Skip to contentSkip to conversation

2.6.14.2 kernel oops, looks acpi-related

DISCUSSION Geoff Mishkin 3 participants 4 messages Nov 22, 2005

lore
Share & tools
lore
#0 lore raw

About once a day, since upgrading from 2.6.12.5 to 2.6.14.2, I have a problem with the following symptoms. This is on a ThinkPad T42, and I'm using swsusp2 2.2-rc11.

o Kicker (the KDE panel) becomes unresponsive. The process is still there,
but I can't kill it (even with -9).
o My screen session gets weirded out. Every time I open a new window inside
of screen, or close one, I get an error at the top of my xterm saying "can't write /var/run/utmp" or something like this

At this point, I check dmesg and find a bunch of errors looking like this:

ACPI-0213: *** Error: Method reached maximum reentrancy limit (255)
ACPI-0508: *** Error: Method execution failed
[\_SB_.PCI0.LPC_.EC__.BAT0._BIF] (Node c20f9480), AE_AML_METHOD_LIMIT
ACPI-0213: *** Error: Method reached maximum reentrancy limit (255)
ACPI-0508: *** Error: Method execution failed
[\_SB_.PCI0.LPC_.EC__.BAT0._BIF] (Node c20f9480), AE_AML_METHOD_LIMIT
ACPI-0213: *** Error: Method reached maximum reentrancy limit (255)
ACPI-0508: *** Error: Method execution failed
[\_SB_.PCI0.LPC_.EC__.BAT0._BIF] (Node c20f9480), AE_AML_METHOD_LIMIT
ACPI-0213: *** Error: Method reached maximum reentrancy limit (255)
ACPI-0508: *** Error: Method execution failed
[\_SB_.PCI0.LPC_.EC__.BAT0._BIF] (Node c20f9480), AE_AML_METHOD_LIMIT

It's not always the exact same error condition, but the format is always like this.

At this point I reboot to clear up these problems. What will happen is that laptop_mode won't shut down properly, but I am able to kill it from another vt with just a SIGTERM. Then shutdown continues.

After the reboot, sometimes my touchpad will be on another device node /dev/input/event# and X won't start as a result. The touchpad still works if I swich the line in xorg.conf, BUT if I reboot again the touchpad goes back to its normal node. This happens about half the time that I've run into the overall problem.

Today, for the first time that I noticed at least, I also got an oops. Here it is:

Unable to handle kernel paging request at virtual address 4b87ad6e
 printing eip:
c01f6b76
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: uhci_hcd ehci_hcd vmnet parport_pc parport vmmon arc4 

ieee80211_crypt_wep ieee80211_crypt nfs lockd sunrpc cisco_ipsec ipt_state iptable_filter iptable_nat ip_nat ip_conntrack iptable_mangle ip_tables pcmcia yenta_socket rsrc_nonstatic pcmcia_core snd_pcm_oss snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc nls_iso8859_1 ntfs nls_base nvram usbhid usb_storage usbcore tun e1000 evdev fglrx intel_agp agpgart rtc speedstep_centrino freq_table ibm_acpi

CPU:    0
EIP:    0060:[<c01f6b76>]    Tainted: P   M  VLI
EFLAGS: 00210202   (2.6.14.2)

EIP is at acpi_ut_update_object_reference+0x26/0x140
eax: c20d75c0 ebx: 4b87ad6e ecx: d9ff5d78 edx: db6fad84
esi: db6fad84 edi: d9ff5d78 ebp: 00000000 esp: d9ff5d74
ds: 007b es: 007b ss: 0068
Process acpi (pid: 24336, threadinfo=d9ff4000 task=d9d2b030)

Stack: 00000000 db6fac64 c210818c c210818c 00000000 db64f1c0 c01f6ca8 c210818c
       00000000 c210818c c01ecd39 c210818c c210818c db64f1c0 db64f000 00000001
       00000009 c01e8874 db64f1c0 db64f000 db64f1c0 db64f000 c01e43d8 db64f1c0
Call Trace:
 [<c01f6ca8>] acpi_ut_add_reference+0x18/0x1c
 [<c01ecd39>] acpi_ex_resolve_node_to_value+0xc1/0xdc
 [<c01e8874>] acpi_ex_resolve_to_value+0x40/0x4b
 [<c01e43d8>] acpi_ds_resolve_operands+0x1e/0x38
 [<c01e3694>] acpi_ds_exec_end_op+0x1aa/0x332
 [<c01f11c6>] acpi_ps_parse_loop+0x53e/0x844
 [<c01f0b01>] acpi_ps_parse_aml+0x4e/0x1d5
 [<c01f19b2>] acpi_ps_execute_pass+0x72/0x84
 [<c01f18e2>] acpi_ps_execute_method+0x4e/0x70
 [<c01eeffd>] acpi_ns_execute_control_method+0x3e/0x4b
 [<c01eefa8>] acpi_ns_evaluate_by_handle+0x76/0x8d
 [<c01eeea5>] acpi_ns_evaluate_relative+0xa9/0xc5
 [<c01ee77f>] acpi_evaluate_object+0x10b/0x1cc
 [<c01c0003>] digest+0x73/0x110
 [<c01f93cc>] acpi_battery_get_info+0x64/0x115
 [<c01f9746>] acpi_battery_read_info+0x41/0x1e0
 [<c017f372>] seq_read+0xf2/0x2e0
 [<c015f636>] vfs_read+0xb6/0x180
 [<c015f9e1>] sys_read+0x51/0x80
 [<c0102fef>] sysenter_past_esp+0x54/0x75
Code: fe ff ff 58 c3 55 57 56 53 51 51 8b 5c 24 1c 0f b7 6c 24 20 c7 44 24 04 

00 00 00 00 85 db c7 04 24 00 00 00 00 0f 84 e6 00 00 00 <80> 3b 0f 0f 84 dd 00 00 00 0f b6 43 01 83 e8 04 83 f8 10 0f 87

As you can see, it looks like there some ACPI stuff going on there. This whole thing is more of an annoyance than anything else, but I figured since I got an oops this time I should report it.

--Geoff Mishkin <[email protected]>

#1 lore raw
Quoted 14 lines · Geoff Mishkin
Unable to handle kernel paging request at virtual address 4b87ad6e
 printing eip:
c01f6b76
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: uhci_hcd ehci_hcd vmnet parport_pc parport vmmon arc4 

ieee80211_crypt_wep ieee80211_crypt nfs lockd sunrpc cisco_ipsec ipt_state iptable_filter iptable_nat ip_nat ip_conntrack iptable_mangle ip_tables pcmcia yenta_socket rsrc_nonstatic pcmcia_core snd_pcm_oss snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc nls_iso8859_1 ntfs nls_base nvram usbhid usb_storage usbcore tun e1000 evdev fglrx intel_agp agpgart rtc speedstep_centrino freq_table ibm_acpi

Quoted 2 lines
CPU:    0
EIP:    0060:[<c01f6b76>]    Tainted: P   M  VLI

Unfortunatly you have loaded the vmware drivers into your kernel, so we really can't help out much here. If you can duplicate this without those drivers loaded, please file a bug at bugzilla.kernel.org.

thanks,

greg k-h

#2 lore raw

Alright, I can do that. What about cisco_ipsec and fglrx? cisco_ipsec I could live without, but fglrx would be difficult. I've never been able to get the radeon driver working on my system, at all.

Sorry if that sounds silly and the VMware drivers are the only ones that could be possibly causing the problem.

Thanks for the response.

--Geoff Mishkin <[email protected]>

Quoted 25 lines · you
Quoted 14 lines · Geoff Mishkin
Unable to handle kernel paging request at virtual address 4b87ad6e
 printing eip:
c01f6b76
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: uhci_hcd ehci_hcd vmnet parport_pc parport vmmon arc4

ieee80211_crypt_wep ieee80211_crypt nfs lockd sunrpc cisco_ipsec ipt_state iptable_filter iptable_nat ip_nat ip_conntrack iptable_mangle ip_tables pcmcia yenta_socket rsrc_nonstatic pcmcia_core snd_pcm_oss snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc nls_iso8859_1 ntfs nls_base nvram usbhid usb_storage usbcore tun e1000 evdev fglrx intel_agp agpgart rtc speedstep_centrino freq_table ibm_acpi

Quoted 2 lines
CPU:    0
EIP:    0060:[<c01f6b76>]    Tainted: P   M  VLI

Unfortunatly you have loaded the vmware drivers into your kernel, so we really can't help out much here. If you can duplicate this without those drivers loaded, please file a bug at bugzilla.kernel.org.

thanks,

greg k-h

#3 lore raw
Quoted 3 lines · Geoff Mishkin

Alright, I can do that. What about cisco_ipsec and fglrx? cisco_ipsec I could live without, but fglrx would be difficult. I've never been able to get the radeon driver working on my system, at all.

Please reproduce it without any binary-only drivers loaded.

Pekka