On Saturday 28 August 2004 12:33 pm, Darren Marshall wrote: > On Sat, Aug 28, 2004 at 11:58:00AM -0700, David Brownell wrote: > > On Saturday 28 August 2004 11:40 am, Darren Marshall wrote: > > > David Brownell wrote: > > > > > > > > > > > > >This suggests a chip bug to me; there were others in > > > >AMD-756 OHCI. Does it work if you disable CONFIG_PM? > > > > > > > > > > > Disabling CONFIG_PM improved things slightly, I can now do 4 or 5 "cats" > > > before a hang occurs. > > > > Looks like there is inded a chip-related bug in that suspend/resume logic > > then. But that's not the only thing going on, since when you turned off > > all the related logic, you got another failure:
Hmm, erratum 10 seems the closest match to this, but it's not quite right: things were broken really early, before any kind of wakeup; and the chip never tried to wake up the OS. Can you post "lspci -vvx" info for all functions on your south bridge? (Run as root.) > > OK, so if it's not working in that case, then there's some other bug. > > As the "async" file empty at this point? > > > > > > > homer # cat /sys/class/usb_host/usb1/registers > > > bus pci, device 0000:00:07.4 > > >... > > > donehead 1730e180 > > > > This value suggests a lost IRQ ... the donelist should > > be flushed at most every 6 msec. Actually that's "lost" in the "never issued" sense ... the HC lost it, not the driver. > Yes, the "async" file is still empty. Huh, puzzling. I'll have to look at the bk1/bk2 delta to see if something suggests itself as a reason to hit an undocumted erratum on that old chip. (Is your board "Slot A" or "Socket A"?) - Dave ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
