Staging: batman-adv: Reduce max characters on a line to 80

Documentation/CodingStyle sets a strongly prefered limit of 80
characters per line in "Chapter 2: Breaking long lines and strings".

Strings must be broken into smaller parts and long statements must be
rewritten.

Reported-by: Mikal Sande <mikal.sande@gmail.com>
Reported-by: Mark Rankilor <reodge@gmail.com>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Fix whitespace problems criticized by checkpatch.pl

Trailing spaces at the end of a line or before a tab are against
Documentation/CodingStyle "3.1: Spaces" and should be avoided. It is
also common style to add a single space after commas unless it is
followed either by a newline or a tab.

Reported-by: Mikal Sande <mikal.sande@gmail.com>
Reported-by: Luis de Bethencourt <luisbg@ubuntu.com>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Update TODO file to reflect current state.

Not much left to do on the TODO list :-)

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Update pointer to ethhdr after skb_copy

We must ensure that all pointer to a socket buffer are updated when we
copy a socket buffer and free our reference to the old one.
Another part of the kernel could also free its reference which maybe
removes the buffer completely. In that situation we would would feed
wrong information to the routing algorithm after the memory area is
written again by someone else.

Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: kfree_skb() in interface_tx() in error case

As we always return that the we consumed the skb, we should also free the skb
in the case of an error.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Limit queue lengths for batman and broadcast packets

This patch limits the queue lengths of batman and broadcast packets. BATMAN
packets are held back for aggregation and jittered to avoid interferences.
Broadcast packets are stored to be sent out multiple times to increase
the probability to be received by other nodes in lossy environments.

Especially in extreme cases like broadcast storms, the queues have been seen
to run full, eating up all the memory and triggering the infamous OOM killer.
With the queue length limits introduced in this patch, this problem is
avoided.

Each queue is limited to 256 entries for now, resulting in 1 MB of maximum
space available in total for typical setups (assuming one packet including
overhead does not require more than 2000 byte). This should also be reasonable
for smaller routers, otherwise the defines can be tweaked later.

This third version of the patch does not increase the local broadcast
sequence number when the queue is already full.

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Reorganize sequence number handling

BATMAN and broadcast packets are tracked with a sequence number window of
currently 64 entries to measure and avoid duplicates. Packets which have a
sequence number smaller than the newest received packet minus 64 are not
within this sequence number window anymore and are called "old packets"
from now on.

When old packets are received, the routing code assumes that the host of the
originator has been restarted. This assumption however might be wrong as
packets can also be delayed by NIC drivers, e.g. because of long queues or
collision detection in dense WiFi? environments. This behaviour can be
reproduced by doing a broadcast ping flood in a dense node environment.

The effect is that the sequence number window is jumping forth and back,
accepting and forwarding any packet (because packets are assumed to be "new")
and causing loops.

To overcome this problem, the sequence number handling has been reorganized.
When an old packet is received, the window is reset back only once. Other old
packets are dropped for (currently) 30 seconds to "protect" the new sequence
number and avoid the hopping as described above.

The reorganization brings some code cleanups (at least i hope you feel the
same) and also fixes a bug in count_real_packets() which falsely updated
the last_real_seqno for slightly older packets within the seqno window
if they are no duplicates.

This second version of the patch also fixes a problem where for seq_diff==64
bit_shift() reads from outside of the seqno window, and removes the loop
for seq_diff == -64 which was present in the first patch.

The third iteration also adds a window for the next expected sequence numbers.
This minimizes sequence number flapping for packets with very big differences
(e.g. 3 packets with seqno 0, 25000 and 50000 might still cause problems
without this window).

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: fix whitespace style issues

This patch fixes unnecessary whitespaces before a quoted
newline that the remaining batman-adv files had.

Reported-by: Luis de Bethencourt <luisbg@ubuntu.com>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: move /proc interface handling to /sys

Instead of having a single /proc file "interfaces" in which you have
to echo the wanted interface batman-adv will create a subfolder in each
suitable /sys/class/net folder. This subfolder contains files for the
interface specific settings. For example, mesh_iface to add/remove an
interface from a virtual mesh network (at the moment only bat0 is
supported).

Example:
echo bat0 > /sys/class/net/eth0/batman-adv/mesh_iface

to deactivate:
echo none > /sys/class/net/eth0/batman-adv/mesh_iface

Interfaces which are not compatible with batman-adv won't contain the
batman-adv folder, therefore can't be activated. Not supported are:
loopback, non-ethernet, non-ARP and virtual mesh network interfaces

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: remove redundant pointer to originator interface

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: move originator interval setting from /proc to /sys

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: convert more files from /proc to /sys

converted files:
vis_mode, vis_data

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: convert multiple /proc files to use sysfs

This is the first patch in a series of patches which aim to convert
all batman-adv /proc files to sysfs. To keep the changes in a
digestable size it has been split up into smaller chunks. During
the transition period batman-adv will use /proc as well as sysfs.

As a first step the following files have been converted:
aggregate_ogm, originators, transtable_global, transtable_local

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: fix whitespace style issues

This patch fixes the 31 unnecessary whitespaces before a quoted
newline that the batman-adv files had.

Signed-off-by: Luis de Bethencourt <luisbg@ubuntu.com>
[sven.eckelmann@gmx.de: Redone to apply against current version]
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: cleanup: change test for end of array

The code here is testing to see if "i" is passed the end of the array.
The original code works probably, but it's not the cleanest way.

Andrew Lunn suggested that I also remove all the hard coded references
to 256 so I have done that as well.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Changing version to 0.2.2-beta

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Update README about vis raw output

We are now having a newer, more neutral vis output so that we won't have
to change the kernelmodule for adding support of new vis output formats.
This patch adds an explanation about this in the README file of
batman-adv and removes the description about the dot/json format (they
will be added to the README of batctl).

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Add 0.2.1 changes to the CHANGELOG

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Remove dead max addr and obsolete VIS_FORMAT strings

Max address is not being used anywhere and just misleading, therefore
removing it. VIS_FORMAT string is now obsolete, so also remove it.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: remove the beta from main.h for release

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Update copyright years

Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Fix aggregation direct-link bug

So far, neighbour's secondary interface OGMs can involuntarily
piggyback on primary interface OGMs that arrived on the same secondary
interface before. Secondary interface OGMs should NEVER leave their
direct neighbour broadcast domain! This patch ensures that secondary
interface OGMs can only be aggregated to other secondary interface OGMs.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: fix aggregation timing bug

batman-adv aggregates routing packets to reduce the number of packets in
the air. Every outgoing packet is compared with other packets in the
buffer to determine whether it can be aggregated or not. Packets sent
at a lower interval can be held back longer to maximize the aggregation.
Due to insufficient checking batman-adv held back all packets for a
certain time depending on its own lowest interval rate which slowed
down all other nodes.

Reported-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: Clone shared bat packets before modifying them

"tcpdump" and "batctl td" will receive packets with a wrong sequence
number on systems with a different endianess than network byte order.
This happens due to the reordering of bytes in the function which
handles aggregated bat packets. The function which receives the bat
packets must ensure that these buffers aren't shared with anything else
before that function tries to write into it. Otherwise it has to copy
the buffers so it is save again to change them.

Reported-by: Kevin Steen <batman@kevinsteen.net>
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: batman-adv: only modify hna-table on active module

If we haven't set the module to MODULE_ACTIVE state before (in general,
no interface has yet been added to batman-adv) then the hna table is not
initialised yet. If the kernel changes the mac address of the bat0
interface at this moment then an hna_local_add() called by interface_set_mac_addr()
then resulted in a null pointer derefernce. With this patch we are now
explicitly checking before if the state is MODULE_ACTIVE right now so
that we can assume having an initialised hna table.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rt2870: add device ID of MelCo.,Inc. WLI-UC-G301N

My wireless LAN module 'MelCo.,Inc. WLI-UC-G301N' works fine,
if the following line is added into 2870_main_dev.c.

Signed-off-by: Nobhiro KUSUNO <n-kusuno@fc4.so-net.ne.jp>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rar_register: depends on PCI

rar_register driver uses PCI interfaces and PCI devices, so it
should depend on PCI.

Also format the Kconfig help text as normally done.

drivers/staging/rar_register/rar_register.c:623: error: implicit declaration of function 'pci_dev_get'
drivers/staging/rar_register/rar_register.c:623: warning: assignment makes pointer from integer without a cast

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: Fixed long line lengths in comedi.h

This patches fixes some of the long line lengths that checkpatch.pl was
complaining about in comedi.h

Signed-off-by: Mark Rankilor <reodge@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: Fixed long line length in comedidev.h

This patch fixes a long line length in comedidev.h to make checkpatch.pl happy

Signed-off-by: Mark Rankilor <reodge@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: Fixed printk call lengths and log levels

This is a patch to include log levels and fix some over length lines in
printk calls in drivers.c

Signed-off-by: Mark Rankilor <reodge@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: crystalhd: fix missing semicolon

A semicolon is missing at the end of a statement, but it does compile
fine without it as the macro BCMLOG_ERR expands to a do {...} while (0);

Signed-off-by: Charles Clément <caratorn@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: netwave: delete the driver

It has sat in the staging directory since October of 2009, and no one
has stepped up to take it over, so odds are, no one cares about it
anymore.  So, it is now deleted as scheduled, and documented in the TODO
file.

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

Staging: wavelan: delete the driver

It has sat in the staging directory since October of 2009, and no one
has stepped up to take it over, so odds are, no one cares about it
anymore.  So, it is now deleted as scheduled, and documented in the TODO
file.

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

Staging: strip: delete the driver

It has sat in the staging directory since October of 2009, and no one
has stepped up to take it over, so odds are, no one cares about it
anymore.  So, it is now deleted as scheduled, and documented in the TODO
file.

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

Staging: arlan: delete the driver

It has sat in the staging directory since October of 2009, and no one
has stepped up to take it over, so odds are, no one cares about it
anymore.  So, it is now deleted as scheduled, and documented in the TODO
file.

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

Staging: hv: rename VersionInfo.h to version_info.h

The great renaming of the hv code is now complete.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename VmbusPacketFormat.h to vmbus_packet_format.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename VmbusChannelInterface.h to vmbus_channel_interface.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename VmbusPrivate.h to vmbus_private.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename VmbusApi.h to vmbus_api.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename StorVscApi.h to storvsc_api.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename NetVscApi.h to netvsc_api.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename ChannelInterface.c and .h to channel_interface.c and .h

All of the uppercase .c files are now gone.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename ChannelMgmt.c and .h to channel_mgmt.c and .h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename RndisFilter.c and .h to rndis_filter.c and .h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename StorVsc.c to storvsc.c

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename NetVsc.c and .h to netvsc.c and .h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename Vmbus.c to vmbus.c

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename BlkVsc.c to blkvsc.c

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename RingBuffer.c and .h to ring_buffer.c and .h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename Connection.c to connection.c

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename Channel.c and .h to channel.c and .h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename Hv.h to hv.h

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename Hv.c to hv.c

No CamelCase in file names.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: rename hyperv_utils.c to hv_utils.c

As the module only has one .c file in it, just name the file
the same as the desired module.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: util.h: fix up space mess again

Again, use tabs, not spaces, it's not difficult to remember...

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove typedef that was just added.

It's a u64, so use a u64, it's not some special typedef.

Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Add Time Sync feature to hv_utils module.

The Time Sync feature synchronizes guest time to host UTC time after reboot,
and restore from saved/paused state.

Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Channel.c: fix up compiler warning

In the series of ASSERT removals, somehow we ended up with a compiler
warning in Channel.c.  This patch fixes that up.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT in logging.h

ASSERT is no longer used in hv, so remove the define

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT() and return -EINVAL in NetVsc.c

return -EINVAL instead of calling ASSERT()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: replace ASSERT() with WARN_ON() in NetVsc.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in vmbus_drv.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s and return -EINVAL in RingBuffer.c

return -EINVAL instead of calling ASSERT() for these conditionals.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: check return value of RingBufferInit()

RingBufferInit() would always return sucess and instead relied on an
ASSERT() to test for an error condition.  Remove the ASSERT() and
return -EINVAL instead.  The return value of RingBufferInit() was also
never checked, so check it.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in RingBuffer.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in RndisFilter.c

return -EINVAL instead of calling ASSERT() for these conditionals.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in RndisFilter.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in StorVsc.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in blkvsc_drv.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: make the block driver depend on LBDAF

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: return error instead calling ASSERT in blkvsc_drv.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in storvsc_drv.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

staging: hv: remove ASSERT()s in Channel.c

These ASSERT()s serve no purpose other than for debugging.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: return -EINVAL instead of calling ASSERT()

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ASSERT() in Channel.c

VmbusChannelOpen() will now return -EINVAL if UserDataLen is too big.
Previously this was handled by an assert.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: test return value of VmbusChannelEstablishGpadl()

The return value of VmbusChannelEstablishGpadl() was not examined in
Channel.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ASSERT() in Channel.c

return an error instead of calling ASSERT() if VmbusPostMessage()
fails.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ASSERT() in Channel.c

check memory allocation in VmbusChannelCreateGpadlHeader() and
return -ENOMEM if it fails

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ASSERT()s in Channel.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: test return value of osd_WaitEventCreate()

The return value of osd_WaitEventCreate() was not examined in some
places.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: return correct error values in Connection.c

Also check the kzalloc call return value.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: check return value of osd_PageAlloc()

The return value of osd_PageAlloc() was checked using an ASSERT().
Change that to more useful behaviour.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: remove ASSERT()s in ChannelMgt.c

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: hv: Remove check for NULL before calling kfree()

kfree() knows how to deal with NULL, so there's no reason to check for
NULL before passing something to it.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Hank Janssen <hjanssen@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155v4l: fix some sparse warnings

Mostly some __iomem markings and some static functions as well.

Cc: Marin Mitov <mitov@issp.bas.bg
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: dt3155v4l: Adding the missing linux/delay.h

The prototypes of udelay() and msleep() are in linux/delay.h,
so include it.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Marin Mitov <mitov@issp.bas.bg>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: code cleanup, resolved sparse finding

Cleared sparse warning 'Using plain integer as NULL pointer'

Signed-off-by: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: code cleanup, removed OUT definition

Remoted empty OUT define in ttype.h and its usage across the code.

Signed-off-by: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: vt6656: removed SUCCESS define as it is not used

Signed-off-by: Andres More <more.andres@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: arlan: fixed unnecessary whitespace style issue in arlan-main.c

This is a patch to the arlan-main.c file that fixes the unnecessary whitespace
issues found by the checkpatch.pl tool.

Signed-off-by: Prashant P. Shah <pshah.mumbai@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: cxt1e1: fix up one remaining THIS_MODULE usage

Copied from original Patch by Randy Dunlap

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: rtl8192u: Check kmalloc return value before use the buffer in ieee80211_softmac.c

Check kmalloc return value before use the buffer.

Signed-off-by: Zhitong Wang <zhitong.wangzt@alibaba-inc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: cb_pcidas64: fix up build warnings

Now that the 8255.h file is being included properly, we were
using the incorrect function prototypes, which causes a build
warning now.  This fixes it up and preserves the __iomem markings that
sparse wants to see.

Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: comedi: 8255: fix up previous static markings.

Ian pointed out that exported symbols should not be marked as static :)

Fixed this up by properly including the 8255.h file, and fixing
the function prototypes there, as the CONFIG variables were not getting
set so users of the header file were seeing the incorrect function
prototypes.

Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: line6: driver: fix up sparse warnings

minor stuff.

Cc: Markus Grabner <grabner@icg.tugraz.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

Staging: iio: industrialio-trigger.c: minor fixups

We needed to include a header file that declared the functions that are
being exported in this file.

Also fix up an indentation problem, and some sparse warnings.

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

Staging: iio: ring_sw.c: fix up sparse warnings

NULL usage, static stuff, etc.

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

Staging: iio: industrialio-ring.c: fix up sparse warnings

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