Re: Catching NForce2 lockup with NMI watchdog - found?

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1601 bytes --]

Craig Bradney wrote:
> Sounds great.. maybe you have come across something. Yes, the CPU
> Disconnect function arrived in your BIOS in revision of 2003/03/27
> "6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset
> does not support C2 disconnect; thus, disable C2 function."

I doubt thats related, i run ACPI with powersave anyways... 

> For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can
> see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS
> that has been discussed.

I don't have it either... 

I'm more hopeful about the patch from Mathieu <cheuche+lkml () free ! fr>...

           CPU0
  0:     267486    IO-APIC-edge  timer
  1:       9654    IO-APIC-edge  keyboard
  2:          0          XT-PIC  cascade
  8:          1    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 14:      28252    IO-APIC-edge  ide0
 15:        103    IO-APIC-edge  ide1
 16:     251712   IO-APIC-level  eth0
 17:      90632   IO-APIC-level  EMU10K1
 19:     415529   IO-APIC-level  nvidia
 20:          0   IO-APIC-level  usb-ohci
 21:        153   IO-APIC-level  ehci_hcd
 22:      58257   IO-APIC-level  usb-ohci
NMI:        479
LOC:     265875
ERR:          0
MIS:          0

this far and it feels like a closer match to what windows does from what
i have read on the ml. 

I haven't even come close to testing this yet, I've only been up 45 mins
but i'll leave it running and do what i usually do when it hangs... =)

I'll get back to you about how it goes... 

-- 
Ian Kumlien <pomac@vapor.com>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

Ok.. I decided I would try the patch out.. here are the results:

Interrupts before patch: (booted with nmi-watchdog=2)

           CPU0
  0:    2863919          XT-PIC  timer
  1:       7851    IO-APIC-edge  i8042
  2:          0          XT-PIC  cascade
  8:          2    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
12:     104453    IO-APIC-edge  i8042
14:      26951    IO-APIC-edge  ide0
15:       8490    IO-APIC-edge  ide1
19:     211976   IO-APIC-level  radeon@PCI:3:0:0
21:      17303   IO-APIC-level  ehci_hcd, NVidia nForce2, eth0
22:          3   IO-APIC-level  ohci1394
NMI:          0
LOC:    2863756
ERR:          0
MIS:          0


Interrupts after patch: (booted with nmi-watchdog=2)
           CPU0
  0:     435688    IO-APIC-edge  timer
  1:       2714    IO-APIC-edge  i8042
  2:          0          XT-PIC  cascade
  8:          2    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
12:       5490    IO-APIC-edge  i8042
14:      16831    IO-APIC-edge  ide0
15:        392    IO-APIC-edge  ide1
19:       9931   IO-APIC-level  radeon@PCI:3:0:0
21:       3980   IO-APIC-level  ehci_hcd, NVidia nForce2, eth0
22:          3   IO-APIC-level  ohci1394
NMI:          0
LOC:     435246
ERR:          0
MIS:          0

All the interrupts are the same...except:
0, timer is now IO-APIC-edge.

Im not getting any NMI counts.. should I use nmi-watchdog=1?


dmesg differences:

1.
after:
..TIMER: vector=0x31 pin1=2 pin2=0

before:
..TIMER: vector=0x31 pin1=2 pin2=-1

2.
after:
...trying to set up timer (IRQ0) through the 8259A ...
..... (found pin 0) ...works.
number of MP IRQ sources: 16.

before:
...trying to set up timer (IRQ0) through the 8259A ...  failed.
...trying to set up timer as Virtual Wire IRQ... failed.
...trying to set up timer as ExtINT IRQ... works.
number of MP IRQ sources: 15.


Ian, from looking back, you have an A7N8X-X bios 1007.
Interesting that my USB hcis are still sharing IRQs there.
Any idea how I can get them apart, or if I should try.
My system was pretty stable as I've stated.. but the patch has changed
things slightly re the timer.

I've run hdparm about 25 times... and as its almost 3am I'm going to let
this thing idle for a few hours like this.

Craig






On Sat, 2003-12-06 at 01:55, Ian Kumlien wrote: 
> Craig Bradney wrote:
> > Sounds great.. maybe you have come across something. Yes, the CPU
> > Disconnect function arrived in your BIOS in revision of 2003/03/27
> > "6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset
> > does not support C2 disconnect; thus, disable C2 function."
> 
> I doubt thats related, i run ACPI with powersave anyways... 
> 
> > For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can
> > see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS
> > that has been discussed.
> 
> I don't have it either... 
> 
> I'm more hopeful about the patch from Mathieu <cheuche+lkml () free ! fr>...
> 
>            CPU0
>   0:     267486    IO-APIC-edge  timer
>   1:       9654    IO-APIC-edge  keyboard
>   2:          0          XT-PIC  cascade
>   8:          1    IO-APIC-edge  rtc
>   9:          0   IO-APIC-level  acpi
>  14:      28252    IO-APIC-edge  ide0
>  15:        103    IO-APIC-edge  ide1
>  16:     251712   IO-APIC-level  eth0
>  17:      90632   IO-APIC-level  EMU10K1
>  19:     415529   IO-APIC-level  nvidia
>  20:          0   IO-APIC-level  usb-ohci
>  21:        153   IO-APIC-level  ehci_hcd
>  22:      58257   IO-APIC-level  usb-ohci
> NMI:        479
> LOC:     265875
> ERR:          0
> MIS:          0
> 
> this far and it feels like a closer match to what windows does from what
> i have read on the ml. 
> 
> I haven't even come close to testing this yet, I've only been up 45 mins
> but i'll leave it running and do what i usually do when it hangs... =)
> 
> I'll get back to you about how it goes... 

Re: Catching NForce2 lockup with NMI watchdog - found?

[Mike Fedyk]

On Sat, Dec 06, 2003 at 02:52:40AM +0100, Craig Bradney wrote:
> Im not getting any NMI counts.. should I use nmi-watchdog=1?

Yes, try with nmi-watchdog=1 it should work with apic, or io-apic (or both)
enabled.

Re: Catching NForce2 lockup with NMI watchdog - found?

[Josh McKinney]

On approximately Sat, Dec 06, 2003 at 02:52:40AM +0100, Craig Bradney wrote:
> Ok.. I decided I would try the patch out.. here are the results:
> 
[snip]

I tried it out too since doing some greps after 3 days uptime crashed
it again.  Guess I am not so lucky.  So far so good with the new
patch, timer is using apic and no crashes yet, after 30 minutes of 
grep's and hdparms.  Tried booting with nmi_watchdog=1 and got this:

...trying to set up timer (IRQ0) through the 8259A ...
..... (found pin 0) ...works.
activating NMI Watchdog ... done.
testing NMI watchdog ... CPU#0: NMI appears to be stuck!

Going to retry with nmi_watchdog=2 and see if that works.

> 
> Im not getting any NMI counts.. should I use nmi-watchdog=1?
> 
> 
[snip]

-- 
Josh McKinney		     |	Webmaster: http://joshandangie.org
--------------------------------------------------------------------------
                             | They that can give up essential liberty
Linux, the choice       -o)  | to obtain a little temporary safety deserve 
of the GNU generation    /\  | neither liberty or safety. 
                        _\_v |                          -Benjamin Franklin

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bob]

Ian Kumlien wrote:

>           CPU0
>  0:     267486    IO-APIC-edge  timer
>  1:       9654    IO-APIC-edge  keyboard
>  2:          0          XT-PIC  cascade
>  8:          1    IO-APIC-edge  rtc
>  9:          0   IO-APIC-level  acpi
> 14:      28252    IO-APIC-edge  ide0
> 15:        103    IO-APIC-edge  ide1
> 16:     251712   IO-APIC-level  eth0
> 17:      90632   IO-APIC-level  EMU10K1
> 19:     415529   IO-APIC-level  nvidia
> 20:          0   IO-APIC-level  usb-ohci
> 21:        153   IO-APIC-level  ehci_hcd
> 22:      58257   IO-APIC-level  usb-ohci
>NMI:        479
>LOC:     265875
>ERR:          0
>MIS:          0
>
note I have only PIC timer....

           CPU0      
  0:  244393560          XT-PIC  timer
  1:      31963    IO-APIC-edge  i8042
  2:          0          XT-PIC  cascade
  8:          1    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 12:     251884    IO-APIC-edge  i8042
 14:         22    IO-APIC-edge  ide0
 15:         24    IO-APIC-edge  ide1
 16:    4290216   IO-APIC-level  3ware Storage Controller, yenta, yenta
 17:    5929405   IO-APIC-level  eth0
 21:          0   IO-APIC-level  NVidia nForce2
NMI:          0
LOC:  244378698
ERR:          0
MIS:          6

RE: merged in bk5 Re: Catching NForce2 lockup with NMI watchdog - found?

[b]

Is this stuff going to be merged into 2.4 soon? I'd like
to try a 2.4.23/24-bk with these patches.


 >From: Bob
 >Subject: merged in bk5 Re: Catching NForce2 lockup with NMI
 >
 >if you're following this thread, good news--
 >
 >nforce2 fixups have been merged in
 >linux-2.6.0-test11-bk5.patch
 >>  -bk snapshot (patch-2.6.0-test11-bk5)
 >
 >nforce2-disconnect-quirk.patch
 >>  [x86] fix lockups with APIC support on nForce2
 >>
 >>nforce2-apic.patch
 >>  [x86] do not wrongly override mp_ExtINT IRQ
 >
 >plus promise and sis fixes so I don't need to pay
 >for a 3ware controller ;-)   that was another
 >show-stopper for me earlier
 >
 >> We're all trying to get acpi, apic, lapic, io-apic working
 >> when turned on in cmos/bios and kernel.
 >>

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bob]

b@netzentry.com wrote:

> Is this stuff going to be merged into 2.4 soon? I'd like
> to try a 2.4.23/24-bk with these patches.

Care to try working with these for 2.4.22?

http://marc.theaimsgroup.com/?l=linux-kernel&m=107080280512734&w=2

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ross Dickson]

On Monday 08 of December 2003 04:08, Bob wrote: 
 > >>Sounds great.. maybe you have come across something. Yes, the CPU 
 > >>Disconnect function arrived in your BIOS in revision of 2003/03/27 
 > >>"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset 
 > >>does not support C2 disconnect; thus, disable C2 function." 
 > >> 
 > >>For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can 
 > >>see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS 
 > >>that has been discussed. 
 > >> 
 > >>Craig 
 > >
 > >I don't have that in MSI K7N2 MCP2-T near the 
 > >agp and fsb spread spectrum items or anywhere 
 >> else. 
>Use athcool: 
>         http://members.jcom.home.ne.jp/jacobi/linux/softwares.html#athcool 
> or apply kernel patch (2.4 and 2.6 versions were posted already). 
>--bart 

Please take a look at 

Fixes for nforce2 hard lockup, apic, io-apic, udma133 covered

in mailing list.

I approached it from another angle regarding delaying the apic ack in local timer irq
and achieved stability. It would be good to have others try it. Ian Kumlien is also
reporting success so far.
 

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Mon, 2003-12-08 at 04:21, Ross Dickson wrote:
> On Monday 08 of December 2003 04:08, Bob wrote: 
>  > >>Sounds great.. maybe you have come across something. Yes, the CPU 
>  > >>Disconnect function arrived in your BIOS in revision of 2003/03/27 
>  > >>"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset 
>  > >>does not support C2 disconnect; thus, disable C2 function." 
>  > >> 
>  > >>For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can 
>  > >>see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS 
>  > >>that has been discussed. 
>  > >> 
>  > >>Craig 
>  > >
>  > >I don't have that in MSI K7N2 MCP2-T near the 
>  > >agp and fsb spread spectrum items or anywhere 
>  >> else. 
> >Use athcool: 
> >         http://members.jcom.home.ne.jp/jacobi/linux/softwares.html#athcool
> > or apply kernel patch (2.4 and 2.6 versions were posted already). 
> >--bart 
> 
> Please take a look at 
> 
> Fixes for nforce2 hard lockup, apic, io-apic, udma133 covered
> 
> in mailing list.
> 
> I approached it from another angle regarding delaying the apic ack in local timer irq
> and achieved stability. It would be good to have others try it. Ian Kumlien is also
> reporting success so far.
>  

Although I had long uptimes before.. and therefore might achieve them
again fairly easily.. I'm now on 2 days 10 hours which has included a
lot of compilation and a lot of idle time, and plenty of the hdpar and
grep tests. I have used only the IRQ0 IO-APIC edge patch.

Can someone please note all the patches for 2.6 that people have tried
and what they achieve? Im starting to get a bit lost, given the fact
that I'm running stable here with only 1 patch. (so far - this is where
it crashes after I click Send I suppose ;) )

-apic

-io-apic (IRQO set to XT-PIC incorrectly)

-udma133?

-cpu disconnect patch (missing bios option for ACPI Cx states)

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ross Dickson]

On Monday 08 December 2003 21:36, Craig Bradney wrote:
> On Mon, 2003-12-08 at 04:21, Ross Dickson wrote:
> > On Monday 08 of December 2003 04:08, Bob wrote: 
> >  > >>Sounds great.. maybe you have come across something. Yes, the CPU 
> >  > >>Disconnect function arrived in your BIOS in revision of 2003/03/27 
> >  > >>"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset 
> >  > >>does not support C2 disconnect; thus, disable C2 function." 
> >  > >> 
> >  > >>For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can 
> >  > >>see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS 
> >  > >>that has been discussed. 
> >  > >> 
> >  > >>Craig 
> >  > >
> >  > >I don't have that in MSI K7N2 MCP2-T near the 
> >  > >agp and fsb spread spectrum items or anywhere 
> >  >> else. 
> > >Use athcool: 
> > >         http://members.jcom.home.ne.jp/jacobi/linux/softwares.html#athcool
> > > or apply kernel patch (2.4 and 2.6 versions were posted already). 
> > >--bart 
> > 
> > Please take a look at 
> > 
> > Fixes for nforce2 hard lockup, apic, io-apic, udma133 covered
> > 
> > in mailing list.
> > 
> > I approached it from another angle regarding delaying the apic ack in local timer irq
> > and achieved stability. It would be good to have others try it. Ian Kumlien is also
> > reporting success so far.
> >  
> 
> Although I had long uptimes before.. and therefore might achieve them
> again fairly easily.. I'm now on 2 days 10 hours which has included a
> lot of compilation and a lot of idle time, and plenty of the hdpar and
> grep tests. I have used only the IRQ0 IO-APIC edge patch.
> 
> Can someone please note all the patches for 2.6 that people have tried
> and what they achieve? Im starting to get a bit lost, given the fact
> that I'm running stable here with only 1 patch. (so far - this is where
> it crashes after I click Send I suppose ;) )
> 
> -apic
> 
> -io-apic (IRQO set to XT-PIC incorrectly)
> 
> -udma133?
> 
> -cpu disconnect patch (missing bios option for ACPI Cx states)
> 
> Craig
> 
> 
> 
> 
Hi Craig
Here is a link to my original posting

http://linux.derkeiler.com/Mailing-Lists/Kernel/2003-12/1528.html

I have not been working with 2.6, only 2.4.23 and 2.4.22.
My work has been independent of the cpu disconnect function and mpparse
patches spoken of in this posting thread. I think my work may have utility
within the 2.6 environment hence my posting to this thread. 

This followup with Ian may shed some more light.

http://linux.derkeiler.com/Mailing-Lists/Kernel/2003-12/1648.html

My test system has also been running ioapic for days now and I have stress
loaded it at times with gigabyte file copies alongside mplayer playing movies in 
three windows at the same time. It feels good to see the hardware performing
well. My motivation was that I have a digital imaging application where one
of the PCI cards performs poorly with shared interrupts so I had to fix the
ioapic mode. At one stage I got desperate and went back to xtpic mode.
I tried reprogramming the pirq routing registers to get extra pci irqs other 
than just irq11 from the AMD768 docs but of course the nforce2 is different
and it didn't work.

Referring to my posting the stability key I found is the (a) apic ack delay. 
I actually implemented the (b) ioapic timer edge first but my system was 
unstable still. Part (c) the udma133 was a cleanup along the way when I
thought the problem had to do with ide interface timings. It is probably not
a required part of the solution but it was there so I thought I would throw
it into the ring, in fact looking back I wonder if part (a) the apic ack delay 
is enough on its own to get rid of the lockups?

I do not know why the apic timing delay stablises the system.
Perhaps with the bios an SMI event occurs to do with the cpu disconnect 
function? and the apic ack gets lost? I do not know? but hopefully someone
can get to the bottom of it. The concept of it being bios dependent is perhaps
not so strange. I found this posting regarding a bad SMI during my research.

http://www.ussg.iu.edu/hypermail/linux/kernel/0203.2/0698.html

I have really found this flying in the dark with respect to hardware docs
very frustrating. I do not have the appropriate AMD cpu nor the nForce2 docs 
to be able to delve much deeper into the cause. Heck I even dragged out my 
old Intel "Microsystem Components Handbook Volume 1 1984" to look up detail on 
the 8259 auto end of interrupt mode used in the virtual wire mode. I had never seen
anyone use it before.

I hope that whichever solutions best fix the problems end up in the kernel tree 
for all our benefit. This has cost me 2 weeks of little sleep to get this far and being 
in small business it is also two weeks of unpaid time. I wish the hardware industry
had an open documents philosophy as good as linux has open source.

Regards
Ross.

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bob]

We're all trying to get acpi, apic, lapic, io-apic working
when turned on in cmos/bios and kernel.

The three things that each alone have achieved stability
on somebody's system here are 1) bios update 2) cpu
disconnect off either in cmos if available or by athcool
or kernel patch with same 3) timing delay patch

For CPU disconnect you still need athcool or this one
http://www.kernel.org/pub/linux/kernel/people/bart/2.6.0-test11-bart1/broken-out/nforce2-disconnect-quirk.patch

Both patches are for 2.6.0-test11 kernel.

turn on ioapic edge timer--

http://www.kernel.org/pub/linux/kernel/people/bart/2.6.0-test11-bart1/broken-out/nforce2-apic.patch


Other changes offer clues and expose symptoms but
are not helpful or necessary after da fix is in. udma
settings are a kludge of a error symptom, just aspirin.

Other kludges are acpi off, local apic off in kernel,
apic off in cmos/bios. These go away when the real
problem is fixed.

-Bob

Craig Bradney wrote:

>On Mon, 2003-12-08 at 04:21, Ross Dickson wrote:
>  
>
>>On Monday 08 of December 2003 04:08, Bob wrote: 
>> > >>Sounds great.. maybe you have come across something. Yes, the CPU 
>> > >>Disconnect function arrived in your BIOS in revision of 2003/03/27 
>> > >>"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset 
>> > >>does not support C2 disconnect; thus, disable C2 function." 
>> > >> 
>> > >>For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can 
>> > >>see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS 
>> > >>that has been discussed. 
>> > >> 
>> > >>Craig 
>> > >
>> > >I don't have that in MSI K7N2 MCP2-T near the 
>> > >agp and fsb spread spectrum items or anywhere 
>> >> else. 
>>    
>>
>>>Use athcool: 
>>>        http://members.jcom.home.ne.jp/jacobi/linux/softwares.html#athcool
>>>or apply kernel patch (2.4 and 2.6 versions were posted already). 
>>>--bart 
>>>      
>>>
>>Please take a look at 
>>
>>Fixes for nforce2 hard lockup, apic, io-apic, udma133 covered
>>
>>in mailing list.
>>
>>I approached it from another angle regarding delaying the apic ack in local timer irq
>>and achieved stability. It would be good to have others try it. Ian Kumlien is also
>>reporting success so far.
>> 
>>    
>>
>
>Although I had long uptimes before.. and therefore might achieve them
>again fairly easily.. I'm now on 2 days 10 hours which has included a
>lot of compilation and a lot of idle time, and plenty of the hdpar and
>grep tests. I have used only the IRQ0 IO-APIC edge patch.
>
>Can someone please note all the patches for 2.6 that people have tried
>and what they achieve? Im starting to get a bit lost, given the fact
>that I'm running stable here with only 1 patch. (so far - this is where
>it crashes after I click Send I suppose ;) )
>
>-apic
>  
>
   -local apic("lapic")

   -acpi

kernel local apic issue and acpi and apic go together,
but turning off lapic first might achieve stability for
some, updating bios will enable using all bios and
linux apic, acpi, lapic, ioapic for some(me twice).

>-io-apic (IRQO set to XT-PIC incorrectly)
>
>-udma133?
>  
>
udma133 may be a clue but I don't think anyone
achieves stability one way or the other on that. I
flogged every possible hdparm change and tried
three brands of hd controller without every
achieving stability, but once you're stable by
using other means you can use 133 and unmask
irq(not for siig sis?) and other hdparm opts.

>-cpu disconnect patch (missing bios option for ACPI Cx states)
>
>Craig
>  
>

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1.1: Type: text/plain, Size: 386 bytes --]

Hi, i'm now running this patch and it survived my grep in /usr/src.

It's mainly a correction of the apic patch and the ACPI halt disconnect
patch that was originally done for 2.6...

I'll get back to you about uptime, but i think this is it... 

Although i would prefer a not so workaroundish approach =)

-- 
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net

[-- Attachment #1.2: Type: text/x-diff, Size: 1680 bytes --]

--- arch/i386/kernel/mpparse.c.orig	2003-11-28 19:26:19.000000000 +0100
+++ arch/i386/kernel/mpparse.c	2003-12-06 19:35:15.000000000 +0100
@@ -1140,7 +1140,8 @@
 	 */
 	for (i = 0; i < mp_irq_entries; i++) {
 		if ((mp_irqs[i].mpc_dstapic == intsrc.mpc_dstapic) 
-			&& (mp_irqs[i].mpc_srcbusirq == intsrc.mpc_srcbusirq)) {
+			&& (mp_irqs[i].mpc_srcbusirq == intsrc.mpc_srcbusirq)
+			&& (mp_irqs[i].mpc_irqtype == intsrc.mpc_irqtype)) {
 			mp_irqs[i] = intsrc;
 			found = 1;
 			break;
--- arch/i386/kernel/pci-pc.c.orig	2003-12-06 19:32:44.000000000 +0100
+++ arch/i386/kernel/pci-pc.c	2003-12-06 19:33:55.000000000 +0100
@@ -1328,6 +1328,18 @@
 		dev->transparent = 1;
 }
 
+/*
+ * Halt Disconnect and Stop Grant Disconnect (bit 4 at offset 0x6F)
+ * must be disabled when APIC is used (or lockups will happen).
+ */
+static void __devinit pci_fixup_nforce2_disconnect(struct pci_dev *d)
+{
+	u8 t;
+
+	pci_read_config_byte(d, 0x6F, &t);
+	pci_write_config_byte(d, 0x6F, (t & 0xef));
+}
+
 struct pci_fixup pcibios_fixups[] = {
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82451NX,	pci_fixup_i450nx },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82454GX,	pci_fixup_i450gx },
@@ -1343,6 +1355,7 @@
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8367_0,	pci_fixup_via_northbridge_bug },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_NCR,	PCI_DEVICE_ID_NCR_53C810,	pci_fixup_ncr53c810 },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_ANY_ID,			pci_fixup_transparent_bridge },
+	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2,	pci_fixup_nforce2_disconnect },
 	{ 0 }
 };
 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sat, 2003-12-06 at 20:33, Ian Kumlien wrote:
> Hi, i'm now running this patch and it survived my grep in /usr/src.
> 
> It's mainly a correction of the apic patch and the ACPI halt disconnect
> patch that was originally done for 2.6...

Correction? how so? code looks the same, although the line numbers are
completely different for mpparse.c and at that location there is
different code. (Havent checked the disconnect)

Or do u just mean combination of the two patches?

> 
> I'll get back to you about uptime, but i think this is it... 

Why do you think the disconnect is also related? (given some are just
running the APIC patch and having (less/)no issues?
> 
> Although i would prefer a not so workaroundish approach =)

23 hrs now.. 

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1100 bytes --]

On Sun, 2003-12-07 at 01:20, Craig Bradney wrote:
> On Sat, 2003-12-06 at 20:33, Ian Kumlien wrote:
> > Hi, i'm now running this patch and it survived my grep in /usr/src.
> > 
> > It's mainly a correction of the apic patch and the ACPI halt disconnect
> > patch that was originally done for 2.6...
> 
> Correction? how so? code looks the same, although the line numbers are
> completely different for mpparse.c and at that location there is
> different code. (Havent checked the disconnect)

this is for 2.4.23 =)

> Or do u just mean combination of the two patches?

Combination + for 2.4.23

> > I'll get back to you about uptime, but i think this is it... 
> 
> Why do you think the disconnect is also related? (given some are just
> running the APIC patch and having (less/)no issues?

Since i had issues with just the apic patch, as stated in my mail.

> > Although i would prefer a not so workaroundish approach =)
> 
> 23 hrs now.. 

I'm at 5:06 and thats a record for it running with apic.

-- 
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sun, 2003-12-07 at 01:29, Ian Kumlien wrote:
> On Sun, 2003-12-07 at 01:20, Craig Bradney wrote:
> > On Sat, 2003-12-06 at 20:33, Ian Kumlien wrote:
> > > Hi, i'm now running this patch and it survived my grep in /usr/src.
> > > 
> > > It's mainly a correction of the apic patch and the ACPI halt disconnect
> > > patch that was originally done for 2.6...
> > 
> > Correction? how so? code looks the same, although the line numbers are
> > completely different for mpparse.c and at that location there is
> > different code. (Havent checked the disconnect)
> 
> this is for 2.4.23 =)

duh.. ok :)


> > Or do u just mean combination of the two patches?
> 
> Combination + for 2.4.23
> 
> > > I'll get back to you about uptime, but i think this is it... 
> > 
> > Why do you think the disconnect is also related? (given some are just
> > running the APIC patch and having (less/)no issues?
> 
> Since i had issues with just the apic patch, as stated in my mail.

oic.. sorry.. must have missed that.
> 
> > > Although i would prefer a not so workaroundish approach =)
> > 
> > 23 hrs now.. 
> 
> I'm at 5:06 and thats a record for it running with apic.

1d 9h here now..

I remembered I also dont have preempt still, but I am happy with it if
it stays like this. The next time I need to reboot (or get a hang), I'll
recompile with preempt it and try again.

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1445 bytes --]

On Sun, 2003-12-07 at 11:32, Craig Bradney wrote:
> On Sun, 2003-12-07 at 01:29, Ian Kumlien wrote:
> > On Sun, 2003-12-07 at 01:20, Craig Bradney wrote:
> > > Correction? how so? code looks the same, although the line numbers are
> > > completely different for mpparse.c and at that location there is
> > > different code. (Havent checked the disconnect)
> > 
> > this is for 2.4.23 =)
> 
> duh.. ok :)

Since none of my 2.6 running computers can read from the cdrom i have
decided to leave this last one on 2.4 =)
 
> > > Why do you think the disconnect is also related? (given some are just
> > > running the APIC patch and having (less/)no issues?
> > 
> > Since i had issues with just the apic patch, as stated in my mail.
> 
> oic.. sorry.. must have missed that.

Ok, i only got 2h uptime. But now i'm breaking the record with every
second that passes...
 
> > > > Although i would prefer a not so workaroundish approach =)
> > > 
> > > 23 hrs now.. 
> > 
> > I'm at 5:06 and thats a record for it running with apic.
> 
> 1d 9h here now..

19h 33m

> I remembered I also dont have preempt still, but I am happy with it if
> it stays like this. The next time I need to reboot (or get a hang), I'll
> recompile with preempt it and try again.

I'm more interested about getting NMI to work with io-apic right now, if
possible...

-- 
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 993 bytes --]

Craig Bradney wrote:
> All the interrupts are the same...except:
> 0, timer is now IO-APIC-edge.

Same here... 

> Im not getting any NMI counts.. should I use nmi-watchdog=1?

I got nmi counts with nmi_watchdog=2...  I never tested with =1... if
you get nmi's 1 lemme know.

> Ian, from looking back, you have an A7N8X-X bios 1007.
> Interesting that my USB hcis are still sharing IRQs there.

Your? i only see one... But you share it with sound and eth0... 

> Any idea how I can get them apart, or if I should try.

You could always move eth0 to a different slot. Other than that, you can
do manual config for the irq's in the bios, but it shouldn't be
needed...

> My system was pretty stable as I've stated.. but the patch has changed
> things slightly re the timer.

As i stated in my prev email, i had to do 2 full greps at a sizable
amount of data to recreate the crash... =P

And, please CC since i'm not on this ml =P
-- 
Ian Kumlien <pomac@vapor.com>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sat, 2003-12-06 at 03:08, Ian Kumlien wrote: 
> Craig Bradney wrote:
> > All the interrupts are the same...except:
> > 0, timer is now IO-APIC-edge.
> 
> Same here... 
> 
> > Im not getting any NMI counts.. should I use nmi-watchdog=1?
> 
> I got nmi counts with nmi_watchdog=2...  I never tested with =1... if
> you get nmi's 1 lemme know.
> 
> > Ian, from looking back, you have an A7N8X-X bios 1007.
> > Interesting that my USB hcis are still sharing IRQs there.
> 
> Your? i only see one... But you share it with sound and eth0... 
> 
> > Any idea how I can get them apart, or if I should try.
> 
> You could always move eth0 to a different slot. Other than that, you can
> do manual config for the irq's in the bios, but it shouldn't be
> needed...

eth0 is the 3com onboard on the a7n8x deluxe... 

> > My system was pretty stable as I've stated.. but the patch has changed
> > things slightly re the timer.
> 
> As i stated in my prev email, i had to do 2 full greps at a sizable
> amount of data to recreate the crash... =P
> 
> And, please CC since i'm not on this ml =P

Having finally woken up (me not the pc), uptime here is now 12 hours..
(without the CPU Disconnect athcool run, just the kernel patch). I did
run the athcool program to check the result though:

nVIDIA nForce2 (10de 01e0) found
'Halt Disconnect and Stop Grant Disconnect' bit is enabled.

Do others have the same value 10de 01e0 when they run athcool stat? Even
with the same motherboard (a7n8x deluxe)?

Im running a grep -R kernel /usr/* and another grep on my 4gb DVD and
compiling a Qt 3.2.3 upgrade now.

For me idle time never seemed to be a problem which I guess relates to
the CPU Disconnect on low usage/low power issue

Perhaps my motherboard and cpu doesnt have a problem with disconnect and
just the IRQ issue, perhaps because its only a few weeks old. It would
make sense in some ways given that my system has only one of the
problems given the uptime I have been able to reach.

My hangs have always been when I have used the PC.. and often completed
a task and then a few seconds later it goes.

Will see in time I guess

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bartlomiej Zolnierkiewicz]

On Saturday 06 of December 2003 14:36, Craig Bradney wrote:

> nVIDIA nForce2 (10de 01e0) found
> 'Halt Disconnect and Stop Grant Disconnect' bit is enabled.
>
> Do others have the same value 10de 01e0 when they run athcool stat? Even
> with the same motherboard (a7n8x deluxe)?

Yes.

0x10de is a PCI VendorID (nVidia).
0x01e0 is a PCI DeviceID (nForce2).

;-)

--bart

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1635 bytes --]

On Sat, 2003-12-06 at 14:36, Craig Bradney wrote:
> On Sat, 2003-12-06 at 03:08, Ian Kumlien wrote: 
> > You could always move eth0 to a different slot. Other than that, you can
> > do manual config for the irq's in the bios, but it shouldn't be
> > needed...
> 
> eth0 is the 3com onboard on the a7n8x deluxe... 

I find that so odd... My on board network controller is a realtec phys
controlled by a nvidia mac. Same goes for audio.

> Having finally woken up (me not the pc), uptime here is now 12 hours..
> (without the CPU Disconnect athcool run, just the kernel patch). I did
> run the athcool program to check the result though:
> 
> nVIDIA nForce2 (10de 01e0) found
> 'Halt Disconnect and Stop Grant Disconnect' bit is enabled.

nVIDIA nForce2 (10de 01e0) found
'Halt Disconnect and Stop Grant Disconnect' bit is enabled.


> Perhaps my motherboard and cpu doesnt have a problem with disconnect and
> just the IRQ issue, perhaps because its only a few weeks old. It would
> make sense in some ways given that my system has only one of the
> problems given the uptime I have been able to reach.

Mine is 2 weeks old or so.. 

> My hangs have always been when I have used the PC.. and often completed
> a task and then a few seconds later it goes.

Mine always seems to hang after a certain amount of time... When id
didn't do a grep in the kernel lib.

Btw, i have UDMA100 disks.. 2 disks on primary and 2 cdroms on
secondary... I dunno if this could make any difference..

> Will see in time I guess

Good luck =)

-- 
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sat, 2003-12-06 at 15:42, Ian Kumlien wrote:
> On Sat, 2003-12-06 at 14:36, Craig Bradney wrote:
> > On Sat, 2003-12-06 at 03:08, Ian Kumlien wrote: 
> > > You could always move eth0 to a different slot. Other than that, you can
> > > do manual config for the irq's in the bios, but it shouldn't be
> > > needed...
> > 
> > eth0 is the 3com onboard on the a7n8x deluxe... 
> 
> I find that so odd... My on board network controller is a realtec phys
> controlled by a nvidia mac. Same goes for audio.

Yes.. maybe if I was using the onboard nvidia nic it would not share..
Not sure.
> 
> > Having finally woken up (me not the pc), uptime here is now 12 hours..
> > (without the CPU Disconnect athcool run, just the kernel patch). I did
> > run the athcool program to check the result though:
> > 
> > nVIDIA nForce2 (10de 01e0) found
> > 'Halt Disconnect and Stop Grant Disconnect' bit is enabled.
> 
> nVIDIA nForce2 (10de 01e0) found
> 'Halt Disconnect and Stop Grant Disconnect' bit is enabled.

Its a pity that as Bart said, those numbers dont reflect any sort of
revision as that might lead to a conclusion about why it happens on some
and not others.


> > Perhaps my motherboard and cpu doesnt have a problem with disconnect and
> > just the IRQ issue, perhaps because its only a few weeks old. It would
> > make sense in some ways given that my system has only one of the
> > problems given the uptime I have been able to reach.
> 
> Mine is 2 weeks old or so.. 

> > My hangs have always been when I have used the PC.. and often completed
> > a task and then a few seconds later it goes.
> 
> Mine always seems to hang after a certain amount of time... When id
> didn't do a grep in the kernel lib.

I've just come hom to find my grep -R a /usr/* and my 4gb dvd grep have
completed correctly... 14 hours uptime now.

> Btw, i have UDMA100 disks.. 2 disks on primary and 2 cdroms on
> secondary... I dunno if this could make any difference..

1 ata133 primary master and dvdrw and cdrom on secondary here.

> > Will see in time I guess
> 
> Good luck =)

You too :)

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1526 bytes --]

On Sat, 2003-12-06 at 16:33, Craig Bradney wrote:
> On Sat, 2003-12-06 at 15:42, Ian Kumlien wrote:
> Its a pity that as Bart said, those numbers dont reflect any sort of
> revision as that might lead to a conclusion about why it happens on some
> and not others.

Yeah i saw that aswell... 

> > Btw, i have UDMA100 disks.. 2 disks on primary and 2 cdroms on
> > secondary... I dunno if this could make any difference..
> 
> 1 ata133 primary master and dvdrw and cdrom on secondary here.

I'm just wondering if it could be a relation to the disk aswell or so.. 

> > Good luck =)
> 
> You too :)

Heh, thanks =)

Now, about this ACPI powersave thing, wouldn't that be enabled in
windows aswell? So wouldn't this workaround be something that $other_os
doesn't have to do.

(In general i have always had to hack windows into not crashing when
linux worked and doing it the other way around without a real fix dosn't
sound that nice in my ears.. =P)

It would be interesting to hear from nvidia about nmi_watchdog... Since
nmi_watchdog with the 'not-done-correctly' APIC patch claims that nmi is
stalled/locked/doesn't work... If nvidia states that this *should* work,
then we have something to go on.

Also, if Allen Martin (nvidia) could go trough the proc/interrupts and
tell us if something is wrong, like the XT-PIC on timer. Or just give us
a correct listing, since noone had io-apic-edge on timer before afair. 

-- 
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Catching NForce2 lockup with NMI watchdog - found?

[Ian Kumlien]

[-- Attachment #1: Type: text/plain, Size: 1679 bytes --]

Quoting myself... 
> I'm more hopeful about the patch from Mathieu <cheuche+lkml () free ! fr>...

>           CPU0
>   0:     267486    IO-APIC-edge  timer
>   1:       9654    IO-APIC-edge  keyboard
>   2:          0          XT-PIC  cascade
>   8:          1    IO-APIC-edge  rtc
>   9:          0   IO-APIC-level  acpi
>  14:      28252    IO-APIC-edge  ide0
>  15:        103    IO-APIC-edge  ide1
>  16:     251712   IO-APIC-level  eth0
>  17:      90632   IO-APIC-level  EMU10K1
>  19:     415529   IO-APIC-level  nvidia
>  20:          0   IO-APIC-level  usb-ohci
>  21:        153   IO-APIC-level  ehci_hcd
>  22:      58257   IO-APIC-level  usb-ohci
> NMI:        479
> LOC:     265875
> ERR:          0
> MIS:          0

> this far and it feels like a closer match to what windows does from what
> i have read on the ml. 

I think that this is what we want, ie know how windows handles the spic
since i just bet that all the mb manuf. ppl only care about windows and
anything else is secondary. [Can we get some more info from nvidia about
differences in the setup?]

> I haven't even come close to testing this yet, I've only been up 45 mins
> but i'll leave it running and do what i usually do when it hangs... =)

And that some great 2 hours, everything was dandy, screen refreshes
faster (moving windows with contents was snappier and you saw less
trailing refreshes)... but it ended in a beeeeeeeep deadlock.

I later reproduced it again in console mode... It required 2 full grep
-rne test * in my /usr/src, that is, 2.6.0-test11 and 2.4.23*2 + some
rpm's... all in all: 624M + 219M

-- 
Ian Kumlien <pomac@vapor.com>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

RE: Catching NForce2 lockup with NMI watchdog

[Allen Martin]

> -----Original Message-----
> From: Jesse Allen [mailto:the3dfxdude@hotmail.com] 
> Sent: Friday, December 05, 2003 12:36 PM
>
> Do you know whether the nforce2's with apic support the timer 
> (IRQ 0) in 
> IO-APIC mode?  To me, it seems like a bug:
> "Dec  4 20:13:11 tesore kernel: ..MP-BIOS bug: 8254 timer not 
> connected to 
> IO-APIC"
> (This message originates in arch/i386/kernel/io_apic.c)
> 

Yes, Win 9x/2k/XP use the system timer on irq0 and have no problem.  I
haven't looked at this yet.

-Allen

Re: Catching NForce2 lockup with NMI watchdog - found?

[Prakash K. Cheemplavam]

Hi,

*maybe* I found the bugger, at least I got APIC more stable (need to 
test whether oit is really stable, compiling kernel right now...):

It is a problem with CPU disconnect function. I tried various parameters 
in bios and turned cpu disconnect off, and tada, I could do several 
subsequent hdparms and machine is running! As CPU disconnect is a ACPI 
state, if I am not mistkaen, I think there is something broken in ACPI 
right now or in APIC and cpu disconnect triggers the bug.

Maybe now my windows environment is stable, as well. It was much more 
stable with cpu disconnect and apic, nevertheless seldomly locked up.


So gals and guys, try disabling cpu disconnect in bios and see whether 
aopic now runs stable.

I have an Abit NF7-S Rev2.0 with Bios 2.0.

Prakash

Re: Catching NForce2 lockup with NMI watchdog - found?

[Prakash K. Cheemplavam]

Prakash K. Cheemplavam wrote:
> *maybe* I found the bugger, at least I got APIC more stable (need to 
> test whether oit is really stable, compiling kernel right now...):

So, new kernel is up. So far so good:

            CPU0
   0:      47118          XT-PIC  timer
   1:         34    IO-APIC-edge  i8042
   2:          0          XT-PIC  cascade
   8:          3    IO-APIC-edge  rtc
   9:          0   IO-APIC-level  acpi
  12:        864    IO-APIC-edge  i8042
  14:         10    IO-APIC-edge  ide0
  15:         16    IO-APIC-edge  ide1
  16:          0   IO-APIC-level  Skystar2
  18:       7690   IO-APIC-level  libata
  19:       1910   IO-APIC-level  nvidia
  20:         43   IO-APIC-level  ohci_hcd, eth0
  21:        540   IO-APIC-level  NVidia nForce2
  22:          0   IO-APIC-level  ohci_hcd
NMI:          0
LOC:      46934
ERR:          0
MIS:          0


Prakash

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sat, 2003-12-06 at 00:49, Prakash K. Cheemplavam wrote:
> Hi,
> 
> *maybe* I found the bugger, at least I got APIC more stable (need to 
> test whether oit is really stable, compiling kernel right now...):
> 
> It is a problem with CPU disconnect function. I tried various parameters 
> in bios and turned cpu disconnect off, and tada, I could do several 
> subsequent hdparms and machine is running! As CPU disconnect is a ACPI 
> state, if I am not mistkaen, I think there is something broken in ACPI 
> right now or in APIC and cpu disconnect triggers the bug.
> 
> Maybe now my windows environment is stable, as well. It was much more 
> stable with cpu disconnect and apic, nevertheless seldomly locked up.
> 
> 
> So gals and guys, try disabling cpu disconnect in bios and see whether 
> aopic now runs stable.

> I have an Abit NF7-S Rev2.0 with Bios 2.0.

> 
> Prakash

I rebooted and checked in my BIOS, I dont seem to have "CPU Disconnect"?
Is there another name. I also downloaded the motherboard manual for your
NF7-S and cant find it there either?

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Prakash K. Cheemplavam]

Craig Bradney wrote:
> On Sat, 2003-12-06 at 00:49, Prakash K. Cheemplavam wrote:
> 
>>Hi,
>>
>>*maybe* I found the bugger, at least I got APIC more stable (need to 
>>test whether oit is really stable, compiling kernel right now...):
>>
>>It is a problem with CPU disconnect function. I tried various parameters 
>>in bios and turned cpu disconnect off, and tada, I could do several 
>>subsequent hdparms and machine is running! As CPU disconnect is a ACPI 
>>state, if I am not mistkaen, I think there is something broken in ACPI 
>>right now or in APIC and cpu disconnect triggers the bug.
>>
>>Maybe now my windows environment is stable, as well. It was much more 
>>stable with cpu disconnect and apic, nevertheless seldomly locked up.
>>
>>
>>So gals and guys, try disabling cpu disconnect in bios and see whether 
>>aopic now runs stable.
> 
> 
>>I have an Abit NF7-S Rev2.0 with Bios 2.0.
> 
> 
>>Prakash
> 
> 
> I rebooted and checked in my BIOS, I dont seem to have "CPU Disconnect"?
> Is there another name. I also downloaded the motherboard manual for your
> NF7-S and cant find it there either?

th efull name should be "CPU Disconnect Function". it is an the page 
with "enhanced pci performance", "enable system bios caching" ".. video 
bios caching" and all the spread spectrums. I have forgotten the name of 
that page in the main menu. Should the 3 or 4 in the first column.

Perhaps your BIOS is too old. I remember it only came with 1.8 (or 
alike) and later. But usually this setting should be disabled at default.

My machine still hasn't locked, btw. :-)

Prakash

Re: Catching NForce2 lockup with NMI watchdog - found?

[Craig Bradney]

On Sat, 2003-12-06 at 01:21, Prakash K. Cheemplavam wrote:
> Craig Bradney wrote:
> > On Sat, 2003-12-06 at 00:49, Prakash K. Cheemplavam wrote:
> > 
> >>Hi,
> >>
> >>*maybe* I found the bugger, at least I got APIC more stable (need to 
> >>test whether oit is really stable, compiling kernel right now...):
> >>
> >>It is a problem with CPU disconnect function. I tried various parameters 
> >>in bios and turned cpu disconnect off, and tada, I could do several 
> >>subsequent hdparms and machine is running! As CPU disconnect is a ACPI 
> >>state, if I am not mistkaen, I think there is something broken in ACPI 
> >>right now or in APIC and cpu disconnect triggers the bug.
> >>
> >>Maybe now my windows environment is stable, as well. It was much more 
> >>stable with cpu disconnect and apic, nevertheless seldomly locked up.
> >>
> >>
> >>So gals and guys, try disabling cpu disconnect in bios and see whether 
> >>aopic now runs stable.
> > 
> > 
> >>I have an Abit NF7-S Rev2.0 with Bios 2.0.
> > 
> > 
> >>Prakash
> > 
> > 
> > I rebooted and checked in my BIOS, I dont seem to have "CPU Disconnect"?
> > Is there another name. I also downloaded the motherboard manual for your
> > NF7-S and cant find it there either?
> 
> th efull name should be "CPU Disconnect Function". it is an the page 
> with "enhanced pci performance", "enable system bios caching" ".. video 
> bios caching" and all the spread spectrums. I have forgotten the name of 
> that page in the main menu. Should the 3 or 4 in the first column.
> 
> Perhaps your BIOS is too old. I remember it only came with 1.8 (or 
> alike) and later. But usually this setting should be disabled at default.
> 
> My machine still hasn't locked, btw. :-)


Sounds great.. maybe you have come across something. Yes, the CPU
Disconnect function arrived in your BIOS in revision of 2003/03/27
"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset
does not support C2 disconnect; thus, disable C2 function."

For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can
see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS
that has been discussed.

Craig

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bob]

Craig Bradney wrote:

>On Sat, 2003-12-06 at 01:21, Prakash K. Cheemplavam wrote:
>  
>
>>Craig Bradney wrote:
>>    
>>
>>>On Sat, 2003-12-06 at 00:49, Prakash K. Cheemplavam wrote:
>>>
>>>      
>>>
>>>>Hi,
>>>>
>>>>*maybe* I found the bugger, at least I got APIC more stable (need to 
>>>>test whether oit is really stable, compiling kernel right now...):
>>>>
>>>>It is a problem with CPU disconnect function. I tried various parameters 
>>>>in bios and turned cpu disconnect off, and tada, I could do several 
>>>>subsequent hdparms and machine is running! As CPU disconnect is a ACPI 
>>>>state, if I am not mistkaen, I think there is something broken in ACPI 
>>>>right now or in APIC and cpu disconnect triggers the bug.
>>>>
>>>>Maybe now my windows environment is stable, as well. It was much more 
>>>>stable with cpu disconnect and apic, nevertheless seldomly locked up.
>>>>
>>>>
>>>>So gals and guys, try disabling cpu disconnect in bios and see whether 
>>>>aopic now runs stable.
>>>>        
>>>>
>>>      
>>>
>>>>I have an Abit NF7-S Rev2.0 with Bios 2.0.
>>>>        
>>>>
>>>      
>>>
>>>>Prakash
>>>>        
>>>>
>>>I rebooted and checked in my BIOS, I dont seem to have "CPU Disconnect"?
>>>Is there another name. I also downloaded the motherboard manual for your
>>>NF7-S and cant find it there either?
>>>      
>>>
>>th efull name should be "CPU Disconnect Function". it is an the page 
>>with "enhanced pci performance", "enable system bios caching" ".. video 
>>bios caching" and all the spread spectrums. I have forgotten the name of 
>>that page in the main menu. Should the 3 or 4 in the first column.
>>
>>Perhaps your BIOS is too old. I remember it only came with 1.8 (or 
>>alike) and later. But usually this setting should be disabled at default.
>>
>>My machine still hasn't locked, btw. :-)
>>    
>>
>
>
>Sounds great.. maybe you have come across something. Yes, the CPU
>Disconnect function arrived in your BIOS in revision of 2003/03/27
>"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset
>does not support C2 disconnect; thus, disable C2 function."
>
>For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can
>see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS
>that has been discussed.
>
>Craig
>
I don't have that in MSI K7N2 MCP2-T near the
agp and fsb spread spectrum items or anywhere
else.

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bartlomiej Zolnierkiewicz]

On Monday 08 of December 2003 04:08, Bob wrote:
> >Sounds great.. maybe you have come across something. Yes, the CPU
> >Disconnect function arrived in your BIOS in revision of 2003/03/27
> >"6.Adds"CPU Disconnect Function" to adjust C1 disconnects. The Chipset
> >does not support C2 disconnect; thus, disable C2 function."
> >
> >For me though.. Im on an ASUS A7N8X Deluxe v2 BIOS 1007. From what I can
> >see the CPU Disconnect isnt even in the Uber BIOS 1007 for this ASUS
> >that has been discussed.
> >
> >Craig
>
> I don't have that in MSI K7N2 MCP2-T near the
> agp and fsb spread spectrum items or anywhere
> else.

Use athcool:
	http://members.jcom.home.ne.jp/jacobi/linux/softwares.html#athcool
or apply kernel patch (2.4 and 2.6 versions were posted already).

--bart

Re: Catching NForce2 lockup with NMI watchdog - found?

[Bob]

Craig Bradney wrote:

>On Sat, 2003-12-06 at 00:49, Prakash K. Cheemplavam wrote:
>  
>
>>...try disabling cpu disconnect in bios and see whether 
>>aopic now runs stable.
>>    
>>
>
>  
>
>>I have an Abit NF7-S Rev2.0 with Bios 2.0.
>>    
>>
>
>  
>
>>Prakash
>>    
>>
>
>I rebooted and checked in my BIOS, I dont seem to have "CPU Disconnect"?
>Is there another name. I also downloaded the motherboard manual for your
>NF7-S and cant find it there either?
>
>Craig
>
I don't have that either on MSI K7N2  -Bob