Regarding MMC4 01g

keiji_katata at keiji_katata at
Mon Apr 7 21:10:33 PDT 2003

* From the T10 Reflector (t10 at, posted by:
* keiji_katata at

Hi Bill,

I think that MMC4 should say "DBD shall be set to 1."
Actually SPC describes DBD bit. So MMC shall keep the same description and
MMC can have some addition. Pioneer's pure parallel SCSI DVD-R/RW device
supports DBD bit. The drive is MMC device.

In case of MMC device, it is recommended that host application checks BD
length field in header to check BD existence.

Best regards,

Keiji Katata

Bmcferrin at on 2003/04/08 07:30:36

???:     owner-t10 at

??:  mike at
cc:    mtfuji5 at, T10 at
??:  Re: Regarding MMC4 01g

Hello Mr. Ito,
Neither Mode Select nor Mode Sense were specified in MMC1, MMC2, or MMC3,
so MMC4 is the first case where the statement has been made.

It is true that for several revisions, Mt Fuji has specified that DBD shall
be set to one.

With respect to the MODE SENSE command, SPC-2 states: "A disable block
descriptors (DBD) bit of zero indicates that the device server may return
zero or more block descriptors in the returned MODE SENSE data.  A DBD bit
of one specifies that the device server shall not return any block
descriptors in the returned MODE SENSE data."

CASE 1: Drive supports no block descriptors
If the drive does not implement block descriptors, then the drive will
return the same information regardless of the setting of the DBD bit.  In
this case, DBD may be ignored.

Case 2: Drive supports one block descriptor
Since 8020, only "Legacy SCSI" (i.e. SCSI-2) drives are permitted to have a
block descriptor.  In this case, DBD has genuine meaning.  Hoever, in the
main text of MMC-4, we do not wish to address legacy implementations.

Changes to MMC-4:
1. I need to put a sub-clause into ANNEX X - LEGACY SPECIFICATIONS to
describe the legacy case for Block Descriptors.

2. To be precise, I think that I should change the "shall ignore DBD" to
"may ignore DBD".

Kind Regards,
Bill McFerrin

 Hi Bill McFerrin, how are you doing ? This is Ito of Fujitsu.
   Now, I have one comment for mmc4 document.
 MMC4 01g MultiMedia Command Set - 4 (MMC-4) Bill McFerrin PDF (2753278)

 - mmc4r01g.pdf  page240 says;
     Multi-Media Logical Units shall ignore the disable block descriptors
     (DBD) bit. Multi-Media Logical Units do not return Block Descriptors.

 ==> This is not good idea.
     I think;
     1. A lot of devices by which DBD is returned are circulating to the
     2. Other OS and the application to expect BD to return exist.
     3. There is no problem because BIOS can correctly acquire data even if
        device responds BD or not.

   When did you change it ? Bestway is to fix same as Mt.Fuji documentation
 for it
 or previous if it's possible. Please verify Fuji documentation.
 And why change it, please let me know.

 Best regards,
 Mike Ito(Fujitsu),

 Masahiro Ito, Yamagata Fujitsu Limited
 E-Mail:mike at
 FONE:237-43-7150(DID) FAX:237-43-7155

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

More information about the T10 mailing list