Commit Graph

43 Commits

Author SHA1 Message Date
Alexander Aring
c38383530f mac802154: tx: fix synced xmit deadlock
This patch reverts 6001d52 ("mac802154: tx: don't allow if down while
sync tx"). This has side effects with stop callback which flush the
transmit workqueue. The stop callback will wait until the workqueue is
flushed and holding the rtnl lock. That means it can happen that the stop
callback waits forever because it try to lock the rtnl mutex which is
already hold by stop callback.

Cc: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-12-10 19:17:11 +01:00
Alexander Aring
b40988c438 ieee802154: change mtu size behaviour
This patch changes the mtu size of 802.15.4 interfaces. The current
setting is the meaning of the maximum transport unit with mac header,
which is 127 bytes according 802.15.4. The linux meaning of the mtu size
field is the maximum payload of a mac frame. Like in ethernet, which is
1500 bytes.

We have dynamic length of mac frames in 802.15.4, this is why we assume
the minimum header length which is hard_header_len. This contains fc and
sequence fields. These can evaluated by driver layer without additional
checks. We currently don't support to set the FCS from userspace, so we
need to subtract this from mtu size as well.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-09-30 13:21:32 +02:00
Alexander Aring
d58a2fa903 mac802154: add comments for llsec issues
While doing a little test with the llsec implementation I saw these
issues. We should move decryption and encruption somewhere else,
otherwise while capturing with wireshark the mac header shows secuirty
fields but the payload is plaintext.

A complete other issue is what doing with HardMAC drivers where the
payload is always plaintext. I think we need a special handling then in
userspace. We currently doesn't support any HardMAC transceivers, so we
should fix the first issue for SoftMAC transceivers.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-09-30 13:16:44 +02:00
Alexander Aring
02c7b69228 mac802154: tx: add warning if MTU exceeds
Sending over AF_PACKET RAW sockets we can sending frames which exceeds
MTU size. To handling it correct we need to change things in AF_PACKET
which knows on RAW sockets an additional FCS is set by hardware or
mac802154 transmit functionality.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-09-22 11:51:20 +02:00
Alexander Aring
87a93e4ece ieee802154: change needed headroom/tailroom
This patch cleanups needed_headroom, needed_tailroom and hard_header_len
fields for wpan and lowpan interfaces.

For wpan interfaces the worst case mac header len should be part of
needed_headroom, currently this is set as hard_header_len, but
hard_header_len should be set to the minimum header length which xmit
call assumes and this is the minimum frame length of 802.15.4.
The hard_header_len value will check inside send callbacl of AF_PACKET
raw sockets.

For lowpan interfaces, if fragmentation isn't needed the skb will
call dev_hard_header for 802154 layer and queue it afterwards. This
happens without new skb allocation, so we need the same headroom and
tailroom lengths like 802154 inside 802154 6lowpan layer. At least we
assume as minimum header length an ipv6 header size.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-09-22 11:51:20 +02:00
Lennert Buytenhek
c22ff7b4e7 mac802154: Fix memory corruption with global deferred transmit state.
When transmitting a packet via a mac802154 driver that can sleep in
its transmit function, mac802154 defers the call to the driver's
transmit function to a per-device workqueue.

However, mac802154 uses a single global work_struct for this, which
means that if you have more than one registered mac802154 interface
in the system, and you transmit on more than one of them at the same
time, you'll very easily cause memory corruption.

This patch moves the deferred transmit processing state from global
variables to struct ieee802154_local, and this seems to fix the memory
corruption issue.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Acked-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2015-07-30 14:08:55 +02:00
Alexander Aring
61f2dcba9a mac802154: add interframe spacing time handling
This patch adds a new interframe spacing time handling into mac802154
layer. Interframe spacing time is a time period between each transmit.
This patch adds a high resolution timer into mac802154 and starts on
xmit complete with corresponding interframe spacing expire time if
ifs_handling is true. We make it variable because it depends if
interframe spacing time is handled by transceiver or mac802154. At the
timer complete function we wake the netdev queue again. This avoids
new frame transmit in range of interframe spacing time.

For synced driver we add no handling of interframe spacing time. This
is currently a lack of support in all synced xmit drivers. I suppose
it's working because the latency of workqueue which is needed to call
spi_sync.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-11-13 04:51:58 +01:00
Alexander Aring
90386a7e3b mac802154: separate omit tx/rx flags
This patch splits the IEEE802154_HW_OMIT_CKSUM hardware flag into
IEEE802154_HW_TX_OMIT_CKSUM and IEEE802154_HW_RX_OMIT_CKSUM. This is
useful to deliver the received crc from the driver layer to the monitor
interface. At the moment we can't do that without change the xmit
handling.

The received checksum should be visible in monitor mode only.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-29 23:07:45 +01:00
Alexander Aring
59cb300f2b mac802154: use driver-ops function wrappers
This patch replaces all directly called driver ops by previous
introduced driver-ops function wrappers.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-28 23:19:07 +01:00
Alexander Aring
4ca18be54f mac802154: tx: remove monitor receive while xmit
This removes the call of monitor receive funktion when any interface
type call xmit. There exist no such use case that a monitor interface
should receive the actual sending frame. One use case could be that a
wpan interface and monitor interface could be running at the same time
on one phy. Then the monitor interface receives the wpan frames also.
Furthermore we adding support for promiscous mode setting. With
promiscous mode setting we can't run a wpan and monitor interface at the
same time.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-27 18:07:42 +01:00
Alexander Aring
061ef8f915 mac802154: tx: use put_unaligned_le16 for copy crc
This patch replaces the memcpy with a put_unaligned_le16. The placement
of crc inside of PSDU can also be unaligned. With memcpy this can fail
on some architectures.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Reported-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-27 18:07:36 +01:00
Alexander Aring
f81f466ca5 mac802154: tx: make worker information static
This patch moves the worker information struct out of skb control block.
Instead control block we declare it static inside of tx.c file. We can do
that, because the worker can't be used twice at the same time. It's
protected by stop and wake netdev queue.

This patch fix an issue that the "struct ieee802154_xmit_cb" doesn't fit
into the skb control block on some kernel configuartion reported by
kbuild test robot.

It was introduced by commit fe24371d66
("mac802154: tx: remove kmalloc in xmit hotpath").

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 19:18:35 +01:00
Alexander Aring
e5e584fcc2 mac802154: tx: change naming convention
This patch changes the naming convention of the tx functions like
mac80211. Just with an 802154 instead 80211 inside the name.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:05 +01:00
Alexander Aring
409c3b0c5f mac802154: tx: move stats tx increment
This patch moves the stats increment of successful transmitted packets
in the right place when the skb was really successful transmitted.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:05 +01:00
Alexander Aring
b7eec52bcb mac802154: tx: cleanup crc calculation
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:05 +01:00
Alexander Aring
cfa626cb37 mac802154: tx: use netdev print helpers
This patch replace the pr_foo printout function to netdev_foo printout
function. Inside the xmit handling, the interface is already known.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:04 +01:00
Alexander Aring
6001d5223d mac802154: tx: don't allow if down while sync tx
This patch holds rtnl lock while sync xmit inside of workqueue.
Otherwise we could down the interface while worker xmit handling.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:04 +01:00
Alexander Aring
ed0a5dce0c mac802154: tx: add support for xmit_async callback
This patch renames the existsing xmit callback to xmit_sync and
introduces an asynchronous xmit_async function. If ieee802154_ops
doesn't provide the xmit_async callback, then we have a fallback to
the xmit_sync callback.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Cc: Alan Ott <alan@signal11.us>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:04 +01:00
Alexander Aring
cdb66beaa0 mac802154: tx: fix error handling while xmit
In case of an error we should call kfree_skb instead of consume_skb which
is called by ieee802154_xmit_complete function.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:04 +01:00
Alexander Aring
18d60a0d49 mac802154: tx: use queue helpers in xmit worker
This patch uses the queue utility helpers inside the xmit worker of
mac802154 subsystem.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:03 +01:00
Alexander Aring
dc67c6b30f mac802154: tx: remove xmit channel context switch
This patch removes the channel hopping feature before xmit. There are
several issues to provide a real channel hopping (timing requirements,
etc...).

We don't have any known kernelspace protocol which really use this
feature. And I don't know an real user of this feature.
We simply drop this feature now.

This patch removes also the hold of pib lock which isn't needed by any
real driver xmit callback implementation.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:03 +01:00
Alexander Aring
e89e45f22a mac802154: tx: squash multiple dereferencing
This patch introduce some new stack variables to avoid multiple
dereferencing inside the xmit worker function.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:24:02 +01:00
Alexander Aring
fe24371d66 mac802154: tx: remove kmalloc in xmit hotpath
This patch removes the kmalloc allocation for workqueue data. This patch
replaces the kmalloc and uses the control block of skb. The control block
has enough space and isn't use by any other layer in this case.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:23:58 +01:00
Alexander Aring
50c6fb9965 mac802154: tx: move xmit callback to tx file
This patch moves the netdev xmit callback functions into the tx.c file.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-26 17:23:50 +01:00
Alexander Aring
f773054254 mac802154: rename dev_workqueue to workqueue
Small rename to use the name workqueue than dev_workqueue. To bring the
same naming convention like wireless into 802.15.4.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:55:38 +02:00
Alexander Aring
60741361c3 mac802154: introduce hw_to_local function
This patch replace the mac802154_to_priv macro with a static inline
function named hw_to_local. This brings a similar naming convention like
mac80211 stack.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:55:38 +02:00
Alexander Aring
d98be45b36 mac802154: rename sdata slaves and slaves_mtx
This patch renamens the slaves attribute in sdata to interfaces and
slaves_mtx to iflist_mtx. This is similar like the mac80211 stack naming
convention.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:55:38 +02:00
Alexander Aring
036562f9c4 mac802154: rename mac802154_sub_if_data
Like wireless this structure should named ieee802154_sub_if_data and not
mac802154_sub_if_data. This patch renames the struct and variables to
sdata instead priv sometimes.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:55:37 +02:00
Alexander Aring
a5e1ec538f mac802154: rename mac802154_priv to ieee802154_local
This patch rename the mac802154_priv to ieee802154_local. The
mac802154_priv structure is like ieee80211_local and so we name it
ieee802154_local.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:55:37 +02:00
Alexander Aring
5ad60d3699 ieee802154: move wpan-phy.h to cfg802154.h
The wpan-phy header contains the wpan_phy struct information. Later this
header will be have similar function like cfg80211 header. The cfg80211
header contains the wiphy struct which is identically the wpan_phy
struct inside 802.15.4 subsystem.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Cc: Alan Ott <alan@signal11.us>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:39:56 +02:00
Alexander Aring
0f1556bc2b mac802154: move mac802154.h to ieee802154_i.h
This patch moves the mac802154.h internal header to ieee802154_i.h like
the wireless stack ieee80211_i.h file. This avoids confusing with the
not internal header include/net/mac802154.h header. Additional we get
the same naming conversion like mac80211 for this file.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 21:39:55 +02:00
Alexander Aring
b3020f0a35 ieee802154: mac802154: remove FSF address
This patch removes the FSF address in files which belongs to ieee802154
and mac802154.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Cc: Alan Ott <alan@signal11.us>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-10-25 08:07:30 +02:00
Varka Bhadram
f55889128a mac802154: common tx error path
This patch introduce the common error path on failure of Tx by
inserting the label 'err_tx'.

Signed-off-by: Varka Bhadram <varkab@cdac.in>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2014-08-14 08:49:25 +02:00
Varka Bhadram
4710d806fc 6lowpan: mac802154: fix coding style issues
This patch fixed the coding style issues reported by checkpatch.pl

following issues fixed:
	CHECK: Alignment should match open parenthesis
	WARNING: line over 80 characters
	CHECK: Blank lines aren't necessary before a close brace '}'
	WARNING: networking block comments don't use an empty /* line, use /* Comment...
	WARNING: Missing a blank line after declarations
	WARNING: networking block comments start with * on subsequent lines
	CHECK: braces {} should be used on all arms of this statement

Signed-off-by: Varka Bhadram <varkab@cdac.in>
Tested-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-07 20:55:22 -07:00
Alan Ott
9f7f78b479 mac802154: Keep track of the channel when changed
Two sections checked whether the current channel != the new channel
without ever setting the current channel variables.

1. net/mac802154/tx.c: Prevent set_channel() from getting called every
time a packet is sent.

2. net/mac802154/mib.c: Lock (pib_lock) accesses to current_channel and
current_page and make sure they are updated when the channel has been
changed.

Signed-off-by: Alan Ott <alan@signal11.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-08 12:09:18 -04:00
Alan Ott
b5992fe962 mac802154: Use netif flow control
Use netif_stop_queue() and netif_wake_queue() to control the flow of
packets to mac802154 devices.  Since many IEEE 802.15.4 devices have no
output buffer, and since the mac802154 xmit() function is designed to
block, netif_stop_queue() is called after each packet.

Signed-off-by: Alan Ott <alan@signal11.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-07 17:06:43 -04:00
Alan Ott
7dd43d356e mac802154: Do not try to resend failed packets
When ops->xmit() fails, drop the packet. Devices which support hardware
ack and retry (which include all devices currently supported by mainline),
will automatically retry sending the packet (in the hardware) up to 3
times, per the 802.15.4 spec.  There is no need, and it is incorrect to
try to do it in mac802154.

Signed-off-by: Alan Ott <alan@signal11.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-07 17:06:43 -04:00
Alan Ott
92a2ec72a7 mac802154: use kfree_skb() instead of dev_kfree_skb()
kfree_skb() indicates failure, which is where this is being used.

Signed-off-by: Alan Ott <alan@signal11.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-30 12:19:24 -05:00
Alan Ott
fcefbe9fcb mac802154: fix memory leaks
kfree_skb() was not getting called in the case of some failures.
This was pointed out by Eric Dumazet.

Signed-off-by: Alan Ott <alan@signal11.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-30 12:19:24 -05:00
David S. Miller
b26d344c6b Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Conflicts:
	drivers/net/caif/caif_hsi.c
	drivers/net/usb/qmi_wwan.c

The qmi_wwan merge was trivial.

The caif_hsi.c, on the other hand, was not.  It's a conflict between
1c385f1fdf ("caif-hsi: Replace platform
device with ops structure.") in the net-next tree and commit
39abbaef19 ("caif-hsi: Postpone init of
HIS until open()") in the net tree.

I did my best with that one and will ask Sjur to check it out.

Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 17:37:00 -07:00
alex.bluesman.smirnov@gmail.com
72fd5a8b75 mac802154: add monitor listener to TX datapath
Add monitor receive callback to the TX datapath to catch all the
data sent to transceivers.

Signed-off-by: Alexander Smirnov <alex.bluesman.smirnov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-26 21:06:33 -07:00
alex.bluesman.smirnov@gmail.com
8a8e28b8e2 mac802154: add missed braces
Add missed braces after 'if' operator.

Signed-off-by: Alexander Smirnov <alex.bluesman.smirnov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-25 16:35:30 -07:00
alex.bluesman.smirnov@gmail.com
5b641ebeec mac802154: TX data path
Main TX data path implementation between upper and physical layers.

Signed-off-by: Alexander Smirnov <alex.bluesman.smirnov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-16 15:16:49 -04:00