On Wed, 2007-04-04 at 01:29 -0300, Marcelo Tosatti wrote: > On Mon, Apr 02, 2007 at 04:56:08PM -0600, Jordan Crouse wrote: > > Hi Richard - > > > > Power management is proceeding along nicely, and we're finally to the > > point where we can start to discuss the design and implementation of the > > OLPC Power Management Daemon. > > > > We have started to list some possible real-world power management scenarios > > here: > > > > http://wiki.laptop.org/go/Power_Management_Scenarios > > > > We should start to discuss how to go about implementing these in the power > > manager - we have a bunch of eager B2 owners out there ready to start saving > > power! :) > > > > Everybody, please feel free to comment on the page I linked above - we've > > only defined those scenarios that are currently within our reach, but there > > are others that we eventually want to support. > > The userspace burn-CPU-madness which happens after resume is partially > due to removal/addition of the eth0,msh0 and lo files in /sys/class/net, > which in the end triggers ifdown/ifup and who knows what else
This is due to (I think) udev rules triggering the network service in /etc/init.d/network. That is actually not necessary if we're running NetworkManager, which handles all that via HAL anyway. So we can 'chkconfig --level 123456 network off' and it shouldn't be a problem any more. Dan > inbetween. The down/up procedures are not necessary if this was a fast > suspend/resume, but necessary if it was a long suspend/resume (so we > possibly moved the machine to another access point network and need to > rerun dhclient). > > This should be handled by udev/HAL/PMD - need to get work on that > started ASAP. > _______________________________________________ > Devel mailing list > [email protected] > http://mailman.laptop.org/mailman/listinfo/devel _______________________________________________ Devel mailing list [email protected] http://mailman.laptop.org/mailman/listinfo/devel
