On Sat, 2008-07-12 at 03:21 +0200, Mildred wrote:
> Hi,
> 
> I have an AR5418 chipset. It is supported (with some bugs) by the
> (closed source) Madwifi driver. Is there anything related to this
> chipset in the ath5k driver ?
> 
> If there isn't, perhaps I can help ...
> Is there anything I could do ?

As it stands now, we need to port the tracing functionality available in
the madwifi-trace branch to the new tracing infrastructure scheduled to
appear in Linux 2.6.27.  Right now, it's present in the ftrace
repository:

git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-ftrace.git

It's also available in the linux-next repository, but it was very
unstable a few days ago.

That would allow us to make traces from ndiswrapper and all madwifi
versions without changing their code.

I realize that it's a big task, but I think it's the blocker.  Other
users are asking how to help, and all we can do is to ask them to
downgrade to Linux 2.6.23 and use madwifi-trace branch, which is not
working well.  The result is usually silence.

Another development task is improving ath_info.  It should not be so
verbose by default.  We need a database of known combinations of MAC,
PHY, EEPROM and so on.  We need a way to support more than one access
method to EEPROM and more than one EEPROM layout.  Reading EEPROM for
AR5418 has been reverse engineered, but the code is not ready for
committing yet.  And write access for AR5418 has not been reverse
engineered.

-- 
Regards,
Pavel Roskin
_______________________________________________
ath5k-users mailing list
[email protected]
https://lists.ath5k.org/mailman/listinfo/ath5k-users

Reply via email to