From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Talpey Subject: Re: [PATCH RFC 2/3] svcrdma: Use device rdma_read_access_flags Date: Tue, 10 Nov 2015 10:16:34 -0500 Message-ID: <56420A52.1080309@talpey.com> References: <1447152255-28231-1-git-send-email-sagig@mellanox.com> <1447152255-28231-3-git-send-email-sagig@mellanox.com> <20151110114145.GA2810@infradead.org> <5641D920.5000409@mellanox.com> <20151110120432.GA8230@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Chuck Lever , Christoph Hellwig Cc: Sagi Grimberg , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org On 11/10/2015 9:34 AM, Chuck Lever wrote: > 2. Is there a reason IB is not using the allphys MR for RDMA Read? > That would be much more efficient. On the NFS server that MR isn't > exposed, thus it is safe to use. True, but only if the layout of the buffer's physical pages do not exceed the local RDMA Read scatter limit. If the size is large then the privileged MR will require additional RDMA Read operations, which can be more expensive. It's certainly a valid optimization, if the buffer "fits". -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html