From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [PATCH 39 of 39] IB/ipath - use streaming copy in RDMA interrupt handler to reduce packet loss Date: Thu, 29 Jun 2006 17:28:50 -0700 Message-ID: <44A47042.8060203@hp.com> References: <1151618377.10886.23.camel@chalcedony.pathscale.com> <20060629.150319.104035601.davem@davemloft.net> <1151624063.10886.34.camel@chalcedony.pathscale.com> <20060629.164623.59469884.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: bos@pathscale.com, akpm@osdl.org, rdreier@cisco.com, mst@mellanox.co.il, openib-general@openib.org, linux-kernel@vger.kernel.org, netdev@vger.kernel.org Return-path: Received: from palrel11.hp.com ([156.153.255.246]:7064 "EHLO palrel11.hp.com") by vger.kernel.org with ESMTP id S964813AbWF3A2w (ORCPT ); Thu, 29 Jun 2006 20:28:52 -0400 To: David Miller In-Reply-To: <20060629.164623.59469884.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org > If you bypass the L2 cache, it's pointless because the next > agent (PCI controller, CPU thread, etc.) is going to need the > data in the L2 cache. > > It's better in that kind of setup to eat the L2 cache miss overhead in > memcpy since memcpy can usually prefetch and store buffer in order to > absorb some of the L2 miss costs. I thought that most PCI controllers (that is to say the things bridging PCI to the rest of the system) could do prefetching and/or that PCI-X (if not PCI, no idea about PCI-e) cards could issue multiple transactions anyway? rick jones