From 75993300d008f418ee2569a632185fc1d7d50674 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 15 Jul 2015 15:26:19 +0300 Subject: virtio_net: don't require ANY_LAYOUT with VERSION_1 ANY_LAYOUT is a compatibility feature. It's implied for VERSION_1 devices, and non-transitional devices might not offer it. Change code to behave accordingly. Signed-off-by: Michael S. Tsirkin Reviewed-by: Paolo Bonzini Reviewed-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/net/virtio_net.c') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 63c7810e1545..7fbca37a1adf 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1828,7 +1828,8 @@ static int virtnet_probe(struct virtio_device *vdev) else vi->hdr_len = sizeof(struct virtio_net_hdr); - if (virtio_has_feature(vdev, VIRTIO_F_ANY_LAYOUT)) + if (virtio_has_feature(vdev, VIRTIO_F_ANY_LAYOUT) || + virtio_has_feature(vdev, VIRTIO_F_VERSION_1)) vi->any_header_sg = true; if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ)) -- cgit v1.2.3 From 0fbd050a7d262b74527a289ae75a33626d1060a8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 31 Jul 2015 18:25:17 +0200 Subject: virtio_net: add gro capability Straightforward patch to add GRO processing to virtio_net. napi_complete_done() usage allows more aggressive aggregation, opted-in by setting /sys/class/net/xxx/gro_flush_timeout Tested: Setting /sys/class/net/xxx/gro_flush_timeout to 1000 nsec, Rick Jones reported following results. One VM of each on a pair of OpenStack compute nodes with E5-2650Lv3 CPUs and Intel 82599ES-based NICs. So, two "before" and two "after" VMs. The OpenStack compute nodes were running OpenStack Kilo, with VxLAN encapsulation being used through OVS so no GRO coming-up the host stack. The compute nodes themselves were running a 3.14-based kernel. Single-stream netperf, CPU utilizations and thus service demands are based on intra-guest reported CPU. Throughput Mbit/s, bigger is better Min Median Average Max 4.2.0-rc3+ 1364 1686 1678 1938 4.2.0-rc3+flush1k 1824 2269 2275 2647 Send Service Demand, smaller is better Min Median Average Max 4.2.0-rc3+ 0.236 0.558 0.524 0.802 4.2.0-rc3+flush1k 0.176 0.503 0.471 0.738 Receive Service Demand, smaller is better. Min Median Average Max 4.2.0-rc3+ 1.906 2.188 2.191 2.531 4.2.0-rc3+flush1k 0.448 0.529 0.533 0.692 Signed-off-by: Eric Dumazet Tested-by: Rick Jones Cc: "Michael S. Tsirkin" Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/virtio_net.c') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 7fbca37a1adf..66f08f622dc6 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -518,7 +518,7 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq, skb_mark_napi_id(skb, &rq->napi); - netif_receive_skb(skb); + napi_gro_receive(&rq->napi, skb); return; frame_err: @@ -756,7 +756,7 @@ static int virtnet_poll(struct napi_struct *napi, int budget) /* Out of packets? */ if (received < budget) { r = virtqueue_enable_cb_prepare(rq->vq); - napi_complete(napi); + napi_complete_done(napi, received); if (unlikely(virtqueue_poll(rq->vq, r)) && napi_schedule_prep(napi)) { virtqueue_disable_cb(rq->vq); -- cgit v1.2.3 From 48900cb6af4282fa0fb6ff4d72a81aa3dadb5c39 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Wed, 5 Aug 2015 10:34:04 +0800 Subject: virtio-net: drop NETIF_F_FRAGLIST virtio declares support for NETIF_F_FRAGLIST, but assumes that there are at most MAX_SKB_FRAGS + 2 fragments which isn't always true with a fraglist. A longer fraglist in the skb will make the call to skb_to_sgvec overflow the sg array, leading to memory corruption. Drop NETIF_F_FRAGLIST so we only get what we can handle. Cc: Michael S. Tsirkin Signed-off-by: Jason Wang Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/virtio_net.c') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 7fbca37a1adf..237f8e5e493d 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1756,9 +1756,9 @@ static int virtnet_probe(struct virtio_device *vdev) /* Do we support "hardware" checksums? */ if (virtio_has_feature(vdev, VIRTIO_NET_F_CSUM)) { /* This opens up the world of extra features. */ - dev->hw_features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST; + dev->hw_features |= NETIF_F_HW_CSUM | NETIF_F_SG; if (csum) - dev->features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST; + dev->features |= NETIF_F_HW_CSUM | NETIF_F_SG; if (virtio_has_feature(vdev, VIRTIO_NET_F_GSO)) { dev->hw_features |= NETIF_F_TSO | NETIF_F_UFO -- cgit v1.2.3 From 5377d75823ff90c0d5d52d69087707b2e54be21b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 19 Aug 2015 09:48:40 +0200 Subject: virtio_net: use DECLARE_EWMA Instead of using the out-of-line EWMA calculation, use DECLARE_EWMA() to create static inlines. On x86/64 this results in no change in code size for me, but reduces the struct receive_queue size by the two unsigned long values that store the parameters. Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'drivers/net/virtio_net.c') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 546b669fbfdd..9b950f2db836 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -40,12 +40,12 @@ module_param(gso, bool, 0444); #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN) #define GOOD_COPY_LEN 128 -/* Weight used for the RX packet size EWMA. The average packet size is used to - * determine the packet buffer size when refilling RX rings. As the entire RX - * ring may be refilled at once, the weight is chosen so that the EWMA will be - * insensitive to short-term, transient changes in packet size. +/* RX packet size EWMA. The average packet size is used to determine the packet + * buffer size when refilling RX rings. As the entire RX ring may be refilled + * at once, the weight is chosen so that the EWMA will be insensitive to short- + * term, transient changes in packet size. */ -#define RECEIVE_AVG_WEIGHT 64 +DECLARE_EWMA(pkt_len, 1, 64) /* Minimum alignment for mergeable packet buffers. */ #define MERGEABLE_BUFFER_ALIGN max(L1_CACHE_BYTES, 256) @@ -85,7 +85,7 @@ struct receive_queue { struct page *pages; /* Average packet length for mergeable receive buffers. */ - struct ewma mrg_avg_pkt_len; + struct ewma_pkt_len mrg_avg_pkt_len; /* Page frag for packet buffer allocation. */ struct page_frag alloc_frag; @@ -407,7 +407,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, } } - ewma_add(&rq->mrg_avg_pkt_len, head_skb->len); + ewma_pkt_len_add(&rq->mrg_avg_pkt_len, head_skb->len); return head_skb; err_skb: @@ -600,12 +600,12 @@ static int add_recvbuf_big(struct virtnet_info *vi, struct receive_queue *rq, return err; } -static unsigned int get_mergeable_buf_len(struct ewma *avg_pkt_len) +static unsigned int get_mergeable_buf_len(struct ewma_pkt_len *avg_pkt_len) { const size_t hdr_len = sizeof(struct virtio_net_hdr_mrg_rxbuf); unsigned int len; - len = hdr_len + clamp_t(unsigned int, ewma_read(avg_pkt_len), + len = hdr_len + clamp_t(unsigned int, ewma_pkt_len_read(avg_pkt_len), GOOD_PACKET_LEN, PAGE_SIZE - hdr_len); return ALIGN(len, MERGEABLE_BUFFER_ALIGN); } @@ -1615,7 +1615,7 @@ static int virtnet_alloc_queues(struct virtnet_info *vi) napi_hash_add(&vi->rq[i].napi); sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg)); - ewma_init(&vi->rq[i].mrg_avg_pkt_len, 1, RECEIVE_AVG_WEIGHT); + ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len); sg_init_table(vi->sq[i].sg, ARRAY_SIZE(vi->sq[i].sg)); } @@ -1658,7 +1658,7 @@ static ssize_t mergeable_rx_buffer_size_show(struct netdev_rx_queue *queue, { struct virtnet_info *vi = netdev_priv(queue->dev); unsigned int queue_index = get_netdev_rx_queue_index(queue); - struct ewma *avg; + struct ewma_pkt_len *avg; BUG_ON(queue_index >= vi->max_queue_pairs); avg = &vi->rq[queue_index].mrg_avg_pkt_len; -- cgit v1.2.3 From 547c890cfd26010e01b3faf98f29ca9b2252c8d8 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 27 Aug 2015 14:53:06 +0800 Subject: virtio-net: avoid unnecessary sg initialzation Usually an skb does not have up to MAX_SKB_FRAGS frags. So no need to initialize the unuse part of sg. This patch initialize the sg based on the real number it will used: - during xmit, it could be inferred from nr_frags and can_push. - for small receive buffer, it will also be 2. Cc: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/virtio_net.c') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 9b950f2db836..d8838dedb7a4 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -540,7 +540,7 @@ static int add_recvbuf_small(struct virtnet_info *vi, struct receive_queue *rq, skb_put(skb, GOOD_PACKET_LEN); hdr = skb_vnet_hdr(skb); - sg_init_table(rq->sg, MAX_SKB_FRAGS + 2); + sg_init_table(rq->sg, 2); sg_set_buf(rq->sg, hdr, vi->hdr_len); skb_to_sgvec(skb, rq->sg + 1, 0, skb->len); @@ -893,7 +893,7 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb) if (vi->mergeable_rx_bufs) hdr->num_buffers = 0; - sg_init_table(sq->sg, MAX_SKB_FRAGS + 2); + sg_init_table(sq->sg, skb_shinfo(skb)->nr_frags + (can_push ? 1 : 2)); if (can_push) { __skb_push(skb, hdr_len); num_sg = skb_to_sgvec(skb, sq->sg, 0, skb->len); -- cgit v1.2.3