[TCP]: Add pluggable congestion control algorithm infrastructure.

Allow TCP to have multiple pluggable congestion control algorithms.
Algorithms are defined by a set of operations and can be built in
or modules.  The legacy "new RENO" algorithm is used as a starting
point and fallback.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 11 insertions, 47 deletions

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 614e939c78a4..72965bfe6cfb 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -333,21 +333,14 @@ enum
         NET_TCP_FRTO=92,
         NET_TCP_LOW_LATENCY=93,
         NET_IPV4_IPFRAG_SECRET_INTERVAL=94,
-        NET_TCP_WESTWOOD=95,
         NET_IPV4_IGMP_MAX_MSF=96,
         NET_TCP_NO_METRICS_SAVE=97,
-        NET_TCP_VEGAS=98,
-        NET_TCP_VEGAS_ALPHA=99,
-        NET_TCP_VEGAS_BETA=100,
-        NET_TCP_VEGAS_GAMMA=101,
-        NET_TCP_BIC=102,
-        NET_TCP_BIC_FAST_CONVERGENCE=103,
-        NET_TCP_BIC_LOW_WINDOW=104,
         NET_TCP_DEFAULT_WIN_SCALE=105,
         NET_TCP_MODERATE_RCVBUF=106,
         NET_TCP_TSO_WIN_DIVISOR=107,
         NET_TCP_BIC_BETA=108,
         NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109,
+        NET_TCP_CONG_CONTROL=110,
 };
 
 enum {
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 97a7c9e03df5..3ea75dd6640a 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -203,13 +203,6 @@ struct tcp_sack_block {
         __u32   end_seq;
 };
 
-enum tcp_congestion_algo {
-        TCP_RENO=0,
-        TCP_VEGAS,
-        TCP_WESTWOOD,
-        TCP_BIC,
-};
-
 struct tcp_options_received {
 /*      PAWS/RTTM data  */
         long    ts_recent_stamp;/* Time we stored ts_recent (for aging) */
@@ -305,7 +298,7 @@ struct tcp_sock {
         __u8    reordering;     /* Packet reordering metric.            */
         __u8    frto_counter;   /* Number of new acks after RTO */
 
-        __u8    adv_cong;       /* Using Vegas, Westwood, or BIC */
+        __u8    unused;
         __u8    defer_accept;   /* User waits for some data after accept() */
 
 /* RTT measurement */
@@ -401,37 +394,10 @@ struct tcp_sock {
                 __u32   time;
         } rcvq_space;
 
-/* TCP Westwood structure */
-        struct {
-                __u32    bw_ns_est;        /* first bandwidth estimation..not too smoothed 8) */
-                __u32    bw_est;           /* bandwidth estimate */
-                __u32    rtt_win_sx;       /* here starts a new evaluation... */
-                __u32    bk;
-                __u32    snd_una;          /* used for evaluating the number of acked bytes */
-                __u32    cumul_ack;
-                __u32    accounted;
-                __u32    rtt;
-                __u32    rtt_min;          /* minimum observed RTT */
-        } westwood;
-
-/* Vegas variables */
-        struct {
-                __u32   beg_snd_nxt;    /* right edge during last RTT */
-                __u32   beg_snd_una;    /* left edge  during last RTT */
-                __u32   beg_snd_cwnd;   /* saves the size of the cwnd */
-                __u8    doing_vegas_now;/* if true, do vegas for this RTT */
-                __u16   cntRTT;         /* # of RTTs measured within last RTT */
-                __u32   minRTT;         /* min of RTTs measured within last RTT (in usec) */
-                __u32   baseRTT;        /* the min of all Vegas RTT measurements seen (in usec) */
-        } vegas;
-
-        /* BI TCP Parameters */
-        struct {
-                __u32   cnt;            /* increase cwnd by 1 after this number of ACKs */
-                __u32   last_max_cwnd;  /* last maximium snd_cwnd */
-                __u32   last_cwnd;      /* the last snd_cwnd */
-                __u32   last_stamp;     /* time when updated last_cwnd */
-        } bictcp;
+        /* Pluggable TCP congestion control hook */
+        struct tcp_congestion_ops *ca_ops;
+        u32     ca_priv[16];
+#define TCP_CA_PRIV_SIZE        (16*sizeof(u32))
 };
 
 static inline struct tcp_sock *tcp_sk(const struct sock *sk)
@@ -439,6 +405,11 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk)
         return (struct tcp_sock *)sk;
 }
 
+static inline void *tcp_ca(const struct tcp_sock *tp)
+{
+        return (void *) tp->ca_priv;
+}
+
 #endif
 
 #endif  /* _LINUX_TCP_H */