4:45 p.m.
For a long time, the Atheros 5007EG hasn't worked with the ath5k module,
and those of us who had it used MadWifi.
Today, I upgraded my kernel, and the latest special Madwifi snapshot
wouldn't build. This has happened in the past, last time, someone on
this list had submitted a patch to fix it.
Just for fun, I decided to try modprobing ath5k and seeing if the card
might be recognized and work.
It seems that it does. This is running
2.6.27-0.244.rc2.git1.fc10.x86_64. I modprobed ath5k and saw that I now
had a visible wlan0 card. Did ifconfig up and iwlist scan and it saw
the networks. I then adjusted wpa_supplicant (which I use rather than
NetworkManager--I use fluxbox most of the time and often just boot into
runlevel 3) and Lo! and Behold! it worked.
I know that at least one other person on this list uses the card. I'd
be interested if other folks with that card (The older EEE PC uses it,
though I believe the 900 series uses a realink of some sort, but it's
fairly popular in lower end laptops) are also having success. Thanks
for any input.
To the kernel developers, thank you, great work.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Buffy: Well, that works out great. You won't tell anyone
that I'm the Slayer, and I won't tell anyone you're a moron.
5:05 p.m.
On Sat, Aug 9, 2008 at 4:45 PM, Scott Robbins <scottro(a)nyc.rr.com> wrote:
For a long time, the Atheros 5007EG hasn't worked with the ath5k
module,
and those of us who had it used MadWifi.
Today, I upgraded my kernel, and the latest special Madwifi snapshot
wouldn't build. This has happened in the past, last time, someone on
this list had submitted a patch to fix it.
Just for fun, I decided to try modprobing ath5k and seeing if the card
might be recognized and work.
It seems that it does. This is running
2.6.27-0.244.rc2.git1.fc10.x86_64. I modprobed ath5k and saw that I now
had a visible wlan0 card. Did ifconfig up and iwlist scan and it saw
the networks. I then adjusted wpa_supplicant (which I use rather than
NetworkManager--I use fluxbox most of the time and often just boot into
runlevel 3) and Lo! and Behold! it worked.
I know that at least one other person on this list uses the card. I'd
be interested if other folks with that card (The older EEE PC uses it,
though I believe the 900 series uses a realink of some sort, but it's
fairly popular in lower end laptops) are also having success. Thanks
for any input.
To the kernel developers, thank you, great work.
I am very interested to hear this.
As you can see from this bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=453124, I am having quite
a bit of problems with the madwifi driver + network manager... I'll
have to give that kernel a try, see if it works with the machine.
--
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )
5:49 p.m.
On Sat, Aug 09, 2008 at 05:05:54PM -0500, Arthur Pemberton wrote:
On Sat, Aug 9, 2008 at 4:45 PM, Scott Robbins
<scottro(a)nyc.rr.com> wrote:
> For a long time, the Atheros 5007EG hasn't worked with the ath5k module,
> and those of us who had it used MadWifi.
>
>
> It seems that it does. This is running
> 2.6.27-0.244.rc2.git1.fc10.x86_64. I modprobed ath5k and saw that I now
> had a visible wlan0 card. Did ifconfig up and iwlist scan and it saw
> the networks. I then adjusted wpa_supplicant (which I use rather than
> NetworkManager--I use fluxbox most of the time and often just boot into
> runlevel 3) and Lo! and Behold! it worked.
>
I am very interested to hear this.
As you can see from this bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=453124, I am having quite
a bit of problems with the madwifi driver + network manager... I'll
have to give that kernel a try, see if it works with the machine.
I don't use resume at all, so I'm not sure if it will fix the problem
you're having. It would be nice if it did. I'm wondering if this ath5k
module is going to be backported.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Spike: You're not friends. You'll never be friends. You'll be in
love 'til it kills you both. You'll fight, and you'll shag, and
you'll hate each other 'til it makes you quiver, but you'll never
be friends. Real love isn't brains, children. It's blood. It's
blood screaming inside you to work its will. I may be love's
bitch, but at least I'm man enough to admit it.
6:17 p.m.
2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
On Sat, Aug 09, 2008 at 05:05:54PM -0500, Arthur Pemberton wrote:
> On Sat, Aug 9, 2008 at 4:45 PM, Scott Robbins <scottro(a)nyc.rr.com> wrote:
> > For a long time, the Atheros 5007EG hasn't worked with the ath5k module,
> > and those of us who had it used MadWifi.
> >
> >
> > It seems that it does. This is running
> > 2.6.27-0.244.rc2.git1.fc10.x86_64. I modprobed ath5k and saw that I now
> > had a visible wlan0 card. Did ifconfig up and iwlist scan and it saw
> > the networks. I then adjusted wpa_supplicant (which I use rather than
> > NetworkManager--I use fluxbox most of the time and often just boot into
> > runlevel 3) and Lo! and Behold! it worked.
> >
>
>
> I am very interested to hear this.
>
> As you can see from this bug report:
> https://bugzilla.redhat.com/show_bug.cgi?id=453124, I am having quite
> a bit of problems with the madwifi driver + network manager... I'll
> have to give that kernel a try, see if it works with the machine.
I don't use resume at all, so I'm not sure if it will fix the problem
you're having. It would be nice if it did. I'm wondering if this ath5k
module is going to be backported.
To what? It's in Fedora 9 already.
--
Christopher Brown
http://www.chruz.com
7:34 p.m.
On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
>
> I don't use resume at all, so I'm not sure if it will fix the problem
> you're having. It would be nice if it did. I'm wondering if this ath5k
> module is going to be backported.
To what? It's in Fedora 9 already.
In Fedora 9, I'd been using the livna-testing repo's version of MadWifi.
Thanks for the information, I just tried the F9 version. However, it
doesn't work for me with that card, and I had to go back to the livna
madwifi.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Whistler: Bottom line is, even if you see them coming, you're not
ready for the big moments. No one asks for their life to change,
not really. But it does. So, what are we, helpless? Puppets? Nah.
The big moments are gonna come, you can't help that. It's what
you do afterwards that counts. That's when you find out who you
are.
8 p.m.
On Sat, Aug 9, 2008 at 7:34 PM, Scott Robbins <scottro(a)nyc.rr.com> wrote:
On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
> 2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
> >
> > I don't use resume at all, so I'm not sure if it will fix the problem
> > you're having. It would be nice if it did. I'm wondering if this
ath5k
> > module is going to be backported.
>
> To what? It's in Fedora 9 already.
In Fedora 9, I'd been using the livna-testing repo's version of MadWifi.
Thanks for the information, I just tried the F9 version. However, it
doesn't work for me with that card, and I had to go back to the livna
madwifi.
Aside from the suspend issues, the livna drivers work. But not being
able to standby/suspend is a fairly big issue for a laptop.
--
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )
6:41 a.m.
2008/8/10 Arthur Pemberton <pemboa(a)gmail.com>:
On Sat, Aug 9, 2008 at 7:34 PM, Scott Robbins
<scottro(a)nyc.rr.com> wrote:
> On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
>> 2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
>
>> >
>> > I don't use resume at all, so I'm not sure if it will fix the
problem
>> > you're having. It would be nice if it did. I'm wondering if this
ath5k
>> > module is going to be backported.
>>
>> To what? It's in Fedora 9 already.
>
> In Fedora 9, I'd been using the livna-testing repo's version of MadWifi.
> Thanks for the information, I just tried the F9 version. However, it
> doesn't work for me with that card, and I had to go back to the livna
> madwifi.
Aside from the suspend issues, the livna drivers work. But not being
able to standby/suspend is a fairly big issue for a laptop.
And as such it never will. The livna drivers are deprecated and all
work is now being carried out on the ath5k/ath9k tree.
--
Christopher Brown
http://www.chruz.com
8:48 a.m.
On Sun, Aug 10, 2008 at 12:41:36PM +0100, Christopher Brown wrote:
> On Sat, Aug 9, 2008 at 7:34 PM, Scott Robbins
<scottro(a)nyc.rr.com> wrote:
>> On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
>>> 2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
>>
>> In Fedora 9, I'd been using the livna-testing repo's version of
MadWifi.
>> Thanks for the information, I just tried the F9 version. However, it
>> doesn't work for me with that card, and I had to go back to the livna
>> madwifi.
And as such it never will. The livna drivers are deprecated and all
work is now being carried out on the ath5k/ath9k tree.
Hello Christopher--to reply to both your posts in one post (the other
one suggested I file a more specific bug report)
At present, the card not working with 2.6.25.x (and 2.6.26.x) kernels is
expected behavior, consistent across several distributions. So it isn't
a Fedora bug. Therefore, I don't really think a bug report will be
useful. There are a variety of bug reports on MadWifi concerning the
card, though their ticket 1192 is the one that is now the most important
one. Work is progressing, and the 2.6.27 kernel should have the card
supported.
That ticket also indicates that it is exactly as you say, work is going
to be done on ath5k in the future. Those of us who have the card are,
however, certainly grateful to FreeBSD's Sam Leffler for enabling us to
use it on both 32 and 64 bit systems. (I might add my own gratitude to
the gentleman on this list who made a patch that was accepted that fixed
another problem with another Rawhide kernel.)
So, it seems to me that filing a Fedora bug report would be simply
reporting expected behavior, and behavior that isn't a Fedora specific
issue.
Thanks.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Buffy: Will, wait. I'm really sorry.
Willow: Buff, I'm storming off. It doesn't really work if you come
with me.
9:04 a.m.
Scott Robbins wrote:
So, it seems to me that filing a Fedora bug report would be simply
reporting expected behavior, and behavior that isn't a Fedora specific
issue.
Regardless of whether it is Fedora specific or not, issues that affect
Fedora should be reported as bugs. If they are major ones, they should
probably go into the Fedora bugzilla. Otherwise, atleast upstream should
be notified.
Rahul
10:10 a.m.
On Sun, Aug 10, 2008 at 07:34:55PM +0530, Rahul Sundaram wrote:
Regardless of whether it is Fedora specific or not, issues that
affect Fedora
should be reported as bugs. If they are major ones, they should probably go
into the Fedora bugzilla. Otherwise, atleast upstream should be notified.
Upstream is aware of the issue.
At any rate, in the near future, it will become a moot point, thanks to
the developers (both kernel and MadWifi).
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Kendra: I call it Mr. Pointy.
Buffy: You named your stake?
Kendra: Yes.
Buffy: Remind me to get you a stuffed animal.
3:08 p.m.
On Sun, Aug 10, 2008 at 6:41 AM, Christopher Brown
<snecklifter(a)gmail.com> wrote:
2008/8/10 Arthur Pemberton <pemboa(a)gmail.com>:
> On Sat, Aug 9, 2008 at 7:34 PM, Scott Robbins <scottro(a)nyc.rr.com> wrote:
>> On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
>>> 2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
>>
>>> >
>>> > I don't use resume at all, so I'm not sure if it will fix the
problem
>>> > you're having. It would be nice if it did. I'm wondering if
this ath5k
>>> > module is going to be backported.
>>>
>>> To what? It's in Fedora 9 already.
>>
>> In Fedora 9, I'd been using the livna-testing repo's version of MadWifi.
>> Thanks for the information, I just tried the F9 version. However, it
>> doesn't work for me with that card, and I had to go back to the livna
>> madwifi.
>
> Aside from the suspend issues, the livna drivers work. But not being
> able to standby/suspend is a fairly big issue for a laptop.
And as such it never will. The livna drivers are deprecated and all
work is now being carried out on the ath5k/ath9k tree.
This is even more useful information. Would have loved to know about
this before, since I've just been waiting on the bug reports hping
something gets updated. As soon as I try the ath5k version, I'll
report back.
--
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )
4:04 p.m.
On Sun, Aug 10, 2008 at 03:08:11PM -0500, Arthur Pemberton wrote:
>
> And as such it never will. The livna drivers are deprecated and all
> work is now being carried out on the ath5k/ath9k tree.
This is even more useful information. Would have loved to know about
this before, since I've just been waiting on the bug reports hping
something gets updated. As soon as I try the ath5k version, I'll
report back.
Heh, I'm glad that I'm not the only one who missed that in ticket 1192.
Kevmf (who posted on that ticket) called it to my attention on Fedora
forums. It's towards the bottom where they state that it will probably
be in place for the 2.6.27 kernel.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Buffy: She's playing you. She tried to kill you.
Angel: That was just. . . That was just a cry for help.
Buffy: A cry for help is when you say Help in a loud
voice.
4:21 p.m.
On Sun, Aug 10, 2008 at 4:04 PM, Scott Robbins <scottro(a)nyc.rr.com> wrote:
On Sun, Aug 10, 2008 at 03:08:11PM -0500, Arthur Pemberton wrote:
> >
> > And as such it never will. The livna drivers are deprecated and all
> > work is now being carried out on the ath5k/ath9k tree.
>
>
> This is even more useful information. Would have loved to know about
> this before, since I've just been waiting on the bug reports hping
> something gets updated. As soon as I try the ath5k version, I'll
> report back.
Heh, I'm glad that I'm not the only one who missed that in ticket 1192.
Kevmf (who posted on that ticket) called it to my attention on Fedora
forums. It's towards the bottom where they state that it will probably
be in place for the 2.6.27 kernel.
In case anyone was wondering: http://madwifi.org/ticket/1192
--
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )
6:40 a.m.
2008/8/10 Scott Robbins <scottro(a)nyc.rr.com>:
On Sun, Aug 10, 2008 at 12:17:36AM +0100, Christopher Brown wrote:
> 2008/8/9 Scott Robbins <scottro(a)nyc.rr.com>:
> >
> > I don't use resume at all, so I'm not sure if it will fix the problem
> > you're having. It would be nice if it did. I'm wondering if this
ath5k
> > module is going to be backported.
>
> To what? It's in Fedora 9 already.
In Fedora 9, I'd been using the livna-testing repo's version of MadWifi.
Thanks for the information, I just tried the F9 version. However, it
doesn't work for me with that card, and I had to go back to the livna
madwifi.
Do you have any further information than "it doesn't work for me". Can
you file a bug with dmesg, lspci and lsmod output perhaps?
--
Christopher Brown
http://www.chruz.com
5:26 p.m.
--- On Sat, 8/9/08, Scott Robbins <scottro(a)nyc.rr.com> wrote:
From: Scott Robbins <scottro(a)nyc.rr.com>
Subject: Latest kernel, ath5k and Atheros5007EG
To: "For testers of Fedora Core development releases"
<fedora-test-list(a)redhat.com>
Date: Saturday, August 9, 2008, 4:45 PM
For a long time, the Atheros 5007EG hasn't worked with
the ath5k module,
and those of us who had it used MadWifi.
Today, I upgraded my kernel, and the latest special Madwifi
snapshot
wouldn't build. This has happened in the past, last
time, someone on
this list had submitted a patch to fix it.
Just for fun, I decided to try modprobing ath5k and seeing
if the card
might be recognized and work.
It seems that it does. This is running
2.6.27-0.244.rc2.git1.fc10.x86_64. I modprobed ath5k and
saw that I now
had a visible wlan0 card. Did ifconfig up and iwlist scan
and it saw
the networks. I then adjusted wpa_supplicant (which I use
rather than
NetworkManager--I use fluxbox most of the time and often
just boot into
runlevel 3) and Lo! and Behold! it worked.
I know that at least one other person on this list uses the
card. I'd
be interested if other folks with that card (The older EEE
PC uses it,
though I believe the 900 series uses a realink of some
sort, but it's
fairly popular in lower end laptops) are also having
success. Thanks
for any input.
To the kernel developers, thank you, great work.
--
Scott Robbins
Works with my cheap Atheros-based wireless card [Atheros
Communications Inc. AR2413 802.11bg NIC (rev 01)] too!
5761
days inactive
5762
days old
14 comments
5 participants
participants (5)
-
Arthur Pemberton -
Christopher Brown -
Rahul Sundaram -
Scott Robbins -
Steven I Usdansky