sock_map: clean up and refactor code for BPF_SK_SKB_VERDICT
Subsystems (11)
v4 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
---
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 (5):
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
include/linux/bpf.h | 20 +- include/linux/bpf_types.h | 2 - include/linux/skbuff.h | 3 + include/linux/skmsg.h | 80 ++++++-- include/net/tcp.h | 38 +--- include/net/udp.h | 4 +- init/Kconfig | 1 + net/Kconfig | 6 +- net/core/Makefile | 2 +- net/core/filter.c | 48 +++-- net/core/skmsg.c | 191 +++++++++--------- net/core/sock_map.c | 70 ++++--- net/ipv4/Makefile | 2 +- net/ipv4/tcp_bpf.c | 4 +- net/ipv4/udp_bpf.c | 2 + .../selftests/bpf/prog_tests/sockmap_listen.c | 8 +- .../selftests/bpf/progs/test_sockmap_listen.c | 4 +- 17 files changed, 253 insertions(+), 232 deletions(-)
Signature · 1 line
2.25.1
PATCH v4 1/5 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 leave CONFIG_NET_SOCK_MSG untouched, as it is used by non-sockmap cases.
--- include/linux/bpf.h | 20 +++--- include/linux/bpf_types.h | 2 - include/linux/skmsg.h | 22 ++++++ include/net/tcp.h | 13 ++-- include/net/udp.h | 4 +- init/Kconfig | 1 + net/Kconfig | 6 +- net/core/Makefile | 2 +- net/core/skmsg.c | 139 ++++++++++++++++++++------------------ net/core/sock_map.c | 2 + net/ipv4/Makefile | 2 +- net/ipv4/tcp_bpf.c | 4 +- net/ipv4/udp_bpf.c | 2 + 13 files changed, 121 insertions(+), 98 deletions(-)
include/linux/bpf.h +9 -11
@@ -1741,6 +1741,14 @@ static inline bool bpf_map_is_dev_bound(struct bpf_map *map)struct bpf_map *bpf_map_offload_map_alloc (union bpf_attr *attr);void bpf_map_offload_map_free (struct bpf_map *map); + +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); +void sock_map_unhash (struct sock *sk); +void sock_map_close (struct sock *sk,long timeout);#else static inline int bpf_prog_offload_init (struct bpf_prog *prog,union bpf_attr *attr)
@@ -1766,17 +1774,7 @@ static inline struct bpf_map *bpf_map_offload_map_alloc(union bpf_attr *attr)static inline void bpf_map_offload_map_free (struct bpf_map *map) { } -#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */ -#if defined(CONFIG_BPF_STREAM_PARSER ) -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); -void sock_map_unhash (struct sock *sk); -void sock_map_close (struct sock *sk,long timeout); -#else static inline int sock_map_prog_update (struct bpf_map *map,struct bpf_prog *prog,struct bpf_prog *old,u32 which)
@@ -1801,7 +1799,7 @@ static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void {return -EOPNOTSUPP ; } -#endif /* CONFIG_BPF_STREAM_PARSER */ +#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */ #if defined(CONFIG_INET ) && defined(CONFIG_BPF_SYSCALL )void bpf_sk_reuseport_detach (struct sock *sk);
include/linux/bpf_types.h +0 -2
@@ -103,10 +103,8 @@ 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)
include/linux/skmsg.h +22 -0
@@ -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 +307,29 @@ 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); +void sk_psock_done_strp (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) +{ +} + +static inline void sk_psock_done_strp (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 +5 -8
@@ -2204,23 +2204,20 @@ 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 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 */ -#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); +#else +static inline void tcp_bpf_clone (const struct sock *sk,struct sock *newsk) +{ +}#endif /* CONFIG_NET_SOCK_MSG */ #ifdef CONFIG_CGROUP_BPF
include/net/udp.h +2 -2
@@ -511,9 +511,9 @@ static inline struct sk_buff *udp_rcv_segment(struct sock *sk, return segs; } -#ifdef CONFIG_BPF_STREAM_PARSER +#ifdef CONFIG_NET_SOCK_MSG struct sk_psock ;struct proto *udp_bpf_get_proto (struct sock *sk,struct sk_psock *psock); -#endif /* BPF_STREAM_PARSER */ +#endif /* CONFIG_NET_SOCK_MSG */ #endif /* _UDP_H */
init/Kconfig +1 -0
@@ -1703,6 +1703,7 @@ config BPF_SYSCALL select BPF select IRQ_WORK select TASKS_TRACE_RCU + select NET_SOCK_MSG if NET default n help Enable the bpf() system call that allows to manipulate eBPF
net/Kconfig +1 -5
@@ -313,13 +313,9 @@ 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 +1 -1
@@ -28,10 +28,10 @@ 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 +obj-$(CONFIG_BPF_SYSCALL) += sock_map.o obj-$(CONFIG_BPF_SYSCALL) += bpf_sk_storage.o
net/core/skmsg.c +73 -66
181 lines not on this page · raw message
net/core/sock_map.c +2 -0
@@ -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
@@ -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_NET_SOCK_MSG) += 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
@@ -229,7 +229,6 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, }EXPORT_SYMBOL_GPL (tcp_bpf_sendmsg_redir); -#ifdef CONFIG_BPF_STREAM_PARSER static bool tcp_bpf_stream_read (const struct sock *sk) {struct sk_psock *psock;
@@ -561,8 +560,10 @@ static void tcp_bpf_rebuild_protos(struct proto prot[TCP_BPF_NUM_CFGS], struct proto *base) { prot[TCP_BPF_BASE ] = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot[TCP_BPF_BASE ].unhash = sock_map_unhash; prot[TCP_BPF_BASE ].close = sock_map_close; +#endif prot[TCP_BPF_BASE ].recvmsg = tcp_bpf_recvmsg; prot[TCP_BPF_BASE ].stream_memory_read = tcp_bpf_stream_read;
@@ -629,4 +630,3 @@ 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 */
net/ipv4/udp_bpf.c +2 -0
@@ -18,8 +18,10 @@ static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS];static void udp_bpf_rebuild_protos (struct proto *prot,const struct proto *base) { *prot = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot->unhash = sock_map_unhash; prot->close = sock_map_close; +#endif }static void udp_bpf_check_v6_needs_rebuild (struct proto *ops)
Signature · 1 line
2.25.1
Jakub Sitnicki
maintainerRe: [Patch bpf-next v4 1/5] bpf: clean up sockmap related Kconfigs
Quoted 11 lines · Cong Wang
From: Cong Wang <[email protected]>
As suggested by John, clean up sockmap related Kconfigs:
Reduce the scope of
CONFIG_BPF_STREAM_PARSERdown to TCP stream parser, to reflect its name.Make the rest sockmap code simply depend on
CONFIG_BPF_SYSCALL. And leaveCONFIG_NET_SOCK_MSGuntouched, as it is used by non-sockmap cases.
Quoted 6 lines
Cc Daniel Borkmann <[email protected]>Cc Jakub Sitnicki <[email protected]>Reviewed-by Lorenz Bauer <[email protected]>Acked-by John Fastabend <[email protected]>Signed-off-by Cong Wang <[email protected]>---
Sorry for the delay. There's a lot happening here. Took me a while to dig through it.
I have a couple of nit-picks, which easily can be addressed as follow-ups, and one comment.
sock_map_prog_update and sk_psock_done_strp are only used in net/core/sock_map.c and can be static.
[...]
Quoted 24 lines
net/ipv4/tcp_bpf.c +2
-1@@ -229,7 +229,6 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, }EXPORT_SYMBOL_GPL (tcp_bpf_sendmsg_redir); -#ifdef CONFIG_BPF_STREAM_PARSER static bool tcp_bpf_stream_read (const struct sock *sk) {struct sk_psock *psock;@@ -561,8 +560,10 @@ static void tcp_bpf_rebuild_protos(struct proto prot[TCP_BPF_NUM_CFGS], struct proto *base) { prot[TCP_BPF_BASE ] = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot[TCP_BPF_BASE ].unhash = sock_map_unhash; prot[TCP_BPF_BASE ].close = sock_map_close; +#endif prot[TCP_BPF_BASE ].recvmsg = tcp_bpf_recvmsg; prot[TCP_BPF_BASE ].stream_memory_read = tcp_bpf_stream_read;
net/ipv4/tcp_bpf.c void tcp_bpf_clone(const struct sock *sk, struct sock *newsk) · line 632
Quoted 4 lines
net/ipv4/tcp_bpf.c +0
-1@@ -629,4 +630,3 @@ 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 */
net/core/sock_map.o now is built only when CONFIG_BPF_SYSCALL is set. While tcp_bpf_get_proto is only called from net/core/sock_map.o.
Seems there is no sense in compiling tcp_bpf_get_proto, and everything it depends on which was enclosed by CONFIG_BPF_STREAM_PARSER check, when CONFIG_BPF_SYSCALL is unset.
Quoted 5 lines
net/ipv4/udp_bpf.c +0
-0
net/ipv4/udp_bpf.c static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS]; · line 27
Quoted 10 lines
net/ipv4/udp_bpf.c +2
-0@@ -18,8 +18,10 @@ static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS];static void udp_bpf_rebuild_protos (struct proto *prot,const struct proto *base) { *prot = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot->unhash = sock_map_unhash; prot->close = sock_map_close; +#endif }static void udp_bpf_check_v6_needs_rebuild (struct proto *ops)
Same situation here but for udp_bpf_get_proto.
Re: [Patch bpf-next v4 1/5] bpf: clean up sockmap related Kconfigs
Quoted 20 lines · Jakub Sitnicki
Quoted 11 lines · Cong Wang
From: Cong Wang <[email protected]>
As suggested by John, clean up sockmap related Kconfigs:
Reduce the scope of
CONFIG_BPF_STREAM_PARSERdown to TCP stream parser, to reflect its name.Make the rest sockmap code simply depend on
CONFIG_BPF_SYSCALL. And leaveCONFIG_NET_SOCK_MSGuntouched, as it is used by non-sockmap cases.Quoted 6 lines
Cc Daniel Borkmann <[email protected]>Cc Jakub Sitnicki <[email protected]>Reviewed-by Lorenz Bauer <[email protected]>Acked-by John Fastabend <[email protected]>Signed-off-by Cong Wang <[email protected]>---
Quoted 5 lines
Sorry for the delay. There's a lot happening here. Took me a while to dig through it.
I have a couple of nit-picks, which easily can be addressed as follow-ups, and one comment.
No problem, it is not merged, so V5 is definitely not a problem.
Quoted 3 lines
sock_map_prog_updateandsk_psock_done_strpare only used innet/core/sock_map.cand can be static.
1. This seems to be unrelated to this patch? But I am still happy to address it.
2. sk_psock_done_strp() is in skmsg.c, hence why it is non-static. And I believe it fits in skmsg.c better than in sock_map.c, because it operates on psock rather than sock_map itself.
So, I can make sock_map_prog_update() static in a separate patch and carry it in V5.
Quoted 35 lines
[...]
Quoted 24 lines
net/ipv4/tcp_bpf.c +2
-1@@ -229,7 +229,6 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, }EXPORT_SYMBOL_GPL (tcp_bpf_sendmsg_redir); -#ifdef CONFIG_BPF_STREAM_PARSER static bool tcp_bpf_stream_read (const struct sock *sk) {struct sk_psock *psock;@@ -561,8 +560,10 @@ static void tcp_bpf_rebuild_protos(struct proto prot[TCP_BPF_NUM_CFGS], struct proto *base) { prot[TCP_BPF_BASE ] = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot[TCP_BPF_BASE ].unhash = sock_map_unhash; prot[TCP_BPF_BASE ].close = sock_map_close; +#endif prot[TCP_BPF_BASE ].recvmsg = tcp_bpf_recvmsg; prot[TCP_BPF_BASE ].stream_memory_read = tcp_bpf_stream_read;net/ipv4/tcp_bpf.c void tcp_bpf_clone(const struct sock *sk, struct sock *newsk) · line 632
Quoted 4 lines
net/ipv4/tcp_bpf.c +0
-1@@ -629,4 +630,3 @@ 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 */
net/core/sock_map.onow is built only whenCONFIG_BPF_SYSCALLis set. Whiletcp_bpf_get_protois only called fromnet/core/sock_map.o.
Quoted 3 lines
Seems there is no sense in compiling
tcp_bpf_get_proto, and everything it depends on which was enclosed byCONFIG_BPF_STREAM_PARSERcheck, whenCONFIG_BPF_SYSCALLis unset.
I can try but I am definitely not sure whether kTLS is happy about it, clearly kTLS at least uses __tcp_bpf_recvmsg() and tcp_bpf_sendmsg_redir().
Quoted 15 lines
Quoted 15 lines
net/ipv4/udp_bpf.c +2
-0@@ -18,8 +18,10 @@ static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS];static void udp_bpf_rebuild_protos (struct proto *prot,const struct proto *base) { *prot = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot->unhash = sock_map_unhash; prot->close = sock_map_close; +#endif }
Quoted 3 lines
net/ipv4/udp_bpf.c static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS]; · line 27
Quoted 1 line
net/ipv4/udp_bpf.c +0
-0static void udp_bpf_check_v6_needs_rebuild (struct proto *ops)Same situation here but for
udp_bpf_get_proto.
UDP is different, as kTLS certainly doesn't and won't use it. I think udp_bpf.c can be just put under CONFIG_BPF_SYSCALL.
Thanks.
Jakub Sitnicki
maintainerRe: [Patch bpf-next v4 1/5] bpf: clean up sockmap related Kconfigs
Quoted 27 lines · Cong Wang
Quoted 20 lines · Jakub Sitnicki
Quoted 11 lines · Cong Wang
From: Cong Wang <[email protected]>
As suggested by John, clean up sockmap related Kconfigs:
Reduce the scope of
CONFIG_BPF_STREAM_PARSERdown to TCP stream parser, to reflect its name.Make the rest sockmap code simply depend on
CONFIG_BPF_SYSCALL. And leaveCONFIG_NET_SOCK_MSGuntouched, as it is used by non-sockmap cases.Quoted 6 lines
Cc Daniel Borkmann <[email protected]>Cc Jakub Sitnicki <[email protected]>Reviewed-by Lorenz Bauer <[email protected]>Acked-by John Fastabend <[email protected]>Signed-off-by Cong Wang <[email protected]>---
Quoted 5 lines
Sorry for the delay. There's a lot happening here. Took me a while to dig through it.
I have a couple of nit-picks, which easily can be addressed as follow-ups, and one comment.
Quoted 8 lines
No problem, it is not merged, so V5 is definitely not a problem.
Quoted 3 lines
sock_map_prog_updateandsk_psock_done_strpare only used innet/core/sock_map.cand can be static.1. This seems to be unrelated to this patch? But I am still happy to address it.
Completely unrelated. Just a nit-pick. Feel free to ignore.
Quoted 3 lines
2.
sk_psock_done_strp()is inskmsg.c, hence why it is non-static. And I believe it fits inskmsg.cbetter than insock_map.c, because it operates on psock rather thansock_mapitself.
I wrote that sk_psock_done_strp is used only in net/core/sock_map.c, while I should have written that it's used only in net/core/skmsg.c.
Sorry, a mistake when copying from my notes. Also, feel free to ignore.
Quoted 2 lines
So, I can make
sock_map_prog_update()static in a separate patch and carry it in V5.
Completely up to you. I don't insist.
Quoted 35 lines
Quoted 35 lines
[...]
Quoted 24 lines
net/ipv4/tcp_bpf.c +2
-1@@ -229,7 +229,6 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, }EXPORT_SYMBOL_GPL (tcp_bpf_sendmsg_redir); -#ifdef CONFIG_BPF_STREAM_PARSER static bool tcp_bpf_stream_read (const struct sock *sk) {struct sk_psock *psock;@@ -561,8 +560,10 @@ static void tcp_bpf_rebuild_protos(struct proto prot[TCP_BPF_NUM_CFGS], struct proto *base) { prot[TCP_BPF_BASE ] = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot[TCP_BPF_BASE ].unhash = sock_map_unhash; prot[TCP_BPF_BASE ].close = sock_map_close; +#endif prot[TCP_BPF_BASE ].recvmsg = tcp_bpf_recvmsg; prot[TCP_BPF_BASE ].stream_memory_read = tcp_bpf_stream_read;net/ipv4/tcp_bpf.c void tcp_bpf_clone(const struct sock *sk, struct sock *newsk) · line 632
Quoted 4 lines
net/ipv4/tcp_bpf.c +0
-1@@ -629,4 +630,3 @@ 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 */
net/core/sock_map.onow is built only whenCONFIG_BPF_SYSCALLis set. Whiletcp_bpf_get_protois only called fromnet/core/sock_map.o.
Quoted 7 lines
Quoted 3 lines
Seems there is no sense in compiling
tcp_bpf_get_proto, and everything it depends on which was enclosed byCONFIG_BPF_STREAM_PARSERcheck, whenCONFIG_BPF_SYSCALLis unset.I can try but I am definitely not sure whether kTLS is happy about it, clearly kTLS at least uses
__tcp_bpf_recvmsg()andtcp_bpf_sendmsg_redir().
I think kTLS will be fine because that's the situation today.
__tcp_bpf_recvmsg and tcp_bpf_sendmsg_redir need to be left out, naturally, is it is now.
(Although I think they could event be stubbed out. But that would be unrelated to this change.)
After all how would we end up on code path in kTLS that utilizes sockmap API, if sockmap can't be created when CONFIG_BPF_SYSCALL is unset.
Quoted 24 lines
Quoted 16 lines
Quoted 16 lines
net/ipv4/udp_bpf.c +2
-0@@ -18,8 +18,10 @@ static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS];static void udp_bpf_rebuild_protos (struct proto *prot,const struct proto *base) { *prot = *base; +#if defined(CONFIG_BPF_SYSCALL ) prot->unhash = sock_map_unhash; prot->close = sock_map_close; +#endif }Quoted 3 lines
net/ipv4/udp_bpf.c static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS]; · line 27
Quoted 1 line
net/ipv4/udp_bpf.c +0
-0static void udp_bpf_check_v6_needs_rebuild (struct proto *ops)Same situation here but for
udp_bpf_get_proto.UDP is different, as kTLS certainly doesn't and won't use it. I think
udp_bpf.ccan be just put underCONFIG_BPF_SYSCALL.Thanks.
PATCH v4 2/5 skmsg: get rid of struct sk_psock_parser
In reply to Cong WangFrom: 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
@@ -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;
@@ -422,8 +417,8 @@ 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); }
@@ -462,6 +457,6 @@ 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
@@ -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
PATCH v4 3/5 bpf: compute data_end dynamically with JIT code
In reply to Cong WangFrom: 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
@@ -885,18 +885,12 @@ 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
@@ -744,7 +744,6 @@ 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
PATCH v4 4/5 skmsg: move sk_redir from TCP_SKB_CB to skb
In reply to Cong WangFrom: 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 | 35 +++++++++++++++++++++++++++++++++++ include/net/tcp.h | 19 ------------------- net/core/skmsg.c | 32 ++++++++++++++++++++------------ net/core/sock_map.c | 8 ++------ 5 files changed, 60 insertions(+), 37 deletions(-)
include/linux/skbuff.h +3 -0
@@ -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 +35 -0
@@ -459,4 +459,39 @@ 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 ) +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; + + sk_redir &= ~0x1UL ; +return (struct sock *)sk_redir; +} + +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
@@ -882,30 +882,11 @@ 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 +20 -12
@@ -494,6 +494,8 @@ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skbstatic int sk_psock_handle_skb (struct sk_psock *psock,struct sk_buff *skb,u32 off,u32 len,bool ingress) { +skb_bpf_redirect_clear (skb); +if (!ingress) {if (!sock_writeable (psock->sk))return -EAGAIN ;
@@ -525,7 +527,7 @@ 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); do { ret = -EIO; if (likely(psock->sk->sk_socket))
@@ -631,7 +633,12 @@ 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); }
@@ -752,7 +759,7 @@ 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. */
@@ -802,9 +809,10 @@ 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);
@@ -816,7 +824,6 @@ 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 ;
@@ -828,8 +835,7 @@ 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
@@ -890,9 +896,10 @@ 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:
@@ -1005,9 +1012,10 @@ 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
@@ -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
RE: [Patch bpf-next v4 4/5] skmsg: move sk_redir from TCP_SKB_CB to skb
Quoted 14 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.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.
Quoted 6 lines
Cc John Fastabend <[email protected]>Cc Daniel Borkmann <[email protected]>Cc Jakub Sitnicki <[email protected]>Cc Lorenz Bauer <[email protected]>Signed-off-by Cong Wang <[email protected]>---
Great, its likely we have more space in sk_buff we can use if needed as well. queue_mapping, skb_iif, vlan*, fields come to mind. Couple comments inline, but looks good to me.
[...]
Quoted 44 lines
net/core/skmsg.c +10
-3@@ -494,6 +494,8 @@ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skbstatic int sk_psock_handle_skb (struct sk_psock *psock,struct sk_buff *skb,u32 off,u32 len,bool ingress) { +skb_bpf_redirect_clear (skb); +if (!ingress) {if (!sock_writeable (psock->sk))return -EAGAIN ;@@ -525,7 +527,7 @@ 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); do { ret = -EIO; if (likely(psock->sk->sk_socket))@@ -631,7 +633,12 @@ 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); }@@ -752,7 +759,7 @@ 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. */
net/core/skmsg.c int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb) · line 813
Quoted 6 lines
net/core/skmsg.c +2
-1@@ -802,9 +809,10 @@ 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);
Do we really need the skb_dst_drop() I thought we would have already dropped this here but I've not had time to check yet.
Quoted 25 lines
net/core/skmsg.c +2
-4ret =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);@@ -816,7 +824,6 @@ 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 ;@@ -828,8 +835,7 @@ 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
net/core/skmsg.c static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb) · line 899
Quoted 5 lines
net/core/skmsg.c +1
-1@@ -890,9 +896,10 @@ 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);
Same here.
net/core/skmsg.c static int sk_psock_verdict_recv(read_descriptor_t *desc, struct sk_buff *skb, · line 1015
Quoted 13 lines
net/core/skmsg.c +3
-2+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:@@ -1005,9 +1012,10 @@ 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);
And here.
net/core/skmsg.c static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb) · line 904
Quoted 6 lines
net/core/skmsg.c +2
-1+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);
Thanks for doing this.
Re: [Patch bpf-next v4 4/5] skmsg: move sk_redir from TCP_SKB_CB to skb
Quoted 10 lines · John Fastabend
net/core/skmsg.c int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb) · line 813
Quoted 7 lines
net/core/skmsg.c +2
-1@@ -802,9 +809,10 @@ 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);Do we really need the
skb_dst_drop()I thought we would have already dropped this here but I've not had time to check yet.
Yes, I got some serious complaints from dst_release() when I didn't add skb_dst_drop().
Thanks.
John Fastabend
maintainerRe: [Patch bpf-next v4 4/5] skmsg: move sk_redir from TCP_SKB_CB to skb
Quoted 13 lines · Cong Wang
Quoted 10 lines · John Fastabend
net/core/skmsg.c int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb) · line 813
Quoted 7 lines
net/core/skmsg.c +2
-1@@ -802,9 +809,10 @@ 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);Do we really need the
skb_dst_drop()I thought we would have already dropped this here but I've not had time to check yet.
Quoted 4 lines
Yes, I got some serious complaints from
dst_release()when I didn't addskb_dst_drop().Thanks.
OK thanks.
PATCH v4 5/5 sock_map: rename skb_parser and skb_verdict
In reply to Cong WangFrom: 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
@@ -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 {
@@ -447,8 +447,8 @@ 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
@@ -690,9 +690,9 @@ 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 );
@@ -802,7 +802,7 @@ 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
@@ -894,7 +894,7 @@ 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);
@@ -918,7 +918,7 @@ 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);
@@ -981,7 +981,7 @@ void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)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); }#endif
@@ -1010,7 +1010,7 @@ 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
@@ -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
@@ -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