diff mbox

[net] sctp: identify chunks that need to be fragmented at IP level

Message ID d3a35eae75b431862cf96b55dd3a3614a11df8a3.1473328451.git.lucien.xin@gmail.com
State Accepted, archived
Delegated to: David Miller
Headers show

Commit Message

Xin Long Sept. 8, 2016, 9:54 a.m. UTC 
From: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>

Previously, without GSO, it was easy to identify it: if the chunk didn't
fit and there was no data chunk in the packet yet, we could fragment at
IP level. So if there was an auth chunk and we were bundling a big data
chunk, it would fragment regardless of the size of the auth chunk. This
also works for the context of PMTU reductions.

But with GSO, we cannot distinguish such PMTU events anymore, as the
packet is allowed to exceed PMTU.

So we need another check: to ensure that the chunk that we are adding,
actually fits the current PMTU. If it doesn't, trigger a flush and let
it be fragmented at IP level in the next round.

Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
---
 net/sctp/output.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

Comments

David Miller Sept. 10, 2016, 2:18 a.m. UTC | #1 
From: Xin Long <lucien.xin@gmail.com>
Date: Thu,  8 Sep 2016 17:54:11 +0800

> From: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
> 
> Previously, without GSO, it was easy to identify it: if the chunk didn't
> fit and there was no data chunk in the packet yet, we could fragment at
> IP level. So if there was an auth chunk and we were bundling a big data
> chunk, it would fragment regardless of the size of the auth chunk. This
> also works for the context of PMTU reductions.
> 
> But with GSO, we cannot distinguish such PMTU events anymore, as the
> packet is allowed to exceed PMTU.
> 
> So we need another check: to ensure that the chunk that we are adding,
> actually fits the current PMTU. If it doesn't, trigger a flush and let
> it be fragmented at IP level in the next round.
> 
> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>

Applied.
diff mbox

Patch

diff --git a/net/sctp/output.c b/net/sctp/output.c
index 1f1682b..31b7bc3 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -878,7 +878,7 @@  static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet,
 					struct sctp_chunk *chunk,
 					u16 chunk_len)
 {
-	size_t psize, pmtu;
+	size_t psize, pmtu, maxsize;
 	sctp_xmit_t retval = SCTP_XMIT_OK;
 
 	psize = packet->size;
@@ -906,6 +906,17 @@  static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet,
 			goto out;
 		}
 
+		/* Similarly, if this chunk was built before a PMTU
+		 * reduction, we have to fragment it at IP level now. So
+		 * if the packet already contains something, we need to
+		 * flush.
+		 */
+		maxsize = pmtu - packet->overhead;
+		if (packet->auth)
+			maxsize -= WORD_ROUND(packet->auth->skb->len);
+		if (chunk_len > maxsize)
+			retval = SCTP_XMIT_PMTU_FULL;
+
 		/* It is also okay to fragment if the chunk we are
 		 * adding is a control chunk, but only if current packet
 		 * is not a GSO one otherwise it causes fragmentation of