From owner-freebsd-scsi Thu Mar 11 10:20: 4 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id B8DBD14D3D for ; Thu, 11 Mar 1999 10:20:01 -0800 (PST) (envelope-from mjacob@feral.com) Received: from localhost (mjacob@localhost) by feral.com (8.8.7/8.8.7) with ESMTP id KAA28919; Thu, 11 Mar 1999 10:19:21 -0800 Date: Thu, 11 Mar 1999 10:19:21 -0800 (PST) From: Matthew Jacob X-Sender: mjacob@feral-gw Reply-To: mjacob@feral.com To: Tom Torrance at home Cc: scsi@freebsd.org Subject: Re: 3.1-STABLE: nrsa0 T4000 doesn't honor "no rewind"? SCSI errs in logs In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > I mentioned this because a half-written tape is worse than useless > (fit only for the scratch bin). It might have some value if there > were a mechanism to close it off in some way. > > Definitely EOM should not trigger this routine. TM, trailer labels and TMs > have to be written, possibly following a short erase over the > reflective spot area. > EOM by itself doesn't trigger a 'loss of position'. However, FreeBSD (and NetBSD) insist on calling EOM (really 'early warning') an error and EIO has to be propagated back to the application. There was a lot of discussion about this some months back. The consensus (which I didn't agree with) was that EIO should still be propagated at early warning (the EOM bit in Sense Data- not the VOLUME OVERFLOW which is hard physical EOT) rather than using a (possibly deferred) residual count to an I/O operation to provide the signification. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message