diff options
| author | Jesper Dangaard Brouer <brouer@redhat.com> | 2018-05-24 16:46:12 +0200 |
|---|---|---|
| committer | Alexei Starovoitov <ast@kernel.org> | 2018-05-24 18:36:15 -0700 |
| commit | 735fc4054b3a25034445c6713d259da0f96f8131 (patch) | |
| tree | 355f7a0672e6239fa4227d562f7d5b65fac9c011 /net/core | |
| parent | 389ab7f01af988c2a1ec5617eb0c7e220df1ef1c (diff) | |
| download | op-kernel-dev-735fc4054b3a25034445c6713d259da0f96f8131.zip op-kernel-dev-735fc4054b3a25034445c6713d259da0f96f8131.tar.gz | |
xdp: change ndo_xdp_xmit API to support bulking
This patch change the API for ndo_xdp_xmit to support bulking
xdp_frames.
When kernel is compiled with CONFIG_RETPOLINE, XDP sees a huge slowdown.
Most of the slowdown is caused by DMA API indirect function calls, but
also the net_device->ndo_xdp_xmit() call.
Benchmarked patch with CONFIG_RETPOLINE, using xdp_redirect_map with
single flow/core test (CPU E5-1650 v4 @ 3.60GHz), showed
performance improved:
for driver ixgbe: 6,042,682 pps -> 6,853,768 pps = +811,086 pps
for driver i40e : 6,187,169 pps -> 6,724,519 pps = +537,350 pps
With frames avail as a bulk inside the driver ndo_xdp_xmit call,
further optimizations are possible, like bulk DMA-mapping for TX.
Testing without CONFIG_RETPOLINE show the same performance for
physical NIC drivers.
The virtual NIC driver tun sees a huge performance boost, as it can
avoid doing per frame producer locking, but instead amortize the
locking cost over the bulk.
V2: Fix compile errors reported by kbuild test robot <lkp@intel.com>
V4: Isolated ndo, driver changes and callers.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Diffstat (limited to 'net/core')
| -rw-r--r-- | net/core/filter.c | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/net/core/filter.c b/net/core/filter.c index 36cf2f8..1d75f93 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3039,7 +3039,7 @@ static int __bpf_tx_xdp(struct net_device *dev, u32 index) { struct xdp_frame *xdpf; - int err; + int sent; if (!dev->netdev_ops->ndo_xdp_xmit) { return -EOPNOTSUPP; @@ -3049,9 +3049,9 @@ static int __bpf_tx_xdp(struct net_device *dev, if (unlikely(!xdpf)) return -EOVERFLOW; - err = dev->netdev_ops->ndo_xdp_xmit(dev, xdpf); - if (err) - return err; + sent = dev->netdev_ops->ndo_xdp_xmit(dev, 1, &xdpf); + if (sent <= 0) + return sent; dev->netdev_ops->ndo_xdp_flush(dev); return 0; } |
