MOTU Midi Express XT driver

MOTU Midi Express XT driver

[Torsten Schenk]

Hello all,

I'm currently trying to write a driver for the MOTU Midi Express XT. The problem I have is that the device has same idVendor and idProduct as the MOTU FastLane Quad MIDI Interface (for which already a driver exists in snd-usbmidi). If I do an lsusb, I see that the field iProduct identifies the device as a Midi Express XT. But I found no way of matching to iProduct in the usb_device_id structure. Is there a possibility to do so? Otherwise I'm thinking of patching snd-usbmidi in a way that it differentiates between the devices there.

Thanks,
Torsten


Here the lsusb output:
Bus 004 Device 003: ID 07fd:0001 Mark of the Unicorn FastLane Quad MIDI Interface
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               1.00
  bDeviceClass          255 Vendor Specific Class
  bDeviceSubClass         1
  bDeviceProtocol         0
  bMaxPacketSize0         8
  idVendor           0x07fd Mark of the Unicorn
  idProduct          0x0001 FastLane Quad MIDI Interface
  bcdDevice            1.32
  iManufacturer           0
  iProduct              119 MIDI Express XT
  iSerial                 0
  bNumConfigurations      1
...

Re: MOTU Midi Express XT driver

[Clemens Ladisch]

Torsten Schenk wrote:
> I'm currently trying to write a driver for the MOTU Midi Express XT.

Isn't that one of those insane devices that use isochronous transfers?

> The problem I have is that the device has same idVendor and idProduct
> as the MOTU FastLane Quad MIDI Interface (for which already a driver
> exists in snd-usbmidi).

Match on bDeviceSubClass.


Regards,
Clemens

Re: MOTU Midi Express XT driver

[Torsten Schenk]

 > Isn't that one of those insane devices that use isochronous transfers? 
There are 4 interfaces, two that look equal (isoc for output, interrupt for input), one with interrupt for in and out and one with isoc for out and in. Seems you can choose... Actually it should reduce latency if you use isoc? Would be my guess.

 > Match on bDeviceSubClass. 
Ok, I found your post from 2005 where you did that for the Fastlane already.

Thanks, Torsten

Re: MOTU Midi Express XT driver

[Clemens Ladisch]

Torsten Schenk wrote:
> > Isn't that one of those insane devices that use isochronous transfers?
> 
> There are 4 interfaces, ...  Seems you can choose.

IIRC the Windows driver uses only iso transfers; I don't know if the
others are implemented (or used for other things than MIDI).

> Actually it should reduce latency if you use isoc?

Iso packets must be scheduled in advance, so it increases latency.


Regards,
Clemens