This message is from the T13 list server.
< Hale Landis stated>
"But Mr Hatfield's document does provide a good explaination of the
problem. And everyone should realize that if a new ID word is added that
contains the "logical sector offset", the value in that word will never
be the "correct value" - my quess is at least 50% of the time it should
be a different value. So that means it will need to be added to the
drive ID data that DCO can modify. Then you have yet another "data
integrity" problem."
Questions arise:
1) Why would it 'never be the "correct value" ?
Is this referring to the fact (acknowleged) that different
operating environments require different values ?
What if there were a tool that would 'fix' your file system
to match the alignment of the device that is actually present ?
Such a tool would, of course, have to be 'very' careful about
what it does, but there exists software today that mucks around
with partitions and such and finds a way to protect user data.
2) The proposal states that the alignment is not changeable in the field
So, a DCO to change it would be prohibited.
There would be no data integrity issue due to changing the alignment.
Thank You !!!
-----------------------------------------------------------------
Jim Hatfield
Seagate Technology LLC
e-mail: [EMAIL PROTECTED]
s-mail: 389 Disc Drive; Longmont, CO 80503 USA
voice: 720-684-2120
fax....: 720-684-2711
==========================================
Hale Landis
<[EMAIL PROTECTED]
m> To
No Phone Info [EMAIL PROTECTED]
Available cc
Subject
06/03/2005 05:49 Re: [t13] e05120r1 Logical Sector
AM Alignment
Please respond to
Hale Landis
<[EMAIL PROTECTED]
i.com>
[EMAIL PROTECTED] wrote:
> This message is from the T13 list server.
> Please see the new revision of Seagate's "Logical Sector Alignment"
> proposal which was just posted.
>>http://t13.org/docs2005/e05120r1-ACS-logical_sector_alignment.pdf
It is sad that this document is even necessary. It is sad that someone
thinks LBA 0 should not be the "first sector in the first physical
sector". But I'm not surprised... I've never found anyone at the world's
largest OS vendor that seems to understand that larger physical sectors
(containing one than one 512-byte logical sector) would have an impact
on how their "fdisk" and "format" programs layout file systems and on
how the file system drives would access the data in the file system.
...Sigh...
But Mr Hatfield's document does provide a good explaination of the
problem. And everyone should realize that if a new ID word is added that
contains the "logical sector offset", the value in that word will never
be the "correct value" - my quess is at least 50% of the time it should
be a different value. So that means it will need to be added to the
drive ID data that DCO can modify. Then you have yet another "data
integrity" problem.
Hale
--
++ Hale Landis ++ www.ata-atapi.com ++
