Staging: rtl8192e: Use skb_tail_pointer

rtl8192e uses skb->tail directly. This patch uses the tail pointer macros
instead.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rt2860: Use skb_tail_pointer

rt2860 uses skb->tail directly. This patch uses the tail pointer macros
instead.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ramzswap: Free memory when create_device is failed

If create_device is failed, it can't free gendisk and request_queue of
preceding devices. It cause memory leak.

This patch fixes it.

Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
Acked-by: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: make i2c device id constant

The id_table field of the struct i2c_driver is constant in <linux/i2c.h>
so it is worth to make the initialization data also constant.

The semantic match that finds this kind of pattern is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r@
disable decl_init,const_decl_init;
identifier I1, I2, x;
@@
struct I1 {
  ...
  const struct I2 *x;
  ...
};
@s@
identifier r.I1, y;
identifier r.x, E;
@@
struct I1 y = {
  .x = E,
};
@c@
identifier r.I2;
identifier s.E;
@@
const struct I2 E[] = ... ;
@depends on !c@
identifier r.I2;
identifier s.E;
@@
+ const
struct I2 E[] = ...;
// </smpl>

Signed-off-by: Németh Márton <nm127@freemail.hu>
Cc: Julia Lawall <julia@diku.dk>
Cc: cocci@diku.dk
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: make USB device id constant

The id_table field of the struct usb_device_id is constant in <linux/usb.h>
so it is worth to make the initialization data also constant.

The semantic match that finds this kind of pattern is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r@
disable decl_init,const_decl_init;
identifier I1, I2, x;
@@
struct I1 {
  ...
  const struct I2 *x;
  ...
};
@s@
identifier r.I1, y;
identifier r.x, E;
@@
struct I1 y = {
  .x = E,
};
@c@
identifier r.I2;
identifier s.E;
@@
const struct I2 E[] = ... ;
@depends on !c@
identifier r.I2;
identifier s.E;
@@
+ const
struct I2 E[] = ...;
// </smpl>

Signed-off-by: Németh Márton <nm127@freemail.hu>
Cc: Julia Lawall <julia@diku.dk>
Cc: cocci@diku.dk
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: make PCI device id constant

The id_table field of the struct pci_driver is constant in <linux/pci.h>
so it is worth to make the initialization data also constant.

The semantic match that finds this kind of pattern is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r@
disable decl_init,const_decl_init;
identifier I1, I2, x;
@@
struct I1 {
  ...
  const struct I2 *x;
  ...
};
@s@
identifier r.I1, y;
identifier r.x, E;
@@
struct I1 y = {
  .x = E,
};
@c@
identifier r.I2;
identifier s.E;
@@
const struct I2 E[] = ... ;
@depends on !c@
identifier r.I2;
identifier s.E;
@@
+ const
struct I2 E[] = ...;
// </smpl>

Signed-off-by: Németh Márton <nm127@freemail.hu>
Cc: Julia Lawall <julia@diku.dk>
Cc: cocci@diku.dk
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: iio: Fix incorrect existence check for a shared event pointer.

A second smatch detected error. First part fixes in a typo
in the comment directly above that I noticed whilst trying
to remember what this code actually does. Second part is
the actual fix.  I'm fairly amazed this one never caused
trouble in testing as it is in one of the most common paths.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: iio: Ensure mutex is correctly unlocked in __iio_push_event

This error was picked up by running the smatch static
checker over all the IIO subsytem.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: zfTkipInit(): increment another pointer

Increment the pointer rather than its value.
These appear to be logic errors.

Thanks to Dan Carpenter for the first hunk of this change.

Cc: Dan Carpenter <error27@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: Fix ZM_SEQ_DEBUG macro for no-debug case

tcp_seq is only initialised in case where ZM_ENABLE_PERFORMANCE_EVALUATION
is defined. So move the call to ZM_SEQ_DEBUG() and the decleration of
tcp_seq in there too.  This allows ZM_SEQ_DEBUG() to be removed from the
non-ZM_ENABLE_PERFORMANCE_EVALUATION case in the header file.

This resolves several compile warnings for the
non-ZM_ENABLE_PERFORMANCE_EVALUATION case.
However, the ZM_ENABLE_PERFORMANCE_EVALUATION
case seems to be completely broken.

$ gcc (Debian 4.4.2-8) 4.4.2
Copyright (C) 2009 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ make
...
drivers/staging/otus/80211core/cagg.c: In function 'zfAggRxEnabled':
drivers/staging/otus/80211core/cagg.c:1872: warning: left-hand operand of
comma expression has no effect
drivers/staging/otus/80211core/cagg.c:1872: warning: left-hand operand of
comma expression has no effect
drivers/staging/otus/80211core/cagg.c:1872: warning: statement with no
effect
...

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: remove unused constants

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: remove unused variables

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: dont mix declarations and code

$ gcc (Debian 4.4.2-8) 4.4.2
Copyright (C) 2009 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ make
...
drivers/staging/otus/ioctl.c: In function 'usbdrv_wpa_ioctl':
drivers/staging/otus/ioctl.c:2253: warning: ISO C90 forbids mixed declarations and code
...

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: crystalhd: trim register defines

We don't actually need most of the register defines to build,
and most of the ones we don't need aren't currently interesting.

We'll leave a full copy of all of them in libcrystalhd's source,
and only include what we need and/or think might be interesting
in the driver.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: crystalhd: add missing fixes for userspace lib build

I somehow managed to not actually include these two fixes in the submission
that was committed to the staging tree. libcrystalhd should eventually be
built against the kernel-provided header, and needs the stdint.h include. The
VOID bit is to keep things in sync with the Mac OS X driver and library that
Scott Davilla is also working on.

Signed-off-by: Scott Davilla <davilla@4pi.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: wlan-ng: Add select WEXT_PRIV to Kconfig to prevent build failure

Without WEXT_PRIV set the p80211wext.c fails to build due to unknown fields in
the iw_handler_def struct.
Those fields are enclosed in WEXT_PRIV conditionals in the prototype
of iw_handler_def in include/net/iw_handler.h

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Simon Horman <horms@verge.net.au>
Cc: stable <stable@kernel.org> [2.6.33 only]
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192u: Add select WEXT_PRIV to Kconfig to prevent build failure

Without WEXT_PRIV set the driver fails to build due to unknown fields in
the iw_handler_def struct.
Those fields are enclosed in WEXT_PRIV conditionals in the prototype
of iw_handler_def in include/net/iw_handler.h

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: Add select WEXT_PRIV to Kconfig to prevent build failure

Without WEXT_PRIV set the driver fails to build due to unknown fields in
the iw_handler_def struct.
Those fields are enclosed in WEXT_PRIV conditionals in the prototype
of iw_handler_def in include/net/iw_handler.h

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: asus_oled: Add NULL test for kmalloc

This patch adds a NULL test to check wether kmalloc was successful or
not.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: cx25821: avoid NULL pointer dereference

Assign dev only if fh is not NULL.

Signed-off-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: v56655: add missing parentheses (in comment)

`!' has a higher precedence than `&' so parentheses are required.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Update the README for mainline usage.

The module now only compiles inside mainline, so remove the comments
about different versions of the kernel it can be used with. Also
update comments about building with debug enabled and how to use the
VIS data now that it no longer natively outputs dot or JSON.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: Fixed pohmelfs regression because of per-bdi writeback.

Since pohmelfs isn't tied to a single block device, it needs to setup a
backing dev like nfs/btrfs/etc do.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: pohmelfs: avoid null dereference

err_out_put is for when the netfs_trans_alloc() succeeded.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: pohmelfs: Remove uses of NIPQUAD, use %pI4

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: sep: Fix use of legacy ioctl fop

SEP doesn't need lock_kernel.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: sep: Fix build problems from header changes

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: 80211core: Fix &&/|| confusion

This always evaluates to true.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rar: Fix all coding style issues.

This fixes all coding style issues and some spelling mistakes.

Signed-off-by: Benjamin Adolphi <b.adolphi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su: buffer overflow in r8192U_core.c

There was a buffer overflow in the original code.  rf_path was 2 and it should
have been only 0 or 1.

I don't have the hardware for this, so I can't test it.

Looking at the code, there are two almost identical sections for updating the
hal variables.  The first one was clearly wrong and had the array overflow as
well.  The second one looked correct.  I decided to use the second section as
is except for whitespace changes.

The differences between the two original sections:
1)  The second one had more debug output.
2)  The second one looped over rf_path instead of corrupting data.
3)  The second one had these additional assigments.
                      if (rf_path == 0) {
                              priv->TxPowerLevelOFDM24G[i] = priv->RfTxPwrLevelOfdm1T[rf_path][i] ;
                              priv->TxPowerLevelCCK[i] = priv->RfTxPwrLevelCck[rf_path][i];
                      }

Signed-off-by: Dan Carpenter <error27@gmail.com>
Cc: Jerry chuang <wlanfae@realtek.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6655/device_main.c: use %pM to shown MAC address

Trivial, use the %pM kernel extension to display the MAC address.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Forest Bond <forest@alittletooquiet.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

r8192U_core.c: use %pM to shown MAC address

Trivial, use the %pM kernel extension to display the MAC address.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: r8192E_core.c: use %pM to shown MAC address

Trivial, use the %pM kernel extension to display the MAC address.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: et131x_netdev.c: use %pM to shown MAC address

Trivial, use the %pM kernel extension to display the MAC address.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Olaf Hartmann <o.hartmann@telovital.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: off by one in usbdrvwext_siwessid()

A 33 char ESSID is too long and it could cause a buffer overflow
a couple lines below when we put a NULL terminator on the end.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cx25821: off by one in cx25821_enum_input()

The INPUT(n) macro indexes an array of size 2.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cx25821: fix double unlock in medusa_video_init()

medusa_set_videostandard() takes the lock but it always drops it before
returning.

This was found with a static checker and compile tested only.  :/

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: reorder check in pcmmio_attach()

The check for dio_num_asics is used to determine if there is more than 1 irq.
If it is false then irq[1] is past the end of the array.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: off by one in doDevConfig()

Changed > to >= to avoid array overflow.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ramzswap: Minor spelling fixes

Also removed an extra semicolon.

Signed-off-by: Cyp <cyp561@gmail.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: remove compat code

Remove compatibility code as this is not an older version of the kernel.

Signed-off-by: Simon Horman <horms@verge.net.au>
Cc: Scott Smedley <ss@aao.gov.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su, rtl8192u: use min_t() in store_debug_level()

sizeof() returns a size_t but the other types involved
are unsigned long, so using min() results in a warning.

As sizeof() is called on an 11 character buffer defined
immediately above unsigned long is obviously wide enough
for the result.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192e: remove some functions from the __exit section

ieee80211_crypto_tkip_exit(), ieee80211_crypto_deinit() and
ieee80211_crypto_ccmp_exit() are called by ieee80211_rtl_init()
which are in section __init, so they can't be in section __exit.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192e: print the elements of tx_pn and rx_pn not the arrays themselves

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rt2870: Remove unnecessary forward declarations

Remove unnecessary forward declaration of rtusb_probe() and rtusb_disconnect()

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rt2870: rtusb_probe() should be in section __devinit

$ make CONFIG_DEBUG_SECTION_MISMATCH=y
WARNING: drivers/staging/rt2870/rt2870sta.o(.text+0x2f4c5): Section
mismatch in reference from the function rtusb_probe() to the function .devinit.text:rt2870_probe()
The function rtusb_probe() references
the function __devinit rt2870_probe().
This is often because rtusb_probe lacks a __devinit
annotation or the annotation of rt2870_probe is wrong.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: DT3155: Use pci_get_device()

The use of pci_find_device() is deprecated.

Signed-off-by: Simon Horman <horms@verge.net.au>
Cc: Scott Smedley <ss@aao.gov.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: match on DMI values to know if we should run.

The HV core mucks around with specific irqs and other low-level stuff
and takes forever to determine that it really shouldn't be running on a
machine.  So instead, trigger off of the DMI system information and
error out much sooner.  This also allows the module loading tools to
recognize that this code should be loaded on this type of system.

Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: add a pci device table

This allows the HV core to be properly found and autoloaded
by the system tools.

It uses the Microsoft virtual VGA device to trigger this.

Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix brace coding style issue in Channel.c

This is a patch to the Channel.c file that fixes up a brace
warning found by checkpatch.pl

Signed-off-by: Chris Nicholson <chris.nicholson@cnick.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Rename struct device_context and re-arrange the fields inside

Rename struct device_context and re-arrange the fields inside.

Rename struct device_context to struct vm_device, and move struct device
field to the end according to Document/driver-model standard.

Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Add proper versioning to HV drivers

Provide proper versioning information for all HV drivers.

With removal of build time/date/and Minor number as requested by Greg KH

Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyang@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: fix various coding style issues in RingBuffer.c

This is a patch to the RingBuffer.c file that corrects various coding style
warnings and errors found by checkpatch.pl

[ The real solution here is to get rid of this file entirely, and use the
kernel's internal ring buffer api, but until then, make these changes so as to
make checkpatch.pl happy, and keep others from continuously sending this type
of patch. - gkh]

Signed-off-by: Craig Bartlett <c-bartlett@hotmail.co.uk>
Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Nicolas Palix <npalix@diku.dk>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove xen legacy code and check for Hyper-V

Removed legacy XEN layer from hypervisor setup, and made sure only
Hyper-V is Is a valid hypervisor to run on.

Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove unneeded OnClose callback

This callback only calls one function, so just call the function
instead, no need for indirection at all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove unneeded OnOpen callback

This callback only calls one function, so just call the function
instead, no need for indirection at all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove unneeded OnHostReset callback

This callback only calls one function, so just call the function
instead, no need for indirection at all.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: storvsc_drv: move unused functions

Now that the callback pointer was removed, we can remove
the code itself, as it is never used.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove unneeded OnHostRescan callback

This callback was never called, so delete the thing.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Dont deactivate aggregation on wrong input

A non-integer changes the aggregation mode. Therefore this patch changes
the behaviour to explicitly check strict_strtoul()'s return code.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Remove compat.h

Since we are now part of mainline, we don't need compat.h to allow
building of the module with old versions of the kernel.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Use printk(%pM) for MAC addresses

printk() since kernel version 2.6.29 has supported printing MAC
addresses directly, as an extension to the %p processing. This patch
makes use of this for printk() and bat_dbg(). This will remove the
overhead of using addr_to_string() which is normally never actually
output.

Fixed a typo found by Gus Wirth.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Allow the MAC address to be set

Some embedded devices have very limited sources of entropy for the
random number generator. It has been observed that the random MAC
address on the interface bat0 is not always random. When testing with
a collection of identical hardware, sometimes the bat0 device the same
MAC address on multiple devices, causing mayhem. This patch allows the
MAC address to be set by the user.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: receive packets directly using skbs

This patch removes the (ugly and racy) packet receiving thread and the
kernel socket usage. Instead, packets are received directly by registering
the ethernet type and handling skbs instead of self-allocated buffers.

Some consequences and comments:

 * we don't copy the payload data when forwarding/sending/receiving data
   anymore. This should boost performance.
 * packets from/to different interfaces can be (theoretically) processed
   simultaneously. Only the big originator hash lock might be in the way.
 * no more polling or sleeping/wakeup/scheduling issues when receiving
   packets
 * this might introduce new race conditions.
 * aggregation and vis code still use packet buffers and are not (yet)
   converted.
 * all spinlocks were converted to irqsave/restore versions to solve
   some lifelock issues when preempted. This might be overkill, some
   of these locks might be reverted later.
 * skb copies are only done if neccesary to avoid overhead

performance differences:

 * we made some "benchmarks" with intel laptops.
 * bandwidth on Gigabit Ethernet increased from ~500 MBit/s to ~920 MBit/s
 * ping latency decresed from ~2ms to ~0.2 ms

I did some tests on my 9 node qemu environment and could confirm that
usual sending/receiving, forwarding, vis, batctl ping etc works.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Acked-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Acked-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: check all kmalloc()s

there are some kmallocs left which are not checked whether they succeeds or
not, which might lead to corrupted data structures if the system memory is
full. This patch should clean up the remaining unchecked kmalloc()s.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: splitting /proc vis file into vis_server and vis_data

The /proc vis file was used to enable/disable the vis server and to output
the vis data at the same time. This behaviour was confusing and lacked a
proper method to display the current vis server status.
This patch seperates the 2 functionalities:
* use vis_server to enable/disable the vis server and to retrieve its status
* use vis_data to retrieve the vis raw data (if the server is enabled)

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: remove obsoleted vis_format /proc file

batman-adv used to export the vis data using different formats that were
switchable via /proc/net/batman-adv/vis_format. The various formats moved
to user space and rendered this configuration switch useless.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: moving vis output formats out of the kernel

The batman-adv kernel module is able to output visualization data using the
dot draw or JSON format. This patch transforms the output into a generic
format (called vis raw). User space tool may convert the raw data to support
a variety of formats without the need of modifying the kernel module.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: fix minor orig table layout bug

The originator table contained a TAB instead of a space which broke
the layout as well as the batctl parser.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Fixes rounding issues in vis.c

This patches fixes two rounding issues in vis.c for sending and
purging vis packets. Before, the timers and timeouts always got
rounded down to seconds, though we want a precision in
milliseconds.

This also fixes a kernel panic that occures when lowering the
timer for sending vis packets (vis_interval) to less than 1000ms
in main.c manually.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Acked-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Use forw_bcast_list_lock always with disabled interrupts

forw_bcast_list_lock is spin_locked in both process and softirq context.
SoftIRQ calls the spinlock with disabled IRQ and normal process context
with enabled IRQs.

When process context is inside an spin_locked area protected by
forw_bcast_list_lock and gets interrupted by an IRQ, it could happen
that something tries to lock forw_bcast_list_lock again in SoftIRQ
context. It cannot proceed further since the lock is already taken
somewhere else, but no reschedule will happen inside the SoftIRQ
context. This leads to an complete kernel hang without any chance of
resurrection.

All functions called in process context must disable IRQs when they try
to get get that lock to to prevent any reschedule due to IRQs.

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Acked-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: consistent spelling of "neighbors"

we have written "neighbors", "neighbours" and bad spelled versions of this
word, this patch should make it consistent.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: initialize static hash iterators

instead of dynamically registering hash iterators, calling functions are
changed to register the iterator objects statically. The two advantages are:

 * no memory leaks when aborting from hash_iterate()
 * no calls to kmalloc/kfree, therefore a little faster/safer

Tested with 9 QEMU instances, no obvious regression found.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Replace KERN_DEBUG with bat_dbg

Ameya Palande requested we replace printk(KERN_DEBUG "") by pr_debug()
I decided it was better to use our debug macro.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: update README for latest kernel

state in the README that we also (compile) support 2.6.32

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Remove NULL pointer check

It is safe to call kfree(NULL) which makes this extra check unneeded. It
was found using checkpatch.pl from linux-2.6

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rt2860: test off by one in RtmpAsicSendCommandToMcu()

`i' reaches 101 after the loop, so if it was 100 then it succeeded in
the last iteration. This is probably unlikely to cause problems.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192u: ieee80211: add missing parentheses

not(!) has a higher precedence than bit and(&).

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: add Broadcom Crystal HD driver

This patch supersedes the earlier ones sent by Manu Abraham to add
the Broadcom Crystal HD driver to the staging tree, per discussion
with him about it. I've been working with Broadcom's Naren Sankar
on this driver for a number of months, and had already talked Naren
about submitting this on Broadcom's behalf, didn't expect anyone
else to jump on submitting it as quickly as Manu did. ;)

This version is a one-shot deal, incorporating the original driver,
Manu's coding style clean-ups, udev device creation support from
Edgar 'gimli' Hucek, and a number of other small tweaks from myself
and Scott Davilla, the other individual who has been working closely
on this code with Naren and I.

I've tested this iteration of the code lightly on a mini pci-e board
in a ThinkPad T61p running x86_64 Fedora 12, with the expected results,
and will test further on other systems with other variants of the card
(I have three varieties of this device currently in hand). Scott has
also tested on assorted primarily i686 varieties of Ubuntu, and Naren
has tested with both Fedora and Ubuntu, iirc.

Note: only the 70012 is currently supported by this driver, 70015
support will follow later. Also note that Blu-Ray support isn't
enabled (at the firmware level), due to misc fun related to the
BD encryption scheme, DRM, etc. :\

I *do* have a git tree containing the driver, lib, gst plugin and
firmware that I'm working from at the moment[*], as there are inter-
dependencies between the driver and lib, and the driver can be used
with kernels going a ways back (I've only tested back to 2.6.18 as
it exists in Red Hat Enterprise Linux 5). I'm exporting from there,
into a linux-next tree, then generating patches from there. The goal
is to feed everything upstream as quickly as possible, but there are
users who want this code for earlier kernels too...

The firmware will be submitted for inclusion in dwmw2's linux-firmware
tree once there is a suitable redistribution-no-modification type of
license on it (I believe Naren is working with Broadcom legal to get
that in place).

Changelog from initial Broadcom release to here:

commit d20475d444610c5683d09e63f707f5bb22359062
Author: Jarod Wilson <jarod@redhat.com>
Date:   Mon Jan 4 13:55:16 2010 -0500

    include: lib doesn't build w/o the removed stdint include

    So add it back...

Signed-off-by: Jarod Wilson <jarod@redhat.com>
commit c181070a330530b792d2b80e3ec6ab12a5a57394
Author: Scott Davilla <davilla@4pi.com>
Date:   Mon Jan 4 13:38:37 2010 -0500

    include: don't define VOID if its already defined

Signed-off-by: Scott Davilla <davilla@4pi.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
commit 33d8a2b691e81212e398f53770578d79650bf0bc
Author: Jarod Wilson <jarod@redhat.com>
Date:   Mon Jan 4 13:12:10 2010 -0500

    driver: create crystalhd device using udev

    Based on:
    http://sourceforge.net/apps/trac/archvdr/browser/trunk/archvdr/crystalhd/use_udev.patch

Signed-off-by: Edgar ( gimli ) Hucek <ebsi4711 at gmail dot com>
    Formatting tweaks, error-handling path fixups and any bugs added by Jarod.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
commit c44c64dea5537814796fcbe2d9db0209383c78b9
Author: Manu Abraham <abraham.manu@gmail.com>
Date:   Mon Jan 4 10:32:47 2010 -0500

    crystalhd: coding style cleanups

Signed-off-by: Manu Abraham <abraham.manu@gmail.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
commit cffa6da7467ff697a656d1dfff54bb0513a053dc
Author: Jarod Wilson <jarod@redhat.com>
Date:   Mon Jan 4 10:17:27 2010 -0500

    crystalhd: run dos2unix over everything, this is linux source...

Signed-off-by: Jarod Wilson <jarod@redhat.com>
commit 7fa38a282db7af5a5746055f7c6cef8a9b8ee138
Author: Jarod Wilson <jarod@redhat.com>
Date:   Mon Jan 4 10:02:33 2010 -0500

    crystalhd: initial import of released Broadcom code

    Straight import of:
    http://www.broadcom.com/docs/support/crystalhd/crystalhd_linux_20091229.zip

    Unfortunately, we're unable to publicly publish all the history that got
    us from the initial internal code to what was released here, but such is
    life, we can just be happy we've got this open-sourced now. :)

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Naren Sankar <nsankar@broadcom.com>
Signed-off-by: Scott Davilla <davilla@4pi.com>
Signed-off-by: Manu Abraham <abraham.manu@gmail.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: altpciechdma: remove driver

No one seems to be able to maintain this, or merge it into mainline, so
remove it.

Acked-by: Leon Woestenberg <leon@sidebranch.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: p9auth: remove driver from tree

No one seems to be maintaining this anymore, and it is not on any
track to be merged to mainline.

Cc: Ashwin Ganti <ashwin.ganti@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: remove the b3dfg driver

It has no users, and no developers to maintain it to get
it merged into mainline.

So sad.

Cc: Daniel Drake <ddrake@brontes3d.com>
Cc: Justin Bronder <jsbronder@brontes3d.com>
Cc: Duane Griffin <duaneg@dghda.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: coding style cleanups for the .h files

This cleans up some of the coding style issues in the .h files.

More remains to be done.

Cc: Scott Smedley <ss@aao.gov.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: coding style cleanups for allocator code

This fixes up the worst of the coding style errors for the
allocator code.

Cc: Scott Smedley <ss@aao.gov.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: wlan-ng: initialise mibitem

$ gcc --version

gcc (Debian 4.4.2-5) 4.4.2
Copyright (C) 2009 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ make
...
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_dorequest’:
drivers/staging/wlan-ng/p80211wext.c:139: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:139: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwretry’:
drivers/staging/wlan-ng/p80211wext.c:967: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:967: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:967: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwretry’:
drivers/staging/wlan-ng/p80211wext.c:1057: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:1057: warning: ‘mibitem.status’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwtxpow’:
drivers/staging/wlan-ng/p80211wext.c:1149: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:1149: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:1149: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwtxpow’:
drivers/staging/wlan-ng/p80211wext.c:1123: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:1123: warning: ‘mibitem.status’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwfrag’:
drivers/staging/wlan-ng/p80211wext.c:891: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:891: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:891: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwfrag’:
drivers/staging/wlan-ng/p80211wext.c:933: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:933: warning: ‘mibitem.status’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwrts’:
drivers/staging/wlan-ng/p80211wext.c:826: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:826: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:826: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwrts’:
drivers/staging/wlan-ng/p80211wext.c:866: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:866: warning: ‘mibitem.status’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwrate’:
drivers/staging/wlan-ng/p80211wext.c:775: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:775: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:775: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_giwfreq’:
drivers/staging/wlan-ng/p80211wext.c:273: warning: ‘mibitem.len’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:273: warning: ‘mibitem.status’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:273: warning: ‘mibitem.data’ is used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwfreq’:
drivers/staging/wlan-ng/p80211wext.c:320: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c: In function ‘p80211wext_siwmode’:
drivers/staging/wlan-ng/p80211wext.c:401: warning: ‘mibitem.len’ may be used uninitialized in this function
drivers/staging/wlan-ng/p80211wext.c:401: warning: ‘mibitem.status’ may be used uninitialized in this function
...

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: prism2-usb: Build should select WIRELESS_EXT

The prevailing idiom is to select WIRELESS_EXT not depend on it.

Depending leaves this driver in a situation where it can only
be built if another driver that selects WIRELESS_EXT has been enabled.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su: Use z format qualifier for output of sizeof()

$ gcc --version
gcc (Debian 4.4.2-5) 4.4.2
Copyright (C) 2009 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ make
...
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c: In function ‘ieee80211_rx_ADDBAReq’:
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c:342: warning: format ‘%ld’ expects type ‘long int’, but argument 3 has type ‘unsigned int’
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c: In function ‘ieee80211_rx_ADDBARsp’:
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c:443: warning: format ‘%ld’ expects type ‘long int’, but argument 3 has type ‘unsigned int’
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c: In function ‘ieee80211_rx_DELBA’:
drivers/staging/rtl8192su/ieee80211/rtl819x_BAProc.c:573: warning: format ‘%ld’ expects type ‘long int’, but argument 3 has type ‘unsigned int’
...

Also some style fixes for these lines:
+ Fix excessively long lines
+ Remove leading space before struct
+ Remove unnecessary parentheses

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su: Build should select WIRELESS_EXT

The prevailing idiom is to select WIRELESS_EXT not depend on it.

Depending leaves this driver in a situation where it can only
be built if another driver that selects WIRELESS_EXT has been enabled.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: s5k3e2fx.c: simplify complexity by factoring

the code was looping, setting s_move[i] to the following calculations

if (actual_step>= 0)
         s_move[i] = ((((i + 1) * gain + 0x200) - (i * gain + 0x200)) / 0x400);
else
         s_move[i] = ((((i + 1) * gain - 0x200) - (i * gain - 0x200)) / 0x400);

but, this code reduces to the expression
s_move[i] = gain>>  10;

The reason for the complexity was to generate a step function with
integer division and rounding to land on specific values. But these calculations
can be simplified to the following code:

gain = ((actual_step<<  10) / 5)>>  10;
for (i = 0; i<= 4; i++)
s_move[i] = gain;

Signed-off-by: Justin Madru<jdm64@gawab.com>
Reviewed-by: Ray Lee<ray-lk@madrabbit.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Remove unused typedefs MLME_DEAUTHREQ_PARA and MLME_DISASSOCREQ_PARA

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Remove unused typedef RXLAYER1

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Remove unused typedef TXRETRY_REC

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Remove unused typedef CHAN_LIST

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Convert typedef ChanInfo to struct chan_info

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Remove unused typedef _EVENTLOG

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Convert typedef RadioOff to struct radio_off

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: w35und: Convert typedef LOCAL_PARA to struct wb_local_para

Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: Fix misspelling of "invocation" in comment.

A comment misspells "invocation"; this fixes it. No code changes.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: Fix misspelling of "should" and "shouldn't" in comments.

Some comments misspell "should" or "shouldn't"; this fixes them. No code changes.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: Fix misspelling of "successful" and variants.

Some comments and one message misspell "successful" or variants of
the word; this fixes them. No change in functionality.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>