On Fri, May 19, 2000, Al Borchers <[EMAIL PROTECTED]> wrote:
> Linux is hanging when we test our USB serial converter
> digi_acceleport.c on an SMP machine.  We are running 2.3.99-pre9-2
> SMP on an ASUS P2B-D motherboard with two PII processors.  We
> are using usb-uhci.
> 
> We run a test that uses select/read/write to send data through
> one of our serial ports as fast as possible.  On a UP kernel
> the test does not hang the system, but on an SMP kernel the
> test hangs the system almost every time.  Both processors are
> hung, there is no response from the system, and we must reboot.
> 
> One processor is stuck in wait_on_irq() waiting for an IRQ handler
> to finish on the other processor.  The function wait_on_irq prints
> out a stack backtrace after waiting for a while; that stack trace,
> sent through ksymoops, is attached.  The trace shows the other
> processor is in usb-uhci/usbcore--I have not tried to track down
> what is happening here.  The stack trace on this other processor
> is quite long--the first few lines are
> 
> CPU Trace; c8824000 <_end+85147b0/8514800>
> Trace; c882c000 <[usbcore]__kstrtab_usb_bulk_msg+ca6/cf6>
> Trace; c882c8f8 <[usb-uhci]append_qh+0/44>
> Trace; c88313a1 <[usb-uhci].data.end+a/4d9>
> Trace; c882c428 <[usb-uhci]uhci_show_status+0/234>
> Trace; c88313ab <[usb-uhci].data.end+14/4d9>
> Trace; c882d948 <[usb-uhci]uhci_clean_iso_step1+0/34>
> Trace; c88313bc <[usb-uhci].data.end+25/4d9>
> Trace; c882d97c <[usb-uhci]uhci_clean_iso_step2+0/28>
> Trace; c88313d1 <[usb-uhci].data.end+3a/4d9>
> Trace; c882cb68 <[usb-uhci]insert_qh+0/a4>
> Trace; c88313e6 <[usb-uhci].data.end+4f/4d9>
> Trace; c882ccec <[usb-uhci]cleanup_skel+0/128>
> Trace; c88313f0 <[usb-uhci].data.end+59/4d9>
> Trace; c882ca10 <[usb-uhci]unlink_td+0/84>
> 
> I have also attached /var/log/messages with USB verbose debugging
> set.
> 
> Unfortunately I don't have much time to look into this right now--but
> I would be glad to provide any further information if anyone wants it.
> 
> Thanks for any help with this problem.

Have you tried using the alternate UHCI driver?

JE


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to