Re: diagnosing hard-locks [was memtest+ won't load]




Andrew Sackville-West wrote:
On Fri, Jul 07, 2006 at 09:22:59PM -0700, Willie Wonka wrote:
ACPI hmm.. Let's see output of 'dmesg | grep ACPI'

andrew@basement:~$ dmesg | grep ACPI
BIOS-e820: 000000000dfec000 - 000000000dfef000 (ACPI data)
BIOS-e820: 000000000dfff000 - 000000000e000000 (ACPI NVS)
[snip]

For comparison reasons only -- here's mine;

---------------
~$ dmesg | grep ACPI
BIOS-e820: 000000000bffdc00 - 000000000bfffc00 (ACPI data)
BIOS-e820: 000000000bfffc00 - 000000000c000000 (ACPI NVS)
ACPI disabled because your bios is from 2000 and too old
ACPI: Subsystem revision 20040326
ACPI: Interpreter disabled.
-----------------

Whoops!!
I forgot to add one VERY IMPORTANT point when running *update-grub*

Remember to Copy down ANY/ALL Kernel and Initrd CUSTOM Boot settings!!

They are within the AutoMagic area (usually) - and running 'update-grub' gives
grub the authority to alter any custom settings to defaults (I.e.; it'll wipe
out your custom options and replace them with the defaults).

Fortunately for me, I only had that one kernel option (acpi=force) that I can
recall and easily add back (using 'sudo nano /boot/grub/menu.lst' to edit and
save the file)

So;
I'll refrain from posting mine anyway (once I reboot after customizing menu.lst
again ;-)) -- since you already have seemingly solved your issue ...

Do you use an 'acpi=force' kernel boot option in GRUB/Lilo ??
I do on this ~1999 PII, 350MHz 100FSB, 192MB RAM;

nope. there is some acpi setting in the BIOS that is turned on, can't
remember what it is exactly, will post in after next
reboot. meanwhile, I found I had apcid turned on from some acpi stuff
I was playing with a few months ago. I've update-rc.d remove'd it and
haven't had a problem in a day and a half (knock wood). I"m going to
go at this thing from the ground up again starting with all BIOS
settings and... heh... logging my actions, there's a novel thought.

Excellent!
Glad to hear it -- my system is one of the first ACPIU compliant, and there's
no BIOS setting for enabling/disabling ACPI, but there is the "Power" page,
which controls APM settings.

well. that's quite a tale. All I can say is, if the keyboard doesn't
seem to work right, maybe next time try a new keyboard first? he
he.

Yeah - but I'm stubborn as an Ox, and I needed to test my skills ;-0
Last resort (replace hardware) was such that if it didn't work, I'd be really
fuming -- perhaps that's why I held off for so long, and was trying to decipher
these type kernel messages;

-------------
Jul 3 08:17:04 localhost kernel: atkbd.c: Keyboard on isa0060/serio0 reports
too many keys pressed
..
Jul 3 08:17:39 localhost last message repeated 7 times
Jul 3 08:18:38 localhost last message repeated 12 times
Jul 3 08:19:37 localhost last message repeated 12 times
Jul 3 08:20:41 localhost last message repeated 13 times
Jul 3 08:21:41 localhost last message repeated 12 times
------------

ad-nauseum... :-(

on that note, i've heard several tales of keyboards being
resurrected by putting them through the dishwasher... I'd google it
first, but if you have a dead keyboard, it certainly wouldn't
hurt. I'm sure it'll void the warranty though.

Yeah, I know -- Thanks -- and that's definitely something I'd *do* ;-)
But she's already got too many bullet holes in her :-p
Besides the SpaceBar key had been semi-broken (physically) for quite sometime
already.

Regards

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com


--
To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx
with a subject of "unsubscribe". Trouble? Contact listmaster@xxxxxxxxxxxxxxxx



Relevant Pages

  • [ANNOUNCE] ACPI BIOS Guideline for Linux
    ... Linux is urgently missing and should get published. ... I mainly try to get feedback from OEMs and BIOS developers, ... ACPI BIOS checks whether Linux is running...). ...
    (Linux-Kernel)
  • Re: 2.6.25-rc1 ahci regression
    ... AMI BIOS detected: BIOS may corrupt low RAM, ... ACPI Warning: 32/64 FACS address mismatch in FADT - two FACS tables! ... pcieport 0000:00:01.0: Requesting control of PCIe PME from ACPI BIOS ... 8250/16550 driver, 4 ports, IRQ sharing enabled ...
    (Linux-Kernel)
  • Re: Event Viewer errors
    ... Windows is complaining it can't find the file. ... the Registry "call" for it behind. ... upgraded the BIOS per their instructions - successfully:). ... Source: ACPI ...
    (microsoft.public.windowsxp.general)
  • Re: FreeBSD 6.1-STABLE: Unexplained power off
    ... playing with BIOS and kernel options. ... I can only disable some ACPI ... <ACPI PCI bus> on pcib0 ... 2 ports with 2 removable, ...
    (freebsd-stable)
  • Re: motherboard recommendations
    ... A BIOS rev and you could be screwed. ... While with Supermicro boards I have never had a Bios problem when I have ... However they do not fully support the ACPI 3.0 ...
    (comp.unix.solaris)