07 Dec, 2010
17 commits
-
Instead of tying mesh activity to interface up,
add join and leave commands for mesh. Since we
must be backward compatible, let cfg80211 handle
joining a mesh if a mesh ID was pre-configured
when the device goes up.Note that this therefore must modify mac80211 as
well since mac80211 needs to lose the logic to
start the mesh on interface up.We now allow querying mesh parameters before the
mesh is connected, which simply returns defaults.
Setting them (internally renamed to "update") is
only allowed while connected. Specify them with
the new mesh join command instead where needed.In mac80211, beaconing must now also follow the
mesh enabled/not enabled state, which is done
by testing the mesh ID.Signed-off-by: Javier Cardona
Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
I'm going to need this in a new place later.
Tested-by: Javier Cardona
Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
cfg80211 used to do all its bookkeeping in
the notifier, but some new stuff will have
to use local variables so make the callback
return the netdev pointer.Tested-by: Javier Cardona
Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Logically, the filter adjusting belongs with
starting/stopping mesh, not interface up/down,
so move it there.Tested-by: Javier Cardona
Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
The TTL in path selection information elements is different from
the mesh ttl used in mesh data frames. Version 7.03 of the 11s
draft calls this ttl 'Element TTL'.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
* Since we report tx acks to the protocol stack, add the needed
flag to hw_flags. This way we'll also use the new AP probing mechanism.Signed-off-by: Nick Kossifidis
Tested-by: Sedat Dilek
Signed-off-by: John W. Linville -
* Correctly report failure to stop RX DMA
Signed-off-by: Nick Kossifidis
Tested-by: Sedat Dilek
Signed-off-by: John W. Linville -
* Stop ANI durring reset to prevent false PHY error reports
Signed-off-by: Nick Kossifidis
Tested-by: Sedat Dilek
Signed-off-by: John W. Linville -
* Always free tx buffers before reset, since we also empty hw queues.
If we don't and a queue gets stuck, we'll never decrease txq_len and sw
will keep thinking the queue is still stuck even after reset.Signed-off-by: Nick Kossifidis
Tested-by: Sedat Dilek
Signed-off-by: John W. Linville -
* By skipping tx power table calibration we also skip setting
tx power table on hw. Make sure we always write tx power table
on hw since it gets cleared on reset.Signed-off-by: Nick Kossifidis
Tested-by: Sedat Dilek
Signed-off-by: John W. Linville -
If the nullfunc frame used to probe the AP was not acked, there is no point
in waiting for the probe timeout, so advance to the next try (or disconnect)
immediately.
If we do reach the probe timeout without having received a tx status, the
connection is probably really bad and worth disconnecting.Signed-off-by: Felix Fietkau
Signed-off-by: John W. Linville -
ieee80211_is_nullfunc() implies ieee80211_is_data()
Signed-off-by: Felix Fietkau
Signed-off-by: John W. Linville -
Using bitshifts instead of division and multiplication should improve
performance. That requires weight and factor to be powers of two, but i think
this is something we can live with.Thanks to Peter Zijlstra for the improved formula!
Signed-off-by: Bruno Randolf
--
v2: use log2.h functions
Signed-off-by: John W. Linville -
This prepares the only place which uses the EWMA library so far for the
performance improved implementation coming up, which requires factor and
weight to be a power of two.Signed-off-by: Bruno Randolf
Signed-off-by: John W. Linville -
This file pattern is not currently shown as maintained
by atheros. Perhaps it should be?Signed-off-by: Joe Perches
Signed-off-by: John W. Linville -
The last_tx_rate field was also updated for non-data frames that are
often sent with a lower rate (for example management frames at 1 Mbps).
This is confusing when the data rate is actually much higher.Hence, only update the last_tx_rate field with tx rate information
gathered from last data frames.If the rate control algorithm filled in txrc.reported_rate we don't need
to verify this information.Cc: Johannes Berg
Signed-off-by: Helmut Schaa
Signed-off-by: John W. Linville -
…t/iwlwifi/iwlwifi-2.6
03 Dec, 2010
23 commits
-
Conflicts:
drivers/net/wireless/ath/ath9k/ar9003_eeprom.c -
On WiSoc we cannot access mac register before it is resetted.
It will crash hardware otherwise.Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
AHB specific functions are now in ahb.c file. AHB bus is
compiled in when CONFIG_ATHEROS_AR231X is set in kernel.
All other platforms will use PCI bus.Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Add bus dependent revision read function which is used to
determine chipset's MAC before hardware is initialized.Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
To be able to support other busses than PCI check if pci device
structure is initialized.Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Split pci initialization into hardware specific
functions and softc structure initialization.
Make function naming similar to ones ath9k.
Introduce ath_bus_opts in ath5k for later
AHB bus integration.Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Signed-off-by: Felix Fietkau
Signed-off-by: Wojciech Dubowik
Signed-off-by: John W. Linville -
Signed-off-by: Andy Shevchenko
Cc: "John W. Linville"
Cc: linux-wireless@vger.kernel.org
Signed-off-by: John W. Linville -
The last patch with the same title was for mac80211 ops, accidentally.
Signed-off-by: Bruno Randolf
Signed-off-by: John W. Linville -
Create a new chapter about SMPS in the
driver API part of the mac80211 book and
populate it with the existing docs.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Add some existing documentation about station
handling to a new chapter about advanced APIs.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Publish the documentation for the interface
iteration functions in the right section of
the book.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Include the aggregation documentation in the
docbook, split up by public and internal docs.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Add an API chapter to the rate control part of
the mac80211 book and populate it with some
existing documentation.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
There isn't a clear TX documentation yet, so
put these into the misc section for now.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Put them along with their "regular" versions for
now, we should write more docs in the future.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Create a new chapter for the mac80211 workqueue
that contains the documentation written for it.Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Just include them in the list of functions,
there aren't any more detailed docs (yet).Signed-off-by: Johannes Berg
Signed-off-by: John W. Linville -
Signed-off-by: Sujith Manoharan
Signed-off-by: John W. Linville
