From: Sridhar Samudrala <sridhar.samudrala@intel.com>
To: magnus.karlsson@intel.com, bjorn.topel@intel.com,
netdev@vger.kernel.org, bpf@vger.kernel.org,
sridhar.samudrala@intel.com, intel-wired-lan@lists.osuosl.org,
maciej.fijalkowski@intel.com, tom.herbert@intel.com
Subject: [PATCH bpf-next 4/5] ixgbe: Enable XDP_SKIP_BPF option for AF_XDP sockets
Date: Wed, 14 Aug 2019 20:46:22 -0700 [thread overview]
Message-ID: <1565840783-8269-5-git-send-email-sridhar.samudrala@intel.com> (raw)
In-Reply-To: <1565840783-8269-1-git-send-email-sridhar.samudrala@intel.com>
This patch skips calling BPF program in the receive path if
the queue is associated with UMEM that is not shared and
bound to an AF_XDP socket that has enabled skip bpf during
bind() call.
Signed-off-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 20 +++++++++++++++++--
drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 16 +++++++++++++--
2 files changed, 32 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dc7b128c780e..594792860cdd 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -2197,6 +2197,7 @@ static struct sk_buff *ixgbe_run_xdp(struct ixgbe_adapter *adapter,
int err, result = IXGBE_XDP_PASS;
struct bpf_prog *xdp_prog;
struct xdp_frame *xdpf;
+ struct xdp_umem *umem;
u32 act;
rcu_read_lock();
@@ -2207,6 +2208,13 @@ static struct sk_buff *ixgbe_run_xdp(struct ixgbe_adapter *adapter,
prefetchw(xdp->data_hard_start); /* xdp_frame write */
+ umem = xdp_get_umem_from_qid(rx_ring->netdev, rx_ring->queue_index);
+ if (xsk_umem_skip_bpf(umem)) {
+ err = xsk_umem_rcv(umem, xdp);
+ result = !err ? IXGBE_XDP_REDIR : IXGBE_XDP_CONSUMED;
+ goto xdp_out;
+ }
+
act = bpf_prog_run_xdp(xdp_prog, xdp);
switch (act) {
case XDP_PASS:
@@ -2400,8 +2408,16 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
total_rx_packets++;
}
- if (xdp_xmit & IXGBE_XDP_REDIR)
- xdp_do_flush_map();
+ if (xdp_xmit & IXGBE_XDP_REDIR) {
+ struct xdp_umem *umem;
+
+ umem = xdp_get_umem_from_qid(rx_ring->netdev,
+ rx_ring->queue_index);
+ if (xsk_umem_skip_bpf(umem))
+ xsk_umem_flush(umem);
+ else
+ xdp_do_flush_map();
+ }
if (xdp_xmit & IXGBE_XDP_TX) {
struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
index 6b609553329f..9ea8a769d7a8 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
@@ -148,6 +148,12 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter,
struct xdp_frame *xdpf;
u32 act;
+ if (xsk_umem_skip_bpf(rx_ring->xsk_umem)) {
+ err = xsk_umem_rcv(rx_ring->xsk_umem, xdp);
+ result = !err ? IXGBE_XDP_REDIR : IXGBE_XDP_CONSUMED;
+ return result;
+ }
+
rcu_read_lock();
xdp_prog = READ_ONCE(rx_ring->xdp_prog);
act = bpf_prog_run_xdp(xdp_prog, xdp);
@@ -527,8 +533,14 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector,
ixgbe_rx_skb(q_vector, skb);
}
- if (xdp_xmit & IXGBE_XDP_REDIR)
- xdp_do_flush_map();
+ if (xdp_xmit & IXGBE_XDP_REDIR) {
+ struct xdp_umem *umem = rx_ring->xsk_umem;
+
+ if (xsk_umem_skip_bpf(umem))
+ xsk_umem_flush(umem);
+ else
+ xdp_do_flush_map();
+ }
if (xdp_xmit & IXGBE_XDP_TX) {
struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
--
2.20.1
next prev parent reply other threads:[~2019-08-15 3:46 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-15 3:46 [PATCH bpf-next 0/5] Add support for SKIP_BPF flag for AF_XDP sockets Sridhar Samudrala
2019-08-15 3:46 ` [PATCH bpf-next 1/5] xsk: Convert bool 'zc' field in struct xdp_umem to a u32 bitmap Sridhar Samudrala
2019-08-15 3:46 ` [PATCH bpf-next 2/5] xsk: Introduce XDP_SKIP_BPF bind option Sridhar Samudrala
2019-08-15 3:46 ` [PATCH bpf-next 3/5] i40e: Enable XDP_SKIP_BPF option for AF_XDP sockets Sridhar Samudrala
2019-08-16 9:21 ` kbuild test robot
2019-08-15 3:46 ` Sridhar Samudrala [this message]
2019-08-15 3:46 ` [PATCH bpf-next 5/5] xdpsock_user: Add skip_bpf option Sridhar Samudrala
2019-08-15 11:12 ` [PATCH bpf-next 0/5] Add support for SKIP_BPF flag for AF_XDP sockets Toke Høiland-Jørgensen
2019-08-15 16:25 ` Samudrala, Sridhar
2019-08-15 17:11 ` Toke Høiland-Jørgensen
2019-08-16 6:12 ` Samudrala, Sridhar
2019-08-15 12:51 ` Björn Töpel
2019-08-15 16:46 ` Samudrala, Sridhar
2019-08-16 13:32 ` [Intel-wired-lan] " Björn Töpel
2019-08-16 22:08 ` Jonathan Lemon
2019-08-19 7:39 ` Björn Töpel
2019-08-15 19:28 ` Jakub Kicinski
2019-08-16 6:25 ` Samudrala, Sridhar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1565840783-8269-5-git-send-email-sridhar.samudrala@intel.com \
--to=sridhar.samudrala@intel.com \
--cc=bjorn.topel@intel.com \
--cc=bpf@vger.kernel.org \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=tom.herbert@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).