vxlan dev currently ignores lowerdev's gso_max_size, which adversely affects TSO performance of liquidio if it's the lowerdev. Egress TCP packets' skb->len often exceed liquidio's advertised gso_max_size. This may happen on other NIC drivers.
Fix it by assigning lowerdev's gso_max_size to that of vxlan dev. Might as well do likewise for gso_max_segs. Single flow TSO throughput of liquidio as lowerdev (using iperf3): Before the patch: 139 Mbps After the patch : 8.68 Gbps Percent increase: 6,144 % Signed-off-by: Felix Manlunas <felix.manlu...@cavium.com> Signed-off-by: Satanand Burla <satananda.bu...@cavium.com> --- drivers/net/vxlan.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 1e54fb5..714f74f 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2926,6 +2926,11 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, return -EINVAL; } + if (lowerdev) { + dev->gso_max_size = lowerdev->gso_max_size; + dev->gso_max_segs = lowerdev->gso_max_segs; + } + if (conf->mtu) { int max_mtu = ETH_MAX_MTU;