[openib-general] How about ib_send_page() ?

[Libor Michalek]

On Mon, May 16, 2005 at 03:54:18PM -0700, Jeff Carr wrote:
> Libor Michalek wrote:
> > On Mon, May 16, 2005 at 03:26:57PM -0700, Jeff Carr wrote:
> > 
> >>It seems to me it would be useful to have a simple ib_send_page() function.
> >>
> >>This is essentially what I'm going to end up writing for what I need IB 
> >>to do. If there is anyone else that has similar needs or interests I'd 
> >>be happy to work with them.
> >>
> >>The CM works well enough to allow me to initiate connections between 
> >>HCA's. I'm now going to spend some time trying to write something to 
> >>kmalloc pages and test the different ways to initiate IB communication 
> >>between the hosts. Hopefully I can get something smart enough to perform 
> >>better than IPoIB does.
> > 
> > FYI, this is what SDP in regular sockets mode is doing. The internal
> > buffers are allocated using __get_free_page and then userspace data
> > is copied to/from the buffer which are transmitted/received by posting
> > send/receive on the allocated pages.
> 
> OK. Well I would rather make something generic. Besides, wasn't there 
> some MS patent issue? The last thread on the subject that I read kinda 
> made it sound like you were going to look into the issue and respond. 
> Maybe I missed the response; there's a lot of mail in the archives for 
> this list.

  Sorry, I discussed it at the openib.org workshop. The short answer is
that after going through the code we didn't feel that the license was
necessary. The question is whether there is any code which relies on MS 
patents, aside from trying to prove a negative, it does not appear to
be the case.

> In any case, I'd rather just make something more generic that perhaps 
> can be used with NBD.

  I wasn't trying to discourage you, just let you know what existing code
is doing, either as a solution, or as an example.

-Libor