GitHub/moto-9609/android_kernel_motorola_exynos9610.git
16 years agonet: Add STP demux layer
Patrick McHardy [Sun, 6 Jul 2008 04:25:39 +0000 (21:25 -0700)]
net: Add STP demux layer

Add small STP demux layer for demuxing STP PDUs based on MAC address.
This is needed to run both GARP and STP in parallel (or even load the
modules) since both use LLC_SAP_BSPAN.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMIB: add struct net to UDP6_INC_STATS_BH
Pavel Emelyanov [Sun, 6 Jul 2008 04:19:40 +0000 (21:19 -0700)]
MIB: add struct net to UDP6_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMIB: add struct net to UDP6_INC_STATS_USER
Pavel Emelyanov [Sun, 6 Jul 2008 04:19:20 +0000 (21:19 -0700)]
MIB: add struct net to UDP6_INC_STATS_USER

As simple as the patch #1 in this set.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMIB: add struct net to UDP_INC_STATS_BH
Pavel Emelyanov [Sun, 6 Jul 2008 04:18:48 +0000 (21:18 -0700)]
MIB: add struct net to UDP_INC_STATS_BH

Two special cases here - one is rxrpc - I put init_net there
explicitly, since we haven't touched this part yet. The second
place is in __udp4_lib_rcv - we already have a struct net there,
but I have to move its initialization above to make it ready
at the "drop" label.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMIB: add struct net to UDP_INC_STATS_USER
Pavel Emelyanov [Sun, 6 Jul 2008 04:18:07 +0000 (21:18 -0700)]
MIB: add struct net to UDP_INC_STATS_USER

Nothing special - all the places already have a struct sock
at hands, so use the sock_net() net.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMerge branch 'davem-next' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
David S. Miller [Sun, 6 Jul 2008 04:03:31 +0000 (21:03 -0700)]
Merge branch 'davem-next' of /linux/kernel/git/jgarzik/netdev-2.6

16 years agonetns: selective flush of rt_cache
Denis V. Lunev [Sun, 6 Jul 2008 02:06:12 +0000 (19:06 -0700)]
netns: selective flush of rt_cache

dst cache is marked as expired on the per/namespace basis by previous
path. Right now we have to implement selective cache shrinking. This
procedure has been ported from older OpenVz codebase.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetns: place rt_genid into struct net
Denis V. Lunev [Sun, 6 Jul 2008 02:04:32 +0000 (19:04 -0700)]
netns: place rt_genid into struct net

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoipv4: pass current value of rt_genid into rt_hash
Denis V. Lunev [Sun, 6 Jul 2008 02:04:09 +0000 (19:04 -0700)]
ipv4: pass current value of rt_genid into rt_hash

Basically, there is no difference to atomic_read internally or pass it as
a parameter as rt_hash is inline.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetns: add struct net parameter to rt_cache_invalidate
Denis V. Lunev [Sun, 6 Jul 2008 02:03:31 +0000 (19:03 -0700)]
netns: add struct net parameter to rt_cache_invalidate

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetns: make rt_secret_rebuild timer per namespace
Denis V. Lunev [Sun, 6 Jul 2008 02:02:59 +0000 (19:02 -0700)]
netns: make rt_secret_rebuild timer per namespace

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetns: register net.ipv4.route.flush in each namespace
Denis V. Lunev [Sun, 6 Jul 2008 02:02:33 +0000 (19:02 -0700)]
netns: register net.ipv4.route.flush in each namespace

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoipv4: remove static flush_delay variable
Denis V. Lunev [Sun, 6 Jul 2008 02:02:06 +0000 (19:02 -0700)]
ipv4: remove static flush_delay variable

flush delay is used as an external storage for net.ipv4.route.flush sysctl
entry. It is write-only.

The ctl_table->data for this entry is used once. Fix this case to point
to the stack to remove global variable. Do this to avoid additional
variable on struct net in the next patch.

Possible race (as it was before) accessing this local variable is removed
using flush_mutex.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonet: add fib_rules_ops to flush_cache method
Denis V. Lunev [Sun, 6 Jul 2008 02:01:28 +0000 (19:01 -0700)]
net: add fib_rules_ops to flush_cache method

This is required to pass namespace context into rt_cache_flush called from
->flush_cache.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetns: add namespace parameter to rt_cache_flush
Denis V. Lunev [Sun, 6 Jul 2008 02:00:44 +0000 (19:00 -0700)]
netns: add namespace parameter to rt_cache_flush

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago8390: Split 8390 support into a pausing and a non pausing driver core
Alan Cox [Fri, 4 Jul 2008 06:43:12 +0000 (23:43 -0700)]
8390: Split 8390 support into a pausing and a non pausing driver core

Only a few ISA controllers need the pausing version of the 8390 core
while PCMCIA, later ISA and PCI do not. More importantly the ISA delays
can break non ISA boxes so we must use a different build of 8390.c for
the two sets of controllers.

No changes since last time as all the points of concerns raised proved to
be invalid

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoWAN: Remove no longer needed pvc_desc in FR code.
Krzysztof Halasa [Tue, 1 Jul 2008 12:42:07 +0000 (14:42 +0200)]
WAN: Remove no longer needed pvc_desc in FR code.

Remove no longer needed struct pvc_desc in FR code.
Requires "WAN: convert drivers to use built-in netdev_stats" patch.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoWAN: convert drivers to use built-in netdev_stats
Krzysztof Halasa [Mon, 30 Jun 2008 21:26:53 +0000 (23:26 +0200)]
WAN: convert drivers to use built-in netdev_stats

There is no point in using separate net_device_stats structs when
the one in struct net_device is present. Compiles.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: add NAPI Rx queue support
PJ Waskiewicz [Fri, 27 Jun 2008 18:00:39 +0000 (11:00 -0700)]
igb: add NAPI Rx queue support

Update the NAPI implementation to use the new napi_struct infrstructure.
This removes the need of multiple net_device structs to implement a
multiqueue NAPI.

Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: fix init on 82575 with MNG enabled
Alexander Duyck [Fri, 27 Jun 2008 18:00:29 +0000 (11:00 -0700)]
igb: fix init on 82575 with MNG enabled

This patch resolves an issue seen on 82575 adapters with managability
pass-thru enabled, which could cause the system to panic.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: eliminate hw from the hw_dbg macro arguments
Auke Kok [Fri, 27 Jun 2008 18:00:18 +0000 (11:00 -0700)]
igb: eliminate hw from the hw_dbg macro arguments

Various cosmetic cleanups. Comment fixes. Eliminate the hw part out
of the hw_dbg macro since it's always used.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: fix parameter options
Alexander Duyck [Fri, 27 Jun 2008 18:00:07 +0000 (11:00 -0700)]
igb: fix parameter options

Currently there are items in the driver to support parameters, but there
are no parameters being added to the driver.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: cleanup function header comments
Jeff Kirsher [Fri, 27 Jun 2008 17:59:59 +0000 (10:59 -0700)]
igb: cleanup function header comments

Function header comments do not match function name.

Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: Remove adapter struct from these function call parameters
Mitch Williams [Fri, 27 Jun 2008 17:59:48 +0000 (10:59 -0700)]
igb: Remove adapter struct from these function call parameters

Simplify a bunch of calls to these functions by moving the adapter
struct out of the function arguments since that is not needed.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: limit EEPROM access
Jeff Kirsher [Fri, 27 Jun 2008 17:59:33 +0000 (10:59 -0700)]
igb: limit EEPROM access

EEPROM access beyond the base eeprom page is unsupported.

Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: run-time configurable jumbo frame support
Pekka Enberg [Mon, 23 Jun 2008 11:36:56 +0000 (14:36 +0300)]
ipg: run-time configurable jumbo frame support

Make jumbo frame support configurable via ifconfig mtu option as suggested by
Stephen Hemminger.

Cc: Stephen Hemminger <stephen.hemminger@vyatta.com>
Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: per-device max_rxframe_size
Pekka Enberg [Mon, 23 Jun 2008 11:36:35 +0000 (14:36 +0300)]
ipg: per-device max_rxframe_size

Add a ->max_rxframe member to struct ipg_nic_private and convert the users of
IPG_MAX_RXFRAME_SIZE to use it instead to enable per-device jumbo frame
configuration.

Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: per-device rxsupport_size
Pekka Enberg [Mon, 23 Jun 2008 11:36:18 +0000 (14:36 +0300)]
ipg: per-device rxsupport_size

Add a ->max_rxframe member to struct ipg_nic_private and convert the users of
IPG_RXSUPPORT_SIZE to use it instead to enable per-device jumbo frame
configuration.

Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: per-device rxfrag_size
Pekka Enberg [Mon, 23 Jun 2008 11:36:00 +0000 (14:36 +0300)]
ipg: per-device rxfrag_size

Add a ->max_rxframe member to struct ipg_nic_private and convert the users of
IPG_RXFRAG_SIZE to use it instead to enable per-device jumbo frame
configuration.

Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: remove jumbo frame #ifdef from mtu
Pekka Enberg [Mon, 23 Jun 2008 11:35:37 +0000 (14:35 +0300)]
ipg: remove jumbo frame #ifdef from mtu

Remove JUMBO_FRAME #ifdef from dev->mtu setting in ipg_nic_open() so that we
can make IPG_TXFRAG_SIZE configurable.

Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoipg: always compile in jumbo frame support
Pekka Enberg [Mon, 23 Jun 2008 11:35:16 +0000 (14:35 +0300)]
ipg: always compile in jumbo frame support

Add a ->is_jumbo boolean to struct ipg_nic_private and fix up
ipg_interrupt_handler() to call the jumbo frame version of ipg_nic_rx() if the
boolean is set to true. Also remove the JUMBO_FRAME #ifdefs so we can always
compile in support for jumbo frames.

Tested-by: Andrew Savchenko <Bircoph@list.ru>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoMerge branch 'r8169-next' of git://git.kernel.org/pub/scm/linux/kernel/git/romieu...
Jeff Garzik [Fri, 4 Jul 2008 12:35:57 +0000 (08:35 -0400)]
Merge branch 'r8169-next' of git://git./linux/kernel/git/romieu/netdev-2.6 into upstream-next

16 years agotun: Allow GSO using virtio_net_hdr
Rusty Russell [Thu, 3 Jul 2008 10:48:02 +0000 (03:48 -0700)]
tun: Allow GSO using virtio_net_hdr

Add a IFF_VNET_HDR flag.  This uses the same ABI as virtio_net
(ie. prepending struct virtio_net_hdr to packets) to indicate GSO and
checksum information.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agotun: TUNSETFEATURES to set gso features.
Rusty Russell [Thu, 3 Jul 2008 10:46:16 +0000 (03:46 -0700)]
tun: TUNSETFEATURES to set gso features.

ethtool is useful for setting (some) device fields, but it's
root-only.  Finer feature control is available through a tun-specific
ioctl.

(Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agotun: Interface to query tun/tap features.
Rusty Russell [Thu, 3 Jul 2008 10:45:32 +0000 (03:45 -0700)]
tun: Interface to query tun/tap features.

The problem with introducing checksum offload and gso to tun is they
need to set dev->features to enable GSO and/or checksumming, which is
supposed to be done before register_netdevice(), ie. as part of
TUNSETIFF.

Unfortunately, TUNSETIFF has always just ignored flags it doesn't
understand, so there's no good way of detecting whether the kernel
supports new IFF_ flags.

This patch implements a TUNGETFEATURES ioctl which returns all the valid IFF
flags.  It could be extended later to include other features.

Here's an example program which uses it:

#include <linux/if_tun.h>
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <err.h>
#include <stdio.h>

static struct {
unsigned int flag;
const char *name;
} known_flags[] = {
{ IFF_TUN, "TUN" },
{ IFF_TAP, "TAP" },
{ IFF_NO_PI, "NO_PI" },
{ IFF_ONE_QUEUE, "ONE_QUEUE" },
};

int main()
{
unsigned int features, i;

int netfd = open("/dev/net/tun", O_RDWR);
if (netfd < 0)
err(1, "Opening /dev/net/tun");

if (ioctl(netfd, TUNGETFEATURES, &features) != 0) {
printf("Kernel does not support TUNGETFEATURES, guessing\n");
features = (IFF_TUN|IFF_TAP|IFF_NO_PI|IFF_ONE_QUEUE);
}
printf("Available features are: ");
for (i = 0; i < sizeof(known_flags)/sizeof(known_flags[0]); i++) {
if (features & known_flags[i].flag) {
features &= ~known_flags[i].flag;
printf("%s ", known_flags[i].name);
}
}
if (features)
printf("(UNKNOWN %#x)", features);
printf("\n");
return 0;
}

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoMerge branch 'net-next-2.6-v6ready-20080703' of git://git.linux-ipv6.org/gitroot...
David S. Miller [Thu, 3 Jul 2008 10:07:58 +0000 (03:07 -0700)]
Merge branch 'net-next-2.6-v6ready-20080703' of git://git.linux-ipv6.org/gitroot/yoshfuji/linux-2.6-next

16 years agoipv4,ipv6 mroute: Add some helper inline functions to remove ugly ifdefs.
YOSHIFUJI Hideaki [Thu, 3 Jul 2008 07:51:22 +0000 (16:51 +0900)]
ipv4,ipv6 mroute: Add some helper inline functions to remove ugly ifdefs.

ip{,v6}_mroute_{set,get}sockopt() should not matter by optimization but
it would be better not to depend on optimization semantically.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv4: Do cleanup for ip_mr_init
Wang Chen [Thu, 3 Jul 2008 04:13:36 +0000 (12:13 +0800)]
ipv4: Do cleanup for ip_mr_init

Same as ip6_mr_init(), make ip_mr_init() return errno if fails.
But do not do error handling in inet_init(), just print a msg.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6: Do cleanup for ip6_mr_init.
Wang Chen [Thu, 3 Jul 2008 04:13:30 +0000 (12:13 +0800)]
ipv6: Do cleanup for ip6_mr_init.

If do not do it, we will get following issues:
1. Leaving junks after inet6_init failing halfway.
2. Leaving proc and notifier junks after ipv6 modules unloading.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6 route: Prefer outgoing interface with source address assigned.
YOSHIFUJI Hideaki [Wed, 2 Jul 2008 09:30:18 +0000 (18:30 +0900)]
ipv6 route: Prefer outgoing interface with source address assigned.

Outgoing interface is selected by the route decision if unspecified.
Let's prefer routes via interface(s) with the address assigned if we
have multiple routes with same cost.
With help from Naohiro Ooiwa <nooiwa@miraclelinux.com>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6 addrconf: add accept_dad sysctl to control DAD operation.
YOSHIFUJI Hideaki [Sat, 28 Jun 2008 05:18:38 +0000 (14:18 +0900)]
ipv6 addrconf: add accept_dad sysctl to control DAD operation.

- If 0, disable DAD.
- If 1, perform DAD (default).
- If >1, perform DAD and disable IPv6 operation if DAD for MAC-based
  link-local address has been failed (RFC4862 5.4.5).

We do not follow RFC4862 by default.  Refer to the netdev thread entitled
"Linux IPv6 DAD not full conform to RFC 4862 ?"
http://www.spinics.net/lists/netdev/msg52027.html

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6: Add disable_ipv6 sysctl to disable IPv6 operaion on specific interface.
YOSHIFUJI Hideaki [Sat, 28 Jun 2008 05:17:11 +0000 (14:17 +0900)]
ipv6: Add disable_ipv6 sysctl to disable IPv6 operaion on specific interface.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6: Assume the loopback address in link-local scope.
YOSHIFUJI Hideaki [Wed, 25 Jun 2008 07:58:17 +0000 (16:58 +0900)]
ipv6: Assume the loopback address in link-local scope.

Handle interface property strictly when looking up a route
for the loopback address (RFC4291 2.5.3).

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6: Do not forward packets with the unspecified source address.
YOSHIFUJI Hideaki [Wed, 25 Jun 2008 07:55:26 +0000 (16:55 +0900)]
ipv6: Do not forward packets with the unspecified source address.

RFC4291 2.5.2.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agoipv6: Do not assign non-valid address on interface.
YOSHIFUJI Hideaki [Wed, 25 Jun 2008 07:26:47 +0000 (16:26 +0900)]
ipv6: Do not assign non-valid address on interface.

Check the type of the address when adding a new one on interface.
- the unspecified address (::) is always disallowed (RFC4291 2.5.2)
- the loopback address is disallowed unless the interface is (one of)
  loopback (RFC4291 2.5.3).
- multicast addresses are disallowed.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
16 years agotcp: de-bloat a bit with factoring NET_INC_STATS_BH out
Pavel Emelyanov [Thu, 3 Jul 2008 08:05:41 +0000 (01:05 -0700)]
tcp: de-bloat a bit with factoring NET_INC_STATS_BH out

There are some places in TCP that select one MIB index to
bump snmp statistics like this:

if (<something>)
NET_INC_STATS_BH(<some_id>);
else if (<something_else>)
NET_INC_STATS_BH(<some_other_id>);
...
else
NET_INC_STATS_BH(<default_id>);

or in a more tricky but still similar way.

On the other hand, this NET_INC_STATS_BH is a camouflaged
increment of percpu variable, which is not that small.

Factoring those cases out de-bloats 235 bytes on non-preemptible
i386 config and drives parts of the code into 80 columns.

add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-235 (-235)
function                                     old     new   delta
tcp_fastretrans_alert                       1437    1424     -13
tcp_dsack_set                                137     124     -13
tcp_xmit_retransmit_queue                    690     676     -14
tcp_try_undo_recovery                        283     265     -18
tcp_sacktag_write_queue                     1550    1515     -35
tcp_update_reordering                        162     106     -56
tcp_retransmit_timer                         990     904     -86

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoniu: Add support for rx flow hash configuration.
Santwona Behera [Wed, 2 Jul 2008 10:49:11 +0000 (03:49 -0700)]
niu: Add support for rx flow hash configuration.

Implemented ethtool callback functions for configuring receive flow
hashing in the niu driver.

Signed-off-by: Santwona Behera <santwona.behera@sun.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetdev: Add support for rx flow hash configuration, using ethtool.
Santwona Behera [Wed, 2 Jul 2008 10:47:41 +0000 (03:47 -0700)]
netdev: Add support for rx flow hash configuration, using ethtool.

Added new interfaces to ethtool to configure receive network flow
distribution across multiple rx rings using hashing.

Signed-off-by: Santwona Behera <santwona.behera@sun.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agosctp: Mark GET_PEER|LOCAL_ADDR_OLD deprecated.
Vlad Yasevich [Wed, 2 Jul 2008 03:06:22 +0000 (20:06 -0700)]
sctp: Mark GET_PEER|LOCAL_ADDR_OLD deprecated.

Socket options SCTP_GET_PEER_ADDR_OLD, SCTP_GET_PEER_ADDR_NUM_OLD,
SCTP_GET_LOCAL_ADDR_OLD, and SCTP_GET_PEER_LOCAL_ADDR_NUM_OLD
have been replaced by newer versions a since 2005.  It's time
to officially deprecate them and schedule them for removal.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoicmp: fix units for ratelimit
Stephen Hemminger [Wed, 2 Jul 2008 02:29:07 +0000 (19:29 -0700)]
icmp: fix units for ratelimit

Convert the sysctl values for icmp ratelimit to use milliseconds instead
of jiffies which is based on kernel configured HZ.
Internal kernel jiffies are not a proper unit for any userspace API.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agor8169: remove non-napi code
Francois Romieu [Sun, 11 May 2008 12:51:00 +0000 (14:51 +0200)]
r8169: remove non-napi code

It will almost unavoidably cause some breakage but it
is long overdue.

The driver identification string has been updated, a
lost tabulation and some unused code have been removed.
Otherwise the code paths should stay the same.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Cc: Edward Hsu <edward_hsu@realtek.com.tw>
16 years agor8169: multicast register update (sync with Realtek's 8.004.00 8168 driver)
Francois Romieu [Wed, 26 Dec 2007 21:46:05 +0000 (22:46 +0100)]
r8169: multicast register update (sync with Realtek's 8.004.00 8168 driver)

The layout of the 8168 serie is different from that of the 8110 one.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Cc: Edward Hsu <edward_hsu@realtek.com.tw>
16 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/linville/wireles...
David S. Miller [Sun, 29 Jun 2008 05:57:58 +0000 (22:57 -0700)]
Merge branch 'master' of /linux/kernel/git/linville/wireless-next-2.6

16 years agoMerge branch 'davem-next' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
David S. Miller [Sun, 29 Jun 2008 04:28:46 +0000 (21:28 -0700)]
Merge branch 'davem-next' of /linux/kernel/git/jgarzik/netdev-2.6

16 years ago[netdrvr] kill sync_irq-before-freq_irq pattern
Jeff Garzik [Fri, 27 Jun 2008 06:20:20 +0000 (02:20 -0400)]
[netdrvr] kill sync_irq-before-freq_irq pattern

synchronize_irq() is superfluous when free_irq() call immediately follows it,
because free_irq() also does a synchronize_irq() call of its own.

Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years ago[netdrvr] fealnx: clean up nasty mess of arch ifdefs
Jeff Garzik [Fri, 27 Jun 2008 06:18:50 +0000 (02:18 -0400)]
[netdrvr] fealnx: clean up nasty mess of arch ifdefs

Clean up config/burst value arch-specific setup.

* bcrvalue only varied by its big-endian bit
* crvalue only varied for certain types of x86-32 chips

This should make fealnx quite a bit more portable, without any behavior
change.

Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agotulip: remove wrapper around get_unaligned
Harvey Harrison [Wed, 28 May 2008 23:51:04 +0000 (16:51 -0700)]
tulip: remove wrapper around get_unaligned

DE_UNALIGNED_16 is always being passed a u16 *, no need to have the
wrapper with two casts in it, just call get_unaligned directly.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoFix forcedeth hibernate/wake-on-lan problems
Tobias Diedrich [Sat, 31 May 2008 23:20:05 +0000 (01:20 +0200)]
Fix forcedeth hibernate/wake-on-lan problems

This patch is the minimal amount of code needed to support
wake-on-lan in platform mode properly (i.e. "ethtool -s eth0 wol g"
is sufficient, no additional magic needed) for me.

This is derived from David Brownells patch
(http://lists.laptop.org/pipermail/devel/2007-April/004691.html).
However I decided to move the hook into pci-acpi.c since the other
two pci hooks also live there and pci and acpi are the only users of
the platform_enable_wakeup-hook.

As a 'side-effect' this also makes wake on usb activity work for me
and I had to disable usb wakeup (which is enabled by default) using
the power/wakeup sysfs functionality ("echo disabled >
${sysfs_path_to_device}/power/wakeup").

(BTW I first thought the 'immediate reboot because of usb wake' effect is
caused by the optical mouse generating a wake event, but it rather
seems to be a problem with a flaky secondary usb host controller,
which sees a connected device where nothing is attached)

Signed-off-by: Tobias Diedrich <ranma+kernel@tdiedrich.de>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoFix forcedeth hibernate/wake-on-lan problems
Tobias Diedrich [Sat, 31 May 2008 22:54:42 +0000 (00:54 +0200)]
Fix forcedeth hibernate/wake-on-lan problems

We currently don't signal the kernel we that this device can wake
the system.  Call device_init_wakeup() to correct this.
Without this device_can_wakeup and device_may_wakeup will return
incorrect values.
Together with the minimized acpi wakeup patch (6/4 ;)), which will
follow in the next mail, this really makes wake-on-lan work for me
as expected (i.e. "ethtool -s eth0 wol g" is sufficient, no
additional magic needed).

Signed-off-by: Tobias Diedrich <ranma+kernel@tdiedrich.de>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years ago8139too: some style cleanups
Márton Németh [Sun, 15 Jun 2008 07:52:30 +0000 (09:52 +0200)]
8139too: some style cleanups

Clean up the following errors and warnings reported by checkpatch.pl:
 + ERROR: Macros with complex values should be enclosed in parenthesis
 + WARNING: __func__ should be used instead of gcc specific __FUNCTION__
 + WARNING: plain inline is preferred over __inline__
 + WARNING: Use #include <linux/io.h> instead of <asm/io.h>
 + WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h>

The changes were verified with by comparing the "objdump -d 8139too.ko"
output which is exactly the same for the old and new version in case of
config CONFIG_8139TOO=m, CONFIG_8139TOO_PIO=n, CONFIG_8139TOO_TUNE_TWISTER=n,
CONFIG_8139TOO_8129=n, CONFIG_8139_OLD_RX_RESET=n.
Software versions used: gcc 4.2.3, objdump 2.18.0.20080103, on elf32-i386.

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agorndis_host: pass buffer length to rndis_command
Jussi Kivilinna [Wed, 18 Jun 2008 12:40:12 +0000 (15:40 +0300)]
rndis_host: pass buffer length to rndis_command

Pass buffer length to rndis_command so that rndis_command can read full
response buffer from device instead of max CONTROL_BUFFER_SIZE bytes.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agonet: sh_eth: Fix compile error sh_eth
Nobuhiro Iwamatsu [Wed, 18 Jun 2008 09:32:09 +0000 (18:32 +0900)]
net: sh_eth: Fix compile error sh_eth

Fix compile error on sh_eth and remove base address macro.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoe1000: remove e1000_clean_tx_irq call from e1000_netpoll
Andy Gospodarek [Thu, 19 Jun 2008 21:19:12 +0000 (17:19 -0400)]
e1000: remove e1000_clean_tx_irq call from e1000_netpoll

The call to e1000_clean_tx_irq in e1000_netpoll can race with the call
to e1000_clean_tx_irq in e1000_clean.  With a small bit of tweaking to
to netpoll_send_skb to simulate a system that was under extreme stress,
I was able to reproduce these concurrent calls.  This can result in
multiple frees to the skbs on the tx ring buffer.

Dropping this call from e1000_netpoll should be fine since we can rely
on the calls in e1000_clean to do what is needed since napi will poll
the hardware just after calling poll_controller.

Signed-off-by: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoigb: make ioport free
Taku Izumi [Fri, 20 Jun 2008 03:10:30 +0000 (12:10 +0900)]
igb: make ioport free

This patch makes igb driver ioport-free.
This corrects behavior in probe function so as not to request ioport
resources as long as they are not really needed.

Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoe1000e: make ioport free
Taku Izumi [Fri, 20 Jun 2008 02:57:02 +0000 (11:57 +0900)]
e1000e: make ioport free

This patch makes e1000e driver ioport-free.
This corrects behavior in probe function so as not to request ioport
resources as long as they are not really needed.

Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoe1000: remove PCI Express device IDs
Auke Kok [Sun, 22 Jun 2008 22:21:29 +0000 (15:21 -0700)]
e1000: remove PCI Express device IDs

We do not want to prolong the situation much longer that e1000
and e1000e support these devices at the same time. As a result,
take out the bandage that was added for the interim period
and remove all the PCI Express device IDs from e1000.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
David S. Miller [Sat, 28 Jun 2008 08:19:40 +0000 (01:19 -0700)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6

Conflicts:

drivers/net/wireless/iwlwifi/iwl4965-base.c

16 years agoipv6 route: Convert rt6_device_match() to use RT6_LOOKUP_F_xxx flags.
YOSHIFUJI Hideaki [Sat, 28 Jun 2008 03:14:54 +0000 (20:14 -0700)]
ipv6 route: Convert rt6_device_match() to use RT6_LOOKUP_F_xxx flags.

The commit 77d16f450ae0452d7d4b009f78debb1294fb435c ("[IPV6] ROUTE:
Unify RT6_F_xxx and RT6_SELECT_F_xxx flags") intended to pass various
routing lookup hints around RT6_LOOKUP_F_xxx flags, but conversion was
missing for rt6_device_match().

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetlabel: Fix a problem when dumping the default IPv6 static labels
Paul Moore [Sat, 28 Jun 2008 03:12:32 +0000 (20:12 -0700)]
netlabel: Fix a problem when dumping the default IPv6 static labels

There is a missing "!" in a conditional statement which is causing entries to
be skipped when dumping the default IPv6 static label entries.  This can be
demonstrated by running the following:

 # netlabelctl unlbl add default address:::1 \
                                 label:system_u:object_r:unlabeled_t:s0
 # netlabelctl -p unlbl list

... you will notice that the entry for the IPv6 localhost address is not
displayed but does exist (works correctly, causes collisions when attempting
to add duplicate entries, etc.).

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonet/inet_lro: remove setting skb->ip_summed when not LRO-able
Eli Cohen [Sat, 28 Jun 2008 03:09:00 +0000 (20:09 -0700)]
net/inet_lro: remove setting skb->ip_summed when not LRO-able

When an SKB cannot be chained to a session, the current code attempts
to "restore" its ip_summed field from lro_mgr->ip_summed. However,
lro_mgr->ip_summed does not hold the original value; in fact, we'd
better not touch skb->ip_summed since it is not modified by the code
in the path leading to a failure to chain it.  Also use a cleaer
comment to the describe the ip_summed field of struct net_lro_mgr.

Issue raised by Or Gerlitz <ogerlitz@voltaire.com>

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoinet fragments: fix race between inet_frag_find and inet_frag_secret_rebuild
Pavel Emelyanov [Sat, 28 Jun 2008 03:06:08 +0000 (20:06 -0700)]
inet fragments: fix race between inet_frag_find and inet_frag_secret_rebuild

The problem is that while we work w/o the inet_frags.lock even
read-locked the secret rebuild timer may occur (on another CPU, since
BHs are still disabled in the inet_frag_find) and change the rnd seed
for ipv4/6 fragments.

It was caused by my patch fd9e63544cac30a34c951f0ec958038f0529e244
([INET]: Omit double hash calculations in xxx_frag_intern) late
in the 2.6.24 kernel, so this should probably be queued to -stable.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoCONNECTOR: add a proc entry to list connectors
Li Zefan [Sat, 28 Jun 2008 03:03:24 +0000 (20:03 -0700)]
CONNECTOR: add a proc entry to list connectors

I got a problem when I wanted to check if the kernel supports process
event connector, and It seems there's no way to do this check.

At best I can check if the kernel supports connector or not, by looking
into /proc/net/netlink, or maybe checking the return value of bind() to
see if it's ENOENT.

So it would be useful to add /proc/net/connector to list all supported
connectors:
 # cat /proc/net/connector
 Name            ID
 connector       4294967295:4294967295
 cn_proc         1:1
 w1              3:1

Changelog:
- fix memory leak: s/seq_release/single_release
- use spin_lock_bh instead of spin_lock_irqsave

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetlink: Fix some doc comments in net/netlink/attr.c
Julius Volz [Sat, 28 Jun 2008 03:02:14 +0000 (20:02 -0700)]
netlink: Fix some doc comments in net/netlink/attr.c

Fix some doc comments to match function and attribute names in
net/netlink/attr.c.

Signed-off-by: Julius Volz <juliusv@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agotcp: /proc/net/tcp rto,ato values not scaled properly (v2)
Stephen Hemminger [Sat, 28 Jun 2008 03:00:19 +0000 (20:00 -0700)]
tcp: /proc/net/tcp rto,ato values not scaled properly (v2)

I found another case where we are sending information to userspace
in the wrong HZ scale.  This should have been fixed back in 2.5 :-(

This means an ABI change but as it stands there is no way for an application
like ss to get the right value.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoinclude/linux/netdevice.h: don't export MAX_HEADER to userspace
Adrian Bunk [Sat, 28 Jun 2008 02:54:54 +0000 (19:54 -0700)]
include/linux/netdevice.h: don't export MAX_HEADER to userspace

Due to the CONFIG_'s the value is anyway not correct in userspace.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agopkt_sched: Remove CONFIG_NET_SCH_RR
Adrian Bunk [Sat, 28 Jun 2008 02:54:05 +0000 (19:54 -0700)]
pkt_sched: Remove CONFIG_NET_SCH_RR

Commit d62733c8e437fdb58325617c4b3331769ba82d70
([SCHED]: Qdisc changes and sch_rr added for multiqueue)
added a NET_SCH_RR option that was unused since the code
went unconditionally into sch_prio.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agopkt_sched: ERR_PTR() ususally encodes an negative errno, not positive.
WANG Cong [Sat, 28 Jun 2008 02:51:35 +0000 (19:51 -0700)]
pkt_sched: ERR_PTR() ususally encodes an negative errno, not positive.

Note, in the following patch, 'err' is initialized as:

int err = -ENOBUFS;

Signed-off-by: WANG Cong <wcong@critical-links.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetdevice: Fix typo of dev_unicast_add() comment
Wang Chen [Sat, 28 Jun 2008 02:35:16 +0000 (19:35 -0700)]
netdevice: Fix typo of dev_unicast_add() comment

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoaf_unix: fix 'poll for write'/connected DGRAM sockets
Rainer Weikusat [Sat, 28 Jun 2008 02:34:18 +0000 (19:34 -0700)]
af_unix: fix 'poll for write'/connected DGRAM sockets

For n:1 'datagram connections' (eg /dev/log), the unix_dgram_sendmsg
routine implements a form of receiver-imposed flow control by
comparing the length of the receive queue of the 'peer socket' with
the max_ack_backlog value stored in the corresponding sock structure,
either blocking the thread which caused the send-routine to be called
or returning EAGAIN. This routine is used by both SOCK_DGRAM and
SOCK_SEQPACKET sockets. The poll-implementation for these socket types
is datagram_poll from core/datagram.c. A socket is deemed to be
writeable by this routine when the memory presently consumed by
datagrams owned by it is less than the configured socket send buffer
size. This is always wrong for PF_UNIX non-stream sockets connected to
server sockets dealing with (potentially) multiple clients if the
abovementioned receive queue is currently considered to be full.
'poll' will then return, indicating that the socket is writeable, but
a subsequent write result in EAGAIN, effectively causing an (usual)
application to 'poll for writeability by repeated send request with
O_NONBLOCK set' until it has consumed its time quantum.

The change below uses a suitably modified variant of the datagram_poll
routines for both type of PF_UNIX sockets, which tests if the
recv-queue of the peer a socket is connected to is presently
considered to be 'full' as part of the 'is this socket
writeable'-checking code. The socket being polled is additionally
put onto the peer_wait wait queue associated with its peer, because the
unix_dgram_recvmsg routine does a wake up on this queue after a
datagram was received and the 'other wakeup call' is done implicitly
as part of skb destruction, meaning, a process blocked in poll
because of a full peer receive queue could otherwise sleep forever
if no datagram owned by its socket was already sitting on this queue.
Among this change is a small (inline) helper routine named
'unix_recvq_full', which consolidates the actual testing code (in three
different places) into a single location.

Signed-off-by: Rainer Weikusat <rweikusat@mssgmbh.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agotcp: fix for splice receive when used with software LRO
Octavian Purdila [Sat, 28 Jun 2008 00:27:21 +0000 (17:27 -0700)]
tcp: fix for splice receive when used with software LRO

If an skb has nr_frags set to zero but its frag_list is not empty (as
it can happen if software LRO is enabled), and a previous
tcp_read_sock has consumed the linear part of the skb, then
__skb_splice_bits:

(a) incorrectly reports an error and

(b) forgets to update the offset to account for the linear part

Any of the two problems will cause the subsequent __skb_splice_bits
call (the one that handles the frag_list skbs) to either skip data,
or, if the unadjusted offset is greater then the size of the next skb
in the frag_list, make tcp_splice_read loop forever.

Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agotcp: calculate tcp_mem based on low memory instead of all memory
Miquel van Smoorenburg [Sat, 28 Jun 2008 00:23:57 +0000 (17:23 -0700)]
tcp: calculate tcp_mem based on low memory instead of all memory

The tcp_mem array which contains limits on the total amount of memory
used by TCP sockets is calculated based on nr_all_pages.  On a 32 bits
x86 system, we should base this on the number of lowmem pages.

Signed-off-by: Miquel van Smoorenburg <miquels@cistron.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agohamradio: remove unused variable
Andre Haupt [Sat, 28 Jun 2008 00:22:08 +0000 (17:22 -0700)]
hamradio: remove unused variable

Signed-off-by: Andre Haupt <andre@bitwigglers.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agomac80211: fix an oops in several failure paths in key allocation
Emmanuel Grumbach [Thu, 26 Jun 2008 09:13:46 +0000 (12:13 +0300)]
mac80211: fix an oops in several failure paths in key allocation

This patch fixes an oops in several failure paths in key allocation. This
Oops occurs when freeing a key that has not been linked yet, so the
key->sdata is not set.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agoprism: islpci_eth.c endianness fix
Harvey Harrison [Wed, 25 Jun 2008 21:20:37 +0000 (14:20 -0700)]
prism: islpci_eth.c endianness fix

clock is already cpu-endian (see le32_to_cpu slightly before), so
le64_to_cpu doesn't make much sense.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agort2x00: Fix lock dependency errror
Ivo van Doorn [Wed, 25 Jun 2008 19:27:00 +0000 (21:27 +0200)]
rt2x00: Fix lock dependency errror

This fixes a circular locking dependency in the workqueue handling.
The interface work task uses the mac80211 function
ieee80211_iterate_active_interfaces() which grabs the RTNL lock.

However when the interface is brough down, this happens under the RTNL
lock as well, this causes problems because mac80211 will flush the workqueue
during the ifdown event. This causes mac80211 to wait until the driver has
completed all work which can't finish because it is waiting on the RTNL lock.

This is fixed by moving rt2x00 workqueue tasks on a different workqueue,
this workqueue can be flushed when the ieee80211_hw structure is removed
by the driver (when the driver is unloaded) which does not happen under the
RTNL lock.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agowireless: remove RFKILL_STATE_HARD_BLOCKED warnings
John W. Linville [Fri, 27 Jun 2008 14:27:47 +0000 (10:27 -0400)]
wireless: remove RFKILL_STATE_HARD_BLOCKED warnings

  CC [M]  drivers/net/wireless/b43/rfkill.o
drivers/net/wireless/b43/rfkill.c: In function ‘b43_rfkill_soft_toggle’:
drivers/net/wireless/b43/rfkill.c:90: warning: enumeration value ‘RFKILL_STATE_HARD_BLOCKED’ not handled in switch

  CC [M]  drivers/net/wireless/b43legacy/rfkill.o
drivers/net/wireless/b43legacy/rfkill.c: In function ‘b43legacy_rfkill_soft_toggle’:
drivers/net/wireless/b43legacy/rfkill.c:92: warning: enumeration value ‘RFKILL_STATE_HARD_BLOCKED’ not handled in switch

  CC [M]  drivers/net/wireless/iwlwifi/iwl-rfkill.o
drivers/net/wireless/iwlwifi/iwl-rfkill.c: In function ‘iwl_rfkill_soft_rf_kill’:
drivers/net/wireless/iwlwifi/iwl-rfkill.c:56: warning: enumeration value ‘RFKILL_STATE_HARD_BLOCKED’ not handled in switch

Also handle RFKILL_STATE_{ON,OFF} -> RFKILL_STATE_{UNBLOCKED,SOFT_BLOCKED}
conversion since I'm already here...

Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agoath5k: remove now unused variable declared in ath5k_tx
John W. Linville [Fri, 27 Jun 2008 13:31:29 +0000 (09:31 -0400)]
ath5k: remove now unused variable declared in ath5k_tx

  CC [M]  drivers/net/wireless/ath5k/base.o
drivers/net/wireless/ath5k/base.c: In function ‘ath5k_tx’:
drivers/net/wireless/ath5k/base.c:2598: warning: unused variable ‘info’

Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: fix tx fragmentation
Johannes Berg [Wed, 25 Jun 2008 11:36:27 +0000 (14:36 +0300)]
mac80211: fix tx fragmentation

This patch fixes TX fragmentation caused by
tx handlers reordering and 'tx info to cb' patches

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: make workqueue freezable
Johannes Berg [Thu, 26 Jun 2008 17:59:56 +0000 (19:59 +0200)]
mac80211: make workqueue freezable

This patch makes the mac80211 workqueue freezable making it
interact a bit better with system suspend and not try to ping
the AP while the hardware is down.

This doesn't really help with implementing proper suspend in
any way but makes some bad things trigger less.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agowireless: Small cleanups
Pavel Machek [Wed, 25 Jun 2008 10:25:53 +0000 (12:25 +0200)]
wireless: Small cleanups

Small whitespace cleanups for wireless drivers

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agoiwlwifi: fix build for CONFIG_INPUT=n
Randy Dunlap [Wed, 25 Jun 2008 04:02:46 +0000 (21:02 -0700)]
iwlwifi: fix build for CONFIG_INPUT=n

Fix iwlwifi so that it builds cleanly with CONFIG_INPUT=n.
Also free the input device on exit.

drivers/built-in.o: In function `iwl_rfkill_unregister':
(.text+0xbf430): undefined reference to `input_unregister_device'
drivers/built-in.o: In function `iwl_rfkill_init':
(.text+0xbf51c): undefined reference to `input_allocate_device'
drivers/built-in.o: In function `iwl_rfkill_init':
(.text+0xbf5bf): undefined reference to `input_register_device'
drivers/built-in.o: In function `iwl_rfkill_init':
(.text+0xbf5e9): undefined reference to `input_free_device'
net/built-in.o: In function `rfkill_disconnect':
rfkill-input.c:(.text+0xe71e1): undefined reference to `input_close_device'
rfkill-input.c:(.text+0xe71e9): undefined reference to `input_unregister_handle'
net/built-in.o: In function `rfkill_connect':
rfkill-input.c:(.text+0xe723e): undefined reference to `input_register_handle'
rfkill-input.c:(.text+0xe724d): undefined reference to `input_open_device'
rfkill-input.c:(.text+0xe725c): undefined reference to `input_unregister_handle'
net/built-in.o: In function `rfkill_handler_init':
rfkill-input.c:(.init.text+0x36ec): undefined reference to `input_register_handler'
net/built-in.o: In function `rfkill_handler_exit':
rfkill-input.c:(.exit.text+0x112c): undefined reference to `input_unregister_handler'
make[1]: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: Add RTNL warning for workqueue
Ivo van Doorn [Tue, 24 Jun 2008 17:23:36 +0000 (19:23 +0200)]
mac80211: Add RTNL warning for workqueue

The workqueue provided by mac80211 should not be used for
scheduled tasks that acquire the RTNL lock. This could be done
when the driver uses the function ieee80211_iterate_active_interfaces()
within the scheduled work. Such behavior will end in locking
dependencies problems when an interface is being removed.

This patch will add a notification about the RTNL locking and
the mac80211 workqueue to prevent driver developers from
blindly using it.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: add phy information to giwname
Tomas Winkler [Tue, 24 Jun 2008 12:50:17 +0000 (15:50 +0300)]
mac80211: add phy information to giwname

This patch add phy information to giwname.

Quoting:
It's not useless, it's supposed to tell you about the protocol
capability of the device, like "IEEE 802.11b" or "IEEE 802.11abg"

Jean

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: update the authentication method
Emmanuel Grumbach [Tue, 24 Jun 2008 10:37:59 +0000 (13:37 +0300)]
mac80211: update the authentication method

This patch updates the authentication method upon giwencode ioctl.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: don't return -EINVAL upon iwconfig wlan0 rts auto
Emmanuel Grumbach [Tue, 24 Jun 2008 10:37:58 +0000 (13:37 +0300)]
mac80211: don't return -EINVAL upon iwconfig wlan0 rts auto

This patch avoids returning -EINVAL upon iwconfig wlan0 rts auto. If
rts->fixed is 0, then we should choose a default value instead of failing.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agob43: Fix PIO skb clobber
Michael Buesch [Tue, 24 Jun 2008 10:22:05 +0000 (12:22 +0200)]
b43: Fix PIO skb clobber

This fixes a clobber of the skb that was introduced by the
tx_control->cb conversion patches.
This bug causes a crash when the skb destructor is invoked. That happens
on skb_orphan or skb_kfree.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agort2x00: kill URB for all TX queues during disable_radio()
Ivo van Doorn [Mon, 23 Jun 2008 17:56:50 +0000 (19:56 +0200)]
rt2x00: kill URB for all TX queues during disable_radio()

During rt2x00usb_disable_radio() all pending urb's should
be killed and not only those from the RX queue.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: mlme.c use new frame control helpers
Harvey Harrison [Sun, 22 Jun 2008 23:45:32 +0000 (16:45 -0700)]
mac80211: mlme.c use new frame control helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: rx.c use new frame control helpers
Harvey Harrison [Sun, 22 Jun 2008 23:45:29 +0000 (16:45 -0700)]
mac80211: rx.c use new frame control helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
16 years agomac80211: tx.c use new frame control helpers
Harvey Harrison [Sun, 22 Jun 2008 23:45:27 +0000 (16:45 -0700)]
mac80211: tx.c use new frame control helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>