staging: mei: fix register access function comments

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

staging: mei: removing dependency between WD and AMTHI init function.

AMTHI need to be initialized after WD Client was initialized, moving the AMTHI outside
of the WD initialization function.
in order to remove the coupling between the initialization of those clients.

AMTHI is getting initialized (getting connected to the FW feature/client) in two ways:
 1) if mei driver fails to send connect message to watchdog client (WD initialization), then
    immediately the AMTHI client getting initialized right after the watchdog initialization function.
 2) if Watchdog client success to send connect message to watchdog client, then only after
    the driver is getting the connect response message the AMTHI client is getting initialized

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

staging: vme: change static device array to pointers

Change the static array of 'struct device''s in struct vme_bridge
to instead use an array of pointers. This is in accordance with the
requirement that all kobjects be dynamically allocated (see
Documentation/kobject.txt) and never be statically allocated.

Signed-off-by: Manohar Vanga <manohar.vanga@cern.ch>
Acked-by: Martyn Welch <martyn.welch@ge.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Most return-values changed from -1 to proper errno macros.

The ioctl handlers were frequently returning -1 upon failure. Most of
these have now been changed to proper errno macros.
The few remaining ones have been left untouched because either the
handler is not called (and so cannot be tested), or the function never
fails (and so cannot be system-tested), or requires new code to
distinguish its failures.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Cc: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Tracking kmemleak false positives.

The frame buffers allocated in _r8712_init_recv_priv() are not tracked
by kmemleak, but they are freed when the driver is unloaded. To help
find any real leaks, these false positives are annotated for kmemleak.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Tx aggregation.

Tx Aggregation (CONFIG_R8712_TX_AGGR, known as CONFIG_USB_TX_AGGR in the
Realtek tarball) is now added.

However, its tests have not been successful! The default in the Realtek
tarball is to not build it -- and the Release Notes does not seem to list this
as a feature. I have tested the driver with and without this feature; the
former does not successfully associate when WPA2 is used.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Various fixes.

In r8712_generate_ie(), HT check is no longer done.
Removed r8712_add_ht_addt_info().
Changes to defragmentation handling, and queue selection.
Get TSSI command.
Added an Ad-Hoc check to is_desired_network()
r8712_ind_disconnect() now checks for Linked state.
r8712_xmit_bh() now schedules an xmit tasklet.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Various fixes.

passive_mode added to private ioctls.
New private ioctls added for RSSI, link speed, mac address, scan type,
and DCE-D.
In r8711_wx_get_scan(), we now sleep only 100 times, instead of 1000.
In r8711_wx_set_essid(), added a check for Ad-Hoc state.
In r8711_wx_get_rate(), added a check for RTL8712_RF_2T2R RF type.
Added Set chplan.
In r871x_get_wireless_stats(), updated the mechanism for displaying
link quality.
Added SetDIG and SetRA commands.
r8712_joinbss_cmd() no longer checks for Ad-Hoc mode.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). New decl's and defines.

Removed unnecessary casts in kfree calls.
A few new types, defines and prototypes.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). tx: NULL _in_ spin lock.

In transmit path, r8712_free_xmitframe(), the pkt pointer is now
NULLed _within_ the spin lock.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Rework efuse data handling.

Rework efuse data handling.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Tx-code refactoring.

Refactor the TX code to prepare for aggregation.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Use stack, not malloc.

translate_scan() now uses the stack for some variables, instead of mallocing.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). stop_drv_timers moved.

stop_drv_timers() is now done in r871xu_dev_remove(), not netdev_close().

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). No power tracking OID.

The call to oid_rt_pro_set_power_tracking_hdl() is no longer available.
The function's definition remains, though it is currently unused.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Removed redundant .mem.

recv_frame.mem
mp_xmit_frame.mem
mp_xmit_packet.mem were not used, and so are now removed.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Updated include directives.

These are the new include directives for header files.
The following were (somehow!) not explicity in Realtek's, but proved necessary:
"linux/usb.h" is needed for usb_alloc_urb();
"linux/interrupt.h" is needed for tasklet_struct. It was often a nested
include.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Redundant NULL check.

Removed the redundant NULL check in LedControl871x.
As noted by Dan Carpenter, the address of a stack variable is never NULL.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Removed unused MSECS.

MSECS was never used in this driver. So removed.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Removed _usb_submit_urb.

Replaced (Realtek's) _usb_submit_urb by (linux's own) usb_submit_urb.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Removed _usb_alloc_urb.

Replaced (Realtek's) _usb_alloc_urb by (linux's own) usb_alloc_urb.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Removed os_free_netdev.

Replaced (Realtek's)  os_free_netdev by (linux's own) free_netdev.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Renamed _netdev_priv.

Renamed _netdev_priv to netdev_priv.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: r8712u: Merging Realtek's latest (v2.6.6). Copyright banners.

Only copyright banners have been added to these files. No functional changes.

Signed-off-by: Ali Bahar <ali@internetDog.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: zram: prevent accessing an unallocated table when init fails early

When the allocation of zram->table fails, we set zram->disksize to zero
to prevent accessing the unallocated table entries during cleanup.
However, we currently don't take this precaution when the initialization
fails earlier.

Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: zram: fix zram locking

Currently init_lock only prevents concurrent execution of zram_init_device()
and zram_reset_device() but not zram_make_request() nor sysfs store functions.

This patch changes init_lock into a rw_semaphore. A write lock is taken by
init, reset and store functions, a read lock is taken by zram_make_request().
Also, avoids to release the lock before calling __zram_reset_device() for
cleaning after a failed init, thus preventing any concurrent task to see an
inconsistent state of zram.

Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Update README file

I believe that the driver is up to date with the latest network driver changes, so removing this line from the TODO to see if anyone else disagrees.

Also all useless typecasts are now dead.

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

staging: et131x: Remove ai_force_[duplex|speed] from et131x_adapter

The ai_force settings were used to set the ethernet speed and duplex manually.
This is now being done by the phy_device, so remove.

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

staging: et131x: Remove struct mi_regs from et1310_phy.h

This structure is no longer used, and references registers defined in mii.h and other parts of et1310_phy.h

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

staging: et131x: Replace magic numbers in et1310_phy.c with defines

Replaced et131x_phy.c magic numbers specifying registers and their values with defines, defined in et131x_phy.h

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

staging: et131x: Remove duplicated register defines from et1310_phy.h

The VMI_* set of register defines are not used, and duplicate the set of PHY_* registers from the same file, which are used - hence removing.

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

staging: et131x: Remove cached_mask_value from et131x_adapter

cached_mask_value is only ever assigned, never read. Remove it.

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

staging: et131x: Remove registry_rx_mem_end from struct et131x_adapter

registry_rx_mem_end is not referenced anywhere in the driver.

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

staging: et131x: remove calls to netif_carrier_[on|off] from et131x_mii_check

As et131x_adjust_link is called from the phydev in response to a netif_carrier_[on|off], and is the only caller of et131x_mii_check, we don't need to call netif_carrier_[on|off] again.

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

staging: et131x: Remove PHY interrupt handling code from driver isr handler

The PHY interrupt is now handled by the phy_device, and the equivalent code is present in et131x_adjust_link, called from the phy_device when needed.

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

staging: et131x: Remove redundant struct adapter members

adapter->speed_duplex was never referenced.
adapter->registry_phy_coma was always set to zero, even in the vendor driver.

Removing all member references and associated dead code.

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

staging: et131x: Remove private adapter->duplex_mode and use phydev->duplex instead

The phy device keeps a note of the duplex mode, so use that value instead of duplex mode.

Also use the phydev defines for duplex modes, and remove the driver private ones.

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

staging: et131x: Remove private adapter->linkspeed and use phydev->speed instead

The phy device keeps a note of the link speed, so use that value instead of the driver private one.

Also use the phydev defines for link speeds, and remove the driver private ones.

adapter->hw_errs was never used, even in the vendor driver, so remove that too.

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

staging: et131x: Amend README file

I think the kernel style cleanups are done now.
Adding myself as a recipient of patches.

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

staging: et131x: Fix checkpatch whitespace warnings

Some of my previous hacking attempts have not been following the rules.
All fixes either lines > 80 chars or whitespace corrections (spaces->tabs etc).

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

staging: rtl8192e: Fix format warning

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Use MII register defines from mii.h

Use defines from include/linux/mii.h instead of et131x_phy.h and delete the latter defines.

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

staging/cxt1e1: Header cleanup

Remove dead and unused code in sbecom_inline_linux.h. This also fixes a
build failure under Turkish locale, where include/config/modversions.h
is incorrectly named as modversIons.h (an unrelated kbuild bug) and the
driver tries to needlessly include the empty file.

Reported-by: Serdar KÖYLÜ <s.koylux@gmail.com>
Cc: Bob Beers <bob.beers@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: sep: call to sep_ioctl() may leave driver in unusable state

If sep_ioctl() is called from a process that does not own
current transaction, it unlocks unheld sep->ioctl_mutex and
returns -EACCES leaving sep->sep_mutex acquired.

The patch fixes the mutex lock-unlock mismatch.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: panel: Fixed checkpatch warning about simple_strtoul()

Fixed the checkpatch warning about sing simple_strtoul instead of
kstrtoul() in panel.c.

Signed-off-by: Pelle Windestam <iceaway@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: et131x: Remove redundant phy code

Now we are using a phy_device, remove driver functionality that is now handled by the phydev.

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

staging: et131x: Remove xcvr_addr and et131x_xcvr_find

Use the phy_device equivalents instead.

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

staging: et131x: converting et131x_ioctl to use phy_mii_ioctl

Handing over ioctls handled by the driver to the phydev.

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

staging: mei: unlock dev->device_lock mutex on error path in mei_open()

mei_open() acquires dev->device_lock mutex and try to allocate mei_cl,
but if the allocation fails it goes to return statement. As a result
dev->device_lock left locked.

The patch fixes goto to unlock dev->device_lock mutex on this path.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: cleaned up softmac srom macro

Substituted macro.

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

staging: brcm80211: cleaned up softmac scb.h macro's

Substituted/removed macro's.

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

staging: brcm80211: cleaned up softmac pub.h macro's

Substituted/removed macro's.

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

staging: brcm80211: remove AP related code from softmac

AP support was only partially implemented and never invoked.
AP related code sections removed.

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

staging: brcm80211: cleaned up more softmac macro's

Substituted/removed macro's.

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

staging: brcm80211: cleaned up softmac nicpci.c/nicpci.h macro's

Substituted macro's.

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

staging: brcm80211: cleaned up softmac phy macro's affecting code flow

Substituted macro's.

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

staging: brcm80211: remove 'default n' from Kconfig

Kconfig options are by default not selected so adding the
statement 'default n' to the option is not needed.

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

staging: brcm80211: remove sparse warning from main.c

The source file main.c contained several sparse warnings which
have been cleaned up.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: use address space qualifier in brcmfmac

Several sparse warning were issued due to missing __user qualifier
for brcmfmac variables. This patch adds those.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: replace 0 integer with NULL pointer

Running sparse showed following warning: Using plain integer as NULL
pointer. This has been fixed.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove extern keyword from function definition

Sparse warning was given when using the extern keyword in the function
definition. This patch removes those instances.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: cleaned up several main.h/main.c related macro's

Moved, deleted or substituted macro's.

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

staging: brcm80211: cleaned up several softmac macro's

Moved, deleted or substituted macro's.

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

staging: brcm80211: cleaned up softmac channel related macro's

Moved, deleted or substituted macro's.

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

staging: brcm80211: macro cleanup in softmac ampdu.c

Moved, deleted or substituted macro's.

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

staging: brcm80211: macro cleanup in softmac aiutils.h

Moved, deleted or substituted macro's.

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

staging: brcm80211: removed redundant SI_INFO() macro from softmac

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

staging: brcm80211: removed redundant macro's in softmac

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

staging: brcm80211: use static qualifier for local symbols in brcmsmac

Cleanup resulted in merging several files and made several symbols
local to a source file. These were found by running sparse as it
gave the warning: symbol 'foo' was not declared. Should it be static?

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove sparse warnings from mac80211_if.c

Some changes have been made to get rid of sparse warnings in the
source file mac80211_if.c.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: reduce sparse messages on brcmsmac sources

Running sparse over the driver sources showed 1356 sparse messages
and quite a number are due to missing __iomem attribute in some
places. This patch reduces the number significantly.

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: use static qualifier for local symbols in brcmfmac

Cleanup resulted in merging several files and made several symbols
local to a source file. These were found by running sparse as it
gave the warning: symbol 'foo' was not declared. Should it be static?

Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: remove static function declaration in dhd_linux

Reshuffle function order in dhd_linux of fullmac to get rid of
static function declaration

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

staging: brcm80211: remove static function declaration in bcmsdh_sdmmc

Reshuffle function order in bcmsdh_sdmmc of fullmac to get rid of
static function declaration

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

staging: brcm80211: remove unused sdioh command type code in fullmac

sdioh command type is not used in brcmf_sdioh_request_word. This patch
removes the related code.

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

staging: brcm80211: remove unused sdioh data mode code in fullmac

sdioh data mode is not used in brcmf_sdioh_request_buffer. The patch
removes the related code.

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

staging: brcm80211: remove function wrapper of getting sb window address

The address stored in sdiodev can be used directly instead of using
the wrapper function brcmf_sdcard_cur_sbwad.

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

staging: brcm80211: remove duplicate set sb window address function

In fullmac, brcmf_sdbrcm_set_siaddr_window and brcmf_sdcard_set_sbaddr_window
have identical fucntionality. Remove the one in dhd_sdio.c

Reported-by: Rafal Milecki <zajec5@gmail.com>
Reviewed-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: consolidate SHOW_EVENTS code under BCMDBG in fullmac

Use unify config marco for debug info. Related event debug print out can be
enabled/disabled by the dedicated event message level.

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

staging: brcm80211: replaced fullmac BRCMF_PM_RESUME_WAIT macro with function

Code cleanup. Advantage is more readable code and better type checking.

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

staging: brcm80211: cleaned up fullmac macro BRCMF_PM_RESUME_RETURN_ERROR

Macros impacting the code flow are sort of frowned upon. No other macro's
in the source code impact the code flow. Replaced macro by function.

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

staging: brcm80211: remove static function prototypes from main.c

Function prototypes for static functions are not strictly needed and
considered unwanted by linux community. This patch reorders the
functions in brcmsmac/main.c and gets rid of the prototypes.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: use min_t() instead of min()

Also eliminate a cast to int that could cause issues with very large values
of di->rxbufsize (as suggested by Dan Carpenter).

Signed-off-by: Henry Ptasinski <henryp@broadcom.com>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in 3 files

nicpci.c, srom.c and stf.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in mac80211_if.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in dma.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in channel.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in antsel.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in ampdu.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in alloc.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: brcm80211: removed static function declarations in aiutils.c

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix various problems noted by smatch

Smatch reports the following problems:

  CHECK   drivers/staging/rtl8192e/rtllib_softmac.c
drivers/staging/rtl8192e/rtllib_softmac.c +3143 rtllib_softmac_free(4) info: redundant null check on ieee->pDot11dInfo calling kfree()

  CHECK   drivers/staging/rtl8192e/rtllib_module.c
drivers/staging/rtl8192e/rtllib_module.c +198 free_rtllib(6) info: redundant null check on ieee->pHTInfo calling kfree()

  CHECK   drivers/staging/rtl8192e/rtl819x_TSProc.c
drivers/staging/rtl8192e/rtl819x_TSProc.c +280 SearchAdmitTRStream(52) error: potential null derefence 'pRet'.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch warnings regarding large arrays on stack

Smatch reports the follwing warnings:

  CHECK   drivers/staging/rtl8192e/rtllib_rx.c
drivers/staging/rtl8192e/rtllib_rx.c +552 RxReorderIndicatePacket(6) warn: 'prxbIndicateArray' puts 1024 bytes on stack

  CHECK   drivers/staging/rtl8192e/rtl819x_TSProc.c
drivers/staging/rtl8192e/rtl819x_TSProc.c +40 RxPktPendingTimeout(9) warn: 'stats_IndicateArray' puts 1024 bytes on stack

  CHECK   drivers/staging/rtl8192e/r8192E_phy.c
drivers/staging/rtl8192e/r8192E_phy.c +859 rtl8192_phy_SwChnlStepByStep(11) warn: function puts 797 bytes on stack

  CHECK   drivers/staging/rtl8192e/rtllib_rx.c
drivers/staging/rtl8192e/rtllib_rx.c +552 RxReorderIndicatePacket(6) warn: 'prxbIndicateArray' puts 1024 bytes on stack

These are fixed by adding the arrays in question as a union in a struct used
by nearly all routines.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch error in r8192E_phy.c

Smatch reports the following:

  CHECK   drivers/staging/rtl8192e/r8192E_phy.c
drivers/staging/rtl8192e/r8192E_phy.c +940 rtl8192_phy_SwChnlStepByStep(92) error: potential null derefence 'CurrentCmd'.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch warning in rtl819x_BAProc.c

Smatch outputs the following warning:

  CHECK   drivers/staging/rtl8192e/rtl819x_BAProc.c
drivers/staging/rtl8192e/rtl819x_BAProc.c +88 rtllib_ADDBA(12) warn: variable dereferenced before check 'ieee'

As ieee cannot be NULL at this point, remove all such checks.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Silence smatch info in r8192E_cmdpkt.c

Smatch outputs the following informational message:

  CHECK   drivers/staging/rtl8192e/r8192E_cmdpkt.c
drivers/staging/rtl8192e/r8192E_cmdpkt.c +419 cmpk_message_handle_rx(77) info: ignoring unreachable code.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch warning in rtl_wx.c

Smatch reports the following warning:

  CHECK   drivers/staging/rtl8192e/rtl_wx.c

drivers/staging/rtl8192e/rtl_wx.c +262 r8192se_wx_set_radio(16) warn: inconsistent returns sem:&priv->wx_sem: locked (258) unlocked (262)

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch info output for r8192E_dev.c

Smatch reports the following informational output:

  CHECK   drivers/staging/rtl8192e/r8192E_dev.c
drivers/staging/rtl8192e/r8192E_dev.c +2404 ActUpdateChannelAccessSetting(13) info: ignoring unreachable code.
drivers/staging/rtl8192e/r8192E_dev.c +2404 ActUpdateChannelAccessSetting(13) info: ignoring unreachable code.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Fix smatch error and warning in rtl_core.c

Smatch shows the following errors:

  CHECK   drivers/staging/rtl8192e/rtl_core.c
drivers/staging/rtl8192e/rtl_core.c +600 rtl8192_qos_activate(7) warn: variable dereferenced before check 'priv'
drivers/staging/rtl8192e/rtl_core.c +1345 rtl8192_init(40) warn: 'dev->irq' was not released on error
drivers/staging/rtl8192e/rtl_core.c +2120 rtl8192_alloc_rx_desc_ring(43) error: potential null derefence 'entry'.
drivers/staging/rtl8192e/rtl_core.c +3010 rtl8192_pci_probe(153) warn: 'pmem_start' was not released on error

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: rtl8192e: Remove dead code involving MOVE_INTO_HANDLER

File rtl_core.c contains the statement "#define MOVE_INTO_HANDLER". As a
result, everything inside an "ifndef MOVE_INTO_HANDLER" is dead code and
can be removed.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>