From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] virtio-net: mergeable buffer size should include virtio-net header Date: Thu, 14 Nov 2013 17:23:08 -0500 (EST) Message-ID: <20131114.172308.1881893130228720832.davem@davemloft.net> References: <1384454464-19923-1-git-send-email-mwdalton@google.com> <1384455552.28716.28.camel@edumazet-glaptop2.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1384455552.28716.28.camel@edumazet-glaptop2.roam.corp.google.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: eric.dumazet@gmail.com Cc: mwdalton@google.com, mst@redhat.com, netdev@vger.kernel.org, dborkman@redhat.com, virtualization@lists.linux-foundation.org, edumazet@google.com List-Id: virtualization@lists.linuxfoundation.org From: Eric Dumazet Date: Thu, 14 Nov 2013 10:59:12 -0800 > On Thu, 2013-11-14 at 10:41 -0800, Michael Dalton wrote: >> Commit 2613af0ed18a ("virtio_net: migrate mergeable rx buffers to page >> frag allocators") changed the mergeable receive buffer size from PAGE_SIZE >> to MTU-size. However, the merge buffer size does not take into account the >> size of the virtio-net header. Consequently, packets that are MTU-size >> will take two buffers intead of one (to store the virtio-net header), >> substantially decreasing the throughput of MTU-size traffic due to TCP >> window / SKB truesize effects. >> >> This commit changes the mergeable buffer size to include the virtio-net >> header. The buffer size is cacheline-aligned because skb_page_frag_refill >> will not automatically align the requested size. >> >> Benchmarks taken from an average of 5 netperf 30-second TCP_STREAM runs >> between two QEMU VMs on a single physical machine. Each VM has two VCPUs and >> vhost enabled. All VMs and vhost threads run in a single 4 CPU cgroup >> cpuset, using cgroups to ensure that other processes in the system will not >> be scheduled on the benchmark CPUs. Transmit offloads and mergeable receive >> buffers are enabled, but guest_tso4 / guest_csum are explicitly disabled to >> force MTU-sized packets on the receiver. >> >> next-net trunk before 2613af0ed18a (PAGE_SIZE buf): 3861.08Gb/s >> net-next trunk (MTU 1500- packet uses two buf due to size bug): 4076.62Gb/s >> net-next trunk (MTU 1480- packet fits in one buf): 6301.34Gb/s >> net-next trunk w/ size fix (MTU 1500 - packet fits in one buf): 6445.44Gb/s >> >> Suggested-by: Eric Northup >> Signed-off-by: Michael Dalton >> --- > > Acked-by: Eric Dumazet Applied, thanks.