Johannes Berg [Wed, 1 Jan 2014 20:40:40 +0000 (21:40 +0100)]
backports: backport multicast list handling in iwlwifi mvm
Backport multicast list handling for kernel < 2.6.35 in iwlwifi mvm
driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 20:36:31 +0000 (21:36 +0100)]
backports: remove usage of addr_assign_type in 6lowpan.c
The bluetooth 6lowpan code now accesses the addr_assign_type member of
struct netdev, which is not available in older kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 20:35:15 +0000 (21:35 +0100)]
backports: add include/net/af_ieee802154.h file
This file was added in kernel 2.6.32 and is now needed by bluetooth.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 20:33:49 +0000 (21:33 +0100)]
backports: remove usage of members of struct property for kernel < 2.6.39
struct property is missing in the general header files in older kernel
versions. There are different versions of this header depending on the
architecture code in use. This removed the access to any members of
struct property for older kernel version when CONFIG_OF is not set.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 19:17:54 +0000 (20:17 +0100)]
backports: make b44 depend on kernel > 2.6.28
The mdio bus functions where introduced in that version.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 18:16:50 +0000 (19:16 +0100)]
backports: fix number of arguments of phy_connect()
The flags argument was removed from phy_connect in this commit:
commit
f9a8f83b04e0c362a2fc660dbad980d24af209fc
Author: Florian Fainelli <florian@openwrt.org>
Date: Mon Jan 14 00:52:52 2013 +0000
net: phy: remove flags argument from phy_{attach, connect, connect_direct}
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 18:10:17 +0000 (19:10 +0100)]
backports: add include/trace/events/v4l2.h file
v4l2 now supports tracing and needs a new header file that was missing.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 18:09:48 +0000 (19:09 +0100)]
backports: add ARPHRD_6LOWPAN
This adds ARPHRD_6LOWPAN to if_arp.h
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 18:07:05 +0000 (19:07 +0100)]
backports: add netdev_notify_peers()
netif_notify_peers() was renamed to netdev_notify_peers() in kernel
3.7. Replace the calls to netdev_notify_peers() with
netif_notify_peers().
Some older kernel versions do not even have netif_notify_peers(), there
we just remove the call, currently it is only used in the 6lowpan code
and I assume it will still work, but not so good.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 1 Jan 2014 18:03:52 +0000 (19:03 +0100)]
backports: add devm_kmalloc()
devm_kmalloc() was added in kernel 3.13. To backport it we replace the
calls with devm_kzalloc() instead.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Wed, 1 Jan 2014 19:37:13 +0000 (20:37 +0100)]
gentree: combine spatches (unless using --gitdebug)
Since spatch is rather slow, but can handle multiple spatches
concatenated in a single file, just do that and run it only
once rather than for each spatch. That shaves off some of the
runtime (startup etc. and finding affected files.)
On my system, I go from
real 9m42.616s
user 8m48.352s
sys 0m22.884s
to
real 9m1.948s
user 8m40.108s
sys 0m12.088s
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Fri, 20 Dec 2013 20:47:26 +0000 (21:47 +0100)]
backports: backport hex2bin()
This changed from not having a return value to having one,
so backport the version that has it.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Luis R. Rodriguez [Mon, 9 Dec 2013 03:12:41 +0000 (04:12 +0100)]
backports: refresh patches for next-
20131206 a second time
After a few changes were made we need a refresh of patches
against next-
20131206, this refreshes the patches and test
compiles against all supported kernels again.
$ time ckmake --allyesconfig
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.2.52 [ OK ]
19 3.3.8 [ OK ]
20 3.4.70 [ OK ]
21 3.5.7 [ OK ]
22 3.6.11 [ OK ]
23 3.7.10 [ OK ]
24 3.8.13 [ OK ]
25 3.9.11 [ OK ]
26 3.10.20 [ OK ]
27 3.11.9 [ OK ]
28 3.12.1 [ OK ]
29 3.13-rc1 [ OK ]
real 33m3.982s
user 892m21.988s
sys 128m11.492s
Current run time *without* the last two cocci patches applied:
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py \
--verbose \
--clean \
--refresh \
/home/mcgrof/linux-next/ \
/home/mcgrof/build/next-
20131206
real 3m13.530s
user 2m56.104s
sys 0m16.180s
With the last two cocci patch files applied and spatch version 1.0.0-rc17:
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py \
--verbose \
--clean \
--refresh \
/home/mcgrof/linux-next/ \
/home/mcgrof/build/next-
20131206
real 23m2.521s
user 19m19.436s
sys 3m3.840s
With the last two cocci patch files applied and spatch version 1.0.0-rc19:
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py \
--verbose \
--clean \
--refresh \
/home/mcgrof/linux-next/ \
/home/mcgrof/build/next-
20131206
real 8m6.077s
user 7m46.076s
sys 0m18.124s
Moving forward please make sure to have present at least spatch
1.0.0-rc19 on your system to generate releases. The gains of extending
how we do backports more automatically with Coccinelle for the
short cost observed on 1.0.0-rc19 makes it worth it to continue to
embrace SmPL to backport further collateral evolutions.
Cc: Peter Senna <peter.senna@gmail.com>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Cc: Gilles Muller <Gilles.Muller@lip6.fr>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Sun, 8 Dec 2013 18:21:12 +0000 (19:21 +0100)]
backports: convert 11-dev-pm-ops patch series to SmPL
The collateral evolution on pci power management ops can easily
be expressed with SmPL, use that instead.
spatch version 1.0.0-rc19 with Python support and with PCRE support
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py \
--verbose \
--clean \
--refresh \
/home/mcgrof/linux-next/ \
/home/mcgrof/build/next-
20131206
real 8m6.077s
user 7m46.076s
sys 0m18.124s
There are three exceptions that this patch does not address:
libertas: this should be made a separate collateral evolution
specifically for spi drivers
iwlegacy: we'll deal with this as an evolution to our spatch,
it deals with having to rely on grammar rules which
go beyond just one file.
mwifiex: coccinelle is not doing the transormation due to
an isomorphism on how pm suspend ops are declared,
although we may be able to deal with this within
Coccinelle to teach it this isomorphism we can
also address this by streamlining how pm ops are
defined upstream in the meantime.
Cc: Peter Senna <peter.senna@gmail.com>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Cc: Gilles Muller <Gilles.Muller@lip6.fr>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Sun, 8 Dec 2013 18:10:48 +0000 (19:10 +0100)]
backports: convert the 62-usb_driver_lpm patch series to SmPL
This is easily expressed instead as an SmPL patch for
interpretation with Coccinelle. This replaces 336 lines
patch code with 9 lines of SmPL.
Cc: Peter Senna <peter.senna@gmail.com>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Cc: Gilles Muller <Gilles.Muller@lip6.fr>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Fri, 6 Dec 2013 01:34:12 +0000 (02:34 +0100)]
backports: add Python based backports-update-manager
This replaces the old shell scripts:
* get-compat-kernels
* get-compat-trees
with a shiny central Python based backpor update manager:
* backports-update-manager
The new version provides a few features which were
simply a difficult to implement and quite frankly
pointless to try to implement in shell. Things worth
mentioning:
* This will now monitor your old directories
and packages and remove any stale old directory
or packages. You should feel comfortable with
now just getting the latest code from the repo
and always just running the script, it will do
everything you wished it did for you.
* This will always check the URLs and ensure your
file size matches the target upstream size, we
used to only check if you had the file or not
* The compat-ksrc directory was renamed to
ksrc-backports which goes inline with our
rebranding crusade. The tool is smart enough
to figure out if you had the old name and
rename it for you. Its also smart enough to
figure out if you had your old compat-ksrc
as a symlink and preserve things just as you
intended following the old real path.
* The debs/ directory is now moved into
ksrc-backports, making it easier to manage
and detect what is or not part of backports.
* I found no way to deal with objects to extract
GNU archive items with Python, so we're stuck with
requiring GNU ar, this provides a simple wrapper
for our usage.
* For the tree updater we instead or relying on
Linus' tree to be a remote we leave it separately
but always use --reference and allow users to provide
their own --reference override for new clones.
It should be easy to extend this should others have
other custom work spaces through a config file or
the like.
* To keep backports up to date on a system after
an initial run one should only need to run through
a cronjob:
backports-update-manager --force
To only update the git trees:
backports-update-manager --force --git-trees-only
Relying on an continued update list is dumb though,
in the future we should look into putting out new
builds through without requiring GNU ar and also
updating the list through a json file just as the
kernel has its own now:
https://www.kernel.org/releases.json
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:14 +0000 (16:08 +0100)]
backports: do not call dev_hw_addr_random()
SLES 11 SP3 does not have dev_hw_addr_random() so this patch includes
the stuff which would be done in dev_hw_addr_random() in
eth_hw_addr_random().
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:13 +0000 (16:08 +0100)]
backports: fix skb_add_rx_fragi() for SLES 11 SP3
SLES 11 SP3 backported skb_add_rx_frag() with 6 arguments and we have
to use that function with the truesize parameter. This removes the
define for SLES 11 SP3 like it is done for RHEL.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:12 +0000 (16:08 +0100)]
backports: add missing LINUX_BACKPORT() on various places
This fixes lots of compile problems on RHEL 6.5.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:11 +0000 (16:08 +0100)]
backports: so not add netdev_features_t on RHEL 6.5
RHEL 6.5 already contains netdev_features_t which will collide with
backports definition.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:10 +0000 (16:08 +0100)]
backports: remove unused workqueue backports
These backports are not used by any driver code and some of them are
conflicting with code from RHEL 6.5 headers.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:09 +0000 (16:08 +0100)]
backports: add LINUX_BACKPORT infront of functions
This is needed to add support for SLES 11 SP3 support.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:08 +0000 (16:08 +0100)]
backports: add missing include for linux/of.h
This fixes a compile error with some config options.
Closes bug: https://bugzilla.kernel.org/show_bug.cgi?id=66601
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Reported-by: Richard Röjfors <richard.rojfors@gmail.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 8 Dec 2013 15:08:07 +0000 (16:08 +0100)]
backports: remove DMI_EXACT_MATCH
DMI_EXACT_MATCH was only used by the Intel gpu driver which was removed.
Kernel 3.10.21 also has DMI_EXACT_MATCH which collides with backports
definition.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Fri, 6 Dec 2013 18:21:50 +0000 (19:21 +0100)]
backports: refresh patches for next-
20131206
Run time of generation of release:
real 5m28.543s
user 3m4.536s
sys 2m2.940s
spatch version 1.0.0-rc17 with Python support and with PCRE support
== ckmake-report.log ==
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.2.52 [ OK ]
19 3.3.8 [ OK ]
20 3.4.70 [ OK ]
21 3.5.7 [ OK ]
22 3.6.11 [ OK ]
23 3.7.10 [ OK ]
24 3.8.13 [ OK ]
25 3.9.11 [ OK ]
26 3.10.20 [ OK ]
27 3.11.9 [ OK ]
28 3.12.1 [ OK ]
29 3.13-rc1 [ OK ]
real 37m43.297s
user 1030m36.392s
sys 138m31.368s
Cc: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Fri, 6 Dec 2013 17:08:33 +0000 (18:08 +0100)]
backports: backport getrawmonotonic() with do_posix_clock_monotonic_gettime()
We cannot properly backport getrawmonotonic() as it requires some
in-kernel changes which cannot be backported so the best thing we
can do is provide at least the monotonic time. This will only be
done as a compromise for kernels that didn't have getrawmonotonic()
which are kernels < 2.6.28, which hopefully we can just deprecate
completely soon. The example users which pushes us to backport this
is change via linux-next
09d8e315d for ath9k, titled:
"ath9k_hw: fix TSF save/restore around chip reset"
The original commit that added getrawmonotonic() was
2d42244ae.
mcgrof@cerro ~/linux-next (git::master)$ git describe --contains
2d42244ae
v2.6.28-rc1~82^2~1
commit
2d42244ae71d6c7b0884b5664cf2eda30fb2ae68
Author: John Stultz <johnstul@us.ibm.com>
Date: Wed Aug 20 16:37:30 2008 -0700
clocksource: introduce CLOCK_MONOTONIC_RAW
In talking with Josip Loncaric, and his work on clock synchronization (see
btime.sf.net), he mentioned that for really close synchronization, it is
useful to have access to "hardware time", that is a notion of time that is
not in any way adjusted by the clock slewing done to keep close time sync.
Part of the issue is if we are using the kernel's ntp adjusted
representation of time in order to measure how we should correct time, we
can run into what Paul McKenney aptly described as "Painting a road using
the lines we're painting as the guide".
I had been thinking of a similar problem, and was trying to come up with a
way to give users access to a purely hardware based time representation
that avoided users having to know the underlying frequency and mask values
needed to deal with the wide variety of possible underlying hardware
counters.
My solution is to introduce CLOCK_MONOTONIC_RAW. This exposes a
nanosecond based time value, that increments starting at bootup and has no
frequency adjustments made to it what so ever.
The time is accessed from userspace via the posix_clock_gettime() syscall,
passing CLOCK_MONOTONIC_RAW as the clock_id.
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Felix Fietkau <nbd@openwrt.org>
Cc: John Stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Fri, 6 Dec 2013 16:26:13 +0000 (17:26 +0100)]
backports: backport ktime_to_ms()
This backports ktime_to_ms()
mcgrof@cerro ~/linux-next (git::master)$ git describe --contains
f56916b9
v2.6.35-rc1~490^2~16
commit
f56916b97fe2031761ca611f0a342efd913afb33
Author: Chuck Lever <chuck.lever@oracle.com>
Date: Fri May 7 13:34:37 2010 -0400
ktime: introduce ktime_to_ms()
To report ktime statistics to user space in milliseconds, a new helper
is required.
When considering how to do this conversion, I didn't immediately see
why the extra step of converting ktime to a timeval was needed. To
make that more clear, introduce a couple of large comments.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Emmanuel Grumbach [Tue, 3 Dec 2013 14:10:52 +0000 (15:10 +0100)]
backports: add support for prandom_bytes
This was added in 3.8 and not used until now.
Since it it being used by the wireless stack, backport it.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
[mcgrof: fixed renamed prandom_bytes_state to prandom_bytes on symbol]
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Tue, 3 Dec 2013 10:02:41 +0000 (11:02 +0100)]
ckmake: sort kernel releases properly
Ever since 3.10, kernel releases weren't sorted properly.
Put them into a parsed list first and then sort them to
get the 3.10/11/12 etc. to the bottom of the list.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sat, 30 Nov 2013 19:54:43 +0000 (20:54 +0100)]
backports: refresh patches on next-
20131129
== ckmake-report.log ==
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.10.20 [ OK ]
19 3.11.9 [ OK ]
20 3.12.1 [ OK ]
21 3.13-rc1 [ OK ]
22 3.2.52 [ OK ]
23 3.3.8 [ OK ]
24 3.4.70 [ OK ]
25 3.5.7 [ OK ]
26 3.6.11 [ OK ]
27 3.7.10 [ OK ]
28 3.8.13 [ OK ]
29 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sat, 30 Nov 2013 19:54:42 +0000 (20:54 +0100)]
backports: add backport_ in front of sign_extend32()
sign_extend32() is also backported by the RHEL kernel, this fixes a
compile problem.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sat, 30 Nov 2013 19:54:41 +0000 (20:54 +0100)]
backports: remove duplicate allyesconfig from help
allyesconfig was mentioned in the help dialog twice.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sat, 30 Nov 2013 19:54:40 +0000 (20:54 +0100)]
backports: add backport_ prefix in front of BQL functions
Sometimes these functions are backported in the kernel, add the
backports_ prefix to prevent the names from colliding.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sat, 30 Nov 2013 19:54:39 +0000 (20:54 +0100)]
backports: check for define in module_driver and not kernel version
In some places backports checked for the kernel version on weather to
add a own define for module_driver or not. This patch makes it check if
there is already one or not.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Sun, 24 Nov 2013 19:06:01 +0000 (20:06 +0100)]
backports: remove own version of include/linux/unaligned/*.h
These files are now copied from the kernel like normal files we take
from the kernel.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 19:03:18 +0000 (20:03 +0100)]
backports: remove own version of include/linux/wireless.h
This file is now copied from the kernel like a normal file we take from
the kernel.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 19:01:59 +0000 (20:01 +0100)]
backports: remove backport/backport-include/linux/vgaarb.h
This file was never included.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 19:01:24 +0000 (20:01 +0100)]
backports: remove usb_unlink_anchored_urbs()
kref_get_unless_zero() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:49:38 +0000 (19:49 +0100)]
backports: remove kref_get_unless_zero()
kref_get_unless_zero() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:48:06 +0000 (19:48 +0100)]
backports: remove hex_byte_pack()
hex_byte_pack() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:44:20 +0000 (19:44 +0100)]
backports: remove vlan_hw_offload_capable()
vlan_hw_offload_capable() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:40:34 +0000 (19:40 +0100)]
backports: remove fb_enable_skip_vt_switch()
fb_enable_skip_vt_switch() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:39:25 +0000 (19:39 +0100)]
backports: remove ethtool_rxfh_indir_default()
ethtool_rxfh_indir_default() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:37:40 +0000 (19:37 +0100)]
backports: remove efi
Non of these defines or functions where used by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:26:48 +0000 (19:26 +0100)]
backports: remove irq_set_affinity_hint()
irq_set_affinity_hint() was never called by any code. This makes it
possible to remove struct cpumask.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:23:35 +0000 (19:23 +0100)]
backports: remove atomic_inc_not_zero_hint()
atomic_inc_not_zero_hint() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:16:45 +0000 (19:16 +0100)]
backports: remove acpi_video_register_with_quirks()
acpi_video_register_with_quirks() was never called by any code.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:12:43 +0000 (19:12 +0100)]
backports: remove eeprom_93cx6.h header
The driver is not included in backports and we should use the in kernel
headers.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 18:12:11 +0000 (19:12 +0100)]
backports: remove mdio driver code
There is some code for the mdio driver in backports, but it does not
get build on any kernel, just remove it till someone wants to make it
work.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 23 Nov 2013 12:41:14 +0000 (13:41 +0100)]
backports: refresh on next-
20131122
== ckmake-report.log ==
1 2.6.24 [ FAIL ]
2 2.6.25 [ OK ]
3 2.6.26 [ OK ]
4 2.6.27 [ OK ]
5 2.6.28 [ OK ]
6 2.6.29 [ OK ]
7 2.6.30 [ OK ]
8 2.6.31 [ OK ]
9 2.6.32 [ OK ]
10 2.6.33 [ OK ]
11 2.6.34 [ OK ]
12 2.6.35 [ OK ]
13 2.6.36 [ OK ]
14 2.6.37 [ OK ]
15 2.6.38 [ OK ]
16 2.6.39 [ OK ]
17 3.0.101 [ OK ]
18 3.1.10 [ OK ]
19 3.10.20 [ OK ]
20 3.11.9 [ OK ]
21 3.12.1 [ OK ]
22 3.13-rc1 [ OK ]
23 3.2.52 [ OK ]
24 3.3.8 [ OK ]
25 3.4.70 [ OK ]
26 3.5.7 [ OK ]
27 3.6.11 [ OK ]
28 3.7.10 [ OK ]
29 3.8.13 [ OK ]
30 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 24 Nov 2013 00:01:04 +0000 (01:01 +0100)]
backports: add current_user_ns()
current_user_ns() is used by seq_user_ns() if CONFIG_USER_NS is set and
kernel version <= 2.6.38.
This was added to the kernel in version 2.6.29.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 23 Nov 2013 12:40:24 +0000 (13:40 +0100)]
backports: update test kernels
Add kernel 3.13-rc1 and update the other kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Fri, 22 Nov 2013 19:21:42 +0000 (20:21 +0100)]
backports: fix genetlink backport
On kernels that already have pre_doit/post_doit, we don't
fake them by re-assigning all ops and their doit, but then
we clearly need to copy the pointers. Fix that oversight.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 21 Nov 2013 16:53:08 +0000 (17:53 +0100)]
backports: refresh on next-
20131121
== ckmake-report.log ==
1 2.6.24 [ FAIL ]
2 2.6.25 [ OK ]
3 2.6.26 [ OK ]
4 2.6.27 [ OK ]
5 2.6.28 [ OK ]
6 2.6.29 [ OK ]
7 2.6.30 [ OK ]
8 2.6.31 [ OK ]
9 2.6.32 [ OK ]
10 2.6.33 [ OK ]
11 2.6.34 [ OK ]
12 2.6.35 [ OK ]
13 2.6.36 [ OK ]
14 2.6.37 [ OK ]
15 2.6.38 [ OK ]
16 2.6.39 [ OK ]
17 3.0.101 [ OK ]
18 3.1.10 [ OK ]
19 3.10.17 [ OK ]
20 3.11.6 [ OK ]
21 3.12-rc7 [ OK ]
22 3.2.52 [ OK ]
23 3.3.8 [ OK ]
24 3.4.67 [ OK ]
25 3.5.7 [ OK ]
26 3.6.11 [ OK ]
27 3.7.10 [ OK ]
28 3.8.13 [ OK ]
29 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 21 Nov 2013 21:41:29 +0000 (22:41 +0100)]
backports: activate REGULATOR_ARIZONA only on >= 3.12
ARIZONA needs struct regulator_linear_range which was added in kernel 3.12.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 21 Nov 2013 20:52:46 +0000 (21:52 +0100)]
backports: remove swiotlb_nr_tbl()
This is not used anywhere.
The Debian Wheezy kernel also includes this function and it conflicted
with backports version.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Thu, 21 Nov 2013 19:51:23 +0000 (20:51 +0100)]
backports: support new generic netlink APIs
I recently changed the generic netlink registration and multicast
group APIs to be safer. Backport these changes by implementing all
the new APIs in terms of what was in the kernel before.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Thu, 21 Nov 2013 16:24:50 +0000 (17:24 +0100)]
backports: replace netdev ops patch with spatch
There's no need to carry a small netdev ops patch for
each driver any more now, we can just write a small
semantic patch. Only the ndo_ioctl() part is easier
to keep as a separate patch.
As some stupid drivers seem to set their ops to NULL
just before freeing the netdev anyway, the attach
function must gracefully handle that (but not doesn't
really need to do anything.)
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Thu, 21 Nov 2013 16:11:22 +0000 (17:11 +0100)]
backports: move part of multicast handling into an spatch
The netdev_for_each_mc_addr() handling can be done with a
semantic patch and a simple macro encapsulating the addr
dereferencing for the two cases.
In fact, this already fixed a bug in dvb_net.c, although
I think that depends on a newer kernel anyway.
Thanks Julia for the help with iterator name!
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Jason Mobarak [Tue, 19 Nov 2013 23:19:38 +0000 (15:19 -0800)]
backports: add defconfig for wcn36xx
Signed-off-by: Jason Mobarak <jam@cozybit.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Thu, 21 Nov 2013 14:51:57 +0000 (15:51 +0100)]
backports: replace linux/bitops.h header
Instead of copying bitops.h from upstream just add what's missing in
older kernel.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Tue, 12 Nov 2013 22:47:15 +0000 (23:47 +0100)]
backports: refresh on next-
20131113
== ckmake-report.log ==
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.10.17 [ OK ]
19 3.11.6 [ OK ]
20 3.12-rc7 [ OK ]
21 3.2.52 [ OK ]
22 3.3.8 [ OK ]
23 3.4.67 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Tue, 12 Nov 2013 22:42:47 +0000 (23:42 +0100)]
backports: add replace_fops()
This is needed by drivers/media/dvb-core/dvbdev.c and was added in:
commit
e84f9e57b90ca89664d733a7cef19aa7ccd832f3
Author: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun Sep 22 14:17:15 2013 -0400
consolidate the reassignments of ->f_op in ->open() instances
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 17:06:39 +0000 (18:06 +0100)]
backports: refresh on next-
20131107
== ckmake-report.log ==
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.10.17 [ OK ]
19 3.11.6 [ OK ]
20 3.12-rc7 [ OK ]
21 3.2.52 [ OK ]
22 3.3.8 [ OK ]
23 3.4.67 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 23:37:45 +0000 (00:37 +0100)]
backports: remove soc_camera_power_{on,off}()
backports already exports these functions in
drivers/media/platform/soc_camera/soc_camera.c
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 21:19:37 +0000 (22:19 +0100)]
backports: remove usage of ipv6_stub
We can not easily backport ipv6_stub which provides some IPv6 function
callbacks. These patches are removing usage of ipv6_stub which was
introduces in this commit:
commit
5f81bd2e5d804ca93f3ec8873451b22d2f454721
Author: Cong Wang <amwang@redhat.com>
Date: Sat Aug 31 13:44:30 2013 +0800
ipv6: export a stub for IPv6 symbols used by vxlan
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 18:40:43 +0000 (19:40 +0100)]
backports: fix kfifo_put()
kfifo_put() now gets the second parameter by value and not by reference
any more. This replaces the in kernel implementation of kfifo_put()
with a version which is compatible to the in kernel structures but
takes the value and not a reference to the value. We wanted to do this:
#define kfifo_put(fifo, val) kfifo_put(fifo, &val)
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 18:34:10 +0000 (19:34 +0100)]
backports: update kfifo
This brings kfifo to the state of next-
20131106.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 17:22:40 +0000 (18:22 +0100)]
backports: fix signature of __vlan_find_dev_deep()
__vlan_find_dev_deep() got an extra parameter in recent kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 17:17:06 +0000 (18:17 +0100)]
backports: add ipv6_addr_is_solict_mult()
This is needed by drivers/net/usb/cdc_mbim.c
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 17:13:58 +0000 (18:13 +0100)]
backports: add of_property_read_u8_array()
This is needed by drivers/bluetooth/btmrvl_main.c
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 23:26:18 +0000 (00:26 +0100)]
backports: always activate the of.h backports
This removes the KERNEL_HAS_OF_SUPPORT option from backports.
We can not include linux/of.h on kernel < 2.6.34, because this depends
on some sparc and powerpc only headers. The backport code should still
be included in this header file for such kernel versions, because some
recent drivers are using these functions now on all platforms, but they
will use the empty implementations then.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 16:53:50 +0000 (17:53 +0100)]
backports: do not build tty backport if it is not in the kernel
The backported parts of tty are depending on tty being build into the
kernel, deactivate them if tty was not activated in the kernel config.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 6 Nov 2013 13:21:14 +0000 (14:21 +0100)]
backports: only add regulator code when regulator is activated in kernel
The backported parts of the regulator subsystem are depending on the
regulator subsystem being build into the kernel, deactivate them if the
regulator subsystem was not activated in the kernel config.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 31 Oct 2013 20:21:58 +0000 (21:21 +0100)]
backports: refresh on next-
20131031
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.101 [ OK ]
17 3.1.10 [ OK ]
18 3.10.17 [ OK ]
19 3.11.6 [ OK ]
20 3.12-rc7 [ OK ]
21 3.2.52 [ OK ]
22 3.3.8 [ OK ]
23 3.4.67 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 31 Oct 2013 20:55:22 +0000 (21:55 +0100)]
backports: add reinit_completion()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 31 Oct 2013 20:17:50 +0000 (21:17 +0100)]
backports: do not add debugfs_create_x64() on RHEL 6.4
debugfs_create_x64() is provided by the RHEL 6.4 kernel and should not
be added to the build on such a kernel.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Wed, 30 Oct 2013 21:13:49 +0000 (22:13 +0100)]
backports: replace netlink portid by semantic patch
Introduce the infrastructure to apply semantic patches
and make use of it for the (frequently breaking) portid
patch for netlink.
Unfortunately this is significantly slower:
before:
real 0m14.312s
user 0m12.144s
sys 0m1.784s
after:
real 0m25.932s
user 0m21.768s
sys 0m2.932s
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 28 Oct 2013 22:04:08 +0000 (23:04 +0100)]
backports: update test kernel versions
This brings the kernels test test against to the latest versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 28 Oct 2013 21:31:52 +0000 (22:31 +0100)]
gentree.py: add v as prefix for stable folder name
The folder name did not started with an v.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Sujith Manoharan [Mon, 28 Oct 2013 03:55:54 +0000 (09:25 +0530)]
backports: Add a defconfig for ath9k debugging
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 28 Oct 2013 20:23:52 +0000 (21:23 +0100)]
genlib: add missing slash to korg_path for stable
Without this slash it wants to use the directory /backportsstable/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 27 Oct 2013 14:24:51 +0000 (15:24 +0100)]
backports: refresh on next-
20131025
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.98 [ OK ]
17 3.1.10 [ OK ]
18 3.10.14 [ OK ]
19 3.11.3 [ OK ]
20 3.12-rc3 [ OK ]
21 3.2.51 [ OK ]
22 3.3.8 [ OK ]
23 3.4.64 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 27 Oct 2013 14:20:36 +0000 (15:20 +0100)]
backports: add missing header for struct in6_addr
fw-api-d3.h in the iwlwifi driver uses struct in6_addr and somehow it
does not get included on all kernel versions like 2.6.25. This patch
adds the missing include for linux/in6.h to fix the problem.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 27 Oct 2013 14:12:08 +0000 (15:12 +0100)]
backports: add strtobool()
This was added in the following upstream commit:
commit
d0f1fed29e6e73d9d17f4c91a5896a4ce3938d45
Author: Jonathan Cameron <jic23@cam.ac.uk>
Date: Tue Apr 19 12:43:45 2011 +0100
Add a strtobool function matching semantics of existing in kernel equivalents
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 27 Oct 2013 13:58:09 +0000 (14:58 +0100)]
backports: add list_first_entry_or_null
This was added in the following upstream commit:
commit
6d7581e62f8be462440d7b22c6361f7c9fa4902b
Author: Jiri Pirko <jiri@resnulli.us>
Date: Wed May 29 05:02:56 2013 +0000
list: introduce list_first_entry_or_null
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 27 Oct 2013 13:35:31 +0000 (14:35 +0100)]
backports: add some new options to defconfigs/wifi
This adds ath10k, wcn36xx and rt3573 to the default wifi config.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 26 Oct 2013 10:19:17 +0000 (12:19 +0200)]
backports: refresh on next-
20131018
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.98 [ OK ]
17 3.1.10 [ OK ]
18 3.10.14 [ OK ]
19 3.11.3 [ OK ]
20 3.12-rc3 [ OK ]
21 3.2.51 [ OK ]
22 3.3.8 [ OK ]
23 3.4.64 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 26 Oct 2013 10:42:13 +0000 (12:42 +0200)]
backports: backport new sysfs API for bcma and ssb
bcma and ssb are using the new sysfs API starting with kernel 3.12.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 26 Oct 2013 12:40:59 +0000 (14:40 +0200)]
backports: use sysfs api backport independent from kernel version
In struct class the new sysfs dev_groups member was added with kernel
3.11, but in other structs, like struct bus_type the new member was
added with kernel 3.12. This backport is now needed on different kernel
versions depending on the struct where is was used.
With this patch ATTRIBUTE_GROUPS() will always be used with the new API
and ATTRIBUTE_GROUPS_BACKPORT() will always be used used with the old
API and the patch has to add a KERNEL_VERSION check to add the correct
macro.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 26 Oct 2013 12:43:49 +0000 (14:43 +0200)]
backports: add REGULATOR_LINEAR_RANGE
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Fri, 25 Oct 2013 22:08:18 +0000 (00:08 +0200)]
backports: refresh patches on next-
20131011
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.98 [ OK ]
17 3.1.10 [ OK ]
18 3.10.14 [ OK ]
19 3.11.3 [ OK ]
20 3.12-rc3 [ OK ]
21 3.2.51 [ OK ]
22 3.3.8 [ OK ]
23 3.4.64 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Fri, 25 Oct 2013 22:40:32 +0000 (00:40 +0200)]
backports: remove struct platform_device_id from wcn36xx
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Fri, 25 Oct 2013 22:40:13 +0000 (00:40 +0200)]
backports: add resource_size()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Fri, 25 Oct 2013 22:26:47 +0000 (00:26 +0200)]
backports: add VLAN_PRIO_MASK and VLAN_PRIO_SHIFT
It should be safe to add them for older kernel versions, there these
values were hard coded in other places.
This was added in linux-2.6.33.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 24 Oct 2013 21:22:12 +0000 (23:22 +0200)]
backports: refresh patches on next-
20131004
Manual fixes where needed in:
patches/collateral-evolutions/network/09-threaded-irq/drivers_net_wireless_ti_wlcore_main.patch
patches/collateral-evolutions/network/40-netdev-hw-features/drivers_net_ethernet_atheros_atl1e_atl1e_main.patch
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.98 [ OK ]
17 3.1.10 [ OK ]
18 3.10.14 [ OK ]
19 3.11.3 [ OK ]
20 3.12-rc3 [ OK ]
21 3.2.51 [ OK ]
22 3.3.8 [ OK ]
23 3.4.64 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 24 Oct 2013 22:15:12 +0000 (00:15 +0200)]
backports: add NETIF_F_RXFCS and NETIF_F_RXALL
When these are set to 0 on kernels where they are not supported it
should result in them not being used at all.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 24 Oct 2013 21:50:42 +0000 (23:50 +0200)]
backports: fix usb-sg for usbnet
This adapts the changes from:
commit
60e453a940ac678565b6641d65f8c18541bb9f28
Author: Ming Lei <ming.lei@canonical.com>
Date: Mon Sep 23 20:59:35 2013 +0800
USBNET: fix handling padding packet
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 24 Oct 2013 21:36:30 +0000 (23:36 +0200)]
backports: add dma_set_mask_and_coherent()
dma_set_mask_and_coherent() is used by some Broadcom drivers.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de
Hauke Mehrtens [Thu, 24 Oct 2013 19:58:39 +0000 (21:58 +0200)]
backports: refresh on next-
20130927
== ckmake-report.log ==
1 2.6.25 [ OK ]
2 2.6.26 [ OK ]
3 2.6.27 [ OK ]
4 2.6.28 [ OK ]
5 2.6.29 [ OK ]
6 2.6.30 [ OK ]
7 2.6.31 [ OK ]
8 2.6.32 [ OK ]
9 2.6.33 [ OK ]
10 2.6.34 [ OK ]
11 2.6.35 [ OK ]
12 2.6.36 [ OK ]
13 2.6.37 [ OK ]
14 2.6.38 [ OK ]
15 2.6.39 [ OK ]
16 3.0.98 [ OK ]
17 3.1.10 [ OK ]
18 3.10.14 [ OK ]
19 3.11.3 [ OK ]
20 3.12-rc3 [ OK ]
21 3.2.51 [ OK ]
22 3.3.8 [ OK ]
23 3.4.64 [ OK ]
24 3.5.7 [ OK ]
25 3.6.11 [ OK ]
26 3.7.10 [ OK ]
27 3.8.13 [ OK ]
28 3.9.11 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
projects / openwrt / staging / blogic.git / log