From owner-freebsd-scsi Tue Mar 7 6:50:16 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from bluebottle.calcaphon.com (calcaphon.demon.co.uk [193.237.19.5]) by hub.freebsd.org (Postfix) with ESMTP id 041D237BE53 for ; Tue, 7 Mar 2000 06:49:07 -0800 (PST) (envelope-from n_hibma@calcaphon.com) Received: from henny.webweaving.org (dhcp36.calcaphon.com [10.0.1.36]) by bluebottle.calcaphon.com (8.9.3/8.9.1) with ESMTP id OAA09667; Tue, 7 Mar 2000 14:49:50 GMT (envelope-from n_hibma@calcaphon.com) Received: from localhost (localhost [127.0.0.1]) by henny.webweaving.org (8.9.3/8.9.3) with ESMTP id OAA01427; Tue, 7 Mar 2000 14:46:42 GMT (envelope-from n_hibma@calcaphon.com) Date: Tue, 7 Mar 2000 14:46:42 +0000 (GMT) From: Nick Hibma X-Sender: n_hibma@localhost Reply-To: Nick Hibma To: Matthew Jacob Cc: FreeBSD SCSI Mailing List Subject: Re: first INQUIRY goes round in circles (fwd) 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 > > The solution to all this Jim pointed out was to not request sense after > > INQUIRY. This I can live with as it is a valid solution. If INQUIRY > > fails, the data stage of the USB transfer will fail and you have another > > indicator that things went wrong. > > Note that might be a case in which it's possible for an unadorned INQUIRY* to > generate a CHECK CONDITION- that's when there's a parity error on the IDENTIFY > message at the start of the command or a parity error sending the CDB. I > haven't done this level of SCSI for 10 years and don't have the spec in front > of me, but IIRC it's up to the target as to how it decides it wants to cope > with received bytes with parity errors so the mechanisms that cope with parity > errors while the command is active may not be used and a CHECK CONDITION with > the ASC/ASCQ may be latched up. Transport problems (so-called wire errors) are handled in the USB part and are signalled differently. They are reported as CAM_REQ_CMP_ERR. > *The second INQUIRY that FreeBSD's xpt_probe always sends that asks for the > VPD info will generally always get a CHECK CONDITION for devices that don't > support VDP info, so blindly saying "don't request sense after an INQUIRY" is > a Bad Thing To Do (tm) as it will leave the latched up Illegal Bit in CDB for > the next command to have to cope with- which will be the poor old Test Unit > Ready. In that case the command or data transport will fail. This means that SENSE will be requested. Nick -- n_hibma@webweaving.org n_hibma@freebsd.org USB project http://www.etla.net/~n_hibma/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message