On Sun, 4 Jun 2006, Dan Sandberg wrote: > Hi Alan, > > I tried doing the rmod/modprobe fix for the ehci-hcd problem but had no > luck. > > Anything else I can do to figure out the story? > > I also changed all the usb-related drivers to be modules rather than > compiled-in as they were originally. > > I re-confirmed that the drive works fine once linux has finished > booting. I also re-confirmed that the > same on-disk configuration ( kernel, initrd, ... ) boots fine on the > Lexar but not on the Sandisk. > > The Sandisk boots properly on my Thinkpad T30. So the problem seems to > only occur at boot-up ( or only when the Sandisk is the first drive to > be used with ehci-hcd ) on the VIA EPIA MII10000 motherboard.
You seem to be contradicting yourself. Above you confirmed that the drive works fine once Linux has finished booting, but here you say that the problem occurs when the Sandisk is the first drive to be used with ehci-hcd. So if you boot over the network and then plug in the Sandisk, what happens? > Here's how I modified the /init inside the initrd to do what you suggested: > > log_begin_msg "Loading modules" > load_modules > log_end_msg > > log_begin_msg "Unload and reloading ehci-hcd" > rmmod ehci-hcd > rmmod ehci-hcd > sleep 20 > modprobe ehci-hcd > sleep 10 > log_end_msg At this point I'm baffled. This seems to be some weird combination of hardware and/or driver bugs, but I have no idea where they are or how to find them. You wouldn't happen to have access to a USB bus analyzer? That could at least show whether the problem is on the host side or the device side. Apparently the easiest solution is to replace the Sandisk with another Lexar unit. Alan Stern _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
