Re: [RFC,PATCH 4/14] knfsd: has_wspace per transport

[Neil Brown <neilb@suse.de>]

On Thursday May 17, Thomas.Talpey@netapp.com wrote:
> At 06:30 AM 5/17/2007, Neil Brown wrote:
> >And I gather by the fact that you test "->sko_has_wspace" that RDMA
> >doesn't have such a function?  Do that mean that RDMA will never
> >reject a write due to lack of space?  That seems unlikely.
> 
> RDMA uses credits, which are preallocated per-connection resources in
> the hardware. RDMA transports don't generally support flow control, that
> is left to the upper layer. As such, if the client is obeying the protocol, the
> send will succeed. And if it's not, then the client only loses since the failure
> will cause the connection to be lost.
> 
> Someday we might want to support this kind of test. There are shared
> resource schemes in RDMA, more for the receive side, but it's not out of the
> question for sends. As a default though, a NULL has_wspace makes sense. 

Thanks for the explanation.
So presumably the serve communicates to the client how much buffer
space it has via these credits, so that we can be sure a write will
never fail (as long as the protocol is being honoured).  That sounds
reasonable.

I think I would rather
   int rdma_has_wspace(struct svc_sock *svsk) { return 1; }
than NULL.  It removes special cases in the generic code, and gives an
obvious place for a comment explaining that RDMA doesn't need flow
control because it uses credits.

NeilBrown

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs