Hi. > Well, the logs are partially usable, the first one fails probably > before the image data is sent from the scanner (Broken pipe), and the > second does just show only the very beginning of a scan sequence. > > However, what is very curious, is that the first log shows the Pixma > dialect used here is different from all the ones implemented in the > pixma backend. > But the first scanimage log you sent (which fails when image data is > about to be sent by scanner) uses the ImageClass Pixma dialect, so I > guess this model understands 2 different Pixma protocols ! Canon seems > to have reinvented the wheel again !
Hmm, that's funny indeed. Maybe to clarify, I do not know if it is related or not. The scanner has two modes, "distant scan" and "connected to a computer". Not quite sure of the translation. I attach a new log in "computer" mode under windows, but I fear the log is incomplete, even if the scan was ok. I compile your modified version of pixma_imageclass.c and attach a new log (scan.log). The scanner has the same behaviour as before. Hope you could find some clue. Cheers. Mickael. -------------- next part -------------- A non-text attachment was scrubbed... Name: scan.log Type: text/x-log Size: 18332 bytes Desc: not available URL: <http://lists.alioth.debian.org/pipermail/sane-devel/attachments/20100724/bcef5e2f/attachment-0001.bin> -------------- next part -------------- A non-text attachment was scrubbed... Name: USBLog3.usblog Type: application/octet-stream Size: 954 bytes Desc: not available URL: <http://lists.alioth.debian.org/pipermail/sane-devel/attachments/20100724/bcef5e2f/attachment-0001.obj>
