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>
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>
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>
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>
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>
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>
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>
* 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
...
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>
When we encounter an error while looking up the dst the second
time we need to drop the first dst. This patch is pretty much
the same as the one for IPv4.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
From RFC341:
A temporary address is created only if this calculated Preferred
Lifetime is greater than REGEN_ADVANCE time units. In particular, an
implementation must not create a temporary address with a zero
Preferred Lifetime.
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When receiving a prefix information from a routeur, only update the
lifetimes of the temporary address associated with that prefix.
Otherwise if one deprecated prefix is advertized, all your temporary
addresses will become deprecated.
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
rose_release() doesn't release sockets properly, e.g. it skips
sock_orphan(), so OOPSes are triggered in sock_def_write_space(),
which was observed especially while ROSE skbs were kfreed from
ax25_frames_acked(). There is also sock_hold() and lock_sock() added -
similarly to ax25_release(). Thanks to Bernard Pidoux for substantial
help in debugging this problem.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Reported-and-tested-by: Bernard Pidoux <bpidoux@free.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
ERP IE bit for preamble mode is 0 for short and 1 for long, not the other
way around. This fixes the value reported to the driver via
bss_conf->use_short_preamble field.
Signed-off-by: Vladimir Koutny <vlado@ksp.sk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
ieee80211_sta_work is disabled while network interface
is down. Therefore, if you configure wireless parameters
before bringing the interface up, these configurations are
not yet effective and association fails.
A workaround from userspace is calling a command like
'iwconfig wlan0 ap any' after the interface is brought up.
To fix this behaviour, trigger execution of ieee80211_sta_work from
ieee80211_open when in STA or IBSS mode.
Signed-off-by: Jan Niehusmann <jan@gondor.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Allocate the skb for llc responses with the received packet size by
using the size adjustable llc_frame_alloc.
Don't allocate useless extra payload.
Cleanup magic numbers.
So, this fixes oops.
Reported by Jim Westfall:
kernel: skb_over_panic: text:c0541fc7 len:1000 put:997 head:c166ac00 data:c166ac2f tail:0xc166b017 end:0xc166ac80 dev:eth0
kernel: ------------[ cut here ]------------
kernel: kernel BUG at net/core/skbuff.c:95!
Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In ip6_mc_input(), we need to check whether we have listener(s) for
the packet.
After commit ae7bf20a63, all packets
for multicast destinations are delivered to upper layer if
IFF_PROMISC or IFF_ALLMULTI is set.
In fact, bug was rather ancient; the original (before the commit)
intent of the dev->flags check was to skip the ipv6_chk_mcast_addr()
call, assuming L2 filters packets appropriately, but it was even not
true.
Let's explicitly check our multicast list.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Acked-by: David L Stevens <dlstevens@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
kill unnecessary llc_station_mac_sa.
Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
discard llc packet which has bogus packet length.
Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The qdisc_run loop is currently unbounded and runs entirely in a
softirq. This is bad as it may create an unbounded softirq run.
This patch fixes this by calling need_resched and breaking out if
necessary.
It also adds a break out if the jiffies value changes since that would
indicate we've been transmitting for too long which starves other
softirqs.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 9af3912ec9 ("[NET] Move DF check
to ip_forward") added a new check to send ICMP fragmentation needed
for large packets.
Unlike the check in ip_finish_output(), it doesn't check for GSO.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The older RW_LOCK_UNLOCKED macros defeat lockdep state tracing so
replace them with the newer __RW_LOCK_UNLOCKED macros.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
LLC currently allows users to inject raw frames, including IP packets
encapsulated in SNAP. While Linux doesn't handle IP over SNAP, other
systems do. Restrict LLC sockets to root similar to packet sockets.
[ Modified Patrick's patch to use CAP_NEW_RAW --DaveM ]
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This elliminates infamous race during module loading when one could lookup
proc entry without proc_fops assigned.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
ESP does not account for the IV size when calling pskb_may_pull() to
ensure everything it accesses directly is within the linear part of a
potential fragment. This results in a BUG() being triggered when the
both the IPv4 and IPv6 ESP stack is fed with an skb where the first
fragment ends between the end of the esp header and the end of the IV.
This bug was found by Dirk Nehring <dnehring@gmx.net> .
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
The IPv6 BEET output function is incorrectly including the inner
header in the payload to be protected. This causes a crash as
the packet doesn't actually have that many bytes for a second
header.
The IPv4 BEET output on the other hand is broken when it comes
to handling an inner IPv6 header since it always assumes an
inner IPv4 header.
This patch fixes both by making sure that neither BEET output
function touches the inner header at all. All access is now
done through the protocol-independent cb structure. Two new
attributes are added to make this work, the IP header length
and the IPv4 option length. They're filled in by the inner
mode's output function.
Thanks to Joakim Koskela for finding this problem.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RDMACTXT_F_LAST_CTXT bit was getting set incorrectly
when the last chunk in the read-list spanned multiple pages. This
resulted in a kernel panic when the wrong context was used to
build the RPC iovec page list.
RDMA_READ is used to fetch RPC data from the client for
NFS_WRITE requests. A scatter-gather is used to map the
advertised client side buffer to the server-side iovec and
associated page list.
WR contexts are used to convey which scatter-gather entries are
handled by each WR. When the write data is large, a single RPC may
require multiple RDMA_READ requests so the contexts for a single RPC
are chained together in a linked list. The last context in this list
is marked with a bit RDMACTXT_F_LAST_CTXT so that when this WR completes,
the CQ handler code can enqueue the RPC for processing.
The code in rdma_read_xdr was setting this bit on the last two
contexts on this list when the last read-list chunk spanned multiple
pages. This caused the svc_rdma_recvfrom logic to incorrectly build
the RPC and caused the kernel to crash because the second-to-last
context doesn't contain the iovec page list.
Modified the condition that sets this bit so that it correctly detects
the last context for the RPC.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Tested-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 8b7817f3a9 ([IPSEC]: Add ICMP host
relookup support) introduced some dst leaks on error paths: the rt
pointer can be forgotten to be put. Fix it bu going to a proper label.
Found after net namespace's lo refused to unregister :) Many thanks to
Den for valuable help during debugging.
Herbert pointed out, that xfrm_lookup() will put the rtable in case
of error itself, so the first goto fix is redundant.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Given that there are no apparent calls to lock_kernel() or
unlock_kernel() under net/ax25, delete the TODO reference related to
that.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
SIOCADDMULTI/SIOCDELMULTI check whether the driver has a set_multicast_list
method to determine whether it supports multicast. Drivers implementing
secondary unicast support use set_rx_mode however.
Check for both dev->set_multicast_mode and dev->set_rx_mode to determine
multicast capabilities.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
It should be a "struct ktermios" not a "struct termios".
Based upon a build warning reported by Stephen Rothwell.
Signed-off-by: David S. Miller <davem@davemloft.net>
Changing these flags requires to use dev_set_allmulti/dev_set_promiscuity
or dev_change_flags. Setting it directly causes two unwanted effects:
- the next dev_change_flags call will notice a difference between
dev->gflags and the actual flags, enable promisc/allmulti
mode and incorrectly update dev->gflags
- this keeps the underlying device in promisc/allmulti mode until
the VLAN device is deleted
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Proxy neighbors do not have any reference counting, so any caller
of pneigh_lookup (unless it's a netlink triggered add/del routine)
should _not_ perform any actions on the found proxy entry.
There's one exception from this rule - the ipv6's ndisc_recv_ns()
uses found entry to check the flags for NTF_ROUTER.
This creates a race between the ndisc and pneigh_delete - after
the pneigh is returned to the caller, the nd_tbl.lock is dropped
and the deleting procedure may proceed.
One of the fixes would be to add a reference counting, but this
problem exists for ndisc only. Besides such a patch would be too
big for -rc4.
So I propose to introduce a __pneigh_lookup() which is supposed
to be called with the lock held and use it in ndisc code to check
the flags on alive pneigh entry.
Changes from v2:
As David noticed, Exported the __pneigh_lookup() to ipv6 module.
The checkpatch generates a warning on it, since the EXPORT_SYMBOL
does not follow the symbol itself, but in this file all the
exports come at the end, so I decided no to break this harmony.
Changes from v1:
Fixed comments from YOSHIFUJI - indentation of prototype in header
and the pndisc_check_router() name - and a compilation fix, pointed
by Daniel - the is_routed was (falsely) considered as uninitialized
by gcc.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
sch_htb: fix "too many events" situation
connector: convert to single-threaded workqueue
[ATM]: When proc_create() fails, do some error handling work and return -ENOMEM.
[SUNGEM]: Fix NAPI assertion failure.
BNX2X: prevent ethtool from setting port type
[9P] net/9p/trans_fd.c: remove unused variable
[IPV6] net/ipv6/ndisc.c: remove unused variable
[IPV4] fib_trie: fix warning from rcu_assign_poinger
[TCP]: Let skbs grow over a page on fast peers
[DLCI]: Fix tiny race between module unload and sock_ioctl.
[SCTP]: Fix build warnings with IPV6 disabled.
[IPV4]: Fix null dereference in ip_defrag
The iWARP protocol limits RDMA read requests to a single scatter
entry. NFS/RDMA has code in rdma_read_max_sge() that is supposed to
limit the sge_count for RDMA read requests to 1, but the code to do
that is inside an #ifdef RDMA_TRANSPORT_IWARP block. In the mainline
kernel at least, RDMA_TRANSPORT_IWARP is an enum and not a
preprocessor #define, so the #ifdef'ed code is never compiled.
In my test of a kernel build with -j8 on an NFS/RDMA mount, this
problem eventually leads to trouble starting with:
svcrdma: Error posting send = -22
svcrdma : RDMA_READ error = -22
and things go downhill from there.
The trivial fix is to delete the #ifdef guard. The check seems to be
a remnant of when the NFS/RDMA code was not merged and needed to
compile against multiple kernel versions, although I don't think it
ever worked as intended. In any case now that the code is upstream
there's no need to test whether the RDMA_TRANSPORT_IWARP constant is
defined or not.
Without this patch, my kernel build on an NFS/RDMA mount using NetEffect
adapters quickly and 100% reproducibly failed with an error like:
ld: final link failed: Software caused connection abort
With the patch applied I was able to complete a kernel build on the
same setup.
(Tom Tucker says this is "actually an _ancient_ remnant when it had to
compile against iWARP vs. non-iWARP enabled OFA trees.")
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Acked-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
HTB is event driven algorithm and part of its work is to apply
scheduled events at proper times. It tried to defend itself from
livelock by processing only limited number of events per dequeue.
Because of faster computers some users already hit this hardcoded
limit.
This patch limits processing up to 2 jiffies (why not 1 jiffie ?
because it might stop prematurely when only fraction of jiffie
remains).
Signed-off-by: Martin Devera <devik@cdi.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
The variable cb is initialized but never used otherwise.
The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@@
type T;
identifier i;
constant C;
@@
(
extern T i;
|
- T i;
<+... when != i
- i = C;
...+>
)
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The variable hlen is initialized but never used otherwise.
The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@@
type T;
identifier i;
constant C;
@@
(
extern T i;
|
- T i;
<+... when != i
- i = C;
...+>
)
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
This gets rid of a warning caused by the test in rcu_assign_pointer.
I tried to fix rcu_assign_pointer, but that devolved into a long set
of discussions about doing it right that came to no real solution.
Since the test in rcu_assign_pointer for constant NULL would never
succeed in fib_trie, just open code instead.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While testing the virtio-net driver on KVM with TSO I noticed
that TSO performance with a 1500 MTU is significantly worse
compared to the performance of non-TSO with a 16436 MTU. The
packet dump shows that most of the packets sent are smaller
than a page.
Looking at the code this actually is quite obvious as it always
stop extending the packet if it's the first packet yet to be
sent and if it's larger than the MSS. Since each extension is
bound by the page size, this means that (given a 1500 MTU) we're
very unlikely to construct packets greater than a page, provided
that the receiver and the path is fast enough so that packets can
always be sent immediately.
The fix is also quite obvious. The push calls inside the loop
is just an optimisation so that we don't end up doing all the
sending at the end of the loop. Therefore there is no specific
reason why it has to do so at MSS boundaries. For TSO, the
most natural extension of this optimisation is to do the pushing
once the skb exceeds the TSO size goal.
This is what the patch does and testing with KVM shows that the
TSO performance with a 1500 MTU easily surpasses that of a 16436
MTU and indeed the packet sizes sent are generally larger than
16436.
I don't see any obvious downsides for slower peers or connections,
but it would be prudent to test this extensively to ensure that
those cases don't regress.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ilpo Järvinen