Skip to contentSkip to conversation

sock_map: clean up and refactor code for BPF_SK_SKB_VERDICT

PATCH Applied Cong Wang 4 participants 14 messages Feb 23, 2021 – Mar 1, 2021

Subsystems (11)
lore
Share & tools
lore

Cong Wang

author
#0 lore raw

v7 cover sock_map: clean up and refactor code for BPF_SK_SKB_VERDICT

From: Cong Wang <[email protected]>

This patchset is the first series of patches separated out from the original large patchset, to make reviews easier. This patchset does not add any new feature or change any functionality but merely cleans up the existing sockmap and skmsg code and refactors it, to prepare for the patches followed up. This passed all BPF selftests.

To see the big picture, the original whole patchset is available on github: https://github.com/congwang/linux/tree/sockmap

and this patchset is also available on github:
https://github.com/congwang/linux/tree/sockmap1

---
v7: add 1 trivial cleanup patch
define a mask for sk_redir
fix CONFIG_BPF_SYSCALL in include/net/udp.h make sk_psock_done_strp() static
move skb_bpf_redirect_clear() to sk_psock_backlog()

v6: fix !CONFIG_INET case

v5: improve CONFIG_BPF_SYSCALL dependency
add 3 trivial cleanup patches

v4: reuse skb dst instead of skb ext
fix another Kconfig error

v3: fix a few Kconfig compile errors
remove an unused variable
add a comment for bpf_convert_data_end_access()

v2: split the original patchset
compute data_end with bpf_convert_data_end_access() get rid of psock->bpf_running
reduce the scope of CONFIG_BPF_STREAM_PARSER do not add CONFIG_BPF_SOCK_MAP

Cong Wang (9):
bpf: clean up sockmap related Kconfigs
skmsg: get rid of struct sk_psock_parser
bpf: compute data_end dynamically with JIT code
skmsg: move sk_redir from TCP_SKB_CB to skb
sock_map: rename skb_parser and skb_verdict
sock_map: make sock_map_prog_update() static
skmsg: make __sk_psock_purge_ingress_msg() static
skmsg: get rid of sk_psock_bpf_run()
skmsg: remove unused sk_psock_stop() declaration

 include/linux/bpf.h                           |  29 +--
 include/linux/bpf_types.h                     |   6 +-
 include/linux/skbuff.h                        |   3 +
 include/linux/skmsg.h                         |  82 +++++--
 include/net/tcp.h                             |  41 +---
 include/net/udp.h                             |   4 +-
 init/Kconfig                                  |   1 +
 net/Kconfig                                   |   6 +-
 net/core/Makefile                             |   6 +-
 net/core/filter.c                             |  48 ++--
 net/core/skmsg.c                              | 212 +++++++++---------
 net/core/sock_map.c                           |  77 +++----
 net/ipv4/Makefile                             |   2 +-
 net/ipv4/tcp_bpf.c                            |   4 +-
 .../selftests/bpf/prog_tests/sockmap_listen.c |   8 +-
 .../selftests/bpf/progs/test_sockmap_listen.c |   4 +-
 16 files changed, 274 insertions(+), 259 deletions(-)
Signature · 1 line
2.25.1
author
#1 lore raw

PATCH v7 1/9 bpf: clean up sockmap related Kconfigs

From: Cong Wang <[email protected]>

As suggested by John, clean up sockmap related Kconfigs:

Reduce the scope of CONFIG_BPF_STREAM_PARSER down to TCP stream parser, to reflect its name.

Make the rest sockmap code simply depend on CONFIG_BPF_SYSCALL and CONFIG_INET, the latter is still needed at this point because of TCP/UDP proto update. And leave CONFIG_NET_SOCK_MSG untouched, as it is used by non-sockmap cases.

---
 include/linux/bpf.h       |  26 ++++---
 include/linux/bpf_types.h |   6 +-
 include/linux/skmsg.h     |  18 +++++
 include/net/tcp.h         |  16 +++--
 include/net/udp.h         |   4 +-
 init/Kconfig              |   1 +
 net/Kconfig               |   6 +-
 net/core/Makefile         |   6 +-
 net/core/skmsg.c          | 145 +++++++++++++++++++++-----------------
 net/core/sock_map.c       |   2 +
 net/ipv4/Makefile         |   2 +-
 net/ipv4/tcp_bpf.c        |   4 +-
 12 files changed, 133 insertions(+), 103 deletions(-)
include/linux/bpf.h +12 -14
index cccaef1088ea..813f30ef44ff 100644
@@ -1768,7 +1768,7 @@ static inline void bpf_map_offload_map_free(struct bpf_map *map)
 }
 #endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */
 
-#if defined(CONFIG_BPF_STREAM_PARSER)
+#if defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL)
 int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
 			 struct bpf_prog *old, u32 which);
 int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
@@ -1776,7 +1776,18 @@ int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
 int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
 void sock_map_unhash(struct sock *sk);
 void sock_map_close(struct sock *sk, long timeout);
+
+void bpf_sk_reuseport_detach(struct sock *sk);
+int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map, void *key,
+				       void *value);
+int bpf_fd_reuseport_array_update_elem(struct bpf_map *map, void *key,
+				       void *value, u64 map_flags);
 #else
+static inline void bpf_sk_reuseport_detach(struct sock *sk)
+{
+}
+
+#ifdef CONFIG_BPF_SYSCALL
 static inline int sock_map_prog_update(struct bpf_map *map,
 				       struct bpf_prog *prog,
 				       struct bpf_prog *old, u32 which)
@@ -1801,20 +1812,7 @@ static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void
 {
 	return -EOPNOTSUPP;
 }
-#endif /* CONFIG_BPF_STREAM_PARSER */
 
-#if defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL)
-void bpf_sk_reuseport_detach(struct sock *sk);
-int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map, void *key,
-				       void *value);
-int bpf_fd_reuseport_array_update_elem(struct bpf_map *map, void *key,
-				       void *value, u64 map_flags);
-#else
-static inline void bpf_sk_reuseport_detach(struct sock *sk)
-{
-}
-
-#ifdef CONFIG_BPF_SYSCALL
 static inline int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map,
 						     void *key, void *value)
 {
include/linux/bpf_types.h +2 -4
index 99f7fd657d87..38fd98901ba9 100644
@@ -103,10 +103,6 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_HASH_OF_MAPS, htab_of_maps_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_DEVMAP, dev_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_DEVMAP_HASH, dev_map_hash_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_SK_STORAGE, sk_storage_map_ops)
-#if defined(CONFIG_BPF_STREAM_PARSER)
-BPF_MAP_TYPE(BPF_MAP_TYPE_SOCKMAP, sock_map_ops)
-BPF_MAP_TYPE(BPF_MAP_TYPE_SOCKHASH, sock_hash_ops)
-#endif
 #ifdef CONFIG_BPF_LSM
 BPF_MAP_TYPE(BPF_MAP_TYPE_INODE_STORAGE, inode_storage_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_TASK_STORAGE, task_storage_map_ops)
@@ -116,6 +112,8 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_CPUMAP, cpu_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_XSKMAP, xsk_map_ops)
 #endif
 #ifdef CONFIG_INET
+BPF_MAP_TYPE(BPF_MAP_TYPE_SOCKMAP, sock_map_ops)
+BPF_MAP_TYPE(BPF_MAP_TYPE_SOCKHASH, sock_hash_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_REUSEPORT_SOCKARRAY, reuseport_array_ops)
 #endif
 #endif
include/linux/skmsg.h +18 -0
index 8edbbf5f2f93..db7a08be4725 100644
@@ -71,7 +71,9 @@ struct sk_psock_link {
 };
 
 struct sk_psock_parser {
+#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
 	struct strparser		strp;
+#endif
 	bool				enabled;
 	void (*saved_data_ready)(struct sock *sk);
 };
@@ -305,9 +307,25 @@ static inline void sk_psock_report_error(struct sk_psock *psock, int err)
 
 struct sk_psock *sk_psock_init(struct sock *sk, int node);
 
+#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
 int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock);
 void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock);
 void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock);
+#else
+static inline int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock)
+{
+}
+
+static inline void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
+{
+}
+#endif
+
 void sk_psock_start_verdict(struct sock *sk, struct sk_psock *psock);
 void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock);
 
include/net/tcp.h +9 -7
index 963cd86d12dd..c00e125dcfb9 100644
@@ -2222,25 +2222,27 @@ void tcp_update_ulp(struct sock *sk, struct proto *p,
 	__MODULE_INFO(alias, alias_userspace, name);		\
 	__MODULE_INFO(alias, alias_tcp_ulp, "tcp-ulp-" name)
 
+#ifdef CONFIG_NET_SOCK_MSG
 struct sk_msg;
 struct sk_psock;
 
-#ifdef CONFIG_BPF_STREAM_PARSER
+#ifdef CONFIG_BPF_SYSCALL
 struct proto *tcp_bpf_get_proto(struct sock *sk, struct sk_psock *psock);
 void tcp_bpf_clone(const struct sock *sk, struct sock *newsk);
-#else
-static inline void tcp_bpf_clone(const struct sock *sk, struct sock *newsk)
-{
-}
-#endif /* CONFIG_BPF_STREAM_PARSER */
+#endif /* CONFIG_BPF_SYSCALL */
 
-#ifdef CONFIG_NET_SOCK_MSG
 int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, u32 bytes,
 			  int flags);
 int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,
 		      struct msghdr *msg, int len, int flags);
 #endif /* CONFIG_NET_SOCK_MSG */
 
+#if !defined(CONFIG_BPF_SYSCALL) || !defined(CONFIG_NET_SOCK_MSG)
+static inline void tcp_bpf_clone(const struct sock *sk, struct sock *newsk)
+{
+}
+#endif
+
 #ifdef CONFIG_CGROUP_BPF
 static inline void bpf_skops_init_skb(struct bpf_sock_ops_kern *skops,
 				      struct sk_buff *skb,
include/net/udp.h +2 -2
index a132a02b2f2c..d4d064c59232 100644
@@ -515,9 +515,9 @@ static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
 	return segs;
 }
 
-#ifdef CONFIG_BPF_STREAM_PARSER
+#ifdef CONFIG_BPF_SYSCALL
 struct sk_psock;
 struct proto *udp_bpf_get_proto(struct sock *sk, struct sk_psock *psock);
-#endif /* BPF_STREAM_PARSER */
+#endif
 
 #endif	/* _UDP_H */
init/Kconfig +1 -0
index 096e1af5c586..66cef5eac275 100644
@@ -1702,6 +1702,7 @@ config BPF_SYSCALL
 	select BPF
 	select IRQ_WORK
 	select TASKS_TRACE_RCU
+	select NET_SOCK_MSG if INET
 	default n
 	help
 	  Enable the bpf() system call that allows to manipulate eBPF
net/Kconfig +1 -5
index 8cea808ad9e8..0ead7ec0d2bd 100644
@@ -317,13 +317,9 @@ config BPF_STREAM_PARSER
 	select STREAM_PARSER
 	select NET_SOCK_MSG
 	help
-	  Enabling this allows a stream parser to be used with
+	  Enabling this allows a TCP stream parser to be used with
 	  BPF_MAP_TYPE_SOCKMAP.
 
-	  BPF_MAP_TYPE_SOCKMAP provides a map type to use with network sockets.
-	  It can be used to enforce socket policy, implement socket redirects,
-	  etc.
-
 config NET_FLOW_LIMIT
 	bool
 	depends on RPS
net/core/Makefile +4 -2
index 3e2c378e5f31..0c2233c826fd 100644
@@ -16,7 +16,6 @@ obj-y		     += dev.o dev_addr_lists.o dst.o netevent.o \
 obj-y += net-sysfs.o
 obj-$(CONFIG_PAGE_POOL) += page_pool.o
 obj-$(CONFIG_PROC_FS) += net-procfs.o
-obj-$(CONFIG_NET_SOCK_MSG) += skmsg.o
 obj-$(CONFIG_NET_PKTGEN) += pktgen.o
 obj-$(CONFIG_NETPOLL) += netpoll.o
 obj-$(CONFIG_FIB_RULES) += fib_rules.o
@@ -28,10 +27,13 @@ obj-$(CONFIG_CGROUP_NET_PRIO) += netprio_cgroup.o
 obj-$(CONFIG_CGROUP_NET_CLASSID) += netclassid_cgroup.o
 obj-$(CONFIG_LWTUNNEL) += lwtunnel.o
 obj-$(CONFIG_LWTUNNEL_BPF) += lwt_bpf.o
-obj-$(CONFIG_BPF_STREAM_PARSER) += sock_map.o
 obj-$(CONFIG_DST_CACHE) += dst_cache.o
 obj-$(CONFIG_HWBM) += hwbm.o
 obj-$(CONFIG_NET_DEVLINK) += devlink.o
 obj-$(CONFIG_GRO_CELLS) += gro_cells.o
 obj-$(CONFIG_FAILOVER) += failover.o
+ifeq ($(CONFIG_INET),y)
+obj-$(CONFIG_NET_SOCK_MSG) += skmsg.o
+obj-$(CONFIG_BPF_SYSCALL) += sock_map.o
+endif
 obj-$(CONFIG_BPF_SYSCALL) += bpf_sk_storage.o
net/core/skmsg.c +79 -66

193 lines not on this page · raw message

net/core/sock_map.c +2 -0
index d758fb83c884..ee3334dd3a38 100644
@@ -1461,9 +1461,11 @@ int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
 	case BPF_SK_MSG_VERDICT:
 		pprog = &progs->msg_parser;
 		break;
+#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
 	case BPF_SK_SKB_STREAM_PARSER:
 		pprog = &progs->skb_parser;
 		break;
+#endif
 	case BPF_SK_SKB_STREAM_VERDICT:
 		pprog = &progs->skb_verdict;
 		break;
net/ipv4/Makefile +1 -1
index 5b77a46885b9..bbdd9c44f14e 100644
@@ -62,7 +62,7 @@ obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
 obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
 obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
 obj-$(CONFIG_NET_SOCK_MSG) += tcp_bpf.o
-obj-$(CONFIG_BPF_STREAM_PARSER) += udp_bpf.o
+obj-$(CONFIG_BPF_SYSCALL) += udp_bpf.o
 obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
net/ipv4/tcp_bpf.c +2 -2
index bc7d2a586e18..17c322b875fd 100644
@@ -229,7 +229,7 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg,
 }
 EXPORT_SYMBOL_GPL(tcp_bpf_sendmsg_redir);
 
-#ifdef CONFIG_BPF_STREAM_PARSER
+#ifdef CONFIG_BPF_SYSCALL
 static bool tcp_bpf_stream_read(const struct sock *sk)
 {
 	struct sk_psock *psock;
@@ -629,4 +629,4 @@ void tcp_bpf_clone(const struct sock *sk, struct sock *newsk)
 	if (prot == &tcp_bpf_prots[family][TCP_BPF_BASE])
 		newsk->sk_prot = sk->sk_prot_creator;
 }
-#endif /* CONFIG_BPF_STREAM_PARSER */
+#endif /* CONFIG_BPF_SYSCALL */
Signature · 1 line
2.25.1
author
#2 lore raw

PATCH v7 2/9 skmsg: get rid of struct sk_psock_parser

In reply to Cong Wang

From: Cong Wang <[email protected]>

struct sk_psock_parser is embedded in sk_psock, it is unnecessary as skb verdict also uses ->saved_data_ready. We can simply fold these fields into sk_psock, and get rid of ->enabled.

---
 include/linux/skmsg.h | 19 ++++++----------
 net/core/skmsg.c      | 53 +++++++++++++------------------------------
 net/core/sock_map.c   |  8 +++----
 3 files changed, 27 insertions(+), 53 deletions(-)
include/linux/skmsg.h +7 -12
index db7a08be4725..22e26f82de33 100644
@@ -70,14 +70,6 @@ struct sk_psock_link {
 	void				*link_raw;
 };
 
-struct sk_psock_parser {
-#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
-	struct strparser		strp;
-#endif
-	bool				enabled;
-	void (*saved_data_ready)(struct sock *sk);
-};
-
 struct sk_psock_work_state {
 	struct sk_buff			*skb;
 	u32				len;
@@ -92,7 +84,9 @@ struct sk_psock {
 	u32				eval;
 	struct sk_msg			*cork;
 	struct sk_psock_progs		progs;
-	struct sk_psock_parser		parser;
+#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
+	struct strparser		strp;
+#endif
 	struct sk_buff_head		ingress_skb;
 	struct list_head		ingress_msg;
 	unsigned long			state;
@@ -102,6 +96,7 @@ struct sk_psock {
 	void (*saved_unhash)(struct sock *sk);
 	void (*saved_close)(struct sock *sk, long timeout);
 	void (*saved_write_space)(struct sock *sk);
+	void (*saved_data_ready)(struct sock *sk);
 	struct proto			*sk_proto;
 	struct sk_psock_work_state	work_state;
 	struct work_struct		work;
@@ -418,8 +413,8 @@ static inline void sk_psock_put(struct sock *sk, struct sk_psock *psock)
 
 static inline void sk_psock_data_ready(struct sock *sk, struct sk_psock *psock)
 {
-	if (psock->parser.enabled)
-		psock->parser.saved_data_ready(sk);
+	if (psock->saved_data_ready)
+		psock->saved_data_ready(sk);
 	else
 		sk->sk_data_ready(sk);
 }
@@ -458,6 +453,6 @@ static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
 {
 	if (!psock)
 		return false;
-	return psock->parser.enabled;
+	return !!psock->saved_data_ready;
 }
 #endif /* _LINUX_SKMSG_H */
net/core/skmsg.c +16 -37

109 lines not on this page · raw message

net/core/sock_map.c +4 -4
index ee3334dd3a38..1a28a5c2c61e 100644
@@ -148,9 +148,9 @@ static void sock_map_del_link(struct sock *sk,
 			struct bpf_map *map = link->map;
 			struct bpf_stab *stab = container_of(map, struct bpf_stab,
 							     map);
-			if (psock->parser.enabled && stab->progs.skb_parser)
+			if (psock->saved_data_ready && stab->progs.skb_parser)
 				strp_stop = true;
-			if (psock->parser.enabled && stab->progs.skb_verdict)
+			if (psock->saved_data_ready && stab->progs.skb_verdict)
 				verdict_stop = true;
 			list_del(&link->list);
 			sk_psock_free_link(link);
@@ -283,14 +283,14 @@ static int sock_map_link(struct bpf_map *map, struct sk_psock_progs *progs,
 		goto out_drop;
 
 	write_lock_bh(&sk->sk_callback_lock);
-	if (skb_parser && skb_verdict && !psock->parser.enabled) {
+	if (skb_parser && skb_verdict && !psock->saved_data_ready) {
 		ret = sk_psock_init_strp(sk, psock);
 		if (ret)
 			goto out_unlock_drop;
 		psock_set_prog(&psock->progs.skb_verdict, skb_verdict);
 		psock_set_prog(&psock->progs.skb_parser, skb_parser);
 		sk_psock_start_strp(sk, psock);
-	} else if (!skb_parser && skb_verdict && !psock->parser.enabled) {
+	} else if (!skb_parser && skb_verdict && !psock->saved_data_ready) {
 		psock_set_prog(&psock->progs.skb_verdict, skb_verdict);
 		sk_psock_start_verdict(sk,psock);
 	}
Signature · 1 line
2.25.1
author
#3 lore raw

PATCH v7 3/9 bpf: compute data_end dynamically with JIT code

In reply to Cong Wang

From: Cong Wang <[email protected]>

Currently, we compute ->data_end with a compile-time constant offset of skb. But as Jakub pointed out, we can actually compute it in eBPF JIT code at run-time, so that we can competely get rid of ->data_end. This is similar to skb_shinfo(skb) computation in bpf_convert_shinfo_access().

---
 include/net/tcp.h |  6 ------
 net/core/filter.c | 48 +++++++++++++++++++++++++++--------------------
 net/core/skmsg.c  |  1 -
 3 files changed, 28 insertions(+), 27 deletions(-)
include/net/tcp.h +0 -6
index c00e125dcfb9..947ef5da6867 100644
@@ -886,18 +886,12 @@ struct tcp_skb_cb {
 		struct {
 			__u32 flags;
 			struct sock *sk_redir;
-			void *data_end;
 		} bpf;
 	};
 };
 
 #define TCP_SKB_CB(__skb)	((struct tcp_skb_cb *)&((__skb)->cb[0]))
 
-static inline void bpf_compute_data_end_sk_skb(struct sk_buff *skb)
-{
-	TCP_SKB_CB(skb)->bpf.data_end = skb->data + skb_headlen(skb);
-}
-
 static inline bool tcp_skb_bpf_ingress(const struct sk_buff *skb)
 {
 	return TCP_SKB_CB(skb)->bpf.flags & BPF_F_INGRESS;
net/core/filter.c +28 -20

87 lines not on this page · raw message

net/core/skmsg.c +0 -1
index d00c9a4b47e7..8822001ab3dc 100644
@@ -746,7 +746,6 @@ EXPORT_SYMBOL_GPL(sk_psock_msg_verdict);
 static int sk_psock_bpf_run(struct sk_psock *psock, struct bpf_prog *prog,
 			    struct sk_buff *skb)
 {
-	bpf_compute_data_end_sk_skb(skb);
 	return bpf_prog_run_pin_on_cpu(prog, skb);
 }
 
Signature · 1 line
2.25.1
author
#4 lore raw

PATCH v7 4/9 skmsg: move sk_redir from TCP_SKB_CB to skb

In reply to Cong Wang

From: Cong Wang <[email protected]>

Currently TCP_SKB_CB() is hard-coded in skmsg code, it certainly does not work for any other non-TCP protocols. We can move them to skb ext, but it introduces a memory allocation on fast path.

Fortunately, we only need to a word-size to store all the information, because the flags actually only contains 1 bit so can be just packed into the lowest bit of the "pointer", which is stored as unsigned long.

Inside struct sk_buff, '_skb_refdst' can be reused because skb dst is no longer needed after ->sk_data_ready() so we can just drop it.

---
 include/linux/skbuff.h |  3 +++
 include/linux/skmsg.h  | 38 ++++++++++++++++++++++++++++++++++++++
 include/net/tcp.h      | 19 -------------------
 net/core/skmsg.c       | 31 +++++++++++++++++++------------
 net/core/sock_map.c    |  8 ++------
 5 files changed, 62 insertions(+), 37 deletions(-)
include/linux/skbuff.h +3 -0
index 6d0a33d1c0db..bd84f799c952 100644
@@ -755,6 +755,9 @@ struct sk_buff {
 			void		(*destructor)(struct sk_buff *skb);
 		};
 		struct list_head	tcp_tsorted_anchor;
+#ifdef CONFIG_NET_SOCK_MSG
+		unsigned long		_sk_redir;
+#endif
 	};
 
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
include/linux/skmsg.h +38 -0
index 22e26f82de33..e0de45527bb6 100644
@@ -455,4 +455,42 @@ static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
 		return false;
 	return !!psock->saved_data_ready;
 }
+
+#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
+
+/* We only have one bit so far. */
+#define BPF_F_PTR_MASK ~(BPF_F_INGRESS)
+
+static inline bool skb_bpf_ingress(const struct sk_buff *skb)
+{
+	unsigned long sk_redir = skb->_sk_redir;
+
+	return sk_redir & BPF_F_INGRESS;
+}
+
+static inline void skb_bpf_set_ingress(struct sk_buff *skb)
+{
+	skb->_sk_redir |= BPF_F_INGRESS;
+}
+
+static inline void skb_bpf_set_redir(struct sk_buff *skb, struct sock *sk_redir,
+				     bool ingress)
+{
+	skb->_sk_redir = (unsigned long)sk_redir;
+	if (ingress)
+		skb->_sk_redir |= BPF_F_INGRESS;
+}
+
+static inline struct sock *skb_bpf_redirect_fetch(const struct sk_buff *skb)
+{
+	unsigned long sk_redir = skb->_sk_redir;
+
+	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
+}
+
+static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
+{
+	skb->_sk_redir = 0;
+}
+#endif /* CONFIG_NET_SOCK_MSG */
 #endif /* _LINUX_SKMSG_H */
include/net/tcp.h +0 -19
index 947ef5da6867..075de26f449d 100644
@@ -883,30 +883,11 @@ struct tcp_skb_cb {
 			struct inet6_skb_parm	h6;
 #endif
 		} header;	/* For incoming skbs */
-		struct {
-			__u32 flags;
-			struct sock *sk_redir;
-		} bpf;
 	};
 };
 
 #define TCP_SKB_CB(__skb)	((struct tcp_skb_cb *)&((__skb)->cb[0]))
 
-static inline bool tcp_skb_bpf_ingress(const struct sk_buff *skb)
-{
-	return TCP_SKB_CB(skb)->bpf.flags & BPF_F_INGRESS;
-}
-
-static inline struct sock *tcp_skb_bpf_redirect_fetch(struct sk_buff *skb)
-{
-	return TCP_SKB_CB(skb)->bpf.sk_redir;
-}
-
-static inline void tcp_skb_bpf_redirect_clear(struct sk_buff *skb)
-{
-	TCP_SKB_CB(skb)->bpf.sk_redir = NULL;
-}
-
 extern const struct inet_connection_sock_af_ops ipv4_specific;
 
 #if IS_ENABLED(CONFIG_IPV6)
net/core/skmsg.c +19 -12
index 8822001ab3dc..409258367bea 100644
@@ -525,7 +525,8 @@ static void sk_psock_backlog(struct work_struct *work)
 		len = skb->len;
 		off = 0;
 start:
-		ingress = tcp_skb_bpf_ingress(skb);
+		ingress = skb_bpf_ingress(skb);
+		skb_bpf_redirect_clear(skb);
 		do {
 			ret = -EIO;
 			if (likely(psock->sk->sk_socket))
@@ -631,7 +632,12 @@ void __sk_psock_purge_ingress_msg(struct sk_psock *psock)
 
 static void sk_psock_zap_ingress(struct sk_psock *psock)
 {
-	__skb_queue_purge(&psock->ingress_skb);
+	struct sk_buff *skb;
+
+	while ((skb = __skb_dequeue(&psock->ingress_skb)) != NULL) {
+		skb_bpf_redirect_clear(skb);
+		kfree_skb(skb);
+	}
 	__sk_psock_purge_ingress_msg(psock);
 }
 
@@ -754,7 +760,7 @@ static void sk_psock_skb_redirect(struct sk_buff *skb)
 	struct sk_psock *psock_other;
 	struct sock *sk_other;
 
-	sk_other = tcp_skb_bpf_redirect_fetch(skb);
+	sk_other = skb_bpf_redirect_fetch(skb);
 	/* This error is a buggy BPF program, it returned a redirect
 	 * return code, but then didn't set a redirect interface.
 	 */
@@ -804,9 +810,10 @@ int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
 		 * TLS context.
 		 */
 		skb->sk = psock->sk;
-		tcp_skb_bpf_redirect_clear(skb);
+		skb_dst_drop(skb);
+		skb_bpf_redirect_clear(skb);
 		ret = sk_psock_bpf_run(psock, prog, skb);
-		ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 		skb->sk = NULL;
 	}
 	sk_psock_tls_verdict_apply(skb, psock->sk, ret);
@@ -818,7 +825,6 @@ EXPORT_SYMBOL_GPL(sk_psock_tls_strp_read);
 static void sk_psock_verdict_apply(struct sk_psock *psock,
 				   struct sk_buff *skb, int verdict)
 {
-	struct tcp_skb_cb *tcp;
 	struct sock *sk_other;
 	int err = -EIO;
 
@@ -830,8 +836,7 @@ static void sk_psock_verdict_apply(struct sk_psock *psock,
 			goto out_free;
 		}
 
-		tcp = TCP_SKB_CB(skb);
-		tcp->bpf.flags |= BPF_F_INGRESS;
+		skb_bpf_set_ingress(skb);
 
 		/* If the queue is empty then we can submit directly
 		 * into the msg queue. If its not empty we have to
@@ -892,9 +897,10 @@ static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb)
 	skb_set_owner_r(skb, sk);
 	prog = READ_ONCE(psock->progs.skb_verdict);
 	if (likely(prog)) {
-		tcp_skb_bpf_redirect_clear(skb);
+		skb_dst_drop(skb);
+		skb_bpf_redirect_clear(skb);
 		ret = sk_psock_bpf_run(psock, prog, skb);
-		ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 	}
 	sk_psock_verdict_apply(psock, skb, ret);
 out:
@@ -1011,9 +1017,10 @@ static int sk_psock_verdict_recv(read_descriptor_t *desc, struct sk_buff *skb,
 	skb_set_owner_r(skb, sk);
 	prog = READ_ONCE(psock->progs.skb_verdict);
 	if (likely(prog)) {
-		tcp_skb_bpf_redirect_clear(skb);
+		skb_dst_drop(skb);
+		skb_bpf_redirect_clear(skb);
 		ret = sk_psock_bpf_run(psock, prog, skb);
-		ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 	}
 	sk_psock_verdict_apply(psock, skb, ret);
 out:
net/core/sock_map.c +2 -6
index 1a28a5c2c61e..dbfcd7006338 100644
@@ -657,7 +657,6 @@ const struct bpf_func_proto bpf_sock_map_update_proto = {
 BPF_CALL_4(bpf_sk_redirect_map, struct sk_buff *, skb,
 	   struct bpf_map *, map, u32, key, u64, flags)
 {
-	struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
 	struct sock *sk;
 
 	if (unlikely(flags & ~(BPF_F_INGRESS)))
@@ -667,8 +666,7 @@ BPF_CALL_4(bpf_sk_redirect_map, struct sk_buff *, skb,
 	if (unlikely(!sk || !sock_map_redirect_allowed(sk)))
 		return SK_DROP;
 
-	tcb->bpf.flags = flags;
-	tcb->bpf.sk_redir = sk;
+	skb_bpf_set_redir(skb, sk, flags & BPF_F_INGRESS);
 	return SK_PASS;
 }
 
@@ -1250,7 +1248,6 @@ const struct bpf_func_proto bpf_sock_hash_update_proto = {
 BPF_CALL_4(bpf_sk_redirect_hash, struct sk_buff *, skb,
 	   struct bpf_map *, map, void *, key, u64, flags)
 {
-	struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
 	struct sock *sk;
 
 	if (unlikely(flags & ~(BPF_F_INGRESS)))
@@ -1260,8 +1257,7 @@ BPF_CALL_4(bpf_sk_redirect_hash, struct sk_buff *, skb,
 	if (unlikely(!sk || !sock_map_redirect_allowed(sk)))
 		return SK_DROP;
 
-	tcb->bpf.flags = flags;
-	tcb->bpf.sk_redir = sk;
+	skb_bpf_set_redir(skb, sk, flags & BPF_F_INGRESS);
 	return SK_PASS;
 }
 
Signature · 1 line
2.25.1
#5 lore raw

Re: [Patch bpf-next v7 4/9] skmsg: move sk_redir from TCP_SKB_CB to skb

Quoted 7 lines · Cong Wang

From: Cong Wang <[email protected]>

Currently TCP_SKB_CB() is hard-coded in skmsg code, it certainly does not work for any other non-TCP protocols. We can move them to skb ext, but it introduces a memory allocation on fast path.

Quoted 7 lines

Fortunately, we only need to a word-size to store all the information, because the flags actually only contains 1 bit so can be just packed into the lowest bit of the "pointer", which is stored as unsigned long.

Inside struct sk_buff, '_skb_refdst' can be reused because skb dst is no longer needed after ->sk_data_ready() so we can just drop it.

Hi Cong Wang,

I saw this on patchwork:

include/linux/skbuff.h:932: warning: Function parameter or member '_sk_redir' not described in 'sk_buff'
New warnings added
0a1

Quoted 1 line

include/linux/skbuff.h:932: warning: Function parameter or member '_sk_redir' not described in 'sk_buff'

Per-file breakdown

Source: https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/

Maybe something to follow up on, I'm not sure what the conventions are here.

Quoted 247 lines
---
 include/linux/skbuff.h |  3 +++
 include/linux/skmsg.h  | 38 ++++++++++++++++++++++++++++++++++++++
 include/net/tcp.h      | 19 -------------------
 net/core/skmsg.c       | 31 +++++++++++++++++++------------
 net/core/sock_map.c    |  8 ++------
 5 files changed, 62 insertions(+), 37 deletions(-)
include/linux/skbuff.h +3 -0
index 6d0a33d1c0db..bd84f799c952 100644
@@ -755,6 +755,9 @@ struct sk_buff {
                        void            (*destructor)(struct sk_buff *skb);
                };
                struct list_head        tcp_tsorted_anchor;
+#ifdef CONFIG_NET_SOCK_MSG
+               unsigned long           _sk_redir;
+#endif
        };
 
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
include/linux/skmsg.h +38 -0
index 22e26f82de33..e0de45527bb6 100644
@@ -455,4 +455,42 @@ static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
                return false;
        return !!psock->saved_data_ready;
 }
+
+#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
+
+/* We only have one bit so far. */
+#define BPF_F_PTR_MASK ~(BPF_F_INGRESS)
+
+static inline bool skb_bpf_ingress(const struct sk_buff *skb)
+{
+       unsigned long sk_redir = skb->_sk_redir;
+
+       return sk_redir & BPF_F_INGRESS;
+}
+
+static inline void skb_bpf_set_ingress(struct sk_buff *skb)
+{
+       skb->_sk_redir |= BPF_F_INGRESS;
+}
+
+static inline void skb_bpf_set_redir(struct sk_buff *skb, struct sock *sk_redir,
+                                    bool ingress)
+{
+       skb->_sk_redir = (unsigned long)sk_redir;
+       if (ingress)
+               skb->_sk_redir |= BPF_F_INGRESS;
+}
+
+static inline struct sock *skb_bpf_redirect_fetch(const struct sk_buff *skb)
+{
+       unsigned long sk_redir = skb->_sk_redir;
+
+       return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
+}
+
+static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
+{
+       skb->_sk_redir = 0;
+}
+#endif /* CONFIG_NET_SOCK_MSG */
 #endif /* _LINUX_SKMSG_H */
include/net/tcp.h +0 -19
index 947ef5da6867..075de26f449d 100644
@@ -883,30 +883,11 @@ struct tcp_skb_cb {
                        struct inet6_skb_parm   h6;
 #endif
                } header;       /* For incoming skbs */
-               struct {
-                       __u32 flags;
-                       struct sock *sk_redir;
-               } bpf;
        };
 };
 
 #define TCP_SKB_CB(__skb)      ((struct tcp_skb_cb *)&((__skb)->cb[0]))
 
-static inline bool tcp_skb_bpf_ingress(const struct sk_buff *skb)
-{
-       return TCP_SKB_CB(skb)->bpf.flags & BPF_F_INGRESS;
-}
-
-static inline struct sock *tcp_skb_bpf_redirect_fetch(struct sk_buff *skb)
-{
-       return TCP_SKB_CB(skb)->bpf.sk_redir;
-}
-
-static inline void tcp_skb_bpf_redirect_clear(struct sk_buff *skb)
-{
-       TCP_SKB_CB(skb)->bpf.sk_redir = NULL;
-}
-
 extern const struct inet_connection_sock_af_ops ipv4_specific;
 
 #if IS_ENABLED(CONFIG_IPV6)
net/core/skmsg.c +19 -12
index 8822001ab3dc..409258367bea 100644
@@ -525,7 +525,8 @@ static void sk_psock_backlog(struct work_struct *work)
                len = skb->len;
                off = 0;
 start:
-               ingress = tcp_skb_bpf_ingress(skb);
+               ingress = skb_bpf_ingress(skb);
+               skb_bpf_redirect_clear(skb);
                do {
                        ret = -EIO;
                        if (likely(psock->sk->sk_socket))
@@ -631,7 +632,12 @@ void __sk_psock_purge_ingress_msg(struct sk_psock *psock)
 
 static void sk_psock_zap_ingress(struct sk_psock *psock)
 {
-       __skb_queue_purge(&psock->ingress_skb);
+       struct sk_buff *skb;
+
+       while ((skb = __skb_dequeue(&psock->ingress_skb)) != NULL) {
+               skb_bpf_redirect_clear(skb);
+               kfree_skb(skb);
+       }
        __sk_psock_purge_ingress_msg(psock);
 }
 
@@ -754,7 +760,7 @@ static void sk_psock_skb_redirect(struct sk_buff *skb)
        struct sk_psock *psock_other;
        struct sock *sk_other;
 
-       sk_other = tcp_skb_bpf_redirect_fetch(skb);
+       sk_other = skb_bpf_redirect_fetch(skb);
        /* This error is a buggy BPF program, it returned a redirect
         * return code, but then didn't set a redirect interface.
         */
@@ -804,9 +810,10 @@ int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
                 * TLS context.
                 */
                skb->sk = psock->sk;
-               tcp_skb_bpf_redirect_clear(skb);
+               skb_dst_drop(skb);
+               skb_bpf_redirect_clear(skb);
                ret = sk_psock_bpf_run(psock, prog, skb);
-               ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+               ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
                skb->sk = NULL;
        }
        sk_psock_tls_verdict_apply(skb, psock->sk, ret);
@@ -818,7 +825,6 @@ EXPORT_SYMBOL_GPL(sk_psock_tls_strp_read);
 static void sk_psock_verdict_apply(struct sk_psock *psock,
                                   struct sk_buff *skb, int verdict)
 {
-       struct tcp_skb_cb *tcp;
        struct sock *sk_other;
        int err = -EIO;
 
@@ -830,8 +836,7 @@ static void sk_psock_verdict_apply(struct sk_psock *psock,
                        goto out_free;
                }
 
-               tcp = TCP_SKB_CB(skb);
-               tcp->bpf.flags |= BPF_F_INGRESS;
+               skb_bpf_set_ingress(skb);
 
                /* If the queue is empty then we can submit directly
                 * into the msg queue. If its not empty we have to
@@ -892,9 +897,10 @@ static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb)
        skb_set_owner_r(skb, sk);
        prog = READ_ONCE(psock->progs.skb_verdict);
        if (likely(prog)) {
-               tcp_skb_bpf_redirect_clear(skb);
+               skb_dst_drop(skb);
+               skb_bpf_redirect_clear(skb);
                ret = sk_psock_bpf_run(psock, prog, skb);
-               ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+               ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
        }
        sk_psock_verdict_apply(psock, skb, ret);
 out:
@@ -1011,9 +1017,10 @@ static int sk_psock_verdict_recv(read_descriptor_t *desc, struct sk_buff *skb,
        skb_set_owner_r(skb, sk);
        prog = READ_ONCE(psock->progs.skb_verdict);
        if (likely(prog)) {
-               tcp_skb_bpf_redirect_clear(skb);
+               skb_dst_drop(skb);
+               skb_bpf_redirect_clear(skb);
                ret = sk_psock_bpf_run(psock, prog, skb);
-               ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
+               ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
        }
        sk_psock_verdict_apply(psock, skb, ret);
 out:
net/core/sock_map.c +2 -7
index 1a28a5c2c61e..dbfcd7006338 100644
@@ -657,7 +657,6 @@ const struct bpf_func_proto bpf_sock_map_update_proto = {
 BPF_CALL_4(bpf_sk_redirect_map, struct sk_buff *, skb,
           struct bpf_map *, map, u32, key, u64, flags)
 {
-       struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
        struct sock *sk;
 
        if (unlikely(flags & ~(BPF_F_INGRESS)))
@@ -667,8 +666,7 @@ BPF_CALL_4(bpf_sk_redirect_map, struct sk_buff *, skb,
        if (unlikely(!sk || !sock_map_redirect_allowed(sk)))
                return SK_DROP;
 
-       tcb->bpf.flags = flags;
-       tcb->bpf.sk_redir = sk;
+       skb_bpf_set_redir(skb, sk, flags & BPF_F_INGRESS);
        return SK_PASS;
 }
 
@@ -1250,7 +1248,6 @@ const struct bpf_func_proto bpf_sock_hash_update_proto = {
 BPF_CALL_4(bpf_sk_redirect_hash, struct sk_buff *, skb,
           struct bpf_map *, map, void *, key, u64, flags)
 {
-       struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
        struct sock *sk;
 
        if (unlikely(flags & ~(BPF_F_INGRESS)))
@@ -1260,8 +1257,7 @@ BPF_CALL_4(bpf_sk_redirect_hash, struct sk_buff *, skb,
        if (unlikely(!sk || !sock_map_redirect_allowed(sk)))
                return SK_DROP;
 
-       tcb->bpf.flags = flags;
-       tcb->bpf.sk_redir = sk;
+       skb_bpf_set_redir(skb, sk, flags & BPF_F_INGRESS);
        return SK_PASS;
 }
 
--

2.25.1

Signature · 4 lines
Lorenz Bauer  |  Systems Engineer
6th Floor, County Hall/The Riverside Building, SE1 7PB, UK

www.cloudflare.com
author
#6 lore raw

Re: [Patch bpf-next v7 4/9] skmsg: move sk_redir from TCP_SKB_CB to skb

Quoted 19 lines · Lorenz Bauer
Quoted 7 lines · Cong Wang

From: Cong Wang <[email protected]>

Currently TCP_SKB_CB() is hard-coded in skmsg code, it certainly does not work for any other non-TCP protocols. We can move them to skb ext, but it introduces a memory allocation on fast path.

Quoted 7 lines

Fortunately, we only need to a word-size to store all the information, because the flags actually only contains 1 bit so can be just packed into the lowest bit of the "pointer", which is stored as unsigned long.

Inside struct sk_buff, '_skb_refdst' can be reused because skb dst is no longer needed after ->sk_data_ready() so we can just drop it.

Hi Cong Wang,

Quoted 8 lines

I saw this on patchwork:

include/linux/skbuff.h:932: warning: Function parameter or member '_sk_redir' not described in 'sk_buff'
New warnings added
0a1

Quoted 1 line

include/linux/skbuff.h:932: warning: Function parameter or member '_sk_redir' not described in 'sk_buff'

Per-file breakdown

Ah, I didn't know the function doc is mandatory now.

Quoted 4 lines

Source: https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/

Maybe something to follow up on, I'm not sure what the conventions are here.

It is already merged, so we definitely need a one-line followup fix.

include/linux/skbuff.h +1 -0
index bd84f799c952..0503c917d773 100644
@@ -656,6 +656,7 @@ typedef unsigned char *sk_buff_data_t;
  *     @protocol: Packet protocol from driver
  *     @destructor: Destruct function
  *     @tcp_tsorted_anchor: list structure for TCP (tp->tsorted_sent_queue)
+ *     @_sk_redir: socket redirection information for skmsg
  *     @_nfct: Associated connection, if any (with nfctinfo bits)
  *     @nf_bridge: Saved data about a bridged frame - see br_netfilter.c
  *     @skb_iif: ifindex of device we arrived on

Thanks.

author
#7 lore raw

PATCH v7 5/9 sock_map: rename skb_parser and skb_verdict

In reply to Cong Wang

From: Cong Wang <[email protected]>

These two eBPF programs are tied to BPF_SK_SKB_STREAM_PARSER and BPF_SK_SKB_STREAM_VERDICT, rename them to reflect the fact they are only used for TCP. And save the name 'skb_verdict' for general use later.

---
 include/linux/skmsg.h                         |  8 +--
 net/core/skmsg.c                              | 14 ++---
 net/core/sock_map.c                           | 60 +++++++++----------
 .../selftests/bpf/prog_tests/sockmap_listen.c |  8 +--
 .../selftests/bpf/progs/test_sockmap_listen.c |  4 +-
 5 files changed, 47 insertions(+), 47 deletions(-)
include/linux/skmsg.h +4 -4
index e0de45527bb6..d9f6ec4a9cf2 100644
@@ -56,8 +56,8 @@ struct sk_msg {
 
 struct sk_psock_progs {
 	struct bpf_prog			*msg_parser;
-	struct bpf_prog			*skb_parser;
-	struct bpf_prog			*skb_verdict;
+	struct bpf_prog			*stream_parser;
+	struct bpf_prog			*stream_verdict;
 };
 
 enum sk_psock_state_bits {
@@ -443,8 +443,8 @@ static inline int psock_replace_prog(struct bpf_prog **pprog,
 static inline void psock_progs_drop(struct sk_psock_progs *progs)
 {
 	psock_set_prog(&progs->msg_parser, NULL);
-	psock_set_prog(&progs->skb_parser, NULL);
-	psock_set_prog(&progs->skb_verdict, NULL);
+	psock_set_prog(&progs->stream_parser, NULL);
+	psock_set_prog(&progs->stream_verdict, NULL);
 }
 
 int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb);
net/core/skmsg.c +7 -7
index 409258367bea..35f9caa3b125 100644
@@ -691,9 +691,9 @@ void sk_psock_drop(struct sock *sk, struct sk_psock *psock)
 	write_lock_bh(&sk->sk_callback_lock);
 	sk_psock_restore_proto(sk, psock);
 	rcu_assign_sk_user_data(sk, NULL);
-	if (psock->progs.skb_parser)
+	if (psock->progs.stream_parser)
 		sk_psock_stop_strp(sk, psock);
-	else if (psock->progs.skb_verdict)
+	else if (psock->progs.stream_verdict)
 		sk_psock_stop_verdict(sk, psock);
 	write_unlock_bh(&sk->sk_callback_lock);
 	sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED);
@@ -803,7 +803,7 @@ int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
 	int ret = __SK_PASS;
 
 	rcu_read_lock();
-	prog = READ_ONCE(psock->progs.skb_verdict);
+	prog = READ_ONCE(psock->progs.stream_verdict);
 	if (likely(prog)) {
 		/* We skip full set_owner_r here because if we do a SK_PASS
 		 * or SK_DROP we can skip skb memory accounting and use the
@@ -895,7 +895,7 @@ static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb)
 		goto out;
 	}
 	skb_set_owner_r(skb, sk);
-	prog = READ_ONCE(psock->progs.skb_verdict);
+	prog = READ_ONCE(psock->progs.stream_verdict);
 	if (likely(prog)) {
 		skb_dst_drop(skb);
 		skb_bpf_redirect_clear(skb);
@@ -919,7 +919,7 @@ static int sk_psock_strp_parse(struct strparser *strp, struct sk_buff *skb)
 	int ret = skb->len;
 
 	rcu_read_lock();
-	prog = READ_ONCE(psock->progs.skb_parser);
+	prog = READ_ONCE(psock->progs.stream_parser);
 	if (likely(prog)) {
 		skb->sk = psock->sk;
 		ret = sk_psock_bpf_run(psock, prog, skb);
@@ -982,7 +982,7 @@ void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
 static void sk_psock_done_strp(struct sk_psock *psock)
 {
 	/* Parser has been stopped */
-	if (psock->progs.skb_parser)
+	if (psock->progs.stream_parser)
 		strp_done(&psock->strp);
 }
 #else
@@ -1015,7 +1015,7 @@ static int sk_psock_verdict_recv(read_descriptor_t *desc, struct sk_buff *skb,
 		goto out;
 	}
 	skb_set_owner_r(skb, sk);
-	prog = READ_ONCE(psock->progs.skb_verdict);
+	prog = READ_ONCE(psock->progs.stream_verdict);
 	if (likely(prog)) {
 		skb_dst_drop(skb);
 		skb_bpf_redirect_clear(skb);
net/core/sock_map.c +30 -30

115 lines not on this page · raw message

tools/testing/selftests/bpf/prog_tests/sockmap_listen.c +4 -4
index d7d65a700799..c26e6bf05e49 100644
@@ -1014,8 +1014,8 @@ static void test_skb_redir_to_connected(struct test_sockmap_listen *skel,
 					struct bpf_map *inner_map, int family,
 					int sotype)
 {
-	int verdict = bpf_program__fd(skel->progs.prog_skb_verdict);
-	int parser = bpf_program__fd(skel->progs.prog_skb_parser);
+	int verdict = bpf_program__fd(skel->progs.prog_stream_verdict);
+	int parser = bpf_program__fd(skel->progs.prog_stream_parser);
 	int verdict_map = bpf_map__fd(skel->maps.verdict_map);
 	int sock_map = bpf_map__fd(inner_map);
 	int err;
@@ -1125,8 +1125,8 @@ static void test_skb_redir_to_listening(struct test_sockmap_listen *skel,
 					struct bpf_map *inner_map, int family,
 					int sotype)
 {
-	int verdict = bpf_program__fd(skel->progs.prog_skb_verdict);
-	int parser = bpf_program__fd(skel->progs.prog_skb_parser);
+	int verdict = bpf_program__fd(skel->progs.prog_stream_verdict);
+	int parser = bpf_program__fd(skel->progs.prog_stream_parser);
 	int verdict_map = bpf_map__fd(skel->maps.verdict_map);
 	int sock_map = bpf_map__fd(inner_map);
 	int err;
tools/testing/selftests/bpf/progs/test_sockmap_listen.c +2 -2
index a3a366c57ce1..fa221141e9c1 100644
@@ -31,13 +31,13 @@ struct {
 static volatile bool test_sockmap; /* toggled by user-space */
 
 SEC("sk_skb/stream_parser")
-int prog_skb_parser(struct __sk_buff *skb)
+int prog_stream_parser(struct __sk_buff *skb)
 {
 	return skb->len;
 }
 
 SEC("sk_skb/stream_verdict")
-int prog_skb_verdict(struct __sk_buff *skb)
+int prog_stream_verdict(struct __sk_buff *skb)
 {
 	unsigned int *count;
 	__u32 zero = 0;
Signature · 1 line
2.25.1
author
#8 lore raw

PATCH v7 6/9 sock_map: make sock_map_prog_update() static

In reply to Cong Wang

From: Cong Wang <[email protected]>

It is only used within sock_map.c so can become static.

---
 include/linux/bpf.h | 9 ---------
 net/core/sock_map.c | 7 +++++--
 2 files changed, 5 insertions(+), 11 deletions(-)
include/linux/bpf.h +0 -9
index 813f30ef44ff..521b75a81aa6 100644
@@ -1769,8 +1769,6 @@ static inline void bpf_map_offload_map_free(struct bpf_map *map)
 #endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */
 
 #if defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL)
-int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
-			 struct bpf_prog *old, u32 which);
 int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
 int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
 int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
@@ -1788,13 +1786,6 @@ static inline void bpf_sk_reuseport_detach(struct sock *sk)
 }
 
 #ifdef CONFIG_BPF_SYSCALL
-static inline int sock_map_prog_update(struct bpf_map *map,
-				       struct bpf_prog *prog,
-				       struct bpf_prog *old, u32 which)
-{
-	return -EOPNOTSUPP;
-}
-
 static inline int sock_map_get_from_fd(const union bpf_attr *attr,
 				       struct bpf_prog *prog)
 {
net/core/sock_map.c +5 -2
index 69785070f02d..dd53a7771d7e 100644
@@ -24,6 +24,9 @@ struct bpf_stab {
 #define SOCK_CREATE_FLAG_MASK				\
 	(BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
 
+static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
+				struct bpf_prog *old, u32 which);
+
 static struct bpf_map *sock_map_alloc(union bpf_attr *attr)
 {
 	struct bpf_stab *stab;
@@ -1444,8 +1447,8 @@ static struct sk_psock_progs *sock_map_progs(struct bpf_map *map)
 	return NULL;
 }
 
-int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
-			 struct bpf_prog *old, u32 which)
+static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
+				struct bpf_prog *old, u32 which)
 {
 	struct sk_psock_progs *progs = sock_map_progs(map);
 	struct bpf_prog **pprog;
Signature · 1 line
2.25.1
author
#9 lore raw

PATCH v7 7/9 skmsg: make __sk_psock_purge_ingress_msg() static

In reply to Cong Wang

From: Cong Wang <[email protected]>

It is only used within skmsg.c so can become static.

Cc Daniel Borkmann <[email protected]>
Cc Lorenz Bauer <[email protected]>
Cc John Fastabend <[email protected]>
---
 include/linux/skmsg.h | 2 --
 net/core/skmsg.c      | 2 +-
 2 files changed, 1 insertion(+), 3 deletions(-)
include/linux/skmsg.h +0 -2
index d9f6ec4a9cf2..676d48e08159 100644
@@ -340,8 +340,6 @@ static inline void sk_psock_free_link(struct sk_psock_link *link)
 
 struct sk_psock_link *sk_psock_link_pop(struct sk_psock *psock);
 
-void __sk_psock_purge_ingress_msg(struct sk_psock *psock);
-
 static inline void sk_psock_cork_free(struct sk_psock *psock)
 {
 	if (psock->cork) {
net/core/skmsg.c +1 -1
index 35f9caa3b125..46e29d2c0c48 100644
@@ -619,7 +619,7 @@ struct sk_psock_link *sk_psock_link_pop(struct sk_psock *psock)
 	return link;
 }
 
-void __sk_psock_purge_ingress_msg(struct sk_psock *psock)
+static void __sk_psock_purge_ingress_msg(struct sk_psock *psock)
 {
 	struct sk_msg *msg, *tmp;
 
Signature · 1 line
2.25.1
author
#10 lore raw

PATCH v7 8/9 skmsg: get rid of sk_psock_bpf_run()

In reply to Cong Wang

From: Cong Wang <[email protected]>

It is now nearly identical to bpf_prog_run_pin_on_cpu() and it has an unused parameter 'psock', so we can just get rid of it and call bpf_prog_run_pin_on_cpu() directly.

Cc Daniel Borkmann <[email protected]>
Cc Lorenz Bauer <[email protected]>
Cc John Fastabend <[email protected]>
---
 net/core/skmsg.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)
net/core/skmsg.c +4 -10
index 46e29d2c0c48..07f54015238a 100644
@@ -749,12 +749,6 @@ int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock,
 }
 EXPORT_SYMBOL_GPL(sk_psock_msg_verdict);
 
-static int sk_psock_bpf_run(struct sk_psock *psock, struct bpf_prog *prog,
-			    struct sk_buff *skb)
-{
-	return bpf_prog_run_pin_on_cpu(prog, skb);
-}
-
 static void sk_psock_skb_redirect(struct sk_buff *skb)
 {
 	struct sk_psock *psock_other;
@@ -812,7 +806,7 @@ int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
 		skb->sk = psock->sk;
 		skb_dst_drop(skb);
 		skb_bpf_redirect_clear(skb);
-		ret = sk_psock_bpf_run(psock, prog, skb);
+		ret = bpf_prog_run_pin_on_cpu(prog, skb);
 		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 		skb->sk = NULL;
 	}
@@ -899,7 +893,7 @@ static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb)
 	if (likely(prog)) {
 		skb_dst_drop(skb);
 		skb_bpf_redirect_clear(skb);
-		ret = sk_psock_bpf_run(psock, prog, skb);
+		ret = bpf_prog_run_pin_on_cpu(prog, skb);
 		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 	}
 	sk_psock_verdict_apply(psock, skb, ret);
@@ -922,7 +916,7 @@ static int sk_psock_strp_parse(struct strparser *strp, struct sk_buff *skb)
 	prog = READ_ONCE(psock->progs.stream_parser);
 	if (likely(prog)) {
 		skb->sk = psock->sk;
-		ret = sk_psock_bpf_run(psock, prog, skb);
+		ret = bpf_prog_run_pin_on_cpu(prog, skb);
 		skb->sk = NULL;
 	}
 	rcu_read_unlock();
@@ -1019,7 +1013,7 @@ static int sk_psock_verdict_recv(read_descriptor_t *desc, struct sk_buff *skb,
 	if (likely(prog)) {
 		skb_dst_drop(skb);
 		skb_bpf_redirect_clear(skb);
-		ret = sk_psock_bpf_run(psock, prog, skb);
+		ret = bpf_prog_run_pin_on_cpu(prog, skb);
 		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
 	}
 	sk_psock_verdict_apply(psock, skb, ret);
Signature · 1 line
2.25.1
author
#11 lore raw

PATCH v7 9/9 skmsg: remove unused sk_psock_stop() declaration

In reply to Cong Wang

From: Cong Wang <[email protected]>

It is not defined or used anywhere.

Cc Daniel Borkmann <[email protected]>
Cc Lorenz Bauer <[email protected]>
Cc John Fastabend <[email protected]>
Cc Jakub Sitnicki <[email protected]>
---
 include/linux/skmsg.h | 1 -
 1 file changed, 1 deletion(-)
include/linux/skmsg.h +0 -1
index 676d48e08159..6c09d94be2e9 100644
@@ -400,7 +400,6 @@ static inline struct sk_psock *sk_psock_get(struct sock *sk)
 	return psock;
 }
 
-void sk_psock_stop(struct sock *sk, struct sk_psock *psock);
 void sk_psock_drop(struct sock *sk, struct sk_psock *psock);
 
 static inline void sk_psock_put(struct sock *sk, struct sk_psock *psock)
Signature · 1 line
2.25.1
maintainer Acked-by on v7 9/9
#12 lore raw

Re: [Patch bpf-next v7 9/9] skmsg: remove unused sk_psock_stop() declaration

Quoted 10 lines · On Tue, 23 Feb 2021 10:49:34 -0800 Cong Wang <[email protected]> wrote:

From: Cong Wang <[email protected]>

It is not defined or used anywhere.

Cc Daniel Borkmann <[email protected]>
Cc Lorenz Bauer <[email protected]>
Cc John Fastabend <[email protected]>
Cc Jakub Sitnicki <[email protected]>

---

#13 lore raw
In reply to Cong Wang
Quoted 15 lines · Cong Wang

From: Cong Wang <[email protected]>

This patchset is the first series of patches separated out from the original large patchset, to make reviews easier. This patchset does not add any new feature or change any functionality but merely cleans up the existing sockmap and skmsg code and refactors it, to prepare for the patches followed up. This passed all BPF selftests.

To see the big picture, the original whole patchset is available on github: https://github.com/congwang/linux/tree/sockmap

and this patchset is also available on github:
https://github.com/congwang/linux/tree/sockmap1

Quoted 6 lines

---
v7: add 1 trivial cleanup patch
define a mask for sk_redir
fix CONFIG_BPF_SYSCALL in include/net/udp.h make sk_psock_done_strp() static
move skb_bpf_redirect_clear() to sk_psock_backlog()

Applied. Thanks