Linux

ACPI not working -- Laptop doesn't boot without acpi=off

Bug #38476 reported by carsten
12
Affects Status Importance Assigned to Milestone
Linux
Fix Released
Medium
linux-source-2.6.15 (Ubuntu)
Fix Released
Medium
Unassigned

Bug Description

On my Toshiba Satellite a100-533 the kernel doesn't work without using acpi=off. I will attach a kernel-oops-foto I took by using Knoppix 5.0. I guess it is the same bug (aka unsupported device), but in Kubuntu (I was using the flight5 and flight6 live-cd) I don't get the oops-debug-output, it just crashes.

I will also attach lspci -vvv

What other kind of information do you need?

Revision history for this message
carsten (cniehaus-kde) wrote :

Sorry, of course this is not against 2.4.x but against the kernel used in Flight6 (2.6.15-something)... I cannot find a way how to change the product this bug is filed against...

Revision history for this message
carsten (cniehaus-kde) wrote : Foto of the kernel-oops

Sorry for the bad quality, on the mini-LCD it looked better.

Revision history for this message
Chuck Short (zulcss) wrote :

Please attach the outputs of sudo dmidecode > dmi.txt.

Thanks
chuck

Changed in kernel-image-2.4.27-i386:
status: Unconfirmed → Needs Info
Revision history for this message
carsten (cniehaus-kde) wrote : Output of dmidecode

Output of dmidecode

Revision history for this message
carsten (cniehaus-kde) wrote : Output of lspci

This is lspci -vvv.

Revision history for this message
carsten (cniehaus-kde) wrote :

I retested:
In Knoppix only failsafe-more works. But then even KDE starts with sound... They are using 2.6.15 as well.
In Flight6 only the combination

  acpi=off noapic nolapic

seems to work.

Also, I changed the product to kernel-image-2.6.15-18-386-di. This was the most appropiate package, no idea what the "-di" stands for, sorry.

Revision history for this message
carsten (cniehaus-kde) wrote :

I hope this is the correct source-package...

Revision history for this message
Matthew Garrett (mjg59) wrote :

Looks like either ata_host_add or request_irq is failing in ata_device_add in drivers/scsi/libata-core.c - I'd go for request_irq, given that it works without acpi. Does booting with pci=noacpi as the only extra boot option help?

Revision history for this message
carsten (cniehaus-kde) wrote :

I cannot test right now but would like to ask before testing: "only" means

live pci=noacpi

or also those
  root=... usplash rw --

and so on parameters? The default is a quite long promt so I would really like to know what exactly to enter as promt...

I hope I can test tomorrow or next monday.

Revision history for this message
Matthew Garrett (mjg59) wrote :

Sorry, with all the normal boot options and also pci=noacpi, but not acpi=off noapic nolapic

Revision history for this message
carsten (cniehaus-kde) wrote :

ok, with the rc of dapper this problem is gone, so you have fixed the bug :)

Revision history for this message
Dennis Kaarsemaker (dennis) wrote : Re: [Bug 38476] Re: ACPI not working -- Laptop doesn't boot without acpi=off

Good to hear!

 status FixReleased

Changed in linux-source-2.6.15:
status: Needs Info → Fix Released
Bug Watch Updater (bug-watch-updater)
Changed in linux:
status: Unconfirmed → Confirmed
Bug Watch Updater (bug-watch-updater)
Changed in linux:
status: Confirmed → Rejected
Bug Watch Updater (bug-watch-updater)
Changed in linux:
status: Invalid → Fix Released
Bug Watch Updater (bug-watch-updater)
Changed in linux:
importance: Unknown → Medium
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.