2008-01-11 17:57:09 +03:00
|
|
|
/* SCTP kernel implementation
|
2005-04-17 02:20:36 +04:00
|
|
|
* (C) Copyright IBM Corp. 2001, 2004
|
|
|
|
* Copyright (c) 1999-2000 Cisco, Inc.
|
|
|
|
* Copyright (c) 1999-2001 Motorola, Inc.
|
|
|
|
*
|
2008-01-11 17:57:09 +03:00
|
|
|
* This file is part of the SCTP kernel implementation
|
2005-04-17 02:20:36 +04:00
|
|
|
*
|
|
|
|
* These functions handle output processing.
|
|
|
|
*
|
2008-01-11 17:57:09 +03:00
|
|
|
* This SCTP implementation is free software;
|
2005-04-17 02:20:36 +04:00
|
|
|
* you can redistribute it and/or modify it under the terms of
|
|
|
|
* the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation; either version 2, or (at your option)
|
|
|
|
* any later version.
|
|
|
|
*
|
2008-01-11 17:57:09 +03:00
|
|
|
* This SCTP implementation is distributed in the hope that it
|
2005-04-17 02:20:36 +04:00
|
|
|
* will be useful, but WITHOUT ANY WARRANTY; without even the implied
|
|
|
|
* ************************
|
|
|
|
* warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
|
|
|
|
* See the GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
2013-12-06 18:28:48 +04:00
|
|
|
* along with GNU CC; see the file COPYING. If not, see
|
|
|
|
* <http://www.gnu.org/licenses/>.
|
2005-04-17 02:20:36 +04:00
|
|
|
*
|
|
|
|
* Please send any bug reports or fixes you make to the
|
|
|
|
* email address(es):
|
2013-07-23 16:51:47 +04:00
|
|
|
* lksctp developers <linux-sctp@vger.kernel.org>
|
2005-04-17 02:20:36 +04:00
|
|
|
*
|
|
|
|
* Written or modified by:
|
|
|
|
* La Monte H.P. Yarroll <piggy@acm.org>
|
|
|
|
* Karl Knutson <karl@athena.chicago.il.us>
|
|
|
|
* Jon Grimm <jgrimm@austin.ibm.com>
|
|
|
|
* Sridhar Samudrala <sri@us.ibm.com>
|
|
|
|
*/
|
|
|
|
|
2010-08-24 17:21:08 +04:00
|
|
|
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/wait.h>
|
|
|
|
#include <linux/time.h>
|
|
|
|
#include <linux/ip.h>
|
|
|
|
#include <linux/ipv6.h>
|
|
|
|
#include <linux/init.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 11:04:11 +03:00
|
|
|
#include <linux/slab.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <net/inet_ecn.h>
|
2009-03-21 23:41:09 +03:00
|
|
|
#include <net/ip.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <net/icmp.h>
|
2008-07-17 07:20:11 +04:00
|
|
|
#include <net/net_namespace.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
#include <linux/socket.h> /* for sa_family_t */
|
|
|
|
#include <net/sock.h>
|
|
|
|
|
|
|
|
#include <net/sctp/sctp.h>
|
|
|
|
#include <net/sctp/sm.h>
|
2007-12-17 01:06:41 +03:00
|
|
|
#include <net/sctp/checksum.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Forward declarations for private helpers. */
|
2012-07-02 13:59:24 +04:00
|
|
|
static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk);
|
2009-08-07 18:43:07 +04:00
|
|
|
static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet,
|
2005-04-17 02:20:36 +04:00
|
|
|
struct sctp_chunk *chunk);
|
2009-08-07 18:43:07 +04:00
|
|
|
static void sctp_packet_append_data(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk);
|
|
|
|
static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk,
|
|
|
|
u16 chunk_len);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2009-09-04 10:34:06 +04:00
|
|
|
static void sctp_packet_reset(struct sctp_packet *packet)
|
|
|
|
{
|
|
|
|
packet->size = packet->overhead;
|
|
|
|
packet->has_cookie_echo = 0;
|
|
|
|
packet->has_sack = 0;
|
|
|
|
packet->has_data = 0;
|
|
|
|
packet->has_auth = 0;
|
|
|
|
packet->ipfragok = 0;
|
|
|
|
packet->auth = NULL;
|
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Config a packet.
|
|
|
|
* This appears to be a followup set of initializations.
|
|
|
|
*/
|
|
|
|
struct sctp_packet *sctp_packet_config(struct sctp_packet *packet,
|
|
|
|
__u32 vtag, int ecn_capable)
|
|
|
|
{
|
2016-06-02 21:05:43 +03:00
|
|
|
struct sctp_transport *tp = packet->transport;
|
|
|
|
struct sctp_association *asoc = tp->asoc;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
packet->vtag = vtag;
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
if (asoc && tp->dst) {
|
|
|
|
struct sock *sk = asoc->base.sk;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
if (__sk_dst_get(sk) != tp->dst) {
|
|
|
|
dst_hold(tp->dst);
|
|
|
|
sk_setup_caps(sk, tp->dst);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (sk_can_gso(sk)) {
|
|
|
|
struct net_device *dev = tp->dst->dev;
|
|
|
|
|
|
|
|
packet->max_size = dev->gso_max_size;
|
|
|
|
} else {
|
|
|
|
packet->max_size = asoc->pathmtu;
|
|
|
|
}
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
} else {
|
|
|
|
packet->max_size = tp->pathmtu;
|
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
if (ecn_capable && sctp_packet_empty(packet)) {
|
2016-06-02 21:05:43 +03:00
|
|
|
struct sctp_chunk *chunk;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* If there a is a prepend chunk stick it on the list before
|
2007-02-09 17:25:18 +03:00
|
|
|
* any other chunks get appended.
|
|
|
|
*/
|
2016-06-02 21:05:43 +03:00
|
|
|
chunk = sctp_get_ecne_prepend(asoc);
|
2005-04-17 02:20:36 +04:00
|
|
|
if (chunk)
|
|
|
|
sctp_packet_append_chunk(packet, chunk);
|
|
|
|
}
|
|
|
|
|
|
|
|
return packet;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Initialize the packet structure. */
|
|
|
|
struct sctp_packet *sctp_packet_init(struct sctp_packet *packet,
|
|
|
|
struct sctp_transport *transport,
|
|
|
|
__u16 sport, __u16 dport)
|
|
|
|
{
|
|
|
|
struct sctp_association *asoc = transport->asoc;
|
|
|
|
size_t overhead;
|
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("%s: packet:%p transport:%p\n", __func__, packet, transport);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
packet->transport = transport;
|
|
|
|
packet->source_port = sport;
|
|
|
|
packet->destination_port = dport;
|
2005-07-09 08:47:49 +04:00
|
|
|
INIT_LIST_HEAD(&packet->chunk_list);
|
2005-04-17 02:20:36 +04:00
|
|
|
if (asoc) {
|
2007-02-09 17:25:18 +03:00
|
|
|
struct sctp_sock *sp = sctp_sk(asoc->base.sk);
|
|
|
|
overhead = sp->pf->af->net_header_len;
|
2005-04-17 02:20:36 +04:00
|
|
|
} else {
|
|
|
|
overhead = sizeof(struct ipv6hdr);
|
|
|
|
}
|
|
|
|
overhead += sizeof(struct sctphdr);
|
|
|
|
packet->overhead = overhead;
|
2009-09-04 10:34:06 +04:00
|
|
|
sctp_packet_reset(packet);
|
2005-04-17 02:20:36 +04:00
|
|
|
packet->vtag = 0;
|
2013-04-19 01:59:37 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
return packet;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Free a packet. */
|
|
|
|
void sctp_packet_free(struct sctp_packet *packet)
|
|
|
|
{
|
2005-07-09 08:47:49 +04:00
|
|
|
struct sctp_chunk *chunk, *tmp;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("%s: packet:%p\n", __func__, packet);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2005-07-09 08:47:49 +04:00
|
|
|
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
|
|
|
|
list_del_init(&chunk->list);
|
2005-04-17 02:20:36 +04:00
|
|
|
sctp_chunk_free(chunk);
|
2005-07-09 08:47:49 +04:00
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
/* This routine tries to append the chunk to the offered packet. If adding
|
|
|
|
* the chunk causes the packet to exceed the path MTU and COOKIE_ECHO chunk
|
|
|
|
* is not present in the packet, it transmits the input packet.
|
|
|
|
* Data can be bundled with a packet containing a COOKIE_ECHO chunk as long
|
|
|
|
* as it can fit in the packet, but any more data that does not fit in this
|
|
|
|
* packet can be sent only after receiving the COOKIE_ACK.
|
|
|
|
*/
|
|
|
|
sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet,
|
2008-06-20 03:08:18 +04:00
|
|
|
struct sctp_chunk *chunk,
|
2016-03-11 00:33:07 +03:00
|
|
|
int one_packet, gfp_t gfp)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
sctp_xmit_t retval;
|
|
|
|
|
2016-06-04 08:53:26 +03:00
|
|
|
pr_debug("%s: packet:%p size:%Zu chunk:%p size:%d\n", __func__,
|
2016-06-02 21:05:44 +03:00
|
|
|
packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) {
|
|
|
|
case SCTP_XMIT_PMTU_FULL:
|
|
|
|
if (!packet->has_cookie_echo) {
|
sctp: save transmit error to sk_err in sctp_outq_flush
Every time when sctp calls sctp_outq_flush, it sends out the chunks of
control queue, retransmit queue and data queue. Even if some trunks are
failed to transmit, it still has to flush all the transports, as it's
the only chance to clean that transmit_list.
So the latest transmit error here should be returned back. This transmit
error is an internal error of sctp stack.
I checked all the places where it uses the transmit error (the return
value of sctp_outq_flush), most of them are actually just save it to
sk_err.
Except for sctp_assoc/endpoint_bh_rcv, they will drop the chunk if
it's failed to send a REPLY, which is actually incorrect, as we can't
be sure the error that sctp_outq_flush returns is from sending that
REPLY.
So it's meaningless for sctp_outq_flush to return error back.
This patch is to save transmit error to sk_err in sctp_outq_flush, the
new error can update the old value. Eventually, sctp_wait_for_* would
check for it.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-13 21:04:21 +03:00
|
|
|
int error = 0;
|
|
|
|
|
2016-03-11 00:33:07 +03:00
|
|
|
error = sctp_packet_transmit(packet, gfp);
|
2005-04-17 02:20:36 +04:00
|
|
|
if (error < 0)
|
|
|
|
chunk->skb->sk->sk_err = -error;
|
|
|
|
|
|
|
|
/* If we have an empty packet, then we can NOT ever
|
|
|
|
* return PMTU_FULL.
|
|
|
|
*/
|
2008-06-20 03:08:18 +04:00
|
|
|
if (!one_packet)
|
|
|
|
retval = sctp_packet_append_chunk(packet,
|
|
|
|
chunk);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case SCTP_XMIT_RWND_FULL:
|
|
|
|
case SCTP_XMIT_OK:
|
2014-07-22 12:59:14 +04:00
|
|
|
case SCTP_XMIT_DELAY:
|
2005-04-17 02:20:36 +04:00
|
|
|
break;
|
2007-04-21 04:09:22 +04:00
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2007-09-17 06:32:45 +04:00
|
|
|
/* Try to bundle an auth chunk into the packet. */
|
|
|
|
static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt,
|
|
|
|
struct sctp_chunk *chunk)
|
|
|
|
{
|
|
|
|
struct sctp_association *asoc = pkt->transport->asoc;
|
|
|
|
struct sctp_chunk *auth;
|
|
|
|
sctp_xmit_t retval = SCTP_XMIT_OK;
|
|
|
|
|
|
|
|
/* if we don't have an association, we can't do authentication */
|
|
|
|
if (!asoc)
|
|
|
|
return retval;
|
|
|
|
|
|
|
|
/* See if this is an auth chunk we are bundling or if
|
|
|
|
* auth is already bundled.
|
|
|
|
*/
|
2009-09-05 02:21:00 +04:00
|
|
|
if (chunk->chunk_hdr->type == SCTP_CID_AUTH || pkt->has_auth)
|
2007-09-17 06:32:45 +04:00
|
|
|
return retval;
|
|
|
|
|
|
|
|
/* if the peer did not request this chunk to be authenticated,
|
|
|
|
* don't do it
|
|
|
|
*/
|
|
|
|
if (!chunk->auth)
|
|
|
|
return retval;
|
|
|
|
|
|
|
|
auth = sctp_make_auth(asoc);
|
|
|
|
if (!auth)
|
|
|
|
return retval;
|
|
|
|
|
2012-07-02 13:59:24 +04:00
|
|
|
retval = __sctp_packet_append_chunk(pkt, auth);
|
|
|
|
|
|
|
|
if (retval != SCTP_XMIT_OK)
|
|
|
|
sctp_chunk_free(auth);
|
2007-09-17 06:32:45 +04:00
|
|
|
|
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Try to bundle a SACK with the packet. */
|
|
|
|
static sctp_xmit_t sctp_packet_bundle_sack(struct sctp_packet *pkt,
|
|
|
|
struct sctp_chunk *chunk)
|
|
|
|
{
|
|
|
|
sctp_xmit_t retval = SCTP_XMIT_OK;
|
|
|
|
|
|
|
|
/* If sending DATA and haven't aleady bundled a SACK, try to
|
|
|
|
* bundle one in to the packet.
|
|
|
|
*/
|
|
|
|
if (sctp_chunk_is_data(chunk) && !pkt->has_sack &&
|
|
|
|
!pkt->has_cookie_echo) {
|
|
|
|
struct sctp_association *asoc;
|
sctp: Fix piggybacked ACKs
This patch corrects the conditions under which a SACK will be piggybacked
on a DATA packet. The previous condition was incorrect due to a
misinterpretation of RFC 4960 and/or RFC 2960. Specifically, the
following paragraph from section 6.2 had not been implemented correctly:
Before an endpoint transmits a DATA chunk, if any received DATA
chunks have not been acknowledged (e.g., due to delayed ack), the
sender should create a SACK and bundle it with the outbound DATA
chunk, as long as the size of the final SCTP packet does not exceed
the current MTU. See Section 6.2.
When about to send a DATA chunk, the code now checks to see if the SACK
timer is running. If it is, we know we have a SACK to send to the
peer, so we append the SACK (assuming available space in the packet)
and turn off the timer. For a simple request-response scenario, this
will result in the SACK being bundled with the response, meaning the
the SACK is received quickly by the client, and also meaning that no
separate SACK packet needs to be sent by the server to acknowledge the
request. Prior to this patch, a separate SACK packet would have been
sent by the server SCTP only after its delayed-ACK timer had expired
(usually 200ms). This is wasteful of bandwidth, and can also have a
major negative impact on performance due the interaction of delayed ACKs
with the Nagle algorithm.
Signed-off-by: Doug Graham <dgraham@nortel.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-07-29 20:05:57 +04:00
|
|
|
struct timer_list *timer;
|
2005-04-17 02:20:36 +04:00
|
|
|
asoc = pkt->transport->asoc;
|
sctp: Fix piggybacked ACKs
This patch corrects the conditions under which a SACK will be piggybacked
on a DATA packet. The previous condition was incorrect due to a
misinterpretation of RFC 4960 and/or RFC 2960. Specifically, the
following paragraph from section 6.2 had not been implemented correctly:
Before an endpoint transmits a DATA chunk, if any received DATA
chunks have not been acknowledged (e.g., due to delayed ack), the
sender should create a SACK and bundle it with the outbound DATA
chunk, as long as the size of the final SCTP packet does not exceed
the current MTU. See Section 6.2.
When about to send a DATA chunk, the code now checks to see if the SACK
timer is running. If it is, we know we have a SACK to send to the
peer, so we append the SACK (assuming available space in the packet)
and turn off the timer. For a simple request-response scenario, this
will result in the SACK being bundled with the response, meaning the
the SACK is received quickly by the client, and also meaning that no
separate SACK packet needs to be sent by the server to acknowledge the
request. Prior to this patch, a separate SACK packet would have been
sent by the server SCTP only after its delayed-ACK timer had expired
(usually 200ms). This is wasteful of bandwidth, and can also have a
major negative impact on performance due the interaction of delayed ACKs
with the Nagle algorithm.
Signed-off-by: Doug Graham <dgraham@nortel.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-07-29 20:05:57 +04:00
|
|
|
timer = &asoc->timers[SCTP_EVENT_TIMEOUT_SACK];
|
2005-04-17 02:20:36 +04:00
|
|
|
|
sctp: Fix piggybacked ACKs
This patch corrects the conditions under which a SACK will be piggybacked
on a DATA packet. The previous condition was incorrect due to a
misinterpretation of RFC 4960 and/or RFC 2960. Specifically, the
following paragraph from section 6.2 had not been implemented correctly:
Before an endpoint transmits a DATA chunk, if any received DATA
chunks have not been acknowledged (e.g., due to delayed ack), the
sender should create a SACK and bundle it with the outbound DATA
chunk, as long as the size of the final SCTP packet does not exceed
the current MTU. See Section 6.2.
When about to send a DATA chunk, the code now checks to see if the SACK
timer is running. If it is, we know we have a SACK to send to the
peer, so we append the SACK (assuming available space in the packet)
and turn off the timer. For a simple request-response scenario, this
will result in the SACK being bundled with the response, meaning the
the SACK is received quickly by the client, and also meaning that no
separate SACK packet needs to be sent by the server to acknowledge the
request. Prior to this patch, a separate SACK packet would have been
sent by the server SCTP only after its delayed-ACK timer had expired
(usually 200ms). This is wasteful of bandwidth, and can also have a
major negative impact on performance due the interaction of delayed ACKs
with the Nagle algorithm.
Signed-off-by: Doug Graham <dgraham@nortel.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-07-29 20:05:57 +04:00
|
|
|
/* If the SACK timer is running, we have a pending SACK */
|
|
|
|
if (timer_pending(timer)) {
|
2005-04-17 02:20:36 +04:00
|
|
|
struct sctp_chunk *sack;
|
2012-06-30 07:04:26 +04:00
|
|
|
|
|
|
|
if (pkt->transport->sack_generation !=
|
|
|
|
pkt->transport->asoc->peer.sack_generation)
|
|
|
|
return retval;
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
asoc->a_rwnd = asoc->rwnd;
|
|
|
|
sack = sctp_make_sack(asoc);
|
|
|
|
if (sack) {
|
2012-07-02 13:59:24 +04:00
|
|
|
retval = __sctp_packet_append_chunk(pkt, sack);
|
|
|
|
if (retval != SCTP_XMIT_OK) {
|
|
|
|
sctp_chunk_free(sack);
|
|
|
|
goto out;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
asoc->peer.sack_needed = 0;
|
sctp: Fix piggybacked ACKs
This patch corrects the conditions under which a SACK will be piggybacked
on a DATA packet. The previous condition was incorrect due to a
misinterpretation of RFC 4960 and/or RFC 2960. Specifically, the
following paragraph from section 6.2 had not been implemented correctly:
Before an endpoint transmits a DATA chunk, if any received DATA
chunks have not been acknowledged (e.g., due to delayed ack), the
sender should create a SACK and bundle it with the outbound DATA
chunk, as long as the size of the final SCTP packet does not exceed
the current MTU. See Section 6.2.
When about to send a DATA chunk, the code now checks to see if the SACK
timer is running. If it is, we know we have a SACK to send to the
peer, so we append the SACK (assuming available space in the packet)
and turn off the timer. For a simple request-response scenario, this
will result in the SACK being bundled with the response, meaning the
the SACK is received quickly by the client, and also meaning that no
separate SACK packet needs to be sent by the server to acknowledge the
request. Prior to this patch, a separate SACK packet would have been
sent by the server SCTP only after its delayed-ACK timer had expired
(usually 200ms). This is wasteful of bandwidth, and can also have a
major negative impact on performance due the interaction of delayed ACKs
with the Nagle algorithm.
Signed-off-by: Doug Graham <dgraham@nortel.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-07-29 20:05:57 +04:00
|
|
|
if (del_timer(timer))
|
2005-04-17 02:20:36 +04:00
|
|
|
sctp_association_put(asoc);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2012-07-02 13:59:24 +04:00
|
|
|
out:
|
2005-04-17 02:20:36 +04:00
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2012-07-02 13:59:24 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Append a chunk to the offered packet reporting back any inability to do
|
|
|
|
* so.
|
|
|
|
*/
|
2012-07-02 13:59:24 +04:00
|
|
|
static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
sctp_xmit_t retval = SCTP_XMIT_OK;
|
|
|
|
__u16 chunk_len = WORD_ROUND(ntohs(chunk->chunk_hdr->length));
|
|
|
|
|
2009-08-07 18:43:07 +04:00
|
|
|
/* Check to see if this chunk will fit into the packet */
|
|
|
|
retval = sctp_packet_will_fit(packet, chunk, chunk_len);
|
|
|
|
if (retval != SCTP_XMIT_OK)
|
|
|
|
goto finish;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2009-08-07 18:43:07 +04:00
|
|
|
/* We believe that this chunk is OK to add to the packet */
|
2007-09-17 06:32:45 +04:00
|
|
|
switch (chunk->chunk_hdr->type) {
|
2013-12-23 08:16:52 +04:00
|
|
|
case SCTP_CID_DATA:
|
2009-08-07 18:43:07 +04:00
|
|
|
/* Account for the data being in the packet */
|
|
|
|
sctp_packet_append_data(packet, chunk);
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Disallow SACK bundling after DATA. */
|
|
|
|
packet->has_sack = 1;
|
2007-09-17 06:32:45 +04:00
|
|
|
/* Disallow AUTH bundling after DATA */
|
|
|
|
packet->has_auth = 1;
|
|
|
|
/* Let it be knows that packet has DATA in it */
|
|
|
|
packet->has_data = 1;
|
2009-01-23 01:53:01 +03:00
|
|
|
/* timestamp the chunk for rtx purposes */
|
|
|
|
chunk->sent_at = jiffies;
|
2016-07-09 14:47:43 +03:00
|
|
|
/* Mainly used for prsctp RTX policy */
|
|
|
|
chunk->sent_count++;
|
2007-09-17 06:32:45 +04:00
|
|
|
break;
|
2013-12-23 08:16:52 +04:00
|
|
|
case SCTP_CID_COOKIE_ECHO:
|
2005-04-17 02:20:36 +04:00
|
|
|
packet->has_cookie_echo = 1;
|
2007-09-17 06:32:45 +04:00
|
|
|
break;
|
|
|
|
|
2013-12-23 08:16:52 +04:00
|
|
|
case SCTP_CID_SACK:
|
2005-04-17 02:20:36 +04:00
|
|
|
packet->has_sack = 1;
|
2012-12-01 08:49:42 +04:00
|
|
|
if (chunk->asoc)
|
|
|
|
chunk->asoc->stats.osacks++;
|
2007-09-17 06:32:45 +04:00
|
|
|
break;
|
|
|
|
|
2013-12-23 08:16:52 +04:00
|
|
|
case SCTP_CID_AUTH:
|
2007-09-17 06:32:45 +04:00
|
|
|
packet->has_auth = 1;
|
|
|
|
packet->auth = chunk;
|
|
|
|
break;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* It is OK to send this chunk. */
|
2005-07-09 08:47:49 +04:00
|
|
|
list_add_tail(&chunk->list, &packet->chunk_list);
|
2005-04-17 02:20:36 +04:00
|
|
|
packet->size += chunk_len;
|
|
|
|
chunk->transport = packet->transport;
|
|
|
|
finish:
|
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2012-07-02 13:59:24 +04:00
|
|
|
/* Append a chunk to the offered packet reporting back any inability to do
|
|
|
|
* so.
|
|
|
|
*/
|
|
|
|
sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk)
|
|
|
|
{
|
|
|
|
sctp_xmit_t retval = SCTP_XMIT_OK;
|
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk);
|
2012-07-02 13:59:24 +04:00
|
|
|
|
|
|
|
/* Data chunks are special. Before seeing what else we can
|
|
|
|
* bundle into this packet, check to see if we are allowed to
|
|
|
|
* send this DATA.
|
|
|
|
*/
|
|
|
|
if (sctp_chunk_is_data(chunk)) {
|
|
|
|
retval = sctp_packet_can_append_data(packet, chunk);
|
|
|
|
if (retval != SCTP_XMIT_OK)
|
|
|
|
goto finish;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Try to bundle AUTH chunk */
|
|
|
|
retval = sctp_packet_bundle_auth(packet, chunk);
|
|
|
|
if (retval != SCTP_XMIT_OK)
|
|
|
|
goto finish;
|
|
|
|
|
|
|
|
/* Try to bundle SACK chunk */
|
|
|
|
retval = sctp_packet_bundle_sack(packet, chunk);
|
|
|
|
if (retval != SCTP_XMIT_OK)
|
|
|
|
goto finish;
|
|
|
|
|
|
|
|
retval = __sctp_packet_append_chunk(packet, chunk);
|
|
|
|
|
|
|
|
finish:
|
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2012-09-03 08:27:42 +04:00
|
|
|
static void sctp_packet_release_owner(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
sk_free(skb->sk);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void sctp_packet_set_owner_w(struct sk_buff *skb, struct sock *sk)
|
|
|
|
{
|
|
|
|
skb_orphan(skb);
|
|
|
|
skb->sk = sk;
|
|
|
|
skb->destructor = sctp_packet_release_owner;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The data chunks have already been accounted for in sctp_sendmsg(),
|
|
|
|
* therefore only reserve a single byte to keep socket around until
|
|
|
|
* the packet has been transmitted.
|
|
|
|
*/
|
|
|
|
atomic_inc(&sk->sk_wmem_alloc);
|
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* All packets are sent to the network through this function from
|
|
|
|
* sctp_outq_tail().
|
|
|
|
*
|
|
|
|
* The return value is a normal kernel error return value.
|
|
|
|
*/
|
2016-03-11 00:33:07 +03:00
|
|
|
int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
struct sctp_transport *tp = packet->transport;
|
|
|
|
struct sctp_association *asoc = tp->asoc;
|
|
|
|
struct sctphdr *sh;
|
2016-06-02 21:05:43 +03:00
|
|
|
struct sk_buff *nskb = NULL, *head = NULL;
|
2005-07-09 08:47:49 +04:00
|
|
|
struct sctp_chunk *chunk, *tmp;
|
2005-04-17 02:20:36 +04:00
|
|
|
struct sock *sk;
|
|
|
|
int err = 0;
|
|
|
|
int padding; /* How much padding do we need? */
|
2016-06-02 21:05:43 +03:00
|
|
|
int pkt_size;
|
2005-04-17 02:20:36 +04:00
|
|
|
__u8 has_data = 0;
|
2016-06-02 21:05:43 +03:00
|
|
|
int gso = 0;
|
|
|
|
int pktcount = 0;
|
2013-12-26 09:55:56 +04:00
|
|
|
struct dst_entry *dst;
|
2007-09-17 06:32:45 +04:00
|
|
|
unsigned char *auth = NULL; /* pointer to auth in skb data */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("%s: packet:%p\n", __func__, packet);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Do NOT generate a chunkless packet. */
|
2005-07-09 08:47:49 +04:00
|
|
|
if (list_empty(&packet->chunk_list))
|
2005-04-17 02:20:36 +04:00
|
|
|
return err;
|
|
|
|
|
|
|
|
/* Set up convenience variables... */
|
2005-07-09 08:47:49 +04:00
|
|
|
chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list);
|
2005-04-17 02:20:36 +04:00
|
|
|
sk = chunk->skb->sk;
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/* Allocate the head skb, or main one if not in GSO */
|
|
|
|
if (packet->size > tp->pathmtu && !packet->ipfragok) {
|
|
|
|
if (sk_can_gso(sk)) {
|
|
|
|
gso = 1;
|
|
|
|
pkt_size = packet->overhead;
|
|
|
|
} else {
|
|
|
|
/* If this happens, we trash this packet and try
|
|
|
|
* to build a new one, hopefully correct this
|
|
|
|
* time. Application may notice this error.
|
|
|
|
*/
|
|
|
|
pr_err_once("Trying to GSO but underlying device doesn't support it.");
|
|
|
|
goto nomem;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
pkt_size = packet->size;
|
|
|
|
}
|
|
|
|
head = alloc_skb(pkt_size + MAX_HEADER, gfp);
|
|
|
|
if (!head)
|
2005-04-17 02:20:36 +04:00
|
|
|
goto nomem;
|
2016-06-02 21:05:43 +03:00
|
|
|
if (gso) {
|
|
|
|
NAPI_GRO_CB(head)->last = head;
|
|
|
|
skb_shinfo(head)->gso_type = sk->sk_gso_type;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Make sure the outbound skb has enough header room reserved. */
|
2016-06-02 21:05:43 +03:00
|
|
|
skb_reserve(head, packet->overhead + MAX_HEADER);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Set the owning socket so that we know where to get the
|
|
|
|
* destination IP address.
|
|
|
|
*/
|
2016-06-02 21:05:43 +03:00
|
|
|
sctp_packet_set_owner_w(head, sk);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2012-05-04 09:24:54 +04:00
|
|
|
if (!sctp_transport_dst_check(tp)) {
|
2006-06-18 09:57:28 +04:00
|
|
|
sctp_transport_route(tp, NULL, sctp_sk(sk));
|
|
|
|
if (asoc && (asoc->param_flags & SPP_PMTUD_ENABLE)) {
|
2012-07-16 14:57:14 +04:00
|
|
|
sctp_assoc_sync_pmtu(sk, asoc);
|
2006-06-18 09:57:28 +04:00
|
|
|
}
|
|
|
|
}
|
2009-06-02 09:19:30 +04:00
|
|
|
dst = dst_clone(tp->dst);
|
2009-06-29 02:49:37 +04:00
|
|
|
if (!dst)
|
2006-06-18 09:57:28 +04:00
|
|
|
goto no_route;
|
2016-06-02 21:05:43 +03:00
|
|
|
skb_dst_set(head, dst);
|
2006-06-18 09:57:28 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Build the SCTP header. */
|
2016-06-02 21:05:43 +03:00
|
|
|
sh = (struct sctphdr *)skb_push(head, sizeof(struct sctphdr));
|
|
|
|
skb_reset_transport_header(head);
|
2005-04-17 02:20:36 +04:00
|
|
|
sh->source = htons(packet->source_port);
|
|
|
|
sh->dest = htons(packet->destination_port);
|
|
|
|
|
|
|
|
/* From 6.8 Adler-32 Checksum Calculation:
|
|
|
|
* After the packet is constructed (containing the SCTP common
|
|
|
|
* header and one or more control or DATA chunks), the
|
|
|
|
* transmitter shall:
|
|
|
|
*
|
|
|
|
* 1) Fill in the proper Verification Tag in the SCTP common
|
|
|
|
* header and initialize the checksum field to 0's.
|
|
|
|
*/
|
|
|
|
sh->vtag = htonl(packet->vtag);
|
|
|
|
sh->checksum = 0;
|
|
|
|
|
net: sctp: rework debugging framework to use pr_debug and friends
We should get rid of all own SCTP debug printk macros and use the ones
that the kernel offers anyway instead. This makes the code more readable
and conform to the kernel code, and offers all the features of dynamic
debbuging that pr_debug() et al has, such as only turning on/off portions
of debug messages at runtime through debugfs. The runtime cost of having
CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing,
is negligible [1]. If kernel debugging is completly turned off, then these
statements will also compile into "empty" functions.
While we're at it, we also need to change the Kconfig option as it /now/
only refers to the ifdef'ed code portions in outqueue.c that enable further
debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code
was enabled with this Kconfig option and has now been removed, we
transform those code parts into WARNs resp. where appropriate BUG_ONs so
that those bugs can be more easily detected as probably not many people
have SCTP debugging permanently turned on.
To turn on all SCTP debugging, the following steps are needed:
# mount -t debugfs none /sys/kernel/debug
# echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control
This can be done more fine-grained on a per file, per line basis and others
as described in [2].
[1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf
[2] Documentation/dynamic-debug-howto.txt
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-28 21:49:40 +04:00
|
|
|
pr_debug("***sctp_transmit_packet***\n");
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
do {
|
|
|
|
/* Set up convenience variables... */
|
|
|
|
chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list);
|
|
|
|
pktcount++;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/* Calculate packet size, so it fits in PMTU. Leave
|
|
|
|
* other chunks for the next packets.
|
|
|
|
*/
|
|
|
|
if (gso) {
|
|
|
|
pkt_size = packet->overhead;
|
|
|
|
list_for_each_entry(chunk, &packet->chunk_list, list) {
|
|
|
|
int padded = WORD_ROUND(chunk->skb->len);
|
|
|
|
|
|
|
|
if (pkt_size + padded > tp->pathmtu)
|
|
|
|
break;
|
|
|
|
pkt_size += padded;
|
sctp: Fix mis-ordering of user space data when multihoming in use
Recently had a bug reported to me, in which the user was sending
packets with a payload containing a sequence number. The packets
were getting delivered in order according the chunk TSN values, but
the sequence values in the payload were arriving out of order. At
first I thought it must be an application error, but we eventually
found it to be a problem on the transmit side in the sctp stack.
The conditions for the error are that multihoming must be in use,
and it helps if each transport has a different pmtu. The problem
occurs in sctp_outq_flush. Basically we dequeue packets from the
data queue, and attempt to append them to the orrered packet for a
given transport. After we append a data chunk we add the trasport
to the end of a list of transports to have their packets sent at
the end of sctp_outq_flush. The problem occurs when a data chunks
fills up a offered packet on a transport. The function that does
the appending (sctp_packet_transmit_chunk), will try to call
sctp_packet_transmit on the full packet, and then append the chunk
to a new packet. This call to sctp_packet_transmit, sends that
packet ahead of the others that may be queued in the transport_list
in sctp_outq_flush. The result is that frames that were sent in one
order from the user space sending application get re-ordered prior
to tsn assignment in sctp_packet_transmit, resulting in mis-sequencing
of data payloads, even though tsn ordering is correct.
The fix is to change where we assign a tsn. By doing this earlier,
we are then free to place chunks in packets, whatever way we
see fit and the protocol will make sure to do all the appropriate
re-ordering on receive as is needed.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: William Reich <reich@ulticom.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-11-23 23:54:00 +03:00
|
|
|
}
|
2013-11-25 07:26:57 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/* Allocate a new skb. */
|
|
|
|
nskb = alloc_skb(pkt_size + MAX_HEADER, gfp);
|
|
|
|
if (!nskb)
|
|
|
|
goto nomem;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/* Make sure the outbound skb has enough header
|
|
|
|
* room reserved.
|
|
|
|
*/
|
|
|
|
skb_reserve(nskb, packet->overhead + MAX_HEADER);
|
|
|
|
} else {
|
|
|
|
nskb = head;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/**
|
|
|
|
* 3.2 Chunk Field Descriptions
|
|
|
|
*
|
|
|
|
* The total length of a chunk (including Type, Length and
|
|
|
|
* Value fields) MUST be a multiple of 4 bytes. If the length
|
|
|
|
* of the chunk is not a multiple of 4 bytes, the sender MUST
|
|
|
|
* pad the chunk with all zero bytes and this padding is not
|
|
|
|
* included in the chunk length field. The sender should
|
|
|
|
* never pad with more than 3 bytes.
|
|
|
|
*
|
|
|
|
* [This whole comment explains WORD_ROUND() below.]
|
2007-09-17 06:32:45 +04:00
|
|
|
*/
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
pkt_size -= packet->overhead;
|
|
|
|
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
|
|
|
|
list_del_init(&chunk->list);
|
|
|
|
if (sctp_chunk_is_data(chunk)) {
|
|
|
|
/* 6.3.1 C4) When data is in flight and when allowed
|
|
|
|
* by rule C5, a new RTT measurement MUST be made each
|
|
|
|
* round trip. Furthermore, new RTT measurements
|
|
|
|
* SHOULD be made no more than once per round-trip
|
|
|
|
* for a given destination transport address.
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (!chunk->resent && !tp->rto_pending) {
|
|
|
|
chunk->rtt_in_progress = 1;
|
|
|
|
tp->rto_pending = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
has_data = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len;
|
|
|
|
if (padding)
|
|
|
|
memset(skb_put(chunk->skb, padding), 0, padding);
|
|
|
|
|
|
|
|
/* if this is the auth chunk that we are adding,
|
|
|
|
* store pointer where it will be added and put
|
|
|
|
* the auth into the packet.
|
|
|
|
*/
|
|
|
|
if (chunk == packet->auth)
|
|
|
|
auth = skb_tail_pointer(nskb);
|
|
|
|
|
|
|
|
memcpy(skb_put(nskb, chunk->skb->len),
|
2006-06-18 09:57:28 +04:00
|
|
|
chunk->skb->data, chunk->skb->len);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, rtt_in_progress:%d\n",
|
|
|
|
chunk,
|
|
|
|
sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)),
|
|
|
|
chunk->has_tsn ? "TSN" : "No TSN",
|
|
|
|
chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0,
|
|
|
|
ntohs(chunk->chunk_hdr->length), chunk->skb->len,
|
|
|
|
chunk->rtt_in_progress);
|
|
|
|
|
|
|
|
/* If this is a control chunk, this is our last
|
|
|
|
* reference. Free data chunks after they've been
|
|
|
|
* acknowledged or have failed.
|
|
|
|
* Re-queue auth chunks if needed.
|
|
|
|
*/
|
|
|
|
pkt_size -= WORD_ROUND(chunk->skb->len);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-07-07 15:39:29 +03:00
|
|
|
if (!sctp_chunk_is_data(chunk) && chunk != packet->auth)
|
2016-06-02 21:05:43 +03:00
|
|
|
sctp_chunk_free(chunk);
|
|
|
|
|
|
|
|
if (!pkt_size)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* SCTP-AUTH, Section 6.2
|
|
|
|
* The sender MUST calculate the MAC as described in RFC2104 [2]
|
|
|
|
* using the hash function H as described by the MAC Identifier and
|
|
|
|
* the shared association key K based on the endpoint pair shared key
|
|
|
|
* described by the shared key identifier. The 'data' used for the
|
|
|
|
* computation of the AUTH-chunk is given by the AUTH chunk with its
|
|
|
|
* HMAC field set to zero (as shown in Figure 6) followed by all
|
|
|
|
* chunks that are placed after the AUTH chunk in the SCTP packet.
|
|
|
|
*/
|
|
|
|
if (auth)
|
|
|
|
sctp_auth_calculate_hmac(asoc, nskb,
|
|
|
|
(struct sctp_auth_chunk *)auth,
|
|
|
|
gfp);
|
|
|
|
|
2016-07-07 15:39:29 +03:00
|
|
|
if (packet->auth) {
|
|
|
|
if (!list_empty(&packet->chunk_list)) {
|
|
|
|
/* We will generate more packets, so re-queue
|
|
|
|
* auth chunk.
|
|
|
|
*/
|
2016-07-30 08:58:35 +03:00
|
|
|
list_add(&packet->auth->list,
|
|
|
|
&packet->chunk_list);
|
2016-07-07 15:39:29 +03:00
|
|
|
} else {
|
|
|
|
sctp_chunk_free(packet->auth);
|
|
|
|
packet->auth = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
if (!gso)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (skb_gro_receive(&head, nskb))
|
|
|
|
goto nomem;
|
|
|
|
nskb = NULL;
|
|
|
|
if (WARN_ON_ONCE(skb_shinfo(head)->gso_segs >=
|
|
|
|
sk->sk_gso_max_segs))
|
|
|
|
goto nomem;
|
|
|
|
} while (!list_empty(&packet->chunk_list));
|
2007-09-17 06:32:45 +04:00
|
|
|
|
|
|
|
/* 2) Calculate the Adler-32 checksum of the whole packet,
|
|
|
|
* including the SCTP common header and all the
|
|
|
|
* chunks.
|
|
|
|
*
|
|
|
|
* Note: Adler-32 is no longer applicable, as has been replaced
|
|
|
|
* by CRC32-C as described in <draft-ietf-tsvwg-sctpcsum-02.txt>.
|
2016-06-02 21:05:43 +03:00
|
|
|
*
|
|
|
|
* If it's a GSO packet, it's postponed to sctp_skb_segment.
|
2007-09-17 06:32:45 +04:00
|
|
|
*/
|
2016-06-02 21:05:43 +03:00
|
|
|
if (!sctp_checksum_disable || gso) {
|
|
|
|
if (!gso && (!(dst->dev->features & NETIF_F_SCTP_CRC) ||
|
|
|
|
dst_xfrm(dst) || packet->ipfragok)) {
|
|
|
|
sh->checksum = sctp_compute_cksum(head, 0);
|
2011-07-13 18:10:29 +04:00
|
|
|
} else {
|
2011-03-31 05:57:33 +04:00
|
|
|
/* no need to seed pseudo checksum for SCTP */
|
2016-06-02 21:05:43 +03:00
|
|
|
head->ip_summed = CHECKSUM_PARTIAL;
|
|
|
|
head->csum_start = skb_transport_header(head) - head->head;
|
|
|
|
head->csum_offset = offsetof(struct sctphdr, checksum);
|
2009-04-28 02:35:52 +04:00
|
|
|
}
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* IP layer ECN support
|
|
|
|
* From RFC 2481
|
|
|
|
* "The ECN-Capable Transport (ECT) bit would be set by the
|
|
|
|
* data sender to indicate that the end-points of the
|
|
|
|
* transport protocol are ECN-capable."
|
|
|
|
*
|
|
|
|
* Now setting the ECT bit all the time, as it should not cause
|
|
|
|
* any problems protocol-wise even if our peer ignores it.
|
|
|
|
*
|
|
|
|
* Note: The works for IPv6 layer checks this bit too later
|
|
|
|
* in transmission. See IP6_ECN_flow_xmit().
|
|
|
|
*/
|
2016-06-02 21:05:43 +03:00
|
|
|
tp->af_specific->ecn_capable(sk);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Set up the IP options. */
|
|
|
|
/* BUG: not implemented
|
|
|
|
* For v4 this all lives somewhere in sk->sk_opt...
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Dump that on IP! */
|
2012-12-01 08:49:42 +04:00
|
|
|
if (asoc) {
|
2016-06-02 21:05:43 +03:00
|
|
|
asoc->stats.opackets += pktcount;
|
2012-12-01 08:49:42 +04:00
|
|
|
if (asoc->peer.last_sent_to != tp)
|
|
|
|
/* Considering the multiple CPU scenario, this is a
|
|
|
|
* "correcter" place for last_sent_to. --xguo
|
|
|
|
*/
|
|
|
|
asoc->peer.last_sent_to = tp;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
if (has_data) {
|
|
|
|
struct timer_list *timer;
|
|
|
|
unsigned long timeout;
|
|
|
|
|
|
|
|
/* Restart the AUTOCLOSE timer when sending data. */
|
2013-12-10 15:48:15 +04:00
|
|
|
if (sctp_state(asoc, ESTABLISHED) &&
|
|
|
|
asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) {
|
2005-04-17 02:20:36 +04:00
|
|
|
timer = &asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE];
|
|
|
|
timeout = asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE];
|
|
|
|
|
|
|
|
if (!mod_timer(timer, jiffies + timeout))
|
|
|
|
sctp_association_hold(asoc);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
pr_debug("***sctp_transmit_packet*** skb->len:%d\n", head->len);
|
|
|
|
|
|
|
|
if (gso) {
|
|
|
|
/* Cleanup our debris for IP stacks */
|
|
|
|
memset(head->cb, 0, max(sizeof(struct inet_skb_parm),
|
|
|
|
sizeof(struct inet6_skb_parm)));
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
skb_shinfo(head)->gso_segs = pktcount;
|
|
|
|
skb_shinfo(head)->gso_size = GSO_BY_FRAGS;
|
|
|
|
|
|
|
|
/* We have to refresh this in case we are xmiting to
|
|
|
|
* more than one transport at a time
|
|
|
|
*/
|
|
|
|
rcu_read_lock();
|
|
|
|
if (__sk_dst_get(sk) != tp->dst) {
|
|
|
|
dst_hold(tp->dst);
|
|
|
|
sk_setup_caps(sk, tp->dst);
|
|
|
|
}
|
|
|
|
rcu_read_unlock();
|
|
|
|
}
|
|
|
|
head->ignore_df = packet->ipfragok;
|
|
|
|
tp->af_specific->sctp_xmit(head, tp);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
out:
|
2009-09-02 09:05:33 +04:00
|
|
|
sctp_packet_reset(packet);
|
2005-04-17 02:20:36 +04:00
|
|
|
return err;
|
|
|
|
no_route:
|
2016-06-02 21:05:43 +03:00
|
|
|
kfree_skb(head);
|
|
|
|
if (nskb != head)
|
|
|
|
kfree_skb(nskb);
|
2015-06-29 11:41:03 +03:00
|
|
|
|
|
|
|
if (asoc)
|
|
|
|
IP_INC_STATS(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* FIXME: Returning the 'err' will effect all the associations
|
|
|
|
* associated with a socket, although only one of the paths of the
|
|
|
|
* association is unreachable.
|
|
|
|
* The real failure of a transport or association can be passed on
|
|
|
|
* to the user via notifications. So setting this error may not be
|
|
|
|
* required.
|
|
|
|
*/
|
|
|
|
/* err = -EHOSTUNREACH; */
|
|
|
|
err:
|
|
|
|
/* Control chunks are unreliable so just drop them. DATA chunks
|
|
|
|
* will get resent or dropped later.
|
|
|
|
*/
|
|
|
|
|
2005-07-09 08:47:49 +04:00
|
|
|
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
|
|
|
|
list_del_init(&chunk->list);
|
2005-04-17 02:20:36 +04:00
|
|
|
if (!sctp_chunk_is_data(chunk))
|
2007-02-09 17:25:18 +03:00
|
|
|
sctp_chunk_free(chunk);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
goto out;
|
|
|
|
nomem:
|
2016-07-07 15:39:29 +03:00
|
|
|
if (packet->auth && list_empty(&packet->auth->list))
|
|
|
|
sctp_chunk_free(packet->auth);
|
2005-04-17 02:20:36 +04:00
|
|
|
err = -ENOMEM;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************
|
|
|
|
* 2nd Level Abstractions
|
|
|
|
********************************************************************/
|
|
|
|
|
2009-08-07 18:43:07 +04:00
|
|
|
/* This private function check to see if a chunk can be added */
|
|
|
|
static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet,
|
2005-04-17 02:20:36 +04:00
|
|
|
struct sctp_chunk *chunk)
|
|
|
|
{
|
2009-08-07 18:43:07 +04:00
|
|
|
size_t datasize, rwnd, inflight, flight_size;
|
2005-04-17 02:20:36 +04:00
|
|
|
struct sctp_transport *transport = packet->transport;
|
|
|
|
struct sctp_association *asoc = transport->asoc;
|
|
|
|
struct sctp_outq *q = &asoc->outqueue;
|
|
|
|
|
|
|
|
/* RFC 2960 6.1 Transmission of DATA Chunks
|
|
|
|
*
|
|
|
|
* A) At any given time, the data sender MUST NOT transmit new data to
|
|
|
|
* any destination transport address if its peer's rwnd indicates
|
|
|
|
* that the peer has no buffer space (i.e. rwnd is 0, see Section
|
|
|
|
* 6.2.1). However, regardless of the value of rwnd (including if it
|
|
|
|
* is 0), the data sender can always have one DATA chunk in flight to
|
|
|
|
* the receiver if allowed by cwnd (see rule B below). This rule
|
|
|
|
* allows the sender to probe for a change in rwnd that the sender
|
|
|
|
* missed due to the SACK having been lost in transit from the data
|
|
|
|
* receiver to the data sender.
|
|
|
|
*/
|
|
|
|
|
|
|
|
rwnd = asoc->peer.rwnd;
|
2009-08-07 18:43:07 +04:00
|
|
|
inflight = q->outstanding_bytes;
|
|
|
|
flight_size = transport->flight_size;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
datasize = sctp_data_size(chunk);
|
|
|
|
|
2014-07-22 12:59:08 +04:00
|
|
|
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;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* RFC 2960 6.1 Transmission of DATA Chunks
|
|
|
|
*
|
|
|
|
* B) At any given time, the sender MUST NOT transmit new data
|
|
|
|
* to a given transport address if it has cwnd or more bytes
|
|
|
|
* of data outstanding to that transport address.
|
|
|
|
*/
|
|
|
|
/* RFC 7.2.4 & the Implementers Guide 2.8.
|
|
|
|
*
|
|
|
|
* 3) ...
|
|
|
|
* When a Fast Retransmit is being performed the sender SHOULD
|
|
|
|
* ignore the value of cwnd and SHOULD NOT delay retransmission.
|
|
|
|
*/
|
2014-07-22 12:59:08 +04:00
|
|
|
if (chunk->fast_retransmit != SCTP_NEED_FRTX &&
|
|
|
|
flight_size >= transport->cwnd)
|
|
|
|
return SCTP_XMIT_RWND_FULL;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* 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.
|
|
|
|
*/
|
|
|
|
|
2014-07-22 12:59:08 +04:00
|
|
|
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.
|
|
|
|
*/
|
2016-04-01 20:05:48 +03:00
|
|
|
if (chunk->skb->len + q->out_qlen >
|
|
|
|
transport->pathmtu - packet->overhead - sizeof(sctp_data_chunk_t) - 4)
|
2014-07-22 12:59:08 +04:00
|
|
|
/* 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 */
|
2014-07-22 12:59:14 +04:00
|
|
|
return SCTP_XMIT_DELAY;
|
2009-08-07 18:43:07 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
/* This private function does management things when adding DATA chunk */
|
|
|
|
static void sctp_packet_append_data(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk)
|
|
|
|
{
|
|
|
|
struct sctp_transport *transport = packet->transport;
|
|
|
|
size_t datasize = sctp_data_size(chunk);
|
|
|
|
struct sctp_association *asoc = transport->asoc;
|
|
|
|
u32 rwnd = asoc->peer.rwnd;
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Keep track of how many bytes are in flight over this transport. */
|
|
|
|
transport->flight_size += datasize;
|
|
|
|
|
|
|
|
/* Keep track of how many bytes are in flight to the receiver. */
|
|
|
|
asoc->outqueue.outstanding_bytes += datasize;
|
|
|
|
|
2011-12-19 08:11:40 +04:00
|
|
|
/* Update our view of the receiver's rwnd. */
|
2005-04-17 02:20:36 +04:00
|
|
|
if (datasize < rwnd)
|
|
|
|
rwnd -= datasize;
|
|
|
|
else
|
|
|
|
rwnd = 0;
|
|
|
|
|
|
|
|
asoc->peer.rwnd = rwnd;
|
|
|
|
/* Has been accepted for transmission. */
|
|
|
|
if (!asoc->peer.prsctp_capable)
|
|
|
|
chunk->msg->can_abandon = 0;
|
sctp: Fix mis-ordering of user space data when multihoming in use
Recently had a bug reported to me, in which the user was sending
packets with a payload containing a sequence number. The packets
were getting delivered in order according the chunk TSN values, but
the sequence values in the payload were arriving out of order. At
first I thought it must be an application error, but we eventually
found it to be a problem on the transmit side in the sctp stack.
The conditions for the error are that multihoming must be in use,
and it helps if each transport has a different pmtu. The problem
occurs in sctp_outq_flush. Basically we dequeue packets from the
data queue, and attempt to append them to the orrered packet for a
given transport. After we append a data chunk we add the trasport
to the end of a list of transports to have their packets sent at
the end of sctp_outq_flush. The problem occurs when a data chunks
fills up a offered packet on a transport. The function that does
the appending (sctp_packet_transmit_chunk), will try to call
sctp_packet_transmit on the full packet, and then append the chunk
to a new packet. This call to sctp_packet_transmit, sends that
packet ahead of the others that may be queued in the transport_list
in sctp_outq_flush. The result is that frames that were sent in one
order from the user space sending application get re-ordered prior
to tsn assignment in sctp_packet_transmit, resulting in mis-sequencing
of data payloads, even though tsn ordering is correct.
The fix is to change where we assign a tsn. By doing this earlier,
we are then free to place chunks in packets, whatever way we
see fit and the protocol will make sure to do all the appropriate
re-ordering on receive as is needed.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: William Reich <reich@ulticom.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-11-23 23:54:00 +03:00
|
|
|
sctp_chunk_assign_tsn(chunk);
|
|
|
|
sctp_chunk_assign_ssn(chunk);
|
2009-08-07 18:43:07 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet,
|
|
|
|
struct sctp_chunk *chunk,
|
|
|
|
u16 chunk_len)
|
|
|
|
{
|
2016-09-08 12:54:11 +03:00
|
|
|
size_t psize, pmtu, maxsize;
|
2009-08-07 18:43:07 +04:00
|
|
|
sctp_xmit_t retval = SCTP_XMIT_OK;
|
|
|
|
|
|
|
|
psize = packet->size;
|
2016-06-02 21:05:43 +03:00
|
|
|
if (packet->transport->asoc)
|
|
|
|
pmtu = packet->transport->asoc->pathmtu;
|
|
|
|
else
|
|
|
|
pmtu = packet->transport->pathmtu;
|
2009-08-07 18:43:07 +04:00
|
|
|
|
|
|
|
/* Decide if we need to fragment or resubmit later. */
|
2016-06-02 21:05:43 +03:00
|
|
|
if (psize + chunk_len > pmtu) {
|
|
|
|
/* It's OK to fragment at IP level if any one of the following
|
2009-08-07 18:43:07 +04:00
|
|
|
* is true:
|
2016-06-02 21:05:43 +03:00
|
|
|
* 1. The packet is empty (meaning this chunk is greater
|
|
|
|
* the MTU)
|
|
|
|
* 2. The packet doesn't have any data in it yet and data
|
|
|
|
* requires authentication.
|
2009-08-07 18:43:07 +04:00
|
|
|
*/
|
2016-06-02 21:05:43 +03:00
|
|
|
if (sctp_packet_empty(packet) ||
|
2009-08-07 18:43:07 +04:00
|
|
|
(!packet->has_data && chunk->auth)) {
|
|
|
|
/* We no longer do re-fragmentation.
|
|
|
|
* Just fragment at the IP layer, if we
|
|
|
|
* actually hit this condition
|
|
|
|
*/
|
|
|
|
packet->ipfragok = 1;
|
2016-06-02 21:05:43 +03:00
|
|
|
goto out;
|
2009-08-07 18:43:07 +04:00
|
|
|
}
|
2016-06-02 21:05:43 +03:00
|
|
|
|
2016-09-08 12:54:11 +03:00
|
|
|
/* 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;
|
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
/* 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
|
|
|
|
* a large frame. So in this case we allow the
|
|
|
|
* fragmentation by forcing it to be in a new packet.
|
|
|
|
*/
|
|
|
|
if (!sctp_chunk_is_data(chunk) && packet->has_data)
|
|
|
|
retval = SCTP_XMIT_PMTU_FULL;
|
|
|
|
|
|
|
|
if (psize + chunk_len > packet->max_size)
|
|
|
|
/* Hit GSO/PMTU limit, gotta flush */
|
|
|
|
retval = SCTP_XMIT_PMTU_FULL;
|
|
|
|
|
|
|
|
if (!packet->transport->burst_limited &&
|
|
|
|
psize + chunk_len > (packet->transport->cwnd >> 1))
|
|
|
|
/* Do not allow a single GSO packet to use more
|
|
|
|
* than half of cwnd.
|
|
|
|
*/
|
|
|
|
retval = SCTP_XMIT_PMTU_FULL;
|
|
|
|
|
|
|
|
if (packet->transport->burst_limited &&
|
|
|
|
psize + chunk_len > (packet->transport->burst_limited >> 1))
|
|
|
|
/* Do not allow a single GSO packet to use more
|
|
|
|
* than half of original cwnd.
|
|
|
|
*/
|
|
|
|
retval = SCTP_XMIT_PMTU_FULL;
|
|
|
|
/* Otherwise it will fit in the GSO packet */
|
2009-08-07 18:43:07 +04:00
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2016-06-02 21:05:43 +03:00
|
|
|
out:
|
2005-04-17 02:20:36 +04:00
|
|
|
return retval;
|
|
|
|
}
|