Daniel Seither [Mon, 21 Jun 2010 23:25:54 +0000 (01:25 +0200)]
Staging: batman-adv: record route for ICMP messages
The standard layer 3 ping utility can use the record route (RR) option
of IP to collect route data for sent ping messages (ping -R). This
patch introduces comparable functionality for batman-adv ICMP messages.
The patch adds a second batman ICMP packet format (icmp_packet_rr) such
that up to 17 MAC addresses can be recorded (sufficient for up to 8
hops per direction). When no RR is wanted, the old icmp_packet without
the RR overhead can be sent.
Signed-off-by: Daniel Seither <post@tiwoc.de> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Simon Wunderlich [Mon, 21 Jun 2010 23:25:53 +0000 (01:25 +0200)]
Staging: batman-adv: bonding and interface alternating
This patch adds interface alternating to the new bonding feature. By
default, we now try to avoid forwarding packets on the receiving
interface, instead choosing alternative interfaces. This feature
works only on nodes which have multiple interfaces connected to the
mesh. This approach should reduce problems of the half-duplex nature
of WiFi Hardware and thus increase performance.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Acked-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Simon Wunderlich [Mon, 21 Jun 2010 23:25:52 +0000 (01:25 +0200)]
Staging: batman-adv: Add bonding functionality
This patch introduces bonding functionality to batman-advanced, targeted
for the 0.3 release. As we are able to route the payload traffic as we
want, we may use multiple interfaces on multihomed hosts to transfer data
to achieve higher bandwidth. This can be considered as "light Multi Path
Routing" for single hop connections.
To detect which interfaces of a peer node belong to the same host, a
new flag PRIMARIES_FIRST_HOP is introduced. This flag is set on the first hop
of OGMs of the primary (first) interface, which is broadcasted on all
interfaces. When receiving such an OGM, we can learn which interfaces
belong to the same host (by assigning them to the primary originator).
Bonding works by sending packets in a round-robin fashion to the available
interfaces of a neighbor host, if multiple interfaces are available. The
neighbor interfaces should be almost equally good to reach.
To avoid interferences (i.e. sending on the same channel), only neighbor
interfaces with different mac addresses and different outgoing interfaces
are considered as candidates.
Bonding is deactivated by default, and can be activated by
echo 1 > /sys/class/net/bat0/mesh/bonding
for each individual node.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Rework on top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Simon Wunderlich [Mon, 21 Jun 2010 23:25:51 +0000 (01:25 +0200)]
Staging: batman-adv: 32bit sequence number and TTL for broadcasts
This patch changes the sequence number range from 8 or 16 bit to 32 bit.
This should avoid problems with the sequence number sliding window algorithm
which we had seen in the past for broadcast floods or malicious packet
injections. We can not assure 100% security with this patch, but it is quite
an improvement over the old 16 bit sequence numbers:
* expected window size can be increased (4096 -> 65536)
* 64k packets in the right order would now be needed to cause a loop,
which seems practically impossible.
Furthermore, a TTL field has been added to the broadcast packet type, just to
make sure.
These changes required to increase the compatibility level once again.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Change atomic64_* back to atomic_*, Rework on
top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Linus Lüssing [Mon, 21 Jun 2010 23:25:49 +0000 (01:25 +0200)]
Staging: batman-adv: permit setting ogm interval to JITTER*2
When trying to set the originator interval to 40ms, we are asked to set
it to a minimum value of 40ms. This patch permits setting an
originator interval of JITTER*2 (40ms by default) now.
Signed-off-by: Linus Lüssing <linus.luessing@web.de> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Mon, 21 Jun 2010 23:25:48 +0000 (01:25 +0200)]
Staging: batman-adv: Ignore debugfs on kernels without debugfs support
All code for debugfs is ignored when the creation of the batman-adv root
directory in debugfs fails. It must also be ignored when the
debugfs_create_dir tells us that debugfs is not implemented to prevent
possible crashes in the future.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Joe Perches [Mon, 21 Jun 2010 23:25:46 +0000 (01:25 +0200)]
Staging: batman-adv: Convert MAC_FMT to %pM
Remove the last uses of MAC_FMT
Signed-off-by: Joe Perches <joe@perches.com>
[sven.eckelmann@gmx.de: Adapted for current batman-adv version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Marek Lindner [Mon, 21 Jun 2010 23:25:45 +0000 (01:25 +0200)]
Staging: batman-adv: remove superfluous hint to "translation table"
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
[sven.eckelmann@gmx.de: Rework on top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Mon, 21 Jun 2010 23:25:43 +0000 (01:25 +0200)]
Staging: batman-adv: Move tables from sysfs to debugfs
Files which represent more than a single attribute aren't allowed in
sysfs. As we have some files which aren't essential and are lists or
tables aggregated from data from different places inside batman-adv, we
must place them in a filesystem without such a restriction.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Mon, 21 Jun 2010 23:25:42 +0000 (01:25 +0200)]
Staging: batman-adv: Move device for icmp injection to debugfs
batctl uses /dev/batman-adv to send special batman-adv icmp packets to
other nodes in the mesh. To get it working with multiple batX devices we
must ensure that every mesh device can have their own socket which is
used to inject those packets in exactly one mesh.
The current implementation still doesn't allow to use complete separated
meshes as we rely on structures which are not part of the private data
of a batman device.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Staging: batman-adv: convert vis_interval into define
vis_interval does not neccesarily needs to be a variable, as there is
no way to change it anyway (and probably no need to). We can therefore
remove yet another global variable.
Thanks Marek for pointing this out.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
[sven.eckelmann@gmx.de: Rework on top of current version] Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
batman-adv is receiving and sending the packets of its own ether type
on a very early/low level. Therefore we need to add explicit hooks to
give netfilter/ebtables a chance to filter them.
Reported-by: Antonio Quartulli <ordex@ritirata.org> Signed-off-by: Linus Lüssing <linus.luessing@web.de> Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Marin Mitov [Sun, 6 Jun 2010 11:16:07 +0000 (14:16 +0300)]
Staging: dt3155v4l: correcting a pointer mismatch bug and cleanups
The third argument of dma_alloc_coherent() is of type dma_addr_t and is defined
as u64 for x86_64, while the member 'device_base' of struct dma_coherent_mem is
defined as u32.
While at it, clean up the code. Now if dev->dma_mem != NULL all the fields/members
are already properly initialized.
Dan Carpenter [Mon, 7 Jun 2010 21:44:57 +0000 (23:44 +0200)]
Staging: otus: fix strcpy() overflow
wrq->name is only 16 characters long but "IEEE 802.11-MIMO" is 16
characters + a NULL character, so it's too long. This patch changes it
to "IEEE 802.11abgn".
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Joe Perches [Mon, 7 Jun 2010 19:11:35 +0000 (12:11 -0700)]
Staging: otus: TODO: Don't bother, for documentation purposes only
On Mon, 2010-06-07 at 12:02 -0700, Luis R. Rodriguez wrote:
> this driver is unmaintained and its only purpose is as a
> source of documentation for developers working on ar9170 and carl9170.
> Once carl9170 gets 11n support and merged upstream then this driver
> can be removed.
Then the TODO file should be updated.
Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Karl Relton [Thu, 3 Jun 2010 22:04:06 +0000 (23:04 +0100)]
Staging: wlan-ng: Switch from wext to cfg80211
Switch driver over from wext to cfg80211 interface.
Some Notes:
- This patch moves the driver wholesale from wext to cfg80211. Wext
support is still provided through the cfg80211 provided wext
compatability layer.
- Currently only infrastructure mode is implemented. Ad hoc mode is not
yet implemented, but can be added.
- It does not support connecting to a specified bssid, instead roaming
is handled by the card itself. This matches the behaviour of the
existing driver.
- It has been tested using NetworkManager (via wpa_supplicant)
configured to use the wext compatability layer, and then again with the
native nl80211 layer.
Signed-off-by: Karl Relton <karllinuxtest.relton@ntlworld.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ben Hutchings [Tue, 25 May 2010 03:25:57 +0000 (04:25 +0100)]
Staging: rtl8192su: Fix procfs code for interfaces not named wlan0
The current code creates directories in procfs named after interfaces,
but doesn't handle renaming. This can result in name collisions and
consequent WARNINGs. It also means that the interface name cannot
reliably be used to remove the directory - in fact the current code
doesn't even try, and always uses "wlan0"!
Since the name of a proc_dir_entry is embedded in it, use that when
removing it.
Add a netdev notifier to catch interface renaming, and remove and
re-add the directory at this point.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Mike Frysinger [Sun, 23 May 2010 07:10:35 +0000 (03:10 -0400)]
Staging: iio: pull in slab.h for kmalloc funcs
These drivers use kzalloc() but don't include slab.h. They currently
build though because the spi.h header will pull in slab.h for us. But
rather than rely on that behavior forever, include slab.h explicitly.
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Acked-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Mike Frysinger [Sun, 23 May 2010 07:10:30 +0000 (03:10 -0400)]
Staging: iio: kill off spurious semicolons
A bunch of IIO files contain spurious semicolons after function
definitions and case statements and if statements. Guess people
really like this thing, but kill them anyways so they'll stop
spreading via copy & paste with new drivers.
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Acked-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Standardize the spacing/style across the IIO build files:
- comment block in Kconfigs
- newlines at ends of files
- trailing lines at ends of files
- indent with one tab, not spaces or mixed
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Acked-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Nitin Gupta [Tue, 1 Jun 2010 08:01:26 +0000 (13:31 +0530)]
Staging: zram: Rename ramzswap to zram in documentation
Related changes:
- Included example to show usage as generic
(non-swap) disk with ext4 filesystem.
- Renamed rzscontrol to zramconfig to match
with new device naming.
Nitin Gupta [Tue, 1 Jun 2010 08:01:24 +0000 (13:31 +0530)]
Staging: Rename ramzswap files to zram
Related changes:
- Modify revelant Kconfig and Makefile accordingly.
- Change include filenames in code.
- Remove dependency on CONFIG_SWAP in Kconfig as zram usage
is no longer limited to swap disks.
Nitin Gupta [Tue, 1 Jun 2010 08:01:23 +0000 (13:31 +0530)]
Staging: ramzswap: Support generic I/O requests
Currently, ramzwap devices (/dev/ramzswapX) can only
be used as swap disks since it was hard-coded to consider
only the first request in bio vector.
Now, we iterate over all the segments in an incoming
bio which allows us to handle all kinds of I/O requests.
ramzswap devices can still handle PAGE_SIZE aligned and
multiple of PAGE_SIZE sized I/O requests only. To ensure
that we get always get such requests only, we set following
request_queue attributes to PAGE_SIZE:
- physical_block_size
- logical_block_size
- io_min
- io_opt
Note: physical and logical block sizes were already set
equal to PAGE_SIZE and that seems to be sufficient to get
PAGE_SIZE aligned I/O.
Since we are no longer limited to handling swap requests
only, the next few patches rename ramzswap to zram. So,
the devices will then be called /dev/zram{0, 1, 2, ...}
Usage/Examples:
1) Use as /tmp storage
- mkfs.ext4 /dev/zram0
- mount /dev/zram0 /tmp
2) Use as swap:
- mkswap /dev/zram0
- swapon /dev/zram0 -p 10 # give highest priority to zram0
Performance:
- I/O benchamark done with 'dd' command. Details can be
found here:
http://code.google.com/p/compcache/wiki/zramperf
Summary:
- Maximum read speed (approx):
- ram disk: 1200 MB/sec
- zram disk: 600 MB/sec
- Maximum write speed (approx):
- ram disk: 500 MB/sec
- zram disk: 160 MB/sec
Issues:
- Double caching: We can potentially waste memory by having
two copies of a page -- one in page cache (uncompress) and
second in the device memory (compressed). However, during
reclaim, clean page cache pages are quickly freed, so this
does not seem to be a big problem.
- Stale data: Not all filesystems support issuing 'discard'
requests to underlying block devices. So, if such filesystems
are used over zram devices, we can accumulate lot of stale
data in memory. Even for filesystems to do support discard
(example, ext4), we need to see how effective it is.
- Scalability: There is only one (per-device) de/compression
buffer stats. This can lead to significant contention, especially
when used for generic (non-swap) purposes.
Naveen Jain [Wed, 9 Jun 2010 12:20:40 +0000 (07:20 -0500)]
Staging: ti-st: fix protocol counting
Older mechanism of counting the number of protocols
registered with ST was slow, in-efficient.
It used to check the protocol data for NULL for each
registration/unregistration.
With this change, counting protocols in maintained by
a single counter protos_registered.
Naveen Jain [Wed, 9 Jun 2010 08:45:32 +0000 (03:45 -0500)]
Staging: ti-st: cleanup sysfs code
Remove the older way of communicating with user-space
daemon UIM.
The older way involved creating of a new kobj and pid
sysfs file under it, and sending signal using that pid.
Now we communicate via rfkill to user-space UIM.
Background: UIM is the user-space daemon which upon
notification from ldisc driver, opens the tty, sets
default baud and then installs (tiocsetd) the ldisc.
Pavan Savoy [Mon, 24 May 2010 14:14:09 +0000 (19:44 +0530)]
Staging: Update TI-ST's TODO
Update the TODO of TI-ST to make sure it reflects current list
of activities that need to be done.
Also point it out to the user-space app code relevant to it.
Otavio Salvador [Fri, 21 May 2010 22:08:14 +0000 (19:08 -0300)]
Staging: vt6656: fix up built-in compilation together with rt2870
Fix this build error:
drivers/staging/vt6656/built-in.o: In function `rotr1':
(.text+0x1a878): multiple definition of `rotr1'
drivers/staging/rt2870/built-in.o:(.text+0x106c2): first defined here
drivers/staging/vt6656/built-in.o: In function `tkip_sbox':
(.text+0x1a848): multiple definition of `tkip_sbox'
drivers/staging/rt2870/built-in.o:(.text+0x10697): first defined here
drivers/staging/vt6656/built-in.o: In function `xor_32':
(.text+0x1ec24): multiple definition of `xor_32'
drivers/staging/rt2870/built-in.o:(.text+0x111c4): first defined here
drivers/staging/vt6656/built-in.o: In function `xor_128':
(.text+0x1ec00): multiple definition of `xor_128'
drivers/staging/rt2870/built-in.o:(.text+0x111dd): first defined here
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Commit 902d2411298a6e0b8a473a0163aa4bf6d5c8a073 converted U32 to u32
which you would think would be just fine. Unfortunatly, it seems that
VIA only builds their code on a 32bit processor (which makes sense if
you think about it), but this doesn't work on x86-64. So fix up the few
places where this really wanted to be an unsigned long width.
Cc: Charles Clément <caratorn@gmail.com> Cc: Forest Bond <forest@alittletooquiet.net>, Cc: Andres More <more.andres@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / mv-sheeva.git / log