[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <158945342911.97035.11214251236208648808.stgit@firesoul>
Date: Thu, 14 May 2020 12:50:29 +0200
From: Jesper Dangaard Brouer <brouer@...hat.com>
To: sameehj@...zon.com
Cc: Jesper Dangaard Brouer <brouer@...hat.com>, netdev@...r.kernel.org,
bpf@...r.kernel.org,
Toke Høiland-Jørgensen <toke@...hat.com>,
Daniel Borkmann <borkmann@...earbox.net>,
Alexei Starovoitov <alexei.starovoitov@...il.com>,
"David S. Miller" <davem@...emloft.net>,
John Fastabend <john.fastabend@...il.com>,
Alexander Duyck <alexander.duyck@...il.com>,
Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
David Ahern <dsahern@...il.com>,
Ilias Apalodimas <ilias.apalodimas@...aro.org>,
Lorenzo Bianconi <lorenzo@...nel.org>,
Saeed Mahameed <saeedm@...lanox.com>,
Tariq Toukan <tariqt@...lanox.com>
Subject: [PATCH net-next v4 18/33] nfp: add XDP frame size to netronome driver
The netronome nfp driver use PAGE_SIZE when xdp_prog is set, but
xdp.data_hard_start begins at offset NFP_NET_RX_BUF_HEADROOM.
Thus, adjust for this when setting xdp.frame_sz, as it counts
from data_hard_start.
When doing XDP_TX this driver is smart and instead of a full DMA-map
does a DMA-sync on with packet length. As xdp_adjust_tail can now
grow packet length, add checks to make sure that grow size is within
the DMA-mapped size.
Cc: Jakub Kicinski <kuba@...nel.org>
Signed-off-by: Jesper Dangaard Brouer <brouer@...hat.com>
Reviewed-by: Jakub Kicinski <kuba@...nel.org>
---
.../net/ethernet/netronome/nfp/nfp_net_common.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 9bfb3b077bc1..0e0cc3d58bdc 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1741,10 +1741,15 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
struct nfp_net_rx_buf *rxbuf, unsigned int dma_off,
unsigned int pkt_len, bool *completed)
{
+ unsigned int dma_map_sz = dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA;
struct nfp_net_tx_buf *txbuf;
struct nfp_net_tx_desc *txd;
int wr_idx;
+ /* Reject if xdp_adjust_tail grow packet beyond DMA area */
+ if (pkt_len + dma_off > dma_map_sz)
+ return false;
+
if (unlikely(nfp_net_tx_full(tx_ring, 1))) {
if (!*completed) {
nfp_net_xdp_complete(tx_ring);
@@ -1817,6 +1822,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
rcu_read_lock();
xdp_prog = READ_ONCE(dp->xdp_prog);
true_bufsz = xdp_prog ? PAGE_SIZE : dp->fl_bufsz;
+ xdp.frame_sz = PAGE_SIZE - NFP_NET_RX_BUF_HEADROOM;
xdp.rxq = &rx_ring->xdp_rxq;
tx_ring = r_vec->xdp_ring;
Powered by blists - more mailing lists