From cc5453a5b7e90c39f713091a7ebc53c1f87d1700 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Aug 2020 16:15:58 +0200 Subject: netfilter: conntrack: allow sctp hearbeat after connection re-use If an sctp connection gets re-used, heartbeats are flagged as invalid because their vtag doesn't match. Handle this in a similar way as TCP conntrack when it suspects that the endpoints and conntrack are out-of-sync. When a HEARTBEAT request fails its vtag validation, flag this in the conntrack state and accept the packet. When a HEARTBEAT_ACK is received with an invalid vtag in the reverse direction after we allowed such a HEARTBEAT through, assume we are out-of-sync and re-set the vtag info. v2: remove left-over snippet from an older incarnation that moved new_state/old_state assignments, thats not needed so keep that as-is. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_sctp.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h index 9a33f171aa82..625f491b95de 100644 --- a/include/linux/netfilter/nf_conntrack_sctp.h +++ b/include/linux/netfilter/nf_conntrack_sctp.h @@ -9,6 +9,8 @@ struct ip_ct_sctp { enum sctp_conntrack state; __be32 vtag[IP_CT_DIR_MAX]; + u8 last_dir; + u8 flags; }; #endif /* _NF_CONNTRACK_SCTP_H */ -- cgit v1.2.3 From 4700c4d80b7bb171f6996016ef121e1508860b42 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 19 Aug 2020 23:29:16 +0100 Subject: rxrpc: Fix loss of RTT samples due to interposed ACK The Rx protocol has a mechanism to help generate RTT samples that works by a client transmitting a REQUESTED-type ACK when it receives a DATA packet that has the REQUEST_ACK flag set. The peer, however, may interpose other ACKs before transmitting the REQUESTED-ACK, as can be seen in the following trace excerpt: rxrpc_tx_data: c=00000044 DATA d0b5ece8:00000001 00000001 q=00000001 fl=07 rxrpc_rx_ack: c=00000044 00000001 PNG r=00000000 f=00000002 p=00000000 n=0 rxrpc_rx_ack: c=00000044 00000002 REQ r=00000001 f=00000002 p=00000001 n=0 ... DATA packet 1 (q=xx) has REQUEST_ACK set (bit 1 of fl=xx). The incoming ping (labelled PNG) hard-acks the request DATA packet (f=xx exceeds the sequence number of the DATA packet), causing it to be discarded from the Tx ring. The ACK that was requested (labelled REQ, r=xx references the serial of the DATA packet) comes after the ping, but the sk_buff holding the timestamp has gone and the RTT sample is lost. This is particularly noticeable on RPC calls used to probe the service offered by the peer. A lot of peers end up with an unknown RTT because we only ever sent a single RPC. This confuses the server rotation algorithm. Fix this by caching the information about the outgoing packet in RTT calculations in the rxrpc_call struct rather than looking in the Tx ring. A four-deep buffer is maintained and both REQUEST_ACK-flagged DATA and PING-ACK transmissions are recorded in there. When the appropriate response ACK is received, the buffer is checked for a match and, if found, an RTT sample is recorded. If a received ACK refers to a packet with a later serial number than an entry in the cache, that entry is presumed lost and the entry is made available to record a new transmission. ACKs types other than REQUESTED-type and PING-type cause any matching sample to be cancelled as they don't necessarily represent a useful measurement. If there's no space in the buffer on ping/data transmission, the sample base is discarded. Fixes: 50235c4b5a2f ("rxrpc: Obtain RTT data by requesting ACKs on DATA packets") Signed-off-by: David Howells --- include/trace/events/rxrpc.h | 27 ++++++++--- net/rxrpc/ar-internal.h | 13 +++--- net/rxrpc/call_object.c | 1 + net/rxrpc/input.c | 104 +++++++++++++++++++++++++------------------ net/rxrpc/output.c | 82 +++++++++++++++++++++++++--------- net/rxrpc/rtt.c | 3 +- 6 files changed, 154 insertions(+), 76 deletions(-) (limited to 'include') diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 059b6e45a028..c33079b986e8 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -138,11 +138,16 @@ enum rxrpc_recvmsg_trace { }; enum rxrpc_rtt_tx_trace { + rxrpc_rtt_tx_cancel, rxrpc_rtt_tx_data, + rxrpc_rtt_tx_no_slot, rxrpc_rtt_tx_ping, }; enum rxrpc_rtt_rx_trace { + rxrpc_rtt_rx_cancel, + rxrpc_rtt_rx_lost, + rxrpc_rtt_rx_obsolete, rxrpc_rtt_rx_ping_response, rxrpc_rtt_rx_requested_ack, }; @@ -339,10 +344,15 @@ enum rxrpc_tx_point { E_(rxrpc_recvmsg_wait, "WAIT") #define rxrpc_rtt_tx_traces \ + EM(rxrpc_rtt_tx_cancel, "CNCE") \ EM(rxrpc_rtt_tx_data, "DATA") \ + EM(rxrpc_rtt_tx_no_slot, "FULL") \ E_(rxrpc_rtt_tx_ping, "PING") #define rxrpc_rtt_rx_traces \ + EM(rxrpc_rtt_rx_cancel, "CNCL") \ + EM(rxrpc_rtt_rx_obsolete, "OBSL") \ + EM(rxrpc_rtt_rx_lost, "LOST") \ EM(rxrpc_rtt_rx_ping_response, "PONG") \ E_(rxrpc_rtt_rx_requested_ack, "RACK") @@ -1087,38 +1097,43 @@ TRACE_EVENT(rxrpc_recvmsg, TRACE_EVENT(rxrpc_rtt_tx, TP_PROTO(struct rxrpc_call *call, enum rxrpc_rtt_tx_trace why, - rxrpc_serial_t send_serial), + int slot, rxrpc_serial_t send_serial), - TP_ARGS(call, why, send_serial), + TP_ARGS(call, why, slot, send_serial), TP_STRUCT__entry( __field(unsigned int, call ) __field(enum rxrpc_rtt_tx_trace, why ) + __field(int, slot ) __field(rxrpc_serial_t, send_serial ) ), TP_fast_assign( __entry->call = call->debug_id; __entry->why = why; + __entry->slot = slot; __entry->send_serial = send_serial; ), - TP_printk("c=%08x %s sr=%08x", + TP_printk("c=%08x [%d] %s sr=%08x", __entry->call, + __entry->slot, __print_symbolic(__entry->why, rxrpc_rtt_tx_traces), __entry->send_serial) ); TRACE_EVENT(rxrpc_rtt_rx, TP_PROTO(struct rxrpc_call *call, enum rxrpc_rtt_rx_trace why, + int slot, rxrpc_serial_t send_serial, rxrpc_serial_t resp_serial, u32 rtt, u32 rto), - TP_ARGS(call, why, send_serial, resp_serial, rtt, rto), + TP_ARGS(call, why, slot, send_serial, resp_serial, rtt, rto), TP_STRUCT__entry( __field(unsigned int, call ) __field(enum rxrpc_rtt_rx_trace, why ) + __field(int, slot ) __field(rxrpc_serial_t, send_serial ) __field(rxrpc_serial_t, resp_serial ) __field(u32, rtt ) @@ -1128,14 +1143,16 @@ TRACE_EVENT(rxrpc_rtt_rx, TP_fast_assign( __entry->call = call->debug_id; __entry->why = why; + __entry->slot = slot; __entry->send_serial = send_serial; __entry->resp_serial = resp_serial; __entry->rtt = rtt; __entry->rto = rto; ), - TP_printk("c=%08x %s sr=%08x rr=%08x rtt=%u rto=%u", + TP_printk("c=%08x [%d] %s sr=%08x rr=%08x rtt=%u rto=%u", __entry->call, + __entry->slot, __print_symbolic(__entry->why, rxrpc_rtt_rx_traces), __entry->send_serial, __entry->resp_serial, diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 6d29a3603a3e..884cff7bb169 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -488,7 +488,6 @@ enum rxrpc_call_flag { RXRPC_CALL_RX_LAST, /* Received the last packet (at rxtx_top) */ RXRPC_CALL_TX_LAST, /* Last packet in Tx buffer (at rxtx_top) */ RXRPC_CALL_SEND_PING, /* A ping will need to be sent */ - RXRPC_CALL_PINGING, /* Ping in process */ RXRPC_CALL_RETRANS_TIMEOUT, /* Retransmission due to timeout occurred */ RXRPC_CALL_BEGAN_RX_TIMER, /* We began the expect_rx_by timer */ RXRPC_CALL_RX_HEARD, /* The peer responded at least once to this call */ @@ -673,9 +672,13 @@ struct rxrpc_call { rxrpc_seq_t ackr_consumed; /* Highest packet shown consumed */ rxrpc_seq_t ackr_seen; /* Highest packet shown seen */ - /* ping management */ - rxrpc_serial_t ping_serial; /* Last ping sent */ - ktime_t ping_time; /* Time last ping sent */ + /* RTT management */ + rxrpc_serial_t rtt_serial[4]; /* Serial number of DATA or PING sent */ + ktime_t rtt_sent_at[4]; /* Time packet sent */ + unsigned long rtt_avail; /* Mask of available slots in bits 0-3, + * Mask of pending samples in 8-11 */ +#define RXRPC_CALL_RTT_AVAIL_MASK 0xf +#define RXRPC_CALL_RTT_PEND_SHIFT 8 /* transmission-phase ACK management */ ktime_t acks_latest_ts; /* Timestamp of latest ACK received */ @@ -1037,7 +1040,7 @@ static inline bool __rxrpc_abort_eproto(struct rxrpc_call *call, /* * rtt.c */ -void rxrpc_peer_add_rtt(struct rxrpc_call *, enum rxrpc_rtt_rx_trace, +void rxrpc_peer_add_rtt(struct rxrpc_call *, enum rxrpc_rtt_rx_trace, int, rxrpc_serial_t, rxrpc_serial_t, ktime_t, ktime_t); unsigned long rxrpc_get_rto_backoff(struct rxrpc_peer *, bool); void rxrpc_peer_init_rtt(struct rxrpc_peer *); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 38a46167523f..a40fae013942 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -153,6 +153,7 @@ struct rxrpc_call *rxrpc_alloc_call(struct rxrpc_sock *rx, gfp_t gfp, call->cong_ssthresh = RXRPC_RXTX_BUFF_SIZE - 1; call->rxnet = rxnet; + call->rtt_avail = RXRPC_CALL_RTT_AVAIL_MASK; atomic_inc(&rxnet->nr_calls); return call; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index a7699e56eac8..19ddfc9807e8 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -608,36 +608,57 @@ unlock: } /* - * Process a requested ACK. + * See if there's a cached RTT probe to complete. */ -static void rxrpc_input_requested_ack(struct rxrpc_call *call, - ktime_t resp_time, - rxrpc_serial_t orig_serial, - rxrpc_serial_t ack_serial) +static void rxrpc_complete_rtt_probe(struct rxrpc_call *call, + ktime_t resp_time, + rxrpc_serial_t acked_serial, + rxrpc_serial_t ack_serial, + enum rxrpc_rtt_rx_trace type) { - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; + rxrpc_serial_t orig_serial; + unsigned long avail; ktime_t sent_at; - int ix; + bool matched = false; + int i; - for (ix = 0; ix < RXRPC_RXTX_BUFF_SIZE; ix++) { - skb = call->rxtx_buffer[ix]; - if (!skb) - continue; + avail = READ_ONCE(call->rtt_avail); + smp_rmb(); /* Read avail bits before accessing data. */ - sent_at = skb->tstamp; - smp_rmb(); /* Read timestamp before serial. */ - sp = rxrpc_skb(skb); - if (sp->hdr.serial != orig_serial) + for (i = 0; i < ARRAY_SIZE(call->rtt_serial); i++) { + if (!test_bit(i + RXRPC_CALL_RTT_PEND_SHIFT, &avail)) continue; - goto found; - } - return; + sent_at = call->rtt_sent_at[i]; + orig_serial = call->rtt_serial[i]; + + if (orig_serial == acked_serial) { + clear_bit(i + RXRPC_CALL_RTT_PEND_SHIFT, &call->rtt_avail); + smp_mb(); /* Read data before setting avail bit */ + set_bit(i, &call->rtt_avail); + if (type != rxrpc_rtt_rx_cancel) + rxrpc_peer_add_rtt(call, type, i, acked_serial, ack_serial, + sent_at, resp_time); + else + trace_rxrpc_rtt_rx(call, rxrpc_rtt_rx_cancel, i, + orig_serial, acked_serial, 0, 0); + matched = true; + } + + /* If a later serial is being acked, then mark this slot as + * being available. + */ + if (after(acked_serial, orig_serial)) { + trace_rxrpc_rtt_rx(call, rxrpc_rtt_rx_obsolete, i, + orig_serial, acked_serial, 0, 0); + clear_bit(i + RXRPC_CALL_RTT_PEND_SHIFT, &call->rtt_avail); + smp_wmb(); + set_bit(i, &call->rtt_avail); + } + } -found: - rxrpc_peer_add_rtt(call, rxrpc_rtt_rx_requested_ack, - orig_serial, ack_serial, sent_at, resp_time); + if (!matched) + trace_rxrpc_rtt_rx(call, rxrpc_rtt_rx_lost, 9, 0, acked_serial, 0, 0); } /* @@ -682,27 +703,11 @@ static void rxrpc_input_check_for_lost_ack(struct rxrpc_call *call) */ static void rxrpc_input_ping_response(struct rxrpc_call *call, ktime_t resp_time, - rxrpc_serial_t orig_serial, + rxrpc_serial_t acked_serial, rxrpc_serial_t ack_serial) { - rxrpc_serial_t ping_serial; - ktime_t ping_time; - - ping_time = call->ping_time; - smp_rmb(); - ping_serial = READ_ONCE(call->ping_serial); - - if (orig_serial == call->acks_lost_ping) + if (acked_serial == call->acks_lost_ping) rxrpc_input_check_for_lost_ack(call); - - if (before(orig_serial, ping_serial) || - !test_and_clear_bit(RXRPC_CALL_PINGING, &call->flags)) - return; - if (after(orig_serial, ping_serial)) - return; - - rxrpc_peer_add_rtt(call, rxrpc_rtt_rx_ping_response, - orig_serial, ack_serial, ping_time, resp_time); } /* @@ -869,12 +874,23 @@ static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb) first_soft_ack, prev_pkt, summary.ack_reason, nr_acks); - if (buf.ack.reason == RXRPC_ACK_PING_RESPONSE) + switch (buf.ack.reason) { + case RXRPC_ACK_PING_RESPONSE: rxrpc_input_ping_response(call, skb->tstamp, acked_serial, ack_serial); - if (buf.ack.reason == RXRPC_ACK_REQUESTED) - rxrpc_input_requested_ack(call, skb->tstamp, acked_serial, - ack_serial); + rxrpc_complete_rtt_probe(call, skb->tstamp, acked_serial, ack_serial, + rxrpc_rtt_rx_ping_response); + break; + case RXRPC_ACK_REQUESTED: + rxrpc_complete_rtt_probe(call, skb->tstamp, acked_serial, ack_serial, + rxrpc_rtt_rx_requested_ack); + break; + default: + if (acked_serial != 0) + rxrpc_complete_rtt_probe(call, skb->tstamp, acked_serial, ack_serial, + rxrpc_rtt_rx_cancel); + break; + } if (buf.ack.reason == RXRPC_ACK_PING) { _proto("Rx ACK %%%u PING Request", ack_serial); diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 1ba43c3df4ad..3cfff7922ba8 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -123,6 +123,49 @@ static size_t rxrpc_fill_out_ack(struct rxrpc_connection *conn, return top - hard_ack + 3; } +/* + * Record the beginning of an RTT probe. + */ +static int rxrpc_begin_rtt_probe(struct rxrpc_call *call, rxrpc_serial_t serial, + enum rxrpc_rtt_tx_trace why) +{ + unsigned long avail = call->rtt_avail; + int rtt_slot = 9; + + if (!(avail & RXRPC_CALL_RTT_AVAIL_MASK)) + goto no_slot; + + rtt_slot = __ffs(avail & RXRPC_CALL_RTT_AVAIL_MASK); + if (!test_and_clear_bit(rtt_slot, &call->rtt_avail)) + goto no_slot; + + call->rtt_serial[rtt_slot] = serial; + call->rtt_sent_at[rtt_slot] = ktime_get_real(); + smp_wmb(); /* Write data before avail bit */ + set_bit(rtt_slot + RXRPC_CALL_RTT_PEND_SHIFT, &call->rtt_avail); + + trace_rxrpc_rtt_tx(call, why, rtt_slot, serial); + return rtt_slot; + +no_slot: + trace_rxrpc_rtt_tx(call, rxrpc_rtt_tx_no_slot, rtt_slot, serial); + return -1; +} + +/* + * Cancel an RTT probe. + */ +static void rxrpc_cancel_rtt_probe(struct rxrpc_call *call, + rxrpc_serial_t serial, int rtt_slot) +{ + if (rtt_slot != -1) { + clear_bit(rtt_slot + RXRPC_CALL_RTT_PEND_SHIFT, &call->rtt_avail); + smp_wmb(); /* Clear pending bit before setting slot */ + set_bit(rtt_slot, &call->rtt_avail); + trace_rxrpc_rtt_tx(call, rxrpc_rtt_tx_cancel, rtt_slot, serial); + } +} + /* * Send an ACK call packet. */ @@ -136,7 +179,7 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, rxrpc_serial_t serial; rxrpc_seq_t hard_ack, top; size_t len, n; - int ret; + int ret, rtt_slot = -1; u8 reason; if (test_bit(RXRPC_CALL_DISCONNECTED, &call->flags)) @@ -196,18 +239,8 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, if (_serial) *_serial = serial; - if (ping) { - call->ping_serial = serial; - smp_wmb(); - /* We need to stick a time in before we send the packet in case - * the reply gets back before kernel_sendmsg() completes - but - * asking UDP to send the packet can take a relatively long - * time. - */ - call->ping_time = ktime_get_real(); - set_bit(RXRPC_CALL_PINGING, &call->flags); - trace_rxrpc_rtt_tx(call, rxrpc_rtt_tx_ping, serial); - } + if (ping) + rtt_slot = rxrpc_begin_rtt_probe(call, serial, rxrpc_rtt_tx_ping); ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); conn->params.peer->last_tx_at = ktime_get_seconds(); @@ -221,8 +254,7 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, if (call->state < RXRPC_CALL_COMPLETE) { if (ret < 0) { - if (ping) - clear_bit(RXRPC_CALL_PINGING, &call->flags); + rxrpc_cancel_rtt_probe(call, serial, rtt_slot); rxrpc_propose_ACK(call, pkt->ack.reason, ntohl(pkt->ack.serial), false, true, @@ -321,7 +353,7 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb, struct kvec iov[2]; rxrpc_serial_t serial; size_t len; - int ret; + int ret, rtt_slot = -1; _enter(",{%d}", skb->len); @@ -397,6 +429,8 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb, sp->hdr.serial = serial; smp_wmb(); /* Set serial before timestamp */ skb->tstamp = ktime_get_real(); + if (whdr.flags & RXRPC_REQUEST_ACK) + rtt_slot = rxrpc_begin_rtt_probe(call, serial, rxrpc_rtt_tx_data); /* send the packet by UDP * - returns -EMSGSIZE if UDP would have to fragment the packet @@ -408,12 +442,15 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb, conn->params.peer->last_tx_at = ktime_get_seconds(); up_read(&conn->params.local->defrag_sem); - if (ret < 0) + if (ret < 0) { + rxrpc_cancel_rtt_probe(call, serial, rtt_slot); trace_rxrpc_tx_fail(call->debug_id, serial, ret, rxrpc_tx_point_call_data_nofrag); - else + } else { trace_rxrpc_tx_packet(call->debug_id, &whdr, rxrpc_tx_point_call_data_nofrag); + } + rxrpc_tx_backoff(call, ret); if (ret == -EMSGSIZE) goto send_fragmentable; @@ -422,7 +459,6 @@ done: if (ret >= 0) { if (whdr.flags & RXRPC_REQUEST_ACK) { call->peer->rtt_last_req = skb->tstamp; - trace_rxrpc_rtt_tx(call, rxrpc_rtt_tx_data, serial); if (call->peer->rtt_count > 1) { unsigned long nowj = jiffies, ack_lost_at; @@ -469,6 +505,8 @@ send_fragmentable: sp->hdr.serial = serial; smp_wmb(); /* Set serial before timestamp */ skb->tstamp = ktime_get_real(); + if (whdr.flags & RXRPC_REQUEST_ACK) + rtt_slot = rxrpc_begin_rtt_probe(call, serial, rxrpc_rtt_tx_data); switch (conn->params.local->srx.transport.family) { case AF_INET6: @@ -487,12 +525,14 @@ send_fragmentable: BUG(); } - if (ret < 0) + if (ret < 0) { + rxrpc_cancel_rtt_probe(call, serial, rtt_slot); trace_rxrpc_tx_fail(call->debug_id, serial, ret, rxrpc_tx_point_call_data_frag); - else + } else { trace_rxrpc_tx_packet(call->debug_id, &whdr, rxrpc_tx_point_call_data_frag); + } rxrpc_tx_backoff(call, ret); up_write(&conn->params.local->defrag_sem); diff --git a/net/rxrpc/rtt.c b/net/rxrpc/rtt.c index 928d8b34a3ee..1221b0637a7e 100644 --- a/net/rxrpc/rtt.c +++ b/net/rxrpc/rtt.c @@ -146,6 +146,7 @@ static void rxrpc_ack_update_rtt(struct rxrpc_peer *peer, long rtt_us) * exclusive access to the peer RTT data. */ void rxrpc_peer_add_rtt(struct rxrpc_call *call, enum rxrpc_rtt_rx_trace why, + int rtt_slot, rxrpc_serial_t send_serial, rxrpc_serial_t resp_serial, ktime_t send_time, ktime_t resp_time) { @@ -162,7 +163,7 @@ void rxrpc_peer_add_rtt(struct rxrpc_call *call, enum rxrpc_rtt_rx_trace why, peer->rtt_count++; spin_unlock(&peer->rtt_input_lock); - trace_rxrpc_rtt_rx(call, why, send_serial, resp_serial, + trace_rxrpc_rtt_rx(call, why, rtt_slot, send_serial, resp_serial, peer->srtt_us >> 3, peer->rto_j); } -- cgit v1.2.3 From 1d4adfaf65746203861c72d9d78de349eb97d528 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 20 Aug 2020 15:13:00 +0100 Subject: rxrpc: Make rxrpc_kernel_get_srtt() indicate validity Fix rxrpc_kernel_get_srtt() to indicate the validity of the returned smoothed RTT. If we haven't had any valid samples yet, the SRTT isn't useful. Fixes: c410bf01933e ("rxrpc: Fix the excessive initial retransmission timeout") Signed-off-by: David Howells --- fs/afs/fs_probe.c | 4 ++-- fs/afs/vl_probe.c | 4 ++-- include/net/af_rxrpc.h | 2 +- net/rxrpc/peer_object.c | 16 +++++++++++++--- 4 files changed, 18 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/fs/afs/fs_probe.c b/fs/afs/fs_probe.c index 5d9ef517cf81..e7e98ad63a91 100644 --- a/fs/afs/fs_probe.c +++ b/fs/afs/fs_probe.c @@ -161,8 +161,8 @@ responded: } } - rtt_us = rxrpc_kernel_get_srtt(call->net->socket, call->rxcall); - if (rtt_us < server->probe.rtt) { + if (rxrpc_kernel_get_srtt(call->net->socket, call->rxcall, &rtt_us) && + rtt_us < server->probe.rtt) { server->probe.rtt = rtt_us; server->rtt = rtt_us; alist->preferred = index; diff --git a/fs/afs/vl_probe.c b/fs/afs/vl_probe.c index e3aa013c2177..081b7e5b13f5 100644 --- a/fs/afs/vl_probe.c +++ b/fs/afs/vl_probe.c @@ -92,8 +92,8 @@ responded: } } - rtt_us = rxrpc_kernel_get_srtt(call->net->socket, call->rxcall); - if (rtt_us < server->probe.rtt) { + if (rxrpc_kernel_get_srtt(call->net->socket, call->rxcall, &rtt_us) && + rtt_us < server->probe.rtt) { server->probe.rtt = rtt_us; alist->preferred = index; have_result = true; diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h index 91eacbdcf33d..f6abcc0bbd6e 100644 --- a/include/net/af_rxrpc.h +++ b/include/net/af_rxrpc.h @@ -59,7 +59,7 @@ bool rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *, void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *); void rxrpc_kernel_get_peer(struct socket *, struct rxrpc_call *, struct sockaddr_rxrpc *); -u32 rxrpc_kernel_get_srtt(struct socket *, struct rxrpc_call *); +bool rxrpc_kernel_get_srtt(struct socket *, struct rxrpc_call *, u32 *); int rxrpc_kernel_charge_accept(struct socket *, rxrpc_notify_rx_t, rxrpc_user_attach_call_t, unsigned long, gfp_t, unsigned int); diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index ca29976bb193..68396d052052 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -502,11 +502,21 @@ EXPORT_SYMBOL(rxrpc_kernel_get_peer); * rxrpc_kernel_get_srtt - Get a call's peer smoothed RTT * @sock: The socket on which the call is in progress. * @call: The call to query + * @_srtt: Where to store the SRTT value. * - * Get the call's peer smoothed RTT. + * Get the call's peer smoothed RTT in uS. */ -u32 rxrpc_kernel_get_srtt(struct socket *sock, struct rxrpc_call *call) +bool rxrpc_kernel_get_srtt(struct socket *sock, struct rxrpc_call *call, + u32 *_srtt) { - return call->peer->srtt_us >> 3; + struct rxrpc_peer *peer = call->peer; + + if (peer->rtt_count == 0) { + *_srtt = 1000000; /* 1S */ + return false; + } + + *_srtt = call->peer->srtt_us >> 3; + return true; } EXPORT_SYMBOL(rxrpc_kernel_get_srtt); -- cgit v1.2.3 From da9125df854ea48a6240c66e8a67be06e2c12c03 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 20 Aug 2020 14:12:55 +0200 Subject: netfilter: nf_tables: incorrect enum nft_list_attributes definition This should be NFTA_LIST_UNSPEC instead of NFTA_LIST_UNPEC, all other similar attribute definitions are postfixed with _UNSPEC. Fixes: 96518518cc41 ("netfilter: add nftables") Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 42f351c1f5c5..2b8e12f7a4a6 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -133,7 +133,7 @@ enum nf_tables_msg_types { * @NFTA_LIST_ELEM: list element (NLA_NESTED) */ enum nft_list_attributes { - NFTA_LIST_UNPEC, + NFTA_LIST_UNSPEC, NFTA_LIST_ELEM, __NFTA_LIST_MAX }; -- cgit v1.2.3 From 1e105e6afa6c3d32bfb52c00ffa393894a525c27 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 20 Aug 2020 21:05:50 +0200 Subject: netfilter: nf_tables: fix destination register zeroing Following bug was reported via irc: nft list ruleset set knock_candidates_ipv4 { type ipv4_addr . inet_service size 65535 elements = { 127.0.0.1 . 123, 127.0.0.1 . 123 } } .. udp dport 123 add @knock_candidates_ipv4 { ip saddr . 123 } udp dport 123 add @knock_candidates_ipv4 { ip saddr . udp dport } It should not have been possible to add a duplicate set entry. After some debugging it turned out that the problem is the immediate value (123) in the second-to-last rule. Concatenations use 32bit registers, i.e. the elements are 8 bytes each, not 6 and it turns out the kernel inserted inet firewall @knock_candidates_ipv4 element 0100007f ffff7b00 : 0 [end] element 0100007f 00007b00 : 0 [end] Note the non-zero upper bits of the first element. It turns out that nft_immediate doesn't zero the destination register, but this is needed when the length isn't a multiple of 4. Furthermore, the zeroing in nft_payload is broken. We can't use [len / 4] = 0 -- if len is a multiple of 4, index is off by one. Skip zeroing in this case and use a conditional instead of (len -1) / 4. Fixes: 49499c3e6e18 ("netfilter: nf_tables: switch registers to 32 bit addressing") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 2 ++ net/netfilter/nft_payload.c | 4 +++- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index bf9491b77d16..224d194ad29d 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -143,6 +143,8 @@ static inline u64 nft_reg_load64(const u32 *sreg) static inline void nft_data_copy(u32 *dst, const struct nft_data *src, unsigned int len) { + if (len % NFT_REG32_SIZE) + dst[len / NFT_REG32_SIZE] = 0; memcpy(dst, src, len); } diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index ed7cb9f747f6..7a2e59638499 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -87,7 +87,9 @@ void nft_payload_eval(const struct nft_expr *expr, u32 *dest = ®s->data[priv->dreg]; int offset; - dest[priv->len / NFT_REG32_SIZE] = 0; + if (priv->len % NFT_REG32_SIZE) + dest[priv->len / NFT_REG32_SIZE] = 0; + switch (priv->base) { case NFT_PAYLOAD_LL_HEADER: if (!skb_mac_header_was_set(skb)) -- cgit v1.2.3 From 3622adb02623cde94ea85c68e9a968d0059de545 Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Mon, 24 Aug 2020 13:46:22 +0200 Subject: ipv6: ndisc: adjust ndisc_ifinfo_sysctl_change prototype Commit 32927393dc1c ("sysctl: pass kernel pointers to ->proc_handler") changed ndisc_ifinfo_sysctl_change to take a kernel pointer. Adjust its prototype in net/ndisc.h as well to fix the following sparse warning: net/ipv6/ndisc.c:1838:5: error: symbol 'ndisc_ifinfo_sysctl_change' redeclared with different type (incompatible argument 3 (different address spaces)): net/ipv6/ndisc.c:1838:5: int extern [addressable] [signed] [toplevel] ndisc_ifinfo_sysctl_change( ... ) net/ipv6/ndisc.c: note: in included file (through include/net/ipv6.h): ./include/net/ndisc.h:496:5: note: previously declared as: ./include/net/ndisc.h:496:5: int extern [addressable] [signed] [toplevel] ndisc_ifinfo_sysctl_change( ... ) net/ipv6/ndisc.c: note: in included file (through include/net/ip6_route.h): Fixes: 32927393dc1c ("sysctl: pass kernel pointers to ->proc_handler") Cc: Christoph Hellwig Signed-off-by: Tobias Klauser Signed-off-by: David S. Miller --- include/net/ndisc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 9205a76d967a..38e4094960ce 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -494,7 +494,7 @@ int igmp6_event_report(struct sk_buff *skb); #ifdef CONFIG_SYSCTL int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, - void __user *buffer, size_t *lenp, loff_t *ppos); + void *buffer, size_t *lenp, loff_t *ppos); int ndisc_ifinfo_sysctl_strategy(struct ctl_table *ctl, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen); -- cgit v1.2.3 From be769db2f95861cc8c7c8fedcc71a8c39b803b10 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 22 Aug 2020 08:23:29 +1000 Subject: net: Get rid of consume_skb when tracing is off The function consume_skb is only meaningful when tracing is enabled. This patch makes it conditional on CONFIG_TRACEPOINTS. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/skbuff.h | 9 +++++++++ net/core/skbuff.c | 2 ++ 2 files changed, 11 insertions(+) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 46881d902124..e8bca74857a3 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1056,7 +1056,16 @@ void kfree_skb(struct sk_buff *skb); void kfree_skb_list(struct sk_buff *segs); void skb_dump(const char *level, const struct sk_buff *skb, bool full_pkt); void skb_tx_error(struct sk_buff *skb); + +#ifdef CONFIG_TRACEPOINTS void consume_skb(struct sk_buff *skb); +#else +static inline void consume_skb(struct sk_buff *skb) +{ + return kfree_skb(skb); +} +#endif + void __consume_stateless_skb(struct sk_buff *skb); void __kfree_skb(struct sk_buff *skb); extern struct kmem_cache *skbuff_head_cache; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index e18184ffa9c3..6faf73d6a0f7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -820,6 +820,7 @@ void skb_tx_error(struct sk_buff *skb) } EXPORT_SYMBOL(skb_tx_error); +#ifdef CONFIG_TRACEPOINTS /** * consume_skb - free an skbuff * @skb: buffer to free @@ -837,6 +838,7 @@ void consume_skb(struct sk_buff *skb) __kfree_skb(skb); } EXPORT_SYMBOL(consume_skb); +#endif /** * consume_stateless_skb - free an skbuff, assuming it is stateless -- cgit v1.2.3 From 645f08975f49441b3e753d8dc5b740cbcb226594 Mon Sep 17 00:00:00 2001 From: Miaohe Lin Date: Thu, 27 Aug 2020 07:27:49 -0400 Subject: net: Fix some comments Fix some comments, including wrong function name, duplicated word and so on. Signed-off-by: Miaohe Lin Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- include/uapi/linux/in.h | 2 +- net/core/sock.c | 2 +- net/ipv4/raw.c | 2 +- net/l3mdev/l3mdev.c | 2 +- net/socket.c | 4 ++-- 6 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e8bca74857a3..8d9ab50b08c9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -71,7 +71,7 @@ * NETIF_F_IPV6_CSUM - Driver (device) is only able to checksum plain * TCP or UDP packets over IPv6. These are specifically * unencapsulated packets of the form IPv6|TCP or - * IPv4|UDP where the Next Header field in the IPv6 + * IPv6|UDP where the Next Header field in the IPv6 * header is either TCP or UDP. IPv6 extension headers * are not supported with this feature. This feature * cannot be set in features for a device with @@ -2667,7 +2667,7 @@ static inline int pskb_network_may_pull(struct sk_buff *skb, unsigned int len) * * Using max(32, L1_CACHE_BYTES) makes sense (especially with RPS) * to reduce average number of cache lines per packet. - * get_rps_cpus() for example only access one 64 bytes aligned block : + * get_rps_cpu() for example only access one 64 bytes aligned block : * NET_IP_ALIGN(2) + ethernet_header(14) + IP_header(20/40) + ports(8) */ #ifndef NET_SKB_PAD diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h index 3d0d8231dc19..7d6687618d80 100644 --- a/include/uapi/linux/in.h +++ b/include/uapi/linux/in.h @@ -135,7 +135,7 @@ struct in_addr { * this socket to prevent accepting spoofed ones. */ #define IP_PMTUDISC_INTERFACE 4 -/* weaker version of IP_PMTUDISC_INTERFACE, which allos packets to get +/* weaker version of IP_PMTUDISC_INTERFACE, which allows packets to get * fragmented if they exeed the interface mtu */ #define IP_PMTUDISC_OMIT 5 diff --git a/net/core/sock.c b/net/core/sock.c index e4f40b175acb..8eb2c924805a 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3254,7 +3254,7 @@ void sk_common_release(struct sock *sk) sk->sk_prot->destroy(sk); /* - * Observation: when sock_common_release is called, processes have + * Observation: when sk_common_release is called, processes have * no access to socket. But net still has. * Step one, detach it from networking: * diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 6fd4330287c2..407956be7deb 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -610,7 +610,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } else if (!ipc.oif) { ipc.oif = inet->uc_index; } else if (ipv4_is_lbcast(daddr) && inet->uc_index) { - /* oif is set, packet is to local broadcast and + /* oif is set, packet is to local broadcast * and uc_index is set. oif is most likely set * by sk_bound_dev_if. If uc_index != oif check if the * oif is an L3 master and uc_index is an L3 slave. diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c index e71ca5aec684..864326f150e2 100644 --- a/net/l3mdev/l3mdev.c +++ b/net/l3mdev/l3mdev.c @@ -154,7 +154,7 @@ int l3mdev_master_upper_ifindex_by_index_rcu(struct net *net, int ifindex) EXPORT_SYMBOL_GPL(l3mdev_master_upper_ifindex_by_index_rcu); /** - * l3mdev_fib_table - get FIB table id associated with an L3 + * l3mdev_fib_table_rcu - get FIB table id associated with an L3 * master interface * @dev: targeted interface */ diff --git a/net/socket.c b/net/socket.c index dbbe8ea7d395..0c0144604f81 100644 --- a/net/socket.c +++ b/net/socket.c @@ -3610,7 +3610,7 @@ int kernel_getsockname(struct socket *sock, struct sockaddr *addr) EXPORT_SYMBOL(kernel_getsockname); /** - * kernel_peername - get the address which the socket is connected (kernel space) + * kernel_getpeername - get the address which the socket is connected (kernel space) * @sock: socket * @addr: address holder * @@ -3671,7 +3671,7 @@ int kernel_sendpage_locked(struct sock *sk, struct page *page, int offset, EXPORT_SYMBOL(kernel_sendpage_locked); /** - * kernel_shutdown - shut down part of a full-duplex connection (kernel space) + * kernel_sock_shutdown - shut down part of a full-duplex connection (kernel space) * @sock: socket * @how: connection part * -- cgit v1.2.3 From ee921183557af39c1a0475f982d43b0fcac25e2e Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 23 Aug 2020 13:55:36 +0200 Subject: netfilter: nfnetlink: nfnetlink_unicast() reports EAGAIN instead of ENOBUFS Frontend callback reports EAGAIN to nfnetlink to retry a command, this is used to signal that module autoloading is required. Unfortunately, nlmsg_unicast() reports EAGAIN in case the receiver socket buffer gets full, so it enters a busy-loop. This patch updates nfnetlink_unicast() to turn EAGAIN into ENOBUFS and to use nlmsg_unicast(). Remove the flags field in nfnetlink_unicast() since this is always MSG_DONTWAIT in the existing code which is exactly what nlmsg_unicast() passes to netlink_unicast() as parameter. Fixes: 96518518cc41 ("netfilter: add nftables") Reported-by: Phil Sutter Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nfnetlink.h | 3 +- net/netfilter/nf_tables_api.c | 61 ++++++++++++++++++------------------- net/netfilter/nfnetlink.c | 11 +++++-- net/netfilter/nfnetlink_log.c | 3 +- net/netfilter/nfnetlink_queue.c | 2 +- 5 files changed, 40 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 851425c3178f..89016d08f6a2 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -43,8 +43,7 @@ int nfnetlink_has_listeners(struct net *net, unsigned int group); int nfnetlink_send(struct sk_buff *skb, struct net *net, u32 portid, unsigned int group, int echo, gfp_t flags); int nfnetlink_set_err(struct net *net, u32 portid, u32 group, int error); -int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u32 portid, - int flags); +int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u32 portid); static inline u16 nfnl_msg_type(u8 subsys, u8 msg_type) { diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 71e501c5ad21..b7dc1cbf40ea 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -815,11 +815,11 @@ static int nf_tables_gettable(struct net *net, struct sock *nlsk, nlh->nlmsg_seq, NFT_MSG_NEWTABLE, 0, family, table); if (err < 0) - goto err; + goto err_fill_table_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); -err: +err_fill_table_info: kfree_skb(skb2); return err; } @@ -1563,11 +1563,11 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk, nlh->nlmsg_seq, NFT_MSG_NEWCHAIN, 0, family, table, chain); if (err < 0) - goto err; + goto err_fill_chain_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); -err: +err_fill_chain_info: kfree_skb(skb2); return err; } @@ -3008,11 +3008,11 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, nlh->nlmsg_seq, NFT_MSG_NEWRULE, 0, family, table, chain, rule, NULL); if (err < 0) - goto err; + goto err_fill_rule_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); -err: +err_fill_rule_info: kfree_skb(skb2); return err; } @@ -3968,11 +3968,11 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk, err = nf_tables_fill_set(skb2, &ctx, set, NFT_MSG_NEWSET, 0); if (err < 0) - goto err; + goto err_fill_set_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); -err: +err_fill_set_info: kfree_skb(skb2); return err; } @@ -4860,24 +4860,18 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set, err = -ENOMEM; skb = nlmsg_new(NLMSG_GOODSIZE, GFP_ATOMIC); if (skb == NULL) - goto err1; + return err; err = nf_tables_fill_setelem_info(skb, ctx, ctx->seq, ctx->portid, NFT_MSG_NEWSETELEM, 0, set, &elem); if (err < 0) - goto err2; + goto err_fill_setelem; - err = nfnetlink_unicast(skb, ctx->net, ctx->portid, MSG_DONTWAIT); - /* This avoids a loop in nfnetlink. */ - if (err < 0) - goto err1; + return nfnetlink_unicast(skb, ctx->net, ctx->portid); - return 0; -err2: +err_fill_setelem: kfree_skb(skb); -err1: - /* this avoids a loop in nfnetlink. */ - return err == -EAGAIN ? -ENOBUFS : err; + return err; } /* called with rcu_read_lock held */ @@ -6182,10 +6176,11 @@ static int nf_tables_getobj(struct net *net, struct sock *nlsk, nlh->nlmsg_seq, NFT_MSG_NEWOBJ, 0, family, table, obj, reset); if (err < 0) - goto err; + goto err_fill_obj_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); -err: + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); + +err_fill_obj_info: kfree_skb(skb2); return err; } @@ -7045,10 +7040,11 @@ static int nf_tables_getflowtable(struct net *net, struct sock *nlsk, NFT_MSG_NEWFLOWTABLE, 0, family, flowtable, &flowtable->hook_list); if (err < 0) - goto err; + goto err_fill_flowtable_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); -err: + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); + +err_fill_flowtable_info: kfree_skb(skb2); return err; } @@ -7234,10 +7230,11 @@ static int nf_tables_getgen(struct net *net, struct sock *nlsk, err = nf_tables_fill_gen_info(skb2, net, NETLINK_CB(skb).portid, nlh->nlmsg_seq); if (err < 0) - goto err; + goto err_fill_gen_info; - return nlmsg_unicast(nlsk, skb2, NETLINK_CB(skb).portid); -err: + return nfnetlink_unicast(skb2, net, NETLINK_CB(skb).portid); + +err_fill_gen_info: kfree_skb(skb2); return err; } diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 5f24edf95830..3a2e64e13b22 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -149,10 +149,15 @@ int nfnetlink_set_err(struct net *net, u32 portid, u32 group, int error) } EXPORT_SYMBOL_GPL(nfnetlink_set_err); -int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u32 portid, - int flags) +int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u32 portid) { - return netlink_unicast(net->nfnl, skb, portid, flags); + int err; + + err = nlmsg_unicast(net->nfnl, skb, portid); + if (err == -EAGAIN) + err = -ENOBUFS; + + return err; } EXPORT_SYMBOL_GPL(nfnetlink_unicast); diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index f02992419850..b35e8d9a5b37 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -356,8 +356,7 @@ __nfulnl_send(struct nfulnl_instance *inst) goto out; } } - nfnetlink_unicast(inst->skb, inst->net, inst->peer_portid, - MSG_DONTWAIT); + nfnetlink_unicast(inst->skb, inst->net, inst->peer_portid); out: inst->qlen = 0; inst->skb = NULL; diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index dadfc06245a3..d1d8bca03b4f 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -681,7 +681,7 @@ __nfqnl_enqueue_packet(struct net *net, struct nfqnl_instance *queue, *packet_id_ptr = htonl(entry->id); /* nfnetlink_unicast will either free the nskb or add it to a socket */ - err = nfnetlink_unicast(nskb, net, queue->peer_portid, MSG_DONTWAIT); + err = nfnetlink_unicast(nskb, net, queue->peer_portid); if (err < 0) { if (queue->flags & NFQA_CFG_F_FAIL_OPEN) { failopen = 1; -- cgit v1.2.3