Subject: Re: st.c broken ? (may have been: Re: CVS commit: src/sys - remove b_flags manipulation)
To: Frank Kardel <kardel@netbsd.org>
From: Juergen Hannken-Illjes <hannken@eis.cs.tu-bs.de>
List: source-changes
Date: 08/25/2007 15:58:15
On Sat, Aug 25, 2007 at 12:35:48PM +0200, Frank Kardel wrote:
> Juergen Hannken-Illjes wrote:
> >On Sat, Aug 25, 2007 at 10:24:01AM +0200, Frank Kardel wrote:
> >  
> >>Hi Andrew, * !
> >>
> >>I am experiencing quite a few strange error with my DAT tape since about
> >>this checkin. A kernel from the beginning of July works flawlessly with 
> >>the
> >>drive and the tape.
> >>
> >>[snip]
> >>    
> >[snip]
> >
> >Same here, reading a tape file yields EIO on end-of-file.  Looks like the
> >change to physio_done() removed the EOM handling.  Before we got
> >
> >    mbp->b_error = 0;
> >    mbp->b_flags |= B_ERROR;
> >
> >now we get
> >
> >    mbp->b_error = EIO;
> >
> >  
> Neat - seems that there is a little bit left to do in order to be able 
> again to append to tapes...
> Do we need an internal EEOM error code now ?

At least scsi disks are also affected.  Reading a raw partition also
gives an EIO at end of partition.

-- 
Juergen Hannken-Illjes - hannken@eis.cs.tu-bs.de - TU Braunschweig (Germany)