offset count and IU transfers

Richard Moore r_moore at qlc.com
Fri Aug 18 09:58:06 PDT 2000


* From the T10 Reflector (t10 at t10.org), posted by:
* Richard Moore <r_moore at qlc.com>
*
The reason for the requirement in data group transfers is that,
if multiple outstanding CRC REQs are allowed, initiators will
have to maintain an additional counter for the offset of the
first CRC and the offset of the second CRC (and an additional
counter for each additional CRC beyond the first two). This is
not an issue in packetized, where target and initiator keep
independent CRC interval counters and no offset tracking is
required for the CRC.

 -- Richard Moore
    QLogic Corp.

>-----Original Message-----
>From: Sriram Srinivasan [mailto:srirams at lsil.com]
>Sent: Friday, August 18, 2000 9:33 AM
>To: t10 at t10.org
>Subject: offset count and IU transfers
>
>
>* From the T10 Reflector (t10 at t10.org), posted by:
>* Sriram Srinivasan <srirams at lsil.com>
>*
>   SPI-4 (rev0) requires that the offset go to zero between 
>SPI IUs in DT DATA 
>OUT phases (see last-but-one para. in 10.8.3.3.2) and nothing 
>is mandated in the 
>other direction.  Whereas in data group transfers, the REQs 
>for pad or pCRC 
>bytes cannot be sent for the current data group until ACKs 
>have been received 
>for the pCRC bytes of the previous data group regardless of 
>the direction of 
>data transfer (see 4) in 10.8.3.3.6 and 3rd para. in 
>10.8.3.3.7).  Why is this 
>different between IU transfers and data group transfers?
>   
>   Thanx,
>   Sriram
>   
>
>*
>* For T10 Reflector information, send a message with
>* 'info t10' (no quotes) in the message body to majordomo at t10.org
>
*
* 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