Bug 141957 - kpowersave doesn't start
Summary: kpowersave doesn't start
Status: RESOLVED FIXED
Alias: None
Product: SUSE Linux 10.1
Classification: openSUSE
Component: Other (show other bugs)
Version: Beta 5
Hardware: Other Other
: P5 - None : Normal (vote)
Target Milestone: ---
Assignee: Thomas Renninger
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-01-08 00:48 UTC by Andreas Schwab
Modified: 2006-04-21 14:04 UTC (History)
1 user (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Andreas Schwab 2006-01-08 00:48:52 UTC 
When the system does not have ACPI then kpowersave exists right away.
Comment 1 Thomas Renninger 2006-01-09 10:27:36 UTC 
This may be because /dev/apm does not exist on PPC anymore?
I will have a look ...
Comment 2 Thomas Renninger 2006-01-09 10:51:19 UTC 
Strange, kpowersave should check for /proc/apm, which should also exists on a G4/5 PPC. If neither /proc/acpi nor /proc/apm exists, it's not worth to start kpowersave (and also powersaved), this is intended.
Comment 3 Thomas Renninger 2006-01-09 10:53:20 UTC 
Andreas can you check for /dev/apm_bios (seems to not exist anymore on new PPC versions, but shouldn't matter anymore) and /proc/apm (this one must exist on a non ACPI system to get (k)powersave started).
Comment 4 Andreas Schwab 2006-01-09 12:10:51 UTC 
Why do you need apm or acpi?  It's completely useless for cpufreq.
Comment 5 Thomas Renninger 2006-01-12 14:16:40 UTC 
I don't know any machine that does cpufreq and not at least APM or ACPI.
The powersave daemon is written to at least have an APM or ACPI interface... Holger and I discussed to add a dummy interface/class, but until now I haven't seen a machine where this should be needed.
Comment 6 Andreas Schwab 2006-01-12 14:21:39 UTC 
Cpufreq has nothing at all to do with apm or acpi.  That's a fact.
Comment 7 Thomas Renninger 2006-01-12 14:26:55 UTC 
I know. Give me a machine where no APM and ACPI is provided and cpufreq works and we can add the dummy interface.
Comment 8 Andreas Schwab 2006-01-12 14:32:46 UTC 
Just remove that useless check.
Comment 9 Holger Macht 2006-02-13 11:26:40 UTC 
The powersave daemon should run even in the no acpi/no apm case. Does kpowersave any checks for this, Danny?
Comment 10 Danny Al-Gaaf 2006-02-28 10:25:45 UTC 
I do not remove the check now for CODE 10, because this is complete untested and we don't know currently if there sidekicks within KPowersave. 

I has saw your patch for 0.5.8. This patch is incorrect/incomplete. I enable the patch only for ppc and i64 (I don't know ix86/x86_64 machines where this is needed), but note: this is complete untested and I don't support the patch atm.
Comment 11 Holger Macht 2006-02-28 11:14:05 UTC 
But can't see any sidekicks atm. We also do not exit in the daemon in the "no apm, no acpi" case. There should be enough time to remove the patch if promblems come up. I still would go for removing this check ;-) But it's up to you Danny...
Comment 12 Thomas Renninger 2006-04-21 14:04:59 UTC 
Why is this one set to "needinfo from dkukawka"?
All modern x86_64/i386 machines have acpi so adding this for ia64/ppc should be sufficient. New x86_64/i386 machines need ACPI to get cpufreq working anyway. I just close this now.
Reopen if there still should be concerns.