packet: account statistics only in tpacket_stats_u

Currently, packet_sock has a struct tpacket_stats stats member for
TPACKET_V1 and TPACKET_V2 statistic accounting, and with TPACKET_V3
``union tpacket_stats_u stats_u'' was introduced, where however only
statistics for TPACKET_V3 are held, and when copied to user space,
TPACKET_V3 does some hackery and access also tpacket_stats' stats,
although everything could have been done within the union itself.

Unify accounting within the tpacket_stats_u union so that we can
remove 8 bytes from packet_sock that are there unnecessary. Note that
even if we switch to TPACKET_V3 and would use non mmap(2)ed option,
this still works due to the union with same types + offsets, that are
exposed to the user space.

Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ba8309a3e01beede11dc5adc01e26dc72fda15b5..dd5cd49b0e09e9ac777484aeae3c7ade6755a3b7 100644 (file)
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -528,7 +528,7 @@ static void init_prb_bdqc(struct packet_sock *po,
        p1->hdrlen = po->tp_hdrlen;
        p1->version = po->tp_version;
        p1->last_kactive_blk_num = 0;
-       po->stats_u.stats3.tp_freeze_q_cnt = 0;
+       po->stats.stats3.tp_freeze_q_cnt = 0;
        if (req_u->req3.tp_retire_blk_tov)
                p1->retire_blk_tov = req_u->req3.tp_retire_blk_tov;
        else
@@ -696,7 +696,7 @@ static void prb_close_block(struct tpacket_kbdq_core *pkc1,
        struct tpacket3_hdr *last_pkt;
        struct tpacket_hdr_v1 *h1 = &pbd1->hdr.bh1;
 
-       if (po->stats.tp_drops)
+       if (po->stats.stats3.tp_drops)
                status |= TP_STATUS_LOSING;
 
        last_pkt = (struct tpacket3_hdr *)pkc1->prev;
@@ -801,7 +801,7 @@ static void prb_freeze_queue(struct tpacket_kbdq_core *pkc,
                                  struct packet_sock *po)
 {
        pkc->reset_pending_on_curr_blk = 1;
-       po->stats_u.stats3.tp_freeze_q_cnt++;
+       po->stats.stats3.tp_freeze_q_cnt++;
 }
 
 #define TOTAL_PKT_LEN_INCL_ALIGN(length) (ALIGN((length), V3_ALIGNMENT))
@@ -1687,7 +1687,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev,
        nf_reset(skb);
 
        spin_lock(&sk->sk_receive_queue.lock);
-       po->stats.tp_packets++;
+       po->stats.stats1.tp_packets++;
        skb->dropcount = atomic_read(&sk->sk_drops);
        __skb_queue_tail(&sk->sk_receive_queue, skb);
        spin_unlock(&sk->sk_receive_queue.lock);
@@ -1696,7 +1696,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev,
 
 drop_n_acct:
        spin_lock(&sk->sk_receive_queue.lock);
-       po->stats.tp_drops++;
+       po->stats.stats1.tp_drops++;
        atomic_inc(&sk->sk_drops);
        spin_unlock(&sk->sk_receive_queue.lock);
 
@@ -1796,10 +1796,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
         * Anyways, moving it for V1/V2 only as V3 doesn't need this
         * at packet level.
         */
-               if (po->stats.tp_drops)
+               if (po->stats.stats1.tp_drops)
                        status |= TP_STATUS_LOSING;
        }
-       po->stats.tp_packets++;
+       po->stats.stats1.tp_packets++;
        if (copy_skb) {
                status |= TP_STATUS_COPY;
                __skb_queue_tail(&sk->sk_receive_queue, copy_skb);
@@ -1898,7 +1898,7 @@ drop:
        return 0;
 
 ring_is_full:
-       po->stats.tp_drops++;
+       po->stats.stats1.tp_drops++;
        spin_unlock(&sk->sk_receive_queue.lock);
 
        sk->sk_data_ready(sk, 0);
@@ -3247,8 +3247,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
        struct sock *sk = sock->sk;
        struct packet_sock *po = pkt_sk(sk);
        void *data = &val;
-       struct tpacket_stats st;
-       union tpacket_stats_u st_u;
+       union tpacket_stats_u st;
 
        if (level != SOL_PACKET)
                return -ENOPROTOOPT;
@@ -3262,22 +3261,18 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
        switch (optname) {
        case PACKET_STATISTICS:
                spin_lock_bh(&sk->sk_receive_queue.lock);
+               memcpy(&st, &po->stats, sizeof(st));
+               memset(&po->stats, 0, sizeof(po->stats));
+               spin_unlock_bh(&sk->sk_receive_queue.lock);
+
                if (po->tp_version == TPACKET_V3) {
                        lv = sizeof(struct tpacket_stats_v3);
-                       memcpy(&st_u.stats3, &po->stats,
-                              sizeof(struct tpacket_stats));
-                       st_u.stats3.tp_freeze_q_cnt =
-                                       po->stats_u.stats3.tp_freeze_q_cnt;
-                       st_u.stats3.tp_packets += po->stats.tp_drops;
-                       data = &st_u.stats3;
+                       data = &st.stats3;
                } else {
                        lv = sizeof(struct tpacket_stats);
-                       st = po->stats;
-                       st.tp_packets += st.tp_drops;
-                       data = &st;
+                       data = &st.stats1;
                }
-               memset(&po->stats, 0, sizeof(st));
-               spin_unlock_bh(&sk->sk_receive_queue.lock);
+
                break;
        case PACKET_AUXDATA:
                val = po->auxdata;

diff --git a/net/packet/internal.h b/net/packet/internal.h
index 650751b9a7ee9c3f92382f37f3a5339562a4c733..c4e4b4561207354c49c9797dd5ea405867de0bcd 100644 (file)
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -93,8 +93,7 @@ struct packet_sock {
        /* struct sock has to be the first member of packet_sock */
        struct sock             sk;
        struct packet_fanout    *fanout;
-       struct tpacket_stats    stats;
-       union  tpacket_stats_u  stats_u;
+       union  tpacket_stats_u  stats;
        struct packet_ring_buffer       rx_ring;
        struct packet_ring_buffer       tx_ring;
        int                     copy_thresh;