Opened 8 years ago

Closed 7 years ago

#1423 closed defect (fixed)

USB power managment.

Reported by: jg Owned by: marcelo
Priority: high Milestone: Trial-2
Component: kernel Version:
Keywords: power Cc: rsmith, cjb
Blocked By: Blocking:
Deployments affected: Action Needed:
Verified: no

Description (last modified by rsmith)

This bug is to track issues pertaining to USB power management. First, we need to confirm that our USB class drivers work properly. Marcelo was working on the stack a while back.

This is an active area of Linux development at the moment, so it will likely need continual reverification in the tinderbox.

Arguably, this may need to interact with a power manager daemon, to set policy as to whether the system should suspend.

And how USB devices that can suspend should deal with our fast suspend/resume is anyone's guess.

Its not clear yet on just how to measure this. USB +5 is switched by U56 and there is not a sense resistor there. One possibility is to break the line at the +5V regulator @ L28 and then do subtract out the current across R397. That would give you whats happening on USB. We could take a baseline with no USB devices plugged in and subtract out that as well.

Change History (5)

comment:1 Changed 8 years ago by jg

  • Keywords power added

comment:2 Changed 8 years ago by marcelo

  • Verified unset

USB devices are now detected after resume.

There are still a few problems remaining, most notably that the wireless interface does not come
back from the dead sometimes. Looking into it now.

comment:3 Changed 8 years ago by jg

Have you tried the new firmware from Cozybit of last week?

comment:4 in reply to: ↑ description Changed 8 years ago by rsmith

  • Description modified (diff)

Replying to jg:

This bug is to track issues pertaining to USB power management. First, we need to confirm that our USB class drivers work properly. Marcelo was working on the stack a while back.

This is an active area of Linux development at the moment, so it will likely need continual reverification in the tinderbox.

Arguably, this may need to interact with a power manager daemon, to set policy as to whether the system should suspend.

And how USB devices that can suspend should deal with our fast suspend/resume is anyone's guess.

comment:5 Changed 7 years ago by marcelo

  • Resolution set to fixed
  • Status changed from new to closed

Marking as fixed, if people encounter problems with particular devices they will
open separate tickets.

Note: See TracTickets for help on using tickets.