widely experienced mouse freeze with AMD X2 & nVidia chipset

[mwette]

I have a Dell C521 with AMD X2 3800+ and ATI X1300 graphics.
This box has a motherboard chipset from nVidia (C51 MCP51).

There are a large number of people complaining about mouse
freeze-ups. I have experienced this when I boot with "acpi=off".
Right now I am not using that option but I have only one core
working. This type of problem seems to be happening with multiple
distributions. Some believe that it is some timing issue with the
nVidia hardware. bugzilla.kernel.org bug # is 7397. dell.com
has forums and there are comments there about "USB linux".

Some have had success by placing mouse and keyboard on
a separate PCI-USB board.

[EDIT: kernel 2849 will not boot with "nolapic" nor with "acpi=off"
the 2798 kernel would boot with either of these args]

[EDIT: new news -- someone ran WinXP and got freeze-ups until
they installed the urgent nVidia driver fix. This from dell forums.]

[leigh123linux]

No problems here, why do you have only one core working?

[mwette]

No problems here, why do you have only one core working?

CPU0: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ stepping 02
SMP alternatives: switching to SMP code
Booting processor 1/1 eip 3000
CPU 1 irqstacks, hard=c07b0000 soft=c0790000
Not responding.
Inquiring remote APIC #1...
... APIC #1 ID: 01000000
... APIC #1 VERSION: 80050010
... APIC #1 SPIV: 000000ff
CPU #1 not responding - cannot use it.
Total of 1 processors activated (4010.53 BogoMIPS).

[leigh123linux]

Strange ?

[leigh@localhost ~]$ cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 15
model : 43
model name : AMD Athlon(tm) 64 X2 Dual Core Processor 3800+
stepping : 1
cpu MHz : 2000.000
cache size : 512 KB
physical id : 0
siblings : 2
core id : 0
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt lm 3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4020.32
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 43
model name : AMD Athlon(tm) 64 X2 Dual Core Processor 3800+
stepping : 1
cpu MHz : 2000.000
cache size : 512 KB
physical id : 0
siblings : 2
core id : 1
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt lm 3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4017.84
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

[leigh@localhost ~]$

CPU 1 irqstacks, hard=c07b0000 soft=c0790000 (do these mean hard & soft locks , check in bios and if you have it disable ERR 94 ehn ( I think i got setting name right )
I remember reading some AMD paper for S939 dual core saying this setting could cause problems.
The motherboard may need bios update to reconize processor?


what command did you use to get your processor info.

[mwette]

[leigh@localhost ~]$ cat /proc/cpuinfo
...
cpu cores : 2


what command did you use to get your processor info.

I don't doubt that other machines are running. To get the info I
posted I did "more dmesg". My current guess is that the problem
has to do with the timer interrupt:

# cat /proc/interrupts 
           CPU0       
  0:    2252758          XT-PIC  timer
  8:          1    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 50:       3307   IO-APIC-level  eth0
 58:     131111   IO-APIC-level  fglrx
209:      14480   IO-APIC-level  ohci_hcd:usb1, HDA Intel
217:     227756   IO-APIC-level  ehci_hcd:usb2
225:      40671   IO-APIC-level  libata
233:          0   IO-APIC-level  libata
NMI:          0 
LOC:    2247584 
ERR:          1
MIS:          0

Note that the timer is using legacy routing (XT-PIC). If I boot
the old (1.2798) kernel w/ acpi=off, I find the timer on "IO-APIC-edge".

[leigh123linux]

[root@localhost ~]# cat /proc/interrupts
CPU0 CPU1
0: 333239 300091 IO-APIC-edge timer
1: 196 208 IO-APIC-edge i8042
6: 5 0 IO-APIC-edge floppy
8: 0 0 IO-APIC-edge rtc
9: 0 0 IO-APIC-level acpi
12: 10997 8738 IO-APIC-edge i8042
14: 5576 5209 IO-APIC-edge ide0
15: 458328 464322 IO-APIC-edge ide1
50: 0 0 IO-APIC-level libata
58: 0 0 IO-APIC-level EMU10K1
66: 21258 22560 IO-APIC-level fglrx
217: 198149 0 IO-APIC-level ohci_hcd:usb1, eth0
225: 12999 12153 IO-APIC-level ehci_hcd:usb2
233: 9687 2121 IO-APIC-level libata
NMI: 205 130
LOC: 633164 633092
ERR: 0
MIS: 0


Did you check to see is err 94 enh is enabled in bios if so try disabling it. Worth a check

[Duli]

Hello, I have the same problem on a Dell Dimension C521, with AMD 64 2.2GHz and nvidia chipset (nForce 430 - C51 MCP51).

Mouse freezes everytime. My sound card, altough recognized, won't play any sound.

People have filed a bug report on kernel bugzilla tracker (bug #7397):

http://bugzilla.kernel.org/show_bug.cgi?id=7397

Also, a lot of people with that and other machines with nforce 430 chipset are experiencing the same mouse problem:

http://forums.us.dell.com/supportfor...cending&page=1

As described at several foruns, it happens in every linux distribution!!! Ubuntu, Fedora, Debian, FreeBSD, Gentoo ...

The funny thing?! Dell has announced this machine to be especially designed for those who'd like to use linux:

http://www.dell.com/content/topics/s...=19&l=en&s=dhs

People guess it's a BIOS problem or a kernel problem.

Hope the problem will be fixed soon, because it's impossible to work with linux on that machines!!!

Cheers,
Duli

[mwette]

Did you try to boot with "nosmp" ?

The disappointing solution most people have is to run an externally powered
USB system: either a PCIx card or externally powered hub plugged into a
single usb port.

My machine is currently stable. For some reason is it booting with only a
single core and the interrupts seem fishy (INT0 is "XT-PIC").

[Duli]

Did you try to boot with "nosmp" ?

The disappointing solution most people have is to run an externally powered
USB system: either a PCIx card or externally powered hub plugged into a
single usb port.

My machine is currently stable. For some reason is it booting with only a
single core and the interrupts seem fishy (INT0 is "XT-PIC").

No, my machine is a single core. I think "nosmp" wouldn't differ...

I haven't tried the hub solution yet. I don't have a usb hub...

And what about the sound card? Did yours work right away?! What type is it (MCP51???)?

Thanks!
Duli

[mwette]

No, my machine is a single core. I think "nosmp" wouldn't differ...

I haven't tried the hub solution yet. I don't have a usb hub...

And what about the sound card? Did yours work right away?! What type is it (MCP51???)?

Thanks!
Duli

My sound seems to be working. I believe it is the "MCP51" or whatever the
default on-board device is. I don't know how to help you except maybe
show my modprobe.conf and attach relevant parts of output from "lshal".

/etc/modprobe.conf:

alias eth0 b44
alias scsi_hostadapter sata_nv
alias snd-card-0 snd-hda-intel
options snd-card-0 index=0  
options snd-hda-intel index=0  
remove snd-hda-intel { /usr/sbin/alsactl store 0 >/dev/null 2>&1 || : ; }; /sbin
/modprobe -r --ignore-remove snd-hda-intel
alias net-pf-10 off
alias ipv6 off

[Duli]

My sound seems to be working. I believe it is the "MCP51" or whatever the
default on-board device is. I don't know how to help you except maybe
show my modprobe.conf and attach relevant parts of output from "lshal".

Hey, thanks a lot! Sure it helps!

I'm away from that computer right now because it's stays at work... I'll check the options later.

One last question, what options are you using to boot? Could you copy and paste your entire line of the kernel in /boot/grub/menu.lst?

thanks again!
Cheers!
Duli

[mwette]

Hey, thanks a lot! Sure it helps!

I'm away from that computer right now because it's stays at work... I'll check the options later.

One last question, what options are you using to boot? Could you copy and paste your entire line of the kernel in /boot/grub/menu.lst?

thanks again!
Cheers!
Duli

No problem.

title Fedora Core (2.6.18-1.2849.fc6)
        root (hd0,1)
        kernel /boot/vmlinuz-2.6.18-1.2849.fc6 ro root=LABEL=/ rhgb apic=debug
        initrd /boot/initrd-2.6.18-1.2849.fc6.img

[Duli]

No problem.

Thanks!!!
Duli

[wt6g]

Kernel 2849 appears to be seriously broke. I have two systems that went into hyperspace when I "upgraded" to FC6 2849 from 2798. The problems with USB are really serious - I had keyboards and mice acting like they lived in antarctica and had epilepsy. System load monitor stays high for some unknown reason too. Both systems were fixed by reverting to Version 2798. One is Intel Pentium 4, the other is AMD Athelon. I did try lots of hardware manipulation to get 2849 to work but was unable to get it to function properly on either system.

[Duli]

Kernel 2849 appears to be seriously broke. I have two systems that went into hyperspace when I "upgraded" to FC6 2849 from 2798. The problems with USB are really serious - I had keyboards and mice acting like they lived in antarctica and had epilepsy. System load monitor stays high for some unknown reason too. Both systems were fixed by reverting to Version 2798. One is Intel Pentium 4, the other is AMD Athelon. I did try lots of hardware manipulation to get 2849 to work but was unable to get it to function properly on either system.

Hum, good to know. Truth be told, I haven´t tried yet to see if the mouse problems happens in my FC6 with the previous kernel... I´ll try that! Thanks for the info. Hope kernel developers solve this soon!!!