Commit Graph

7896 Commits

Author SHA1 Message Date
Linus Torvalds
b4b8f57965 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  [TCP]: Add return value indication to tcp_prune_ofo_queue().
  PS3: gelic: fix the oops on the broken IE returned from the hypervisor
  b43legacy: fix DMA mapping leakage
  mac80211: remove message on receiving unexpected unencrypted frames
  Update rt2x00 MAINTAINERS entry
  Add rfkill to MAINTAINERS file
  rfkill: Fix device type check when toggling states
  b43legacy: Fix usage of struct device used for DMAing
  ssb: Fix usage of struct device used for DMAing
  MAINTAINERS: move to generic repository for iwlwifi
  b43legacy: fix initvals loading on bcm4303
  rtl8187: Add missing priv->vif assignments
  netconsole: only set CON_PRINTBUFFER if the user specifies a netconsole
  [CAN]: Update documentation of struct sockaddr_can
  MAINTAINERS: isdn4linux@listserv.isdn4linux.de is subscribers-only
  [TCP]: Fix never pruned tcp out-of-order queue.
  [NET_SCHED] sch_api: fix qdisc_tree_decrease_qlen() loop
2008-04-16 07:44:27 -07:00
Paul Bolle
424b00e2c0 AFS: Do not describe debug parameters with their value
Describe debug parameters with their names (and not their values).

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-16 07:43:48 -07:00
Vitaliy Gusev
56f367bbfd [TCP]: Add return value indication to tcp_prune_ofo_queue().
Returns non-zero if tp->out_of_order_queue was seen non-empty.
This allows tcp_try_rmem_schedule() to return early.

Signed-off-by: Vitaliy Gusev <vgusev@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-15 20:26:34 -07:00
Johannes Berg
b3fc9c6c58 mac80211: remove message on receiving unexpected unencrypted frames
Some people are getting this message a lot, and we have traced it to
broken access points that much too often send completely empty frames
(all bytes zeroed, which they shouldn't do at all.)

Since we cannot do anything about such frames in any case except the
special case where we're debugging an AP, just remove the message.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-04-15 15:04:36 -04:00
Carlos Corbacho
89796f64a2 rfkill: Fix device type check when toggling states
rfkill_switch_all() is supposed to only switch all the interfaces of a
given type, but does not actually do this; instead, it just switches
everything currently in the same state.

Add the necessary type check in.

(This fixes a bug I've been seeing while developing an rfkill laptop
driver, with both bluetooth and wireless simultaneously changing state
after only pressing either KEY_WLAN or KEY_BLUETOOTH).

Signed-off-by: Carlos Corbacho <carlos@strangeworlds.co.uk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-04-15 15:04:35 -04:00
Vitaliy Gusev
b000cd3707 [TCP]: Fix never pruned tcp out-of-order queue.
tcp_prune_queue() doesn't prune an out-of-order queue at all.
Therefore sk_rmem_schedule() can fail but the out-of-order queue isn't
pruned . This can lead to tcp deadlock state if the next two
conditions are held:

1. There are a sequence hole between last received in
   order segment and segments enqueued to the out-of-order queue.

2. Size of all segments in the out-of-order queue is more than tcp_mem[2].

Signed-off-by: Vitaliy Gusev <vgusev@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-15 00:33:38 -07:00
Jarek Poplawski
066a3b5b23 [NET_SCHED] sch_api: fix qdisc_tree_decrease_qlen() loop
TC_H_MAJ(parentid) for root classes is the same as for ingress, and if
ingress qdisc is created qdisc_lookup() returns its pointer (without
ingress NULL is returned). After this all qdisc_lookups give the same,
and we get endless loop. (I don't know how this could hide for so long
- it should trigger with every leaf class deleted if it's qdisc isn't
empty.)

After this fix qdisc_lookup() is omitted both for ingress and root
parents, but looking for root is only wasting a little time here...
Many thanks to Enrico Demarin for finding a test for catching this
bug, which probably bothered quite a lot of admins.
 
Reported-by: Enrico Demarin <enrico@superclick.com>,
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-14 15:10:42 -07:00
Linus Torvalds
533bb8a4d7 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (31 commits)
  [BRIDGE]: Fix crash in __ip_route_output_key with bridge netfilter
  [NETFILTER]: ipt_CLUSTERIP: fix race between clusterip_config_find_get and _entry_put
  [IPV6] ADDRCONF: Don't generate temporary address for ip6-ip6 interface.
  [IPV6] ADDRCONF: Ensure disabling multicast RS even if privacy extensions are disabled.
  [IPV6]: Use appropriate sock tclass setting for routing lookup.
  [IPV6]: IPv6 extension header structures need to be packed.
  [IPV6]: Fix ipv6 address fetching in raw6_icmp_error().
  [NET]: Return more appropriate error from eth_validate_addr().
  [ISDN]: Do not validate ISDN net device address prior to interface-up
  [NET]: Fix kernel-doc for skb_segment
  [SOCK] sk_stamp: should be initialized to ktime_set(-1L, 0)
  net: check for underlength tap writes
  net: make struct tun_struct private to tun.c
  [SCTP]: IPv4 vs IPv6 addresses mess in sctp_inet[6]addr_event.
  [SCTP]: Fix compiler warning about const qualifiers
  [SCTP]: Fix protocol violation when receiving an error lenght INIT-ACK
  [SCTP]: Add check for hmac_algo parameter in sctp_verify_param()
  [NET_SCHED] cls_u32: refcounting fix for u32_delete()
  [DCCP]: Fix skb->cb conflicts with IP
  [AX25]: Potential ax25_uid_assoc-s leaks on module unload.
  ...
2008-04-14 07:56:24 -07:00
Patrick McHardy
159d83363b [BRIDGE]: Fix crash in __ip_route_output_key with bridge netfilter
The bridge netfilter code attaches a fake dst_entry with a pointer to a
fake net_device structure to skbs it passes up to IPv4 netfilter. This
leads to crashes when the skb is passed to __ip_route_output_key when
dereferencing the namespace pointer.

Since bridging can currently only operate in the init_net namespace,
the easiest fix for now is to initialize the nd_net pointer of the
fake net_device struct to &init_net.

Should fix bugzilla 10323: http://bugzilla.kernel.org/show_bug.cgi?id=10323

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-14 00:46:01 -07:00
Pavel Emelyanov
4dee959723 [NETFILTER]: ipt_CLUSTERIP: fix race between clusterip_config_find_get and _entry_put
Consider we are putting a clusterip_config entry with the "entries"
count == 1, and on the other CPU there's a clusterip_config_find_get
in progress:

CPU1:							CPU2:
clusterip_config_entry_put:				clusterip_config_find_get:
if (atomic_dec_and_test(&c->entries)) {
	/* true */
							read_lock_bh(&clusterip_lock);
							c = __clusterip_config_find(clusterip);
							/* found - it's still in list */
							...
							atomic_inc(&c->entries);
							read_unlock_bh(&clusterip_lock);

	write_lock_bh(&clusterip_lock);
	list_del(&c->list);
	write_unlock_bh(&clusterip_lock);
	...
	dev_put(c->dev);

Oops! We have an entry returned by the clusterip_config_find_get,
which is a) not in list b) has a stale dev pointer.

The problems will happen when the CPU2 will release the entry - it
will remove it from the list for the 2nd time, thus spoiling it, and
will put a stale dev pointer.

The fix is to make atomic_dec_and_test under the clusterip_lock.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2008-04-14 00:44:52 -07:00
YOSHIFUJI Hideaki
9625ed72e8 [IPV6] ADDRCONF: Don't generate temporary address for ip6-ip6 interface.
As far as I can remember, I was going to disable privacy extensions
on all "tunnel" interfaces.  Disable it on ip6-ip6 interface as well.

Also, just remove ifdefs for SIT for simplicity.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 23:47:11 -07:00
YOSHIFUJI Hideaki
b077d7abab [IPV6] ADDRCONF: Ensure disabling multicast RS even if privacy extensions are disabled.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 23:42:18 -07:00
David S. Miller
b45e9189c0 [IPV6]: Fix ipv6 address fetching in raw6_icmp_error().
Fixes kernel bugzilla 10437

Based almost entirely upon a patch by Dmitry Butskoy.

When deciding what raw sockets to deliver the ICMPv6
to, we should use the addresses in the ICMPv6 quoted
IPV6 header, not the top-level one.

Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 23:14:15 -07:00
Patrick McHardy
2ed9926e16 [NET]: Return more appropriate error from eth_validate_addr().
Paul Bolle wrote:
> http://bugzilla.kernel.org/show_bug.cgi?id=9923 would have been much easier to
> track down if eth_validate_addr() would somehow complain aloud if an address 
> is invalid. Shouldn't it make at least some noise?

I guess it should return -EADDRNOTAVAIL similar to eth_mac_addr()
when validation fails.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 22:45:40 -07:00
Ben Hutchings
4c821d753d [NET]: Fix kernel-doc for skb_segment
The kernel-doc comment for skb_segment is clearly wrong.  This states
what it actually does.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 21:52:48 -07:00
Eric Dumazet
f37f0afb29 [SOCK] sk_stamp: should be initialized to ktime_set(-1L, 0)
Problem spotted by Andrew Brampton

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 21:39:26 -07:00
Pavel Emelyanov
a40a7d15ba [SCTP]: IPv4 vs IPv6 addresses mess in sctp_inet[6]addr_event.
All IP addresses that are present in a system are duplicated on
struct sctp_sockaddr_entry. They are linked in the global list
called sctp_local_addr_list. And this struct unions IPv4 and IPv6
addresses.

So, there can be rare case, when a sockaddr_in.sin_addr coincides
with the corresponding part of the sockaddr_in6 and the notifier
for IPv4 will carry away an IPv6 entry.

The fix is to check the family before comparing the addresses.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:40:38 -07:00
Vlad Yasevich
ab38fb04c9 [SCTP]: Fix compiler warning about const qualifiers
Fix 3 warnings about discarding const qualifiers:

net/sctp/ulpevent.c:862: warning: passing argument 1 of 'sctp_event2skb' discards qualifiers from pointer target type
net/sctp/sm_statefuns.c:4393: warning: passing argument 1 of 'SCTP_ASOC' discards qualifiers from pointer target type
net/sctp/socket.c:5874: warning: passing argument 1 of 'cmsg_nxthdr' discards qualifiers from pointer target type

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:40:06 -07:00
Gui Jianfeng
f4ad85ca3e [SCTP]: Fix protocol violation when receiving an error lenght INIT-ACK
When receiving an error length INIT-ACK during COOKIE-WAIT,
a 0-vtag ABORT will be responsed. This action violates the
protocol apparently. This patch achieves the following things.
1 If the INIT-ACK contains all the fixed parameters, use init-tag
  recorded from INIT-ACK as vtag.
2 If the INIT-ACK doesn't contain all the fixed parameters,
  just reflect its vtag.

Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:39:34 -07:00
Wei Yongjun
72da7b3860 [SCTP]: Add check for hmac_algo parameter in sctp_verify_param()
RFC 4890 has the following text:

  The HMAC algorithm based on SHA-1 MUST be supported and
  included in the HMAC-ALGO parameter.

As a result, we need to check in sctp_verify_param() that HMAC_SHA1 is
present in the list.  If not, we should probably treat this as a
protocol violation.

It should also be a protocol violation if the HMAC parameter is empty.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:39:19 -07:00
Jarek Poplawski
e56cfad132 [NET_SCHED] cls_u32: refcounting fix for u32_delete()
Deleting of nonroot hnodes mostly doesn't work in u32_delete():
refcnt == 1 is expected, but such hnodes' refcnts are initialized
with 0 and charged only with "link" nodes. Now they'll start with
1 like usual. Thanks to Patrick McHardy for an improving suggestion.

Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:37:13 -07:00
Patrick McHardy
028b027524 [DCCP]: Fix skb->cb conflicts with IP
dev_queue_xmit() and the other IP output functions expect to get a skb
with clear or properly initialized skb->cb. Unlike TCP and UDP, the
dccp_skb_cb doesn't contain a struct inet_skb_parm at the beginning,
so the DCCP-specific data is interpreted by the IP output functions.
This can cause false negatives for the conditional POST_ROUTING hook
invocation, making the packet bypass the hook.

Add a inet_skb_parm/inet6_skb_parm union to the beginning of
dccp_skb_cb to avoid clashes. Also add a BUILD_BUG_ON to make
sure it fits in the cb.

[ Combined with patch from Gerrit Renker to remove two now unnecessary
  memsets of IPCB(skb)->opt ]

Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:35:41 -07:00
Pavel Emelyanov
ae1b6a31b1 [AX25]: Potential ax25_uid_assoc-s leaks on module unload.
The ax25_uid_free call walks the ax25_uid_list and releases entries
from it. The problem is that after the fisrt call to hlist_del_init
the hlist_for_each_entry (which hides behind the ax25_uid_for_each)
will consider the current position to be the last and will return.

Thus, the whole list will be left not freed.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-12 18:33:50 -07:00
Linus Torvalds
14897e35fd Merge branch 'docs' of git://git.lwn.net/linux-2.6
* 'docs' of git://git.lwn.net/linux-2.6:
  Add additional examples in Documentation/spinlocks.txt
  Move sched-rt-group.txt to scheduler/
  Documentation: move rpc-cache.txt to filesystems/
  Documentation: move nfsroot.txt to filesystems/
  Spell out behavior of atomic_dec_and_lock() in kerneldoc
  Fix a typo in highres.txt
  Fixes to the seq_file document
  Fill out information on patch tags in SubmittingPatches
  Add the seq_file documentation
2008-04-11 13:24:16 -07:00
J. Bruce Fields
6ded55da6b Documentation: move nfsroot.txt to filesystems/
Documentation/ is a little large, and filesystems/ seems an obvious
place for this file.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-04-11 13:18:01 -06:00
Linus Torvalds
90768c09bc Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  [NETNS][IPV6] tcp - assign the netns for timewait sockets
  [IPV4]: Fix byte value boundary check in do_ip_getsockopt().
  BNX2X: Correct bringing chip out of reset
  [NETFILTER]: nf_nat: autoload IPv4 connection tracking
  [NETFILTER]: xt_hashlimit: fix mask calculation
  [XFRM]: xfrm_user: fix selector family initialization
  rt61pci: rt61pci_beacon_update do not free skb twice
  ssb-mipscore: Fix interrupt vectors
  ssb-pcicore: Fix IRQ TPS flag handling
  mac80211: use short_preamble mode from capability if ERP IE not present
  [NET]: Undo code bloat in hot paths due to print_mac().
  [TCP]: Don't allow FRTO to take place while MTU is being probed
  [TCP]: tcp_simple_retransmit can cause S+L
  [TCP]: Fix NewReno's fast rexmit/recovery problems with GSOed skb
  [TCP]: Restore 2.6.24 mark_head_lost behavior for newreno/fack
  nl80211: fix STA AID bug
  b43legacy: fix bcm4303 crash
  iwlwifi: fix n-band association problem
  ipw2200: set MAC address on radiotap interface
  libertas: fix mode initialization problem
2008-04-11 08:10:24 -07:00
Daniel Lezcano
7951f0b03a [NETNS][IPV6] tcp - assign the netns for timewait sockets
Copy the network namespace from the socket to the timewait socket.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Mark Lord <mlord@pobox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-10 20:53:10 -07:00
David S. Miller
951e07c930 [IPV4]: Fix byte value boundary check in do_ip_getsockopt().
This fixes kernel bugzilla 10371.

As reported by M.Piechaczek@osmosys.tv, if we try to grab a
char sized socket option value, as in:

  unsigned char ttl = 255;
  socklen_t     len = sizeof(ttl);
  setsockopt(socket, IPPROTO_IP, IP_MULTICAST_TTL, &ttl, &len);

  getsockopt(socket, IPPROTO_IP, IP_MULTICAST_TTL, &ttl, &len);

The ttl returned will be wrong on big-endian, and on both little-
endian and big-endian the next three bytes in userspace are written
with garbage.

It's because of this test in do_ip_getsockopt():

	if (len < sizeof(int) && len > 0 && val>=0 && val<255) {

It should allow a 'val' of 255 to pass here, but it doesn't so it
copies a full 'int' back to userspace.

On little-endian that will write the correct value into the location
but it spams on the next three bytes in userspace.  On big endian it
writes the wrong value into the location and spams the next three
bytes.

Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-10 01:29:36 -07:00
Jan Engelhardt
475959d477 [NETFILTER]: nf_nat: autoload IPv4 connection tracking
Without this patch, the generic L3 tracker would kick in
if nf_conntrack_ipv4 was not loaded before nf_nat, which
would lead to translation problems with ICMP errors.

NAT does not make sense without IPv4 connection tracking
anyway, so just add a call to need_ipv4_conntrack().

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-09 15:14:58 -07:00
Patrick McHardy
1b9b70ea2e [NETFILTER]: xt_hashlimit: fix mask calculation
Shifts larger than the data type are undefined, don't try to shift
an u32 by 32. Also remove some special-casing of bitmasks divisible
by 32.

Based on patch by Jan Engelhardt <jengelh@computergmbh.de>.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-09 15:14:18 -07:00
David S. Miller
b41f5bfff7 Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6 2008-04-09 15:10:14 -07:00
Patrick McHardy
bcf0dda8d2 [XFRM]: xfrm_user: fix selector family initialization
Commit df9dcb45 ([IPSEC]: Fix inter address family IPsec tunnel handling)
broke openswan by removing the selector initialization for tunnel mode
in case it is uninitialized.

This patch restores the initialization, fixing openswan, but probably
breaking inter-family tunnels again (unknown since the patch author
disappeared). The correct thing for inter-family tunnels is probably
to simply initialize the selector family explicitly.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-09 15:08:24 -07:00
Vladimir Koutny
216bce90b8 mac80211: use short_preamble mode from capability if ERP IE not present
When associating to a b-only AP where there is no ERP IE, short preamble
mode is left at previous state (probably also protection mode). In this
case, disable protection and use short preamble mode as specified in
capability field. The same is done if capability field is changed on-the-fly.

Signed-off-by: Vladimir Koutny <vlado@ksp.sk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-04-09 10:33:49 -04:00
Chuck Lever
ed13c27e54 SUNRPC: Fix a memory leak in rpc_create()
Commit 510deb0d was supposed to move the xprt_create_transport() call in
rpc_create(), but neglected to remove the old call site.  This resulted in
a transport leak after every rpc_create() call.

This leak is present in 2.6.24 and 2.6.25.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2008-04-08 21:07:00 -04:00
Trond Myklebust
daeba89d43 SUNRPC: don't call flush_dcache_page() with an invalid pointer
Fix a problem in _copy_to_pages(), whereby it may call flush_dcache_page()
with an invalid pointer due to the fact that 'pgto' gets incremented
beyond the end of the page array. Fix is to exit the loop without this
unnecessary increment of pgto.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2008-04-08 21:06:50 -04:00
David S. Miller
21f644f3ea [NET]: Undo code bloat in hot paths due to print_mac().
If print_mac() is used inside of a pr_debug() the compiler
can't see that the call is redundant so still performs it
even of pr_debug() ends up being a nop.

So don't use print_mac() in such cases in hot code paths,
use MAC_FMT et al. instead.

As noted by Joe Perches, pr_debug() could be modified to
handle this better, but that is a change to an interface
used by the entire kernel and thus needs to be validated
carefully.  This here is thus the less risky fix for
2.6.25

Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-08 16:50:44 -07:00
Ilpo Järvinen
6adb4f733e [TCP]: Don't allow FRTO to take place while MTU is being probed
MTU probe can cause some remedies for FRTO because the normal
packet ordering may be violated allowing FRTO to make a wrong
decision (it might not be that serious threat for anything
though). Thus it's safer to not run FRTO while MTU probe is
underway.

It seems that the basic FRTO variant should also look for an
skb at probe_seq.start to check if that's retransmitted one
but I didn't implement it now (plain seqno in window check
isn't robust against wraparounds).

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-07 22:33:57 -07:00
Ilpo Järvinen
882bebaaca [TCP]: tcp_simple_retransmit can cause S+L
This fixes Bugzilla #10384

tcp_simple_retransmit does L increment without any checking
whatsoever for overflowing S+L when Reno is in use.

The simplest scenario I can currently think of is rather
complex in practice (there might be some more straightforward
cases though). Ie., if mss is reduced during mtu probing, it
may end up marking everything lost and if some duplicate ACKs
arrived prior to that sacked_out will be non-zero as well,
leading to S+L > packets_out, tcp_clean_rtx_queue on the next
cumulative ACK or tcp_fastretrans_alert on the next duplicate
ACK will fix the S counter.

More straightforward (but questionable) solution would be to
just call tcp_reset_reno_sack() in tcp_simple_retransmit but
it would negatively impact the probe's retransmission, ie.,
the retransmissions would not occur if some duplicate ACKs
had arrived.

So I had to add reno sacked_out reseting to CA_Loss state
when the first cumulative ACK arrives (this stale sacked_out
might actually be the explanation for the reports of left_out
overflows in kernel prior to 2.6.23 and S+L overflow reports
of 2.6.24). However, this alone won't be enough to fix kernel
before 2.6.24 because it is building on top of the commit
1b6d427bb7 ([TCP]: Reduce sacked_out with reno when purging
write_queue) to keep the sacked_out from overflowing.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Reported-by: Alessandro Suardi <alessandro.suardi@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-07 22:33:07 -07:00
Ilpo Järvinen
c137f3dda0 [TCP]: Fix NewReno's fast rexmit/recovery problems with GSOed skb
Fixes a long-standing bug which makes NewReno recovery crippled.
With GSO the whole head skb was marked as LOST which is in
violation of NewReno procedure that only wants to mark one packet
and ended up breaking our TCP code by causing counter overflow
because our code was built on top of assumption about valid
NewReno procedure. This manifested as triggering a WARN_ON for
the overflow in a number of places.

It seems relatively safe alternative to just do nothing if
tcp_fragment fails due to oom because another duplicate ACK is
likely to be received soon and the fragmentation will be retried.

Special thanks goes to Soeren Sonnenburg <kernel@nn7.de> who was
lucky enough to be able to reproduce this so that the warning
for the overflow was hit. It's not as easy task as it seems even
if this bug happens quite often because the amount of outstanding
data is pretty significant for the mismarkings to lead to an
overflow.

Because it's very late in 2.6.25-rc cycle (if this even makes in
time), I didn't want to touch anything with SACK enabled here.
Fragmenting might be useful for it as well but it's more or less
a policy decision rather than mandatory fix. Thus there's no need
to rush and we can postpone considering tcp_fragment with SACK
for 2.6.26.

In 2.6.24 and earlier, this very same bug existed but the effect
is slightly different because of a small changes in the if
conditions that fit to the patch's context. With them nothing
got lost marker and thus no retransmissions happened.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-07 22:32:38 -07:00
Ilpo Järvinen
1b69d74539 [TCP]: Restore 2.6.24 mark_head_lost behavior for newreno/fack
The fast retransmission can be forced locally to the rfc3517
branch in tcp_update_scoreboard instead of making such fragile
constructs deeper in tcp_mark_head_lost.

This is necessary for the next patch which must not have
loopholes for cnt > packets check. As one can notice,
readability got some improvements too because of this :-).

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-07 22:31:38 -07:00
Johannes Berg
16f2e85d31 nl80211: fix STA AID bug
This fixes the STA AID setting and actually makes hostapd/mac80211
work properly in presence of power-saving stations.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-04-07 22:19:30 -04:00
Linus Torvalds
4cac04dd63 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  fix endian lossage in forcedeth
  net/tokenring/olympic.c section fixes
  net: marvell.c fix sparse shadowed variable warning
  [VLAN]: Fix egress priority mappings leak.
  [TG3]: Add PHY workaround for 5784
  [NET]: srandom32 fixes for networking v2
  [IPV6]: Fix refcounting for anycast dst entries.
  [IPV6]: inet6_dev on loopback should be kept until namespace stop.
  [IPV6]: Event type in addrconf_ifdown is mis-used.
  [ICMP]: Ensure that ICMP relookup maintains status quo
2008-04-07 08:36:57 -07:00
Pavel Emelyanov
23556323b2 [VLAN]: Fix egress priority mappings leak.
These entries are allocated in vlan_dev_set_egress_priority, 
but are never released and leaks on vlan device removal.

Drop these in vlan's ->uninit callback - after the device is 
brought down and everyone is notified about it is going to
be unregistered.

Found during testing vlan netnsization patchset.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-04 12:45:12 -07:00
Sven Schnelle
9f389f4b20 rxrpc: remove smp_processor_id() from debug macro
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-03 15:40:54 -07:00
Denis V. Lunev
84f59370c5 [IPV6]: Fix refcounting for anycast dst entries.
Anycast DST entries allocated inside ipv6_dev_ac_inc are leaked when
network device is stopped without removing IPv6 addresses from it. The
bug has been observed in the reality on 2.6.18-rhel5 kernel.

In the above case addrconf_ifdown marks all entries as obsolete and
ip6_del_rt called from __ipv6_dev_ac_dec returns ENOENT. The
referrence is not dropped.

The fix is simple. DST entry should not keep referrence when stored in
the FIB6 tree.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-03 13:33:00 -07:00
Denis V. Lunev
eb86757931 [IPV6]: inet6_dev on loopback should be kept until namespace stop.
In the other case it will be destroyed when last address will be removed
from lo inside a namespace. This will break IPv6 in several places. The
most obvious one is ip6_dst_ifdown.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-03 13:31:53 -07:00
Denis V. Lunev
439e23857a [IPV6]: Event type in addrconf_ifdown is mis-used.
addrconf_ifdown is broken in respect to the usage of how
parameter. This function is called with (event != NETDEV_DOWN) and (2)
on the IPv6 stop.  It the latter case inet6_dev from loopback device
should be destroyed.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-03 13:30:17 -07:00
Herbert Xu
af2681828a [ICMP]: Ensure that ICMP relookup maintains status quo
The ICMP relookup path is only meant to modify behaviour when
appropriate IPsec policies are in place and marked as requiring
relookups.  It is certainly not meant to modify behaviour when
IPsec policies don't exist at all.

However, due to an oversight on the error paths existing behaviour
may in fact change should one of the relookup steps fail.

This patch corrects this by redirecting all errors on relookup
failures to the previous code path.  That is, if the initial
xfrm_lookup let the packet pass, we will stand by that decision
should the relookup fail due to an error.

This should be safe from a security point-of-view because compliant
systems must install a default deny policy so the packet would'nt
have passed in that case.

Many thanks to Julian Anastasov for pointing out this error.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-03 12:52:19 -07:00
Linus Torvalds
2f819ae881 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (45 commits)
  [VLAN]: Proc entry is not renamed when vlan device name changes.
  [IPV6]: Fix ICMP relookup error path dst leak
  [ATM] drivers/atm/iphase.c: compilation warning fix
  IPv6: do not create temporary adresses with too short preferred lifetime
  IPv6: only update the lifetime of the relevant temporary address
  bluetooth : __rfcomm_dlc_close lock fix
  bluetooth : use lockdep sub-classes for diffrent bluetooth protocol
  [ROSE/AX25] af_rose: rose_release() fix
  mac80211: correct use_short_preamble handling
  b43: Fix PCMCIA IRQ routing
  b43: Add DMA mapping failure messages
  mac80211: trigger ieee80211_sta_work after opening interface
  [LLC]: skb allocation size for responses
  [IP] UDP: Use SEQ_START_TOKEN.
  [NET]: Remove Documentation/networking/sk98lin.txt
  [ATM] atm/idt77252.c: Make 2 functions static
  [ATM]: Make atm/he.c:read_prom_byte() static
  [IPV6] MCAST: Ensure to check multicast listener(s).
  [LLC]: Kill llc_station_mac_sa symbol export.
  forcedeth: fix locking bug with netconsole
  ...
2008-04-02 07:46:18 -07:00
Pavel Emelyanov
802fb176d8 [VLAN]: Proc entry is not renamed when vlan device name changes.
This may lead to situations, when each of two proc entries produce
data for the other's device.

Looks like a BUG, so this patch is for net-2.6. It will not apply to
net-2.6.26 since dev->nd_net access is replaced with dev_net(dev)
one.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-02 00:08:01 -07:00