Staging: rt3070: remove dead code

While merging rt30[79]0 support into rt28[67]0, drivers/staging/rt3070/md4.h
has been forgotten, while it isn't referenced from the remaining drivers
anymore; remove it.

Signed-off-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Acked-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Matti Lammi <mattij.lammi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: RT2860 Fixed all errors in mlme.h and all but one in rtmp.h.

I have fixed all errors in mlme.h and fixed all bar one in rtmp.h.
I think I can fix rtmp.h too by moving a brace up two lines, I just dont know if it will work.
If someone can confirm it will work I will change it.

Signed-off-by: Neil Munro <neilmunro@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: increase VBIOS_VER_MAX_LENGTH to 5

VBIOS_VER_MAX_LENGTH was set to 4, but the value "0.84" is strcpy'd
into it.  That value is 5 long including the trailing NULL.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: remove externs in XGI_accel.c

Remove 2 extern variables definations from XGI_accel.c that were
defined in other .h files.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: remove __KERNEL__ #ifdef in XGIfb.h

The code inside this #ifdef declared a couple extern functions that
weren't used anywhere, so remove the whole #ifdef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: remove XGI.h

This was not included anywhere and just defined a debugging macro.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove XGI_TV_PLUG typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove typedefs in XGIfb.h

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove typedefs in vb_struct.h

This removes all the remaining typedefs in vb_struct.h

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove XGI_HW_DEVICE_INFO and PXGI_HW_DEVICE_INFO typedefs

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove XGI_DSReg and PXGI_DSReg typedefs

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove XGI_LCD_TYPE and XGI_VB_CHIP_TYPE typedefs

There is no need for these to be typedefs as a simple enum will do.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove SHORT, UCHAR, and LONG typedef

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove USHORT, ULONG, BOOLEAN, and VOID typedefs

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove unused code

Remove code that was removed using #if 0

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove PUSHORT, PLONGU, and PVOID typedefs

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove PUCHAR typedef

PUCHAR was unsigned char *, use that instead

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove defines for TRUE, FALSE, and NULL

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove port macros in osdef.h

The port related macros in osdef.h are replaced with calls to out?()
and in?().  This removes the last macros defined in osdef.h, so this
file is no longer needed.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove XGI_SetMemory and XGI_MemoryCopy

These were just macros for memset and memcpy, so use those instead.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove use of LINUX_KERNEL define

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove use of LINUX_XF86 define

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove use of TC define

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove use of WINCE_HEADER define

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: Remove code for WIN2000

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: remove use of LINUX_VERSION_CODE

Remove ifdefs that check LINUX_VERSION_CODE as this is
not needed in the staging tree.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: xgifb: use %lx instead of %x in printk format

This fixes compiler warning  format '%x' expects type 'unsigned int',
but argument 2 has type 'long unsigned int'

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: record route for ICMP messages

The standard layer 3 ping utility can use the record route (RR) option
of IP to collect route data for sent ping messages (ping -R). This
patch introduces comparable functionality for batman-adv ICMP messages.

The patch adds a second batman ICMP packet format (icmp_packet_rr) such
that up to 17 MAC addresses can be recorded (sufficient for up to 8
hops per direction). When no RR is wanted, the old icmp_packet without
the RR overhead can be sent.

Signed-off-by: Daniel Seither <post@tiwoc.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: bonding and interface alternating

This patch adds interface alternating to the new bonding feature. By
default, we now try to avoid forwarding packets on the receiving
interface, instead choosing alternative interfaces. This feature
works only on nodes which have multiple interfaces connected to the
mesh. This approach should reduce problems of the half-duplex nature
of WiFi Hardware and thus increase performance.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Acked-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Add bonding functionality

This patch introduces bonding functionality to batman-advanced, targeted
for the 0.3 release. As we are able to route the payload traffic as we
want, we may use multiple interfaces on multihomed hosts to transfer data
to achieve higher bandwidth. This can be considered as "light Multi Path
Routing" for single hop connections.

To detect which interfaces of a peer node belong to the same host, a
new flag PRIMARIES_FIRST_HOP is introduced. This flag is set on the first hop
of OGMs of the primary (first) interface, which is broadcasted on all
interfaces. When receiving such an OGM, we can learn which interfaces
belong to the same host (by assigning them to the primary originator).

Bonding works by sending packets in a round-robin fashion to the available
interfaces of a neighbor host, if multiple interfaces are available. The
neighbor interfaces should be almost equally good to reach.

To avoid interferences (i.e. sending on the same channel), only neighbor
interfaces with different mac addresses and different outgoing interfaces
are considered as candidates.

Bonding is deactivated by default, and can be activated by

echo 1 > /sys/class/net/bat0/mesh/bonding

for each individual node.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Rework on top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: 32bit sequence number and TTL for broadcasts

This patch changes the sequence number range from 8 or 16 bit to 32 bit.
This should avoid problems with the sequence number sliding window algorithm
which we had seen in the past for broadcast floods or malicious packet
injections. We can not assure 100% security with this patch, but it is quite
an improvement over the old 16 bit sequence numbers:

 * expected window size can be increased (4096 -> 65536)
 * 64k packets in the right order would now be needed to cause a loop,
   which seems practically impossible.

Furthermore, a TTL field has been added to the broadcast packet type, just to
make sure.

These changes required to increase the compatibility level once again.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Change atomic64_* back to atomic_*, Rework on
top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Add release information for version 2010.0.0

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

Staging: batman-adv: permit setting ogm interval to JITTER*2

When trying to set the originator interval to 40ms, we are asked to set
it to a minimum value of 40ms. This patch permits setting an
originator interval of JITTER*2 (40ms by default) now.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Ignore debugfs on kernels without debugfs support

All code for debugfs is ignored when the creation of the batman-adv root
directory in debugfs fails. It must also be ignored when the
debugfs_create_dir tells us that debugfs is not implemented to prevent
possible crashes in the future.

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Mark locally used symbols as static

Functions and variables which are used only inside one object file can
be declared as static. This helped to find unused functions/variables

 * mainIfAddr_default
 * main_if_was_up

and functions with declarations but missing definitions

 * hash_debug
 * orig_find
 * send_own_packet_work

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Convert MAC_FMT to %pM

Remove the last uses of MAC_FMT

Signed-off-by: Joe Perches <joe@perches.com>
[sven.eckelmann@gmx.de: Adapted for current batman-adv version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: remove superfluous hint to "translation table"

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: convert all sysfs files to single value files

This patch removes the extra usage output which came when
the sysfs files were read.

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

Staging: batman-adv: Move tables from sysfs to debugfs

Files which represent more than a single attribute aren't allowed in
sysfs. As we have some files which aren't essential and are lists or
tables aggregated from data from different places inside batman-adv, we
must place them in a filesystem without such a restriction.

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Move device for icmp injection to debugfs

batctl uses /dev/batman-adv to send special batman-adv icmp packets to
other nodes in the mesh. To get it working with multiple batX devices we
must ensure that every mesh device can have their own socket which is
used to inject those packets in exactly one mesh.

The current implementation still doesn't allow to use complete separated
meshes as we rely on structures which are not part of the private data
of a batman device.

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: fix function prototype

In today linux-next I got a compile warning in staging/batman-adv.

This is due a struct bin_attribute read function prototype change and the driver was not updated.

This patch solves the issue

Signed-off-by: Javier Martinez Canillas <martinez.javier@gmail.com>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: return -EFAULT on copy_to_user errors

copy_to_user() returns the number of bites remaining but we want to
return a negative error code here.

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

Staging: batman-adv: convert vis_interval into define

vis_interval does not neccesarily needs to be a variable, as there is
no way to change it anyway (and probably no need to). We can therefore
remove yet another global variable.

Thanks Marek for pointing this out.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Rework on top of current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Add information about batman-adv sysfs entries

Add documents about new sysfs entries in
 * /sys/class/net/<iface>/batman-adv/mesh_iface
 * /sys/class/net/<mesh_iface>/mesh/vis_mode

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

Staging: batman-adv: Adding netfilter-bridge hooks

batman-adv is receiving and sending the packets of its own ether type
on a very early/low level. Therefore we need to add explicit hooks to
give netfilter/ebtables a chance to filter them.

Reported-by: Antonio Quartulli <ordex@ritirata.org>
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: remove redundant struct declaration

The hardif_attr and the bat_attr struct share the same attributes,
hence it is not necessary to specify 2 different structs.

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

Staging: MSM: remove duplicated #include

Remove duplicated #include('s) in
  drivers/staging/msm/mddihost.h
  drivers/staging/msm/msm_fb_def.h
  drivers/staging/msm/staging-devices.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: XGI: remove duplicated #include

Remove duplicated #include('s) in
  drivers/staging/xgifb/XGI_main_26.c
  drivers/staging/xgifb/vb_init.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: pohmelfs: fix compile warning

A really trivial patch to solve a warning when compiling staging/pohmelfs

Signed-off-by: Javier Martinez Canillas <martinez.javier@gmail.com>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: remove DT_3155_* errno defines

Remove the DT_3155_{SUCCESS/FAILURE} errno defines and use the
kernel provided ones.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Scott Smedley <ss@aao.gov.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: fix different address spaces noise in dt3155_drv.c

This fixes the different address spaces noise when copying data to/from
user space to kernel space.

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

Staging: dt3155v4l: correcting a pointer mismatch bug and cleanups

The third argument of dma_alloc_coherent() is of type dma_addr_t and is defined
as u64 for x86_64, while the member 'device_base' of struct dma_coherent_mem is
defined as u32.

While at it, clean up the code. Now if dev->dma_mem != NULL all the fields/members
are already properly initialized.

Signed-off-by: Marin Mitov <mitov@issp.bas.bg>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192u: fix checkpatch.pl issues to line 514

Fixed most problems pointed out by checkpatch.pl in file r8192U_core.c
up to line 500

Signed-off-by: Mike Gilks <mike.kernel@gilksonline.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155: dt3155_isr.c: Fix checkpatch.pl issues.

Fix resolves checkpatch.pl issues.

Signed-off-by: Nanakos Chrysostomos <nanakos@wired-net.gr>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl81xx: Fix build problems when CONFIG_CRYPTO=n

Select CONFIG_CRYPTO for all rtl81xx wireless drivers

... to avoid build problems like:

ERROR: "crypto_destroy_tfm" [drivers/staging/rtl8187se/r8187se.ko] undefined!
ERROR: "crypto_alloc_base" [drivers/staging/rtl8187se/r8187se.ko] undefined!
ERROR: "crypto_destroy_tfm" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
ERROR: "crypto_alloc_base" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
ERROR: "crypto_destroy_tfm" [drivers/staging/rtl8192su/r8192s_usb.ko] undefined!
ERROR: "crypto_alloc_base" [drivers/staging/rtl8192su/r8192s_usb.ko] undefined!
ERROR: "crypto_destroy_tfm" [drivers/staging/rtl8192e/r8192e_pci.ko] undefined!
ERROR: "crypto_alloc_base" [drivers/staging/rtl8192e/r8192e_pci.ko] undefined!

when drivers are built as modules but CONFIG_CRYPTO=n.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: crystalhd: removed kfree(NULL) checks

Removed kfree(NULL checks) that were not necessary

Signed-off-by: Scott Kidder <scott.kidder11@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dream: pmem.c: Fix checkpatch.pl issues.

This patch resolves checkpatch.pl issues in the pmem.c file.

Signed-off-by: Nanakos Chrysostomos <nanakos@wired-net.gr>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cx25821: added KERN_ facility levels

Fixed issues found by checkpatch.pl including adding KERN_
facility levels to printk() calls and C99 comments.

Signed-off-by: Scott Kidder <scott.kidder11@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: Removed one unnecessary space

Nothing special really, just a small fix.

Signed-off-by: Matti Lammi <matti.lammi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: otus: fix strcpy() overflow

wrq->name is only 16 characters long but "IEEE 802.11-MIMO" is 16
characters + a NULL character, so it's too long.  This patch changes it
to "IEEE 802.11abgn".

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

Staging: otus: TODO: Don't bother, for documentation purposes only

On Mon, 2010-06-07 at 12:02 -0700, Luis R. Rodriguez wrote:
> this driver is unmaintained and its only purpose is as a
> source of documentation for developers working on ar9170 and carl9170.
> Once carl9170 gets 11n support and merged upstream then this driver
> can be removed.

Then the TODO file should be updated.

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

Staging: otus: fix style issues in zdusb.c

This is a patch to zdusb.c that fixes whitespace,
C99 comment, and other style issues found by the
checkpatch.pl tool

Signed-off-by: Scott Kidder <scott.kidder11@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: wlan-ng: silence a sparse warning

This doesn't change the behavior.  It just silences a sparse warning.

drivers/staging/wlan-ng/hfa384x_usb.c:2810:62: warning: dubious: !x | !y

The point of the bitwise OR is so that a logical OR could short circuit
the second call to test_and_set_bit().

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

Merge branch 'staging-next' into work

Staging: winbond: fix build errors

Some errors crept in due to a previous patch that I missed.

This fixes them up so the driver continues to build, sorry about that.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: wlan-ng: Switch from wext to cfg80211

Switch driver over from wext to cfg80211 interface.

Some Notes:

- This patch moves the driver wholesale from wext to cfg80211. Wext
support is still provided through the cfg80211 provided wext
compatability layer.

- Currently only infrastructure mode is implemented. Ad hoc mode is not
yet implemented, but can be added.

- It does not support connecting to a specified bssid, instead roaming
is handled by the card itself. This matches the behaviour of the
existing driver.

- It has been tested using NetworkManager (via wpa_supplicant)
configured to use the wext compatability layer, and then again with the
native nl80211 layer.

Signed-off-by: Karl Relton <karllinuxtest.relton@ntlworld.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cxt1e1: remove code guarded by GENERIC_HDLC_VERSION

Current linux kernel has GENERIC_HDLC_VERSION equal to 4, no need to
have dead code in-kernel which was there for ancient kernel versions.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su: remove all code dependent on LINUX_VERSION_CODE

Remove all code which is dead for in-kernel driver due to being
ifdefed by LINUX_VERSION_CODE.

While at it, also remove surrounding code which is commented out,
or '#if 1' nops.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cxt1e1: remove all code dependent on LINUX_VERSION_CODE

Remove all code which is dead for in-kernel driver due to being
ifdefed by LINUX_VERSION_CODE.

While at it, also remove surrounding code which is commented out,
or '#if 1' nops.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192su: Fix procfs code for interfaces not named wlan0

The current code creates directories in procfs named after interfaces,
but doesn't handle renaming.  This can result in name collisions and
consequent WARNINGs.  It also means that the interface name cannot
reliably be used to remove the directory - in fact the current code
doesn't even try, and always uses "wlan0"!

Since the name of a proc_dir_entry is embedded in it, use that when
removing it.

Add a netdev notifier to catch interface renaming, and remove and
re-add the directory at this point.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

Staging: rtl8192su: Clean up in case of an error in module initialisation

Currently various resources may be leaked in case of an error.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dream: camera: Use memdup_user

Use memdup_user when user data is immediately copied into the
allocated region.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression from,to,size,flag;
position p;
identifier l1,l2;
@@

-  to = \(kmalloc@p\|kzalloc@p\)(size,flag);
+  to = memdup_user(from,size);
   if (
-      to==NULL
+      IS_ERR(to)
                 || ...) {
   <+... when != goto l1;
-  -ENOMEM
+  PTR_ERR(to)
   ...+>
   }
-  if (copy_from_user(to, from, size) != 0) {
-    <+... when != goto l2;
-    -EFAULT
-    ...+>
-  }
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vme: bridges: Add missing unlocks

Add a spin_unlock and mutex_unlock missing on the error path.

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

// <smpl>
@@
expression E1;
@@

* spin_lock(E1,...);
  <+... when != E1
  if (...) {
    ... when != E1
*   return ...;
  }
  ...+>
* spin_unlock(E1,...);
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Martyn Welch <martyn.welch@ge.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vme: devices: vme_user.c: Fix checkpatch.pl issues.

Fix resolves checkpatch.pl issues for vme_user.c file.

Signed-off-by: Nanakos Chrysostomos <nanakos@wired-net.gr>
Signed-off-by: Martyn Welch <martyn.welch@ge.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: adis16255: missing spacess in log messages added

Signed-off-by: Matthias Brugger <mensch0815@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: adis16209/220/240/350: tuning spi delay to make hardware more stable

Signed-off-by: Barry Song <21cnbao@gmail.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: adis16400: fix some minor issues and clean-up

1. move adis16400_spi_read_burst() to adis16400_ring.c since it is only
   called there
2. add the lost calling to adis16400_self_test()
3. codes cleanup

Signed-off-by: Barry Song <21cnbao@gmail.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: adis16300: fix some minor issues and clean-up

1. add delay between spi transfers
2. move burst read to ring function
3. clean-up

Signed-off-by: Barry Song <21cnbao@gmail.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: add file pointer to sysfs callbacks

The sysfs attribute call backs take a file pointer these days.  This was
added in 2c3c8bea6088 "sysfs: add struct file* to bin_attr callbacks"

Signed-off-by: Dan Carpenter <error27@gmail.com>
Cc: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: Add mailing list to the TODO file.

As suggested by Charles Clement

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

Staging: iio: pull in slab.h for kmalloc funcs

These drivers use kzalloc() but don't include slab.h.  They currently
build though because the spi.h header will pull in slab.h for us.  But
rather than rely on that behavior forever, include slab.h explicitly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: kill off spurious semicolons

A bunch of IIO files contain spurious semicolons after function
definitions and case statements and if statements.  Guess people
really like this thing, but kill them anyways so they'll stop
spreading via copy & paste with new drivers.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: standardize kconfig/makefile spacing/style

Standardize the spacing/style across the IIO build files:
- comment block in Kconfigs
- newlines at ends of files
- trailing lines at ends of files
- indent with one tab, not spaces or mixed

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: zram: Rename ramzswap to zram in documentation

Related changes:
 - Included example to show usage as generic
(non-swap) disk with ext4 filesystem.
 - Renamed rzscontrol to zramconfig to match
with new device naming.

Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: zram: Rename ramzswap to zram in code

Automated renames in code:
 - rzs* -> zram*
 - RZS* -> ZRAM*
 - ramzswap* -> zram*

Manual changes:
 - Edited comments/messages mentioning "swap"

Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: Rename ramzswap files to zram

Related changes:
 - Modify revelant Kconfig and Makefile accordingly.
 - Change include filenames in code.
 - Remove dependency on CONFIG_SWAP in Kconfig as zram usage
is no longer limited to swap disks.

Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ramzswap: Support generic I/O requests

Currently, ramzwap devices (/dev/ramzswapX) can only
be used as swap disks since it was hard-coded to consider
only the first request in bio vector.

Now, we iterate over all the segments in an incoming
bio which allows us to handle all kinds of I/O requests.

ramzswap devices can still handle PAGE_SIZE aligned and
multiple of PAGE_SIZE sized I/O requests only. To ensure
that we get always get such requests only, we set following
request_queue attributes to PAGE_SIZE:
 - physical_block_size
 - logical_block_size
 - io_min
 - io_opt

Note: physical and logical block sizes were already set
equal to PAGE_SIZE and that seems to be sufficient to get
PAGE_SIZE aligned I/O.

Since we are no longer limited to handling swap requests
only, the next few patches rename ramzswap to zram. So,
the devices will then be called /dev/zram{0, 1, 2, ...}

Usage/Examples:
 1) Use as /tmp storage
 - mkfs.ext4 /dev/zram0
 - mount /dev/zram0 /tmp

 2) Use as swap:
 - mkswap /dev/zram0
 - swapon /dev/zram0 -p 10 # give highest priority to zram0

Performance:

 - I/O benchamark done with 'dd' command. Details can be
found here:
http://code.google.com/p/compcache/wiki/zramperf
Summary:
 - Maximum read speed (approx):
   - ram disk: 1200 MB/sec
   - zram disk: 600 MB/sec
 - Maximum write speed (approx):
   - ram disk: 500 MB/sec
   - zram disk: 160 MB/sec

Issues:

 - Double caching: We can potentially waste memory by having
two copies of a page -- one in page cache (uncompress) and
second in the device memory (compressed). However, during
reclaim, clean page cache pages are quickly freed, so this
does not seem to be a big problem.

 - Stale data: Not all filesystems support issuing 'discard'
requests to underlying block devices. So, if such filesystems
are used over zram devices, we can accumulate lot of stale
data in memory. Even for filesystems to do support discard
(example, ext4), we need to see how effective it is.

 - Scalability: There is only one (per-device) de/compression
buffer stats. This can lead to significant contention, especially
when used for generic (non-swap) purposes.

Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ti-st: fix protocol counting

Older mechanism of counting the number of protocols
registered with ST was slow, in-efficient.
It used to check the protocol data for NULL for each
registration/unregistration.

With this change, counting protocols in maintained by
a single counter protos_registered.

Counting protocols is not just for debug purposes

Signed-off-by: Naveen Jain <naveen_jain@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ti-st: add version info to sysfs

Add version information to be available under the sysfs group
for kim.

Signed-off-by: Naveen Jain <naveen_jain@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: ti-st: cleanup sysfs code

Remove the older way of communicating with user-space
daemon UIM.
The older way involved creating of a new kobj and pid
sysfs file under it, and sending signal using that pid.

Now we communicate via rfkill to user-space UIM.

Background: UIM is the user-space daemon which upon
notification from ldisc driver, opens the tty, sets
default baud and then installs (tiocsetd) the ldisc.

Signed-off-by: Naveen Jain <naveen_jain@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: Update TI-ST's TODO

Update the TODO of TI-ST to make sure it reflects current list
of activities that need to be done.
Also point it out to the user-space app code relevant to it.

Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: easycap: add easycap driver

This adds the easycap USB video adapter driver to
the staging directory.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Merge git://git.infradead.org/users/dwmw2/spectra-2.6 into work

Staging: vt6655: struct pci_driver cleanup

Rename driver struct and callbacks to vt6655_* instead of device_* and add
__devinit/__devexit directives.

Signed-off-by: Charles Clément <caratorn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt665*: fix typos concerning "management"

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: fix coding style issues in 80211mgr.c

This is a patch to the 80211mgr.c file that fixes up warnings
found by the checkpatch.pl tool

Signed-off-by: Timofey Trofimov <tumoxep@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: fix up built-in compilation together with rt2870

Fix this build error:

  drivers/staging/vt6656/built-in.o: In function `rotr1':
  (.text+0x1a878): multiple definition of `rotr1'
  drivers/staging/rt2870/built-in.o:(.text+0x106c2): first defined here
  drivers/staging/vt6656/built-in.o: In function `tkip_sbox':
  (.text+0x1a848): multiple definition of `tkip_sbox'
  drivers/staging/rt2870/built-in.o:(.text+0x10697): first defined here
  drivers/staging/vt6656/built-in.o: In function `xor_32':
  (.text+0x1ec24): multiple definition of `xor_32'
  drivers/staging/rt2870/built-in.o:(.text+0x111c4): first defined here
  drivers/staging/vt6656/built-in.o: In function `xor_128':
  (.text+0x1ec00): multiple definition of `xor_128'
  drivers/staging/rt2870/built-in.o:(.text+0x111dd): first defined here

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: code cleanup, resolved checkpatch findings in headers

Resolved whitespace-related checkpatch findings in .h files

Signed-off-by: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: code cleanup, fixed 'for' statements

Resolved checkpatch findings, but some long lines warnings.
ERROR: space required before the open parenthesis '('
ERROR: spaces required around that

Signed-off-by: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6655: fix up U32 conversion

Commit 902d2411298a6e0b8a473a0163aa4bf6d5c8a073 converted U32 to u32
which you would think would be just fine.  Unfortunatly, it seems that
VIA only builds their code on a 32bit processor (which makes sense if
you think about it), but this doesn't work on x86-64.  So fix up the few
places where this really wanted to be an unsigned long width.

Cc: Charles Clément <caratorn@gmail.com>
Cc: Forest Bond <forest@alittletooquiet.net>,
Cc: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>