[argyllcms] Re: tcgetattr failed with 'Input/output error'

  • From: Ludovico Fischer <ranucel@xxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sun, 09 Dec 2007 16:57:43 +0100

Yes, it is a revision D instrument. I tried the patch you suggested (I
suppose you meant i1pro.c instead of i1dissp.c, as there is no such line
in that file). With both the patched and unpatched code, if I disconnect
the eye-one pro, and reconnect it every time, it is always recognised.
With the patched code, the behaviour of 'dispread -k target.cal Screen'
now does not affect dispread -?, which always displays the instrument as
connected, whether the other command is run beforehand or not. But
'dispread -k target.cal -Screen', works only the first time after the
instrument is reconnected. Afterwards, it gives this error:
dispread: Error - Configuring USB port 'usb:/bus0/dev14 (GretagMacbeth
i1 Pro)' to 1 failed with -71 (could not set config 1: Protocol error).

To make it work again, you are now forced to disconnect and reconnect.
This happens even though dispread -? always displays
 1 = 'usb:/bus0/dev14 (GretagMacbeth i1 Pro)'



Ludovico Fischer
 
On Sat, 2007-12-08 at 11:26 +1100, Graeme Gill wrote:
> Ludovico Fischer wrote:
> > Ok, I had misdiagnosed the problem. Whether the tcgeattr error appears
> > is not dependent on whether I use a space or not. It seemed so only
> > because I was giving the two syntaxes always on the same order.
> 
> Right, so the instrument is not being recognized after being used.
> Is it recognized every time if you disconnect and reconnect the
> instrument before trying to use it ?
> 
> Which revision Eye-One Pro do you have (it's on the label on
> the bottom of the instrument) ?
> 
> [ I had a problem like this on Linux with the Rev D instrument,
> but I thought I'd worked around it by reseting the instrument
> rather than closing it. In fact, the Rev D would cause my version
> of Linux to crash without this workaround.]
> 
> Since you're compiling from source, you could try commenting out
> the line 100 in spectro/i1disp.c:
> 
>       usbflags |= icomuf_reset_not_close;
> 
> and see if that affects the behaviour.
> 
> Graeme Gill.
> 


Other related posts: