From 39bc9de532672e8cc3b0a4b349df260b54c9c5fd Mon Sep 17 00:00:00 2001 From: Lawrence Stewart Date: Tue, 28 Dec 2010 12:13:30 +0000 Subject: [PATCH] - Add some helper hook points to the TCP stack. The hooks allow Khelp modules to access inbound/outbound events and associated data for established TCP connections. The hooks only run if at least one hook function is registered for the hook point, ensuring the impact on the stack is effectively nil when no TCP Khelp modules are loaded. struct tcp_hhook_data is passed as contextual data to any registered Khelp module hook functions. - Add an OSD (Object Specific Data) pointer to struct tcpcb to allow Khelp modules to associate per-connection data with the TCP control block. - Bump __FreeBSD_version and add a note to UPDATING regarding to ABI changes introduced by this commit and r216753. In collaboration with: David Hayes and Grenville Armitage Sponsored by: FreeBSD Foundation Reviewed by: bz, others along the way MFC after: 3 months --- UPDATING | 7 +++++++ sys/netinet/tcp_input.c | 29 +++++++++++++++++++++++++++++ sys/netinet/tcp_output.c | 28 ++++++++++++++++++++++++++++ sys/netinet/tcp_subr.c | 20 ++++++++++++++++++++ sys/netinet/tcp_var.h | 20 ++++++++++++++++++++ sys/sys/param.h | 2 +- 6 files changed, 105 insertions(+), 1 deletion(-) diff --git a/UPDATING b/UPDATING index 0323738cb4c0..b4c302d4c76f 100644 --- a/UPDATING +++ b/UPDATING @@ -22,6 +22,13 @@ NOTE TO PEOPLE WHO THINK THAT FreeBSD 9.x IS SLOW: machines to maximize performance. (To disable malloc debugging, run ln -s aj /etc/malloc.conf.) +20101228: + The TCP stack has been modified to allow Khelp modules to interact with + it via helper hook points and store per-connection data in the TCP + control block. Bump __FreeBSD_version to 900029. User space tools that + rely on the size of struct tcpcb in tcp_var.h (e.g. sockstat) need to + be recompiled. + 20101114: Generic IEEE 802.3 annex 31B full duplex flow control support has been added to mii(4) and bge(4), bce(4), msk(4), nfe(4) and stge(4) along diff --git a/sys/netinet/tcp_input.c b/sys/netinet/tcp_input.c index 4dbc0380f131..e2be41afeaaf 100644 --- a/sys/netinet/tcp_input.c +++ b/sys/netinet/tcp_input.c @@ -54,6 +54,7 @@ __FBSDID("$FreeBSD$"); #include #include +#include #include #include #include /* for proc0 declaration */ @@ -218,6 +219,8 @@ static void inline cc_ack_received(struct tcpcb *tp, struct tcphdr *th, uint16_t type); static void inline cc_conn_init(struct tcpcb *tp); static void inline cc_post_recovery(struct tcpcb *tp, struct tcphdr *th); +static void inline hhook_run_tcp_est_in(struct tcpcb *tp, + struct tcphdr *th, struct tcpopt *to); /* * Kernel module interface for updating tcpstat. The argument is an index @@ -233,6 +236,24 @@ kmod_tcpstat_inc(int statnum) (*((u_long *)&V_tcpstat + statnum))++; } +/* + * Wrapper for the TCP established input helper hook. + */ +static void inline +hhook_run_tcp_est_in(struct tcpcb *tp, struct tcphdr *th, struct tcpopt *to) +{ + struct tcp_hhook_data hhook_data; + + if (V_tcp_hhh[HHOOK_TCP_EST_IN]->hhh_nhooks > 0) { + hhook_data.tp = tp; + hhook_data.th = th; + hhook_data.to = to; + + hhook_run_hooks(V_tcp_hhh[HHOOK_TCP_EST_IN], &hhook_data, + tp->osd); + } +} + /* * CC wrapper hook functions */ @@ -1486,6 +1507,10 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so, ticks - tp->t_rtttime); } acked = BYTES_THIS_ACK(tp, th); + + /* Run HHOOK_TCP_ESTABLISHED_IN helper hooks. */ + hhook_run_tcp_est_in(tp, th, &to); + TCPSTAT_INC(tcps_rcvackpack); TCPSTAT_ADD(tcps_rcvackbyte, acked); sbdrop(&so->so_snd, acked); @@ -2199,6 +2224,10 @@ tcp_do_segment(struct mbuf *m, struct tcphdr *th, struct socket *so, ((to.to_flags & TOF_SACK) || !TAILQ_EMPTY(&tp->snd_holes))) tcp_sack_doack(tp, &to, th->th_ack); + + /* Run HHOOK_TCP_ESTABLISHED_IN helper hooks. */ + hhook_run_tcp_est_in(tp, th, &to); + if (SEQ_LEQ(th->th_ack, tp->snd_una)) { if (tlen == 0 && tiwin == tp->snd_wnd) { TCPSTAT_INC(tcps_rcvdupack); diff --git a/sys/netinet/tcp_output.c b/sys/netinet/tcp_output.c index b7ec6f7e21d9..55d3c8e7b968 100644 --- a/sys/netinet/tcp_output.c +++ b/sys/netinet/tcp_output.c @@ -40,6 +40,7 @@ __FBSDID("$FreeBSD$"); #include #include #include +#include #include #include #include @@ -126,8 +127,32 @@ SYSCTL_VNET_INT(_net_inet_tcp, OID_AUTO, sendbuf_max, CTLFLAG_RW, &VNET_NAME(tcp_autosndbuf_max), 0, "Max size of automatic send buffer"); +static void inline hhook_run_tcp_est_out(struct tcpcb *tp, + struct tcphdr *th, struct tcpopt *to, + long len, int tso); static void inline cc_after_idle(struct tcpcb *tp); +/* + * Wrapper for the TCP established ouput helper hook. + */ +static void inline +hhook_run_tcp_est_out(struct tcpcb *tp, struct tcphdr *th, + struct tcpopt *to, long len, int tso) +{ + struct tcp_hhook_data hhook_data; + + if (V_tcp_hhh[HHOOK_TCP_EST_OUT]->hhh_nhooks > 0) { + hhook_data.tp = tp; + hhook_data.th = th; + hhook_data.to = to; + hhook_data.len = len; + hhook_data.tso = tso; + + hhook_run_hooks(V_tcp_hhh[HHOOK_TCP_EST_OUT], &hhook_data, + tp->osd); + } +} + /* * CC wrapper hook functions */ @@ -1134,6 +1159,9 @@ tcp_output(struct tcpcb *tp) tp->snd_max = tp->snd_nxt + len; } + /* Run HHOOK_TCP_ESTABLISHED_OUT helper hooks. */ + hhook_run_tcp_est_out(tp, th, &to, len, tso); + #ifdef TCPDEBUG /* * Trace. diff --git a/sys/netinet/tcp_subr.c b/sys/netinet/tcp_subr.c index 73456615f723..00d266d27b2d 100644 --- a/sys/netinet/tcp_subr.c +++ b/sys/netinet/tcp_subr.c @@ -41,7 +41,9 @@ __FBSDID("$FreeBSD$"); #include #include #include +#include #include +#include #include #include #include @@ -214,6 +216,8 @@ SYSCTL_INT(_net_inet_tcp, OID_AUTO, soreceive_stream, CTLFLAG_RDTUN, VNET_DEFINE(uma_zone_t, sack_hole_zone); #define V_sack_hole_zone VNET(sack_hole_zone) +VNET_DEFINE(struct hhook_head *, tcp_hhh[HHOOK_TCP_LAST+1]); + static struct inpcb *tcp_notify(struct inpcb *, int); static void tcp_isn_tick(void *); static char * tcp_log_addr(struct in_conninfo *inc, struct tcphdr *th, @@ -239,6 +243,7 @@ struct tcpcb_mem { struct tcpcb tcb; struct tcp_timer tt; struct cc_var ccv; + struct osd osd; }; static VNET_DEFINE(uma_zone_t, tcpcb_zone); @@ -278,6 +283,13 @@ tcp_init(void) { int hashsize; + if (hhook_head_register(HHOOK_TYPE_TCP, HHOOK_TCP_EST_IN, + &V_tcp_hhh[HHOOK_TCP_EST_IN], HHOOK_NOWAIT|HHOOK_HEADISINVNET) != 0) + printf("%s: WARNING: unable to register helper hook\n", __func__); + if (hhook_head_register(HHOOK_TYPE_TCP, HHOOK_TCP_EST_OUT, + &V_tcp_hhh[HHOOK_TCP_EST_OUT], HHOOK_NOWAIT|HHOOK_HEADISINVNET) != 0) + printf("%s: WARNING: unable to register helper hook\n", __func__); + hashsize = TCBHASHSIZE; TUNABLE_INT_FETCH("net.inet.tcp.tcbhashsize", &hashsize); if (!powerof2(hashsize)) { @@ -661,6 +673,12 @@ tcp_newtcpcb(struct inpcb *inp) return (NULL); } + tp->osd = &tm->osd; + if (khelp_init_osd(HELPER_CLASS_TCP, tp->osd)) { + uma_zfree(V_tcpcb_zone, tm); + return (NULL); + } + #ifdef VIMAGE tp->t_vnet = inp->inp_vnet; #endif @@ -894,6 +912,8 @@ tcp_discardcb(struct tcpcb *tp) if (CC_ALGO(tp)->cb_destroy != NULL) CC_ALGO(tp)->cb_destroy(tp->ccv); + khelp_destroy_osd(tp->osd); + CC_ALGO(tp) = NULL; inp->inp_ppcb = NULL; tp->t_inpcb = NULL; diff --git a/sys/netinet/tcp_var.h b/sys/netinet/tcp_var.h index 3305b1a6bbaf..4a43213fc57f 100644 --- a/sys/netinet/tcp_var.h +++ b/sys/netinet/tcp_var.h @@ -201,6 +201,7 @@ struct tcpcb { int t_bytes_acked; /* # bytes acked during current RTT */ struct cc_algo *cc_algo; /* congestion control algorithm */ struct cc_var *ccv; + struct osd *osd; /* storage for Khelp module data */ int t_ispare; /* explicit pad for 64bit alignment */ void *t_pspare2[4]; /* 4 TBD */ @@ -501,6 +502,22 @@ struct tcpstat { void kmod_tcpstat_inc(int statnum); #define KMOD_TCPSTAT_INC(name) \ kmod_tcpstat_inc(offsetof(struct tcpstat, name) / sizeof(u_long)) + +/* + * TCP specific helper hook point identifiers. + */ +#define HHOOK_TCP_EST_IN 0 +#define HHOOK_TCP_EST_OUT 1 +#define HHOOK_TCP_LAST HHOOK_TCP_EST_OUT + +struct tcp_hhook_data { + struct tcpcb *tp; + struct tcphdr *th; + struct tcpopt *to; + long len; + int tso; + tcp_seq curack; +}; #endif /* @@ -608,6 +625,9 @@ VNET_DECLARE(int, tcp_ecn_maxretries); #define V_tcp_do_ecn VNET(tcp_do_ecn) #define V_tcp_ecn_maxretries VNET(tcp_ecn_maxretries) +VNET_DECLARE(struct hhook_head *, tcp_hhh[HHOOK_TCP_LAST+1]); +#define V_tcp_hhh VNET(tcp_hhh) + int tcp_addoptions(struct tcpopt *, u_char *); int tcp_ccalgounload(struct cc_algo *unload_algo); struct tcpcb * diff --git a/sys/sys/param.h b/sys/sys/param.h index 29114d301635..7ccd3ea4bbfc 100644 --- a/sys/sys/param.h +++ b/sys/sys/param.h @@ -58,7 +58,7 @@ * in the range 5 to 9. */ #undef __FreeBSD_version -#define __FreeBSD_version 900028 /* Master, propagated to newvers */ +#define __FreeBSD_version 900029 /* Master, propagated to newvers */ #ifdef _KERNEL #define P_OSREL_SIGSEGV 700004