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>
Current implementation of max.burst ends up limiting new
data during cwnd decay period. The decay is happening becuase
the connection is idle and we are allowed to fill the congestion
window. The point of max.burst is to limit micro-bursts in response
to large acks. This still happens, as max.burst is still applied
to each transmit opportunity. It will also apply if a very large
send is made (greater then allowed by burst).
Tested-by: Florian Niederbacher <florian.niederbacher@student.uibk.ac.at>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The transport last_time_used variable is rather useless.
It was only used when determining if CWND needs to be updated
due to idle transport. However, idle transport detection was
based on a Heartbeat timer and last_time_used was not incremented
when sending Heartbeats. As a result the check for cwnd reduction
was always true. We can get rid of the variable and just base
our cwnd manipulation on the HB timer (like the code comment sais).
We also have to call into the cwnd manipulation function regardless
of whether HBs are enabled or not. That way we will detect idle
transports if the user has disabled Heartbeats.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
We currently track if AUTH has been bundled using the 'auth'
pointer to the chunk. However, AUTH is disallowed after DATA
is already in the packet, so we need to instead use the
'has_auth' field.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The packet information does not reset after packet transmit, this
may cause some problems such as following DATA chunk be sent without
AUTH chunk, even if the authentication of DATA chunk has been
requested by the peer.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
SCTP will delay the last part of a large write due to NAGLE, if that
part is smaller then MTU. Since we are doing large writes, we might
as well send the last portion now instead of waiting untill the next
large write happens. The small portion will be sent as is regardless,
so it's better to not delay it.
This is a result of much discussions with Wei Yongjun <yjwei@cn.fujitsu.com>
and Doug Graham <dgraham@nortel.com>. Many thanks go out to them.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The decision to delay due to Nagle should be based on the path mtu
and future packet size. We currently incorrectly base it on
'frag_point' which is the SCTP DATA segment size, and also we do
not count DATA chunk header overhead in the computation. This
actuall allows situations where a user can set low 'frag_point',
and then send small messages without delay.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
We are now trying to bundle SACKs when we have outbound
DATA to send. However, there are situations where this
outbound DATA will not be sent (due to congestion or
available window). In such cases it's ok to wait for the
timer to expire. This patch refactors the sending code
so that betfore attempting to bundle the SACK we check
to see if the DATA will actually be transmitted.
Based on eirlier works for Doug Graham <dgraham@nortel.com> and
Wei Youngjun <yjwei@cn.fujitsu.com>.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
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>
Commit 'net: skb->dst accessors'(adf30907d6)
broken the sctp protocol stack, the sctp packet can never be sent out after
Eric Dumazet's patch, which have typo in the sctp code.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Vlad Yasevich <vladisalv.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define three accessors to get/set dst attached to a skb
struct dst_entry *skb_dst(const struct sk_buff *skb)
void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
void skb_dst_drop(struct sk_buff *skb)
This one should replace occurrences of :
dst_release(skb->dst)
skb->dst = NULL;
Delete skb->dst field
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
this is the sctp code to enable hardware crc32c offload for
adapters that support it.
Originally by: Vlad Yasevich <vladislav.yasevich@hp.com>
modified by Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove 2 TEST_FRAME hacks that are no longer needed. These allowed
sctp regression tests to compile before, but are no longer needed.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sctp crc32c checksum is always generated in little endian.
So, we clean up the code to treat it as little endian and remove
all the __force casts.
Suggested by Herbert Xu.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a new version of my patch, now using a module parameter instead
of a sysctl, so that the option is harder to find. Please note that,
once the module is loaded, it is still possible to change the value of
the parameter in /sys/module/sctp/parameters/, which is useful if you
want to do performance comparisons without rebooting.
Computation of SCTP checksums significantly affects the performance of
SCTP. For example, using two dual-Opteron 246 connected using a Gbe
network, it was not possible to achieve more than ~730 Mbps, compared to
941 Mbps after disabling SCTP checksums.
Unfortunately, SCTP checksum offloading in NICs is not commonly
available (yet).
By default, checksums are still enabled, of course.
Signed-off-by: Lucas Nussbaum <lucas.nussbaum@ens-lyon.fr>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Base versions handle constant folding now.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Recent changes to the retransmit code exposed a long standing
bug where it was possible for a chunk to be time stamped
after the retransmit timer was reset. This caused a rare
situation where the retrnamist timer has expired, but
nothing was marked for retrnasmission because all of
timesamps on data were less then 1 rto ago. As result,
the timer was never restarted since nothing was retransmitted,
and this resulted in a hung association that did couldn't
complete the data transfer. The solution is to timestamp
the chunk when it's added to the packet for transmission
purposes. After the packet is trsnmitted the rtx timer
is restarted. This guarantees that when the timer expires,
there will be data to retransmit.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sctp_chunks should be put on a diet. This is some of the low hanging
fruit that we can strip out. Changes all the __s8/__u8 flags to
bitfields. Saves 12 bytes per chunk.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Loopback used to clobber the ip_summed filed which sctp then used
to figure out if it needed to do checksumming or not. Now that
loopback doesn't do that any more, sctp needs to set the ip_summed
field correctly.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ipfragok flag controls whether the packet may be fragmented
either on the local host on beyond. The latter is only valid on
IPv4.
In fact, we never want to do the latter even on IPv4 when PMTU is
enabled. This is because even though we can't fragment packets
within SCTP due to the prtocol's inherent faults, we can still
fragment it at IP layer. By setting the DF bit we will improve
the PMTU process.
RFC 2960 only says that we SHOULD clear the DF bit in this case,
so we're compliant even if we set the DF bit. In fact RFC 4960
no longer has this statement.
Once we make this change, we only need to control the local
fragmentation. There is already a bit in the skb which controls
that, local_df. So this patch sets that instead of using the
ipfragok argument.
The only complication is that there isn't a struct sock object
per transport, so for IPv4 we have to resort to changing the
pmtudisc field for every packet. This should be safe though
as the protocol is single-threaded.
Note that after this patch we can remove ipfragok from the rest
of the stack too.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
RFC 4960, Section 11.4. Protection of Non-SCTP-Capable Hosts
When an SCTP stack receives a packet containing multiple control or
DATA chunks and the processing of the packet requires the sending of
multiple chunks in response, the sender of the response chunk(s) MUST
NOT send more than one packet. If bundling is supported, multiple
response chunks that fit into a single packet MAY be bundled together
into one single response packet. If bundling is not supported, then
the sender MUST NOT send more than one response chunk and MUST
discard all other responses. Note that this rule does NOT apply to a
SACK chunk, since a SACK chunk is, in itself, a response to DATA and
a SACK does not require a response of more DATA.
We implement this by not servicing our outqueue until we reach the end
of the packet. This enables maximum bundling. We also identify
'response' chunks and make sure that we only send 1 packet when sending
such chunks.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit e9df2e8fd8 ("[IPV6]: Use
appropriate sock tclass setting for routing lookup.") also changed the
way that ECN capable transports mark this capability in IPv6. As a
result, SCTP was not marking ECN capablity because the traffic class
was never set. This patch brings back the markings for IPv6 traffic.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I was notified by Randy Stewart that lksctp claims to be
"the reference implementation". First of all, "the
refrence implementation" was the original implementation
of SCTP in usersapce written ty Randy and a few others.
Second, after looking at the definiton of 'reference implementation',
we don't really meet the requirements.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The crc32c library used an identical table and algorithm
as SCTP. Switch to using the library instead of carrying
our own table. Using crypto layer proved to have too
much overhead compared to using the library directly.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
SCTP-AUTH, Section 6.2:
Endpoints MUST send all requested chunks authenticated where this has
been requested by the peer. The other chunks MAY be sent
authenticated or not. If endpoint pair shared keys are used, one of
them MUST be selected for authentication.
To send chunks in an authenticated way, the sender MUST include these
chunks after an AUTH chunk. This means that a sender MUST bundle
chunks in order to authenticate them.
If the endpoint has no endpoint pair shared key for the peer, it MUST
use Shared Key Identifier 0 with an empty endpoint pair shared key.
If there are multiple endpoint shared keys the sender selects one and
uses the corresponding Shared Key Identifier
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch initializes AUTH related members of the generic SCTP
structures and provides a way to enable/disable auth extension.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Spring cleaning time...
There seems to be a lot of places in the network code that have
extra bogus semicolons after conditionals. Most commonly is a
bogus semicolon after: switch() { }
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently if the sender is sending small messages, it can cause a receiver
to run out of receive buffer space even when the advertised receive window
is still open and results in packet drops and retransmissions. Including
a overhead while updating the sender's view of peer receive window will
reduce the chances of receive buffer space overshooting the receive window.
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
SCTP used to "fast retransmit" a TSN every time we hit the number
of missing reports for the TSN. However the Implementers Guide
specifies that we should only "fast retransmit" a given TSN once.
Subsequent retransmits should be timeouts only. Also change the
number of missing reports to 3 as per the latest IG(similar to TCP).
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds support to set/get heartbeat interval, maximum number of
retransmissions, pathmtu, sackdelay time for a particular transport/
association/socket as per the latest SCTP sockets api draft11.
Signed-off-by: Frank Filz <ffilz@us.ibm.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Neil Horman
Vlad Yasevich
Wei Yongjun
Doug Graham
Eric Dumazet
Jesse Brandeburg
Lucas Nussbaum
Harvey Harrison
Herbert Xu
Pavel Emelyanov
Stephen Hemminger
YOSHIFUJI Hideaki
Sridhar Samudrala
Frank Filz
David S. Miller
Linus Torvalds