On Thu, 2007-09-06 at 09:33 +0200, Mattias Nissler wrote:
quoted text
> On Wed, 2007-09-05 at 16:44 -0400, Alan Stern wrote:
> > On Wed, 5 Sep 2007, Mattias Nissler wrote:
> > 
> > > Hey, I got another suspicion: Have you noticed that the UHCI and EHCI
> > > controller suspend/resume in different order in the two logs?
> > 
> > Not quite correct.  The controllers (1d.0 and 1d.7) are in the same 
> > order, because that is determined by the PCI stack.  However the root 
> > hubs (usb2, usb3, etc.) _are_ in a different order because it is 
> > determined by the USB stack.
> 
> Ah, right. I hadn't checked for the pci resume messages...
> 
> > 
> > >  That is
> > > probably because after I had made ehci-hcd a module, it was loaded
> > > after.
> > 
> > Yes, it is.
> > 
> > >  In my syslog I have verified that ehci is loaded before uhci when
> > > both are compiled in. Does that sound sensible? At least both drivers
> > > talk to interrelated hardware...
> > > 
> > > I'll try to compile them both as modules and control the load order.
> > 
> > Okay.  It may not be the order itself so much as the time delay 
> > between resuming the controller and resuming the root hub.  Since the 
> > controllers get resumed in the same order, the delays are different.
> 
> Seems you are right, it's the timing: This is a core 2 duo based
> machine, I have had it running overclocked with the FSB raised from 266
> to 400 Mhz. This has been perfectly stable for me during the last 4
> months. Now I have reduced the bus frequency back to more conservative
> values, and the usb problem is gone :-)
> 
> Anyway, I'll play with some sleeping, to see where the hardware needs
> more time. Will only make the driver more robust. However, I don't have
> much spare time at the moment, not sure when I get around to do it.

Oh, I forgot. Thanks for all your friendly help with this.

Mattias


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel