sockmap: add sockmap support for unix datagram socket
Subsystems (9)
v4 cover sockmap: add sockmap support for unix datagram socket
From: Cong Wang <[email protected]>
This is the last patchset of the original large patchset. In the previous patchset, a new BPF sockmap program BPF_SK_SKB_VERDICT was introduced and UDP began to support it too. In this patchset, we add BPF_SK_SKB_VERDICT support to Unix datagram socket, so that we can finally splice Unix datagram socket and UDP socket. Please check each patch description for more details.
To see the big picture, the previous patchsets are available here:
https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/commit/?id=1e0ab70778bd86a90de438cc5e1535c115a7c396
https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/commit/?id=89d69c5d0fbcabd8656459bc8b1a476d6f1efee4
and this patchset is available here:
https://github.com/congwang/linux/tree/sockmap3
---
v4: fix af_unix disconnect case
add unix_unhash()
split out two small patches
reduce u->iolock critical section
remove an unused parameter of __unix_dgram_recvmsg()
v3: fix Kconfig dependency
make unix_read_sock() static
fix a UAF in unix_release()
add a missing header unix_bpf.c
v2: separate out from the original large patchset
rebase to the latest bpf-next
clean up unix_read_sock()
export sock_map_close()
factor out some helpers in selftests for code reuse
Cong Wang (12):
sock_map: relax config dependency to CONFIG_NET
af_unix: implement ->read_sock() for sockmap
af_unix: set TCP_ESTABLISHED for datagram sockets too
sock_map: export symbols for af_unix module
af_unix: prepare for sockmap support
af_unix: implement ->psock_update_sk_prot()
af_unix: implement unix_dgram_bpf_recvmsg()
sock_map: update sock type checks for AF_UNIX selftests/bpf: factor out udp_socketpair() selftests/bpf: factor out add_to_sockmap() selftests/bpf: add a test case for unix sockmap selftests/bpf: add test cases for redirection between udp and unix
MAINTAINERS | 1 + include/linux/bpf.h | 38 +- include/net/af_unix.h | 13 + init/Kconfig | 2 +- net/core/Makefile | 2 - net/core/sock_map.c | 10 + net/unix/Makefile | 1 + net/unix/af_unix.c | 90 +++- net/unix/unix_bpf.c | 95 +++++ .../selftests/bpf/prog_tests/sockmap_listen.c | 386 ++++++++++++++---- 10 files changed, 535 insertions(+), 103 deletions(-) create mode 100644 net/unix/unix_bpf.c
Signature · 1 line
2.25.1
PATCH v4 1/12 sock_map: relax config dependency to CONFIG_NET
From: Cong Wang <[email protected]>
Currently sock_map still has hard dependency on CONFIG_INET, but there is no actual functional dependency on it after we introduce ->psock_update_sk_prot().
We have to extend it to CONFIG_NET now as we are going to support AF_UNIX.
--- include/linux/bpf.h | 38 ++++++++++++++++++++------------------ init/Kconfig | 2 +- net/core/Makefile | 2 -- 3 files changed, 21 insertions(+), 21 deletions(-)
include/linux/bpf.h +20 -18
@@ -1826,6 +1826,12 @@ 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_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)
@@ -1851,24 +1857,6 @@ 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_INET ) && defined(CONFIG_BPF_SYSCALL ) -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); - -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_get_from_fd(const union bpf_attr *attr,
@@ -1888,7 +1876,21 @@ static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void {return -EOPNOTSUPP ; } +#endif /* CONFIG_BPF_SYSCALL */ +#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */ +#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) {
init/Kconfig +1 -1
@@ -1723,7 +1723,7 @@ config BPF_SYSCALL select IRQ_WORK select TASKS_TRACE_RCU select BINARY_PRINTF - select NET_SOCK_MSG if INET + select NET_SOCK_MSG if NET default n help Enable the bpf() system call that allows to manipulate eBPF
net/core/Makefile +0 -2
@@ -33,8 +33,6 @@ 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
Signature · 1 line
2.25.1
PATCH v4 2/12 af_unix: implement ->read_sock() for sockmap
In reply to Cong WangFrom: Cong Wang <[email protected]>
Implement ->read_sock() for AF_UNIX datagram socket, it is pretty much similar to udp_read_sock().
--- net/unix/af_unix.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+)
net/unix/af_unix.c +35 -0
@@ -661,6 +661,8 @@ static ssize_t unix_stream_splice_read(struct socket *, loff_t *ppos, unsigned int flags);static int unix_dgram_sendmsg (struct socket *,struct msghdr *,size_t );static int unix_dgram_recvmsg (struct socket *,struct msghdr *,size_t ,int ); +static int unix_read_sock (struct sock *sk,read_descriptor_t *desc, +sk_read_actor_t recv_actor);static int unix_dgram_connect (struct socket *,struct sockaddr *,int ,int );static int unix_seqpacket_sendmsg (struct socket *,struct msghdr *,size_t );
@@ -738,6 +740,7 @@ static const struct proto_ops unix_dgram_ops = { .listen = sock_no_listen, .shutdown = unix_shutdown, .sendmsg = unix_dgram_sendmsg, + .read_sock = unix_read_sock, .recvmsg = unix_dgram_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage,
@@ -2183,6 +2186,38 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,return err; } +static int unix_read_sock (struct sock *sk,read_descriptor_t *desc, +sk_read_actor_t recv_actor) +{ +int copied =0 ; + +while (1 ) { +struct unix_sock *u =unix_sk (sk); +struct sk_buff *skb; +int used, err; + +mutex_lock (&u->iolock); + skb =skb_recv_datagram (sk,0 ,1 , &err); +mutex_unlock (&u->iolock); +if (!skb) +return err; + + used =recv_actor (desc, skb,0 , skb->len); +if (used <=0 ) { +if (!copied) + copied = used; +break ; + }else if (used <= skb->len) { + copied += used; + } + +if (!desc->count) +break ; + } + +return copied; +} +/* * Sleep until more data has arrived. But check for races.. */
Signature · 1 line
2.25.1
Re: [Patch bpf-next v4 02/12] af_unix: implement ->read_sock() for sockmap
Quoted 15 lines · Cong Wang
net/unix/af_unix.c +15
-0+static int unix_read_sock (struct sock *sk,read_descriptor_t *desc, +sk_read_actor_t recv_actor) +{ +int copied =0 ; + +while (1 ) { +struct unix_sock *u =unix_sk (sk); +struct sk_buff *skb; +int used, err; + +mutex_lock (&u->iolock); + skb =skb_recv_datagram (sk,0 ,1 , &err); +mutex_unlock (&u->iolock); +if (!skb) +return err;
net/unix/af_unix.c static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, · line 2213
Quoted 10 lines
net/unix/af_unix.c +10
-0+ + used =recv_actor (desc, skb,0 , skb->len); +if (used <=0 ) { +if (!copied) + copied = used; +break ; + }else if (used <= skb->len) { + copied += used; + } +
Like udp_read_sock(), we should also free the skb here. I will update this and send V5 soon.
Thanks.
PATCH v4 3/12 af_unix: set TCP_ESTABLISHED for datagram sockets too
In reply to Cong WangFrom: Cong Wang <[email protected]>
Currently only unix stream socket sets TCP_ESTABLISHED, datagram socket can set this too when they connect to its peer socket. At least __ip4_datagram_connect() does the same.
This will be used by a later patch to determine whether an AF_UNIX datagram socket can be redirected in sockmap.
--- net/unix/af_unix.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
net/unix/af_unix.c +8 -6
@@ -487,6 +487,7 @@ static void unix_dgram_disconnected(struct sock *sk, struct sock *other) other->sk_error_report (other); } } + sk->sk_state = other->sk_state =TCP_CLOSE ; }static void unix_sock_destructor (struct sock *sk)
@@ -1197,6 +1198,9 @@ static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr,unix_peer (sk) = other;unix_state_double_unlock (sk, other); } + +if (unix_peer (sk)) + sk->sk_state = other->sk_state =TCP_ESTABLISHED ;return 0 ; out_unlock:
@@ -1429,12 +1433,10 @@ static int unix_socketpair(struct socket *socka, struct socket *sockb)init_peercred (ska);init_peercred (skb); -if (ska->sk_type !=SOCK_DGRAM ) { - ska->sk_state =TCP_ESTABLISHED ; - skb->sk_state =TCP_ESTABLISHED ; - socka->state =SS_CONNECTED ; - sockb->state =SS_CONNECTED ; - } + ska->sk_state =TCP_ESTABLISHED ; + skb->sk_state =TCP_ESTABLISHED ; + socka->state =SS_CONNECTED ; + sockb->state =SS_CONNECTED ;return 0 ; }
Signature · 1 line
2.25.1
PATCH v4 4/12 sock_map: export symbols for af_unix module
In reply to Cong WangFrom: Cong Wang <[email protected]>
Before af_unix there is no need to support modules. Now we have to export these two functions for af_unix module.
--- net/core/sock_map.c | 2 ++ 1 file changed, 2 insertions(+)
net/core/sock_map.c +2 -0
@@ -1513,6 +1513,7 @@ void sock_map_unhash(struct sock *sk)rcu_read_unlock ();saved_unhash (sk); } +EXPORT_SYMBOL_GPL (sock_map_unhash);void sock_map_close (struct sock *sk,long timeout) {
@@ -1536,6 +1537,7 @@ void sock_map_close(struct sock *sk, long timeout)release_sock (sk);saved_close (sk, timeout); } +EXPORT_SYMBOL_GPL (sock_map_close); static int sock_map_iter_attach_target(struct bpf_prog *prog, union bpf_iter_link_info *linfo,
Signature · 1 line
2.25.1
PATCH v4 5/12 af_unix: prepare for sockmap support
In reply to Cong WangFrom: Cong Wang <[email protected]>
Unlike af_inet, unix_proto is very different, it does not even have a ->close() or ->unhash(). We have to add some dummy implementations to satisfy sockmap. Normally they are just nops, they are only used for the sockmap replacements.
--- net/unix/af_unix.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
net/unix/af_unix.c +16 -0
@@ -487,6 +487,8 @@ static void unix_dgram_disconnected(struct sock *sk, struct sock *other) other->sk_error_report (other); } } + sk->sk_prot->unhash (sk); + other->sk_prot->unhash (other); sk->sk_state = other->sk_state =TCP_CLOSE ; }
@@ -773,10 +775,23 @@ static const struct proto_ops unix_seqpacket_ops = { .show_fdinfo = unix_show_fdinfo, }; +/* Nothing to do here, unix socket is not unhashed when disconnecting, +* and does not need a ->close(). These are merely for sockmap. +*/ +static void unix_unhash (struct sock *sk) +{ +} + +static void unix_close (struct sock *sk,long timeout) +{ +} +static struct proto unix_proto = { .name ="UNIX" , .owner =THIS_MODULE , .obj_size =sizeof (struct unix_sock ), + .unhash = unix_unhash, + .close = unix_close, };static struct sock *unix_create1 (struct net *net,struct socket *sock,int kern)
@@ -860,6 +875,7 @@ static int unix_release(struct socket *sock)if (!sk)return 0 ; + sk->sk_prot->close (sk,0 );unix_release_sock (sk,0 ); sock->sk =NULL ;
Signature · 1 line
2.25.1
PATCH v4 6/12 af_unix: implement ->psock_update_sk_prot()
In reply to Cong WangFrom: Cong Wang <[email protected]>
Now we can implement unix_bpf_update_proto() to update sk_prot, particularly prot->close() and prot->unhash().
--- MAINTAINERS | 1 + include/net/af_unix.h | 10 +++++++++ net/unix/Makefile | 1 + net/unix/af_unix.c | 6 +++++- net/unix/unix_bpf.c | 48 +++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 65 insertions(+), 1 deletion(-) create mode 100644 net/unix/unix_bpf.c
MAINTAINERS +1 -0
@@ -10162,6 +10162,7 @@ F: net/core/skmsg.c F: net/core/sock_map.c F: net/ipv4/tcp_bpf.c F: net/ipv4/udp_bpf.c +F: net/unix/unix_bpf.c LANTIQ / INTEL Ethernet drivers M: Hauke Mehrtens <[email protected]>
include/net/af_unix.h +10 -0
@@ -89,4 +89,14 @@ void unix_sysctl_unregister(struct net *net);static inline int unix_sysctl_register (struct net *net) {return 0 ; }static inline void unix_sysctl_unregister (struct net *net) {}#endif + +#ifdef CONFIG_BPF_SYSCALL +extern struct proto unix_proto; + +int unix_bpf_update_proto (struct sock *sk,struct sk_psock *psock,bool restore); +void __init unix_bpf_build_proto (void ); +#else +static inline void __init unix_bpf_build_proto (void ) +{} +#endif #endif
net/unix/Makefile +1 -0
@@ -7,6 +7,7 @@ obj-$(CONFIG_UNIX) += unix.o unix-y := af_unix.o garbage.o unix-$(CONFIG_SYSCTL) += sysctl_net_unix.o +unix-$(CONFIG_BPF_SYSCALL) += unix_bpf.o obj-$(CONFIG_UNIX_DIAG) += unix_diag.o unix_diag-y := diag.o
net/unix/af_unix.c +5 -1
@@ -786,12 +786,15 @@ static void unix_close(struct sock *sk, long timeout) { } -static struct proto unix_proto = { +struct proto unix_proto = { .name ="UNIX" , .owner =THIS_MODULE , .obj_size =sizeof (struct unix_sock ), .unhash = unix_unhash, .close = unix_close, +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = unix_bpf_update_proto, +#endif };static struct sock *unix_create1 (struct net *net,struct socket *sock,int kern)
@@ -2973,6 +2976,7 @@ static int __init af_unix_init(void)sock_register (&unix_family_ops);register_pernet_subsys (&unix_net_ops); +unix_bpf_build_proto (); out:return rc; }
net/unix/unix_bpf.c new file +48 -0
48 lines not on this page · raw message
Signature · 1 line
2.25.1
PATCH v4 7/12 af_unix: implement unix_dgram_bpf_recvmsg()
In reply to Cong WangFrom: Cong Wang <[email protected]>
We have to implement unix_dgram_bpf_recvmsg() to replace the original ->recvmsg() to retrieve skmsg from ingress_msg.
AF_UNIX is again special here because the lack of sk_prot->recvmsg(). I simply add a special case inside unix_dgram_recvmsg() to call sk->sk_prot->recvmsg() directly.
--- include/net/af_unix.h | 3 +++ net/unix/af_unix.c | 19 ++++++++++++++--- net/unix/unix_bpf.c | 47 +++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 66 insertions(+), 3 deletions(-)
include/net/af_unix.h +3 -0
@@ -82,6 +82,9 @@ static inline struct unix_sock *unix_sk(const struct sock *sk)long unix_inq_len (struct sock *sk);long unix_outq_len (struct sock *sk); +int __unix_dgram_recvmsg (struct sock *sk,struct msghdr *msg,size_t size, +int nonblock,int flags); +#ifdef CONFIG_SYSCTL int unix_sysctl_register (struct net *net);void unix_sysctl_unregister (struct net *net);
net/unix/af_unix.c +16 -3
@@ -2100,11 +2100,11 @@ static void unix_copy_addr(struct msghdr *msg, struct sock *sk) } } -static int unix_dgram_recvmsg (struct socket *sock,struct msghdr *msg, -size_t size,int flags) +int __unix_dgram_recvmsg (struct sock *sk,struct msghdr *msg,size_t size, +int nonblock,int flags) {struct scm_cookie scm; -struct sock *sk = sock->sk; +struct socket *sock = sk->sk_socket;struct unix_sock *u =unix_sk (sk);struct sk_buff *skb, *last;long timeo;
@@ -2207,6 +2207,19 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,return err; } +static int unix_dgram_recvmsg (struct socket *sock,struct msghdr *msg,size_t size, +int flags) +{ +struct sock *sk = sock->sk; + +#ifdef CONFIG_BPF_SYSCALL +if (sk->sk_prot != &unix_proto) +return sk->sk_prot->recvmsg (sk, msg, size, flags &MSG_DONTWAIT , + flags & ~MSG_DONTWAIT ,NULL ); +#endif +return __unix_dgram_recvmsg (sk, msg, size, flags &MSG_DONTWAIT , flags); +} +static int unix_read_sock (struct sock *sk,read_descriptor_t *desc,sk_read_actor_t recv_actor) {
net/unix/unix_bpf.c +47 -0
59 lines not on this page · raw message
Signature · 1 line
2.25.1
PATCH v4 8/12 sock_map: update sock type checks for AF_UNIX
In reply to Cong WangFrom: Cong Wang <[email protected]>
Now AF_UNIX datagram supports sockmap and redirection, we can update the sock type checks for them accordingly.
--- net/core/sock_map.c | 8 ++++++++ 1 file changed, 8 insertions(+)
net/core/sock_map.c +8 -0
@@ -533,6 +533,12 @@ static bool sk_is_udp(const struct sock *sk) sk->sk_protocol ==IPPROTO_UDP ; } +static bool sk_is_unix (const struct sock *sk) +{ +return sk->sk_type ==SOCK_DGRAM && + sk->sk_family ==AF_UNIX ; +} +static bool sock_map_redirect_allowed (const struct sock *sk) {if (sk_is_tcp (sk))
@@ -552,6 +558,8 @@ static bool sock_map_sk_state_allowed(const struct sock *sk)return (1 << sk->sk_state) & (TCPF_ESTABLISHED |TCPF_LISTEN ); else if (sk_is_udp(sk)) return sk_hashed(sk); + else if (sk_is_unix(sk)) + return sk->sk_state == TCP_ESTABLISHED;return false; }
Signature · 1 line
2.25.1
PATCH v4 9/12 selftests/bpf: factor out udp_socketpair()
In reply to Cong WangFrom: Cong Wang <[email protected]>
Factor out a common helper udp_socketpair() which creates a pair of connected UDP sockets.
--- .../selftests/bpf/prog_tests/sockmap_listen.c | 76 ++++++++++--------- 1 file changed, 39 insertions(+), 37 deletions(-)
tools/testing/selftests/bpf/prog_tests/sockmap_listen.c +39 -37
@@ -1603,32 +1603,27 @@ static void test_reuseport(struct test_sockmap_listen *skel, } } -static void udp_redir_to_connected (int family,int sotype,int sock_mapfd, -int verd_mapfd,enum redir_mode mode) +static int udp_socketpair (int family,int *s,int *c) { -const char *log_prefix =redir_mode_str (mode);struct sockaddr_storage addr; -int c0, c1, p0, p1; -unsigned int pass;socklen_t len; -int err, n; -u64 value; -u32 key; -char b; - -zero_verdict_count (verd_mapfd); +int p0, c0; +int err; - p0 =socket_loopback (family, sotype |SOCK_NONBLOCK ); + p0 =socket_loopback (family,SOCK_DGRAM |SOCK_NONBLOCK );if (p0 <0 ) -return ; +return p0; + len =sizeof (addr); err =xgetsockname (p0,sockaddr (&addr), &len);if (err) goto close_peer0; - c0 =xsocket (family, sotype |SOCK_NONBLOCK ,0 ); -if (c0 <0 ) + c0 =xsocket (family,SOCK_DGRAM |SOCK_NONBLOCK ,0 ); +if (c0 <0 ) { + err = c0; goto close_peer0; + } err =xconnect (c0,sockaddr (&addr), len);if (err) goto close_cli0;
@@ -1639,25 +1634,36 @@ static void udp_redir_to_connected(int family, int sotype, int sock_mapfd,if (err) goto close_cli0; - p1 =socket_loopback (family, sotype |SOCK_NONBLOCK ); -if (p1 <0 ) - goto close_cli0; - err =xgetsockname (p1,sockaddr (&addr), &len); -if (err) - goto close_cli0; + *s = p0; + *c = c0; +return 0 ; - c1 =xsocket (family, sotype |SOCK_NONBLOCK ,0 ); -if (c1 <0 ) - goto close_peer1; - err =xconnect (c1,sockaddr (&addr), len); -if (err) - goto close_cli1; - err =xgetsockname (c1,sockaddr (&addr), &len); +close_cli0: +xclose (c0); +close_peer0: +xclose (p0); +return err; +} + +static void udp_redir_to_connected (int family,int sock_mapfd,int verd_mapfd, +enum redir_mode mode) +{ +const char *log_prefix =redir_mode_str (mode); +int c0, c1, p0, p1; +unsigned int pass; +int err, n; +u64 value; +u32 key; +char b; + +zero_verdict_count (verd_mapfd); + + err =udp_socketpair (family, &p0, &c0);if (err) - goto close_cli1; - err =xconnect (p1,sockaddr (&addr), len); +return ; + err =udp_socketpair (family, &p1, &c1);if (err) - goto close_cli1; + goto close_cli0; key =0 ; value = p0;
@@ -1694,11 +1700,9 @@ static void udp_redir_to_connected(int family, int sotype, int sock_mapfd, close_cli1:xclose (c1); -close_peer1:xclose (p1); close_cli0:xclose (c0); -close_peer0:xclose (p0); }
@@ -1715,11 +1719,9 @@ static void udp_skb_redir_to_connected(struct test_sockmap_listen *skel,return ; skel->bss->test_ingress = false; -udp_redir_to_connected (family,SOCK_DGRAM , sock_map, verdict_map, -REDIR_EGRESS ); +udp_redir_to_connected (family, sock_map, verdict_map,REDIR_EGRESS ); skel->bss->test_ingress = true; -udp_redir_to_connected (family,SOCK_DGRAM , sock_map, verdict_map, -REDIR_INGRESS ); +udp_redir_to_connected (family, sock_map, verdict_map,REDIR_INGRESS );xbpf_prog_detach2 (verdict, sock_map,BPF_SK_SKB_VERDICT ); }
Signature · 1 line
2.25.1
PATCH v4 10/12 selftests/bpf: factor out add_to_sockmap()
In reply to Cong WangFrom: Cong Wang <[email protected]>
Factor out a common helper add_to_sockmap() which adds two sockets into a sockmap.
--- .../selftests/bpf/prog_tests/sockmap_listen.c | 59 +++++++------------ 1 file changed, 21 insertions(+), 38 deletions(-)
tools/testing/selftests/bpf/prog_tests/sockmap_listen.c +21 -38
@@ -919,6 +919,23 @@ static const char *redir_mode_str(enum redir_mode mode) } } +static int add_to_sockmap (int sock_mapfd,int fd1,int fd2) +{ +u64 value; +u32 key; +int err; + + key =0 ; + value = fd1; + err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); +if (err) +return err; + + key =1 ; + value = fd2; +return xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); +} +static void redir_to_connected (int family,int sotype,int sock_mapfd,int verd_mapfd,enum redir_mode mode) {
@@ -928,7 +945,6 @@ static void redir_to_connected(int family, int sotype, int sock_mapfd,unsigned int pass;socklen_t len;int err, n; -u64 value;u32 key;char b;
@@ -965,15 +981,7 @@ static void redir_to_connected(int family, int sotype, int sock_mapfd,if (p1 <0 ) goto close_cli1; - key =0 ; - value = p0; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); -if (err) - goto close_peer1; - - key =1 ; - value = p1; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); + err =add_to_sockmap (sock_mapfd, p0, p1);if (err) goto close_peer1;
@@ -1061,7 +1069,6 @@ static void redir_to_listening(int family, int sotype, int sock_mapfd,int s, c, p, err, n;unsigned int drop;socklen_t len; -u64 value;u32 key;zero_verdict_count (verd_mapfd);
@@ -1086,15 +1093,7 @@ static void redir_to_listening(int family, int sotype, int sock_mapfd,if (p <0 ) goto close_cli; - key =0 ; - value = s; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); -if (err) - goto close_peer; - - key =1 ; - value = p; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); + err =add_to_sockmap (sock_mapfd, s, p);if (err) goto close_peer;
@@ -1346,7 +1345,6 @@ static void test_reuseport_mixed_groups(int family, int sotype, int sock_map,int s1, s2, c, err;unsigned int drop;socklen_t len; -u64 value;u32 key;zero_verdict_count (verd_map);
@@ -1360,16 +1358,10 @@ static void test_reuseport_mixed_groups(int family, int sotype, int sock_map,if (s2 <0 ) goto close_srv1; - key =0 ; - value = s1; - err =xbpf_map_update_elem (sock_map, &key, &value,BPF_NOEXIST ); + err =add_to_sockmap (sock_map, s1, s2);if (err) goto close_srv2; - key =1 ; - value = s2; - err =xbpf_map_update_elem (sock_map, &key, &value,BPF_NOEXIST ); -/* Connect to s2, reuseport BPF selects s1 via sock_map[0] */ len =sizeof (addr); err =xgetsockname (s2,sockaddr (&addr), &len);
@@ -1652,7 +1644,6 @@ static void udp_redir_to_connected(int family, int sock_mapfd, int verd_mapfd,int c0, c1, p0, p1;unsigned int pass;int err, n; -u64 value;u32 key;char b;
@@ -1665,15 +1656,7 @@ static void udp_redir_to_connected(int family, int sock_mapfd, int verd_mapfd,if (err) goto close_cli0; - key =0 ; - value = p0; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); -if (err) - goto close_cli1; - - key =1 ; - value = p1; - err =xbpf_map_update_elem (sock_mapfd, &key, &value,BPF_NOEXIST ); + err =add_to_sockmap (sock_mapfd, p0, p1);if (err) goto close_cli1;
Signature · 1 line
2.25.1
PATCH v4 11/12 selftests/bpf: add a test case for unix sockmap
In reply to Cong WangFrom: Cong Wang <[email protected]>
Add a test case to ensure redirection between two AF_UNIX datagram sockets work.
--- .../selftests/bpf/prog_tests/sockmap_listen.c | 92 +++++++++++++++++++ 1 file changed, 92 insertions(+)
tools/testing/selftests/bpf/prog_tests/sockmap_listen.c +92 -0
@@ -1433,6 +1433,8 @@ static const char *family_str(sa_family_t family)return "IPv4" ;case AF_INET6 :return "IPv6" ; +case AF_UNIX : +return "Unix" ;default :return "unknown" ; }
@@ -1555,6 +1557,94 @@ static void test_redir(struct test_sockmap_listen *skel, struct bpf_map *map, } } +static void unix_redir_to_connected (int sotype,int sock_mapfd, +int verd_mapfd,enum redir_mode mode) +{ +const char *log_prefix =redir_mode_str (mode); +int c0, c1, p0, p1; +unsigned int pass; +int err, n; +int sfd[2 ]; +u32 key; +char b; + +zero_verdict_count (verd_mapfd); + +if (socketpair (AF_UNIX , sotype |SOCK_NONBLOCK ,0 , sfd)) +return ; + c0 = sfd[0 ], p0 = sfd[1 ]; + +if (socketpair (AF_UNIX , sotype |SOCK_NONBLOCK ,0 , sfd)) + goto close0; + c1 = sfd[0 ], p1 = sfd[1 ]; + + err =add_to_sockmap (sock_mapfd, p0, p1); +if (err) + goto close; + + n =write (c1,"a" ,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: write" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete write" , log_prefix); +if (n <1 ) + goto close; + + key =SK_PASS ; + err =xbpf_map_lookup_elem (verd_mapfd, &key, &pass); +if (err) + goto close; +if (pass !=1 ) +FAIL ("%s: want pass count 1, have %d" , log_prefix, pass); + + n =read (mode ==REDIR_INGRESS ? p0 : c0, &b,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: read" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete read" , log_prefix); + +close: +xclose (c1); +xclose (p1); +close0: +xclose (c0); +xclose (p0); +} + +static void unix_skb_redir_to_connected (struct test_sockmap_listen *skel, +struct bpf_map *inner_map,int sotype) +{ +int verdict =bpf_program__fd (skel->progs.prog_skb_verdict); +int verdict_map =bpf_map__fd (skel->maps.verdict_map); +int sock_map =bpf_map__fd (inner_map); +int err; + + err =xbpf_prog_attach (verdict, sock_map,BPF_SK_SKB_VERDICT ,0 ); +if (err) +return ; + + skel->bss->test_ingress = false; +unix_redir_to_connected (sotype, sock_map, verdict_map,REDIR_EGRESS ); + skel->bss->test_ingress = true; +unix_redir_to_connected (sotype, sock_map, verdict_map,REDIR_INGRESS ); + +xbpf_prog_detach2 (verdict, sock_map,BPF_SK_SKB_VERDICT ); +} + +static void test_unix_redir (struct test_sockmap_listen *skel,struct bpf_map *map, +int sotype) +{ +const char *family_name, *map_name; +char s[MAX_TEST_NAME ]; + + family_name =family_str (AF_UNIX ); + map_name =map_type_str (map); +snprintf (s,sizeof (s),"%s %s %s" , map_name, family_name, __func__); +if (!test__start_subtest (s)) +return ; +unix_skb_redir_to_connected (skel, map, sotype); +} +static void test_reuseport (struct test_sockmap_listen *skel,struct bpf_map *map,int family,int sotype) {
@@ -1747,10 +1837,12 @@ void test_sockmap_listen(void) skel->bss->test_sockmap = true;run_tests (skel, skel->maps.sock_map,AF_INET );run_tests (skel, skel->maps.sock_map,AF_INET6 ); +test_unix_redir (skel, skel->maps.sock_map,SOCK_DGRAM ); skel->bss->test_sockmap = false;run_tests (skel, skel->maps.sock_hash,AF_INET );run_tests (skel, skel->maps.sock_hash,AF_INET6 ); +test_unix_redir (skel, skel->maps.sock_hash,SOCK_DGRAM );test_sockmap_listen__destroy (skel); }
Signature · 1 line
2.25.1
PATCH v4 12/12 selftests/bpf: add test cases for redirection between udp and unix
In reply to Cong WangFrom: Cong Wang <[email protected]>
Add two test cases to ensure redirection between udp and unix work bidirectionally.
--- .../selftests/bpf/prog_tests/sockmap_listen.c | 165 ++++++++++++++++++ 1 file changed, 165 insertions(+)
tools/testing/selftests/bpf/prog_tests/sockmap_listen.c +165 -0
@@ -1813,6 +1813,170 @@ static void test_udp_redir(struct test_sockmap_listen *skel, struct bpf_map *mapudp_skb_redir_to_connected (skel, map, family); } +static void udp_unix_redir_to_connected (int family,int sock_mapfd, +int verd_mapfd,enum redir_mode mode) +{ +const char *log_prefix =redir_mode_str (mode); +int c0, c1, p0, p1; +unsigned int pass; +int err, n; +int sfd[2 ]; +u32 key; +char b; + +zero_verdict_count (verd_mapfd); + +if (socketpair (AF_UNIX ,SOCK_DGRAM |SOCK_NONBLOCK ,0 , sfd)) +return ; + c0 = sfd[0 ], p0 = sfd[1 ]; + + err =udp_socketpair (family, &p1, &c1); +if (err) + goto close; + + err =add_to_sockmap (sock_mapfd, p0, p1); +if (err) + goto close_cli1; + + n =write (c1,"a" ,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: write" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete write" , log_prefix); +if (n <1 ) + goto close_cli1; + + key =SK_PASS ; + err =xbpf_map_lookup_elem (verd_mapfd, &key, &pass); +if (err) + goto close_cli1; +if (pass !=1 ) +FAIL ("%s: want pass count 1, have %d" , log_prefix, pass); + + n =read (mode ==REDIR_INGRESS ? p0 : c0, &b,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: read" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete read" , log_prefix); + +close_cli1: +xclose (c1); +xclose (p1); +close: +xclose (c0); +xclose (p0); +} + +static void udp_unix_skb_redir_to_connected (struct test_sockmap_listen *skel, +struct bpf_map *inner_map,int family) +{ +int verdict =bpf_program__fd (skel->progs.prog_skb_verdict); +int verdict_map =bpf_map__fd (skel->maps.verdict_map); +int sock_map =bpf_map__fd (inner_map); +int err; + + err =xbpf_prog_attach (verdict, sock_map,BPF_SK_SKB_VERDICT ,0 ); +if (err) +return ; + + skel->bss->test_ingress = false; +udp_unix_redir_to_connected (family, sock_map, verdict_map,REDIR_EGRESS ); + skel->bss->test_ingress = true; +udp_unix_redir_to_connected (family, sock_map, verdict_map,REDIR_INGRESS ); + +xbpf_prog_detach2 (verdict, sock_map,BPF_SK_SKB_VERDICT ); +} + +static void unix_udp_redir_to_connected (int family,int sock_mapfd, +int verd_mapfd,enum redir_mode mode) +{ +const char *log_prefix =redir_mode_str (mode); +int c0, c1, p0, p1; +unsigned int pass; +int err, n; +int sfd[2 ]; +u32 key; +char b; + +zero_verdict_count (verd_mapfd); + + err =udp_socketpair (family, &p0, &c0); +if (err) +return ; + +if (socketpair (AF_UNIX ,SOCK_DGRAM |SOCK_NONBLOCK ,0 , sfd)) + goto close_cli0; + c1 = sfd[0 ], p1 = sfd[1 ]; + + err =add_to_sockmap (sock_mapfd, p0, p1); +if (err) + goto close; + + n =write (c1,"a" ,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: write" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete write" , log_prefix); +if (n <1 ) + goto close; + + key =SK_PASS ; + err =xbpf_map_lookup_elem (verd_mapfd, &key, &pass); +if (err) + goto close; +if (pass !=1 ) +FAIL ("%s: want pass count 1, have %d" , log_prefix, pass); + + n =read (mode ==REDIR_INGRESS ? p0 : c0, &b,1 ); +if (n <0 ) +FAIL_ERRNO ("%s: read" , log_prefix); +if (n ==0 ) +FAIL ("%s: incomplete read" , log_prefix); + +close: +xclose (c1); +xclose (p1); +close_cli0: +xclose (c0); +xclose (p0); + +} + +static void unix_udp_skb_redir_to_connected (struct test_sockmap_listen *skel, +struct bpf_map *inner_map,int family) +{ +int verdict =bpf_program__fd (skel->progs.prog_skb_verdict); +int verdict_map =bpf_map__fd (skel->maps.verdict_map); +int sock_map =bpf_map__fd (inner_map); +int err; + + err =xbpf_prog_attach (verdict, sock_map,BPF_SK_SKB_VERDICT ,0 ); +if (err) +return ; + + skel->bss->test_ingress = false; +unix_udp_redir_to_connected (family, sock_map, verdict_map,REDIR_EGRESS ); + skel->bss->test_ingress = true; +unix_udp_redir_to_connected (family, sock_map, verdict_map,REDIR_INGRESS ); + +xbpf_prog_detach2 (verdict, sock_map,BPF_SK_SKB_VERDICT ); +} + +static void test_udp_unix_redir (struct test_sockmap_listen *skel,struct bpf_map *map, +int family) +{ +const char *family_name, *map_name; +char s[MAX_TEST_NAME ]; + + family_name =family_str (family); + map_name =map_type_str (map); +snprintf (s,sizeof (s),"%s %s %s" , map_name, family_name, __func__); +if (!test__start_subtest (s)) +return ; +udp_unix_skb_redir_to_connected (skel, map, family); +unix_udp_skb_redir_to_connected (skel, map, family); +} +static void run_tests (struct test_sockmap_listen *skel,struct bpf_map *map,int family) {
@@ -1822,6 +1986,7 @@ static void run_tests(struct test_sockmap_listen *skel, struct bpf_map *map,test_reuseport (skel, map, family,SOCK_STREAM );test_reuseport (skel, map, family,SOCK_DGRAM );test_udp_redir (skel, map, family); +test_udp_unix_redir (skel, map, family); }void test_sockmap_listen (void )
Signature · 1 line
2.25.1