usb/121052: Microsoft Notebook Optical Mouse 3000 (model 1049) doesn't work

[Kai Wang]

On Mon, Feb 25, 2008 at 07:56:47AM -0800, Duane H. Hesser wrote:
> I am attaching a rather verbose connection report for an ms3000 which
> is generated by a revised ums driver which I hope to post to my ISP
> account within a couple of days.  I am also attaching a "mouse report"
> generated by a new program ("mouse_report") which takes advantage of
> information which the revised driver makes available.
> 
> (I don't recall whether these lists are allowing attachments, so I
> will post the files at
> 
>   http://accima.com/members/dhesser/fbsd_mouse_stuff/ms3000-report.txt
> and
>   http://accima.com/members/dhesser/fbsd_mouse_stuff/ms3000connectmsgs.txt
> 
> for those who wish to see).

The format of these report is really nice ;-)

> The revised driver is intended to support all mice which reasonably
> conform to the HID spec, including those which offer multiple input
> reports (and including the Microsoft 3000).
> 
> The 3000 has a couple of anomalies.  It reports a tilt wheel, but
> does not offer tilt.  It reports 5 buttons, but has only 3.  The
> mouse data is provided in an input with ID 17 (0x11), but a private
> input report with ID 21 (0x15) is also provided whether it is
> requested or not, at the polling frequency.  The mouse ignores any
> attempt to "set_idle(interface,0,0) which is intended to cause
> input reports to be sent only when there is new data.  The odd thing
> is that the "real" input report behaves properly, but the private
> one won't shut up.  The revised driver ignores the private report.

To figure out what these other report IDs really does, we could probably
sniff the USB traffic under Windows with the mouse driver installed.

Kai