Furthermore I've experienced that this problem does *not* occur
with my ScanExpress 12000SP if at most 2 read_data commands are
queued (even with 600 * 1200 dpi ^2). However, this way scanning
becomes incredible slow. (the scanner has only 128kB internal buffer)
The only cure I'know at the moment would be increasing the max.
request size from 127kB to 2MB, say.
But I don't know whether this is possible with the Linux sg.
Any kernel hacker around?
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com