From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH for-next 02/10] IB/iser: Default to fastreg instead of fmr Date: Tue, 17 Nov 2015 12:09:26 +0200 Message-ID: <564AFCD6.309@mellanox.com> References: <1447691861-3796-1-git-send-email-sagig@mellanox.com> <1447691861-3796-3-git-send-email-sagig@mellanox.com> <564ADA1C.4090905@mellanox.com> <564AF4CB.9060302@dev.mellanox.co.il> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <564AF4CB.9060302@dev.mellanox.co.il> Sender: target-devel-owner@vger.kernel.org To: Sagi Grimberg , Sagi Grimberg , linux-rdma@vger.kernel.org, target-devel@vger.kernel.org Cc: "Nicholas A. Bellinger" , Steve Wise List-Id: linux-rdma@vger.kernel.org On 11/17/2015 11:35 AM, Sagi Grimberg wrote: >> On 11/16/2015 6:37 PM, Sagi Grimberg wrote: >>> This is a pre-step before adding remote invalidate support. >> >> Wouldn't that introduce performance regression on ConnectX3 devices? > > With remote invalidate it shouldn't make much of a difference. Why? the invalidate is just one part of the story, we are doing a mapping on IO submission and CX3 has strong ordering on FRWRs, right? > Plus, I'd really like us to start phasing out from FMRs... We should make sure not to introduce performance regression for HW which has such a big existing install base and is well selling in 2015/16