Comment 21 for bug 709245
Revision history for this message
| warmcat (andy-warmcat) wrote Re: panda: USB disk IO slow | #21 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
a60fb26
(Get the code!)
Maybe the 1ms threshold thing is just to do with a threshold heuristic where it stops using the scheduler and sleeping and instead spins udelay() style, hence the 100% CPU. That's not to write off its effect on the bug just a possible explanation of why it cares about 1ms or less.
Reading these new findings yesterday and today I wonder if it's somehow related to when the ehci stuff schedules the URB. Like if the CPU is totally up it can pack USB transactions to go out immediately, but somehow when "idle", although what exactly is in what kind of idle is unclear, it bumps it to the next frame or so.