More Streaming device questions

Burn Alting burn at goldweb.com.au
Thu Aug 26 14:58:33 PDT 2004


* From the T10 Reflector (t10 at t10.org), posted by:
* Burn Alting <burn at goldweb.com.au>
*
Kevin,

Thanks for your response.

Paul Entzel of Quantum pointed out that Section 4.2.11.3 of the standard
describes my error condition.

As I said to Paul, I must read the standard twice and then another to
times just to be sure .... although I'd suggest to the editor that
perhaps since 99% of Illegal CDB conditions seem to be described in the
relevant command description reference, perhaps 100% should be or a "see
section X.Y.Z for additional generic error conditions" be added.

Burn
On Fri, 2004-08-27 at 03:06, Kevin D Butt wrote:
> Burn,
> 
> The way this is supposed to work is:
> 
> A Block Length value of 0 indicates that only variable block transfers
> are allowed.
> If the Fixed bit is set to 1, the Block Length field set to 0, and the
> Transfer Length field is not 0, Check Condition status is returned
> with Illegal Field in CDB (5/2400h).
> 
> I have to admit that I don't find this in SSC-3.
> 
> Dave,
> 
> Do you know where this is specified?  If we have dropped it, or it is
> not specified, we should add it.
> 
> Thanks,
> 
> Kevin D. Butt
> Fibre Channel & SCSI Architect, IBM Tape Microcode, 
> 6TYA, 9000 S. Rita Rd., Tucson, AZ  85744
> Tie-line 321; Office: 520-799-5280, Lab: 799-5751, Fax: 799-4138,
> Email: kdbutt at us.ibm.com
> 
> 
> Burn Alting <burn at goldweb.com.au>
> Sent by: owner-t10 at t10.org
> 
> 08/25/2004 03:53 PM
>                To
> t10 at t10.org
>                cc
> 
>           Subject
> Re: More
> Streaming device
> questions
> 
> 
> 
> 
> * From the T10 Reflector (t10 at t10.org), posted by:
> * Burn Alting <burn at goldweb.com.au>
> *
> Can one of the SSC-3 guru's comment.
> 
> Is this a (perverse) case of the last paragraph on page 64 of
> T10/1611-D
> Revision 01b (12 Aug 2004)? The initiator has set the logical block
> size
> to zero and has issued a read with the fixed bit set. Is it a case of
> an
> incorrect-length logical block attempting to be read?
> 
> Burn
> 
> 
> On Thu, 2004-08-26 at 03:01, Tim Jones wrote:
> > * From the T10 Reflector (t10 at t10.org), posted by:
> > * Tim Jones <tjmac at tolisgroup.com>
> > *
> > I suspect that this should cause an ILI since the fixed bit changes
> the 
> > meaning of the transfer length field (1 = Blocks, 0 = Bytes).  But, 
> > since the tape drivers have to actually control the writing (rather 
> > than a user app - disclaimer: on Unix systems), the driver should be
> > aware of the state of the fixed condition and modify its CDB 
> > accordingly.
> > 
> > Tim
> > --
> > Tim Jones                                                          
> tjmac at tolisgroup.com
> > 
> > On Aug 25, 2004, at 3:35 AM, Burn Alting wrote:
> > 
> > > * From the T10 Reflector (t10 at t10.org), posted by:
> > > * Burn Alting <burn at goldweb.com.au>
> > > *
> > > Yet more questions.
> > >
> > > When I start my tape target, I set the logical block size to a
> default
> > > value.
> > >
> > > For my block limits, I set my minimum block length to 512 and my 
> > > maximum
> > > to some large multiple of 512 bytes and since I want all IO in
> 512-byte
> > > quantities, I set the granularity field to be 9 (ie 2^9 = 512).
> > >
> > > If I want to perform IO in variable block mode, am I required to
> first
> > > set the logical block size (via a Mode Select changing the Block
> Length
> > > field in the Mode Parameter Block Descriptor) to zero and then
> issue my
> > > Read or Write commands with the fixed bit set to zero and the
> transfer
> > > length to be the number of bytes I want to transfer?
> > >
> > > If this is the case, ie I have set the logical block size to 0,
> what
> > > happens if a read or write command has it's fixed bit set to 1?
> > >
> > > Regards and thanks to all those who have been replying to my
> questions.
> > >
> > > Burn
> > >
> > > *
> > > * For T10 Reflector information, send a message with
> > > * 'info t10' (no quotes) in the message body to majordomo at t10.org
> > >
> > *
> > * For T10 Reflector information, send a message with
> > * 'info t10' (no quotes) in the message body to majordomo at t10.org
> > 
> 
> *
> * For T10 Reflector information, send a message with
> * 'info t10' (no quotes) in the message body to majordomo at t10.org
> 

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list