Subject: Re: kern/2086: SCSI tape support broken
To: der Mouse <mouse@collatz.mcrcim.mcgill.edu>
From: Matthias Scheler <tron@colwyn.owl.de>
List: netbsd-bugs
Date: 02/17/1996 16:22:38
der Mouse said:
> Anyone who sends a "bug report" with that little information,
> especially without stating all the negative information (like, no
> console messages) ...

Ok, next time I'll mention that there were no kernel messages.

> That's quite bizarre.  Something _somewhere_ along the line is
> returning EIO.

You are sure that this message is only generated for EIO? I only added
printf statements were I found "return EIO;" or something like that.

> Perhaps scatter printfs along the call chain from write() all the way down
> to st.c, and find out where the EIO is being generated?

Could you tell me which source files are involved, please?

> Define "broken", then.

See below.

> (Just as obviously not completely working, either.)

"Not completely working" means "broken" for me (no offence intended).

-- 
Matthias Scheler
tron@colwyn.owl.de