From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [RFC][PATCH v4 00/18] Provide a zero-copy method on KVM virtio-net. Date: Sun, 25 Apr 2010 02:55:29 -0700 (PDT) Message-ID: <20100425.025529.123989625.davem@davemloft.net> References: <1272187206-18534-17-git-send-email-xiaohui.xin@intel.com> <1272187206-18534-18-git-send-email-xiaohui.xin@intel.com> <1272187206-18534-19-git-send-email-xiaohui.xin@intel.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, mst@redhat.com, mingo@elte.hu, jdike@linux.intel.com To: xiaohui.xin@intel.com Return-path: In-Reply-To: <1272187206-18534-19-git-send-email-xiaohui.xin@intel.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org From: xiaohui.xin@intel.com Date: Sun, 25 Apr 2010 17:20:06 +0800 > The idea is simple, just to pin the guest VM user space and then let > host NIC driver has the chance to directly DMA to it. Isn't it much easier to map the RX ring of the network device into the guest's address space, have DMA map calls translate guest addresses to physical/DMA addresses as well as do all of this crazy page pinning stuff, and provide the translations and protections via the IOMMU? What's being proposed here looks a bit over-engineered.