https://bugzilla.kernel.org/show_bug.cgi?id=214021

--- Comment #5 from Chris Chiu (chris.chiu@canonical.com) ---

I add printks to trace where the error happens. The dmesg show as down below

[   44.488980] xhci_hcd 0000:00:14.0: FIXME: allocate a command ring segment
[   44.488981] usb_alloc_dev: hcd driver alloc dev exist but return 0
[   44.488982] usb usb1-port4: couldn't allocate usb_device

The first line show for the "xhci_dbg(xhci, "FIXME: allocate a command ring
segment\n");" in xhci_alloc_dev()

The second line shows the printk I add in usb_alloc_dev()

       /* Root hubs aren't true devices, so don't allocate HCD resources */
        if (usb_hcd->driver->alloc_dev && parent &&
                !usb_hcd->driver->alloc_dev(usb_hcd, dev)) {
+               printk("%s: hcd driver alloc dev exist but return 0",
__func__);
                usb_put_hcd(bus_to_hcd(bus));
                kfree(dev);
                return NULL;
        }

The usb_hcd->driver->alloc_dev is equal to xhci_alloc_dev and fails the
xhci_queue_slot_control(). Does that mean anything regarding this?

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.