Hauke Mehrtens [Thu, 10 Jul 2014 20:34:09 +0000 (22:34 +0200)]
backports: add cc2520 driver
This adds the cc2520.h and disables the driver on kernel < 3.5.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Thu, 10 Jul 2014 20:33:45 +0000 (22:33 +0200)]
backports: add U16_MAX
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:48:47 +0000 (17:48 +0200)]
backports: remove parameter from dma_buf_export()
dma_buf_export() got a new parameter which is not available in older
kernel versions.
This was once removed in this commit and is needed again:
commit
9981d1abf4b52e84a432fc9bdfbdd30d1838819d
Author: Hauke Mehrtens <hauke@hauke-m.de>
Date: Mon Jun 16 21:02:25 2014 +0200
backports: Revert "remove parameter from dma_buf_export()"
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Wed, 23 Jul 2014 08:38:02 +0000 (10:38 +0200)]
backports: bump minimum kernel version for ADV7604 to 3.13
While doing a make allmodconfig I noticed the following:
CC [M] /dev/shm/backports-next/drivers/media/i2c/adv7604.o
/dev/shm/backports-next/drivers/media/i2c/adv7604.c:31:33: fatal error: linux/gpio/consumer.h: No such file or directory
linux/gpio/consumer.h cannot be backported currently, as functions rely
on in-kernel functionality (GPIO_DEVRES). Bumping the minimum required
kernel version for ADV7604 to 3.13 where the header got introduced.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
[add comment with reason]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Luis R. Rodriguez [Thu, 17 Jul 2014 21:27:16 +0000 (14:27 -0700)]
Revert "backports: include gpio consumer.h"
This is provides in-kernel functionality, to include
it assuems we have provided a backport for this functionality
and that's not true.
This reverts commit
16472f0c6924f0fa54c68a024a55d89689b413a0.
Stefan Assmann [Thu, 10 Jul 2014 13:31:50 +0000 (15:31 +0200)]
backports: include gpio consumer.h
While doing a make allmodconfig I noticed the following:
CC [M] /dev/shm/backports-next/drivers/media/i2c/adv7604.o
/dev/shm/backports-next/drivers/media/i2c/adv7604.c:31:33: fatal error: linux/gpio/consumer.h: No such file or directory
Including the header fixes the error.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Thu, 17 Jul 2014 19:23:05 +0000 (19:23 +0000)]
backports: make regulator palmas driver depend on v3.15
The regulator palmas driver makes use of the new struct regulator_desc
enable_val and disable_val added via commit
ca5d1b352 on v3.15 to help
overcome an issue on assumptions on how regulator drivers enable /
disable the regulator with only one value / mask. We can't port this
so just require at least v3.15 for it.
mcgrof@ergon ~/linux (git::master)$ git describe --contains
ca5d1b352
v3.15-rc1~148^2~4^3
commit
ca5d1b3524b4d90a2e2f1f71583c1dca6b96fd92
Author: Carlo Caione <carlo@caione.org>
Date: Wed Mar 5 22:11:29 2014 +0100
regulator: helpers: Modify helpers enabling multi-bit control
This patch extends the regulator helpers to account for device that use
multiple bits for control when using regmap enable/disable/bypass ops.
The actual regulator helpers wrongly assume that the regulator control
is always performed using single bits, using in the regulator_desc
struct only two parameters *_reg and *_mask defining register and mask
for control.
This patch extends this struct and introduces the helpers to take into
account devices where control is performed using multiple bits and
specific multi-bit values are used for enabling/disabling/bypassing the
regulator.
Signed-off-by: Carlo Caione <carlo@caione.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Hauke Mehrtens [Mon, 16 Jun 2014 19:15:10 +0000 (21:15 +0200)]
backports: update test kernel versions
This adds kernel 3.16 and updates some other kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 21:51:21 +0000 (23:51 +0200)]
backports: refresh patches on next-
20140616
== ckmake-report.log ==
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.60 [ OK ]
4 3.3.8 [ OK ]
5 3.4.93 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.43 [ OK ]
12 3.11.10 [ OK ]
13 3.12.22 [ OK ]
14 3.13.11 [ OK ]
15 3.14.7 [ OK ]
16 3.15.0 [ OK ]
17 3.16-rc1 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 16 Jun 2014 20:51:59 +0000 (22:51 +0200)]
backports: do not include net_ieee802154_lowpan in kernel >= 3.16
In kernel >= 3.16 this backport is not needed at all. When
CPTCFG_IEEE802154_6LOWPAN is not set it should not get build, we do not
have to check if we are kernel > 3.5, because in an other case
ieee802.15.4 will not get build. Kernel version 3.15 needs a different
backport than the older versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 16 Jun 2014 19:02:25 +0000 (21:02 +0200)]
backports: Revert "remove parameter from dma_buf_export()"
The linux commit which made backports need this patch was removed from
linux-next.
This reverts commit
6e1b080e83f1537d8d198917180d0a8bdfd7e0c9.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 22:17:15 +0000 (00:17 +0200)]
backports: add __pskb_copy_fclone()
__pskb_copy() was renamed to __pskb_copy_fclone() in most places and
got an extra parameter, it should be safe to just ignore this extra
parameter.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:52:19 +0000 (17:52 +0200)]
backports: refresh patches on next-
20140606
There were manual changes ineeded in the following files:
patches/collateral-evolutions/network/0034-ndo_set_vf_spoofchk/igb_ndo_set_vf_spoofchk.patch
patches/collateral-evolutions/network/0037-ethtool_get_module/igb_get_module.patch
patches/collateral-evolutions/network/0038-ethtool_rxfh_indir/igb_rxfh_indir.patch
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.59 [ OK ]
4 3.3.8 [ OK ]
5 3.4.91 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.40 [ OK ]
12 3.11.10 [ OK ]
13 3.12.20 [ OK ]
14 3.13.11 [ OK ]
15 3.14.4 [ OK ]
16 3.15-rc7 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 16:57:29 +0000 (18:57 +0200)]
backports: remove useless git patch header
When exporting a patch from git it adds a header to it, but this header
is not needed by backports, remove it from the patches.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 14 Jun 2014 21:23:17 +0000 (23:23 +0200)]
backports: activate NFC_TRF7970A on kernel 3.4
The function devm_gpio_request_one() was backported some time ago.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 20:53:36 +0000 (22:53 +0200)]
backports: make some regulator drivers depend on more recent kernel versions
These drivers got new and not so easy backportable dependencies.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 20:53:36 +0000 (22:53 +0200)]
backports: add hashtable.h
This header file is used by net/mac802154/llsec.h and does not provide
any function prototype so it is save to copy it.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 17:08:25 +0000 (19:08 +0200)]
backports: remove const from pci error handlers on kernel < 3.7
This fixes a warning on these older kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 20:53:06 +0000 (22:53 +0200)]
backports: do not access user_ns on struct net
The ieee802154 subsystem wants to access user_ns on struct net, but
this member was added in kernel 3.8, check the general capabilities
instead.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:49:43 +0000 (17:49 +0200)]
backports: handle rename of ndo_set_vf_tx_rate to ndo_set_vf_rate
This member of struct net_device_ops was renamed and the function
signature was also changed a bit.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:49:14 +0000 (17:49 +0200)]
backports: remove usage of get_ts_info in cdc_ncm.c
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:46:55 +0000 (17:46 +0200)]
backports: copy include/linux/platform_data/st21nfca.h
This is needed by the st21nfca nfc driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 20:52:39 +0000 (22:52 +0200)]
backports: add split_page()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 14 Jun 2014 21:23:00 +0000 (23:23 +0200)]
backports: add of_match_ptr()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 20:52:18 +0000 (22:52 +0200)]
backports: add of_node_put()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:48:47 +0000 (17:48 +0200)]
backports: remove parameter from dma_buf_export()
dma_buf_export() got a new parameter which is not available in older
kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:47:50 +0000 (17:47 +0200)]
backports: add crc7_be()
This was added in:
commit
1836eea209546b870dd83f3f4ef234d6598a560d
Author: George Spelvin <linux@horizon.com>
Date: Sat May 10 10:32:57 2014 -0400
lib/crc7: Shift crc7() output left 1 bit
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:45:44 +0000 (17:45 +0200)]
backports: add prandom_u32_max()
This was added in:
commit
f337db64af059c9a94278a8b0ab97d87259ff62f
Author: Daniel Borkmann <dborkman@redhat.com>
Date: Wed Jan 22 02:29:39 2014 +0100
random32: add prandom_u32_max and convert open coded users
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 15:44:47 +0000 (17:44 +0200)]
backports: add U32_MAX
This was added in:
commit
89a0714106aac7309c7dfa0f004b39e1e89d2942
Author: Alex Elder <alex.elder@linaro.org>
Date: Thu Jan 23 15:54:00 2014 -0800
kernel.h: define u8, s8, u32, etc. limits
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 14:28:20 +0000 (16:28 +0200)]
backports: remove USHORT_MAX from kernel.h
This define is not in the kernel any more.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 8 Jun 2014 14:11:27 +0000 (16:11 +0200)]
backports: remove upstream patches
These two patches are now in upstream linux kernel and not needed in
backports any more
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Emmanuel Grumbach [Tue, 13 May 2014 13:19:12 +0000 (16:19 +0300)]
backport: update drivers_net_wireless_iwlwifi_iwl-debug.patch
The driver was updated - need to update the patch accordingly
commit
7f715626a9ddfa7df13f076f59fc62e3c99066cf
Author: Liad Kaufman <liad.kaufman@intel.com>
Date: Thu May 8 15:24:54 2014 +0300
iwlwifi: use dev_printk instead of dev_dbg for debug logs
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:31:39 +0000 (14:31 +0200)]
backports: activate some more media drivers
These new drivers are compile tested only.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:57:57 +0000 (14:57 +0200)]
backports: fix a compile warning in devnode
The mode parameter type was changed in the callback function devnode in
struct device_type.
commit
2c9ede55ecec58099b72e4bb8eab719f32f72c31
Author: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat Jul 23 20:24:48 2011 -0400
switch device_get_devnode() and ->devnode() to umode_t *
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:42:31 +0000 (14:42 +0200)]
backports: add I2C_CLIENT_SCCB
This is needed for some newly activated media driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:40:30 +0000 (14:40 +0200)]
backports: add i2c_add_mux_adapter()
This adds i2c_add_mux_adapter() also for kernel < 3.5. This is needed
for some newly activated media driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:02:44 +0000 (14:02 +0200)]
backports: remove backport of DMA shared buffers
This DMA shared buffer backport was used for kernel < 3.5, more recent
kernel versions already have a working version of the DMA shared
buffers.
The DMA shared buffers are only used by some media drivers, but non of
these media drivers, which are actually using this API, are compiled on
kernel < 3.5. The DMA shared buffer part would need some more attention
in the next backports version, because there are some bigger changes
and instead we removed it.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 15 Jun 2014 12:15:05 +0000 (14:15 +0200)]
backports: disable DMA shared buffer support for kernel < 3.5 in v4l2
v4l2 uses the DMA shared buffers API of kernel >= 3.5. When the v4l2
backport is be compiled against older kernel versions a backport of the
DMA shared buffers is used. Instead of using this backport this patch
will remove usage of the DMA shared buffers on these old kernel
versions. This allows us to remove DMA shared buffers from backports.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 14:16:48 +0000 (16:16 +0200)]
backports: refresh patches on next-
20140516
== ckmake-report.log ==
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.59 [ OK ]
4 3.3.8 [ OK ]
5 3.4.91 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.40 [ OK ]
12 3.11.10 [ OK ]
13 3.12.20 [ OK ]
14 3.13.11 [ OK ]
15 3.14.4 [ OK ]
16 3.15-rc7 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 17:09:14 +0000 (19:09 +0200)]
backports: add RSI driver
This adds Redpine Signals Inc 91x WLAN driver support
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 17:50:05 +0000 (19:50 +0200)]
backports: add mmc_card_hs()
This is needed by the RSI driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 17:49:45 +0000 (19:49 +0200)]
backports: add ether_addr_copy()
This is needed by the rsi driver.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 18:32:57 +0000 (20:32 +0200)]
backports: rename irq_get_trigger_type()
This should avoid possible name conflicts.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 14:15:59 +0000 (16:15 +0200)]
backports: handle changes to vlan_rx_add/kill pointers in cdc_mbim.c
These callback pointers are now also used in cdc_mbim.c
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 7 Jun 2014 14:08:55 +0000 (16:08 +0200)]
backports: revert rename of __vlan_find_dev_deep()
This function was renamed in this commit in mainline linux:
commit
f06c7f9f92295faf701a9628b383156c4efb6119
Author: dingtianhong <dingtianhong@huawei.com>
Date: Fri May 9 14:58:05 2014 +0800
vlan: rename __vlan_find_dev_deep() to __vlan_find_dev_deep_rcu()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Fri, 9 May 2014 06:23:16 +0000 (08:23 +0200)]
backports: fix class/bus attribute group spatch conversion
The spatch conversion (commit
05d44557c6a83b4ecb935cb6306f62,
"backports: transform group attribute backport to SmPL form")
added an #ifdef that shouldn't be there - remove it.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 1 Jun 2014 21:30:01 +0000 (23:30 +0200)]
backports: set defconfig media to m
Set the default option to module.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sun, 1 Jun 2014 21:22:09 +0000 (23:22 +0200)]
backports: update default configs
Remove some config options which were not available in the kernel any
more and add some missing new options.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:29:08 +0000 (17:29 +0200)]
backports: update test kernel versions
This updates the kernel versions to test against to the most recent
versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:23:22 +0000 (17:23 +0200)]
backports: refresh patches on next-
20140501
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.59 [ OK ]
4 3.3.8 [ OK ]
5 3.4.91 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.40 [ OK ]
12 3.11.10 [ OK ]
13 3.12.20 [ OK ]
14 3.13.11 [ OK ]
15 3.14.4 [ OK ]
16 3.15-rc7 [ OK ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:21:40 +0000 (17:21 +0200)]
backports: remove unused functions from bluetooth
hidp_set_raw_report() and hidp_output_report() are not used with kernel
< 3.15.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:19:48 +0000 (17:19 +0200)]
backports: add irq_get_trigger_type()
This function is needed by some drivers.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:18:28 +0000 (17:18 +0200)]
backports: add devm_gpio_request{_one}()
These two functions are used by some drivers, but not available in all
kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 15:17:05 +0000 (17:17 +0200)]
backports: add patch to prevent section mismatch again
This patch was removed in:
commit
be4a0f9ad7e17670d7a30c9e94d5dd918425f90a
Author: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Date: Fri Apr 11 01:45:20 2014 +0000
backports: nuke support for kernels < 3.0
It is still needed for some still supported kernel versions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 14:11:24 +0000 (16:11 +0200)]
backports: move net_device_priv_flags
These flags are now defined in linux/netdevice.h and not in linux/if.h
any more. This patch also removed some defined in all kernel versions
since 3.0.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 13:52:42 +0000 (15:52 +0200)]
backports: only add net_get_random_once() when not available
net_get_random_once() was added unconditionally also when the kernel we
are compiling against already provided this function, this patch checks
that first.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 13:34:55 +0000 (15:34 +0200)]
backports: fix build if CONFIG_PTP_1588_CLOCK=m
ifdef just checked if this was build into the kernel and not if it was
build as a module.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 13:33:41 +0000 (15:33 +0200)]
backports: raise kernel minimal kernel version for some regulator drivers
Compiling with older kernel versions causes some problems.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 13:31:29 +0000 (15:31 +0200)]
backports: select_queue patch: fix callback kernel version
The patch changing the callback function signature was backported to
kernel 3.14, this was already fixed in backports for mac80211, but that
commit missed mwifiex.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Sat, 31 May 2014 13:24:18 +0000 (15:24 +0200)]
backports: add smp_mb__after_atomic()
smp_mb__after_clear_bit() was renamed to smp_mb__after_atomic() in more
recent kernel version, this patch adds compatibility for older kernel
versions.
In addition it also removes a useless include of asm/atomic.h which
would include this same file.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Jonathan DE CESCO [Mon, 26 May 2014 14:15:36 +0000 (17:15 +0300)]
backports: Avoid multiple definition of DEVICE_ATTR_RW for older kernels
This definition may exists on non-vanilla kernels.
Signed-off-by: Jonathan DE CESCO <jonathan.de.cesco@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Wed, 21 May 2014 07:44:43 +0000 (09:44 +0200)]
git-tracker: remove shortlog for empty changes
When the input tree had changes but the output doesn't get
any changes at all, then the shortlog was being generated
over all commits since no files were specified.
Catch this special case and instead of generating a shortlog
just add a note that no commits changed generated code.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:24 +0000 (13:21 +0200)]
backports: handle pci_error_handlers change to const
In kernel 3.7 struct pci_driver changed pci_error_handlers to const.
Address this by putting ifdef around the code.
commit
494530284f16298050ab99f54b7b12dd7d1418a1
Author: Stephen Hemminger <shemminger@vyatta.com>
Date: Fri Sep 7 09:33:14 2012 -0700
PCI: Make pci_error_handlers const
git describe --contains
494530284f16298050ab99f54b7b12dd7d1418a1
v3.7-rc1~177^2~9^2~3
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:23 +0000 (13:21 +0200)]
backports: handle hwmon_device_register_with_groups API calls
The new hwmon_device_register_with_groups API has not been backported yet.
Address this by putting ifdef around the code using the API.
commit
74188cba088192e14cd7fd5433876e8c947bcdd8
Author: Guenter Roeck <linux@roeck-us.net>
Date: Thu Jul 11 20:00:12 2013 -0700
hwmon: Provide managed hwmon registration
git describe --contains
74188cba088192e14cd7fd5433876e8c947bcdd8
v3.13-rc1~126^2~30
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:22 +0000 (13:21 +0200)]
backports: handle struct ethtool_cmd MDI changes
In kernel 3.7 struct ethtool_cmd received eth_tp_mdix and eth_tp_mdix_ctrl.
Address this by putting ifdef around the code using these struct components.
commit
6f6bbc186dc8e4e0c628db7decbd1a5e02cb5fd8
Author: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Thu Jul 26 02:30:53 2012 +0000
ethtool.h: MDI setting support
git describe --contains
6f6bbc186dc8e4e0c628db7decbd1a5e02cb5fd8
v3.7-rc1~145^2~300^2~5
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:21 +0000 (13:21 +0200)]
backports: handle argument type change of get_rxnfc in struct ethtool_ops
In kernel 3.2 an argument type of function pointer get_rxnfc in
struct ethtool_ops changed from void to u32.
Address this by putting ifdef around the code.
commit
815c7db5c809ea3d5735de3131ecdf758b0e14ff
Author: Ben Hutchings <bhutchings@solarflare.com>
Date: Tue Sep 6 13:49:12 2011 +0000
ethtool: Clean up definitions of rule location arrays in RX NFC
git describe --contains
815c7db5c809ea3d5735de3131ecdf758b0e14ff
v3.2-rc1~129^2~272
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:20 +0000 (13:21 +0200)]
backports: handle new rxfh_ndir pointers in struct ethtool_ops
In kernel 3.3 struct ethtool_ops received 3 new function pointers
get_rxfh_indir_size, get_rxfh_indir, set_rxfh_indir.
Address this by putting ifdef around the code.
commit
7850f63f1620512631445b901ae11cd149e7375c
Author: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu Dec 15 13:55:01 2011 +0000
ethtool: Centralise validation of ETHTOOL_{G, S}RXFHINDIR parameters
git describe --contains
7850f63f1620512631445b901ae11cd149e7375c
v3.3-rc1~182^2~192
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:19 +0000 (13:21 +0200)]
backports: handle new get_module_info and get_module_eeprom pointers in struct ethtool_ops
In kernel 3.5 struct ethtool_ops received 2 new function pointers
get_module_info and get_module_eeprom. Address this by putting ifdef around
the code.
commit
41c3cb6d20f0252308e9796fa4f3dacb4960de91
Author: Stuart Hodgson <smhodgson@solarflare.com>
Date: Thu Apr 19 09:44:42 2012 +0100
ethtool: Extend the ethtool API to obtain plugin module eeprom data
git describe --contains
41c3cb6d20f0252308e9796fa4f3dacb4960de91
v3.5-rc1~109^2~124^2~2
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:18 +0000 (13:21 +0200)]
backports: handle new get_ts_info pointer in struct ethtool_ops
In kernel 3.5 struct ethtool_ops received a new function pointer
get_ts_info. Address this by putting ifdef around the code.
commit
c8f3a8c31069137fe0100e6920558f1a7487ef3c
Author: Richard Cochran <richardcochran@gmail.com>
Date: Tue Apr 3 22:59:17 2012 +0000
ethtool: Introduce a method for getting time stamping capabilities.
git describe --contains
c8f3a8c31069137fe0100e6920558f1a7487ef3c
v3.5-rc1~109^2~632
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:17 +0000 (13:21 +0200)]
backports: handle new get_eee and set_eee pointers in struct ethtool_ops
In kernel 3.6 struct ethtool_ops received 2 new function pointers get_eee
and set_eee. Address this by putting ifdef around the code.
commit
80f12eccce775dc6bb93dba9b52529740f929237
Author: Yuval Mintz <yuvalmin@broadcom.com>
Date: Wed Jun 6 17:13:06 2012 +0000
Added kernel support in EEE Ethtool commands
git describe --contains
80f12eccce775dc6bb93dba9b52529740f929237
v3.6-rc1~125^2~587
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:16 +0000 (13:21 +0200)]
backports: handle no_fcs change in struct sk_buff
In kernel 3.4 a new field no_fcs was introduced to struct sk_buff.
Address this by putting ifdef around the code.
commit
3bdc0eba0b8b47797f4a76e377dd8360f317450f
Author: Ben Greear <greearb@candelatech.com>
Date: Sat Feb 11 15:39:30 2012 +0000
net: Add framework to allow sending packets with customized CRC.
git describe --contains
3bdc0eba0b8b47797f4a76e377dd8360f317450f
v3.4-rc1~177^2~207
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:15 +0000 (13:21 +0200)]
backports: handle new ndo_set_vf_spoofchk pointer in struct net_device_ops
In kernel 3.2 struct net_device_ops got a new function pointer
ndo_set_vf_spoofchk. Address this by putting ifdef around the code.
commit
5f8444a3fa617076f8da51a3e8ecce01a5d7f738
Author: Greg Rose <gregory.v.rose@intel.com>
Date: Sat Oct 8 03:05:24 2011 +0000
if_link: Add additional parameter to IFLA_VF_INFO for spoof checking
git describe --contains
5f8444a3fa617076f8da51a3e8ecce01a5d7f738
v3.2-rc1~129^2~87
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:14 +0000 (13:21 +0200)]
backports: handle changes to vlan_rx_add/kill pointers in struct net_device_ops
In kernel 3.3 and 3.10 struct net_device_ops function pointers
ndo_vlan_rx_add_vid and ndo_vlan_rx_kill_vid changed.
Address this by putting ifdef around the code.
commit
8e586137e6b63af1e881b328466ab5ffbe562510
Author: Jiri Pirko <jpirko@redhat.com>
Date: Thu Dec 8 19:52:37 2011 -0500
net: make vlan ndo_vlan_rx_[add/kill]_vid return error value
git describe --contains
8e586137e6b63af1e881b328466ab5ffbe562510
v3.3-rc1~182^2~291
commit
80d5c3689b886308247da295a228a54df49a44f6
Author: Patrick McHardy <kaber@trash.net>
Date: Fri Apr 19 02:04:28 2013 +0000
net: vlan: prepare for 802.1ad VLAN filtering offload
git describe --contains
80d5c3689b886308247da295a228a54df49a44f6
v3.10-rc1~66^2~97^2~4
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:13 +0000 (13:21 +0200)]
backports: handle new sriov_configure pointer in struct pci_driver
In kernel 3.8 struct pci_driver got a new function pointer sriov_configure.
Address this by putting ifdef around the code.
commit
1789382a72a537447d65ea4131d8bcc1ad85ce7b
Author: Donald Dutile <ddutile@redhat.com>
Date: Mon Nov 5 15:20:36 2012 -0500
PCI: SRIOV control and status via sysfs
git describe --contains
1789382a72a537447d65ea4131d8bcc1ad85ce7b
v3.8-rc1~128^2~6^2~4
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Stefan Assmann [Fri, 16 May 2014 11:21:12 +0000 (13:21 +0200)]
backports: remove old igb patches
Reworking the igb patches and thus removing the old ones.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Thu, 15 May 2014 09:22:06 +0000 (11:22 +0200)]
backport: CCM: set correct dependencies
If these aren't actually compiled in, then the mac80211
code will fail at runtime. Add these dependencies so we
at least get a warning at compile time about this.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Mon, 5 May 2014 12:46:11 +0000 (14:46 +0200)]
backports: provide net_ieee802154_lowpan only when needed
If 802.15.4 isn't configured, there's no need to provide the
network namespace support for it, so don't.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
[use IS_ENABLED to hanle CPTCFG_IEEE802154_6LOWPAN=m]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Mon, 5 May 2014 14:58:38 +0000 (16:58 +0200)]
gentree: remove stray print
My previous commit to restrict the defconfig files left
a debugging print statement, remove it.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Richard Röjfors [Sat, 3 May 2014 14:09:37 +0000 (16:09 +0200)]
backports: backport-3.15: Include fix
In this version the correct linux string.h is included.
Include string.h to avoid compile issues if warnings are treated as errors:
backports-3.14-1/compat/backport-3.15.c: In function 'backport_devm_kstrdup':
backports-3.14-1/compat/backport-3.15.c:33:2: error: implicit declaration of function 'strlen' [-Werror=implicit-function-declaration]
size = strlen(s) + 1;
^
backports-3.14-1/compat/backport-3.15.c:33:9: warning: incompatible implicit declaration of built-in function 'strlen' [enabled by default]
size = strlen(s) + 1;
^
backports-3.14-1/compat/backport-3.15.c:36:3: error: implicit declaration of function 'memcpy' [-Werror=implicit-function-declaration]
memcpy(buf, s, size);
^
backports-3.14-1/compat/backport-3.15.c:36:3: warning: incompatible implicit declaration of built-in function 'memcpy' [enabled by default]
Signed-off-by: Richard Röjfors <richard@puffinpack.se>
[refreshed patch on current master]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Johannes Berg [Wed, 30 Apr 2014 22:10:25 +0000 (00:10 +0200)]
backports: select_queue patch: fix callback kernel version
The callback argument has been passed since kernel 3.14, not
only 3.15 - fix that.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Arend van Spriel [Fri, 2 May 2014 01:03:37 +0000 (18:03 -0700)]
backports: lib: remove raising ExecutionError exception
In spatch a ExecutionError was raised, but that causes the thread
to finish without closing the outfile and more importantly without
putting (ret, fn) tuple on the return queue. This results in the
threaded_spatch routine to get stuck on the ret_q.get() call. This
patch removes raising the ExecutionError and just return the tuple.
The non-zero return code will result in ExecutionErrorThread exception
anyway.
Signed-off-by: Arend van Spriel <arend@broadcom.com>
[mcgrof]: rebased to fit into pycocci
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 20:48:23 +0000 (13:48 -0700)]
backports: fix igb patch for next-
20140501
Manual fixes were needed for:
patches/the-way-not-to-do-backports/0001-ethernet-igb/0001-igb_net_device_ops.patch
We currently hit a compile error as of next-
20140501 because of the
introduction of smp_mb__after_atomic(). This is not yet addressed.
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 20:45:19 +0000 (13:45 -0700)]
backports: devel/backports-update-manager: fix for older shutil.move
Older shutil.move does not seem to like it when there are
symlinks on a target, to account for this and to make things
clearer just split up the work on the lib/modules and usr/src
directory.
This fixes running devel/backports-update-manager on OpenSUSE 13.1
base install.
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 16:25:38 +0000 (09:25 -0700)]
backports: remove two upstream patches
Two changes to help with backports were merged upstream so nuke 'em.
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 16:04:58 +0000 (09:04 -0700)]
backports: add ckmake binary requirements
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 16:02:06 +0000 (09:02 -0700)]
backports: run autopep8 on ckmake
This converts rel-html.py to conform to the PEP 8 style
guide using autopep8. If using vi, consider adding this to
your .vimrc:
filetype indent plugin on
au FileType python setlocal tabstop=8 expandtab shiftwidth=4 softtabstop=4
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 15:30:34 +0000 (08:30 -0700)]
backports: add gentree.py binary requirements
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 15:11:03 +0000 (08:11 -0700)]
backports: add backports-update-manager binary requirements
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Luis R. Rodriguez [Thu, 1 May 2014 14:58:04 +0000 (07:58 -0700)]
backports: add a Python binary package dependency library
This lets us use this in our python scripts for
program dependencies that they might have, instead
of getting nasty stack traces you get output that's
a very useful and clear. This is also important as
we move along with Coccinelle and require newer
versions with new bells and whisteles added upstream
to Coccinelle.
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Avery Pennarun [Wed, 30 Apr 2014 01:29:06 +0000 (21:29 -0400)]
Makefile: add savedefconfig target.
This writes a file called 'defconfig' which contains only the deltas from
the default configuration. It's suitable for use as an input defconfig file
for later builds.
Signed-off-by: Avery Pennarun <apenwarr@gmail.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Tue, 29 Apr 2014 09:31:28 +0000 (11:31 +0200)]
gentree: strip useless defconfig files
If a defconfig file has symbols that can't be satisfied
with the given copy-list file, then there's no point in
shipping it, so just copy selectively the ones needed.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Tue, 29 Apr 2014 09:31:27 +0000 (11:31 +0200)]
backports: remove IWLWIFI_P2P from iwlwifi defconfig
This symbol is obsolete, it was removed upstream, so just
remove it from the defconfig file as well.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Mon, 28 Apr 2014 13:12:15 +0000 (15:12 +0200)]
backports: remove more old ifdef/compat code
Since support for kernels < 3.0 was removed, there's now a bunch
of dead code that can be removed - do it.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Mon, 28 Apr 2014 12:06:58 +0000 (14:06 +0200)]
backports: remove mac80211.h
As support for old kernels has been removed, there's no need to keep
around the mac80211.h file that redefines the ieee80211_rx() symbol;
In 2.6.32 this name was changed to belong to mac80211.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Stefan Assmann [Fri, 25 Apr 2014 07:14:30 +0000 (09:14 +0200)]
backports: ifdef code already present in RHEL7.0
Avoid conflicts with RHEL7.0 as DEVICE_ATTR_* defines,
enum pkt_hash_types, skb_set_hash(), dma_common_get_sgtable()
are already present.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Stefan Assmann [Fri, 25 Apr 2014 07:14:29 +0000 (09:14 +0200)]
backports: add linux/of_graph.h to copy-list
Required by CONFIG_VIDEO_ADV7343 and others.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Stefan Assmann [Fri, 25 Apr 2014 07:14:28 +0000 (09:14 +0200)]
backports: wrap dma_set_mask_and_coherent in LINUX_BACKPORT()
To avoid conflicts with any distribution kernel wrap
dma_set_mask_and_coherent with LINUX_BACKPORT().
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Johannes Berg [Thu, 24 Apr 2014 19:19:38 +0000 (21:19 +0200)]
git-tracker: restrict shortlog to existing files
When the git-tracker updates through a merge commit or
otherwise multiple kernel commits, it appends a shortlog
of all the changes. That's fine, but is often unhelpful
since it's so large - restrict it to the files that are
actually backported to make it more readable.
While at it, rewrite the bpgit.status() helper function
to actually do something useful.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Tue, 15 Apr 2014 20:07:36 +0000 (20:07 +0000)]
backports: refresh patches for next-
20140415
This update required one collateral evolution to be addressed,
the rest was just a patch refresh.
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py --clean
--refresh /home/mcgrof/linux-next /home/mcgrof/build/next-
20140415
Copy original source files ...
Apply patches ...
Modify Kconfig tree ...
Rewrite Makefiles and Kconfig files ...
Done!
real 1m16.527s
user 10m47.068s
sys 0m37.412s
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.54 [ OK ]
4 3.3.8 [ OK ]
5 3.4.79 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.29 [ OK ]
12 3.11.10 [ OK ]
13 3.12.10 [ OK ]
14 3.13.2 [ OK ]
15 3.14-rc1 [ OK ]
real 18m42.577s
user 498m48.572s
sys 64m0.560s
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Luis R. Rodriguez [Tue, 15 Apr 2014 21:03:31 +0000 (21:03 +0000)]
backports: formalize struct sock sk_data_ready() backport with SmPL
Commit
676d2369 by David removed the skb->len arguments passed onto
the struct sock sk_data_ready() callback. This was done as its racy,
a few drivers were passing 0 to it, and it was not really used.
By removing it the raciness is addresed but to backport this we are
going to have to deal with the races as-is on older kernels. This was
merged as of v3.15:
mcgrof@ergon ~/linux-next (git::master)$ git describe --contains
676d2369
v3.15-rc1~8^2~10
Since this is not a define or static inline we can't easily replace this with
the backports module or header files, instead we use SmPL grammar to generalize
the backport for all use cases. Note that in order to backport this we won't
know what older kernel drivers were using before this change, it could have
been 0 or skb->len for the length parameter, since we have to infer something
we choose skb->len *iff* skb_queue_tail() was used right before it, otherwise
we infer to throw 0.
Adding this SmPL patch to our series only incurs an additional ~9 seconds
on run time code generation.
mcgrof@drvbp1 ~/backports (git::master)$ time ./gentree.py --clean
/home/mcgrof/linux-next /home/mcgrof/build/next-
20140415-clean
Copy original source files ...
Apply patches ...
Modify Kconfig tree ...
Rewrite Makefiles and Kconfig files ...
Done!
real 1m25.128s
user 12m49.380s
sys 0m44.892s
1 3.0.101 [ OK ]
2 3.1.10 [ OK ]
3 3.2.54 [ OK ]
4 3.3.8 [ OK ]
5 3.4.79 [ OK ]
6 3.5.7 [ OK ]
7 3.6.11 [ OK ]
8 3.7.10 [ OK ]
9 3.8.13 [ OK ]
10 3.9.11 [ OK ]
11 3.10.29 [ OK ]
12 3.11.10 [ OK ]
13 3.12.10 [ OK ]
14 3.13.2 [ OK ]
15 3.14-rc1 [ OK ]
real 18m42.577s
user 498m48.572s
sys 64m0.560s
commit
676d23690fb62b5d51ba5d659935e9f7d9da9f8e
Author: David S. Miller <davem@davemloft.net>
Date: Fri Apr 11 16:15:36 2014 -0400
net: Fix use after free by removing length arg from sk_data_ready callbacks.
Several spots in the kernel perform a sequence like:
skb_queue_tail(&sk->s_receive_queue, skb);
sk->sk_data_ready(sk, skb->len);
But at the moment we place the SKB onto the socket receive queue it
can be consumed and freed up. So this skb->len access is potentially
to freed up memory.
Furthermore, the skb->len can be modified by the consumer so it is
possible that the value isn't accurate.
And finally, no actual implementation of this callback actually uses
the length argument. And since nobody actually cared about it's
value, lots of call sites pass arbitrary values in such as '0' and
even '1'.
So just remove the length argument from the callback, that way there
is no confusion whatsoever and all of these use-after-free cases get
fixed as a side effect.
Based upon a patch by Eric Dumazet and his suggestion to audit this
issue tree-wide.
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: David S. Miller <davem@davemloft.net>
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>