Bug 6578 - regression since 2.6.14.4: Tyan S2875 refuses to power off completly and deadlocks.
Summary: regression since 2.6.14.4: Tyan S2875 refuses to power off completly and dead...
Status: REJECTED INSUFFICIENT_DATA
Alias: None
Product: ACPI
Classification: Unclassified
Component: Power-Off (show other bugs)
Hardware: i386 Linux
: P2 high
Assignee: acpi_power-off
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-05-18 01:47 UTC by Keven Tipping
Modified: 2006-11-16 14:43 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.13
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Keven Tipping 2006-05-18 01:47:46 UTC
Most recent kernel where this bug did not occur: 2.6.13/2.6.14
Distribution: Fedora Core 4/5
Hardware Environment: Tyan S2875, Nvidia 5950 Ultra AGP
Software Environment: Bios V3.02
Problem Description:
Tyan S2875 will not kill the ATX power after ACPI power off. Linux terminates, however, the Port 80 
LCD displays "FF". Given roughly a minute, the system emits an extremely loud alternating alarm- 
"HIGH-LOW-HIGH-LOW-HIGH-LOW".

Changing ACPI options under the BIOS has absolutely no effect. The system will completly cease up 
aside from the alarm, Reset button has NO effect, nether does holding the ATX power button. You have 
to pull the mains at the back of the computer to kill it. Returning the mains power allows the system to 
boot normally.

Kernels 2.6.13 and 2.6.14.4 worked- ie, the system cut the ATX power and remains off, and responds 
to pushing the power button. Kernels 2.6.12 and previous, as well as 2.6.15 and later, exhibit the same 
problem. Problem is not evident under Windows (any version). Linux kernels were 64 Bit, SMP enabled.

System board is not at fault- as the problem dissapeared in 2.6.13 through to 2.6.14.4.

Steps to reproduce:
Shut down the system with any other kernel then 2.6.13-2.6.14.4. This is not a faulty board problem, 
as the problem is not evident in other OS's.

Running the Kororaa LiveCD 0.2 (2.6.14.4) reveals that the system will power down properly.
Comment 1 Dan Carpenter 2006-05-22 19:50:13 UTC
That sounds like the intrusion detection alarm.

I read where you said it wasn't a BIOS issue, but what BIOS are you using?  It
could still be a BIOS issue even if other OSs aren't affected.

Comment 2 Luming Yu 2006-06-04 19:31:30 UTC
Please make sure you are using latest BIOS
Comment 3 Luming Yu 2006-06-07 08:20:03 UTC
>Steps to reproduce:
>Shut down the system with any other kernel then 2.6.13-2.6.14.4

Sounds like a regression since 2.6.14.4.
Comment 4 Luming Yu 2006-06-12 00:57:56 UTC
Could you do bisection to find out the patch causes this regression?
Comment 5 Adrian Bunk 2006-11-16 14:43:33 UTC
Please reopen this bug if:
- it is still present in kernel 2.6.19-rc6 and
- you can provide the requested information.

Note You need to log in before you can comment on or make changes to this bug.