multi-touch-dev team mailing list archive

[Rafi Rubin <rafi@xxxxxxxxxxxxxx>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



Vikram Dhillon wrote:
> On Wed, Mar 31, 2010 at 06:09:40PM -0400, Rafi Rubin wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Just an FYI, Stéphane and I have been seeing an odd behavior where the n-trig
>> hardware stops sending events.  The devices look like they should be functioning
>> normally and just show no sign of activity (no events coming down from the usb
>> device).
>>
>> I've seen this behavior twice now, the first time just after changing the
>> firmware, and it didn't register, and a second time just yesterday.  I have not
>> had a chance to figure out how to encourage it to fail more regularly.
>>
>> Stéphane has been seeing it more regularly since a firmware update on his HP
>> last week.  Hopefully its deterministic enough to identify the source and a
>> solution quickly.
>>
>>
>> We just want you to be aware there is a potential problem.
>>
>>
>> On a side note we are both wondering if there is (and where to find) a
>> wiki/table of known hardware and its characteristics (physical/supported
>> features, stability testing status, etc).
>>
>> Rafi
> 
> Thanks for the update on this Rafi, I don't have a multi-touch (yet) but I have seen some similar behaviour on a regular touchpad lately. I wonder if similar problems are occuring on both architectures. It seems to be like a hardware interrupt (the thing responsible for sending the x,y coordinates after some z seconds) is unable to send those through the buffer. But without the architecture its not possible to say if that's exactly it. 

What brand/type of touchpad do you have, and is it serial or usb?

I've tried unloading a number of modules and unbinding the usb device, then
binding and loading.  That does not fix the stuck behavior.  I would think that
should have cleared most buffers.

Next time I will also try unloading and resetting the usb module and hosts.

I also want to try some of signals that the windows driver sends it during reset
and mode changes.

> About the wiki that you were talking about, there are a couple of wikis on different hardwares that are present but none on the current state of how they work, one of the ways I can think of collecting more info. would be checking mailing lists like this one. I will look into opensolaris and freebsd side to see if I get anything useful. 

For now I'm mostly interested in picking a good way to present our observations
going forward.

Also whatever testing pages we have for QA should point to whatever database of
support status is most appropriate.

Rafi
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkuz4O8ACgkQwuRiAT9o609CTgCg9ub166r/LF61haZSsFvMxvxr
HZgAnjNyGvB3JizpWlf/AsAL3l0e/cs0
=4r1m
-----END PGP SIGNATURE-----