staging: xgifb: make RAM type tables const

RAM type tables are read-only and can be made const.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move XGINew_DataBusWidth into vb_device_info

Move the memory bus width info to vb_device_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move XGINew_ChannelAB into vb_device_info

Move the memory channel info into vb_device_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move XGINew_RAMType into vb_device_info

The RAM type is device specific, so move it into vb_device_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate global XGI_Pr data

Move vb_device_info into xgifb_video_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: make XGIbios_mode const

Comments are wrong, the table is read-only and can be made const.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate filter_tb global variable

filter_tb is only used inside a single function, and it does not need
to be static.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate XGIfb_fix

Eliminate XGIfb_fix and initialize needed fields of fb_info->fix
in probe().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate default_var

Eliminate default_var and initialize needed fields of fb_info->var
in probe().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate XGIfb_CRT2_write_enable variable

Replace XGIfb_CRT2_write_enable with IND_XGI_CRT2_WRITE_ENABLE_315.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move pseudo_palette into xgifb_video_info

pseudo_palette should be dynamically allocated for each fb.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move XGIfb_rate_idx into xgifb_video_info

The current refresh rate index should be stored in the device-specific
data allocated in probe().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: move xgifb_mode_idx into xgifb_video_info

The current mode index should be stored in the device-specific data
allocated in probe().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate XGIfb_mode_no

We can get the information from the table with mode index.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: dynamically allocate xgifb_video_info

Delete global xgi_video_info and dynamically allocate the data.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: avoid direct references xgi_video_info

Avoid direct references to global xgi_video_info. This will help changing
the data from statically allocated to dynamically allocated.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: rename struct video_info to xgifb_video_info

Rename struct video_info to xgifb_video_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: add a global variable for the default refresh rate

Add a global variable for the default refresh rate. This is done to
get rid of references to xgi_video_info before the probe routine, which
should allocate the xgi_video_info dynamically in the future.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: inline XGIfb_query_VGA_config_space()

XGIfb_query_VGA_config_space() is used only once during the init and
can be replaced with a single PCI configuration space read.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate global XGIhw_ext data

Move xgi_hw_device_info into xgi_video_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: eliminate global fb_info variable

Move fb_info into xgi_video_info.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: xgifb: use PCI drvdata

Don't access the global xgi_video_info directly in xgifb_remove().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: zcache: reduce tmem bucket lock contention

tmem uses hash buckets each with their own rbtree and lock to
quickly lookup tmem objects.  tmem has TMEM_HASH_BUCKETS (256)
buckets per pool.  However, because of the way the tmem_oid is
generated for frontswap pages, only 16 unique tmem_oids are being
generated, resulting in only 16 of the 256 buckets being used.
This cause high lock contention for the per bucket locks.

This patch changes SWIZ_BITS to include more bits of the offset.
The result is that all 256 hash buckets are potentially used resulting in a
95% drop in hash bucket lock contention.

Signed-off-by: Seth Jennings <sjenning@linux.vnet.ibm.com>
Acked-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove brcm80211 driver from the staging tree

With the mainline patch being applied to the wireless-next repository
by John Linville this driver is no longer needed under the staging
directory. This patch ends its life under the staging tree.

Cc: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Fix coding style issues in InterfaceMisc.c

This patch cleans up several code style issues found
in InterfaceMisc.c reported by checkpatch.pl.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Remove assignment from if conditions reported by checkpatch.pl

This patch removes an assignment from three if
conditions.  In all three cases, the line of code was
attempting to allocate memory, and check
if the memory was allocated in the if statement.
This issue was reported by checkpatch.pl.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Fix coding style issues reported by checkpatch.pl

This patch cleans up several code style issues found
in InterfaceInit.c reported by checkpatch.pl.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Fix assignment issue in if statement reported by checkpatch.pl.

This patch fixes an error where an assignment "="
was being used in an if statement to determine if
Firmware was downloaded. This patch removes that
assignment, and places it above the if statement.
The if statement then evaluates the status to
verify if "0" successful, or != 0 failure.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Remove assignment in if condition reported by checkpatch.pl

There is a case where an assignment was being done
in an if condition. This patch removes the assignment
from the if condition and places the assignment above
the if statement; thereby improving the readability of
the code.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Replace dated variable __FUNCTION__ reported by checkpatch.pl

This patch replaces the obsolete variable, __FUNCTION__, that
holds the name of the current function with variable, __func__.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: bcm: Fix coding style issues reported by checkpatch.pl

This patch cleans up several hundred code style issues found
in Misc.c reported by checkpatch.pl.

Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Update TODO list

Added some more tasks to the TODO list, as highlighted by Francois
Romieu <romieu@fr.zoreil.com>.

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Remove unused mac defines

Remove some unused defines from et1310_mac.c.

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Fix et131x smatch issues

This fixes the following issues:

et131x_initpci.c +488 et131x_adjust_link(45) error: we previously assumed 'phydev' could be null.
et131x_initpci.c +504 et131x_adjust_link(61) warn: variable dereferenced before check 'phydev'

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Fix et131x sparse warnings

This fixes the following warnings:

et1310_mac.c:375:44: warning: incorrect type in initializer (different address spaces)
et1310_mac.c:382:28: warning: incorrect type in argument 2 (different address spaces)
et1310_mac.c:384:31: warning: incorrect type in argument 2 (different address spaces)
et131x_initpci.c:555:5: warning: symbol 'et131x_mii_probe' was not declared. Should it be static?
et131x_netdev.c:267:5: warning: symbol 'et131x_ioctl' was not declared. Should it be static?
et131x_netdev.c:285:5: warning: symbol 'et131x_set_packet_filter' was not declared. Should it be static?
et131x_netdev.c:347:6: warning: symbol 'et131x_multicast' was not declared. Should it be static?
et131x_netdev.c:420:5: warning: symbol 'et131x_tx' was not declared. Should it be static?
et131x_netdev.c:453:6: warning: symbol 'et131x_tx_timeout' was not declared. Should it be static?
et131x_netdev.c:511:5: warning: symbol 'et131x_change_mtu' was not declared. Should it be static?
et131x_netdev.c:564:5: warning: symbol 'et131x_set_mac_addr' was not declared. Should it be static?

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Remove file et131x_version.h

Removed defines in et131x_version.h and replaced them by actual
strings where convinient, or moved them to et131x.h

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Move pm calls from pci device to driver device

Move the pci driver suspend/resume calls up to the driver.pm ops
structure, as they are not pci device specific.

Thanks to Francois Romieu <romieu@fr.zoreil.com> for pointing this out.

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Remove error path from suspend/resume code

Removing an error path from et131x suspend/resume functions.

Also added a call to phy_stop() to complement the phy_start() call
during device start/stop routine.

Thanks to Francois Romieu <romieu@fr.zoreil.com> for pointing this out.

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Removing Olaf Hartmann's email as it bounces

On 3 October 2011 18:51, Uwe Ranft <uwe.ranft@telozo.com> wrote:
> Hello,
> olaf is not more longer owner of this e-mailadress. he has left our
> company.
> Please remove olaf fron the mailing list!
> Best Regards
> Uwe

Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Fix possible out-of-bounds index with TKIP and AES keys

Array XGrpKey has only 2 elements and uses (keyid - 1) as the index, which
allows the possibility of memory corruption from an out-of-bounds index.

This problem was reported by a new version of smatch.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: zcache: fix crash on cpu remove

In the case that a cpu is taken offline before zcache_do_preload() is
ever called on the cpu, the per-cpu zcache_preloads structure will
be uninitialized.  In the CPU_DEAD case for zcache_cpu_notifier(),
kp->obj is not checked before calling kmem_cache_free() on it.
If it is NULL, a crash results.

This patch ensures that both kp->obj and kp->page are not NULL before
calling the respective free functions. In practice, just checking
one or the other should be sufficient since they are assigned together
in zcache_do_preload(), but I check both for safety.

Signed-off-by: Seth Jennings <sjenning@linux.vnet.ibm.com>
Acked-by: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155v4l: silence a compiler warning

The '!' is dead code that was left over from a previous version.
It causes a compiler warning:
drivers/staging/dt3155v4l/dt3155v4l.c:335:3: warning: value computed
is not used

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

easycap: fix ntsc module parameter description

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

staging:iio:dac:ad5638: Fix channel address

Commit c6fc806247 ("staging:iio: ABI rework - add in_ or out_ prefix to
channnels") added the AD5868_CHANNEL macro to simplify channel initialization.
Unfortunately the macro hardcodes the channel's address to AD5686_ADDR_DAC0. As
a result writing to any of the channels will change the value of the first
channel.

This patch fixes the issue by calculating the channel address based on the
channel number.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:dac/dds/impedance dev_info to indio_dev for consistency

We had a random missmatch of these two.  Lets pick the most common
and get rid of the other.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
For adxxxx parts
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:adc:naming: dev_info to indio_dev for consistency

We had a random missmatch of these two.  Lets pick the most common
and get rid of the other.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:accel:sca3000 dev_info to indio_dev renaming.

Again, there is a mix of the two names for the struct iio_dev.
Lets pick one and run with it.

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

staging:iio:core:naming: dev_info to indio_dev for consistency

We had a random missmatch of these two.  Lets pick the most common
and get rid of the other.  This patch covers the core.  Others
will clean up the drivers.

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

Staging: iio/adc/ad7150: release lock on error

The error_ret label should have been before the mutex_unlock().  It's
a typo.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio/light/tsl2563: fix compile warning

This fixes a compile warning:
drivers/staging/iio/light/tsl2563.c:696:2:
warning: initialization from incompatible pointer type [enabled by default]
drivers/staging/iio/light/tsl2563.c:696:2:
warning: (near initialization for ‘tsl2563_info.write_event_value’) [enabled by default]

The tsl2563_write_thresh() function returns zero on success and error
codes on failure, so nothing is lost by making the return type int
instead of ssize_t.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio/light/tsl2563: unlock on an error path

We need to unlock here before returning.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: remove struct hv_device_info from hyperv.h

This is only used/needed by the vmbus core code, so move it out of the
hyperv.h file and into the .c file that uses it.

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

hv: remove free_channel() from hyperv.h

This function is only used in the file it is declared in
(channel_mgmt.c) so make it static and remove it from the hyperv.h file.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: hyperv.h: remove unneeded forward declarations of structures

This file was created by mushing different .h files together and it
shows.  This change removes some unneeded forward declarations.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: hyperv.h: remove unused module macros

I have no idea what these were ever for, but they aren't used, so delete
them.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: remove unused LOWORD and HIWORD macros from hyperv.h

They aren't used anywhere anymore now that the debugging macros are
gone, so remove it from hyperv.h as well.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove vmbus_loglevel as it is not used at all anymore

As there is no user of this variable, it's time to delete it.  For
dynamic debugging of the hyperv code, use the standard dynamic debug
kernel interface.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove last user of DPRINT() macro

This also removed the unused function hv_dump_ring_info().

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: storvsc: remove last usage of DPRINT_WARN

Used the correct dev_warn() call instead.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: remove a bunch of unused debug macros from hyperv.h

These aren't used by anyone anymore, so remove them before someone tries
to use them again.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

hv: rename prep_negotiate_resp() to vmbus_prep_negotiate_resp()

It's a global symbol, so properly prefix it and use the proper EXPORT
value as well.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove unneeded asm include file in hyperv.h

No one outside of the hyperv core needs to include the asm/hyperv.h
file, so don't put it in the "global" include/linux/hyperv.h file.

Cc: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: move hyperv code out of staging directory

After many years wandering the desert, it is finally time for the
Microsoft HyperV code to move out of the staging directory.  Or at least
the core hyperv bus code, and the utility driver, the rest still have
some review to get through by the various subsystem maintainers.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>

Staging: hv: netvsc: Fix a dereferencing issue

net_dev is being de-referenced without proper checking; fix it.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: storvsc: ignore SET_WINDOW scsi command

Some commands sent by smartd will offline the device.
With this change applied, smartd sill not monitor the device anymore.

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Acked-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: util: Invoke cn_netlink_send() in a work context

Invoke cn_netlink_send() in a work context as opposed being called
in the context of  channel callback. On entry into the channel callback
code the channel inbound spin lock is held and deferring to a work
context avoids having to invoke cn_netlink_send() while holding
the inbound lock. As part of this adjustment, also increase the
timeout value for waiting for the user level component of KVP.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: util: Fix a bug in kvp implementation

The host gurantees that there can be only one kvp transaction active
against the guest. So, the transaction active state is needed only to
protect against spurious user level calls. The current code had a race
condition where the guest could prematurely return because the previous
transaction state was not cleared - this state was being cleared after
sending the response to the host and there was a window where the host
could notify the guest of a new transaction before the transaction active
state was properly set.
Also deal with the case when the user mode component
does not respond in a timely fashion correctly.
I would like to thank Long Li <longli@microsoft.com>
for identifying the problem.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Diagnosed-by: Long Li <longli@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s1210 cleanup gpio handling.

Basically make use of the gpio array functions.
Technically the free doesn't need as much info
as given here, but this keeps it clean and easy
to follow.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s1210 drop raw config register access

This should never have been exposed to userspace

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s1210 minimal chan spec conversion.

Just convert the raw reads in this patch.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s1210 ensure iio_dev->name is set

This is needed to ensure the required name attribute is
created. Using an id table is the route most consistent
with other drivers.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver rename ad2s120x ->ad2s1200

There are far too many possible part numbers that are
not covered in here to have such a generic name.
So move to naming after one of the supported parts.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s120x chan spec conversion

Straight forward conversion.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:Documentation gyro -> anglvel updates in attribute names

Follows directly from change made in previous patch.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio: rename gyro channels to anglvel

Ensure naming reflects what is measured, not how it is done.
Resolvers can measure the same thing for starters.
IIO_GYRO->IIO_ANGL_VEL to ensure consistent naming.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s90 chan spec conversion

Trivial conversion.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s90 ensure name is passed to iio_core.

Scraps the pointless name define and adds an id table.
It's not technically required in drivers with only one supported
part but it does make the probe code more consistent.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:resolver:ad2s90 fix registration of null pointer

When the new allocation code was introduced a stray pointer
to iio_dev structure was left in the chip state structure.
This was never set but was then registered with the core.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging:iio:lisght:isl29018 use IIO_PROCESSED enum value.

No functional change, just a trivial tidy up so all drivers
do the same thing.

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

staging:iio:adc:ad7314 removal. Supported via hwmon.

Driver ported over to hwmon where it fits much better.

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove brcmf_op_if from fullmac

Absorb related code into brcmf_add_if/brcmf_del_if. This is part
of the net device interface clean up.

Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: clean up struct brcmf_if in fullmac

Some elements in structure brcmf_if are unused at all or
redundant. This patch cleans it up.

Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: cleanup driver variable references

Throughout the data structures within the driver several references
to the driver variables were stored. As the storage and access
functions are now consolidated into srom.c those references are
no longer needed and have been removed.

Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove parameter 'off' from _initvars_srom_pci()

The function is called once with parameter 'off' fixed to zero. This
parameter has been removed.

Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: replace string based variable storage by linked list

The storage of variables obtained from srom are now stored in a linked
list and lookup is done based on the enumerated identifier.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: use enum identifiers in srom variable tables

In the srom variable tables fields were identified using string
identifiers. This has been replaced by enumeration identifiers.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: use identifiers instead of string for srom lookup

The driver variables from srom are stored in memory with string
identifiers and the caller needed to use this string identifier
to retrieve a driver variable. The commit changes the calling
code replacing strings with enumerated identifiers.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove invalid variable lookup from srom

In the driver several driver variables were looked up that are
not supported by the srom code. These have been removed.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Brett Rudley <brudley@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: fix thread blocking issue in brcmf_sdbrcm_bus_stop()

The function brcmf_sdbrcm_bus_stop() terminates the watchdog and dpc
thread, but this function can be called from the dpc thread itself.
Stopping the dpc thread is only done when it is not the 'current'
thread.

Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove locking macro definitions

The driver contained locking macros which map directly to the
lock function provided by the kernel. It does not provide any
additional value so these have been removed.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: change parameter in driver variable lookup

The functions getvar() and getintvar() had to pass the buffer
containing the driver variables. Now they pass the structure
containing this buffer hiding what type of buffer/container
is used for storing the driver variables.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove code duplication for driver variable lookup

The phy code used its own driver variable lookup functions which were
duplicates of those in the brcmsmac driver (in srom.c). The phy code
now uses the functions in srom.c through the phy_shim interface.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove redundant bus register layer from fullmac

Remove some single line functions for bus register/unregister interface

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove threads_only code from fullmac

threads_only is always true as we never use the tasklet implementation.
So related code is removed.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: move driver variable functions to srom.c

The driver uses variables which are stored in string format. Using
strings as variable identifiers is disliked by the community. The
driver has been cleaned up and the only module providing these
variables is srom.c. The variable retrieval functions have been
moved to srom.c in preparation of a more likable way to store and
lookup these driver variables.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: smac: fix for 64 bit systems

A bug was introduced by the following commit (Sep 13):

    staging: brcm80211: use endian annotated structures in brcmsmac

Result was that 64 bits systems will not be able to acquire an IP address.
Also the rmmod crashed. This has been fixed and retested on a Sparc64.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: softmac: added event tracing

Dpc and timer events can now be traced. Combined with Mac80211
driver callback tracing, all entry points into the driver can now
be traced.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Alwin Beukers <alwin@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove N mode validation function

Removed the brcms_c_nmode_validate function as it only checks N-mode
capability, which should always be true.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: combined if statements

Combined if statements in brcms_c_edcf_setparams and brcms_c_wme_setparams functions.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove brcms_b_dotxstatus wrapper function

- removed brcms_b_dotxstatus
- changed 'fatal' argument of brcms_b_txstatus from INOUT to OUT.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>