6 years agohyperv: Fix page buffer handling in rndis_filter_send_request()
Haiyang Zhang [Tue, 2 Oct 2012 05:30:21 +0000]
hyperv: Fix page buffer handling in rndis_filter_send_request()

To prevent possible data corruption in RNDIS requests, add another
page buffer if the request message crossed page boundary.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agohyperv: Fix the missing return value in rndis_filter_set_packet_filter()
Haiyang Zhang [Tue, 2 Oct 2012 05:30:20 +0000]
hyperv: Fix the missing return value in rndis_filter_set_packet_filter()

Return ETIMEDOUT when the reply message is not received in time.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agohyperv: Fix the max_xfer_size in RNDIS initialization
Haiyang Zhang [Tue, 2 Oct 2012 05:30:19 +0000]
hyperv: Fix the max_xfer_size in RNDIS initialization

According to RNDIS specs, Windows sets this size to
0x4000. I use the same value here.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agovxlan: put UDP socket in correct namespace
stephen hemminger [Mon, 1 Oct 2012 18:49:21 +0000]
vxlan: put UDP socket in correct namespace

Move vxlan UDP socket to correct network namespace

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agovxlan: Depend on CONFIG_INET
David S. Miller [Tue, 2 Oct 2012 18:37:31 +0000]
vxlan: Depend on CONFIG_INET

Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoMerge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/bwh/sfc...
David S. Miller [Tue, 2 Oct 2012 01:32:55 +0000]
Merge branch 'for-davem' of git://git./linux/kernel/git/bwh/sfc-next

Ben Hutchings says:

====================
Some bug fixes that should go into 3.7:

1. Fix oops when removing device with SR-IOV enabled.  (This regression
was introduced by the last set of changes, so the fix does not need to
be applied to any earlier kernel versions.)
2. Fix firmware structure field lookup bug that resulted in missing
sensor information.
3. Fix bug that makes self-test do very little in some configurations.
4. Fix the numbering of ethtool RX flow steering filters to reflect the
real hardware priorities.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosfc: Fix the reported priorities of different filter types
Ben Hutchings [Fri, 24 Aug 2012 19:12:20 +0000]
sfc: Fix the reported priorities of different filter types

Each RX filter table contains filters with two different levels of
specificity: TCP/IPv4 and UDP/IPv4 filters match the local address and
port and optionally the remote address and port; Ethernet filters
match the local address and optionally the VID.  The more specific
filters always override less specific filters within the same table,
and should be numbered accordingly.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

6 years agosfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP
Ben Hutchings [Tue, 4 Sep 2012 17:57:25 +0000]
sfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP

This filter flag cannot yet be set through the ethtool command and
will not be supported on future hardware.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

6 years agosfc: Fix loopback self-test with separate_tx_channels=1
Ben Hutchings [Thu, 13 Sep 2012 01:22:52 +0000]
sfc: Fix loopback self-test with separate_tx_channels=1

The loopback self-test iterates over all the TX queues of channel 0,
which is not very interesting when that's an RX-only channel.

Signed-off-by: Ben Hutchings <bhutchings@solarflre.com>

6 years agosfc: Fix MCDI structure field lookup
Ben Hutchings [Tue, 11 Sep 2012 20:46:41 +0000]
sfc: Fix MCDI structure field lookup

The least significant bit number (LBN) of a field within an MCDI
structure is counted from the start of the structure, not the
containing dword.  In MCDI_ARRAY_FIELD() we need to mask it rather
than using the usual EFX_DWORD_FIELD() macro.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

6 years agosfc: Add parentheses around use of bitfield macro arguments
Ben Hutchings [Tue, 11 Sep 2012 20:37:36 +0000]
sfc: Add parentheses around use of bitfield macro arguments

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

6 years agosfc: Fix null function pointer in efx_sriov_channel_type
Ben Hutchings [Tue, 2 Oct 2012 00:43:45 +0000]
sfc: Fix null function pointer in efx_sriov_channel_type

Commit c31e5f9 ('sfc: Add channel specific receive_skb handler and
post_remove callback') added the function pointer field
efx_channel_type::post_remove and an unconditional call through it.

This field should have been initialised to efx_channel_dummy_op_void
in the existing instances of efx_channel_type, but this was only done
in efx_default_channel_type.  Consequently, if a device has SR-IOV
enabled then removing the driver or device will result in an oops.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

6 years agovxlan: virtual extensible lan
stephen hemminger [Mon, 1 Oct 2012 12:32:35 +0000]
vxlan: virtual extensible lan

This is an implementation of Virtual eXtensible Local Area Network
as described in draft RFC:
  http://tools.ietf.org/html/draft-mahalingam-dutt-dcops-vxlan-02

The driver integrates a Virtual Tunnel Endpoint (VTEP) functionality
that learns MAC to IP address mapping.

This implementation has not been tested only against the Linux
userspace implementation using TAP, not against other vendor's
equipment.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoigmp: export symbol ip_mc_leave_group
stephen hemminger [Mon, 1 Oct 2012 12:32:34 +0000]
igmp: export symbol ip_mc_leave_group

Needed for VXLAN.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agonetlink: add attributes to fdb interface
stephen hemminger [Mon, 1 Oct 2012 12:32:33 +0000]
netlink: add attributes to fdb interface

Later changes need to be able to refer to neighbour attributes
when doing fdb_add.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: unconditionally select HWMON support when tg3 is enabled.
Paul Gortmaker [Mon, 1 Oct 2012 15:43:49 +0000]
tg3: unconditionally select HWMON support when tg3 is enabled.

There is the seldom used corner case where HWMON=m at the same
time as TIGON3=y (typically randconfigs) which will cause a link
fail like:

drivers/built-in.o: In function `tg3_close':
tg3.c:(.text+0x16bd86): undefined reference to `hwmon_device_unregister'
drivers/built-in.o: In function `tg3_hwmon_open':
tg3.c:(.text+0x16fc4b): undefined reference to `hwmon_device_register'
make[1]: *** [vmlinux] Error 1

Fix it as suggested by DaveM[1] by having the Kconfig logic simply
select HWMON when TIGON3 is selected.  This gets rid of all the
extra IS_ENABLED ifdeffery in tg3.c as a side benefit.

[1] http://marc.info/?l=linux-netdev&m=134250573718151&w=2

Cc: Michael Chan <mchan@broadcom.com>
Reported-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reported-by: Anisse Astier <anisse@astier.eu>
Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoRevert "net: ti cpsw ethernet: allow reading phy interface mode from DT"
David S. Miller [Mon, 1 Oct 2012 21:39:31 +0000]
Revert "net: ti cpsw ethernet: allow reading phy interface mode from DT"

This reverts commit d7559982701ac500662b2e8e150ff34f7faf0281.

It wasn't meant to be applied, commit
342b7b741d76bc8aadeff844634348bb2a343d19 ("net: ti cpsw ethernet: set
IFCTL_A bit in MACCONTROL") was redone in such a way to make this
commit unnecessary.

Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agogre: fix sparse warning
stephen hemminger [Mon, 1 Oct 2012 05:21:14 +0000]
gre: fix sparse warning

Use be16 consistently when looking at flags.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoasix: Adds support for Lenovo 10/100 USB dongle.
Quinlan Pfiffer [Fri, 28 Sep 2012 19:58:44 +0000]
asix: Adds support for Lenovo 10/100 USB dongle.

This dongle ships with the X1 Carbon, and has an AX88772B
usb to ethernet chip in it.

Signed-off-by: Quinlan Pfiffer <qpfiffer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoflexcan: disable bus error interrupts for the i.MX28
Wolfgang Grandegger [Fri, 28 Sep 2012 03:17:15 +0000]
flexcan: disable bus error interrupts for the i.MX28

Due to a bug in most Flexcan cores, the bus error interrupt needs
to be enabled. Otherwise we don't get any error warning or passive
interrupts. This is _not_ necessary for the i.MX28 and this patch
disables bus error interrupts if "berr-reporting" is not requested.
This avoids bus error flooding, which might harm, especially on
low-end systems.

To handle such quirks of the Flexcan cores, a hardware feature flag
has been introduced, also replacing the "hw_ver" variable. So far
nobody could tell what Flexcan core version is available on what
Freescale SOC, apart from the i.MX6Q and P1010, and which bugs or
features are present on the various "hw_rev".

CC: Hui Wang <jason77.wang@gmail.com>
CC: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agonet/key/af_key.c: add range checks on ->sadb_x_policy_len
Dan Carpenter [Thu, 27 Sep 2012 22:21:19 +0000]
net/key/af_key.c: add range checks on ->sadb_x_policy_len

Because sizeof() is size_t then if "len" is negative, it counts as a
large positive value.

The call tree looks like:
pfkey_sendmsg()
-> pfkey_process()
   -> pfkey_spdadd()
      -> parse_ipsecrequests()

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoIB/ipoib: Add more rtnl_link_ops callbacks
Or Gerlitz [Thu, 27 Sep 2012 12:06:02 +0000]
IB/ipoib: Add more rtnl_link_ops callbacks

Add the rtnl_link_ops changelink and fill_info callbacks, through
which the admin can now set/get the driver mode, etc policies.
Maintain the proprietary sysfs entries only for legacy childs.

For child devices, set dev->iflink to point to the parent
device ifindex, such that user space tools can now correctly
show the uplink relation as done for vlan, macvlan, etc
devices. Pointed out by Patrick McHardy <kaber@trash.net>

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agonet: ti cpsw ethernet: allow reading phy interface mode from DT
Daniel Mack [Wed, 26 Sep 2012 07:24:27 +0000]
net: ti cpsw ethernet: allow reading phy interface mode from DT

Allow users to specify the phy interface of the CPSW slaves. The new
node parameter is called "phy_if_mode" and is optional. The original
behaviour of the driver is preserved when not given.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Cc: Mugunthan V N <mugunthanvnm@ti.com>
Cc: Vaibhav Hiremath <hvaibhav@ti.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agonet: ti cpsw ethernet: set IFCTL_A bit in MACCONTROL
Daniel Mack [Thu, 27 Sep 2012 09:19:34 +0000]
net: ti cpsw ethernet: set IFCTL_A bit in MACCONTROL

For RMII/RGMII mode operation in 100Mbps, the CPSW needs to set the
IFCTL_A bits in the MACCONTROL register. For all other PHY modes, this
bit is unused, so setting it unconditionally shouldn't cause any
trouble.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Cc: Mugunthan V N <mugunthanvnm@ti.com>
Cc: Vaibhav Hiremath <hvaibhav@ti.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agomlx4: dont orphan skbs in mlx4_en_xmit()
Eric Dumazet [Fri, 28 Sep 2012 07:53:26 +0000]
mlx4: dont orphan skbs in mlx4_en_xmit()

After commit e22979d96a55d (mlx4_en: Moving to Interrupts for TX
completions) we no longer need to orphan skbs in mlx4_en_xmit()
since skb wont stay a long time in TX ring before their release.

Orphaning skbs in ndo_start_xmit() should be avoided as much as
possible, since it breaks TCP Small Queue or other flow control
mechanisms (per socket limits)

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Yevgeny Petrilin <yevgenyp@mellanox.com>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoipv4: gre: add GRO capability
Eric Dumazet [Thu, 27 Sep 2012 02:48:50 +0000]
ipv4: gre: add GRO capability

Add GRO capability to IPv4 GRE tunnels, using the gro_cells
infrastructure.

Tested using IPv4 and IPv6 TCP traffic inside this tunnel, and
checking GRO is building large packets.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agonet: add gro_cells infrastructure
Eric Dumazet [Thu, 27 Sep 2012 19:29:05 +0000]
net: add gro_cells infrastructure

This adds a new include file (include/net/gro_cells.h), to bring GRO
(Generic Receive Offload) capability to tunnels, in a modular way.

Because tunnels receive path is lockless, and GRO adds a serialization
using a napi_struct, I chose to add an array of up to
DEFAULT_MAX_NUM_RSS_QUEUES cells, so that multi queue devices wont be
slowed down because of GRO layer.

skb_get_rx_queue() is used as selector.

In the future, we might add optional fanout capabilities, using rxhash
for example.

With help from Ben Hutchings who reminded me
netif_get_num_default_rss_queues() function.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotcp: gro: add checksuming helpers
Eric Dumazet [Thu, 27 Sep 2012 02:14:53 +0000]
tcp: gro: add checksuming helpers

skb with CHECKSUM_NONE cant currently be handled by GRO, and
we notice this deep in GRO stack in tcp[46]_gro_receive()

But there are cases where GRO can be a benefit, even with a lack
of checksums.

This preliminary work is needed to add GRO support
to tunnels.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoipv6: del unreachable route when an addr is deleted on lo
Nicolas Dichtel [Wed, 26 Sep 2012 00:04:55 +0000]
ipv6: del unreachable route when an addr is deleted on lo

When an address is added on loopback (ip -6 a a 2002::1/128 dev lo), two routes
are added:
 - one in the local table:
    local 2002::1 via :: dev lo  proto none  metric 0
 - one the in main table (for the prefix):
    unreachable 2002::1 dev lo  proto kernel  metric 256  error -101

When the address is deleted, the route inserted in the main table remains
because we use rt6_lookup(), which returns NULL when dst->error is set, which
is the case here! Thus, it is better to use ip6_route_lookup() to avoid this
kind of filter.

Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobnx2x: update version to 1.78.00-0.
Yuval Mintz [Mon, 1 Oct 2012 03:46:20 +0000]
bnx2x: update version to 1.78.00-0.

Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobnx2x,cnic: use FW 7.8.2
Yuval Mintz [Mon, 1 Oct 2012 03:46:19 +0000]
bnx2x,cnic: use FW 7.8.2

This patch moves the bnx2x and cnic drivers into using FW 7.8.2
which was recently submitted into the linux-firmware tree.

A short summary of minor bugs fixed by this FW:
 1. In switch dependent mode, fix several issues regarding inner vlan
    vs. DCB priorities.
 2. iSCSI - not all packets were completed on a forward channel.
 3. DCB - fixed for 4-port devices.
 4. Fixed false parity reported in CAM memories when operating near -5%
    on the 1.0V core supply.
 5. ETS default settings are set to fairness between traffic classes
    (rather than strict priority), and uses the same chip receive buffer
    configuration for both PFC and pause.

For a complete list of fixes made by this FW, see commit 236367db
in the linux-firmware git repository.

Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobe2net: fix vfs enumeration
Ivan Vecera [Mon, 1 Oct 2012 01:56:55 +0000]
be2net: fix vfs enumeration

Current VFs enumeration algorithm used in be_find_vfs does not take domain
number into the match. The match found in igb/ixgbe is more elegant and
safe.

This 2nd version uses pci_physfn instead of checking dev->physfn directly.

Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agouse skb_end_offset() in skb_try_coalesce()
Weiping Pan [Fri, 28 Sep 2012 20:15:30 +0000]
use skb_end_offset() in skb_try_coalesce()

Commit ec47ea824774(skb: Add inline helper for getting the skb end offset from
head) introduces this helper function, skb_end_offset(),
we should make use of it.

Signed-off-by: Weiping Pan <wpan@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
John W. Linville [Mon, 1 Oct 2012 11:39:36 +0000]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-next into for-davem

6 years agoMerge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/linville...
David S. Miller [Sun, 30 Sep 2012 06:30:16 +0000]
Merge branch 'for-davem' of git://git./linux/kernel/git/linville/wireless-next

John W. Linville says:

====================
Here is another batch of updates intended for 3.7...

Highlights include an hci_connect re-write in Bluetooth, HCI/LLC
layer separation in NFC, removal of the raw pn544 NFC driver, NFC LLCP
raw sockets support, improved IBSS auth frame handling in mac80211,
full-MAC AP mode notification support in mac80211, a lot of attention
paid to brcmfmac, and the usual level of updates to iwlwifi, ath9k,
mwifiex, and rt2x00, and various other updates.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobe2net: fixup log messages
Sathya Perla [Fri, 28 Sep 2012 04:39:44 +0000]
be2net: fixup log messages

Added and modified a few log messages mostly in probe path.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobe2net: cleanup code related to be_link_status_query()
Sathya Perla [Fri, 28 Sep 2012 04:39:43 +0000]
be2net: cleanup code related to be_link_status_query()

1) link_status_query() is always called to query the link-speed (speed
after applying qos). When there is no qos setting, link-speed is derived from
port-speed. Do all this inside this routine and hide this from the callers.

2) adpater->phy.forced_port_speed is not being set anywhere after being
initialized. Get rid of this variable.

3) Ignore async link_speed notifications till the initial value has been
fetched from FW.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobe2net: fix wrong handling of be_setup() failure in be_probe()
Sathya Perla [Fri, 28 Sep 2012 04:39:42 +0000]
be2net: fix wrong handling of be_setup() failure in be_probe()

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agobe2net: remove type argument of be_cmd_mac_addr_query()
Sathya Perla [Fri, 28 Sep 2012 04:39:41 +0000]
be2net: remove type argument of be_cmd_mac_addr_query()

All invocations of this routine use the same type value.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Disable multiple TX rings by default due to hardware flaw
Michael Chan [Fri, 28 Sep 2012 07:12:43 +0000]
tg3: Disable multiple TX rings by default due to hardware flaw

Simple round-robin hardware TX scheduling can cause starvation of TX rings
with small packets when other TX rings have large TSO or jumbo packets.

In the simplest case, consider 2 TCP streams running in opposite
directions.  The TSO TX traffic will hash to one ring and the ACKs for the
incoming data on a different TCP connection will hash to a different TX
ring.  The hardware fetches one complete TSO packet (up to 64K data)
before servicing the other TX ring.  When it gets to the other TX ring, it
will only fetch one packet (64-byte ACK packet in this case).  After that,
it will switch back to the 1st ring filled with more TSO packets.  Because
only one ACK can go out roughly every 500 usec in this case, the incoming
data rate becomes very low.

Update version to 3.125.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Add support for ethtool -L|-l to get/set the number of rings.
Michael Chan [Fri, 28 Sep 2012 07:12:42 +0000]
tg3: Add support for ethtool -L|-l to get/set the number of rings.

Default remains the same.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Refactor tg3_close()
Michael Chan [Fri, 28 Sep 2012 07:12:41 +0000]
tg3: Refactor tg3_close()

by introducing tg3_stop() that does the opposite of tg3_start().  This
function will be useful when adding the support for changing the numbe
of rx and tx rings.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Refactor tg3_open()
Michael Chan [Fri, 28 Sep 2012 07:12:40 +0000]
tg3: Refactor tg3_open()

by introducing tg3_start() that handles all initialization steps from
IRQ allocation.  This function will be needed when adding support for
changing the number of rx and tx rings.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Separate coalescing setup for rx and tx
Michael Chan [Fri, 28 Sep 2012 07:12:39 +0000]
tg3: Separate coalescing setup for rx and tx

since the number of rings can be different.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Allow number of rx and tx rings to be set independently.
Michael Chan [Fri, 28 Sep 2012 07:12:38 +0000]
tg3: Allow number of rx and tx rings to be set independently.

irq_cnt is no longer necessarily equal to the number rx or tx rings.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agotg3: Introduce separate functions to allocate/free RX/TX rings.
Michael Chan [Fri, 28 Sep 2012 07:12:37 +0000]
tg3: Introduce separate functions to allocate/free RX/TX rings.

This is preparation work to allow the number of RX and TX rings to be
configured separately.

Reviewed-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoipv6: return errno pointers consistently for fib6_add_1()
Lin Ming [Tue, 25 Sep 2012 15:17:07 +0000]
ipv6: return errno pointers consistently for fib6_add_1()

fib6_add_1() should consistently return errno pointers,
rather than a mixture of NULL and errno pointers.

Signed-off-by: Lin Ming <mlin@ss.pku.edu.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoMerge branch 'for-davem' of git://gitorious.org/linux-can/linux-can-next
David S. Miller [Fri, 28 Sep 2012 22:46:29 +0000]
Merge branch 'for-davem' of git://gitorious.org/linux-can/linux-can-next

Marc Kleine-Budde says:

====================
this pull request is for net-next, for the v3.7 release cycle.
AnilKumar Ch contributed a fix for a segfault in the c_can driver,
which is triggered by an earlier commit [1] in net-next (so no backport
is needed).
...
[1] 4cdd34b can: c_can: Add runtime PM support to Bosch C_CAN/D_CAN controller
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc75xx: add wol magic packet support
Steve Glendinning [Fri, 28 Sep 2012 00:57:53 +0000]
smsc75xx: add wol magic packet support

This patch enables wake from system suspend on magic packet.

Patch updated to change BUG_ON to WARN_ON_ONCE.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc75xx: enable power saving mode during system suspend
Steve Glendinning [Fri, 28 Sep 2012 00:57:52 +0000]
smsc75xx: enable power saving mode during system suspend

This patch instructs the device to enter its lowest power SUSPEND2
state during system suspend.

This patch also explicitly wakes the device after resume, which
should address reports of the device not automatically coming
back after system suspend:

Patch updated to change BUG_ON to WARN_ON_ONCE.

http://code.google.com/p/chromium-os/issues/detail?id=31871

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc75xx: add explicit test that device is READY
Steve Glendinning [Fri, 28 Sep 2012 00:57:51 +0000]
smsc75xx: add explicit test that device is READY

This patch adds an explicit test that the READY bit is set on
the device when attempting to initialize it.

If this bit is clear then the device hasn't succesfully started
all its clocks, and this patch helps make the resulting logged
error more helpful.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: add wol magic packet support
Steve Glendinning [Fri, 28 Sep 2012 00:07:12 +0000]
smsc95xx: add wol magic packet support

This patch enables wake from system suspend on magic packet.

Patch updated to replace BUG_ON with WARN_ON_ONCE and return.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: enable power saving mode during system suspend
Steve Glendinning [Fri, 28 Sep 2012 00:07:11 +0000]
smsc95xx: enable power saving mode during system suspend

This patch enables the device to enter its lowest power SUSPEND2
state during system suspend, instead of staying up using full power.

Patch updated to not add two pointers to .suspend & .resume.

Patch updated to replace BUG_ON with WARN_ON_ONCE and return.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: fix resume when usb device is reset
Steve Glendinning [Fri, 28 Sep 2012 00:07:10 +0000]
smsc95xx: fix resume when usb device is reset

This patch fixes an issue on some systems, where after suspend the
link is re-established but the ethernet interface does not resume.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: check return code from control messages
Steve Glendinning [Fri, 28 Sep 2012 00:07:09 +0000]
smsc95xx: check return code from control messages

This patch adds additional checks of the values returned by
smsc95xx_(read|write)_reg, and wraps their common patterns
in macros.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: remove unnecessary variables
Steve Glendinning [Fri, 28 Sep 2012 00:07:08 +0000]
smsc95xx: remove unnecessary variables

Removes unnecessary variables as smsc95xx_write_reg takes its
value by parameter.  Early versions passed this parameter by
reference.

Also replace hardcoded interrupt status value with a #define

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agosmsc95xx: sleep before read for lengthy operations
Steve Glendinning [Fri, 28 Sep 2012 00:07:07 +0000]
smsc95xx: sleep before read for lengthy operations

During init, the device reset is unexpected to complete immediately,
so sleep before testing the condition rather than after it.

Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Fri, 28 Sep 2012 18:40:49 +0000]
Merge git://git./linux/kernel/git/davem/net

Conflicts:
drivers/net/team/team.c
drivers/net/usb/qmi_wwan.c
net/batman-adv/bat_iv_ogm.c
net/ipv4/fib_frontend.c
net/ipv4/route.c
net/l2tp/l2tp_netlink.c

The team, fib_frontend, route, and l2tp_netlink conflicts were simply
overlapping changes.

qmi_wwan and bat_iv_ogm were of the "use HEAD" variety.

With help from Antonio Quartulli.

Signed-off-by: David S. Miller <davem@davemloft.net>

6 years agoath9k: Fix BTCOEX weight initialization
Sujith Manoharan [Fri, 28 Sep 2012 05:43:51 +0000]
ath9k: Fix BTCOEX weight initialization

The WLAN/BT weights have to set correctly before BTCOEX
is initialized. Currently, this is not done for all chips
in the AR9003 family. This patch fixes this issue by setting
the weights in the init path. While at it, rename ar9462_wlan_weights
to mci_wlan_weights since it is common to both AR9462 and AR9565.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: convert to use le16_add_cpu()
Wei Yongjun [Fri, 28 Sep 2012 04:59:43 +0000]
mwifiex: convert to use le16_add_cpu()

Convert cpu_to_le16(le16_to_cpu(E1) + E2) to use le16_add_cpu().

dpatch engine is used to auto generate this patch.
(https://github.com/weiyj/dpatch)

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: enhance RX reordering to avoid packet drop during host sleep
Avinash Patil [Fri, 28 Sep 2012 02:00:10 +0000]
mwifiex: enhance RX reordering to avoid packet drop during host sleep

Once hostsleep is activated, this patch takes care of not
dropping packets in RX reorder table.

Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k_hw: Handle AR_INTR_SYNC_HOST1_(FATAL|PERR) on AR9003
Sven Eckelmann [Thu, 27 Sep 2012 14:41:02 +0000]
ath9k_hw: Handle AR_INTR_SYNC_HOST1_(FATAL|PERR) on AR9003

Interrupts with the sync_cause AR_INTR_SYNC_HOST1_FATAL and
AR_INTR_SYNC_HOST1_PERR have to be handled using a chip reset. Otherwise a
interrupt storm with unhandled interrupts will cause a hang or crash of the
machine.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: remove brcmf_update_prof() function
Arend van Spriel [Thu, 27 Sep 2012 12:17:57 +0000]
brcmfmac: remove brcmf_update_prof() function

The function brcmf_update_prof() only add complexity that is not
needed.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: remove brcmf_read_prof() function
Arend van Spriel [Thu, 27 Sep 2012 12:17:56 +0000]
brcmfmac: remove brcmf_read_prof() function

The function brcmf_read_prof() provided access to fields
with brcmf_cfg80211_profile structure using a field id,
but the calling function can easily access that structure
directly. This patch removes the function.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: rename structure brcmf_cfg80211_priv
Arend van Spriel [Thu, 27 Sep 2012 12:17:55 +0000]
brcmfmac: rename structure brcmf_cfg80211_priv

Renamed structure to brcmf_cfg80211_info as it is not really a
private structure since it is exposed in header file.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: add hostap supoort.
Hante Meuleman [Thu, 27 Sep 2012 12:17:54 +0000]
brcmfmac: add hostap supoort.

This patch adds support for host AP mode.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: remove two obsolete structure definitions
Arend van Spriel [Thu, 27 Sep 2012 12:17:53 +0000]
brcmfmac: remove two obsolete structure definitions

The structures brcmf_cfg80211_dev and brcmf_cfg80211_iface are only
adding complexity and provide no additional value. These have been
removed in this patch.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: use brcmf_cfg80211_priv to interface with wl_cfg80211 code
Arend van Spriel [Thu, 27 Sep 2012 12:17:52 +0000]
brcmfmac: use brcmf_cfg80211_priv to interface with wl_cfg80211 code

The low-level driver part interfaces with wl_cfg80211 part using
brcmf_cfg80211_dev structure as handle. As brcmf_cfg80211_priv
is defined in interface it is more efficient to use that as handle
in function calls.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: get rid of extern keyword in wl_cfg80211.h
Arend van Spriel [Thu, 27 Sep 2012 12:17:51 +0000]
brcmfmac: get rid of extern keyword in wl_cfg80211.h

The function declarations in wl_cfg80211.h used the extern keyword
but it is redundant as that is implicit so they were removed.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: notify common driver about usb tx completion.
Hante Meuleman [Thu, 27 Sep 2012 12:17:50 +0000]
brcmfmac: notify common driver about usb tx completion.

brcmf_usb_tx_complete now calls brcmf_txcomplete to correct
eapol counter.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: use define instead of hardcoded values.
Hante Meuleman [Thu, 27 Sep 2012 12:17:49 +0000]
brcmfmac: use define instead of hardcoded values.

brcmf_parse_tlvs was using hardcoded values.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: use different fw api for encryption,auth. config
Hante Meuleman [Thu, 27 Sep 2012 12:17:48 +0000]
brcmfmac: use different fw api for encryption,auth. config

This patch changes the commands being used to configure
encryption and authentication. These new methods are needed
for when p2p and hostap support are added.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: use wait_event_timeout for fw control packets over usb.
Hante Meuleman [Thu, 27 Sep 2012 12:17:47 +0000]
brcmfmac: use wait_event_timeout for fw control packets over usb.

firmware control packets over usb are async acked. This is
awaited for by using interruptible wait. This can be
interrupted by signal which results in undesirable behavior.
Instead wait_event_timeout is used.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agob43legacy: Fix crash on unload when firmware not available
Larry Finger [Wed, 26 Sep 2012 17:32:02 +0000]
b43legacy: Fix crash on unload when firmware not available

When b43legacy is loaded without the firmware being available, a following
unload generates a kernel NULL pointer dereference BUG as follows:

[  214.330789] BUG: unable to handle kernel NULL pointer dereference at 0000004c
[  214.330997] IP: [<c104c395>] drain_workqueue+0x15/0x170
[  214.331179] *pde = 00000000
[  214.331311] Oops: 0000 [#1] SMP
[  214.331471] Modules linked in: b43legacy(-) ssb pcmcia mac80211 cfg80211 af_packet mperf arc4 ppdev sr_mod cdrom sg shpchp yenta_socket pcmcia_rsrc pci_hotplug pcmcia_core battery parport_pc parport floppy container ac button edd autofs4 ohci_hcd ehci_hcd usbcore usb_common thermal processor scsi_dh_rdac scsi_dh_hp_sw scsi_dh_emc scsi_dh_alua scsi_dh fan thermal_sys hwmon ata_generic pata_ali libata [last unloaded: cfg80211]
[  214.333421] Pid: 3639, comm: modprobe Not tainted 3.6.0-rc6-wl+ #163 Source Technology VIC 9921/ALI Based Notebook
[  214.333580] EIP: 0060:[<c104c395>] EFLAGS: 00010246 CPU: 0
[  214.333687] EIP is at drain_workqueue+0x15/0x170
[  214.333788] EAX: c162ac40 EBX: cdfb8360 ECX: 0000002a EDX: 00002a2a
[  214.333890] ESI: 00000000 EDI: 00000000 EBP: cd767e7c ESP: cd767e5c
[  214.333957]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[  214.333957] CR0: 8005003b CR2: 0000004c CR3: 0c96a000 CR4: 00000090
[  214.333957] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[  214.333957] DR6: ffff0ff0 DR7: 00000400
[  214.333957] Process modprobe (pid: 3639, ti=cd766000 task=cf802e90 task.ti=cd766000)
[  214.333957] Stack:
[  214.333957]  00000292 cd767e74 c12c5e09 00000296 00000296 cdfb8360 cdfb9220 00000000
[  214.333957]  cd767e90 c104c4fd cdfb8360 cdfb9220 cd682800 cd767ea4 d0c10184 cd682800
[  214.333957]  cd767ea4 cba31064 cd767eb8 d0867908 cba31064 d087e09c cd96f034 cd767ec4
[  214.333957] Call Trace:
[  214.333957]  [<c12c5e09>] ? skb_dequeue+0x49/0x60
[  214.333957]  [<c104c4fd>] destroy_workqueue+0xd/0x150
[  214.333957]  [<d0c10184>] ieee80211_unregister_hw+0xc4/0x100 [mac80211]
[  214.333957]  [<d0867908>] b43legacy_remove+0x78/0x80 [b43legacy]
[  214.333957]  [<d083654d>] ssb_device_remove+0x1d/0x30 [ssb]
[  214.333957]  [<c126f15a>] __device_release_driver+0x5a/0xb0
[  214.333957]  [<c126fb07>] driver_detach+0x87/0x90
[  214.333957]  [<c126ef4c>] bus_remove_driver+0x6c/0xe0
[  214.333957]  [<c1270120>] driver_unregister+0x40/0x70
[  214.333957]  [<d083686b>] ssb_driver_unregister+0xb/0x10 [ssb]
[  214.333957]  [<d087c488>] b43legacy_exit+0xd/0xf [b43legacy]
[  214.333957]  [<c1089dde>] sys_delete_module+0x14e/0x2b0
[  214.333957]  [<c110a4a7>] ? vfs_write+0xf7/0x150
[  214.333957]  [<c1240050>] ? tty_write_lock+0x50/0x50
[  214.333957]  [<c110a6f8>] ? sys_write+0x38/0x70
[  214.333957]  [<c1397c55>] syscall_call+0x7/0xb
[  214.333957] Code: bc 27 00 00 00 00 a1 74 61 56 c1 55 89 e5 e8 a3 fc ff ff 5d c3 90 55 89 e5 57 56 89 c6 53 b8 40 ac 62 c1 83 ec 14 e8 bb b7 34 00 <8b> 46 4c 8d 50 01 85 c0 89 56 4c 75 03 83 0e 40 80 05 40 ac 62
[  214.333957] EIP: [<c104c395>] drain_workqueue+0x15/0x170 SS:ESP 0068:cd767e5c
[  214.333957] CR2: 000000000000004c
[  214.341110] ---[ end trace c7e90ec026d875a6 ]---Index: wireless-testing/drivers/net/wireless/b43legacy/main.c

The problem is fixed by making certain that the ucode pointer is not NULL
before deregistering the driver in mac80211.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Stable <stable@vger.kernel.org> [v 3.3.0+]
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k: Enable MCI for AR9565
Bala Shanmugam [Wed, 26 Sep 2012 11:34:26 +0000]
ath9k: Enable MCI for AR9565

Signed-off-by: Bala Shanmugam <bkamatch@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobrcmfmac: use kcalloc() to prevent integer overflow
Dan Carpenter [Wed, 26 Sep 2012 07:21:48 +0000]
brcmfmac: use kcalloc() to prevent integer overflow

The multiplication here looks like it could overflow.  I've changed it
to use kcalloc() to prevent that.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k: Fix chainmask selection for AR9462
Sujith Manoharan [Wed, 26 Sep 2012 06:52:08 +0000]
ath9k: Fix chainmask selection for AR9462

Force chain 1 to be used for CCK rates since the target power
table stored in EEPROM is too high to transmit with both chains.
This is needed to avoid regulatory violation.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: add inactivity deauth support for ap
Kevin Gan [Wed, 26 Sep 2012 03:23:45 +0000]
mwifiex: add inactivity deauth support for ap

The firmware has support for this feature, so we offload it to
firmware. In start_ap, driver passes the inactivity timeout
value to firmware via TLVs and firmware will report STA_DEAUTH
event to driver when inactivity timer is fired.

Signed-off-by: Kevin Gan <ganhy@marvell.com>
Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Nishant Sarmukadam <nishants@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: set txpd when send a mgmt frame for AP and GO mode
Stone Piao [Wed, 26 Sep 2012 03:23:44 +0000]
mwifiex: set txpd when send a mgmt frame for AP and GO mode

Set packet type and packet offset in txpd when send a mgmt frame
in AP and GO mode.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kevin Gan <ganhy@marvell.com>
Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: parse P2P IEs from beacon_data
Stone Piao [Wed, 26 Sep 2012 03:23:43 +0000]
mwifiex: parse P2P IEs from beacon_data

Parse P2P IEs and set them to FW with related masks.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: add support for P2P GO in interface type change
Stone Piao [Wed, 26 Sep 2012 03:23:42 +0000]
mwifiex: add support for P2P GO in interface type change

When cfg80211 calls to change interface type for P2P GO, send
P2P mode config commands to firmware and set bss role and bss
mode accordingly.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: add support for P2P client in interface type change
Stone Piao [Wed, 26 Sep 2012 03:23:41 +0000]
mwifiex: add support for P2P client in interface type change

When cfg80211 calls to change interface type for P2P client,
send P2P mode config commands to firmware.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: add P2P interface
Stone Piao [Wed, 26 Sep 2012 03:23:40 +0000]
mwifiex: add P2P interface

Due to firmware design, driver needs to add a default P2P interface
to implement find phase and action frame handshake.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: send firmware initialization commands synchronously
Stone Piao [Wed, 26 Sep 2012 03:23:39 +0000]
mwifiex: send firmware initialization commands synchronously

The driver will send some commands to firmware during the
initialization. Currently these commands are sent asynchronously,
which means that we firstly insert all of them to a pre-allocated
command queue, and then start to process them one by one. The
command queue will soon be exhausted if we keep adding new
initialization commands.

This issue can be resolved by sending initialization commands
synchronously because each command is consumed and the buffer is
recycled before queuing next command.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: append each IE into a seperate IE buffer
Stone Piao [Wed, 26 Sep 2012 03:23:38 +0000]
mwifiex: append each IE into a seperate IE buffer

When scan is triggered from cfg80211, the request contains some IEs
and we should parse all the IEs and append each IE into a seperate
IE buffer.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: process remain on channel expired event
Stone Piao [Wed, 26 Sep 2012 03:23:37 +0000]
mwifiex: process remain on channel expired event

Process remain on channel expired event, and notify cfg80211.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: implement remain_on_channel and cancel_remain_on_channel
Stone Piao [Wed, 26 Sep 2012 03:23:36 +0000]
mwifiex: implement remain_on_channel and cancel_remain_on_channel

Add a new command to implement remain_on_channel and
cancel_remain_on_channel.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: report received management frames to cfg80211
Stone Piao [Wed, 26 Sep 2012 03:23:35 +0000]
mwifiex: report received management frames to cfg80211

Process the management frames received from firmware and report
them to cfg80211.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Kevin Gan <ganhy@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: implement cfg80211 mgmt_frame_register handler
Stone Piao [Wed, 26 Sep 2012 03:23:34 +0000]
mwifiex: implement cfg80211 mgmt_frame_register handler

Add a new command to implement mgmt_frame_register.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: advertise mgmt_stype to cfg80211
Stone Piao [Wed, 26 Sep 2012 03:23:33 +0000]
mwifiex: advertise mgmt_stype to cfg80211

Advertise supported management frame types to cfg80211.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: implement cfg80211 mgmt_tx handler
Stone Piao [Wed, 26 Sep 2012 03:23:32 +0000]
mwifiex: implement cfg80211 mgmt_tx handler

Implement mgmt_tx in cfg80211 ops through data path.
Advertise probe resp offload and skip to send probe resp in AP
or GO mode.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Yogesh Ashok Powar <yogeshp@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agomwifiex: fix coding style issue
Stone Piao [Wed, 26 Sep 2012 03:23:31 +0000]
mwifiex: fix coding style issue

Remove unnecessary blank lines and extra tab.

Signed-off-by: Stone Piao <piaoyun@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k: Add a debugfs file to adjust antenna diversity
Sujith Manoharan [Wed, 26 Sep 2012 02:25:18 +0000]
ath9k: Add a debugfs file to adjust antenna diversity

Location: /<debugfs>/ieee80211/phy#/ath9k/diversity

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k: Handle errors properly in MCI initialization
Sujith Manoharan [Wed, 26 Sep 2012 02:24:43 +0000]
ath9k: Handle errors properly in MCI initialization

The MCI initialization path has various points of failures,
handle these to ensure that we bail out correctly in such
cases.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoath9k: Fix rx filtering issue for older chips
Thomas Wagner [Tue, 25 Sep 2012 16:02:55 +0000]
ath9k: Fix rx filtering issue for older chips

We need to have the promiscuous mode enabled for older
chipsets so that the olderchips hardware does not
filters out some valid/necessary frames that need
to be sent to mac80211. Fix this by enabling promiscus
mode for all the chipsets whose macversion <= AR9160
chipsets. This should fix
https://bugzilla.kernel.org/show_bug.cgi?id=45591

shafi: made the fix generic by having the frame filtering
disabled for chipsets older than AR9280.

Cc: Javier Cardona <javier@cozybit.com>
Signed-off-by: Thomas Wagner <Thomas.Wagner@hs-rm.de>
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agobcma: change delays to follow timers-howto guide
Rafał Miłecki [Tue, 25 Sep 2012 08:17:22 +0000]
bcma: change delays to follow timers-howto guide

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

6 years agoMerge branch 'for-linville' of git://git.kernel.org/pub/scm/linux/kernel/git/luca...
John W. Linville [Fri, 28 Sep 2012 17:49:21 +0000]
Merge branch 'for-linville' of git://git./linux/kernel/git/luca/wl12xx

6 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 28 Sep 2012 17:09:33 +0000]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David S Miller:

 1) Netfilter xt_limit module can use uninitialized rules, from Jan
    Engelhardt.

 2) Wei Yongjun has found several more spots where error pointers were
    treated as NULL/non-NULL and vice versa.

 3) bnx2x was converted to pci_io{,un}map() but one remaining plain
    iounmap() got missed.  From Neil Horman.

 4) Due to a fence-post type error in initialization of inetpeer entries
    (which is where we store the ICMP rate limiting information), we can
    erroneously drop ICMPs if the inetpeer was created right around when
    jiffies wraps.

    Fix from Nicolas Dichtel.

 5) smsc75xx resume fix from Steve Glendinnig.

 6) LAN87xx smsc chips need an explicit hardware init, from Marek Vasut.

 7) qlcnic uses msleep() with locks held, fix from Narendra K.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  netdev: octeon: fix return value check in octeon_mgmt_init_phy()
  inetpeer: fix token initialization
  qlcnic: Fix scheduling while atomic bug
  bnx2: Clean up remaining iounmap
  net: phy: smsc: Implement PHY config_init for LAN87xx
  smsc75xx: fix resume after device reset
  netdev: pasemi: fix return value check in pasemi_mac_phy_init()
  team: fix return value check
  l2tp: fix return value check
  netfilter: xt_limit: have r->cost != 0 case work

6 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Fri, 28 Sep 2012 17:02:53 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull vfs fixes from Al Viro:
 "A couple of fixes; one for automount/lazy umount race, another a
  classic "we don't protect the refcount transition to zero with the
  lock that protects looking for object in hash" kind of crap in lockd."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  close the race in nlmsvc_free_block()
  do_add_mount()/umount -l races

6 years agoMerge branch 'for-linus-3.6-rc-final' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 28 Sep 2012 17:00:52 +0000]
Merge branch 'for-linus-3.6-rc-final' of git://git./linux/kernel/git/rw/uml

Pull UML fixes from Richard Weinberger.

* 'for-linus-3.6-rc-final' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: Preinclude include/linux/kern_levels.h
  um: Fix IPC on um
  um: kill thread->forking
  um: let signal_delivered() do SIGTRAP on singlestepping into handler
  um: don't leak floating point state and segment registers on execve()
  um: take cleaning singlestep to start_thread()