diff options
| author | David Howells <dhowells@redhat.com> | 2016-09-24 19:05:27 +0200 |
|---|---|---|
| committer | David Howells <dhowells@redhat.com> | 2016-09-25 00:49:46 +0200 |
| commit | 57494343cb5d66962bb197878fb1cc576177db31 (patch) | |
| tree | 97da848d2149916a2ff1f410fc912e139991a56e /net/rxrpc/call_object.c | |
| parent | rxrpc: Schedule an ACK if the reply to a client call appears overdue (diff) | |
| download | linux-57494343cb5d66962bb197878fb1cc576177db31.tar.xz linux-57494343cb5d66962bb197878fb1cc576177db31.zip | |
rxrpc: Implement slow-start
Implement RxRPC slow-start, which is similar to RFC 5681 for TCP. A
tracepoint is added to log the state of the congestion management algorithm
and the decisions it makes.
Notes:
(1) Since we send fixed-size DATA packets (apart from the final packet in
each phase), counters and calculations are in terms of packets rather
than bytes.
(2) The ACK packet carries the equivalent of TCP SACK.
(3) The FLIGHT_SIZE calculation in RFC 5681 doesn't seem particularly
suited to SACK of a small number of packets. It seems that, almost
inevitably, by the time three 'duplicate' ACKs have been seen, we have
narrowed the loss down to one or two missing packets, and the
FLIGHT_SIZE calculation ends up as 2.
(4) In rxrpc_resend(), if there was no data that apparently needed
retransmission, we transmit a PING ACK to ask the peer to tell us what
its Rx window state is.
Signed-off-by: David Howells <dhowells@redhat.com>
Diffstat (limited to 'net/rxrpc/call_object.c')
| -rw-r--r-- | net/rxrpc/call_object.c | 13 |
1 files changed, 13 insertions, 0 deletions
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index a53f4c2c0025..d4b3293b78fa 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -160,6 +160,14 @@ struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) call->rx_winsize = rxrpc_rx_window_size; call->tx_winsize = 16; call->rx_expect_next = 1; + + if (RXRPC_TX_SMSS > 2190) + call->cong_cwnd = 2; + else if (RXRPC_TX_SMSS > 1095) + call->cong_cwnd = 3; + else + call->cong_cwnd = 4; + call->cong_ssthresh = RXRPC_RXTX_BUFF_SIZE - 1; return call; nomem_2: @@ -176,6 +184,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call(struct sockaddr_rxrpc *srx, gfp_t gfp) { struct rxrpc_call *call; + ktime_t now; _enter(""); @@ -185,6 +194,9 @@ static struct rxrpc_call *rxrpc_alloc_client_call(struct sockaddr_rxrpc *srx, call->state = RXRPC_CALL_CLIENT_AWAIT_CONN; call->service_id = srx->srx_service; call->tx_phase = true; + now = ktime_get_real(); + call->acks_latest_ts = now; + call->cong_tstamp = now; _leave(" = %p", call); return call; @@ -325,6 +337,7 @@ void rxrpc_incoming_call(struct rxrpc_sock *rx, call->state = RXRPC_CALL_SERVER_ACCEPTING; if (sp->hdr.securityIndex > 0) call->state = RXRPC_CALL_SERVER_SECURING; + call->cong_tstamp = skb->tstamp; /* Set the channel for this call. We don't get channel_lock as we're * only defending against the data_ready handler (which we're called |