lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date:   Tue,  1 Sep 2020 11:00:07 +0200
From:   Petr Malat <oss@...at.biz>
To:     linux-sctp@...r.kernel.org
Cc:     vyasevich@...il.com, nhorman@...driver.com,
        marcelo.leitner@...il.com, davem@...emloft.net, kuba@...nel.org,
        Petr Malat <oss@...at.biz>
Subject: [PATCH] sctp: Honour SCTP_PARTIAL_DELIVERY_POINT even under memory pressure

Command SCTP_CMD_PART_DELIVER issued under memory pressure calls
sctp_ulpq_partial_delivery(), which tries to fetch and partially deliver
the first message it finds without checking if the message is longer than
SCTP_PARTIAL_DELIVERY_POINT. According to the RFC 6458 paragraph 8.1.21.
such a behavior is invalid. Fix it by returning the first message only if
its part currently available is longer than SCTP_PARTIAL_DELIVERY_POINT.

Signed-off-by: Petr Malat <oss@...at.biz>
---
 net/sctp/ulpqueue.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c
index 1c6c640607c5..cada0b7f1548 100644
--- a/net/sctp/ulpqueue.c
+++ b/net/sctp/ulpqueue.c
@@ -610,6 +610,7 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq)
 	struct sctp_ulpevent *cevent;
 	__u32 ctsn, next_tsn;
 	struct sctp_ulpevent *retval;
+	size_t pd_point, pd_len = 0;
 
 	/* The chunks are held in the reasm queue sorted by TSN.
 	 * Walk through the queue sequentially and look for a sequence of
@@ -633,8 +634,9 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq)
 				first_frag = pos;
 				next_tsn = ctsn + 1;
 				last_frag = pos;
+				pd_len = pos->len;
 			} else
-				goto done;
+				goto check;
 			break;
 
 		case SCTP_DATA_MIDDLE_FRAG:
@@ -643,15 +645,19 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq)
 			if (ctsn == next_tsn) {
 				next_tsn++;
 				last_frag = pos;
+				pd_len += pos->len;
 			} else
-				goto done;
+				goto check;
 			break;
 
 		case SCTP_DATA_LAST_FRAG:
 			if (!first_frag)
 				return NULL;
-			else
+			if (ctsn == next_tsn) {
+				last_frag = pos;
 				goto done;
+			} else
+				goto check;
 			break;
 
 		default:
@@ -659,6 +665,11 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq)
 		}
 	}
 
+check:
+	pd_point = sctp_sk(ulpq->asoc->base.sk)->pd_point;
+	if (pd_point && pd_point > pd_len)
+		return NULL;
+
 	/* We have the reassembled event. There is no need to look
 	 * further.
 	 */
-- 
2.20.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ