[t13] Fwd: why not H = C = D ModeSense6 for just the header

Hale Landis hlandis at indra.com
Wed Feb 12 12:27:53 PST 2003


This message is from the T13 list server.


This message to both T13 and T10 - Hale.

On Wed, 12 Feb 2003 10:52:36 -0800, Eschmann, Michael K wrote:
>This message is from the T13 list server.
>[...]
>You'll see that the allocation length (or Parameter List Length)data doesn't line up, so good drivers will convert from 6-to-10 and pad the CDB out (with zeros) to a 12 or 16 byte CDB so bad drives (many, I suppose) will function.

All this tells me is that the "host", aka the application and the OS
driver stack, must understand the command and what the command
requires and what action(s) the command performs. This is especially
true if the OS driver stack is converting the command to some other
command (or set of commands), or, if the OS driver stack "emulates"
the command in host software. 

None of this has anything to do with the physcial interface (ATA or
SCSI) and how that interface transfers data for a command. The "host"
and device shall understand what the command means and both shall
have the same expectations for any "data stream" transferred by that
command on the physical interface.

It sounds to be like Microsoft may have some homework to do,
especially if they do not prohibit applications from generating I/O
requests that the OS driver stack can not correctly execute.

Hale



*** Hale Landis *** www.ata-atapi.com ***






More information about the T10 mailing list