- remove the constant '1'
- when the device is not connected, the protocol error
  HIDPP_ERROR_RESOURCE_ERROR is raised. We should not warn the user about
  it because it is somewhat expected as an answer when we check if the
  device is connected.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Reported by Dan Carpenter:

drivers/hid/hid-logitech-hidpp.c:359 hidpp_root_get_protocol_version() warn: should this return really be negated?
drivers/hid/hid-logitech-hidpp.c:398 hidpp_devicenametype_get_count() warn: should this return really be negated?
drivers/hid/hid-logitech-hidpp.c:417 hidpp_devicenametype_get_device_name() warn: should this return really be negated?
drivers/hid/hid-logitech-hidpp.c:524 hidpp_touchpad_get_raw_info() warn: should this return really be negated?

The problem lies in hidpp_send_message_sync() which can return 2 types of
errors depending of their sign. Adding a comment there to clarify what is
happening.

To solve that, print an error in case of a protocol problem, and raise
-EPROTO instead.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Shift the allocation down a few lines to avoid a memory leak and also
add a check for allocation failure.

Fixes: 2f31c525 ('HID: Introduce hidpp, a module to handle Logitech hid++ devices')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

The TK820 presents both a keyboard and a touchpad on the same
physical (and logical device). Use the generic hid-input
processing for the keyboard part. The keyboard input device is created
when the receiver is plugged in, so no events are missed on connect.

When the device actaully connects, we can set it to use the raw
multitouch reporting to have a consistent user experience accross
all Logitech touchpads.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

This touchpad differs from the T650 in several ways:
- the resolution is not correctly returned by the device
- it presents physical buttons, so the button flag in the raw touch report
  is not filled.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

All the bits are now in place to add the support of the
Touchpad T650.
The creation/population of the input device is delayed until
the device is ready.

The T650 uses the special HID++ reporting protocol, so activate
this on connect.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Now that the receiver forwards the connect/disconnect events, we can
know when the device is available to communicate with us.

When it is ready, we can for instance retrieve its full name, which
guarantee that we always have the same name for the DJ device (the DJ
name is somewhat shorter than the HID++ name).

This mechanism is mandatory for the touchpads line, which has the
min/max information stored in the device. This information can only
be retrieved when the device is connected. So we can not populate
the input device until we are sure that the device is connected.

This patch creates a new input device for such devices. However,
this input is not bound to hid directly, so the various drivers
which wants to use it are required to process completely the
incoming reports in .raw_event().

Note that the patch in itself just adds the bits for the next
ones, and this feature is disabled by default.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

The receiver can send HID++ notifications to the DJ devices when the
physical devices are connected/disconnected.
Enable this feature by default.

This command uses a HID++ command instead of a DJ one, so use a direct
call to usbhid instead of using logi_dj_recv_send_report()
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

The names of the DJ devices are stored in the receiver. These names
can be retrieved through a HID++ command. However, the protocol says
that you have to ask the receiver for that, not the device iteself.

Introduce a special case in the DJ handling where a device can request
its unifying name, and when such a name is given, forward it also to
the corresponding device.

On the HID++ side, the receiver talks only HID++ 1.0, so we need to
implement this part of the protocol in the module.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

HID++ is a Logitech-specific protocol for communicating with HID
devices. DJ devices implement HID++, and so we can add the HID++
collection in the report descriptor and forward the incoming
reports from the receiver to the appropriate DJ device.

The same can be done in the other way, if someone calls a
.raw_request(), we can forward it to the correct dj device
by overriding the device_index in the HID++ report.
Signed-off-by: Benjamin Tisssoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Devices connected through the Logitech Wireless Receiver are HID++ devices.
We can handle them here to benefit from this new module and activate
enhaced support of the various wireless touchpad or mice with touch
sensors on them.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Logitech devices use a vendor protocol to communicate various
information with the device. This protocol is called HID++,
and an exerpt can be found here:
https://drive.google.com/folderview?id=0BxbRzx7vEV7eWmgwazJ3NUFfQ28&usp=shar

The main difficulty which is related to this protocol is that
it is a synchronous protocol using the input reports.
So when we want to get some information from the device, we need
to wait for a matching input report.

This driver introduce this capabilities to be able to support
the multitouch mode of the Logitech Wireless Touchpad T651
(the bluetooth one). The multitouch data is available directly
from the mouse input reports, and we just need to query the device
on connect about its caracteristics.

HID++ and the touchpad features has a specific reporting mode
which uses pure HID++ reports, but Logitech told us not to use
it for this specific device. During QA, they detected that
some bluetooth input reports where lost, and so the only supported
mode is the pointer mode.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

There is no point in keeping the header in a separate file, nobody
but hid-logitech-dj should have access to its content.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Several benefits here:
- we can drop the macro is_dj_device: I never been really conviced by
  this macro as we could fall into a null pointer anytime. Anyway time
  showed that this never happened.
- we can simplify the hid driver logitech-djdevice, and make it aware
  of any new receiver VID/PID.
- we can use the Wireless PID of the DJ device as the product id of the
  hid device, this way the sysfs will differentiate between different
  DJ devices.
Signed-off-by: Benjamin Tisssoires <benjamin.tissoires@redhat.com>
Tested-by: Andrew de los Reyes <adlr@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

This allows the transport layer (I have in mind hid-logitech-dj and uhid)
to set the group before it is added to the hid bus. This way, it can
bypass the hid_scan_report() call, and choose in advance which driver
will handle the newly created hid device.
Signed-off-by: Benjamin Tisssoires <benjamin.tissoires@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

While merging wacom from the input to the hid tree, some
comments have been duplicated. We can also integrate the
test for Synaptics devices in the switch case below, so
it is clear that there will be only one place for such
quirks.

No functional changes are expected in this commit.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Currently unknown consumer page codes are ignored, which means that they cannot
later be mapped from userspace using udev / hwdb. Map them to KEY_UNKNOWN, so
that userspace can remap them for keyboards which make up their own consumer
page codes.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>

Pull HID updates from Jiri Kosina:

 - quirk for devices that need to be pulled in much more aggresive way
   than mandated, by Johan Hovold

 - robustification of sanity checking of incoming reports in RMI driver,
   by Benjamin Tissoires

 - fixes, updates, and new HW support to SONY driver, by Frank Praznik

 - port of uHID to the new transport layer layout, by David Herrmann

 - robustification of Clear-Halt/reset in USB HID, by Alan Stern

 - native support for hopefully any future HID compliant wacom tablet.
   Those found on the various laptops (ISDv4/5) already are HID
   compliant and they should work in the future without any modification
   of the kernel.  Written by Benjamin Tissoires.

 - a lot more simple fixes and device ID additions all over the place

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: (45 commits)
  HID: uHID: fix excepted report type
  HID: usbhid: add another mouse that needs QUIRK_ALWAYS_POLL
  HID: wacom: implement the finger part of the HID generic handling
  HID: wacom: implement generic HID handling for pen generic devices
  HID: wacom: move allocation of inputs earlier
  HID: wacom: split out input allocation and registration
  HID: wacom: rename failN with some meaningful information
  HID: sony: Update the DualShock 4 touchpad resolution
  HID: wacom: fix timeout on probe for some wacoms
  HID: sony: Set touchpad bits in the input_configured callback
  HID: sony: Update file header and correct comments
  HID: sony: Corrections for the DualShock 4 HID descriptor
  HID: rmi: check sanity of the incoming report
  HID: wacom: make the WL connection friendly for the desktop
  HID: wacom - enable LED support for Wireless Intuos5/Pro
  HID: wacom - remove report_id from wacom_get_report interface
  HID: wacom - Clean up of sysfs
  HID: wacom - Add default permission defines for sysfs attributes
  HID: usbhid: fix PIXART optical mouse
  HID: Add Holtek USB ID 04d9:a0c2 ETEKCITY Scroll
  ...

Pull "trivial tree" updates from Jiri Kosina:
 "Usual pile from trivial tree everyone is so eagerly waiting for"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (39 commits)
  Remove MN10300_PROC_MN2WS0038
  mei: fix comments
  treewide: Fix typos in Kconfig
  kprobes: update jprobe_example.c for do_fork() change
  Documentation: change "&" to "and" in Documentation/applying-patches.txt
  Documentation: remove obsolete pcmcia-cs from Changes
  Documentation: update links in Changes
  Documentation: Docbook: Fix generated DocBook/kernel-api.xml
  score: Remove GENERIC_HAS_IOMAP
  gpio: fix 'CONFIG_GPIO_IRQCHIP' comments
  tty: doc: Fix grammar in serial/tty
  dma-debug: modify check_for_stack output
  treewide: fix errors in printk
  genirq: fix reference in devm_request_threaded_irq comment
  treewide: fix synchronize_rcu() in comments
  checkstack.pl: port to AArch64
  doc: queue-sysfs: minor fixes
  init/do_mounts: better syntax description
  MIPS: fix comment spelling
  powerpc/simpleboot: fix comment
  ...

Pull documentation updates from Jiri Kosina:
 "Updates to kernel documentation.

  I took this over (hopefully temporarily) from Randy who was not
  willing to maintain it any longer.  This pile mostly is a relay of
  queue that Randy already had in his tree"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/doc:
  Documentation: fix broken v4l-utils URL
  Documentation: update include path for mpssd
  Documentation: correct parameter error for dma_mapping_error
  MAINTAINERS: update location of linux-doc tree
  Documentation: remove networking/.gitignore
  tools: add more endian.h macros
  Make Documenation depend on headers_install
  Docs: this_cpu_ops: remove redundant add forms
  Documentation: disable vdso_test to avoid breakage with old glibc
  Documentation: update vDSO makefile to build portable examples
  Documentation: update .gitignore files
  Documentation: support glibc versions without htole macros
  v4l2-pci-skeleton: Only build if PCI is available
  Documentation: fix misc. warnings
  Documentation: make functions static to avoid prototype warnings
  Documentation: add makefiles for more targets
  Documentation: use subdir-y to avoid unnecessary built-in.o files

Pull spi updates from Mark Brown:
 "A quiet release for SPI, mainly driver updates and not too many of
  them:

   - Support for dummy transfers (for delays on startup) in drivers
     using transfer_one().
   - Lots of enhancements to the Designware driver to support new Intel
     SoCs.
   - Support for newer Renesas chips.
   - DMA support for the i.MX driver.
   - One new driver for Broadcom BCM53xx chips"

* tag 'spi-v3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (64 commits)
  spi: spi-mxs: fix a tiny typo in a comment
  spi: dw-mid: follow new DMAengine workflow
  spi: dw-mid: convert to use DMAengine wrappers
  spi: dw-mid: change magic numbers to the constants
  spi: orion: support armada extended baud rates
  spi: fsl: Sort include headers alphabetically
  spi: bcm53xx: Add missing module information
  spi: bcm53xx: Fix module dependency
  spi/rockchip: fix bug that cause the failure to read data in DMA mode
  spi: fsl-dspi: Remove probe info message
  spi: pl022: Add support for chip select extension
  spi: Fix possible ZERO_SIZE_PTR pointer dereferencing error.
  spi: dw: fix style of code in few places
  spi: dw: introduce support of loopback mode
  spi: dw-mid: terminate ongoing transfers at exit
  spi: dw-mid: respect 8 bit mode
  spi: clps711x: Migrate to the new clk subsystem
  spi: pl022: Add missing error check for devm_kzalloc
  spi: spi-imx: add DMA support
  spi: davinci: add support for adding delay between word's transmissions
  ...

Pull regulator updates from Mark Brown:
 "This time around most of the changes are a lot of new drivers along
  with the standard set of fixes and cleanups (thanks again largely to
  Axel Lin).  We do have one nice new feature in the core which factors
  out the disappointingly tricky code around DT parsing, only a couple
  of drivers have been converted so far:

   - Factor out the code for parsing the standard bindings for a set of
     regulators out of DT, making the probe part of a lot of drivers
     simplier.
   - New drivers for Dialog DA9213, HiSilicon HI6420, Intersil
     ISL9305/H, Ricoh RN5T618, Rockchip RK808, Skyworks SKY81452,
     Silergy SYR82x, and Qualcomm RPM"

* tag 'regulator-v3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator: (71 commits)
  regulator: da9211: Fix a bug in update of mask bit
  regulator: pwm-regulator: add devicetree bindings for pwm regulator
  regulator: pwm-regulator: get voltage and duty table from dts
  regulator: qcom_rpm: Fix FORCE_MODE_IS_2_BITS macro
  regulator: qcom_rpm: Don't explicitly initialise the first field of config
  regulator: ltc3589: fix broken voltage transitions
  regulator: qcom-rpm: Regulator driver for the Qualcomm RPM
  regulator: axp20x: Use parent device as regulator configuration device
  regulator: fan53555: Fix null pointer dereference
  regulator: fan53555: Fixup report wrong vendor message
  regulator: fan53555: fix wrong cast in probe
  regulator: fan53555: add support for Silergy SYR82x regulators
  regulator: fan53555: add devicetree support
  regulator: add devicetree bindings for Fairchild FAN53555 regulators
  regulator: rk808: Add function for ramp delay for buck1/buck2
  regulator: fan53555: use set_ramp_delay to set the ramp up slew rate
  regulator: fan53555: enable vin supply
  regulator: rk808: Fix missing of_node_put
  regulator: rk808: Remove unused variables
  regulator: of: Add stub OF match function for !OF case
  ...

Pull regmap updates from Mark Brown:
 "The main update this time around is the addition of a standard DT
  binding for specifying the endianness of devices.  This allows drivers
  to support any endianness of device register map without any code,
  useful for configurable IP blocks.

  There's also a few bug fixes that I didn't get round to sending, none
  of them terribly severe or new, and a reduction in size for struct
  regmap"

* tag 'regmap-v3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
  regmap: Fix debugfs-file 'registers' mode
  regmap: fix possible ZERO_SIZE_PTR pointer dereferencing error.
  regmap: debugfs: fix possbile NULL pointer dereference
  regmap: fix NULL pointer dereference in _regmap_write/read
  regmap: fix NULL pointer dereference in regmap_get_val_endian
  regmap: cache: Do not fail silently from regcache_sync calls
  regmap: change struct regmap's internal locks as union
  regmap: Split regmap_get_endian() in two functions
  regmap: of_regmap_get_endian() cleanup
  regmap: Fix DT endianess parsing logic
  regmap: Add explicit dependencies to catch "select" misuse
  regmap: Restore L: linux-kernel@vger.kernel.org entry
  regmap: Add the DT binding documentation for endianness
  regmap: add DT endianness binding support.

Pull pin control changes from Linus Walleij:
 "This is the bulk of pin control changes for the v3.18 development
  series:

   - New drivers for the Freescale i.MX21, Qualcomm APQ8084 pin
     controllers.

   - Incremental new features on the Rockchip, atlas 6, OMAP, AM437x,
     APQ8064, prima2, AT91, Tegra, i.MX, Berlin and Nomadik.

   - Push Freescale drivers down into their own subdirectory.

   - Assorted sprays of syntax and semantic fixes"

* tag 'pinctrl-v3.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (48 commits)
  pinctrl: specify bindings for pins and groups
  pinctrl: nomadik: improve GPIO debug prints
  pinctrl: abx500: refactor DT parser to take two paths
  pinctrl: abx500: use helpers for map allocation/free
  pinctrl: alter device tree bindings for functions
  pinctrl: nomadik: refactor DT parser to take two paths
  pinctrl: nomadik: use utils map free function
  pinctrl: nomadik: use util function to reserve maps
  pinctrl: qcom: use restart_notifier mechanism for ps_hold
  pinctrl: sh-pfc: sh73a0: Remove unnecessary SoC data allocation
  pinctrl: berlin: fix the dt_free_map function
  pinctrl: at91: disable PD or PU before enabling PU or PD
  pinctrl: st: remove gpiochip in failure cases
  pinctrl: at91: Fix error handling while doing gpiochio_irqchip_add
  pinctrl: at91: Fix failure path in at91_gpio_probe path
  pinctrl: lantiq: Release gpiochip resources in fail case
  pinctrl: imx: detect uninitialized pins
  pinctrl: tegra: Add MIPI pad control
  pinctrl: at91: Switch to using managed clk_get
  pinctrl: adi2: Remove duplicate gpiochip_remove_pin_ranges
  ...

Pull EDAC fixes from Borislav Petkov:
 "Misc small fixes/improvements accumulated from the summer months,
  nothing earth-shattering.

  The mpc85xx_edac fix is tagged for stable but since it got confirmed
  only days before the final 3.17 release, I didn't rush a last minute
  pull request to you.  I hope that's ok as it will trickle down to
  stable anyway and we'll make sure it lands in 3.17-stable too"

* tag 'edac_for_3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp:
  mpc85xx_edac: Make L2 interrupt shared too
  amd64_edac: Modify usage of amd64_read_dct_pci_cfg()
  ppc4xx_edac: Fix build error caused by wrong member access
  EDAC: Fix mem_types strings type

Pull dmaengine updates from Dan Williams:
 "Even though this has fixes marked for -stable, given the size and the
  needed conflict resolutions this is 3.18-rc1/merge-window material.

  These patches have been languishing in my tree for a long while.  The
  fact that I do not have the time to do proper/prompt maintenance of
  this tree is a primary factor in the decision to step down as
  dmaengine maintainer.  That and the fact that the bulk of drivers/dma/
  activity is going through Vinod these days.

  The net_dma removal has not been in -next.  It has developed simple
  conflicts against mainline and net-next (for-3.18).

  Continuing thanks to Vinod for staying on top of drivers/dma/.

  Summary:

   1/ Step down as dmaengine maintainer see commit 08223d80
      "dmaengine maintainer update"

   2/ Removal of net_dma, as it has been marked 'broken' since 3.13
      (commit 77873803 "net_dma: mark broken"), without reports of
      performance regression.

   3/ Miscellaneous fixes"

* tag 'dmaengine-3.17' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/dmaengine:
  net: make tcp_cleanup_rbuf private
  net_dma: revert 'copied_early'
  net_dma: simple removal
  dmaengine maintainer update
  dmatest: prevent memory leakage on error path in thread
  ioat: Use time_before_jiffies()
  dmaengine: fix xor sources continuation
  dma: mv_xor: Rename __mv_xor_slot_cleanup() to mv_xor_slot_cleanup()
  dma: mv_xor: Remove all callers of mv_xor_slot_cleanup()
  dma: mv_xor: Remove unneeded mv_xor_clean_completed_slots() call
  ioat: Use pci_enable_msix_exact() instead of pci_enable_msix()
  drivers: dma: Include appropriate header file in dca.c
  drivers: dma: Mark functions as static in dma_v3.c
  dma: mv_xor: Add DMA API error checks
  ioat/dca: Use dev_is_pci() to check whether it is pci device

Pull module update from Rusty Russell:
 "Nothing major: support for compressing modules, and auto-tainting
  params.

  PS. My virtio-next tree is empty: DaveM took the patches I had.  There
      might be a virtio-rng starvation fix, but so far it's a bit voodoo
      so I will get to that in the next two days or it will wait"

* tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
  moduleparam: Resolve missing-field-initializer warning
  kbuild: handle module compression while running 'make modules_install'.
  modinst: wrap long lines in order to enhance cmd_modules_install
  modsign: lookup lines ending in .ko in .mod files
  modpost: simplify file name generation of *.mod.c files
  modpost: reduce visibility of symbols and constify r/o arrays
  param: check for tainting before calling set op.
  drm/i915: taint the kernel if unsafe module parameters are set
  module: add module_param_unsafe and module_param_named_unsafe
  module: make it possible to have unsafe, tainting module params
  module: rename KERNEL_PARAM_FL_NOARG to avoid confusion

Merge ipmi patches from Corey Minyard:
 "These have been in linux-next for a while, ready for 3.18"

* emailed patches from Corey Minyard <minyard@acm.org>:
  ipmi: Clear drvdata when interface is removed
  ipmi: work around gcc-4.9 build warning
  ipmi/of: Don't use unavailable interfaces
  ipmi: Clean up the error handling for channel config errors

This patch fixes a bug on hotmod removing.

After ipmi interface is removed using hotmod, kernel panic occurs when
rmmod impi_si.  For example, try this:

 # echo "remove,"`cat /proc/ipmi/0/params` > \
 /sys/module/ipmi_si/parameters/hotmod
 # rmmod ipmi_si

Then, rmmod fails with the following messages.

  ------------[ cut here ]------------
  WARNING: CPU: 12 PID: 10819 at /mnt/repos/linux/lib/list_debug.c:53 __list_del_entry+0x63/0xd0()
  CPU: 12 PID: 10819 Comm: rmmod Not tainted 3.17.0-rc1 #19
  Hardware name: FUJITSU-SV PRIMERGY BX920 S2/D3030, BIOS 080015 Rev.3D81.3030 02/10/2012
  Call Trace:
    dump_stack+0x45/0x56
    warn_slowpath_common+0x7d/0xa0
    warn_slowpath_fmt+0x4c/0x50
    __list_del_entry+0x63/0xd0
    list_del+0xd/0x30
    cleanup_one_si+0x2a/0x230 [ipmi_si]
    ipmi_pnp_remove+0x15/0x20 [ipmi_si]
    pnp_device_remove+0x24/0x40
    __device_release_driver+0x7f/0xf0
    driver_detach+0xb0/0xc0
    bus_remove_driver+0x55/0xd0
    driver_unregister+0x2c/0x50
    pnp_unregister_driver+0x12/0x20
    cleanup_ipmi_si+0xbc/0xf0 [ipmi_si]
    SyS_delete_module+0x132/0x1c0
    system_call_fastpath+0x16/0x1b
  ---[ end trace 70b4377268f85c23 ]---

list_del in cleanup_one_si() fails because the smi_info is already
removed when hotmod removing.

When ipmi interface is removed by hotmod, smi_info is removed by
cleanup_one_si(), but is is still set in drvdata.  Therefore when rmmod
ipmi_si, ipmi_pnp_remove tries to remove it again and fails.

By this patch, a pointer to smi_info in drvdata is cleared when hotmod
removing so that it will be not accessed when rmmod.

changelog:
v2:
- Clear drvdata in cleanup_one_si
- Change subject

v1:
https://lkml.org/lkml/2014/9/8/741Signed-off-by: Takao Indoh <indou.takao@jp.fujitsu.com>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Building ipmi on arm with gcc-4.9 results in this warning for an
allmodconfig build:

  drivers/char/ipmi/ipmi_si_intf.c: In function 'ipmi_thread':
  include/linux/time.h:28:5: warning: 'busy_until.tv_sec' may be used uninitialized in this function [-Wmaybe-uninitialized]
    if (lhs->tv_sec > rhs->tv_sec)
       ^
  drivers/char/ipmi/ipmi_si_intf.c:1007:18: note: 'busy_until.tv_sec' was declared here
    struct timespec busy_until;
                    ^

The warning is bogus and this case can not occur.  Apparently this is a
false positive resulting from gcc getting a little smarter about
tracking assignments but not smart enough.

Marking the ipmi_thread_busy_wait function as inline gives the gcc
optimization logic enough information to figure out for itself that the
case cannot happen, which gets rid of the warning without adding any
fake initialization.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

If an IPMI controller is used by the firmware and as such marked with
a reserved status, we shouldn't use it.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The code to send the channel config errors was missing an error report
in one place and needed some more information in another, and had an
extraneous bit of code.  Clean all that up.
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull "tinification" patches from Josh Triplett.

Work on making smaller kernels.

* tag 'tiny/for-3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/josh/linux:
  bloat-o-meter: Ignore syscall aliases SyS_ and compat_SyS_
  mm: Support compiling out madvise and fadvise
  x86: Support compiling out human-friendly processor feature names
  x86: Drop support for /proc files when !CONFIG_PROC_FS
  x86, boot: Don't compile early_serial_console.c when !CONFIG_EARLY_PRINTK
  x86, boot: Don't compile aslr.c when !CONFIG_RANDOMIZE_BASE
  x86, boot: Use the usual -y -n mechanism for objects in vmlinux
  x86: Add "make tinyconfig" to configure the tiniest possible kernel
  x86, platform, kconfig: move kvmconfig functionality to a helper

Merge branches 'for-3.18/always-poll-quirk', 'for-3.18/logitech', 'for-3.18/picolcd', 'for-3.18/rmi', 'for-3.18/sony', 'for-3.18/uhid', 'for-3.18/upstream' and 'for-3.18/wacom' into for-linus

Pull SCSI fixes from James Bottomley:
 "This is a set of two small fixes, both to code which went in during
  the merge window: cxgb4i has a scheduling in atomic bug in its new
  ipv6 code and uas fails to work properly with the new scsi-mq code"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  [SCSI] uas: disable use of blk-mq I/O path
  [SCSI] cxgb4i: avoid holding mutex in interrupt context

Pull kconfig fixes for tiny setups from Josh Triplett:
 "Two Kconfig bugfixes for 3.17 related to tinification.  These fixes
  make the Kconfig "General Setup" menu much more usable"

* tag 'tiny/kconfig-for-3.17' of https://git.kernel.org/pub/scm/linux/kernel/git/josh/linux:
  init/Kconfig: Fix HAVE_FUTEX_CMPXCHG to not break up the EXPERT menu
  init/Kconfig: Hide printk log config if CONFIG_PRINTK=n

commit 03b8c7b6 ("futex: Allow
architectures to skip futex_atomic_cmpxchg_inatomic() test") added the
HAVE_FUTEX_CMPXCHG symbol right below FUTEX.  This placed it right in
the middle of the options for the EXPERT menu.  However,
HAVE_FUTEX_CMPXCHG does not depend on EXPERT or FUTEX, so Kconfig stops
placing items in the EXPERT menu, and displays the remaining several
EXPERT items (starting with EPOLL) directly in the General Setup menu.

Since both users of HAVE_FUTEX_CMPXCHG only select it "if FUTEX", make
HAVE_FUTEX_CMPXCHG itself depend on FUTEX.  With this change, the
subsequent items display as part of the EXPERT menu again; the EMBEDDED
menu now appears as the next top-level item in the General Setup menu,
which makes General Setup much shorter and more usable.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Cc: stable <stable@vger.kernel.org>

The buffers sized by CONFIG_LOG_BUF_SHIFT and
CONFIG_LOG_CPU_MAX_BUF_SHIFT do not exist if CONFIG_PRINTK=n, so don't
ask about their size at all.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Cc: stable <stable@vger.kernel.org>

Pull i2c fixes from Wolfram Sang:
 "Two i2c driver bugfixes"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: qup: Fix order of runtime pm initialization
  i2c: rk3x: fix 0 length write transfers