Skip to contentSkip to conversation

Introduce preferred busy-polling

PATCH Superseded Björn Töpel 7 participants 33 messages Nov 19, 2020 – Nov 5, 2021

Subsystems (32)
Networking Network drivers VFS MIPS +28 more subsystems

v3 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

v3 cover Introduce preferred busy-polling

This series introduces three new features:

1. A new "heavy traffic" busy-polling variant that works in concert with the existing napi_defer_hard_irqs and gro_flush_timeout knobs.

2. A new socket option that let a user change the busy-polling NAPI budget.

3. Allow busy-polling to be performed on XDP sockets.

The existing busy-polling mode, enabled by the SO_BUSY_POLL socket option or system-wide using the /proc/sys/net/core/busy_read knob, is an opportunistic. That means that if the NAPI context is not scheduled, it will poll it. If, after busy-polling, the budget is exceeded the busy-polling logic will schedule the NAPI onto the regular softirq handling.

One implication of the behavior above is that a busy/heavy loaded NAPI context will never enter/allow for busy-polling. Some applications prefer that most NAPI processing would be done by busy-polling.

This series adds a new socket option, SO_PREFER_BUSY_POLL, that works in concert with the napi_defer_hard_irqs and gro_flush_timeout knobs. The napi_defer_hard_irqs and gro_flush_timeout knobs were introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature"), and allows for a user to defer interrupts to be enabled and instead schedule the NAPI context from a watchdog timer. When a user enables the SO_PREFER_BUSY_POLL, again with the other knobs enabled, and the NAPI context is being processed by a softirq, the softirq NAPI processing will exit early to allow the busy-polling to be performed.

If the application stops performing busy-polling via a system call, the watchdog timer defined by gro_flush_timeout will timeout, and regular softirq handling will resume.

In summary; Heavy traffic applications that prefer busy-polling over softirq processing should use this option.

Patch 6 touches a lot of drivers, so the Cc: list is grossly long.

Example usage:

  $ echo 2 | sudo tee /sys/class/net/ens785f1/napi_defer_hard_irqs
  $ echo 200000 | sudo tee /sys/class/net/ens785f1/gro_flush_timeout

Note that the timeout should be larger than the userspace processing window, otherwise the watchdog will timeout and fall back to regular softirq processing.

Enable the SO_BUSY_POLL/SO_PREFER_BUSY_POLL options on your socket.

Performance simple UDP ping-pong:

A packet generator blasts UDP packets from a packet generator to a certain {src,dst}IP/port, so a dedicated ksoftirq will be busy handling the packets at a certain core.

A simple UDP test program that simply does recvfrom/sendto is running at the host end. Throughput in pps and RTT latency is measured at the packet generator.

/proc/sys/net/core/busy_read is set (20).

Min Max Avg (usec)

1. Blocking 2-cores: 490Kpps
1218.192 1335.427 1271.083

2. Blocking, 1-core: 155Kpps
1327.195 17294.855 4761.367

3. Non-blocking, 2-cores: 475Kpps
1221.197 1330.465 1270.740

4. Non-blocking, 1-core: 3Kpps
29006.482 37260.465 33128.367

5. Non-blocking, prefer busy-poll, 1-core: 420Kpps 1202.535 5494.052 4885.443

Scenario 2 and 5 shows when the new option should be used. Throughput go from 155 to 420Kpps, average latency are similar, but the tail latencies are much better for the latter.

Performance XDP sockets:

Again, a packet generator blasts UDP packets from a packet generator to a certain {src,dst}IP/port.

Today, running XDP sockets sample on the same core as the softirq handling, performance tanks mainly because we do not yield to user-space when the XDP socket Rx queue is full.

  # taskset -c 5 ./xdpsock -i ens785f1 -q 5 -n 1 -r
  Rx: 64Kpps

# # biased busy-polling, budget 8

  # taskset -c 5 ./xdpsock -i ens785f1 -q 5 -n 1 -r -B -b 8
  Rx 9.9Mpps
  # # biased busy-polling, budget 64
  # taskset -c 5 ./xdpsock -i ens785f1 -q 5 -n 1 -r -B -b 64
  Rx: 19.3Mpps
  # # biased busy-polling, budget 256
  # taskset -c 5 ./xdpsock -i ens785f1 -q 5 -n 1 -r -B -b 256
  Rx: 21.4Mpps
  # # biased busy-polling, budget 512
  # taskset -c 5 ./xdpsock -i ens785f1 -q 5 -n 1 -r -B -b 512
  Rx: 21.7Mpps

Compared to the two-core case:

  # taskset -c 4 ./xdpsock -i ens785f1 -q 20 -n 1 -r
  Rx: 20.7Mpps

We're getting better single-core performance than two, for this naïve drop scenario.

Performance netperf UDP_RR:

Note that netperf UDP_RR is not a heavy traffic tests, and preferred busy-polling is not typically something we want to use here.

  $ echo 20 | sudo tee /proc/sys/net/core/busy_read
  $ netperf -H 192.168.1.1 -l 30 -t UDP_RR -v 2 -- \
      -o min_latency,mean_latency,max_latency,stddev_latency,transaction_rate

busy-polling blocking sockets: 12,13.33,224,0.63,74731.177

I hacked netperf to use non-blocking sockets and re-ran:

busy-polling non-blocking sockets: 12,13.46,218,0.72,73991.172
prefer busy-polling non-blocking sockets: 12,13.62,221,0.59,73138.448

Using the preferred busy-polling mode does not impact performance.

The above tests was done for the 'ice' driver.

Thanks to Jakub for suggesting this busy-polling addition [1], and Eric for all input/review!

Changes:

rfc-v1 [2] -> rfc-v2:
* Changed name from bias to prefer.
* Base the work on Eric's/Luigi's defer irq/gro timeout work.
* Proper GRO flushing.
* Build issues for some XDP drivers.

rfc-v2 [3] -> v1:
* Fixed broken qlogic build.
* Do not trigger an IPI (XDP socket wakeup) when busy-polling is enabled.

v1 [4] -> v2:
* Added napi_id to socionext driver, and added Ilias Acked-by:. (Ilias)
* Added a samples patch to improve busy-polling for xdpsock/l2fwd.
* Correctly mark atomic operations with {WRITE,READ}_ONCE, to make KCSAN and the code readers happy. (Eric)
* Check NAPI budget not to exceed U16_MAX. (Eric)
* Added kdoc.

v2 [5] -> v3:
* Collected Acked-by:.
* Check NAPI disable prior prefer busy-polling. (Jakub)
* Added napi_id registration for virtio-net. (Michael)
* Added napi_id registration for veth.

[1] https://lore.kernel.org/netdev/20200925120652.10b8d7c5@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com/
[2] https://lore.kernel.org/bpf/[email protected]/
[3] https://lore.kernel.org/bpf/[email protected]/
[4] https://lore.kernel.org/bpf/[email protected]/
[5] https://lore.kernel.org/bpf/[email protected]/

Björn Töpel (10):
net: introduce preferred busy-polling
net: add SO_BUSY_POLL_BUDGET socket option
xsk: add support for recvmsg()
xsk: check need wakeup flag in sendmsg()
xsk: add busy-poll support for {recv,send}msg()
xsk: propagate napi_id to XDP socket Rx path samples/bpf: use recvfrom() in xdpsock/rxdrop samples/bpf: use recvfrom() in xdpsock/l2fwd samples/bpf: add busy-poll support to xdpsock samples/bpf: add option to set the busy-poll budget

 arch/alpha/include/uapi/asm/socket.h          |  3 +
 arch/mips/include/uapi/asm/socket.h           |  3 +
 arch/parisc/include/uapi/asm/socket.h         |  3 +
 arch/sparc/include/uapi/asm/socket.h          |  3 +
 drivers/net/ethernet/amazon/ena/ena_netdev.c  |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     |  2 +-
 .../ethernet/cavium/thunder/nicvf_queues.c    |  2 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  2 +-
 drivers/net/ethernet/intel/ice/ice_base.c     |  4 +-
 drivers/net/ethernet/intel/ice/ice_txrx.c     |  2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  2 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/ethernet/marvell/mvneta.c         |  2 +-
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   |  4 +-
 drivers/net/ethernet/mellanox/mlx4/en_rx.c    |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 .../ethernet/netronome/nfp/nfp_net_common.c   |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_main.c  |  2 +-
 drivers/net/ethernet/sfc/rx_common.c          |  2 +-
 drivers/net/ethernet/socionext/netsec.c       |  2 +-
 drivers/net/ethernet/ti/cpsw_priv.c           |  2 +-
 drivers/net/hyperv/netvsc.c                   |  2 +-
 drivers/net/tun.c                             |  2 +-
 drivers/net/veth.c                            | 12 ++-
 drivers/net/virtio_net.c                      |  2 +-
 drivers/net/xen-netfront.c                    |  2 +-
 fs/eventpoll.c                                |  3 +-
 include/linux/netdevice.h                     | 35 +++++---
 include/net/busy_poll.h                       | 27 ++++--
 include/net/sock.h                            |  6 ++
 include/net/xdp.h                             |  3 +-
 include/uapi/asm-generic/socket.h             |  3 +
 net/core/dev.c                                | 89 ++++++++++++++-----
 net/core/sock.c                               | 19 ++++
 net/core/xdp.c                                |  3 +-
 net/xdp/xsk.c                                 | 53 ++++++++++-
 net/xdp/xsk_buff_pool.c                       | 13 ++-
 samples/bpf/xdpsock_user.c                    | 78 ++++++++++------
 40 files changed, 299 insertions(+), 107 deletions(-)

base-commit: 4e99d115d865d45e17e83478d757b58d8fa66d3c

Signature · 1 line
2.27.0
#1 lore raw

PATCH v3 1/10 net: introduce preferred busy-polling

From: Björn Töpel <[email protected]>

The existing busy-polling mode, enabled by the SO_BUSY_POLL socket option or system-wide using the /proc/sys/net/core/busy_read knob, is an opportunistic. That means that if the NAPI context is not scheduled, it will poll it. If, after busy-polling, the budget is exceeded the busy-polling logic will schedule the NAPI onto the regular softirq handling.

One implication of the behavior above is that a busy/heavy loaded NAPI context will never enter/allow for busy-polling. Some applications prefer that most NAPI processing would be done by busy-polling.

This series adds a new socket option, SO_PREFER_BUSY_POLL, that works in concert with the napi_defer_hard_irqs and gro_flush_timeout knobs. The napi_defer_hard_irqs and gro_flush_timeout knobs were introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature"), and allows for a user to defer interrupts to be enabled and instead schedule the NAPI context from a watchdog timer. When a user enables the SO_PREFER_BUSY_POLL, again with the other knobs enabled, and the NAPI context is being processed by a softirq, the softirq NAPI processing will exit early to allow the busy-polling to be performed.

If the application stops performing busy-polling via a system call, the watchdog timer defined by gro_flush_timeout will timeout, and regular softirq handling will resume.

In summary; Heavy traffic applications that prefer busy-polling over softirq processing should use this option.

Example usage:

  $ echo 2 | sudo tee /sys/class/net/ens785f1/napi_defer_hard_irqs
  $ echo 200000 | sudo tee /sys/class/net/ens785f1/gro_flush_timeout

Note that the timeout should be larger than the userspace processing window, otherwise the watchdog will timeout and fall back to regular softirq processing.

Enable the SO_BUSY_POLL/SO_PREFER_BUSY_POLL options on your socket.

---
 arch/alpha/include/uapi/asm/socket.h  |  2 +
 arch/mips/include/uapi/asm/socket.h   |  2 +
 arch/parisc/include/uapi/asm/socket.h |  2 +
 arch/sparc/include/uapi/asm/socket.h  |  2 +
 fs/eventpoll.c                        |  2 +-
 include/linux/netdevice.h             | 35 +++++++-----
 include/net/busy_poll.h               |  5 +-
 include/net/sock.h                    |  4 ++
 include/uapi/asm-generic/socket.h     |  2 +
 net/core/dev.c                        | 78 +++++++++++++++++++++------
 net/core/sock.c                       |  9 ++++
 11 files changed, 111 insertions(+), 32 deletions(-)
arch/alpha/include/uapi/asm/socket.h +2 -0
index de6c4df61082..538359642554 100644
@@ -124,6 +124,8 @@
 
 #define SO_DETACH_REUSEPORT_BPF 68
 
+#define SO_PREFER_BUSY_POLL	69
+
 #if !defined(__KERNEL__)
 
 #if __BITS_PER_LONG == 64
arch/mips/include/uapi/asm/socket.h +2 -0
index d0a9ed2ca2d6..e406e73b5e6e 100644
@@ -135,6 +135,8 @@
 
 #define SO_DETACH_REUSEPORT_BPF 68
 
+#define SO_PREFER_BUSY_POLL	69
+
 #if !defined(__KERNEL__)
 
 #if __BITS_PER_LONG == 64
arch/parisc/include/uapi/asm/socket.h +2 -0
index 10173c32195e..1bc46200889d 100644
@@ -116,6 +116,8 @@
 
 #define SO_DETACH_REUSEPORT_BPF 0x4042
 
+#define SO_PREFER_BUSY_POLL	0x4043
+
 #if !defined(__KERNEL__)
 
 #if __BITS_PER_LONG == 64
arch/sparc/include/uapi/asm/socket.h +2 -0
index 8029b681fc7c..99688cf673a4 100644
@@ -117,6 +117,8 @@
 
 #define SO_DETACH_REUSEPORT_BPF  0x0047
 
+#define SO_PREFER_BUSY_POLL	 0x0048
+
 #if !defined(__KERNEL__)
 
 
fs/eventpoll.c +1 -1
index 4df61129566d..e11fab3a0b9e 100644
@@ -397,7 +397,7 @@ static void ep_busy_loop(struct eventpoll *ep, int nonblock)
 	unsigned int napi_id = READ_ONCE(ep->napi_id);
 
 	if ((napi_id >= MIN_NAPI_ID) && net_busy_loop_on())
-		napi_busy_loop(napi_id, nonblock ? NULL : ep_busy_loop_end, ep);
+		napi_busy_loop(napi_id, nonblock ? NULL : ep_busy_loop_end, ep, false);
 }
 
 static inline void ep_reset_busy_poll_napi_id(struct eventpoll *ep)
include/linux/netdevice.h +21 -14
index 7ce648a564f7..52d1cc2bd8a7 100644
@@ -350,23 +350,25 @@ struct napi_struct {
 };
 
 enum {
-	NAPI_STATE_SCHED,	/* Poll is scheduled */
-	NAPI_STATE_MISSED,	/* reschedule a napi */
-	NAPI_STATE_DISABLE,	/* Disable pending */
-	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
-	NAPI_STATE_LISTED,	/* NAPI added to system lists */
-	NAPI_STATE_NO_BUSY_POLL,/* Do not add in napi_hash, no busy polling */
-	NAPI_STATE_IN_BUSY_POLL,/* sk_busy_loop() owns this NAPI */
+	NAPI_STATE_SCHED,		/* Poll is scheduled */
+	NAPI_STATE_MISSED,		/* reschedule a napi */
+	NAPI_STATE_DISABLE,		/* Disable pending */
+	NAPI_STATE_NPSVC,		/* Netpoll - don't dequeue from poll_list */
+	NAPI_STATE_LISTED,		/* NAPI added to system lists */
+	NAPI_STATE_NO_BUSY_POLL,	/* Do not add in napi_hash, no busy polling */
+	NAPI_STATE_IN_BUSY_POLL,	/* sk_busy_loop() owns this NAPI */
+	NAPI_STATE_PREFER_BUSY_POLL,	/* prefer busy-polling over softirq processing*/
 };
 
 enum {
-	NAPIF_STATE_SCHED	 = BIT(NAPI_STATE_SCHED),
-	NAPIF_STATE_MISSED	 = BIT(NAPI_STATE_MISSED),
-	NAPIF_STATE_DISABLE	 = BIT(NAPI_STATE_DISABLE),
-	NAPIF_STATE_NPSVC	 = BIT(NAPI_STATE_NPSVC),
-	NAPIF_STATE_LISTED	 = BIT(NAPI_STATE_LISTED),
-	NAPIF_STATE_NO_BUSY_POLL = BIT(NAPI_STATE_NO_BUSY_POLL),
-	NAPIF_STATE_IN_BUSY_POLL = BIT(NAPI_STATE_IN_BUSY_POLL),
+	NAPIF_STATE_SCHED		= BIT(NAPI_STATE_SCHED),
+	NAPIF_STATE_MISSED		= BIT(NAPI_STATE_MISSED),
+	NAPIF_STATE_DISABLE		= BIT(NAPI_STATE_DISABLE),
+	NAPIF_STATE_NPSVC		= BIT(NAPI_STATE_NPSVC),
+	NAPIF_STATE_LISTED		= BIT(NAPI_STATE_LISTED),
+	NAPIF_STATE_NO_BUSY_POLL	= BIT(NAPI_STATE_NO_BUSY_POLL),
+	NAPIF_STATE_IN_BUSY_POLL	= BIT(NAPI_STATE_IN_BUSY_POLL),
+	NAPIF_STATE_PREFER_BUSY_POLL	= BIT(NAPI_STATE_PREFER_BUSY_POLL),
 };
 
 enum gro_result {
@@ -437,6 +439,11 @@ static inline bool napi_disable_pending(struct napi_struct *n)
 	return test_bit(NAPI_STATE_DISABLE, &n->state);
 }
 
+static inline bool napi_prefer_busy_poll(struct napi_struct *n)
+{
+	return test_bit(NAPI_STATE_PREFER_BUSY_POLL, &n->state);
+}
+
 bool napi_schedule_prep(struct napi_struct *n);
 
 /**
include/net/busy_poll.h +3 -2
index b001fa91c14e..0292b8353d7e 100644
@@ -43,7 +43,7 @@ bool sk_busy_loop_end(void *p, unsigned long start_time);
 
 void napi_busy_loop(unsigned int napi_id,
 		    bool (*loop_end)(void *, unsigned long),
-		    void *loop_end_arg);
+		    void *loop_end_arg, bool prefer_busy_poll);
 
 #else /* CONFIG_NET_RX_BUSY_POLL */
 static inline unsigned long net_busy_loop_on(void)
@@ -105,7 +105,8 @@ static inline void sk_busy_loop(struct sock *sk, int nonblock)
 	unsigned int napi_id = READ_ONCE(sk->sk_napi_id);
 
 	if (napi_id >= MIN_NAPI_ID)
-		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk);
+		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk,
+			       READ_ONCE(sk->sk_prefer_busy_poll));
 #endif
 }
 
include/net/sock.h +4 -0
index a5c6ae78df77..d49b89b071b6 100644
@@ -301,6 +301,7 @@ struct bpf_local_storage;
   *	@sk_ack_backlog: current listen backlog
   *	@sk_max_ack_backlog: listen backlog set in listen()
   *	@sk_uid: user id of owner
+  *	@sk_prefer_busy_poll: prefer busypolling over softirq processing
   *	@sk_priority: %SO_PRIORITY setting
   *	@sk_type: socket type (%SOCK_STREAM, etc)
   *	@sk_protocol: which protocol this socket belongs in this network family
@@ -479,6 +480,9 @@ struct sock {
 	u32			sk_ack_backlog;
 	u32			sk_max_ack_backlog;
 	kuid_t			sk_uid;
+#ifdef CONFIG_NET_RX_BUSY_POLL
+	u8			sk_prefer_busy_poll;
+#endif
 	struct pid		*sk_peer_pid;
 	const struct cred	*sk_peer_cred;
 	long			sk_rcvtimeo;
include/uapi/asm-generic/socket.h +2 -0
index 77f7c1638eb1..7dd02408b7ce 100644
@@ -119,6 +119,8 @@
 
 #define SO_DETACH_REUSEPORT_BPF 68
 
+#define SO_PREFER_BUSY_POLL	69
+
 #if !defined(__KERNEL__)
 
 #if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))
net/core/dev.c +63 -15

149 lines not on this page · raw message

net/core/sock.c +9 -0
index 727ea1cc633c..e05f2e52b5a8 100644
@@ -1159,6 +1159,12 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 				sk->sk_ll_usec = val;
 		}
 		break;
+	case SO_PREFER_BUSY_POLL:
+		if (valbool && !capable(CAP_NET_ADMIN))
+			ret = -EPERM;
+		else
+			WRITE_ONCE(sk->sk_prefer_busy_poll, valbool);
+		break;
 #endif
 
 	case SO_MAX_PACING_RATE:
@@ -1523,6 +1529,9 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 	case SO_BUSY_POLL:
 		v.val = sk->sk_ll_usec;
 		break;
+	case SO_PREFER_BUSY_POLL:
+		v.val = READ_ONCE(sk->sk_prefer_busy_poll);
+		break;
 #endif
 
 	case SO_MAX_PACING_RATE:
Signature · 1 line
2.27.0
maintainer
#2 lore raw

Re: [PATCH bpf-next v3 01/10] net: introduce preferred busy-polling

net/core/dev.c static int napi_poll(struct napi_struct *n, struct list_head *repoll) · line 6830

Quoted 12 lines · On Thu, 19 Nov 2020 09:30:15 +0100 Björn Töpel wrote:
net/core/dev.c +12 -0
+	/* The NAPI context has more processing work, but busy-polling
+	 * is preferred. Exit early.
+	 */
+	if (napi_prefer_busy_poll(n)) {
+		if (napi_complete_done(n, work)) {
+			/* If timeout is not set, we need to make sure
+			 * that the NAPI is re-scheduled.
+			 */
+			napi_schedule(n);
+		}
+		goto out_unlock;
+	}

Do we really need to go through napi_complete_done() here?

Isn't it sufficient to check:

	if (napi_prefer_busy_poll(n) && 
	    hrtimer_active(&n->timer)) // not 100% sure this is the
	                               // right helper for the check

If timer is scheduled it will fire and worst case sirq will kick back in after timeout. napi_complete_done() should had been called by the driver already to schedule the timer. If the driver doesn't call napi_complete_done() we should not allow it to use busy_poll() anyway.

#3 lore raw

Re: [PATCH bpf-next v3 01/10] net: introduce preferred busy-polling

Quoted 18 lines · Jakub Kicinski

net/core/dev.c static int napi_poll(struct napi_struct *n, struct list_head *repoll) · line 6830

Quoted 12 lines · On Thu, 19 Nov 2020 09:30:15 +0100 Björn Töpel wrote:
net/core/dev.c +12 -0
+	/* The NAPI context has more processing work, but busy-polling
+	 * is preferred. Exit early.
+	 */
+	if (napi_prefer_busy_poll(n)) {
+		if (napi_complete_done(n, work)) {
+			/* If timeout is not set, we need to make sure
+			 * that the NAPI is re-scheduled.
+			 */
+			napi_schedule(n);
+		}
+		goto out_unlock;
+	}

Do we really need to go through napi_complete_done() here?

Isn't it sufficient to check:

Quoted 9 lines
	if (napi_prefer_busy_poll(n) &&
	    hrtimer_active(&n->timer)) // not 100% sure this is the
	                               // right helper for the check

If timer is scheduled it will fire and worst case sirq will kick back in after timeout. napi_complete_done() should had been called by the driver already to schedule the timer. If the driver doesn't call napi_complete_done() we should not allow it to use busy_poll() anyway.

No, it's not. For a heavy traffic load, the napi_complete_done() will never be called by the driver. It'll just keep on spinning in the ksoftirqd. This code is to force out of that loop, so we need to call napi_complete_done() explicitly (which will set the timeout).

Without the explicit napi_complete_done(), the ksoftirqd will not stop, and the busy-polling will never allow to enter.

Björn

maintainer
#4 lore raw

Re: [PATCH bpf-next v3 01/10] net: introduce preferred busy-polling

In reply to Björn Töpel

include/net/busy_poll.h static inline void sk_busy_loop(struct sock *sk, int nonblock) · line 109

Quoted 7 lines · On Thu, 19 Nov 2020 09:30:15 +0100 Björn Töpel wrote:
include/net/busy_poll.h +2 -1
@@ -105,7 +105,8 @@ static inline void sk_busy_loop(struct sock *sk, int nonblock)
 	unsigned int napi_id = READ_ONCE(sk->sk_napi_id);
 
 	if (napi_id >= MIN_NAPI_ID)
-		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk);
+		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk,
+			       READ_ONCE(sk->sk_prefer_busy_poll));

Perhaps a noob question, but aren't all accesses to the new sk members under the socket lock? Do we really need the READ_ONCE() / WRITE_ONCE()?

#5 lore raw

Re: [PATCH bpf-next v3 01/10] net: introduce preferred busy-polling

Quoted 4 lines · Jakub Kicinski
Quoted 3 lines · On Thu, 19 Nov 2020 09:30:15 +0100 Björn Töpel wrote:
include/net/busy_poll.h +0 -0
@@ -105,7 +105,8 @@ static inline void sk_busy_loop(struct sock *sk, int nonblock)
  	unsigned int napi_id = READ_ONCE(sk->sk_napi_id);
 
Quoted 8 lines

include/net/busy_poll.h static inline void sk_busy_loop(struct sock *sk, int nonblock) · line 109

Quoted 4 lines
include/net/busy_poll.h +2 -1
  	if (napi_id >= MIN_NAPI_ID)
-		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk);
+		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk,
+			       READ_ONCE(sk->sk_prefer_busy_poll));

Perhaps a noob question, but aren't all accesses to the new sk members under the socket lock? Do we really need the READ_ONCE() / WRITE_ONCE()?

No, only when setting them via sock_setsockopt. Reading is done outside the lock.

Björn

maintainer Reviewed-by on v3 1/10
#6 lore raw

Re: [PATCH bpf-next v3 01/10] net: introduce preferred busy-polling

In reply to Björn Töpel
Quoted 35 lines · On Thu, 19 Nov 2020 09:30:15 +0100 Björn Töpel wrote:

From: Björn Töpel <[email protected]>

The existing busy-polling mode, enabled by the SO_BUSY_POLL socket option or system-wide using the /proc/sys/net/core/busy_read knob, is an opportunistic. That means that if the NAPI context is not scheduled, it will poll it. If, after busy-polling, the budget is exceeded the busy-polling logic will schedule the NAPI onto the regular softirq handling.

One implication of the behavior above is that a busy/heavy loaded NAPI context will never enter/allow for busy-polling. Some applications prefer that most NAPI processing would be done by busy-polling.

This series adds a new socket option, SO_PREFER_BUSY_POLL, that works in concert with the napi_defer_hard_irqs and gro_flush_timeout knobs. The napi_defer_hard_irqs and gro_flush_timeout knobs were introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature"), and allows for a user to defer interrupts to be enabled and instead schedule the NAPI context from a watchdog timer. When a user enables the SO_PREFER_BUSY_POLL, again with the other knobs enabled, and the NAPI context is being processed by a softirq, the softirq NAPI processing will exit early to allow the busy-polling to be performed.

If the application stops performing busy-polling via a system call, the watchdog timer defined by gro_flush_timeout will timeout, and regular softirq handling will resume.

In summary; Heavy traffic applications that prefer busy-polling over softirq processing should use this option.

Example usage:

  $ echo 2 | sudo tee /sys/class/net/ens785f1/napi_defer_hard_irqs
  $ echo 200000 | sudo tee /sys/class/net/ens785f1/gro_flush_timeout
Quoted 7 lines

Note that the timeout should be larger than the userspace processing window, otherwise the watchdog will timeout and fall back to regular softirq processing.

Enable the SO_BUSY_POLL/SO_PREFER_BUSY_POLL options on your socket.

#7 lore raw

PATCH v3 2/10 net: add SO_BUSY_POLL_BUDGET socket option

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

This option lets a user set a per socket NAPI budget for busy-polling. If the options is not set, it will use the default of 8.

---
 arch/alpha/include/uapi/asm/socket.h  |  1 +
 arch/mips/include/uapi/asm/socket.h   |  1 +
 arch/parisc/include/uapi/asm/socket.h |  1 +
 arch/sparc/include/uapi/asm/socket.h  |  1 +
 fs/eventpoll.c                        |  3 ++-
 include/net/busy_poll.h               |  7 +++++--
 include/net/sock.h                    |  2 ++
 include/uapi/asm-generic/socket.h     |  1 +
 net/core/dev.c                        | 21 ++++++++++-----------
 net/core/sock.c                       | 10 ++++++++++
 10 files changed, 34 insertions(+), 14 deletions(-)
arch/alpha/include/uapi/asm/socket.h +1 -0
index 538359642554..57420356ce4c 100644
@@ -125,6 +125,7 @@
 #define SO_DETACH_REUSEPORT_BPF 68
 
 #define SO_PREFER_BUSY_POLL	69
+#define SO_BUSY_POLL_BUDGET	70
 
 #if !defined(__KERNEL__)
 
arch/mips/include/uapi/asm/socket.h +1 -0
index e406e73b5e6e..2d949969313b 100644
@@ -136,6 +136,7 @@
 #define SO_DETACH_REUSEPORT_BPF 68
 
 #define SO_PREFER_BUSY_POLL	69
+#define SO_BUSY_POLL_BUDGET	70
 
 #if !defined(__KERNEL__)
 
arch/parisc/include/uapi/asm/socket.h +1 -0
index 1bc46200889d..f60904329bbc 100644
@@ -117,6 +117,7 @@
 #define SO_DETACH_REUSEPORT_BPF 0x4042
 
 #define SO_PREFER_BUSY_POLL	0x4043
+#define SO_BUSY_POLL_BUDGET	0x4044
 
 #if !defined(__KERNEL__)
 
arch/sparc/include/uapi/asm/socket.h +1 -0
index 99688cf673a4..848a22fbac20 100644
@@ -118,6 +118,7 @@
 #define SO_DETACH_REUSEPORT_BPF  0x0047
 
 #define SO_PREFER_BUSY_POLL	 0x0048
+#define SO_BUSY_POLL_BUDGET	 0x0049
 
 #if !defined(__KERNEL__)
 
fs/eventpoll.c +2 -1
index e11fab3a0b9e..73c346e503d7 100644
@@ -397,7 +397,8 @@ static void ep_busy_loop(struct eventpoll *ep, int nonblock)
 	unsigned int napi_id = READ_ONCE(ep->napi_id);
 
 	if ((napi_id >= MIN_NAPI_ID) && net_busy_loop_on())
-		napi_busy_loop(napi_id, nonblock ? NULL : ep_busy_loop_end, ep, false);
+		napi_busy_loop(napi_id, nonblock ? NULL : ep_busy_loop_end, ep, false,
+			       BUSY_POLL_BUDGET);
 }
 
 static inline void ep_reset_busy_poll_napi_id(struct eventpoll *ep)
include/net/busy_poll.h +5 -2
index 0292b8353d7e..2f8f51807b83 100644
@@ -23,6 +23,8 @@
  */
 #define MIN_NAPI_ID ((unsigned int)(NR_CPUS + 1))
 
+#define BUSY_POLL_BUDGET 8
+
 #ifdef CONFIG_NET_RX_BUSY_POLL
 
 struct napi_struct;
@@ -43,7 +45,7 @@ bool sk_busy_loop_end(void *p, unsigned long start_time);
 
 void napi_busy_loop(unsigned int napi_id,
 		    bool (*loop_end)(void *, unsigned long),
-		    void *loop_end_arg, bool prefer_busy_poll);
+		    void *loop_end_arg, bool prefer_busy_poll, u16 budget);
 
 #else /* CONFIG_NET_RX_BUSY_POLL */
 static inline unsigned long net_busy_loop_on(void)
@@ -106,7 +108,8 @@ static inline void sk_busy_loop(struct sock *sk, int nonblock)
 
 	if (napi_id >= MIN_NAPI_ID)
 		napi_busy_loop(napi_id, nonblock ? NULL : sk_busy_loop_end, sk,
-			       READ_ONCE(sk->sk_prefer_busy_poll));
+			       READ_ONCE(sk->sk_prefer_busy_poll),
+			       READ_ONCE(sk->sk_busy_poll_budget) ?: BUSY_POLL_BUDGET);
 #endif
 }
 
include/net/sock.h +2 -0
index d49b89b071b6..77ba2c2737db 100644
@@ -302,6 +302,7 @@ struct bpf_local_storage;
   *	@sk_max_ack_backlog: listen backlog set in listen()
   *	@sk_uid: user id of owner
   *	@sk_prefer_busy_poll: prefer busypolling over softirq processing
+  *	@sk_busy_poll_budget: napi processing budget when busypolling
   *	@sk_priority: %SO_PRIORITY setting
   *	@sk_type: socket type (%SOCK_STREAM, etc)
   *	@sk_protocol: which protocol this socket belongs in this network family
@@ -482,6 +483,7 @@ struct sock {
 	kuid_t			sk_uid;
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	u8			sk_prefer_busy_poll;
+	u16			sk_busy_poll_budget;
 #endif
 	struct pid		*sk_peer_pid;
 	const struct cred	*sk_peer_cred;
include/uapi/asm-generic/socket.h +1 -0
index 7dd02408b7ce..4dcd13d097a9 100644
@@ -120,6 +120,7 @@
 #define SO_DETACH_REUSEPORT_BPF 68
 
 #define SO_PREFER_BUSY_POLL	69
+#define SO_BUSY_POLL_BUDGET	70
 
 #if !defined(__KERNEL__)
 
net/core/dev.c +10 -11
index 6f8d2cffb7c5..7a1e5936c67f 100644
@@ -6496,8 +6496,6 @@ static struct napi_struct *napi_by_id(unsigned int napi_id)
 
 #if defined(CONFIG_NET_RX_BUSY_POLL)
 
-#define BUSY_POLL_BUDGET 8
-
 static void __busy_poll_stop(struct napi_struct *napi, bool skip_schedule)
 {
 	if (!skip_schedule) {
@@ -6517,7 +6515,8 @@ static void __busy_poll_stop(struct napi_struct *napi, bool skip_schedule)
 	clear_bit(NAPI_STATE_SCHED, &napi->state);
 }
 
-static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, bool prefer_busy_poll)
+static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, bool prefer_busy_poll,
+			   u16 budget)
 {
 	bool skip_schedule = false;
 	unsigned long timeout;
@@ -6549,21 +6548,21 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, bool
 	/* All we really want here is to re-enable device interrupts.
 	 * Ideally, a new ndo_busy_poll_stop() could avoid another round.
 	 */
-	rc = napi->poll(napi, BUSY_POLL_BUDGET);
+	rc = napi->poll(napi, budget);
 	/* We can't gro_normal_list() here, because napi->poll() might have
 	 * rearmed the napi (napi_complete_done()) in which case it could
 	 * already be running on another CPU.
 	 */
-	trace_napi_poll(napi, rc, BUSY_POLL_BUDGET);
+	trace_napi_poll(napi, rc, budget);
 	netpoll_poll_unlock(have_poll_lock);
-	if (rc == BUSY_POLL_BUDGET)
+	if (rc == budget)
 		__busy_poll_stop(napi, skip_schedule);
 	local_bh_enable();
 }
 
 void napi_busy_loop(unsigned int napi_id,
 		    bool (*loop_end)(void *, unsigned long),
-		    void *loop_end_arg, bool prefer_busy_poll)
+		    void *loop_end_arg, bool prefer_busy_poll, u16 budget)
 {
 	unsigned long start_time = loop_end ? busy_loop_current_time() : 0;
 	int (*napi_poll)(struct napi_struct *napi, int budget);
@@ -6606,8 +6605,8 @@ void napi_busy_loop(unsigned int napi_id,
 			have_poll_lock = netpoll_poll_lock(napi);
 			napi_poll = napi->poll;
 		}
-		work = napi_poll(napi, BUSY_POLL_BUDGET);
-		trace_napi_poll(napi, work, BUSY_POLL_BUDGET);
+		work = napi_poll(napi, budget);
+		trace_napi_poll(napi, work, budget);
 		gro_normal_list(napi);
 count:
 		if (work > 0)
@@ -6620,7 +6619,7 @@ void napi_busy_loop(unsigned int napi_id,
 
 		if (unlikely(need_resched())) {
 			if (napi_poll)
-				busy_poll_stop(napi, have_poll_lock, prefer_busy_poll);
+				busy_poll_stop(napi, have_poll_lock, prefer_busy_poll, budget);
 			preempt_enable();
 			rcu_read_unlock();
 			cond_resched();
@@ -6631,7 +6630,7 @@ void napi_busy_loop(unsigned int napi_id,
 		cpu_relax();
 	}
 	if (napi_poll)
-		busy_poll_stop(napi, have_poll_lock, prefer_busy_poll);
+		busy_poll_stop(napi, have_poll_lock, prefer_busy_poll, budget);
 	preempt_enable();
 out:
 	rcu_read_unlock();
net/core/sock.c +10 -0
index e05f2e52b5a8..d422a6808405 100644
@@ -1165,6 +1165,16 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 		else
 			WRITE_ONCE(sk->sk_prefer_busy_poll, valbool);
 		break;
+	case SO_BUSY_POLL_BUDGET:
+		if (val > READ_ONCE(sk->sk_busy_poll_budget) && !capable(CAP_NET_ADMIN)) {
+			ret = -EPERM;
+		} else {
+			if (val < 0 || val > U16_MAX)
+				ret = -EINVAL;
+			else
+				WRITE_ONCE(sk->sk_busy_poll_budget, val);
+		}
+		break;
 #endif
 
 	case SO_MAX_PACING_RATE:
Signature · 1 line
2.27.0
#9 lore raw

PATCH v3 3/10 xsk: add support for recvmsg()

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Add support for non-blocking recvmsg() to XDP sockets. Previously, only sendmsg() was supported by XDP socket. Now, for symmetry and the upcoming busy-polling support, recvmsg() is added.

---
 net/xdp/xsk.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
net/xdp/xsk.c +21 -1
index b0141973f23e..56a52ec75696 100644
@@ -531,6 +531,26 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 	return __xsk_sendmsg(sk);
 }
 
+static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int flags)
+{
+	bool need_wait = !(flags & MSG_DONTWAIT);
+	struct sock *sk = sock->sk;
+	struct xdp_sock *xs = xdp_sk(sk);
+
+	if (unlikely(!(xs->dev->flags & IFF_UP)))
+		return -ENETDOWN;
+	if (unlikely(!xs->rx))
+		return -ENOBUFS;
+	if (unlikely(!xsk_is_bound(xs)))
+		return -ENXIO;
+	if (unlikely(need_wait))
+		return -EOPNOTSUPP;
+
+	if (xs->pool->cached_need_wakeup & XDP_WAKEUP_RX && xs->zc)
+		return xsk_wakeup(xs, XDP_WAKEUP_RX);
+	return 0;
+}
+
 static __poll_t xsk_poll(struct file *file, struct socket *sock,
 			     struct poll_table_struct *wait)
 {
@@ -1191,7 +1211,7 @@ static const struct proto_ops xsk_proto_ops = {
 	.setsockopt	= xsk_setsockopt,
 	.getsockopt	= xsk_getsockopt,
 	.sendmsg	= xsk_sendmsg,
-	.recvmsg	= sock_no_recvmsg,
+	.recvmsg	= xsk_recvmsg,
 	.mmap		= xsk_mmap,
 	.sendpage	= sock_no_sendpage,
 };
Signature · 1 line
2.27.0
Acked-by on v3 3/10
#10 lore raw

Re: [PATCH bpf-next v3 03/10] xsk: add support for recvmsg()

Quoted 11 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Add support for non-blocking recvmsg() to XDP sockets. Previously, only sendmsg() was supported by XDP socket. Now, for symmetry and the upcoming busy-polling support, recvmsg() is added.

---
 net/xdp/xsk.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
Quoted 43 lines
net/xdp/xsk.c +21 -2
index b0141973f23e..56a52ec75696 100644
@@ -531,6 +531,26 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
        return __xsk_sendmsg(sk);
 }
 
+static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int flags)
+{
+       bool need_wait = !(flags & MSG_DONTWAIT);
+       struct sock *sk = sock->sk;
+       struct xdp_sock *xs = xdp_sk(sk);
+
+       if (unlikely(!(xs->dev->flags & IFF_UP)))
+               return -ENETDOWN;
+       if (unlikely(!xs->rx))
+               return -ENOBUFS;
+       if (unlikely(!xsk_is_bound(xs)))
+               return -ENXIO;
+       if (unlikely(need_wait))
+               return -EOPNOTSUPP;
+
+       if (xs->pool->cached_need_wakeup & XDP_WAKEUP_RX && xs->zc)
+               return xsk_wakeup(xs, XDP_WAKEUP_RX);
+       return 0;
+}
+
 static __poll_t xsk_poll(struct file *file, struct socket *sock,
                             struct poll_table_struct *wait)
 {
@@ -1191,7 +1211,7 @@ static const struct proto_ops xsk_proto_ops = {
        .setsockopt     = xsk_setsockopt,
        .getsockopt     = xsk_getsockopt,
        .sendmsg        = xsk_sendmsg,
-       .recvmsg        = sock_no_recvmsg,
+       .recvmsg        = xsk_recvmsg,
        .mmap           = xsk_mmap,
        .sendpage       = sock_no_sendpage,
 };
--

2.27.0

#11 lore raw

PATCH v3 4/10 xsk: check need wakeup flag in sendmsg()

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Add a check for need wake up in sendmsg(), so that if a user calls sendmsg() when no wakeup is needed, do not trigger a wakeup.

To simplify the need wakeup check in the syscall, unconditionally enable the need wakeup flag for Tx. This has a side-effect for poll(); If poll() is called for a socket without enabled need wakeup, a Tx wakeup is unconditionally performed.

The wakeup matrix for AF_XDP now looks like:

need wakeup | poll() | sendmsg() | recvmsg()
------------+--------------+-------------+------------

disabled    | wake Tx      | wake Tx     | nop
enabled     | check flag;  | check flag; | check flag;
            |   wake Tx/Rx |   wake Tx   |   wake Rx
---
 net/xdp/xsk.c           |  6 +++++-
 net/xdp/xsk_buff_pool.c | 13 ++++++-------
 2 files changed, 11 insertions(+), 8 deletions(-)
net/xdp/xsk.c +5 -1
index 56a52ec75696..bf0f5c34af6c 100644
@@ -522,13 +522,17 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 	bool need_wait = !(m->msg_flags & MSG_DONTWAIT);
 	struct sock *sk = sock->sk;
 	struct xdp_sock *xs = xdp_sk(sk);
+	struct xsk_buff_pool *pool;
 
 	if (unlikely(!xsk_is_bound(xs)))
 		return -ENXIO;
 	if (unlikely(need_wait))
 		return -EOPNOTSUPP;
 
-	return __xsk_sendmsg(sk);
+	pool = xs->pool;
+	if (pool->cached_need_wakeup & XDP_WAKEUP_TX)
+		return __xsk_sendmsg(sk);
+	return 0;
 }
 
 static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int flags)
net/xdp/xsk_buff_pool.c +6 -7
index 8a3bf4e1318e..96bb607853ad 100644
@@ -144,14 +144,13 @@ static int __xp_assign_dev(struct xsk_buff_pool *pool,
 	if (err)
 		return err;
 
-	if (flags & XDP_USE_NEED_WAKEUP) {
+	if (flags & XDP_USE_NEED_WAKEUP)
 		pool->uses_need_wakeup = true;
-		/* Tx needs to be explicitly woken up the first time.
-		 * Also for supporting drivers that do not implement this
-		 * feature. They will always have to call sendto().
-		 */
-		pool->cached_need_wakeup = XDP_WAKEUP_TX;
-	}
+	/* Tx needs to be explicitly woken up the first time.  Also
+	 * for supporting drivers that do not implement this
+	 * feature. They will always have to call sendto() or poll().
+	 */
+	pool->cached_need_wakeup = XDP_WAKEUP_TX;
 
 	dev_hold(netdev);
 
Signature · 1 line
2.27.0
Acked-by on v3 4/10
#12 lore raw

Re: [PATCH bpf-next v3 04/10] xsk: check need wakeup flag in sendmsg()

Quoted 19 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Add a check for need wake up in sendmsg(), so that if a user calls sendmsg() when no wakeup is needed, do not trigger a wakeup.

To simplify the need wakeup check in the syscall, unconditionally enable the need wakeup flag for Tx. This has a side-effect for poll(); If poll() is called for a socket without enabled need wakeup, a Tx wakeup is unconditionally performed.

The wakeup matrix for AF_XDP now looks like:

need wakeup | poll() | sendmsg() | recvmsg()
------------+--------------+-------------+------------

disabled    | wake Tx      | wake Tx     | nop
enabled     | check flag;  | check flag; | check flag;
            |   wake Tx/Rx |   wake Tx   |   wake Rx
Quoted 5 lines
---
 net/xdp/xsk.c           |  6 +++++-
 net/xdp/xsk_buff_pool.c | 13 ++++++-------
 2 files changed, 11 insertions(+), 8 deletions(-)
Quoted 51 lines
net/xdp/xsk.c +5 -1
index 56a52ec75696..bf0f5c34af6c 100644
@@ -522,13 +522,17 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
        bool need_wait = !(m->msg_flags & MSG_DONTWAIT);
        struct sock *sk = sock->sk;
        struct xdp_sock *xs = xdp_sk(sk);
+       struct xsk_buff_pool *pool;
 
        if (unlikely(!xsk_is_bound(xs)))
                return -ENXIO;
        if (unlikely(need_wait))
                return -EOPNOTSUPP;
 
-       return __xsk_sendmsg(sk);
+       pool = xs->pool;
+       if (pool->cached_need_wakeup & XDP_WAKEUP_TX)
+               return __xsk_sendmsg(sk);
+       return 0;
 }
 
 static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int flags)
net/xdp/xsk_buff_pool.c +6 -8
index 8a3bf4e1318e..96bb607853ad 100644
@@ -144,14 +144,13 @@ static int __xp_assign_dev(struct xsk_buff_pool *pool,
        if (err)
                return err;
 
-       if (flags & XDP_USE_NEED_WAKEUP) {
+       if (flags & XDP_USE_NEED_WAKEUP)
                pool->uses_need_wakeup = true;
-               /* Tx needs to be explicitly woken up the first time.
-                * Also for supporting drivers that do not implement this
-                * feature. They will always have to call sendto().
-                */
-               pool->cached_need_wakeup = XDP_WAKEUP_TX;
-       }
+       /* Tx needs to be explicitly woken up the first time.  Also
+        * for supporting drivers that do not implement this
+        * feature. They will always have to call sendto() or poll().
+        */
+       pool->cached_need_wakeup = XDP_WAKEUP_TX;
 
        dev_hold(netdev);
 
--

2.27.0

#13 lore raw

PATCH v3 5/10 xsk: add busy-poll support for {recv,send}msg()

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Wire-up XDP socket busy-poll support for recvmsg() and sendmsg(). If the XDP socket prefers busy-polling, make sure that no wakeup/IPI is performed.

---
 net/xdp/xsk.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
net/xdp/xsk.c +24 -0
index bf0f5c34af6c..ecc4579e41ee 100644
@@ -23,6 +23,7 @@
 #include <linux/netdevice.h>
 #include <linux/rculist.h>
 #include <net/xdp_sock_drv.h>
+#include <net/busy_poll.h>
 #include <net/xdp.h>
 
 #include "xsk_queue.h"
@@ -517,6 +518,17 @@ static int __xsk_sendmsg(struct sock *sk)
 	return xs->zc ? xsk_zc_xmit(xs) : xsk_generic_xmit(sk);
 }
 
+static bool xsk_no_wakeup(struct sock *sk)
+{
+#ifdef CONFIG_NET_RX_BUSY_POLL
+	/* Prefer busy-polling, skip the wakeup. */
+	return READ_ONCE(sk->sk_prefer_busy_poll) && READ_ONCE(sk->sk_ll_usec) &&
+		READ_ONCE(sk->sk_napi_id) >= MIN_NAPI_ID;
+#else
+	return false;
+#endif
+}
+
 static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 {
 	bool need_wait = !(m->msg_flags & MSG_DONTWAIT);
@@ -529,6 +541,12 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 	if (unlikely(need_wait))
 		return -EOPNOTSUPP;
 
+	if (sk_can_busy_loop(sk))
+		sk_busy_loop(sk, 1); /* only support non-blocking sockets */
+
+	if (xsk_no_wakeup(sk))
+		return 0;
+
 	pool = xs->pool;
 	if (pool->cached_need_wakeup & XDP_WAKEUP_TX)
 		return __xsk_sendmsg(sk);
@@ -550,6 +568,12 @@ static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int fl
 	if (unlikely(need_wait))
 		return -EOPNOTSUPP;
 
+	if (sk_can_busy_loop(sk))
+		sk_busy_loop(sk, 1); /* only support non-blocking sockets */
+
+	if (xsk_no_wakeup(sk))
+		return 0;
+
 	if (xs->pool->cached_need_wakeup & XDP_WAKEUP_RX && xs->zc)
 		return xsk_wakeup(xs, XDP_WAKEUP_RX);
 	return 0;
Signature · 1 line
2.27.0
Acked-by on v3 5/10
#14 lore raw

Re: [PATCH bpf-next v3 05/10] xsk: add busy-poll support for {recv,send}msg()

Quoted 11 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Wire-up XDP socket busy-poll support for recvmsg() and sendmsg(). If the XDP socket prefers busy-polling, make sure that no wakeup/IPI is performed.

---
 net/xdp/xsk.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
Quoted 59 lines
net/xdp/xsk.c +24 -1
index bf0f5c34af6c..ecc4579e41ee 100644
@@ -23,6 +23,7 @@
 #include <linux/netdevice.h>
 #include <linux/rculist.h>
 #include <net/xdp_sock_drv.h>
+#include <net/busy_poll.h>
 #include <net/xdp.h>
 
 #include "xsk_queue.h"
@@ -517,6 +518,17 @@ static int __xsk_sendmsg(struct sock *sk)
        return xs->zc ? xsk_zc_xmit(xs) : xsk_generic_xmit(sk);
 }
 
+static bool xsk_no_wakeup(struct sock *sk)
+{
+#ifdef CONFIG_NET_RX_BUSY_POLL
+       /* Prefer busy-polling, skip the wakeup. */
+       return READ_ONCE(sk->sk_prefer_busy_poll) && READ_ONCE(sk->sk_ll_usec) &&
+               READ_ONCE(sk->sk_napi_id) >= MIN_NAPI_ID;
+#else
+       return false;
+#endif
+}
+
 static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 {
        bool need_wait = !(m->msg_flags & MSG_DONTWAIT);
@@ -529,6 +541,12 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
        if (unlikely(need_wait))
                return -EOPNOTSUPP;
 
+       if (sk_can_busy_loop(sk))
+               sk_busy_loop(sk, 1); /* only support non-blocking sockets */
+
+       if (xsk_no_wakeup(sk))
+               return 0;
+
        pool = xs->pool;
        if (pool->cached_need_wakeup & XDP_WAKEUP_TX)
                return __xsk_sendmsg(sk);
@@ -550,6 +568,12 @@ static int xsk_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int fl
        if (unlikely(need_wait))
                return -EOPNOTSUPP;
 
+       if (sk_can_busy_loop(sk))
+               sk_busy_loop(sk, 1); /* only support non-blocking sockets */
+
+       if (xsk_no_wakeup(sk))
+               return 0;
+
        if (xs->pool->cached_need_wakeup & XDP_WAKEUP_RX && xs->zc)
                return xsk_wakeup(xs, XDP_WAKEUP_RX);
        return 0;
--

2.27.0

#15 lore raw

PATCH v3 6/10 xsk: propagate napi_id to XDP socket Rx path

In reply to Björn Töpel

From: Bj?rn T?pel <[email protected]>

Add napi_id to the xdp_rxq_info structure, and make sure the XDP socket pick up the napi_id in the Rx path. The napi_id is used to find the corresponding NAPI structure for socket busy polling.

---
 drivers/net/ethernet/amazon/ena/ena_netdev.c  |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     |  2 +-
 .../ethernet/cavium/thunder/nicvf_queues.c    |  2 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  2 +-
 drivers/net/ethernet/intel/ice/ice_base.c     |  4 ++--
 drivers/net/ethernet/intel/ice/ice_txrx.c     |  2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  2 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/ethernet/marvell/mvneta.c         |  2 +-
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   |  4 ++--
 drivers/net/ethernet/mellanox/mlx4/en_rx.c    |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 .../ethernet/netronome/nfp/nfp_net_common.c   |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_main.c  |  2 +-
 drivers/net/ethernet/sfc/rx_common.c          |  2 +-
 drivers/net/ethernet/socionext/netsec.c       |  2 +-
 drivers/net/ethernet/ti/cpsw_priv.c           |  2 +-
 drivers/net/hyperv/netvsc.c                   |  2 +-
 drivers/net/tun.c                             |  2 +-
 drivers/net/veth.c                            | 12 ++++++++----
 drivers/net/virtio_net.c                      |  2 +-
 drivers/net/xen-netfront.c                    |  2 +-
 include/net/busy_poll.h                       | 19 +++++++++++++++----
 include/net/xdp.h                             |  3 ++-
 net/core/dev.c                                |  2 +-
 net/core/xdp.c                                |  3 ++-
 net/xdp/xsk.c                                 |  1 +
 29 files changed, 54 insertions(+), 36 deletions(-)
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -1
index e8131dadc22c..6ad59f0068f6 100644
@@ -416,7 +416,7 @@ static int ena_xdp_register_rxq_info(struct ena_ring *rx_ring)
 {
 	int rc;
 
-	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid);
+	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid, 0);
 
 	if (rc) {
 		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
index 7975f59735d6..725d929eddb1 100644
@@ -2884,7 +2884,7 @@ static int bnxt_alloc_rx_rings(struct bnxt *bp)
 		if (rc)
 			return rc;
 
-		rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i);
+		rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i, 0);
 		if (rc < 0)
 			return rc;
 
drivers/net/ethernet/cavium/thunder/nicvf_queues.c +1 -1
index 7a141ce32e86..f782e6af45e9 100644
@@ -770,7 +770,7 @@ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs,
 	rq->caching = 1;
 
 	/* Driver have no proper error path for failed XDP RX-queue info reg */
-	WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx) < 0);
+	WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx, 0) < 0);
 
 	/* Send a mailbox msg to PF to config RQ */
 	mbx.rq.msg = NIC_MBOX_MSG_RQ_CFG;
drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +1 -1
index cf9400a9886d..40953980e846 100644
@@ -3334,7 +3334,7 @@ static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv,
 		return 0;
 
 	err = xdp_rxq_info_reg(&fq->channel->xdp_rxq, priv->net_dev,
-			       fq->flowid);
+			       fq->flowid, 0);
 	if (err) {
 		dev_err(dev, "xdp_rxq_info_reg failed\n");
 		return err;
drivers/net/ethernet/intel/i40e/i40e_txrx.c +1 -1
index c21548c71bb1..9f73cd7aee09 100644
@@ -1447,7 +1447,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring)
 	/* XDP RX-queue info only needed for RX rings exposed to XDP */
 	if (rx_ring->vsi->type == I40E_VSI_MAIN) {
 		err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-				       rx_ring->queue_index);
+				       rx_ring->queue_index, rx_ring->q_vector->napi.napi_id);
 		if (err < 0)
 			return err;
 	}
drivers/net/ethernet/intel/ice/ice_base.c +2 -2
index fe4320e2d1f2..3124a3bf519a 100644
@@ -306,7 +306,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 		if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
 			/* coverity[check_return] */
 			xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
-					 ring->q_index);
+					 ring->q_index, ring->q_vector->napi.napi_id);
 
 		ring->xsk_pool = ice_xsk_pool(ring);
 		if (ring->xsk_pool) {
@@ -333,7 +333,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 				/* coverity[check_return] */
 				xdp_rxq_info_reg(&ring->xdp_rxq,
 						 ring->netdev,
-						 ring->q_index);
+						 ring->q_index, ring->q_vector->napi.napi_id);
 
 			err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
 							 MEM_TYPE_PAGE_SHARED,
drivers/net/ethernet/intel/ice/ice_txrx.c +1 -1
index eae75260fe20..77d5eae6b4c2 100644
@@ -483,7 +483,7 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring)
 	if (rx_ring->vsi->type == ICE_VSI_PF &&
 	    !xdp_rxq_info_is_reg(&rx_ring->xdp_rxq))
 		if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-				     rx_ring->q_index))
+				     rx_ring->q_index, rx_ring->q_vector->napi.napi_id))
 			goto err;
 	return 0;
 
drivers/net/ethernet/intel/igb/igb_main.c +1 -1
index 5fc2c381da55..6a4ef4934fcf 100644
@@ -4352,7 +4352,7 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, 0) < 0)
 		goto err;
 
 	return 0;
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +1 -1
index 45ae33e15303..50e6b8b6ba7b 100644
@@ -6577,7 +6577,7 @@ int ixgbe_setup_rx_resources(struct ixgbe_adapter *adapter,
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, rx_ring->q_vector->napi.napi_id) < 0)
 		goto err;
 
 	rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +1 -1
index 82fce27f682b..4061cd7db5dd 100644
@@ -3493,7 +3493,7 @@ int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, 0) < 0)
 		goto err;
 
 	rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/marvell/mvneta.c +1 -1
index 183530ed4d1d..ba6dcb19bb1d 100644
@@ -3227,7 +3227,7 @@ static int mvneta_create_page_pool(struct mvneta_port *pp,
 		return err;
 	}
 
-	err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id);
+	err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id, 0);
 	if (err < 0)
 		goto err_free_pp;
 
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 -2
index 3069e192d773..5504cbc24970 100644
@@ -2614,11 +2614,11 @@ static int mvpp2_rxq_init(struct mvpp2_port *port,
 	mvpp2_rxq_status_update(port, rxq->id, 0, rxq->size);
 
 	if (priv->percpu_pools) {
-		err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id);
+		err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id, 0);
 		if (err < 0)
 			goto err_free_dma;
 
-		err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id);
+		err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id, 0);
 		if (err < 0)
 			goto err_unregister_rxq_short;
 
drivers/net/ethernet/mellanox/mlx4/en_rx.c +1 -1
index b0f79a5151cf..40775cb8fb2a 100644
@@ -283,7 +283,7 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 	ring->log_stride = ffs(ring->stride) - 1;
 	ring->buf_size = ring->size * ring->stride + TXBB_SIZE;
 
-	if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index) < 0)
+	if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index, 0) < 0)
 		goto err_ring;
 
 	tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +1 -1
index 527c5f12c5af..427fc376fe1a 100644
@@ -434,7 +434,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c,
 	rq_xdp_ix = rq->ix;
 	if (xsk)
 		rq_xdp_ix += params->num_channels * MLX5E_RQ_GROUP_XSK;
-	err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix);
+	err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix, 0);
 	if (err < 0)
 		goto err_rq_xdp_prog;
 
drivers/net/ethernet/netronome/nfp/nfp_net_common.c +1 -1
index b150da43adb2..b4acf2f41e84 100644
@@ -2533,7 +2533,7 @@ nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring)
 
 	if (dp->netdev) {
 		err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, dp->netdev,
-				       rx_ring->idx);
+				       rx_ring->idx, rx_ring->r_vec->napi.napi_id);
 		if (err < 0)
 			return err;
 	}
drivers/net/ethernet/qlogic/qede/qede_main.c +1 -1
index 05e3a3b60269..9cf960a6d007 100644
@@ -1762,7 +1762,7 @@ static void qede_init_fp(struct qede_dev *edev)
 
 			/* Driver have no error path from here */
 			WARN_ON(xdp_rxq_info_reg(&fp->rxq->xdp_rxq, edev->ndev,
-						 fp->rxq->rxq_id) < 0);
+						 fp->rxq->rxq_id, 0) < 0);
 
 			if (xdp_rxq_info_reg_mem_model(&fp->rxq->xdp_rxq,
 						       MEM_TYPE_PAGE_ORDER0,
drivers/net/ethernet/sfc/rx_common.c +1 -1
index 19cf7cac1e6e..68fc7d317693 100644
@@ -262,7 +262,7 @@ void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
 
 	/* Initialise XDP queue information */
 	rc = xdp_rxq_info_reg(&rx_queue->xdp_rxq_info, efx->net_dev,
-			      rx_queue->core_index);
+			      rx_queue->core_index, 0);
 
 	if (rc) {
 		netif_err(efx, rx_err, efx->net_dev,
drivers/net/ethernet/socionext/netsec.c +1 -1
index 1503cc9ec6e2..27d3c9d9210e 100644
@@ -1304,7 +1304,7 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv)
 		goto err_out;
 	}
 
-	err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0);
+	err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0, priv->napi.napi_id);
 	if (err)
 		goto err_out;
 
drivers/net/ethernet/ti/cpsw_priv.c +1 -1
index 31c5e36ff706..6dd73bd0f458 100644
@@ -1186,7 +1186,7 @@ static int cpsw_ndev_create_xdp_rxq(struct cpsw_priv *priv, int ch)
 	pool = cpsw->page_pool[ch];
 	rxq = &priv->xdp_rxq[ch];
 
-	ret = xdp_rxq_info_reg(rxq, priv->ndev, ch);
+	ret = xdp_rxq_info_reg(rxq, priv->ndev, ch, 0);
 	if (ret)
 		return ret;
 
drivers/net/hyperv/netvsc.c +1 -1
index 0c3de94b5178..fa8341f8359a 100644
@@ -1499,7 +1499,7 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device,
 		u64_stats_init(&nvchan->tx_stats.syncp);
 		u64_stats_init(&nvchan->rx_stats.syncp);
 
-		ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i);
+		ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i, 0);
 
 		if (ret) {
 			netdev_err(ndev, "xdp_rxq_info_reg fail: %d\n", ret);
drivers/net/tun.c +1 -1
index 3d45d56172cb..8867d39db6ac 100644
@@ -780,7 +780,7 @@ static int tun_attach(struct tun_struct *tun, struct file *file,
 	} else {
 		/* Setup XDP RX-queue info, for new tfile getting attached */
 		err = xdp_rxq_info_reg(&tfile->xdp_rxq,
-				       tun->dev, tfile->queue_index);
+				       tun->dev, tfile->queue_index, 0);
 		if (err < 0)
 			goto out;
 		err = xdp_rxq_info_reg_mem_model(&tfile->xdp_rxq,
drivers/net/veth.c +8 -4
index 8c737668008a..9bd37c7151f8 100644
@@ -884,7 +884,6 @@ static int veth_napi_add(struct net_device *dev)
 	for (i = 0; i < dev->real_num_rx_queues; i++) {
 		struct veth_rq *rq = &priv->rq[i];
 
-		netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
 		napi_enable(&rq->xdp_napi);
 	}
 
@@ -926,7 +925,8 @@ static int veth_enable_xdp(struct net_device *dev)
 		for (i = 0; i < dev->real_num_rx_queues; i++) {
 			struct veth_rq *rq = &priv->rq[i];
 
-			err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i);
+			netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
+			err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i, rq->xdp_napi.napi_id);
 			if (err < 0)
 				goto err_rxq_reg;
 
@@ -952,8 +952,12 @@ static int veth_enable_xdp(struct net_device *dev)
 err_reg_mem:
 	xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
 err_rxq_reg:
-	for (i--; i >= 0; i--)
-		xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
+	for (i--; i >= 0; i--) {
+		struct veth_rq *rq = &priv->rq[i];
+
+		xdp_rxq_info_unreg(&rq->xdp_rxq);
+		netif_napi_del(&rq->xdp_napi);
+	}
 
 	return err;
 }
drivers/net/virtio_net.c +1 -1
index 21b71148c532..052975ea0af4 100644
@@ -1485,7 +1485,7 @@ static int virtnet_open(struct net_device *dev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 
-		err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i);
+		err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i, vi->rq[i].napi.napi_id);
 		if (err < 0)
 			return err;
 
drivers/net/xen-netfront.c +1 -1
index 920cac4385bf..b01848ef4649 100644
@@ -2014,7 +2014,7 @@ static int xennet_create_page_pool(struct netfront_queue *queue)
 	}
 
 	err = xdp_rxq_info_reg(&queue->xdp_rxq, queue->info->netdev,
-			       queue->id);
+			       queue->id, 0);
 	if (err) {
 		netdev_err(queue->info->netdev, "xdp_rxq_info_reg failed\n");
 		goto err_free_pp;
include/net/busy_poll.h +15 -4
index 2f8f51807b83..45b3e04b99d3 100644
@@ -135,14 +135,25 @@ static inline void sk_mark_napi_id(struct sock *sk, const struct sk_buff *skb)
 	sk_rx_queue_set(sk, skb);
 }
 
-/* variant used for unconnected sockets */
-static inline void sk_mark_napi_id_once(struct sock *sk,
-					const struct sk_buff *skb)
+static inline void __sk_mark_napi_id_once_xdp(struct sock *sk, unsigned int napi_id)
 {
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	if (!READ_ONCE(sk->sk_napi_id))
-		WRITE_ONCE(sk->sk_napi_id, skb->napi_id);
+		WRITE_ONCE(sk->sk_napi_id, napi_id);
 #endif
 }
 
+/* variant used for unconnected sockets */
+static inline void sk_mark_napi_id_once(struct sock *sk,
+					const struct sk_buff *skb)
+{
+	__sk_mark_napi_id_once_xdp(sk, skb->napi_id);
+}
+
+static inline void sk_mark_napi_id_once_xdp(struct sock *sk,
+					    const struct xdp_buff *xdp)
+{
+	__sk_mark_napi_id_once_xdp(sk, xdp->rxq->napi_id);
+}
+
 #endif /* _LINUX_NET_BUSY_POLL_H */
include/net/xdp.h +2 -1
index 7d48b2ae217a..700ad5db7f5d 100644
@@ -59,6 +59,7 @@ struct xdp_rxq_info {
 	u32 queue_index;
 	u32 reg_state;
 	struct xdp_mem_info mem;
+	unsigned int napi_id;
 } ____cacheline_aligned; /* perf critical, avoid false-sharing */
 
 struct xdp_txq_info {
@@ -226,7 +227,7 @@ static inline void xdp_release_frame(struct xdp_frame *xdpf)
 }
 
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-		     struct net_device *dev, u32 queue_index);
+		     struct net_device *dev, u32 queue_index, unsigned int napi_id);
 void xdp_rxq_info_unreg(struct xdp_rxq_info *xdp_rxq);
 void xdp_rxq_info_unused(struct xdp_rxq_info *xdp_rxq);
 bool xdp_rxq_info_is_reg(struct xdp_rxq_info *xdp_rxq);
net/core/dev.c +1 -1
index 7a1e5936c67f..3b6b0e175fe7 100644
@@ -9810,7 +9810,7 @@ static int netif_alloc_rx_queues(struct net_device *dev)
 		rx[i].dev = dev;
 
 		/* XDP RX-queue setup */
-		err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i);
+		err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i, 0);
 		if (err < 0)
 			goto err_rxq_info;
 	}
net/core/xdp.c +2 -1
index 3d330ebda893..17ffd33c6b18 100644
@@ -158,7 +158,7 @@ static void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq)
 
 /* Returns 0 on success, negative on failure */
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-		     struct net_device *dev, u32 queue_index)
+		     struct net_device *dev, u32 queue_index, unsigned int napi_id)
 {
 	if (xdp_rxq->reg_state == REG_STATE_UNUSED) {
 		WARN(1, "Driver promised not to register this");
@@ -179,6 +179,7 @@ int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
 	xdp_rxq_info_init(xdp_rxq);
 	xdp_rxq->dev = dev;
 	xdp_rxq->queue_index = queue_index;
+	xdp_rxq->napi_id = napi_id;
 
 	xdp_rxq->reg_state = REG_STATE_REGISTERED;
 	return 0;
net/xdp/xsk.c +1 -0
index ecc4579e41ee..d4cb1c5c1abf 100644
@@ -233,6 +233,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp,
 	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index)
 		return -EINVAL;
 
+	sk_mark_napi_id_once_xdp(&xs->sk, xdp);
 	len = xdp->data_end - xdp->data;
 
 	return xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL ?
Signature · 1 line
2.27.0
Acked-by on v3 6/10
#16 lore raw

Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

Quoted 33 lines · Bj?rn T?pel

From: Bj?rn T?pel <[email protected]>

Add napi_id to the xdp_rxq_info structure, and make sure the XDP socket pick up the napi_id in the Rx path. The napi_id is used to find the corresponding NAPI structure for socket busy polling.

---
 drivers/net/ethernet/amazon/ena/ena_netdev.c  |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     |  2 +-
 .../ethernet/cavium/thunder/nicvf_queues.c    |  2 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  2 +-
 drivers/net/ethernet/intel/ice/ice_base.c     |  4 ++--
 drivers/net/ethernet/intel/ice/ice_txrx.c     |  2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  2 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/ethernet/marvell/mvneta.c         |  2 +-
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   |  4 ++--
 drivers/net/ethernet/mellanox/mlx4/en_rx.c    |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 .../ethernet/netronome/nfp/nfp_net_common.c   |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_main.c  |  2 +-
 drivers/net/ethernet/sfc/rx_common.c          |  2 +-
 drivers/net/ethernet/socionext/netsec.c       |  2 +-
 drivers/net/ethernet/ti/cpsw_priv.c           |  2 +-
 drivers/net/hyperv/netvsc.c                   |  2 +-
 drivers/net/tun.c                             |  2 +-
 drivers/net/veth.c                            | 12 ++++++++----
Quoted 8 lines
 drivers/net/virtio_net.c                      |  2 +-
 drivers/net/xen-netfront.c                    |  2 +-
 include/net/busy_poll.h                       | 19 +++++++++++++++----
 include/net/xdp.h                             |  3 ++-
 net/core/dev.c                                |  2 +-
 net/core/xdp.c                                |  3 ++-
 net/xdp/xsk.c                                 |  1 +
 29 files changed, 54 insertions(+), 36 deletions(-)

For Intel drivers:

Quoted 454 lines
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -1
index e8131dadc22c..6ad59f0068f6 100644
@@ -416,7 +416,7 @@ static int ena_xdp_register_rxq_info(struct ena_ring *rx_ring)
 {
        int rc;
 
-       rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid);
+       rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid, 0);
 
        if (rc) {
                netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
index 7975f59735d6..725d929eddb1 100644
@@ -2884,7 +2884,7 @@ static int bnxt_alloc_rx_rings(struct bnxt *bp)
                if (rc)
                        return rc;
 
-               rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i);
+               rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i, 0);
                if (rc < 0)
                        return rc;
 
drivers/net/ethernet/cavium/thunder/nicvf_queues.c +1 -1
index 7a141ce32e86..f782e6af45e9 100644
@@ -770,7 +770,7 @@ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs,
        rq->caching = 1;
 
        /* Driver have no proper error path for failed XDP RX-queue info reg */
-       WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx) < 0);
+       WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx, 0) < 0);
 
        /* Send a mailbox msg to PF to config RQ */
        mbx.rq.msg = NIC_MBOX_MSG_RQ_CFG;
drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +1 -1
index cf9400a9886d..40953980e846 100644
@@ -3334,7 +3334,7 @@ static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv,
                return 0;
 
        err = xdp_rxq_info_reg(&fq->channel->xdp_rxq, priv->net_dev,
-                              fq->flowid);
+                              fq->flowid, 0);
        if (err) {
                dev_err(dev, "xdp_rxq_info_reg failed\n");
                return err;
drivers/net/ethernet/intel/i40e/i40e_txrx.c +1 -1
index c21548c71bb1..9f73cd7aee09 100644
@@ -1447,7 +1447,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring)
        /* XDP RX-queue info only needed for RX rings exposed to XDP */
        if (rx_ring->vsi->type == I40E_VSI_MAIN) {
                err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-                                      rx_ring->queue_index);
+                                      rx_ring->queue_index, rx_ring->q_vector->napi.napi_id);
                if (err < 0)
                        return err;
        }
drivers/net/ethernet/intel/ice/ice_base.c +2 -2
index fe4320e2d1f2..3124a3bf519a 100644
@@ -306,7 +306,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
                if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
                        /* coverity[check_return] */
                        xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
-                                        ring->q_index);
+                                        ring->q_index, ring->q_vector->napi.napi_id);
 
                ring->xsk_pool = ice_xsk_pool(ring);
                if (ring->xsk_pool) {
@@ -333,7 +333,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
                                /* coverity[check_return] */
                                xdp_rxq_info_reg(&ring->xdp_rxq,
                                                 ring->netdev,
-                                                ring->q_index);
+                                                ring->q_index, ring->q_vector->napi.napi_id);
 
                        err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
                                                         MEM_TYPE_PAGE_SHARED,
drivers/net/ethernet/intel/ice/ice_txrx.c +1 -1
index eae75260fe20..77d5eae6b4c2 100644
@@ -483,7 +483,7 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring)
        if (rx_ring->vsi->type == ICE_VSI_PF &&
            !xdp_rxq_info_is_reg(&rx_ring->xdp_rxq))
                if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-                                    rx_ring->q_index))
+                                    rx_ring->q_index, rx_ring->q_vector->napi.napi_id))
                        goto err;
        return 0;
 
drivers/net/ethernet/intel/igb/igb_main.c +1 -1
index 5fc2c381da55..6a4ef4934fcf 100644
@@ -4352,7 +4352,7 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
 
        /* XDP RX-queue info */
        if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-                            rx_ring->queue_index) < 0)
+                            rx_ring->queue_index, 0) < 0)
                goto err;
 
        return 0;
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +1 -1
index 45ae33e15303..50e6b8b6ba7b 100644
@@ -6577,7 +6577,7 @@ int ixgbe_setup_rx_resources(struct ixgbe_adapter *adapter,
 
        /* XDP RX-queue info */
        if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-                            rx_ring->queue_index) < 0)
+                            rx_ring->queue_index, rx_ring->q_vector->napi.napi_id) < 0)
                goto err;
 
        rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +1 -1
index 82fce27f682b..4061cd7db5dd 100644
@@ -3493,7 +3493,7 @@ int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
 
        /* XDP RX-queue info */
        if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-                            rx_ring->queue_index) < 0)
+                            rx_ring->queue_index, 0) < 0)
                goto err;
 
        rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/marvell/mvneta.c +1 -1
index 183530ed4d1d..ba6dcb19bb1d 100644
@@ -3227,7 +3227,7 @@ static int mvneta_create_page_pool(struct mvneta_port *pp,
                return err;
        }
 
-       err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id);
+       err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id, 0);
        if (err < 0)
                goto err_free_pp;
 
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 -2
index 3069e192d773..5504cbc24970 100644
@@ -2614,11 +2614,11 @@ static int mvpp2_rxq_init(struct mvpp2_port *port,
        mvpp2_rxq_status_update(port, rxq->id, 0, rxq->size);
 
        if (priv->percpu_pools) {
-               err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id);
+               err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id, 0);
                if (err < 0)
                        goto err_free_dma;
 
-               err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id);
+               err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id, 0);
                if (err < 0)
                        goto err_unregister_rxq_short;
 
drivers/net/ethernet/mellanox/mlx4/en_rx.c +1 -1
index b0f79a5151cf..40775cb8fb2a 100644
@@ -283,7 +283,7 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
        ring->log_stride = ffs(ring->stride) - 1;
        ring->buf_size = ring->size * ring->stride + TXBB_SIZE;
 
-       if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index) < 0)
+       if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index, 0) < 0)
                goto err_ring;
 
        tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +1 -1
index 527c5f12c5af..427fc376fe1a 100644
@@ -434,7 +434,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c,
        rq_xdp_ix = rq->ix;
        if (xsk)
                rq_xdp_ix += params->num_channels * MLX5E_RQ_GROUP_XSK;
-       err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix);
+       err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix, 0);
        if (err < 0)
                goto err_rq_xdp_prog;
 
drivers/net/ethernet/netronome/nfp/nfp_net_common.c +1 -1
index b150da43adb2..b4acf2f41e84 100644
@@ -2533,7 +2533,7 @@ nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring)
 
        if (dp->netdev) {
                err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, dp->netdev,
-                                      rx_ring->idx);
+                                      rx_ring->idx, rx_ring->r_vec->napi.napi_id);
                if (err < 0)
                        return err;
        }
drivers/net/ethernet/qlogic/qede/qede_main.c +1 -1
index 05e3a3b60269..9cf960a6d007 100644
@@ -1762,7 +1762,7 @@ static void qede_init_fp(struct qede_dev *edev)
 
                        /* Driver have no error path from here */
                        WARN_ON(xdp_rxq_info_reg(&fp->rxq->xdp_rxq, edev->ndev,
-                                                fp->rxq->rxq_id) < 0);
+                                                fp->rxq->rxq_id, 0) < 0);
 
                        if (xdp_rxq_info_reg_mem_model(&fp->rxq->xdp_rxq,
                                                       MEM_TYPE_PAGE_ORDER0,
drivers/net/ethernet/sfc/rx_common.c +1 -1
index 19cf7cac1e6e..68fc7d317693 100644
@@ -262,7 +262,7 @@ void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
 
        /* Initialise XDP queue information */
        rc = xdp_rxq_info_reg(&rx_queue->xdp_rxq_info, efx->net_dev,
-                             rx_queue->core_index);
+                             rx_queue->core_index, 0);
 
        if (rc) {
                netif_err(efx, rx_err, efx->net_dev,
drivers/net/ethernet/socionext/netsec.c +1 -1
index 1503cc9ec6e2..27d3c9d9210e 100644
@@ -1304,7 +1304,7 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv)
                goto err_out;
        }
 
-       err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0);
+       err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0, priv->napi.napi_id);
        if (err)
                goto err_out;
 
drivers/net/ethernet/ti/cpsw_priv.c +1 -1
index 31c5e36ff706..6dd73bd0f458 100644
@@ -1186,7 +1186,7 @@ static int cpsw_ndev_create_xdp_rxq(struct cpsw_priv *priv, int ch)
        pool = cpsw->page_pool[ch];
        rxq = &priv->xdp_rxq[ch];
 
-       ret = xdp_rxq_info_reg(rxq, priv->ndev, ch);
+       ret = xdp_rxq_info_reg(rxq, priv->ndev, ch, 0);
        if (ret)
                return ret;
 
drivers/net/hyperv/netvsc.c +1 -1
index 0c3de94b5178..fa8341f8359a 100644
@@ -1499,7 +1499,7 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device,
                u64_stats_init(&nvchan->tx_stats.syncp);
                u64_stats_init(&nvchan->rx_stats.syncp);
 
-               ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i);
+               ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i, 0);
 
                if (ret) {
                        netdev_err(ndev, "xdp_rxq_info_reg fail: %d\n", ret);
drivers/net/tun.c +1 -1
index 3d45d56172cb..8867d39db6ac 100644
@@ -780,7 +780,7 @@ static int tun_attach(struct tun_struct *tun, struct file *file,
        } else {
                /* Setup XDP RX-queue info, for new tfile getting attached */
                err = xdp_rxq_info_reg(&tfile->xdp_rxq,
-                                      tun->dev, tfile->queue_index);
+                                      tun->dev, tfile->queue_index, 0);
                if (err < 0)
                        goto out;
                err = xdp_rxq_info_reg_mem_model(&tfile->xdp_rxq,
drivers/net/veth.c +8 -4
index 8c737668008a..9bd37c7151f8 100644
@@ -884,7 +884,6 @@ static int veth_napi_add(struct net_device *dev)
        for (i = 0; i < dev->real_num_rx_queues; i++) {
                struct veth_rq *rq = &priv->rq[i];
 
-               netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
                napi_enable(&rq->xdp_napi);
        }
 
@@ -926,7 +925,8 @@ static int veth_enable_xdp(struct net_device *dev)
                for (i = 0; i < dev->real_num_rx_queues; i++) {
                        struct veth_rq *rq = &priv->rq[i];
 
-                       err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i);
+                       netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
+                       err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i, rq->xdp_napi.napi_id);
                        if (err < 0)
                                goto err_rxq_reg;
 
@@ -952,8 +952,12 @@ static int veth_enable_xdp(struct net_device *dev)
 err_reg_mem:
        xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
 err_rxq_reg:
-       for (i--; i >= 0; i--)
-               xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
+       for (i--; i >= 0; i--) {
+               struct veth_rq *rq = &priv->rq[i];
+
+               xdp_rxq_info_unreg(&rq->xdp_rxq);
+               netif_napi_del(&rq->xdp_napi);
+       }
 
        return err;
 }
drivers/net/virtio_net.c +1 -1
index 21b71148c532..052975ea0af4 100644
@@ -1485,7 +1485,7 @@ static int virtnet_open(struct net_device *dev)
                        if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
                                schedule_delayed_work(&vi->refill, 0);
 
-               err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i);
+               err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i, vi->rq[i].napi.napi_id);
                if (err < 0)
                        return err;
 
drivers/net/xen-netfront.c +1 -1
index 920cac4385bf..b01848ef4649 100644
@@ -2014,7 +2014,7 @@ static int xennet_create_page_pool(struct netfront_queue *queue)
        }
 
        err = xdp_rxq_info_reg(&queue->xdp_rxq, queue->info->netdev,
-                              queue->id);
+                              queue->id, 0);
        if (err) {
                netdev_err(queue->info->netdev, "xdp_rxq_info_reg failed\n");
                goto err_free_pp;
include/net/busy_poll.h +15 -4
index 2f8f51807b83..45b3e04b99d3 100644
@@ -135,14 +135,25 @@ static inline void sk_mark_napi_id(struct sock *sk, const struct sk_buff *skb)
        sk_rx_queue_set(sk, skb);
 }
 
-/* variant used for unconnected sockets */
-static inline void sk_mark_napi_id_once(struct sock *sk,
-                                       const struct sk_buff *skb)
+static inline void __sk_mark_napi_id_once_xdp(struct sock *sk, unsigned int napi_id)
 {
 #ifdef CONFIG_NET_RX_BUSY_POLL
        if (!READ_ONCE(sk->sk_napi_id))
-               WRITE_ONCE(sk->sk_napi_id, skb->napi_id);
+               WRITE_ONCE(sk->sk_napi_id, napi_id);
 #endif
 }
 
+/* variant used for unconnected sockets */
+static inline void sk_mark_napi_id_once(struct sock *sk,
+                                       const struct sk_buff *skb)
+{
+       __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
+}
+
+static inline void sk_mark_napi_id_once_xdp(struct sock *sk,
+                                           const struct xdp_buff *xdp)
+{
+       __sk_mark_napi_id_once_xdp(sk, xdp->rxq->napi_id);
+}
+
 #endif /* _LINUX_NET_BUSY_POLL_H */
include/net/xdp.h +2 -1
index 7d48b2ae217a..700ad5db7f5d 100644
@@ -59,6 +59,7 @@ struct xdp_rxq_info {
        u32 queue_index;
        u32 reg_state;
        struct xdp_mem_info mem;
+       unsigned int napi_id;
 } ____cacheline_aligned; /* perf critical, avoid false-sharing */
 
 struct xdp_txq_info {
@@ -226,7 +227,7 @@ static inline void xdp_release_frame(struct xdp_frame *xdpf)
 }
 
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-                    struct net_device *dev, u32 queue_index);
+                    struct net_device *dev, u32 queue_index, unsigned int napi_id);
 void xdp_rxq_info_unreg(struct xdp_rxq_info *xdp_rxq);
 void xdp_rxq_info_unused(struct xdp_rxq_info *xdp_rxq);
 bool xdp_rxq_info_is_reg(struct xdp_rxq_info *xdp_rxq);
net/core/dev.c +1 -1
index 7a1e5936c67f..3b6b0e175fe7 100644
@@ -9810,7 +9810,7 @@ static int netif_alloc_rx_queues(struct net_device *dev)
                rx[i].dev = dev;
 
                /* XDP RX-queue setup */
-               err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i);
+               err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i, 0);
                if (err < 0)
                        goto err_rxq_info;
        }
net/core/xdp.c +2 -1
index 3d330ebda893..17ffd33c6b18 100644
@@ -158,7 +158,7 @@ static void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq)
 
 /* Returns 0 on success, negative on failure */
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-                    struct net_device *dev, u32 queue_index)
+                    struct net_device *dev, u32 queue_index, unsigned int napi_id)
 {
        if (xdp_rxq->reg_state == REG_STATE_UNUSED) {
                WARN(1, "Driver promised not to register this");
@@ -179,6 +179,7 @@ int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
        xdp_rxq_info_init(xdp_rxq);
        xdp_rxq->dev = dev;
        xdp_rxq->queue_index = queue_index;
+       xdp_rxq->napi_id = napi_id;
 
        xdp_rxq->reg_state = REG_STATE_REGISTERED;
        return 0;
net/xdp/xsk.c +1 -1
index ecc4579e41ee..d4cb1c5c1abf 100644
@@ -233,6 +233,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp,
        if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index)
                return -EINVAL;
 
+       sk_mark_napi_id_once_xdp(&xs->sk, xdp);
        len = xdp->data_end - xdp->data;
 
        return xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL ?
--

2.27.0

maintainer Acked-by on v3 6/10
#17 lore raw

Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

In reply to Björn Töpel
Quoted 9 lines · Bj??rn T??pel

From: Bj??rn T??pel <[email protected]>

Add napi_id to the xdp_rxq_info structure, and make sure the XDP socket pick up the napi_id in the Rx path. The napi_id is used to find the corresponding NAPI structure for socket busy polling.

For virtio:

Quoted 485 lines
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c  |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     |  2 +-
 .../ethernet/cavium/thunder/nicvf_queues.c    |  2 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  2 +-
 drivers/net/ethernet/intel/ice/ice_base.c     |  4 ++--
 drivers/net/ethernet/intel/ice/ice_txrx.c     |  2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  2 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/ethernet/marvell/mvneta.c         |  2 +-
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   |  4 ++--
 drivers/net/ethernet/mellanox/mlx4/en_rx.c    |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 .../ethernet/netronome/nfp/nfp_net_common.c   |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_main.c  |  2 +-
 drivers/net/ethernet/sfc/rx_common.c          |  2 +-
 drivers/net/ethernet/socionext/netsec.c       |  2 +-
 drivers/net/ethernet/ti/cpsw_priv.c           |  2 +-
 drivers/net/hyperv/netvsc.c                   |  2 +-
 drivers/net/tun.c                             |  2 +-
 drivers/net/veth.c                            | 12 ++++++++----
 drivers/net/virtio_net.c                      |  2 +-
 drivers/net/xen-netfront.c                    |  2 +-
 include/net/busy_poll.h                       | 19 +++++++++++++++----
 include/net/xdp.h                             |  3 ++-
 net/core/dev.c                                |  2 +-
 net/core/xdp.c                                |  3 ++-
 net/xdp/xsk.c                                 |  1 +
 29 files changed, 54 insertions(+), 36 deletions(-)
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -1
index e8131dadc22c..6ad59f0068f6 100644
@@ -416,7 +416,7 @@ static int ena_xdp_register_rxq_info(struct ena_ring *rx_ring)
 {
 	int rc;
 
-	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid);
+	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid, 0);
 
 	if (rc) {
 		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
drivers/net/ethernet/broadcom/bnxt/bnxt.c +1 -1
index 7975f59735d6..725d929eddb1 100644
@@ -2884,7 +2884,7 @@ static int bnxt_alloc_rx_rings(struct bnxt *bp)
 		if (rc)
 			return rc;
 
-		rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i);
+		rc = xdp_rxq_info_reg(&rxr->xdp_rxq, bp->dev, i, 0);
 		if (rc < 0)
 			return rc;
 
drivers/net/ethernet/cavium/thunder/nicvf_queues.c +1 -1
index 7a141ce32e86..f782e6af45e9 100644
@@ -770,7 +770,7 @@ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs,
 	rq->caching = 1;
 
 	/* Driver have no proper error path for failed XDP RX-queue info reg */
-	WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx) < 0);
+	WARN_ON(xdp_rxq_info_reg(&rq->xdp_rxq, nic->netdev, qidx, 0) < 0);
 
 	/* Send a mailbox msg to PF to config RQ */
 	mbx.rq.msg = NIC_MBOX_MSG_RQ_CFG;
drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +1 -1
index cf9400a9886d..40953980e846 100644
@@ -3334,7 +3334,7 @@ static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv,
 		return 0;
 
 	err = xdp_rxq_info_reg(&fq->channel->xdp_rxq, priv->net_dev,
-			       fq->flowid);
+			       fq->flowid, 0);
 	if (err) {
 		dev_err(dev, "xdp_rxq_info_reg failed\n");
 		return err;
drivers/net/ethernet/intel/i40e/i40e_txrx.c +1 -1
index c21548c71bb1..9f73cd7aee09 100644
@@ -1447,7 +1447,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring)
 	/* XDP RX-queue info only needed for RX rings exposed to XDP */
 	if (rx_ring->vsi->type == I40E_VSI_MAIN) {
 		err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-				       rx_ring->queue_index);
+				       rx_ring->queue_index, rx_ring->q_vector->napi.napi_id);
 		if (err < 0)
 			return err;
 	}
drivers/net/ethernet/intel/ice/ice_base.c +2 -2
index fe4320e2d1f2..3124a3bf519a 100644
@@ -306,7 +306,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 		if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
 			/* coverity[check_return] */
 			xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
-					 ring->q_index);
+					 ring->q_index, ring->q_vector->napi.napi_id);
 
 		ring->xsk_pool = ice_xsk_pool(ring);
 		if (ring->xsk_pool) {
@@ -333,7 +333,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 				/* coverity[check_return] */
 				xdp_rxq_info_reg(&ring->xdp_rxq,
 						 ring->netdev,
-						 ring->q_index);
+						 ring->q_index, ring->q_vector->napi.napi_id);
 
 			err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
 							 MEM_TYPE_PAGE_SHARED,
drivers/net/ethernet/intel/ice/ice_txrx.c +1 -1
index eae75260fe20..77d5eae6b4c2 100644
@@ -483,7 +483,7 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring)
 	if (rx_ring->vsi->type == ICE_VSI_PF &&
 	    !xdp_rxq_info_is_reg(&rx_ring->xdp_rxq))
 		if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-				     rx_ring->q_index))
+				     rx_ring->q_index, rx_ring->q_vector->napi.napi_id))
 			goto err;
 	return 0;
 
drivers/net/ethernet/intel/igb/igb_main.c +1 -1
index 5fc2c381da55..6a4ef4934fcf 100644
@@ -4352,7 +4352,7 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, 0) < 0)
 		goto err;
 
 	return 0;
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +1 -1
index 45ae33e15303..50e6b8b6ba7b 100644
@@ -6577,7 +6577,7 @@ int ixgbe_setup_rx_resources(struct ixgbe_adapter *adapter,
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, rx_ring->q_vector->napi.napi_id) < 0)
 		goto err;
 
 	rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +1 -1
index 82fce27f682b..4061cd7db5dd 100644
@@ -3493,7 +3493,7 @@ int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
 
 	/* XDP RX-queue info */
 	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-			     rx_ring->queue_index) < 0)
+			     rx_ring->queue_index, 0) < 0)
 		goto err;
 
 	rx_ring->xdp_prog = adapter->xdp_prog;
drivers/net/ethernet/marvell/mvneta.c +1 -1
index 183530ed4d1d..ba6dcb19bb1d 100644
@@ -3227,7 +3227,7 @@ static int mvneta_create_page_pool(struct mvneta_port *pp,
 		return err;
 	}
 
-	err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id);
+	err = xdp_rxq_info_reg(&rxq->xdp_rxq, pp->dev, rxq->id, 0);
 	if (err < 0)
 		goto err_free_pp;
 
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 -2
index 3069e192d773..5504cbc24970 100644
@@ -2614,11 +2614,11 @@ static int mvpp2_rxq_init(struct mvpp2_port *port,
 	mvpp2_rxq_status_update(port, rxq->id, 0, rxq->size);
 
 	if (priv->percpu_pools) {
-		err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id);
+		err = xdp_rxq_info_reg(&rxq->xdp_rxq_short, port->dev, rxq->id, 0);
 		if (err < 0)
 			goto err_free_dma;
 
-		err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id);
+		err = xdp_rxq_info_reg(&rxq->xdp_rxq_long, port->dev, rxq->id, 0);
 		if (err < 0)
 			goto err_unregister_rxq_short;
 
drivers/net/ethernet/mellanox/mlx4/en_rx.c +1 -1
index b0f79a5151cf..40775cb8fb2a 100644
@@ -283,7 +283,7 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 	ring->log_stride = ffs(ring->stride) - 1;
 	ring->buf_size = ring->size * ring->stride + TXBB_SIZE;
 
-	if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index) < 0)
+	if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index, 0) < 0)
 		goto err_ring;
 
 	tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +1 -1
index 527c5f12c5af..427fc376fe1a 100644
@@ -434,7 +434,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c,
 	rq_xdp_ix = rq->ix;
 	if (xsk)
 		rq_xdp_ix += params->num_channels * MLX5E_RQ_GROUP_XSK;
-	err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix);
+	err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix, 0);
 	if (err < 0)
 		goto err_rq_xdp_prog;
 
drivers/net/ethernet/netronome/nfp/nfp_net_common.c +1 -1
index b150da43adb2..b4acf2f41e84 100644
@@ -2533,7 +2533,7 @@ nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring)
 
 	if (dp->netdev) {
 		err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, dp->netdev,
-				       rx_ring->idx);
+				       rx_ring->idx, rx_ring->r_vec->napi.napi_id);
 		if (err < 0)
 			return err;
 	}
drivers/net/ethernet/qlogic/qede/qede_main.c +1 -1
index 05e3a3b60269..9cf960a6d007 100644
@@ -1762,7 +1762,7 @@ static void qede_init_fp(struct qede_dev *edev)
 
 			/* Driver have no error path from here */
 			WARN_ON(xdp_rxq_info_reg(&fp->rxq->xdp_rxq, edev->ndev,
-						 fp->rxq->rxq_id) < 0);
+						 fp->rxq->rxq_id, 0) < 0);
 
 			if (xdp_rxq_info_reg_mem_model(&fp->rxq->xdp_rxq,
 						       MEM_TYPE_PAGE_ORDER0,
drivers/net/ethernet/sfc/rx_common.c +1 -1
index 19cf7cac1e6e..68fc7d317693 100644
@@ -262,7 +262,7 @@ void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
 
 	/* Initialise XDP queue information */
 	rc = xdp_rxq_info_reg(&rx_queue->xdp_rxq_info, efx->net_dev,
-			      rx_queue->core_index);
+			      rx_queue->core_index, 0);
 
 	if (rc) {
 		netif_err(efx, rx_err, efx->net_dev,
drivers/net/ethernet/socionext/netsec.c +1 -1
index 1503cc9ec6e2..27d3c9d9210e 100644
@@ -1304,7 +1304,7 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv)
 		goto err_out;
 	}
 
-	err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0);
+	err = xdp_rxq_info_reg(&dring->xdp_rxq, priv->ndev, 0, priv->napi.napi_id);
 	if (err)
 		goto err_out;
 
drivers/net/ethernet/ti/cpsw_priv.c +1 -1
index 31c5e36ff706..6dd73bd0f458 100644
@@ -1186,7 +1186,7 @@ static int cpsw_ndev_create_xdp_rxq(struct cpsw_priv *priv, int ch)
 	pool = cpsw->page_pool[ch];
 	rxq = &priv->xdp_rxq[ch];
 
-	ret = xdp_rxq_info_reg(rxq, priv->ndev, ch);
+	ret = xdp_rxq_info_reg(rxq, priv->ndev, ch, 0);
 	if (ret)
 		return ret;
 
drivers/net/hyperv/netvsc.c +1 -1
index 0c3de94b5178..fa8341f8359a 100644
@@ -1499,7 +1499,7 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device,
 		u64_stats_init(&nvchan->tx_stats.syncp);
 		u64_stats_init(&nvchan->rx_stats.syncp);
 
-		ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i);
+		ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i, 0);
 
 		if (ret) {
 			netdev_err(ndev, "xdp_rxq_info_reg fail: %d\n", ret);
drivers/net/tun.c +1 -1
index 3d45d56172cb..8867d39db6ac 100644
@@ -780,7 +780,7 @@ static int tun_attach(struct tun_struct *tun, struct file *file,
 	} else {
 		/* Setup XDP RX-queue info, for new tfile getting attached */
 		err = xdp_rxq_info_reg(&tfile->xdp_rxq,
-				       tun->dev, tfile->queue_index);
+				       tun->dev, tfile->queue_index, 0);
 		if (err < 0)
 			goto out;
 		err = xdp_rxq_info_reg_mem_model(&tfile->xdp_rxq,
drivers/net/veth.c +8 -4
index 8c737668008a..9bd37c7151f8 100644
@@ -884,7 +884,6 @@ static int veth_napi_add(struct net_device *dev)
 	for (i = 0; i < dev->real_num_rx_queues; i++) {
 		struct veth_rq *rq = &priv->rq[i];
 
-		netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
 		napi_enable(&rq->xdp_napi);
 	}
 
@@ -926,7 +925,8 @@ static int veth_enable_xdp(struct net_device *dev)
 		for (i = 0; i < dev->real_num_rx_queues; i++) {
 			struct veth_rq *rq = &priv->rq[i];
 
-			err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i);
+			netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
+			err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i, rq->xdp_napi.napi_id);
 			if (err < 0)
 				goto err_rxq_reg;
 
@@ -952,8 +952,12 @@ static int veth_enable_xdp(struct net_device *dev)
 err_reg_mem:
 	xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
 err_rxq_reg:
-	for (i--; i >= 0; i--)
-		xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
+	for (i--; i >= 0; i--) {
+		struct veth_rq *rq = &priv->rq[i];
+
+		xdp_rxq_info_unreg(&rq->xdp_rxq);
+		netif_napi_del(&rq->xdp_napi);
+	}
 
 	return err;
 }
drivers/net/virtio_net.c +1 -1
index 21b71148c532..052975ea0af4 100644
@@ -1485,7 +1485,7 @@ static int virtnet_open(struct net_device *dev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 
-		err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i);
+		err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i, vi->rq[i].napi.napi_id);
 		if (err < 0)
 			return err;
 
drivers/net/xen-netfront.c +1 -1
index 920cac4385bf..b01848ef4649 100644
@@ -2014,7 +2014,7 @@ static int xennet_create_page_pool(struct netfront_queue *queue)
 	}
 
 	err = xdp_rxq_info_reg(&queue->xdp_rxq, queue->info->netdev,
-			       queue->id);
+			       queue->id, 0);
 	if (err) {
 		netdev_err(queue->info->netdev, "xdp_rxq_info_reg failed\n");
 		goto err_free_pp;
include/net/busy_poll.h +15 -4
index 2f8f51807b83..45b3e04b99d3 100644
@@ -135,14 +135,25 @@ static inline void sk_mark_napi_id(struct sock *sk, const struct sk_buff *skb)
 	sk_rx_queue_set(sk, skb);
 }
 
-/* variant used for unconnected sockets */
-static inline void sk_mark_napi_id_once(struct sock *sk,
-					const struct sk_buff *skb)
+static inline void __sk_mark_napi_id_once_xdp(struct sock *sk, unsigned int napi_id)
 {
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	if (!READ_ONCE(sk->sk_napi_id))
-		WRITE_ONCE(sk->sk_napi_id, skb->napi_id);
+		WRITE_ONCE(sk->sk_napi_id, napi_id);
 #endif
 }
 
+/* variant used for unconnected sockets */
+static inline void sk_mark_napi_id_once(struct sock *sk,
+					const struct sk_buff *skb)
+{
+	__sk_mark_napi_id_once_xdp(sk, skb->napi_id);
+}
+
+static inline void sk_mark_napi_id_once_xdp(struct sock *sk,
+					    const struct xdp_buff *xdp)
+{
+	__sk_mark_napi_id_once_xdp(sk, xdp->rxq->napi_id);
+}
+
 #endif /* _LINUX_NET_BUSY_POLL_H */
include/net/xdp.h +2 -1
index 7d48b2ae217a..700ad5db7f5d 100644
@@ -59,6 +59,7 @@ struct xdp_rxq_info {
 	u32 queue_index;
 	u32 reg_state;
 	struct xdp_mem_info mem;
+	unsigned int napi_id;
 } ____cacheline_aligned; /* perf critical, avoid false-sharing */
 
 struct xdp_txq_info {
@@ -226,7 +227,7 @@ static inline void xdp_release_frame(struct xdp_frame *xdpf)
 }
 
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-		     struct net_device *dev, u32 queue_index);
+		     struct net_device *dev, u32 queue_index, unsigned int napi_id);
 void xdp_rxq_info_unreg(struct xdp_rxq_info *xdp_rxq);
 void xdp_rxq_info_unused(struct xdp_rxq_info *xdp_rxq);
 bool xdp_rxq_info_is_reg(struct xdp_rxq_info *xdp_rxq);
net/core/dev.c +1 -1
index 7a1e5936c67f..3b6b0e175fe7 100644
@@ -9810,7 +9810,7 @@ static int netif_alloc_rx_queues(struct net_device *dev)
 		rx[i].dev = dev;
 
 		/* XDP RX-queue setup */
-		err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i);
+		err = xdp_rxq_info_reg(&rx[i].xdp_rxq, dev, i, 0);
 		if (err < 0)
 			goto err_rxq_info;
 	}
net/core/xdp.c +2 -1
index 3d330ebda893..17ffd33c6b18 100644
@@ -158,7 +158,7 @@ static void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq)
 
 /* Returns 0 on success, negative on failure */
 int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
-		     struct net_device *dev, u32 queue_index)
+		     struct net_device *dev, u32 queue_index, unsigned int napi_id)
 {
 	if (xdp_rxq->reg_state == REG_STATE_UNUSED) {
 		WARN(1, "Driver promised not to register this");
@@ -179,6 +179,7 @@ int xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq,
 	xdp_rxq_info_init(xdp_rxq);
 	xdp_rxq->dev = dev;
 	xdp_rxq->queue_index = queue_index;
+	xdp_rxq->napi_id = napi_id;
 
 	xdp_rxq->reg_state = REG_STATE_REGISTERED;
 	return 0;
net/xdp/xsk.c +1 -0
index ecc4579e41ee..d4cb1c5c1abf 100644
@@ -233,6 +233,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp,
 	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index)
 		return -EINVAL;
 
+	sk_mark_napi_id_once_xdp(&xs->sk, xdp);
 	len = xdp->data_end - xdp->data;
 
 	return xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL ?
Signature · 1 line
2.27.0
#18 lore raw

Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

In reply to Björn Töpel
I love your patch! Yet something to improve:

Hi "Björn,

I love your patch! Yet something to improve:

[auto build test ERROR on 4e99d115d865d45e17e83478d757b58d8fa66d3c]

url:    https://github.com/0day-ci/linux/commits/Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934
base:   4e99d115d865d45e17e83478d757b58d8fa66d3c
config: um-kunit_defconfig (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0

reproduce (this is a W=1 build):
# https://github.com/0day-ci/linux/commit/f481c00164924dd5d782a92cc67897cc7f804502
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934 git checkout f481c00164924dd5d782a92cc67897cc7f804502
# save the attached .config to linux build tree
mkdir build_dir
make W=1 O=build_dir ARCH=um SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate

All errors (new ones prefixed by >>):

cc1: warning: arch/um/include/uapi: No such file or directory [-Wmissing-include-dirs] In file included from fs/select.c:32:
include/net/busy_poll.h: In function 'sk_mark_napi_id_once':

>> include/net/busy_poll.h:150:36: error: 'const struct sk_buff' has no member named 'napi_id'
     150 |  __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
         |                                    ^~

vim +150 include/net/busy_poll.h

145
146 /* variant used for unconnected sockets */
147 static inline void sk_mark_napi_id_once(struct sock *sk,
148 const struct sk_buff *skb)
149 {

 > 150		__sk_mark_napi_id_once_xdp(sk, skb->napi_id);

151 }
152

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

#19 lore raw

Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

Quoted 35 lines · kernel test robot

Hi "Björn,

I love your patch! Yet something to improve:

[auto build test ERROR on 4e99d115d865d45e17e83478d757b58d8fa66d3c]

url:    https://github.com/0day-ci/linux/commits/Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934
base:   4e99d115d865d45e17e83478d757b58d8fa66d3c
config: um-kunit_defconfig (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0

reproduce (this is a W=1 build):
# https://github.com/0day-ci/linux/commit/f481c00164924dd5d782a92cc67897cc7f804502
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934 git checkout f481c00164924dd5d782a92cc67897cc7f804502
# save the attached .config to linux build tree
mkdir build_dir
make W=1 O=build_dir ARCH=um SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate

All errors (new ones prefixed by >>):

cc1: warning: arch/um/include/uapi: No such file or directory [-Wmissing-include-dirs] In file included from fs/select.c:32:
include/net/busy_poll.h: In function 'sk_mark_napi_id_once':

Quoted 1 line
Quoted 1 line

include/net/busy_poll.h:150:36: error: 'const struct sk_buff' has no member named 'napi_id'

     150 |  __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
         |                                    ^~

vim +150 include/net/busy_poll.h

Quoted 8 lines

145
146 /* variant used for unconnected sockets */
147 static inline void sk_mark_napi_id_once(struct sock *sk,
148 const struct sk_buff *skb)
149 {

Quoted 1 line

150 __sk_mark_napi_id_once_xdp(sk, skb->napi_id);

151 }
152

It seems that the robot tested an old commit and that this was already fixed by Daniel 10 months ago. Slow mail delivery, a robot glitch, or am I missing something?

commit ba0581749fec389e55c9d761f2716f8fcbefced5

Author: Daniel Borkmann <[email protected]>
Date:   Tue Dec 1 15:22:59 2020 +0100

net, xdp, xsk: fix __sk_mark_napi_id_once napi_id error

Stephen reported the following build error for !CONFIG_NET_RX_BUSY_POLL built kernels:

In file included from fs/select.c:32:
include/net/busy_poll.h: In function 'sk_mark_napi_id_once':
include/net/busy_poll.h:150:36: error: 'const struct sk_buff'
has no member named 'napi_id'
150 | __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
| ^~

Fix it by wrapping a CONFIG_NET_RX_BUSY_POLL around the helpers.

Quoted 3 lines

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

#20 lore raw

Re: [kbuild-all] Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

Quoted 45 lines · Magnus Karlsson
Quoted 35 lines · kernel test robot

Hi "Björn,

I love your patch! Yet something to improve:

[auto build test ERROR on 4e99d115d865d45e17e83478d757b58d8fa66d3c]

url:    https://github.com/0day-ci/linux/commits/Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934
base:   4e99d115d865d45e17e83478d757b58d8fa66d3c
config: um-kunit_defconfig (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0

reproduce (this is a W=1 build):
# https://github.com/0day-ci/linux/commit/f481c00164924dd5d782a92cc67897cc7f804502
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934 git checkout f481c00164924dd5d782a92cc67897cc7f804502
# save the attached .config to linux build tree
mkdir build_dir
make W=1 O=build_dir ARCH=um SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate

All errors (new ones prefixed by >>):

cc1: warning: arch/um/include/uapi: No such file or directory [-Wmissing-include-dirs] In file included from fs/select.c:32:
include/net/busy_poll.h: In function 'sk_mark_napi_id_once':

Quoted 1 line
Quoted 1 line

include/net/busy_poll.h:150:36: error: 'const struct sk_buff' has no member named 'napi_id'

     150 |  __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
         |                                    ^~

vim +150 include/net/busy_poll.h

Quoted 8 lines

145
146 /* variant used for unconnected sockets */
147 static inline void sk_mark_napi_id_once(struct sock *sk,
148 const struct sk_buff *skb)
149 {

Quoted 1 line

150 __sk_mark_napi_id_once_xdp(sk, skb->napi_id);

151 }
152

Quoted 3 lines

It seems that the robot tested an old commit and that this was already fixed by Daniel 10 months ago. Slow mail delivery, a robot glitch, or am I missing something?

sorry for the noise, we got storage crash and it seems to have old data back in test queue after recovery. We will do cleanup to avoid this kind of old commit being tested.

Quoted 33 lines

commit ba0581749fec389e55c9d761f2716f8fcbefced5

Author: Daniel Borkmann <[email protected]>
Date:   Tue Dec 1 15:22:59 2020 +0100

net, xdp, xsk: fix __sk_mark_napi_id_once napi_id error

Stephen reported the following build error for !CONFIG_NET_RX_BUSY_POLL built kernels:

In file included from fs/select.c:32:
include/net/busy_poll.h: In function 'sk_mark_napi_id_once':
include/net/busy_poll.h:150:36: error: 'const struct sk_buff'
has no member named 'napi_id'
150 | __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
| ^~

Fix it by wrapping a CONFIG_NET_RX_BUSY_POLL around the helpers.

Quoted 3 lines

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

_______________________________________________
kbuild-all mailing list -- kbuild-all(a)lists.01.org To unsubscribe send an email to kbuild-all-leave(a)lists.01.org

#21 lore raw

Re: [PATCH bpf-next v3 06/10] xsk: propagate napi_id to XDP socket Rx path

In reply to Björn Töpel
I love your patch! Yet something to improve:

Hi "Björn,

I love your patch! Yet something to improve:

[auto build test ERROR on 4e99d115d865d45e17e83478d757b58d8fa66d3c]

url:    https://github.com/0day-ci/linux/commits/Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934
base:   4e99d115d865d45e17e83478d757b58d8fa66d3c
config: riscv-randconfig-r042-20210929 (attached as .config)
compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project dc6e8dfdfe7efecfda318d43a06fae18b40eb498)

reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# install riscv cross compiling tool for clang build
# apt-get install binutils-riscv64-linux-gnu
# https://github.com/0day-ci/linux/commit/f481c00164924dd5d782a92cc67897cc7f804502
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Bj-rn-T-pel/Introduce-preferred-busy-polling/20210929-234934 git checkout f481c00164924dd5d782a92cc67897cc7f804502
# save the attached .config to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=riscv SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate

All errors (new ones prefixed by >>):

In file included from fs/select.c:32:
In file included from include/net/busy_poll.h:15:
In file included from include/linux/netdevice.h:37:
In file included from include/linux/ethtool.h:18:
In file included from include/uapi/linux/ethtool.h:19:
In file included from include/linux/if_ether.h:19:
include/linux/skbuff.h:4622:26: error: implicit declaration of function 'skb_ext_add' [-Werror,-Wimplicit-function-declaration]

                   u64 *kcov_handle_ptr = skb_ext_add(skb, SKB_EXT_KCOV_HANDLE);
                                          ^
   include/linux/skbuff.h:4622:43: error: use of undeclared identifier 'SKB_EXT_KCOV_HANDLE'
                   u64 *kcov_handle_ptr = skb_ext_add(skb, SKB_EXT_KCOV_HANDLE);
                                                           ^
   include/linux/skbuff.h:4631:21: error: implicit declaration of function 'skb_ext_find' [-Werror,-Wimplicit-function-declaration]
           u64 *kcov_handle = skb_ext_find(skb, SKB_EXT_KCOV_HANDLE);
                              ^
   include/linux/skbuff.h:4631:39: error: use of undeclared identifier 'SKB_EXT_KCOV_HANDLE'
           u64 *kcov_handle = skb_ext_find(skb, SKB_EXT_KCOV_HANDLE);
                                                ^
   In file included from fs/select.c:32:
>> include/net/busy_poll.h:150:38: error: no member named 'napi_id' in 'struct sk_buff'
           __sk_mark_napi_id_once_xdp(sk, skb->napi_id);
                                          ~~~  ^

5 errors generated.

vim +150 include/net/busy_poll.h

145
146 /* variant used for unconnected sockets */
147 static inline void sk_mark_napi_id_once(struct sock *sk,
148 const struct sk_buff *skb)
149 {

 > 150		__sk_mark_napi_id_once_xdp(sk, skb->napi_id);

151 }
152

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

#22 lore raw

PATCH v3 7/10 samples/bpf: use recvfrom() in xdpsock/rxdrop

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Start using recvfrom() the rxdrop scenario.

---
 samples/bpf/xdpsock_user.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
samples/bpf/xdpsock_user.c +4 -4
index 2567f0db5aca..f90111b95b2e 100644
@@ -1170,7 +1170,7 @@ static inline void complete_tx_only(struct xsk_socket_info *xsk,
 	}
 }
 
-static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
+static void rx_drop(struct xsk_socket_info *xsk)
 {
 	unsigned int rcvd, i;
 	u32 idx_rx = 0, idx_fq = 0;
@@ -1180,7 +1180,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.rx_empty_polls++;
-			ret = poll(fds, num_socks, opt_timeout);
+			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
 		return;
 	}
@@ -1191,7 +1191,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
 			exit_with_error(-ret);
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.fill_fail_polls++;
-			ret = poll(fds, num_socks, opt_timeout);
+			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
 		ret = xsk_ring_prod__reserve(&xsk->umem->fq, rcvd, &idx_fq);
 	}
@@ -1233,7 +1233,7 @@ static void rx_drop_all(void)
 		}
 
 		for (i = 0; i < num_socks; i++)
-			rx_drop(xsks[i], fds);
+			rx_drop(xsks[i]);
 
 		if (benchmark_done)
 			break;
Signature · 1 line
2.27.0
Acked-by on v3 7/10
#23 lore raw

Re: [PATCH bpf-next v3 07/10] samples/bpf: use recvfrom() in xdpsock/rxdrop

Quoted 9 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Start using recvfrom() the rxdrop scenario.

---
 samples/bpf/xdpsock_user.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
Quoted 43 lines
samples/bpf/xdpsock_user.c +4 -5
index 2567f0db5aca..f90111b95b2e 100644
@@ -1170,7 +1170,7 @@ static inline void complete_tx_only(struct xsk_socket_info *xsk,
        }
 }
 
-static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
+static void rx_drop(struct xsk_socket_info *xsk)
 {
        unsigned int rcvd, i;
        u32 idx_rx = 0, idx_fq = 0;
@@ -1180,7 +1180,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
        if (!rcvd) {
                if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.rx_empty_polls++;
-                       ret = poll(fds, num_socks, opt_timeout);
+                       recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
                return;
        }
@@ -1191,7 +1191,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
                        exit_with_error(-ret);
                if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.fill_fail_polls++;
-                       ret = poll(fds, num_socks, opt_timeout);
+                       recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
                ret = xsk_ring_prod__reserve(&xsk->umem->fq, rcvd, &idx_fq);
        }
@@ -1233,7 +1233,7 @@ static void rx_drop_all(void)
                }
 
                for (i = 0; i < num_socks; i++)
-                       rx_drop(xsks[i], fds);
+                       rx_drop(xsks[i]);
 
                if (benchmark_done)
                        break;
--

2.27.0

#24 lore raw

PATCH v3 8/10 samples/bpf: use recvfrom() in xdpsock/l2fwd

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Start using recvfrom() the l2fwd scenario, instead of poll() which is more expensive and need additional knobs for busy-polling.

---
 samples/bpf/xdpsock_user.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)
samples/bpf/xdpsock_user.c +11 -14
index f90111b95b2e..24aa7511c4c8 100644
@@ -1098,8 +1098,7 @@ static void kick_tx(struct xsk_socket_info *xsk)
 	exit_with_error(errno);
 }
 
-static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
-				     struct pollfd *fds)
+static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk)
 {
 	struct xsk_umem_info *umem = xsk->umem;
 	u32 idx_cq = 0, idx_fq = 0;
@@ -1134,7 +1133,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
 				exit_with_error(-ret);
 			if (xsk_ring_prod__needs_wakeup(&umem->fq)) {
 				xsk->app_stats.fill_fail_polls++;
-				ret = poll(fds, num_socks, opt_timeout);
+				recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 			}
 			ret = xsk_ring_prod__reserve(&umem->fq, rcvd, &idx_fq);
 		}
@@ -1331,19 +1330,19 @@ static void tx_only_all(void)
 		complete_tx_only_all();
 }
 
-static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
+static void l2fwd(struct xsk_socket_info *xsk)
 {
 	unsigned int rcvd, i;
 	u32 idx_rx = 0, idx_tx = 0;
 	int ret;
 
-	complete_tx_l2fwd(xsk, fds);
+	complete_tx_l2fwd(xsk);
 
 	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.rx_empty_polls++;
-			ret = poll(fds, num_socks, opt_timeout);
+			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
 		return;
 	}
@@ -1353,7 +1352,7 @@ static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
 	while (ret != rcvd) {
 		if (ret < 0)
 			exit_with_error(-ret);
-		complete_tx_l2fwd(xsk, fds);
+		complete_tx_l2fwd(xsk);
 		if (xsk_ring_prod__needs_wakeup(&xsk->tx)) {
 			xsk->app_stats.tx_wakeup_sendtos++;
 			kick_tx(xsk);
@@ -1388,22 +1387,20 @@ static void l2fwd_all(void)
 	struct pollfd fds[MAX_SOCKS] = {};
 	int i, ret;
 
-	for (i = 0; i < num_socks; i++) {
-		fds[i].fd = xsk_socket__fd(xsks[i]->xsk);
-		fds[i].events = POLLOUT | POLLIN;
-	}
-
 	for (;;) {
 		if (opt_poll) {
-			for (i = 0; i < num_socks; i++)
+			for (i = 0; i < num_socks; i++) {
+				fds[i].fd = xsk_socket__fd(xsks[i]->xsk);
+				fds[i].events = POLLOUT | POLLIN;
 				xsks[i]->app_stats.opt_polls++;
+			}
 			ret = poll(fds, num_socks, opt_timeout);
 			if (ret <= 0)
 				continue;
 		}
 
 		for (i = 0; i < num_socks; i++)
-			l2fwd(xsks[i], fds);
+			l2fwd(xsks[i]);
 
 		if (benchmark_done)
 			break;
Signature · 1 line
2.27.0
Acked-by on v3 8/10
#25 lore raw

Re: [PATCH bpf-next v3 08/10] samples/bpf: use recvfrom() in xdpsock/l2fwd

Quoted 10 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Start using recvfrom() the l2fwd scenario, instead of poll() which is more expensive and need additional knobs for busy-polling.

---
 samples/bpf/xdpsock_user.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)
Quoted 86 lines
samples/bpf/xdpsock_user.c +11 -15
index f90111b95b2e..24aa7511c4c8 100644
@@ -1098,8 +1098,7 @@ static void kick_tx(struct xsk_socket_info *xsk)
        exit_with_error(errno);
 }
 
-static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
-                                    struct pollfd *fds)
+static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk)
 {
        struct xsk_umem_info *umem = xsk->umem;
        u32 idx_cq = 0, idx_fq = 0;
@@ -1134,7 +1133,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
                                exit_with_error(-ret);
                        if (xsk_ring_prod__needs_wakeup(&umem->fq)) {
                                xsk->app_stats.fill_fail_polls++;
-                               ret = poll(fds, num_socks, opt_timeout);
+                               recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                        }
                        ret = xsk_ring_prod__reserve(&umem->fq, rcvd, &idx_fq);
                }
@@ -1331,19 +1330,19 @@ static void tx_only_all(void)
                complete_tx_only_all();
 }
 
-static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
+static void l2fwd(struct xsk_socket_info *xsk)
 {
        unsigned int rcvd, i;
        u32 idx_rx = 0, idx_tx = 0;
        int ret;
 
-       complete_tx_l2fwd(xsk, fds);
+       complete_tx_l2fwd(xsk);
 
        rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
        if (!rcvd) {
                if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.rx_empty_polls++;
-                       ret = poll(fds, num_socks, opt_timeout);
+                       recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
                return;
        }
@@ -1353,7 +1352,7 @@ static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
        while (ret != rcvd) {
                if (ret < 0)
                        exit_with_error(-ret);
-               complete_tx_l2fwd(xsk, fds);
+               complete_tx_l2fwd(xsk);
                if (xsk_ring_prod__needs_wakeup(&xsk->tx)) {
                        xsk->app_stats.tx_wakeup_sendtos++;
                        kick_tx(xsk);
@@ -1388,22 +1387,20 @@ static void l2fwd_all(void)
        struct pollfd fds[MAX_SOCKS] = {};
        int i, ret;
 
-       for (i = 0; i < num_socks; i++) {
-               fds[i].fd = xsk_socket__fd(xsks[i]->xsk);
-               fds[i].events = POLLOUT | POLLIN;
-       }
-
        for (;;) {
                if (opt_poll) {
-                       for (i = 0; i < num_socks; i++)
+                       for (i = 0; i < num_socks; i++) {
+                               fds[i].fd = xsk_socket__fd(xsks[i]->xsk);
+                               fds[i].events = POLLOUT | POLLIN;
                                xsks[i]->app_stats.opt_polls++;
+                       }
                        ret = poll(fds, num_socks, opt_timeout);
                        if (ret <= 0)
                                continue;
                }
 
                for (i = 0; i < num_socks; i++)
-                       l2fwd(xsks[i], fds);
+                       l2fwd(xsks[i]);
 
                if (benchmark_done)
                        break;
--

2.27.0

#26 lore raw

PATCH v3 9/10 samples/bpf: add busy-poll support to xdpsock

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Add a new option to xdpsock, 'B', for busy-polling. This option will also set the batching size, 'b' option, to the busy-poll budget.

---
 samples/bpf/xdpsock_user.c | 40 +++++++++++++++++++++++++++++++-------
 1 file changed, 33 insertions(+), 7 deletions(-)
samples/bpf/xdpsock_user.c +33 -7
index 24aa7511c4c8..cb1eaee8a32b 100644
@@ -95,6 +95,7 @@ static int opt_timeout = 1000;
 static bool opt_need_wakeup = true;
 static u32 opt_num_xsks = 1;
 static u32 prog_id;
+static bool opt_busy_poll;
 
 struct xsk_ring_stats {
 	unsigned long rx_npkts;
@@ -911,6 +912,7 @@ static struct option long_options[] = {
 	{"quiet", no_argument, 0, 'Q'},
 	{"app-stats", no_argument, 0, 'a'},
 	{"irq-string", no_argument, 0, 'I'},
+	{"busy-poll", no_argument, 0, 'B'},
 	{0, 0, 0, 0}
 };
 
@@ -949,6 +951,7 @@ static void usage(const char *prog)
 		"  -Q, --quiet          Do not display any stats.\n"
 		"  -a, --app-stats	Display application (syscall) statistics.\n"
 		"  -I, --irq-string	Display driver interrupt statistics for interface associated with irq-string.\n"
+		"  -B, --busy-poll      Busy poll.\n"
 		"\n";
 	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
 		opt_batch_size, MIN_PKT_SIZE, MIN_PKT_SIZE,
@@ -964,7 +967,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:xQaI:",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:xQaI:B",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -1062,7 +1065,9 @@ static void parse_command_line(int argc, char **argv)
 				fprintf(stderr, "ERROR: Failed to get irqs for %s\n", opt_irq_str);
 				usage(basename(argv[0]));
 			}
-
+			break;
+		case 'B':
+			opt_busy_poll = 1;
 			break;
 		default:
 			usage(basename(argv[0]));
@@ -1131,7 +1136,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk)
 		while (ret != rcvd) {
 			if (ret < 0)
 				exit_with_error(-ret);
-			if (xsk_ring_prod__needs_wakeup(&umem->fq)) {
+			if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&umem->fq)) {
 				xsk->app_stats.fill_fail_polls++;
 				recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 			}
@@ -1177,7 +1182,7 @@ static void rx_drop(struct xsk_socket_info *xsk)
 
 	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
-		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+		if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.rx_empty_polls++;
 			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
@@ -1188,7 +1193,7 @@ static void rx_drop(struct xsk_socket_info *xsk)
 	while (ret != rcvd) {
 		if (ret < 0)
 			exit_with_error(-ret);
-		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+		if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.fill_fail_polls++;
 			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
@@ -1340,7 +1345,7 @@ static void l2fwd(struct xsk_socket_info *xsk)
 
 	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
-		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+		if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
 			xsk->app_stats.rx_empty_polls++;
 			recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
 		}
@@ -1353,7 +1358,7 @@ static void l2fwd(struct xsk_socket_info *xsk)
 		if (ret < 0)
 			exit_with_error(-ret);
 		complete_tx_l2fwd(xsk);
-		if (xsk_ring_prod__needs_wakeup(&xsk->tx)) {
+		if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->tx)) {
 			xsk->app_stats.tx_wakeup_sendtos++;
 			kick_tx(xsk);
 		}
@@ -1458,6 +1463,24 @@ static void enter_xsks_into_map(struct bpf_object *obj)
 	}
 }
 
+static void apply_setsockopt(struct xsk_socket_info *xsk)
+{
+	int sock_opt;
+
+	if (!opt_busy_poll)
+		return;
+
+	sock_opt = 1;
+	if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_PREFER_BUSY_POLL,
+		       (void *)&sock_opt, sizeof(sock_opt)) < 0)
+		exit_with_error(errno);
+
+	sock_opt = 20;
+	if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL,
+		       (void *)&sock_opt, sizeof(sock_opt)) < 0)
+		exit_with_error(errno);
+}
+
 int main(int argc, char **argv)
 {
 	struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
@@ -1499,6 +1522,9 @@ int main(int argc, char **argv)
 	for (i = 0; i < opt_num_xsks; i++)
 		xsks[num_socks++] = xsk_configure_socket(umem, rx, tx);
 
+	for (i = 0; i < opt_num_xsks; i++)
+		apply_setsockopt(xsks[i]);
+
 	if (opt_bench == BENCH_TXONLY) {
 		gen_eth_hdr_data();
 
Signature · 1 line
2.27.0
Acked-by on v3 9/10
#27 lore raw

Re: [PATCH bpf-next v3 09/10] samples/bpf: add busy-poll support to xdpsock

Quoted 10 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Add a new option to xdpsock, 'B', for busy-polling. This option will also set the batching size, 'b' option, to the busy-poll budget.

---
 samples/bpf/xdpsock_user.c | 40 +++++++++++++++++++++++++++++++-------
 1 file changed, 33 insertions(+), 7 deletions(-)
Quoted 131 lines
samples/bpf/xdpsock_user.c +33 -8
index 24aa7511c4c8..cb1eaee8a32b 100644
@@ -95,6 +95,7 @@ static int opt_timeout = 1000;
 static bool opt_need_wakeup = true;
 static u32 opt_num_xsks = 1;
 static u32 prog_id;
+static bool opt_busy_poll;
 
 struct xsk_ring_stats {
        unsigned long rx_npkts;
@@ -911,6 +912,7 @@ static struct option long_options[] = {
        {"quiet", no_argument, 0, 'Q'},
        {"app-stats", no_argument, 0, 'a'},
        {"irq-string", no_argument, 0, 'I'},
+       {"busy-poll", no_argument, 0, 'B'},
        {0, 0, 0, 0}
 };
 
@@ -949,6 +951,7 @@ static void usage(const char *prog)
                "  -Q, --quiet          Do not display any stats.\n"
                "  -a, --app-stats      Display application (syscall) statistics.\n"
                "  -I, --irq-string     Display driver interrupt statistics for interface associated with irq-string.\n"
+               "  -B, --busy-poll      Busy poll.\n"
                "\n";
        fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
                opt_batch_size, MIN_PKT_SIZE, MIN_PKT_SIZE,
@@ -964,7 +967,7 @@ static void parse_command_line(int argc, char **argv)
        opterr = 0;
 
        for (;;) {
-               c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:xQaI:",
+               c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:xQaI:B",
                                long_options, &option_index);
                if (c == -1)
                        break;
@@ -1062,7 +1065,9 @@ static void parse_command_line(int argc, char **argv)
                                fprintf(stderr, "ERROR: Failed to get irqs for %s\n", opt_irq_str);
                                usage(basename(argv[0]));
                        }
-
+                       break;
+               case 'B':
+                       opt_busy_poll = 1;
                        break;
                default:
                        usage(basename(argv[0]));
@@ -1131,7 +1136,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk)
                while (ret != rcvd) {
                        if (ret < 0)
                                exit_with_error(-ret);
-                       if (xsk_ring_prod__needs_wakeup(&umem->fq)) {
+                       if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&umem->fq)) {
                                xsk->app_stats.fill_fail_polls++;
                                recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                        }
@@ -1177,7 +1182,7 @@ static void rx_drop(struct xsk_socket_info *xsk)
 
        rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
        if (!rcvd) {
-               if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+               if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.rx_empty_polls++;
                        recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
@@ -1188,7 +1193,7 @@ static void rx_drop(struct xsk_socket_info *xsk)
        while (ret != rcvd) {
                if (ret < 0)
                        exit_with_error(-ret);
-               if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+               if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.fill_fail_polls++;
                        recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
@@ -1340,7 +1345,7 @@ static void l2fwd(struct xsk_socket_info *xsk)
 
        rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
        if (!rcvd) {
-               if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
+               if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) {
                        xsk->app_stats.rx_empty_polls++;
                        recvfrom(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, NULL);
                }
@@ -1353,7 +1358,7 @@ static void l2fwd(struct xsk_socket_info *xsk)
                if (ret < 0)
                        exit_with_error(-ret);
                complete_tx_l2fwd(xsk);
-               if (xsk_ring_prod__needs_wakeup(&xsk->tx)) {
+               if (opt_busy_poll || xsk_ring_prod__needs_wakeup(&xsk->tx)) {
                        xsk->app_stats.tx_wakeup_sendtos++;
                        kick_tx(xsk);
                }
@@ -1458,6 +1463,24 @@ static void enter_xsks_into_map(struct bpf_object *obj)
        }
 }
 
+static void apply_setsockopt(struct xsk_socket_info *xsk)
+{
+       int sock_opt;
+
+       if (!opt_busy_poll)
+               return;
+
+       sock_opt = 1;
+       if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_PREFER_BUSY_POLL,
+                      (void *)&sock_opt, sizeof(sock_opt)) < 0)
+               exit_with_error(errno);
+
+       sock_opt = 20;
+       if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL,
+                      (void *)&sock_opt, sizeof(sock_opt)) < 0)
+               exit_with_error(errno);
+}
+
 int main(int argc, char **argv)
 {
        struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
@@ -1499,6 +1522,9 @@ int main(int argc, char **argv)
        for (i = 0; i < opt_num_xsks; i++)
                xsks[num_socks++] = xsk_configure_socket(umem, rx, tx);
 
+       for (i = 0; i < opt_num_xsks; i++)
+               apply_setsockopt(xsks[i]);
+
        if (opt_bench == BENCH_TXONLY) {
                gen_eth_hdr_data();
 
--

2.27.0

#28 lore raw

PATCH v3 10/10 samples/bpf: add option to set the busy-poll budget

In reply to Björn Töpel

From: Björn Töpel <[email protected]>

Support for the SO_BUSY_POLL_BUDGET setsockopt, via the batching option ('b').

---
 samples/bpf/xdpsock_user.c | 5 +++++
 1 file changed, 5 insertions(+)
samples/bpf/xdpsock_user.c +5 -0
index cb1eaee8a32b..deba623e9003 100644
@@ -1479,6 +1479,11 @@ static void apply_setsockopt(struct xsk_socket_info *xsk)
 	if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL,
 		       (void *)&sock_opt, sizeof(sock_opt)) < 0)
 		exit_with_error(errno);
+
+	sock_opt = opt_batch_size;
+	if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL_BUDGET,
+		       (void *)&sock_opt, sizeof(sock_opt)) < 0)
+		exit_with_error(errno);
 }
 
 int main(int argc, char **argv)
Signature · 1 line
2.27.0
Acked-by on v3 10/10
#29 lore raw

Re: [PATCH bpf-next v3 10/10] samples/bpf: add option to set the busy-poll budget

Quoted 10 lines · Björn Töpel

From: Björn Töpel <[email protected]>

Support for the SO_BUSY_POLL_BUDGET setsockopt, via the batching option ('b').

---
 samples/bpf/xdpsock_user.c | 5 +++++
 1 file changed, 5 insertions(+)
Quoted 19 lines
samples/bpf/xdpsock_user.c +5 -1
index cb1eaee8a32b..deba623e9003 100644
@@ -1479,6 +1479,11 @@ static void apply_setsockopt(struct xsk_socket_info *xsk)
        if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL,
                       (void *)&sock_opt, sizeof(sock_opt)) < 0)
                exit_with_error(errno);
+
+       sock_opt = opt_batch_size;
+       if (setsockopt(xsk_socket__fd(xsk->xsk), SOL_SOCKET, SO_BUSY_POLL_BUDGET,
+                      (void *)&sock_opt, sizeof(sock_opt)) < 0)
+               exit_with_error(errno);
 }
 
 int main(int argc, char **argv)
--

2.27.0

#30 lore raw
In reply to Björn Töpel
Quoted 32 lines · Björn Töpel

This series introduces three new features:

1. A new "heavy traffic" busy-polling variant that works in concert with the existing napi_defer_hard_irqs and gro_flush_timeout knobs.

2. A new socket option that let a user change the busy-polling NAPI budget.

3. Allow busy-polling to be performed on XDP sockets.

The existing busy-polling mode, enabled by the SO_BUSY_POLL socket option or system-wide using the /proc/sys/net/core/busy_read knob, is an opportunistic. That means that if the NAPI context is not scheduled, it will poll it. If, after busy-polling, the budget is exceeded the busy-polling logic will schedule the NAPI onto the regular softirq handling.

One implication of the behavior above is that a busy/heavy loaded NAPI context will never enter/allow for busy-polling. Some applications prefer that most NAPI processing would be done by busy-polling.

This series adds a new socket option, SO_PREFER_BUSY_POLL, that works in concert with the napi_defer_hard_irqs and gro_flush_timeout knobs. The napi_defer_hard_irqs and gro_flush_timeout knobs were introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature"), and allows for a user to defer interrupts to be enabled and instead schedule the NAPI context from a watchdog timer. When a user enables the SO_PREFER_BUSY_POLL, again with the other knobs enabled, and the NAPI context is being processed by a softirq, the softirq NAPI processing will exit early to allow the busy-polling to be performed.

Quoted 7 lines

If the application stops performing busy-polling via a system call, the watchdog timer defined by gro_flush_timeout will timeout, and regular softirq handling will resume.

In summary; Heavy traffic applications that prefer busy-polling over softirq processing should use this option.

Eric/Jakub, any more thoughts/input? Tomatoes? :-P

Thank you,
Björn

maintainer
#31 lore raw
Quoted 1 line · On Mon, 23 Nov 2020 14:31:14 +0100 Björn Töpel wrote:

Eric/Jakub, any more thoughts/input? Tomatoes? :-P

Looking now, sorry for the delay. Somehow patches without net in their tag feel like they can wait..

maintainer
#32 lore raw
In reply to Björn Töpel
Quoted 11 lines · On Thu, 19 Nov 2020 09:30:14 +0100 Björn Töpel wrote:

Performance netperf UDP_RR:

Note that netperf UDP_RR is not a heavy traffic tests, and preferred busy-polling is not typically something we want to use here.

  $ echo 20 | sudo tee /proc/sys/net/core/busy_read
  $ netperf -H 192.168.1.1 -l 30 -t UDP_RR -v 2 -- \
      -o min_latency,mean_latency,max_latency,stddev_latency,transaction_rate

busy-polling blocking sockets: 12,13.33,224,0.63,74731.177

Quoted 8 lines

I hacked netperf to use non-blocking sockets and re-ran:

busy-polling non-blocking sockets: 12,13.46,218,0.72,73991.172
prefer busy-polling non-blocking sockets: 12,13.62,221,0.59,73138.448

Using the preferred busy-polling mode does not impact performance.

The above tests was done for the 'ice' driver.

Any interest in this work form ADQ folks? I recall they were using memcache with busy polling for their tests, it'd cool to see how much this helps memcache on P99+ latency!