On Mon, 26 Sep 2005, David Brownell wrote:

> > Okay, but this should be regarded only as a temporary fix.  The old code
> > went to some lengths to carry out an autosuspend from within an interrupt
> > context, and this just piles on more of the same.
> 
> Well, "irqs blocked"; not really an interrupt context.  And I still don't
> see a way to really get around blocking IRQs while entering the OHCI
> suspend state!!

"Atomic context", then.  You run into difficulty trying to do things like 
stop the LED-display work queue, and you duplicate code for stopping the 
root hub status URB.  None of that would be necessary if there was a 
process context available for doing a normal suspend.

> >       Instead, the
> > interrupt-level code should arrange for some thread to do a real suspend,
> > in process context.
> 
> Yet we don't have any such code yet ... and the current approach has worked
> for quite a while, modulo recent breakage fixed in these patches.  What's
> there is just bugfixes for that breakage.  :)
> 
> If we get such usbcore code, I'd much rather see it phase in without breaking
> OHCI; it kind of sucks that 2.6.14 is going to go out with it still broken.

I think it won't be long before we do have such code...

Greg, can you say when you expect to apply those patches from David?

Alan Stern



-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to