RE: Re: Oops during startup

RE: Re: Oops during startup

[Yu, Luming]

 Please try kernel option: acpi_sci={ level | edge | high | low }

--Luming

>-----Original Message-----
>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org 
>[mailto:acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of 
>Andrew D. Keyser
>Sent: Wednesday, June 02, 2004 10:48 AM
>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: [ACPI] Re: Oops during startup
>
>Your problem seems similar to mine; I can't boot any kernel newer than 
>2.6.6-mm4 without using acpi=off. When the kernel boots, it gets to 
>"ACPI: IRQ9 SCI: Level Trigger." and stops. I have an HP ze4560us. The 
>stranger thing though is that I can't even load the Gentoo LiveCD 
>without acpi=off, so something must be broken either in my laptop 
>(however that happened) or in both of the kernel versions...
>Might not be related but the symptoms are similar so I figured I might 
>as well add on.
>
>Andrew
>
>Ethan Glasser-Camp wrote:
>> 2.6.6 oopses on startup for me. acpi=off seems to work OK.
>> 
>> My Thinkpad T-30 has served me faithfully for about 1.5 
>years, but was
>> needing some service so I took it to the shop today. They gave me a
>> replacement laptop (same model, still a T-30), but when I tried to
>> boot up, I got a kernel oops. 2.6.5 boots up fine, and so does 2.6.6
>> when given the boot parameter acpi=off. The machines are basically
>> identical except for the CD-ROM drive. Might this cause a problem? Or
>> should I be pointing a finger at the ACPI tables and other firmware?
>> 
>> Since the kernel oopses before it finishes booting up, I can't even
>> scroll up to the top of it to see what the oops message is. 
>I can copy
>> the callback if it helps. How can I help to diagnose the problem?
>> 
>> Ethan Glasser-Camp
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the one
>installation-authoring solution that does it all. Learn more and
>evaluate today! http://www.installshield.com/Dev2Dev/0504
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>


-------------------------------------------------------
This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the one
installation-authoring solution that does it all. Learn more and
evaluate today! http://www.installshield.com/Dev2Dev/0504

Re: Re: Oops during startup

[Ethan Glasser-Camp]

[-- Attachment #1: Type: text/plain, Size: 1528 bytes --]

On Wed, Jun 02, 2004 at 11:17:57AM +0800, Yu, Luming wrote:
>  Please try kernel option: acpi_sci={ level | edge | high | low }
> 
> --Luming

None of the above resulted in any differences.

Yi Zhu wrote:
> But firstly could you provide whatever be seen to give us a clue?

Here is my reproduction of the oops, as well as I could do it:

   EIP is at acpi_os_read_memory+0x0c/0x60
eax: c0428120  ebx: c13d6b7c  ecx: 00000008  edx: c1233f38
esi: c13d6b94  edi: c1233f38  ebp: 00000008  esp: c1233edc
ds: 007b  es: 0007b  ss: 0068
  Process events/0 (pid:3, threadinfo=c1232000  task=c123cc10)
Stack: 00001a66 c0246116 c0428120 c04280c0 c1233f38 00000008 c13d6b7c c1233f38
       00000282 c1233f3c c13c72bc 00000282 00000002 c03c18d0 c03c1520 c025edc8
       00000008 c1233f38 c13d6b94 c1233f68 c13c72bc c1232000 00000283 00000000

Call trace:
 [<c0246116>] acpi_hw_low_level_read+0x6e/0x124
 [<c025edc8>] acpi_ec_gpe_query+0x128/0x220
 [<c023019d>] acpi_os_execute_deferred+0x5d/0x80
 [<c0230140>] acpi_os_execute_deferred+0x0/0x80
 [<c0130604>] worker_thread+0x298/0x3dc
 [<c013036c>] worker_thread+0x0/0x3dc
 [<c0230140>] acpi_os_execute_deferred+0x0/0x80
 [<c011913c>] default_wake_function+0x0/0x1c
 [<c011913c>] default_wake_function+0x0/0x1c
 [<c013519c>] kthread+0x70/0xa0
 [<c013512c>] kthread+0x0/0xa0
 [<c0103245>] kernel_thread_helper+0x5/0xc

Code: 8a 80 00 00 00 c0 88 02 eb 26 0f b7 80 00 00 00 c0 66 89 02
<6>note: events/0[3] exited with preempt-count 1

Ethan Glasser-Camp

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

RE: Re: Oops during startup

[Li, Shaohua]

Hi,
I don't know if below patch can help you, but it seems there is a
potential error. We should remove gpe handler before address space
hander is removed, since gpe handler may use gpe address space handler.
Please just try.

Thanks
Shaohua

--- 2.6/drivers/acpi/ec.c.orig	2004-06-02 13:31:31.273490992 +0800
+++ 2.6/drivers/acpi/ec.c	2004-06-02 13:32:23.205596112 +0800
@@ -595,11 +595,9 @@ acpi_ec_add (
 	   we now have the *real* EC info, so kill the makeshift one.*/
 	acpi_evaluate_integer(ec->handle, "_UID", NULL, &uid);
 	if (ec_ecdt && ec_ecdt->uid == uid) {
+		acpi_remove_gpe_handler(NULL, ec_ecdt->gpe_bit,
&acpi_ec_gpe_handler);
 		acpi_remove_address_space_handler(ACPI_ROOT_OBJECT,
 			ACPI_ADR_SPACE_EC, &acpi_ec_space_handler);
-	
-		acpi_remove_gpe_handler(NULL, ec_ecdt->gpe_bit,
&acpi_ec_gpe_handler);
-
 		kfree(ec_ecdt);
 	}

>-----Original Message-----
>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ethan Glasser-Camp
>Sent: Wednesday, June 02, 2004 12:29 PM
>To: Yu, Luming; acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: Re: [ACPI] Re: Oops during startup
>
>On Wed, Jun 02, 2004 at 11:17:57AM +0800, Yu, Luming wrote:
>>  Please try kernel option: acpi_sci={ level | edge | high | low }
>>
>> --Luming
>
>None of the above resulted in any differences.
>
>Yi Zhu wrote:
>> But firstly could you provide whatever be seen to give us a clue?
>
>Here is my reproduction of the oops, as well as I could do it:
>
>   EIP is at acpi_os_read_memory+0x0c/0x60
>eax: c0428120  ebx: c13d6b7c  ecx: 00000008  edx: c1233f38
>esi: c13d6b94  edi: c1233f38  ebp: 00000008  esp: c1233edc
>ds: 007b  es: 0007b  ss: 0068
>  Process events/0 (pid:3, threadinfo=c1232000  task=c123cc10)
>Stack: 00001a66 c0246116 c0428120 c04280c0 c1233f38 00000008 c13d6b7c
>c1233f38
>       00000282 c1233f3c c13c72bc 00000282 00000002 c03c18d0 c03c1520
>c025edc8
>       00000008 c1233f38 c13d6b94 c1233f68 c13c72bc c1232000 00000283
>00000000
>
>Call trace:
> [<c0246116>] acpi_hw_low_level_read+0x6e/0x124
> [<c025edc8>] acpi_ec_gpe_query+0x128/0x220
> [<c023019d>] acpi_os_execute_deferred+0x5d/0x80
> [<c0230140>] acpi_os_execute_deferred+0x0/0x80
> [<c0130604>] worker_thread+0x298/0x3dc
> [<c013036c>] worker_thread+0x0/0x3dc
> [<c0230140>] acpi_os_execute_deferred+0x0/0x80
> [<c011913c>] default_wake_function+0x0/0x1c
> [<c011913c>] default_wake_function+0x0/0x1c
> [<c013519c>] kthread+0x70/0xa0
> [<c013512c>] kthread+0x0/0xa0
> [<c0103245>] kernel_thread_helper+0x5/0xc
>
>Code: 8a 80 00 00 00 c0 88 02 eb 26 0f b7 80 00 00 00 c0 66 89 02
><6>note: events/0[3] exited with preempt-count 1
>
>Ethan Glasser-Camp


-------------------------------------------------------
This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the one
installation-authoring solution that does it all. Learn more and
evaluate today! http://www.installshield.com/Dev2Dev/0504

Re: Oops during startup

[Andrew D. Keyser]

I tried all of them, none of them made any difference (except the edge 
parameter makes it say "ACPI: IRQ9 SCI: Level Trigger set to Egde mode." 
or something).

Andrew

Yu, Luming wrote:
>  Please try kernel option: acpi_sci={ level | edge | high | low }
> 
> --Luming
> 
> 
>>-----Original Message-----
>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org 
>>[mailto:acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of 
>>Andrew D. Keyser
>>Sent: Wednesday, June 02, 2004 10:48 AM
>>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>Subject: [ACPI] Re: Oops during startup
>>
>>Your problem seems similar to mine; I can't boot any kernel newer than 
>>2.6.6-mm4 without using acpi=off. When the kernel boots, it gets to 
>>"ACPI: IRQ9 SCI: Level Trigger." and stops. I have an HP ze4560us. The 
>>stranger thing though is that I can't even load the Gentoo LiveCD 
>>without acpi=off, so something must be broken either in my laptop 
>>(however that happened) or in both of the kernel versions...
>>Might not be related but the symptoms are similar so I figured I might 
>>as well add on.
>>
>>Andrew
>>
>>Ethan Glasser-Camp wrote:
>>
>>>2.6.6 oopses on startup for me. acpi=off seems to work OK.
>>>
>>>My Thinkpad T-30 has served me faithfully for about 1.5 
>>
>>years, but was
>>
>>>needing some service so I took it to the shop today. They gave me a
>>>replacement laptop (same model, still a T-30), but when I tried to
>>>boot up, I got a kernel oops. 2.6.5 boots up fine, and so does 2.6.6
>>>when given the boot parameter acpi=off. The machines are basically
>>>identical except for the CD-ROM drive. Might this cause a problem? Or
>>>should I be pointing a finger at the ACPI tables and other firmware?
>>>
>>>Since the kernel oopses before it finishes booting up, I can't even
>>>scroll up to the top of it to see what the oops message is. 
>>
>>I can copy
>>
>>>the callback if it helps. How can I help to diagnose the problem?
>>>
>>>Ethan Glasser-Camp
>>
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by the new InstallShield X.
> 
>>From Windows to Linux, servers to mobile, InstallShield X is the one
> 
>>installation-authoring solution that does it all. Learn more and
>>evaluate today! http://www.installshield.com/Dev2Dev/0504
>>_______________________________________________
>>Acpi-devel mailing list
>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by the new InstallShield X.
> From Windows to Linux, servers to mobile, InstallShield X is the one
> installation-authoring solution that does it all. Learn more and
> evaluate today! http://www.installshield.com/Dev2Dev/0504



-------------------------------------------------------
This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the one
installation-authoring solution that does it all. Learn more and
evaluate today! http://www.installshield.com/Dev2Dev/0504

RE: Re: Oops during startup

[Yu, Luming]

Andrew,

Hmmm.., it  sound like a regression.  If you suspect ACPI did it,
Please take a look at
 http://www.kernel.org/pub/linux/kernel/people/lenb/acpi/patches/

And I think you can tracker down which the patch broke your laptop.

Good Luck!

Thanks,
Luming

>>>Your problem seems similar to mine; I can't boot any kernel 
>newer than 
>>>2.6.6-mm4 without using acpi=off. When the kernel boots, it gets to 
>>>"ACPI: IRQ9 SCI: Level Trigger." and stops. I have an HP 
>ze4560us. The 
>>>stranger thing though is that I can't even load the Gentoo LiveCD 
>>>without acpi=off, so something must be broken either in my laptop 
>>>(however that happened) or in both of the kernel versions...
>>>Might not be related but the symptoms are similar so I 
>figured I might 


-------------------------------------------------------
This SF.Net email is sponsored by: GNOME Foundation
Hackers Unite!  GUADEC: The world's #1 Open Source Desktop Event.
GNOME Users and Developers European Conference, 28-30th June in Norway
http://2004/guadec.org