staging/lustre/llite: fix mkdir endless loop

Running on 3.11-rc4 kernel, I got below endless loop. It turns to be that Lustre
always saves the first page of a dir inode mapping at index ~0UL.  And after
commit 5a720394 (mm: teach truncate_inode_pages_range() to handle non page
aligned ranges), truncate_inode_pages_range() _NO LONGER_ truncates the page
that is sitting at index ~0UL.

[16768.998006] mkdir           R  running task        0  2717   2716 0x00000080
[16768.998073]  000000000000000e 0000000000000000 0000000000000000 ffff88000be00460
[16768.998157]  ffff88000ea65908 ffffffff810fec3e ffff88000ea65968 ffff8800229e7750
[16768.998241]  ffff88000ea658b8 0000000000000000 0000000000000000 ffff88000ea65958
[16768.998326] Call Trace:
[16768.998401]  [<ffffffff810fc6ed>] ? rcu_read_unlock+0x1c/0x2d
[16768.998473]  [<ffffffff810fec3e>] ? find_get_pages+0xf5/0x11b
[16768.998530]  [<ffffffff811078f0>] ? pagevec_lookup+0x20/0x2a
[16768.998586]  [<ffffffff8110920e>] ? truncate_inode_pages_range.part.2+0x161/0x39a
[16768.998680]  [<ffffffffa02ad5dc>] ? ll_md_blocking_ast+0x338/0x62f [lustre]
[16768.998744]  [<ffffffff8110947f>] ? truncate_inode_pages_range+0x38/0x3f
[16768.998805]  [<ffffffff811094f8>] ? truncate_inode_pages+0x12/0x14
[16768.998871]  [<ffffffffa02ad6e8>] ? ll_md_blocking_ast+0x444/0x62f [lustre]
[16768.998948]  [<ffffffff810981b5>] ? arch_local_irq_save+0x9/0xc
[16768.999022]  [<ffffffffa07ee0e8>] ? ldlm_cancel_callback+0x67/0x12a [ptlrpc]
[16768.999100]  [<ffffffffa07f85b2>] ? ldlm_cli_cancel_local+0xf3/0x2bc [ptlrpc]
[16768.999176]  [<ffffffffa07f9163>] ? ldlm_cli_cancel_list_local+0x7e/0x1e4 [ptlrpc]
[16768.999268]  [<ffffffffa07f9473>] ? ldlm_cancel_resource_local+0x1aa/0x1b9 [ptlrpc]
[16768.999385]  [<ffffffffa0657bad>] ? mdc_resource_get_unused+0xf8/0x115 [mdc]
[16768.999472]  [<ffffffff8109c887>] ? trace_hardirqs_on+0xd/0xf
[16768.999533]  [<ffffffffa06583d8>] ? mdc_create+0x11e/0x4db [mdc]
[16768.999597]  [<ffffffff8152ed84>] ? mutex_unlock+0xe/0x10
[16768.999654]  [<ffffffffa0350e99>] ? lmv_create+0x355/0x3e9 [lmv]
[16768.999712]  [<ffffffff811553b7>] ? final_putname+0x35/0x39
[16768.999775]  [<ffffffffa02ae167>] ? ll_new_node+0x33b/0x3ff [lustre]
[16768.999841]  [<ffffffffa02ae62c>] ? ll_mkdir+0xf2/0x127 [lustre]
[16768.999897]  [<ffffffff81156996>] ? vfs_mkdir+0x84/0xc9
[16768.999961]  [<ffffffff81158cf8>] ? SyS_mkdirat+0x77/0xad
[16769.000014]  [<ffffffff81158d47>] ? SyS_mkdir+0x19/0x1b
[16769.000066]  [<ffffffff81538652>] ? system_call_fastpath+0x16/0x1b

Signed-off-by: Peng Tao <bergwolf@gmail.com>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

lustre/fld: remove fld_thread_key

It is left over when porting Lustre tree patch in commit (e62e5d92) and should be
removed.

Signed-off-by: Peng Tao <bergwolf@gmail.com>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Remove typedef and update cfs_hash_bucket struct

Remove typedef keyword and rename the cfs_hash_bucket_t struct to
cfs_hash_bucket in libcfs_hash.h. These changes resolve the
"Do not add new typedefs" warning generated by checkpatch.pl and
meet kernel coding style.

The struct variables in hash.c are updated to reflect this change
as well.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: sb105x: Fix a possible null dereference

This patch fixes the following error in sb_pc_mp.c-

drivers/staging/sb105x/sb_pci_mp.c:546 mp_startup() error: we previously assumed 'info->tty' could be null (see line 525)

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ashmem: Fix ASHMEM_PURGE_ALL_CACHES return value

Hopefully this isn't too late for 3.12.

In commit 7dc19d5aff (convert shrinkers to new count/scan API)
the return value to PURGE_ALL_CACHES was dropped, causing -EPERM
to always be returned.

This patch re-adds the ret assignment, setting it to the the
ashmem_shrink_count(), which is the lru_count.

(Sorry this was missed in the review!)

Fixes: 7dc19d5affd7 ("convert shrinkers to new count/scan API")
Cc: Colin Cross <ccross@android.com>
Cc: Android Kernel Team <kernel-team@android.com>
Cc: Glauber Costa <glommer@openvz.org>
Reported-by: YongQin Liu <yongqin.liu@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: stable <stable@kernel.org> # 3.12
Acked-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Remove typedef and update cfs_hash_lock union

Remove typedef keyword and rename the cfs_hash_lock_t
union to cfs_hash_lock in libcfs_hash.h. These changes
resolve the "Do not add new typedefs" warning
generated by checkpatch.pl and meet kernel coding style.

The union variables in hash.c are updated to reflect this
change as well.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: replace foo* with foo *

Fix checkpatch.pl warning of "(foo*)" should be "(foo *)" in lustre/lov

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: Fix externs should be avoided in .c files warning

Fix the checkpatch.pl warning 'externs should be avoided in .c files' in
lustre/lov.

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: Fix space before open parenthesis

Fix checkpatch.pl error "space required before the open parenthesis '('"
in lustre/lov.

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: Replace max() with max_t()

Replace max() with max_t() in accordance to chekpatch.pl warning in
lov_obd.c.

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: Fix C99 style comments

Fix C99 style comments in lov_obd.c in accordance to checkpatch.pl
warning.

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:lustre: Remove space between function name and open parenthesis

Remove checkpatch.pl warning of space prohibited between function name
and open parenthesis '('in /lustre/lov.

Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: libcfs: Remove bitmap.h file

The functions in the bitmap.h header file are not being used by
any source file in the lustre driver.

Unless there are any objections, please remove the bitmap.h
header file as it contains unused code in the kernel.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8187se: checkpatch cleanups

Fix the following checkpatch.pl errors:

drivers/staging/rtl8187se/ieee80211/ieee80211_softmac.c:890: "foo* bar" should be "foo *bar"
drivers/staging/rtl8187se/ieee80211/ieee80211_softmac.c:944: "foo* bar" should be "foo *bar"

Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8187se: fix static sparse warnings

This patch fixes sparse warnings regarding functions that should be
static in driver rtl8187se.

Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8187se: add missing __iomem annotation

Fix the following type of sparse warnings:

drivers/staging/rtl8187se/r8180_core.c:163:48: warning: incorrect type in argument 1 (different address spaces)
drivers/staging/rtl8187se/r8180_core.c:163:48:    expected void const volatile [noderef] <asn:2>*addr
drivers/staging/rtl8187se/r8180_core.c:163:48:    got unsigned char [usertype] *

Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8187se: use gfp_t instead of int

Fix the following type of sparse warnings:

drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17: warning: incorrect type in argument 2 (different base types)
drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17:    expected restricted gfp_t [usertype] flags
drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17:    got int [signed] gfp_mask

Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: sb105x: Remove redundant null check

This patch fixes the following Sparse information in sb_pci_mp.c-

~/git/kernels/linux/drivers/staging/sb105x/sb_pci_mp.c:1813
mp_unregister_driver() info: redundant null check on drv->state calling kfree()

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: line6: replacing -EINVAL with retval in driver.c

drivers/staging/line6/driver.c:208 line6_send_raw_message_async_part()
info: why not propagate 'retval' from usb_submit_urb() instead of (-22)?

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: line6: replacing -EINVAL with retval

drivers/staging/line6/midi.c:147 send_midi_async() info: why not
propagate 'retval' from usb_submit_urb() instead of (-22)?

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: speakup: replacing simple_strtoul with kstrtoul

Fix checkpatch warning: WARNING: simple_strtoul is obsolete, use
kstrtoul instead

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: sep: Fix Smatch Warning in sep_crypto.c

This patch fixes the following smatch warning-

~/git/kernels/linux/drivers/staging/sep/sep_crypto.c:1137
sep_crypto_block_data() info: why not propagate 'int_error' from sep_oddball_pages() instead of (-12)?
~/git/kernels/linux/drivers/staging/sep/sep_crypto.c:1152
sep_crypto_block_data() info: why not propagate 'int_error' from sep_oddball_pages() instead of (-12)?

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: Fix C99 commenting style in michael.h

This patch fixes C99 commenting style in staging:vt6655:michael.h

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: slicoss: Fix quoted string split across lines

This patch fixes the following checkpatch.pl warning in slicoss.c-
WARNING: quoted string split across lines

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: slicoss: Refactor code by introducing new function slic_interrupt_card_up()

This patch fixes the following checkpatch.pl warning in slicoss.c:
WARNING: Too many leading tabs - consider code refactoring

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: slicoss: Replace printk() with pr_debug() or dev_dbg() as required

This patch fixes the following checkpatch.pl warning in slicoss.c:
WARNING: Prefer netdev_dbg(netdev, ... then dev_dbg(dev, ... then
pr_debug(...  to printk(KERN_DEBUG ...

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: slicoss: Removal of if check since debugfs_remove(NULL) is safe

This patch fixes the following checkpatch.pl warning in slicoss.c:
WARNING: debugfs_remove(NULL) is safe this check is probably not
required

Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: dgnc: avoiding buffer overflow

drivers/staging/dgnc/dgnc_neo.c:1969 neo_vpd() error: buffer overflow 'brd->vpd' 128 <= 128
drivers/staging/dgnc/dgnc_neo.c:1970 neo_vpd() warn: buffer overflow 'brd->vpd' 128 <= 130
drivers/staging/dgnc/dgnc_neo.c:1970 neo_vpd() warn: buffer overflow 'brd->vpd' 128 <= 130

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: dgnc: removing the comparison 'board<0'

Fix dgnc_mgmt_ioctl() warn: unsigned 'board' is never less than zero.

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: braces {} are not necessary for single statement in dgnc_driver.c

Fixes "braces {} are not necessary for single statement blocks" warning
in dgnc_driver.c

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc:Removed unecessary error messages in dgnc_driver.c

Fixed removal of unecessary error messages in staging:dgnc:dgnc_driver.c

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wlags49_h2: Fixes the sparse warning in sta_h2.c

This patch fixes sparse warning "Using plain integer as NULL pointer" in
sta_h2.c

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wlags49_h2: fixed space prohibited warnings in hcf.h

This patch fixes the following warnings in hcf.c:
"space prohibited between function name and open parenthesis '(' "

"space prohibited before that close parenthesis ')' ".

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: bitfield manipulation macros for CRA, CRB and setup

Some bits of the 'CRB' register have different functions when read or
written, so add macros to define the read-only parts.

Add macros to define the widths of the bitfields in the 'CRA' and 'CRB'
registers and the standard encoder setup value.

Add macros to construct and extract parts of the 'CRA' and 'CRB'
register values and the standard encoder setup value, along with a
couple of general helper macros for the above.

Redefine the bitfield mask macros for 'CRA', 'CRB' and standard encoder
setup using the above.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: add missing bits for 'CRB' register

There are some bits in the 'CRB' register not defined in "s626.h".
Three of these are read-only bits that overlay the write-only interrupt
control bits.  Another missing bit controls whether counter 'B' is
cleared when counter 'A' overflows.  Add the missing bit definitions for
completeness.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: correct S626_CRAMSK_CLKPOL_A macro (unused)

The counter 'A' clock polarity field in the 'CRA' register is only 1 bit
wide, but the `S626_CRAMSK_CLKPOL_A` macro shows it as 2 bits wide,
which would overlap with the counter 'A' interrupt source field.  This
is harmless as the macro isn't actually used yet, but correct it anyway
as I want to use it!

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: distinguish counter src from encoder mode

The macros `S626_CLKSRC_COUNTER`, `S626_CLKSRC_TIMER` and
`S626_CLKSRC_EXTENDER` are used logically to set the operating mode of
an encoder channel.  `S626_CLKSRC_COUNTER` (0) is also used as a 2-bit
physical value to set the counter source of an encoder channel to
"encoder".

Rename the macros to `S626_ENCMODE_COUNTER`, `S626_ENCMODE_TIMER` and
`S626_ENCMODE_EXTENDER` and rename some other macros and (unused)
functions relating to the encoder mode for consistency.

Define new macros to specify the physical counter source values for the
'CRA' register and rename the corresponding bitshift and mask macros
accordingly.  The physical values for the counter source are:

  S626_CNTSRC_ENCODER = 0      // encoder
  S626_CNTSRC_DIGIN = 1        // digital inputs
  S626_CNTSRC_SYSCLK = 2       // system clock up
  S626_CNTSRC_SYSCLK_DOWN = 3  // system clock down

Also use the `S626_CNTSRC_SYSCLK` value as a bitmask (bit 1) to indicate
either of the system clock values, with the direction (bit 0) indicated
separately in this case.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: correct a comment in s626_get_mode_b()

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: specify bitshift for encoder A clock source

When setting the clock source for one of the 'A' encoders to operate in
"counter" mode in `s626_set_mode_a()`, bitshift the clock source value by
`S626_CRABIT_CLKSRC_A` for consistency with the other modes.  This has
no effect on the value since `S626_CRABIT_CLKSRC_A` is 0.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: s626: clock polarity and direction are the same

When setting up an encoder channel, the setup value includes a polarity
and direction, but these are the same bit of the setup value:

  S626_CLKPOL_POS = S626_CNTDIR_UP = 0

  S626_CLKPOL_NEG = S626_CNTDIR_DOWN = 1

In the construction of the setup value, both the CLKPOL and the CNTDIR
constants are shifted by the same amount `S626_BF_CLKPOL`.  Only the
following combinations are set up currently (this may change if user
configuration of the encoder is implemented properly):

  (S626_CLKPOL_POS << S626_BF_CLKPOL)

  (S626_CLKPOL_POS << S626_BF_CLKPOL) |
  (S626_CNTDIR_UP << S626_BF_CLKPOL)

  (S626_CLKPOL_POS << S626_BF_CLKPOL) |
  (S626_CNTDIR_DOWN << S626_BF_CLKPOL)

The first two are used in "counter" mode and is equivalent to:

  (S626_CLKPOL_POS << S626_BF_CLKPOL)

The last one is used in "timer" mode and is equivalent to:

  (S626_CNTDIR_DOWN << S626_BF_CLKPOL)

Use the shorter equivalents.  The comments in "s626.h" indicate that the
'CLKPOL' constants make more sense for the "counter" mode (when the
encoders operate as up/down counters) and the 'CNTDIR' constants make
more sense for the "timer" mode.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: rtd520: use normal bitfield functions

The `unsigned char chan_is_bipolar[]` member of `struct rtd_private` is
used with some macros as a packed array of 1-bit values that indicate
whether the corresponding entries in the hardware's "channel-gain" table
have been set to a bipolar (1) or unipolar (0) range, as the raw samples
from the hardware need to be cooked differently in each case.

Replace the declaration of the member with a standard Linux bitfield
using `DECLARE_BITFIELD()`, and replace the home-grown macros used
access the bitfield with the standard Linux non-atomic bitop functions,
`__set_bit()`, `__clear_bit()` and `test_bit()`.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: ft1000: fixed two coding style issues

Fixed two coding style issues, specifically:

ft1000_proc.c:35: ERROR: space required before the open parenthesis '('
ft1000_proc.c:42: ERROR: space required before the open parenthesis '('

Signed-off-by: Aldo Iljazi <me@aldo.io>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xillybus: Use dev_* functions instead of pr_* in xillybus_core

Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xillybus: Use dev_* functions instead of pr_* in xillybus_of

Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xillybus: Use dev_* functions instead of pr_* in xillybus_pcie

Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xillybus: Remember device pointer for use with dev_* functions

This is necessary so that xillybus_core uses the correct device pointer
for PCIe devices in diagnostic message calls (dev_err, dev_warn and dev_info)

Signed-off-by: Eli Billauer <eli.billauer@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Remove unnecessary spaces in router_proc.c

Remove unnecessary spaces between function names and open
parentheses in router_proc.c to meet kernel coding style.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Fix static initialization error in router_proc.c

Fixed the static initialization error generated by checkpatch.pl to
meet kernel coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Reformat pointer variable in lib-move.c

Reformat a pointer variable in lib-move.c to meet kernel
coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Remove () from return statements in lib-move.c

Remove unnecessary parentheses from return statements to
eliminate errors generated by checkpatch.pl and meet kernel
coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Adjust lines to meet 80 column limit in lib-move.c

Fixed lines to not exceed more than 80 columns per line to meet
kernel coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Move open braces to previous lines in lib-move.c

Move open braces to previous lines in lib-move.c to meet kernel
coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet: Remove unnecessary whitespace in lib-move.c

Remove unnecessary whitespace around open parentheses in lib-move.c
to meet kernel coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet/selftest: Remove unnecessary spaces in brw_test.c

Remove spaces between function names and open parentheses to
eliminate warnings generated by checkpatch.pl and meet kernel
coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre/lnet/selftest: Fix trailing statements in brw_test.c

Fix trailing statement errors generated by checkpatch.pl in
brw_test.c to meet kernel coding standards.

Signed-off-by: Lisa Nguyen <lisa@xenapiadmin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge 3.12-rc6 into staging-next.

We want these fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Linux 3.12-rc6

staging: rtl8192u: fix sparse warning, make a function static

Fix the following sparse warning:
drivers/staging/rtl8192u/ieee80211/ieee80211_tx.c:240:22: warning: symbol 'ieee80211_alloc_txb' was not declared. Should it be static?

Signed-off-by: Teodora Baluta <teobaluta@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: speakup: removing jiffies comparison in speakup_apollo.c

Fix checkpatch warning: WARNING: Comparing jiffies is almost always
wrong; prefer time_after, time_before and friends

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: speakup: removing jiffies comparison using time_after_eq()

Fix checkpatch warning:  Comparing jiffies is almost always wrong;
prefer time_after, time_before and friends

Signed-off-by: Ashvini Varatharaj <ashvinivaratharaj@gmail.com>
Reviewed-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'iio-for-3.12d' of git://git./linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

Fourth round of IIO new drivers, functionality and cleanups for the 3.13 cycle.

New Drivers
* cm36651 combined RGB light and proximity sensor.

Core improvements

* Some more fixes and cleanups related to buffers.  These include the second
  half of a series which went is as fixes.  The basis for delaying until the
  next merge window is that some are too invasive for this late in a cycle
  and others only effect code paths current unused in the mainline tree.
  In this case we have:
   * protecting against concurrent userspace access
   * fixing a memory leak if a device goes away
   * avoiding always reallocating the buffer whether or not it has changed
     (a bug fix, but one with no functional changes other than a small speed
     improvement.)
   * Add reference counting for buffers to ensure they hang around if open
     from userspace or in kernel when the device is forcefully removed.
   * Return -ENODEV for buffer access operations when the device has gone
     away.
   * Add proper locking for iio_update_buffers (currently we only have one
     buffer per device in mainline, but an input bridge driver is under
     development which would make this bug 'real'.)
   * Wake up anyone waiting on a buffer if the device is unregistered.  A
     subsequent read will fail, notifying userspace that the device is no
     longer there rather than having it wait possibly for ever.

* Move the iio_sw_preenable functionality into the core.  This avoids drivers
  having to 'know' about how the buffers are implemented and is called by
  almost all drivers anyway.  Those that don't call it are not harmed by it
  being called.
* New registration approach for information (i.e. sysfs attributes) about
  events.  Much more generic and now similar to how the equivalent is
  handled for channel information.  The events infrastructure had been left
  behind by other changes so this brings it back in line.
* Using the new events registration approach, add a hysterisis event_info
  element and apply this to those drivers with this property.
* A little unitialized variable bug in the generic_buffer.c example.
* Factor out the code for freeing lists of IIO Device attributes to avoid
  some repitition.

Driver cleanups
* At91 driver gains touch screen support and some related fixes.
* Follow up series of patches removing the now redundant
  iio_sw_buffer_preenable calls.
* Lots of conversions to the new event registration methods.
* Another round of hmc5843 cleanups as that driver moves towards graduating
  from staging.
* Make some SoC drivers buildable if COMPILE_TEST is used.  Follow up fixes
  for a few bits and bobs that revealed.
* Add explicit includes of linux/of.h to those drivers making us of linux/of.h

Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Pull btrfs fix from Chris Mason:
 "Sage hit a deadlock with ceph on btrfs, and Josef tracked it down to a
  regression in our initial rc1 pull.  When doing nocow writes we were
  sometimes starting a transaction with locks held"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: release path before starting transaction in can_nocow_extent

Merge tag 'pm+acpi-3.12-rc6' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI and power management fixes from Rafael Wysocki:

 - intel_pstate fix for misbehavior after system resume if sysfs
   attributes are set in a specific way before the corresponding suspend
   from Dirk Brandewie.

 - A recent intel_pstate fix has no effect if unsigned long is 32-bit,
   so fix it up to cover that case as well.

 - The s3c64xx cpufreq driver was not updated when the index field of
   struct cpufreq_frequency_table was replaced with driver_data, so
   update it now.  From Charles Keepax.

 - The Kconfig help text for ACPI_BUTTON still refers to
   /proc/acpi/event that has been dropped recently, so modify it to
   remove that reference.  From Krzysztof Mazur.

 - A Lan Tianyu's change adds a missing mutex unlock to an error code
   path in acpi_resume_power_resources().

 - Some code related to ACPI power resources, whose very purpose is
   questionable to put it lightly, turns out to cause problems to happen
   during testing on real systems, so remove it completely (we may
   revisit that in the future if there's a compelling enough reason).
   From Rafael J Wysocki and Aaron Lu.

* tag 'pm+acpi-3.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI / PM: Drop two functions that are not used any more
  ATA / ACPI: remove power dependent device handling
  cpufreq: s3c64xx: Rename index to driver_data
  ACPI / power: Drop automaitc resume of power resource dependent devices
  intel_pstate: Fix type mismatch warning
  cpufreq / intel_pstate: Fix max_perf_pct on resume
  ACPI: remove /proc/acpi/event from ACPI_BUTTON help
  ACPI / power: Release resource_lock after acpi_power_get_state() return error

Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Ingo Molnar:
 "Two fixlets:

   - fix a (rare-config) build bug
   - fix a next-gen SGI/UV hw/firmware enumeration bug"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86: Update UV3 hub revision ID
  x86/microcode: Correct Kconfig dependencies

iio: light: gp2ap020a00f: Include linux/of.h header

'of_match_ptr' is defined in linux/of.h. Include it explicitly to
avoid build breakage in the future.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

iio: adc: ti-adc081c: Include linux/of.h header

'of_match_ptr' is defined in linux/of.h. Include it explicitly to
avoid build breakage in the future.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

iio: adc: nau7802: Include linux/of.h header

'of_match_ptr' is defined in linux/of.h. Include it explicitly to
avoid build breakage in the future.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Cc: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

iio: adc: mcp3422: Include linux/of.h header

'of_match_ptr' is defined in linux/of.h. Include it explicitly to
avoid build breakage in the future.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

Btrfs: release path before starting transaction in can_nocow_extent

We can't be holding tree locks while we try to start a transaction, we will
deadlock.  Thanks,

Reported-by: Sage Weil <sage@inktank.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>

Merge branch 'acpi-fixes'

* acpi-fixes:
  ACPI / PM: Drop two functions that are not used any more
  ATA / ACPI: remove power dependent device handling
  ACPI / power: Drop automaitc resume of power resource dependent devices
  ACPI: remove /proc/acpi/event from ACPI_BUTTON help
  ACPI / power: Release resource_lock after acpi_power_get_state() return error

Merge branch 'pm-fixes'

* pm-fixes:
  cpufreq: s3c64xx: Rename index to driver_data
  intel_pstate: Fix type mismatch warning
  cpufreq / intel_pstate: Fix max_perf_pct on resume

Merge branch 'for-linus' of git://git.samba.org/sfrench/cifs-2.6

Pull CIFS fixes from Steve French:
 "Five small cifs fixes (includes fixes for: unmount hang, 2 security
  related, symlink, large file writes)"

* 'for-linus' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: ntstatus_to_dos_map[] is not terminated
  cifs: Allow LANMAN auth method for servers supporting unencapsulated authentication methods
  cifs: Fix inability to write files >2GB to SMB2/3 shares
  cifs: Avoid umount hangs with smb2 when server is unresponsive
  do not treat non-symlink reparse points as valid symlinks

iio: at91: introduce touch screen support in iio adc driver

AT91 ADC hardware integrate touch screen support. So this patch add touch
screen support for at91 adc iio driver.
To enable touch screen support in adc, you need to add the dt parameters:
  1. which type of touch are used? (4 or 5 wires), sample period time.
  2. correct pressure detect threshold value.

In the meantime, since touch screen will use a interal period trigger of adc,
so it is conflict to other hardware triggers. Driver will disable the hardware
trigger support if touch screen is enabled.

This driver has been tested in AT91SAM9X5-EK and SAMA5D3x-EK.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: devicetree@vger.kernel.org
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Trivial cleanup

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Check initialization and chip identifier

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Introduce _set_range_gain()

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Rename _set_rate() to _set_samp_freq()

move locking inside _set() function

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Reorganize _set_meas_conf()

move locking inside _set() function

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Rename _configure() to _set_mode()

and be consistent with other setter functions in that first argument
is hmc5843_data

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Remove ability to change operating mode

only continuous mode is supported for now; the driver could/should
be switched to single conversion mode

operating mode should be determined by the way IIO accesses the device
and not exposed explicitly

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Add trigger handling

v3:
* use __be16 instead of s16
v2 (thanks to Jonathan Cameron):
* drop dynamic buffer allocation, buffer is in hmc5842_data
* grab timestamp near data acquisition
* restrict available scan masks (only read all axis)

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Always read all channels values otherwise no updates

v2:
* use __be16 instead of s16

Split out data ready/wait for read measurement
fix bug in case reading status register fails

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Rename _check_samp_freq to get_samp_freq_index

and drop/inline helper functions _check_int_plus_micros() and
_show_int_plus_micros()

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:hmc5843: Use SCALE instead of magn_range

v3:
* rename _check_scale() to _get_scale_index()
v2:
* use SCALE instead of CALIBSCALE to control the range/gain
  of measurements

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

DT: Add documentation for cm36651 proximity/light sensor

This patch adds device tree binding documentation for CM36651 proximity/light sensor.

Signed-off-by: Beomho Seo <beomho.seo@samsung.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

iio: cm36651: Add CM36651 proximity/light sensor

This patch adds a new driver for Capella CM36651 proximity and RGB sensor.

Signed-off-by: Beomho Seo <beomho.seo@samsung.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:spear_adc: Fix IRQ check

The test in the spear_adc driver which checks whether the IRQ number returned
by platform_get_irq() has multiple problems. It accepts 0 even though this is
an invalid IRQ. It also rejects IRQ numbers that are larger or equal than
NR_IRQS. First of all drivers should never need to reference NR_IRQS and
secondly with CONFIG_SPARSE_IRQ NR_IRQS is not the upper limit, so the check
might reject valid IRQ numbers. This patch modifies the check to only test
against less or equal to 0.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging:iio:lpc32xx_adc: Fix IRQ check

The test in the lpc32xx_adc driver which checks whether the IRQ number returned
by platform_get_irq() has multiple problems. It accepts 0 even though this is an
invalid IRQ. It also rejects IRQ numbers that are larger or equal than NR_IRQS.
First of all drivers should never need to reference NR_IRQS and secondly with
CONFIG_SPARSE_IRQ NR_IRQS is not the upper limit, so the check might reject
valid IRQ numbers. This patch modifies the check to only test against less or
equal to 0.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Cc: Roland Stigge <stigge@antcom.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>

staging: vt6656: removed line over 80 chars warning

Removed the checkpatch warning of line over 80 chars
by breaking the long line into sensible chunks of 2 lines
to comply with coding style

Signed-off-by: Nandini Hanumanthagowda <nandu.hgowda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6656: removed C99-style comments

Linux style for comment is C89 style "/* */" and it
doesn't prefer C99-style comment "//...". Hence replaced
C99-style comments used in code by C89 style comment to
comply with linux coding style

Signed-off-by: Nandini Hanumanthagowda <nandu.hgowda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6656: removed unnecessary parentheses in return statement

There was parentheses around return statement's value which
was not required since return statement is not a function.
Hence removed the parentheses to eliminate the checkpatch error
which states:
ERROR: return is not a function, parentheses are not required

Signed-off-by: Nandini Hanumanthagowda <nandu.hgowda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6656: removal of braces under if-control statement

Removed braces under if-else control flow statement whenever
there is only one statement under if-else control statement
to comply with linux coding style

Signed-off-by: Nandini Hanumanthagowda <nandu.hgowda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6656: indentation and removal of unnecessary spaces

Removed unnecessary white spaces at beginning of line
and added proper indentation to fix checkpatch warnings/errors
to improve the readability of code

Signed-off-by: Nandini Hanumanthagowda <nandu.hgowda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655:removed incorrect casting in hostap.c

This patch fixes the following type of sparse warnings:

drivers/staging/vt6655/hostap.c:733:42: warning: cast from restricted gfp_t
drivers/staging/vt6655/hostap.c:733:42: warning: incorrect type in argument 2 (different base types)
drivers/staging/vt6655/hostap.c:733:42:    expected restricted gfp_t [usertype] flags
drivers/staging/vt6655/hostap.c:733:42:    got int [signed] <noident>

Signed-off-by: Archana kumari <archanakumari959@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'driver-core-3.12-rc6' of git://git./linux/kernel/git/gregkh/driver-core

Pull driver core fix from Greg KH:
 "Here is one fix for the hotplug memory path that resolves a regression
  when removing memory that showed up in 3.12-rc1"

* tag 'driver-core-3.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  driver core: Release device_hotplug_lock when store_mem_state returns EINVAL

Merge tag 'usb-3.12-rc6' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are some USB fixes and new device ids for 3.12-rc6

  The largest change here is a bunch of new device ids for the option
  USB serial driver for new Huawei devices.  Other than that, just some
  small bug fixes for issues that people have reported (run-time and
  build-time), nothing major"

* tag 'usb-3.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: usb_phy_gen: refine conditional declaration of usb_nop_xceiv_register
  usb: misc: usb3503: Fix compile error due to incorrect regmap depedency
  usb/chipidea: fix oops on memory allocation failure
  usb-storage: add quirk for mandatory READ_CAPACITY_16
  usb: serial: option: blacklist Olivetti Olicard200
  USB: quirks: add touchscreen that is dazzeled by remote wakeup
  Revert "usb: musb: gadget: fix otg active status flag"
  USB: quirks.c: add one device that cannot deal with suspension
  USB: serial: option: add support for Inovia SEW858 device
  USB: serial: ti_usb_3410_5052: add Abbott strip port ID to combined table as well.
  USB: support new huawei devices in option.c
  usb: musb: start musb on the udc side, too
  xhci: Fix spurious wakeups after S5 on Haswell
  xhci: fix write to USB3_PSSEN and XUSB2PRM pci config registers
  xhci: quirk for extra long delay for S4
  xhci: Don't enable/disable RWE on bus suspend/resume.

Merge tag 'tty-3.12-rc6' of git://git./linux/kernel/git/gregkh/tty

Pull serial driver fixes from Greg KH:
 "Here are two serial driver fixes for your tree.  One is a revert of a
  patch that causes a build error, the other is a fix to provide the
  correct brace placement which resolves a bug where the driver was not
  working properly"

* tag 'tty-3.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: vt8500: add missing braces
  Revert "serial: i.MX: evaluate linux,stdout-path property"

Merge tag 'char-misc-3.12-rc6' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are some small iio and w1 driver fixes for 3.12-rc6.

  There is also a hyper-v fix in here, which turned out to be incorrect,
  so it was reverted.  That will probably have to wait unto 3.13-rc1 to
  get accepted as it's still being discussed"

* tag 'char-misc-3.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  Revert "Drivers: hv: vmbus: Fix a bug in channel rescind code"
  Drivers: hv: vmbus: Fix a bug in channel rescind code
  iio:buffer: Free active scan mask in iio_disable_all_buffers()
  iio: frequency: adf4350: add missing clk_disable_unprepare() on error in adf4350_probe()
  w1 - call request_module with w1 master mutex unlocked
  w1 - fix fops in w1_bus_notify

Merge tag 'sound-3.12' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "All reasonably small fixes as rc6: a HD-audio mic fix, a us122l mmap
  regression fix, and kernel memory leak fix in hdsp driver.  Hopefully
  this will be the last pull request for 3.12..."

* tag 'sound-3.12' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hdsp - info leak in snd_hdsp_hwdep_ioctl()
  ALSA: us122l: Fix pcm_usb_stream mmapping regression
  ALSA: hda - Fix inverted internal mic not indicated on some machines

Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/linux-security

Pull apparmor fixes from James Morris:
 "A couple more regressions fixed"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
  apparmor: fix bad lock balance when introspecting policy
  apparmor: fix memleak of the profile hash