openwrt/staging/blogic.git
13 years agoipv4: Kill RT_CACHE_DEBUG
David S. Miller [Wed, 18 May 2011 22:23:21 +0000 (18:23 -0400)]
ipv4: Kill RT_CACHE_DEBUG

It's way past it's usefulness.  And this gets rid of a bunch
of stray ->rt_{dst,src} references.

Even the comment documenting the macro was inaccurate (stated
default was 1 when it's 0).

If reintroduced, it should be done properly, with dynamic debug
facilities.

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobe2net: Enable SR-IOV for Lancer
Mammatha Edhala [Wed, 18 May 2011 03:26:22 +0000 (03:26 +0000)]
be2net: Enable SR-IOV for Lancer

Enable SR-IOV for Lancer

Signed-off-by: Mammatha Edhala <mammatha.edhala@emulex.com>
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobnx2x: add support for retrieving dcb peer configuration
Shmulik Ravid [Wed, 18 May 2011 02:55:31 +0000 (02:55 +0000)]
bnx2x: add support for retrieving dcb peer configuration

This patch adds support to the bnx2x for retrieving dcb peer (remote)
configuration from the embedded DCBX stack.

Signed-off-by: Shmulik Ravid <shmulikr@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoqeth: use ndo_set_features callback for initial setup and recovery
Frank Blaschka [Wed, 18 May 2011 03:28:35 +0000 (03:28 +0000)]
qeth: use ndo_set_features callback for initial setup and recovery

This patch uses the ndo_set_features callback during normal device
startup or recovery to turn on hardware RX checksum. Patch was done
with much help from Michal Miroslaw, thx!!!

Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Reviewed-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agowl12xx: fix compilation error when CONFIG_PM is not set
Luciano Coelho [Wed, 18 May 2011 20:51:26 +0000 (16:51 -0400)]
wl12xx: fix compilation error when CONFIG_PM is not set

There was a compilation error when PM is not enabled:

  CC [M]  drivers/net/wireless/wl12xx/main.o
drivers/net/wireless/wl12xx/main.c:3653: error: unknown field 'suspend' specified in initializer
drivers/net/wireless/wl12xx/main.c:3653: warning: initialization from incompatible pointer type
drivers/net/wireless/wl12xx/main.c:3654: error: unknown field 'resume' specified in initializer
drivers/net/wireless/wl12xx/main.c:3654: warning: initialization from incompatible pointer type

Fix this by adding #ifdef's in the appropriate places.

Cc: Eliad Peller <eliad@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agosfc: Don't use enums as a bitmask.
David S. Miller [Tue, 17 May 2011 21:53:22 +0000 (17:53 -0400)]
sfc: Don't use enums as a bitmask.

This fixes:

drivers/net/sfc/mcdi_mac.c: In function ‘efx_mcdi_set_mac’:
drivers/net/sfc/mcdi_mac.c:36:2: warning: case value ‘3’ not in enumerated type ‘enum efx_fc_type’

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
David S. Miller [Tue, 17 May 2011 21:33:11 +0000 (17:33 -0400)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6

Conflicts:
drivers/net/vmxnet3/vmxnet3_ethtool.c
net/core/dev.c

13 years agoipv4: Don't use enums as bitmasks in ip_fragment.c
David S. Miller [Tue, 17 May 2011 21:28:02 +0000 (17:28 -0400)]
ipv4: Don't use enums as bitmasks in ip_fragment.c

Noticed by Joe Perches.

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: ethtool: fix IPV6 checksum feature name string
Michał Mirosław [Tue, 17 May 2011 20:50:02 +0000 (16:50 -0400)]
net: ethtool: fix IPV6 checksum feature name string

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: Change netdev_fix_features messages loglevel
Michael S. Tsirkin [Mon, 16 May 2011 10:37:39 +0000 (10:37 +0000)]
net: Change netdev_fix_features messages loglevel

Cool, how about we make 'Features changed' debug as well?
This way userspace can't fill up the log just by tweaking tun features
with an ioctl.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: recvmmsg: Strip MSG_WAITFORONE when calling recvmsg
Anton Blanchard [Tue, 17 May 2011 19:38:57 +0000 (15:38 -0400)]
net: recvmmsg: Strip MSG_WAITFORONE when calling recvmsg

recvmmsg fails on a raw socket with EINVAL. The reason for this is
packet_recvmsg checks the incoming flags:

        err = -EINVAL;
        if (flags & ~(MSG_PEEK|MSG_DONTWAIT|MSG_TRUNC|MSG_CMSG_COMPAT|MSG_ERRQUEUE))
                goto out;

This patch strips out MSG_WAITFORONE when calling recvmmsg which
fixes the issue.

Signed-off-by: Anton Blanchard <anton@samba.org>
Cc: stable@kernel.org [2.6.34+]
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoxen: netback: use __CONST_RING_SIZE not __RING_SIZE
Ian Campbell [Mon, 16 May 2011 23:59:48 +0000 (23:59 +0000)]
xen: netback: use __CONST_RING_SIZE not __RING_SIZE

The later causes warnings with gcc 4.5+. __CONST_RING_SIZE was introduced in
667c78afaec0 to fix this but as netback wasn't upstream at the time it did not
benefit, hence:

  CC      drivers/net/xen-netback/netback.o
drivers/net/xen-netback/netback.c:110:37: warning: variably modified 'grant_copy_op' at file scope [enabled by default]
drivers/net/xen-netback/netback.c:111:30: warning: variably modified 'meta' at file scope [enabled by default]
drivers/net/xen-netback/netback.c: In function 'xen_netbk_rx_action':
drivers/net/xen-netback/netback.c:584:6: warning: variable 'irq' set but not used [-Wunused-but-set-variable]

Thanks to Witold Baryluk for pointing this out.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Witold Baryluk <baryluk@smp.if.uj.edu.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/kaber/nf-2.6
David S. Miller [Tue, 17 May 2011 18:18:25 +0000 (14:18 -0400)]
Merge branch 'master' of git://git./linux/kernel/git/kaber/nf-2.6

13 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/bwh/sfc-next-2.6
David S. Miller [Tue, 17 May 2011 18:17:05 +0000 (14:17 -0400)]
Merge branch 'master' of git://git./linux/kernel/git/bwh/sfc-next-2.6

13 years agonet: ping: fix build failure
Vasiliy Kulikov [Tue, 17 May 2011 00:16:56 +0000 (00:16 +0000)]
net: ping: fix build failure

If CONFIG_PROC_SYSCTL=n the building process fails:

    ping.c:(.text+0x52af3): undefined reference to `inet_get_ping_group_range_net'

Moved inet_get_ping_group_range_net() to ping.c.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: use hlist_del_rcu() in dev_change_name()
Eric Dumazet [Tue, 17 May 2011 17:56:59 +0000 (13:56 -0400)]
net: use hlist_del_rcu() in dev_change_name()

Using plain hlist_del() in dev_change_name() is wrong since a
concurrent reader can crash trying to dereference LIST_POISON1.

Bug introduced in commit 72c9528bab94 (net: Introduce
dev_get_by_name_rcu())

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobluetooth: Fix warnings in l2cap_core.c
David S. Miller [Tue, 17 May 2011 03:09:26 +0000 (23:09 -0400)]
bluetooth: Fix warnings in l2cap_core.c

net/bluetooth/l2cap_core.c: In function ‘l2cap_recv_frame’:
net/bluetooth/l2cap_core.c:3758:15: warning: ‘sk’ may be used uninitialized in this function
net/bluetooth/l2cap_core.c:3758:15: note: ‘sk’ was declared here
net/bluetooth/l2cap_core.c:3791:15: warning: ‘sk’ may be used uninitialized in this function
net/bluetooth/l2cap_core.c:3791:15: note: ‘sk’ was declared here

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/linville...
David S. Miller [Tue, 17 May 2011 02:49:41 +0000 (22:49 -0400)]
Merge branch 'for-davem' of git://git./linux/kernel/git/linville/wireless-next-2.6

13 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
John W. Linville [Mon, 16 May 2011 18:55:42 +0000 (14:55 -0400)]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-next-2.6 into for-davem

Conflicts:
drivers/net/wireless/iwlwifi/iwl-agn-tx.c
net/mac80211/sta_info.h

13 years agosfc: Use netif_device_{detach,attach}() around reset and self-test
Ben Hutchings [Mon, 16 May 2011 17:51:24 +0000 (18:51 +0100)]
sfc: Use netif_device_{detach,attach}() around reset and self-test

We need to keep the TX queues stopped throughout a reset, without
triggering the TX watchdog and regardless of the link state.  The
proper way to do this is to use netif_device_{detach,attach}() just as
we do around suspend/resume, rather than the current bodge of faking
link-down.

Since we also need to do this during an offline self-test and we
perform a reset during that, add these function calls outside of
efx_reset_down() and efx_reset_up().

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
13 years agonet: Change netdev_fix_features messages loglevel
Michał Mirosław [Mon, 16 May 2011 19:14:21 +0000 (15:14 -0400)]
net: Change netdev_fix_features messages loglevel

Those reduced to DEBUG can possibly be triggered by unprivileged processes
and are nothing exceptional. Illegal checksum combinations can only be
caused by driver bug, so promote those messages to WARN.

Since GSO without SG will now only cause DEBUG message from
netdev_fix_features(), remove the workaround from register_netdevice().

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agovmxnet3: Fix inconsistent LRO state after initialization
Thomas Jarosch [Mon, 16 May 2011 06:28:15 +0000 (06:28 +0000)]
vmxnet3: Fix inconsistent LRO state after initialization

During initialization of vmxnet3, the state of LRO
gets out of sync with netdev->features.

This leads to very poor TCP performance in a IP forwarding
setup and is hitting many VMware users.

Simplified call sequence:
1. vmxnet3_declare_features() initializes "adapter->lro" to true.

2. The kernel automatically disables LRO if IP forwarding is enabled,
so vmxnet3_set_flags() gets called. This also updates netdev->features.

3. Now vmxnet3_setup_driver_shared() is called. "adapter->lro" is still
set to true and LRO gets enabled again, even though
netdev->features shows it's disabled.

Fix it by updating "adapter->lro", too.

The private vmxnet3 adapter flags are scheduled for removal
in net-next, see commit a0d2730c9571aeba793cb5d3009094ee1d8fda35
"net: vmxnet3: convert to hw_features".

Patch applies to 2.6.37 / 2.6.38 and 2.6.39-rc6.

Please CC: comments.

Signed-off-by: Thomas Jarosch <thomas.jarosch@intra2net.com>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agosfc: Fix oops in register dump after mapping change
Ben Hutchings [Mon, 16 May 2011 06:13:49 +0000 (06:13 +0000)]
sfc: Fix oops in register dump after mapping change

Commit 747df2258b1b9a2e25929ef496262c339c380009 ('sfc: Always map MCDI
shared memory as uncacheable') introduced a separate mapping for the
MCDI shared memory (MC_TREG_SMEM).  This means we can no longer easily
include it in the register dump.  Since it is not particularly useful
in debugging, substitute a recognisable dummy value.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: convert to new cpumask API
KOSAKI Motohiro [Mon, 16 May 2011 18:53:49 +0000 (11:53 -0700)]
net: convert to new cpumask API

We plan to remove cpu_xx() old api later. Thus this patch
convert it.

This patch has no functional change.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoipv4: more compliant RFC 3168 support
Eric Dumazet [Mon, 16 May 2011 08:37:37 +0000 (08:37 +0000)]
ipv4: more compliant RFC 3168 support

Commit 6623e3b24a5e (ipv4: IP defragmentation must be ECN aware) was an
attempt to not lose "Congestion Experienced" (CE) indications when
performing datagram defragmentation.

Stefanos Harhalakis raised the point that RFC 3168 requirements were not
completely met by this commit.

In particular, we MUST detect invalid combinations and eventually drop
illegal frames.

Reported-by: Stefanos Harhalakis <v13@v13.gr>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobe2net: FW download for Lancer
Shripad Nunjundarao [Mon, 16 May 2011 07:36:59 +0000 (07:36 +0000)]
be2net: FW download for Lancer

Added implementation of FW download feature for Lancer.

Signed-off-by: Shripad Nunjundarao <shripad.nunjundarao@emulex.com>
Signed-off-by: Sevin Xavier <selvin.xavier@emulex.com>
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobe2net: Stats for Lancer
Selvin Xavier [Mon, 16 May 2011 07:36:35 +0000 (07:36 +0000)]
be2net: Stats for Lancer

Added Lancer stats implementation.

Signed-off-by: Selvin Xavier <selvin.xavier@emulex.com>
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobe2net: Support for version 1 of stats for BE3
Ajit Khaparde [Mon, 16 May 2011 07:36:18 +0000 (07:36 +0000)]
be2net: Support for version 1 of stats for BE3

Added support to get version 1 of the stats for BE3.
Use old stats command for BE2.

Signed-off-by: Ajit Khaparde <ajit.khaparde@emulex.com>
Signed-off-by: Selvin Xavier <selvin.xavier@emulex.com>
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agomwifiex: use ieee80211_amsdu_to_8023s routine
Yogesh Ashok Powar [Fri, 13 May 2011 18:22:32 +0000 (11:22 -0700)]
mwifiex: use ieee80211_amsdu_to_8023s routine

mwifiex was using its own implementation of converting 802.11n
AMSDU to 802.3s.  This patch removes mwifiex specific
implementation and uses existing ieee80211_amsdu_to_8023s
routine.

Signed-off-by: Yogesh Ashok Powar <yogeshp@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agocfg80211: make stripping of 802.11 header optional from AMSDU
Yogesh Ashok Powar [Fri, 13 May 2011 18:22:31 +0000 (11:22 -0700)]
cfg80211: make stripping of 802.11 header optional from AMSDU

Currently the devices that have already stripped IEEE 802.11
header from the AMSDU SKB can not use ieee80211_amsdu_to_8023s
routine. This patch enhances ieee80211_amsdu_to_8023s() API by
changing mandatory removing of IEEE 802.11 header from AMSDU
to optional.

Signed-off-by: Yogesh Ashok Powar <yogeshp@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agonl80211: Move peer link state definition to nl80211
Javier Cardona [Fri, 13 May 2011 17:45:43 +0000 (10:45 -0700)]
nl80211: Move peer link state definition to nl80211

These definitions need to be exposed now that we can set the peer link
states via NL80211_ATTR_STA_PLINK_STATE.  They were already being
(opaquely) reported by NL80211_STA_INFO_PLINK_STATE.

Signed-off-by: Javier Cardona <javier@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k: make npending frames check as bool
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:29:42 +0000 (20:59 +0530)]
ath9k: make npending frames check as bool

we are not doing anything by tracking the number of pending frames.
bail out when we first find a pending frame in any one of the 10 queues.

Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k: make sure main_rssi is positive
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:01:40 +0000 (20:31 +0530)]
ath9k: make sure main_rssi is positive

some times the rssi control descriptor for the main antenna may be
negative like that of alternate antenna, hence before  incrementing
packet counts/rssi of main/alternate antenna make sure both main_rssi
and alt_rssi are positive only. this avoids wrong selection of antenna
due to diversity

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k: configure fast_div_bias based on diversity group
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:01:23 +0000 (20:31 +0530)]
ath9k: configure fast_div_bias based on diversity group

configure fast diversity bias based on the antenna diversity group and
based on main/alt LNA configurations. also configure main antenna and
alternate antenna to gain-table 0 for diversity group 2(AR9485)

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k: Implement an API to swap main/ALT LNA's
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:01:09 +0000 (20:31 +0530)]
ath9k: Implement an API to swap main/ALT LNA's

for the diversity group 2(AR9485) we swap the LNA's of main/ALT antenna
based on alternate antenna's rssi average in comparision with main
antenna's rssi, while for AR9285(antenna diversity group 0)we still
follow the older method of looking at the packet count in alternate
antenna

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: define antenna diversity group
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:00:56 +0000 (20:30 +0530)]
ath9k_hw: define antenna diversity group

AR9285 belongs to diversity group 0 and AR9485 belongs to diversity
group 2. Based on the diversity group we configure certain antenna
diversity paramaters such as lna1_lna2_delta and fast diversity
bias values. For AR9485 we have some gain table parameter which
selects the gain table 0/1 for main and alternate antenna

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: define modules to get/set Antenna diversity paramaters
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:00:41 +0000 (20:30 +0530)]
ath9k_hw: define modules to get/set Antenna diversity paramaters

these are the two important modules that will be called by the antenna
diversity algorithm module in the rx. this will continuosly configure
the hardware based on the current diversity status obtained
from the algorithm

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: config diversity based on eeprom contents
Mohammed Shafi Shajakhan [Fri, 13 May 2011 15:00:27 +0000 (20:30 +0530)]
ath9k_hw: config diversity based on eeprom contents

* enable LNA-diversity, fast-diversity for AR9485 based  on
  the value read from EEPROM content
* if antenna diversity/combining is supported, set LNA1 for the main
  antenna and LNA2 for the alternate antenna based on the new diversity
  algorithm

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Cc: Luis Rodriguez <Luis.Rodriguez@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: define registers/macros to support Antenna diversity
Mohammed Shafi Shajakhan [Fri, 13 May 2011 14:59:53 +0000 (20:29 +0530)]
ath9k_hw: define registers/macros to support Antenna diversity

define few registers and macros to configure/enable Antenna diversity
parameters in AR9485

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: enable Antenna diversity for AR9485
Mohammed Shafi Shajakhan [Fri, 13 May 2011 14:59:31 +0000 (20:29 +0530)]
ath9k_hw: enable Antenna diversity for AR9485

read antenna diversity and combining information from the EEPROM.
Enable antenna diversity/combining feature only when both LNA
diversity and fast diversity are supported

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Tested-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoath9k_hw: make antenna diversity modules chip specific
Mohammed Shafi Shajakhan [Fri, 13 May 2011 14:59:04 +0000 (20:29 +0530)]
ath9k_hw: make antenna diversity modules chip specific

this is necessary to support Antenna diversity and combining in new chip
sets such as AR9485, previously Antenna diversity support is available
only in AR9285

Cc: Gabriel Tseng <Gabriel.Tseng@Atheros.com>
Cc: Senthilkumar Balasubramanian <Senthilkumar.Balasubramanian@Atheros.com>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agonet/rfkill/core.c: Avoid leaving freed data in a list
Julia Lawall [Fri, 13 May 2011 13:52:10 +0000 (15:52 +0200)]
net/rfkill/core.c: Avoid leaving freed data in a list

The list_for_each_entry loop can fail, in which case the list element is
not removed from the list rfkill_fds.  Since this list is not accessed by
the loop, the addition of &data->list into the list is just moved after the
loop.

The sematic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression E,E1,E2;
identifier l;
@@

*list_add(&E->l,E1);
... when != E1
    when != list_del(&E->l)
    when != list_del_init(&E->l)
    when != E = E2
*kfree(E);// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agomac80211: sparse RCU annotations
Johannes Berg [Fri, 13 May 2011 12:15:49 +0000 (14:15 +0200)]
mac80211: sparse RCU annotations

This adds sparse RCU annotations to most of
mac80211, only the mesh code remains to be
done.

Due the the previous patches, the annotations
are pretty simple. The only thing that this
actually changes is removing the RCU usage of
key->sta in debugfs since this pointer isn't
actually an RCU-managed pointer (it only has
a single assignment done before the key even
goes live). As that is otherwise harmless, I
decided to make it part of this patch.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agomac80211: fix TX a-MPDU locking
Johannes Berg [Fri, 13 May 2011 11:35:40 +0000 (13:35 +0200)]
mac80211: fix TX a-MPDU locking

During my quest to make mac80211 not have any RCU
warnings from sparse, I came across the a-MPDU code
again and it wasn't quite clear why it isn't racy.
So instead of assigning the tid_tx array with just
the spinlock held in ieee80211_start_tx_ba_session
use a separate temporary array protected only by
the spinlock and protect all assignments to the
"live" array by both the spinlock and the mutex so
that other code is easily verified to be correct.

Due to pointer assignment atomicity I don't think
this is a real issue, but I'm not sure, especially
on Alpha the current code might be problematic.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agocfg80211: advertise possible interface combinations
Johannes Berg [Fri, 13 May 2011 08:58:57 +0000 (10:58 +0200)]
cfg80211: advertise possible interface combinations

Add the ability to advertise interface combinations in nl80211.
This allows the driver to indicate what the combinations are
that it supports. "Combinations" of just a single interface are
implicit, as previously. Note that cfg80211 will enforce that
the restrictions are met, but not for all drivers yet (once all
drivers are updated, we can remove the flag and enforce for all).

When no combinations are actually supported, an empty list will
be exported so that userspace can know if the kernel exported
this info or not (although it isn't clear to me what tools using
the info should do if the kernel didn't export it).

Since some interface types are purely virtual/software and don't
fit the restrictions, those are exposed in a new list of pure SW
types, not subject to restrictions. This mainly exists to handle
AP-VLAN and monitor interfaces in mac80211.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoiwlwifi: Silence DEBUG_STRICT_USER_COPY_CHECKS=y warning
Stephen Boyd [Thu, 12 May 2011 23:50:05 +0000 (16:50 -0700)]
iwlwifi: Silence DEBUG_STRICT_USER_COPY_CHECKS=y warning

Enabling DEBUG_STRICT_USER_COPY_CHECKS causes the following
warning:

In file included from arch/x86/include/asm/uaccess.h:573,
                 from include/net/checksum.h:25,
                 from include/linux/skbuff.h:28,
                 from drivers/net/wireless/iwlwifi/iwl-agn-rs.c:28:
In function 'copy_from_user',
    inlined from 'rs_sta_dbgfs_scale_table_write' at
    drivers/net/wireless/iwlwifi/iwl-agn-rs.c:3099:
arch/x86/include/asm/uaccess_64.h:65:
warning: call to 'copy_from_user_overflow' declared with
attribute warning: copy_from_user() buffer size is not provably
correct

presumably due to buf_size being signed causing GCC to fail to
see that buf_size can't become negative.

Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoiwlegacy: Silence DEBUG_STRICT_USER_COPY_CHECKS=y warning
Stephen Boyd [Thu, 12 May 2011 23:50:04 +0000 (16:50 -0700)]
iwlegacy: Silence DEBUG_STRICT_USER_COPY_CHECKS=y warning

Enabling DEBUG_STRICT_USER_COPY_CHECKS causes the following
warning:

In file included from arch/x86/include/asm/uaccess.h:573,
                 from include/net/checksum.h:25,
                 from include/linux/skbuff.h:28,
                 from drivers/net/wireless/iwlegacy/iwl-4965-rs.c:28:
In function 'copy_from_user',
    inlined from 'iwl4965_rs_sta_dbgfs_scale_table_write' at
    drivers/net/wireless/iwlegacy/iwl-4965-rs.c:2616:
arch/x86/include/asm/uaccess_64.h:65:
warning: call to 'copy_from_user_overflow' declared with
attribute warning: copy_from_user() buffer size is not provably
correct

presumably due to buf_size being signed causing GCC to fail to
see that buf_size can't become negative.

Cc: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agobnx2x: Allow ethtool to enable/disable loopback.
Mahesh Bandewar [Fri, 13 May 2011 15:08:49 +0000 (15:08 +0000)]
bnx2x: Allow ethtool to enable/disable loopback.

This patch updates the bnx2x_set_features() to handle loopback mode.
When enabled; it sets internal-MAC loopback.

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoipv4: Trivial rt->rt_src conversions in net/ipv4/route.c
David S. Miller [Fri, 13 May 2011 22:01:21 +0000 (18:01 -0400)]
ipv4: Trivial rt->rt_src conversions in net/ipv4/route.c

At these points we have a fully filled in value via the IP
header the form of ip_hdr(skb)->saddr

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agosfc: Fix TX queue numbering when separate_tx_channels=1
Ben Hutchings [Mon, 16 May 2011 16:32:39 +0000 (17:32 +0100)]
sfc: Fix TX queue numbering when separate_tx_channels=1

This option appears to have been broken by commit
8313aca38b3937947fffebca6e34bac8e24300c8 ('sfc: Allocate each channel
separately, along with its RX and TX queues').

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
13 years agonet: ping: dont call udp_ioctl()
Eric Dumazet [Sun, 15 May 2011 21:26:31 +0000 (21:26 +0000)]
net: ping: dont call udp_ioctl()

udp_ioctl() really handles UDP and UDPLite protocols.

1) It can increment UDP_MIB_INERRORS in case first_packet_length() finds
a frame with bad checksum.

2) It has a dependency on sizeof(struct udphdr), not applicable to
ICMP/PING

If ping sockets need to handle SIOCINQ/SIOCOUTQ ioctl, this should be
done differently.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Vasiliy Kulikov <segoon@openwall.com>
Acked-by: Vasiliy Kulikov <segoon@openwall.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonetfilter: nf_ct_sip: fix SDP parsing in TCP SIP messages for some Cisco phones
Patrick McHardy [Mon, 16 May 2011 12:45:39 +0000 (14:45 +0200)]
netfilter: nf_ct_sip: fix SDP parsing in TCP SIP messages for some Cisco phones

Some Cisco phones do not place the Content-Length field at the end of the
SIP message. This is valid, due to a misunderstanding of the specification
the parser expects the SDP body to start directly after the Content-Length
field. Fix the parser to scan for \r\n\r\n to locate the beginning of the
SDP body.

Reported-by: Teresa Kang <teresa_kang@gemtek.com.tw>
Signed-off-by: Patrick McHardy <kaber@trash.net>
13 years agonetfilter: nf_ct_sip: validate Content-Length in TCP SIP messages
Patrick McHardy [Mon, 16 May 2011 12:42:26 +0000 (14:42 +0200)]
netfilter: nf_ct_sip: validate Content-Length in TCP SIP messages

Verify that the message length of a single SIP message, which is calculated
based on the Content-Length field contained in the SIP message, does not
exceed the packet boundaries.

Signed-off-by: Patrick McHardy <kaber@trash.net>
13 years agoMerge branch 'pablo/nf-2.6-updates' of git://1984.lsi.us.es/net-2.6
David S. Miller [Sun, 15 May 2011 22:14:02 +0000 (18:14 -0400)]
Merge branch 'pablo/nf-2.6-updates' of git://1984.lsi.us.es/net-2.6

13 years agonet: drivers: kill two unused macro definitions
Shan Wei [Fri, 13 May 2011 21:08:47 +0000 (21:08 +0000)]
net: drivers: kill two unused macro definitions

Signed-off-by: Shan Wei <shanwei@cn.fujitsu.com>
Reviewed-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: remove unesesarry exports
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:08 +0000 (02:44 +0000)]
caif: remove unesesarry exports

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Bugfix debugfs directory name must be unique.
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:07 +0000 (02:44 +0000)]
caif: Bugfix debugfs directory name must be unique.

Race condition caused debugfs_create_dir() to fail due to duplicate
name. Use atomic counter to create unique directory name.

net_ratelimit() is introduced to limit debug printouts.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Handle dev_queue_xmit errors.
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:06 +0000 (02:44 +0000)]
caif: Handle dev_queue_xmit errors.

Do proper handling of dev_queue_xmit errors in order to
avoid double free of skb and leaks in error conditions.
In cfctrl pending requests are removed when CAIF Link layer goes down.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: prepare support for namespaces
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:05 +0000 (02:44 +0000)]
caif: prepare support for namespaces

Use struct net to reference CAIF configuration object instead of static variables.
Refactor functions caif_connect_client, caif_disconnect_client and squach
files cfcnfg.c and caif_config_utils.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Protected in-flight packets using dev or sock refcont.
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:04 +0000 (02:44 +0000)]
caif: Protected in-flight packets using dev or sock refcont.

CAIF Socket Layer and ip-interface registers reference counters
in CAIF service layer. The functions sock_hold, sock_put and
dev_hold, dev_put are used by CAIF Stack to protect from freeing
memory while packets are in-flight.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Move refcount from service layer to sock and dev.
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:03 +0000 (02:44 +0000)]
caif: Move refcount from service layer to sock and dev.

Instead of having reference counts in caif service layers,
we hook into existing refcount handling in socket layer and netdevice.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Add ref-count to framing layer
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:02 +0000 (02:44 +0000)]
caif: Add ref-count to framing layer

Introduce Per-cpu reference for lower part of CAIF Stack.
Before freeing payload is disabled, synchronize_rcu() is called,
and then ref-count verified to be zero.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Use RCU and lists in cfcnfg.c for managing caif link layers
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:01 +0000 (02:44 +0000)]
caif: Use RCU and lists in cfcnfg.c for managing caif link layers

RCU lists are used for handling the link layers instead of array.
When generating CAIF phy-id, ifindex is used as base. Legal range is 1-6.
Introduced set_phy_state() for managing CAIF Link layer state.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Use RCU instead of spin-lock in caif_dev.c
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:44:00 +0000 (02:44 +0000)]
caif: Use RCU instead of spin-lock in caif_dev.c

RCU read_lock and refcount is used to protect in-flight packets.

Use RCU and counters to manage freeing lower part of the CAIF stack if
CAIF-link layer is removed. Old solution based on delaying removal of
device is removed.

When CAIF link layer goes down the use of CAIF link layer is disabled
(by calling caif_set_phy_state()), but removal and freeing of the
lower part of the CAIF stack is done when Link layer is unregistered.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agocaif: Use rcu_read_lock in CAIF mux layer.
sjur.brandeland@stericsson.com [Fri, 13 May 2011 02:43:59 +0000 (02:43 +0000)]
caif: Use rcu_read_lock in CAIF mux layer.

Replace spin_lock with rcu_read_lock when accessing lists to layers
and cache. While packets are in flight rcu_read_lock should not be held,
instead ref-counters are used in combination with RCU.

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoIPVS: fix netns if reading ip_vs_* procfs entries
Hans Schillstrom [Sun, 15 May 2011 15:20:29 +0000 (17:20 +0200)]
IPVS: fix netns if reading ip_vs_* procfs entries

Without this patch every access to ip_vs in procfs will increase
the netns count i.e. an unbalanced get_net()/put_net().
(ipvsadm commands also use procfs.)
The result is you can't exit a netns if reading ip_vs_* procfs entries.

Signed-off-by: Hans Schillstrom <hans.schillstrom@ericsson.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
13 years agobridge: fix forwarding of IPv6
Stephen Hemminger [Fri, 13 May 2011 20:03:24 +0000 (16:03 -0400)]
bridge: fix forwarding of IPv6

The commit 6b1e960fdbd75dcd9bcc3ba5ff8898ff1ad30b6e
    bridge: Reset IPCB when entering IP stack on NF_FORWARD
broke forwarding of IPV6 packets in bridge because it would
call bp_parse_ip_options with an IPV6 packet.

Reported-by: Noah Meyerhans <noahm@debian.org>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
13 years agonet: ping: small changes
Eric Dumazet [Fri, 13 May 2011 22:59:19 +0000 (22:59 +0000)]
net: ping: small changes

ping_table is not __read_mostly, since it contains one rwlock,
and is static to ping.c

ping_port_rover & ping_v4_lookup are static

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Vasiliy Kulikov <segoon@openwall.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/jkirsher/net...
David S. Miller [Sun, 15 May 2011 05:08:23 +0000 (01:08 -0400)]
Merge branch 'master' of /linux/kernel/git/jkirsher/net-next-2.6

13 years agoMerge branch 'batman-adv/next' of git://git.open-mesh.org/ecsv/linux-merge
David S. Miller [Sun, 15 May 2011 02:47:51 +0000 (22:47 -0400)]
Merge branch 'batman-adv/next' of git://git.open-mesh.org/ecsv/linux-merge

13 years agoixgbe: Add support for new 82599 adapter
Don Skidmore [Sat, 14 May 2011 06:36:35 +0000 (06:36 +0000)]
ixgbe: Add support for new 82599 adapter

This patch adds support for a new adapter in the 82599 family.  Included
in that support is a new media_type ixgbe_media_type_fiber_lco.

Signed-of-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: fix sparse warning
Emil Tantilov [Sat, 7 May 2011 06:49:18 +0000 (06:49 +0000)]
ixgbe: fix sparse warning

error: bad constant expression

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Tested-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: cleanup some minor issues in ixgbe_down()
Alexander Duyck [Fri, 22 Apr 2011 04:08:14 +0000 (04:08 +0000)]
ixgbe: cleanup some minor issues in ixgbe_down()

This patch cleans up two minor issues in ixgbe_down.  Specifically it
addresses the fact that the VFs should not be pinged until after interrupts
are disabled otherwise they might still get a response.  It also drops the
use of the txdctl temporary variable since the only bit we should be
writing to the TXDCTL registers during a shutdown is the flush bit.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: Merge over-temp task into service task
Alexander Duyck [Fri, 22 Apr 2011 04:08:09 +0000 (04:08 +0000)]
ixgbe: Merge over-temp task into service task

This change merges the over-temp task into the service task.  As a result
all tasklets are finally combined into once single tasklet for easier
management.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: Merge ATR reinit into the service task
Alexander Duyck [Wed, 27 Apr 2011 09:25:34 +0000 (09:25 +0000)]
ixgbe: Merge ATR reinit into the service task

This change merges the ATR table reinitialization into the service task.
This is yet another opportunity to avoid any race conditions as we don't
want to be attempting to reinitialize the table during a possible reset.

In addition this change adds a counter for table reinitialization so that
it can be tracked as part of the regular statistics.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: merge reset task into service task
Alexander Duyck [Wed, 27 Apr 2011 09:21:16 +0000 (09:21 +0000)]
ixgbe: merge reset task into service task

This change is meant to further help to reduce possible configuration
collisions between the various tasklets.  This change combines the device
reset with the service task.  As a result it is now not possible to be
updating the link on the device while also resetting the part.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: Merge watchdog functionality into service task
Alexander Duyck [Fri, 22 Apr 2011 04:07:54 +0000 (04:07 +0000)]
ixgbe: Merge watchdog functionality into service task

This patch is meant to merge the functionality of the ixgbe watchdog task
into the service task.  By doing this all link state functionality will be
controlled by a single task.  As a result the reliability of the interface
will be improved as the likelihood of any race conditions is further
reduced.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: Combine SFP and multi-speed fiber task into single service task
Alexander Duyck [Wed, 27 Apr 2011 09:13:56 +0000 (09:13 +0000)]
ixgbe: Combine SFP and multi-speed fiber task into single service task

This change is meant to address several race conditions with multi-speed
fiber SFP+ modules in 82599 adapters.  Specifically issues have been seen
in which both the SFP configuration and the multi-speed fiber configuration
are running simultaneously which will result in the device getting into an
erroneous link down state.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: move flags and state into the same cacheline
Alexander Duyck [Fri, 22 Apr 2011 04:07:43 +0000 (04:07 +0000)]
ixgbe: move flags and state into the same cacheline

This change moves flags and state into the same cacheline.  The reason for
this change is because both are frequently read around the same time and
infrequently written.  By combining them into the same cacheline this
should help to reduce memory utilization.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: force unlock on timeout
Emil Tantilov [Fri, 8 Apr 2011 01:23:59 +0000 (01:23 +0000)]
ixgbe: force unlock on timeout

The semaphore can be in locked state upon driver load, particularly
on 82598 if a machine is rebooted due to panic and the semaphore was
acquired just prior to the panic.

This patch unlocks the semaphore if it times out.

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbe: Add macvlan support for VF
Greg Rose [Fri, 13 May 2011 01:33:48 +0000 (01:33 +0000)]
ixgbe: Add macvlan support for VF

Add infrastructure in the PF driver to support macvlan in the VF driver.

Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoixgbevf: Add macvlan support in the set rx mode op
Greg Rose [Fri, 13 May 2011 01:33:42 +0000 (01:33 +0000)]
ixgbevf: Add macvlan support in the set rx mode op

Implement setup of unicast address list in the VF driver's set_rx_mode
netdev op.  Unicast addresses are sent to the PF via a mailbox message
and the PF will check if it has room in the RAR table and if so set the
filter for the VF.

Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agoe1000e: minor comment cleanups
Bruce Allan [Fri, 13 May 2011 07:19:58 +0000 (07:19 +0000)]
e1000e: minor comment cleanups

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
13 years agobatman-adv: reset broadcast flood protection on error
Marek Lindner [Sat, 14 May 2011 18:01:22 +0000 (20:01 +0200)]
batman-adv: reset broadcast flood protection on error

The broadcast flood protection should be reset to its original value
if the primary interface could not be retrieved.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
13 years agobatman-adv: Add missing hardif_free_ref in forw_packet_free
Sven Eckelmann [Wed, 11 May 2011 18:59:06 +0000 (20:59 +0200)]
batman-adv: Add missing hardif_free_ref in forw_packet_free

add_bcast_packet_to_list increases the refcount for if_incoming but the
reference count is never decreased. The reference count must be
increased for all kinds of forwarded packets which have the primary
interface stored and forw_packet_free must decrease them. Also
purge_outstanding_packets has to invoke forw_packet_free when a work
item was really cancelled.

This regression was introduced in
32ae9b221e788413ce68feaae2ca39e406211a0a.

Reported-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
13 years agosfc: Fix return value from efx_ethtool_set_rx_ntuple()
Ben Hutchings [Sat, 14 May 2011 01:35:25 +0000 (02:35 +0100)]
sfc: Fix return value from efx_ethtool_set_rx_ntuple()

ethtool_ops::set_rx_ntuple is supposed to return 0 on success, but it
currently returns the filter ID when it inserts or modifies a filter.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
13 years agoipv4: Remove rt->rt_dst reference from ip_forward_options().
David S. Miller [Fri, 13 May 2011 21:31:02 +0000 (17:31 -0400)]
ipv4: Remove rt->rt_dst reference from ip_forward_options().

At this point iph->daddr equals what rt->rt_dst would hold.

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoipv4: Remove route key identity dependencies in ip_rt_get_source().
David S. Miller [Fri, 13 May 2011 21:29:41 +0000 (17:29 -0400)]
ipv4: Remove route key identity dependencies in ip_rt_get_source().

Pass in the sk_buff so that we can fetch the necessary keys from
the packet header when working with input routes.

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoipv4: Always call ip_options_build() after rest of IP header is filled in.
David S. Miller [Fri, 13 May 2011 21:21:27 +0000 (17:21 -0400)]
ipv4: Always call ip_options_build() after rest of IP header is filled in.

This will allow ip_options_build() to reliably look at the values of
iph->{daddr,saddr}

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoipv4: Kill spurious write to iph->daddr in ip_forward_options().
David S. Miller [Fri, 13 May 2011 21:15:50 +0000 (17:15 -0400)]
ipv4: Kill spurious write to iph->daddr in ip_forward_options().

This code block executes when opt->srr_is_hit is set.  It will be
set only by ip_options_rcv_srr().

ip_options_rcv_srr() walks until it hits a matching nexthop in the SRR
option addresses, and when it matches one 1) looks up the route for
that nexthop and 2) on route lookup success it writes that nexthop
value into iph->daddr.

ip_forward_options() runs later, and again walks the SRR option
addresses looking for the option matching the destination of the route
stored in skb_rtable().  This route will be the same exact one looked
up for the nexthop by ip_options_rcv_srr().

Therefore "rt->rt_dst == iph->daddr" must be true.

All it really needs to do is record the route's source address in the
matching SRR option adddress.  It need not write iph->daddr again,
since that has already been done by ip_options_rcv_srr() as detailed
above.

Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoolympic: convert to seq_file
Alexey Dobriyan [Fri, 13 May 2011 20:50:49 +0000 (16:50 -0400)]
olympic: convert to seq_file

->read_proc interface is going away, switch to seq_file.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet:set valid name before calling ndo_init()
Peter Pan(潘卫平) [Thu, 12 May 2011 15:46:56 +0000 (15:46 +0000)]
net:set valid name before calling ndo_init()

In commit 1c5cae815d19 (net: call dev_alloc_name from register_netdevice),
a bug of bonding was involved, see example 1 and 2.

In register_netdevice(), the name of net_device is not valid until
dev_get_valid_name() is called. But dev->netdev_ops->ndo_init(that is
bond_init) is called before dev_get_valid_name(),
and it uses the invalid name of net_device.

I think register_netdevice() should make sure that the name of net_device is
valid before calling ndo_init().

example 1:
modprobe bonding
ls  /proc/net/bonding/bond%d

ps -eLf
root      3398     2  3398  0    1 21:34 ?        00:00:00 [bond%d]

example 2:
modprobe bonding max_bonds=3

[  170.100292] bonding: Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011)
[  170.101090] bonding: Warning: either miimon or arp_interval and arp_ip_target module parameters must be specified, otherwise bonding will not detect link failures! see bonding.txt for details.
[  170.102469] ------------[ cut here ]------------
[  170.103150] WARNING: at /home/pwp/net-next-2.6/fs/proc/generic.c:586 proc_register+0x126/0x157()
[  170.104075] Hardware name: VirtualBox
[  170.105065] proc_dir_entry 'bonding/bond%d' already registered
[  170.105613] Modules linked in: bonding(+) sunrpc ipv6 uinput microcode ppdev parport_pc parport joydev e1000 pcspkr i2c_piix4 i2c_core [last unloaded: bonding]
[  170.108397] Pid: 3457, comm: modprobe Not tainted 2.6.39-rc2+ #14
[  170.108935] Call Trace:
[  170.109382]  [<c0438f3b>] warn_slowpath_common+0x6a/0x7f
[  170.109911]  [<c051a42a>] ? proc_register+0x126/0x157
[  170.110329]  [<c0438fc3>] warn_slowpath_fmt+0x2b/0x2f
[  170.110846]  [<c051a42a>] proc_register+0x126/0x157
[  170.111870]  [<c051a4dd>] proc_create_data+0x82/0x98
[  170.112335]  [<f94e6af6>] bond_create_proc_entry+0x3f/0x73 [bonding]
[  170.112905]  [<f94dd806>] bond_init+0x77/0xa5 [bonding]
[  170.113319]  [<c0721ac6>] register_netdevice+0x8c/0x1d3
[  170.113848]  [<f94e0e30>] bond_create+0x6c/0x90 [bonding]
[  170.114322]  [<f94f4763>] bonding_init+0x763/0x7b1 [bonding]
[  170.114879]  [<c0401240>] do_one_initcall+0x76/0x122
[  170.115317]  [<f94f4000>] ? 0xf94f3fff
[  170.115799]  [<c0463f1e>] sys_init_module+0x1286/0x140d
[  170.116879]  [<c07c6d9f>] sysenter_do_call+0x12/0x28
[  170.117404] ---[ end trace 64e4fac3ae5fff1a ]---
[  170.117924] bond%d: Warning: failed to register to debugfs
[  170.128728] ------------[ cut here ]------------
[  170.129360] WARNING: at /home/pwp/net-next-2.6/fs/proc/generic.c:586 proc_register+0x126/0x157()
[  170.130323] Hardware name: VirtualBox
[  170.130797] proc_dir_entry 'bonding/bond%d' already registered
[  170.131315] Modules linked in: bonding(+) sunrpc ipv6 uinput microcode ppdev parport_pc parport joydev e1000 pcspkr i2c_piix4 i2c_core [last unloaded: bonding]
[  170.133731] Pid: 3457, comm: modprobe Tainted: G        W   2.6.39-rc2+ #14
[  170.134308] Call Trace:
[  170.134743]  [<c0438f3b>] warn_slowpath_common+0x6a/0x7f
[  170.135305]  [<c051a42a>] ? proc_register+0x126/0x157
[  170.135820]  [<c0438fc3>] warn_slowpath_fmt+0x2b/0x2f
[  170.137168]  [<c051a42a>] proc_register+0x126/0x157
[  170.137700]  [<c051a4dd>] proc_create_data+0x82/0x98
[  170.138174]  [<f94e6af6>] bond_create_proc_entry+0x3f/0x73 [bonding]
[  170.138745]  [<f94dd806>] bond_init+0x77/0xa5 [bonding]
[  170.139278]  [<c0721ac6>] register_netdevice+0x8c/0x1d3
[  170.139828]  [<f94e0e30>] bond_create+0x6c/0x90 [bonding]
[  170.140361]  [<f94f4763>] bonding_init+0x763/0x7b1 [bonding]
[  170.140927]  [<c0401240>] do_one_initcall+0x76/0x122
[  170.141494]  [<f94f4000>] ? 0xf94f3fff
[  170.141975]  [<c0463f1e>] sys_init_module+0x1286/0x140d
[  170.142463]  [<c07c6d9f>] sysenter_do_call+0x12/0x28
[  170.142974] ---[ end trace 64e4fac3ae5fff1b ]---
[  170.144949] bond%d: Warning: failed to register to debugfs

Signed-off-by: Weiping Pan <panweiping3@gmail.com>
Reviewed-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agostmmac: fix autoneg in set_pauseparam
Giuseppe CAVALLARO [Thu, 12 May 2011 20:28:05 +0000 (20:28 +0000)]
stmmac: fix autoneg in set_pauseparam

This patch fixes a bug in the set_pauseparam
function that didn't well manage the ANE
field and returned broken values when use
ethtool -A|-a.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agostmmac: don't go through ethtool to start auto-negotiation
David Decotigny [Thu, 12 May 2011 20:28:04 +0000 (20:28 +0000)]
stmmac: don't go through ethtool to start auto-negotiation

The driver used to call phy's ethtool configuration routine to start
auto-negotiation. This change has it call directly phy's routine to
start auto-negotiation.

The initial version was hiding phy_start_aneg() return value,
this patch returns it (<0 upon error).

Tested: module compiles, tested on STM HDK7108 STB.

Signed-off-by: David Decotigny <decot@google.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agodrivers/isdn/hisax: Drop unused list
Julia Lawall [Fri, 13 May 2011 04:15:39 +0000 (04:15 +0000)]
drivers/isdn/hisax: Drop unused list

The file st5481_init.c locally defines and initializes the adapter_list
variable, but does not use it for anything.  Removing the list makes it
possible to remove the list field from the st5481_adapter data structure.
In the function probe_st5481, it also makes it possible to free the locally
allocated adapter value on an error exit.

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agonet: ipv4: add IPPROTO_ICMP socket kind
Vasiliy Kulikov [Fri, 13 May 2011 10:01:00 +0000 (10:01 +0000)]
net: ipv4: add IPPROTO_ICMP socket kind

This patch adds IPPROTO_ICMP socket kind.  It makes it possible to send
ICMP_ECHO messages and receive the corresponding ICMP_ECHOREPLY messages
without any special privileges.  In other words, the patch makes it
possible to implement setuid-less and CAP_NET_RAW-less /bin/ping.  In
order not to increase the kernel's attack surface, the new functionality
is disabled by default, but is enabled at bootup by supporting Linux
distributions, optionally with restriction to a group or a group range
(see below).

Similar functionality is implemented in Mac OS X:
http://www.manpagez.com/man/4/icmp/

A new ping socket is created with

    socket(PF_INET, SOCK_DGRAM, PROT_ICMP)

Message identifiers (octets 4-5 of ICMP header) are interpreted as local
ports. Addresses are stored in struct sockaddr_in. No port numbers are
reserved for privileged processes, port 0 is reserved for API ("let the
kernel pick a free number"). There is no notion of remote ports, remote
port numbers provided by the user (e.g. in connect()) are ignored.

Data sent and received include ICMP headers. This is deliberate to:
1) Avoid the need to transport headers values like sequence numbers by
other means.
2) Make it easier to port existing programs using raw sockets.

ICMP headers given to send() are checked and sanitized. The type must be
ICMP_ECHO and the code must be zero (future extensions might relax this,
see below). The id is set to the number (local port) of the socket, the
checksum is always recomputed.

ICMP reply packets received from the network are demultiplexed according
to their id's, and are returned by recv() without any modifications.
IP header information and ICMP errors of those packets may be obtained
via ancillary data (IP_RECVTTL, IP_RETOPTS, and IP_RECVERR). ICMP source
quenches and redirects are reported as fake errors via the error queue
(IP_RECVERR); the next hop address for redirects is saved to ee_info (in
network order).

socket(2) is restricted to the group range specified in
"/proc/sys/net/ipv4/ping_group_range".  It is "1 0" by default, meaning
that nobody (not even root) may create ping sockets.  Setting it to "100
100" would grant permissions to the single group (to either make
/sbin/ping g+s and owned by this group or to grant permissions to the
"netadmins" group), "0 4294967295" would enable it for the world, "100
4294967295" would enable it for the users, but not daemons.

The existing code might be (in the unlikely case anyone needs it)
extended rather easily to handle other similar pairs of ICMP messages
(Timestamp/Reply, Information Request/Reply, Address Mask Request/Reply
etc.).

Userspace ping util & patch for it:
http://openwall.info/wiki/people/segoon/ping

For Openwall GNU/*/Linux it was the last step on the road to the
setuid-less distro.  A revision of this patch (for RHEL5/OpenVZ kernels)
is in use in Owl-current, such as in the 2011/03/12 LiveCD ISOs:
http://mirrors.kernel.org/openwall/Owl/current/iso/

Initially this functionality was written by Pavel Kankovsky for
Linux 2.4.32, but unfortunately it was never made public.

All ping options (-b, -p, -Q, -R, -s, -t, -T, -M, -I), are tested with
the patch.

PATCH v3:
    - switched to flowi4.
    - minor changes to be consistent with raw sockets code.

PATCH v2:
    - changed ping_debug() to pr_debug().
    - removed CONFIG_IP_PING.
    - removed ping_seq_fops.owner field (unused for procfs).
    - switched to proc_net_fops_create().
    - switched to %pK in seq_printf().

PATCH v1:
    - fixed checksumming bug.
    - CAP_NET_RAW may not create icmp sockets anymore.

RFC v2:
    - minor cleanups.
    - introduced sysctl'able group range to restrict socket(2).

Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobridge: fix forwarding of IPv6
Stephen Hemminger [Fri, 13 May 2011 20:03:24 +0000 (16:03 -0400)]
bridge: fix forwarding of IPv6

The commit 6b1e960fdbd75dcd9bcc3ba5ff8898ff1ad30b6e
    bridge: Reset IPCB when entering IP stack on NF_FORWARD
broke forwarding of IPV6 packets in bridge because it would
call bp_parse_ip_options with an IPV6 packet.

Reported-by: Noah Meyerhans <noahm@debian.org>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'for-linville' of git://git.kernel.org/pub/scm/linux/kernel/git/luca...
John W. Linville [Fri, 13 May 2011 19:18:35 +0000 (15:18 -0400)]
Merge branch 'for-linville' of git://git./linux/kernel/git/luca/wl12xx

13 years agobonding,llc: Fix structure sizeof incompatibility for some PDUs
Vitalii Demianets [Thu, 12 May 2011 23:04:29 +0000 (23:04 +0000)]
bonding,llc: Fix structure sizeof incompatibility for some PDUs

With some combinations of arch/compiler (e.g. arm-linux-gcc) the sizeof
operator on structure returns value greater than expected. In cases when the
structure is used for mapping PDU fields it may lead to unexpected results
(such as holes and alignment problems in skb data). __packed prevents this
undesired behavior.

Signed-off-by: Vitalii Demianets <vitas@nppfactor.kiev.ua>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agossb: fix pcicore build breakage
John W. Linville [Fri, 13 May 2011 13:23:47 +0000 (09:23 -0400)]
ssb: fix pcicore build breakage

drivers/ssb/main.c:1336: error: 'SSB_PCICORE_BCAST_ADDR' undeclared (first use in this function)
drivers/ssb/main.c:1337: error: 'SSB_PCICORE_BCAST_DATA' undeclared (first use in this function)
drivers/ssb/main.c:1349: error: 'struct ssb_pcicore' has no member named 'dev'

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>