Bug 5446

Summary:	Audio bangs on poweroff
Product:	Drivers	Reporter:	Oleksij Rempel (fishor) (bug-track)
Component:	Sound(ALSA)	Assignee:	Jaroslav Kysela (perex)
Status:	REJECTED INSUFFICIENT_DATA
Severity:	normal	CC:	acpi-bugzilla, perex, protasnb, tiwai
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.16	Subsystem:
Regression:	---	Bisected commit-id:
Attachments:	Add reboot notifier Add PCI shutdown callback

Description Oleksij Rempel (fishor) 2005-10-15 00:07:13 UTC

Most recent kernel where this bug did not occur: 2.6.12; 2.6.13; 2.6.14-rc3
Distribution:Ubuntu Breezy
Hardware Environment:laptop samsung nv5000, sound ES1988 Allegro-1
Software Environment:kernel + alsa
Problem Description:
Power_off process is not correct, sound device mas to be disabled before power
off. This is realy dangerous, if device is not disabled it's make laut sound
impuls wich can damage hardware.

Comment 1 Oleksij Rempel (fishor) 2005-10-15 00:09:35 UTC

Actualy i don't have rigt to say it's not correct but dangerous :o)

Comment 2 Oleksij Rempel (fishor) 2005-10-15 04:48:21 UTC

I added "modprobe -r snd-maestro3" to halt script. It's halp power_off properly,
with out any krash_sound but it's need alsa compiled as module.

Comment 3 Takashi Iwai 2005-10-17 03:14:57 UTC

Does the patch below help?

Comment 4 Takashi Iwai 2005-10-17 03:15:59 UTC

Created attachment 6323 [details]
Add reboot notifier

Comment 5 Takashi Iwai 2005-10-17 06:39:47 UTC

Created attachment 6325 [details]
Add PCI shutdown callback

Comment 6 Takashi Iwai 2005-10-17 06:40:46 UTC

The second one (in comment #5) is better.

Comment 7 Oleksij Rempel (fishor) 2005-10-17 07:04:53 UTC

I tryed therst one but it's not helping.
The secend one i will try leater. I tryed poweroff with "alsactl power off"
added to the halt script ... it's working too.

Comment 8 Oleksij Rempel (fishor) 2005-10-18 00:47:34 UTC

I tryed second one, result negative.

Comment 9 Oleksij Rempel (fishor) 2005-10-18 00:48:18 UTC

Both together don't help too.

Comment 10 Len Brown 2005-10-19 18:29:44 UTC

I'm confused about the kernel version.
The Kernel version box above says 2.6.12, 2.6.13, 2.6.14-rc3
Does the problem exist or not exist in those kernel versions?

Is there a version where the problem does not exist?
If yes, what is the most recent version where there is no problem?

Comment 11 Oleksij Rempel (fishor) 2005-10-19 22:21:43 UTC

This is tested wersions and all of them with this problem. 

> Is there a version where the problem does not exist?
i didn't found it. I think no.

Sory if it's confusing you.

Comment 12 Oleksij Rempel (fishor) 2005-10-31 01:07:39 UTC

I ting Bug 5398 depends on power managment like this one too.

Comment 13 Natalie Protasevich 2008-03-14 23:31:27 UTC

Alexey, please confirm if bug exists with latest kernel.