Henning Meier-Geinitz wrote: > Hi, > > On Sun, Sep 01, 2002 at 11:12:30PM +0200, abel deuring wrote: > >>>Can you do that? You are a lot more familiar with the SCSI code :-) >> >>In theory, no problem. I'm currently a bit busy with other stuff, so I >>might simply forget it... If nothing happened till, say, next sunday, >>feel free to kick me in the *** ;) > > > Ok. If you set a default value: after lots of testing the person with > the problem reported 99 seconds to be the maximum so we should set it > to 120 seconds. But with an environment variable I can also add a > comment to the manpage how to set it.
I've commited changes of sanei_scsi.c sane sane-scsi.man to the CVS server. The default timeout value is now 120 seconds; can be changed by setting the environment variable SANE_SCSICMD_TIMEOUT. I can test the stuff only on Linux, so it would be fine, if others could test, if I messed up the implementation for other OSes. Can be interesting to select a very small value ;) With one second timeout, the Sharp backend segfaults. (Didn't fix that yet...) Abel
