOMMANDS
if(p->scb->cus & CU_ACTIVE) /* COMMAND-UNIT active? */
{
@@ -1164,7 +1164,7 @@ static int ni52_send_packet(struct sk_bu
#ifndef NO_NOPCOMMANDS
int next_nop;
#endif
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(skb->len > XMIT_BUFF_SIZE)
{
@@ -1273,7 +1273,7 @@ #endif
static struct net_device_stats *ni52_get_stats(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
unsigned short crc,aln,rsc,ovrn;
crc = p->scb->crc_errs; /* get error-statistic from the ni82586 */
diff --git a/drivers/net/pci-skeleton.c b/drivers/net/pci-skeleton.c
index 3cdbe11..ef1902f 100644
--- a/drivers/net/pci-skeleton.c
+++ b/drivers/net/pci-skeleton.c
@@ -606,7 +606,7 @@ static int __devinit netdrv_init_board (
}
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &pdev->dev);
- tp = dev->priv;
+ tp = netdev_priv(dev);
/* enable device (incl. PCI PM wakeup), and bus-mastering */
rc = pci_enable_device (pdev);
@@ -761,7 +761,7 @@ #endif
return i;
}
- tp = dev->priv;
+ tp = netdev_priv(dev);
assert (ioaddr != NULL);
assert (dev != NULL);
@@ -786,7 +786,7 @@ #endif
dev->base_addr = (unsigned long) ioaddr;
/* dev->priv/tp zeroed and aligned in alloc_etherdev */
- tp = dev->priv;
+ tp = netdev_priv(dev);
/* note: tp->chipset set in netdrv_init_board */
tp->drv_flags = PCI_COMMAND_IO | PCI_COMMAND_MEMORY |
@@ -848,7 +848,7 @@ static void __devexit netdrv_remove_one
assert (dev != NULL);
- np = dev->priv;
+ np = netdev_priv(dev);
assert (np != NULL);
unregister_netdev (dev);
@@ -981,7 +981,7 @@ static void mdio_sync (void *mdio_addr)
static int mdio_read (struct net_device *dev, int phy_id, int location)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *mdio_addr = tp->mmio_addr + Config4;
int mii_cmd = (0xf6 << 10) | (phy_id << 5) | location;
int retval = 0;
@@ -1024,7 +1024,7 @@ static int mdio_read (struct net_device
static void mdio_write (struct net_device *dev, int phy_id, int location,
int value)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *mdio_addr = tp->mmio_addr + Config4;
int mii_cmd =
(0x5002 << 16) | (phy_id << 23) | (location << 18) | value;
@@ -1067,7 +1067,7 @@ static void mdio_write (struct net_devic
static int netdrv_open (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
int retval;
#ifdef NETDRV_DEBUG
void *ioaddr = tp->mmio_addr;
@@ -1128,7 +1128,7 @@ #endif
/* Start the hardware at open or resume. */
static void netdrv_hw_start (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
u32 i;
@@ -1198,7 +1198,7 @@ static void netdrv_hw_start (struct net_
/* Initialize the Rx and Tx rings, along with various 'dev' bits. */
static void netdrv_init_ring (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
int i;
DPRINTK ("ENTER\n");
@@ -1220,7 +1220,7 @@ static void netdrv_init_ring (struct net
static void netdrv_timer (unsigned long data)
{
struct net_device *dev = (struct net_device *) data;
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
int next_tick = 60 * HZ;
int mii_lpa;
@@ -1285,7 +1285,7 @@ static void netdrv_tx_clear (struct netd
static void netdrv_tx_timeout (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
int i;
u8 tmp8;
@@ -1332,7 +1332,7 @@ static void netdrv_tx_timeout (struct ne
static int netdrv_start_xmit (struct sk_buff *skb, struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
int entry;
@@ -1655,7 +1655,7 @@ static void netdrv_weird_interrupt (stru
static irqreturn_t netdrv_interrupt (int irq, void *dev_instance)
{
struct net_device *dev = (struct net_device *) dev_instance;
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
int boguscnt = max_interrupt_work;
void *ioaddr = tp->mmio_addr;
int status = 0, link_changed = 0; /* avoid bogus "uninit" warning */
@@ -1718,7 +1718,7 @@ static irqreturn_t netdrv_interrupt (int
static int netdrv_close (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
unsigned long flags;
@@ -1769,7 +1769,7 @@ static int netdrv_close (struct net_devi
static int netdrv_ioctl (struct net_device *dev, struct ifreq *rq, int cmd)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
struct mii_ioctl_data *data = if_mii(rq);
unsigned long flags;
int rc = 0;
@@ -1810,7 +1810,7 @@ static int netdrv_ioctl (struct net_devi
static struct net_device_stats *netdrv_get_stats (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
DPRINTK ("ENTER\n");
@@ -1837,7 +1837,7 @@ static struct net_device_stats *netdrv_g
static void netdrv_set_rx_mode (struct net_device *dev)
{
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
u32 mc_filter[2]; /* Multicast hash filter */
int i, rx_mode;
@@ -1894,7 +1894,7 @@ #ifdef CONFIG_PM
static int netdrv_suspend (struct pci_dev *pdev, pm_message_t state)
{
struct net_device *dev = pci_get_drvdata (pdev);
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
void *ioaddr = tp->mmio_addr;
unsigned long flags;
@@ -1924,7 +1924,7 @@ static int netdrv_suspend (struct pci_de
static int netdrv_resume (struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata (pdev);
- struct netdrv_private *tp = dev->priv;
+ struct netdrv_private *tp = netdev_priv(dev);
if (!netif_running(dev))
return 0;
diff --git a/drivers/net/plip.c b/drivers/net/plip.c
index 8754cf3..6a8968c 100644
--- a/drivers/net/plip.c
+++ b/drivers/net/plip.c
@@ -236,7 +236,7 @@ static inline void enable_parport_interr
if (dev->irq != -1)
{
struct parport *port =
- ((struct net_local *)dev->priv)->pardev->port;
+ ((struct net_local *)netdev_priv(dev))->pardev->port;
port->ops->enable_irq (port);
}
}
@@ -246,7 +246,7 @@ static inline void disable_parport_inter
if (dev->irq != -1)
{
struct parport *port =
- ((struct net_local *)dev->priv)->pardev->port;
+ ((struct net_local *)netdev_priv(dev))->pardev->port;
port->ops->disable_irq (port);
}
}
@@ -254,7 +254,7 @@ static inline void disable_parport_inter
static inline void write_data (struct net_device *dev, unsigned char data)
{
struct parport *port =
- ((struct net_local *)dev->priv)->pardev->port;
+ ((struct net_local *)netdev_priv(dev))->pardev->port;
port->ops->write_data (port, data);
}
@@ -262,7 +262,7 @@ static inline void write_data (struct ne
static inline unsigned char read_status (struct net_device *dev)
{
struct parport *port =
- ((struct net_local *)dev->priv)->pardev->port;
+ ((struct net_local *)netdev_priv(dev))->pardev->port;
return port->ops->read_status (port);
}
diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 69da95b..b152f23 100755
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3753,7 +3753,7 @@ static int ql3xxx_open(struct net_device
static struct net_device_stats *ql3xxx_get_stats(struct net_device *dev)
{
- struct ql3_adapter *qdev = (struct ql3_adapter *)dev->priv;
+ struct ql3_adapter *qdev = netdev_priv(dev);
return &qdev->stats;
}
diff --git a/drivers/net/rionet.c b/drivers/net/rionet.c
index df6b738..1a4a37c 100644
--- a/drivers/net/rionet.c
+++ b/drivers/net/rionet.c
@@ -93,7 +93,7 @@ #define RIONET_GET_DESTID(x) (*(u16 *)(x
static struct net_device_stats *rionet_stats(struct net_device *ndev)
{
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
return &rnet->stats;
}
@@ -101,7 +101,7 @@ static int rionet_rx_clean(struct net_de
{
int i;
int error = 0;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
void *data;
i = rnet->rx_slot;
@@ -139,7 +139,7 @@ static int rionet_rx_clean(struct net_de
static void rionet_rx_fill(struct net_device *ndev, int end)
{
int i;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
i = rnet->rx_slot;
do {
@@ -158,7 +158,7 @@ static void rionet_rx_fill(struct net_de
static int rionet_queue_tx_msg(struct sk_buff *skb, struct net_device *ndev,
struct rio_dev *rdev)
{
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
rio_add_outb_message(rnet->mport, rdev, 0, skb->data, skb->len);
rnet->tx_skb[rnet->tx_slot] = skb;
@@ -182,7 +182,7 @@ static int rionet_queue_tx_msg(struct sk
static int rionet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
{
int i;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
struct ethhdr *eth = (struct ethhdr *)skb->data;
u16 destid;
unsigned long flags;
@@ -221,7 +221,7 @@ static void rionet_dbell_event(struct ri
u16 info)
{
struct net_device *ndev = dev_id;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
struct rionet_peer *peer;
if (netif_msg_intr(rnet))
@@ -249,7 +249,7 @@ static void rionet_inb_msg_event(struct
{
int n;
struct net_device *ndev = dev_id;
- struct rionet_private *rnet = (struct rionet_private *)ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
if (netif_msg_intr(rnet))
printk(KERN_INFO "%s: inbound message event, mbox %d slot %d\n",
@@ -264,7 +264,7 @@ static void rionet_inb_msg_event(struct
static void rionet_outb_msg_event(struct rio_mport *mport, void *dev_id, int mbox, int slot)
{
struct net_device *ndev = dev_id;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
spin_lock(&rnet->lock);
@@ -293,7 +293,7 @@ static int rionet_open(struct net_device
int i, rc = 0;
struct rionet_peer *peer, *tmp;
u32 pwdcsr;
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
if (netif_msg_ifup(rnet))
printk(KERN_INFO "%s: open\n", DRV_NAME);
@@ -357,7 +357,7 @@ static int rionet_open(struct net_device
static int rionet_close(struct net_device *ndev)
{
- struct rionet_private *rnet = (struct rionet_private *)ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
struct rionet_peer *peer, *tmp;
int i;
@@ -404,7 +404,7 @@ static void rionet_remove(struct rio_dev
static void rionet_get_drvinfo(struct net_device *ndev,
struct ethtool_drvinfo *info)
{
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
@@ -414,14 +414,14 @@ static void rionet_get_drvinfo(struct ne
static u32 rionet_get_msglevel(struct net_device *ndev)
{
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
return rnet->msg_enable;
}
static void rionet_set_msglevel(struct net_device *ndev, u32 value)
{
- struct rionet_private *rnet = ndev->priv;
+ struct rionet_private *rnet = netdev_priv(ndev);
rnet->msg_enable = value;
}
@@ -450,7 +450,7 @@ static int rionet_setup_netdev(struct ri
}
/* Set up private area */
- rnet = (struct rionet_private *)ndev->priv;
+ rnet = netdev_priv(ndev);
rnet->mport = mport;
/* Set the default MAC address */
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 2be0a0f..7743bd2 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -330,7 +330,7 @@ #define S2IO_TIMER_CONF(timer, handle, a
static void s2io_vlan_rx_register(struct net_device *dev,
struct vlan_group *grp)
{
- struct s2io_nic *nic = dev->priv;
+ struct s2io_nic *nic = netdev_priv(dev);
unsigned long flags;
spin_lock_irqsave(&nic->tx_lock, flags);
@@ -2581,7 +2581,7 @@ static void free_rx_buffers(struct s2io_
static int s2io_poll(struct net_device *dev, int *budget)
{
- struct s2io_nic *nic = dev->priv;
+ struct s2io_nic *nic = netdev_priv(dev);
int pkt_cnt = 0, org_pkts_to_process;
struct mac_info *mac_control;
struct config_param *config;
@@ -2655,7 +2655,7 @@ #ifdef CONFIG_NET_POLL_CONTROLLER
*/
static void s2io_netpoll(struct net_device *dev)
{
- struct s2io_nic *nic = dev->priv;
+ struct s2io_nic *nic = netdev_priv(dev);
struct mac_info *mac_control;
struct config_param *config;
struct XENA_dev_config __iomem *bar0 = nic->bar0;
@@ -2926,7 +2926,7 @@ static void tx_intr_handler(struct fifo_
static void s2io_mdio_write(u32 mmd_type, u64 addr, u16 value, struct net_device *dev)
{
u64 val64 = 0x0;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
//address transaction
@@ -2975,7 +2975,7 @@ static u64 s2io_mdio_read(u32 mmd_type,
{
u64 val64 = 0x0;
u64 rval64 = 0x0;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
/* address transaction */
@@ -3079,7 +3079,7 @@ static void s2io_updt_xpak_counter(struc
u64 val64 = 0x0;
u64 addr = 0x0;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct stat_block *stat_info = sp->mac_control.stats_info;
/* Check the communication with the MDIO slave */
@@ -3799,7 +3799,7 @@ static int s2io_enable_msi_x(struct s2io
static int s2io_open(struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
int err = 0;
/*
@@ -3858,7 +3858,7 @@ hw_init_failed:
static int s2io_close(struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
netif_stop_queue(dev);
/* Reset card, kill tasklet and free Tx and Rx buffers. */
@@ -3882,7 +3882,7 @@ static int s2io_close(struct net_device
static int s2io_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
u16 frg_cnt, frg_len, i, queue, queue_len, put_off, get_off;
register u64 val64;
struct TxD *txdp;
@@ -4194,7 +4194,7 @@ static void s2io_txpic_intr_handle(struc
static irqreturn_t s2io_isr(int irq, void *dev_id)
{
struct net_device *dev = (struct net_device *) dev_id;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
int i;
u64 reason = 0;
@@ -4323,7 +4323,7 @@ static void s2io_updt_stats(struct s2io_
static struct net_device_stats *s2io_get_stats(struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct mac_info *mac_control;
struct config_param *config;
@@ -4365,7 +4365,7 @@ static void s2io_set_multicast(struct ne
{
int i, j, prev_cnt;
struct dev_mc_list *mclist;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
u64 val64 = 0, multi_mac = 0x010203040506ULL, mask =
0xfeffffffffffULL;
@@ -4537,7 +4537,7 @@ static void s2io_set_multicast(struct ne
static int s2io_set_mac_addr(struct net_device *dev, u8 * addr)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
register u64 val64, mac_addr = 0;
int i;
@@ -4601,7 +4601,7 @@ static int s2io_set_mac_addr(struct net_
static int s2io_ethtool_sset(struct net_device *dev,
struct ethtool_cmd *info)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
if ((info->autoneg == AUTONEG_ENABLE) ||
(info->speed != SPEED_10000) || (info->duplex != DUPLEX_FULL))
return -EINVAL;
@@ -4627,7 +4627,7 @@ static int s2io_ethtool_sset(struct net_
static int s2io_ethtool_gset(struct net_device *dev, struct ethtool_cmd *info)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
info->supported = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE);
info->advertising = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE);
info->port = PORT_FIBRE;
@@ -4660,7 +4660,7 @@ static int s2io_ethtool_gset(struct net_
static void s2io_ethtool_gdrvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
strncpy(info->driver, s2io_driver_name, sizeof(info->driver));
strncpy(info->version, s2io_driver_version, sizeof(info->version));
@@ -4696,7 +4696,7 @@ static void s2io_ethtool_gregs(struct ne
int i;
u64 reg;
u8 *reg_space = (u8 *) space;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
regs->len = XENA_REG_SPACE;
regs->version = sp->pdev->subsystem_device;
@@ -4756,7 +4756,7 @@ static void s2io_phy_id(unsigned long da
static int s2io_ethtool_idnic(struct net_device *dev, u32 data)
{
u64 val64 = 0, last_gpio_ctrl_val;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
u16 subid;
@@ -4794,7 +4794,7 @@ static int s2io_ethtool_idnic(struct net
static void s2io_ethtool_gringparam(struct net_device *dev,
struct ethtool_ringparam *ering)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
int i,tx_desc_count=0,rx_desc_count=0;
if (sp->rxd_mode == RXD_MODE_1)
@@ -4837,7 +4837,7 @@ static void s2io_ethtool_getpause_data(s
struct ethtool_pauseparam *ep)
{
u64 val64;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
val64 = readq(&bar0->rmac_pause_cfg);
@@ -4864,7 +4864,7 @@ static int s2io_ethtool_setpause_data(st
struct ethtool_pauseparam *ep)
{
u64 val64;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct XENA_dev_config __iomem *bar0 = sp->bar0;
val64 = readq(&bar0->rmac_pause_cfg);
@@ -5094,7 +5094,7 @@ static int s2io_ethtool_geeprom(struct n
{
u32 i, valid;
u64 data;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
eeprom->magic = sp->pdev->vendor | (sp->pdev->device << 16);
@@ -5132,7 +5132,7 @@ static int s2io_ethtool_seeprom(struct n
{
int len = eeprom->len, cnt = 0;
u64 valid = 0, data;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
if (eeprom->magic != (sp->pdev->vendor | (sp->pdev->device << 16))) {
DBG_PRINT(ERR_DBG,
@@ -5512,7 +5512,7 @@ static void s2io_ethtool_test(struct net
struct ethtool_test *ethtest,
uint64_t * data)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
int orig_state = netif_running(sp->dev);
if (ethtest->flags == ETH_TEST_FL_OFFLINE) {
@@ -5568,7 +5568,7 @@ static void s2io_get_ethtool_stats(struc
u64 * tmp_stats)
{
int i = 0;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
struct stat_block *stat_info = sp->mac_control.stats_info;
s2io_updt_stats(sp);
@@ -5829,14 +5829,14 @@ static int s2io_ethtool_get_regs_len(str
static u32 s2io_ethtool_get_rx_csum(struct net_device * dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
return (sp->rx_csum);
}
static int s2io_ethtool_set_rx_csum(struct net_device *dev, u32 data)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
if (data)
sp->rx_csum = 1;
@@ -5860,7 +5860,7 @@ static void s2io_ethtool_get_strings(str
u32 stringset, u8 * data)
{
int stat_size = 0;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
switch (stringset) {
case ETH_SS_TEST:
@@ -5882,7 +5882,7 @@ static void s2io_ethtool_get_strings(str
}
static int s2io_ethtool_get_stats_count(struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
int stat_count = 0;
switch(sp->device_type) {
case XFRAME_I_DEVICE:
@@ -5982,7 +5982,7 @@ static int s2io_ioctl(struct net_device
static int s2io_change_mtu(struct net_device *dev, int new_mtu)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
if ((new_mtu < MIN_MTU) || (new_mtu > S2IO_JUMBO_SIZE)) {
DBG_PRINT(ERR_DBG, "%s: MTU size is invalid.\n",
@@ -6026,7 +6026,7 @@ static int s2io_change_mtu(struct net_de
static void s2io_tasklet(unsigned long dev_addr)
{
struct net_device *dev = (struct net_device *) dev_addr;
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
int i, ret;
struct mac_info *mac_control;
struct config_param *config;
@@ -6621,7 +6621,7 @@ out_unlock:
static void s2io_tx_watchdog(struct net_device *dev)
{
- struct s2io_nic *sp = dev->priv;
+ struct s2io_nic *sp = netdev_priv(dev);
if (netif_carrier_ok(dev)) {
sp->mac_control.stats_info->sw_stat.watchdog_timer_cnt++;
@@ -7077,7 +7077,7 @@ s2io_init_nic(struct pci_dev *pdev, cons
SET_NETDEV_DEV(dev, &pdev->dev);
/* Private member variable initialized to s2io NIC structure */
- sp = dev->priv;
+ sp = netdev_priv(dev);
memset(sp, 0, sizeof(struct s2io_nic));
sp->dev = dev;
sp->pdev = pdev;
@@ -7438,7 +7438,7 @@ static void __devexit s2io_rem_nic(struc
flush_scheduled_work();
- sp = dev->priv;
+ sp = netdev_priv(dev);
unregister_netdev(dev);
free_shared_mem(sp);
@@ -7806,7 +7806,7 @@ static pci_ers_result_t s2io_io_error_de
pci_channel_state_t state)
{
struct net_device *netdev = pci_get_drvdata(pdev);
- struct s2io_nic *sp = netdev->priv;
+ struct s2io_nic *sp = netdev_priv(netdev);
netif_device_detach(netdev);
@@ -7831,7 +7831,7 @@ static pci_ers_result_t s2io_io_error_de
static pci_ers_result_t s2io_io_slot_reset(struct pci_dev *pdev)
{
struct net_device *netdev = pci_get_drvdata(pdev);
- struct s2io_nic *sp = netdev->priv;
+ struct s2io_nic *sp = netdev_priv(netdev);
if (pci_enable_device(pdev)) {
printk(KERN_ERR "s2io: "
@@ -7855,7 +7855,7 @@ static pci_ers_result_t s2io_io_slot_res
static void s2io_io_resume(struct pci_dev *pdev)
{
struct net_device *netdev = pci_get_drvdata(pdev);
- struct s2io_nic *sp = netdev->priv;
+ struct s2io_nic *sp = netdev_priv(netdev);
if (netif_running(netdev)) {
if (s2io_card_up(sp)) {
diff --git a/drivers/net/sgiseeq.c b/drivers/net/sgiseeq.c
index 384b468..d69b2d7 100644
--- a/drivers/net/sgiseeq.c
+++ b/drivers/net/sgiseeq.c
@@ -577,7 +577,7 @@ static struct net_device_stats *sgiseeq_
static void sgiseeq_set_multicast(struct net_device *dev)
{
- struct sgiseeq_private *sp = (struct sgiseeq_private *) dev->priv;
+ struct sgiseeq_private *sp = netdev_priv(dev);
unsigned char oldmode = sp->mode;
if(dev->flags & IFF_PROMISC)
diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c
index 4c3d98f..14bdba0 100644
--- a/drivers/net/shaper.c
+++ b/drivers/net/shaper.c
@@ -133,7 +133,7 @@ static void shaper_setspeed(struct shape
static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct shaper *shaper = dev->priv;
+ struct shaper *shaper = netdev_priv(dev);
struct sk_buff *ptr;
spin_lock(&shaper->lock);
@@ -288,7 +288,7 @@ static void shaper_kick(struct shaper *s
static int shaper_open(struct net_device *dev)
{
- struct shaper *shaper=dev->priv;
+ struct shaper *shaper=netdev_priv(dev);
/*
* Can't open until attached.
@@ -309,7 +309,7 @@ static int shaper_open(struct net_device
static int shaper_close(struct net_device *dev)
{
- struct shaper *shaper=dev->priv;
+ struct shaper *shaper=netdev_priv(dev);
struct sk_buff *skb;
while ((skb = skb_dequeue(&shaper->sendq)) != NULL)
@@ -331,14 +331,14 @@ static int shaper_close(struct net_devic
static struct net_device_stats *shaper_get_stats(struct net_device *dev)
{
- struct shaper *sh=dev->priv;
+ struct shaper *sh=netdev_priv(dev);
return &sh->stats;
}
static int shaper_header(struct sk_buff *skb, struct net_device *dev,
unsigned short type, void *daddr, void *saddr, unsigned len)
{
- struct shaper *sh=dev->priv;
+ struct shaper *sh=netdev_priv(dev);
int v;
if(sh_debug)
printk("Shaper header\n");
@@ -476,7 +476,7 @@ #endif
static int shaper_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
struct shaperconf *ss= (struct shaperconf *)&ifr->ifr_ifru;
- struct shaper *sh=dev->priv;
+ struct shaper *sh=netdev_priv(dev);
if(ss->ss_cmd == SHAPER_SET_DEV || ss->ss_cmd == SHAPER_SET_SPEED)
{
@@ -493,7 +493,7 @@ static int shaper_ioctl(struct net_devic
return -ENODEV;
if(sh->dev)
return -EBUSY;
- return shaper_attach(dev,dev->priv, them);
+ return shaper_attach(dev,netdev_priv(dev), them);
}
case SHAPER_GET_DEV:
if(sh->dev==NULL)
@@ -513,7 +513,7 @@ static int shaper_ioctl(struct net_devic
static void shaper_init_priv(struct net_device *dev)
{
- struct shaper *sh = dev->priv;
+ struct shaper *sh = netdev_priv(dev);
skb_queue_head_init(&sh->sendq);
init_timer(&sh->timer);
diff --git a/drivers/net/sis900.c b/drivers/net/sis900.c
index 7c6e480..36162d5 100644
--- a/drivers/net/sis900.c
+++ b/drivers/net/sis900.c
@@ -439,7 +439,7 @@ #endif
if (ret)
goto err_out;
- sis_priv = net_dev->priv;
+ sis_priv = netdev_priv(net_dev);
net_dev->base_addr = ioaddr;
net_dev->irq = pci_dev->irq;
sis_priv->pci_dev = pci_dev;
@@ -575,7 +575,7 @@ #endif
static int __devinit sis900_mii_probe(struct net_device * net_dev)
{
- struct sis900_private * sis_priv = net_dev->priv;
+ struct sis900_private * sis_priv = netdev_priv(net_dev);
const char *dev_name = pci_name(sis_priv->pci_dev);
u16 poll_bit = MII_STAT_LINK, status = 0;
unsigned long timeout = jiffies + 5 * HZ;
@@ -703,7 +703,7 @@ static int __devinit sis900_mii_probe(st
static u16 sis900_default_phy(struct net_device * net_dev)
{
- struct sis900_private * sis_priv = net_dev->priv;
+ struct sis900_private * sis_priv = netdev_priv(net_dev);
struct mii_phy *phy = NULL, *phy_home = NULL,
*default_phy = NULL, *phy_lan = NULL;
u16 status;
@@ -1004,7 +1004,7 @@ #endif
static int
sis900_open(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
int ret;
@@ -1060,7 +1060,7 @@ sis900_open(struct net_device *net_dev)
static void
sis900_init_rxfilter (struct net_device * net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
u32 rfcrSave;
u32 i;
@@ -1098,7 +1098,7 @@ sis900_init_rxfilter (struct net_device
static void
sis900_init_tx_ring(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
int i;
@@ -1132,7 +1132,7 @@ sis900_init_tx_ring(struct net_device *n
static void
sis900_init_rx_ring(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
int i;
@@ -1203,7 +1203,7 @@ sis900_init_rx_ring(struct net_device *n
static void sis630_set_eq(struct net_device *net_dev, u8 revision)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
u16 reg14h, eq_value=0, max_value=0, min_value=0;
int i, maxcount=10;
@@ -1276,7 +1276,7 @@ static void sis630_set_eq(struct net_dev
static void sis900_timer(unsigned long data)
{
struct net_device *net_dev = (struct net_device *)data;
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
struct mii_phy *mii_phy = sis_priv->mii;
static const int next_tick = 5*HZ;
u16 status;
@@ -1346,7 +1346,7 @@ static void sis900_timer(unsigned long d
static void sis900_check_mode(struct net_device *net_dev, struct mii_phy *mii_phy)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
int speed, duplex;
@@ -1425,7 +1425,7 @@ #endif
static void sis900_auto_negotiate(struct net_device *net_dev, int phy_addr)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
int i = 0;
u32 status;
@@ -1460,7 +1460,7 @@ static void sis900_auto_negotiate(struct
static void sis900_read_mode(struct net_device *net_dev, int *speed, int *duplex)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
struct mii_phy *phy = sis_priv->mii;
int phy_addr = sis_priv->cur_phy;
u32 status;
@@ -1515,7 +1515,7 @@ static void sis900_read_mode(struct net_
static void sis900_tx_timeout(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
unsigned long flags;
int i;
@@ -1574,7 +1574,7 @@ static void sis900_tx_timeout(struct net
static int
sis900_start_xmit(struct sk_buff *skb, struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
unsigned int entry;
unsigned long flags;
@@ -1644,7 +1644,7 @@ sis900_start_xmit(struct sk_buff *skb, s
static irqreturn_t sis900_interrupt(int irq, void *dev_instance)
{
struct net_device *net_dev = dev_instance;
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
int boguscnt = max_interrupt_work;
long ioaddr = net_dev->base_addr;
u32 status;
@@ -1706,7 +1706,7 @@ static irqreturn_t sis900_interrupt(int
static int sis900_rx(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
unsigned int entry = sis_priv->cur_rx % NUM_RX_DESC;
u32 rx_status = sis_priv->rx_ring[entry].cmdsts;
@@ -1856,7 +1856,7 @@ refill_rx_ring:
static void sis900_finish_xmit (struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
for (; sis_priv->dirty_tx != sis_priv->cur_tx; sis_priv->dirty_tx++) {
struct sk_buff *skb;
@@ -1925,7 +1925,7 @@ static void sis900_finish_xmit (struct n
static int sis900_close(struct net_device *net_dev)
{
long ioaddr = net_dev->base_addr;
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
struct sk_buff *skb;
int i;
@@ -1980,7 +1980,7 @@ static int sis900_close(struct net_devic
static void sis900_get_drvinfo(struct net_device *net_dev,
struct ethtool_drvinfo *info)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
strcpy (info->driver, SIS900_MODULE_NAME);
strcpy (info->version, SIS900_DRV_VERSION);
@@ -1989,26 +1989,26 @@ static void sis900_get_drvinfo(struct ne
static u32 sis900_get_msglevel(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
return sis_priv->msg_enable;
}
static void sis900_set_msglevel(struct net_device *net_dev, u32 value)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
sis_priv->msg_enable = value;
}
static u32 sis900_get_link(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
return mii_link_ok(&sis_priv->mii_info);
}
static int sis900_get_settings(struct net_device *net_dev,
struct ethtool_cmd *cmd)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
spin_lock_irq(&sis_priv->lock);
mii_ethtool_gset(&sis_priv->mii_info, cmd);
spin_unlock_irq(&sis_priv->lock);
@@ -2018,7 +2018,7 @@ static int sis900_get_settings(struct ne
static int sis900_set_settings(struct net_device *net_dev,
struct ethtool_cmd *cmd)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
int rt;
spin_lock_irq(&sis_priv->lock);
rt = mii_ethtool_sset(&sis_priv->mii_info, cmd);
@@ -2028,7 +2028,7 @@ static int sis900_set_settings(struct ne
static int sis900_nway_reset(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
return mii_nway_restart(&sis_priv->mii_info);
}
@@ -2045,7 +2045,7 @@ static int sis900_nway_reset(struct net_
static int sis900_set_wol(struct net_device *net_dev, struct ethtool_wolinfo *wol)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long pmctrl_addr = net_dev->base_addr + pmctrl;
u32 cfgpmcsr = 0, pmctrl_bits = 0;
@@ -2116,7 +2116,7 @@ static const struct ethtool_ops sis900_e
static int mii_ioctl(struct net_device *net_dev, struct ifreq *rq, int cmd)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
struct mii_ioctl_data *data = if_mii(rq);
switch(cmd) {
@@ -2148,7 +2148,7 @@ static int mii_ioctl(struct net_device *
static struct net_device_stats *
sis900_get_stats(struct net_device *net_dev)
{
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
return &sis_priv->stats;
}
@@ -2165,7 +2165,7 @@ sis900_get_stats(struct net_device *net_
static int sis900_set_config(struct net_device *dev, struct ifmap *map)
{
- struct sis900_private *sis_priv = dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(dev);
struct mii_phy *mii_phy = sis_priv->mii;
u16 status;
@@ -2288,7 +2288,7 @@ static inline u16 sis900_mcast_bitnr(u8
static void set_rx_mode(struct net_device *net_dev)
{
long ioaddr = net_dev->base_addr;
- struct sis900_private * sis_priv = net_dev->priv;
+ struct sis900_private * sis_priv = netdev_priv(net_dev);
u16 mc_filter[16] = {0}; /* 256/128 bits multicast hash table */
int i, table_entries;
u32 rx_mode;
@@ -2363,7 +2363,7 @@ static void set_rx_mode(struct net_devic
static void sis900_reset(struct net_device *net_dev)
{
- struct sis900_private * sis_priv = net_dev->priv;
+ struct sis900_private * sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
int i = 0;
u32 status = TxRCMP | RxRCMP;
@@ -2396,7 +2396,7 @@ static void sis900_reset(struct net_devi
static void __devexit sis900_remove(struct pci_dev *pci_dev)
{
struct net_device *net_dev = pci_get_drvdata(pci_dev);
- struct sis900_private * sis_priv = net_dev->priv;
+ struct sis900_private * sis_priv = netdev_priv(net_dev);
struct mii_phy *phy = NULL;
while (sis_priv->first_mii) {
@@ -2440,7 +2440,7 @@ static int sis900_suspend(struct pci_dev
static int sis900_resume(struct pci_dev *pci_dev)
{
struct net_device *net_dev = pci_get_drvdata(pci_dev);
- struct sis900_private *sis_priv = net_dev->priv;
+ struct sis900_private *sis_priv = netdev_priv(net_dev);
long ioaddr = net_dev->base_addr;
if(!netif_running(net_dev))
diff --git a/drivers/net/smc9194.c b/drivers/net/smc9194.c
index 36c1eba..af14f24 100644
--- a/drivers/net/smc9194.c
+++ b/drivers/net/smc9194.c
@@ -1052,7 +1052,7 @@ #endif
printk("%2.2x \n", dev->dev_addr[5] );
/* set the private data to zero by default */
- memset(dev->priv, 0, sizeof(struct smc_local));
+ memset(netdev_priv(dev), 0, sizeof(struct smc_local));
/* Grab the IRQ */
retval = request_irq(dev->irq, &smc_interrupt, 0, DRV_NAME, dev);
@@ -1127,7 +1127,7 @@ static int smc_open(struct net_device *d
int i; /* used to set hw ethernet address */
/* clear out all the junk that was put here before... */
- memset(dev->priv, 0, sizeof(struct smc_local));
+ memset(netdev_priv(dev), 0, sizeof(struct smc_local));
/* reset the hardware */
@@ -1183,7 +1183,7 @@ static void smc_timeout(struct net_devic
smc_enable( dev->base_addr );
dev->trans_start = jiffies;
/* clear anything saved */
- ((struct smc_local *)dev->priv)->saved_skb = NULL;
+ ((struct smc_local *)netdev_priv(dev))->saved_skb = NULL;
netif_wake_queue(dev);
}
diff --git a/drivers/net/sonic.h b/drivers/net/sonic.h
index 7db13e4..bbc0ddc 100644
--- a/drivers/net/sonic.h
+++ b/drivers/net/sonic.h
@@ -371,7 +371,7 @@ #endif
static inline void sonic_cda_put(struct net_device* dev, int entry,
int offset, __u16 val)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
sonic_buf_put(lp->cda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_CD) + offset, val);
}
@@ -379,27 +379,27 @@ static inline void sonic_cda_put(struct
static inline __u16 sonic_cda_get(struct net_device* dev, int entry,
int offset)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
return sonic_buf_get(lp->cda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_CD) + offset);
}
static inline void sonic_set_cam_enable(struct net_device* dev, __u16 val)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
sonic_buf_put(lp->cda, lp->dma_bitmode, SONIC_CDA_CAM_ENABLE, val);
}
static inline __u16 sonic_get_cam_enable(struct net_device* dev)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
return sonic_buf_get(lp->cda, lp->dma_bitmode, SONIC_CDA_CAM_ENABLE);
}
static inline void sonic_tda_put(struct net_device* dev, int entry,
int offset, __u16 val)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
sonic_buf_put(lp->tda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_TD) + offset, val);
}
@@ -407,7 +407,7 @@ static inline void sonic_tda_put(struct
static inline __u16 sonic_tda_get(struct net_device* dev, int entry,
int offset)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
return sonic_buf_get(lp->tda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_TD) + offset);
}
@@ -415,7 +415,7 @@ static inline __u16 sonic_tda_get(struct
static inline void sonic_rda_put(struct net_device* dev, int entry,
int offset, __u16 val)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
sonic_buf_put(lp->rda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_RD) + offset, val);
}
@@ -423,7 +423,7 @@ static inline void sonic_rda_put(struct
static inline __u16 sonic_rda_get(struct net_device* dev, int entry,
int offset)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
return sonic_buf_get(lp->rda, lp->dma_bitmode,
(entry * SIZEOF_SONIC_RD) + offset);
}
@@ -431,7 +431,7 @@ static inline __u16 sonic_rda_get(struct
static inline void sonic_rra_put(struct net_device* dev, int entry,
int offset, __u16 val)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
sonic_buf_put(lp->rra, lp->dma_bitmode,
(entry * SIZEOF_SONIC_RR) + offset, val);
}
@@ -439,7 +439,7 @@ static inline void sonic_rra_put(struct
static inline __u16 sonic_rra_get(struct net_device* dev, int entry,
int offset)
{
- struct sonic_local* lp = (struct sonic_local *) dev->priv;
+ struct sonic_local* lp = netdev_priv(dev);
return sonic_buf_get(lp->rra, lp->dma_bitmode,
(entry * SIZEOF_SONIC_RR) + offset);
}
diff --git a/drivers/net/sun3_82586.c b/drivers/net/sun3_82586.c
index b77ab6e..a94f98c 100644
--- a/drivers/net/sun3_82586.c
+++ b/drivers/net/sun3_82586.c
@@ -247,7 +247,7 @@ static int check586(struct net_device *d
*/
static void alloc586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
sun3_reset586();
DELAY(1);
@@ -364,17 +364,17 @@ static int __init sun3_82586_probe1(stru
goto out;
}
- ((struct priv *) (dev->priv))->memtop = (char *)dvma_btov(dev->mem_start);
- ((struct priv *) (dev->priv))->base = (unsigned long) dvma_btov(0);
+ ((struct priv *) (netdev_priv(dev)))->memtop = (char *)dvma_btov(dev->mem_start);
+ ((struct priv *) (netdev_priv(dev)))->base = (unsigned long) dvma_btov(0);
alloc586(dev);
/* set number of receive-buffs according to memsize */
if(size == 0x2000)
- ((struct priv *) dev->priv)->num_recv_buffs = NUM_RECV_BUFFS_8;
+ ((struct priv *)netdev_priv(dev))->num_recv_buffs = NUM_RECV_BUFFS_8;
else if(size == 0x4000)
- ((struct priv *) dev->priv)->num_recv_buffs = NUM_RECV_BUFFS_16;
+ ((struct priv *)netdev_priv(dev))->num_recv_buffs = NUM_RECV_BUFFS_16;
else
- ((struct priv *) dev->priv)->num_recv_buffs = NUM_RECV_BUFFS_32;
+ ((struct priv *)netdev_priv(dev))->num_recv_buffs = NUM_RECV_BUFFS_32;
printk("Memaddr: 0x%lx, Memsize: %d, IRQ %d\n",dev->mem_start,size, dev->irq);
@@ -398,7 +398,7 @@ static int init586(struct net_device *de
{
void *ptr;
int i,result=0;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
volatile struct configure_cmd_struct *cfg_cmd;
volatile struct iasetup_cmd_struct *ias_cmd;
volatile struct tdr_cmd_struct *tdr_cmd;
@@ -635,7 +635,7 @@ static void *alloc_rfa(struct net_device
volatile struct rfd_struct *rfd = (struct rfd_struct *)ptr;
volatile struct rbd_struct *rbd;
int i;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
memset((char *) rfd,0,sizeof(struct rfd_struct)*(p->num_recv_buffs+rfdadd));
p->rfd_first = rfd;
@@ -687,7 +687,7 @@ static irqreturn_t sun3_82586_interrupt(
printk ("sun3_82586-interrupt: irq %d for unknown device.\n",irq);
return IRQ_NONE;
}
- p = (struct priv *) dev->priv;
+ p = netdev_priv(dev);
if(debuglevel > 1)
printk("I");
@@ -757,7 +757,7 @@ static void sun3_82586_rcv_int(struct ne
unsigned short totlen;
struct sk_buff *skb;
struct rbd_struct *rbd;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(debuglevel > 0)
printk("R");
@@ -875,7 +875,7 @@ #endif
static void sun3_82586_rnr_int(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
p->stats.rx_errors++;
@@ -899,7 +899,7 @@ static void sun3_82586_rnr_int(struct ne
static void sun3_82586_xmt_int(struct net_device *dev)
{
int status;
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(debuglevel > 0)
printk("X");
@@ -949,7 +949,7 @@ #endif
static void startrecv586(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
WAIT_4_SCB_CMD();
WAIT_4_SCB_CMD_RUC();
@@ -961,7 +961,7 @@ static void startrecv586(struct net_devi
static void sun3_82586_timeout(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
#ifndef NO_NOPCOMMANDS
if(p->scb->cus & CU_ACTIVE) /* COMMAND-UNIT active? */
{
@@ -1003,7 +1003,7 @@ static int sun3_82586_send_packet(struct
#ifndef NO_NOPCOMMANDS
int next_nop;
#endif
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
if(skb->len > XMIT_BUFF_SIZE)
{
@@ -1112,7 +1112,7 @@ #endif
static struct net_device_stats *sun3_82586_get_stats(struct net_device *dev)
{
- struct priv *p = (struct priv *) dev->priv;
+ struct priv *p = netdev_priv(dev);
unsigned short crc,aln,rsc,ovrn;
crc = swab16(p->scb->crc_errs); /* get error-statistic from the ni82586 */
diff --git a/drivers/net/sunbmac.c b/drivers/net/sunbmac.c
index b3e0158..376ec86 100644
--- a/drivers/net/sunbmac.c
+++ b/drivers/net/sunbmac.c
@@ -914,7 +914,7 @@ static irqreturn_t bigmac_interrupt(int
static int bigmac_open(struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
int ret;
ret = request_irq(dev->irq, &bigmac_interrupt, IRQF_SHARED, dev->name, bp);
@@ -931,7 +931,7 @@ static int bigmac_open(struct net_device
static int bigmac_close(struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
del_timer(&bp->bigmac_timer);
bp->timer_state = asleep;
@@ -945,7 +945,7 @@ static int bigmac_close(struct net_devic
static void bigmac_tx_timeout(struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
bigmac_init_hw(bp, 0);
netif_wake_queue(dev);
@@ -954,7 +954,7 @@ static void bigmac_tx_timeout(struct net
/* Put a packet on the wire. */
static int bigmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
int len, entry;
u32 mapping;
@@ -986,7 +986,7 @@ static int bigmac_start_xmit(struct sk_b
static struct net_device_stats *bigmac_get_stats(struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
bigmac_get_counters(bp, bp->bregs);
return &bp->enet_stats;
@@ -994,7 +994,7 @@ static struct net_device_stats *bigmac_g
static void bigmac_set_multicast(struct net_device *dev)
{
- struct bigmac *bp = (struct bigmac *) dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
void __iomem *bregs = bp->bregs;
struct dev_mc_list *dmi = dev->mc_list;
char *addrs;
@@ -1051,7 +1051,7 @@ static void bigmac_set_multicast(struct
/* Ethtool support... */
static void bigmac_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct bigmac *bp = dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
strcpy(info->driver, "sunbmac");
strcpy(info->version, "2.0");
@@ -1061,7 +1061,7 @@ static void bigmac_get_drvinfo(struct ne
static u32 bigmac_get_link(struct net_device *dev)
{
- struct bigmac *bp = dev->priv;
+ struct bigmac *bp = netdev_priv(dev);
spin_lock_irq(&bp->lock);
bp->sw_bmsr = bigmac_tcvr_read(bp, bp->tregs, BIGMAC_BMSR);
@@ -1097,7 +1097,7 @@ static int __init bigmac_ether_init(stru
dev->dev_addr[i] = idprom->id_ethaddr[i];
/* Setup softc, with backpointers to QEC and BigMAC SBUS device structs. */
- bp = dev->priv;
+ bp = netdev_priv(dev);
bp->qec_sdev = qec_sdev;
bp->bigmac_sdev = qec_sdev->child;
diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c
index 4328038..1896e62 100644
--- a/drivers/net/sungem.c
+++ b/drivers/net/sungem.c
@@ -164,7 +164,7 @@ static u16 __phy_read(struct gem *gp, in
static inline int _phy_read(struct net_device *dev, int mii_id, int reg)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
return __phy_read(gp, mii_id, reg);
}
@@ -197,7 +197,7 @@ static void __phy_write(struct gem *gp,
static inline void _phy_write(struct net_device *dev, int mii_id, int reg, int val)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
__phy_write(gp, mii_id, reg, val & 0xffff);
}
@@ -880,7 +880,7 @@ static int gem_rx(struct gem *gp, int wo
static int gem_poll(struct net_device *dev, int *budget)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
/*
@@ -934,7 +934,7 @@ static int gem_poll(struct net_device *d
static irqreturn_t gem_interrupt(int irq, void *dev_id)
{
struct net_device *dev = dev_id;
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
/* Swallow interrupts when shutting the chip down, though
@@ -980,7 +980,7 @@ #endif
static void gem_tx_timeout(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
printk(KERN_ERR "%s: transmit timed out, resetting\n", dev->name);
if (!gp->running) {
@@ -1019,7 +1019,7 @@ static __inline__ int gem_intme(int entr
static int gem_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
int entry;
u64 ctrl;
unsigned long flags;
@@ -2192,7 +2192,7 @@ static void gem_stop_phy(struct gem *gp,
static int gem_do_start(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
spin_lock_irqsave(&gp->lock, flags);
@@ -2239,7 +2239,7 @@ static int gem_do_start(struct net_devic
static void gem_do_stop(struct net_device *dev, int wol)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
spin_lock_irqsave(&gp->lock, flags);
@@ -2315,7 +2315,7 @@ static void gem_reset_task(struct work_s
static int gem_open(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
int rc = 0;
mutex_lock(&gp->pm_mutex);
@@ -2332,7 +2332,7 @@ static int gem_open(struct net_device *d
static int gem_close(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
/* Note: we don't need to call netif_poll_disable() here because
* our caller (dev_close) already did it for us
@@ -2353,7 +2353,7 @@ #ifdef CONFIG_PM
static int gem_suspend(struct pci_dev *pdev, pm_message_t state)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
mutex_lock(&gp->pm_mutex);
@@ -2417,7 +2417,7 @@ static int gem_suspend(struct pci_dev *p
static int gem_resume(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned long flags;
printk(KERN_INFO "%s: resuming\n", dev->name);
@@ -2492,7 +2492,7 @@ #endif /* CONFIG_PM */
static struct net_device_stats *gem_get_stats(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
struct net_device_stats *stats = &gp->net_stats;
spin_lock_irq(&gp->lock);
@@ -2528,7 +2528,7 @@ static struct net_device_stats *gem_get_
static int gem_set_mac_address(struct net_device *dev, void *addr)
{
struct sockaddr *macaddr = (struct sockaddr *) addr;
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unsigned char *e = &dev->dev_addr[0];
if (!is_valid_ether_addr(macaddr->sa_data))
@@ -2556,7 +2556,7 @@ static int gem_set_mac_address(struct ne
static void gem_set_multicast(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
u32 rxcfg, rxcfg_new;
int limit = 10000;
@@ -2605,7 +2605,7 @@ #endif
static int gem_change_mtu(struct net_device *dev, int new_mtu)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
if (new_mtu < GEM_MIN_MTU || new_mtu > GEM_MAX_MTU)
return -EINVAL;
@@ -2636,7 +2636,7 @@ static int gem_change_mtu(struct net_dev
static void gem_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
@@ -2645,7 +2645,7 @@ static void gem_get_drvinfo(struct net_d
static int gem_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
if (gp->phy_type == phy_mii_mdio0 ||
gp->phy_type == phy_mii_mdio1) {
@@ -2691,7 +2691,7 @@ static int gem_get_settings(struct net_d
static int gem_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
/* Verify the settings we care about. */
if (cmd->autoneg != AUTONEG_ENABLE &&
@@ -2722,7 +2722,7 @@ static int gem_set_settings(struct net_d
static int gem_nway_reset(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
if (!gp->want_autoneg)
return -EINVAL;
@@ -2739,13 +2739,13 @@ static int gem_nway_reset(struct net_dev
static u32 gem_get_msglevel(struct net_device *dev)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
return gp->msg_enable;
}
static void gem_set_msglevel(struct net_device *dev, u32 value)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
gp->msg_enable = value;
}
@@ -2757,7 +2757,7 @@ #define WOL_SUPPORTED_MASK (WAKE_MAGIC)
static void gem_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
/* Add more when I understand how to program the chip */
if (gp->has_wol) {
@@ -2771,7 +2771,7 @@ static void gem_get_wol(struct net_devic
static int gem_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
if (!gp->has_wol)
return -EOPNOTSUPP;
@@ -2793,7 +2793,7 @@ static const struct ethtool_ops gem_etht
static int gem_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
struct mii_ioctl_data *data = if_mii(ifr);
int rc = -EOPNOTSUPP;
unsigned long flags;
@@ -2925,7 +2925,7 @@ static void gem_remove_one(struct pci_de
struct net_device *dev = pci_get_drvdata(pdev);
if (dev) {
- struct gem *gp = dev->priv;
+ struct gem *gp = netdev_priv(dev);
unregister_netdev(dev);
@@ -3029,7 +3029,7 @@ static int __devinit gem_init_one(struct
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &pdev->dev);
- gp = dev->priv;
+ gp = netdev_priv(dev);
err = pci_request_regions(pdev, DRV_NAME);
if (err) {
diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c
index 8b35f13..1049740 100644
--- a/drivers/net/sunhme.c
+++ b/drivers/net/sunhme.c
@@ -2136,7 +2136,7 @@ static irqreturn_t quattro_sbus_interrup
for (i = 0; i < 4; i++) {
struct net_device *dev = qp->happy_meals[i];
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
u32 happy_status = hme_read32(hp, hp->gregs + GREG_STAT);
HMD(("quattro_interrupt: status=%08x ", happy_status));
@@ -2181,7 +2181,7 @@ #endif
static int happy_meal_open(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
int res;
HMD(("happy_meal_open: "));
@@ -2213,7 +2213,7 @@ static int happy_meal_open(struct net_de
static int happy_meal_close(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
spin_lock_irq(&hp->happy_lock);
happy_meal_stop(hp, hp->gregs);
@@ -2242,7 +2242,7 @@ #endif
static void happy_meal_tx_timeout(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
printk (KERN_ERR "%s: transmit timed out, resetting\n", dev->name);
tx_dump_log();
@@ -2260,7 +2260,7 @@ static void happy_meal_tx_timeout(struct
static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
int entry;
u32 tx_flags;
@@ -2349,7 +2349,7 @@ static int happy_meal_start_xmit(struct
static struct net_device_stats *happy_meal_get_stats(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
spin_lock_irq(&hp->happy_lock);
happy_meal_get_counters(hp, hp->bigmacregs);
@@ -2360,7 +2360,7 @@ static struct net_device_stats *happy_me
static void happy_meal_set_multicast(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
void __iomem *bregs = hp->bigmacregs;
struct dev_mc_list *dmi = dev->mc_list;
char *addrs;
@@ -2410,7 +2410,7 @@ static void happy_meal_set_multicast(str
/* Ethtool support... */
static int hme_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
cmd->supported =
(SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full |
@@ -2455,7 +2455,7 @@ static int hme_get_settings(struct net_d
static int hme_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
/* Verify the settings we care about. */
if (cmd->autoneg != AUTONEG_ENABLE &&
@@ -2479,7 +2479,7 @@ static int hme_set_settings(struct net_d
static void hme_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
strcpy(info->driver, "sunhme");
strcpy(info->version, "2.02");
@@ -2498,7 +2498,7 @@ #endif
static u32 hme_get_link(struct net_device *dev)
{
- struct happy_meal *hp = dev->priv;
+ struct happy_meal *hp = netdev_priv(dev);
spin_lock_irq(&hp->happy_lock);
hp->sw_bmcr = happy_meal_tcvr_read(hp, hp->tcvregs, MII_BMCR);
@@ -2709,7 +2709,7 @@ static int __devinit happy_meal_sbus_pro
memcpy(dev->dev_addr, idprom->id_ethaddr, 6);
}
- hp = dev->priv;
+ hp = netdev_priv(dev);
hp->happy_dev = sdev;
@@ -3030,7 +3030,7 @@ #endif
dev->base_addr = (long) pdev;
- hp = (struct happy_meal *)dev->priv;
+ hp = netdev_priv(dev);
memset(hp, 0, sizeof(*hp));
hp->happy_dev = pdev;
diff --git a/drivers/net/sunqe.c b/drivers/net/sunqe.c
index 1b65ae8..49498fd 100644
--- a/drivers/net/sunqe.c
+++ b/drivers/net/sunqe.c
@@ -510,7 +510,7 @@ static irqreturn_t qec_interrupt(int irq
static int qe_open(struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
qep->mconfig = (MREGS_MCONFIG_TXENAB |
MREGS_MCONFIG_RXENAB |
@@ -520,7 +520,7 @@ static int qe_open(struct net_device *de
static int qe_close(struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
qe_stop(qep);
return 0;
@@ -546,7 +546,7 @@ static void qe_tx_reclaim(struct sunqe *
static void qe_tx_timeout(struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
int tx_full;
spin_lock_irq(&qep->lock);
@@ -572,7 +572,7 @@ out:
/* Get a packet queued to go onto the wire. */
static int qe_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
struct sunqe_buffers *qbufs = qep->buffers;
__u32 txbuf_dvma, qbufs_dvma = qep->buffers_dvma;
unsigned char *txbuf;
@@ -624,14 +624,14 @@ static int qe_start_xmit(struct sk_buff
static struct net_device_stats *qe_get_stats(struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
return &qep->net_stats;
}
static void qe_set_multicast(struct net_device *dev)
{
- struct sunqe *qep = (struct sunqe *) dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
struct dev_mc_list *dmi = dev->mc_list;
u8 new_mconfig = qep->mconfig;
char *addrs;
@@ -696,7 +696,7 @@ static void qe_set_multicast(struct net_
/* Ethtool support... */
static void qe_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct sunqe *qep = dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
strcpy(info->driver, "sunqe");
strcpy(info->version, "3.0");
@@ -706,7 +706,7 @@ static void qe_get_drvinfo(struct net_de
static u32 qe_get_link(struct net_device *dev)
{
- struct sunqe *qep = dev->priv;
+ struct sunqe *qep = netdev_priv(dev);
void __iomem *mregs = qep->mregs;
u8 phyconfig;
diff --git a/drivers/net/tc35815.c b/drivers/net/tc35815.c
index ec41469..e91cf34 100644
--- a/drivers/net/tc35815.c
+++ b/drivers/net/tc35815.c
@@ -614,7 +614,7 @@ static int __devinit tc35815_mac_match(s
static int __devinit tc35815_read_plat_dev_addr(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct device *pd = bus_find_device(&platform_bus_type, NULL,
lp->pci_dev, tc35815_mac_match);
if (pd) {
@@ -684,7 +684,7 @@ static int __devinit tc35815_init_one (s
}
SET_MODULE_OWNER(dev);
SET_NETDEV_DEV(dev, &pdev->dev);
- lp = dev->priv;
+ lp = netdev_priv(dev);
/* enable device (incl. PCI PM wakeup), and bus-mastering */
rc = pci_enable_device (pdev);
@@ -749,7 +749,7 @@ #endif
dev->base_addr = (unsigned long) ioaddr;
/* dev->priv/lp zeroed and aligned in alloc_etherdev */
- lp = dev->priv;
+ lp = netdev_priv(dev);
spin_lock_init(&lp->lock);
lp->pci_dev = pdev;
lp->boardtype = ent->driver_data;
@@ -827,7 +827,7 @@ static void __devexit tc35815_remove_one
static int
tc35815_init_queues(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int i;
unsigned long fd_addr;
@@ -964,7 +964,7 @@ #endif
static void
tc35815_clear_queues(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int i;
for (i = 0; i < TX_FD_NUM; i++) {
@@ -995,7 +995,7 @@ #endif
static void
tc35815_free_queues(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int i;
if (lp->tfd_base) {
@@ -1109,7 +1109,7 @@ #ifdef DEBUG
static void
panic_queues(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int i;
printk("TxFD base %p, start %u, end %u\n",
@@ -1143,13 +1143,13 @@ static void print_eth(char *add)
static int tc35815_tx_full(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
return ((lp->tfd_start + 1) % TX_FD_NUM == lp->tfd_end);
}
static void tc35815_restart(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
int do_phy_reset = 1;
del_timer(&lp->timer); /* Kill if running */
@@ -1180,7 +1180,7 @@ static void tc35815_restart(struct net_d
static void tc35815_tx_timeout(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
@@ -1219,7 +1219,7 @@ static void tc35815_tx_timeout(struct ne
static int
tc35815_open(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
/*
* This is used if the interrupt line can turned off (shared).
@@ -1257,7 +1257,7 @@ tc35815_open(struct net_device *dev)
*/
static int tc35815_send_packet(struct sk_buff *skb, struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct TxFD *txfd;
unsigned long flags;
@@ -1371,7 +1371,7 @@ #else
static int tc35815_do_interrupt(struct net_device *dev, u32 status)
#endif
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
int ret = -1;
@@ -1456,7 +1456,7 @@ #ifdef TC35815_NAPI
}
return IRQ_NONE;
#else
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int handled;
u32 status;
@@ -1488,7 +1488,7 @@ static void
tc35815_rx(struct net_device *dev)
#endif
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
unsigned int fdctl;
int i;
int buf_free_count = 0;
@@ -1729,7 +1729,7 @@ #ifdef TC35815_NAPI
static int
tc35815_poll(struct net_device *dev, int *budget)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
int limit = min(*budget, dev->quota);
@@ -1773,7 +1773,7 @@ #endif
static void
tc35815_check_tx_stat(struct net_device *dev, int status)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
const char *msg = NULL;
/* count collisions */
@@ -1850,7 +1850,7 @@ #endif
static void
tc35815_txdone(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct TxFD *txfd;
unsigned int fdctl;
@@ -1948,7 +1948,7 @@ #endif
static int
tc35815_close(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
netif_stop_queue(dev);
/* Flush the Tx and disable Rx here. */
@@ -1969,7 +1969,7 @@ tc35815_close(struct net_device *dev)
*/
static struct net_device_stats *tc35815_get_stats(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
if (netif_running(dev)) {
@@ -1982,7 +1982,7 @@ static struct net_device_stats *tc35815_
static void tc35815_set_cam_entry(struct net_device *dev, int index, unsigned char *addr)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
int cam_index = index * 6;
@@ -2041,7 +2041,7 @@ tc35815_set_multicast_list(struct net_de
#ifdef WORKAROUND_100HALF_PROMISC
/* With some (all?) 100MHalf HUB, controller will hang
* if we enabled promiscuous mode before linkup... */
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
if (!(tc_mdio_read(dev, pid, MII_BMSR) & BMSR_LSTATUS))
return;
@@ -2081,7 +2081,7 @@ #endif
static void tc35815_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
strcpy(info->driver, MODNAME);
strcpy(info->version, DRV_VERSION);
strcpy(info->bus_info, pci_name(lp->pci_dev));
@@ -2089,7 +2089,7 @@ static void tc35815_get_drvinfo(struct n
static int tc35815_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
spin_lock_irq(&lp->lock);
mii_ethtool_gset(&lp->mii, cmd);
spin_unlock_irq(&lp->lock);
@@ -2098,7 +2098,7 @@ static int tc35815_get_settings(struct n
static int tc35815_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int rc;
#if 1 /* use our negotiation method... */
/* Verify the settings we care about. */
@@ -2128,7 +2128,7 @@ #endif
static int tc35815_nway_reset(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int rc;
spin_lock_irq(&lp->lock);
rc = mii_nway_restart(&lp->mii);
@@ -2138,7 +2138,7 @@ static int tc35815_nway_reset(struct net
static u32 tc35815_get_link(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int rc;
spin_lock_irq(&lp->lock);
rc = mii_link_ok(&lp->mii);
@@ -2148,25 +2148,25 @@ static u32 tc35815_get_link(struct net_d
static u32 tc35815_get_msglevel(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
return lp->msg_enable;
}
static void tc35815_set_msglevel(struct net_device *dev, u32 datum)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
lp->msg_enable = datum;
}
static int tc35815_get_stats_count(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
return sizeof(lp->lstats) / sizeof(int);
}
static void tc35815_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
data[0] = lp->lstats.max_tx_qlen;
data[1] = lp->lstats.tx_ints;
data[2] = lp->lstats.rx_ints;
@@ -2202,7 +2202,7 @@ static const struct ethtool_ops tc35815_
static int tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int rc;
if (!netif_running(dev))
@@ -2274,7 +2274,7 @@ static void tc_mdio_write(struct net_dev
static int tc35815_try_next_permutation(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short bmcr;
@@ -2303,7 +2303,7 @@ static int tc35815_try_next_permutation(
static void
tc35815_display_link_mode(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short lpa, bmcr;
char *speed = "", *duplex = "";
@@ -2329,7 +2329,7 @@ tc35815_display_link_mode(struct net_dev
static void tc35815_display_forced_link_mode(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short bmcr;
char *speed = "", *duplex = "";
@@ -2351,7 +2351,7 @@ static void tc35815_display_forced_link_
static void tc35815_set_link_modes(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
int pid = lp->phy_addr;
@@ -2418,7 +2418,7 @@ #endif
static void tc35815_timer(unsigned long data)
{
struct net_device *dev = (struct net_device *)data;
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short bmsr, bmcr, lpa;
int restart_timer = 0;
@@ -2626,7 +2626,7 @@ out:
static void tc35815_start_auto_negotiation(struct net_device *dev,
struct ethtool_cmd *ep)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short bmsr, bmcr, advertize;
int timeout;
@@ -2750,7 +2750,7 @@ force_link:
static void tc35815_find_phy(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short id0;
@@ -2779,7 +2779,7 @@ static void tc35815_find_phy(struct net_
static void tc35815_phy_chip_init(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
int pid = lp->phy_addr;
unsigned short bmcr;
struct ethtool_cmd ecmd, *ep;
@@ -2853,7 +2853,7 @@ static void tc35815_chip_reset(struct ne
static void tc35815_chip_init(struct net_device *dev)
{
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
struct tc35815_regs __iomem *tr =
(struct tc35815_regs __iomem *)dev->base_addr;
unsigned long txctl = TX_CTL_CMD;
@@ -2915,7 +2915,7 @@ #ifdef CONFIG_PM
static int tc35815_suspend(struct pci_dev *pdev, pm_message_t state)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
unsigned long flags;
pci_save_state(pdev);
@@ -2933,7 +2933,7 @@ static int tc35815_suspend(struct pci_de
static int tc35815_resume(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
- struct tc35815_local *lp = dev->priv;
+ struct tc35815_local *lp = netdev_priv(dev);
unsigned long flags;
pci_restore_state(pdev);
diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c
index 0358720..76ad294 100644
--- a/drivers/net/typhoon.c
+++ b/drivers/net/typhoon.c
@@ -1818,7 +1818,7 @@ static irqreturn_t
typhoon_interrupt(int irq, void *dev_instance)
{
struct net_device *dev = (struct net_device *) dev_instance;
- struct typhoon *tp = dev->priv;
+ struct typhoon *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->ioaddr;
u32 intr_status;
diff --git a/drivers/net/znet.c b/drivers/net/znet.c
index 4032e9f..27e060e 100644
--- a/drivers/net/znet.c
+++ b/drivers/net/znet.c
@@ -169,7 +169,7 @@ static void znet_tx_timeout (struct net_
/* Request needed resources */
static int znet_request_resources (struct net_device *dev)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
unsigned long flags;
if (request_irq (dev->irq, &znet_interrupt, 0, "ZNet", dev))
@@ -203,7 +203,7 @@ static int znet_request_resources (struc
static void znet_release_resources (struct net_device *dev)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
unsigned long flags;
release_region (znet->sia_base, znet->sia_size);
@@ -218,7 +218,7 @@ static void znet_release_resources (stru
/* Keep the magical SIA stuff in a single function... */
static void znet_transceiver_power (struct net_device *dev, int on)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
unsigned char v;
/* Turn on/off the 82501 SIA, using zenith-specific magic. */
@@ -237,7 +237,7 @@ static void znet_transceiver_power (stru
Also used from hardware_init. */
static void znet_set_multicast_list (struct net_device *dev)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
short ioaddr = dev->base_addr;
struct i82593_conf_block *cfblk = &znet->i593_init;
@@ -390,7 +390,7 @@ static int __init znet_probe (void)
SET_MODULE_OWNER (dev);
- znet = dev->priv;
+ znet = netdev_priv(dev);
netinfo = (struct netidblk *)p;
dev->base_addr = netinfo->iobase1;
@@ -535,7 +535,7 @@ static void znet_tx_timeout (struct net_
static int znet_send_packet(struct sk_buff *skb, struct net_device *dev)
{
int ioaddr = dev->base_addr;
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
unsigned long flags;
short length = skb->len;
@@ -605,7 +605,7 @@ static int znet_send_packet(struct sk_bu
static irqreturn_t znet_interrupt(int irq, void *dev_id)
{
struct net_device *dev = dev_id;
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
int ioaddr;
int boguscnt = 20;
int handled = 0;
@@ -683,7 +683,7 @@ static irqreturn_t znet_interrupt(int ir
static void znet_rx(struct net_device *dev)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
int ioaddr = dev->base_addr;
int boguscount = 1;
short next_frame_end_offset = 0; /* Offset of next frame start. */
@@ -833,7 +833,7 @@ static int znet_close(struct net_device
closed. */
static struct net_device_stats *net_get_stats(struct net_device *dev)
{
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
return &znet->stats;
}
@@ -842,7 +842,7 @@ static void show_dma(struct net_device *
{
short ioaddr = dev->base_addr;
unsigned char stat = inb (ioaddr);
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
unsigned long flags;
short dma_port = ((znet->tx_dma&3)<<2) + IO_DMA2_BASE;
unsigned addr = inb(dma_port);
@@ -865,7 +865,7 @@ static void hardware_init(struct net_dev
{
unsigned long flags;
short ioaddr = dev->base_addr;
- struct znet_private *znet = dev->priv;
+ struct znet_private *znet = netdev_priv(dev);
znet->rx_cur = znet->rx_start;
znet->tx_cur = znet->tx_start;
@@ -927,7 +927,7 @@ static void update_stop_hit(short ioaddr
static __exit void znet_cleanup (void)
{
if (znet_dev) {
- struct znet_private *znet = znet_dev->priv;
+ struct znet_private *znet = netdev_priv(znet_dev);
unregister_netdev (znet_dev);
kfree (znet->rx_start);
^ permalink raw reply related
* [patch 2.6.23-rc1] add xt_statistic.h to the header list for usermode programs
From: Chuck Ebbert @ 2007-08-03 17:28 UTC (permalink / raw)
To: David Miller; +Cc: Netdev
Add xt_statistic.h to the list of headers to install.
Apparently needed to build newer versions of iptables.
Signed-off-by: Chuck Ebbert <cebbert@redhat.com>
---
include/linux/netfilter/Kbuild | 1 +
1 file changed, 1 insertion(+)
--- linux-2.6.22.noarch.orig/include/linux/netfilter/Kbuild
+++ linux-2.6.22.noarch/include/linux/netfilter/Kbuild
@@ -28,6 +28,7 @@ header-y += xt_policy.h
header-y += xt_realm.h
header-y += xt_sctp.h
header-y += xt_state.h
+header-y += xt_statistic.h
header-y += xt_string.h
header-y += xt_tcpmss.h
header-y += xt_tcpudp.h
^ permalink raw reply
* [PATCH] lro: myri10ge example how to use LRO
From: Andrew Gallatin @ 2007-08-03 17:19 UTC (permalink / raw)
To: netdev
Cc: raisch, themann, linux-kernel, linuxppc-dev, meder, tklein,
stefan.roscher, jeff, davem
[-- Attachment #1: Type: text/plain, Size: 268 bytes --]
To follow up on Jan-Bernd Themann's LRO patch earlier today,
this patch shows how the generic LRO interface can be used for
page based drivers.
Again, many thanks to Jan-Bernd Themann for leading this effort.
Drew
Singed off by: Andrew Gallatin <gallatin@myri.com>
[-- Attachment #2: myri10ge_lro.diff --]
[-- Type: text/plain, Size: 6263 bytes --]
diff -urNp a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c
--- a/drivers/net/myri10ge/myri10ge.c 2007-07-24 15:57:12.000000000 -0400
+++ b/drivers/net/myri10ge/myri10ge.c 2007-08-03 13:07:48.000000000 -0400
@@ -48,6 +48,7 @@
#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
+#include <linux/inet_lro.h>
#include <linux/ip.h>
#include <linux/inet.h>
#include <linux/in.h>
@@ -62,6 +63,8 @@
#include <linux/io.h>
#include <linux/log2.h>
#include <net/checksum.h>
+#include <net/ip.h>
+#include <net/tcp.h>
#include <asm/byteorder.h>
#include <asm/io.h>
#include <asm/processor.h>
@@ -89,6 +92,7 @@ MODULE_LICENSE("Dual BSD/GPL");
#define MYRI10GE_EEPROM_STRINGS_SIZE 256
#define MYRI10GE_MAX_SEND_DESC_TSO ((65536 / 2048) * 2)
+#define MYRI10GE_MAX_LRO_DESCRIPTORS 8
#define MYRI10GE_NO_CONFIRM_DATA htonl(0xffffffff)
#define MYRI10GE_NO_RESPONSE_RESULT 0xffffffff
@@ -151,6 +155,8 @@ struct myri10ge_rx_done {
dma_addr_t bus;
int cnt;
int idx;
+ struct net_lro_mgr lro_mgr;
+ struct net_lro_desc lro_desc[MYRI10GE_MAX_LRO_DESCRIPTORS];
};
struct myri10ge_priv {
@@ -276,6 +282,14 @@ static int myri10ge_debug = -1; /* defau
module_param(myri10ge_debug, int, 0);
MODULE_PARM_DESC(myri10ge_debug, "Debug level (0=none,...,16=all)");
+static int myri10ge_lro = 1;
+module_param(myri10ge_lro, int, S_IRUGO);
+MODULE_PARM_DESC(myri10ge_lro, "Enable large receive offload\n");
+
+static int myri10ge_lro_max_pkts = MYRI10GE_LRO_MAX_PKTS;
+module_param(myri10ge_lro_max_pkts, int, S_IRUGO);
+MODULE_PARM_DESC(myri10ge_lro, "Number of LRO packets to be aggregated\n");
+
static int myri10ge_fill_thresh = 256;
module_param(myri10ge_fill_thresh, int, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(myri10ge_fill_thresh, "Number of empty rx slots allowed\n");
@@ -1019,6 +1033,15 @@ myri10ge_rx_done(struct myri10ge_priv *m
remainder -= MYRI10GE_ALLOC_SIZE;
}
+ if (mgp->csum_flag && myri10ge_lro) {
+ rx_frags[0].page_offset += MXGEFW_PAD;
+ rx_frags[0].size -= MXGEFW_PAD;
+ len -= MXGEFW_PAD;
+ lro_receive_frags(&mgp->rx_done.lro_mgr, rx_frags,
+ len, len, (void *)(unsigned long)csum, csum);
+ return 1;
+ }
+
hlen = MYRI10GE_HLEN > len ? len : MYRI10GE_HLEN;
/* allocate an skb to attach the page(s) to. */
@@ -1137,6 +1160,9 @@ static inline void myri10ge_clean_rx_don
mgp->stats.rx_packets += rx_packets;
mgp->stats.rx_bytes += rx_bytes;
+ if (myri10ge_lro)
+ lro_flush_all(&rx_done->lro_mgr);
+
/* restock receive rings if needed */
if (mgp->rx_small.fill_cnt - mgp->rx_small.cnt < myri10ge_fill_thresh)
myri10ge_alloc_rx_pages(mgp, &mgp->rx_small,
@@ -1378,7 +1404,8 @@ static const char myri10ge_gstrings_stat
"dropped_pause", "dropped_bad_phy", "dropped_bad_crc32",
"dropped_unicast_filtered", "dropped_multicast_filtered",
"dropped_runt", "dropped_overrun", "dropped_no_small_buffer",
- "dropped_no_big_buffer"
+ "dropped_no_big_buffer", "LRO aggregated", "LRO flushed",
+ "LRO avg aggr", "LRO no_desc"
};
#define MYRI10GE_NET_STATS_LEN 21
@@ -1444,6 +1471,14 @@ myri10ge_get_ethtool_stats(struct net_de
data[i++] = (unsigned int)ntohl(mgp->fw_stats->dropped_overrun);
data[i++] = (unsigned int)ntohl(mgp->fw_stats->dropped_no_small_buffer);
data[i++] = (unsigned int)ntohl(mgp->fw_stats->dropped_no_big_buffer);
+ data[i++] = mgp->rx_done.lro_mgr.stats.aggregated;
+ data[i++] = mgp->rx_done.lro_mgr.stats.flushed;
+ if (mgp->rx_done.lro_mgr.stats.flushed)
+ data[i++] = mgp->rx_done.lro_mgr.stats.aggregated /
+ mgp->rx_done.lro_mgr.stats.flushed;
+ else
+ data[i++] = 0;
+ data[i++] = mgp->rx_done.lro_mgr.stats.no_desc;
}
static void myri10ge_set_msglevel(struct net_device *netdev, u32 value)
@@ -1717,10 +1752,69 @@ static void myri10ge_free_irq(struct myr
pci_disable_msi(pdev);
}
+static int
+myri10ge_get_frag_header(struct skb_frag_struct *frag, void **mac_hdr,
+ void **ip_hdr, void **tcpudp_hdr,
+ u64 * hdr_flags, void *priv)
+{
+ struct ethhdr *eh;
+ struct vlan_ethhdr *veh;
+ struct iphdr *iph;
+ u8 *va = page_address(frag->page) + frag->page_offset;
+ unsigned long ll_hlen;
+ __wsum csum = (__wsum) (unsigned long)priv;
+
+ /* find the mac header, aborting if not IPv4 */
+
+ eh = (struct ethhdr *)va;
+ *mac_hdr = eh;
+ ll_hlen = ETH_HLEN;
+ if (eh->h_proto != htons(ETH_P_IP)) {
+ if (eh->h_proto == htons(ETH_P_8021Q)) {
+ veh = (struct vlan_ethhdr *)va;
+ if (veh->h_vlan_encapsulated_proto != htons(ETH_P_IP))
+ return -1;
+
+ ll_hlen += VLAN_HLEN;
+
+ /*
+ * HW checksum starts ETH_HLEN bytes into
+ * frame, so we must subtract off the VLAN
+ * header's checksum before csum can be used
+ */
+ csum = csum_sub(csum, csum_partial(va + ETH_HLEN,
+ VLAN_HLEN, 0));
+ } else {
+ return -1;
+ }
+ }
+ *hdr_flags = LRO_IPV4;
+
+ iph = (struct iphdr *)(va + ll_hlen);
+ *ip_hdr = iph;
+ if (iph->protocol != IPPROTO_TCP)
+ return -1;
+ *hdr_flags |= LRO_TCP;
+ *tcpudp_hdr = (u8 *) (*ip_hdr) + (iph->ihl << 2);
+
+ /* verify the IP checksum */
+ if (unlikely(ip_fast_csum((u8 *) iph, iph->ihl)))
+ return -1;
+
+ /* verify the checksum */
+ if (unlikely(csum_tcpudp_magic(iph->saddr, iph->daddr,
+ ntohs(iph->tot_len) - (iph->ihl << 2),
+ IPPROTO_TCP, csum)))
+ return -1;
+
+ return 0;
+}
+
static int myri10ge_open(struct net_device *dev)
{
struct myri10ge_priv *mgp;
struct myri10ge_cmd cmd;
+ struct net_lro_mgr *lro_mgr;
int status, big_pow2;
mgp = netdev_priv(dev);
@@ -1852,6 +1946,18 @@ static int myri10ge_open(struct net_devi
mgp->link_state = htonl(~0U);
mgp->rdma_tags_available = 15;
+ lro_mgr = &mgp->rx_done.lro_mgr;
+ lro_mgr->dev = dev;
+ lro_mgr->features = LRO_F_NAPI;
+ lro_mgr->ip_summed = CHECKSUM_COMPLETE;
+ lro_mgr->ip_summed_aggr = CHECKSUM_UNNECESSARY;
+ lro_mgr->max_desc = MYRI10GE_MAX_LRO_DESCRIPTORS;
+ lro_mgr->lro_arr = mgp->rx_done.lro_desc;
+ lro_mgr->get_frag_header = myri10ge_get_frag_header;
+ lro_mgr->max_aggr = myri10ge_lro_max_pkts;
+ if (lro_mgr->max_aggr > MAX_SKB_FRAGS)
+ lro_mgr->max_aggr = MAX_SKB_FRAGS;
+
netif_poll_enable(mgp->dev); /* must happen prior to any irq */
status = myri10ge_send_cmd(mgp, MXGEFW_CMD_ETHERNET_UP, &cmd, 0);
^ permalink raw reply
* Re: strange tcp behavior
From: Simon Arlott @ 2007-08-03 16:51 UTC (permalink / raw)
To: Evgeniy Polyakov; +Cc: john, netdev, David Miller
In-Reply-To: <20070803120920.GA23166@2ka.mipt.ru>
[-- Attachment #1: Type: text/plain, Size: 3937 bytes --]
On 03/08/07 13:09, Evgeniy Polyakov wrote:
> On Fri, Aug 03, 2007 at 01:03:46PM +0100, Simon Arlott (simon@fire.lp0.eu) wrote:
>> On Fri, August 3, 2007 12:56, Evgeniy Polyakov wrote:
>> > On Fri, Aug 03, 2007 at 12:21:46PM +0100, Simon Arlott (simon@fire.lp0.eu) wrote:
>> >> Since the connection is considered closed, couldn't another socket re-use it?
>> >>
>> >> Socket A: Recv data (unread)
>> >> Socket A: Recv RST
>> >> Socket B: Reuses connection (same IPs/ports)
>> >> Socket A: Close
>> >>
>> >> Wouldn't that disrupt socket B's use of the connection?
>> >
>> > Then it will drop our data, since there were no appropriate handhsake.
>>
>> Couldn't the sequence numbers be close enough to make the RST valid?
>
> It does not matter - if connection is not in synchronized state all
> unrelated data is dropped, so remote side is only allowed to receive syn
> flag only, anything else must be dropped. If remote side does not do
> that, it violates RFC.
Except the remote side has a connection, because another one can be made
before the existing connection is closed:
17:37:37.377571 IP 192.168.7.4.50550 > 192.168.7.8.2500: S 134077329:134077329(0) win 1500 (raw)
17:37:37.382352 IP 192.168.7.8.2500 > 192.168.7.4.50550: S 3460060233:3460060233(0) ack 134077330 win 14360 <mss 7180> (accept)
17:37:37.377966 IP 192.168.7.4.50550 > 192.168.7.8.2500: . ack 1 win 1500 (raw)
17:37:37.378128 IP 192.168.7.4.50550 > 192.168.7.8.2500: P 1:17(16) ack 1 win 1500 (raw)
17:37:37.378162 IP 192.168.7.8.2500 > 192.168.7.4.50550: . ack 17 win 14360
17:37:37.378131 IP 192.168.7.4.50550 > 192.168.7.8.2500: R 134077346:134077346(0) win 1500 (raw)
17:37:37.412709 IP 192.168.7.4.50550 > 192.168.7.8.2500: SWE 3257207813:3257207813(0) win 14280 <mss 7140,sackOK,timestamp 3601441543 0,nop,wscale 5> (connect)
17:37:37.412785 IP 192.168.7.8.2500 > 192.168.7.4.50550: SE 3495384256:3495384256(0) ack 3257207814 win 14336 <mss 7180,sackOK,timestamp 4294812905 3601441543,nop,wscale 6> (accept)
17:37:37.412960 IP 192.168.7.4.50550 > 192.168.7.8.2500: . ack 1 win 447 <nop,nop,timestamp 3601441543 4294812905>
17:37:38.383085 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 4259643274:4259643274(0) ack 1171836829 win 14360 (close (previous connection))
17:37:47.417649 IP 192.168.7.8.2500 > 192.168.7.4.50550: F 1:1(0) ack 1 win 224 <nop,nop,timestamp 4294822910 3601441543> (close)
17:37:47.417993 IP 192.168.7.4.50550 > 192.168.7.8.2500: F 1:1(0) ack 2 win 447 <nop,nop,timestamp 3601444045 4294822910> (read returned)
17:37:47.418466 IP 192.168.7.8.2500 > 192.168.7.4.50550: . ack 2 win 224 <nop,nop,timestamp 4294822911 3601444045>
The second connection also modified the RST|ACK that was sent compared to no second connection:
17:38:03.532703 IP 192.168.7.4.50550 > 192.168.7.8.2500: S 82517575:82517575(0) win 1500 (raw)
17:38:03.532832 IP 192.168.7.8.2500 > 192.168.7.4.50550: S 3495449795:3495449795(0) ack 82517576 win 14360 <mss 7180> (accept)
17:38:03.533388 IP 192.168.7.4.50550 > 192.168.7.8.2500: . ack 1 win 1500 (raw)
17:38:03.533457 IP 192.168.7.4.50550 > 192.168.7.8.2500: P 1:17(16) ack 1 win 1500 (raw)
17:38:03.533597 IP 192.168.7.8.2500 > 192.168.7.4.50550: . ack 17 win 14360
17:38:03.533589 IP 192.168.7.4.50550 > 192.168.7.8.2500: R 82517592:82517592(0) win 1500 (raw)
17:38:04.536277 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 1:1(0) ack 17 win 14360 (close)
17:38:04.536277 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 1:1(0) ack 17 win 14360
vs
17:37:38.383085 IP 192.168.7.8.2500 > 192.168.7.4.50550: R 4259643274:4259643274(0) ack 1171836829 win 14360
What happened there ?
On the server, run tcptest-server.c, which waits for 1s on the first connection then 10s on the second connection.
On the client, run:
iptables -I INPUT -i eth0 -p tcp --dport 50550 -j DROP; ./client; iptables -D INPUT -i eth0 -p tcp --dport 50550 -j DROP; ./tcptest-client
(client.c from john's original email)
--
Simon Arlott
[-- Attachment #2: tcptest-server.c --]
[-- Type: text/x-csrc, Size: 1712 bytes --]
#include <sys/types.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <poll.h>
#include <fcntl.h>
#define PORT 2500
#define xerror(str) do { perror(str); exit(1); } while (0)
int main(void) {
struct sockaddr_in sa;
int l, s, tmp;
int t = 0;
memset(&sa, 0, sizeof(sa));
l = socket(PF_INET, SOCK_STREAM, IPPROTO_TCP);
if (!l)
xerror("socket");
sa.sin_family = AF_INET;
sa.sin_addr.s_addr = htonl(INADDR_ANY);
sa.sin_port = htons(PORT);
tmp = 1;
setsockopt(l, SOL_SOCKET, SO_REUSEADDR, (char*)&tmp, sizeof(tmp));
if (bind(l, (struct sockaddr*)&sa, sizeof(sa)) != 0)
xerror("bind");
if (listen(l, 0) != 0)
xerror("listen");
printf("server %d ready...\n", getpid());
for (t = 1; t <= 2; t++) {
s = accept(l, NULL, NULL);
switch (fork()) {
case -1:
xerror("fork");
break;
case 0:
switch (t) {
case 1:
printf("server %d accepted connection\n", getpid());
#if 0
tmp = fcntl(s, F_GETFL, 0);
if (fcntl(s, F_SETFL, tmp | O_NONBLOCK) != 0)
xerror("fcntl");
if (send(s, "AAAAAAA", 7, 0) != 7)
xerror("send");
#endif
printf("server %d waiting for 1 second...\n", getpid());
sleep(1);
printf("server %d closing connection\n", getpid());
close(s);
return 0;
break;
case 2:
printf("server %d accepted connection\n", getpid());
printf("server %d waiting for 10 seconds...\n", getpid());
sleep(10);
printf("server %d closing connection\n", getpid());
close(s);
return 0;
break;
}
break;
default:
close(s);
}
}
wait(NULL);
wait(NULL);
return 0;
}
[-- Attachment #3: tcptest-client.c --]
[-- Type: text/x-csrc, Size: 1408 bytes --]
#include <sys/types.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#define SERVER_IP "192.168.7.8"
#define SRC_PORT 50550
#define DST_PORT 2500
#define xerror(str) do { perror(str); exit(1); } while (0)
int main(void) {
struct sockaddr_in sa;
int s;
int t = 0;
char buf[32];
int ret, tmp;
memset(&sa, 0, sizeof(sa));
for (t = 1; t <= 1; t++) {
switch (fork()) {
case -1:
xerror("fork");
break;
case 0:
printf("client %d binding port...\n", getpid());
s = socket(PF_INET, SOCK_STREAM, IPPROTO_TCP);
if (!s)
xerror("socket");
sa.sin_family = AF_INET;
sa.sin_addr.s_addr = htonl(INADDR_ANY);
sa.sin_port = htons(SRC_PORT);
tmp = 1;
if (setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char*)&tmp, sizeof(tmp)) != 0)
xerror("setsockopt");
if (bind(s, (struct sockaddr*)&sa, sizeof(sa)) != 0)
xerror("bind");
printf("client %d connecting...\n", getpid());
sa.sin_family = AF_INET;
sa.sin_addr.s_addr = inet_addr(SERVER_IP);
sa.sin_port = htons(DST_PORT);
if (connect(s, (struct sockaddr*)&sa, sizeof(sa)) != 0)
xerror("connect");
printf("client %d waiting in read()...\n", getpid());
ret = read(s, buf, 32);
printf("client %d read() returned %d\n", getpid(), ret);
close(s);
return 0;
}
wait(NULL);
}
return 0;
}
^ permalink raw reply
* Re: Distributed storage.
From: Peter Zijlstra @ 2007-08-03 14:53 UTC (permalink / raw)
To: Evgeniy Polyakov
Cc: Daniel Phillips, netdev, linux-kernel, linux-fsdevel,
Arnaldo Carvalho de Melo
In-Reply-To: <20070803134943.GA21221@2ka.mipt.ru>
On Fri, 2007-08-03 at 17:49 +0400, Evgeniy Polyakov wrote:
> On Fri, Aug 03, 2007 at 02:27:52PM +0200, Peter Zijlstra (peterz@infradead.org) wrote:
> > On Fri, 2007-08-03 at 14:57 +0400, Evgeniy Polyakov wrote:
> >
> > > For receiving situation is worse, since system does not know in advance
> > > to which socket given packet will belong to, so it must allocate from
> > > global pool (and thus there must be independent global reserve), and
> > > then exchange part of the socket's reserve to the global one (or just
> > > copy packet to the new one, allocated from socket's reseve is it was
> > > setup, or drop it otherwise). Global independent reserve is what I
> > > proposed when stopped to advertise network allocator, but it seems that
> > > it was not taken into account, and reserve was always allocated only
> > > when system has serious memory pressure in Peter's patches without any
> > > meaning for per-socket reservation.
> >
> > This is not true. I have a global reserve which is set-up a priori. You
> > cannot allocate a reserve when under pressure, that does not make sense.
>
> I probably did not cut enough details - my main position is to allocate
> per socket reserve from socket's queue, and copy data there from main
> reserve, all of which are allocated either in advance (global one) or
> per sockoption, so that there would be no fairness issues what to mark
> as special and what to not.
>
> Say we have a page per socket, each socket can assign a reserve for
> itself from own memory, this accounts both tx and rx side. Tx is not
> interesting, it is simple, rx has global reserve (always allocated on
> startup or sometime way before reclaim/oom)where data is originally
> received (including skb, shared info and whatever is needed, page is
> just an exmaple), then it is copied into per-socket reserve and reused
> for the next packet. Having per-socket reserve allows to have progress
> in any situation not only in cases where single action must be
> received/processed, and allows to be completely fair for all users, but
> not only special sockets, thus admin for example would be allowed to
> login, ipsec would work and so on...
Ah, I think I understand now. Yes this is indeed a good idea!
It would be quite doable to implement this on top of that I already
have. We would need to extend the socket with a sock_opt that would
reserve a specified amount of data for that specific socket. And then on
socket demux check if the socket has a non zero reserve and has not yet
exceeded said reserve. If so, process the packet.
This would also quite neatly work for -rt where we would not want
incomming packet processing to be delayed by memory allocations.
^ permalink raw reply
* Re: Distributed storage.
From: Evgeniy Polyakov @ 2007-08-03 13:49 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Daniel Phillips, netdev, linux-kernel, linux-fsdevel,
Arnaldo Carvalho de Melo
In-Reply-To: <1186144072.11797.55.camel@lappy>
On Fri, Aug 03, 2007 at 02:27:52PM +0200, Peter Zijlstra (peterz@infradead.org) wrote:
> On Fri, 2007-08-03 at 14:57 +0400, Evgeniy Polyakov wrote:
>
> > For receiving situation is worse, since system does not know in advance
> > to which socket given packet will belong to, so it must allocate from
> > global pool (and thus there must be independent global reserve), and
> > then exchange part of the socket's reserve to the global one (or just
> > copy packet to the new one, allocated from socket's reseve is it was
> > setup, or drop it otherwise). Global independent reserve is what I
> > proposed when stopped to advertise network allocator, but it seems that
> > it was not taken into account, and reserve was always allocated only
> > when system has serious memory pressure in Peter's patches without any
> > meaning for per-socket reservation.
>
> This is not true. I have a global reserve which is set-up a priori. You
> cannot allocate a reserve when under pressure, that does not make sense.
I probably did not cut enough details - my main position is to allocate
per socket reserve from socket's queue, and copy data there from main
reserve, all of which are allocated either in advance (global one) or
per sockoption, so that there would be no fairness issues what to mark
as special and what to not.
Say we have a page per socket, each socket can assign a reserve for
itself from own memory, this accounts both tx and rx side. Tx is not
interesting, it is simple, rx has global reserve (always allocated on
startup or sometime way before reclaim/oom)where data is originally
received (including skb, shared info and whatever is needed, page is
just an exmaple), then it is copied into per-socket reserve and reused
for the next packet. Having per-socket reserve allows to have progress
in any situation not only in cases where single action must be
received/processed, and allows to be completely fair for all users, but
not only special sockets, thus admin for example would be allowed to
login, ipsec would work and so on...
--
Evgeniy Polyakov
^ permalink raw reply
* Re: [patch 0/5][RFC] Update network drivers to use devres
From: Stephen Hemminger @ 2007-08-03 13:44 UTC (permalink / raw)
To: Tejun Heo; +Cc: Brandon Philips, netdev, teheo
In-Reply-To: <46B31270.7090503@gmail.com>
On Fri, 03 Aug 2007 20:33:04 +0900
Tejun Heo <htejun@gmail.com> wrote:
> Hello,
>
> Stephen Hemminger wrote:
> >> Skimming through drivers... via-rhine doesn't disable PCI device on
> >> init failure path but does so on removal. sky2 doesn't free
> >> consistent memory if sky2_init() fails. acenic calls iounmap() with
> >> NULL parameter which I'm not sure whether it's safe or not. natsemi
> >> doesn't disable PCI device on failure or removal.
> >
> > Did you report these to the developers?
>
> Just skimmed through. I'm pretty sure Brandon will pick those up later.
>
> >> Devres makes low level drivers simpler, easier to get right and
> >> maintain. Writing new drivers becomes easier too. So, why not?
> >>
> >>> Network devices seem to work fine thanks, and the resource requirements
> >>> are different. If ain't broke, don't fix it.
> >> Care to enlighten me on how the resource requirments are different
> >> from ATA drivers?
> >
> > I was thinking of the hot remove (no mod ref counts) and lingering
> > /sys open issues. ATA drivers use ref counts.
>
> I guess the hot removing is done by severing netdev from the actual
> device, right? I don't see how that affects usage of devres on network
> drivers. Am I missing something?
The issue is that device may be removed at any time. So you can't rely
on module ref counts to save you. And netdevice structure must still
linger after module is removed, till dev ref count goes to zero.
> On a separate note, can you explain lingering /sys open issue to me a
> bit? With recent sysfs changes, sysfs nodes are disconnected
> immediately on deletion. Would that make any difference to netdevs?
Examples are in Documentation/networking/netdevices.txt
> > My take on devres is that it is similar to talloc() for device drivers.
> > Not a bad idea in itself, but the real advantage of hierarchical allocation
> > is that it makes exception handling easier if things are layered deeply.
>
> Yeah, devres made layering easier in libata, especially SFF stuff.
> Dunno how much of that is applicable to netdev but, with or without
> layering, it'll be a nice cleanup and I don't see much negative side.
> Conversion would take some work and bugs might be introduced in the
> process as with any changes but the good thing about devres is that
> you're very likely to get failure/release paths right if you get the
> init path right, and if you get the init path wrong, it will stand out
> like a sore thumb - easy to spot, easy to fix.
>
> So, I think using devres on net drivers is a good idea, well, for that
> matter, for any driver, but me being the devres writer, that isn't
> really surprising, is it?
>
> Thanks.
>
> --
> tejun
^ permalink raw reply
* Re: [PATCH 1/1] lro: Generic Large Receive Offload for TCP traffic
From: Jörn Engel @ 2007-08-03 13:41 UTC (permalink / raw)
To: Jan-Bernd Themann
Cc: David Miller, Christoph Raisch, Jan-Bernd Themann, linux-kernel,
linux-ppc, Marcus Eder, Thomas Klein, netdev, Andrew Gallatin,
Jeff Garzik, Stefan Roscher
In-Reply-To: <200708031441.20632.ossthema@de.ibm.com>
On Fri, 3 August 2007 14:41:19 +0200, Jan-Bernd Themann wrote:
>
> This patch provides generic Large Receive Offload (LRO) functionality
> for IPv4/TCP traffic.
>
> LRO combines received tcp packets to a single larger tcp packet and
> passes them then to the network stack in order to increase performance
> (throughput). The interface supports two modes: Drivers can either pass
> SKBs or fragment lists to the LRO engine.
Maybe this is a stupid question, but why is LRO done at the device
driver level?
If it is a unversal performance benefit, I would have expected it to be
done generically, i.e. have all packets moved into network layer pass
through LRO instead.
> +void lro_flush_pkt(struct net_lro_mgr *lro_mgr,
> + struct iphdr *iph, struct tcphdr *tcph);
In particular this bit looks like it should be driven by a timeout,
which would be settable via /proc/sys/net/core/lro_timeout or similar.
Jörn
--
Rules of Optimization:
Rule 1: Don't do it.
Rule 2 (for experts only): Don't do it yet.
-- M.A. Jackson
^ permalink raw reply
* [PATCH] lro: eHEA example how to use LRO
From: Jan-Bernd Themann @ 2007-08-03 12:41 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, Andrew Gallatin,
Jeff Garzik, David Miller
This patch shows how the generic LRO interface is used for SKB mode
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
drivers/net/Kconfig | 1 +
drivers/net/ehea/ehea.h | 9 ++++-
drivers/net/ehea/ehea_ethtool.c | 15 +++++++
drivers/net/ehea/ehea_main.c | 84 +++++++++++++++++++++++++++++++++++---
4 files changed, 101 insertions(+), 8 deletions(-)
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index f8a602c..fec4004 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2399,6 +2399,7 @@ config CHELSIO_T3
config EHEA
tristate "eHEA Ethernet support"
depends on IBMEBUS
+ select INET_LRO
---help---
This driver supports the IBM pSeries eHEA ethernet adapter.
diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h
index d67f97b..70e33fe 100644
--- a/drivers/net/ehea/ehea.h
+++ b/drivers/net/ehea/ehea.h
@@ -33,13 +33,14 @@
#include <linux/ethtool.h>
#include <linux/vmalloc.h>
#include <linux/if_vlan.h>
+#include <linux/inet_lro.h>
#include <asm/ibmebus.h>
#include <asm/abs_addr.h>
#include <asm/io.h>
#define DRV_NAME "ehea"
-#define DRV_VERSION "EHEA_0073"
+#define DRV_VERSION "EHEA_0074"
/* eHEA capability flags */
#define DLPAR_PORT_ADD_REM 1
@@ -58,6 +59,7 @@
#define EHEA_SMALL_QUEUES
#define EHEA_NUM_TX_QP 1
+#define EHEA_LRO_MAX_AGGR 64
#ifdef EHEA_SMALL_QUEUES
#define EHEA_MAX_CQE_COUNT 1023
@@ -84,6 +86,8 @@
#define EHEA_RQ2_PKT_SIZE 1522
#define EHEA_L_PKT_SIZE 256 /* low latency */
+#define MAX_LRO_DESCRIPTORS 8
+
/* Send completion signaling */
/* Protection Domain Identifier */
@@ -376,6 +380,8 @@ struct ehea_port_res {
u64 tx_packets;
u64 rx_packets;
u32 poll_counter;
+ struct net_lro_mgr lro_mgr;
+ struct net_lro_desc lro_desc[MAX_LRO_DESCRIPTORS];
};
@@ -427,6 +433,7 @@ struct ehea_port {
u32 msg_enable;
u32 sig_comp_iv;
u32 state;
+ u32 lro_max_aggr;
u8 full_duplex;
u8 autoneg;
u8 num_def_qps;
diff --git a/drivers/net/ehea/ehea_ethtool.c b/drivers/net/ehea/ehea_ethtool.c
index decec8c..29ef7a9 100644
--- a/drivers/net/ehea/ehea_ethtool.c
+++ b/drivers/net/ehea/ehea_ethtool.c
@@ -183,6 +183,9 @@ static char ehea_ethtool_stats_keys[][ETH_GSTRING_LEN] = {
{"PR5 free_swqes"},
{"PR6 free_swqes"},
{"PR7 free_swqes"},
+ {"LRO aggregated"},
+ {"LRO flushed"},
+ {"LRO no_desc"},
};
static void ehea_get_strings(struct net_device *dev, u32 stringset, u8 *data)
@@ -239,6 +242,18 @@ static void ehea_get_ethtool_stats(struct net_device *dev,
for (k = 0; k < 8; k++)
data[i++] = atomic_read(&port->port_res[k].swqe_avail);
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.aggregated;
+ data[i++] = tmp;
+
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.flushed;
+ data[i++] = tmp;
+
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.no_desc;
+ data[i++] = tmp;
+
}
const struct ethtool_ops ehea_ethtool_ops = {
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index 9756211..fbaa395 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -52,6 +52,8 @@ static int rq2_entries = EHEA_DEF_ENTRIES_RQ2;
static int rq3_entries = EHEA_DEF_ENTRIES_RQ3;
static int sq_entries = EHEA_DEF_ENTRIES_SQ;
static int use_mcs = 0;
+static int use_lro = 0;
+static int lro_max_aggr = EHEA_LRO_MAX_AGGR;
static int num_tx_qps = EHEA_NUM_TX_QP;
module_param(msg_level, int, 0);
@@ -60,6 +62,8 @@ module_param(rq2_entries, int, 0);
module_param(rq3_entries, int, 0);
module_param(sq_entries, int, 0);
module_param(use_mcs, int, 0);
+module_param(use_lro, int, 0);
+module_param(lro_max_aggr, int, 0);
module_param(num_tx_qps, int, 0);
MODULE_PARM_DESC(num_tx_qps, "Number of TX-QPS");
@@ -77,6 +81,10 @@ MODULE_PARM_DESC(sq_entries, " Number of entries for the Send Queue "
"[2^x - 1], x = [6..14]. Default = "
__MODULE_STRING(EHEA_DEF_ENTRIES_SQ) ")");
MODULE_PARM_DESC(use_mcs, " 0:NAPI, 1:Multiple receive queues, Default = 1 ");
+MODULE_PARM_DESC(lro_max_aggr, " LRO: Max packets to be aggregated. Default = "
+ __MODULE_STRING(EHEA_LRO_MAX_AGGR));
+MODULE_PARM_DESC(use_lro, " Large Receive Offload, 1: enable, 0: disable, "
+ "Default = 0");
static int port_name_cnt = 0;
static LIST_HEAD(adapter_list);
@@ -389,6 +397,60 @@ static int ehea_treat_poll_error(struct ehea_port_res *pr, int rq,
return 0;
}
+static int get_skb_hdr(struct sk_buff *skb, void **iphdr,
+ void **tcph, u64 *hdr_flags, void *priv)
+{
+ struct ehea_cqe *cqe = priv;
+ unsigned int ip_len;
+ struct iphdr *iph;
+
+ /* non tcp/udp packets */
+ if (!cqe->header_length)
+ return -1;
+
+ /* non tcp packet */
+ skb_reset_network_header(skb);
+ iph = ip_hdr(skb);
+ if (iph->protocol != IPPROTO_TCP)
+ return -1;
+
+ ip_len = ip_hdrlen(skb);
+ skb_set_transport_header(skb, ip_len);
+ *tcph = tcp_hdr(skb);
+
+ /* check if ip header and tcp header are complete */
+ if (iph->tot_len < ip_len + tcp_hdrlen(skb))
+ return -1;
+
+ *hdr_flags = LRO_IPV4 | LRO_TCP;
+ *iphdr = iph;
+
+ return 0;
+}
+
+static void ehea_proc_skb(struct ehea_port_res *pr, struct ehea_cqe *cqe,
+ struct sk_buff *skb)
+{
+ int vlan_extracted = (cqe->status & EHEA_CQE_VLAN_TAG_XTRACT)
+ && pr->port->vgrp;
+
+ if (use_lro) {
+ if (vlan_extracted)
+ lro_vlan_hwaccel_receive_skb(&pr->lro_mgr, skb,
+ pr->port->vgrp,
+ cqe->vlan_tag,
+ cqe);
+ else
+ lro_receive_skb(&pr->lro_mgr, skb, cqe);
+ } else {
+ if (vlan_extracted)
+ vlan_hwaccel_receive_skb(skb, pr->port->vgrp,
+ cqe->vlan_tag);
+ else
+ netif_receive_skb(skb);
+ }
+}
+
static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
struct ehea_port_res *pr,
int *budget)
@@ -460,13 +522,7 @@ static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
processed_rq3++;
}
- if ((cqe->status & EHEA_CQE_VLAN_TAG_XTRACT)
- && port->vgrp)
- vlan_hwaccel_receive_skb(skb, port->vgrp,
- cqe->vlan_tag);
- else
- netif_receive_skb(skb);
-
+ ehea_proc_skb(pr, cqe, skb);
dev->last_rx = jiffies;
} else {
pr->p_stats.poll_receive_errors++;
@@ -478,6 +534,8 @@ static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
}
cqe = ehea_poll_rq1(qp, &wqe_index);
}
+ if (use_lro)
+ lro_flush_all(&pr->lro_mgr);
pr->rx_packets += processed;
*budget -= processed;
@@ -1233,6 +1291,15 @@ static int ehea_init_port_res(struct ehea_port *port, struct ehea_port_res *pr,
set_bit(__LINK_STATE_START, &pr->d_netdev->state);
strcpy(pr->d_netdev->name, port->netdev->name);
+ pr->lro_mgr.max_aggr = pr->port->lro_max_aggr;
+ pr->lro_mgr.max_desc = MAX_LRO_DESCRIPTORS;
+ pr->lro_mgr.lro_arr = pr->lro_desc;
+ pr->lro_mgr.get_skb_header = get_skb_hdr;
+ pr->lro_mgr.features = LRO_F_NAPI | LRO_F_EXTRACT_VLAN_ID;
+ pr->lro_mgr.dev = port->netdev;
+ pr->lro_mgr.ip_summed = CHECKSUM_UNNECESSARY;
+ pr->lro_mgr.ip_summed_aggr = CHECKSUM_UNNECESSARY;
+
ret = 0;
goto out;
@@ -1712,6 +1779,7 @@ static int ehea_change_mtu(struct net_device *dev, int new_mtu)
if ((new_mtu < 68) || (new_mtu > EHEA_MAX_PACKET_SIZE))
return -EINVAL;
dev->mtu = new_mtu;
+
return 0;
}
@@ -2669,6 +2737,8 @@ struct ehea_port *ehea_setup_single_port(struct ehea_adapter *adapter,
goto out_dereg_bc;
}
+ port->lro_max_aggr = lro_max_aggr;
+
ret = ehea_get_jumboframe_status(port, &jumbo);
if (ret)
ehea_error("failed determining jumbo frame status for %s",
--
1.5.2
^ permalink raw reply related
* [PATCH 1/1] lro: Generic Large Receive Offload for TCP traffic
From: Jan-Bernd Themann @ 2007-08-03 12:41 UTC (permalink / raw)
To: David Miller
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, netdev, Andrew Gallatin, Jeff Garzik,
Stefan Roscher
This patch provides generic Large Receive Offload (LRO) functionality
for IPv4/TCP traffic.
LRO combines received tcp packets to a single larger tcp packet and
passes them then to the network stack in order to increase performance
(throughput). The interface supports two modes: Drivers can either pass
SKBs or fragment lists to the LRO engine.
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
include/linux/inet_lro.h | 177 ++++++++++++++
net/ipv4/Kconfig | 8 +
net/ipv4/Makefile | 1 +
net/ipv4/inet_lro.c | 600 ++++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 786 insertions(+), 0 deletions(-)
create mode 100644 include/linux/inet_lro.h
create mode 100644 net/ipv4/inet_lro.c
diff --git a/include/linux/inet_lro.h b/include/linux/inet_lro.h
new file mode 100644
index 0000000..e1fc1d1
--- /dev/null
+++ b/include/linux/inet_lro.h
@@ -0,0 +1,177 @@
+/*
+ * linux/include/linux/inet_lro.h
+ *
+ * Large Receive Offload (ipv4 / tcp)
+ *
+ * (C) Copyright IBM Corp. 2007
+ *
+ * Authors:
+ * Jan-Bernd Themann <themann@de.ibm.com>
+ * Christoph Raisch <raisch@de.ibm.com>
+ *
+ *
+ * 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, 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.
+ *
+ * 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __INET_LRO_H_
+#define __INET_LRO_H_
+
+#include <net/ip.h>
+#include <net/tcp.h>
+
+/*
+ * LRO statistics
+ */
+
+struct net_lro_stats {
+ unsigned long aggregated;
+ unsigned long flushed;
+ unsigned long no_desc;
+};
+
+/*
+ * LRO descriptor for a tcp session
+ */
+struct net_lro_desc {
+ struct sk_buff *parent;
+ struct sk_buff *last_skb;
+ struct skb_frag_struct *next_frag;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ struct vlan_group *vgrp;
+ __wsum data_csum;
+ u32 tcp_rcv_tsecr;
+ u32 tcp_rcv_tsval;
+ u32 tcp_ack;
+ u32 tcp_next_seq;
+ u32 skb_tot_frags_len;
+ u16 ip_tot_len;
+ u16 tcp_saw_tstamp; /* timestamps enabled */
+ u16 tcp_window;
+ u16 vlan_tag;
+ int pkt_aggr_cnt; /* counts aggregated packets */
+ int vlan_packet;
+ int mss;
+ int active;
+};
+
+/*
+ * Large Receive Offload (LRO) Manager
+ *
+ * Fields must be set by driver
+ */
+
+struct net_lro_mgr {
+ struct net_device *dev;
+ struct net_lro_stats stats;
+
+ /* LRO features */
+ unsigned long features;
+#define LRO_F_NAPI 1 /* Pass packets to stack via NAPI */
+#define LRO_F_EXTRACT_VLAN_ID 2 /* Set flag if VLAN IDs are extracted
+ from received packets and eth protocol
+ is still ETH_P_8021Q */
+
+ u32 ip_summed; /* Set in non generated SKBs in page mode */
+ u32 ip_summed_aggr; /* Set in aggregated SKBs: CHECKSUM_UNNECESSARY
+ * or CHECKSUM_NONE */
+
+ int max_desc; /* Max number of LRO descriptors */
+ int max_aggr; /* Max number of LRO packets to be aggregated */
+
+ struct net_lro_desc *lro_arr; /* Array of LRO descriptors */
+
+ /*
+ * Optimized driver functions
+ *
+ * get_skb_header: returns tcp and ip header for packet in SKB
+ */
+ int (*get_skb_header)(struct sk_buff *skb, void **ip_hdr,
+ void **tcpudp_hdr, u64 *hdr_flags, void *priv);
+
+ /* hdr_flags: */
+#define LRO_IPV4 1 /* ip_hdr is IPv4 header */
+#define LRO_TCP 2 /* tcpudp_hdr is TCP header */
+
+ /*
+ * get_frag_header: returns mac, tcp and ip header for packet in SKB
+ *
+ * @hdr_flags: Indicate what kind of LRO has to be done
+ * (IPv4/IPv6/TCP/UDP)
+ */
+ int (*get_frag_header)(struct skb_frag_struct *frag, void **mac_hdr,
+ void **ip_hdr, void **tcpudp_hdr, u64 *hdr_flags,
+ void *priv);
+};
+
+/*
+ * Processes a SKB
+ *
+ * @lro_mgr: LRO manager to use
+ * @skb: SKB to aggregate
+ * @priv: Private data that may be used by driver functions
+ * (for example get_tcp_ip_hdr)
+ */
+
+void lro_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ void *priv);
+
+/*
+ * Processes a SKB with VLAN HW acceleration support
+ */
+
+void lro_vlan_hwaccel_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv);
+
+/*
+ * Processes a fragment list
+ *
+ * This functions aggregate fragments and generate SKBs do pass
+ * the packets to the stack.
+ *
+ * @lro_mgr: LRO manager to use
+ * @frags: Fragment to be processed. Must contain entire header in first
+ * element.
+ * @len: Length of received data
+ * @true_size: Actual size of memory the fragment is consuming
+ * @priv: Private data that may be used by driver functions
+ * (for example get_tcp_ip_hdr)
+ */
+
+void lro_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size, void *priv, __wsum sum);
+
+void lro_vlan_hwaccel_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv, __wsum sum);
+
+/*
+ * Forward all aggregated SKBs held by lro_mgr to network stack
+ */
+
+void lro_flush_all(struct net_lro_mgr *lro_mgr);
+
+void lro_flush_pkt(struct net_lro_mgr *lro_mgr,
+ struct iphdr *iph, struct tcphdr *tcph);
+
+#endif
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index fb79097..d894f61 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -394,6 +394,14 @@ config INET_XFRM_MODE_BEET
If unsure, say Y.
+config INET_LRO
+ tristate "Large Receive Offload (ipv4/tcp)"
+
+ ---help---
+ Support for Large Receive Offload (ipv4/tcp).
+
+ If unsure, say Y.
+
config INET_DIAG
tristate "INET: socket monitoring interface"
default y
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index fbf1674..a02c36d 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_INET_ESP) += esp4.o
obj-$(CONFIG_INET_IPCOMP) += ipcomp.o
obj-$(CONFIG_INET_XFRM_TUNNEL) += xfrm4_tunnel.o
obj-$(CONFIG_INET_XFRM_MODE_BEET) += xfrm4_mode_beet.o
+obj-$(CONFIG_INET_LRO) += inet_lro.o
obj-$(CONFIG_INET_TUNNEL) += tunnel4.o
obj-$(CONFIG_INET_XFRM_MODE_TRANSPORT) += xfrm4_mode_transport.o
obj-$(CONFIG_INET_XFRM_MODE_TUNNEL) += xfrm4_mode_tunnel.o
diff --git a/net/ipv4/inet_lro.c b/net/ipv4/inet_lro.c
new file mode 100644
index 0000000..3309adc
--- /dev/null
+++ b/net/ipv4/inet_lro.c
@@ -0,0 +1,600 @@
+/*
+ * linux/net/ipv4/inet_lro.c
+ *
+ * Large Receive Offload (ipv4 / tcp)
+ *
+ * (C) Copyright IBM Corp. 2007
+ *
+ * Authors:
+ * Jan-Bernd Themann <themann@de.ibm.com>
+ * Christoph Raisch <raisch@de.ibm.com>
+ *
+ *
+ * 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, 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.
+ *
+ * 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+
+#include <linux/module.h>
+#include <linux/if_vlan.h>
+#include <linux/inet_lro.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Jan-Bernd Themann <themann@de.ibm.com>");
+MODULE_DESCRIPTION("Large Receive Offload (ipv4 / tcp)");
+
+#define TCP_HDR_LEN(tcph) (tcph->doff << 2)
+#define IP_HDR_LEN(iph) (iph->ihl << 2)
+#define TCP_PAYLOAD_LENGTH(iph, tcph) \
+ (ntohs(iph->tot_len) - IP_HDR_LEN(iph) - TCP_HDR_LEN(tcph))
+
+#define IPH_LEN_WO_OPTIONS 5
+#define TCPH_LEN_WO_OPTIONS 5
+#define TCPH_LEN_W_TIMESTAMP 8
+
+#define LRO_MAX_PG_HLEN 64
+
+#define LRO_INC_STATS(lro_mgr, attr) { lro_mgr->stats.attr++; }
+
+/*
+ * Basic tcp checks whether packet is suitable for LRO
+ */
+
+static int lro_tcp_ip_check(struct iphdr *iph, struct tcphdr *tcph,
+ int len, struct net_lro_desc *lro_desc)
+{
+ /* check ip header: don't aggregate padded frames */
+ if (ntohs(iph->tot_len) != len)
+ return -1;
+
+ if (TCP_PAYLOAD_LENGTH(iph, tcph) == 0)
+ return -1;
+
+ if (iph->ihl != IPH_LEN_WO_OPTIONS)
+ return -1;
+
+ if (tcph->cwr || tcph->ece || tcph->urg || !tcph->ack
+ || tcph->rst || tcph->syn || tcph->fin)
+ return -1;
+
+ if (INET_ECN_is_ce(ipv4_get_dsfield(iph)))
+ return -1;
+
+ if (tcph->doff != TCPH_LEN_WO_OPTIONS
+ && tcph->doff != TCPH_LEN_W_TIMESTAMP)
+ return -1;
+
+ /* check tcp options (only timestamp allowed) */
+ if (tcph->doff == TCPH_LEN_W_TIMESTAMP) {
+ u32 *topt = (u32 *)(tcph + 1);
+
+ if (*topt != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
+ | (TCPOPT_TIMESTAMP << 8)
+ | TCPOLEN_TIMESTAMP))
+ return -1;
+
+ /* timestamp should be in right order */
+ topt++;
+ if (lro_desc && after(ntohl(lro_desc->tcp_rcv_tsval),
+ ntohl(*topt)))
+ return -1;
+
+ /* timestamp reply should not be zero */
+ topt++;
+ if (*topt == 0)
+ return -1;
+ }
+
+ return 0;
+}
+
+static void lro_update_tcp_ip_header(struct net_lro_desc *lro_desc)
+{
+ struct iphdr *iph = lro_desc->iph;
+ struct tcphdr *tcph = lro_desc->tcph;
+ u32 *p;
+ __wsum tcp_hdr_csum;
+
+ tcph->ack_seq = lro_desc->tcp_ack;
+ tcph->window = lro_desc->tcp_window;
+
+ if (lro_desc->tcp_saw_tstamp) {
+ p = (u32 *)(tcph + 1);
+ *(p+2) = lro_desc->tcp_rcv_tsecr;
+ }
+
+ iph->tot_len = htons(lro_desc->ip_tot_len);
+
+ iph->check = 0;
+ iph->check = ip_fast_csum((u8 *)lro_desc->iph, iph->ihl);
+
+ tcph->check = 0;
+ tcp_hdr_csum = csum_partial((u8 *)tcph, TCP_HDR_LEN(tcph), 0);
+ lro_desc->data_csum = csum_add(lro_desc->data_csum, tcp_hdr_csum);
+ tcph->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
+ lro_desc->ip_tot_len -
+ IP_HDR_LEN(iph), IPPROTO_TCP,
+ lro_desc->data_csum);
+}
+
+static __wsum lro_tcp_data_csum(struct iphdr *iph, struct tcphdr *tcph, int len)
+{
+ __wsum tcp_csum;
+ __wsum tcp_hdr_csum;
+ __wsum tcp_ps_hdr_csum;
+
+ tcp_csum = ~csum_unfold(tcph->check);
+ tcp_hdr_csum = csum_partial((u8 *)tcph, TCP_HDR_LEN(tcph), tcp_csum);
+
+ tcp_ps_hdr_csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
+ len + TCP_HDR_LEN(tcph),
+ IPPROTO_TCP, 0);
+
+ return csum_sub(csum_sub(tcp_csum, tcp_hdr_csum),
+ tcp_ps_hdr_csum);
+}
+
+static void lro_init_desc(struct net_lro_desc *lro_desc, struct sk_buff *skb,
+ struct iphdr *iph, struct tcphdr *tcph,
+ u16 vlan_tag, struct vlan_group *vgrp)
+{
+ int nr_frags;
+ u32 *ptr;
+ u32 tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ nr_frags = skb_shinfo(skb)->nr_frags;
+ lro_desc->parent = skb;
+ lro_desc->next_frag = &(skb_shinfo(skb)->frags[nr_frags]);
+ lro_desc->iph = iph;
+ lro_desc->tcph = tcph;
+ lro_desc->tcp_next_seq = ntohl(tcph->seq) + tcp_data_len;
+ lro_desc->tcp_ack = ntohl(tcph->ack_seq);
+ lro_desc->tcp_window = tcph->window;
+
+ lro_desc->pkt_aggr_cnt = 1;
+ lro_desc->ip_tot_len = ntohs(iph->tot_len);
+
+ if (tcph->doff == 8) {
+ ptr = (u32 *)(tcph+1);
+ lro_desc->tcp_saw_tstamp = 1;
+ lro_desc->tcp_rcv_tsval = *(ptr+1);
+ lro_desc->tcp_rcv_tsecr = *(ptr+2);
+ }
+
+ lro_desc->mss = tcp_data_len;
+ lro_desc->vgrp = vgrp;
+ lro_desc->vlan_tag = vlan_tag;
+ lro_desc->active = 1;
+
+ lro_desc->data_csum = lro_tcp_data_csum(iph, tcph,
+ tcp_data_len);
+}
+
+static inline void lro_clear_desc(struct net_lro_desc *lro_desc)
+{
+ memset(lro_desc, 0, sizeof(struct net_lro_desc));
+}
+
+static void lro_add_common(struct net_lro_desc *lro_desc, struct iphdr *iph,
+ struct tcphdr *tcph, int tcp_data_len)
+{
+ struct sk_buff *parent = lro_desc->parent;
+ u32 *topt;
+
+ lro_desc->pkt_aggr_cnt++;
+ lro_desc->ip_tot_len += tcp_data_len;
+ lro_desc->tcp_next_seq += tcp_data_len;
+ lro_desc->tcp_window = tcph->window;
+ lro_desc->tcp_ack = tcph->ack_seq;
+
+ /* don't update tcp_rcv_tsval, would not work with PAWS */
+ if (lro_desc->tcp_saw_tstamp) {
+ topt = (u32 *) (tcph + 1);
+ lro_desc->tcp_rcv_tsecr = *(topt + 2);
+ }
+
+ lro_desc->data_csum = csum_block_add(lro_desc->data_csum,
+ lro_tcp_data_csum(iph, tcph,
+ tcp_data_len),
+ parent->len);
+
+ parent->len += tcp_data_len;
+ parent->data_len += tcp_data_len;
+ if (tcp_data_len > lro_desc->mss)
+ lro_desc->mss = tcp_data_len;
+}
+
+static void lro_add_packet(struct net_lro_desc *lro_desc, struct sk_buff *skb,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct sk_buff *parent = lro_desc->parent;
+ int tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ lro_add_common(lro_desc, iph, tcph, tcp_data_len);
+
+ skb_pull(skb, (skb->len - tcp_data_len));
+ parent->truesize += skb->truesize;
+
+ if (lro_desc->last_skb)
+ lro_desc->last_skb->next = skb;
+ else
+ skb_shinfo(parent)->frag_list = skb;
+
+ lro_desc->last_skb = skb;
+}
+
+static void lro_add_frags(struct net_lro_desc *lro_desc,
+ int len, int hlen, int truesize,
+ struct skb_frag_struct *skb_frags,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct sk_buff *skb = lro_desc->parent;
+ int tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ lro_add_common(lro_desc, iph, tcph, tcp_data_len);
+
+ skb->truesize += truesize;
+
+ skb_frags[0].page_offset += hlen;
+ skb_frags[0].size -= hlen;
+
+ while (tcp_data_len > 0) {
+ *(lro_desc->next_frag) = *skb_frags;
+ tcp_data_len -= skb_frags->size;
+ lro_desc->next_frag++;
+ skb_frags++;
+ skb_shinfo(skb)->nr_frags++;
+ }
+}
+
+static int lro_check_tcp_conn(struct net_lro_desc *lro_desc,
+ struct iphdr *iph,
+ struct tcphdr *tcph)
+{
+ if ((lro_desc->iph->saddr != iph->saddr)
+ || (lro_desc->iph->daddr != iph->daddr)
+ || (lro_desc->tcph->source != tcph->source)
+ || (lro_desc->tcph->dest != tcph->dest))
+ return -1;
+ return 0;
+}
+
+static struct net_lro_desc *lro_get_desc(struct net_lro_mgr *lro_mgr,
+ struct net_lro_desc *lro_arr,
+ struct iphdr *iph,
+ struct tcphdr *tcph)
+{
+ struct net_lro_desc *lro_desc = NULL;
+ struct net_lro_desc *tmp;
+ int max_desc = lro_mgr->max_desc;
+ int i;
+
+ for (i = 0; i < max_desc; i++) {
+ tmp = &lro_arr[i];
+ if (tmp->active)
+ if (!lro_check_tcp_conn(tmp, iph, tcph)) {
+ lro_desc = tmp;
+ goto out;
+ }
+ }
+
+ for (i = 0; i < max_desc; i++) {
+ if (!lro_arr[i].active) {
+ lro_desc = &lro_arr[i];
+ goto out;
+ }
+ }
+
+ LRO_INC_STATS(lro_mgr, no_desc);
+out:
+ return lro_desc;
+}
+
+static void lro_flush(struct net_lro_mgr *lro_mgr,
+ struct net_lro_desc *lro_desc)
+{
+ if (lro_desc->pkt_aggr_cnt > 1)
+ lro_update_tcp_ip_header(lro_desc);
+
+ skb_shinfo(lro_desc->parent)->gso_size = lro_desc->mss;
+
+ if (lro_desc->vgrp) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(lro_desc->parent,
+ lro_desc->vgrp,
+ lro_desc->vlan_tag);
+ else
+ vlan_hwaccel_rx(lro_desc->parent,
+ lro_desc->vgrp,
+ lro_desc->vlan_tag);
+
+ } else {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(lro_desc->parent);
+ else
+ netif_rx(lro_desc->parent);
+ }
+
+ LRO_INC_STATS(lro_mgr, flushed);
+ lro_clear_desc(lro_desc);
+}
+
+static int __lro_proc_skb(struct net_lro_mgr *lro_mgr, struct sk_buff *skb,
+ struct vlan_group *vgrp, u16 vlan_tag, void *priv)
+{
+ struct net_lro_desc *lro_desc;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ u64 flags;
+ int vlan_hdr_len = 0;
+
+ if (!lro_mgr->get_skb_header
+ || lro_mgr->get_skb_header(skb, (void *)&iph, (void *)&tcph,
+ &flags, priv))
+ goto out;
+
+ if (!(flags & LRO_IPV4) || !(flags & LRO_TCP))
+ goto out;
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (!lro_desc)
+ goto out;
+
+ if ((skb->protocol == htons(ETH_P_8021Q))
+ && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ vlan_hdr_len = VLAN_HLEN;
+
+ if (!lro_desc->active) { /* start new lro session */
+ if (lro_tcp_ip_check(iph, tcph, skb->len - vlan_hdr_len, NULL))
+ goto out;
+
+ skb->ip_summed = lro_mgr->ip_summed_aggr;
+ lro_init_desc(lro_desc, skb, iph, tcph, vlan_tag, vgrp);
+ LRO_INC_STATS(lro_mgr, aggregated);
+ return 0;
+ }
+
+ if (lro_desc->tcp_next_seq != ntohl(tcph->seq))
+ goto out2;
+
+ if (lro_tcp_ip_check(iph, tcph, skb->len, lro_desc))
+ goto out2;
+
+ lro_add_packet(lro_desc, skb, iph, tcph);
+ LRO_INC_STATS(lro_mgr, aggregated);
+
+ if ((lro_desc->pkt_aggr_cnt >= lro_mgr->max_aggr) ||
+ lro_desc->parent->len > (0xFFFF - lro_mgr->dev->mtu))
+ lro_flush(lro_mgr, lro_desc);
+
+ return 0;
+
+out2: /* send aggregated SKBs to stack */
+ lro_flush(lro_mgr, lro_desc);
+
+out: /* Original SKB has to be posted to stack */
+ skb->ip_summed = lro_mgr->ip_summed;
+ return 1;
+}
+
+
+static struct sk_buff *lro_gen_skb(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ void *mac_hdr,
+ int hlen, __wsum sum,
+ u32 ip_summed)
+{
+ struct sk_buff *skb;
+ struct skb_frag_struct *skb_frags;
+ int data_len = len;
+ int hdr_len = min(len, hlen);
+
+ skb = netdev_alloc_skb(lro_mgr->dev, hlen);
+ if (!skb)
+ return NULL;
+
+ skb->len = len;
+ skb->data_len = len - hdr_len;
+ skb->truesize += true_size;
+ skb->tail += hdr_len;
+
+ memcpy(skb->data, mac_hdr, hdr_len);
+
+ skb_frags = skb_shinfo(skb)->frags;
+ while (data_len > 0) {
+ *skb_frags = *frags;
+ data_len -= frags->size;
+ skb_frags++;
+ frags++;
+ skb_shinfo(skb)->nr_frags++;
+ }
+
+ skb_shinfo(skb)->frags[0].page_offset += hdr_len;
+ skb_shinfo(skb)->frags[0].size -= hdr_len;
+
+ skb->ip_summed = ip_summed;
+ skb->csum = sum;
+ skb->protocol = eth_type_trans(skb, lro_mgr->dev);
+ return skb;
+}
+
+static struct sk_buff *__lro_proc_segment(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag, void *priv, __wsum sum)
+{
+ struct net_lro_desc *lro_desc;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ struct sk_buff *skb;
+ u64 flags;
+ void *mac_hdr;
+ int mac_hdr_len;
+ int hdr_len = LRO_MAX_PG_HLEN;
+ int vlan_hdr_len = 0;
+
+ if (!lro_mgr->get_frag_header
+ || lro_mgr->get_frag_header(frags, (void *)&mac_hdr, (void *)&iph,
+ (void *)&tcph, &flags, priv)) {
+ mac_hdr = page_address(frags->page) + frags->page_offset;
+ goto out1;
+ }
+
+ if (!(flags & LRO_IPV4) || !(flags & LRO_TCP))
+ goto out1;
+
+ hdr_len = (int)((void *)(tcph) + TCP_HDR_LEN(tcph) - mac_hdr);
+ mac_hdr_len = (int)((void *)(iph) - mac_hdr);
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (!lro_desc)
+ goto out1;
+
+ if (!lro_desc->active) { /* start new lro session */
+ if (lro_tcp_ip_check(iph, tcph, len - mac_hdr_len, NULL))
+ goto out1;
+
+ skb = lro_gen_skb(lro_mgr, frags, len, true_size, mac_hdr,
+ hdr_len, 0, lro_mgr->ip_summed_aggr);
+ if (!skb)
+ goto out;
+
+ if ((skb->protocol == htons(ETH_P_8021Q))
+ && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ vlan_hdr_len = VLAN_HLEN;
+
+ iph = (void *)(skb->data + vlan_hdr_len);
+ tcph = (void *)((u8 *)skb->data + vlan_hdr_len
+ + IP_HDR_LEN(iph));
+
+ lro_init_desc(lro_desc, skb, iph, tcph, 0, NULL);
+ LRO_INC_STATS(lro_mgr, aggregated);
+ return 0;
+ }
+
+ if (lro_desc->tcp_next_seq != ntohl(tcph->seq))
+ goto out2;
+
+ if (lro_tcp_ip_check(iph, tcph, len - mac_hdr_len, lro_desc))
+ goto out2;
+
+ lro_add_frags(lro_desc, len, hdr_len, true_size, frags, iph, tcph);
+ LRO_INC_STATS(lro_mgr, aggregated);
+
+ if ((skb_shinfo(lro_desc->parent)->nr_frags >= lro_mgr->max_aggr) ||
+ lro_desc->parent->len > (0xFFFF - lro_mgr->dev->mtu))
+ lro_flush(lro_mgr, lro_desc);
+
+ return NULL;
+
+out2: /* send aggregated packets to the stack */
+ lro_flush(lro_mgr, lro_desc);
+
+out1: /* Original packet has to be posted to the stack */
+ skb = lro_gen_skb(lro_mgr, frags, len, true_size, mac_hdr,
+ hdr_len, sum, lro_mgr->ip_summed);
+out:
+ return skb;
+}
+
+void lro_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ void *priv)
+{
+ if (__lro_proc_skb(lro_mgr, skb, NULL, 0, priv)) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(skb);
+ else
+ netif_rx(skb);
+ }
+}
+EXPORT_SYMBOL(lro_receive_skb);
+
+void lro_vlan_hwaccel_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv)
+{
+ if (__lro_proc_skb(lro_mgr, skb, vgrp, vlan_tag, priv)) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
+ else
+ vlan_hwaccel_rx(skb, vgrp, vlan_tag);
+ }
+}
+EXPORT_SYMBOL(lro_vlan_hwaccel_receive_skb);
+
+void lro_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size, void *priv, __wsum sum)
+{
+ struct sk_buff *skb;
+
+ skb = __lro_proc_segment(lro_mgr, frags, len, true_size, NULL, 0,
+ priv, sum);
+ if (!skb)
+ return;
+
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(skb);
+ else
+ netif_rx(skb);
+}
+EXPORT_SYMBOL(lro_receive_frags);
+
+void lro_vlan_hwaccel_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag, void *priv, __wsum sum)
+{
+ struct sk_buff *skb;
+
+ skb = __lro_proc_segment(lro_mgr, frags, len, true_size, vgrp,
+ vlan_tag, priv, sum);
+ if (!skb)
+ return;
+
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
+ else
+ vlan_hwaccel_rx(skb, vgrp, vlan_tag);
+}
+EXPORT_SYMBOL(lro_vlan_hwaccel_receive_frags);
+
+void lro_flush_all(struct net_lro_mgr *lro_mgr)
+{
+ int i;
+ struct net_lro_desc *lro_desc = lro_mgr->lro_arr;
+
+ for (i = 0; i < lro_mgr->max_desc; i++) {
+ if (lro_desc[i].active)
+ lro_flush(lro_mgr, &lro_desc[i]);
+ }
+}
+EXPORT_SYMBOL(lro_flush_all);
+
+void lro_flush_pkt(struct net_lro_mgr *lro_mgr,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct net_lro_desc *lro_desc;
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (lro_desc->active)
+ lro_flush(lro_mgr, lro_desc);
+}
+EXPORT_SYMBOL(lro_flush_pkt);
--
1.5.2
^ permalink raw reply related
* [PATCH 0/1] lro: Generic Large Receive Offload for TCP traffic
From: Jan-Bernd Themann @ 2007-08-03 12:41 UTC (permalink / raw)
To: David Miller
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Andrew Gallatin, Jeff Garzik,
Stefan Roscher, netdev
Hi,
I think this patch could be the final version for now. It has been tested
on two platforms (power and x86_64) and works very well.
Apart from David Miller and Evgeniy Polaykov, we'd like to thank especially
Andrew Gallatin for his great reviews and help to make that happen.
After some discussion we decided to post the LRO patch separately from the
driver patches. Our final driver patches for LRO will be posted later with
some additional fixes for upstream inclusion to the netdev git.
However, I'll also post our LRO patch for the driver today as an example
of how to use this interface.
Thanks a lot,
Jan-Bernd
[PATCH 1/1] lro: Generic Large Receive Offload for TCP traffic
Changes to http://www.spinics.net/lists/netdev/msg37084.html
1) Fixed the LRO_MAX_PG_HLEN bug
2) skb->ip_summed can now be defined by driver for aggregated packets
3) The problem that the "ramp up" for tcp connections between machines
with different MTU size (1500 vs 9000) is very slow has been fixed
by setting skb->gso_size.
4) Checksum problem for little endian machines has been fixed
5) missing additon of vlan_hdr_len for TCP header determination has been added.
^ permalink raw reply
* Re: Distributed storage.
From: Peter Zijlstra @ 2007-08-03 12:27 UTC (permalink / raw)
To: Evgeniy Polyakov
Cc: Daniel Phillips, netdev, linux-kernel, linux-fsdevel,
Arnaldo Carvalho de Melo
In-Reply-To: <20070803105747.GE10089@2ka.mipt.ru>
On Fri, 2007-08-03 at 14:57 +0400, Evgeniy Polyakov wrote:
> For receiving situation is worse, since system does not know in advance
> to which socket given packet will belong to, so it must allocate from
> global pool (and thus there must be independent global reserve), and
> then exchange part of the socket's reserve to the global one (or just
> copy packet to the new one, allocated from socket's reseve is it was
> setup, or drop it otherwise). Global independent reserve is what I
> proposed when stopped to advertise network allocator, but it seems that
> it was not taken into account, and reserve was always allocated only
> when system has serious memory pressure in Peter's patches without any
> meaning for per-socket reservation.
This is not true. I have a global reserve which is set-up a priori. You
cannot allocate a reserve when under pressure, that does not make sense.
Let me explain my approach once again.
At swapon(8) time we allocate a global reserve. And associate the needed
sockets with it. The size of this global reserve is make up of two
parts:
- TX
- RX
The RX pool is the most interresting part. It again is made up of two
parts:
- skb
- auxilary data
The skb part is scaled such that it can overflow the IP fragment
reassembly, the aux pool such that it can overflow the route cache (that
was the largest other allocator in the RX path)
All (reserve) RX skb allocations are accounted, so as to never allocate
more than we reserved.
All packets are received (given the limit) and are processed up to
socket demux. At that point all packets not targeted at an associated
socket are dropped and the skb memory freed - ready for another packet.
All packets targeted for associated sockets get processed. This requires
that this packet processing happens in-kernel. Since we are swapping
user-space might be waiting for this data, and we'd deadlock.
I'm not quite sure why you need per socket reservations.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox