staging: ozwpan: Remove useless return variables

This patch remove variables that are initialized with a constant,
are never updated, and are only used as parameter of return.
Return the constant instead of using a variable.

Verified by compilation only.

The coccinelle script that find and fixes this issue is:
// <smpl>
@@
type T;
constant C;
identifier ret;
@@
- T ret = C;
... when != ret
- return ret;
+ return C;
// </smpl>

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ozwpan: Add missing blanklines after declarations

Signed-off-by: James A Shackleford <shack@linux.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ozwpan: Change Maintainer

I will be not able to work on ozwpan driver anymore.
Remove my name & add Tateno as maintainer.

Signed-off-by: Rupesh Gujare <rupesh.gujare@atmel.com>
Acked-by: Shigekatsu Tateno <shigekatsu.tateno@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ozwpan: ozproto.c fix for "WARNING: Prefer ether_addr_copy() over memcpy() if the Ethernet addresses are __aligned(2)"

Fixed the warning message by replacing memcpy() with ether_addr_copy()

Signed-off-by: Anil Belur <askb23@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ozwpan: remove redundant NULL check for devs

The "devs" is a pointer to g_net_dev in ozmain.c.
g_net_dev has a default value as empty string.
So "devs" cannot be NULL, removes NULL check for "devs".

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Use time_before()

To be future-proof and for better readability the time comparisons are modified
to use time_before() instead of plain, error-prone math.

Signed-off-by: Manuel Schölling <manuel.schoelling@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lustre: ptlrpc: sec_config.c: Fix for possible null pointer dereference

There is otherwise a risk of a possible null pointer dereference.

Was largely found by using a static code analysis program called cppcheck.

Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lustre: lov: lov_obd.c: Fix for possible null pointer dereference

There is otherwise a risk of a possible null pointer dereference.

Was largely found by using a static code analysis program called cppcheck.

Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: check length of char array

This fixes the following sparse error:

drivers/staging/lustre/lustre/obdclass/llog_ioctl.c:388:39: error: incompatible types for operation (>)
drivers/staging/lustre/lustre/obdclass/llog_ioctl.c:388:39:    left side has type char *ioc_inlbuf2
drivers/staging/lustre/lustre/obdclass/llog_ioctl.c:388:39:    right side has type int

Signed-off-by: Andreas Platschek <platschek@ict.tuwien.ac.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: socklnd: Remove unnecessary return brackets

Remove unnecessary brackets from return statements in socklnd.c,
socklnd.h, socklnd_cb.c and socklnd_lib-linux.c to meet
kernel coding style.

Signed-off-by: Masaru Nomura <massa.nomura@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: socklnd: Remove unnecessary line continuation

Remove unnecessary line continuation '\' in socklnd.c to meet
kernel coding style.

Signed-off-by: Masaru Nomura <massa.nomura@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: socklnd: Fix indenting

Fixed indenting of if statement in socklnd.c to meet kernel coding style.

Signed-off-by: Masaru Nomura <massa.nomura@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: socklnd: Add a required space

Added a required space before open brace and parenthsis
in socklnd.c to meet kernel coding style.

Signed-off-by: Masaru Nomura <massa.nomura@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: socklnd: Remove unnecessary () from return statement

Remove unnecessary brackets from return statements in socklnd.c
to meet kernel coding style.

Signed-off-by: Masaru Nomura <massa.nomura@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre: fix sparse warnings in o2iblnd_cb.c

This patch fixes the following sparse warnings:

drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:44:1: warning: symbol 'kiblnd_tx_done' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:102:10: warning: symbol 'kiblnd_get_idle_tx' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:131:1: warning: symbol 'kiblnd_drop_rx' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:212:10: warning: symbol 'kiblnd_find_waiting_tx_locked' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:238:1: warning: symbol 'kiblnd_handle_completion' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:277:1: warning: symbol 'kiblnd_send_completion' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:296:1: warning: symbol 'kiblnd_handle_rx' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:457:1: warning: symbol 'kiblnd_rx_complete' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:527:13: warning: symbol 'kiblnd_kvaddr_to_page' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:699:1: warning: symbol 'kiblnd_setup_rd_iov' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:752:1: warning: symbol 'kiblnd_setup_rd_kiov' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:792:1: warning: symbol 'kiblnd_post_tx_locked' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:996:1: warning: symbol 'kiblnd_tx_complete' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1270:1: warning: symbol 'kiblnd_connect_peer' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1627:1: warning: symbol 'kiblnd_reply' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1814:1: warning: symbol 'kiblnd_thread_fini' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1828:1: warning: symbol 'kiblnd_peer_notify' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1934:1: warning: symbol 'kiblnd_handle_early_rxs' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1957:1: warning: symbol 'kiblnd_abort_txs' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:1993:1: warning: symbol 'kiblnd_finalise_conn' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2167:1: warning: symbol 'kiblnd_reject' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2178:1: warning: symbol 'kiblnd_passive_connect' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2452:1: warning: symbol 'kiblnd_reconnect' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2516:1: warning: symbol 'kiblnd_rejected' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2655:1: warning: symbol 'kiblnd_check_connreply' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:2754:1: warning: symbol 'kiblnd_active_connect' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:3025:1: warning: symbol 'kiblnd_check_conns' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:3108:1: warning: symbol 'kiblnd_disconnect_conn' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:3247:1: warning: symbol 'kiblnd_complete' was not declared. Should it be static?
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c:904:20: warning: context imbalance in 'kiblnd_post_tx_locked' - unexpected unlock

Signed-off-by: Zi Shen Lim <zlim.lnx@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre: fix sparse warning: Using plain integer as NULL pointer

Signed-off-by: Brilliantov Kirill Vladimirovich <brilliantov@inbox.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: fix sparse warning "should it be static"

Fix the following warning in lib-eq.c
warning: symbol 'lnet_eq_dequeue_event' was not declared. Should it be static?

Signed-off-by: Toby Smith <toby@tismith.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: fix sparse warning "unexpected unlock"

Fix the following warning in lib-eq.c
warning: context imbalance in 'lnet_eq_wait_locked' - unexpected unlock

Signed-off-by: Toby Smith <toby@tismith.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/lustre: remove Peng Tao from Lustre contact list

I am leaving emc for a new company where I cannot have much
time working on Lustre. So remove myself from the contact
list. I will still be around and continue to contribute as a
hobbyist though.

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

staging: bcm: fix lines over 80 characters

Signed-off-by: Greg DeAngelis <gdeangel@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: Indentation fixes in Bcmchar.c

Signed-off-by: Matthias Beyer <mail@beyermatthias.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: Outsourced IsFlash2x() handling

This patch outsources the code from the IsFlash2x() check in
bcm_char_ioctl_nvm_rw() function to shorten it.

Signed-off-by: Matthias Beyer <mail@beyermatthias.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: Outsourced cmd handling to function

This patch outsourced the true-branch for the IOCTL_BCM_NVM_READ command
handling to shorten the bcm_char_ioctl_nvm_rw() function.

Signed-off-by: Matthias Beyer <mail@beyermatthias.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: Fixed line lengths

Signed-off-by: Matthias Beyer <mail@beyermatthias.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm: use kzalloc instead of kmalloc/memset

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: binder: fix usage of uninit scalar in binder_transaction()

Fix the error path when a cookie mismatch is detected. In that case the
function jumps to the exit label without setting the uninitialized, local
variable 'return_error'. Detected by Coverity - CID 201453.

Signed-off-by: Christian Engelmayer <cengelma@gmx.at>
Acked-by: Arve <arve@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: binder: cleanup dereference of noderef expressions

Clean up sparse warnings for cred struct dereference.

Signed-off-by: Jerry Snitselaar <dev@snitselaar.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: remove rts5139 driver code

Roger writes:
Since all patches have been applied and the device is now
supported by the new driver, would you remove the former staging
one at drivers/staging/rts5139?

Cc: Roger Tseng <rogerable@realtek.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

imx-hdmi: Make checkpatch happy

WARNING: Missing a blank line after declarations
#160: FILE: drivers/staging/imx-drm/imx-hdmi.c:160:
+ u8 val = hdmi_readb(hdmi, reg) & ~mask;
+ val |= data & mask;

WARNING: Missing a blank line after declarations
#1609: FILE: drivers/staging/imx-drm/imx-hdmi.c:1609:
+ const struct platform_device_id *device_id = of_id->data;
+ hdmi->dev_type = device_id->driver_data;

total: 0 errors, 2 warnings, 1767 lines checked

drivers/staging/imx-drm/imx-hdmi.c has style problems, please review.

If any of these errors are false positives, please report
them to the maintainer, see CHECKPATCH in MAINTAINERS.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: et131x: Remove empty file Module.symvers

It was added by commit e0349d5ba (staging: et131x: Remove unused rcv_pend_lock spinlock)

Signed-off-by: Daniele Forsi <dforsi@gmail.com>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: et131x: add blank lines after declarations

Signed-off-by: Maarten de Jonge <mdejonge1990@gmail.com>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: et131x: Remove useless return variables

This patch remove variables that are initialized with a constant,
are never updated, and are only used as parameter of return.
Return the constant instead of using a variable.

Verified by compilation only.

The coccinelle script that find and fixes this issue is:
// <smpl>
@@
type T;
constant C;
identifier ret;
@@
- T ret = C;
... when != ret
- return ret;
+ return C;
// </smpl>

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: silicom: Remove unused pointer in bypass_init_module()

Pointer 'pbpctl_dev_c' in function bypass_init_module() is unused. Thus remove
it. With the last variable declaration gone, there is no more need for an own
block. Remove it and adapt the indenting accordingly.

Signed-off-by: Christian Engelmayer <cengelma@gmx.at>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: silicom: Remove needless calls of get_status_port_fn()

Remove a needless pointer initialisation and call to get_status_port_fn()
in functions remove_bypass_tpl_auto() and set_tpl_fn(). Variable
'pbpctl_dev_b' is set correctly later in the function before first use.

Signed-off-by: Christian Engelmayer <cengelma@gmx.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: silicom: add blank line after declarations

Add blank line after declarations and delete extra blank line at the
beginning of the function

Signed-off-by: Fernando Apesteguia <fernando.apesteguia@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxt1e1: remove set a value to static variable

cleanup checkpatch.pl error:
 ERROR: do not initialise statics to 0 or NULL

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxt1e1: Fix line length over 80 characters in musycc.c

clean up checkpatch.pl warning:
 WARNING: Line length over 80 characters

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxt1e1: remove redundant curly braces in musycc.c

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxt1e1: remove dead code in musycc.c

Removes "#if 0" blocks.

And the musycc_dump_rxbuffer_ring(ch, 0) which is commented
out puts in RLD_DEBUG block and uncommented.
Because this function may be used for debugging.

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxt1e1: Fix no spaces at the start of a line in musycc.c

clean up checkpatch.pl warning:
 WARNING: please no spaces at the start of a line in

Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: panel: (coding style) Line alignments and malloc sizeof

Style-only modifications to comply with checkpatch.pl --strict --file.
. Correctly realigns the lines that needed to be ;
. Suppress useless blank rows ;
. Fix sizeof() issues in various -malloc() functions.

Signed-off-by: Dominique van den Broeck <domdevlin@free.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: panel: (coding style) Matching braces

Style-only modifications to comply with checkpatch.pl --strict --file.
. Adds every missing brace in condition statements.

Signed-off-by: Dominique van den Broeck <domdevlin@free.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: panel: fix regression in lcd_write

This patch fix a regression in lcd_write caused by commit
70a8c3eb8546cefe40fb0bc7991e8899b7b91075

Signed-off-by: Bastien Armand <armand.bastien@laposte.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging/speakup/: avoid world-writable sysfs files.

In line with practice for module parameters, we're adding a build-time
check that sysfs files aren't world-writable.

Cc: Christopher Brannon <chris@the-brannons.com>
Cc: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rtl8192e: userspace ptr deref + incorrect declarations

. userspace pointer dereference ;

These issues have been fixed by a concurrent patch:
. missing inclusions of needed header files (fixed by concurrent patch);
. unrequired static function declaration (confusing another *.c file).

Signed-off-by: Dominique van den Broeck <domdevlin@free.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: r8188eu: Add support for Dlink DWA-121 802.11n Wireless N 150 Pico Adapter

This is a new device supported by this driver.

Signed-off-by: Kim Nylund <kim@pratsam.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: r8188eu: Fix some sparse warnings

In a patch entitles "staging: r8188eu: Fix case where ethtype was never obtained
and always be checked against 0" (commit ID unknown), I introduce an endian error.
This patch fixes that, and removes two additional sparse warnings.

drivers/staging/rtl8188eu/core/rtw_recv.c:653:6: warning: symbol 'process_pwrbit_data' was not declared. Should it be static?
drivers/staging/rtl8188eu/core/rtw_recv.c:1828:5: warning: symbol 'enqueue_reorder_recvframe' was not declared. Should it be static?

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

Staging: unisys: visorchipset: Remove filexfer.c

There is not any code using the functionality in filexfer.c so I removed it
and filexfer.h.

Signed-off-by: Ken Cox <jkc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: unisys: Fixup sparse warnings for dereferencing noderef types.

Fixed the usage of the following so they don't try to dereference
pointers to iomem.
CHANNEL_U64_MISMATCH
CHANNEL_U32_MISMATCH
wait_for_valid_guid

Signed-off-by: Ken Cox <jkc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: unisys: Remove unused macros from visorchannel/visorchannel.h

VISORCHANNEL_CHANGE_SERVER_STATE and VISORCHANNEL_CHANGE_CLIENT_STATE
are never used in any of the source so they have been removed.

VISORCHANNEL_CHANGE_CLIENT_STATE would have caused a broken kernel build
after commit  a8d7f21d, but since it was never used the kernel continued
to build.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Ken Cox <jkc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: unisys: virthba: declare virthba_queue_command as static

virthba_queue_command() is only used inside virthba.c so declare it static.

Signed-off-by: Ken Cox <jkc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: unisys: Fix sparse warnings

Fix sparse warnings caused by incorrect references to IO space.

Signed-off-by: Ken Cox <jkc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: handle errors from slic_config_get

slic_config_get() can fail. Change the return type from void to
int and handle the error in slic_card_init(). So now, instead of
silently failing (and then timing out waiting for the config data),
the driver will fail loudly at request time.

Signed-off-by: David Matlack <matlackdavid@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fail on corrupt eeprom

Remove fail_on_bad_eeprom, which was always 0 and thus being used
to ignore incorrect checksumming. This means devices with corrupt
eeprom will now cause the driver to fail.

Since fail_on_bad_eeprom was the last member in use of struct
slic_reg_params, remove that struct altogether.

Signed-off-by: David Matlack <matlackdavid@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix eeprom checksum code

Rewrite slic_eeprom_cksum() to fix bugs and make readable. This patch
technically has no effect on the user as failed eeprom checksums are
ignored anyway.

The original implementation had the following issues:

  1. 2 of the 3 unrolled loops had the following bug:

       while ((len -= 32) >= 0) {
               [...]
               sum += w[15];
               w = (u16 *)((ulong) w + 16);    /* verify */
       }

     This processes 32-bytes of data but only increments the word
     pointer by 16 bytes. Fixing both of these bugs seems to fix
     slic_eeprom_cksum().

  2. Non-descriptive variable names, use of unions, and macros that
     change local state make the code difficult to read.

  3. The checksum loop is unrolled which makes the code harder to
     reason about while providing small performance improvement:
      - max eeprom length is 0x80 bytes (MAX_EECODE_SIZE), that's
        only 0x40 iterations
      - checksum is only computed during pci probe(), so not very
        often

Tested on Mojave card

Signed-off-by: David Matlack <matlackdavid@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix use-after-free bug in slic_entry_remove

This patch fixes a use-after-free bug that causes a null pointer
dereference in slic_entry_halt.

Since unregister_netdev() will ultimately call slic_entry_halt (the
net_device ndo_stop() virtual function for this device), we should
call it before freeing the memory used by slic_entry_halt.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix 64-bit isr address bug

This patch fixes a bug that only manifests when the physical address of
the interrupt status register is >4GB. Specifically, the driver was only
telling the device about the lower 32 bits of the ISR. This patch adds
the upper 32 bits.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix dma memory leak

This patch fixes a memory leak in slic_card_init. If the driver fails
to poll for an interrupt after requesting config data from the device
the dma memory is never freed.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: remove gratuitous debug infrastructure

As per the TODO file, this patch removes the gratuitous debug
infrastructure. As an extra incentive for removing this code,
the debugfs files are not cleaned up properly. For example, if
register_netdev() fails in slic_entry_probe() then all debugfs
files get left behind, even after the driver module is unloaded.
Touching these files quickly leads to an oops.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: remove unused members of struct adapter

This patch removes two fields from the private "struct adapter".
Specifically,

memorybase      duplicate of slic_regs

memorylength    written once and never read. This field is trivially
                computed with pci_resource_len if it's ever needed in
                the future.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix multiple free-after-free in slic_entry_remove

This patch fixes two free-after-free bugs in slic_entry_remove.
Specifically, slic_unmap_mmio_space() iounmaps adapter->slic_regs,
which is the same region of memory as dev->base_addr (iounmap-ed
a few lines later).

Next, both release_mem_region() and pci_release_regions() are called
on the same pci_dev struct.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: slicoss: fix use-after-free in slic_entry_probe

This patch fixes a use-after-free bug that can cause a kernel
oops. If slic_card_init fails then slic_entry_probe (the pci
probe() function for this device) will return error without
cleaning up memory.

Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:Octeon-usb:octeon-hcd.c return value

return value instead of function.

Signed-off-by: Paul McQuade <paulmcquad@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Use kmemdup() instead of memcpy() to duplicate memory

This issue was reported by coccicheck using the semantic patch
at scripts/coccinelle/api/memdup.cocci, and tested by compilation.

Signed-off-by: Benoit Taine <benoit.taine@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: realtek: Convert /n to \n

Use proper line terminations.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_sta_mgt: change rtw_alloc_stainfo23a to use gfp_t instead if int

This should fix a few sparse warnings like the following:

  CHECK   drivers/staging/rtl8723au/core/rtw_ap.c
drivers/staging/rtl8723au/core/rtw_ap.c:1054:45: warning: incorrect type in argument 3 (different base types)
drivers/staging/rtl8723au/core/rtw_ap.c:1054:45:    expected int [signed] gfp
drivers/staging/rtl8723au/core/rtw_ap.c:1054:45:    got restricted gfp_t

Signed-off-by: Karim Raslan <karim.allah.ahmed@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Fixup some more unreadable formatting

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Improve bad formatting in event handlers

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Declare rtw_site_survey() static

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Remove obsolete rtw_action_frame_parse23a()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_cfg80211_rx_action(): Use ieee80211_mgmt to parse action frame

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_cfg80211_monitor_if_xmit_entry() use struct ieee80211_mgmt

Use struct ieee80211_mgmt to parse action frame. In addition remove
unused functions rtw_cfg80211_rx_p2p_action_public() and
rtw_cfg80211_rx_action_p2p()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_get_bcn_info23a(): Don't zero variables unncessarily

In addition remove excessive brackets.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_get_bcn_info23a(): Use cfg80211_find_ie()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_ieee80211.c: Remove a number of unused functions

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_cmd.h: Remove some more unused #defines

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Get rid of some unused tdls definitions

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Move dummy_event_callback() to rtw_mlme_ext.c

In addition get rid of a couple of empty *_event_callback() handlers
and use dummy_event_callback() instead.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Get rid of unused get_bsstype23a()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Declare SetBWMode23a() static

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Don't compile rtw_ap.c when CONFIG_8723AU_AP_MODE=n

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Fix compile warnings

Fixup #ifdefs to not build issue_assocrsp() and OnAuth23a() when
CONFIG_8723AU_AP_MODE is not set.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Declare a number of functions static

These are all just used locally, no point cluttering the kernel
namespace with them.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: wifi.h: Get rid of last ugly frame_control manipulation macros

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtl8723a_cmd.c: Get rid of SetToDs()/SetFrDs() usage

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_xmit.c: Get rid of SetToDs()/SetFrDs() usage

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Get rid of SetToDs()/SetFrDs() usage

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Get rid of duplicate ieee80211 frame_control defines

This is all defined nicely in ieee80211.h, this was all duplication
serving no other purpose than to obfuscate the code.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Fix issue_assocrsp() to use IEEE80211_STYPE_* defines

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Declare issue_asocrsp() static

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_mlme_ext.c: Get rid of most WIFI_* frame type usage

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: ioctl_cfg80211.c: Use ieee80211.h provided fctl bits

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtl8723a_cmd.c: Use ieee80211.h provided fctl bits

v2: Fix endian bug reported by Dan Carpenter <dan.carpenter@oracle.com>

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtw_xmit.c: Use IEEE80211_[FS]TYPE_ defines

In addition pkt_attrib->type a u16 and get rid of pattrib->stype

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Remove unused SetFrameType() macro

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: Get rid of obsolete SetPriority()/SetEOSP()/SetAckpolicy()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: _issue_qos_nulldata23a(): Use struct ieee80211_qos_hdr

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rl8723au: rtw_make_wlanhdr(): Use ieee80211_qos_hdr

Use the kernel provided struct and defines to fill in qos_ctrl

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtl8723a_cmd.c: Use ieee80211.h defines to set qos_ctrl

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: rtl8723a_update_ramask(): Get rid of annoying debug message

There really is no need to print this message for anyone not trying to
debug the driver:

  rtl8723a_update_ramask => mac_id:0, networkType:0x0b, mask:0x000fffff

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>