We now timeout also if the state machine does not change within the
given time. For that, the driver-specific completion-functions are
extended to return true or false depending on the timeout. This then
gets checked in the algorithm.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

Waiting for a free bus now accepts the timeout value in jiffies and does
proper checking using time_before.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

Add support for the PCA9665 I2C controller.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

According to kerneljanitors todo list all printk calls (beginning
a new line) should have an according KERN_* constant.
Those are the changes to the debug macros in the i2c subsystem
to meet this requirement. Also changing no-debug statements
to raw printks again.
Signed-off-by: NFrank Seidel <frank@f-seidel.de>
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Tested-by: NWolfram Sang <w.sang@pengutronix.de>

Properly set the adapter timeout value in jiffies, and then use that
value in the driver, rather than a hard-coded constant.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Tested-by: NTroy Kisky <troy.kisky@boundarydevices.com>
Cc: Kevin Hilman <khilman@mvista.com>

i2c_adapter.timeout is in jiffies. Fix all drivers which thought
otherwise. It didn't really matter as long as the value was only used
inside the driver, but soon i2c-core will use it too so it must have
the proper unit.

Note: for the i2c-mpc driver, this fixes a bug in polling mode.
Timeout would trigger after 1 jiffy, which is most probably not what
the author wanted.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Cc: Clifford Wolf <clifford@clifford.at>
Acked-by: NSean MacLennan <smaclennan@pikatech.com>
Cc: Stefan Roese <sr@denx.de>
Acked-by: NLennert Buytenhek <kernel@wantstofly.org>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Grant Likely <grant.likely@secretlab.ca>
Acked-by: NMark A. Greer <mgreer@mvista.com>

Setting a default timeout value on a per-algo basis doesn't make any
sense. Move the default value setting to i2c-core. Individual adapter
drivers can specify a different (non-zero) value if they wish.

Also express the timeout value in a way which results in the same
duration regarless of the value of HZ.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Acked-by: NWolfram Sang <w.sang@pengutronix.de>

According to kerneljanitors todo list all printk calls (beginning
a new line) should have an according KERN_* constant.
Those are the missing pieces here for the i2c subsystem.
Signed-off-by: NFrank Seidel <frank@f-seidel.de>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

With a postfix decrement these timeouts reach -1 rather than 0, but after the
loop it is tested whether they have become 0.

As pointed out by Jean Delvare, the msg_num should be tested before the timeout.
With the current order, you could exit with a timeout error while all the
messages were successfully transferred.
Signed-off-by: NRoel Kluin <roel.kluin@gmail.com>
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Acked-by: NEric Brower <ebrower@gmail.com>

cleanup whitespace, fix comments and remove the unused STUB_I2C.
Signed-off-by: NRoel Kluin <roel.kluin@gmail.com>
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Acked-by: NEric Brower <ebrower@gmail.com>

This driver has been widely used since inclusion and no problems have
been reported.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NJean Delvare <khali@linux-fr.org>

The MCP78S and MCP79 appear to be compatible with the previous nForce
chips as far as the SMBus controller is concerned. The MCP67 and MCP73
were not tested yet but I'd be very surprised if they weren't
compatible too.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Cc: Oleg Ryjkov <olegr@olegr.ca>
Cc: Malcolm Lalkaka <mlalkaka@gmail.com>
Cc: Zbigniew Luszpinski <zbiggy@o2.pl>

The automatic removal of i2c clients only affects the clients which
were created automatically in the first place. Add a comment saying
that to avoid any confusion.
Signed-off-by: NJean Delvare <khali@linux-fr.org>

As suggested by Mauro Carvalho Chehab.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>

On popular demand, here comes some documentation about how to
instantiate i2c devices in the new (standard) i2c device driver
binding model.

I have also clarified how the class bitfield lets driver authors
control which buses are probed in the auto-detect case, and warned
more loudly against the abuse of this method.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Acked-by: NMichael Lawnick <nospam_lawnick@gmx.de>
Acked-by: NHans Verkuil <hverkuil@xs4all.nl>

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (166 commits)
  Revert "ax25: zero length frame filtering in AX25"
  Revert "netrom: zero length frame filtering in NetRom"
  cfg80211: default CONFIG_WIRELESS_OLD_REGULATORY to n
  mac80211/iwlwifi: move virtual A-MDPU queue bookkeeping to iwlwifi
  mac80211: fix aggregation to not require queue stop
  mac80211: add skb length sanity checking
  mac80211: unify and fix TX aggregation start
  mac80211: clean up __ieee80211_tx args
  mac80211: rework the pending packets code
  mac80211: fix A-MPDU queue assignment
  mac80211: rewrite fragmentation
  iwlwifi: show current driver status in user readable format
  b43: Add BCM4307 PCI-ID
  cfg80211: fix locking in nl80211_set_wiphy
  mac80211: fix RX path
  ath5k: properly drop packets from ops->tx
  ar9170: single module build
  ath9k: fix dma mapping leak of rx buffer upon rmmod
  rt2x00: New USB ID for rt73usb
  ath5k: warn and correct rate for unknown hw rate indexes
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc64: We need to use compat_sys_ustat() as well.
  sparc64: Fix MM refcount check in smp_flush_tlb_pending().

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: (53 commits)
  DVB: firedtv: FireDTV S2 problems with tuning solved
  DVB: firedtv: fix printk format mismatch
  ieee1394: constify device ID tables
  ieee1394: raw1394: add sparse annotations to raw1394_compat_write
  ieee1394: Storage class should be before const qualifier
  ieee1394: sbp2: follow up on "ieee1394: inherit ud vendor_id from node vendor_id"
  firewire: core: optimize propagation of BROADCAST_CHANNEL
  firewire: core: simplify broadcast channel allocation
  firewire: core: increase bus manager grace period
  firewire: core: drop unused call parameters of close_transaction
  firewire: cdev: add closure to async stream ioctl
  firewire: cdev: simplify FW_CDEV_IOC_SEND_REQUEST return value
  firewire: cdev: fix race of ioctl_send_request with bus reset
  firewire: cdev: secure add_descriptor ioctl
  firewire: cdev: amendment to "add ioctl to query maximum transmission speed"
  firewire: broadcast channel support
  firewire: implement asynchronous stream transmission
  firewire: core: normalize a function argument name
  firewire: normalize a variable name
  firewire: core: remove condition which is always false
  ...

Sparc was missed in commit 2b1c6bd7
("generic compat_sys_ustat").  We definitely need it, since our
__kernel_ino_t is "unsigned long".
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit f99bcff7.

Like netrom, Alan Cox says that zero lengths have real meaning
and are useful in this protocol.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit a3ac80a1.

Alan Cox says that zero length writes do have special meaning
and are useful in this protocol.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

And update description and feature-removal schedule according
to the new plan.
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This patch removes all the virtual A-MPDU-queue bookkeeping from
mac80211. Curiously, iwlwifi already does its own bookkeeping, so
it doesn't require much changes except where it needs to handle
starting and stopping the queues in mac80211.

To handle the queue stop/wake properly, we rewrite the software
queue number for aggregation frames and internally to iwlwifi keep
track of the queues that map into the same AC queue, and only talk
to mac80211 about the AC queue. The implementation requires calling
two new functions, iwl_stop_queue and iwl_wake_queue instead of the
mac80211 counterparts.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Cc: Reinette Chattre <reinette.chatre@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Instead of stopping the entire AC queue when enabling aggregation
(which was only done for hardware with aggregation queues) buffer
the packets for each station, and release them to the pending skb
queue once aggregation is turned on successfully.

We get a little more code, but it becomes conceptually simpler and
we can remove the entire virtual queue mechanism from mac80211 in
a follow-up patch.

This changes how mac80211 behaves towards drivers that support
aggregation but have no hardware queues -- those drivers will now
not be handed packets while the aggregation session is being
established, but only after it has been fully established.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We just found a bug in zd1211rw where it would reject
packets in the ->tx() method but leave them modified,
which would cause retransmit attempts with completely
bogus skbs, eventually leading to a panic due to not
having enough headroom in those.

This patch adds a sanity check to mac80211 to catch
such driver mistakes; in this case we warn and drop
the skb.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

When TX aggregation becomes operational, we do a number of steps:
 1) print a debug message
 2) wake the virtual queue
 3) notify the driver

Unfortunately, 1) and 3) are only done if the driver is first to
reply to the aggregation request, it is, however, possible that the
remote station replies before the driver! Thus, unify the code for
this and call the new function ieee80211_agg_tx_operational in both
places where TX aggregation can become operational.

Additionally, rename the driver notification from
IEEE80211_AMPDU_TX_RESUME to IEEE80211_AMPDU_TX_OPERATIONAL.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

__ieee80211_tx takes a struct ieee80211_tx_data argument, but only
uses a few of its members, namely 'skb' and 'sta'. Make that explicit,
so that less internal knowledge is required in ieee80211_tx_pending
and the possibility of introducing errors here is removed.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Reviewed-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The pending packets code is quite incomprehensible, uses memory barriers
nobody really understands, etc. This patch reworks it entirely, using
the queue spinlock, proper stop bits and the skb queues themselves to
indicate whether packets are pending or not (rather than a separate
variable like before).
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Reviewed-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Internally, mac80211 requires the skb's queue mapping to be set
to the AC queue, not the virtual A-MPDU queue. This is not done
correctly currently, this patch moves the code down to directly
before the driver is invoked and adds a comment that it will be
moved into the driver later.

Since this requires __ieee80211_tx() to have the sta pointer,
make sure to provide it in ieee80211_tx_pending().
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Reviewed-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Fragmentation currently uses an allocated array to store the
fragment skbs, and then keeps track of which have been sent
and which are still pending etc. This is rather complicated;
make it simpler by just chaining the fragments into skb->next
and removing from that list when sent. Also simplifies all
code that needs to touch fragments, since it now only needs
to walk the skb->next list.

This is a prerequisite for fixing the stored packet code,
which I need to do for proper aggregation packet storing.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Reviewed-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

change the display of current driver status bit to user readable format
for better and easier debugging
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NReinette Chatre <reinette.chatre@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

00:09.0 Network controller [0280]: Broadcom Corporation BCM4307 Ethernet Controller [14e4:4306] (rev 03)
    Subsystem: Broadcom Corporation BCM4307 Ethernet Controller [14e4:4306]
    Flags: bus master, fast devsel, latency 32, IRQ 10
    Memory at d7000000 (32-bit, non-prefetchable) [size=8K]
    Kernel driver in use: b43-pci-bridge
    Kernel modules: ssb
Reported-by: Nyoann <yoann71@msn.com>
Signed-off-by: NMichael Buesch <mb@bu3sch.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Luis reports that there's a circular locking dependency;
this is because cfg80211_dev_rename() will acquire the
cfg80211_mutex while the device mutex is held, while
this normally is done the other way around. The solution
is to open-code the device-getting in nl80211_set_wiphy
and require holding the mutex around cfg80211_dev_rename
rather than acquiring it within.

Also fix a bug -- rtnl locking is expected by drivers so
we need to provide it.
Reported-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

My previous patch ("mac80211: remove mixed-cell and userspace MLME code")
was too obvious to me, so obvious that a stupid bug crept in. The IBSS
RX function must be invoked for IBSS, of course, not anything != IBSS.
Reported-by: NLarry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Tested-by: NLarry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We shouldn't return NETDEV_TX_BUSY from the TX callback, especially
after we've mucked with the sk_buffs.  Drop the packets and return
NETDEV_TX_OK.

Changes-licensed-under: 3-Clause-BSD
Signed-off-by: NBob Copeland <me@bobcopeland.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This patch restores all-in-one module build procedure for ar9170.
Signed-off-by: NChristian Lamparter <chunkeey@web.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We were claiming DMA buffers on the RX tasklet but never
upon a simple module removal.

Cc: stable@kernel.org
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Signed-off-by: NIvo van Doorn <IvDoorn@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>