Al Viro [Tue, 19 Sep 2006 20:23:19 +0000 (13:23 -0700)]
[IPV4]: ipip and ip_gre encapsulation bugs
Handling of ipip and ip_gre ICMP error relaying is b0rken; it accesses
8bit field + 3 reserved octets as host-endian 32bit, does comparison,
subtraction and stuffs the result back. That breaks on big-endian.
Fixed, made endian-clean.
[ Note that this effected code is permanently commented out with
and ifdef, so this error couldn't actually cause problems for
anyone. -DaveM ]
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:15:33 +0000 (13:15 -0700)]
[DCCP] CCID2: Add helper functions for changing important CCID2 state
Introduce methods which manipulate interesting congestion control
state such as pipe and rtt estimate. This is useful for people
wishing to monitor the variables of CCID and instrument the code
[perhaps using Kprobes]. Personally, I am a fan of
encapsulation---that justifies this change =D.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:14:43 +0000 (13:14 -0700)]
[DCCP] CCID2: Halve cwnd once upon multiple losses in a single RTT
When multiple losses occur in one RTT, the window should be halved
only once [a single "congestion event"]. This is now implemented,
although not perfectly. Slightly changed the interface for changing
the cwnd: pass hctx instead of dp. This is required in order to allow
for change_cwnd to be called from _init().
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:13:37 +0000 (13:13 -0700)]
[DCCP] CCID2: Allocate seq records on demand
Allocate more sequence state on demand. Each time a packet is sent
out by CCID2, a record of it needs to be kept. This list of records
grows proportionally to cwnd. Previously, the length of this list was
hardcored and therefore the cwnd could only grow to this value (of
128). Now, records are allocated on demand as necessary---cwnd may
grow as it wishes. The exceptional case of when memory is not
available is not handled gracefully. Perhaps, cwnd should be capped
at that point.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:12:44 +0000 (13:12 -0700)]
[DCCP] CCID2: Add Kconfig option for CCID2 debug
Allow the user to choose whether or not to enable CCID2 debugging via
Kconfig.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:10:11 +0000 (13:10 -0700)]
[DCCP] CCID2: Tell DCCP to quickly check whether cwnd is available
If not enough cwnd is available, tell the sender to check again as
soon as possible. This will increase CPU utilization (polling
frequently for cwnd) but will improve network performance. That is,
the sender will need to wait less before detecting the increase of
cwnd. A better architecture would be for the CCID to call-back (or
dequeue) from DCCP when it is able to transmit traffic -- not the
other way around as it currently occurs.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:07:20 +0000 (13:07 -0700)]
[DCCP] CCID2: Initialize ssthresh to infinity
Initialize the slow-start threshold to infinity. This way, upon connection
initiation, slow-start will be exited only upon a packet loss. This patch will
allow connections to quickly gain speed.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:06:46 +0000 (13:06 -0700)]
[DCCP] CCID2: Fix jiffie wrap issues
Jiffies are now handled correctly (I hope) in CCID2. If they wrap, no
problem.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:06:16 +0000 (13:06 -0700)]
[DCCP] ackvec: Remove unused variables
Get rid of unused variables in ackvector state.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:05:35 +0000 (13:05 -0700)]
[DCCP] ackvec: Fix how DCCP_ACKVEC_STATE_NOT_RECEIVED is used
Fix the way state is masked out. DCCP_ACKVEC_STATE_NOT_RECEIVED is
defined as appears in the packet, therefore bit shifting is not
required. This fix allows CCID2 to correctly detect losses.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrea Bittau [Tue, 19 Sep 2006 20:04:54 +0000 (13:04 -0700)]
[DCCP] ACKVEC: fix ackvector length calculation
Fix ackvector length calculation upon receiving an "ack-of-ack". This
patch avoids the ackvector from growing too large which causes it to
not be inserted into packets.
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Tue, 19 Sep 2006 19:57:34 +0000 (12:57 -0700)]
[XFRM]: Fix wildcard as tunnel source
Hashing SAs by source address breaks templates with wildcards as tunnel
source since the source address used for hashing/lookup is still 0/0.
Move source address lookup to xfrm_tmpl_resolve_one() so we can use the
real address in the lookup.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
It does not affect either mss-sized connections (obviously) or
connections controlled by Nagle (because there is only one small
segment in flight).
The idea is to record the fact that a small segment arrives on a
connection, where one small segment has already been received and
still not-ACKed. In this case ACK is forced after tcp_recvmsg() drains
receive buffer.
In other words, it is a "soft" each-2nd-segment ACK, which is enough
to preserve ACK clock even when ABC is enabled.
Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
Adrian Bunk [Mon, 18 Sep 2006 07:40:38 +0000 (00:40 -0700)]
[SCTP]: Cleanups
This patch contains the following cleanups:
- make the following needlessly global function static:
- socket.c: sctp_apply_peer_addr_params()
- add proper prototypes for the several global functions in
include/net/sctp/sctp.h
Note that this fixes wrong prototypes for the following functions:
- sctp_snmp_proc_exit()
- sctp_eps_proc_exit()
- sctp_assocs_proc_exit()
The latter was spotted by the GNU C compiler and reported
by David Woodhouse.
Signed-off-by: Adrian Bunk <bunk@stusta.de> Acked-by: Sridhar Samudrala <sri@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Thomas Graf [Mon, 18 Sep 2006 07:13:46 +0000 (00:13 -0700)]
[IPV6] address: Support NLM_F_EXCL when adding addresses
iproute2 doesn't provide the NLM_F_CREATE flag when adding addresses,
it is assumed to be implied. The existing code issues a check on
said flag when the modify operation fails (likely due to ENOENT)
before continueing to create it, this leads to a hard to predict
result, therefore the NLM_F_CREATE check is removed.
Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Thomas Graf [Mon, 18 Sep 2006 07:01:59 +0000 (00:01 -0700)]
[GENL]: Provide more information to userspace about registered genl families
Additionaly exports the following information when providing
the list of registered generic netlink families:
- protocol version
- header size
- maximum number of attributes
- list of available operations including
- id
- flags
- avaiability of policy and doit/dumpit function
libnl HEAD provides a utility to read this new information:
0x0010 nlctrl version 1
hdrsize 0 maxattr 6
op GETFAMILY (0x03) [POLICY,DOIT,DUMPIT]
0x0011 NLBL_MGMT version 1
hdrsize 0 maxattr 0
op unknown (0x02) [DOIT]
op unknown (0x03) [DOIT]
....
Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 14 Sep 2006 03:35:36 +0000 (20:35 -0700)]
[RTNETLINK]: Fix netdevice name corruption
When changing a device by ifindex without including a IFLA_IFNAME
attribute, the ifname variable contains random garbage and is used
to change the device name.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
James Morris [Fri, 1 Sep 2006 07:32:12 +0000 (00:32 -0700)]
[XFRM]: remove xerr_idxp from __xfrm_policy_check()
It seems that during the MIPv6 respin, some code which was originally
conditionally compiled around CONFIG_XFRM_ADVANCED was accidently left
in after the config option was removed.
This patch removes an extraneous pointer (xerr_idxp) which is no
longer needed.
Signed-off-by: James Morris <jmorris@namei.org> Acked-by: Masahide NAKAMURA <nakam@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
[IPSEC]: output mode to take an xfrm state as input param
Expose IPSEC modes output path to take an xfrm state as input param.
This makes it consistent with the input mode processing (which already
takes the xfrm state as a param).
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
Dmitry Mishin [Thu, 31 Aug 2006 22:28:39 +0000 (15:28 -0700)]
[NET]: Fix sk->sk_filter field access
Function sk_filter() is called from tcp_v{4,6}_rcv() functions with arg
needlock = 0, while socket is not locked at that moment. In order to avoid
this and similar issues in the future, use rcu for sk->sk_filter field read
protection.
Herbert Xu [Thu, 31 Aug 2006 22:11:02 +0000 (15:11 -0700)]
[TCP]: Fix rcv mss estimate for LRO
By passing a Linux-generated TSO packet straight back into Linux, Xen
becomes our first LRO user :) Unfortunately, there is at least one spot
in our stack that needs to be changed to cope with this.
The receive MSS estimate is computed from the raw packet size. This is
broken if the packet is GSO/LRO. Fortunately the real MSS can be found
in gso_size so we simply need to use that if it is non-zero.
Real LRO NICs should of course set the gso_size field in future.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
Ian McDonald [Wed, 30 Aug 2006 00:50:19 +0000 (17:50 -0700)]
[DCCP]: Tidyup CCID3 list handling
As Arnaldo Carvalho de Melo points out I should be using list_entry in case
the structure changes in future. Current code functions but is reliant
on position and requires type cast.
Noticed when doing this that I have one more variable than I needed so
removing that also.
Signed off by: Ian McDonald <ian.mcdonald@jandi.co.nz> Signed-off-by: David S. Miller <davem@davemloft.net>
If CONFIG_NETFILTER_DEBUG is enabled, it shouldn't change the
actions of the filtering. The message about skb->dst being NULL
is commonly triggered by dhclient, so it is useless. Make sure all
messages end in newline.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Do some simple optimization on the nf_bridge_pad() function
and don't use magic constants. Eliminate a double call and
the #ifdef'd code for CONFIG_BRIDGE_NETFILTER.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
[NETFILTER] bridge: code rearrangement for clarity
Cleanup and rearrangement for better style and clarity:
Split the function nf_bridge_maybe_copy_header into two pieces
Move copy portion out of line.
Use Ethernet header size macros.
Use header file to handle CONFIG_NETFILTER_BRIDGE differences
Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Benoit Boissinot [Tue, 29 Aug 2006 00:50:37 +0000 (17:50 -0700)]
[NETFILTER]: xt_CONNMARK.c build fix
net/netfilter/xt_CONNMARK.c: In function 'target':
net/netfilter/xt_CONNMARK.c:59: warning: implicit declaration of
function 'nf_conntrack_event_cache'
The warning is due to the following .config:
CONFIG_IP_NF_CONNTRACK=m
CONFIG_IP_NF_CONNTRACK_MARK=y
# CONFIG_IP_NF_CONNTRACK_EVENTS is not set
CONFIG_IP_NF_CONNTRACK_NETLINK=m
Proposed solution (based on the define in
include/net/netfilter/nf_conntrack_compat.h:
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org> Acked-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Thomas Graf [Sun, 27 Aug 2006 03:11:47 +0000 (20:11 -0700)]
[NETLINK]: Improve string attribute validation
Introduces a new attribute type NLA_NUL_STRING to support NUL
terminated strings. Attributes of this kind require to carry
a terminating NUL within the maximum specified in the policy.
The `old' NLA_STRING which is not required to be NUL terminated
is extended to provide means to specify a maximum length of the
string.
Aims at easing the pain with using nla_strlcpy() on temporary
buffers.
Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Gerrit Renker [Sun, 27 Aug 2006 03:06:05 +0000 (20:06 -0700)]
[UDP]: Unify UDPv4 and UDPv6 ->get_port()
This patch creates one common function which is called by
udp_v4_get_port() and udp_v6_get_port(). As a result,
* duplicated code is removed
* udp_port_rover and local port lookup can now be removed from udp.h
* further savings follow since the same function will be used by UDP-Litev4
and UDP-Litev6
In contrast to the patch sent in response to Yoshifujis comments
(fixed by this variant), the code below also removes the
EXPORT_SYMBOL(udp_port_rover), since udp_port_rover can now remain
local to net/ipv4/udp.c.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg [Sun, 27 Aug 2006 02:17:53 +0000 (19:17 -0700)]
[NETLINK]: remove third bogus argument from NLA_PUT_FLAG
This patch removes the 'value' argument from NLA_PUT_FLAG which is
unused anyway. The documentation comment was already correct so it
doesn't need an update :)
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Sat, 26 Aug 2006 23:50:20 +0000 (16:50 -0700)]
[IPV6]: Fix routing by fwmark
Fix mark comparison, also dump the mask to userspace when the mask is
zero, but the mark is not (in which case the mark is dumped, so the
mask is needed to make sense of it).
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Fri, 25 Aug 2006 23:11:42 +0000 (16:11 -0700)]
[NET_SCHED]: Add mask support to fwmark classifier
Support masking the nfmark value before the search. The mask value is
global for all filters contained in one instance. It can only be set
when a new instance is created, all filters must specify the same mask.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
- Remove redundant code. Pointed out by Brian Haley <brian.haley@hp.com>.
- Unify code paths with/without CONFIG_IPV6_MIP.
- Use NIP6_FMT for IPv6 address textual presentation.
- Fold long line. Pointed out by David Miller <davem@davemloft.net>.
David S. Miller [Thu, 24 Aug 2006 11:45:07 +0000 (04:45 -0700)]
[XFRM]: Hash policies when non-prefixed.
This idea is from Alexey Kuznetsov.
It is common for policies to be non-prefixed. And for
that case we can optimize lookups, insert, etc. quite
a bit.
For each direction, we have a dynamically sized policy
hash table for non-prefixed policies. We also have a
hash table on policy->index.
For prefixed policies, we have a list per-direction which
we will consult on lookups when a non-prefix hashtable
lookup fails.
This still isn't as efficient as I would like it. There
are four immediate problems:
1) Lots of excessive refcounting, which can be fixed just
like xfrm_state was
2) We do 2 hash probes on insert, one to look for dups and
one to allocate a unique policy->index. Althought I wonder
how much this matters since xfrm_state inserts do up to
3 hash probes and that seems to perform fine.
3) xfrm_policy_insert() is very complex because of the priority
ordering and entry replacement logic.
4) Lots of counter bumping, in addition to policy refcounts,
in the form of xfrm_policy_count[]. This is merely used
to let code path(s) know that some IPSEC rules exist. So
this count is indexed per-direction, maybe that is overkill.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 24 Aug 2006 10:54:22 +0000 (03:54 -0700)]
[XFRM]: Kill excessive refcounting of xfrm_state objects.
The refcounting done for timers and hash table insertions
are just wasted cycles. We can eliminate all of this
refcounting because:
1) The implicit refcount when the xfrm_state object is active
will always be held while the object is in the hash tables.
We never kfree() the xfrm_state until long after we've made
sure that it has been unhashed.
2) Timers are even easier. Once we mark that x->km.state as
anything other than XFRM_STATE_VALID (__xfrm_state_delete
sets it to XFRM_STATE_DEAD), any timer that fires will
do nothing and return without rearming the timer.
Therefore we can defer the del_timer calls until when the
object is about to be freed up during GC. We have to use
del_timer_sync() and defer it to GC because we can't do
a del_timer_sync() while holding x->lock which all callers
of __xfrm_state_delete hold.
This makes SA changes even more light-weight.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 24 Aug 2006 10:30:28 +0000 (03:30 -0700)]
[XFRM]: Purge dst references to deleted SAs passively.
Just let GC and other normal mechanisms take care of getting
rid of DST cache references to deleted xfrm_state objects
instead of walking all the policy bundles.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 24 Aug 2006 10:18:09 +0000 (03:18 -0700)]
[XFRM]: Add generation count to xfrm_state and xfrm_dst.
Each xfrm_state inserted gets a new generation counter
value. When a bundle is created, the xfrm_dst objects
get the current generation counter of the xfrm_state
they will attach to at dst->xfrm.
xfrm_bundle_ok() will return false if it sees an
xfrm_dst with a generation count different from the
generation count of the xfrm_state that dst points to.
This provides a facility by which to passively and
cheaply invalidate cached IPSEC routes during SA
database changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 24 Aug 2006 10:08:07 +0000 (03:08 -0700)]
[XFRM]: Dynamic xfrm_state hash table sizing.
The grow algorithm is simple, we grow if:
1) we see a hash chain collision at insert, and
2) we haven't hit the hash size limit (currently 1*1024*1024 slots), and
3) the number of xfrm_state objects is > the current hash mask
All of this needs some tweaking.
Remove __initdata from "hashdist" so we can use it safely at run time.
Signed-off-by: David S. Miller <davem@davemloft.net>
projects / linux-beck.git / log