[git] batman-adv branch, master, created. ea0a7b50fbd7b711fd23a2b3e8cae62b41011eae - commits - lists.open-mesh.org

31 Dec 2009


      The branch, master has been created
        at  ea0a7b50fbd7b711fd23a2b3e8cae62b41011eae (commit)
- Shortlog ------------------------------------------------------------
commit ea0a7b50fbd7b711fd23a2b3e8cae62b41011eae
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 30 04:38:07 2009 +0000
batman-adv: fix regression in the regression fix
null-initialized fields instead of the source fields were used to
    allocate and copy the incoming HNA buffer ...
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 773bed14e9ec59cc2d86ddb415447423d2e5215f
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 30 03:36:43 2009 +0000
batman-adv: fix kmalloc regressions
The previous patch kmalloc introduced some regressions that this
    patch tries to solve.
Signed-off-by: Marek Lindner lindner_marek@yahoo.de
    Acked-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 4f987bc12fd99a5094a773c6c5c9e504cc42dbaf
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 30 01:33:46 2009 +0000
batman-adv: check all kmalloc()s
there are some kmallocs left which are not checked whether they succeeds or
    not, which might lead to corrupted data structures if the system memory is
    full. This patch should clean up the remaining unchecked kmalloc()s.
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 36aefc6f9c381b8228a0d5d6ed2458329781eb1d
Author: Andrew Lunn andrew@lunn.ch
Date:   Wed Dec 30 00:28:34 2009 +0000
Staging: batman-adv: Allow the MAC address to be set
Some embedded devices have very limited sources of entropy for the
    random number generator. It has been observed that the random MAC
    address on the interface bat0 is not always random. When testing with
    a collection of identical hardware, sometimes the bat0 device the same
    MAC address on multiple devices, causing mayhem. This patch allows the
    MAC address to be set by the user.
Signed-off-by: Andrew Lunn andrew@lunn.ch
    Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit fe28a94c01e14fc1a1a7b1306eaac518a474427f
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Dec 29 15:47:31 2009 +0000
batman-adv: receive packets directly using skbs
This patch removes the (ugly and racy) packet receiving thread and the
    kernel socket usage. Instead, packets are received directly by registering
    the ethernet type and handling skbs instead of self-allocated buffers.
Some consequences and comments:
* we don't copy the payload data when forwarding/sending/receiving data
       anymore. This should boost performance.
     * packets from/to different interfaces can be (theoretically) processed
       simultaneously. Only the big originator hash lock might be in the way.
     * no more polling or sleeping/wakeup/scheduling issues when receiving
       packets
     * this might introduce new race conditions.
     * aggregation and vis code still use packet buffers and are not (yet)
       converted.
     * all spinlocks were converted to irqsave/restore versions to solve
       some lifelock issues when preempted. This might be overkill, some
       of these locks might be reverted later.
     * skb copies are only done if neccesary to avoid overhead
performance differences:
* we made some "benchmarks" with intel laptops.
     * bandwidth on Gigabit Ethernet increased from ~500 MBit/s to ~920 MBit/s
     * ping latency decresed from ~2ms to ~0.2 ms
I did some tests on my 9 node qemu environment and could confirm that
    usual sending/receiving, forwarding, vis, batctl ping etc works.
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
    Acked-by: Sven Eckelmann sven.eckelmann@gmx.de
    Acked-by: Marek Lindner lindner_marek@yahoo.de
    Acked-by: Linus Lüssing linus.luessing@web.de
commit 7a61e77fd20d19e693690f779174cd5bfca84d72
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 29 13:23:35 2009 +0000
batman-adv: mark trunk as 0.3.0 alpha
The git repository contains a newly created maint(enance) branch which
    becomes the batman-adv stable branch. Only bugfixes and lightweight
    interface changes needed for the linux kernel integration will be made
    available there. All new features go into the 0.3-branch.
Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit d719903078c5bfb6f27845471718b6b115d10bb6
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 29 13:02:17 2009 +0000
batman-adv: splitting /proc vis file into vis_server and vis_data
The /proc vis file was used to enable/disable the vis server and to output
    the vis data at the same time. This behaviour was confusing and lacked a
    proper method to display the current vis server status.
    This patch seperates the 2 functionalities:
    * use vis_server to enable/disable the vis server and to retrieve its status
    * use vis_data to retrieve the vis raw data (if the server is enabled)
Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit d939f29f47dccf9cac13ed7561701fb6f82cef89
Author: Linus Luessing linus.luessing@web.de
Date:   Tue Dec 29 13:02:15 2009 +0000
batman-adv: remove obsoleted vis_format /proc file
batman-adv used to export the vis data using different formats that were
    switchable via /proc/net/batman-adv/vis_format. The various formats moved
    to user space and rendered this configuration switch useless.
Signed-off-by: Linus Luessing linus.luessing@web.de
    Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit 02115bc7dd68c75f6e003137be77caaec0ccaa79
Author: Andrew Lunn andrew@lunn.ch
Date:   Tue Dec 29 13:02:14 2009 +0000
batman-adv: moving vis output formats out of the kernel land
The batman-adv kernel module is able to output visualization data using the
    dot draw or JSON format. This patch transforms the output into a generic
    format (called vis raw). User space tool may convert the raw data to support
    a variety of formats without the need of modifying the kernel module.
Signed-off-by: Andrew Lunn andrew@lunn.ch
    Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit b584ad15eac939974fa1bb84499096c82e47fe5e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Dec 28 03:56:37 2009 +0000
batman-adv: fix minor orig table layout bug
The originator table contained a TAB instead of a space which broke
    the layout as well as the batctl parser.
Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit d4a00d03f8f595bd79b59e0505026480b9affded
Author: Linus Lüssing linus.luessing@web.de
Date:   Thu Dec 17 23:09:28 2009 +0000
batman-adv: Fixes rounding issues in vis.c
This patches fixes two rounding issues in vis.c for sending and
    purging vis packets. Before, the timers and timeouts always got
    rounded down to seconds, though we want a precision in
    milliseconds.
    This also fixes a kernel panic that occures when lowering the
    timer for sending vis packets (vis_interval) to less than 1000ms
    in main.c manually.
Signed-off-by: Linus Lüssing linus.luessing@web.de
    Acked-by: Marek Lindner lindner_marek@yahoo.de
commit c140c90d6272d4ce6885c9ceb12e10d2ab75e4c2
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Thu Dec 17 13:23:35 2009 +0000
batman-adv: Use forw_bcast_list_lock always in disabled interrupt context
forw_bcast_list_lock is spin_locked in both process and softirq context.
    SoftIRQ calls the spinlock with disabled IRQ and normal process context
    with enabled IRQs.
When process context is inside an spin_locked area protected by
    forw_bcast_list_lock and gets interrupted by an IRQ, it could happen
    that something tries to lock forw_bcast_list_lock again in SoftIRQ
    context. It cannot proceed further since the lock is already taken
    somewhere else, but no reschedule will happen inside the SoftIRQ
    context. This leads to an complete kernel hang without any chance of
    resurrection.
All functions called in process context must disable IRQs when they try
    to get get that lock to to prevent any reschedule due to IRQs.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Acked-by: Marek Lindner lindner_marek@yahoo.de
commit cf8eb943c67c47237bd81f667f5f291086ce5a9f
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 16 23:00:03 2009 +0000
staging: batman-adv: more consistent spelling
forgot one more neighbor to convert
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit d3029433f5f0452560b9b05a8e29f074bbb39b8d
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 16 16:55:59 2009 +0000
staging: batman-adv: consistent spelling of "neighbors"
we have written "neighbors", "neighbours" and bad spelled versions of this
    word, this patch should make it consistent.
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit abd39334c1f93db2abb641af7e433b1d517f169a
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Dec 13 21:58:12 2009 +0000
staging: batman-adv: initialize static hash iterators
instead of dynamically registering hash iterators, calling functions are
    changed to register the iterator objects statically. The two advantages are:
* no memory leaks when aborting from hash_iterate()
     * no calls to kmalloc/kfree, therefore a little faster/safer
Tested with 9 QEMU instances, no obvious regression found.
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit a14ebcf895d89378707f8e9902e6475e6d4e0be3
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Dec 13 20:33:45 2009 +0000
batman-adv: update the date of README state
Thanks Andrew for pointing this out.
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 49c63ede4e7af87bd7a88a54095d5f14f980961b
Author: Andrew Lunn andrew@lunn.ch
Date:   Sun Dec 13 20:09:03 2009 +0000
Staging: batman-adv: Replace KERN_DEBUG with bat_dbg
Ameya Palande requested we replace printk(KERN_DEBUG "") by pr_debug()
    I decided it was better to use our debug macro.
Signed-off-by: Andrew Lunn andrew@lunn.ch
    Acked-by: Marek Lindner lindner_marek@yahoo.de
commit ea765d04920173c63dbc01087d28dfa95353ba25
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Dec 13 18:40:52 2009 +0000
batman-adv: update README for latest kernel
state in the README that we also (compile) support 2.6.32
Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit dbaf431ff450a4b09fe00211319d236f07f88a73
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sun Dec 13 18:30:30 2009 +0000
batman-adv: Remove whitespace style errors
checkpatch.pl found different smaller errors regarding trailing
    whitespaces or trailing extra newlines before EOF. These should be
    removed to follow linux-2.6 CodingStyle.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Signed-off-by: Simon Wundelrich siwu@hrz.tu-chemnitz.de
commit 0cfed9bd2facca39be6e43416bef227cf67edc21
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sun Dec 13 18:21:03 2009 +0000
batman-adv: Remove NULL pointer check
It is safe to call kfree(NULL) which makes this extra check unneeded. It
    was found using checkpatch.pl from linux-2.6
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Acked-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 6af67a802eed53000acfb11eb7b62a6e38c5fe6d
Author: Julia Lawall julia@diku.dk
Date:   Sun Dec 13 18:16:07 2009 +0000
batman-adv: introduce missing kfree
Error handling code following a kzalloc should free the allocated data.
    Similarly for usb-alloc urb.
The semantic match that finds the first problem is as follows:
    (http://www.emn.fr/x-info/coccinelle/)
// <smpl>
    @r exists@
    local idexpression x;
    statement S;
    expression E;
    identifier f,f1,l;
    position p1,p2;
    expression *ptr != NULL;
    @@
x@p1 = (kmalloc|kzalloc|kcalloc)(...);
    ...
    if (x == NULL) S
    <... when != x
         when != if (...) { <+...x...+> }
    (
    x->f1 = E
    |
     (x->f1 == NULL || ...)
    |
     f(...,x->f1,...)
    )
    ...>
    (
     return (0|<+...x...+>|ptr);
    |
     return@p2 ...;
    )
@script:python@
    p1 << r.p1;
    p2 << r.p2;
    @@
print "* file: %s kmalloc %s return %s" % (p1[0].file,p1[0].line,p2[0].line)
    // </smpl>
Signed-off-by: Julia Lawall julia@diku.dk
    Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 9274deaa5a53d6369cfb342119902f0d899662f9
Author: Andrew Lunn andrew@lunn.ch
Date:   Sat Dec 5 15:43:00 2009 +0000
[batman-adv] correctly name the configuration option for debugging
Signed-off-by: Andrew Lunn andrew@lunn.ch
    Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit 8038183add3d635bd10dc9ba7091e84e9ed9ac31
Author: Linus Luessing linus.luessing@web.de
Date:   Thu Dec 3 03:29:56 2009 +0000
Updating orig_str for debug
--===============1664300015==
    Content-Type: multipart/signed; micalg=pgp-sha1;
    	protocol="application/pgp-signature"; boundary="3V7upXqbjpZ4EhLz"
    Content-Disposition: inline
--3V7upXqbjpZ4EhLz
    Content-Type: text/plain; charset=utf-8
    Content-Disposition: inline
    Content-Transfer-Encoding: quoted-printable
In commit 1488 the orig_str has been removed from
    purge_orig_neighbours(). Apparently, this breaks compilation when
    enabling debug-flags, as bat_dbg() is expecting this variable as a
    parameter. Actually, this variable has never been initialised also
    before commit 1488 (which was probably the reason for mistakenly
    removing it).
    This tiny patch adds the orig_str again, but initialises it now as
    well.
Signed-off-by: Linus Luessing linus.luessing@web.de
commit 22cde43cdb9dadb6d75c20d4736307d628116094
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 28 20:11:09 2009 +0000
if discarding a packet, we need to read it under any condition.
commit 768b4f81b24b0d58341b6e61d49bbebef27129f5
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 28 20:09:17 2009 +0000
split originator handling parts off routing.c
commit 0da533406524a39c40b8326ebdb9935399c50c93
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 28 18:44:04 2009 +0000
stop permanent warnings if a device is added but deactivated.
commit d48bf92cea1a84943f5d0b919f5b7687d835207a
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Nov 28 18:41:31 2009 +0000
Andrews refactored routing.c file patch: "I broke a lot of the big
    functions up into many smaller ones. This helps with readability and
    there is now a lot less code squashed against the right margin."
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
    Acked-by: Marek Lindner lindner_marek@yahoo.de
    Acked-by: Simon Wunderlich simon.wunderlich@s2003.tu-chemnitz.de
commit cdc8c00c2412f49bfd730227dd282f5566181e03
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Nov 14 06:11:37 2009 +0000
[batctl adapt batctl to the new logging mechanisms
commit 2610523c9305d9607bbbc0e73d1c240b33223903
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Nov 14 06:11:32 2009 +0000
[batman-adv] replace internal logging mechanism with standard kernel logging
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 4f375745f3fa377eb3eec9b57b175716e5d7128c
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Nov 9 20:09:12 2009 +0000
avoid (or at least decrease) route flapping in perfect environments
commit 8ba762453114eb8bfca5687183dc55ca4e93cef1
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Nov 8 14:27:43 2009 +0000
update version strings in trunk for upcoming releases
commit 037fa527f6c6148a38440c588b9c14b1f511ed31
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Nov 8 01:12:51 2009 +0000
style adjustment: add a newline in CHANGELOG
commit b51e5362ca7ebf77a417217136691c63ce94f794
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Nov 7 14:10:40 2009 +0000
[batman-adv] update Changelog
commit de005feb6e17132a67f4445f5e4d2f6b8c103aa2
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 14:04:18 2009 +0000
unlock in bat_device_add_packet() before returning (thanks Sven)
commit fa11c627cf75e58b5a30f0b0e804a2f5e65e0ad3
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sat Nov 7 13:50:40 2009 +0000
Remove unknown escape string {
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit 1b439e3ac028ed0648688e556715b49ed40deccd
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sat Nov 7 13:45:41 2009 +0000
Mark local symbols as static
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit 700d0946ed2270f0c7a0971db072dc0832b81978
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 13:29:35 2009 +0000
fix memory leak when there is a problem adding an interface
commit f91a8a47c04e2d83c456afb37c902f8d9c8b0c63
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 13:21:17 2009 +0000
remove obsolete functions from bitarray.h (thanks Sven)
commit 43ce74a5b78f51b54ca9833a80650a73c8cf552a
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 11:44:54 2009 +0000
remove dead code
commit bb95208d663318f141740127fbfc445a330fa88e
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 11:23:28 2009 +0000
update version string to 0.2
commit 47ed27f9706554ce99d93d76f976222ffc328cdc
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Nov 7 10:59:00 2009 +0000
update README and adjust year numbers in copyright notice
commit b0b2aaf6115bb16dad5096acae349ac904f43d98
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Nov 4 23:12:34 2009 +0000
[batman-adv] avoid temporary routing loops
commit 3a8deaa9b7db8764e4a6711c487efd573d2c78dc
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sun Oct 25 00:13:47 2009 +0000
[batman-adv] Start to send originator messages when interface gets active
The module gets in a inactive state when all interfaces are down. This
    stops the sending of new packets. When a interface gets activated again
    it must reactivate the module and then start the sending of new
    originator messages.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit 531035e31bee1d989638a976dcb212009ffe0ee3
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Oct 18 11:47:08 2009 +0000
[batman-adv] remove last LINUX_VERSION_CODE check outside of compat.h
commit c7560658b16bd85bed27ad278dae93a88cf856ee
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Oct 17 22:11:17 2009 +0000
[batman-adv] print a warning when an existing mac address is added again
commit 47cd66d85d746949f8fab9b091aecd13695ddbe3
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 22:01:20 2009 +0000
drop false packets which are found in the padding of Ethernet frames.
commit 8d669c4000307ca11f4069bcd474fb56f12163b6
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 19:28:57 2009 +0000
announce compile-compatibility for tested kernels in the README
commit 9e1227fa9b2dc533100e0641650fd1472926b55f
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 18:08:34 2009 +0000
as routing.c uses cancel_delayed_work_sync(), add compat.h to support kernels < 2.6.23
commit 8d19b35117de99a2826091da19ea6fe6dff39ce1
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 17:58:14 2009 +0000
remove warning when unregistering device as it won't be checked for newer kernels.
    Adjust style according to checkpatchs (2.6.30) recommendations
commit 1758daf0292c960b80f4007ba50b98d8ba49259c
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 17:56:02 2009 +0000
enable pre2.6.25 kernels again by adding compatibility for strict_strto[u]l()
commit 55ed866868150aa4d7312bed0945e9287a31fe87
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Oct 17 16:28:30 2009 +0000
use random ethernet address generator from the kernel
commit fca94633426e9ed14aeb6136bd91119f69631505
Author: Andrea Ghittino a.mailevent@gmail.com
Date:   Fri Oct 9 17:11:51 2009 +0000
[batman-adv] remove no checksum flag on bat0 to fix checksumming on certain architectures
It is related to the TCP checksum offload function: I am using as embedded
    platform PCEngines Alix 3d2, that do not support this function (and also
    qemu virtual machines do not support it).
This patch disable the tx checksum offload and everything now works fine for
    me!
Signed-off-by: Andrea Ghittino a.mailevent@gmail.com
commit f421723929dedbff03e689c281e7141dc2e4a4bf
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Sep 27 02:21:54 2009 +0000
[batman-adv] fixing race condition while handling lots of data
The callback batman_data_ready() set the variable data_ready_cond
    to true before waking up the packet handling thread. When the
    thread has completed its work it will sleep and sets the variable
    back to false. When the system has to handle large quantities of
    traffic the callback might be called several times while the
    packet handling thread runs. The status variable will be
    overwritten once the thread is done which might leave packets in
    the socket buffer. If the socket buffer is full no more packets
    are accepted and the callback won't be triggered. Therefore the
    packet receive thread has to reset the variable at the very
    beginning.
commit f93fa86ab835ebf64e3788afbc051a8ec3a6d098
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sat Sep 12 17:28:23 2009 +0000
[batman-adv] Remove compatibility code for	pre 2.6 kernel versions
The earliest version of linux which is compatible to
    batman-adv-kernelland is 2.6.20. Their are patches available in debian
    for batman-adv-kernelland 0.1 to make it build against linux 2.6.15, but
    they cannot be ported to current trunk due to changes to the working
    queues. To have "support" for much older versions isn't needed and can
    be removed without loosing any functionality.
    As result the only remaining LINUX_VERSION_CODE dependend precomiler
    code beside compat.h is for netdevice_ops support in soft-interface.c
    and result code checking of unregister_chrdev in device.c. For a
    possible merge into linux compat.h and all appearances of
    LINUX_VERSION_CODE should be removed according to checkpatch.pl.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit 1438a0797c67a212e532114e4b821fe0d79df3a1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Sep 11 17:56:49 2009 +0000
[batman-adv] check whether we have a valid "previous ttl" to avoid a TTL of 255
commit f792ca5f4c4a97bf56d3b394275c45ffc0e9dc5d
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Sep 10 16:31:08 2009 +0000
[batman-adv] can't sleep while holding a spinlock
commit 203475cdc3a2c9c02274ec3c2a80ba43ef6f272f
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Sep 9 15:07:52 2009 +0000
[batman-adv] rename old_orig to prev_sender to be consistent with the documentation
commit 7a9251378aadb205937e6f3fdba0f6d1e2bee1c3
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Sep 7 20:01:29 2009 +0000
[batman-adv] activate module after adding a new interface
This fixes a regression that entered the code after switching to
    the event based interface handling.
commit 4ec798258734ef8aa037053e0d9faad6e303256a
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Sep 6 19:21:41 2009 +0000
[batman-adv] fix crash in vis client after an interface went down
generate_vis_packet() checks if the originator has an interface but
    does not check whether this interface is up & running (while
    generating the vis packet). After an interface goes down there is a
    short period in which an interface is selected but down.
commit 88861ea9acb7c7605796188438102c9218026f50
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Sep 6 17:44:31 2009 +0000
[batman-adv] replacing if up/down timer with kernel notifications
Batman ran a timer that checked the interfaces for status changes
    and then reactinging accordingly. If applications did not shutdown
    their interfaces properly before removing it this timer might lead
    to crashes. The kernel notification should eliminate these race
    conditions.
commit 3f6cea4f8cd1d050e971cacde58663745b4d1479
Author: Linus Lüssing linus.luessing@web.de
Date:   Fri Sep 4 20:59:47 2009 +0000
This is Linus' final version with the rewind to top fix.
Original message (1):
This is a small review of Linus patch. It contains:
     * minor style adaptions
     * optional compile flag VIS_SUBCLUSTERS_DISABLED to prevent vis parser breakages
     * sanity checks for kmalloc()
Linus, Andrew: please give it a try on your systems. I currently don't have
    machines set up to test, so this is a dry run. :)
Original message (2):
Pfeh... long night again, but here you go. With this patch,
    batman-adv now supports the subgraphing-feature of the
    dot-file-format. The vis-output can then be parsed with
    "fdp -Tsvg -Gcharset=utf8 /proc/net/batman-adv/vis > test.svg"
    for example.
Interfaces belonging to one BATMAN-node can be found
    inside a box, the primary interface of a BATMAN-node is
    double-circled. The output also got more detailed, in that
    connections between nodes are being displayed with the exact
    mac address of the used interface instead of replacing one end
    with the primary interface's mac (see [0] for the problems I had
    posted before).
Cheers, Linus
PS: Please check if I got that kmalloc stuff right.
    PPS: I had to introduce a src-field in the vis-packet-struct,
    therefore the compatibility version had to be increased as well.
[0] https://lists.open-mesh.net/pipermail/b.a.t.m.a.n/2009-August/002832.html
Signed-off-by: Simon Wunderlich simon.wunderlich@s2003.tu-chemnitz.de
    From: Linus Lüssing linus.luessing@web.de
commit 1a60eadda913b4a24bd113fd69c81071c4c4428e
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Aug 29 21:09:13 2009 +0000
fix bug in vis (thanks sven)
commit 1b164d4b7a7d6d095920152554f3d76000ae1cc5
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 10:22:02 2009 +0000
[batman-adv] don't read from uninitialized socket
commit b8eb853f386a8b8acef848d89c08fbdb77481ea5
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:22:15 2009 +0000
[batman-adv] avoid spin-dead-lock in vis_quit()
vis_quit() also tries to aquire the vis_hash_lock, hence vis_init()
    has to give up that lock before calling vis_quit().
commit 256c85a32836304387d4f2d0a0f449fc88b498dd
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:00:27 2009 +0000
[batman-adv] don't call activate_module() twice to avoid race conditions
proc_interfaces_write() would call activate_module() manually after
    adding an interface with hardif_add_interface(). hardif runs a periodic
    check for new interfaces which also calls activate_module(). If both
    calls are close to each other spinlockups may occur due to the fact
    that activate_module() can sleep.
commit d3959759e07251a52ea0b9dc6123e7961cd27000
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:00:23 2009 +0000
[batman-adv] replace GFP_KERNEL by GFP_ATOMIC to avoid sleeping in interrupt context
commit dce4ec15fdebe4adb3a5fd025fcdd2a155a6e1c1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:00:19 2009 +0000
[batman-adv] locking in originator_init() and originator_free() spares us some (rare) crashes
commit 1d0fa92bfeebbc8de2f0103d1f1a120ebcd4e2b5
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:00:16 2009 +0000
[batman-adv] properly lock vis_init() and check its return value
commit cdd590786e462cf890b2825a299e156de864b1aa
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 21 06:00:12 2009 +0000
[batman-adv] check return value of bat_device_setup()
commit be7f78400cb493af28e510a1bd1504788f095b77
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 19:48:10 2009 +0000
[batman-adv] minor version display fix
commit 0ceefe7469bd9836f3790317bd31d309916ae6d9
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 19:33:12 2009 +0000
[batman-adv] show batman-adv version in /sys/module/batman_adv
commit 21e838760727761c1065f70557a0a61d63f6c2b6
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 17:01:21 2009 +0000
[batman-adv] fix various race conditions during startup & shutdown
* mesh hases are created/destroyed when the module is (de)activated
    * bat0 interface should not process packets for the module is active
    * bat device also should not process packets while the module is
      inactive
    * module_state became atomic_t as well due to alignment problems
commit 7441df0d982c891a66d6e0d5fe6ea7662bb35507
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 17:01:17 2009 +0000
[batman-adv] make hash init functions more robust
* moving orig_hash init to routing.c to follow the other hashes
    * check if hashes exist before destroying them
    * properly reset hash pointers to NULL after deleting a hash
commit cb2960815aad7ae3e44f8e8c4387f7656418581b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 17:01:13 2009 +0000
[batman-adv] transform hna_local_changed to atomic_t to avoid alignment problems on ARM
commit c9612578c68d5d5689d4772d4e49ea94e147e1cc
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 17:01:10 2009 +0000
[batman-adv] fix kernel oops after local hna add
After adding a new interface or client batman will flood
    the network with its mac address. Therefore it is necessary
    to created a new batman packet with the changed mac data
    inside. Rebuilding the packet also changes the pointer to
    the packet buffer.
commit a02bba100134d34ccba37741718e47c761179d5b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Aug 14 17:01:01 2009 +0000
[batman-adv] add hash sanity checks
commit 1f90bea5fb273cdde18a362041344238b59f0f8e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Aug 11 22:08:19 2009 +0000
[batman-adv] fixing "ghost" entries in originator table
Activating a given interface was not in sync with sending
    outgoing packets which could lead to wrong entries in the
    originator table (that would time out after a while).
    This patch syncs these events and avoids the wrong entries.
commit 0acc62784c5a5c6f7fe845fe73238507ef1704d1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Aug 10 08:00:09 2009 +0000
[batman-adv] fix logging crash
Through code refactoring entered an uninitialized variable that
    needs a value before we print it.  :-)
commit f495e65e5951e3607df2ddd46247206c860780f8
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Aug 9 17:29:49 2009 +0000
[batman-adv] remove extra wifi penalty due to loop risk
Tests have shown that the extra wifi penalty may lead to
    loops if 2 neighbors have a similar link quality.
commit 4e7d42305a4e05a2bbc6c970cb29bb899ac33aa5
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Aug 9 17:29:43 2009 +0000
[batman-adv] adjust coding style in device handling
commit 75a7c934fd6091210496fcaedd20c95b004d81cd
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Aug 9 17:29:39 2009 +0000
[batman-adv] use kernel constants instead of numbers
commit 6398f8e3414357b78d7a26d56834727e2116f5ff
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Aug 9 17:29:27 2009 +0000
[batman-adv] removing ignored parameter from send_raw_packet()
The sender address that had to be specified when calling the
    function send_raw_packet() was always ignored.
commit a00d6cd2d7902cc1f88834fd32c9f2a39135defc
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Aug 6 14:03:15 2009 +0000
[batman-adv] change next hop only when we receive a better TQ value
Andrew Lunn & Yang Su noticed that batman would produce transient
    routing loops due the next hop selection algorithm. When batman
    received an incoming packet from its current best next hop with a
    TQ value that is smaller than this neighbors current TQ value
    batman might change the next hop router. If this neighbor was not
    updated yet it might send its data to us.
    In oder to resolve this batman does not immediately change its
    route when receiving a worse TQ from its best neighbor. If
    alternative neighbors send a better TQ later on the route can
    safely be switched.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
    Signed-off-by: Yang Su Yang.Su@ascom.ch
    Signed-off-by: Marek Lindner lindner_marek@yahoo.de
commit 6c98a8ba8453ae3af0433fd41c25ce6222f8199e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Aug 6 14:03:00 2009 +0000
[batman-adv] reduce global TQ winsize for faster convergence
As suggested by Andrew Lunn & Yang Su the global TQ window size
    should be reduced in order to speed up the convergence time when a
    routing change occurs.
commit b571f83758a1ff84517a0909897b1bb87d2872b1
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sun Jul 26 09:40:32 2009 +0000
Cleanup smaller style problems
Make some files clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with printk/seq_print.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit b8367279d43a5a5825db7f16329572ced167062d
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sun Jul 26 09:37:34 2009 +0000
[batman-adv] cleanup send.[ch]
Make send.[ch] clean with respect to the 2.6.29
    checkpatch script. Exceptions are long lines with debug_log.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit a616e1d60e2bdbe4f96517fcd7bb8e1b0f19f342
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Jul 25 13:22:31 2009 +0000
[batman-adv] Cleanup translation-table.[ch]
Make translation-table.[ch] clean with respect to the 2.6.29
    checkpatch script. Exceptions are long lines with debug_log.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 07542fafce72b79335aa5e32f593275956fe1521
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Jul 25 12:25:31 2009 +0000
[batman-adv] Cleanup soft_interface.[ch]
Make soft-interface.[ch] clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with debug_log and using
    KERNEL_VERSION.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 1986dfc09dff22e4acedcad3020ecf3b43bbee1b
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Jul 25 12:18:34 2009 +0000
[batman-adv] Cleanup ring_buffer.[ch]
Make proc.[ch] clean with respect to the 2.6.29 checkpatch
    script.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 46af15b568c9464fed51a8a7ff9a9714dc9ab95c
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Sat Jul 25 12:05:05 2009 +0000
[batman-adv] Cleanup proc.[ch]
Make proc.[ch] clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with printk/seq_print.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 736aa1c40b510de81e58675431cda9a247ea10e5
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Jul 23 12:52:25 2009 +0000
[batman-adv] Cleanup main.[ch]
Make main.[ch] clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with debug_log and the use of
    KERNEL_VERSION.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit e0769b03d8527856dbced9cfda8450bcecb4cf05
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Jul 23 12:43:37 2009 +0000
[batman-adv] Cleanup log.[ch]
Make log.[ch] clean with respect to the 2.6.29 checkpatch
    script.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit aebffe72d0ad33bc466274cfb48f1f9457113a42
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Jul 23 12:38:59 2009 +0000
[batman-adv] Cleanup hardinterface.[ch]
Make hardinterface.[ch] clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with debug_log.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit ca760d395322f7220825ac8b5e2f6047ab621dea
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Jul 23 12:32:29 2009 +0000
[batman-adv] cleanup device.[ch]
Make device.[ch] clean with respect to the 2.6.29 checkpatch
    script. Exceptions are long lines with debug_log and the use of
    KERNEL_VERSION.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit d893c5d00f6369e5a68515794853baabf502d7d8
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Thu Jul 23 12:21:26 2009 +0000
[batman-adv] Cleanup aggregation.[ch]
Make aggregation.[ch] clean with respect to the 2.6.29 checkpatch script.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 8ce38a0d912cb0976717d098fd1a8b4ba9c7e66e
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Fri Jul 17 20:01:33 2009 +0000
[batman-adv] packet.h coding style cleanup
Make packet.h clean with respect to the 2.6.29 checkpatch script.
    Also adds one macro, BAT_PACKET_LEN, which is used in other patches
    to follow.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit ee0cd70c9292869665918b1c3f75bbcf34373562
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Fri Jul 17 19:49:54 2009 +0000
[batman-adv] vis.[ch] Linux coding style cleanup
Fix all the warnings generated by the 2.6.29 checkpatch script about
    violations of the Linux coding style in the vis.[ch] files.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 84456c1e12f87ad0b4101f18d2026252e13cfdf9
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Fri Jul 17 13:25:59 2009 +0000
[batman-adv] move vis_format from vis.c to proc.c
vis_format is currently in vis.c however this variable is only used in
    proc.c.  Move it into proc, and remove the volatile, which should not
    be needed.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit abbb1a48272cb58645619e6140ce2777cb2d9c5c
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Wed Jul 15 17:07:40 2009 +0000
[batman-adv] hash.[ch] Linux kernel coding style	cleanup
Fix all the warnings generated by the 2.6.29 checkpatch script about
    violations of the Linux coding style in the hash.[ch] files.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit e4cda69027def979d0136843daf6d629c8c185a3
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Wed Jul 15 16:58:35 2009 +0000
[batman-adv] bitarray.[ch] Linux coding style cleanup
Fix all the warnings generated by the 2.6.29 checkpatch script about
    violations of the Linux coding style in the bitarray.[ch]
    files.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit d15221cbcee3f4dde045b07dac73aa710b8b6553
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Tue Jul 14 12:17:52 2009 +0000
[batman-adv] schedule_forward_packet(): Remove	unused parameter
send.c:schedule_forward_packet() takes a parameter unsigned char
    *hna_buff which is not used in the function. Remove this unused
    parameter. No functional change otherwise.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit c836b7c37c0ec3258f6a877c647f6cc28aae1f0e
Author: Andrew Lunn andrew.lunn@ascom.ch
Date:   Tue Jul 14 12:14:15 2009 +0000
[batman-adv] Fix some typo's and spelling errors
Fix a few typo's and spelling errors i noticed while reading through
    the code. No functional changes.
Signed-off-by: Andrew Lunn andrew.lunn@ascom.ch
commit 3ec8f43584882f0a26bf336a1b09ac6a128757f0
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Sat Jun 27 07:59:49 2009 +0000
[batman-adv] Convert to new net_device_ops
The current version of batman-adv-kernelland uses the netdevice
    structure instead of netdevice ops introduced by
    v2.6.28-rc5-510-gd314774. The compatibility version was removed in
    v2.6.30-rc6-851-ge3804cb. We need to change now change to this new
    structure to be able to compile it against linux-2.6.31.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit 59e8c577c7b1e51e89c6e933e4eb8ec2368cfe09
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 22:52:56 2009 +0000
[batman-adv] code beautifications & typo fix
commit 85a0ece5c74aac26d314bfe8398245b37cfe8fd2
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 22:43:07 2009 +0000
[batman-adv] fix layer 2 icmp transmission
commit 249dd0777f11e98a882b8c9cb5567ee403516828
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 17:21:03 2009 +0000
[batman-adv] don't display originator table if primary interface is inactive
commit ca67b172f0293021046105f3dfa0e0a15612b06a
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 17:11:39 2009 +0000
[batman-adv] also show state of interfaces in the interfaces file
commit 46ed0657611bf59fc0e73048076345833e520433
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 07:53:40 2009 +0000
[batman-adv] change adv string to lower case latters
commit 4ebf3604804252848a1493c730cf9987b624e232
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jun 24 07:52:39 2009 +0000
[batman-adv] read packet even though it will be ignore to avoid endless call loop
commit dba95eb270ff9674326865a20d25921c69c04d6c
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jun 21 11:07:18 2009 +0000
[batman-adv] add baticmp version check
commit 92cc19374c9ef0804fcf6b55a6b678b4e790e218
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jun 21 10:37:56 2009 +0000
[batman-adv] move packet format related defines to packet.h as this file is included by batctl
commit b2d041d19d0fe479317ab17d85bd05f46359c6c1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Jun 4 17:31:21 2009 +0000
[batman] aggregate packets from secondary interfaces only with similar packets to avoid spreading information about secondary interfaces further than necessary
commit 11e7f35955b9e04af53663c596159524df037b53
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Jun 4 11:22:22 2009 +0000
[batman-adv] add extra hop penalty if incoming and outgoing interface are the same wifi interface
Some setups use multiple radios link on different channels to increase the throughput. This patch allows batman to detect these setups.
commit 81b506ad3975f5410c55c66465ce51c9a67d4d0f
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon May 18 18:00:17 2009 +0000
[batman-adv] marking the trunk as beta as 0.2 is feature complete now
commit f61d0ac881885ca7f9e77c76ca1b04c9c555a89a
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon May 18 17:58:53 2009 +0000
[batman-adv] (re)transmit payload bcast 3 times in order to increase the probability of a successful transmission
The wifi layer will do the same for unicast packets but broadcast packets have a hard time coming through. The overhead created should be minimal as a node remembers the last received broadcasts and drops duplicates.
commit 692d17294760c38f45cb8c118400f338bfc65453
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat May 16 17:01:20 2009 +0000
[batman-adv] porting json vis output format from Antoine van Gelder to batman adv
change the vis format via /proc/net/batman-adv/vis_format
commit 2bdb9489caeff5a185d45f5b717bc63c62ed287c
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu May 14 15:19:46 2009 +0000
[Makefile] make -C replaced by $(MAKE) -C for better portability
commit 974e3e207348207c26180f065708c1b8195ef688
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat May 9 15:46:01 2009 +0000
remove unused gwflags and padding, raise compatibility level to 7, realign
    icmp packet.
commit 43bf47eecc6edf3bcf5cacb73fdf8dfcd44efc60
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 29 18:22:29 2009 +0000
try to activate the module immediately to avoid misleading warning message
commit c93f9a44f9e10fd8355508afac61813d64ff47cd
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 29 17:45:04 2009 +0000
fix crash on module_shutdown() due to aggregated packets
commit 9f7c3c75f38c74dc5a57ed1b9d6624553037e9a9
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Apr 16 12:24:15 2009 +0000
code beautification: concentrate the aggregation code in aggregation.{c|h}
commit 8480927a2f0f1fd94323572f224644a9e1c86c83
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Apr 14 16:56:48 2009 +0000
lock the whole aggregation process to avoid crashes due to sent packets as hinted by simon
commit b08259ee8c1cd2c91701773ea6f13850421b7876
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Apr 12 13:30:11 2009 +0000
start the timer after the new packet was added to avoid race conditions with removing the packet from the forward list
commit 485fc9e88c388638ab9a8ba92ede11b03b4576a8
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Apr 12 00:04:30 2009 +0000
re-implement packet scheduler by using a timer per packet
commit a70e62de8774ef12c81fd56649ce77947eb46beb
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Apr 10 22:44:30 2009 +0000
remove the "error" from the kernel header folder check as requested by sven
commit 2f3d2f0724360f2253c93607c745a2ce33203d46
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Thu Apr 2 17:54:49 2009 +0000
Don't access random memory after forwarding
B.A.T.M.A.N. advanced iterates over every known interface when receiving
    data and tries to forward as much as possible data from the same
    interface as possible using a while-loop inside the outer loop.
    When it receives a broadcast ethernet frame which needs to be forwarded
    again it will try to send it to every known interface again. This loop
    is inside the first one and used the same pos variable as the outer
    loop. After the inner loop has finished it will point to a memory
    location which is not part of the interface list, but the while loop
    starts again and tries to access this memory region without knowing what
    it is and to what it belongs. This could lead to a kernel oops or any
    kind of other unspecified behavior of the kernel.
    The inner loop should use a seperate position variable to iterate over
    all interfaces for the broadcast.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit c4b719f1958c5218f84487492c341221befe5b59
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Apr 2 12:34:51 2009 +0000
restore irqs when putting a character because __put_user might sleep (thanks sven)
commit fb4f62329e4866e7d73a9863f00ecd89210e23d3
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Wed Feb 25 04:23:02 2009 +0000
[B.A.T.M.A.N.] [PATCH] Fix deletion of aggregate_ogm in
remove_proc_entry needs as second parameter the parent directory and
    not the proc_dir_entry of the file to delete. Otherwise nothing will
    be removed the deletion of /proc/net/batman-adv will fail because it
    still has child nodes.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit eb5e7e5183e7f98054c0f927a4d3bf73f5711b23
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Feb 19 20:10:42 2009 +0000
don't check the lock by using spin_trylock, but pass a parameter to
    control whether workqueue must be rescheduled or not.
commit b93ba89bdeca659e0454311d15a1c5a9580fc5e8
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Feb 13 17:20:57 2009 +0000
make aggregation_enabled changeable via /proc/net/batman-adv/aggregate_ogm
commit 3ae41693bb1ddd8705679057e2dc8f401e41f45e
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Feb 13 15:01:35 2009 +0000
cancel_delayed_work_sync() is only available for linux >= 2.6.23
commit 1a1664695f85f5b0e9d562e821596955690ad908
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Feb 13 11:09:24 2009 +0000
remove obsolete if_lock
commit b780db96954ae1f950b545fa45494ae9a8dee4b4
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Feb 13 10:57:31 2009 +0000
add packet aggregation
    add jitter for rebroadcast of packets if aggregation is disabled
commit 345247203b11351695f20c0f04774a42fca43f6e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Jan 31 21:03:34 2009 +0000
some minor fixes (add forgotten rcu, fix queue name, replace obsolete function call, etc)
commit 40e14c3c0d36a34ad530c511972d331c3599b085
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Jan 31 21:00:46 2009 +0000
some more code cleanups
commit 79c1b0dbd8c75720487978b50c7ab590c7cebcdd
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 25 09:53:51 2009 +0000
more code cleanups for better readability
    preparing for packet aggregation
commit 6e8cc51de1c20470ff0d524508357a44f5912ebb
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 25 09:43:10 2009 +0000
code cleanups for better readability
    preparing for packet aggregation
commit 6f3fb29480773737db6a9fafee989352b3435a67
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 25 09:21:30 2009 +0000
change version string to reflect new major version
    add understandable error messages to icmp packet injection
commit 13e03c8aa34f4f654e788f1bad16c5412094cd75
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Dec 28 00:04:45 2008 +0000
prepare 0.1 release
commit 6b4a0816c4badfb73456be16c44ccd6478879d61
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Dec 5 19:51:41 2008 +0000
decrease JITTER so we can support lower orig_intervals ( > 40)
commit 4b55a35934e6504138d0f8bbb9c786cf84553236
Author: Scott Raynel scottraynel@gmail.com
Date:   Thu Dec 4 02:32:47 2008 +0000
[PATCH] batman-adv-kernelland: Fix memory corruption bug
I've been spending some time tracking down a bug that's been causing
    memory corruption followed by random kernel panics. Thanks to the
    kernel's slab memory debugger I tracked it down to a kfree in send.c
    that was freeing a block of memory that had been written to past the
    end of its allocation.
Turned out to be a simple typo, which I've fixed in the following
    patch. When resizing the packet_buff struct in batman_if, the new
    length was being updated but the old length was being used for the
    kmalloc(), causing something later to think it had more memory
    allocated to write to, hence writing past the end of the allocation.
Signed-off-by: Scott Raynel scottraynel@gmail.com
commit 1bd0020c40b3efec60dab1f97e65160a629a67ca
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Dec 1 16:13:33 2008 +0000
remove (empty) batman-core directory
commit 6a68c39dc78be9c0bd3dde04b10665ab88a9e56b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Nov 16 16:41:27 2008 +0000
remove "core" from the kernel modules filename
    update README
commit d5731b07486506c2971e5fa3297c3606f172c59d
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Nov 12 14:23:33 2008 +0000
remove obsolete 802.11 stuff, move batman-core into batman-adv-kernelland
commit 527017e3d21cd4290ec9c5607e20ab58372c0857
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Wed Nov 5 21:51:19 2008 +0000
Use device_create instead of device_create_drvdata in linux modules
The function device_create_drvdata was renamed to device_create in linux
    v2.6.26-5782-g4e10673. Until v2.6.27-4446-g7fb6b5d there was a compatibility
    define in include/linux/device.h which is removed in linux 2.6.28-rc1.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
commit 589ffa79833f4ed81ba8efe032cb2d10ddacae62
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Sep 25 22:00:35 2008 +0000
change htons to __constant_htons where constants are used.
commit 65d32416ed458a105e79dcf01221c042d16583cd
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Aug 3 18:14:07 2008 +0000
* rename bat_device to soft_device in order to avoid confusion with the
      /dev/batman-adv stuff
    * remove rcu-lock in packet_recv_thread to avoid schedule-while-atomic bugs
    * move creation/deletion of the bat0 device into init_module()/cleanup_module().
      Scripts should not need to wait for creation of bat0, but can configure/bind
      it as soon as the module is loaded.
      (scripts like 'echo eth0 > interfaces; sleep 1; ifconfig bat0 1.2.3.4 up'
       are racy and ugly.)
commit fa137b532bc7673c7b31365421132e1fe035a232
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Aug 2 15:14:30 2008 +0000
FIX various unloading synchronisation things:
    * reload hardif check timer even on failures
    * change module_state semantics (inactive is now really inactive, not "ready for reactivation")
    * module should not be reactivated in the middle of deactivation
    * kthread should wait until parents kthread_stop()
commit 089633cb3e7dc00bea40305e986e212b56fd7898
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Jul 25 13:45:00 2008 +0000
changing more kmallocs to GFP_ATOMIC, add checks
    remove seqno_lock, using atomic_* should be enough
commit 0384ad13fa7bf17e45d5bec0c6034a46a4f2c15b
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Fri Jul 18 11:48:12 2008 +0000
Don't process the icmp packet further when we could not read it
__copy_from_user returns the number of bytes it could not read when it
    fails. It makes no sense to process the data further if we don't have it
    in a complete form.
    access_ok will only check if it may be valid but not if it is definitly valid.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit bc0a5dea857b678ac11a90074fc52122a521709a
Author: Sven Eckelmann sven.eckelmann@gmx.de
Date:   Fri Jul 18 11:41:39 2008 +0000
Move more compatibility code to compat.h
Use the madwifi-like compat.h to keep as much of the #ifLINUX_VERSION..
    stuff out of the code to increase the readability.
    Parts which weren't moved are the device creation/deletion, which is to
    different to move it to compat.h, and the module count.
Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
    Signed-off-by: Simon Wunderlich siwu@hrz.tu-chemnitz.de
commit 4e7a65c455952a5f571bb871a4150880130be113
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Jul 17 08:01:32 2008 +0000
apply Scott Raynels patch to support linux 2.6.20, 2.6.21 (thanks!), updated README
commit 8009e9f7ac4f46e6ecdb24f366164edd64ffc3eb
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Jul 12 13:48:22 2008 +0000
set bat0 MTU according to the MTUs of the hard interfaces
commit de34abc50418202eb20961b3f2e2ca561e53107f
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Jul 11 10:45:29 2008 +0000
update the README
commit e559c76f5306e5c2e8ee90e7f39c04f030322122
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Jul 11 09:47:58 2008 +0000
some functions have been renamed in 2.6.26
commit 0fc206e39de7c3c3bff423a2264d09b2e859c390
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Jul 10 21:32:27 2008 +0000
use GFP_ATOMIC for iterators (called within locks)
commit 5367dc92af4f0fb36154b304aba1a4773ac2f54e
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Jul 9 21:22:27 2008 +0000
do not waste the stack in proc.c
commit 75d0cf51b353fb5149f0983338b8e3ff658a6e2d
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Jul 2 09:36:43 2008 +0000
FIX: spinlock deadlocks. use spin_lock_irqsave() or spin_trylock() instead, so
    non-interruptible functions don't have to wait (forever) for their lock.
    added some more sanity checks
commit 88c2d87e6cf6cdc4f42892596eea9fc08a296f4f
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Jul 1 21:24:50 2008 +0000
do not use so much stack space.
    stack is pretty limited (8k or 4k per process/thread) in the kernel.
commit fd52df281881299ab1b294bdcacde292d73e9ac0
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Jun 30 19:36:33 2008 +0000
FIX: convert timers to delayed work for work queues
    timers are not interruptible, and spinlocks will deadlock in the timer handlers
    when the lock is still hold by another thread.
commit 511f3335ce7863d5f4561a8e1a72adfc9e087ad8
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Jun 30 11:07:29 2008 +0000
clean up send_packet(): do checking on batman interfaces before using them
    add some comments
commit 76c00b2ac315bc64035f60bc754a0d1ff25cedaf
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Jun 25 21:19:17 2008 +0000
FIX: resize orig nodes in hardif_add_interface() so it's resized before added into if_list
    FIX: check for activeness before comparing net_dev->dev_addr (it's NULL if inactive -> BUG)
    clean up locks, NULL workqueue after use
commit 5dd1a00f694186cb86f215027f004412153d24c8
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Jun 25 20:04:11 2008 +0000
FIX: cleanup vis_quit() for corner cases
commit 14caf4399de645fc1e1c6efd7e67b768b3da3286
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Jun 23 10:09:42 2008 +0000
Change Ethertype from 0x0842 to 0x4305.
    This Ethertype is not IEEE registered, but does not collide with something
    else as far as i can see.  (0x0842 = Wake on LAN, also not registered at IEEE)
If you want to sponsor an official Ethertype ($2500) please contact us.
commit 49f203a47f0e1cc68b6db62e0a3fa2625ee595ca
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Jun 22 17:53:33 2008 +0000
use dev_queue_xmit() instead of kernel_sendmsg() to send packets
commit 0949cb1a27872f25e0b4513ab7155e1b5ae83cbe
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Jun 21 10:44:57 2008 +0000
FIX: don't lock orig_hash when receiving packets. recursive deadlocking might happen.
    (i hope this really fixes the "ping of death bug" now ;] )
commit ee57c209930fdc69aba31cf1da7360d5cf0c5758
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Jun 10 09:12:31 2008 +0000
FIX: The "ping of death" bug
    It was originally caused by congestion.
    rtnl_lock()ing the netif_rx() call fixes the problem.
commit 7381990f93c1abb2296d64d6e88ae391e8b75a8c
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Jun 10 08:53:58 2008 +0000
move receive_vis_packet to routing.c
    FIX: ignore unicast packets for others in an hub enviroment
commit f8b8d583b370cbd4e223d99da0ab2d66bb654550
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Jun 7 15:33:47 2008 +0000
FIX: a typo, responsible for bloating the list of neighbours. :(
commit 1821f1b27d3ed5d170fc05851aed390395f9f94b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Jun 5 17:49:27 2008 +0000
FIX: BAT_HEADER_LEN bad brackets
    FIX: fix test for svn/svk binary
commit 88cd813125a65375e6fbbaaa486031dcbabe18e6
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue May 27 12:54:04 2008 +0000
add "version" field to all batman-adv packet formats, increase COMPAT_VERSION
    output "unknown" in Makefile when no svn/svk is installed,
    instead of letting sh throw "svn/svk not found"
commit 17504793b7c77c71d4af37c4460b8cfcccd27e93
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri May 23 15:04:29 2008 +0000
FIX: initialize send_list in own packet
    FIX: check if interface is up when receiving
    change "batman_if->net_dev->name" to "batman_if->dev"
commit c146ca2bfdf3f41f45b80f45fdf2488181c79ede
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue May 20 17:57:01 2008 +0000
remove broken link
commit 283b1928efb2150479c3c09d80d09b4a97035bf2
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun May 4 12:01:36 2008 +0000
don't malloc when incoming packet is invalid
    clean up readability
commit 2df687e24f3214faadfb49fcef903f62a6b922fc
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat May 3 11:40:28 2008 +0000
update batman-adv-kernelland README
commit acc43009bbed11c5616051d97772d5609d009469
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri May 2 11:03:17 2008 +0000
enable vis clients (to be set by writing to vis-procfile), be client by default
commit 64ea78c0d1b3b2fe5455cc8fc6f71788faaf6898
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Apr 26 06:43:56 2008 +0000
fix various vis server bugs
commit 08e56d7b28c5918d38503116f252492ee846746b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Apr 24 10:23:33 2008 +0000
FIX: activate module when hard-interfaces become available (welcome back, chicken-egg)
commit 739a35c80f738e98006dbac9ebf6f3d74bc53676
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Apr 24 03:16:57 2008 +0000
update proc (part 2, forgot to commit ...)
commit aa9b970f72b33ce7cdecfdb751a9f195b237cb17
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Apr 24 02:52:17 2008 +0000
change procinterface to only use single_open/seq_printf
    remove obsolete gateway things from proc
commit 910d84ce666f55be78e75ec3673e914be69cfb41
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 23 08:00:28 2008 +0000
renaming the sources (final part)
commit 700b8b2d0ecb32c9de1d7b55189442b986a82fa4
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 23 07:37:50 2008 +0000
renaming the sources (part II)
commit 629ad07b91def2ab4917564e09e5989f646fc191
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 23 07:24:47 2008 +0000
renaming the sources (part I)
commit 3f96d400a9d0350d252284b0c5f2f74a6ea36d24
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Apr 23 06:59:36 2008 +0000
fix rcu crash, synchronize events
commit 29a2a3dc52f41863b742c5e30599ab075d667a25
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Apr 23 03:47:08 2008 +0000
hard interface rewrite
commit f0fc198ca913d9754632fbc15a516674a987a5eb
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Apr 22 07:00:02 2008 +0000
remember who sent which packets to decrease traffic
commit 91edb4b111886b62f666ae65bf346f3342e7c24b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Apr 21 08:48:00 2008 +0000
sending information to all vis-servers in range, don't use broadcasts
commit d20f8174630ff0ddb3809408f4d91d5871b0a9e6
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Apr 21 08:20:39 2008 +0000
organize vis packets in a send queue, only use one timer
commit 8cebe2910f2d4569ca9c6e789f9a75d7cb994455
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Apr 20 15:11:50 2008 +0000
add _rcu() things (incomplete, but you'll get the picture)
commit 5916bf047d89e2ab1f25e750157d4f3fb9d6dca3
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Apr 12 20:27:39 2008 +0000
FIX: remove locks in send_packet again, this produces deadlocks with send_own_packet() and  del_timer_sync()
commit 8533c5657be8613c1e220ee42d27d04379393648
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Apr 12 19:21:41 2008 +0000
change if_list_lock from spinlock into mutex to avoid "scheduling while atomic" bugs
    (only?) performance-critical if_list_lock'ed function is_my_mac changed to lockfree version
    FIX: removed deadlock: recursive locking batman_raw_send() -> deactivate_interface() -> shutdown_module()
    add comments to functions which requires locks to avoid such deadlocks
    FIX: cannot call sock_release when called from a function which uses this socket; therefore just adding IF_TO_BE_REMOVED, and call deactivate_interface() later.
    add workqueue for check_inactive_interfaces(), because sleeping in timer function is not allowed
commit a50f63ad722cfed934b8891cff6e05735e954891
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Apr 8 22:50:51 2008 +0000
FIX: (try to) activate interface when adding, otherwise it will never come up.
    (it's a chicken-egg problem: inactive timers are check in purge timer, but
     this timer won't be started without the main interface and active module)
    FIX: interfaces_read: use dev name, not netdev name, as netdev may not be available
commit 6137d997fef3145f4bf7a8650b4c104a266fd7a8
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Mar 29 02:07:31 2008 +0000
re-activate deactivated bcast_timers
commit d83571147325710e69a7e38347bc4387117ecb22
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Mar 28 10:20:34 2008 +0000
add some README to describe the usage
commit 888ebb2a1396437a6251ceecf3f03ab595f919a4
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 26 13:48:23 2008 +0000
try to remove timers before (re)adding them.
    rename last_seen to first_seen (it's only set the first time after all ...)
commit 1880a29deee1fef665c5954947392700daccb596
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 26 13:15:22 2008 +0000
add support for vis client update packets
commit dca37c899ff55e669d20060f45c97f17f82aa686
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 26 09:29:14 2008 +0000
use seq_file for long procfs output
    remove obsolete debug things from vis
commit d25863e8ca6002bfc5f13577bacc53181fa16100
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Mar 25 13:51:01 2008 +0000
FIX: don't activate timer for inactive devices
commit 5800a58025a96fc8ec33c3fe3b021f6958c79d90
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Mar 25 12:23:37 2008 +0000
initialize if_active ...
commit ae462b10b2912329ee2a6a117e915f516359cc81
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Mar 24 05:27:01 2008 +0000
fix crash in shutdown_module()
    make kernel module more independent from the main interface
    don't activate the module if the main interface was never up
commit 188a62c576f78771f376fb4569d2459000718f49
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 15:40:23 2008 +0000
after locking, we should UNlock -.-
commit 36e482453f35f1a314d1657abded28a9f2ff7c3a
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 14:53:12 2008 +0000
FIX: never purge own vis-packet
commit 3e63f37d36390acc687fe7293499d16965f978db
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 13:54:30 2008 +0000
add hna to vis-packets
    purge vis-packets after a while
    fix various vis-bugs
commit abea19bf5646403656bc403be2dbdb207f63f191
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 12:22:19 2008 +0000
FIX: forgot to call vis_quit() in shutdown_module
commit fd99b351db8ee41ea3604aaee48223eb75c35eed
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 12:08:24 2008 +0000
QUICK FIX: don't crash if first interface is deactivated.
this will simply not start if first interface is not up.
    some things rely on the first interface as originator mac,
    and expect the net_dev there.
    Maybe a better fix would be to have seperate list for activated
    and deactivated interfaces?
commit a5d6a89ea8cec5ac625a52806fbc158b294cab49
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Thu Mar 20 11:07:21 2008 +0000
add initial meshing vis server
commit 3dc6f392e36608392a5a2a51726ef06c7caf031c
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 19 15:40:04 2008 +0000
add spinlocks to interface_tx (broadcast)
commit 33823ebb768169bb357cb9604ad91ca5a71cd137
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 19 15:24:14 2008 +0000
little cleanup
commit 2a7f1c61fa0eff35c5fbcc95bba5ba97d03106d4
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 19 11:53:57 2008 +0000
move batman_if back into types.h, change #include order
commit 4dd3000e5edc25c081ad5216a3b51e44311f0890
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Mar 19 11:28:32 2008 +0000
keep bat interface while adding new interfaces
commit 6ceb394c34ceb520aecf6c20254ceb607f4ac48b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Mar 19 10:53:55 2008 +0000
(de)activate interface(s) dynamically
commit 3e03bd304533665338f161b73cb91b037d2a740c
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Mar 17 08:57:01 2008 +0000
drop packets SILENT when buffer is full
commit ade63b44d07220d0662b2432d72a23271a20f02e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Mar 15 16:38:34 2008 +0000
remove unsued unidirectional flag
    use the improved default values
    remove perfect_tq_penalty and add hop_penalty
    fix typos
    improve makefile speed
commit 1b4977d5105f511e9616f2af47d056adc0b12631
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Mar 14 15:53:10 2008 +0000
more spinlock vodoo to avoid race conditions
    makefile performance improvements
commit bb7a3b92100e1617baebe5883f428d082911c12f
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Mar 14 15:27:40 2008 +0000
fix typo -> rev1001 => YAY
commit e5d0fc5c4e61790e4f03f92746d61d906b127bc3
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Mar 14 14:50:36 2008 +0000
FIX: securing cleanup_module() with a few spinlocks (it freezed on rmmod),
         please review this patch.
YAY, 1000 commits. cheers ;)
commit cb973eff7f28b73dbbb4f750ca0d50ad8bbf8486
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Mar 14 12:21:49 2008 +0000
- fix overflow in integer expression
commit bfdd2fa2c7f051e46e52bafbae3c01dfa22c4c3d
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Fri Mar 14 11:04:13 2008 +0000
FIX: unreference interfaces when removing interfaces with dev_put()
commit 92cc65d5e8363ee6bba729ba688e5bc32876854b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Mar 12 12:02:41 2008 +0000
FIX: bat0/module got deactivated when there was an error in interfaces add
    e.g.:
    adding eth0 twice:
    echo "eth0" > /proc/net/batman-adv/interfaces
    echo "eth0" > /proc/net/batman-adv/interfaces
    deactivated the module.
commit b50d8bb98ed53ce7b3dabad9297115e546534888
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Mar 11 10:09:01 2008 +0000
count packets which had to be reexpanded
commit 6723ad4d8a77188ac62f6f1b117b34a9d1ca4aab
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Mar 10 16:01:43 2008 +0000
add support for net_namespace stuff in 2.6.24 (init_net etc)
commit e114742128ad75c788564b56a87df91793c58b2b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Mar 10 15:20:23 2008 +0000
use EXTRA_CFLAGS instead of CFLAGS. linux 2.6.24 won't allow CFLAGS anymore
commit ca80ad470b2375c3f323ae546d6ea7438ba372e6
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Feb 7 20:32:13 2008 +0000
fix crash on payload send
    don't allow bat0 mac address change
commit 41d53c908321acad0ac50029a6c49286a49000d6
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 20 23:24:04 2008 +0000
update readme
    make batman adv userspace makefile openwrt compatible
commit 57017c53b71650cd3ba0f9c143bbecf5f39aaa93
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 20 16:22:26 2008 +0000
fix revision string
commit e138980691c4c2e7ac62d569da4033482f3e334a
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Jan 16 18:07:45 2008 +0000
seperated makefiles for openwrt integration
commit 167c2b7336d13ec523b934802f100182d59de9e4
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Jan 15 15:06:44 2008 +0000
clean up batman-adv-kernelland directory, write a README
commit eb0e2d97fc38c5347c80cbd41718ec1345bb7d79
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Jan 14 15:43:00 2008 +0000
it is better to purge the neighbors after timeout :-)
commit 2750431f71625be4417d0600f0e9607771170d46
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Jan 13 22:19:38 2008 +0000
- update author information
commit e568a0457bf83f0e5c73c18b4dd045707f1ca055
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 13 22:14:25 2008 +0000
setting local mac address timeout to 1 hour
    added /proc/net/batman-adv/transtable_local and /proc/net/batman-adv/transtable_global for viewing the current translation table
commit 723899c5147a2b8c1db87d45c64398f5df259fb7
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 13 18:04:50 2008 +0000
fix local hna size
    hide unused gateway file
commit 499761efe5b9a4846fb9573b8ad150128cfef8db
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Jan 13 15:44:30 2008 +0000
small fixes in sending local hna buffer
commit 63f1fe56c7b8851e17be5188cb08e240e2719446
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Jan 12 04:15:17 2008 +0000
purge local hna if global hna received and vice versa
    send num hna with batman packet (ethernet packet size again ...)
commit c1c4ab14623b96205cabb990b8d861041f037c80
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Jan 11 14:34:55 2008 +0000
fix: batif mtu, local hna, device_client_hash, skb->len, rebroadcast of BAT_BCAST
commit 9742bc8f57baaf905e904a038e6eac13838b14cb
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Jan 11 04:10:27 2008 +0000
multicast support added
    local hna broadcast implemented
    translation table implemented
commit 742e786e4a2c6595c952bc82781a34d019a2c066
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Jan 8 20:58:35 2008 +0000
fix receive batping
commit 486a7e3a0a6fe939d12ed56340083dfde8149886
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Jan 8 19:27:17 2008 +0000
fix destination unreachable
commit d09417e211484470731b23c1cd04a069cc7e7779
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Jan 8 18:56:58 2008 +0000
some fixes and beautifications
commit ace8385c73b095ea3a27e0f52cc98747a80e3ab7
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Jan 5 16:50:08 2008 +0000
complete battool connection device (read/write/poll)
commit c163b3b3e4f4375cbbd31272faa69569cce19626
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Jan 5 10:58:29 2008 +0000
fix receive packet
    fix originator string
commit e580066f71a71bbabd95ad3032b2c23702624441
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Mon Dec 31 00:16:45 2007 +0000
- support KERNELPATH parameter (ripped from madwifi) for crosscompiling
commit 1499b1016af91ec5f22daa938530709d17d2dcaf
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sun Dec 30 14:55:29 2007 +0000
- satisfy sparse. yay. thanks sven
commit 32c61c70057c89413e565beaf883ce648c9951de
Author: Andreas Langer a.langer@q-dsl.de
Date:   Fri Dec 28 19:25:21 2007 +0000
porting to 2.6.23 ;)
commit 1d2f0422fbc6221ff1e4d33c6cafc6ad888f7247
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 26 23:45:56 2007 +0000
- FIX: whoohops, forgot to change a line
commit 2203ac1115302df915bdafd3dab5e68a6f17f670
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 26 23:42:47 2007 +0000
- revert back to simple log_level with volatile uint8_t
commit 271e9c8055f752f796b87bc6cfaf0cda5a35befd
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 26 18:57:58 2007 +0000
route icmp packets
commit 001eddaf35c41aa1476687b9313f004d95eaf65d
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 26 16:15:11 2007 +0000
add /dev/batman-adv for battools
commit 9fa1d5ce43f1858ad3a44a0471c6df86bb7a1560
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 26 15:35:57 2007 +0000
remove unused locks
    fix some routing bugs
commit 094a751463a9a14a9e96a02047a1d0f8f15ac317
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 25 22:28:23 2007 +0000
route outgoing traffic
commit 410e194c0dc3d618d38b13999cd56a794767dff3
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 25 11:55:59 2007 +0000
ethtool support added
    unicast and bcast reception implemented
commit 4c9992fc79bcc2171c78b8b951c0e3c5e698a0d1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 25 10:33:52 2007 +0000
fix null pointer dereference
commit 80c0c219aa56987216483d0b8b695d84efed6f94
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Dec 23 14:36:44 2007 +0000
add layer 2 interface (batX)
commit b21378a554f09a943e9b074e3716fef14c23ea80
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Dec 11 18:35:48 2007 +0000
fix minor bugs
commit 937148f0f8a3ef5303a48160553a1f767c9ad06b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Dec 8 20:51:51 2007 +0000
- remove debug-line
commit 26aef36daffc7af57a09dfdb104f145d84ff92a1
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Sat Dec 8 20:50:17 2007 +0000
- FIX: bitnumber is not bitmask of the bit which is to be toggled
commit 1682727a550dbc21bfce6a2beeb5b334129d1aad
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Dec 8 02:50:45 2007 +0000
rename LOG_TYPE_ROUTING to LOG_TYPE_BATMAN
    LOG_TYPE_ROUTES added
    updates_routes() implemented
commit 4cf49ad2312e0a440c33ecc85087c653d9c54cfa
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Dec 8 01:58:22 2007 +0000
make the routing work (testing looks good)
commit 47edf181fbc4fb315743f758a56b403985ca50ce
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sat Dec 8 01:20:40 2007 +0000
fix some freezes and annoyances
commit cdb72a59903d67cd2595f947caa5f6b930a1928b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Dec 6 16:59:10 2007 +0000
purge originators
commit bab8fb678e7aeda8090884dd1884abb282105ab0
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Dec 6 16:05:23 2007 +0000
fix freeze after rmmod
    resize originators when a new interface is added
commit bf24572fe61381283505d594126e5cb23a44ea53
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 5 20:57:27 2007 +0000
fix receive packet - now packets can be received !  ;-)
    rmmod with active interfaces allowed
commit dc3ebe4a53748fcf289ceb2ff23a9f3176b718d5
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 5 19:28:53 2007 +0000
- atomic_set() annotation
commit 251891588fa6e69b068f50edf914a58145e7f97b
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 5 19:16:49 2007 +0000
- fix log_level again: allow it to be cleared again
commit eda29edb45bec37e6bc520260b2900a1f2d46458
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 5 19:00:08 2007 +0000
fix setting log level
commit 6f2ae28bef4a3c6115b333d4cb1326604f61b8b9
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Wed Dec 5 17:40:07 2007 +0000
- add jiffies to batman log, batman-adv: only in kernel log
    - printk()s can probably be also replace by debug_log()s
commit 1b987762f3f95684b2b9960e86fc328205526a9e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Dec 5 14:17:38 2007 +0000
fix various race conditions in orig_interval and log_level
    fix non existent write() in /proc
commit 98b38a6ef12b9f14e9293e7acf7c7155b24056a8
Author: Simon Wunderlich siwu@hrz.tu-chemnitz.de
Date:   Tue Dec 4 22:50:04 2007 +0000
- add log_level proc-file
commit 78078ec10d99c34900db72049d5708c6ded6c111
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Dec 3 16:34:03 2007 +0000
minor fixes in the routing code
commit 58232977fbaf9f137d85079ac78e0dbec6b733cb
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Dec 3 00:37:19 2007 +0000
basic routing code complete (hases and ring buffer added) -> needs testing !
commit 83a8342678a06958cbf11208378b461126c500b9
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Sun Dec 2 01:23:08 2007 +0000
more basic routing code added (forwarding packets / bitarray added)
commit 187cc4e123a62067d437d0be493ab754c46e7097
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Nov 30 22:09:43 2007 +0000
basic routing code added
commit 2462acd6f9a94b1ebd293d4d1c1ac20d1e773d90
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Nov 30 16:36:13 2007 +0000
implemented kernel thread for packet processing
commit b8898fc20f32d2d1fbcb4012d9c3d500cdcda3c5
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Nov 30 01:09:12 2007 +0000
logging infrastructure added -> /proc/net/batman-adv/log
    new logging mechanisms implented (LOG_TYPE_CRIT|LOG_TYPE_WARN|LOG_TYPE_NOTICE|LOG_TYPE_ROUTING)
    preparations for HNA
commit cac6d5565b7719df1421b357a9752f1e730a15a1
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Nov 28 18:27:07 2007 +0000
fix random issue
commit d1fbb61d05345da2e091186296de21f005a07935
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Nov 28 17:38:00 2007 +0000
raw socket operations added: create / destroy / bind / send
    broadcast of own OGMs implemented
    orig interval configurable via /proc/net/batman-adv/orig_interval
commit 6dd7f820835e998514273456cdd1cd2d8c24b36e
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Nov 27 23:24:31 2007 +0000
interfaces can be addded & deleted via /proc/batman-adv/interfaces
commit c4e57f35d3009afd3b7e7b63c51b277be64e241d
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Nov 27 21:24:12 2007 +0000
rename files for multiple file support in the Makefile
commit 7eba64340145c2c85a2eb0f6143e08115ecd8597
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Tue Nov 27 21:19:43 2007 +0000
add procfs configuration support
commit 6987a8eea0bd4193950d9f8c1f87de66b209188b
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Oct 5 11:38:36 2007 +0000
unify naming scheme
commit c7bbcf6069035c5c7600fcd9370074ff21e90308
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Wed Sep 19 21:34:59 2007 +0000
batman adv k: sending beacons
commit d48e3125620a01f5076d70b20b3f12c278e64458
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Sep 17 21:48:06 2007 +0000
forgotten patch  ;-)
commit 1d9c159d043539467cfaa62a70f47dab1041d018
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Mon Sep 17 21:22:31 2007 +0000
batman adv k: intial beacon setup
commit 5ad24eb313d1474f8542cd4ed04bb39898b3423a
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Sep 14 16:11:33 2007 +0000
batman in kernel space initial test setup II
commit dc0a074c450493a7b5e7e01e7a23cc74e7ad9f57
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Sep 13 18:45:37 2007 +0000
batman in kernel space initial test setup
commit c3fa6ee5f33087e6839bb223eccc04ab723ac294
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Fri Sep 7 22:53:56 2007 +0000
improving madwifi patchset
commit e5e275572a9f0a042a05da5df67048a5b6ede092
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Sep 6 21:56:33 2007 +0000
beginning patch framework for madwifi
commit 79a0e4f09f1576b55694ef2bb927f1c9795b25cc
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Sep 6 19:50:09 2007 +0000
preparing trunk structure for batman-adv-kernelland II
commit a6e241203bd1d93a43294cfd3537b843c5b5ad98
Author: Marek Lindner lindner_marek@yahoo.de
Date:   Thu Sep 6 19:46:53 2007 +0000
preparing trunk structure for batman-adv-kernelland
-----------------------------------------------------------------------
-- 
batman-adv