Miscellaneous sas-r03g items

Elliott, Robert (Server Storage) Elliott at hp.com
Thu Apr 24 16:45:38 PDT 2003


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
*
Here are the non-really-trivial items I've received on
sas-r03g with proposed changes.


SL_IR Stop SNTT trigger
-----------------------
Identify Transmitted should be Identify Received.

In 7.9.5.3.3.1, change:
After this state receives an Identify Transmitted message, it shall
send a Stop SNTT request to the phy layer and initialize and start
the Receive Identify Timeout timer. If an Identify Received message
is received before the Receive Identify Timeout timer expires,
this state shall:

to:
After this state receives an Identify Received message, it shall
send a Stop SNTT request to the phy layer.

After this state receives an Identify Transmitted message, it shall
initialize and start the Receive Identify Timeout timer. If an
Identify Received message is received before the Receive Identify
Timeout timer expires, this state shall:



Timer descriptions in Connection sections
-----------------------------------------
a) in 7.12.7 Breaking a connection, add a copy of the Break Timeout
timer sentence.  This section is for breaking an existing
connection, which is different from aborting a connection request. 
Its table refers to the Break Timeout timer, but that timer is not
currently introduced in that section.

b) in 7.12.8 Closing a connection, add a sentence about the Close
Timeout timer: "After transmitting CLOSE, the source phy shall
initialize a Close Timeout timer to 1 ms and start the Close
Timeout timer."

and add a "Close connection responses" table listing the two responses:
"CLOSE:
	This confirms that the connection has been closed.
Close Timeout timer expires:
	The originating phy shall attempt to break the connection (see
7.12.7)."

That would parallel the Connection responses, Aborting connection
request, and Breaking connection sections which all have similar tables.

c) Move 7.12.7 Breaking a connection section after 7.12.8 Closing a
connection section.  Since the order (when both are attempted) is
to try CLOSE first, then try BREAK, the sections should be so ordered.

d) In 7.12.7, delete "If the Break Timeout timer expires before a
break response is received, the source phy may assume the physical
link is unusable."

so the Break Timeout timer paragraph becomes just a Break Timeout
timer sentence.

That sentence is not particularly good advice for aborting a
connection request (it is reasonable advice for breaking a
connection).  An error in the SOF or EOF will lead to that
scenario - the recipient SL or XL is still in idle, so doesn't
reply to the BREAK.


Disparity
---------
6.3.3.1 still says phys must use negative disparity after power
on. This requirement was removed elsewhere.

Change:
After powering on, the transmitter shall initialize the current
RD to negative.

to:
After powering on, the transmitter may initialize the current
RD to positive or negative.


SP state machine diagram
------------------------
page 126, figure 68, The transition out of SP14:SAS_Fail to
SP2:OOB_AwaitCOMX  should be to SP1:OOB_AwaitCOMX   

The text in 6.7.4.7.2 is correct.


etc.
----
4.6.7.5 Expander route index order

Replace << etc. >> in step 5 with:
"for each n greater than 3, devices attached to level n-1 edge
expander devices, except for level n-2 edge expander devices,
are considered level n."


TTS4 to TTS5
------------
9.2.6.3.3.5.2 Transition ST_TTS4:Receive_Data_Out to
ST_TTS5:Prepare_Xfer_Rdy (from proposal 03-164r3)
Change:
b) this state has received a Receive Data-Out transport protocol
service request and first burst data, all of the first burst
data has been processed, but the first burst data did not
satisfy the requested byte count.

to:
b) this state has received a Receive Data-Out transport protocol
service request, first burst is enabled, all the first burst data
has been processed, and the first burst data did not satisfy the
requested byte count.



--
Rob Elliott, elliott at hp.com
Hewlett-Packard Industry Standard Server Storage Advanced Technology
https://ecardfile.com/id/RobElliott



*
* 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