FC-AL-2, FC-DA: out-of-order frame reception

Robert Snively rsnively at brocade.com
Wed Oct 23 13:50:35 PDT 2002


* From the T10 Reflector (t10 at t10.org), posted by:
* Robert Snively <rsnively at brocade.com>
*
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C27AD5.D6080DF0
Content-Type: text/plain;
	charset="iso-8859-1"

There is no known requirement for out-of-order 
implementations in FC-AL-2 or FC-DA, whether 
switch attached or directly attached.  While it 
is possible to design switches that cannot achieve 
in-order delivery, all present switches provide that 
capability and all present disk drives require it. 
Some host adapters, some RAIDs, and the FCP-2 
recovery protocols do allow for out-of-order behavior, 
but the FCP-2 recovery protocol implementations have typically been 
simplified based on the assumption of in-order delivery. 

The net is that out-of-order is not a problem because 
no one does it. 

Bob 

> -----Original Message----- 
> From: Kastein, Kurt [ mailto:kurtk at lsil.com  ] 
> Sent: Wednesday, September 25, 2002 8:46 AM 
> To: T10 Reflector (E-mail) 
> Subject: FC-AL-2, FC-DA: out-of-order frame reception 
> 
> 
> * From the T10 Reflector (t10 at t10.org), posted by: 
> * "Kastein, Kurt" <kurtk at lsil.com> 
> * 
> Has there been discussion about FC-AL-2 target devices 
> needing to support 
> out-of-order frame reception when the loop is part of a 
> fabric?  i.e. Are 
> future switch architectures NOT going to ensure in-order 
> frame transmission? 
> Is this being addressed within FC-DA? 
> 
> Regards, 
> Kurt 
> 
> ************************************************************** 
> ************** 
> * Kurt Kastein               |  kurtk at lsil.com                
>              * 
> * Hardware Design Engineer   |  (970)206-5827                 
>              * 
> * LSI Logic                  |  (970)215-7469 (cell)          
>              * 
> * 2001 Danfield Court        |  (970)206-5280 (FAX)           
>              * 
> * Fort Collins, CO  80525                                     
>              * 
> ************************************************************** 
> ************** 
> * 
> * For T10 Reflector information, send a message with 
> * 'info t10' (no quotes) in the message body to majordomo at t10.org 
> 


------_=_NextPart_001_01C27AD5.D6080DF0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">

 RE: FC-AL-2, FC-DA: out-of-order frame reception There is no known requirement for out-of-order 
implementations in FC-AL-2 or FC-DA, whether 
switch attached or directly attached.  While = it 
is possible to design switches that cannot = achieve 
in-order delivery, all present switches provide = that 
capability and all present disk drives require = it. 
Some host adapters, some RAIDs, and the FCP-2 = 
recovery protocols do allow for out-of-order = behavior, 
but the FCP-2 recovery protocol implementations have = typically been 
simplified based on the assumption of in-order = delivery. The net is that out-of-order is not a problem = because 
no one does it. Bob > -----Original Message----- 
> From: Kastein, Kurt [mailto:kurtk at lsil.com] 
> Sent: Wednesday, September 25, 2002 8:46 = AM 
> To: T10 Reflector (E-mail) 
> Subject: FC-AL-2, FC-DA: out-of-order frame = reception 
> 
> 
> * From the T10 Reflector (t10 at t10.org), posted = by: 
> * ;Kastein, Kurt; = <kurtk at lsil.com> 
> * 
> Has there been discussion about FC-AL-2 target = devices 
> needing to support 
> out-of-order frame reception when the loop is = part of a 
> fabric?  i.e. Are 
> future switch architectures NOT going to ensure = in-order 
> frame transmission? 
> Is this being addressed within FC-DA? 
> 
> Regards, 
> Kurt 
> 
> = ************************************************************** 
> ************** 
> * Kurt = Kastein          ;    |  = kurtk at lsil.com         p;      
>         ;    * 
> * Hardware Design Engineer   |  = (970)206-5827         ;       
>         ;    * 
> * LSI = Logic           nbsp;      |  (970)215-7469 = (cell)          
>         ;    * 
> * 2001 Danfield = Court        |  (970)206-5280 = (FAX)           = 
>         ;    * 
> * Fort Collins, CO  = 80525           nbsp;           nbsp;           nbsp; 
>         ;    * 
> = ************************************************************** 
> ************** 
> * 
> * For T10 Reflector information, send a message = with 
> * 'info t10' (no quotes) in the message body to = majordomo at t10.org 
> 
------_=_NextPart_001_01C27AD5.D6080DF0--




More information about the T10 mailing list