staging:iio:adc:ad7816 move from old to current event handling.

I'm not sure what the event is and am very much against the dodgy
hack to give it a code. However for now, lets just stop it using
the deprecated core handling so we can move on.

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

staging:iio:adc:ad7745 move from old to current event handling.

Nice simple one.  Not sure we actually want devices outputting
datardy signals like that, but I'll leave it for now.
Cleaned up some unneeded functions whilst here.

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

staging:iio:adt7316 get rid of legacy event handling code.

Note this driver is still a long way from being abi compliant.
What I have done here cleans up a few corners, but primarily gets
it away from using the infrastructure that is going away.

Untested

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

staging:iio:dac:ad5504 move from old to new event handling.

Untested, but fairly trivial change here.

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

staging:iio:light:tsl2563 remove old style event registration.

Untested.

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

staging:iio:buffering remove unused parameter dead_offset from read_last_n in all buffer implementations.

This element has been usused by the core for quite some time.  sca3000 set it none the less
until the rewrite in the previous patch (and hence didn't work).

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

staging:iio:sca3000 extract old event handling and move to poll for events from buffer

Fairly substantial rewrite as the code had bitrotted.
A rethink is needed for how to handle variable types in the new chan_spec world.

This patch restores sca3000 buffer usage to a working state.
V3: Rebase fixups.
V2: Move to new version of IIO_CHAN macro

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

staging:iio: lis3l02dq - separate entirely interrupt handling for thesholds from that for the datardy signal.

This removes the one and only real user of the rather complex event list management.
V3: More trivial rebase fixups.
V2: Trivial rebase fixup.

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

staging:iio: Push interrupt setup down into the drivers for event lines.

It is much easier to do in driver, and the core does not add much.
Note all drivers will have to be updated with this patch.
None currently are.

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

staging:iio:lis3l02dq: General cleanup

As Arnd observed, things are clearner if we pass iio_dev into read and write fucntions.

Now uses st for lis3l02dq_state everywhere.

Other bits of trivial tidying.

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

staging:iio: Buffer device flattening.

Given we now only have one device we don't need the extra layer any more.
Hence this patch removes it.

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

staging:iio: remove legacy event chrdev for the buffers

part of sca3000 driver temporarily disabled (buffer won't run
anyway).  This section is replaced later in this patch set.

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

staging:iio: Add polling of events on the ring access chrdev.

Staging one of combining the ring chrdevs.

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

staging:iio: remove ability to escalate events.

Whilst it is possible to output events to say buffers have passed
a particular level there are no obvious reasons to actually do so.

The upshot of this patch is that buffers will only ever have
one threshold turned on at a time.

For now sca3000 has it's ring buffer effectively disabled.
Fixed later in series.

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

staging:iio:max1363 - move to channel_spec registration.

V3: move to single chan registration macro. Also introduce some
local macros to greatly reduce code length when setting up the
chan_spec arrays for all the different devices.

V2: update read_raw for two value approach.

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

staging:iio:lis3l02dq - move to new channel_spec approach.

V3: Move to single IIO_CHAN macro.
V2: Update to two part read_raw value.

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

staging:iio: allow channels to be set up using a table of iio_channel_spec structures.

V8: Add missing address in IIO_CHAN macro. Spotted by Michael Hennerich.
V7: Document additions to iio_dev structure.
V6: Fixup the docs for iio_chan_spec structure.
V5: Actually have the macro handle the _input type channels (oops)
V4: Add ability to do, _input and modified channel naming in a coherent fashion.
    Scrap all the messy IIO_CHAN_* macros and move to only one.

V3: Added more types - intensity and light.

V2: Various fixes - some thanks to Arnd.
    Bug fix for unregistering of event attr group
    Changed iio_read_channel_info to have two part value - use for
    raw value read as well.
    constify the channelspec structures
    raw write support for calibbias and similar
    Additional strings for buidling attribute names.

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

Staging: initial version of the nvec driver

This is an implementation of a NVidia compliant embedded controller
protocol driver. It is used on some ARM-Tegra boards for device
communication.

Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: sm7xx: printk loglevels modified to match the scenario

KERN_ERR should be used in place of KERN_INFO in the case of
error scenarios.

Signed-off-by: anish kumar <anish198519851985@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: sm7xx: preferred form for passing a size to memory allocation routines

The preferred form for passing a size of a struct is the following:
p = kmalloc(sizeof(*p), ...);
Please refer Documentation/Codingstyle chapter 14

Signed-off-by: anish kumar <anish198519851985@gmail.com>
Acked-by: Harry Wei <harryxiyou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: sm7xx: removing extra white spaces,redundant code and using macros

This patch is to remove extra spaces,redundant code and using
ARRAY_SIZE macros.

Signed-off-by: anish kumar <anish198519851985@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: fix mei build when PM is not enabled

Fix mei build when CONFIG_PM is not enabled (i.e., fix typo):

drivers/staging/mei/main.c:1159: error: 'MEI_PM_OPS' undeclared here (not in a function)

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: bind-driver.c: mark remote_host as unused

Use __attribute__((unused)) to suppress error until it can be determined
that remote_host is not needed.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: stub_driver.c: update kernel module name

Change kernel module name to usbip-host.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: support comedi auto-configuration

Add support for automatically associating a vmk8055 device with a
comedi device (previously the user had to use comedi_num_legacy_minors
to reserve device slots and then associate them with vmk8055 devices
using comedi_config).
Tested on multiple K8055s, but not on K8061s.

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: support bits instruction

Calling comedi_dio_bifield2() returns EBUSY permanently. Implementing
the insn_bits call fixes the problem and is good in its own right since
one can then read and write to all the digitial lines at the same time.
Tested on a K8055, but not on a K8061.

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: extend rudimentary_check to check both directions

rudimentary_check() can currently check whether the input or
output direction is currently available (no pending transaction),
but not both at the same time. We need this facility for do_bits().

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: simplify rinsn output calculation

vmk80xx_di_rinsn() and vmk80xx_do_rinsn() extract the required channel
data by inconsistent and overly-complex algorithms. Simplify them both.

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: Digitial I/O should have a maxdata of 1

Digitial input and output sub-devices were reporting a maxdata of
0x1F and 0xFF respectively. They should both be 1.

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

comedi vmk80xx: make rudimentary_check a static function

rudimentary_check is a macro with side-effects (it returns
on error) which is contary to CodingStyle. Replace it with
a static function.

Signed-off-by: J. Ali Harlow <ali@avrc.city.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

drivers/staging/cptm1217/clearpad_tm1217.c: Correct call to input_free_device

This code is in a loop that currently is only executed once.  Because of
this property, the first block of code is currently actually correct.
Nevertheless, the comments associated with the code suggest that the loop
is planned to take more than one iteration in the future, and thus this
patch is made with that case in mind.

In the first block of code, there is currently an immediate abort from the
function.  It is changed to jump to the error handling code at fail, to be
able to unregister and free the resources allocated on previous iterations.

In the second block of code, the input_dev for the current iteration has
been allocated, but has not been registered.  It has also not been stored
in ts->cp_input_info[i].input.  Thus on jumping to fail, it will not be
freed.  In this case, we want to free, but not unregister, so the free for
this most recently allocated resource is put before the jump.

A simplified version of the semantic match that finds this problem is:
(http://coccinelle.lip6.fr/)

// <smpl>
@r exists@
local idexpression struct input_dev * x;
expression ra,rr;
position p1,p2;
@@

x = input_allocate_device@p1(...)
...  when != x = rr
     when != input_free_device(x,...)
     when != if (...) { ... input_free_device(x,...) ...}
if(...) { ... when != x = ra
     when forall
     when != input_free_device(x,...)
 \(return <+...x...+>; \| return@p2...; \) }

@script:python@
p1 << r.p1;
p2 << r.p2;
@@

cocci.print_main("input_allocate_device",p1)
cocci.print_secs("input_free_device",p2)
// </smpl>

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

staging/easycap: easycap_probe: drop more unused variables

wMaxPacketSize are  bEndpointAddress assigned but not used

Cc: Mike Thomas <rmthomas@sciolus.org>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/easycap: probe: simplify the endpoints tests

we are interested only in isochronous in endpoints
so we can simplify the flow

Cc: Mike Thomas <rmthomas@sciolus.org>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/easycap: easycap_probe: drop verbose printouts

reduce printouts of not necessary information

Cc: Mike Thomas <rmthomas@sciolus.org>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: add mei to staging Kbuild

Add mei to Kconfig and Makefile in drivers/staging

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: AMT Watchdog

code that open connection and invoke
heartbeats to the AMT Watchdog client/feature, if exists

Connect to WD Client, if exists Send Start WD Command.
Every 2 secs send heartbeats.
On System shutdown/suspends, send Stop WD command.

This is intermediate stage before moving this code to standalone watchdog
driver.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: mei.h defining user space interface

define IOCTL_MEI_CONNECT_CLIENT and its associated structure

When the user wants to connect to a ME feature/client after
it open a file descriptor to the driver, he need to use Connect
IOCTL.

This IOCTL received a struct that contains a union of 2 other structs.

1st struct - Input Parameters:
UUID - a predefine unique that identify the ME feature, this
 id per feature is constant all over the chipsets
 and versions.

2nd struct Output Parameters:
MaxMessageLen -  maximum message length that allowed
to be send to the feature
ProtocolVersion ME feature current protocol version.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: Hardware and MEI driver internal struct definition

define the MEI protocol msg structs and
HW registers, also define the MEI internal status and struct

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: MEI driver init flow.

Init driver list and queue, MEI Hardware reset flow,
init of driver specific host client.

MEI Init/reset flow:
- Ack all waiting interrupts
- Hardware reset flow (Set Reset Bit, Generate Interrupt, Clear Reset Bit
  Generate Interrupt)
- Wait for ME Ready Bit (done in interrupt thread)
- Set ME Ready Bit (done in interrupt thread)
- Send Start request (done in interrupt thread)
- wait for answer
- Send Enumerate Clients request (done in interrupt thread)
- wait for answer
- Send Get Client property for each client request (done in interrupt thread)
- Wait for answers
- Init Done.

MEI Driver connect internally to 2 ME clients/features:
AMTHI and AMT watchdog.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: MEI link layer

Implementation of the communication between host and ME.
connect/disconnect to/from a client, send MEI message,
read MEI message, flow control handling.

Each MEI message has mei_msg_hdr followed by a payload.
Driver is oblivious the payload.

ME Address/ID  - This is the logical address of the ME
feature/client of that message.
Host Address/ID - This is the logical address of the Host
  client of that message
Length  - This is the Length of message payload in bytes
Reserved  -  reserved for future use.
Message Complete - This bit is used to indicative that
this is the last message of multi message
MEI transfer of a client message that is larger
then the MEI circular buffer.
Payload - Message payload (data) up to 512bytes

The HW data registers are consist two circular buffers,
one for data from ME and other data from Host application.
Each buffer has two pointers, read_ptr (H_CBRP)
and write_ptr (H_CBWP).
The buffers size is defined by depth value that exists
in the status registers (H_CBD and ME_CBD_HRA).
Every read from ME circular buffer cause read_ptr++
Every write to the Host  circular buffer write_ptr++

Flow control MEI message that ME and MEI Driver use to notify
each other that a ME feature/client or Host client buffer is ready
to receive data.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: Interrupt handling.

ISR and interrupt thread for handling incoming data.
e.g. read bus message, read client message, handle reset requests.

quick handler:
As MEI may share interrupt with GFX and/or USB
the HW register need to be checked and acknowledged.

thread handler:
Check if HW has data for read.
Write data to HW if possible.
May init reset flow on error

there can be two types of messages:
1) bus messages:
Management messages between MEI Driver and ME e.g.
Connect request/response,
Disconnect request/response
Enum clients request/response
Flow control request/response
those message are indicated by
ME Address/ID == 0 && Host Address/ID == 0

2) feature/client messages:
message that are sends between ME Feature/Client and
an application, the struct of the message is defined
by the ME Feature Protocol (e.g. APF Protocol, AMTHI Protocol)
those message are indicated by
ME Address/ID != 0 && Host Address/ID != 0

MEI Initialization state machine is also managed by this patch.
After MEI Reset is preform:
Send Start request
wait for answer
Send Enumerate Clients request
wait for answer
Send Get Client property for each client request
wait for answers
Init Done.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: PCI device and char driver support.

contains module entries and PCI driver and char device
definitions (using file_operations, pci_driver struts).

The HW interface is exposed on PCI interface.
PCI:
The MEI HW resources are memory map 32 bit registers
(Host and ME Status Registers and Data Registers)
and interrupt (shared, with Intel GFX on some chipsets
and USB2 controller on others).
The device is part of the chipsets and cannot be hotplugged.
The MEI device present is determined by BIOS configuration.

Probe:
The driver starts the init MEI flow, that is explained
in the patch "MEI driver init flow" [06/10],
then schedules a timer that handles
timeouts and watchdog heartbeats.

Remove:
The driver closes all connections and stops the watchdog.

The driver expose char device that supports:
open, release, write, read, ioctl, poll.

Open:
Upon open the driver allocates HOST data structure
on behalf of application which will resides in the file's
private data and assign a host ID number which
will identify messages between driver client instance
and MEI client.

The driver also checks readiness of the device. The number
of simultaneously opened instances is limited to 253.
(255 - (amthi + watchdog))

Release:
In release the driver sends a Disconnect Command to
ME feature and clean all the data structs.

IOCTL:
MEI adds new IOCTL: (IOCTL_MEI_CONNECT_CLIENT)
The IOCTL links the current file descriptor to ME feature.
This is done by sending MEI Bus command: 'hbm_client_connect_request'
to the ME and waiting for an answer :'hbm_client_connect_response'.
Upon answer reception the driver updates its and HOST data
structures in file structure to indicate that the file
descriptor is associated to ME feature.

Each ME feature is represented by UUID which is given as
an input parameter to the IOCTL, upon success connect command the
IOCTL will return the ME feature properties.
ME can reject CONNECT commands due to several reasons,
most common are:
Invalid UUID ME or feature does not exists in ME.
No More Connection allowed to this is feature,
usually only one connection is allowed.

Write:
Upon write, the driver splits the user data into several MEI
messages up to 512 bytes each and sends it to the HW.
If the user wants to write data to AMTHI ME feature then the
drivers routes the messages through AMTHI queues.

Read:
In read the driver checks is a connection exists to
current file descriptor and then wait until a data is available.
Message might be received (by interrupt from ME) in multiple chunks.
Only complete message is released to the application.
Poll:
Nothing special here. Waiting for see if we have
data available for reading.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Itzhak Tzeel-Krupp <itzhak.tzeel-krupp@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: MEI Driver TODO list

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/mei: MEI Driver documentations

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Oren Weil <oren.jer.weil@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix transport other code style

fix keucr transport.c other coding style but not from checkpatch.pl.
replace ternary conditional "?:" with if/else

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr transport coding style

fix keucr transport.c transport.h coding style

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr smil.h coding style

fix keucr smil.h coding style

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr smilecc.c coding style

fix keucr smilecc.c coding style

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr scsiglue coding style

fix keucr scsiglue.c coding style

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr msscsi coding style

Fix keucr msscsi.c coding style.
Remove externs ,and move MS_SCSIIrp to end,
because there are not necessary to add extern for MS_SCSIIrp function.

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging/keucr: fix keucr ms coding style

fix keucr ms.c and ms.h coding style

Signed-off-by: Cho, Yu-Chen <acho@novell.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: vmbus_drv: Get rid of util synchronixation based on channel counting

Now that we have dealt with this issue differently, get rid of the
old mechanism.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Re-implement the synchronization for util channels

The util module expects that the util channels are fully initialized
when the module loads. To deal with the race condition which can result
in a NULL pointer dereferencing if the util module were to load before
all the util channels are fully initialized, in commit:

commit: 8b5d6d3bd3e34e4cc67d875c8c88007c1c9aa96
Author: Haiyang Zhang <haiyangz@microsoft.com>
Date:   Fri May 28 23:22:44 2010 +000

code was introduced in the vmbus driver to ensure that all the
util channels were fully initialized before returning from the load
of the vmbus driver. This solution has several problems: if for whatever
reason, any util channel were to fail to initialize, vmbus driver would
wait indefinitely. We deal with this synchronization issue very differently
in this patch.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: vmbus_drv: Introduce state in struct vmbus_channel to track util services

In preparation for getting rid of util channel synchronization based on
counting util channels, introduce state in  struct vmbus_channel  to
track util services.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: comedi: addi-data: Get rid of redundant ps_BoardInfo

The ps_BoardInfo pointer in the device private data is redundant as we
can just use the this_board macro to access the same data, as is done
elsewhere in the code.  Get rid of the pointer and change the code to
use the this_board macro instead.

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

staging: comedi: addi-data: Don't overwrite read-only data

The drivers for ADDI-DATA cards can override some static parameters for
the board type using information read from EEPROM.  Unfortunately, they
currently write the parameters from the EEPROM back to the shared,
read-only board data!  The problem has been masked during compilation by
type-casting away the const-ness of the data.

This patch changes the code to use an area in the private data for the
board instance to hold the parameters read from EEPROM (after
initializing the parameters from the static board data).  It also
changes the type-casts to the read-only data to preserve the const
qualifier.

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

staging: comedi: adv_pci_dio: Add a counter subdevice to PCI-1751

The Advantech PCI-1751 has a 8254 counter chip on board.  Add it to the
device as a counter subdevice.  Apparently the counter can generate
interrupts although the driver does not currently use this capability.

Original patch by Ivan Russkih (Иван Русских) <vanekrus at gmail dot
com>.

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

drivers/staging/line6/driver.c: Drop unneeded put functions.

This seems to be the result of patches ab366c1a and 1027f476 crossing each
other.  Patch ab366c1a adds calls to usb_put_intf and usb_put_dev at the
end of the function line6_probe, in the error handling code, while patch
1027f476 moves the calls to the corresponding get function from the
beginning to the end of line6_probe, making the calls to put in the error
handling code unnecessary.

// </smpl>

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

ath6kl: iw dev wlan0 link implementation

implementing the cfg ops that gets called when iw dev wlan0
link is issued by user. The ops that needs to be implemented
is get_station.

kvalo: check the mac address, remove signal_pending(), use ARRAY_SIZE()
and fix style issues

Signed-off-by: Naveen Singh <nsingh@atheros.com>
Signed-off-by: Kalle Valo <kalle.valo@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

drivers/staging/ath6kl/os/linux/cfg80211.c: Add missing call to cfg80211_put_bss

A call to cfg80211_get_bss hould be accompanied by a call to
cfg80211_put_bss in error-handling code.

A simplified version of the semantic match that finds this problem is:
(http://coccinelle.lip6.fr/)

// <smpl>
@r exists@
local idexpression struct cfg80211_bss * x;
expression ra,rr;
position p1,p2;
@@

x = cfg80211_get_bss@p1(...)
...  when != x = rr
     when != cfg80211_put_bss(x,...)
     when != if (...) { ... cfg80211_put_bss(x,...) ...}
if(...) { ... when != x = ra
     when forall
     when != cfg80211_put_bss(x,...)
 \(return <+...x...+>; \| return@p2...; \) }

@script:python@
p1 << r.p1;
p2 << r.p2;
@@

cocci.print_main("cfg80211_get_bss",p1)
cocci.print_secs("return",p2)

// </smpl>

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

staging: brcm80211: use correct pointer types in bcm_pktq_flush

to clarify the intention:

On Sun, May 15, 2011 at 09:54:26PM +0200, Arend van Spriel wrote:
> On 05/15/2011 08:10 PM, Hauke Mehrtens wrote:
[...]
>> Why don't you use the correct pointer type here instead of casting then
>> around and the parameter names are meaningless too.
>>
>> static bool cb_del_ampdu_pkt(struct sk_buff *mpdu, struct
>> cb_del_ampdu_pars *ampdu_pars)
[...]
> Actually this is a generic callback mechanism in which an additional
> parameter can be passed, which can be of *any* type hence a void pointer
> seems justified here and a meaningless name is used. I do agree that the
> txi parameter will always be a struct sk_buff and should be indicated as
> such.

Reported-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Clemens Noss <cnoss@gmx.de>
Acked-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: fix cast to pointer from integer

bcm_pktq_flush and related functions only ever get 0 or a pointer for
arg, so make it a pointer.

This might fix a crash on 64bit.

Signed-off-by: Clemens Noss <cnoss@gmx.de>
Acked-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: echo: Trivial indentation changes

--bcaec50161b317392804a33628dd
Content-Type: text/plain; charset=ISO-8859-1

Hi Greg,

 Please find patch attached .

Thanks,
 Rahul

Hi Greg,<br><br> Please find patch attached .<br><br>Thanks,<br> Rahul<br>

From ba1dbafaef1a87c1338c5bb0e6bba3be06071c90 Mon Sep 17 00:00:00 2001
From: Rahul Tank <rahul.tank@gmail.com>
Date: Sat, 14 May 2011 11:28:25 +0530
Subject: [PATCH] Trivial indentation cleanup
Signed-off-by: Rahul Tank <rahul.tank@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: modify project gcc flags

Add -Werror, remove -Wstrict-prototypes, and change -W to -Wextra.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: move common cflags to configure.ac

There are a standard set of cflags that are used in each makefile so
let's set those in EXTRA_CFLAGS and enforce them for the entire project.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: remove gcc warnings

The warnings were for unused parameters, so __attribute__((unused))
has been added until it can be determined they are truly unneeded.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: remove revision $Id$

This is git not CVS!

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: set libusbip version in configure.ac

Move libusbip version setting to configure.ac so that version
numbers can be found in a single location.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: cleanup makefiles

Remove unneeded comments; change deprecated flag INCLUDE to
AM_CPPFLAGS and put -D option in *_CPPFLAGS; and use "simply
expanded variables" in assignments.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: configure.ac: major overhaul

All parameters have been quoted; and autoscan was rerun so new
headers, types, and functions were added. The deprecated macros
AM_CONFIG_HEADER and AM_PROG_LIBTOOL were changed to
AC_CONFIG_HEADERS and LT_INIT, respectively. The AS_HELP_STRING
macro is used to avoid arbitrary spacing for proper help menu
alignment, and AS_CASE to avoid quoting issues. And finally, the
macros were realigned to allow mere mortals the ability to read them.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: update module name references

Change all references to the kernel modules to correspond with the new
names.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: whitespace cleanup

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: remove unused files

Remove ChangeLog and NEWS.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: rename source directories

Rename cmd/ to src/ and lib/ to libsrc/.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: userspace: move manpages to separate directory

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: usbip: add userspace code

Add userspace code from svn trunk revision 175. Located at
https://usbip.svn.sourceforge.net/svnroot/usbip/linux/trunk/src.

Signed-off-by: matt mooney <mfm@muteddisk.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Make netvsc_drv_exit() the module exit function

Make netvsc_drv_exit() the module exit function.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Make netvsc_drv_init the module init function

Now make netvsc_drv_init the module init function.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Move the dmi_check code to netvsc_drv_init()

In preparation to eliminating netvsc_init(), move the dmi_check code
to netvsc_drv_init().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Change the signature of netvsc_drv_init()

Change the signature of netvsc_drv_init().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Move the dmi table declaration to earlier in the file

Move the dmi table declaration to earlier in the file.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of the used type struct netvsc_driver

Now, get rid of the unused type  struct netvsc_driver.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of the unused function drv_to_netvscdrv()

Get rid of the unused function drv_to_netvscdrv().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of the empty function rndis_filter_init()

Get rid of the empty function rndis_filter_init().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of some unnecessary/dead code from rndis_filter.c

After the cleanup that has been done, some code in rndis_filter.c
is no longer needed; get rid of it.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Make netvsc_drv an instance of struct hv_driver

In preparation to getting rid of struct  netvsc_driver, make
the variable netvsc_drv an instance of struct hv_driver.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of the unnecessary state from struct netvsc_driver

Now get rid of unused state (ring_buf_size) from struct netvsc_driver.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of the unnecessary assignment of ring_buf_size

Since we pass ring size information differently, this assignment is
not needed.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get the ring size information from struct netvsc_device_info

Get the ring size information from struct netvsc_device_info.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Add ring_size element to struct netvsc_device_info

Add ring_size element to struct netvsc_device_info.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of unused state in struct netvsc_driver

Now, get rid of unused state (req_ext_size) from struct netvsc_driver.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of an unnecessary assignment in rndis_filter.c

Now that we compute the size of struct rndis_filter_packet where
it is needed, get rid of the unnecessary assignment.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Directly get the size of rndis_filter_packet

Directly get the size of rndis_filter_packet.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Move the definition of struct rndis_filter_packet to hyperv_net.h

Move the definition of struct rndis_filter_packet to hyperv_net.h.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Get rid of unnecessary state in struct netvsc_driver

Since the function pointers in struct netvsc_driver are no longer used
(as these are invoked directly), get rid of this unused state.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Directly invoke the function rndis_filter_receive()

Directly invoke the function rndis_filter_receive().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Directly invoke the function netvsc_recv_callback()

Directly invoke the function netvsc_recv_callback().

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Invoke netvsc_linkstatus_callback directly

Invoke netvsc_linkstatus_callback directly.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Invoke netvsc_send() directly

Invoke netvsc_send() directly.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: netvsc_drv: Invoke rndis_filter_send() directly

In preparation to getting rid of struct netvsc_driver,
make the function  rndis_filter_send()  non-static.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Abhishek Kane <v-abkane@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>