End-to-end logical block guard checking question

George Penokie gop at us.ibm.com
Fri Jan 23 09:11:40 PST 2004


* From the T10 Reflector (t10 at t10.org), posted by:
* George Penokie <gop at us.ibm.com>
*
This is a multipart message in MIME format.
--=_alternative 005E5F5F86256E24_=
Content-Type: text/plain; charset="US-ASCII"


Kieth, 

Yes there are things in the service delivery subsystem that could
possibly check the protection information. By definition things in the
service delivery subsystem are not defined by SCSI. As a result the
words you suggest are correct but as somewhat redundant to the
definition of a service delivery subsystem. But, more important they do
not address the issue in the current words that assumes all data is in
blocks. Which, if true, means that the logical block guard would always
be valid because CRC in that field is generated on the data in the
block. So if it's always valid then why does the standard say it may not
be valid? 

Bye for now,
George Penokie

Dept 2C6  114-2 N212
E-Mail:    gop at us.ibm.com
Internal:  553-5208
External: 507-253-5208   FAX: 507-253-2880





"Holt, Keith" <keith.holt at lsil.com> 


01/23/2004 10:44 AM 

To
George Penokie/Rochester/IBM at IBMUS, t10 at t10.org 

cc

Subject
RE: End-to-end logical block guard checking question

	





George, 
  
I agree that the current wording is kind of confusing.  Taken by itself,
the statement regarding the EMDP bit seems superfluous.  I assume that
statement is there because of the sentence in the first paragraph:
"Protection information is generated at the application layer and may be
checked by any object along the I_T_L nexus."  I assume that "any
object" includes objects other than the application client and device
server.  Is it a given that objects other than the application client
and device server have special considerations for out of order
transfers?  If so, then perhaps the text should instead read something
like the following: 
  
"If the logical unit is formatted with protection information and the
EMDP bit is set to one in the Disconnect-Reconnect mode page (see
SPC-3), then checking of the logical block reference tag or the logical
block guard within the service delivery subsystem by objects other than
the application client and device server may cause false errors because
logical blocks may be transmitted out of order." 
  
Keith 
  
-- 
Keith Holt 
LSI Logic Storage Systems Inc. 
keith.holt at lsil.com 
316-636-8665 
  
-----Original Message-----
From: George Penokie [mailto:gop at us.ibm.com]
Sent: Friday, January 23, 2004 8:55 AM
To: t10 at t10.org
Subject: End-to-end logical block guard checking question


In the protection information description in SBC-2 section 4.5.1
Protection information overview the following statement is made: 

If the logical unit is formatted with protection information and the
EMDP bit is set to one in the Disconnect-Reconnect mode page (see
SPC-3), then checking of the logical block reference tag or the logical
block guard within the service delivery subsystem may cause false errors
because logical blocks may be transmitted out of order. 

The statement that false errors may occur because blocks may be
transmitted out of order does not apply to the logical block guard as
that is a check of the block itself and does not care what the order of
the blocks is. Although it may be possible that a service delivery
subsystem may see data within a block the is not in the original order
that is not how the above statement reads. 

The question is does it make sense to make a statement about the logical
block guard causing false error? Is yes then the reason has to change
because as currently stated it is illogical.   

Bye for now,
George Penokie

Dept 2C6  114-2 N212
E-Mail:    gop at us.ibm.com
Internal:  553-5208
External: 507-253-5208   FAX: 507-253-2880



--=_alternative 005E5F5F86256E24_=
Content-Type: text/html; charset="US-ASCII"


<br><font size=2 face="sans-serif">Kieth,</font>
<br>
<br><font size=2 face="sans-serif">Yes there are things in the service
delivery subsystem that could possibly check the protection information.
By definition things in the service delivery subsystem are not defined
by SCSI. As a result the words you suggest are correct but as somewhat
redundant to the definition of a service delivery subsystem. But, more
important they do not address the issue in the current words that assumes
all data is in blocks. Which, if true, means that the logical block guard
would always be valid because CRC in that field is generated on the data
in the block. So if it's always valid then why does the standard say it
may not be valid?</font>
<br><font size=2 face="sans-serif"><br>
Bye for now,<br>
George Penokie<br>
<br>
Dept 2C6 &nbsp;114-2 N212<br>
E-Mail: &nbsp; &nbsp;gop at us.ibm.com<br>
Internal: &nbsp;553-5208<br>
External: 507-253-5208 &nbsp; FAX: 507-253-2880<br>
<br>
</font>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td width=40%><font size=1 face="sans-serif"><b>"Holt, Keith"
<keith.holt at lsil.com&gt;</b> </font>
<p><font size=1 face="sans-serif">01/23/2004 10:44 AM</font>
<td width=59%>
<table width=100%>
<tr>
<td>
<div align=right><font size=1 face="sans-serif">To</font></div>
<td valign=top><font size=1 face="sans-serif">George Penokie/Rochester/IBM at IBMUS,
t10 at t10.org</font>
<tr>
<td>
<div align=right><font size=1 face="sans-serif">cc</font></div>
<td valign=top>
<tr>
<td>
<div align=right><font size=1 face="sans-serif">Subject</font></div>
<td valign=top><font size=1 face="sans-serif">RE: End-to-end logical block
guard checking question</font></table>
<br>
<table>
<tr valign=top>
<td>
<td></table>
<br></table>
<br>
<br>
<br><font size=2 color=blue face="Arial">George,</font>
<br><font size=3>&nbsp;</font>
<br><font size=2 color=blue face="Arial">I agree that the current wording
is kind of confusing. &nbsp;Taken by itself, the statement regarding the
EMDP bit seems superfluous. &nbsp;I assume that statement is there because
of the sentence in the first paragraph: &nbsp;"Protection information
is generated at the application layer and may be checked by any object
along the I_T_L nexus." &nbsp;I assume that "any object"
includes objects other than the application client and device server. &nbsp;Is
it a given that objects other than the application client and device server
have special considerations for out of order transfers? &nbsp;If so, then
perhaps the text should instead read something like the following:</font>
<br><font size=3>&nbsp;</font>
<br><font size=2 color=blue face="Arial">"If the logical unit is formatted
with protection information and the EMDP bit is set to one in the Disconnect-Reconnect
mode page (see SPC-3), then checking of the logical block reference tag
or the logical block guard within the service delivery subsystem <b>by
objects other than the application client and device server</b> may cause
false errors because logical blocks may be transmitted out of order."</font>
<br><font size=3>&nbsp;</font>
<br><font size=2 color=blue face="Arial">Keith</font>
<br><font size=3>&nbsp;</font>
<br><font size=2 color=blue face="Arial">-- <br>
Keith Holt <br>
LSI Logic Storage Systems Inc. <br>
keith.holt at lsil.com <br>
316-636-8665 </font>
<br><font size=3>&nbsp;</font>
<br><font size=2 face="Tahoma">-----Original Message-----<b><br>
From:</b> George Penokie [mailto:gop at us.ibm.com]<b><br>
Sent:</b> Friday, January 23, 2004 8:55 AM<b><br>
To:</b> t10 at t10.org<b><br>
Subject:</b> End-to-end logical block guard checking question<br>
</font>
<br><font size=2 face="sans-serif"><br>
In the protection information description in SBC-2 section 4.5.1 Protection
information overview the following statement is made:</font><font size=3>
<br>
</font><font size=2 face="sans-serif"><br>
If the logical unit is formatted with protection information and the EMDP
bit is set to one in the Disconnect-Reconnect mode page (see SPC-3), then
checking of the logical block reference tag or the logical block guard
within the service delivery subsystem may cause false errors because logical
blocks may be transmitted out of order.</font><font size=3> <br>
</font><font size=2 face="sans-serif"><br>
The statement that false errors may occur because blocks may be transmitted
out of order does not apply to the logical block guard as that is a check
of the block itself and does not care what the order of the blocks is.
Although it may be possible that a service delivery subsystem may see data
within a block the is not in the original order that is not how the above
statement reads. </font><font size=3><br>
</font><font size=2 face="sans-serif"><br>
The question is does it make sense to make a statement about the logical
block guard causing false error? Is yes then the reason has to change because
as currently stated it is illogical. &nbsp;</font><font size=3> </font><font size=2 face="sans-serif"><br>
<br>
Bye for now,<br>
George Penokie<br>
<br>
Dept 2C6 &nbsp;114-2 N212<br>
E-Mail: &nbsp; &nbsp;gop at us.ibm.com<br>
Internal: &nbsp;553-5208<br>
External: 507-253-5208 &nbsp; FAX: 507-253-2880<br>
</font>
<br>
--=_alternative 005E5F5F86256E24_=--




More information about the T10 mailing list