Re: Avoiding deadlock in iSCSI

["Matt Wakeley" <matt_wakeley@agilent.com>]

I don't think VI/TCP helps at all.  The VI is implemented on top of a TCP
stream.  If the TCP stream delivers the iSCSI command to VI, and VI has no
place to put it, what is VI going to do?  It has to block the TCP stream, and
that in turn will block and "RDMA" from occurring.

-Matt

Stephen Byan wrote:

> Jim Williams [mailto:jimw@giganet.com] wrote:
>
> > Layering iSCSI on VI avoids the problem below.
> >
> > see:
> > http://search.ietf.org/internet-drafts/draft-csapuntz-ips-iscs
> > ivi-00.txt
> >
> > VI layer flow control will be invoked when target is unable
> > to accept more
> > commands, but this will not impede flow of data in response
> > to RDMA read
> > requests by the target.  Functionally this is equivalent to
> > option 1 below.
>
> I haven't seen much discussion of iSCSI over VI here. I think it makes
> sense. Using VI makes the iSCSI remote DMA hardware acceleration available
> to other applications besides iSCSI. This broader market should greatly
> increase the utility of iSCSI hardware, and thus speed the market
> penetration of iSCSI, bringing the usual benefits of customer acceptance,
> lower cost, and better implementions courtesy of darwinian competition.
>
> And, as an added bonus, you get flow control on commands for free :-)
>
> Regards,
> -Steve
>
> Steve Byan
> <stephen.byan@quantum.com>
> Design Engineer
> MS 1-3/E23
> 333 South Street
> Shrewsbury, MA 01545
> (508)770-3414
> fax: (508)770-2604