[PATCH bpf-next v2 0/2] xdp: Introduce bulking for non-map XDP_REDIRECT

[PATCH bpf-next v2 0/2] xdp: Introduce bulking for non-map XDP_REDIRECT

[Toke Høiland-Jørgensen]

Since commit 96360004b862 ("xdp: Make devmap flush_list common for all map
instances"), devmap flushing is a global operation instead of tied to a
particular map. This means that with a bit of refactoring, we can finally fix
the performance delta between the bpf_redirect_map() and bpf_redirect() helper
functions, by introducing bulking for the latter as well.

This series makes this change by moving the data structure used for the bulking
into struct net_device itself, so we can access it even when there is not
devmap. Once this is done, moving the bpf_redirect() helper to use the bulking
mechanism becomes quite trivial, and brings bpf_redirect() up to the same as
bpf_redirect_map():

                       Before:   After:
1 CPU:
bpf_redirect_map:      8.4 Mpps  8.4 Mpps  (no change)
bpf_redirect:          5.0 Mpps  8.4 Mpps  (+68%)
2 CPUs:
bpf_redirect_map:     15.9 Mpps  16.1 Mpps  (+1% or ~no change)
bpf_redirect:          9.5 Mpps  15.9 Mpps  (+67%)

After this patch series, the only semantics different between the two variants
of the bpf() helper (apart from the absence of a map argument, obviously) is
that the _map() variant will return an error if passed an invalid map index,
whereas the bpf_redirect() helper will succeed, but drop packets on
xdp_do_redirect(). This is because the helper has no reference to the calling
netdev, so unfortunately we can't do the ifindex lookup directly in the helper.

Changelog:

v2:
  - Consolidate code paths and tracepoints for map and non-map redirect variants
    (Björn)
  - Add performance data for 2-CPU test (Jesper)
  - Move fields to avoid shifting cache lines in struct net_device (Eric)

---

Toke Høiland-Jørgensen (2):
      xdp: Move devmap bulk queue into struct net_device
      xdp: Use bulking for non-map XDP_REDIRECT and consolidate code paths


 include/linux/bpf.h        |   13 +++++-
 include/linux/netdevice.h  |   11 +++--
 include/trace/events/xdp.h |  104 +++++++++++++++++++-------------------------
 kernel/bpf/devmap.c        |   94 +++++++++++++++++++++-------------------
 net/core/dev.c             |    2 +
 net/core/filter.c          |   86 +++++++-----------------------------
 6 files changed, 132 insertions(+), 178 deletions(-)

[PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Toke Høiland-Jørgensen]

From: Toke Høiland-Jørgensen <toke@redhat.com>

Commit 96360004b862 ("xdp: Make devmap flush_list common for all map
instances"), changed devmap flushing to be a global operation instead of a
per-map operation. However, the queue structure used for bulking was still
allocated as part of the containing map.

This patch moves the devmap bulk queue into struct net_device. The
motivation for this is reusing it for the non-map variant of XDP_REDIRECT,
which will be changed in a subsequent commit.  To avoid other fields of
struct net_device moving to different cache lines, we also move a couple of
other members around.

We defer the actual allocation of the bulk queue structure until the
NETDEV_REGISTER notification devmap.c. This makes it possible to check for
ndo_xdp_xmit support before allocating the structure, which is not possible
at the time struct net_device is allocated. However, we keep the freeing in
free_netdev() to avoid adding another RCU callback on NETDEV_UNREGISTER.

Because of this change, we lose the reference back to the map that
originated the redirect, so change the tracepoint to always return 0 as the
map ID and index. Otherwise no functional change is intended with this
patch.

Acked-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
 include/linux/netdevice.h  |   11 +++++---
 include/trace/events/xdp.h |    2 +
 kernel/bpf/devmap.c        |   63 +++++++++++++++++++-------------------------
 net/core/dev.c             |    2 +
 4 files changed, 37 insertions(+), 41 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2741aa35bec6..1f24405c1ec5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -876,6 +876,7 @@ enum bpf_netdev_command {
 struct bpf_prog_offload_ops;
 struct netlink_ext_ack;
 struct xdp_umem;
+struct xdp_dev_bulk_queue;
 
 struct netdev_bpf {
 	enum bpf_netdev_command command;
@@ -1986,12 +1987,10 @@ struct net_device {
 	unsigned int		num_tx_queues;
 	unsigned int		real_num_tx_queues;
 	struct Qdisc		*qdisc;
-#ifdef CONFIG_NET_SCHED
-	DECLARE_HASHTABLE	(qdisc_hash, 4);
-#endif
 	unsigned int		tx_queue_len;
 	spinlock_t		tx_global_lock;
-	int			watchdog_timeo;
+
+	struct xdp_dev_bulk_queue __percpu *xdp_bulkq;
 
 #ifdef CONFIG_XPS
 	struct xps_dev_maps __rcu *xps_cpus_map;
@@ -2001,8 +2000,12 @@ struct net_device {
 	struct mini_Qdisc __rcu	*miniq_egress;
 #endif
 
+#ifdef CONFIG_NET_SCHED
+	DECLARE_HASHTABLE	(qdisc_hash, 4);
+#endif
 	/* These may be needed for future network-power-down code. */
 	struct timer_list	watchdog_timer;
+	int			watchdog_timeo;
 
 	int __percpu		*pcpu_refcnt;
 	struct list_head	todo_list;
diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index a7378bcd9928..72bad13d4a3c 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -278,7 +278,7 @@ TRACE_EVENT(xdp_devmap_xmit,
 	),
 
 	TP_fast_assign(
-		__entry->map_id		= map->id;
+		__entry->map_id		= map ? map->id : 0;
 		__entry->act		= XDP_REDIRECT;
 		__entry->map_index	= map_index;
 		__entry->drops		= drops;
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index da9c832fc5c8..030d125c3839 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -53,13 +53,11 @@
 	(BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
 
 #define DEV_MAP_BULK_SIZE 16
-struct bpf_dtab_netdev;
-
-struct xdp_bulk_queue {
+struct xdp_dev_bulk_queue {
 	struct xdp_frame *q[DEV_MAP_BULK_SIZE];
 	struct list_head flush_node;
+	struct net_device *dev;
 	struct net_device *dev_rx;
-	struct bpf_dtab_netdev *obj;
 	unsigned int count;
 };
 
@@ -67,9 +65,8 @@ struct bpf_dtab_netdev {
 	struct net_device *dev; /* must be first member, due to tracepoint */
 	struct hlist_node index_hlist;
 	struct bpf_dtab *dtab;
-	struct xdp_bulk_queue __percpu *bulkq;
 	struct rcu_head rcu;
-	unsigned int idx; /* keep track of map index for tracepoint */
+	unsigned int idx;
 };
 
 struct bpf_dtab {
@@ -219,7 +216,6 @@ static void dev_map_free(struct bpf_map *map)
 
 			hlist_for_each_entry_safe(dev, next, head, index_hlist) {
 				hlist_del_rcu(&dev->index_hlist);
-				free_percpu(dev->bulkq);
 				dev_put(dev->dev);
 				kfree(dev);
 			}
@@ -234,7 +230,6 @@ static void dev_map_free(struct bpf_map *map)
 			if (!dev)
 				continue;
 
-			free_percpu(dev->bulkq);
 			dev_put(dev->dev);
 			kfree(dev);
 		}
@@ -320,10 +315,9 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
+static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 {
-	struct bpf_dtab_netdev *obj = bq->obj;
-	struct net_device *dev = obj->dev;
+	struct net_device *dev = bq->dev;
 	int sent = 0, drops = 0, err = 0;
 	int i;
 
@@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 out:
 	bq->count = 0;
 
-	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
-			      sent, drops, bq->dev_rx, dev, err);
+	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);
 	bq->dev_rx = NULL;
 	__list_del_clearprev(&bq->flush_node);
 	return 0;
@@ -374,7 +367,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 void __dev_map_flush(void)
 {
 	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
-	struct xdp_bulk_queue *bq, *tmp;
+	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
@@ -401,12 +394,12 @@ struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key)
 /* Runs under RCU-read-side, plus in softirq under NAPI protection.
  * Thus, safe percpu variable access.
  */
-static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
+static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
 
 {
 	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
-	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
+	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
 		bq_xmit_all(bq, 0);
@@ -444,7 +437,7 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 	if (unlikely(!xdpf))
 		return -EOVERFLOW;
 
-	return bq_enqueue(dst, xdpf, dev_rx);
+	return bq_enqueue(dev, xdpf, dev_rx);
 }
 
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -483,7 +476,6 @@ static void __dev_map_entry_free(struct rcu_head *rcu)
 	struct bpf_dtab_netdev *dev;
 
 	dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
-	free_percpu(dev->bulkq);
 	dev_put(dev->dev);
 	kfree(dev);
 }
@@ -538,30 +530,15 @@ static struct bpf_dtab_netdev *__dev_map_alloc_node(struct net *net,
 						    u32 ifindex,
 						    unsigned int idx)
 {
-	gfp_t gfp = GFP_ATOMIC | __GFP_NOWARN;
 	struct bpf_dtab_netdev *dev;
-	struct xdp_bulk_queue *bq;
-	int cpu;
 
-	dev = kmalloc_node(sizeof(*dev), gfp, dtab->map.numa_node);
+	dev = kmalloc_node(sizeof(*dev), GFP_ATOMIC | __GFP_NOWARN,
+			   dtab->map.numa_node);
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
-	dev->bulkq = __alloc_percpu_gfp(sizeof(*dev->bulkq),
-					sizeof(void *), gfp);
-	if (!dev->bulkq) {
-		kfree(dev);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	for_each_possible_cpu(cpu) {
-		bq = per_cpu_ptr(dev->bulkq, cpu);
-		bq->obj = dev;
-	}
-
 	dev->dev = dev_get_by_index(net, ifindex);
 	if (!dev->dev) {
-		free_percpu(dev->bulkq);
 		kfree(dev);
 		return ERR_PTR(-EINVAL);
 	}
@@ -721,9 +698,23 @@ static int dev_map_notification(struct notifier_block *notifier,
 {
 	struct net_device *netdev = netdev_notifier_info_to_dev(ptr);
 	struct bpf_dtab *dtab;
-	int i;
+	int i, cpu;
 
 	switch (event) {
+	case NETDEV_REGISTER:
+		if (!netdev->netdev_ops->ndo_xdp_xmit || netdev->xdp_bulkq)
+			break;
+
+		/* will be freed in free_netdev() */
+		netdev->xdp_bulkq =
+			__alloc_percpu_gfp(sizeof(struct xdp_dev_bulk_queue),
+					   sizeof(void *), GFP_ATOMIC);
+		if (!netdev->xdp_bulkq)
+			return NOTIFY_BAD;
+
+		for_each_possible_cpu(cpu)
+			per_cpu_ptr(netdev->xdp_bulkq, cpu)->dev = netdev;
+		break;
 	case NETDEV_UNREGISTER:
 		/* This rcu_read_lock/unlock pair is needed because
 		 * dev_map_list is an RCU list AND to ensure a delete
diff --git a/net/core/dev.c b/net/core/dev.c
index d99f88c58636..e7802a41ae7f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9847,6 +9847,8 @@ void free_netdev(struct net_device *dev)
 
 	free_percpu(dev->pcpu_refcnt);
 	dev->pcpu_refcnt = NULL;
+	free_percpu(dev->xdp_bulkq);
+	dev->xdp_bulkq = NULL;
 
 	netdev_unregister_lockdep_key(dev);
 

[PATCH bpf-next v2 2/2] xdp: Use bulking for non-map XDP_REDIRECT and consolidate code paths

[Toke Høiland-Jørgensen]

From: Toke Høiland-Jørgensen <toke@redhat.com>

Since the bulk queue used by XDP_REDIRECT now lives in struct net_device,
we can re-use the bulking for the non-map version of the bpf_redirect()
helper. This is a simple matter of having xdp_do_redirect_slow() queue the
frame on the bulk queue instead of sending it out with __bpf_tx_xdp().

Unfortunately we can't make the bpf_redirect() helper return an error if
the ifindex doesn't exit (as bpf_redirect_map() does), because we don't
have a reference to the network namespace of the ingress device at the time
the helper is called. So we have to leave it as-is and keep the device
lookup in xdp_do_redirect_slow().

Since this leaves less reason to have the non-map redirect code in a
separate function, so we get rid of the xdp_do_redirect_slow() function
entirely. This does lose us the tracepoint disambiguation, but fortunately
the xdp_redirect and xdp_redirect_map tracepoints use the same tracepoint
entry structures. This means both can contain a map index, so we can just
amend the tracepoint definitions so we always emit the xdp_redirect(_err)
tracepoints, but with the map ID only populated if a map is present. This
means we retire the xdp_redirect_map(_err) tracepoints entirely, but keep
the definitions around in case someone is still listening for them.

With this change, the performance of the xdp_redirect sample program goes
from 5Mpps to 8.4Mpps (a 68% increase).

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
 include/linux/bpf.h        |   13 +++++-
 include/trace/events/xdp.h |  102 +++++++++++++++++++-------------------------
 kernel/bpf/devmap.c        |   31 +++++++++----
 net/core/filter.c          |   86 +++++++------------------------------
 4 files changed, 95 insertions(+), 137 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index b14e51d56a82..25c050202536 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -962,7 +962,9 @@ struct sk_buff;
 
 struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
-void __dev_map_flush(void);
+void __dev_flush(void);
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -1071,13 +1073,20 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 	return NULL;
 }
 
-static inline void __dev_map_flush(void)
+static inline void __dev_flush(void)
 {
 }
 
 struct xdp_buff;
 struct bpf_dtab_netdev;
 
+static inline
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	return 0;
+}
+
 static inline
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx)
diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index 72bad13d4a3c..cf568a38f852 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -79,14 +79,27 @@ TRACE_EVENT(xdp_bulk_tx,
 		  __entry->sent, __entry->drops, __entry->err)
 );
 
+#ifndef __DEVMAP_OBJ_TYPE
+#define __DEVMAP_OBJ_TYPE
+struct _bpf_dtab_netdev {
+	struct net_device *dev;
+};
+#endif /* __DEVMAP_OBJ_TYPE */
+
+#define devmap_ifindex(tgt, map)				\
+	(((map->map_type == BPF_MAP_TYPE_DEVMAP ||	\
+		  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH)) ? \
+	  ((struct _bpf_dtab_netdev *)tgt)->dev->ifindex : 0)
+
+
 DECLARE_EVENT_CLASS(xdp_redirect_template,
 
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
 
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
+	TP_ARGS(dev, xdp, tgt, err, map, index),
 
 	TP_STRUCT__entry(
 		__field(int, prog_id)
@@ -103,90 +116,65 @@ DECLARE_EVENT_CLASS(xdp_redirect_template,
 		__entry->act		= XDP_REDIRECT;
 		__entry->ifindex	= dev->ifindex;
 		__entry->err		= err;
-		__entry->to_ifindex	= to_ifindex;
+		__entry->to_ifindex	= map ? devmap_ifindex(tgt, map) :
+						index;
 		__entry->map_id		= map ? map->id : 0;
-		__entry->map_index	= map_index;
+		__entry->map_index	= map ? index : 0;
 	),
 
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d",
+	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
+		  " map_id=%d map_index=%d",
 		  __entry->prog_id,
 		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
 		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err)
+		  __entry->err, __entry->map_id, __entry->map_index)
 );
 
 DEFINE_EVENT(xdp_redirect_template, xdp_redirect,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
 DEFINE_EVENT(xdp_redirect_template, xdp_redirect_err,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
 #define _trace_xdp_redirect(dev, xdp, to)		\
-	 trace_xdp_redirect(dev, xdp, to, 0, NULL, 0);
+	 trace_xdp_redirect(dev, xdp, NULL, 0, NULL, to);
 
 #define _trace_xdp_redirect_err(dev, xdp, to, err)	\
-	 trace_xdp_redirect_err(dev, xdp, to, err, NULL, 0);
+	 trace_xdp_redirect_err(dev, xdp, NULL, err, NULL, to);
+
+#define _trace_xdp_redirect_map(dev, xdp, to, map, index)		\
+	 trace_xdp_redirect(dev, xdp, to, 0, map, index);
 
-DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map,
+#define _trace_xdp_redirect_map_err(dev, xdp, to, map, index, err)	\
+	 trace_xdp_redirect_err(dev, xdp, to, err, map, index);
+
+/* not used anymore, but kept around so as not to break old programs */
+DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
-		  " map_id=%d map_index=%d",
-		  __entry->prog_id,
-		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
-		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err,
-		  __entry->map_id, __entry->map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
-DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map_err,
+DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map_err,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
-		  " map_id=%d map_index=%d",
-		  __entry->prog_id,
-		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
-		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err,
-		  __entry->map_id, __entry->map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
-#ifndef __DEVMAP_OBJ_TYPE
-#define __DEVMAP_OBJ_TYPE
-struct _bpf_dtab_netdev {
-	struct net_device *dev;
-};
-#endif /* __DEVMAP_OBJ_TYPE */
-
-#define devmap_ifindex(fwd, map)				\
-	((map->map_type == BPF_MAP_TYPE_DEVMAP ||		\
-	  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH) ?		\
-	  ((struct _bpf_dtab_netdev *)fwd)->dev->ifindex : 0)
-
-#define _trace_xdp_redirect_map(dev, xdp, fwd, map, idx)		\
-	 trace_xdp_redirect_map(dev, xdp, devmap_ifindex(fwd, map),	\
-				0, map, idx)
-
-#define _trace_xdp_redirect_map_err(dev, xdp, fwd, map, idx, err)	\
-	 trace_xdp_redirect_map_err(dev, xdp, devmap_ifindex(fwd, map),	\
-				    err, map, idx)
-
 TRACE_EVENT(xdp_cpumap_kthread,
 
 	TP_PROTO(int map_id, unsigned int processed,  unsigned int drops,
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 030d125c3839..db32272c4f77 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -81,7 +81,7 @@ struct bpf_dtab {
 	u32 n_buckets;
 };
 
-static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
+static DEFINE_PER_CPU(struct list_head, dev_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -357,16 +357,16 @@ static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 	goto out;
 }
 
-/* __dev_map_flush is called from xdp_do_flush_map() which _must_ be signaled
+/* __dev_flush is called from xdp_do_flush_map() which _must_ be signaled
  * from the driver before returning from its napi->poll() routine. The poll()
  * routine is called either from busy_poll context or net_rx_action signaled
  * from NET_RX_SOFTIRQ. Either way the poll routine must complete before the
  * net device can be torn down. On devmap tear down we ensure the flush list
  * is empty before completing to ensure all flush operations have completed.
  */
-void __dev_map_flush(void)
+void __dev_flush(void)
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
 	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
@@ -398,7 +398,7 @@ static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
 
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
 	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -419,10 +419,9 @@ static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 	return 0;
 }
 
-int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
-		    struct net_device *dev_rx)
+static inline int _xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+			       struct net_device *dev_rx)
 {
-	struct net_device *dev = dst->dev;
 	struct xdp_frame *xdpf;
 	int err;
 
@@ -440,6 +439,20 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 	return bq_enqueue(dev, xdpf, dev_rx);
 }
 
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	return _xdp_enqueue(dev, xdp, dev_rx);
+}
+
+int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	struct net_device *dev = dst->dev;
+
+	return _xdp_enqueue(dev, xdp, dev_rx);
+}
+
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog)
 {
@@ -762,7 +775,7 @@ static int __init dev_map_init(void)
 	register_netdevice_notifier(&dev_map_notifier);
 
 	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(&per_cpu(dev_map_flush_list, cpu));
+		INIT_LIST_HEAD(&per_cpu(dev_flush_list, cpu));
 	return 0;
 }
 
diff --git a/net/core/filter.c b/net/core/filter.c
index 42fd17c48c5f..f023f3a8f351 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3458,58 +3458,6 @@ static const struct bpf_func_proto bpf_xdp_adjust_meta_proto = {
 	.arg2_type	= ARG_ANYTHING,
 };
 
-static int __bpf_tx_xdp(struct net_device *dev,
-			struct bpf_map *map,
-			struct xdp_buff *xdp,
-			u32 index)
-{
-	struct xdp_frame *xdpf;
-	int err, sent;
-
-	if (!dev->netdev_ops->ndo_xdp_xmit) {
-		return -EOPNOTSUPP;
-	}
-
-	err = xdp_ok_fwd_dev(dev, xdp->data_end - xdp->data);
-	if (unlikely(err))
-		return err;
-
-	xdpf = convert_to_xdp_frame(xdp);
-	if (unlikely(!xdpf))
-		return -EOVERFLOW;
-
-	sent = dev->netdev_ops->ndo_xdp_xmit(dev, 1, &xdpf, XDP_XMIT_FLUSH);
-	if (sent <= 0)
-		return sent;
-	return 0;
-}
-
-static noinline int
-xdp_do_redirect_slow(struct net_device *dev, struct xdp_buff *xdp,
-		     struct bpf_prog *xdp_prog, struct bpf_redirect_info *ri)
-{
-	struct net_device *fwd;
-	u32 index = ri->tgt_index;
-	int err;
-
-	fwd = dev_get_by_index_rcu(dev_net(dev), index);
-	ri->tgt_index = 0;
-	if (unlikely(!fwd)) {
-		err = -EINVAL;
-		goto err;
-	}
-
-	err = __bpf_tx_xdp(fwd, NULL, xdp, 0);
-	if (unlikely(err))
-		goto err;
-
-	_trace_xdp_redirect(dev, xdp_prog, index);
-	return 0;
-err:
-	_trace_xdp_redirect_err(dev, xdp_prog, index, err);
-	return err;
-}
-
 static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 			    struct bpf_map *map, struct xdp_buff *xdp)
 {
@@ -3529,7 +3477,7 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 
 void xdp_do_flush_map(void)
 {
-	__dev_map_flush();
+	__dev_flush();
 	__cpu_map_flush();
 	__xsk_map_flush();
 }
@@ -3568,10 +3516,11 @@ void bpf_clear_redirect_map(struct bpf_map *map)
 	}
 }
 
-static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
-			       struct bpf_prog *xdp_prog, struct bpf_map *map,
-			       struct bpf_redirect_info *ri)
+int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
+		    struct bpf_prog *xdp_prog)
 {
+	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_map *map = READ_ONCE(ri->map);
 	u32 index = ri->tgt_index;
 	void *fwd = ri->tgt_value;
 	int err;
@@ -3580,7 +3529,18 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	ri->tgt_value = NULL;
 	WRITE_ONCE(ri->map, NULL);
 
-	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
+	if (unlikely(!map)) {
+		fwd = dev_get_by_index_rcu(dev_net(dev), index);
+		if (unlikely(!fwd)) {
+			err = -EINVAL;
+			goto err;
+		}
+
+		err = dev_xdp_enqueue(fwd, xdp, dev);
+	} else {
+		err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
+	}
+
 	if (unlikely(err))
 		goto err;
 
@@ -3590,18 +3550,6 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	_trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err);
 	return err;
 }
-
-int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
-		    struct bpf_prog *xdp_prog)
-{
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	struct bpf_map *map = READ_ONCE(ri->map);
-
-	if (likely(map))
-		return xdp_do_redirect_map(dev, xdp, xdp_prog, map, ri);
-
-	return xdp_do_redirect_slow(dev, xdp, xdp_prog, ri);
-}
 EXPORT_SYMBOL_GPL(xdp_do_redirect);
 
 static int xdp_do_generic_redirect_map(struct net_device *dev,

Re: [PATCH bpf-next v2 0/2] xdp: Introduce bulking for non-map XDP_REDIRECT

[Alexei Starovoitov]

On Mon, Jan 13, 2020 at 10:11 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
> Since commit 96360004b862 ("xdp: Make devmap flush_list common for all map
> instances"), devmap flushing is a global operation instead of tied to a
> particular map. This means that with a bit of refactoring, we can finally fix
> the performance delta between the bpf_redirect_map() and bpf_redirect() helper
> functions, by introducing bulking for the latter as well.
>
> This series makes this change by moving the data structure used for the bulking
> into struct net_device itself, so we can access it even when there is not
> devmap. Once this is done, moving the bpf_redirect() helper to use the bulking
> mechanism becomes quite trivial, and brings bpf_redirect() up to the same as
> bpf_redirect_map():
>
>                        Before:   After:
> 1 CPU:
> bpf_redirect_map:      8.4 Mpps  8.4 Mpps  (no change)
> bpf_redirect:          5.0 Mpps  8.4 Mpps  (+68%)
> 2 CPUs:
> bpf_redirect_map:     15.9 Mpps  16.1 Mpps  (+1% or ~no change)
> bpf_redirect:          9.5 Mpps  15.9 Mpps  (+67%)
>
> After this patch series, the only semantics different between the two variants
> of the bpf() helper (apart from the absence of a map argument, obviously) is
> that the _map() variant will return an error if passed an invalid map index,
> whereas the bpf_redirect() helper will succeed, but drop packets on
> xdp_do_redirect(). This is because the helper has no reference to the calling
> netdev, so unfortunately we can't do the ifindex lookup directly in the helper.
>
> Changelog:
>
> v2:
>   - Consolidate code paths and tracepoints for map and non-map redirect variants
>     (Björn)
>   - Add performance data for 2-CPU test (Jesper)
>   - Move fields to avoid shifting cache lines in struct net_device (Eric)

John, since you commented on v1 please review this v2. Thanks!

Re: [PATCH bpf-next v2 2/2] xdp: Use bulking for non-map XDP_REDIRECT and consolidate code paths

[Maciej Fijalkowski]

On Mon, Jan 13, 2020 at 07:10:56PM +0100, Toke Høiland-Jørgensen wrote:
> From: Toke Høiland-Jørgensen <toke@redhat.com>
> 
> Since the bulk queue used by XDP_REDIRECT now lives in struct net_device,
> we can re-use the bulking for the non-map version of the bpf_redirect()
> helper. This is a simple matter of having xdp_do_redirect_slow() queue the
> frame on the bulk queue instead of sending it out with __bpf_tx_xdp().
> 
> Unfortunately we can't make the bpf_redirect() helper return an error if
> the ifindex doesn't exit (as bpf_redirect_map() does), because we don't
> have a reference to the network namespace of the ingress device at the time
> the helper is called. So we have to leave it as-is and keep the device
> lookup in xdp_do_redirect_slow().
> 
> Since this leaves less reason to have the non-map redirect code in a
> separate function, so we get rid of the xdp_do_redirect_slow() function
> entirely. This does lose us the tracepoint disambiguation, but fortunately
> the xdp_redirect and xdp_redirect_map tracepoints use the same tracepoint
> entry structures. This means both can contain a map index, so we can just
> amend the tracepoint definitions so we always emit the xdp_redirect(_err)
> tracepoints, but with the map ID only populated if a map is present. This
> means we retire the xdp_redirect_map(_err) tracepoints entirely, but keep
> the definitions around in case someone is still listening for them.
> 
> With this change, the performance of the xdp_redirect sample program goes
> from 5Mpps to 8.4Mpps (a 68% increase).
> 
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> ---
>  include/linux/bpf.h        |   13 +++++-
>  include/trace/events/xdp.h |  102 +++++++++++++++++++-------------------------
>  kernel/bpf/devmap.c        |   31 +++++++++----
>  net/core/filter.c          |   86 +++++++------------------------------
>  4 files changed, 95 insertions(+), 137 deletions(-)
> 
> diff --git a/include/linux/bpf.h b/include/linux/bpf.h
> index b14e51d56a82..25c050202536 100644
> --- a/include/linux/bpf.h
> +++ b/include/linux/bpf.h
> @@ -962,7 +962,9 @@ struct sk_buff;
>  
>  struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
>  struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
> -void __dev_map_flush(void);
> +void __dev_flush(void);
> +int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
> +		    struct net_device *dev_rx);
>  int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
>  		    struct net_device *dev_rx);
>  int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
> @@ -1071,13 +1073,20 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
>  	return NULL;
>  }
>  
> -static inline void __dev_map_flush(void)
> +static inline void __dev_flush(void)
>  {
>  }
>  
>  struct xdp_buff;
>  struct bpf_dtab_netdev;
>  
> +static inline
> +int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
> +		    struct net_device *dev_rx)
> +{
> +	return 0;
> +}
> +
>  static inline
>  int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
>  		    struct net_device *dev_rx)
> diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
> index 72bad13d4a3c..cf568a38f852 100644
> --- a/include/trace/events/xdp.h
> +++ b/include/trace/events/xdp.h
> @@ -79,14 +79,27 @@ TRACE_EVENT(xdp_bulk_tx,
>  		  __entry->sent, __entry->drops, __entry->err)
>  );
>  
> +#ifndef __DEVMAP_OBJ_TYPE
> +#define __DEVMAP_OBJ_TYPE
> +struct _bpf_dtab_netdev {
> +	struct net_device *dev;
> +};
> +#endif /* __DEVMAP_OBJ_TYPE */
> +
> +#define devmap_ifindex(tgt, map)				\
> +	(((map->map_type == BPF_MAP_TYPE_DEVMAP ||	\
> +		  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH)) ? \
> +	  ((struct _bpf_dtab_netdev *)tgt)->dev->ifindex : 0)
> +

Delete one blank line

> +
>  DECLARE_EVENT_CLASS(xdp_redirect_template,
>  
>  	TP_PROTO(const struct net_device *dev,
>  		 const struct bpf_prog *xdp,
> -		 int to_ifindex, int err,
> -		 const struct bpf_map *map, u32 map_index),
> +		 const void *tgt, int err,
> +		 const struct bpf_map *map, u32 index),
>  
> -	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
> +	TP_ARGS(dev, xdp, tgt, err, map, index),
>  
>  	TP_STRUCT__entry(
>  		__field(int, prog_id)
> @@ -103,90 +116,65 @@ DECLARE_EVENT_CLASS(xdp_redirect_template,
>  		__entry->act		= XDP_REDIRECT;
>  		__entry->ifindex	= dev->ifindex;
>  		__entry->err		= err;
> -		__entry->to_ifindex	= to_ifindex;
> +		__entry->to_ifindex	= map ? devmap_ifindex(tgt, map) :
> +						index;
>  		__entry->map_id		= map ? map->id : 0;
> -		__entry->map_index	= map_index;
> +		__entry->map_index	= map ? index : 0;
>  	),
>  
> -	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d",
> +	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
> +		  " map_id=%d map_index=%d",
>  		  __entry->prog_id,
>  		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
>  		  __entry->ifindex, __entry->to_ifindex,
> -		  __entry->err)
> +		  __entry->err, __entry->map_id, __entry->map_index)
>  );
>  
>  DEFINE_EVENT(xdp_redirect_template, xdp_redirect,
>  	TP_PROTO(const struct net_device *dev,
>  		 const struct bpf_prog *xdp,
> -		 int to_ifindex, int err,
> -		 const struct bpf_map *map, u32 map_index),
> -	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
> +		 const void *tgt, int err,
> +		 const struct bpf_map *map, u32 index),
> +	TP_ARGS(dev, xdp, tgt, err, map, index)
>  );
>  
>  DEFINE_EVENT(xdp_redirect_template, xdp_redirect_err,
>  	TP_PROTO(const struct net_device *dev,
>  		 const struct bpf_prog *xdp,
> -		 int to_ifindex, int err,
> -		 const struct bpf_map *map, u32 map_index),
> -	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
> +		 const void *tgt, int err,
> +		 const struct bpf_map *map, u32 index),
> +	TP_ARGS(dev, xdp, tgt, err, map, index)
>  );
>  
>  #define _trace_xdp_redirect(dev, xdp, to)		\
> -	 trace_xdp_redirect(dev, xdp, to, 0, NULL, 0);
> +	 trace_xdp_redirect(dev, xdp, NULL, 0, NULL, to);
>  
>  #define _trace_xdp_redirect_err(dev, xdp, to, err)	\
> -	 trace_xdp_redirect_err(dev, xdp, to, err, NULL, 0);
> +	 trace_xdp_redirect_err(dev, xdp, NULL, err, NULL, to);
> +
> +#define _trace_xdp_redirect_map(dev, xdp, to, map, index)		\
> +	 trace_xdp_redirect(dev, xdp, to, 0, map, index);
>  
> -DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map,
> +#define _trace_xdp_redirect_map_err(dev, xdp, to, map, index, err)	\
> +	 trace_xdp_redirect_err(dev, xdp, to, err, map, index);
> +
> +/* not used anymore, but kept around so as not to break old programs */
> +DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map,
>  	TP_PROTO(const struct net_device *dev,
>  		 const struct bpf_prog *xdp,
> -		 int to_ifindex, int err,
> -		 const struct bpf_map *map, u32 map_index),
> -	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
> -	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
> -		  " map_id=%d map_index=%d",
> -		  __entry->prog_id,
> -		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
> -		  __entry->ifindex, __entry->to_ifindex,
> -		  __entry->err,
> -		  __entry->map_id, __entry->map_index)
> +		 const void *tgt, int err,
> +		 const struct bpf_map *map, u32 index),
> +	TP_ARGS(dev, xdp, tgt, err, map, index)
>  );
>  
> -DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map_err,
> +DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map_err,
>  	TP_PROTO(const struct net_device *dev,
>  		 const struct bpf_prog *xdp,
> -		 int to_ifindex, int err,
> -		 const struct bpf_map *map, u32 map_index),
> -	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
> -	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
> -		  " map_id=%d map_index=%d",
> -		  __entry->prog_id,
> -		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
> -		  __entry->ifindex, __entry->to_ifindex,
> -		  __entry->err,
> -		  __entry->map_id, __entry->map_index)
> +		 const void *tgt, int err,
> +		 const struct bpf_map *map, u32 index),
> +	TP_ARGS(dev, xdp, tgt, err, map, index)
>  );
>  
> -#ifndef __DEVMAP_OBJ_TYPE
> -#define __DEVMAP_OBJ_TYPE
> -struct _bpf_dtab_netdev {
> -	struct net_device *dev;
> -};
> -#endif /* __DEVMAP_OBJ_TYPE */
> -
> -#define devmap_ifindex(fwd, map)				\
> -	((map->map_type == BPF_MAP_TYPE_DEVMAP ||		\
> -	  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH) ?		\
> -	  ((struct _bpf_dtab_netdev *)fwd)->dev->ifindex : 0)
> -
> -#define _trace_xdp_redirect_map(dev, xdp, fwd, map, idx)		\
> -	 trace_xdp_redirect_map(dev, xdp, devmap_ifindex(fwd, map),	\
> -				0, map, idx)
> -
> -#define _trace_xdp_redirect_map_err(dev, xdp, fwd, map, idx, err)	\
> -	 trace_xdp_redirect_map_err(dev, xdp, devmap_ifindex(fwd, map),	\
> -				    err, map, idx)
> -
>  TRACE_EVENT(xdp_cpumap_kthread,
>  
>  	TP_PROTO(int map_id, unsigned int processed,  unsigned int drops,
> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
> index 030d125c3839..db32272c4f77 100644
> --- a/kernel/bpf/devmap.c
> +++ b/kernel/bpf/devmap.c
> @@ -81,7 +81,7 @@ struct bpf_dtab {
>  	u32 n_buckets;
>  };
>  
> -static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
> +static DEFINE_PER_CPU(struct list_head, dev_flush_list);
>  static DEFINE_SPINLOCK(dev_map_lock);
>  static LIST_HEAD(dev_map_list);
>  
> @@ -357,16 +357,16 @@ static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
>  	goto out;
>  }
>  
> -/* __dev_map_flush is called from xdp_do_flush_map() which _must_ be signaled
> +/* __dev_flush is called from xdp_do_flush_map() which _must_ be signaled
>   * from the driver before returning from its napi->poll() routine. The poll()
>   * routine is called either from busy_poll context or net_rx_action signaled
>   * from NET_RX_SOFTIRQ. Either way the poll routine must complete before the
>   * net device can be torn down. On devmap tear down we ensure the flush list
>   * is empty before completing to ensure all flush operations have completed.
>   */
> -void __dev_map_flush(void)
> +void __dev_flush(void)
>  {
> -	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
> +	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
>  	struct xdp_dev_bulk_queue *bq, *tmp;
>  
>  	rcu_read_lock();
> @@ -398,7 +398,7 @@ static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
>  		      struct net_device *dev_rx)
>  
^^^
While you're at this part of code maybe you could remove another blank
line? :)

>  {
> -	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
> +	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
>  	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
>  
>  	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
> @@ -419,10 +419,9 @@ static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
>  	return 0;
>  }
>  
> -int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
> -		    struct net_device *dev_rx)
> +static inline int _xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
> +			       struct net_device *dev_rx)
>  {
> -	struct net_device *dev = dst->dev;
>  	struct xdp_frame *xdpf;
>  	int err;
>  
> @@ -440,6 +439,20 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
>  	return bq_enqueue(dev, xdpf, dev_rx);
>  }
>  
> +int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
> +		    struct net_device *dev_rx)
> +{
> +	return _xdp_enqueue(dev, xdp, dev_rx);

AFAIK normally the internal functions are prefixed with a double
underscore, no? Could we have it renamed to __xdp_enqueue?

> +}
> +
> +int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
> +		    struct net_device *dev_rx)
> +{
> +	struct net_device *dev = dst->dev;
> +
> +	return _xdp_enqueue(dev, xdp, dev_rx);
> +}
> +
>  int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
>  			     struct bpf_prog *xdp_prog)
>  {
> @@ -762,7 +775,7 @@ static int __init dev_map_init(void)
>  	register_netdevice_notifier(&dev_map_notifier);
>  
>  	for_each_possible_cpu(cpu)
> -		INIT_LIST_HEAD(&per_cpu(dev_map_flush_list, cpu));
> +		INIT_LIST_HEAD(&per_cpu(dev_flush_list, cpu));
>  	return 0;
>  }
>  
> diff --git a/net/core/filter.c b/net/core/filter.c
> index 42fd17c48c5f..f023f3a8f351 100644
> --- a/net/core/filter.c
> +++ b/net/core/filter.c
> @@ -3458,58 +3458,6 @@ static const struct bpf_func_proto bpf_xdp_adjust_meta_proto = {
>  	.arg2_type	= ARG_ANYTHING,
>  };
>  
> -static int __bpf_tx_xdp(struct net_device *dev,
> -			struct bpf_map *map,
> -			struct xdp_buff *xdp,
> -			u32 index)
> -{
> -	struct xdp_frame *xdpf;
> -	int err, sent;
> -
> -	if (!dev->netdev_ops->ndo_xdp_xmit) {
> -		return -EOPNOTSUPP;
> -	}
> -
> -	err = xdp_ok_fwd_dev(dev, xdp->data_end - xdp->data);
> -	if (unlikely(err))
> -		return err;
> -
> -	xdpf = convert_to_xdp_frame(xdp);
> -	if (unlikely(!xdpf))
> -		return -EOVERFLOW;
> -
> -	sent = dev->netdev_ops->ndo_xdp_xmit(dev, 1, &xdpf, XDP_XMIT_FLUSH);
> -	if (sent <= 0)
> -		return sent;
> -	return 0;
> -}
> -
> -static noinline int
> -xdp_do_redirect_slow(struct net_device *dev, struct xdp_buff *xdp,
> -		     struct bpf_prog *xdp_prog, struct bpf_redirect_info *ri)
> -{
> -	struct net_device *fwd;
> -	u32 index = ri->tgt_index;
> -	int err;
> -
> -	fwd = dev_get_by_index_rcu(dev_net(dev), index);
> -	ri->tgt_index = 0;
> -	if (unlikely(!fwd)) {
> -		err = -EINVAL;
> -		goto err;
> -	}
> -
> -	err = __bpf_tx_xdp(fwd, NULL, xdp, 0);
> -	if (unlikely(err))
> -		goto err;
> -
> -	_trace_xdp_redirect(dev, xdp_prog, index);
> -	return 0;
> -err:
> -	_trace_xdp_redirect_err(dev, xdp_prog, index, err);
> -	return err;
> -}
> -
>  static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
>  			    struct bpf_map *map, struct xdp_buff *xdp)
>  {
> @@ -3529,7 +3477,7 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
>  
>  void xdp_do_flush_map(void)
>  {
> -	__dev_map_flush();
> +	__dev_flush();

Hmm maybe it's also time for s/xdp_do_flush_map/xdp_do_flush ? Drivers
changes, though :<

>  	__cpu_map_flush();
>  	__xsk_map_flush();
>  }
> @@ -3568,10 +3516,11 @@ void bpf_clear_redirect_map(struct bpf_map *map)
>  	}
>  }
>  
> -static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
> -			       struct bpf_prog *xdp_prog, struct bpf_map *map,
> -			       struct bpf_redirect_info *ri)
> +int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
> +		    struct bpf_prog *xdp_prog)
>  {
> +	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
> +	struct bpf_map *map = READ_ONCE(ri->map);
>  	u32 index = ri->tgt_index;
>  	void *fwd = ri->tgt_value;
>  	int err;
> @@ -3580,7 +3529,18 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
>  	ri->tgt_value = NULL;
>  	WRITE_ONCE(ri->map, NULL);
>  
> -	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
> +	if (unlikely(!map)) {
> +		fwd = dev_get_by_index_rcu(dev_net(dev), index);
> +		if (unlikely(!fwd)) {
> +			err = -EINVAL;
> +			goto err;
> +		}
> +
> +		err = dev_xdp_enqueue(fwd, xdp, dev);
> +	} else {
> +		err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
> +	}
> +
>  	if (unlikely(err))
>  		goto err;
>  
> @@ -3590,18 +3550,6 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
>  	_trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err);
>  	return err;
>  }
> -
> -int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
> -		    struct bpf_prog *xdp_prog)
> -{
> -	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
> -	struct bpf_map *map = READ_ONCE(ri->map);
> -
> -	if (likely(map))
> -		return xdp_do_redirect_map(dev, xdp, xdp_prog, map, ri);
> -
> -	return xdp_do_redirect_slow(dev, xdp, xdp_prog, ri);
> -}
>  EXPORT_SYMBOL_GPL(xdp_do_redirect);
>  
>  static int xdp_do_generic_redirect_map(struct net_device *dev,
> 

Re: [PATCH bpf-next v2 0/2] xdp: Introduce bulking for non-map XDP_REDIRECT

[John Fastabend]

Alexei Starovoitov wrote:
> On Mon, Jan 13, 2020 at 10:11 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
> >
> > Since commit 96360004b862 ("xdp: Make devmap flush_list common for all map
> > instances"), devmap flushing is a global operation instead of tied to a
> > particular map. This means that with a bit of refactoring, we can finally fix
> > the performance delta between the bpf_redirect_map() and bpf_redirect() helper
> > functions, by introducing bulking for the latter as well.
> >
> > This series makes this change by moving the data structure used for the bulking
> > into struct net_device itself, so we can access it even when there is not
> > devmap. Once this is done, moving the bpf_redirect() helper to use the bulking
> > mechanism becomes quite trivial, and brings bpf_redirect() up to the same as
> > bpf_redirect_map():
> >
> >                        Before:   After:
> > 1 CPU:
> > bpf_redirect_map:      8.4 Mpps  8.4 Mpps  (no change)
> > bpf_redirect:          5.0 Mpps  8.4 Mpps  (+68%)
> > 2 CPUs:
> > bpf_redirect_map:     15.9 Mpps  16.1 Mpps  (+1% or ~no change)
> > bpf_redirect:          9.5 Mpps  15.9 Mpps  (+67%)
> >
> > After this patch series, the only semantics different between the two variants
> > of the bpf() helper (apart from the absence of a map argument, obviously) is
> > that the _map() variant will return an error if passed an invalid map index,
> > whereas the bpf_redirect() helper will succeed, but drop packets on
> > xdp_do_redirect(). This is because the helper has no reference to the calling
> > netdev, so unfortunately we can't do the ifindex lookup directly in the helper.
> >
> > Changelog:
> >
> > v2:
> >   - Consolidate code paths and tracepoints for map and non-map redirect variants
> >     (Björn)
> >   - Add performance data for 2-CPU test (Jesper)
> >   - Move fields to avoid shifting cache lines in struct net_device (Eric)
> 
> John, since you commented on v1 please review this v2. Thanks!

hmm don't think I had an initial comment but will review regardless ;)

RE: [PATCH bpf-next v2 2/2] xdp: Use bulking for non-map XDP_REDIRECT and consolidate code paths

[John Fastabend]

Toke Høiland-Jørgensen wrote:
> From: Toke Høiland-Jørgensen <toke@redhat.com>
> 
> Since the bulk queue used by XDP_REDIRECT now lives in struct net_device,
> we can re-use the bulking for the non-map version of the bpf_redirect()
> helper. This is a simple matter of having xdp_do_redirect_slow() queue the
> frame on the bulk queue instead of sending it out with __bpf_tx_xdp().
> 
> Unfortunately we can't make the bpf_redirect() helper return an error if
> the ifindex doesn't exit (as bpf_redirect_map() does), because we don't
> have a reference to the network namespace of the ingress device at the time
> the helper is called. So we have to leave it as-is and keep the device
> lookup in xdp_do_redirect_slow().
> 
> Since this leaves less reason to have the non-map redirect code in a
> separate function, so we get rid of the xdp_do_redirect_slow() function
> entirely. This does lose us the tracepoint disambiguation, but fortunately
> the xdp_redirect and xdp_redirect_map tracepoints use the same tracepoint
> entry structures. This means both can contain a map index, so we can just
> amend the tracepoint definitions so we always emit the xdp_redirect(_err)
> tracepoints, but with the map ID only populated if a map is present. This
> means we retire the xdp_redirect_map(_err) tracepoints entirely, but keep
> the definitions around in case someone is still listening for them.
> 
> With this change, the performance of the xdp_redirect sample program goes
> from 5Mpps to 8.4Mpps (a 68% increase).
> 
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>

Acked-by: John Fastabend <john.fastabend@gmail.com>

RE: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[John Fastabend]

Toke Høiland-Jørgensen wrote:
> From: Toke Høiland-Jørgensen <toke@redhat.com>
> 
> Commit 96360004b862 ("xdp: Make devmap flush_list common for all map
> instances"), changed devmap flushing to be a global operation instead of a
> per-map operation. However, the queue structure used for bulking was still
> allocated as part of the containing map.
> 
> This patch moves the devmap bulk queue into struct net_device. The
> motivation for this is reusing it for the non-map variant of XDP_REDIRECT,
> which will be changed in a subsequent commit.  To avoid other fields of
> struct net_device moving to different cache lines, we also move a couple of
> other members around.
> 
> We defer the actual allocation of the bulk queue structure until the
> NETDEV_REGISTER notification devmap.c. This makes it possible to check for
> ndo_xdp_xmit support before allocating the structure, which is not possible
> at the time struct net_device is allocated. However, we keep the freeing in
> free_netdev() to avoid adding another RCU callback on NETDEV_UNREGISTER.
> 
> Because of this change, we lose the reference back to the map that
> originated the redirect, so change the tracepoint to always return 0 as the
> map ID and index. Otherwise no functional change is intended with this
> patch.
> 
> Acked-by: Björn Töpel <bjorn.topel@intel.com>
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> ---

LGTM. I didn't check the net_device layout with pahole though so I'm
trusting they are good from v1 discussion.

Acked-by: John Fastabend <john.fastabend@gmail.com>

Re: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Jesper Dangaard Brouer]

On Mon, 13 Jan 2020 19:10:55 +0100
Toke Høiland-Jørgensen <toke@redhat.com> wrote:

> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
> index da9c832fc5c8..030d125c3839 100644
> --- a/kernel/bpf/devmap.c
> +++ b/kernel/bpf/devmap.c
[...]
> @@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
>  out:
>  	bq->count = 0;
>  
> -	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
> -			      sent, drops, bq->dev_rx, dev, err);
> +	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);

Hmm ... I don't like that we lose the map_id and map_index identifier.
This is part of our troubleshooting interface.  

>  	bq->dev_rx = NULL;
>  	__list_del_clearprev(&bq->flush_node);
>  	return 0;



-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer

Re: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Toke Høiland-Jørgensen]

Jesper Dangaard Brouer <brouer@redhat.com> writes:

> On Mon, 13 Jan 2020 19:10:55 +0100
> Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
>> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
>> index da9c832fc5c8..030d125c3839 100644
>> --- a/kernel/bpf/devmap.c
>> +++ b/kernel/bpf/devmap.c
> [...]
>> @@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
>>  out:
>>  	bq->count = 0;
>>  
>> -	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
>> -			      sent, drops, bq->dev_rx, dev, err);
>> +	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);
>
> Hmm ... I don't like that we lose the map_id and map_index identifier.
> This is part of our troubleshooting interface.

Hmm, I guess I can take another look at whether there's a way to avoid
that. Any ideas?

-Toke

RE: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Toke Høiland-Jørgensen]

John Fastabend <john.fastabend@gmail.com> writes:

> Toke Høiland-Jørgensen wrote:
>> From: Toke Høiland-Jørgensen <toke@redhat.com>
>> 
>> Commit 96360004b862 ("xdp: Make devmap flush_list common for all map
>> instances"), changed devmap flushing to be a global operation instead of a
>> per-map operation. However, the queue structure used for bulking was still
>> allocated as part of the containing map.
>> 
>> This patch moves the devmap bulk queue into struct net_device. The
>> motivation for this is reusing it for the non-map variant of XDP_REDIRECT,
>> which will be changed in a subsequent commit.  To avoid other fields of
>> struct net_device moving to different cache lines, we also move a couple of
>> other members around.
>> 
>> We defer the actual allocation of the bulk queue structure until the
>> NETDEV_REGISTER notification devmap.c. This makes it possible to check for
>> ndo_xdp_xmit support before allocating the structure, which is not possible
>> at the time struct net_device is allocated. However, we keep the freeing in
>> free_netdev() to avoid adding another RCU callback on NETDEV_UNREGISTER.
>> 
>> Because of this change, we lose the reference back to the map that
>> originated the redirect, so change the tracepoint to always return 0 as the
>> map ID and index. Otherwise no functional change is intended with this
>> patch.
>> 
>> Acked-by: Björn Töpel <bjorn.topel@intel.com>
>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>> ---
>
> LGTM. I didn't check the net_device layout with pahole though so I'm
> trusting they are good from v1 discussion.

I believe so; looks like this now:

	/* --- cacheline 14 boundary (896 bytes) --- */
	struct netdev_queue *      _tx __attribute__((__aligned__(64))); /*   896     8 */
	unsigned int               num_tx_queues;        /*   904     4 */
	unsigned int               real_num_tx_queues;   /*   908     4 */
	struct Qdisc *             qdisc;                /*   912     8 */
	unsigned int               tx_queue_len;         /*   920     4 */
	spinlock_t                 tx_global_lock;       /*   924     4 */
	struct xdp_dev_bulk_queue * xdp_bulkq;           /*   928     8 */
	struct xps_dev_maps *      xps_cpus_map;         /*   936     8 */
	struct xps_dev_maps *      xps_rxqs_map;         /*   944     8 */
	struct mini_Qdisc *        miniq_egress;         /*   952     8 */
	/* --- cacheline 15 boundary (960 bytes) --- */
	struct hlist_head  qdisc_hash[16];               /*   960   128 */
	/* --- cacheline 17 boundary (1088 bytes) --- */
	struct timer_list  watchdog_timer;               /*  1088    40 */

	/* XXX last struct has 4 bytes of padding */

	int                        watchdog_timeo;       /*  1128     4 */

	/* XXX 4 bytes hole, try to pack */

	int *                      pcpu_refcnt;          /*  1136     8 */
	struct list_head   todo_list;                    /*  1144    16 */
	/* --- cacheline 18 boundary (1152 bytes) was 8 bytes ago --- */

-Toke

Re: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Jesper Dangaard Brouer]

On Wed, 15 Jan 2020 23:11:21 +0100
Toke Høiland-Jørgensen <toke@redhat.com> wrote:

> Jesper Dangaard Brouer <brouer@redhat.com> writes:
> 
> > On Mon, 13 Jan 2020 19:10:55 +0100
> > Toke Høiland-Jørgensen <toke@redhat.com> wrote:
> >  
> >> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
> >> index da9c832fc5c8..030d125c3839 100644
> >> --- a/kernel/bpf/devmap.c
> >> +++ b/kernel/bpf/devmap.c  
> > [...]  
> >> @@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
> >>  out:
> >>  	bq->count = 0;
> >>  
> >> -	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
> >> -			      sent, drops, bq->dev_rx, dev, err);
> >> +	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);  
> >
> > Hmm ... I don't like that we lose the map_id and map_index identifier.
> > This is part of our troubleshooting interface.  
> 
> Hmm, I guess I can take another look at whether there's a way to avoid
> that. Any ideas?

Looking at the code and the other tracepoints...

I will actually suggest to remove these two arguments, because the
trace_xdp_redirect_map tracepoint also contains the ifindex'es, and to
troubleshoot people can record both tracepoints and do the correlation
themselves.

When changing the tracepoint I would like to keep member 'drops' and
'sent' at the same struct offsets.  As our xdp_monitor example reads
these and I hope we can kept it working this way.

I've coded it up, and tested it.  The new xdp_monitor will work on
older kernels, but the old xdp_monitor will fail attaching on newer
kernels. I think this is fair enough, as we are backwards compatible.


[PATCH] devmap: adjust tracepoing after Tokes changes

From: Jesper Dangaard Brouer <brouer@redhat.com>

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
---
 include/trace/events/xdp.h     |   29 ++++++++++++-----------------
 kernel/bpf/devmap.c            |    2 +-
 samples/bpf/xdp_monitor_kern.c |    8 +++-----
 3 files changed, 16 insertions(+), 23 deletions(-)

diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index cf568a38f852..f1e64689ce94 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -247,43 +247,38 @@ TRACE_EVENT(xdp_cpumap_enqueue,
 
 TRACE_EVENT(xdp_devmap_xmit,
 
-	TP_PROTO(const struct bpf_map *map, u32 map_index,
-		 int sent, int drops,
-		 const struct net_device *from_dev,
-		 const struct net_device *to_dev, int err),
+	TP_PROTO(const struct net_device *from_dev,
+		 const struct net_device *to_dev,
+		 int sent, int drops, int err),
 
-	TP_ARGS(map, map_index, sent, drops, from_dev, to_dev, err),
+	TP_ARGS(from_dev, to_dev, sent, drops, err),
 
 	TP_STRUCT__entry(
-		__field(int, map_id)
+		__field(int, from_ifindex)
 		__field(u32, act)
-		__field(u32, map_index)
+		__field(int, to_ifindex)
 		__field(int, drops)
 		__field(int, sent)
-		__field(int, from_ifindex)
-		__field(int, to_ifindex)
 		__field(int, err)
 	),
 
 	TP_fast_assign(
-		__entry->map_id		= map ? map->id : 0;
+		__entry->from_ifindex	= from_dev->ifindex;
 		__entry->act		= XDP_REDIRECT;
-		__entry->map_index	= map_index;
+		__entry->to_ifindex	= to_dev->ifindex;
 		__entry->drops		= drops;
 		__entry->sent		= sent;
-		__entry->from_ifindex	= from_dev->ifindex;
-		__entry->to_ifindex	= to_dev->ifindex;
 		__entry->err		= err;
 	),
 
 	TP_printk("ndo_xdp_xmit"
-		  " map_id=%d map_index=%d action=%s"
+		  " from_ifindex=%d to_ifindex=%d action=%s"
 		  " sent=%d drops=%d"
-		  " from_ifindex=%d to_ifindex=%d err=%d",
-		  __entry->map_id, __entry->map_index,
+		  " err=%d",
+		  __entry->from_ifindex, __entry->to_ifindex,
 		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
 		  __entry->sent, __entry->drops,
-		  __entry->from_ifindex, __entry->to_ifindex, __entry->err)
+		  __entry->err)
 );
 
 /* Expect users already include <net/xdp.h>, but not xdp_priv.h */
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index db32272c4f77..1b4bfe4e06d6 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -340,7 +340,7 @@ static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 out:
 	bq->count = 0;
 
-	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);
+	trace_xdp_devmap_xmit(bq->dev_rx, dev, sent, drops, err);
 	bq->dev_rx = NULL;
 	__list_del_clearprev(&bq->flush_node);
 	return 0;
diff --git a/samples/bpf/xdp_monitor_kern.c b/samples/bpf/xdp_monitor_kern.c
index ad10fe700d7d..39458a44472e 100644
--- a/samples/bpf/xdp_monitor_kern.c
+++ b/samples/bpf/xdp_monitor_kern.c
@@ -222,14 +222,12 @@ struct bpf_map_def SEC("maps") devmap_xmit_cnt = {
  */
 struct devmap_xmit_ctx {
 	u64 __pad;		// First 8 bytes are not accessible by bpf code
-	int map_id;		//	offset:8;  size:4; signed:1;
+	int from_ifindex;	//	offset:8;  size:4; signed:1;
 	u32 act;		//	offset:12; size:4; signed:0;
-	u32 map_index;		//	offset:16; size:4; signed:0;
+	int to_ifindex; 	//	offset:16; size:4; signed:1;
 	int drops;		//	offset:20; size:4; signed:1;
 	int sent;		//	offset:24; size:4; signed:1;
-	int from_ifindex;	//	offset:28; size:4; signed:1;
-	int to_ifindex;		//	offset:32; size:4; signed:1;
-	int err;		//	offset:36; size:4; signed:1;
+	int err;		//	offset:28; size:4; signed:1;
 };
 
 SEC("tracepoint/xdp/xdp_devmap_xmit")

Re: [PATCH bpf-next v2 1/2] xdp: Move devmap bulk queue into struct net_device

[Toke Høiland-Jørgensen]

Jesper Dangaard Brouer <brouer@redhat.com> writes:

> On Wed, 15 Jan 2020 23:11:21 +0100
> Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
>> Jesper Dangaard Brouer <brouer@redhat.com> writes:
>> 
>> > On Mon, 13 Jan 2020 19:10:55 +0100
>> > Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>> >  
>> >> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
>> >> index da9c832fc5c8..030d125c3839 100644
>> >> --- a/kernel/bpf/devmap.c
>> >> +++ b/kernel/bpf/devmap.c  
>> > [...]  
>> >> @@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
>> >>  out:
>> >>  	bq->count = 0;
>> >>  
>> >> -	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
>> >> -			      sent, drops, bq->dev_rx, dev, err);
>> >> +	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);  
>> >
>> > Hmm ... I don't like that we lose the map_id and map_index identifier.
>> > This is part of our troubleshooting interface.  
>> 
>> Hmm, I guess I can take another look at whether there's a way to avoid
>> that. Any ideas?
>
> Looking at the code and the other tracepoints...
>
> I will actually suggest to remove these two arguments, because the
> trace_xdp_redirect_map tracepoint also contains the ifindex'es, and to
> troubleshoot people can record both tracepoints and do the correlation
> themselves.
>
> When changing the tracepoint I would like to keep member 'drops' and
> 'sent' at the same struct offsets.  As our xdp_monitor example reads
> these and I hope we can kept it working this way.
>
> I've coded it up, and tested it.  The new xdp_monitor will work on
> older kernels, but the old xdp_monitor will fail attaching on newer
> kernels. I think this is fair enough, as we are backwards compatible.

SGTM - thanks! I'll respin and include this :)

-Toke