--- Christoph Hellwig [EMAIL PROTECTED] wrote:
On Fri, May 27, 2005 at 06:24:46PM -0700, Caitlin
Bestler wrote:
Oh that's a great development plan for me to
propose
to my business people. Let's go write lots of
code,
and *after* we write it the IB people (our
competitors)
have agreed
Venkata,
How will that work? If the RNIC offloads RDMA and
TCP completely from the Operating System and does not
share any state information then the application
running on the host will never be in the position to
utilize the socket interface to use the communication
logic to send and receive
I completely agree with Michael's viewpoint here.
Doing a clean RDMA focussed implementation would bring
out the exact requirements that need to be changed.
Retrofitting IB/Infiniband based API into the RDMA
based subsystem just to reuse an existing base is not
a good start.
Thanks
SG
---
to be justfied. Did I read that
correctly?
Thanks
SG
--- Michael Krause [EMAIL PROTECTED] wrote:
At 06:40 AM 5/27/2005, Sukanta ganguly wrote:
Venkata,
How will that work? If the RNIC offloads RDMA
and
TCP completely from the Operating System and does
not
share any state information
I agree.
SG
--- Bernard Metzler [EMAIL PROTECTED] wrote:
Sukanta,
without touching any TOE issues (this question is
about RDMA, right?),
after
transforming a TCP connection into RDMA mode and
using an RDMA API,
the socket file descriptor is not longer to be used
for communication.
In
--- Caitlin Bestler [EMAIL PROTECTED] wrote:
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
On Behalf Of
Michael Krause
Sent: Friday, May 27, 2005 7:25 AM
To: Sukanta ganguly
Cc: openib-general@openib.org;
[EMAIL
Venkata,
I understand what you were saying now.
Thanks
SG
--- Venkata Jagana [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] wrote on
05/27/2005 10:25:36
AM:
Mike,
I am not sure I do understand what your were
trying
to communicate. Let me try and decode this. My
basic