We had a bug that we never stored the user-defined value for
MAXSEG when setting the value on an association. Thus future
PMTU events ended up re-writing the frag point and increasing
it past user limit. Additionally, when setting the option on
the socket/endpoint, we effect all current associations, which
is against spec.
Now, we store the user 'maxseg' value along with the computed
'frag_point'. We inherit 'maxseg' from the socket at association
creation and use it as an upper limit for 'frag_point' when its
set.
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 currently set a_rwnd to 0 when faking a SACK from SHUTDOWN.
This results in an hung association if the remote only uses
SHUTDOWNs (which it's allowed to do) to acknowlege DATA when
closing. The reason for that is that we simply honor the a_rwnd
from the sack, but since we faked it to be 0, we enter 0-window
probing. The fix is to use the peers old rwnd and add our flight
size to it.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
SCTP has a problem that when small chunks are used, it is possible
to exhaust the receiver buffer without fully closing receive window.
This happens due to all overhead that we have account for with small
messages. To fix this, when receive buffer is exceeded, we'll drop
the window to 0 and save the 'drop' portion. When application starts
reading data and freeing up recevie buffer space, we'll wait until
we've reached the 'drop' window and then add back this 'drop' one
mtu at a time. This worked well in testing and under stress produced
rather even recovery.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
If T3 timer expires, we are retransmitting data due to timeout any
any fast recovery is null and void. We can clear the fast recovery
flag.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
SCTP RFC 4960 states that unacknowledged HEARTBEATS count as
errors agains a given transport or endpoint. As such, we
should increment the error counts for only for unacknowledged
HB, otherwise we detect failure too soon. This goes for both
the overall error count and the path error count.
Now, there is a difference in how the detection is done
between the two. The path error detection is done after
the increment, so to detect it properly, we actually need
to exceed the path threshold. The overall error detection
is done _BEFORE_ the increment. Thus to detect the failure,
it's enough for the error count to match the threshold.
This is why all the state functions use '>=' to detect failure,
while path detection uses '>'.
Thanks goes to Chunbo Luo <chunbo.luo@windriver.com> who first
proposed patches to fix this issue and made me re-read the spec
and the code to figure out how this cruft really works.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The receiver of the HEARTBEAT should respond with a HEARTBEAT ACK
that contains the Heartbeat Information field copied from the
received HEARTBEAT chunk. So the received HEARTBEAT-ACK chunk
must have a length of:
sizeof(sctp_chunkhdr_t) + sizeof(sctp_sender_hb_info_t)
A badly formatted HB-ACK chunk, it is possible that we may access
invalid memory. We should really make sure that the chunk format
is what we expect, before attempting to touch the data.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
If Cumulative TSN Ack field of SHUTDOWN chunk is less than the
Cumulative TSN Ack Point then drop the SHUTDOWN chunk.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Currenlty, sctp breaks up user messages into fragments and
sends each fragment to the lower layer by itself. This means
that for each fragment we go all the way down the stack
and back up. This also discourages bundling of multiple
fragments when they can fit into a sigle packet (ex: due
to user setting a low fragmentation threashold).
We introduce a new command SCTP_CMD_SND_MSG and hand the
whole message down state machine. The state machine and
the side-effect parser will cork the queue, add all chunks
from the message to the queue, and then un-cork the queue
thus causing the chunks to get transmitted.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
If the association has a SACK timer pending and now DATA queued
to be send, we'll try to bundle the SACK with the next application send.
As such, try encourage bundling by accounting for SACK in the size
of the first chunk fragment.
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>
Since an application may specify the maximum SCTP fragment size
that all data should be fragmented to, we need to fix how
we do segmentation. Right now, if a user specifies a small
fragment size, the segment size can go negative in the presence
of AUTH or COOKIE_ECHO bundling.
What we need to do is track the largest possbile DATA chunk that
can fit into the mtu. Then if the fragment size specified is
bigger then this maximum length, we'll shrink it down. Otherwise,
we just use the smaller segment size without changing it further.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
If a socket has a lot of association that are in the process of
of being closed/aborted, it is possible for a remote to establish
new associations during the time period that the old ones are shutting
down. If this was a result of a close() call, there will be no socket
and will cause a memory leak. We'll prevent this by setting the
socket state to CLOSING and disallow new associations when in this state.
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>
When the sctp transport is marked down, we can release the
cached route and force a new lookup when attempting to use
this transport for anything. This way, if a better route
or source address is available, we'll try to use it.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Update the route and saddr entries for the non-active transports as some
of the added addresses can be used as better source addresses, or may
be there is a better route.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
This patch fix to check the unrecognized ASCONF parameter before
access it.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
The return value of sctp_process_asconf_ack() may be
overwritten while process parameters with no error.
This patch fixed the problem.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Here is a patch which fixes an issue observed when using TCP over IPv6
and AH from IPsec.
When a connection gets closed the 4-way method and the last ACK from
the server gets dropped, the subsequent FINs from the client do not
get ACKed because tcp_v6_send_response does not set the transport
header pointer. This causes ah6_output to try to allocate a lot of
memory, which typically fails, so the ACKs never make it out of the
stack.
I have reproduced the problem on kernel 2.6.7, but after looking at
the latest kernel it seems the problem is still there.
Signed-off-by: Cosmin Ratiu <cratiu@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Its hard to tell if vlans are dropping frames, since
every frame given to vlan_???_start_xmit() functions
is accounted as fully transmitted by lower device.
We can test dev_queue_xmit() return values to
properly account for dropped frames.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove a debugging aid I accidently left in previous 'cleanup' patch
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
vlan_dev_hard_start_xmit() & vlan_dev_hwaccel_hard_start_xmit()
select txqueue number 0, instead of using index provided by
skb_get_queue_mapping().
This is not correct after commit 2e59af3dcb
[vlan: multiqueue vlan device] because
txq->tx_packets & txq->tx_bytes changes are performed on
a single location, and not the right locking.
Fix is to take the appropriate struct netdev_queue pointer
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch removes the call to dev_kfree_skb() when the atm device is busy.
Calling dev_kfree_skb() causes heavy packet loss then the device is under
heavy load, the more correct behavior should be to stop the upper layers,
then when the lower device can queue packets again wake the upper layers.
Signed-off-by: Karl Hiramoto <karl@hiramoto.org>
Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixed a lockdep warning which appeared when doing stress
memory tests over NFS:
inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.
page reclaim => nfs_writepage => tcp_sendmsg => lock sk_lock
mount_root => nfs_root_data => tcp_close => lock sk_lock =>
tcp_send_fin => alloc_skb_fclone => page reclaim
David raised a concern that if the allocation fails in tcp_send_fin(), and it's
GFP_ATOMIC, we are going to yield() (which sleeps) and loop endlessly waiting
for the allocation to succeed.
But fact is, the original GFP_KERNEL also sleeps. GFP_ATOMIC+yield() looks
weird, but it is no worse the implicit sleep inside GFP_KERNEL. Both could
loop endlessly under memory pressure.
CC: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
CC: David S. Miller <davem@davemloft.net>
CC: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds support to flash a firmware image to a device using ethtool.
The driver gets the filename of the firmware image and flashes the image
using the request firmware path.
The region "on the chip" to be flashed can be specified by an option.
It is upto the device driver to enumerate the region number passed by ethtool,
to the region to be flashed.
The default behavior is to flash all the regions on the chip.
Signed-off-by: Ajit Khaparde <ajitk@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.
IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.
This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.
Example after an UDP tx flood
# netstat -s
...
IP:
1487048 outgoing packets dropped
...
Udp:
...
SndbufErrors: 1487048
send() syscalls, do however still return an OK status, to not
break applications.
Note : send() manual page explicitly says for -ENOBUFS error :
"The output queue for a network interface was full.
This generally indicates that the interface has stopped sending,
but may be caused by transient congestion.
(Normally, this does not occur in Linux. Packets are just silently
dropped when a device queue overflows.) "
This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.
Many thanks to Christoph, David, and last but not least, Alexey !
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
vlan devices are currently not multi-queue capable.
We can do that with a new rtnl_link_ops method,
get_tx_queues(), called from rtnl_create_link()
This new method gets num_tx_queues/real_num_tx_queues
from real device.
register_vlan_device() is also handled.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It was recently pointed out to me that the last_rx field of the
net_device structure wasn't updated regularly. In fact only the
bonding driver really uses it currently. Since the drop_monitor code
relies on the last_rx field to detect drops on recevie in hardware, We
need to find a more reliable way to rate limit our drop checks (so
that we don't check for drops on every frame recevied, which would be
inefficient. This patch makes a last_rx timestamp that is private to
the drop monitor code and is updated for every device that we track.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can just display this upon enabling mac80211 with an
'if MAC80211 != n' check.
Cc: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Refer to the wireless wiki for more information.
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
All instances of file_operations should be const.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function block inet_connect_sock_af_ops contains no data
make it constant.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
qdisc drops should be notified to IP_RECVERR enabled sockets, as done in IPV4.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The net_dev of backlog napi is NULL, like below:
__get_cpu_var(softnet_data).backlog.dev == NULL
So, we should check it in napi tracepoint's probe function
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
These are full of unresolved problems, mainly that conversions don't
work 1-1 from hrtimers to tasklet_hrtimers because unlike hrtimers
tasklets can't be killed from softirq context.
And when a qdisc gets reset, that's exactly what we need to do here.
We'll work this out in the net-next-2.6 tree and if warranted we'll
backport that work to -stable.
This reverts the following 3 changesets:
a2cb6a4dd4
("pkt_sched: Fix bogon in tasklet_hrtimer changes.")
38acce2d79
("pkt_sched: Convert CBQ to tasklet_hrtimer.")
ee5f9757ea
("pkt_sched: Convert qdisc_watchdog to tasklet_hrtimer")
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit 2b85a34e91
(net: No more expensive sock_hold()/sock_put() on each tx)
sk_free() frees socks conditionally and depends
on sk_wmem_alloc being set e.g. in sock_init_data(). But in some
cases sk_free() is called earlier, usually after other alloc errors.
Fix is to move sk_wmem_alloc initialization from sock_init_data()
to sk_alloc() itself.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
These tables are never modified at runtime. Move to read-only
section.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct net::ipv6.ip6_dst_ops is separatedly dynamically allocated,
but there is no fundamental reason for it. Embed it directly into
struct netns_ipv6.
For that:
* move struct dst_ops into separate header to fix circular dependencies
I honestly tried not to, it's pretty impossible to do other way
* drop dynamical allocation, allocate together with netns
For a change, remove struct dst_ops::dst_net, it's deducible
by using container_of() given dst_ops pointer.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
RFC 1122 specifies two threshold values R1 and R2 for connection timeouts,
which may represent a number of allowed retransmissions or a timeout value.
Currently linux uses sysctl_tcp_retries{1,2} to specify the thresholds
in number of allowed retransmissions.
For any desired threshold R2 (by means of time) one can specify tcp_retries2
(by means of number of retransmissions) such that TCP will not time out
earlier than R2. This is the case, because the RTO schedule follows a fixed
pattern, namely exponential backoff.
However, the RTO behaviour is not predictable any more if RTO backoffs can be
reverted, as it is the case in the draft
"Make TCP more Robust to Long Connectivity Disruptions"
(http://tools.ietf.org/html/draft-zimmermann-tcp-lcd).
In the worst case TCP would time out a connection after 3.2 seconds, if the
initial RTO equaled MIN_RTO and each backoff has been reverted.
This patch introduces a function retransmits_timed_out(N),
which calculates the timeout of a TCP connection, assuming an initial
RTO of MIN_RTO and N unsuccessful, exponentially backed-off retransmissions.
Whenever timeout decisions are made by comparing the retransmission counter
to some value N, this function can be used, instead.
The meaning of tcp_retries2 will be changed, as many more RTO retransmissions
can occur than the value indicates. However, it yields a timeout which is
similar to the one of an unpatched, exponentially backing off TCP in the same
scenario. As no application could rely on an RTO greater than MIN_RTO, there
should be no risk of a regression.
Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Here, an ICMP host/network unreachable message, whose payload fits to
TCP's SND.UNA, is taken as an indication that the RTO retransmission has
not been lost due to congestion, but because of a route failure
somewhere along the path.
With true congestion, a router won't trigger such a message and the
patched TCP will operate as standard TCP.
This patch reverts one RTO backoff, if an ICMP host/network unreachable
message, whose payload fits to TCP's SND.UNA, arrives.
Based on the new RTO, the retransmission timer is reset to reflect the
remaining time, or - if the revert clocked out the timer - a retransmission
is sent out immediately.
Backoffs are only reverted, if TCP is in RTO loss recovery, i.e. if
there have been retransmissions and reversible backoffs, already.
Changes from v2:
1) Renaming of skb in tcp_v4_err() moved to another patch.
2) Reintroduced tcp_bound_rto() and __tcp_set_rto().
3) Fixed code comments.
Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
This supplementary patch renames skb to icmp_skb in tcp_v4_err() in order to
disambiguate from another sk_buff variable, which will be introduced
in a separate patch.
Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implements the dcbnl netlink setapp/getapp pair. When a setapp/getapp
is received, dcbnl would just pass on to dcbnl_rtnl_op.setapp/getapp
that are supposed to be implemented by the low level drivers.
Signed-off-by: Yi Zou <yi.zou@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vlad Yasevich