[Thread moved to linux-usb-devel, which seems more appropriate]

On Wed, 21 Jun 2006, Maulik Mankad wrote:

> Hello Alan,
> 
> I am a member of the Linux usb user group and am very new to USB device 
> drivers
> and USB specifications.
> 
> I am working on USB host driver porting for Intel's IXDP465 network processor.

Good for you.  It's not an easy job...

> I am using Snap gear Linux Kernel 2.6.12, which runs on IXDP465.
> 
> I am facing the following issues: -
> 
> (1) The USB bus enumeration does not complete fully. The request for getting 
> the
> device descriptor fails with timeout error.
>     "device descriptor read/64, error -110"
> 
> (2) My USB host controller is EHCI compliant. Although it supports only full 
> and
> low speed mode. There are neither transaction translators nor companion
> controllers available with the USB host.

This sounds backward.  EHCI supports only high speed, not full or low 
speed.  If the host doesn't have full-speed capability then it is not 
USB-2.0 compliant.

> (3) I have ported the host driver using the host controller registers. The 
> host
> gets initialized properly and I do get interrupts on plugging / unplugging of 
> a
> USB device (pen drive) to the device port. I also do get the SOF’s occurring
> down the enabled port.
> 
> (4) The asynchronous schedule also executes properly as I get interrupts for
> asynchronous schedule advance.
> 
> (5) The host is not able to set the address of the device. This request also
> times out with error "usb 1-1: device not accepting address 3, error -110".
> 
> The request for the device descriptor and SET_ADDRESS gets submitted properly 
> in
> the queue. But somehow the timer times out and a function timeout_kill () gets
> called. This function then
> calls a function usb_unlink_urb (), which aborts the request and removes it 
> from
> the queue. Does any one know why this can happen?

Certainly.  The timer expires because the request doesn't complete quickly 
enough.  Probably because a bug in your driver prevents the request from 
completing at all.

What does a bus trace reveal?

> Can anyone let me know what can be the issue in this case? I am struggling 
> with
> the above issue from three weeks now. Also has anyone worked with IXDP465 USB
> host controller?
> Please share your experience.
> 
> Please let me know if you need more details.
> 
> Thanks,
> Maulik Mankad

Alan Stern



_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to