So what is happening with this bug? It seems to be the same thing that has broken what was perfectly working wifi on my laptop when I had Lenny installed. But upgrading to Squeeze so I could continue to get security updates has left me with no wireless connection unless I plug in an external device. It doesn't seem to have been marked as fixed, as it should not be if it is still present in the standard Squeeze kernel, but at the end of one of the responses it says a fix has been entered into the 2.6.36~rc5-1~experimental.1 kernel. When will this propagate to the regular kernel issued to users? If it worked perfectly in Lenny, could it be that hard to implement the fix? Can I get a patch somewhere to apply myself while waiting for this to appear in the standard kernel?

On Mon, 2012-05-07 at 14:51 -0700, Paul Zimmerman wrote:
> So what is happening with this bug? It seems to be the same thing that
> has broken what was perfectly working wifi on my laptop when I had
> Lenny installed. But upgrading to Squeeze so I could continue to get
> security updates has left me with no wireless connection unless I plug
> in an external device. It doesn't seem to have been marked as fixed,
> as it should not be if it is still present in the standard Squeeze
> kernel, but at the end of one of the responses it says a fix has been
> entered into the 2.6.36~rc5-1~experimental.1 kernel.
[...]

The bug log is at <http://bugs.debian.org/604613> and there is no claim
that it has been fixed.

Can you test Linux 3.2 from testing/unstable/squeeze-backports? Then we
can see whether it was fixed after Linux 2.6.32. and either report the
bug upstream or try to find that fix.

Ben.

--
Ben Hutchings
Life is what happens to you while you're busy making other plans.
- John Lennon

I can confirm that it works for me on linux 3.2.0 from squeeze-backports, while failing on 2.6.32 from squeeze.
--Robert Wallner

Ben Hutchings <ben@decadent.org.uk> wrote:

>On Mon, 2012-05-07 at 14:51 -0700, Paul Zimmerman wrote:
>> So what is happening with this bug? It seems to be the same thing that
>> has broken what was perfectly working wifi on my laptop when I had
>> Lenny installed. But upgrading to Squeeze so I could continue to get
>> security updates has left me with no wireless connection unless I plug
>> in an external device. It doesn't seem to have been marked as fixed,
>> as it should not be if it is still present in the standard Squeeze
>> kernel, but at the end of one of the responses it says a fix has been
>> entered into the 2.6.36~rc5-1~experimental.1 kernel.
[...]

>The bug log is at <http://bugs.debian.org/604613> and there is no claim
>that it has been fixed.
>
>Can you test Linux 3.2 from testing/unstable/squeeze-backports?* Then
>we can see whether it was fixed after Linux 2.6.32. and either report the
>bug upstream or try to find that fix.

When I first discovered on upgrading that the wifi was broken I tried the 3.2 backport
kernel. It couldn't finish loading, but froze with a black screen. If it has been updated
since early March I could try it again. But when I tried to install it again there was a
dependency problem that I don't recall from my first attempt. It says that the newer
kernel needs an backported initramfs as well. Was initramfs possibly involved in the
problem that caused it to not be able to load before? Would updating that for the
newer kernel make the current stable kernel not able to load? Not knowing for certain
that the backport would work, I would hate to get into a situation where I could not
revert without some major rescue effort.



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 1337220806.51639.YahooMailNeo@web162604.mail.bf1.y ahoo.com">http://lists.debian.org/1337220806.51639.YahooMailNeo@web162604.mail.bf1.y ahoo.com

Paul Zimmerman wrote:

(where "that" = initramfs-tools)
> Would updating that for the
> newer kernel make the current stable kernel not able to load?

No. initramfs-tools from wheezy and squeeze-backports are known to
work well with the kernel from squeeze.

Hope that helps,
Jonathan



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20120517052452.GB12275@burratino">http://lists.debian.org/20120517052452.GB12275@burratino

No go. I don't know how to tell if it has been updated since early March, but the 3.2 kernel does exactly
the same thing now that it did then. It looks like it is booting ok, but when it gets to loading the desktop
it freezes with a black screen. Apparently it runs into some sort of problem starting X on this system.

Admittedly, this is a marginal system. It's a laptop with the VIA P4M800 Pro integrated chipset. The
video system uses the OpenChrome drivers, which actually work more reliably than the Windows drivers.
But it is still buggy. Since I have no idea how to make it work, I have uninstalled the backported kernel,
and tried to uninstall the supporting packages but they seem to have disappeared. Both initramfs and
linux-base had to be upgraded to backports for the 3.2 kernel, but now I cannot find the upgraded
versions to uninstall them. Why would they turn invisible? At least the standard kernel is able to boot
and the system is usable.

For now I have installed a different wifi card I found on eBay with a Realtek chipset. This does work, so
the system is otherwise functional. It's just the ath5k driver that is broken. The performance of this other
card is noticeably less, though, so I am definitely watching and waiting for this driver to be fixed in the
stable kernel for Squeeze.



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 1337375379.43431.YahooMailNeo@web162602.mail.bf1.y ahoo.com">http://lists.debian.org/1337375379.43431.YahooMailNeo@web162602.mail.bf1.y ahoo.com

Hi Paul,

Paul Zimmerman wrote:

> No go. I don't know how to tell if it has been updated since early
> March, but the 3.2 kernel does exactly the same thing now that it
> did then. It looks like it is booting ok, but when it gets to
> loading the desktop it freezes with a black screen. Apparently it
> runs into some sort of problem starting X on this system.

Is the machine completely frozen? Does caps lock work? Does the
magic sysrq key work? Can you access the machine using SSH via a
wired or wireless connection? Can you get logs from this event, for
example by using netconsole or serial console[1] or examining
/var/log/dmesg, /var/log/Xorg.0.log, and /var/log/syslog after a
reboot?

[1] http://kernel-handbook.alioth.debian.org/ch-bugs.html#s9.1.1



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20120518215108.GA13925@burratino">http://lists.debian.org/20120518215108.GA13925@burratino

Jonathan Nieder <jrnieder@gmail.com> wrote:

>Paul Zimmerman wrote:

>> No go. I don't know how to tell if it has been updated since early
>> March, but the 3.2 kernel does exactly the same thing now that it
>> did then. It looks like it is booting ok, but when it gets to
>> loading the desktop it freezes with a black screen. Apparently it
>> runs into some sort of problem starting X on this system.

>Is the machine completely frozen?* Does caps lock work?* Does the
>magic sysrq key work?* Can you access the machine using SSH via a
>wired or wireless connection?* Can you get logs from this event, for
>example by using netconsole or serial console[1] or examining
>/var/log/dmesg, /var/log/Xorg.0.log, and /var/log/syslog after a
.reboot?

>[1] http://kernel-handbook.alioth.debian.org/ch-bugs.html#s9.1.1

I admit I did not try these other options. If it wasn't going to boot to the desktop,
it was hung and I just did a hard power-off and selected the old kernel from
the GRUB menu when restarting. After removing the backported kernel I used
"shutdown -Fr now" to make sure no major problems were left in the file system.

It's fairly simple to go back and look at dmesg, kern.log, syslog and such. The Xorg.log
would be gone by now, after several restarts. Or would it? I'm not sure how many
iterations are kept as .z archives. But I'm not sure if I could figure out how to use
netconsole or serial console. It has to be connected to another machine running Linux,
doesn't it?

Is there some other bug page for this that I should send any reports to?




--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 1337391018.28403.YahooMailNeo@web162605.mail.bf1.y ahoo.com">http://lists.debian.org/1337391018.28403.YahooMailNeo@web162605.mail.bf1.y ahoo.com

[Please always cc the bug address, i.e. reply to all.]

On Sun, 2012-05-13 at 14:48 -0700, Paul Zimmerman wrote:
>
> On Monday, May 7, 2012 7:11 PM, Ben Hutchings <ben@decadent.org.uk> wrote:
>
> >On Mon, 2012-05-07 at 14:51 -0700, Paul Zimmerman wrote:
> >> So what is happening with this bug? It seems to be the same thing that
> >> has broken what was perfectly working wifi on my laptop when I had
> >> Lenny installed. But upgrading to Squeeze so I could continue to get
> >> security updates has left me with no wireless connection unless I plug
> >> in an external device. It doesn't seem to have been marked as fixed,
> >> as it should not be if it is still present in the standard Squeeze
> >> kernel, but at the end of one of the responses it says a fix has been
> >> entered into the 2.6.36~rc5-1~experimental.1 kernel.
> >[...]
>
> >The bug log is at <http://bugs.debian.org/604613> and there is no claim
> >that it has been fixed.
>
> >Can you test Linux 3.2 from testing/unstable/squeeze-backports? Then we
> >can see whether it was fixed after Linux 2.6.32. and either report the
> >bug upstream or try to find that fix.
>
> Well, I don't claim to understand the system, but in a page linked on this one (http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=11;bug=604613) it gives a series of "reassigned" reports, ending with:
>
> reassign 586995 linux-2.6 2.6.32-24
> found 586995 linux-2.6/2.6.32-24
> fixed 586995 2.6.36~rc5-1~experimental.1
> thanks
>
> So it looks like the work on the bug was bounced around to various places,
> perhaps because the same problem was causing more than one bug report.
> The "fixed" line at the end would seem to mean that it was resolved, but
> put into the experimental kernel rather than the stable one issued to users.

I reassigned a large number of bugs with a single control message, one
of which happened to be this one. I also marked another bug fixed.

> As for the backported kernel, I did try that when I first upgraded and found the
> wifi was broken. It would not finish loading, but froze with a black screen when
> it tried to start the desktop. It seems to be a little _too_ unstable for me at this
>
> time. So I simply uninstalled it. But if it has been updated since early March,
> I can try it again and let you know if it works now.

It's updated almost every week in unstable. I can't promise the
graphics bug has been fixed though.

Ben.

--
Ben Hutchings
Experience is directly proportional to the value of equipment destroyed.
- Carolyn Scheppner