#10943 NORM 11.2.0-: powerd not detecting system activity to the extent it probably should
Zarro Boogs per Child
bugtracker at laptop.org
Thu Jun 2 11:19:54 EDT 2011
#10943: powerd not detecting system activity to the extent it probably should
----------------------------------------------+-----------------------------
Reporter: greenfeld | Owner: pgf
Type: defect | Status: new
Priority: normal | Milestone: 11.2.0-final
Component: power manager (powerd) | Version: Development build as of this date
Resolution: | Keywords:
Next_action: diagnose | Verified: 0
Deployment_affected: | Blockedby:
Blocking: |
----------------------------------------------+-----------------------------
Comment(by pgf):
this may be related to improvements made to the video subsystem. i think
jnettlet reported far less cpu used in screen activity -- he was talking
about record, i guess, but perhaps the improvements affect other things as
well? i think we need a better idea of the specific kinds of places where
we sleep now where we didn't sleep before to understand the issue.
as for the startup ring: powerd won't sleep if the dcon is frozen. on
normal boots, the dcon isn't thawed until sugar is running or gnome is
being started from /usr/bin/olpc-session (there's also a 30 second
failsafe timeout). once the dcon has been thawed, the normal 15 second
idle timeout must elapse before we'll sleep. the timing of the first boot
is different, than normal boots, so i guess that should be investigated.
as always, a powerd trace ("powerd-config =trace-on", or "powerd-config
=trace-debug") will help.
--
Ticket URL: <http://dev.laptop.org/ticket/10943#comment:1>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system
More information about the Bugs
mailing list