* [PATCH net-next v3 1/3] Open out the check for Nagle
@ 2014-07-18 17:36 David Laight
2014-07-21 23:25 ` David Miller
0 siblings, 1 reply; 3+ messages in thread
From: David Laight @ 2014-07-18 17:36 UTC (permalink / raw)
To: netdev, 'linux-sctp@vger.kernel.org'; +Cc: 'David Miller'
The check for Nagle contains 6 separate checks all of which must be true
before a data packet is delayed.
Separate out each into its own 'if (test) return SCTP_XMIT_OK' so that
the reasons can be individually described.
Also return directly with SCTP_XMIT_RWND_FULL.
Delete the now-unused 'retval' variable and 'finish' label from
sctp_packet_can_append_data().
Signed-off-by: David Laight <david.laight@aculab.com>
---
Changes from v2:
- Fix inverted test for 'inflight == 0'
net/sctp/output.c | 69 +++++++++++++++++++++++++++++--------------------------
1 file changed, 36 insertions(+), 33 deletions(-)
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 0f4d15f..553ba1d 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -633,7 +633,6 @@ nomem:
static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet,
struct sctp_chunk *chunk)
{
- sctp_xmit_t retval = SCTP_XMIT_OK;
size_t datasize, rwnd, inflight, flight_size;
struct sctp_transport *transport = packet->transport;
struct sctp_association *asoc = transport->asoc;
@@ -658,15 +657,11 @@ static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet,
datasize = sctp_data_size(chunk);
- if (datasize > rwnd) {
- if (inflight > 0) {
- /* We have (at least) one data chunk in flight,
- * so we can't fall back to rule 6.1 B).
- */
- retval = SCTP_XMIT_RWND_FULL;
- goto finish;
- }
- }
+ if (datasize > rwnd && inflight > 0)
+ /* We have (at least) one data chunk in flight,
+ * so we can't fall back to rule 6.1 B).
+ */
+ return SCTP_XMIT_RWND_FULL;
/* RFC 2960 6.1 Transmission of DATA Chunks
*
@@ -680,36 +675,44 @@ static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet,
* When a Fast Retransmit is being performed the sender SHOULD
* ignore the value of cwnd and SHOULD NOT delay retransmission.
*/
- if (chunk->fast_retransmit != SCTP_NEED_FRTX)
- if (flight_size >= transport->cwnd) {
- retval = SCTP_XMIT_RWND_FULL;
- goto finish;
- }
+ if (chunk->fast_retransmit != SCTP_NEED_FRTX &&
+ flight_size >= transport->cwnd)
+ return SCTP_XMIT_RWND_FULL;
/* Nagle's algorithm to solve small-packet problem:
* Inhibit the sending of new chunks when new outgoing data arrives
* if any previously transmitted data on the connection remains
* unacknowledged.
*/
- if (!sctp_sk(asoc->base.sk)->nodelay && sctp_packet_empty(packet) &&
- inflight && sctp_state(asoc, ESTABLISHED)) {
- unsigned int max = transport->pathmtu - packet->overhead;
- unsigned int len = chunk->skb->len + q->out_qlen;
-
- /* Check whether this chunk and all the rest of pending
- * data will fit or delay in hopes of bundling a full
- * sized packet.
- * Don't delay large message writes that may have been
- * fragmeneted into small peices.
- */
- if ((len < max) && chunk->msg->can_delay) {
- retval = SCTP_XMIT_NAGLE_DELAY;
- goto finish;
- }
- }
-finish:
- return retval;
+ if (sctp_sk(asoc->base.sk)->nodelay)
+ /* Nagle disabled */
+ return SCTP_XMIT_OK;
+
+ if (!sctp_packet_empty(packet))
+ /* Append to packet */
+ return SCTP_XMIT_OK;
+
+ if (inflight == 0)
+ /* Nothing unacked */
+ return SCTP_XMIT_OK;
+
+ if (!sctp_state(asoc, ESTABLISHED))
+ return SCTP_XMIT_OK;
+
+ /* Check whether this chunk and all the rest of pending data will fit
+ * or delay in hopes of bundling a full sized packet.
+ */
+ if (chunk->skb->len + q->out_qlen >= transport->pathmtu - packet->overhead)
+ /* Enough data queued to fill a packet */
+ return SCTP_XMIT_OK;
+
+ /* Don't delay large message writes that may have been fragmented */
+ if (!chunk->msg->can_delay)
+ return SCTP_XMIT_OK;
+
+ /* Defer until all data acked or packet full */
+ return SCTP_XMIT_NAGLE_DELAY;
}
/* This private function does management things when adding DATA chunk */
--
1.8.1.2
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH net-next v3 1/3] Open out the check for Nagle
2014-07-18 17:36 [PATCH net-next v3 1/3] Open out the check for Nagle David Laight
@ 2014-07-21 23:25 ` David Miller
2014-07-22 9:03 ` David Laight
0 siblings, 1 reply; 3+ messages in thread
From: David Miller @ 2014-07-21 23:25 UTC (permalink / raw)
To: David.Laight; +Cc: netdev, linux-sctp
From: David Laight <David.Laight@ACULAB.COM>
Date: Fri, 18 Jul 2014 17:36:24 +0000
> The check for Nagle contains 6 separate checks all of which must be true
> before a data packet is delayed.
> Separate out each into its own 'if (test) return SCTP_XMIT_OK' so that
> the reasons can be individually described.
>
> Also return directly with SCTP_XMIT_RWND_FULL.
> Delete the now-unused 'retval' variable and 'finish' label from
> sctp_packet_can_append_data().
>
> Signed-off-by: David Laight <david.laight@aculab.com>
Please add the "sctp: " subsystem prefix to this subject line.
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: [PATCH net-next v3 1/3] Open out the check for Nagle
2014-07-21 23:25 ` David Miller
@ 2014-07-22 9:03 ` David Laight
0 siblings, 0 replies; 3+ messages in thread
From: David Laight @ 2014-07-22 9:03 UTC (permalink / raw)
To: 'David Miller'; +Cc: netdev, linux-sctp
> From: David Laight <David.Laight@ACULAB.COM>
> Date: Fri, 18 Jul 2014 17:36:24 +0000
>
> > The check for Nagle contains 6 separate checks all of which must be true
> > before a data packet is delayed.
> > Separate out each into its own 'if (test) return SCTP_XMIT_OK' so that
> > the reasons can be individually described.
> >
> > Also return directly with SCTP_XMIT_RWND_FULL.
> > Delete the now-unused 'retval' variable and 'finish' label from
> > sctp_packet_can_append_data().
> >
> > Signed-off-by: David Laight <david.laight@aculab.com>
>
> Please add the "sctp: " subsystem prefix to this subject line.
I've resent all 3 patches with net: sctp: added to the subject line
and a comment below the ---.
I must check whether I can send patches directly, instead of pasting
from wordpad into outlook (which seems to preserve any tabs).
David
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2014-07-22 9:04 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-07-18 17:36 [PATCH net-next v3 1/3] Open out the check for Nagle David Laight
2014-07-21 23:25 ` David Miller
2014-07-22 9:03 ` David Laight
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).