P1394 Mail Archive: Re: P1394> MAX_?2?_DATA_SIZE(Was Re: Revised PWG1394 Cmd Set)

Re: P1394> MAX_?2?_DATA_SIZE(Was Re: Revised PWG1394 Cmd Set)

Greg Shue (gregs@sdd.hp.com)
Fri, 24 Jul 1998 09:32:52 -0700 (PDT)

It was acknowledged at the July meeting that these parameters
do NOT need to go across the wire in order to have a functioning
protocol.

A strong desire was expressed to have these parameters go
across the wire in order for initiator and target to more
effeciently use their respective memory resources. These
are there as a concession to that desire.

MAX_I2T_DATA_SIZE indicates the maximum amount of
target-to-initiator data the target can buffer and retransmit for
this connection across an

MAX_T2I_DATA_SIZE is the maximum amount of target-to-initiator
data the target will buffer and retransmit across this
connection if the Task is implicilty or explicitly aborted
(e.g. Bus Reset or TASK SET ABORT). Any memory allocated by
the initiator beyond this amount is wasted, since it will never
be used. Similarly, if an initiator only provides memory
buffers much smaller than those indicated by the target, the
target may be able to manage it's internal memory more
effeciently.

MAX_I2T_DATA_SIZE is there for symmetry.

Greg Shue, HP

Akihiro Shimura wrote:

> > B) MAX_I2T_DATA_SIZE (31 bits unsigned int)
> > C) MAX_T2I_DATA_SIZE (31 bits unsigned int)
>
> It is still not clear for me what is specified for initiator and/or
> target by these parameters.
>
> Size of the "buffer" referred by data_descriptor field of an ORB will
> be determined along with requesting client of the initiator. Once the
> ORB is appended to task list, access right to the "buffer" is granted
> to the target, and the "buffer" looks like (remote) receiving/sending
> buffer of the target. The target can access any part of the "buffer"
> within the address range of the "buffer" in arbitrary size unit the
> target wants to handle.
>
> I think the size of the "buffer" referred by the ORB and the size
> unit that target processes the "buffer" are independent of each other.
>
> Could you explain why these parameters need to go across the wire?
>
> Akihiro Shimura

-- 
Greg Shue
Hewlett-Packard Company
All-in-One Division			        gregs@sdd.hp.com
----------------------------------------------------------------