Return-Path: Received: from mail-wm0-f45.google.com ([74.125.82.45]:37304 "EHLO mail-wm0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753396AbbKXOoh (ORCPT ); Tue, 24 Nov 2015 09:44:37 -0500 Received: by wmww144 with SMTP id w144so29437693wmw.0 for ; Tue, 24 Nov 2015 06:44:35 -0800 (PST) Subject: Re: [PATCH v1 3/9] xprtrdma: Introduce ro_unmap_sync method To: Chuck Lever References: <20151123220627.32702.62667.stgit@manet.1015granger.net> <20151123221414.32702.87638.stgit@manet.1015granger.net> <20151124064556.GA29141@infradead.org> <565442F5.7080400@dev.mellanox.co.il> <4B2D7C66-31AC-44F3-A8CC-22CC7136015C@oracle.com> Cc: Christoph Hellwig , linux-rdma@vger.kernel.org, Linux NFS Mailing List , Sagi Grimberg From: Sagi Grimberg Message-ID: <565477CC.5070309@dev.mellanox.co.il> Date: Tue, 24 Nov 2015 16:44:28 +0200 MIME-Version: 1.0 In-Reply-To: <4B2D7C66-31AC-44F3-A8CC-22CC7136015C@oracle.com> Content-Type: text/plain; charset=windows-1252; format=flowed Sender: linux-nfs-owner@vger.kernel.org List-ID: Hey Chuck, > >> It is painful, too painful. The entire value proposition of RDMA is >> low-latency and waiting for the extra HW round-trip for a local >> invalidation to complete is unacceptable, moreover it adds a huge loads >> of extra interrupts and cache-line pollutions. > > The killer is the extra context switches, I?ve found. That too... > I?ve noticed only a marginal loss of performance on modern > hardware. Would you mind sharing your observations?