[ath9k-devel] [Fwd: Re: still same problem]

[ath9k-devel] [Fwd: Re: still same problem]

[Dmitri Seletski]

Hello Luis.

I am not looking for patch even!

Last time for sake of clarity I have unloaded nvidia module(obviously 
successfully, since no X were runing, GDM and gnome for that matter were 
killed before)

I use gentoo, as such there are no convenient ways of getting all those 
fancy wireless tools, without EXPLICITLY installing them. i didn't 
install anything anyway. Same happens with WAP, none of modules were 
ever loaded. So it's 100% nothing to do with gnome/kde.
Latest ones I have tried was yesterday.

i am somewhat familiar with tools like "ip", "iwconfig", "iwlist", 
"emacs". So I can very well live without graphical environment 
whatsoever, which I actually do for testing matter.

No matter if I am under X or not at all - still computer freezes without 
any error messages. Metalog doesnt have anything in logs either, so it 
looks like instant issue!

I can notice more things, that may somewhat help youbut i am not sure if 
it's any helpfull.

Please do instruct me on what you want me to do. I AM WILLING to do 
anything you ask me. BTW, i am using vesa for video in case if nvidia is 
down, does that matter?

Regards

P.S.: my module list below as well as pstree output

lsmod
Module Size Used by
cdc_acm 17824 0
isofs 35048 1
zlib_inflate 16576 1 isofs
it87 26776 0
w83791d 21788 0
hwmon_vid 3712 2 it87,w83791d
snd_seq_dummy 3268 0
snd_seq_oss 33472 0
snd_seq_midi_event 7488 1 snd_seq_oss
snd_seq 56032 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi_event
snd_seq_device 7252 3 snd_seq_dummy,snd_seq_oss,snd_seq
snd_pcm_oss 43264 0
snd_mixer_oss 16896 1 snd_pcm_oss
loop 17228 0
cpufreq_conservative 7880 0
cpufreq_ondemand 8016 1
cpufreq_powersave 2304 0
cpufreq_performance 2304 0
powernow_k8 17604 1
freq_table 5120 2 cpufreq_ondemand,powernow_k8
parport_pc 33992 0
parport 25056 1 parport_pc
vfat 12352 0
fat 52984 1 vfat
i2c_dev 8400 0
udf 83624 0
cifs 267544 0
ohci1394 31540 0
ieee1394 89024 1 ohci1394
nvidia 7805288 66
ath9k 254576 0
snd_hda_intel 460396 4
snd_pcm 80072 3 snd_pcm_oss,snd_hda_intel
mac80211 235520 1 ath9k
snd_timer 22608 3 snd_seq,snd_pcm
snd 62664 14 
snd_seq_oss,snd_seq,snd_seq_device,snd_pcm_oss,snd_mixer_oss,snd_hda_intel,snd_pcm,snd_timer
ehci_hcd 53784 0
cfg80211 42032 1 mac80211
soundcore 7648 1 snd
pcspkr 2944 0
firewire_ohci 24068 0
usbhid 32032 0
firewire_core 40416 1 firewire_ohci
forcedeth 55568 0
sata_sil24 15492 0
thermal 18656 0
snd_page_alloc 8720 2 snd_hda_intel,snd_pcm
i2c_nforce2 7168 0
k8temp 5184 0
processor 38520 2 powernow_k8,thermal
button 7520 0


Kernel is manually compiled, as well as all modules are.


pstree
init???6*[agetty]
??audacious???5*[{audacious}]
??bonobo-activati???{bonobo-activati}
??cpufreq-applet
??2*[dbus-daemon]
??dbus-launch
??drivemount_appl
??fcron
??firefox???6*[{firefox}]
??gconfd-2
??gdm???gdm???X
? ??gnome-session???gnome-panel
? ??metacity
? ??nautilus
? ??seahorse-agent
? ??ssh-agent
? ??{gnome-session}
??gedit
??gnome-dictionar
??gnome-keyboard-
??gnome-keyring-d
??gnome-power-man
??gnome-screensav
??gnome-settings-???{gnome-settings-}
??gnome-terminal???bash???pstree
? ??bash
? ??gnome-pty-helpe
? ??{gnome-terminal}
??gnome-vfs-daemo
??gnome-volume-ma
??gvfsd
??gvfsd-burn
??gvfsd-computer
??gvfsd-trash
??hald???hald-runner???hald-addon-acpi
? ??hald-addon-cpuf
? ??hald-addon-inpu
? ??2*[hald-addon-stor]
??mdadm
??metalog???metalog
??mozilla-launche???thunderbird-bin???11*[{thunderbird-bin}]
??multiload-apple
??sensors-applet
??skype???7*[{skype}]
??smartd
??udevd
??xgame-gtk2???startx???xinit???X
??xterm???bash

-------------- next part --------------
An embedded message was scrubbed...
From: "Luis R. Rodriguez" <mcgrof@gmail.com>
Subject: Re: [ath9k-devel] still same problem
Date: Sat, 3 Jan 2009 19:34:26 -0800
Size: 5421
Url: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20090104/8dd59a1e/attachment.eml 

[ath9k-devel] [Fwd: Re: still same problem]

[Luis R. Rodriguez]

2009/1/4 Dmitri Seletski <drjoms@gmail.com>:
> Hello Luis.
>
> I am not looking for patch even!
>
> Last time for sake of clarity I have unloaded nvidia module(obviously
> successfully, since no X were runing, GDM and gnome for that matter were
> killed before)
>
> I use gentoo, as such there are no convenient ways of getting all those
> fancy wireless tools, without EXPLICITLY installing them. i didn't install
> anything anyway. Same happens with WAP, none of modules were ever loaded. So
> it's 100% nothing to do with gnome/kde.
> Latest ones I have tried was yesterday.
>
> i am somewhat familiar with tools like "ip", "iwconfig", "iwlist", "emacs".
> So I can very well live without graphical environment whatsoever, which I
> actually do for testing matter.
>
> No matter if I am under X or not at all - still computer freezes without any
> error messages. Metalog doesnt have anything in logs either, so it looks
> like instant issue!
>
> I can notice more things, that may somewhat help youbut i am not sure if
> it's any helpfull.
>
> Please do instruct me on what you want me to do. I AM WILLING to do anything
> you ask me.

Since you are not getting anything on the virtual terminal please try
enabling NMI watchdog and see if that is able to generate an oops
after 5 seconds from when the system hangs. To enable the NMI watchdog
you can use:

nmi_watchdog=1

This assumes you have an IO-APIC (you can cat /proc/interrupts to see
if you have one). If you don't have one use:

nmi_watchdog=2

For more information please see Documentation/nmi_watchdog.txt from
the kernel sources.

If this doesn't work you can perhaps try a netconsole, which you can
read about in Documentation/networking/netconsole.txt.

> BTW, i am using vesa for video in case if nvidia is down, does
> that matter?

Yeah keep proprietary drivers away from your kernel when debugging
issues otherwise your issues can be ignored by some developers. So
using vesa is a good alternative.

  Luis

[ath9k-devel] Fwd: [Fwd: Re: still same problem]

[Luis R. Rodriguez]

---------- Forwarded message ----------
From: Dmitri Seletski <drjoms@gmail.com>
Date: Mon, Jan 5, 2009 at 1:02 PM
Subject: Re: [ath9k-devel] [Fwd: Re: still same problem]
To: "Luis R. Rodriguez" <mcgrof@gmail.com>


Luis R. Rodriguez wrote:
>
> 2009/1/4 Dmitri Seletski <drjoms@gmail.com>:
>
>>
>> Hello Luis.
>>
>> I am not looking for patch even!
>>
>> Last time for sake of clarity I have unloaded nvidia module(obviously
>> successfully, since no X were runing, GDM and gnome for that matter were
>> killed before)
>>
>> I use gentoo, as such there are no convenient ways of getting all those
>> fancy wireless tools, without EXPLICITLY installing them. i didn't install
>> anything anyway. Same happens with WAP, none of modules were ever loaded. So
>> it's 100% nothing to do with gnome/kde.
>> Latest ones I have tried was yesterday.
>>
>> i am somewhat familiar with tools like "ip", "iwconfig", "iwlist", "emacs".
>> So I can very well live without graphical environment whatsoever, which I
>> actually do for testing matter.
>>
>> No matter if I am under X or not at all - still computer freezes without any
>> error messages. Metalog doesnt have anything in logs either, so it looks
>> like instant issue!
>>
>> I can notice more things, that may somewhat help youbut i am not sure if
>> it's any helpfull.
>>
>> Please do instruct me on what you want me to do. I AM WILLING to do anything
>> you ask me.
>>
>
> Since you are not getting anything on the virtual terminal please try
> enabling NMI watchdog and see if that is able to generate an oops
> after 5 seconds from when the system hangs. To enable the NMI watchdog
> you can use:
>
> nmi_watchdog=1
>
> This assumes you have an IO-APIC (you can cat /proc/interrupts to see
> if you have one). If you don't have one use:
>
> nmi_watchdog=2
>
> For more information please see Documentation/nmi_watchdog.txt from
> the kernel sources.
>
> If this doesn't work you can perhaps try a netconsole, which you can
> read about in Documentation/networking/netconsole.txt.
>
>
>>
>> BTW, i am using vesa for video in case if nvidia is down, does
>> that matter?
>>
>
> Yeah keep proprietary drivers away from your kernel when debugging
> issues otherwise your issues can be ignored by some developers. So
> using vesa is a good alternative.
>
>  Luis
>

Hello Luis.

Your advice seemed to help!

Upgrade to 2.27.9 seemed to help!

I will use it, for a while, if no news from me until beginning of new
week - we are good :)

Alternatively, if i come across something, i will let you guys know.

Again, i am glad to help with testing features/debugging stuff. Do ask
me anything you want.

Regards
P.S.: sorry for being naughty at times.