All of lore.kernel.org
 help / color / mirror / Atom feed
* [MPTCP] [PATCH net] mptcp: fix DSS map generation on fin retransmission
@ 2020-07-03 16:06 ` Paolo Abeni
  0 siblings, 0 replies; 5+ messages in thread
From: Paolo Abeni @ 2020-07-03 16:06 UTC (permalink / raw)
  To: mptcp

[-- Attachment #1: Type: text/plain, Size: 1919 bytes --]

The RFC 8684 mandates that no-data DATA FIN packets should carry
a DSS with 0 sequence number and data len equal to 1. Currently,
on FIN retransmission we re-use the existing mapping; if the previous
fin transmission was part of a partially acked data packet, we could
end-up writing in the egress packet a non-compliant DSS.

The above will be detected by a "Bad mapping" warning on the receiver
side.

This change addresses the issue explicitly checking for 0 len packet
when adding the DATA_FIN option.

Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets")
Reported-by: syzbot+42a07faa5923cfaeb9c9(a)syzkaller.appspotmail.com
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Reviewed-by: Christoph Paasch <cpaasch(a)apple.com>
Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/options.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index df9a51425c6f..8f940be42f98 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -449,9 +449,9 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
 }
 
 static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
-				 struct mptcp_ext *ext)
+				 struct sk_buff *skb, struct mptcp_ext *ext)
 {
-	if (!ext->use_map) {
+	if (!ext->use_map || !skb->len) {
 		/* RFC6824 requires a DSS mapping with specific values
 		 * if DATA_FIN is set but no data payload is mapped
 		 */
@@ -503,7 +503,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 			opts->ext_copy = *mpext;
 
 		if (skb && tcp_fin && subflow->data_fin_tx_enable)
-			mptcp_write_data_fin(subflow, &opts->ext_copy);
+			mptcp_write_data_fin(subflow, skb, &opts->ext_copy);
 		ret = true;
 	}
 
-- 
2.26.2

^ permalink raw reply related	[flat|nested] 5+ messages in thread
* [PATCH net] mptcp: fix DSS map generation on fin retransmission
@ 2020-07-03 16:06 ` Paolo Abeni
  0 siblings, 0 replies; 5+ messages in thread
From: Paolo Abeni @ 2020-07-03 16:06 UTC (permalink / raw)
  To: netdev; +Cc: David S. Miller, mptcp

The RFC 8684 mandates that no-data DATA FIN packets should carry
a DSS with 0 sequence number and data len equal to 1. Currently,
on FIN retransmission we re-use the existing mapping; if the previous
fin transmission was part of a partially acked data packet, we could
end-up writing in the egress packet a non-compliant DSS.

The above will be detected by a "Bad mapping" warning on the receiver
side.

This change addresses the issue explicitly checking for 0 len packet
when adding the DATA_FIN option.

Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets")
Reported-by: syzbot+42a07faa5923cfaeb9c9@syzkaller.appspotmail.com
Tested-by: Christoph Paasch <cpaasch@apple.com>
Reviewed-by: Christoph Paasch <cpaasch@apple.com>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/options.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index df9a51425c6f..8f940be42f98 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -449,9 +449,9 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
 }
 
 static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
-				 struct mptcp_ext *ext)
+				 struct sk_buff *skb, struct mptcp_ext *ext)
 {
-	if (!ext->use_map) {
+	if (!ext->use_map || !skb->len) {
 		/* RFC6824 requires a DSS mapping with specific values
 		 * if DATA_FIN is set but no data payload is mapped
 		 */
@@ -503,7 +503,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 			opts->ext_copy = *mpext;
 
 		if (skb && tcp_fin && subflow->data_fin_tx_enable)
-			mptcp_write_data_fin(subflow, &opts->ext_copy);
+			mptcp_write_data_fin(subflow, skb, &opts->ext_copy);
 		ret = true;
 	}
 
-- 
2.26.2


^ permalink raw reply related	[flat|nested] 5+ messages in thread
* [MPTCP] Re: [PATCH net] mptcp: fix DSS map generation on fin retransmission
  2020-07-03 16:06 ` Paolo Abeni
@ 2020-07-07 22:28 ` David Miller
  -1 siblings, 0 replies; 5+ messages in thread
From: David Miller @ 2020-07-07 22:28 UTC (permalink / raw)
  To: mptcp

[-- Attachment #1: Type: text/plain, Size: 1056 bytes --]

From: Paolo Abeni <pabeni(a)redhat.com>
Date: Fri,  3 Jul 2020 18:06:04 +0200

> The RFC 8684 mandates that no-data DATA FIN packets should carry
> a DSS with 0 sequence number and data len equal to 1. Currently,
> on FIN retransmission we re-use the existing mapping; if the previous
> fin transmission was part of a partially acked data packet, we could
> end-up writing in the egress packet a non-compliant DSS.
> 
> The above will be detected by a "Bad mapping" warning on the receiver
> side.
> 
> This change addresses the issue explicitly checking for 0 len packet
> when adding the DATA_FIN option.
> 
> Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets")
> Reported-by: syzbot+42a07faa5923cfaeb9c9(a)syzkaller.appspotmail.com
> Tested-by: Christoph Paasch <cpaasch(a)apple.com>
> Reviewed-by: Christoph Paasch <cpaasch(a)apple.com>
> Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
> Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>

Applied and queued up for -stable, thank you.

^ permalink raw reply	[flat|nested] 5+ messages in thread
* Re: [PATCH net] mptcp: fix DSS map generation on fin retransmission
@ 2020-07-07 22:28 ` David Miller
  0 siblings, 0 replies; 5+ messages in thread
From: David Miller @ 2020-07-07 22:28 UTC (permalink / raw)
  To: pabeni; +Cc: netdev, mptcp

From: Paolo Abeni <pabeni@redhat.com>
Date: Fri,  3 Jul 2020 18:06:04 +0200

> The RFC 8684 mandates that no-data DATA FIN packets should carry
> a DSS with 0 sequence number and data len equal to 1. Currently,
> on FIN retransmission we re-use the existing mapping; if the previous
> fin transmission was part of a partially acked data packet, we could
> end-up writing in the egress packet a non-compliant DSS.
> 
> The above will be detected by a "Bad mapping" warning on the receiver
> side.
> 
> This change addresses the issue explicitly checking for 0 len packet
> when adding the DATA_FIN option.
> 
> Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets")
> Reported-by: syzbot+42a07faa5923cfaeb9c9@syzkaller.appspotmail.com
> Tested-by: Christoph Paasch <cpaasch@apple.com>
> Reviewed-by: Christoph Paasch <cpaasch@apple.com>
> Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
> Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Applied and queued up for -stable, thank you.

^ permalink raw reply	[flat|nested] 5+ messages in thread
* [MPTCP] [PATCH net] mptcp: fix DSS map generation on fin retransmission
@ 2020-06-26  9:10 Paolo Abeni
  0 siblings, 0 replies; 5+ messages in thread
From: Paolo Abeni @ 2020-06-26  9:10 UTC (permalink / raw)
  To: mptcp

[-- Attachment #1: Type: text/plain, Size: 1810 bytes --]

The RFC 8684 mandates that no-data DATA FIN packets should carry
a DSS with 0 sequence number and data len equal to 1. Currently,
on FIN retransmission we re-use the existing mapping; if the previous
fin transmission was part of a partially acked data packet, we could
end-up writing in the egress packet a non-compliant DSS.

The above will be detected by a "Bad mapping" warning on the receiver
side.

This change addresses the issue explicitly checking for 0 len packet
when adding the DATA_FIN option.

Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets")
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/options.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
---
this should go on "top" of export branch, just before the "DO NOT MERGE"
changes.
Hopefully should not conflict with others

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index f464f8669dfc..46470194b8ca 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -451,9 +451,9 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
 }
 
 static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
-				 struct mptcp_ext *ext)
+				 struct sk_buff *skb, struct mptcp_ext *ext)
 {
-	if (!ext->use_map) {
+	if (!ext->use_map || !skb->len) {
 		/* RFC6824 requires a DSS mapping with specific values
 		 * if DATA_FIN is set but no data payload is mapped
 		 */
@@ -505,7 +505,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 			opts->ext_copy = *mpext;
 
 		if (skb && tcp_fin && subflow->data_fin_tx_enable)
-			mptcp_write_data_fin(subflow, &opts->ext_copy);
+			mptcp_write_data_fin(subflow, skb, &opts->ext_copy);
 		ret = true;
 	}
 
-- 
2.26.2

^ permalink raw reply related	[flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-07-07 22:28 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-07-03 16:06 [MPTCP] [PATCH net] mptcp: fix DSS map generation on fin retransmission Paolo Abeni
2020-07-03 16:06 ` Paolo Abeni
  -- strict thread matches above, loose matches on Subject: below --
2020-07-07 22:28 [MPTCP] " David Miller
2020-07-07 22:28 ` David Miller
2020-06-26  9:10 [MPTCP] " Paolo Abeni

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.