On Thursday 25 May 2006 09:03, Jose Paulo Moitinho de Almeida wrote:
>.......
>
> Using the version in the package (build 182) I had problems that seemed
> very similar to your log, but with a (much) older version (build 99) I
> could a scan without problems. From files that I found laying around I
> would say that build 153 was also working.
>
> In any case I promise to try a more rigorous test later on.
Yesterday I tested several versions and I think that at least one of the
problems has to do with the calibration method, when the backend chooses
between "one command" or "send data channel by channel". The driver is using
the second and hangs. I forced the driver to use the first method and it
seems to work.
Rene, if you please check in the correct voodoo (AV_ONE_CALIB_CMD or
AV_MULTI_CALIB_CMD or ...) I can check more thoroughly how the code behaves.
---------------------------- Old (but working)
out_size = format->pixel_per_line * 2;
/* send data in one command? */
if (format->channels == 1 ||
( (dev->hw->feature_type & AV_ONE_CALIB_CMD) ||
! BIT(format->ability1, 0) ) )
/* one command (most scanners) */
{
-------------------------- New (but hanging)
out_size = format->pixel_per_line * 2;
/* send data in one command? */
/* FR: HP5370 reports one-pass, but needs multi (or other format in single)
*/
if (format->channels == 1 ||
( ( (dev->hw->feature_type & AV_ONE_CALIB_CMD) ||
! BIT(format->ability1, 0) ) &&
! (dev->hw->feature_type & AV_MULTI_CALIB_CMD) ) )
/* one command (most scanners) */
{
-------------------------------------------------------
Regards
ZP
