diff options
| author | Eric Dumazet <edumazet@google.com> | 2017-08-29 15:16:01 -0700 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2017-08-29 16:10:50 -0700 |
| commit | eaa72dc47488d599439cd0fd0f8c4f1bcb3906bb (patch) | |
| tree | d35829277414b465b199dcada5af2e70f6183533 /net/ipv4 | |
| parent | 0dd5759dbb1c9a862e7d90c09d6cf398c45f1100 (diff) | |
| download | linux-stable-eaa72dc47488d599439cd0fd0f8c4f1bcb3906bb.tar.gz linux-stable-eaa72dc47488d599439cd0fd0f8c4f1bcb3906bb.tar.bz2 linux-stable-eaa72dc47488d599439cd0fd0f8c4f1bcb3906bb.zip | |
neigh: increase queue_len_bytes to match wmem_default
Florian reported UDP xmit drops that could be root caused to the
too small neigh limit.
Current limit is 64 KB, meaning that even a single UDP socket would hit
it, since its default sk_sndbuf comes from net.core.wmem_default
(~212992 bytes on 64bit arches).
Once ARP/ND resolution is in progress, we should allow a little more
packets to be queued, at least for one producer.
Once neigh arp_queue is filled, a rogue socket should hit its sk_sndbuf
limit and either block in sendmsg() or return -EAGAIN.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4')
| -rw-r--r-- | net/ipv4/arp.c | 2 | ||||
| -rw-r--r-- | net/ipv4/tcp_input.c | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 8b52179ddc6e..7c45b8896709 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -171,7 +171,7 @@ struct neigh_table arp_tbl = { [NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ, [NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ, [NEIGH_VAR_GC_STALETIME] = 60 * HZ, - [NEIGH_VAR_QUEUE_LEN_BYTES] = 64 * 1024, + [NEIGH_VAR_QUEUE_LEN_BYTES] = SK_WMEM_MAX, [NEIGH_VAR_PROXY_QLEN] = 64, [NEIGH_VAR_ANYCAST_DELAY] = 1 * HZ, [NEIGH_VAR_PROXY_DELAY] = (8 * HZ) / 10, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 568ccfd6dd37..7616cd76f6f6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6086,9 +6086,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); struct sock *fastopen_sk = NULL; - struct dst_entry *dst = NULL; struct request_sock *req; bool want_cookie = false; + struct dst_entry *dst; struct flowi fl; /* TW buckets are converted to open requests without |