ath9k working, almost

ath9k working, almost

[Alistair John Strachan]

Hi,

The newly announced ath9k driver works great on an AR5418 Core 2 Duo Macbook, 
WPA2-TKIP/AES et al all work fine. It would be good to see this driver in 
wireless-testing, it completely replaces madwifi for me. I used linux.git 
(2.6.26) + wireless-testing.git + my-wireless-testing.git with success.

The only thing that I couldn't get working was the signal strength meter in 
knetworkmanager. The AP signal strength shows up in iwlist ath0 scanning, but 
whatever NM uses doesn't seem to work. All is good with madwifi, by 
comparison.

Any ideas? NM issue?

-- 
Cheers,
Alistair.

Re: ath9k working, almost

[Luis R. Rodriguez]

On Sat, Jul 26, 2008 at 9:42 AM, Alistair John Strachan
<alistair@devzero.co.uk> wrote:
> Hi,
>
> The newly announced ath9k driver works great on an AR5418 Core 2 Duo Macbook,
> WPA2-TKIP/AES et al all work fine. It would be good to see this driver in
> wireless-testing, it completely replaces madwifi for me. I used linux.git
> (2.6.26) + wireless-testing.git + my-wireless-testing.git with success.

Please keep in mind Linus' tree is just broken for wireless. But glad
it is working for you after your tremendous strange changes on your
tree...

> The only thing that I couldn't get working was the signal strength meter in
> knetworkmanager.

Its part of our TODO list. We rather get you supported with a new
driver, part of the release and review early process.

  Luis

Re: ath9k working, almost

[Alistair John Strachan]

On Saturday 26 July 2008 19:15:51 Luis R. Rodriguez wrote:
> On Sat, Jul 26, 2008 at 9:42 AM, Alistair John Strachan
>
> <alistair@devzero.co.uk> wrote:
> > Hi,
> >
> > The newly announced ath9k driver works great on an AR5418 Core 2 Duo
> > Macbook, WPA2-TKIP/AES et al all work fine. It would be good to see this
> > driver in wireless-testing, it completely replaces madwifi for me. I used
> > linux.git (2.6.26) + wireless-testing.git + my-wireless-testing.git with
> > success.
>
> Please keep in mind Linus' tree is just broken for wireless. But glad
> it is working for you after your tremendous strange changes on your
> tree...

I explicitly said I was not using Linus' tree. Perhaps it is a 
misunderstanding on your part. If you use git pull inside a tree that another 
tree was originally forked from, it is much smaller and faster to check out. 
This is the process I was describing. I am pretty sure there are 
no "tremendous strange changes". :-)

Thanks for the driver..

-- 
Cheers,
Alistair.

Re: ath9k working, almost

[Luis R. Rodriguez]

On Sat, Jul 26, 2008 at 12:30 PM, Alistair John Strachan
<alistair@devzero.co.uk> wrote:
> On Saturday 26 July 2008 19:15:51 Luis R. Rodriguez wrote:
>> On Sat, Jul 26, 2008 at 9:42 AM, Alistair John Strachan
>>
>> <alistair@devzero.co.uk> wrote:
>> > Hi,
>> >
>> > The newly announced ath9k driver works great on an AR5418 Core 2 Duo
>> > Macbook, WPA2-TKIP/AES et al all work fine. It would be good to see this
>> > driver in wireless-testing, it completely replaces madwifi for me. I used
>> > linux.git (2.6.26) + wireless-testing.git + my-wireless-testing.git with
>> > success.
>>
>> Please keep in mind Linus' tree is just broken for wireless. But glad
>> it is working for you after your tremendous strange changes on your
>> tree...
>
> I explicitly said I was not using Linus' tree. Perhaps it is a
> misunderstanding on your part. If you use git pull inside a tree that another
> tree was originally forked from, it is much smaller and faster to check out.
> This is the process I was describing. I am pretty sure there are
> no "tremendous strange changes". :-)

Well my concern was I believe Linus' tree has the new MQ stuff and
wireless-testing doesn't. How did you handle that?

  Luis

Re: ath9k working, almost

[Alistair John Strachan]

On Saturday 26 July 2008 20:36:22 Luis R. Rodriguez wrote:
> On Sat, Jul 26, 2008 at 12:30 PM, Alistair John Strachan
>
> <alistair@devzero.co.uk> wrote:
> > On Saturday 26 July 2008 19:15:51 Luis R. Rodriguez wrote:
> >> On Sat, Jul 26, 2008 at 9:42 AM, Alistair John Strachan
> >>
> >> <alistair@devzero.co.uk> wrote:
> >> > Hi,
> >> >
> >> > The newly announced ath9k driver works great on an AR5418 Core 2 Duo
> >> > Macbook, WPA2-TKIP/AES et al all work fine. It would be good to see
> >> > this driver in wireless-testing, it completely replaces madwifi for
> >> > me. I used linux.git (2.6.26) + wireless-testing.git +
> >> > my-wireless-testing.git with success.
> >>
> >> Please keep in mind Linus' tree is just broken for wireless. But glad
> >> it is working for you after your tremendous strange changes on your
> >> tree...
> >
> > I explicitly said I was not using Linus' tree. Perhaps it is a
> > misunderstanding on your part. If you use git pull inside a tree that
> > another tree was originally forked from, it is much smaller and faster to
> > check out. This is the process I was describing. I am pretty sure there
> > are no "tremendous strange changes". :-)
>
> Well my concern was I believe Linus' tree has the new MQ stuff and
> wireless-testing doesn't. How did you handle that?

git pull will happily merge into a non-HEAD checkout. If you do something
like this (inside your linux-2.6 checkout):

git reset --hard v2.6.26
git pull git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
git pull git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/my-wireless-testing.git ath9k

Then you get a minimal, incremental set of changes based on 2.6.26 (which is
what most wireless-testing users will want). There are some drawbacks but it
sure beats wasting several hundred megs on two git repos.

If you want to go back to Linus you just do:

git reset --hard v2.6.26

And it'll blow away all the merges (since you were never tracking those trees).

-- 
Cheers,
Alistair.

Re: ath9k working, almost

[Dan Williams]

On Sat, 2008-07-26 at 17:42 +0100, Alistair John Strachan wrote:
> Hi,
> 
> The newly announced ath9k driver works great on an AR5418 Core 2 Duo Macbook, 
> WPA2-TKIP/AES et al all work fine. It would be good to see this driver in 
> wireless-testing, it completely replaces madwifi for me. I used linux.git 
> (2.6.26) + wireless-testing.git + my-wireless-testing.git with success.
> 
> The only thing that I couldn't get working was the signal strength meter in 
> knetworkmanager. The AP signal strength shows up in iwlist ath0 scanning, but 
> whatever NM uses doesn't seem to work. All is good with madwifi, by 
> comparison.
> 
> Any ideas? NM issue?

The signal strength code in NM hasn't changed for quite a long time.
Chances are it's the driver not reporting correctly.  What's the
iwconfig output for while you're connected to your AP?

 Link Quality=92/100  Signal level=-36 dBm  Noise level=-91 dBm

That's the line I care about.

Dan

Re: ath9k working, almost

[Luis R. Rodriguez]

On Sun, Jul 27, 2008 at 7:59 AM, Dan Williams <dcbw@redhat.com> wrote:
> On Sat, 2008-07-26 at 17:42 +0100, Alistair John Strachan wrote:
>> Hi,
>>
>> The newly announced ath9k driver works great on an AR5418 Core 2 Duo Macbook,
>> WPA2-TKIP/AES et al all work fine. It would be good to see this driver in
>> wireless-testing, it completely replaces madwifi for me. I used linux.git
>> (2.6.26) + wireless-testing.git + my-wireless-testing.git with success.
>>
>> The only thing that I couldn't get working was the signal strength meter in
>> knetworkmanager. The AP signal strength shows up in iwlist ath0 scanning, but
>> whatever NM uses doesn't seem to work. All is good with madwifi, by
>> comparison.
>>
>> Any ideas? NM issue?
>
> The signal strength code in NM hasn't changed for quite a long time.
> Chances are it's the driver not reporting correctly.  What's the
> iwconfig output for while you're connected to your AP?
>
>  Link Quality=92/100  Signal level=-36 dBm  Noise level=-91 dBm
>
> That's the line I care about.

Its definitely borked, we just have to fix it. Its on our TODO list,
unless someone gets to it first.

  Luis

Re: ath9k working, almost

[Alistair John Strachan]

On Sunday 27 July 2008 16:07:05 Luis R. Rodriguez wrote:
> On Sun, Jul 27, 2008 at 7:59 AM, Dan Williams <dcbw@redhat.com> wrote:
> > On Sat, 2008-07-26 at 17:42 +0100, Alistair John Strachan wrote:
> >> Hi,
> >>
> >> The newly announced ath9k driver works great on an AR5418 Core 2 Duo
> >> Macbook, WPA2-TKIP/AES et al all work fine. It would be good to see this
> >> driver in wireless-testing, it completely replaces madwifi for me. I
> >> used linux.git (2.6.26) + wireless-testing.git + my-wireless-testing.git
> >> with success.
> >>
> >> The only thing that I couldn't get working was the signal strength meter
> >> in knetworkmanager. The AP signal strength shows up in iwlist ath0
> >> scanning, but whatever NM uses doesn't seem to work. All is good with
> >> madwifi, by comparison.
> >>
> >> Any ideas? NM issue?
> >
> > The signal strength code in NM hasn't changed for quite a long time.
> > Chances are it's the driver not reporting correctly.  What's the
> > iwconfig output for while you're connected to your AP?
> >
> >  Link Quality=92/100  Signal level=-36 dBm  Noise level=-91 dBm
> >
> > That's the line I care about.
>
> Its definitely borked, we just have to fix it. Its on our TODO list,
> unless someone gets to it first.

This is probably an issue of not RTFM'ing (sorry), but are there documents 
enabling Joe User to contribute this kind of thing?

How much of the XXX fixups are you planning on doing yourself, and how many 
would you like help with?

-- 
Cheers,
Alistair.

Re: ath9k working, almost

[Luis R. Rodriguez]

On Sun, Jul 27, 2008 at 9:20 AM, Alistair John Strachan
<alistair@devzero.co.uk> wrote:
> On Sunday 27 July 2008 16:07:05 Luis R. Rodriguez wrote:
>> On Sun, Jul 27, 2008 at 7:59 AM, Dan Williams <dcbw@redhat.com> wrote:
>> > On Sat, 2008-07-26 at 17:42 +0100, Alistair John Strachan wrote:
>> >> Hi,
>> >>
>> >> The newly announced ath9k driver works great on an AR5418 Core 2 Duo
>> >> Macbook, WPA2-TKIP/AES et al all work fine. It would be good to see this
>> >> driver in wireless-testing, it completely replaces madwifi for me. I
>> >> used linux.git (2.6.26) + wireless-testing.git + my-wireless-testing.git
>> >> with success.
>> >>
>> >> The only thing that I couldn't get working was the signal strength meter
>> >> in knetworkmanager. The AP signal strength shows up in iwlist ath0
>> >> scanning, but whatever NM uses doesn't seem to work. All is good with
>> >> madwifi, by comparison.
>> >>
>> >> Any ideas? NM issue?
>> >
>> > The signal strength code in NM hasn't changed for quite a long time.
>> > Chances are it's the driver not reporting correctly.  What's the
>> > iwconfig output for while you're connected to your AP?
>> >
>> >  Link Quality=92/100  Signal level=-36 dBm  Noise level=-91 dBm
>> >
>> > That's the line I care about.
>>
>> Its definitely borked, we just have to fix it. Its on our TODO list,
>> unless someone gets to it first.
>
> This is probably an issue of not RTFM'ing (sorry), but are there documents
> enabling Joe User to contribute this kind of thing?

No documents, just the source.

> How much of the XXX fixups are you planning on doing yourself, and how many
> would you like help with?

We plan on doing a lot, I'll post a new series in a few minutes with
about 15 patches from us. If you'd like to work on the link quality
that would be greatly welcomed. What you can expect from us soon is
more enhancements to STA, enabling IBSS, and AP. If it helps we can
try to come up with a list of elements we can take care of and a few
others we can use community help with. But in general patches from the
community are greatly welcomed and encouraged. We all just want the
best driver possible. We can also coordinate if you want to work on
something in particular.

  Luis

Re: ath9k working, almost

[Stefanik Gábor]

On Sun, Jul 27, 2008 at 6:30 PM, Luis R. Rodriguez <mcgrof@gmail.com> w=
rote:
> What you can expect from us soon is more enhancements
> to STA, enabling IBSS, and AP.

Not really related, but is monitor mode also planned to be enabled? Or
is it already enabled?

--G=E1bor

--=20
Vista: [V]iruses, [I]ntruders, [S]pyware, [T]rojans and [A]dware. :-)
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: ath9k working, almost

[Russell Senior]

>>>>> "Stefanik" =3D=3D Stefanik G=E1bor <netrolller.3d@gmail.com> writ=
es:

Luis> What you can expect from us soon is more enhancements to STA,
Luis> enabling IBSS, and AP.

Stefanik> Not really related, but is monitor mode also planned to be
Stefanik> enabled? Or is it already enabled?

And how about WDS?  That's something I'm using on madwifi (on older
atheros hardware) right now.


--=20
Russell Senior         ``I have nine fingers; you have ten.''
seniorr@aracnet.com
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: ath9k working, almost

[Luis R. Rodriguez]
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Re: ath9k working, almost

[Luis R. Rodriguez]

T24gU3VuLCBKdWwgMjcsIDIwMDggYXQgMTI6MTEgUE0sIFN0ZWZhbmlrIEfDoWJvcgo8bmV0cm9s
bGxlci4zZEBnbWFpbC5jb20+IHdyb3RlOgo+IE9uIFN1biwgSnVsIDI3LCAyMDA4IGF0IDY6MzAg
UE0sIEx1aXMgUi4gUm9kcmlndWV6IDxtY2dyb2ZAZ21haWwuY29tPiB3cm90ZToKPj4gV2hhdCB5
b3UgY2FuIGV4cGVjdCBmcm9tIHVzIHNvb24gaXMgbW9yZSBlbmhhbmNlbWVudHMKPj4gdG8gU1RB
LCBlbmFibGluZyBJQlNTLCBhbmQgQVAuCj4KPiBOb3QgcmVhbGx5IHJlbGF0ZWQsIGJ1dCBpcyBt
b25pdG9yIG1vZGUgYWxzbyBwbGFubmVkIHRvIGJlIGVuYWJsZWQ/IE9yCj4gaXMgaXQgYWxyZWFk
eSBlbmFibGVkPwoKSXQgaXMuIE5lZWRzIHRlc3RpbmcuCgogIEx1aXMK

Re: ath9k working, almost

[Luis R. Rodriguez]

On Sun, Jul 27, 2008 at 9:30 AM, Luis R. Rodriguez <mcgrof@gmail.com> wrote:
> On Sun, Jul 27, 2008 at 9:20 AM, Alistair John Strachan

>> How much of the XXX fixups are you planning on doing yourself, and how many
>> would you like help with?
>
> We plan on doing a lot, I'll post a new series in a few minutes with
> about 15 patches from us. If you'd like to work on the link quality
> that would be greatly welcomed. What you can expect from us soon is
> more enhancements to STA, enabling IBSS, and AP. If it helps we can
> try to come up with a list of elements we can take care of and a few
> others we can use community help with. But in general patches from the
> community are greatly welcomed and encouraged. We all just want the
> best driver possible. We can also coordinate if you want to work on
> something in particular.

I'll post a revised TODO list on ath9k-devel. If you'd like to help
please subscribe and feel free to take up an item if you'd like to
help out with the driver. Patches are greatly appreciated and
welcomed.

  Luis