Re: 答复: MM: Non QMI Huawei modems without PPP support?

[Olof Ermis <olof ermis gmail com>]

Hi

lsusb -t gives:

root@oloflinux:/sys/bus/usb/drivers/cdc_ncm# lsusb -t
2-6:1.1: No such file or directory
/:  Bus 08.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 07.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 06.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 05.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
    |__ Port 1: Dev 2, If 0, Class=HID, Driver=usbhid, 1.5M
    |__ Port 1: Dev 2, If 1, Class=HID, Driver=usbhid, 1.5M
    |__ Port 2: Dev 3, If 0, Class=HID, Driver=usbhid, 1.5M
/:  Bus 04.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 03.Port 1: Dev 1, Class=root_hub, Driver=uhci_hcd/2p, 12M
/:  Bus 02.Port 1: Dev 1, Class=root_hub, Driver=ehci_hcd/6p, 480M
    |__ Port 6: Dev 6, If 0, Class=vend., Driver=option, 480M
    |__ Port 6: Dev 6, If 1, Class=vend., Driver=, 480M
    |__ Port 6: Dev 6, If 2, Class=stor., Driver=usb-storage, 480M
    |__ Port 6: Dev 6, If 3, Class=stor., Driver=usb-storage, 480M
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=ehci_hcd/6p, 480M


Maybe this strange

Port 6: Dev 6, If 1, Class=vend., Driver=, 480M

is the problem or? Some driver there with no name? That is 2-6:1.1 right?

I'll give the debugging a shot as well.

Cheers
Olof


2012/11/12 Bjørn Mork <bjorn mork no>:
> Olof Ermis <olof ermis gmail com> writes:
>
>> Update:
>>
>> When loading manually using
>>
>> modprobe -v cdc_ncm
>>
>> I get this in syslog:
>>
>> Nov 12 14:49:23 oloflinux kernel: [ 1840.329871] usb 2-6: bind() failure
>> Nov 12 14:49:23 oloflinux kernel: [ 1840.329900] usbcore: registered
>> new interface driver cdc_ncm
>>
>> So it looks like cdc_ncm somehow tried to attach the dongle there (usb
>> 2-6) but failed.
>
> Yes.  OK, then it makes sense again.
>
>
> We obviously overlooked something.  But that can be fixed.  We just need
> to find out where and why it fails.  It worked for Tommy, which seemed
> likely when looking at the code, but reading code and trying to imagine
> the flow often fails.  For me at least...
>
> Could you add a few extra debugging printk's throughout the
> cdc_ncm_bind() function? There are quite a few "goto error;" and
> "goto error2;".  Could you replace each of these with something which
> makes it possible to pinpoint exactly which one we take?  Something like
> this, but with a unique error message per goto would do:
>
> {
>     dev_err(&intf->dev,"exit 1\n");
>     goto error;
> }
>
>
>
>
> Bjørn
>