The USB controller and eth adapter share interrupts, but I make it hang equally well by copying stuff from DVDs onto the disk.
It's also a different system than the one from which you captured the sysfs files -- that one had a VT6202 (or maybe VT6212), this one had NEC CardBus EHCI.
See if that last patch (with NEC-specific workaround) makes a difference.
I can easily imagine a 100baseT link throwing your IRQ rate past the level I'd expect EHCI to produce for "big file" loads, even if the OHCI controllers also using that IRQ were idle!
I've had the drive in three different machines here; some small and old PII-350's with USB 1.1 and 2.4 kernel (load skyrockets, machines turns into paperweight fairly fast in around ~30 sec).
On a PIII-500 with both USB 1.1 and 2.0 exhibits same behaviour on 2.4 as above.
So, it's pretty clear the "load skyrockets" problem comes from something other than the HCD. As I commented when it was first reported.
- Dave
------------------------------------------------------- This SF.net email is sponsored by: Etnus, makers of TotalView, The best thread debugger on the planet. Designed with thread debugging features you've never dreamed of, try TotalView 6 free at www.etnus.com. _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
