Re: [PATCH] Trivial, add AO subdevice to comedi_test

On Thu, 15 May 2003, Frank Mori Hess wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Thursday 15 May 2003 10:05 am, Calin A. Culianu wrote:
> > Hi guys,
> >
> > I found it useful when devloping/testing comedi applications to
> > sometimes have an AO subdevice available in the comedi_test driver.
> > It's pretty useless, as all it does is loop back AO to the AI channels,
> > and doesn't even support AO commands or anything, but it at least lets
> > one use the comedi_test driver with applications that require analog
> > input _and_ analog output.
> >
> > Let me know if this patch is useful.  It isn't very invasive and
> > hopefully it can make it into comedi cvs...
> >
>
> It's useful, a couple things should be changed though.  You should insert
> the ao readback into the fake_waveform function, adding a new ai channel
> for each ao channel readback.  That way, it will work with ai commands
> also.  Then make the ai insn get its data by calling fake_waveform with
> the current time (this is a fix that isn't really related to your changes,
> but it should be done).  Your waveform_ai_insn_read should be made the ao
> read insn.  If you don't want to do this stuff yourself, tell me and I'll
> try and get around to it in a few days.

Yeah, you're right.. I re-read the code and see that it would be nice for
the regular ai insn's to get their data from the waveform generators, as
well as the loopback to somehow apply itself to commands as well.  Ok.. i
will work on this in the next few days if I find time..

Thanks Dave for applying the patch as is, and I will try and make it more
useful by incorporating Frank's suggestions.  I just need some time to do
this because I have to first read the current waveform generator code and
figure out how it works fully before I invasively modify it :)

-Calin

Received on 2003-05-16Z13:43:50