On Thu, 23 Sep 2004, Oliver Neukum wrote:

> > Sorry, I didn't express myself very well. �What I meant was, there needs 
> > to be a way to exempt some devices (such as the swap device) from an 
> > initial pass during system-wide suspend, then suspend them during a second 
> > pass (after the memory image has been written).
> 
> 1. Where is the advantage of special casing the swap device only?

Nowhere.  The swap device isn't special cased.  The exemption mechanism 
should be general: any device can go on the second pass, not only swap 
devices.

> 2. What distinguishes the first pass from the second in the view of
> a device?

Nothing.  Devices and drivers aren't aware of the difference between 
passes.  The only difference is that the PM core suspends devices on one 
list first, then it suspends devices on the other list.

> 3. Isn't exporting the knowledge which devices are swap devices
> a layering violation?

Yes.  So that knowledge isn't exported.  All that's exported (imported
actually, into the PM core from somewhere else) is a set of devices that
need to go on the second list.

Do you know how the current code handles these issues?  I don't.

Alan Stern



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to