Bug#673496: [lenny -> squeeze regression] wifi broken (?)
>From: Jonathan Nieder <jrnieder@gmail.com>
>To: Paul Zimmerman <aiwanar@yahoo.com>
>Cc: 673496@bugs.debian.org
>Sent: Saturday, July 14, 2012 11:44 AM
>Subject: Re: [lenny -> squeeze regression] wifi broken (?)
>Jonathan Nieder wrote:
>> Paul Zimmerman wrote:
>>> Trying
>>> to use the backported kernel from Wheezy runs into the freeze-up
>>> problem I described.
>>
>> Were you able to boot a 3.x.y kernel in recovery mode?
>Ping. Are you still interested in pursuing this bug? (If you're just
>busy and will get to it in a few weeks or so, that's fine. And if
>you're not interested in it, that's fine, too, but please do let us
>know so we can plan accordingly.)
Well, as I said, it's working now. When I was first trying to figure out the
problem, I started by replying to bug 604613 because it seemed to be
the same symptoms I was having with my system: it had worked with
Lenny, then was unable to get an IP address with the Squeeze kernel.
But sometime between my install of Squeeze and the attempt to use the
backported kernel, it was resolved. Since you say the ath5k drivers have
not been updated I can only presume that something else was wrong,
and it was fixed in one of the updates related to or accompanying the
wicd security fix. Since that bug is no longer occurring for me, I don't
know how I can contribute anything more to this particular bug discussion.
The backported kernel is a completely different problem, which I don't
have to deal with if I don't need that kernel. :D Yes, it did boot in "recovery"
mode on my system. But it apparently did not boot correctly, because the
wifi card was not visible to the command line interface utilities. Even worse,
simply restarting and booting to the current Squeeze kernel was not enough
to fix whatever it did wrong with the hardware on boot. The graphical desktop
could not load until I did a hard shutdown and then restart. Whether this is
specific to GDM or perhaps would also happen with KDM I don't know. But
some registers were altered or some bit set somewhere that caused major
problems even when the backported kernel itself was no longer in memory.
Admittedly, when Wheezy becomes the stable release it will BE a problem
if this has not been fixed. But I don't know what I can do to test this weird
malfunction. It seems to be a a very low hardware level. Would there be
some sort of debugging utility that could read all the hardware registers and
log the results? Is anyone else reporting a problem like this with the Wheezy
kernel? Either in its backported form or in its "native" setting as part of the
Wheezy system?
Reply to: