© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
Marking this as won't fix as the more significant outstanding issues were resolved, namely:
1. Some issues causes by incorrect BIOS settings (see the case of Barteq's issues)
2. Opera causes a lot of wakeups that causes a lot of legitimate rescheduling interrupts - this is not a scheduler problem but an application issue
3. Given details on how to identify which rogue apps are causing a lot of wakeups (debug using top, vmstat, etc)
4. Any improvements upstream are significant patches back to 2.6.24, and the risk of upsetting the scheduler at this point in the release cycle is too high. Some less intrusive patches have been tried by the power consumption increases when reducing rescheduling interrupts.
5. Rescheduling interrupts shown to be a manifestation of other wakeup or interrupt issues (e.g. bad ACPI DSTD, bad apps, etc) and not necessarily a problem with the scheduler.
6. Hardy+1 will pick up a lot of the more subtle scheduler improvements in 2.6.25+