>> init timing. Experiment: leave all debug messages off, but change
>> the first dbg() call in hc_reset() into an err() call. Does that make
>> things better?
>
>
> The answer is yes.
I'll submit a patch to make that arbitrary timeout longer, and make
sure the debug messages won't affect that timing again.
> I feel changing dbg() in err() is a bit worse than full
> CONFIG_USB_DEBUG=y. It certainly did hang more often at boot than with
> CONFIG_USB_DEBUG=y. However :
> * with err() I get about 50% boot chance
> * with err() I've never so far booted with a useless keyboard (sole
> times I booted with unchanged kernel and CONFIG_USB_DEBUG=n keyboard was
> dead)
> * when it hangs with err() the takeover message is printed (never was
> before on hang)
I think there's something funky about your BIOS, causing these boot
time problems. That takeover problem should _never_ happen. Can you
still get BIOS updates for that motherboard?
> Sometimes after 2.5.43 is booted switching to the console freezes the
> usb mouse. Don't know if it's related to the boot hang, but
> chain-restarting gpm will more often result into an oops than a
> recovered mouse. However I've just found that re-pluging it instead of
> restarting gpm unfreezes the mouse cursor.
That'd seem to be a different problem.
> drivers/usb/host/ohci-q.c: 00:07.4 bad entry 5fb7d1e1
> drivers/usb/host/ohci-q.c: 00:07.4 bad entry ffffffe1
> drivers/usb/host/ohci-q.c: 00:07.4 bad entry 5fb7d1e1
> drivers/usb/host/ohci-q.c: 00:07.4 bad entry ffffffe1
> drivers/usb/core/usb.c: USB disconnect on device 4
> drivers/usb/core/hub.c: new USB device 00:07.4-2.3, assigned address 6
> input: USB HID v1.10 Mouse [Logitech USB Mouse] on usb-00:07.4-2.3
>
>
> The « drivers/usb/host/ohci-q.c: 00:07.4 bad entry » are triggerred by
> gpm failing to restart.
Now that's the wierdest clue to that failure I've seen yet! :)
Good that for you it didn't seem to cause other trouble. I'm
starting to think I probably know where that problem must live.
- Dave
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Oct 23 2002 - 22:00:51 EST