commit 2b85a34e91
(net: No more expensive sock_hold()/sock_put() on each tx)
changed initial sk_wmem_alloc value.
We need to take into account this offset when reporting
sk_wmem_alloc to user, in PROC_FS files or various
ioctls (SIOCOUTQ/TIOCOUTQ)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
My previous patch, which explicitly delays freeing of tnodes by adding
them to the list to flush them after the update is finished, isn't
strict enough. It treats exceptionally tnodes without parent, assuming
they are newly created, so "invisible" for the read side yet.
But the top tnode doesn't have parent as well, so we have to exclude
all exceptions (at least until a better way is found). Additionally we
need to move rcu assignment of this node before flushing, so the
return type of the trie_rebalance() function is changed.
Reported-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While doing trie_rebalance(): resize(), inflate(), halve() RCU free
tnodes before updating their parents. It depends on RCU delaying the
real destruction, but if RCU readers start after call_rcu() and before
parent update they could access freed memory.
It is currently prevented with preempt_disable() on the update side,
but it's not safe, except maybe classic RCU, plus it conflicts with
memory allocations with GFP_KERNEL flag used from these functions.
This patch explicitly delays freeing of tnodes by adding them to the
list, which is flushed after the update is finished.
Reported-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPv4:
- make PIM register vifs netns local
- set the netns when a PIM register vif is created
- make PIM available in all network namespaces (if CONFIG_IP_PIMSM_V2)
by adding the protocol handler when multicast routing is initialized
IPv6:
- make PIM register vifs netns local
- make PIM available in all network namespaces (if CONFIG_IPV6_PIMSM_V2)
by adding the protocol handler when multicast routing is initialized
Signed-off-by: Tom Goff <thomas.goff@boeing.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Removed the statements about ARP cache size as this config option does
not affect it. The cache size is controlled by neigh_table gc thresholds.
Remove also expiremental and obsolete markings as the API originally
intended for arp caching is useful for implementing ARP-like protocols
(e.g. NHRP) in user space and has been there for a long enough time.
Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the sake of power saver lovers, use a deferrable timer to fire
rt_check_expire()
As some big routers cache equilibrium depends on garbage collection
done in time, we take into account elapsed time between two
rt_check_expire() invocations to adjust the amount of slots we have to
check.
Based on an initial idea and patch from Tero Kristo
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Tero Kristo <tero.kristo@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix build error introduced by commit bb70dfa5 (netfilter: xtables:
consolidate comefrom debug cast access):
net/ipv4/netfilter/ip_tables.c: In function 'ipt_do_table':
net/ipv4/netfilter/ip_tables.c:421: error: 'comefrom' undeclared (first use in this function)
net/ipv4/netfilter/ip_tables.c:421: error: (Each undeclared identifier is reported only once
net/ipv4/netfilter/ip_tables.c:421: error: for each function it appears in.)
Signed-off-by: Patrick McHardy <kaber@trash.net>
One of the problem with sock memory accounting is it uses
a pair of sock_hold()/sock_put() for each transmitted packet.
This slows down bidirectional flows because the receive path
also needs to take a refcount on socket and might use a different
cpu than transmit path or transmit completion path. So these
two atomic operations also trigger cache line bounces.
We can see this in tx or tx/rx workloads (media gateways for example),
where sock_wfree() can be in top five functions in profiles.
We use this sock_hold()/sock_put() so that sock freeing
is delayed until all tx packets are completed.
As we also update sk_wmem_alloc, we could offset sk_wmem_alloc
by one unit at init time, until sk_free() is called.
Once sk_free() is called, we atomic_dec_and_test(sk_wmem_alloc)
to decrement initial offset and atomicaly check if any packets
are in flight.
skb_set_owner_w() doesnt call sock_hold() anymore
sock_wfree() doesnt call sock_put() anymore, but check if sk_wmem_alloc
reached 0 to perform the final freeing.
Drawback is that a skb->truesize error could lead to unfreeable sockets, or
even worse, prematurely calling __sk_free() on a live socket.
Nice speedups on SMP. tbench for example, going from 2691 MB/s to 2711 MB/s
on my 8 cpu dev machine, even if tbench was not really hitting sk_refcnt
contention point. 5 % speedup on a UDP transmit workload (depends
on number of flows), lowering TX completion cpu usage.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current conntrack code kills the ICMP conntrack entry as soon as
the first reply is received. This is incorrect, as we then see only
the first ICMP echo reply out of several possible duplicates as
ESTABLISHED, while the rest will be INVALID. Also this unnecessarily
increases the conntrackd traffic on H-A firewalls.
Make all the ICMP conntrack entries (including the replied ones)
last for the default of nf_conntrack_icmp{,v6}_timeout seconds.
Signed-off-by: Jan "Yenya" Kasprzak <kas@fi.muni.cz>
Signed-off-by: Patrick McHardy <kaber@trash.net>
The lock "protects" an assignment and a comparision of an integer.
When the caller of device_cmp() evaluates the result, nat->masq_index
may already have been changed (regardless if the lock is there or not).
So, the lock either has to be held during nf_ct_iterate_cleanup(),
or can be removed.
This does the latter.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
ip_mc_drop_socket() method is declared in linux/igmp.h, which
is included anyhow in af_inet.c. So there is no need for this declaration.
This patch removes it from af_inet.c.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define three accessors to get/set dst attached to a skb
struct dst_entry *skb_dst(const struct sk_buff *skb)
void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
void skb_dst_drop(struct sk_buff *skb)
This one should replace occurrences of :
dst_release(skb->dst)
skb->dst = NULL;
Delete skb->dst field
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define skb_rtable(const struct sk_buff *skb) accessor to get rtable from skb
Delete skb->rtable field
Setting rtable is not allowed, just set dst instead as rtable is an alias.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch simplifies the conntrack event caching system by removing
several events:
* IPCT_[*]_VOLATILE, IPCT_HELPINFO and IPCT_NATINFO has been deleted
since the have no clients.
* IPCT_COUNTER_FILLING which is a leftover of the 32-bits counter
days.
* IPCT_REFRESH which is not of any use since we always include the
timeout in the messages.
After this patch, the existing events are:
* IPCT_NEW, IPCT_RELATED and IPCT_DESTROY, that are used to identify
addition and deletion of entries.
* IPCT_STATUS, that notes that the status bits have changes,
eg. IPS_SEEN_REPLY and IPS_ASSURED.
* IPCT_PROTOINFO, that reports that internal protocol information has
changed, eg. the TCP, DCCP and SCTP protocol state.
* IPCT_HELPER, that a helper has been assigned or unassigned to this
entry.
* IPCT_MARK and IPCT_SECMARK, that reports that the mark has changed, this
covers the case when a mark is set to zero.
* IPCT_NATSEQADJ, to report that there's updates in the NAT sequence
adjustment.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
After some discussion offline with Christoph Lameter and David Stevens
regarding multicast behaviour in Linux, I'm submitting a slightly
modified patch from the one Christoph submitted earlier.
This patch provides a new socket option IP_MULTICAST_ALL.
In this case, default behaviour is _unchanged_ from the current
Linux standard. The socket option is set by default to provide
original behaviour. Sockets wishing to receive data only from
multicast groups they join explicitly will need to clear this
socket option.
Signed-off-by: Nivedita Singhvi <niv@us.ibm.com>
Signed-off-by: Christoph Lameter<cl@linux.com>
Acked-by: David Stevens <dlstevens@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Somewhat luckily, I was looking into these parts with very fine
comb because I've made somewhat similar changes on the same
area (conflicts that arose weren't that lucky though). The loop
was very much overengineered recently in commit 915219441d
(tcp: Use SKB queue and list helpers instead of doing it
by-hand), while it basically just wants to know if there are
skbs after 'skb'.
Also it got broken because skb1 = skb->next got translated into
skb1 = skb1->next (though abstracted) improperly. Note that
'skb1' is pointing to previous sk_buff than skb or NULL if at
head. Two things went wrong:
- We'll kfree 'skb' on the first iteration instead of the
skbuff following 'skb' (it would require required SACK reneging
to recover I think).
- The list head case where 'skb1' is NULL is checked too early
and the loop won't execute whereas it previously did.
Conclusion, mostly revert the recent changes which makes the
cset very messy looking but using proper accessor in the
previous-like version.
The effective changes against the original can be viewed with:
git-diff 915219441d566f1da0caa0e262be49b666159e17^ \
net/ipv4/tcp_input.c | sed -n -e '57,70 p'
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
ipgre_tunnel_xmit() might need skb->dst, so tell dev_hard_start_xmit()
to no release it.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ipip_tunnel_xmit() might need skb->dst, so tell dev_hard_start_xmit()
to no release it.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no need to repeatedly check flush when comparing TCP
options for GRO as it will be false 99% of the time where it
matters.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch optimises the IPv4 GRO code by using 32-bit loads
(instead of 16-bit ones) on the ID and length checks in the receive
function.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the overwhelming majority of cases, skb_gro_header's return
value cannot be NULL. Yet we must check it because of its current
form. This patch splits it up into multiple functions in order
to avoid this.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of checking len > mss || len == 0, we can accomplish
both by checking (len - 1) > mss using the unsigned wraparound.
At nearly a million times a second, this might just help.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The window has already been checked as part of the flag word
so there is no need to check it explicitly.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of doing two 16-bit operations for the source/destination
ports, we can do one 32-bit operation to take care both.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes ssthresh accounting issues in tcp_vegas when cwnd decreases
Signed-off-by: Doug Leith <doug.leith@nuim.ie>
Signed-off-by: David S. Miller <davem@davemloft.net>
It seems we can fix this by disabling preemption while we re-balance the
trie. This is with the CONFIG_CLASSIC_RCU. It's been stress-tested at high
loads continuesly taking a full BGP table up/down via iproute -batch.
Note. fib_trie is not updated for CONFIG_PREEMPT_RCU
Reported-by: Andrei Popa
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netlink message header (struct nlmsghdr) is an unused parameter in
fill method of fib_rules_ops struct. This patch removes this
parameter from this method and fixes the places where this method is
called.
(include/net/fib_rules.h)
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexander V. Lukyanov found a regression in 2.6.29 and made a complete
analysis found in http://bugzilla.kernel.org/show_bug.cgi?id=13339
Quoted here because its a perfect one :
begin_of_quotation
2.6.29 patch has introduced flexible route cache rebuilding. Unfortunately the
patch has at least one critical flaw, and another problem.
rt_intern_hash calculates rthi pointer, which is later used for new entry
insertion. The same loop calculates cand pointer which is used to clean the
list. If the pointers are the same, rtable leak occurs, as first the cand is
removed then the new entry is appended to it.
This leak leads to unregister_netdevice problem (usage count > 0).
Another problem of the patch is that it tries to insert the entries in certain
order, to facilitate counting of entries distinct by all but QoS parameters.
Unfortunately, referencing an existing rtable entry moves it to list beginning,
to speed up further lookups, so the carefully built order is destroyed.
For the first problem the simplest patch it to set rthi=0 when rthi==cand, but
it will also destroy the ordering.
end_of_quotation
Problematic commit is 1080d709fb
(net: implement emergency route cache rebulds when gc_elasticity is exceeded)
Trying to keep dst_entries ordered is too complex and breaks the fact that
order should depend on the frequency of use for garbage collection.
A possible fix is to make rt_intern_hash() simpler, and only makes
rt_check_expire() a litle bit smarter, being able to cope with an arbitrary
entries order. The added loop is running on cache hot data, while cpu
is prefetching next object, so should be unnoticied.
Reported-and-analyzed-by: Alexander V. Lukyanov <lav@yar.ru>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rt_check_expire() computes average and standard deviation of chain lengths,
but not correclty reset length to 0 at beginning of each chain.
This probably gives overflows for sum2 (and sum) on loaded machines instead
of meaningful results.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The DHCP spec allows the server to specify the MTU. This can be useful
for netbooting with UDP-based NFS-root on a network using jumbo frames.
This patch allows the kernel IP autoconfiguration to handle this option
correctly.
It would be possible to use initramfs and add a script to set the MTU,
but that seems like a complicated solution if no initramfs is otherwise
necessary, and would bloat the kernel image more than this code would.
This patch was originally submitted to LKML in 2003 by Hans-Peter Jansen.
Signed-off-by: Chris Friesen <cfriesen@nortel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sysctls are unregistered with the rntl_lock held making
it unsafe to unconditionally grab the the rtnl_lock. Instead
we need to call rtnl_trylock and restart the system call
if we can not grab it. Otherwise we could deadlock at unregistration
time.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit e81963b1 ("ipv4: Make INET_LRO a bool instead of tristate.")
changed this config from tristate to bool. Add default so that it is
consistent with the help text.
Signed-off-by: Frans Pop <elendil@planet.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no need for net/icmp.h header in net/ipv4/fib_frontend.c.
This patch removes the #include net/icmp.h from it.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 518a09ef11 (tcp: Fix recvmsg MSG_PEEK influence of
blocking behavior) lets the loop run longer than the race check
did previously expect, so we need to be more careful with this
check and consider the work we have been doing.
I tried my best to deal with urg hole madness too which happens
here:
if (!sock_flag(sk, SOCK_URGINLINE)) {
++*seq;
...
by using additional offset by one but I certainly have very
little interest in testing that part.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Tested-by: Frans Pop <elendil@planet.nl>
Tested-by: Ian Zimmermann <itz@buug.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a DHCP server is delayed, it's possible for the client to receive the
DHCPOFFER after it has already sent out a new DHCPDISCOVER message from
a second interface. The client then sends out a DHCPREQUEST from the
second interface, but the server doesn't recognize the device and
rejects the request.
This patch simply tracks the current device being configured and throws
away the OFFER if it is not intended for the current device. A more
sophisticated approach would be to put the OFFER information into the
struct ic_device rather than storing it globally.
Signed-off-by: Chris Friesen <cfriesen@nortel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This code is used as a library by several device drivers,
which select INET_LRO.
If some are modules and some are statically built into the
kernel, we get build failures if INET_LRO is modular.
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet