MMC / Mt Fuji: Should host software retry I/O on sense code 03/02/00 (no seek complete)?

Keiji_Katata katata at avc-pioneer.com
Wed Oct 1 07:16:55 PDT 2008


* From the T10 Reflector (t10 at t10.org), posted by:
* "Keiji_Katata" <katata at avc-pioneer.com>
*
Hello David,
I think "03/02/00 – no seek complete" is widely used in DVD-RAM writable 
drive (maybe BD-RE too) to avoid some OSs IO timeout on Read10/12 command. 
When a drive cannot response to a Read command within 5 sec (or such time 
length), the drive may return "03/02/00 – no seek complete" when 5 sec (or 
such time length) was passed from the Read command received. Because the 
drive cache is full by write data or the drive performs replacement of the 
write data to spare area and is updating DFLs.
The some OSs may retry 5 times maximum with the same Read command. Then 
drive can have 30 sec for to be ready for the read operation.
Some OSs and a famous UDF writing software driver accept this operation.
I think that Pioneer drives should support this operation because the famous 
UDF writing software driver was bundled.
I have two questions to you.
1. Do you want to suspend the releasing the final version of Fuji 7 until 
you finish your examination of this "03/02/00 – no seek complete"?
Because we will close Fuji 7 discussion on this Friday after releasing the 
final version.
2. Do you want to add some notes or description about this "03/02/00 – no 
seek complete" in Fuji document?
Best regards,
Keiji Katata
PIONEER CORP.
----- Original Message ----- 
送信者 : David Burg
宛先 : mtfuji5 at avc-pioneer.com ; t10 at t10.org
Cc: Ope Aladekomo ; David Walp
送信日時 : 2008年9月30日 7:41
件名 : MMC / Mt Fuji: Should host software retry I/O on sense code 03/02/00
(no seek complete)?
Hello Mt Fuji and MMC members,
I would like to hear the opinion from optical device manufacturers with
regards to the sense code 03/02/00 – no seek complete. Is this a fatal
error
which will be fail again always if the same command is issued again by the
host, or, is this an error that may be recovered from if the same command is
issued again by the host (assume the media is not changed)?
Microsoft would like this information in order to better implement Windows
and make sure that no unnecessary delays is introduced for the user under
this condition.
I understand the behavior may depend of a vendor’s specific implementation
and would like to hear what each device vendor implements. If you do not
feel comfortable discussing your particular implementation in this public
distribution list, feel free to respond directly to Microsoft.
With regards,
David Burg,
Microsoft.
*
* 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