From e086101b150ae8e99e54ab26101ef3835fa9f48d Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Fri, 13 Oct 2017 13:03:16 -0700
Subject: tcp: add a tracepoint for tcp retransmission

We need a real-time notification for tcp retransmission
for monitoring.

Of course we could use ftrace to dynamically instrument this
kernel function too, however we can't retrieve the connection
information at the same time, for example perf-tools [1] reads
/proc/net/tcp for socket details, which is slow when we have
a lots of connections.

Therefore, this patch adds a tracepoint for __tcp_retransmit_skb()
and exposes src/dst IP addresses and ports of the connection.
This also makes it easier to integrate into perf.

Note, I expose both IPv4 and IPv6 addresses at the same time:
for a IPv4 socket, v4 mapped address is used as IPv6 addresses,
for a IPv6 socket, LOOPBACK4_IPV6 is already filled by kernel.
Also, add sk and skb pointers as they are useful for BPF.

1. https://github.com/brendangregg/perf-tools/blob/master/net/tcpretrans

Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Brendan Gregg <bgregg@netflix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 68 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 68 insertions(+)
 create mode 100644 include/trace/events/tcp.h

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
new file mode 100644
index 0000000000000..3d1cbd072b7e5
--- /dev/null
+++ b/include/trace/events/tcp.h
@@ -0,0 +1,68 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM tcp
+
+#if !defined(_TRACE_TCP_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_TCP_H
+
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
+#include <linux/tracepoint.h>
+#include <net/ipv6.h>
+
+TRACE_EVENT(tcp_retransmit_skb,
+
+	TP_PROTO(struct sock *sk, struct sk_buff *skb),
+
+	TP_ARGS(sk, skb),
+
+	TP_STRUCT__entry(
+		__field(void *, skbaddr)
+		__field(void *, skaddr)
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__array(__u8, saddr, 4)
+		__array(__u8, daddr, 4)
+		__array(__u8, saddr_v6, 16)
+		__array(__u8, daddr_v6, 16)
+	),
+
+	TP_fast_assign(
+		struct ipv6_pinfo *np = inet6_sk(sk);
+		struct inet_sock *inet = inet_sk(sk);
+		struct in6_addr *pin6;
+		__be32 *p32;
+
+		__entry->skbaddr = skb;
+		__entry->skaddr = sk;
+
+		__entry->sport = ntohs(inet->inet_sport);
+		__entry->dport = ntohs(inet->inet_dport);
+
+		p32 = (__be32 *) __entry->saddr;
+		*p32 = inet->inet_saddr;
+
+		p32 = (__be32 *) __entry->daddr;
+		*p32 =  inet->inet_daddr;
+
+		if (np) {
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			*pin6 = np->saddr;
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			*pin6 = *(np->daddr_cache);
+		} else {
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_daddr, pin6);
+		}
+	),
+
+	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6 daddrv6=%pI6",
+		  __entry->sport, __entry->dport, __entry->saddr, __entry->daddr,
+		  __entry->saddr_v6, __entry->daddr_v6)
+);
+
+#endif /* _TRACE_TCP_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
-- 
cgit v1.2.3


From 9185a610f8f7f1b4e4d28c9de27d1969cf58e0f1 Mon Sep 17 00:00:00 2001
From: "Steven Rostedt (VMware)" <rostedt@goodmis.org>
Date: Thu, 12 Oct 2017 18:40:02 -0400
Subject: tracing: bpf: Hide bpf trace events when they are not used

All the trace events defined in include/trace/events/bpf.h are only
used when CONFIG_BPF_SYSCALL is defined. But this file gets included by
include/linux/bpf_trace.h which is included by the networking code with
CREATE_TRACE_POINTS defined.

If a trace event is created but not used it still has data structures
and functions created for its use, even though nothing is using them.
To not waste space, do not define the BPF trace events in bpf.h unless
CONFIG_BPF_SYSCALL is defined.

Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/bpf.h | 5 ++++-
 kernel/bpf/core.c          | 3 +++
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include/trace')

diff --git a/include/trace/events/bpf.h b/include/trace/events/bpf.h
index 52c8425d144b3..1fb58faa4a44b 100644
--- a/include/trace/events/bpf.h
+++ b/include/trace/events/bpf.h
@@ -4,6 +4,9 @@
 #if !defined(_TRACE_BPF_H) || defined(TRACE_HEADER_MULTI_READ)
 #define _TRACE_BPF_H
 
+/* These are only used within the BPF_SYSCALL code */
+#ifdef CONFIG_BPF_SYSCALL
+
 #include <linux/filter.h>
 #include <linux/bpf.h>
 #include <linux/fs.h>
@@ -345,7 +348,7 @@ TRACE_EVENT(bpf_map_next_key,
 		  __print_hex(__get_dynamic_array(nxt), __entry->key_len),
 		  __entry->key_trunc ? " ..." : "")
 );
-
+#endif /* CONFIG_BPF_SYSCALL */
 #endif /* _TRACE_BPF_H */
 
 #include <trace/define_trace.h>
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 248961af2421d..8e7c8bf2b687d 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1580,5 +1580,8 @@ int __weak skb_copy_bits(const struct sk_buff *skb, int offset, void *to,
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(xdp_exception);
 
+/* These are only used within the BPF_SYSCALL code */
+#ifdef CONFIG_BPF_SYSCALL
 EXPORT_TRACEPOINT_SYMBOL_GPL(bpf_prog_get_type);
 EXPORT_TRACEPOINT_SYMBOL_GPL(bpf_prog_put_rcu);
+#endif
-- 
cgit v1.2.3


From 9c270af37bb62e708e3e4415d653ce73e713df02 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 16 Oct 2017 12:19:34 +0200
Subject: bpf: XDP_REDIRECT enable use of cpumap

This patch connects cpumap to the xdp_do_redirect_map infrastructure.

Still no SKB allocation are done yet.  The XDP frames are transferred
to the other CPU, but they are simply refcnt decremented on the remote
CPU.  This served as a good benchmark for measuring the overhead of
remote refcnt decrement.  If driver page recycle cache is not
efficient then this, exposes a bottleneck in the page allocator.

A shout-out to MST's ptr_ring, which is the secret behind is being so
efficient to transfer memory pointers between CPUs, without constantly
bouncing cache-lines between CPUs.

V3: Handle !CONFIG_BPF_SYSCALL pointed out by kbuild test robot.

V4: Make Generic-XDP aware of cpumap type, but don't allow redirect yet,
 as implementation require a separate upstream discussion.

V5:
 - Fix a maybe-uninitialized pointed out by kbuild test robot.
 - Restrict bpf-prog side access to cpumap, open when use-cases appear
 - Implement cpu_map_enqueue() as a more simple void pointer enqueue

V6:
 - Allow cpumap type for usage in helper bpf_redirect_map,
   general bpf-prog side restriction moved to earlier patch.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/bpf.h        |  31 +++++++++-
 include/trace/events/xdp.h |  10 +++-
 kernel/bpf/cpumap.c        |  22 ++++++-
 kernel/bpf/verifier.c      |   3 +-
 net/core/filter.c          | 140 +++++++++++++++++++++++++++++++++++----------
 5 files changed, 172 insertions(+), 34 deletions(-)

(limited to 'include/trace')

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 4373125de1f33..6d4dd844828a1 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -355,6 +355,13 @@ struct net_device  *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 void __dev_map_insert_ctx(struct bpf_map *map, u32 index);
 void __dev_map_flush(struct bpf_map *map);
 
+struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
+void __cpu_map_insert_ctx(struct bpf_map *map, u32 index);
+void __cpu_map_flush(struct bpf_map *map);
+struct xdp_buff;
+int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
+		    struct net_device *dev_rx);
+
 /* Return map's numa specified by userspace */
 static inline int bpf_map_attr_numa_node(const union bpf_attr *attr)
 {
@@ -362,7 +369,7 @@ static inline int bpf_map_attr_numa_node(const union bpf_attr *attr)
 		attr->numa_node : NUMA_NO_NODE;
 }
 
-#else
+#else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
 	return ERR_PTR(-EOPNOTSUPP);
@@ -425,6 +432,28 @@ static inline void __dev_map_insert_ctx(struct bpf_map *map, u32 index)
 static inline void __dev_map_flush(struct bpf_map *map)
 {
 }
+
+static inline
+struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
+{
+	return NULL;
+}
+
+static inline void __cpu_map_insert_ctx(struct bpf_map *map, u32 index)
+{
+}
+
+static inline void __cpu_map_flush(struct bpf_map *map)
+{
+}
+
+struct xdp_buff;
+static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu,
+				  struct xdp_buff *xdp,
+				  struct net_device *dev_rx)
+{
+	return 0;
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 #if defined(CONFIG_STREAM_PARSER) && defined(CONFIG_BPF_SYSCALL)
diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index 4e16c43fba10a..eb2ece96c1a2b 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -136,12 +136,18 @@ DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map_err,
 		  __entry->map_id, __entry->map_index)
 );
 
+#define devmap_ifindex(fwd, map)				\
+	(!fwd ? 0 :						\
+	 (!map ? 0 :						\
+	  ((map->map_type == BPF_MAP_TYPE_DEVMAP) ?		\
+	   ((struct net_device *)fwd)->ifindex : 0)))
+
 #define _trace_xdp_redirect_map(dev, xdp, fwd, map, idx)		\
-	 trace_xdp_redirect_map(dev, xdp, fwd ? fwd->ifindex : 0,	\
+	 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, fwd ? fwd->ifindex : 0,	\
+	 trace_xdp_redirect_map_err(dev, xdp, devmap_ifindex(fwd, map),	\
 				    err, map, idx)
 
 #endif /* _TRACE_XDP_H */
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index e1e25ddba038d..768da6a2c2655 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -500,7 +500,7 @@ struct xdp_pkt {
 /* Runs under RCU-read-side, plus in softirq under NAPI protection.
  * Thus, safe percpu variable access.
  */
-int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_pkt *xdp_pkt)
+static int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_pkt *xdp_pkt)
 {
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
@@ -520,6 +520,26 @@ int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_pkt *xdp_pkt)
 	return 0;
 }
 
+int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	struct xdp_pkt *xdp_pkt;
+	int headroom;
+
+	/* For now this is just used as a void pointer to data_hard_start.
+	 * Followup patch will generalize this.
+	 */
+	xdp_pkt = xdp->data_hard_start;
+
+	/* Fake writing into xdp_pkt->data to measure overhead */
+	headroom = xdp->data - xdp->data_hard_start;
+	if (headroom < sizeof(*xdp_pkt))
+		xdp_pkt->data = xdp->data;
+
+	bq_enqueue(rcpu, xdp_pkt);
+	return 0;
+}
+
 void __cpu_map_insert_ctx(struct bpf_map *map, u32 bit)
 {
 	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index cefa64be9a2fc..e4d5136725a2f 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1486,7 +1486,8 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 			goto error;
 		break;
 	case BPF_FUNC_redirect_map:
-		if (map->map_type != BPF_MAP_TYPE_DEVMAP)
+		if (map->map_type != BPF_MAP_TYPE_DEVMAP &&
+		    map->map_type != BPF_MAP_TYPE_CPUMAP)
 			goto error;
 		break;
 	case BPF_FUNC_sk_redirect_map:
diff --git a/net/core/filter.c b/net/core/filter.c
index 140fa9f9c0f41..4d88e0665c413 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2526,10 +2526,36 @@ static int __bpf_tx_xdp(struct net_device *dev,
 	err = dev->netdev_ops->ndo_xdp_xmit(dev, xdp);
 	if (err)
 		return err;
-	if (map)
+	dev->netdev_ops->ndo_xdp_flush(dev);
+	return 0;
+}
+
+static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
+			    struct bpf_map *map,
+			    struct xdp_buff *xdp,
+			    u32 index)
+{
+	int err;
+
+	if (map->map_type == BPF_MAP_TYPE_DEVMAP) {
+		struct net_device *dev = fwd;
+
+		if (!dev->netdev_ops->ndo_xdp_xmit)
+			return -EOPNOTSUPP;
+
+		err = dev->netdev_ops->ndo_xdp_xmit(dev, xdp);
+		if (err)
+			return err;
 		__dev_map_insert_ctx(map, index);
-	else
-		dev->netdev_ops->ndo_xdp_flush(dev);
+
+	} else if (map->map_type == BPF_MAP_TYPE_CPUMAP) {
+		struct bpf_cpu_map_entry *rcpu = fwd;
+
+		err = cpu_map_enqueue(rcpu, xdp, dev_rx);
+		if (err)
+			return err;
+		__cpu_map_insert_ctx(map, index);
+	}
 	return 0;
 }
 
@@ -2539,11 +2565,33 @@ void xdp_do_flush_map(void)
 	struct bpf_map *map = ri->map_to_flush;
 
 	ri->map_to_flush = NULL;
-	if (map)
-		__dev_map_flush(map);
+	if (map) {
+		switch (map->map_type) {
+		case BPF_MAP_TYPE_DEVMAP:
+			__dev_map_flush(map);
+			break;
+		case BPF_MAP_TYPE_CPUMAP:
+			__cpu_map_flush(map);
+			break;
+		default:
+			break;
+		}
+	}
 }
 EXPORT_SYMBOL_GPL(xdp_do_flush_map);
 
+static void *__xdp_map_lookup_elem(struct bpf_map *map, u32 index)
+{
+	switch (map->map_type) {
+	case BPF_MAP_TYPE_DEVMAP:
+		return __dev_map_lookup_elem(map, index);
+	case BPF_MAP_TYPE_CPUMAP:
+		return __cpu_map_lookup_elem(map, index);
+	default:
+		return NULL;
+	}
+}
+
 static inline bool xdp_map_invalid(const struct bpf_prog *xdp_prog,
 				   unsigned long aux)
 {
@@ -2556,8 +2604,8 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	struct redirect_info *ri = this_cpu_ptr(&redirect_info);
 	unsigned long map_owner = ri->map_owner;
 	struct bpf_map *map = ri->map;
-	struct net_device *fwd = NULL;
 	u32 index = ri->ifindex;
+	void *fwd = NULL;
 	int err;
 
 	ri->ifindex = 0;
@@ -2570,7 +2618,7 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 		goto err;
 	}
 
-	fwd = __dev_map_lookup_elem(map, index);
+	fwd = __xdp_map_lookup_elem(map, index);
 	if (!fwd) {
 		err = -EINVAL;
 		goto err;
@@ -2578,7 +2626,7 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	if (ri->map_to_flush && ri->map_to_flush != map)
 		xdp_do_flush_map();
 
-	err = __bpf_tx_xdp(fwd, map, xdp, index);
+	err = __bpf_tx_xdp_map(dev, fwd, map, xdp, index);
 	if (unlikely(err))
 		goto err;
 
@@ -2620,54 +2668,88 @@ err:
 }
 EXPORT_SYMBOL_GPL(xdp_do_redirect);
 
-int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb,
-			    struct bpf_prog *xdp_prog)
+static int __xdp_generic_ok_fwd_dev(struct sk_buff *skb, struct net_device *fwd)
+{
+	unsigned int len;
+
+	if (unlikely(!(fwd->flags & IFF_UP)))
+		return -ENETDOWN;
+
+	len = fwd->mtu + fwd->hard_header_len + VLAN_HLEN;
+	if (skb->len > len)
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+int xdp_do_generic_redirect_map(struct net_device *dev, struct sk_buff *skb,
+				struct bpf_prog *xdp_prog)
 {
 	struct redirect_info *ri = this_cpu_ptr(&redirect_info);
 	unsigned long map_owner = ri->map_owner;
 	struct bpf_map *map = ri->map;
 	struct net_device *fwd = NULL;
 	u32 index = ri->ifindex;
-	unsigned int len;
 	int err = 0;
 
 	ri->ifindex = 0;
 	ri->map = NULL;
 	ri->map_owner = 0;
 
-	if (map) {
-		if (unlikely(xdp_map_invalid(xdp_prog, map_owner))) {
-			err = -EFAULT;
-			map = NULL;
-			goto err;
-		}
-		fwd = __dev_map_lookup_elem(map, index);
-	} else {
-		fwd = dev_get_by_index_rcu(dev_net(dev), index);
+	if (unlikely(xdp_map_invalid(xdp_prog, map_owner))) {
+		err = -EFAULT;
+		map = NULL;
+		goto err;
 	}
+	fwd = __xdp_map_lookup_elem(map, index);
 	if (unlikely(!fwd)) {
 		err = -EINVAL;
 		goto err;
 	}
 
-	if (unlikely(!(fwd->flags & IFF_UP))) {
-		err = -ENETDOWN;
+	if (map->map_type == BPF_MAP_TYPE_DEVMAP) {
+		if (unlikely((err = __xdp_generic_ok_fwd_dev(skb, fwd))))
+			goto err;
+		skb->dev = fwd;
+	} else {
+		/* TODO: Handle BPF_MAP_TYPE_CPUMAP */
+		err = -EBADRQC;
 		goto err;
 	}
 
-	len = fwd->mtu + fwd->hard_header_len + VLAN_HLEN;
-	if (skb->len > len) {
-		err = -EMSGSIZE;
+	_trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index);
+	return 0;
+err:
+	_trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err);
+	return err;
+}
+
+int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb,
+			    struct bpf_prog *xdp_prog)
+{
+	struct redirect_info *ri = this_cpu_ptr(&redirect_info);
+	u32 index = ri->ifindex;
+	struct net_device *fwd;
+	int err = 0;
+
+	if (ri->map)
+		return xdp_do_generic_redirect_map(dev, skb, xdp_prog);
+
+	ri->ifindex = 0;
+	fwd = dev_get_by_index_rcu(dev_net(dev), index);
+	if (unlikely(!fwd)) {
+		err = -EINVAL;
 		goto err;
 	}
 
+	if (unlikely((err = __xdp_generic_ok_fwd_dev(skb, fwd))))
+		goto err;
+
 	skb->dev = fwd;
-	map ? _trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index)
-		: _trace_xdp_redirect(dev, xdp_prog, index);
+	_trace_xdp_redirect(dev, xdp_prog, index);
 	return 0;
 err:
-	map ? _trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err)
-		: _trace_xdp_redirect_err(dev, xdp_prog, index, err);
+	_trace_xdp_redirect_err(dev, xdp_prog, index, err);
 	return err;
 }
 EXPORT_SYMBOL_GPL(xdp_do_generic_redirect);
-- 
cgit v1.2.3


From f9419f7bd7a5318b636a941a0214c5cdfa6f6530 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 16 Oct 2017 12:19:44 +0200
Subject: bpf: cpumap add tracepoints

This adds two tracepoint to the cpumap.  One for the enqueue side
trace_xdp_cpumap_enqueue() and one for the kthread dequeue side
trace_xdp_cpumap_kthread().

To mitigate the tracepoint overhead, these are invoked during the
enqueue/dequeue bulking phases, thus amortizing the cost.

The obvious use-cases are for debugging and monitoring.  The
non-intuitive use-case is using these as a feedback loop to know the
system load.  One can imagine auto-scaling by reducing, adding or
activating more worker CPUs on demand.

V4: tracepoint remove time_limit info, instead add sched info

V8: intro struct bpf_cpu_map_entry members cpu+map_id in this patch

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/xdp.h | 70 ++++++++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/cpumap.c        | 24 ++++++++++++----
 2 files changed, 89 insertions(+), 5 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index eb2ece96c1a2b..0c8dec61987e0 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -150,6 +150,76 @@ DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map_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,
+		 int sched),
+
+	TP_ARGS(map_id, processed, drops, sched),
+
+	TP_STRUCT__entry(
+		__field(int, map_id)
+		__field(u32, act)
+		__field(int, cpu)
+		__field(unsigned int, drops)
+		__field(unsigned int, processed)
+		__field(int, sched)
+	),
+
+	TP_fast_assign(
+		__entry->map_id		= map_id;
+		__entry->act		= XDP_REDIRECT;
+		__entry->cpu		= smp_processor_id();
+		__entry->drops		= drops;
+		__entry->processed	= processed;
+		__entry->sched	= sched;
+	),
+
+	TP_printk("kthread"
+		  " cpu=%d map_id=%d action=%s"
+		  " processed=%u drops=%u"
+		  " sched=%d",
+		  __entry->cpu, __entry->map_id,
+		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
+		  __entry->processed, __entry->drops,
+		  __entry->sched)
+);
+
+TRACE_EVENT(xdp_cpumap_enqueue,
+
+	TP_PROTO(int map_id, unsigned int processed,  unsigned int drops,
+		 int to_cpu),
+
+	TP_ARGS(map_id, processed, drops, to_cpu),
+
+	TP_STRUCT__entry(
+		__field(int, map_id)
+		__field(u32, act)
+		__field(int, cpu)
+		__field(unsigned int, drops)
+		__field(unsigned int, processed)
+		__field(int, to_cpu)
+	),
+
+	TP_fast_assign(
+		__entry->map_id		= map_id;
+		__entry->act		= XDP_REDIRECT;
+		__entry->cpu		= smp_processor_id();
+		__entry->drops		= drops;
+		__entry->processed	= processed;
+		__entry->to_cpu		= to_cpu;
+	),
+
+	TP_printk("enqueue"
+		  " cpu=%d map_id=%d action=%s"
+		  " processed=%u drops=%u"
+		  " to_cpu=%d",
+		  __entry->cpu, __entry->map_id,
+		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
+		  __entry->processed, __entry->drops,
+		  __entry->to_cpu)
+);
+
 #endif /* _TRACE_XDP_H */
 
 #include <trace/define_trace.h>
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index ee7adf4352dd4..b4358d84ddf19 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -24,6 +24,7 @@
 #include <linux/workqueue.h>
 #include <linux/kthread.h>
 #include <linux/capability.h>
+#include <trace/events/xdp.h>
 
 #include <linux/netdevice.h>   /* netif_receive_skb_core */
 #include <linux/etherdevice.h> /* eth_type_trans */
@@ -43,6 +44,8 @@ struct xdp_bulk_queue {
 
 /* Struct for every remote "destination" CPU in map */
 struct bpf_cpu_map_entry {
+	u32 cpu;    /* kthread CPU and map index */
+	int map_id; /* Back reference to map */
 	u32 qsize;  /* Queue size placeholder for map lookup */
 
 	/* XDP can run multiple RX-ring queues, need __percpu enqueue store */
@@ -280,15 +283,16 @@ static int cpu_map_kthread_run(void *data)
 	 * kthread_stop signal until queue is empty.
 	 */
 	while (!kthread_should_stop() || !__ptr_ring_empty(rcpu->queue)) {
-		unsigned int processed = 0, drops = 0;
+		unsigned int processed = 0, drops = 0, sched = 0;
 		struct xdp_pkt *xdp_pkt;
 
 		/* Release CPU reschedule checks */
 		if (__ptr_ring_empty(rcpu->queue)) {
 			__set_current_state(TASK_INTERRUPTIBLE);
 			schedule();
+			sched = 1;
 		} else {
-			cond_resched();
+			sched = cond_resched();
 		}
 		__set_current_state(TASK_RUNNING);
 
@@ -318,6 +322,9 @@ static int cpu_map_kthread_run(void *data)
 			if (++processed == 8)
 				break;
 		}
+		/* Feedback loop via tracepoint */
+		trace_xdp_cpumap_kthread(rcpu->map_id, processed, drops, sched);
+
 		local_bh_enable(); /* resched point, may call do_softirq() */
 	}
 	__set_current_state(TASK_RUNNING);
@@ -354,7 +361,9 @@ struct bpf_cpu_map_entry *__cpu_map_entry_alloc(u32 qsize, u32 cpu, int map_id)
 	if (err)
 		goto free_queue;
 
-	rcpu->qsize = qsize;
+	rcpu->cpu    = cpu;
+	rcpu->map_id = map_id;
+	rcpu->qsize  = qsize;
 
 	/* Setup kthread */
 	rcpu->kthread = kthread_create_on_node(cpu_map_kthread_run, rcpu, numa,
@@ -584,6 +593,8 @@ const struct bpf_map_ops cpu_map_ops = {
 static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
 			     struct xdp_bulk_queue *bq)
 {
+	unsigned int processed = 0, drops = 0;
+	const int to_cpu = rcpu->cpu;
 	struct ptr_ring *q;
 	int i;
 
@@ -599,13 +610,16 @@ static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
 
 		err = __ptr_ring_produce(q, xdp_pkt);
 		if (err) {
-			/* Free xdp_pkt */
-			page_frag_free(xdp_pkt);
+			drops++;
+			page_frag_free(xdp_pkt); /* Free xdp_pkt */
 		}
+		processed++;
 	}
 	bq->count = 0;
 	spin_unlock(&q->producer_lock);
 
+	/* Feedback loop via tracepoints */
+	trace_xdp_cpumap_enqueue(rcpu->map_id, processed, drops, to_cpu);
 	return 0;
 }
 
-- 
cgit v1.2.3


From fb6ff75e18937a20dbec1eb47b5f893f38eabae4 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 16 Oct 2017 14:24:02 -0700
Subject: tcp: Use pI6c in tcp tracepoint

The compact form for IPv6 addresses is more user friendly than the full
version. For example:
   compact: 2001:db8:1::1
      full: 2001:0db8:0001:0000:0000:0000:0000:0004i

Update the tcp tracepoint to show the compact form.

Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 3d1cbd072b7e5..1ffab6d96e943 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -57,7 +57,7 @@ TRACE_EVENT(tcp_retransmit_skb,
 		}
 	),
 
-	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6 daddrv6=%pI6",
+	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c",
 		  __entry->sport, __entry->dport, __entry->saddr, __entry->daddr,
 		  __entry->saddr_v6, __entry->daddr_v6)
 );
-- 
cgit v1.2.3


From 386fd5da401dc6c4b0ab6a54d333609876b699fe Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 16 Oct 2017 15:32:07 -0700
Subject: tcp: Check daddr_cache before use in tracepoint

Running perf in one window to capture tcp_retransmit_skb tracepoint:
    $ perf record -e tcp:tcp_retransmit_skb -a

And causing a retransmission on an active TCP session (e.g., dropping
packets in the receiver, changing MTU on the interface to 500 and back
to 1500) triggers a panic:

[   58.543144] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
[   58.545300] IP: perf_trace_tcp_retransmit_skb+0xd0/0x145
[   58.546770] PGD 0 P4D 0
[   58.547472] Oops: 0000 [#1] SMP
[   58.548328] Modules linked in: vrf
[   58.549262] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.14.0-rc4+ #26
[   58.551004] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014
[   58.554560] task: ffffffff81a0e540 task.stack: ffffffff81a00000
[   58.555817] RIP: 0010:perf_trace_tcp_retransmit_skb+0xd0/0x145
[   58.557137] RSP: 0018:ffff88003fc03d68 EFLAGS: 00010282
[   58.558292] RAX: 0000000000000000 RBX: ffffe8ffffc0ec80 RCX: ffff880038543098
[   58.559850] RDX: 0400000000000000 RSI: ffff88003fc03d70 RDI: ffff88003fc14b68
[   58.561099] RBP: ffff88003fc03da8 R08: 0000000000000000 R09: ffffea0000d3224a
[   58.562005] R10: ffff88003fc03db8 R11: 0000000000000010 R12: ffff8800385428c0
[   58.562930] R13: ffffe8ffffc0e478 R14: ffffffff81a93a40 R15: ffff88003d4f0c00
[   58.563845] FS:  0000000000000000(0000) GS:ffff88003fc00000(0000) knlGS:0000000000000000
[   58.564873] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   58.565613] CR2: 0000000000000008 CR3: 000000003d68f004 CR4: 00000000000606f0
[   58.566538] Call Trace:
[   58.566865]  <IRQ>
[   58.567140]  __tcp_retransmit_skb+0x4ab/0x4c6
[   58.567704]  ? tcp_set_ca_state+0x22/0x3f
[   58.568231]  tcp_retransmit_skb+0x14/0xa3
[   58.568754]  tcp_retransmit_timer+0x472/0x5e3
[   58.569324]  ? tcp_write_timer_handler+0x1e9/0x1e9
[   58.569946]  tcp_write_timer_handler+0x95/0x1e9
[   58.570548]  tcp_write_timer+0x2a/0x58

Check that daddr_cache is non-NULL before de-referencing.

Fixes: e086101b150a ("tcp: add a tracepoint for tcp retransmission")
Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 1ffab6d96e943..f51c130f1e0fc 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -27,7 +27,6 @@ TRACE_EVENT(tcp_retransmit_skb,
 	),
 
 	TP_fast_assign(
-		struct ipv6_pinfo *np = inet6_sk(sk);
 		struct inet_sock *inet = inet_sk(sk);
 		struct in6_addr *pin6;
 		__be32 *p32;
@@ -44,11 +43,12 @@ TRACE_EVENT(tcp_retransmit_skb,
 		p32 = (__be32 *) __entry->daddr;
 		*p32 =  inet->inet_daddr;
 
-		if (np) {
+		/* IPv6 socket ? */
+		if (inet6_sk(sk)) {
 			pin6 = (struct in6_addr *)__entry->saddr_v6;
-			*pin6 = np->saddr;
+			*pin6 = sk->sk_v6_rcv_saddr;
 			pin6 = (struct in6_addr *)__entry->daddr_v6;
-			*pin6 = *(np->daddr_cache);
+			*pin6 = sk->sk_v6_daddr;
 		} else {
 			pin6 = (struct in6_addr *)__entry->saddr_v6;
 			ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
-- 
cgit v1.2.3


From 890056783c60ad9d0789774af2bc10fe4f27dd9d Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Wed, 18 Oct 2017 08:17:29 -0700
Subject: tcp: Remove use of inet6_sk and add IPv6 checks to tracepoint

386fd5da401d ("tcp: Check daddr_cache before use in tracepoint") was the
second version of the tracepoint fixup patch. This patch is the delta
between v2 and v3.  Specifically, remove the use of inet6_sk and check
sk_family as requested by Eric and add IS_ENABLED(CONFIG_IPV6) around
the use of sk_v6_rcv_saddr and sk_v6_daddr as done in sock_common (noted
by Cong).

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index f51c130f1e0fc..c3220d9144752 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -43,13 +43,15 @@ TRACE_EVENT(tcp_retransmit_skb,
 		p32 = (__be32 *) __entry->daddr;
 		*p32 =  inet->inet_daddr;
 
-		/* IPv6 socket ? */
-		if (inet6_sk(sk)) {
+#if IS_ENABLED(CONFIG_IPV6)
+		if (sk->sk_family == AF_INET6) {
 			pin6 = (struct in6_addr *)__entry->saddr_v6;
 			*pin6 = sk->sk_v6_rcv_saddr;
 			pin6 = (struct in6_addr *)__entry->daddr_v6;
 			*pin6 = sk->sk_v6_daddr;
-		} else {
+		} else
+#endif
+		{
 			pin6 = (struct in6_addr *)__entry->saddr_v6;
 			ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
 			pin6 = (struct in6_addr *)__entry->daddr_v6;
-- 
cgit v1.2.3


From b65f164d37cf6d4aac59b0e13c2e5c4cfe293fd2 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Thu, 19 Oct 2017 09:31:43 +0200
Subject: ipv6: let trace_fib6_table_lookup() dereference the fib table

The perf traces for ipv6 routing code show a relevant cost around
trace_fib6_table_lookup(), even if no trace is enabled. This is
due to the fib6_table de-referencing currently performed by the
caller.

Let's the tracing code pay this overhead, passing to the trace
helper the table pointer. This gives small but measurable
performance improvement under UDP flood.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Acked-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/fib6.h |  6 +++---
 net/ipv6/route.c            | 12 ++++++------
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/fib6.h b/include/trace/events/fib6.h
index d60096cddb2a6..b34bed17abc70 100644
--- a/include/trace/events/fib6.h
+++ b/include/trace/events/fib6.h
@@ -12,9 +12,9 @@
 TRACE_EVENT(fib6_table_lookup,
 
 	TP_PROTO(const struct net *net, const struct rt6_info *rt,
-		 u32 tb_id, const struct flowi6 *flp),
+		 struct fib6_table *table, const struct flowi6 *flp),
 
-	TP_ARGS(net, rt, tb_id, flp),
+	TP_ARGS(net, rt, table, flp),
 
 	TP_STRUCT__entry(
 		__field(	u32,	tb_id		)
@@ -34,7 +34,7 @@ TRACE_EVENT(fib6_table_lookup,
 	TP_fast_assign(
 		struct in6_addr *in6;
 
-		__entry->tb_id = tb_id;
+		__entry->tb_id = table->tb6_id;
 		__entry->oif = flp->flowi6_oif;
 		__entry->iif = flp->flowi6_iif;
 		__entry->tos = ip6_tclass(flp->flowlabel);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 074fac966018b..46c59a53c53f3 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -945,7 +945,7 @@ restart:
 
 	rcu_read_unlock();
 
-	trace_fib6_table_lookup(net, rt, table->tb6_id, fl6);
+	trace_fib6_table_lookup(net, rt, table, fl6);
 
 	return rt;
 
@@ -1682,7 +1682,7 @@ redo_rt6_select:
 	if (rt == net->ipv6.ip6_null_entry) {
 		rcu_read_unlock();
 		dst_hold(&rt->dst);
-		trace_fib6_table_lookup(net, rt, table->tb6_id, fl6);
+		trace_fib6_table_lookup(net, rt, table, fl6);
 		return rt;
 	} else if (rt->rt6i_flags & RTF_CACHE) {
 		if (ip6_hold_safe(net, &rt, true)) {
@@ -1690,7 +1690,7 @@ redo_rt6_select:
 			rt6_dst_from_metrics_check(rt);
 		}
 		rcu_read_unlock();
-		trace_fib6_table_lookup(net, rt, table->tb6_id, fl6);
+		trace_fib6_table_lookup(net, rt, table, fl6);
 		return rt;
 	} else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) &&
 			    !(rt->rt6i_flags & RTF_GATEWAY))) {
@@ -1726,7 +1726,7 @@ redo_rt6_select:
 		}
 
 uncached_rt_out:
-		trace_fib6_table_lookup(net, uncached_rt, table->tb6_id, fl6);
+		trace_fib6_table_lookup(net, uncached_rt, table, fl6);
 		return uncached_rt;
 
 	} else {
@@ -1754,7 +1754,7 @@ uncached_rt_out:
 		}
 		local_bh_enable();
 		rcu_read_unlock();
-		trace_fib6_table_lookup(net, pcpu_rt, table->tb6_id, fl6);
+		trace_fib6_table_lookup(net, pcpu_rt, table, fl6);
 		return pcpu_rt;
 	}
 }
@@ -2195,7 +2195,7 @@ out:
 
 	rcu_read_unlock();
 
-	trace_fib6_table_lookup(net, rt, table->tb6_id, fl6);
+	trace_fib6_table_lookup(net, rt, table, fl6);
 	return rt;
 };
 
-- 
cgit v1.2.3


From f6e37b25413cf636369668652e9752ee77c7d9f7 Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:22 -0700
Subject: tcp: add trace event class tcp_event_sk_skb

Introduce event class tcp_event_sk_skb for tcp tracepoints that
have arguments sk and skb.

Existing tracepoint trace_tcp_retransmit_skb() falls into this class.
This patch rewrites the definition of trace_tcp_retransmit_skb() with
tcp_event_sk_skb.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index c3220d9144752..14b0a7083f1da 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -9,7 +9,13 @@
 #include <linux/tracepoint.h>
 #include <net/ipv6.h>
 
-TRACE_EVENT(tcp_retransmit_skb,
+/*
+ * tcp event with arguments sk and skb
+ *
+ * Note: this class requires a valid sk pointer; while skb pointer could
+ *       be NULL.
+ */
+DECLARE_EVENT_CLASS(tcp_event_sk_skb,
 
 	TP_PROTO(struct sock *sk, struct sk_buff *skb),
 
@@ -64,6 +70,13 @@ TRACE_EVENT(tcp_retransmit_skb,
 		  __entry->saddr_v6, __entry->daddr_v6)
 );
 
+DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb,
+
+	TP_PROTO(struct sock *sk, struct sk_buff *skb),
+
+	TP_ARGS(sk, skb)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
-- 
cgit v1.2.3


From 7344e29f285a94b965075599731811c352f3ab40 Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:23 -0700
Subject: tcp: mark trace event arguments sk and skb as const

Some functions that we plan to add trace points require const sk
and/or skb. So we mark these fields as const in the tracepoint.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 14b0a7083f1da..2b6fe72c67814 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -17,13 +17,13 @@
  */
 DECLARE_EVENT_CLASS(tcp_event_sk_skb,
 
-	TP_PROTO(struct sock *sk, struct sk_buff *skb),
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
 
 	TP_ARGS(sk, skb),
 
 	TP_STRUCT__entry(
-		__field(void *, skbaddr)
-		__field(void *, skaddr)
+		__field(const void *, skbaddr)
+		__field(const void *, skaddr)
 		__field(__u16, sport)
 		__field(__u16, dport)
 		__array(__u8, saddr, 4)
@@ -72,7 +72,7 @@ DECLARE_EVENT_CLASS(tcp_event_sk_skb,
 
 DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb,
 
-	TP_PROTO(struct sock *sk, struct sk_buff *skb),
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
 
 	TP_ARGS(sk, skb)
 );
-- 
cgit v1.2.3


From c24b14c46bb88d844275de5c4024c8745ae89d42 Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:24 -0700
Subject: tcp: add tracepoint trace_tcp_send_reset

New tracepoint trace_tcp_send_reset is added and called from
tcp_v4_send_reset(), tcp_v6_send_reset() and tcp_send_active_reset().

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 11 +++++++++++
 net/core/net-traces.c      |  2 ++
 net/ipv4/tcp_ipv4.c        |  6 +++++-
 net/ipv4/tcp_output.c      |  5 +++++
 net/ipv6/tcp_ipv6.c        | 10 ++++++++--
 5 files changed, 31 insertions(+), 3 deletions(-)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 2b6fe72c67814..3e57e1ae1c6b0 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -77,6 +77,17 @@ DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb,
 	TP_ARGS(sk, skb)
 );
 
+/*
+ * skb of trace_tcp_send_reset is the skb that caused RST. In case of
+ * active reset, skb should be NULL
+ */
+DEFINE_EVENT(tcp_event_sk_skb, tcp_send_reset,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+
+	TP_ARGS(sk, skb)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/core/net-traces.c b/net/core/net-traces.c
index f4e4fa2db5057..8dcd9b0be04a4 100644
--- a/net/core/net-traces.c
+++ b/net/core/net-traces.c
@@ -49,3 +49,5 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(br_fdb_update);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb);
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(napi_poll);
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(tcp_send_reset);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index e22439f05e46a..eb3f3b8e1e4b3 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -85,6 +85,8 @@
 #include <crypto/hash.h>
 #include <linux/scatterlist.h>
 
+#include <trace/events/tcp.h>
+
 #ifdef CONFIG_TCP_MD5SIG
 static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key,
 			       __be32 daddr, __be32 saddr, const struct tcphdr *th);
@@ -701,8 +703,10 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	 * routing might fail in this case. No choice here, if we choose to force
 	 * input interface, we will misroute in case of asymmetric route.
 	 */
-	if (sk)
+	if (sk) {
 		arg.bound_dev_if = sk->sk_bound_dev_if;
+		trace_tcp_send_reset(sk, skb);
+	}
 
 	BUILD_BUG_ON(offsetof(struct sock, sk_bound_dev_if) !=
 		     offsetof(struct inet_timewait_sock, tw_bound_dev_if));
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 988733f289c8c..1f01f4c9c7384 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3084,6 +3084,11 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 	/* Send it off. */
 	if (tcp_transmit_skb(sk, skb, 0, priority))
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED);
+
+	/* skb of trace_tcp_send_reset() keeps the skb that caused RST,
+	 * skb here is different to the troublesome skb, so use NULL
+	 */
+	trace_tcp_send_reset(sk, NULL);
 }
 
 /* Send a crossed SYN-ACK during socket establishment.
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index ae83615b7f6d5..0e2529958b52c 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -69,6 +69,8 @@
 #include <crypto/hash.h>
 #include <linux/scatterlist.h>
 
+#include <trace/events/tcp.h>
+
 static void	tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb);
 static void	tcp_v6_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 				      struct request_sock *req);
@@ -890,7 +892,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 	int genhash;
 	struct sock *sk1 = NULL;
 #endif
-	int oif;
+	int oif = 0;
 
 	if (th->rst)
 		return;
@@ -939,7 +941,11 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 		ack_seq = ntohl(th->seq) + th->syn + th->fin + skb->len -
 			  (th->doff << 2);
 
-	oif = sk ? sk->sk_bound_dev_if : 0;
+	if (sk) {
+		oif = sk->sk_bound_dev_if;
+		trace_tcp_send_reset(sk, skb);
+	}
+
 	tcp_v6_send_response(sk, skb, seq, ack_seq, 0, 0, 0, oif, key, 1, 0, 0);
 
 #ifdef CONFIG_TCP_MD5SIG
-- 
cgit v1.2.3


From 5941521c05d69cf3f2b1293eefd21207e083b70f Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:25 -0700
Subject: tcp: add tracepoint trace_tcp_receive_reset

New tracepoint trace_tcp_receive_reset is added and called from
tcp_reset(). This tracepoint is define with a new class tcp_event_sk.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 66 ++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_input.c       |  3 +++
 2 files changed, 69 insertions(+)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 3e57e1ae1c6b0..c83c71187719d 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -88,6 +88,72 @@ DEFINE_EVENT(tcp_event_sk_skb, tcp_send_reset,
 	TP_ARGS(sk, skb)
 );
 
+/*
+ * tcp event with arguments sk
+ *
+ * Note: this class requires a valid sk pointer.
+ */
+DECLARE_EVENT_CLASS(tcp_event_sk,
+
+	TP_PROTO(const struct sock *sk),
+
+	TP_ARGS(sk),
+
+	TP_STRUCT__entry(
+		__field(const void *, skaddr)
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__array(__u8, saddr, 4)
+		__array(__u8, daddr, 4)
+		__array(__u8, saddr_v6, 16)
+		__array(__u8, daddr_v6, 16)
+	),
+
+	TP_fast_assign(
+		struct inet_sock *inet = inet_sk(sk);
+		struct in6_addr *pin6;
+		__be32 *p32;
+
+		__entry->skaddr = sk;
+
+		__entry->sport = ntohs(inet->inet_sport);
+		__entry->dport = ntohs(inet->inet_dport);
+
+		p32 = (__be32 *) __entry->saddr;
+		*p32 = inet->inet_saddr;
+
+		p32 = (__be32 *) __entry->daddr;
+		*p32 =  inet->inet_daddr;
+
+#if IS_ENABLED(CONFIG_IPV6)
+		if (sk->sk_family == AF_INET6) {
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			*pin6 = sk->sk_v6_rcv_saddr;
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			*pin6 = sk->sk_v6_daddr;
+		} else
+#endif
+		{
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_daddr, pin6);
+		}
+	),
+
+	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c",
+		  __entry->sport, __entry->dport,
+		  __entry->saddr, __entry->daddr,
+		  __entry->saddr_v6, __entry->daddr_v6)
+);
+
+DEFINE_EVENT(tcp_event_sk, tcp_receive_reset,
+
+	TP_PROTO(const struct sock *sk),
+
+	TP_ARGS(sk)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index ab3f128982454..c5e64d4b5839a 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -75,6 +75,7 @@
 #include <linux/ipsec.h>
 #include <asm/unaligned.h>
 #include <linux/errqueue.h>
+#include <trace/events/tcp.h>
 
 int sysctl_tcp_fack __read_mostly;
 int sysctl_tcp_max_reordering __read_mostly = 300;
@@ -4010,6 +4011,8 @@ static inline bool tcp_sequence(const struct tcp_sock *tp, u32 seq, u32 end_seq)
 /* When we get a reset we do this. */
 void tcp_reset(struct sock *sk)
 {
+	trace_tcp_receive_reset(sk);
+
 	/* We want the right error as BSD sees it (and indeed as we do). */
 	switch (sk->sk_state) {
 	case TCP_SYN_SENT:
-- 
cgit v1.2.3


From e1a4aa50f47303ebb3ca0cfd01687884551ce03d Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:26 -0700
Subject: tcp: add tracepoint trace_tcp_destroy_sock

This patch adds trace event trace_tcp_destroy_sock.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 7 +++++++
 net/ipv4/tcp_ipv4.c        | 2 ++
 2 files changed, 9 insertions(+)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index c83c71187719d..1724c12c25cf0 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -154,6 +154,13 @@ DEFINE_EVENT(tcp_event_sk, tcp_receive_reset,
 	TP_ARGS(sk)
 );
 
+DEFINE_EVENT(tcp_event_sk, tcp_destroy_sock,
+
+	TP_PROTO(const struct sock *sk),
+
+	TP_ARGS(sk)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index eb3f3b8e1e4b3..23a8100af5ad3 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1869,6 +1869,8 @@ void tcp_v4_destroy_sock(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
+	trace_tcp_destroy_sock(sk);
+
 	tcp_clear_xmit_timers(sk);
 
 	tcp_cleanup_congestion_control(sk);
-- 
cgit v1.2.3


From e8fce23946b7e7eadf25ad78d8207c22903dfe27 Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 23 Oct 2017 09:20:27 -0700
Subject: tcp: add tracepoint trace_tcp_set_state()

This patch adds tracepoint trace_tcp_set_state. Besides usual fields
(s/d ports, IP addresses), old and new state of the socket is also
printed with TP_printk, with __print_symbolic().

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 76 ++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp.c             |  4 +++
 2 files changed, 80 insertions(+)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 1724c12c25cf0..03699ba716230 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -9,6 +9,22 @@
 #include <linux/tracepoint.h>
 #include <net/ipv6.h>
 
+#define tcp_state_name(state)	{ state, #state }
+#define show_tcp_state_name(val)			\
+	__print_symbolic(val,				\
+		tcp_state_name(TCP_ESTABLISHED),	\
+		tcp_state_name(TCP_SYN_SENT),		\
+		tcp_state_name(TCP_SYN_RECV),		\
+		tcp_state_name(TCP_FIN_WAIT1),		\
+		tcp_state_name(TCP_FIN_WAIT2),		\
+		tcp_state_name(TCP_TIME_WAIT),		\
+		tcp_state_name(TCP_CLOSE),		\
+		tcp_state_name(TCP_CLOSE_WAIT),		\
+		tcp_state_name(TCP_LAST_ACK),		\
+		tcp_state_name(TCP_LISTEN),		\
+		tcp_state_name(TCP_CLOSING),		\
+		tcp_state_name(TCP_NEW_SYN_RECV))
+
 /*
  * tcp event with arguments sk and skb
  *
@@ -161,6 +177,66 @@ DEFINE_EVENT(tcp_event_sk, tcp_destroy_sock,
 	TP_ARGS(sk)
 );
 
+TRACE_EVENT(tcp_set_state,
+
+	TP_PROTO(const struct sock *sk, const int oldstate, const int newstate),
+
+	TP_ARGS(sk, oldstate, newstate),
+
+	TP_STRUCT__entry(
+		__field(const void *, skaddr)
+		__field(int, oldstate)
+		__field(int, newstate)
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__array(__u8, saddr, 4)
+		__array(__u8, daddr, 4)
+		__array(__u8, saddr_v6, 16)
+		__array(__u8, daddr_v6, 16)
+	),
+
+	TP_fast_assign(
+		struct inet_sock *inet = inet_sk(sk);
+		struct in6_addr *pin6;
+		__be32 *p32;
+
+		__entry->skaddr = sk;
+		__entry->oldstate = oldstate;
+		__entry->newstate = newstate;
+
+		__entry->sport = ntohs(inet->inet_sport);
+		__entry->dport = ntohs(inet->inet_dport);
+
+		p32 = (__be32 *) __entry->saddr;
+		*p32 = inet->inet_saddr;
+
+		p32 = (__be32 *) __entry->daddr;
+		*p32 =  inet->inet_daddr;
+
+#if IS_ENABLED(CONFIG_IPV6)
+		if (sk->sk_family == AF_INET6) {
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			*pin6 = sk->sk_v6_rcv_saddr;
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			*pin6 = sk->sk_v6_daddr;
+		} else
+#endif
+		{
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			ipv6_addr_set_v4mapped(inet->inet_daddr, pin6);
+		}
+	),
+
+	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c oldstate=%s newstate=%s",
+		  __entry->sport, __entry->dport,
+		  __entry->saddr, __entry->daddr,
+		  __entry->saddr_v6, __entry->daddr_v6,
+		  show_tcp_state_name(__entry->oldstate),
+		  show_tcp_state_name(__entry->newstate))
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8b1fa4dd45384..be07e9b6dbdde 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -282,6 +282,8 @@
 #include <asm/ioctls.h>
 #include <net/busy_poll.h>
 
+#include <trace/events/tcp.h>
+
 int sysctl_tcp_min_tso_segs __read_mostly = 2;
 
 int sysctl_tcp_autocorking __read_mostly = 1;
@@ -2040,6 +2042,8 @@ void tcp_set_state(struct sock *sk, int state)
 {
 	int oldstate = sk->sk_state;
 
+	trace_tcp_set_state(sk, oldstate, state);
+
 	switch (state) {
 	case TCP_ESTABLISHED:
 		if (oldstate != TCP_ESTABLISHED)
-- 
cgit v1.2.3


From e87c6bc3852b981e71c757be20771546ce9f76f3 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Mon, 23 Oct 2017 23:53:08 -0700
Subject: bpf: permit multiple bpf attachments for a single perf event

This patch enables multiple bpf attachments for a
kprobe/uprobe/tracepoint single trace event.
Each trace_event keeps a list of attached perf events.
When an event happens, all attached bpf programs will
be executed based on the order of attachment.

A global bpf_event_mutex lock is introduced to protect
prog_array attaching and detaching. An alternative will
be introduce a mutex lock in every trace_event_call
structure, but it takes a lot of extra memory.
So a global bpf_event_mutex lock is a good compromise.

The bpf prog detachment involves allocation of memory.
If the allocation fails, a dummy do-nothing program
will replace to-be-detached program in-place.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/bpf.h           | 30 +++++++++++++---
 include/linux/trace_events.h  | 43 ++++++++++++++++++++---
 include/trace/perf.h          |  6 ++--
 kernel/bpf/core.c             | 81 ++++++++++++++++++++++++++++++++++++++++++
 kernel/events/core.c          | 26 +++++---------
 kernel/trace/bpf_trace.c      | 82 ++++++++++++++++++++++++++++++++++++++++---
 kernel/trace/trace_kprobe.c   |  6 ++--
 kernel/trace/trace_syscalls.c | 34 ++++++++++--------
 kernel/trace/trace_uprobe.c   |  3 +-
 9 files changed, 255 insertions(+), 56 deletions(-)

(limited to 'include/trace')

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 1e334b248ff65..172be7faf7bae 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -273,18 +273,38 @@ int bpf_prog_array_length(struct bpf_prog_array __rcu *progs);
 int bpf_prog_array_copy_to_user(struct bpf_prog_array __rcu *progs,
 				__u32 __user *prog_ids, u32 cnt);
 
-#define BPF_PROG_RUN_ARRAY(array, ctx, func)		\
+void bpf_prog_array_delete_safe(struct bpf_prog_array __rcu *progs,
+				struct bpf_prog *old_prog);
+int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array,
+			struct bpf_prog *exclude_prog,
+			struct bpf_prog *include_prog,
+			struct bpf_prog_array **new_array);
+
+#define __BPF_PROG_RUN_ARRAY(array, ctx, func, check_non_null)	\
 	({						\
-		struct bpf_prog **_prog;		\
+		struct bpf_prog **_prog, *__prog;	\
+		struct bpf_prog_array *_array;		\
 		u32 _ret = 1;				\
 		rcu_read_lock();			\
-		_prog = rcu_dereference(array)->progs;	\
-		for (; *_prog; _prog++)			\
-			_ret &= func(*_prog, ctx);	\
+		_array = rcu_dereference(array);	\
+		if (unlikely(check_non_null && !_array))\
+			goto _out;			\
+		_prog = _array->progs;			\
+		while ((__prog = READ_ONCE(*_prog))) {	\
+			_ret &= func(__prog, ctx);	\
+			_prog++;			\
+		}					\
+_out:							\
 		rcu_read_unlock();			\
 		_ret;					\
 	 })
 
+#define BPF_PROG_RUN_ARRAY(array, ctx, func)		\
+	__BPF_PROG_RUN_ARRAY(array, ctx, func, false)
+
+#define BPF_PROG_RUN_ARRAY_CHECK(array, ctx, func)	\
+	__BPF_PROG_RUN_ARRAY(array, ctx, func, true)
+
 #ifdef CONFIG_BPF_SYSCALL
 DECLARE_PER_CPU(int, bpf_prog_active);
 
diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
index 2e0f22298fe95..fc6aeca945db7 100644
--- a/include/linux/trace_events.h
+++ b/include/linux/trace_events.h
@@ -271,14 +271,37 @@ struct trace_event_call {
 #ifdef CONFIG_PERF_EVENTS
 	int				perf_refcount;
 	struct hlist_head __percpu	*perf_events;
-	struct bpf_prog			*prog;
-	struct perf_event		*bpf_prog_owner;
+	struct bpf_prog_array __rcu	*prog_array;
 
 	int	(*perf_perm)(struct trace_event_call *,
 			     struct perf_event *);
 #endif
 };
 
+#ifdef CONFIG_PERF_EVENTS
+static inline bool bpf_prog_array_valid(struct trace_event_call *call)
+{
+	/*
+	 * This inline function checks whether call->prog_array
+	 * is valid or not. The function is called in various places,
+	 * outside rcu_read_lock/unlock, as a heuristic to speed up execution.
+	 *
+	 * If this function returns true, and later call->prog_array
+	 * becomes false inside rcu_read_lock/unlock region,
+	 * we bail out then. If this function return false,
+	 * there is a risk that we might miss a few events if the checking
+	 * were delayed until inside rcu_read_lock/unlock region and
+	 * call->prog_array happened to become non-NULL then.
+	 *
+	 * Here, READ_ONCE() is used instead of rcu_access_pointer().
+	 * rcu_access_pointer() requires the actual definition of
+	 * "struct bpf_prog_array" while READ_ONCE() only needs
+	 * a declaration of the same type.
+	 */
+	return !!READ_ONCE(call->prog_array);
+}
+#endif
+
 static inline const char *
 trace_event_name(struct trace_event_call *call)
 {
@@ -435,12 +458,23 @@ trace_trigger_soft_disabled(struct trace_event_file *file)
 }
 
 #ifdef CONFIG_BPF_EVENTS
-unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx);
+unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx);
+int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog);
+void perf_event_detach_bpf_prog(struct perf_event *event);
 #else
-static inline unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx)
+static inline unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx)
 {
 	return 1;
 }
+
+static inline int
+perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void perf_event_detach_bpf_prog(struct perf_event *event) { }
+
 #endif
 
 enum {
@@ -511,6 +545,7 @@ perf_trace_buf_submit(void *raw_data, int size, int rctx, u16 type,
 {
 	perf_tp_event(type, count, raw_data, size, regs, head, rctx, task, event);
 }
+
 #endif
 
 #endif /* _LINUX_TRACE_EVENT_H */
diff --git a/include/trace/perf.h b/include/trace/perf.h
index 04fe68bbe7679..14f127b6acf54 100644
--- a/include/trace/perf.h
+++ b/include/trace/perf.h
@@ -34,7 +34,6 @@ perf_trace_##call(void *__data, proto)					\
 	struct trace_event_call *event_call = __data;			\
 	struct trace_event_data_offsets_##call __maybe_unused __data_offsets;\
 	struct trace_event_raw_##call *entry;				\
-	struct bpf_prog *prog = event_call->prog;			\
 	struct pt_regs *__regs;						\
 	u64 __count = 1;						\
 	struct task_struct *__task = NULL;				\
@@ -46,8 +45,9 @@ perf_trace_##call(void *__data, proto)					\
 	__data_size = trace_event_get_offsets_##call(&__data_offsets, args); \
 									\
 	head = this_cpu_ptr(event_call->perf_events);			\
-	if (!prog && __builtin_constant_p(!__task) && !__task &&	\
-				hlist_empty(head))			\
+	if (!bpf_prog_array_valid(event_call) &&			\
+	    __builtin_constant_p(!__task) && !__task &&			\
+	    hlist_empty(head))						\
 		return;							\
 									\
 	__entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32),\
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 8e7c8bf2b687d..7fe448799d762 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1394,6 +1394,20 @@ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err)
 }
 EXPORT_SYMBOL_GPL(bpf_prog_select_runtime);
 
+static unsigned int __bpf_prog_ret1(const void *ctx,
+				    const struct bpf_insn *insn)
+{
+	return 1;
+}
+
+static struct bpf_prog_dummy {
+	struct bpf_prog prog;
+} dummy_bpf_prog = {
+	.prog = {
+		.bpf_func = __bpf_prog_ret1,
+	},
+};
+
 /* to avoid allocating empty bpf_prog_array for cgroups that
  * don't have bpf program attached use one global 'empty_prog_array'
  * It will not be modified the caller of bpf_prog_array_alloc()
@@ -1463,6 +1477,73 @@ int bpf_prog_array_copy_to_user(struct bpf_prog_array __rcu *progs,
 	return 0;
 }
 
+void bpf_prog_array_delete_safe(struct bpf_prog_array __rcu *progs,
+				struct bpf_prog *old_prog)
+{
+	struct bpf_prog **prog = progs->progs;
+
+	for (; *prog; prog++)
+		if (*prog == old_prog) {
+			WRITE_ONCE(*prog, &dummy_bpf_prog.prog);
+			break;
+		}
+}
+
+int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array,
+			struct bpf_prog *exclude_prog,
+			struct bpf_prog *include_prog,
+			struct bpf_prog_array **new_array)
+{
+	int new_prog_cnt, carry_prog_cnt = 0;
+	struct bpf_prog **existing_prog;
+	struct bpf_prog_array *array;
+	int new_prog_idx = 0;
+
+	/* Figure out how many existing progs we need to carry over to
+	 * the new array.
+	 */
+	if (old_array) {
+		existing_prog = old_array->progs;
+		for (; *existing_prog; existing_prog++) {
+			if (*existing_prog != exclude_prog &&
+			    *existing_prog != &dummy_bpf_prog.prog)
+				carry_prog_cnt++;
+			if (*existing_prog == include_prog)
+				return -EEXIST;
+		}
+	}
+
+	/* How many progs (not NULL) will be in the new array? */
+	new_prog_cnt = carry_prog_cnt;
+	if (include_prog)
+		new_prog_cnt += 1;
+
+	/* Do we have any prog (not NULL) in the new array? */
+	if (!new_prog_cnt) {
+		*new_array = NULL;
+		return 0;
+	}
+
+	/* +1 as the end of prog_array is marked with NULL */
+	array = bpf_prog_array_alloc(new_prog_cnt + 1, GFP_KERNEL);
+	if (!array)
+		return -ENOMEM;
+
+	/* Fill in the new prog array */
+	if (carry_prog_cnt) {
+		existing_prog = old_array->progs;
+		for (; *existing_prog; existing_prog++)
+			if (*existing_prog != exclude_prog &&
+			    *existing_prog != &dummy_bpf_prog.prog)
+				array->progs[new_prog_idx++] = *existing_prog;
+	}
+	if (include_prog)
+		array->progs[new_prog_idx++] = include_prog;
+	array->progs[new_prog_idx] = NULL;
+	*new_array = array;
+	return 0;
+}
+
 static void bpf_prog_free_deferred(struct work_struct *work)
 {
 	struct bpf_prog_aux *aux;
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 9f78a6825bbed..9660ee65fbef3 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -7954,11 +7954,9 @@ void perf_trace_run_bpf_submit(void *raw_data, int size, int rctx,
 			       struct pt_regs *regs, struct hlist_head *head,
 			       struct task_struct *task)
 {
-	struct bpf_prog *prog = call->prog;
-
-	if (prog) {
+	if (bpf_prog_array_valid(call)) {
 		*(struct pt_regs **)raw_data = regs;
-		if (!trace_call_bpf(prog, raw_data) || hlist_empty(head)) {
+		if (!trace_call_bpf(call, raw_data) || hlist_empty(head)) {
 			perf_swevent_put_recursion_context(rctx);
 			return;
 		}
@@ -8147,13 +8145,11 @@ static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd)
 {
 	bool is_kprobe, is_tracepoint, is_syscall_tp;
 	struct bpf_prog *prog;
+	int ret;
 
 	if (event->attr.type != PERF_TYPE_TRACEPOINT)
 		return perf_event_set_bpf_handler(event, prog_fd);
 
-	if (event->tp_event->prog)
-		return -EEXIST;
-
 	is_kprobe = event->tp_event->flags & TRACE_EVENT_FL_UKPROBE;
 	is_tracepoint = event->tp_event->flags & TRACE_EVENT_FL_TRACEPOINT;
 	is_syscall_tp = is_syscall_trace_event(event->tp_event);
@@ -8181,26 +8177,20 @@ static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd)
 			return -EACCES;
 		}
 	}
-	event->tp_event->prog = prog;
-	event->tp_event->bpf_prog_owner = event;
 
-	return 0;
+	ret = perf_event_attach_bpf_prog(event, prog);
+	if (ret)
+		bpf_prog_put(prog);
+	return ret;
 }
 
 static void perf_event_free_bpf_prog(struct perf_event *event)
 {
-	struct bpf_prog *prog;
-
 	if (event->attr.type != PERF_TYPE_TRACEPOINT) {
 		perf_event_free_bpf_handler(event);
 		return;
 	}
-
-	prog = event->tp_event->prog;
-	if (prog && event->tp_event->bpf_prog_owner == event) {
-		event->tp_event->prog = NULL;
-		bpf_prog_put(prog);
-	}
+	perf_event_detach_bpf_prog(event);
 }
 
 #else
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 3126da2f468a5..b65011d320e3d 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -17,7 +17,7 @@
 
 /**
  * trace_call_bpf - invoke BPF program
- * @prog: BPF program
+ * @call: tracepoint event
  * @ctx: opaque context pointer
  *
  * kprobe handlers execute BPF programs via this helper.
@@ -29,7 +29,7 @@
  * 1 - store kprobe event into ring buffer
  * Other values are reserved and currently alias to 1
  */
-unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx)
+unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx)
 {
 	unsigned int ret;
 
@@ -49,9 +49,22 @@ unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx)
 		goto out;
 	}
 
-	rcu_read_lock();
-	ret = BPF_PROG_RUN(prog, ctx);
-	rcu_read_unlock();
+	/*
+	 * Instead of moving rcu_read_lock/rcu_dereference/rcu_read_unlock
+	 * to all call sites, we did a bpf_prog_array_valid() there to check
+	 * whether call->prog_array is empty or not, which is
+	 * a heurisitc to speed up execution.
+	 *
+	 * If bpf_prog_array_valid() fetched prog_array was
+	 * non-NULL, we go into trace_call_bpf() and do the actual
+	 * proper rcu_dereference() under RCU lock.
+	 * If it turns out that prog_array is NULL then, we bail out.
+	 * For the opposite, if the bpf_prog_array_valid() fetched pointer
+	 * was NULL, you'll skip the prog_array with the risk of missing
+	 * out of events when it was updated in between this and the
+	 * rcu_dereference() which is accepted risk.
+	 */
+	ret = BPF_PROG_RUN_ARRAY_CHECK(call->prog_array, ctx, BPF_PROG_RUN);
 
  out:
 	__this_cpu_dec(bpf_prog_active);
@@ -741,3 +754,62 @@ const struct bpf_verifier_ops perf_event_verifier_ops = {
 
 const struct bpf_prog_ops perf_event_prog_ops = {
 };
+
+static DEFINE_MUTEX(bpf_event_mutex);
+
+int perf_event_attach_bpf_prog(struct perf_event *event,
+			       struct bpf_prog *prog)
+{
+	struct bpf_prog_array __rcu *old_array;
+	struct bpf_prog_array *new_array;
+	int ret = -EEXIST;
+
+	mutex_lock(&bpf_event_mutex);
+
+	if (event->prog)
+		goto out;
+
+	old_array = rcu_dereference_protected(event->tp_event->prog_array,
+					      lockdep_is_held(&bpf_event_mutex));
+	ret = bpf_prog_array_copy(old_array, NULL, prog, &new_array);
+	if (ret < 0)
+		goto out;
+
+	/* set the new array to event->tp_event and set event->prog */
+	event->prog = prog;
+	rcu_assign_pointer(event->tp_event->prog_array, new_array);
+	bpf_prog_array_free(old_array);
+
+out:
+	mutex_unlock(&bpf_event_mutex);
+	return ret;
+}
+
+void perf_event_detach_bpf_prog(struct perf_event *event)
+{
+	struct bpf_prog_array __rcu *old_array;
+	struct bpf_prog_array *new_array;
+	int ret;
+
+	mutex_lock(&bpf_event_mutex);
+
+	if (!event->prog)
+		goto out;
+
+	old_array = rcu_dereference_protected(event->tp_event->prog_array,
+					      lockdep_is_held(&bpf_event_mutex));
+
+	ret = bpf_prog_array_copy(old_array, event->prog, NULL, &new_array);
+	if (ret < 0) {
+		bpf_prog_array_delete_safe(old_array, event->prog);
+	} else {
+		rcu_assign_pointer(event->tp_event->prog_array, new_array);
+		bpf_prog_array_free(old_array);
+	}
+
+	bpf_prog_put(event->prog);
+	event->prog = NULL;
+
+out:
+	mutex_unlock(&bpf_event_mutex);
+}
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 8a907e12b6b90..abf92e478cfb5 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1174,13 +1174,12 @@ static void
 kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs)
 {
 	struct trace_event_call *call = &tk->tp.call;
-	struct bpf_prog *prog = call->prog;
 	struct kprobe_trace_entry_head *entry;
 	struct hlist_head *head;
 	int size, __size, dsize;
 	int rctx;
 
-	if (prog && !trace_call_bpf(prog, regs))
+	if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs))
 		return;
 
 	head = this_cpu_ptr(call->perf_events);
@@ -1210,13 +1209,12 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri,
 		    struct pt_regs *regs)
 {
 	struct trace_event_call *call = &tk->tp.call;
-	struct bpf_prog *prog = call->prog;
 	struct kretprobe_trace_entry_head *entry;
 	struct hlist_head *head;
 	int size, __size, dsize;
 	int rctx;
 
-	if (prog && !trace_call_bpf(prog, regs))
+	if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs))
 		return;
 
 	head = this_cpu_ptr(call->perf_events);
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 696afe72d3b14..71a6af34d7a91 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -559,9 +559,10 @@ static DECLARE_BITMAP(enabled_perf_exit_syscalls, NR_syscalls);
 static int sys_perf_refcount_enter;
 static int sys_perf_refcount_exit;
 
-static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs,
-			      struct syscall_metadata *sys_data,
-			      struct syscall_trace_enter *rec) {
+static int perf_call_bpf_enter(struct trace_event_call *call, struct pt_regs *regs,
+			       struct syscall_metadata *sys_data,
+			       struct syscall_trace_enter *rec)
+{
 	struct syscall_tp_t {
 		unsigned long long regs;
 		unsigned long syscall_nr;
@@ -573,7 +574,7 @@ static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs,
 	param.syscall_nr = rec->nr;
 	for (i = 0; i < sys_data->nb_args; i++)
 		param.args[i] = rec->args[i];
-	return trace_call_bpf(prog, &param);
+	return trace_call_bpf(call, &param);
 }
 
 static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
@@ -581,7 +582,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 	struct syscall_metadata *sys_data;
 	struct syscall_trace_enter *rec;
 	struct hlist_head *head;
-	struct bpf_prog *prog;
+	bool valid_prog_array;
 	int syscall_nr;
 	int rctx;
 	int size;
@@ -596,9 +597,9 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 	if (!sys_data)
 		return;
 
-	prog = READ_ONCE(sys_data->enter_event->prog);
 	head = this_cpu_ptr(sys_data->enter_event->perf_events);
-	if (!prog && hlist_empty(head))
+	valid_prog_array = bpf_prog_array_valid(sys_data->enter_event);
+	if (!valid_prog_array && hlist_empty(head))
 		return;
 
 	/* get the size after alignment with the u32 buffer size field */
@@ -614,7 +615,8 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 	syscall_get_arguments(current, regs, 0, sys_data->nb_args,
 			       (unsigned long *)&rec->args);
 
-	if ((prog && !perf_call_bpf_enter(prog, regs, sys_data, rec)) ||
+	if ((valid_prog_array &&
+	     !perf_call_bpf_enter(sys_data->enter_event, regs, sys_data, rec)) ||
 	    hlist_empty(head)) {
 		perf_swevent_put_recursion_context(rctx);
 		return;
@@ -659,8 +661,9 @@ static void perf_sysenter_disable(struct trace_event_call *call)
 	mutex_unlock(&syscall_trace_lock);
 }
 
-static int perf_call_bpf_exit(struct bpf_prog *prog, struct pt_regs *regs,
-			      struct syscall_trace_exit *rec) {
+static int perf_call_bpf_exit(struct trace_event_call *call, struct pt_regs *regs,
+			      struct syscall_trace_exit *rec)
+{
 	struct syscall_tp_t {
 		unsigned long long regs;
 		unsigned long syscall_nr;
@@ -670,7 +673,7 @@ static int perf_call_bpf_exit(struct bpf_prog *prog, struct pt_regs *regs,
 	*(struct pt_regs **)&param = regs;
 	param.syscall_nr = rec->nr;
 	param.ret = rec->ret;
-	return trace_call_bpf(prog, &param);
+	return trace_call_bpf(call, &param);
 }
 
 static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
@@ -678,7 +681,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 	struct syscall_metadata *sys_data;
 	struct syscall_trace_exit *rec;
 	struct hlist_head *head;
-	struct bpf_prog *prog;
+	bool valid_prog_array;
 	int syscall_nr;
 	int rctx;
 	int size;
@@ -693,9 +696,9 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 	if (!sys_data)
 		return;
 
-	prog = READ_ONCE(sys_data->exit_event->prog);
 	head = this_cpu_ptr(sys_data->exit_event->perf_events);
-	if (!prog && hlist_empty(head))
+	valid_prog_array = bpf_prog_array_valid(sys_data->exit_event);
+	if (!valid_prog_array && hlist_empty(head))
 		return;
 
 	/* We can probably do that at build time */
@@ -709,7 +712,8 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 	rec->nr = syscall_nr;
 	rec->ret = syscall_get_return_value(current, regs);
 
-	if ((prog && !perf_call_bpf_exit(prog, regs, rec)) ||
+	if ((valid_prog_array &&
+	     !perf_call_bpf_exit(sys_data->exit_event, regs, rec)) ||
 	    hlist_empty(head)) {
 		perf_swevent_put_recursion_context(rctx);
 		return;
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index 4525e0271a539..153c0e4114611 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -1113,13 +1113,12 @@ static void __uprobe_perf_func(struct trace_uprobe *tu,
 {
 	struct trace_event_call *call = &tu->tp.call;
 	struct uprobe_trace_entry_head *entry;
-	struct bpf_prog *prog = call->prog;
 	struct hlist_head *head;
 	void *data;
 	int size, esize;
 	int rctx;
 
-	if (prog && !trace_call_bpf(prog, regs))
+	if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs))
 		return;
 
 	esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu));
-- 
cgit v1.2.3


From cf34ce3da1e41579296364509266c7dac573822a Mon Sep 17 00:00:00 2001
From: Song Liu <songliubraving@fb.com>
Date: Mon, 30 Oct 2017 14:41:35 -0700
Subject: tcp: add tracepoint trace_tcp_retransmit_synack()

This tracepoint can be used to trace synack retransmits. It maintains
pointer to struct request_sock.

We cannot simply reuse trace_tcp_retransmit_skb() here, because the
sk here is the LISTEN socket. The IP addresses and ports should be
extracted from struct request_sock.

Note that, like many other tracepoints, this patch uses IS_ENABLED
in TP_fast_assign macro, which triggers sparse warning like:

./include/trace/events/tcp.h:274:1: error: directive in argument list
./include/trace/events/tcp.h:281:1: error: directive in argument list

However, there is no good solution to avoid these warnings. To the
best of our knowledge, these warnings are harmless.

Signed-off-by: Song Liu <songliubraving@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 56 ++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_output.c      |  1 +
 2 files changed, 57 insertions(+)

(limited to 'include/trace')

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 03699ba716230..07cccca6cbf17 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -237,6 +237,62 @@ TRACE_EVENT(tcp_set_state,
 		  show_tcp_state_name(__entry->newstate))
 );
 
+TRACE_EVENT(tcp_retransmit_synack,
+
+	TP_PROTO(const struct sock *sk, const struct request_sock *req),
+
+	TP_ARGS(sk, req),
+
+	TP_STRUCT__entry(
+		__field(const void *, skaddr)
+		__field(const void *, req)
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__array(__u8, saddr, 4)
+		__array(__u8, daddr, 4)
+		__array(__u8, saddr_v6, 16)
+		__array(__u8, daddr_v6, 16)
+	),
+
+	TP_fast_assign(
+		struct inet_request_sock *ireq = inet_rsk(req);
+		struct in6_addr *pin6;
+		__be32 *p32;
+
+		__entry->skaddr = sk;
+		__entry->req = req;
+
+		__entry->sport = ireq->ir_num;
+		__entry->dport = ntohs(ireq->ir_rmt_port);
+
+		p32 = (__be32 *) __entry->saddr;
+		*p32 = ireq->ir_loc_addr;
+
+		p32 = (__be32 *) __entry->daddr;
+		*p32 = ireq->ir_rmt_addr;
+
+#if IS_ENABLED(CONFIG_IPV6)
+		if (sk->sk_family == AF_INET6) {
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			*pin6 = ireq->ir_v6_loc_addr;
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			*pin6 = ireq->ir_v6_rmt_addr;
+		} else
+#endif
+		{
+			pin6 = (struct in6_addr *)__entry->saddr_v6;
+			ipv6_addr_set_v4mapped(ireq->ir_loc_addr, pin6);
+			pin6 = (struct in6_addr *)__entry->daddr_v6;
+			ipv6_addr_set_v4mapped(ireq->ir_rmt_addr, pin6);
+		}
+	),
+
+	TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6c daddrv6=%pI6c",
+		  __entry->sport, __entry->dport,
+		  __entry->saddr, __entry->daddr,
+		  __entry->saddr_v6, __entry->daddr_v6)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index a85e8a282d173..06a0c89ffe408 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3782,6 +3782,7 @@ int tcp_rtx_synack(const struct sock *sk, struct request_sock *req)
 		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNRETRANS);
 		if (unlikely(tcp_passive_fastopen(sk)))
 			tcp_sk(sk)->total_retrans++;
+		trace_tcp_retransmit_synack(sk, req);
 	}
 	return res;
 }
-- 
cgit v1.2.3


From a3dcaf17ee54f1d01d22cc2b22cab0b4f60d78cf Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 7 Nov 2017 00:29:27 -0800
Subject: net: allow per netns sysctl_rmem and sysctl_wmem for protos

As we want to gradually implement per netns sysctl_rmem and sysctl_wmem
on per protocol basis, add two new fields in struct proto,
and two new helpers : sk_get_wmem0() and sk_get_rmem0()

First user will be TCP. Then UDP and SCTP can be easily converted,
while DECNET probably wont get this support.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h          | 22 ++++++++++++++++++++++
 include/trace/events/sock.h |  2 +-
 net/core/sock.c             | 10 ++++++----
 3 files changed, 29 insertions(+), 5 deletions(-)

(limited to 'include/trace')

diff --git a/include/net/sock.h b/include/net/sock.h
index 6f1be9726e02b..688a823dccc30 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1101,8 +1101,12 @@ struct proto {
 	 */
 	unsigned long		*memory_pressure;
 	long			*sysctl_mem;
+
 	int			*sysctl_wmem;
 	int			*sysctl_rmem;
+	u32			sysctl_wmem_offset;
+	u32			sysctl_rmem_offset;
+
 	int			max_header;
 	bool			no_autobind;
 
@@ -2390,4 +2394,22 @@ extern int sysctl_optmem_max;
 extern __u32 sysctl_wmem_default;
 extern __u32 sysctl_rmem_default;
 
+static inline int sk_get_wmem0(const struct sock *sk, const struct proto *proto)
+{
+	/* Does this proto have per netns sysctl_wmem ? */
+	if (proto->sysctl_wmem_offset)
+		return *(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset);
+
+	return *proto->sysctl_wmem;
+}
+
+static inline int sk_get_rmem0(const struct sock *sk, const struct proto *proto)
+{
+	/* Does this proto have per netns sysctl_rmem ? */
+	if (proto->sysctl_rmem_offset)
+		return *(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset);
+
+	return *proto->sysctl_rmem;
+}
+
 #endif	/* _SOCK_H */
diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h
index 6d31c0520ef36..ec4dade244660 100644
--- a/include/trace/events/sock.h
+++ b/include/trace/events/sock.h
@@ -48,7 +48,7 @@ TRACE_EVENT(sock_exceed_buf_limit,
 		strncpy(__entry->name, prot->name, 32);
 		__entry->sysctl_mem = prot->sysctl_mem;
 		__entry->allocated = allocated;
-		__entry->sysctl_rmem = prot->sysctl_rmem[0];
+		__entry->sysctl_rmem = sk_get_rmem0(sk, prot);
 		__entry->rmem_alloc = atomic_read(&sk->sk_rmem_alloc);
 	),
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 759400053110f..c59bcf90d9053 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2346,16 +2346,18 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind)
 
 	/* guarantee minimum buffer size under pressure */
 	if (kind == SK_MEM_RECV) {
-		if (atomic_read(&sk->sk_rmem_alloc) < prot->sysctl_rmem[0])
+		if (atomic_read(&sk->sk_rmem_alloc) < sk_get_rmem0(sk, prot))
 			return 1;
 
 	} else { /* SK_MEM_SEND */
+		int wmem0 = sk_get_wmem0(sk, prot);
+
 		if (sk->sk_type == SOCK_STREAM) {
-			if (sk->sk_wmem_queued < prot->sysctl_wmem[0])
+			if (sk->sk_wmem_queued < wmem0)
 				return 1;
-		} else if (refcount_read(&sk->sk_wmem_alloc) <
-			   prot->sysctl_wmem[0])
+		} else if (refcount_read(&sk->sk_wmem_alloc) < wmem0) {
 				return 1;
+		}
 	}
 
 	if (sk_has_memory_pressure(sk)) {
-- 
cgit v1.2.3