Re: Framing Discussion

[Mohan Parthasarathy <Mohan.Parthasarathy@eng.sun.com>]

> 
> >Buffer being allocated by the user application is just one case
> >where applications are doing raw read. I understand that it
> >is a problem when buffers are pre-allocated. But is this
> >the problem space we are addressing here ? 
> 
> Yes, this is what that needs to be addressed. SCSI applications always
> pre-allocate the buffer before sending out the SCSI read request.
>
So, the zero copy problem addresses a specific API model. It
was not obvious reading the document that was sent. May be
this was discussed before in this mailing list.

> 
> >2) iscsi layer posts commands and willing to receive buffers
> >   handed by the transport layer. In this cases, i would assume
> >   the iscsi layer can easily separate the data belonging to
> >   the different commands it posted. I don't see why you need
> >   any other extra mechanism here.
> 
> Since the application has already pre-allocated the buffer, if the
> data is received from the transport layer is in a different buffer,
> the iscsi layer has to copy the data into the application buffer.
>
Again you are assuming the traditional API model, when you say
"application has already pre-allocated the buffer".

-mohan
 
> -JP
>