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>
By separating the freeing code from the refcounting decrementing.
Probably reducing icache pressure when we still have reference counts to
go.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_prequeue() refers to the constant value (TCP_RTO_MIN) regardless of
the actual value might be tuned. The following patches fix this and make
tcp_prequeue get the actual value returns from tcp_rto_min().
Signed-off-by: Satoru SATOH <satoru.satoh@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This should be very safe compared with full enabled, so I see
no reason why it shouldn't be done right away. As ECN can only
be negotiated if the SYN sending party is also supporting it,
somebody in the loop probably knows what he/she is doing. If
SYN does not ask for ECN, the server side SYN-ACK is identical
to what it is without ECN. Thus it's quite safe.
The chosen value is safe w.r.t to existing configs which
choose to currently set manually either 0 or 1 but
silently upgrades those who have not explicitly requested
ECN off.
Whether to just enable both sides comes up time to time but
unless that gets done now we can at least make the servers
aware of ECN already. As there are some known problems to occur
if ECN is enabled, it's currently questionable whether there's
any real gain from enabling clients as servers mostly won't
support it anyway (so we'd hit just the negative sides). After
enabling the servers and getting that deployed, the client end
enable really has some potential gain too.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
The x_tables are organized with a table structure and a per-cpu copies
of the counters and rules. On older kernels there was a reader/writer
lock per table which was a performance bottleneck. In 2.6.30-rc, this
was converted to use RCU and the counters/rules which solved the performance
problems for do_table but made replacing rules much slower because of
the necessary RCU grace period.
This version uses a per-cpu set of spinlocks and counters to allow to
table processing to proceed without the cache thrashing of a global
reader lock and keeps the same performance for table updates.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
These are later assigned to other values without being used meanwhile.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On a brand new GRO skb, we cannot call ip_hdr since the header
may lie in the non-linear area. This patch adds the helper
skb_gro_network_header to handle this.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Right now we have no upper limit on the size of the route cache hash table.
On a 128GB POWER6 box it ends up as 32MB:
IP route cache hash table entries: 4194304 (order: 9, 33554432 bytes)
It would be nice to cap this for memory consumption reasons, but a massive
hashtable also causes a significant spike when measuring OS jitter.
With a 32MB hashtable and 4 million entries, rt_worker_func is taking
5 ms to complete. On another system with more memory it's taking 14 ms.
Even though rt_worker_func does call cond_sched() to limit its impact,
in an HPC environment we want to keep all sources of OS jitter to a minimum.
With the patch applied we limit the number of entries to 512k which
can still be overriden by using the rt_entries boot option:
IP route cache hash table entries: 524288 (order: 6, 4194304 bytes)
With this patch rt_worker_func now takes 0.460 ms on the same system.
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The IP MIB (RFC 4293) defines stats for InOctets, OutOctets, InMcastOctets and
OutMcastOctets:
http://tools.ietf.org/html/rfc4293
But it seems we don't track those in any way that easy to separate from other
protocols. This patch adds those missing counters to the stats file. Tested
successfully by me
With help from Eric Dumazet.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
last_synq_overflow eats 4 or 8 bytes in struct tcp_sock, even
though it is only used when a listening sockets syn queue
is full.
We can (ab)use rx_opt.ts_recent_stamp to store the same information;
it is not used otherwise as long as a socket is in listen state.
Move linger2 around to avoid splitting struct mtu_probe
across cacheline boundary on 32 bit arches.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Just noticed while doing some new work that the recent
mid-wq adjustment logic will misbehave when FACK is not
in use (happens either due sysctl'ed off or auto-detected
reordering) because I forgot the relevant TCPCB tagbit.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet_register_protosw() function is responsible for adding a new
inet protocol into a global table (inetsw[]) that is used with RCU rules.
As soon as the store of the pointer is done, other cpus might see
this new protocol in inetsw[], so we have to make sure new protocol
is ready for use. All pending memory updates should thus be committed
to memory before setting the pointer.
This is correctly done using rcu_assign_pointer()
synchronize_net() is typically used at unregister time, after
unsetting the pointer, to make sure no other cpu is still using
the object we want to dismantle. Using it at register time
is only adding an artificial delay that could hide a real bug,
and this bug could popup if/when synchronize_rcu() can proceed
faster than now.
This saves about 13 ms on boot time on a HZ=1000 8 cpus machine ;)
(4 calls to inet_register_protosw(), and about 3200 us per call)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After calling skb_gro_receive skb->len can no longer be relied
on since if the skb was merged using frags, then its pages will
have been removed and the length reduced.
This caused tcp_gro_receive to prematurely end merging which
resulted in suboptimal performance with ixgbe.
The fix is to store skb->len on the stack.
Reported-by: Mark Wagner <mwagner@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The removal of the SAME target accidentally removed one feature that is
not available from the normal NAT targets so far, having multi-range
mappings that use the same mapping for each connection from a single
client. The current behaviour is to choose the address from the range
based on source and destination IP, which breaks when communicating
with sites having multiple addresses that require all connections to
originate from the same IP address.
Introduce a IP_NAT_RANGE_PERSISTENT option that controls whether the
destination address is taken into account for selecting addresses.
http://bugzilla.kernel.org/show_bug.cgi?id=12954
Signed-off-by: Patrick McHardy <kaber@trash.net>
A long-standing feature in tcp_init_metrics() is such that
any of its goto reset prevents call to tcp_init_cwnd().
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (54 commits)
glge: remove unused #include <version.h>
dnet: remove unused #include <version.h>
tcp: miscounts due to tcp_fragment pcount reset
tcp: add helper for counter tweaking due mid-wq change
hso: fix for the 'invalid frame length' messages
hso: fix for crash when unplugging the device
fsl_pq_mdio: Fix compile failure
fsl_pq_mdio: Revive UCC MDIO support
ucc_geth: Pass proper device to DMA routines, otherwise oops happens
i.MX31: Fixing cs89x0 network building to i.MX31ADS
tc35815: Fix build error if NAPI enabled
hso: add Vendor/Product ID's for new devices
ucc_geth: Remove unused header
gianfar: Remove unused header
kaweth: Fix locking to be SMP-safe
net: allow multiple dev per napi with GRO
r8169: reset IntrStatus after chip reset
ixgbe: Fix potential memory leak/driver panic issue while setting up Tx & Rx ring parameters
ixgbe: fix ethtool -A|a behavior
ixgbe: Patch to fix driver panic while freeing up tx & rx resources
...
It seems that trivial reset of pcount to one was not sufficient
in tcp_retransmit_skb. Multiple counters experience a positive
miscount when skb's pcount gets lowered without the necessary
adjustments (depending on skb's sacked bits which exactly), at
worst a packets_out miscount can crash at RTO if the write queue
is empty!
Triggering this requires mss change, so bidir tcp or mtu probe or
like.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Tested-by: Uwe Bugla <uwe.bugla@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need full-scale adjustment to fix a TCP miscount in the next
patch, so just move it into a helper and call for that from the
other places.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 784544739a
(netfilter: iptables: lock free counters) forgot to disable BH
in arpt_do_table(), ipt_do_table() and ip6t_do_table()
Use rcu_read_lock_bh() instead of rcu_read_lock() cures the problem.
Reported-and-bisected-by: Roman Mindalev <r000n@r000n.net>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6:
smack: Add a new '-CIPSO' option to the network address label configuration
netlabel: Cleanup the Smack/NetLabel code to fix incoming TCP connections
lsm: Remove the socket_post_accept() hook
selinux: Remove the "compat_net" compatibility code
netlabel: Label incoming TCP connections correctly in SELinux
lsm: Relocate the IPv4 security_inet_conn_request() hooks
TOMOYO: Fix a typo.
smack: convert smack to standard linux lists
The current NetLabel/SELinux behavior for incoming TCP connections works but
only through a series of happy coincidences that rely on the limited nature of
standard CIPSO (only able to convey MLS attributes) and the write equality
imposed by the SELinux MLS constraints. The problem is that network sockets
created as the result of an incoming TCP connection were not on-the-wire
labeled based on the security attributes of the parent socket but rather based
on the wire label of the remote peer. The issue had to do with how IP options
were managed as part of the network stack and where the LSM hooks were in
relation to the code which set the IP options on these newly created child
sockets. While NetLabel/SELinux did correctly set the socket's on-the-wire
label it was promptly cleared by the network stack and reset based on the IP
options of the remote peer.
This patch, in conjunction with a prior patch that adjusted the LSM hook
locations, works to set the correct on-the-wire label format for new incoming
connections through the security_inet_conn_request() hook. Besides the
correct behavior there are many advantages to this change, the most significant
is that all of the NetLabel socket labeling code in SELinux now lives in hooks
which can return error codes to the core stack which allows us to finally get
ride of the selinux_netlbl_inode_permission() logic which greatly simplfies
the NetLabel/SELinux glue code. In the process of developing this patch I
also ran into a small handful of AF_INET6 cleanliness issues that have been
fixed which should make the code safer and easier to extend in the future.
Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: James Morris <jmorris@namei.org>
The current placement of the security_inet_conn_request() hooks do not allow
individual LSMs to override the IP options of the connection's request_sock.
This is a problem as both SELinux and Smack have the ability to use labeled
networking protocols which make use of IP options to carry security attributes
and the inability to set the IP options at the start of the TCP handshake is
problematic.
This patch moves the IPv4 security_inet_conn_request() hooks past the code
where the request_sock's IP options are set/reset so that the LSM can safely
manipulate the IP options as needed. This patch intentionally does not change
the related IPv6 hooks as IPv6 based labeling protocols which use IPv6 options
are not currently implemented, once they are we will have a better idea of
the correct placement for the IPv6 hooks.
Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: James Morris <jmorris@namei.org>
Conflicts:
arch/sparc/kernel/time_64.c
drivers/gpu/drm/drm_proc.c
Manual merge to resolve build warning due to phys_addr_t type change
on x86:
drivers/gpu/drm/drm_info.c
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use "hlist_nulls" infrastructure we added in 2.6.29 for RCUification of UDP & TCP.
This permits an easy conversion from call_rcu() based hash lists to a
SLAB_DESTROY_BY_RCU one.
Avoiding call_rcu() delay at nf_conn freeing time has numerous gains.
First, it doesnt fill RCU queues (up to 10000 elements per cpu).
This reduces OOM possibility, if queued elements are not taken into account
This reduces latency problems when RCU queue size hits hilimit and triggers
emergency mode.
- It allows fast reuse of just freed elements, permitting better use of
CPU cache.
- We delete rcu_head from "struct nf_conn", shrinking size of this structure
by 8 or 16 bytes.
This patch only takes care of "struct nf_conn".
call_rcu() is still used for less critical conntrack parts, that may
be converted later if necessary.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Commit e1b4b9f ([NETFILTER]: {ip,ip6,arp}_tables: fix exponential worst-case
search for loops) introduced a regression in the loop detection algorithm,
causing sporadic incorrectly detected loops.
When a chain has already been visited during the check, it is treated as
having a standard target containing a RETURN verdict directly at the
beginning in order to not check it again. The real target of the first
rule is then incorrectly treated as STANDARD target and checked not to
contain invalid verdicts.
Fix by making sure the rule does actually contain a standard target.
Based on patch by Francis Dupont <Francis_Dupont@isc.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
We use same not trivial helper function in four places. We can factorize it.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
The ipv6 version of bind_conflict code calls ipv6_rcv_saddr_equal()
which at times wrongly identified intersections between addresses.
It particularly broke down under a few instances and caused erroneous
bind conflicts.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arches without efficient unaligned access can still perform a loop
assuming 16bit alignment in ifname_compare()
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Discard incoming packets whose ack field iincludes data not yet sent.
This is consistent with RFC 793 Section 3.9.
Change tcp_ack() to distinguish between too-small and too-large ack
field values. Keep segments with too-large ack fields out of the fast
path, and change slow path to discard them.
Reported-by: Oliver Zheng <mailinglists+netdev@oliverzheng.com>
Signed-off-by: John Dykstra <john.dykstra1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_sack_swap seems unnecessary so I pushed swap to the caller.
Also removed comment that seemed then pointless, and added include
when not already there. Compile tested.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
dev can be NULL in ip[6]_frag_reasm for skb's coming from RAW sockets.
Quagga's OSPFD sends fragmented packets on a RAW socket, when netfilter
conntrack reassembles them on the OUTPUT path you hit this code path.
You can test it with something like "hping2 -0 -d 2000 -f AA.BB.CC.DD"
With help from Jarek Poplawski.
Signed-off-by: Jorge Boncompte [DTI2] <jorge@dti2.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch removes an unused parameter (addr_len) from tcp_recv_urg()
method in net/ipv4/tcp.c.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ip_queue module is missing the net-pf-16-proto-3 alias that would
causae it to be auto-loaded when a socket of that type is opened. This
patch adds the alias.
Signed-off-by: Scott James Remnant <scott@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This patch increases the statistics of packets drop if the sequence
adjustment fails in ipv4_confirm().
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This patch modifies nf_log to use a linked list of loggers for each
protocol. This list of loggers is read and write protected with a
mutex.
This patch separates registration and binding. To be used as
logging module, a module has to register calling nf_log_register()
and to bind to a protocol it has to call nf_log_bind_pf().
This patch also converts the logging modules to the new API. For nfnetlink_log,
it simply switchs call to register functions to call to bind function and
adds a call to nf_log_register() during init. For other modules, it just
remove a const flag from the logger structure and replace it with a
__read_mostly.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
It's not too likely to happen, would basically require crafted
packets (must hit the max guard in tcp_bound_to_half_wnd()).
It seems that nothing that bad would happen as there's tcp_mems
and congestion window that prevent runaway at some point from
hurting all too much (I'm not that sure what all those zero
sized segments we would generate do though in write queue).
Preventing it regardless is certainly the best way to go.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Cc: Evgeniy Polyakov <zbr@ioremap.net>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
The results is very unlikely change every so often so we
hardly need to divide again after doing that once for a
connection. Yet, if divide still becomes necessary we
detect that and do the right thing and again settle for
non-divide state. Takes the u16 space which was previously
taken by the plain xmit_size_goal.
This should take care part of the tso vs non-tso difference
we found earlier.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's very little need for most of the callsites to get
tp->xmit_goal_size updated. That will cost us divide as is,
so slice the function in two. Also, the only users of the
tp->xmit_goal_size are directly behind tcp_current_mss(),
so there's no need to store that variable into tcp_sock
at all! The drop of xmit_goal_size currently leaves 16-bit
hole and some reorganization would again be necessary to
change that (but I'm aiming to fill that hole with u16
xmit_goal_size_segs to cache the results of the remaining
divide to get that tso on regression).
Bring xmit_goal_size parts into tcp.c
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Cc: Evgeniy Polyakov <zbr@ioremap.net>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
It seems that no variables clash such that we couldn't do
the check just once later on. Therefore move it.
Also kill dead obvious comment, dead argument and add
unlikely since this mtu probe does not happen too often.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Wow, it was quite tricky to merge that stream of negations
but I think I finally got it right:
check & replace_ts_recent:
(s32)(rcv_tsval - ts_recent) >= 0 => 0
(s32)(ts_recent - rcv_tsval) <= 0 => 0
discard:
(s32)(ts_recent - rcv_tsval) > TCP_PAWS_WINDOW => 1
(s32)(ts_recent - rcv_tsval) <= TCP_PAWS_WINDOW => 0
I toggled the return values of tcp_paws_check around since
the old encoding added yet-another negation making tracking
of truth-values really complicated.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
I've already forgotten what for this was necessary, anyway
it's no longer used (if it ever was).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the pure assignment case, the earlier zeroing is
still in effect.
David S. Miller raised concerns if the ifs are there to avoid
dirtying cachelines. I came to these conclusions:
> We'll be dirty it anyway (now that I check), the first "real" statement
> in tcp_rcv_established is:
>
> tp->rx_opt.saw_tstamp = 0;
>
> ...that'll land on the same dword. :-/
>
> I suppose the blocks are there just because they had more complexity
> inside when they had to calculate the eff_sacks too (maybe it would
> have been better to just remove them in that drop-patch so you would
> have had less head-ache :-)).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Windows (XP at least) hosts on boot, with configured static ip, performing
address conflict detection, which is defined in RFC3927.
Here is quote of important information:
"
An ARP announcement is identical to the ARP Probe described above,
except that now the sender and target IP addresses are both set
to the host's newly selected IPv4 address.
"
But it same time this goes wrong with RFC5227.
"
The 'sender IP address' field MUST be set to all zeroes; this is to avoid
polluting ARP caches in other hosts on the same link in the case
where the address turns out to be already in use by another host.
"
When ARP proxy configured, it must not answer to both cases, because
it is address conflict verification in any case. For Windows it is just
causing to detect false "ip conflict". Already there is code for RFC5227, so
just trivially we just check also if source ip == target ip.
Signed-off-by: Denys Fedoryshchenko <denys@visp.net.lb>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some systems send SYN packets with apparently wrong RFC1323 timestamp
option values [timestamp tsval=0 tsecr=0].
It might be for security reasons (http://www.secuobs.com/plugs/25220.shtml )
Linux TCP stack ignores this option and sends back a SYN+ACK packet
without timestamp option, thus many TCP flows cannot use timestamps
and lose some benefit of RFC1323.
Other operating systems seem to not care about initial tsval value, and let
tcp flows to negotiate timestamp option.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Protocols that use packet_type can be __read_mostly section for better
locality. Elminate any unnecessary initializations of NULL.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To remove the possibility of packets flying around when network
devices are being cleaned up use reisger_pernet_subsys instead of
register_pernet_device.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Recently I had a kernel panic in icmp_send during a network namespace
cleanup. There were packets in the arp queue that failed to be sent
and we attempted to generate an ICMP host unreachable message, but
failed because icmp_sk_exit had already been called.
The network devices are removed from a network namespace and their
arp queues are flushed before we do attempt to shutdown subsystems
so this error should have been impossible.
It turns out icmp_init is using register_pernet_device instead
of register_pernet_subsys. Which resulted in icmp being shut down
while we still had the possibility of packets in flight, making
a nasty NULL pointer deference in interrupt context possible.
Changing this to register_pernet_subsys fixes the problem in
my testing.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The above functions from include/net/tcp.h have been defined with an
argument that they never use. The argument is 'u32 ack' which is never
used inside the function body, and thus it can be removed. The rest of
the patch involves the necessary changes to the function callers of the
above two functions.
Signed-off-by: Hantzis Fotis <xantzis@ceid.upatras.gr>
Signed-off-by: David S. Miller <davem@davemloft.net>
I guess these fields were one day 16-bit in the struct but
nowadays they're just using 8 bits anyway.
This is just a precaution, didn't result any change in my
case but who knows what all those varying gcc versions &
options do. I've been told that 16-bit is not so nice with
some cpus.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
copied was assigned zero right before the goto, so if (copied)
cannot ever be true.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Also fixes insignificant bug that would cause sending of stale
SACK block (would occur in some corner cases).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
It seems that implementation in yeah was inconsistent to what
other did as it would increase cwnd one ack earlier than the
others do.
Size benefits:
bictcp_cong_avoid | -36
tcp_cong_avoid_ai | +52
bictcp_cong_avoid | -34
tcp_scalable_cong_avoid | -36
tcp_veno_cong_avoid | -12
tcp_yeah_cong_avoid | -38
= -104 bytes total
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to what is done elsewhere in TCP code when double
state checks are being done.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Redundant checks made indentation impossible to follow.
However, it might be useful to make this ca_state+is_sack
indexed array.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some comment about its current state added. So far I have
seen very few cases where the thing is actually useful,
usually just marginally (though admittedly I don't usually
see top of window losses where it seems possible that there
could be some gain), instead, more often the cases suffer
from L-marking spike which is certainly not desirable
(I'll bury improving it to my todo list, but on a low
prio position).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arnd Hannemann <hannemann@nets.rwth-aachen.de> noticed and was
puzzled by the fact that !tcp_is_fack(tp) leads to early return
near the beginning and the later on tcp_is_fack(tp) was still
used in an if condition. The later check was a left-over from
RFC3517 SACK stuff (== !tcp_is_fack(tp) behavior nowadays) as
there wasn't clear way how to handle this particular check
cheaply in the spirit of RFC3517 (using only SACK blocks, not
holes + SACK blocks as with FACK). I sort of left it there as
a reminder but since it's confusing other people just remove
it and comment the missing-feature stuff instead.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Cc: Arnd Hannemann <hannemann@nets.rwth-aachen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
If cur_mss grew very recently so that the previously G/TSOed skb
now fits well into a single segment it would get send up in
parts unless we calculate # of segments again. This corner-case
could happen eg. after mtu probe completes or less than
previously sack blocks are required for the opposite direction.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
1) We didn't remove any skbs, so no need to handle stale refs.
2) scoreboard_skb_hint is trivial, no timestamps were changed
so no need to clear that one
3) lost_skb_hint needs tweaking similar to that of
tcp_sacktag_one().
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
If skb can be sent right away, we certainly should do that
if it's in the middle of the queue because it won't get
more data into it.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is possible that lost_cnt_hint gets underflow in
tcp_clean_rtx_queue because the cumulative ACK can cover
the segment where lost_skb_hint points to only partially,
which means that the hint is not cleared, opposite to what
my (earlier) comment claimed.
Also I don't agree what I ended up writing about non-trivial
case there to be what I intented to say. It was not supposed
to happen that the hint won't get cleared and we underflow
in any scenario.
In general, this is quite hard to trigger in practice.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Backtracking to sacked skbs is a horrible performance killer
since the hint cannot be advanced successfully past them...
...And it's totally unnecessary too.
In theory this is 2.6.27..28 regression but I doubt anybody
can make .28 to have worse performance because of other TCP
improvements.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's conflicting assumptions in shifting, the caller assumes
that dupsack results in S'ed skbs (or a part of it) for sure but
never gave a hint to tcp_sacktag_one when dsack is actually in
use. Thus DSACK retrans_out -= pcount was not taken and the
counter became out of sync. Remove obstacle from that information
flow to get DSACKs accounted in tcp_sacktag_one as expected.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Tested-by: Denys Fedoryshchenko <denys@visp.net.lb>
Signed-off-by: David S. Miller <davem@davemloft.net>
Impact: Attribute function with __releases(...)
Fix this sparse warning:
net/ipv4/inet_fragment.c:276:35: warning: context imbalance in 'inet_frag_find' - unexpected unlock
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Impact: build fix
API was changed, but not all usage sites were converted:
net/ipv4/route.c: In function ‘ip_rt_init’:
net/ipv4/route.c:3379: error: too few arguments to function ‘__alloc_percpu’
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The functions time_before is more robust for comparing
jiffies against other values.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The functions time_before is more robust for comparing
jiffies against other values.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch changes the return value of nlmsg_notify() as follows:
If NETLINK_BROADCAST_ERROR is set by any of the listeners and
an error in the delivery happened, return the broadcast error;
else if there are no listeners apart from the socket that
requested a change with the echo flag, return the result of the
unicast notification. Thus, with this patch, the unicast
notification is handled in the same way of a broadcast listener
that has set the NETLINK_BROADCAST_ERROR socket flag.
This patch is useful in case that the caller of nlmsg_notify()
wants to know the result of the delivery of a netlink notification
(including the broadcast delivery) and take any action in case
that the delivery failed. For example, ctnetlink can drop packets
if the event delivery failed to provide reliable logging and
state-synchronization at the cost of dropping packets.
This patch also modifies the rtnetlink code to ignore the return
value of rtnl_notify() in all callers. The function rtnl_notify()
(before this patch) returned the error of the unicast notification
which makes rtnl_set_sk_err() reports errors to all listeners. This
is not of any help since the origin of the change (the socket that
requested the echoing) notices the ENOBUFS error if the notification
fails and should resync itself.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The IP_ADVANCED_ROUTER Kconfig describes the rp_filter
proc option. Recent changes added a loose mode.
Instead of documenting this change too places, refer to
the document describing it:
Documentation/networking/ip-sysctl.txt
I'm considering moving the rp_filter description away
from the Kconfig file into ip-sysctl.txt.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
netns: fix double free at netns creation
veth : add the set_mac_address capability
sunlance: Beyond ARRAY_SIZE of ib->btx_ring
sungem: another error printed one too early
ISDN: fix sc/shmem printk format warning
SMSC: timeout reaches -1
smsc9420: handle magic field of ethtool_eeprom
sundance: missing parentheses?
smsc9420: fix another postfixed timeout
wimax/i2400m: driver loads firmware v1.4 instead of v1.3
vlan: Update skb->mac_header in __vlan_put_tag().
cxgb3: Add support for PCI ID 0x35.
tcp: remove obsoleted comment about different passes
TG3: &&/|| confusion
ATM: misplaced parentheses?
net/mv643xx: don't disable the mib timer too early and lock properly
net/mv643xx: use GFP_ATOMIC while atomic
atl1c: Atheros L1C Gigabit Ethernet driver
net: Kill skb_truesize_check(), it only catches false-positives.
net: forcedeth: Fix wake-on-lan regression
To remove the possibility of packets flying around when network
devices are being cleaned up use reisger_pernet_subsys instead of
register_pernet_device.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Recently I had a kernel panic in icmp_send during a network namespace
cleanup. There were packets in the arp queue that failed to be sent
and we attempted to generate an ICMP host unreachable message, but
failed because icmp_sk_exit had already been called.
The network devices are removed from a network namespace and their
arp queues are flushed before we do attempt to shutdown subsystems
so this error should have been impossible.
It turns out icmp_init is using register_pernet_device instead
of register_pernet_subsys. Which resulted in icmp being shut down
while we still had the possibility of packets in flight, making
a nasty NULL pointer deference in interrupt context possible.
Changing this to register_pernet_subsys fixes the problem in
my testing.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
While going through net/ipv4/Kconfig cleanup whitespaces.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The reverse path filter (rp_filter) will NOT get enabled
when enabling forwarding. Read the code and tested in
in practice.
Most distributions do enable it in startup scripts.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Get rid of compile warning about non-const format
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend existing reverse path filter option to allow strict or loose
filtering. (See http://en.wikipedia.org/wiki/Reverse_path_filtering).
For compatibility with existing usage, the value 1 is chosen for strict mode
and 2 for loose mode.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The CIPSO protocol engine incorrectly stated that the FIPS-188 specification
could be found in the kernel's Documentation directory. This patch corrects
that by removing the comment and directing users to the FIPS-188 documented
hosted online. For the sake of completeness I've also included a link to the
CIPSO draft specification on the NetLabel website.
Thanks to Randy Dunlap for spotting the error and letting me know.
Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
Our TCP stack does not set the urgent flag if the urgent pointer
does not fit in 16 bits, i.e., if it is more than 64K from the
sequence number of a packet.
This behaviour is different from the BSDs, and clearly contradicts
the purpose of urgent mode, which is to send the notification
(though not necessarily the associated data) as soon as possible.
Our current behaviour may in fact delay the urgent notification
indefinitely if the receiver window does not open up.
Simply matching BSD however may break legacy applications which
incorrectly rely on the out-of-band delivery of urgent data, and
conversely the in-band delivery of non-urgent data.
Alexey Kuznetsov suggested a safe solution of following BSD only
if the urgent pointer itself has not yet been transmitted. This
way we guarantee that when the remote end sees the packet with
non-urgent data marked as urgent due to wrap-around we would have
advanced the urgent pointer beyond, either to the actual urgent
data or to an as-yet untransmitted packet.
The only potential downside is that applications on the remote
end may see multiple SIGURG notifications. However, this would
occur anyway with other TCP stacks. More importantly, the outcome
of such a duplicate notification is likely to be harmless since
the signal itself does not carry any information other than the
fact that we're in urgent mode.
Thanks to Ilpo Järvinen for fixing a critical bug in this and
Jeff Chua for reporting that bug.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
While doing oprofile tests I noticed two loops are not properly unrolled by gcc
Using a hand coded unrolled loop provides nice speedup : ipt_do_table
credited of 2.52 % of cpu instead of 3.29 % in tbench.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
The reader/writer lock in ip_tables is acquired in the critical path of
processing packets and is one of the reasons just loading iptables can cause
a 20% performance loss. The rwlock serves two functions:
1) it prevents changes to table state (xt_replace) while table is in use.
This is now handled by doing rcu on the xt_table. When table is
replaced, the new table(s) are put in and the old one table(s) are freed
after RCU period.
2) it provides synchronization when accesing the counter values.
This is now handled by swapping in new table_info entries for each cpu
then summing the old values, and putting the result back onto one
cpu. On a busy system it may cause sampling to occur at different
times on each cpu, but no packet/byte counts are lost in the process.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Sucessfully tested on my dual quad core machine too, but iptables only (no ipv6 here)
BTW, my new "tbench 8" result is 2450 MB/s, (it was 2150 MB/s not so long ago)
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This prepares for a real __alloc_percpu, by adding an alignment argument.
Only one place uses __alloc_percpu directly, and that's for a string.
tj: af_inet also uses __alloc_percpu(), update it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Jens Axboe <axboe@kernel.dk>
Concern has been expressed about the changing Kconfig options.
Provide the old options that forward-select.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This is obsolete since the passes got combined.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Suggested by: James King <t.james.king@gmail.com>
Similarly to commit c9fd496809, merge
TTL and HL. Since HL does not depend on any IPv6-specific function,
no new module dependencies would arise.
With slight adjustments to the Kconfig help text.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
x86 and powerpc can perform long word accesses in an efficient maner.
We can use this to unroll two loops in arp_packet_match(), to
perform arithmetic on long words instead of bytes. This is a win
on x86_64 for example.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Later patches change the locking on xt_table and the initialization of
the lock element is not needed since the lock is always initialized in
xt_table_register anyway.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Impact: syntax fix
Interestingly enough this compiles w/o any complaints:
orphans = percpu_counter_sum_positive(&tcp_orphan_count),
sockets = percpu_counter_sum_positive(&tcp_sockets_allocated),
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instructions for time stamping outgoing packets are take from the
socket layer and later copied into the new skb.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gro: Optimise TCP packet reception
As this function can be called more than half a million times for
10GbE, it's important to optimise it as much as we can.
This patch uses bit ops to logical ops, as well as open coding
memcmp to exploit alignment properties.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
As this function can be called more than half a million times for
10GbE, it's important to optimise it as much as we can.
This patch does some obvious changes to use 2-byte and 4-byte
operations instead of byte-oriented ones where possible. Bit
ops are also used to replace logical ops to reduce branching.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Like the UDP header fix, pskb_may_pull() can potentially
alter the SKB buffer. Thus the saddr and daddr, pointers
may point to the old skb->data buffer.
I haven't seen corruptions, as its only seen if the old
skb->data buffer were reallocated by another user and
written into very quickly (or poison'd by SLAB debugging).
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The UDP header pointer assignment must happen after calling
pskb_may_pull(). As pskb_may_pull() can potentially alter the SKB
buffer.
This was exposted by running multicast traffic through the NIU driver,
as it won't prepull the protocol headers into the linear area on
receive.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 93821778de (udp: Fix rcv socket
locking) accidentally removed sk_drops increments for UDP IPV4
sockets.
This field can be used to detect incorrect sizing of socket receive
buffers.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_alloc now sets sk_family so this is redundant. In fact it caught
my eye because sock_init_data already uses sk_family so this is too
late anyway.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
And switch bsockets to atomic_t since it might be changed in parallel.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Stephen Hemminger <shemminger@vyatta.com>
Fix regression introduced by a9d8f9110d
("inet: Allowing more than 64k connections and heavily optimize
bind(0) time.")
Based upon initial patches and feedback from Evegniy Polyakov and
Eric Dumazet.
From Eric Dumazet:
--------------------
Also there might be a problem at line 175
if (sk->sk_reuse && sk->sk_state != TCP_LISTEN && --attempts >= 0) {
spin_unlock(&head->lock);
goto again;
If we entered inet_csk_get_port() with a non null snum, we can "goto again"
while it was not expected.
--------------------
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds another inet device option to enable gratuitous ARP
when device is brought up or address change. This is handy for
clusters or virtualization.
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Base versions handle constant folding now.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Unfortunately simplicity isn't always the best. The fraginfo
interface turned out to be suboptimal. The problem was quite
obvious. For every packet, we have to copy the headers from
the frags structure into skb->head, even though for 99% of the
packets this part is immediately thrown away after the merge.
LRO didn't have this problem because it directly read the headers
from the frags structure.
This patch attempts to address this by creating an interface
that allows GRO to access the headers in the first frag without
having to copy it. Because all drivers that use frags place the
headers in the first frag this optimisation should be enough.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_splice_data_recv has two lengths to consider: the len parameter it
gets from tcp_read_sock, which specifies the amount of data in the skb,
and rd_desc->count, which is the amount of data the splice caller still
wants. Currently it passes just the latter to skb_splice_bits, which then
splices min(rd_desc->count, skb->len - offset) bytes.
Most of the time this is fine, except when the skb contains urgent data.
In that case len goes only up to the urgent byte and is less than
skb->len - offset. By ignoring len tcp_splice_data_recv may a) splice
data tcp_read_sock told it not to, b) return to tcp_read_sock a value > len.
Now, tcp_read_sock doesn't handle used > len and leaves the socket in a
bad state (both sk_receive_queue and copied_seq are bad at that point)
resulting in duplicated data and corruption.
Fix by passing min(rd_desc->count, len) to skb_splice_bits.
Signed-off-by: Dimitris Michailidis <dm@chelsio.com>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 9088c56095
(udp: Improve port randomization) introduced a regression for UDP bind() syscall
to null port (getting a random port) in case lot of ports are already in use.
This is because we do about 28000 scans of very long chains (220 sockets per chain),
with many spin_lock_bh()/spin_unlock_bh() calls.
Fix this using a bitmap (64 bytes for current value of UDP_HTABLE_SIZE)
so that we scan chains at most once.
Instead of 250 ms per bind() call, we get after patch a time of 2.9 ms
Based on a report from Vitaly Mayatskikh
Reported-by: Vitaly Mayatskikh <v.mayatskih@gmail.com>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Tested-by: Vitaly Mayatskikh <v.mayatskih@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of keeping candidate tunnel device from all categories,
keep only one candidate with best score. This optimizes stack
usage and speeds up exit code.
Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
This last patch makes the appropriate changes to use and propagate the
network namespace where needed in IPv4 multicast routing code.
This consists mainly in replacing all the remaining init_net occurences
with current netns pointer retrieved from sockets, net devices or
mfc_caches depending on the routines' contexts.
Some routines receive a new 'struct net' parameter to propagate the current
netns:
* vif_add/vif_delete
* ipmr_new_tunnel
* mroute_clean_tables
* ipmr_cache_find
* ipmr_cache_report
* ipmr_cache_unresolved
* ipmr_mfc_add/ipmr_mfc_delete
* ipmr_get_route
* rt_fill_info (in route.c)
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Declare IPv4 multicast forwarding /proc/net entries per-namespace:
/proc/net/ip_mr_vif
/proc/net/ip_mr_cache
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv4 multicast routing netns-aware.
Declare variable 'reg_vif_num' per-namespace, move into struct netns_ipv4.
At the moment, this variable is only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv4 multicast routing netns-aware.
Declare IPv multicast routing variables 'mroute_do_assert' and
'mroute_do_pim' per-namespace in struct netns_ipv4.
At the moment, these variables are only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv4 multicast routing netns-aware.
Declare variable cache_resolve_queue_len per-namespace: move it into
struct netns_ipv4.
This variable counts the number of unresolved cache entries queued in the
list mfc_unres_queue. This list is kept global to all netns as the number
of entries per namespace is limited to 10 (hardcoded in routine
ipmr_cache_unresolved).
Entries belonging to different namespaces in mfc_unres_queue will be
identified by matching the mfc_net member introduced previously in
struct mfc_cache.
Keeping this list global to all netns, also allows us to keep a single
timer (ipmr_expire_timer) to handle their expiration.
In some places cache_resolve_queue_len value was tested for arming
or deleting the timer. These tests were equivalent to testing
mfc_unres_queue value instead and are replaced in this patch.
At the moment, cache_resolve_queue_len is only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv4 multicast routing netns-aware.
Dynamically allocate IPv4 multicast forwarding cache, mfc_cache_array,
and move it to struct netns_ipv4.
At the moment, mfc_cache_array is only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch stores into struct mfc_cache the network namespace each
mfc_cache belongs to. The new member is mfc_net.
mfc_net is assigned at cache allocation and doesn't change during
the rest of the cache entry life.
A new net parameter is added to ipmr_cache_alloc/ipmr_cache_alloc_unres.
This will help to retrieve the current netns around the IPv4 multicast
routing code.
At the moment, all mfc_cache are allocated in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv6 multicast routing netns-aware.
Dynamically allocate interface table vif_table and move it to
struct netns_ipv4, and update MIF_EXISTS() macro.
At the moment, vif_table is only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preliminary work to make IPv4 multicast routing netns-aware.
Make IPv4 multicast routing mroute_socket per-namespace,
moves it into struct netns_ipv4.
At the moment, mroute_socket is only referenced in init_net.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check the device on receive path and allow otherwise identical devices
as long as the physical device differs.
This is useful for NBMA tunnels, where you want to use different gre IP
for each public IP available via different physical devices.
Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
With simple extension to the binding mechanism, which allows to bind more
than 64k sockets (or smaller amount, depending on sysctl parameters),
we have to traverse the whole bind hash table to find out empty bucket.
And while it is not a problem for example for 32k connections, bind()
completion time grows exponentially (since after each successful binding
we have to traverse one bucket more to find empty one) even if we start
each time from random offset inside the hash table.
So, when hash table is full, and we want to add another socket, we have
to traverse the whole table no matter what, so effectivelly this will be
the worst case performance and it will be constant.
Attached picture shows bind() time depending on number of already bound
sockets.
Green area corresponds to the usual binding to zero port process, which
turns on kernel port selection as described above. Red area is the bind
process, when number of reuse-bound sockets is not limited by 64k (or
sysctl parameters). The same exponential growth (hidden by the green
area) before number of ports reaches sysctl limit.
At this time bind hash table has exactly one reuse-enbaled socket in a
bucket, but it is possible that they have different addresses. Actually
kernel selects the first port to try randomly, so at the beginning bind
will take roughly constant time, but with time number of port to check
after random start will increase. And that will have exponential growth,
but because of above random selection, not every next port selection
will necessary take longer time than previous. So we have to consider
the area below in the graph (if you could zoom it, you could find, that
there are many different times placed there), so area can hide another.
Blue area corresponds to the port selection optimization.
This is rather simple design approach: hashtable now maintains (unprecise
and racely updated) number of currently bound sockets, and when number
of such sockets becomes greater than predefined value (I use maximum
port range defined by sysctls), we stop traversing the whole bind hash
table and just stop at first matching bucket after random start. Above
limit roughly corresponds to the case, when bind hash table is full and
we turned on mechanism of allowing to bind more reuse-enabled sockets,
so it does not change behaviour of other sockets.
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Tested-by: Denys Fedoryschenko <denys@visp.net.lb>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we get a GSO packet from an untrusted source, we need to
ensure that it is sufficiently long so that we don't end up
crashing.
Based on discovery and patch by Ian Campbell.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As spotted by Willy Tarreau, current splice() from tcp socket to pipe is not
optimal. It processes at most one segment per call.
This results in low performance and very high overhead due to syscall rate
when splicing from interfaces which do not support LRO.
Willy provided a patch inside tcp_splice_read(), but a better fix
is to let tcp_read_sock() process as many segments as possible, so
that tcp_rcv_space_adjust() and tcp_cleanup_rbuf() are called less
often.
With this change, splice() behaves like tcp_recvmsg(), being able
to consume many skbs in one system call. With typical 1460 bytes
of payload per frame, that means splice(SPLICE_F_NONBLOCK) can return
16*1460 = 23360 bytes.
Signed-off-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
An old bug crept back into the ICMP/ICMPv6 conntrack protocols: the timeout
values are defined as unsigned longs, the sysctl's maxsize is set to
sizeof(unsigned int). Use unsigned int for the timeout values as in the
other conntrack protocols.
Reported-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't spam logs for locally generated short packets. these can only
be generated by root.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (22 commits)
ioat: fix self test for multi-channel case
dmaengine: bump initcall level to arch_initcall
dmaengine: advertise all channels on a device to dma_filter_fn
dmaengine: use idr for registering dma device numbers
dmaengine: add a release for dma class devices and dependent infrastructure
ioat: do not perform removal actions at shutdown
iop-adma: enable module removal
iop-adma: kill debug BUG_ON
iop-adma: let devm do its job, don't duplicate free
dmaengine: kill enum dma_state_client
dmaengine: remove 'bigref' infrastructure
dmaengine: kill struct dma_client and supporting infrastructure
dmaengine: replace dma_async_client_register with dmaengine_get
atmel-mci: convert to dma_request_channel and down-level dma_slave
dmatest: convert to dma_request_channel
dmaengine: introduce dma_request_channel and private channels
net_dma: convert to dma_find_channel
dmaengine: provide a common 'issue_pending_all' implementation
dmaengine: centralize channel allocation, introduce dma_find_channel
dmaengine: up-level reference counting to the module level
...
This patch adds GRO support for TCP over IPv6. The code is exactly
the same as the IPv4 version except for the pseudo-header checksum
computation.
Note that I've removed the unused tcphdr argument from tcp_v6_check
rather than invent a bogus value for GRO.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the general-purpose channel allocation provided by dmaengine.
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Simply, if a client wants any dmaengine channel then prevent all dmaengine
modules from being removed. Once the clients are done re-enable module
removal.
Why?, beyond reducing complication:
1/ Tracking reference counts per-transaction in an efficient manner, as
is currently done, requires a complicated scheme to avoid cache-line
bouncing effects.
2/ Per-transaction ref-counting gives the false impression that a
dma-driver can be gracefully removed ahead of its user (net, md, or
dma-slave)
3/ None of the in-tree dma-drivers talk to hot pluggable hardware, but
if such an engine were built one day we still would not need to notify
clients of remove events. The driver can simply return NULL to a
->prep() request, something that is much easier for a client to handle.
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In splice TCP receive, the SPLICE_F_NONBLOCK flag is used
to compute the "timeo" value. So checking it again inside
of the main receive loop to trigger -EAGAIN processing is
entirely unnecessary.
Noticed by Jarek P. and Lennert Buytenhek.
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, setting SPLICE_F_NONBLOCK on splice from a TCP socket
results in masking of EOF (RDHUP) and error conditions on the socket
by an -EAGAIN return. Move the NONBLOCK check in tcp_splice_read()
to be after the EOF and error checks to fix this.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to allow GRO packets without frag_list at all, we need to
store the MSS in the packet itself. The obvious place is gso_size.
The only thing to watch out for is if the packet ends up not being
GRO then we need to clear gso_size before pushing the packet into
the stack.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update the NetLabel kernel API to expose the new features added in kernel
releases 2.6.25 and 2.6.28: the static/fallback label functionality and network
address based selectors.
Signed-off-by: Paul Moore <paul.moore@hp.com>
When we converted the protocol atomic counters such as the orphan
count and the total socket count deadlocks were introduced due to
the mismatch in BH status of the spots that used the percpu counter
operations.
Based on the diagnosis and patch by Peter Zijlstra, this patch
fixes these issues by disabling BH where we may be in process
context.
Reported-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In future all cpumask ops will only be valid (in general) for bit
numbers < nr_cpu_ids. So use that instead of NR_CPUS in iterators
and other comparisons.
This is always safe: no cpu number can be >= nr_cpu_ids, and
nr_cpu_ids is initialized to NR_CPUS at boot.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1429 commits)
net: Allow dependancies of FDDI & Tokenring to be modular.
igb: Fix build warning when DCA is disabled.
net: Fix warning fallout from recent NAPI interface changes.
gro: Fix potential use after free
sfc: If AN is enabled, always read speed/duplex from the AN advertising bits
sfc: When disabling the NIC, close the device rather than unregistering it
sfc: SFT9001: Add cable diagnostics
sfc: Add support for multiple PHY self-tests
sfc: Merge top-level functions for self-tests
sfc: Clean up PHY mode management in loopback self-test
sfc: Fix unreliable link detection in some loopback modes
sfc: Generate unique names for per-NIC workqueues
802.3ad: use standard ethhdr instead of ad_header
802.3ad: generalize out mac address initializer
802.3ad: initialize ports LACPDU from const initializer
802.3ad: remove typedef around ad_system
802.3ad: turn ports is_individual into a bool
802.3ad: turn ports is_enabled into a bool
802.3ad: make ntt bool
ixgbe: Fix set_ringparam in ixgbe to use the same memory pools.
...
Fixed trivial IPv4/6 address printing conflicts in fs/cifs/connect.c due
to the conversion to %pI (in this networking merge) and the addition of
doing IPv6 addresses (from the earlier merge of CIFS).
This patch removes a useless ret variable from the IPv4 ESP/UDP
decapsulation code.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Our TCP stack does not set the urgent flag if the urgent pointer
does not fit in 16 bits, i.e., if it is more than 64K from the
sequence number of a packet.
This behaviour is different from the BSDs, and clearly contradicts
the purpose of urgent mode, which is to send the notification
(though not necessarily the associated data) as soon as possible.
Our current behaviour may in fact delay the urgent notification
indefinitely if the receiver window does not open up.
Simply matching BSD however may break legacy applications which
incorrectly rely on the out-of-band delivery of urgent data, and
conversely the in-band delivery of non-urgent data.
Alexey Kuznetsov suggested a safe solution of following BSD only
if the urgent pointer itself has not yet been transmitted. This
way we guarantee that when the remote end sees the packet with
non-urgent data marked as urgent due to wrap-around we would have
advanced the urgent pointer beyond, either to the actual urgent
data or to an as-yet untransmitted packet.
The only potential downside is that applications on the remote
end may see multiple SIGURG notifications. However, this would
occur anyway with other TCP stacks. More importantly, the outcome
of such a duplicate notification is likely to be harmless since
the signal itself does not carry any information other than the
fact that we're in urgent mode.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch makes the followinf proc entries per-netns:
/proc/net/igmp
/proc/net/mcfilter
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Looks like everything is already ready.
Required for ebtables(8) for one thing.
Also, required for ipmr per-netns (coming soon). (Benjamin)
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The original message was unhelpful and extremely alarming to our poor
users, despite its charm. Make it less frightening.
Signed-off-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the TCP-specific portion of GRO. The criterion for
merging is extremely strict (the TCP header must match exactly apart
from the checksum) so as to allow refragmentation. Otherwise this
is pretty much identical to LRO, except that we support the merging
of ECN packets.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds GRO support for IPv4.
The criteria for merging is more stringent than LRO, in particular,
we require all fields in the IP header to be identical except for
the length, ID and checksum. In addition, the ID must form an
arithmetic sequence with a difference of one.
The ID requirement might seem overly strict, however, most hardware
TSO solutions already obey this rule. Linux itself also obeys this
whether GSO is in use or not.
In future we could relax this rule by storing the IDs (or rather
making sure that we don't drop them when pulling the aggregate
skb's tail).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit e099a17357
(netfilter: netns nat: per-netns NAT table) renamed the
nat_table from __nat_table to nat_table without updating the
__RW_LOCK_UNLOCKED(__nat_table.lock).
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch addresses a book-keeping issue in tcp_vegas.c. At present
tcp_vegas does separate book-keeping of cwnd based on packet sequence
numbers. A mismatch can develop between this book-keeping and
tp->snd_cwnd due, for example, to delayed acks acking multiple
packets. When vegas transitions to reno operation (e.g. following
loss), then this mismatch leads to incorrect behaviour (akin to a cwnd
backoff). This seems mostly to affect operation at low cwnds where
delayed acking can lead to a significant fraction of cwnd being
covered by a single ack, leading to the book-keeping mismatch. This
patch modifies the congestion avoidance update to avoid the need for
separate book-keeping while leaving vegas congestion avoidance
functionally unchanged. A secondary advantage of this modification is
that the use of fixed-point (via V_PARAM_SHIFT) and 64 bit arithmetic
is no longer necessary, simplifying the code.
Some example test measurements with the patched code (confirming no functional
change in the congestion avoidance algorithm) can be seen at:
http://www.hamilton.ie/doug/vegaspatch/
Signed-off-by: Doug Leith <doug.leith@nuim.ie>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since jiffies is unsigned long, the types get expanded into
that and after long enough time the difference will therefore
always be > 1 (and that probably happens near boot as well as
iirc the first jiffies wrap is scheduler close after boot to
find out problems related to that early).
This was originally noted by Bill Fink in Dec'07 but nobody
never ended fixing it.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_minshall_update is not significant difference since it only
checks for not full-sized skb which is BUG'ed on the push_one
path anyway.
tcp_snd_test is tcp_nagle_test+tcp_cwnd_test+tcp_snd_wnd_test,
just the order changed slightly.
net/ipv4/tcp_output.c:
tcp_snd_test | -89
tcp_mss_split_point | -91
tcp_may_send_now | +53
tcp_cwnd_validate | -98
tso_fragment | -239
__tcp_push_pending_frames | -1340
tcp_push_one | -146
7 functions changed, 53 bytes added, 2003 bytes removed, diff: -1950
net/ipv4/tcp_output.c:
tcp_write_xmit | +1772
1 function changed, 1772 bytes added, diff: +1772
tcp_output.o.new:
8 functions changed, 1825 bytes added, 2003 bytes removed, diff: -178
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are just too many args to some sacktag functions. This
idea was first proposed by David S. Miller around a year ago,
and the current situation is much worse that what it was back
then.
tcp_sacktag_one can be made a bit simpler by returning the
new sacked (it can be achieved with a single variable though
the previous code "caching" sacked into a local variable and
therefore it is not exactly equal but the results will be the
same).
codiff on x86_64
tcp_sacktag_one | -15
tcp_shifted_skb | -50
tcp_match_skb_to_sack | -1
tcp_sacktag_walk | -64
tcp_sacktag_write_queue | -59
tcp_urg | +1
tcp_event_data_recv | -1
7 functions changed, 1 bytes added, 190 bytes removed, diff: -189
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
I noticed that since skb->len has nothing to do with actual segment
length with gso, we need to figure it out separately, reuse
a function from the recent shifting stuff (generalize it).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
S|R won't result in S if just SACK is received. DSACK is
another story (but it is covered correctly already).
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes a bug in tcp_vegas.c. At the moment this code leaves
ssthresh untouched. However, this means that the vegas congestion
control algorithm is effectively unable to reduce cwnd below the
ssthresh value (if the vegas update lowers the cwnd below ssthresh,
then slow start is activated to raise it back up). One example where
this matters is when during slow start cwnd overshoots the link
capacity and a flow then exits slow start with ssthresh set to a value
above where congestion avoidance would like to adjust it.
Signed-off-by: Doug Leith <doug.leith@nuim.ie>
Signed-off-by: David S. Miller <davem@davemloft.net>
Today, iproute2 fails to show multicast forwarding unresolved cache
entries while scanning /proc/net/ip_mr_cache.
Indeed, it expects to see -1 in 'Iif' column to identify unresolved
entries but the kernel outputs 65535. It's a signed/unsigned issue:
'Iif', the source interface, is retrieved from member mfc_parent in
struct mfc_cache. mfc_parent is a vifi_t: unsigned short, but is
displayed in ipmr_mfc_seq_show() as "%-3d", signed integer.
In unresolevd entries, the 65535 value (0xFFFF) comes from this define:
#define ALL_VIFS ((vifi_t)(-1))
That may explains why the guy who added support for this in iproute2
thought a -1 should be expected.
I don't know if this must be fixed in kernel or in iproute2. Who is
right? What is the correct API? How was it designed originally?
I let you decide if it should goes in the kernel or be fixed in iproute2.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
/proc/net/ip_mr_cache and /proc/net/ip6_mr_cache displays garbage when
showing unresolved mfc_cache entries.
[root@qemu tests]# cat /proc/net/ip_mr_cache
Group Origin Iif Pkts Bytes Wrong Oifs
014C00EF 010014AC 1 10 10050 0 2:1 3:1
024C00EF 010014AC 65535 514 2 -559067475
The first line is correct. It is a resolved cache entry, 10 packets used it...
The second line represents an unresolved entry, and the columns Pkts(4th),
Bytes(5th) and Wrong(6th) just show garbage.
In struct mfc_cache, there's an union to store data for resolved and
unresolved cases. And what ipmr_mfc_seq_show() is printing in these
columns for the unresolved entries is some bytes from mfc_cache.mfc_un.res.
Bad.
(eg. In our case -559067475 is in fact 0xdead4ead which is the spinlock
magic from mfc_cache.mfc_un.unres.unresolved.lock.magic).
This patch replaces the garbage data written in these columns for the
unresolved entries by '0' (zeros) which is more correct.
This change doesn't break the ABI.
Also, mfc->mfc_un.res.pkt, mfc->mfc_un.res.bytes, mfc->mfc_un.res.wrong_if
are unsigned long.
It applies on top of net-next-2.6.
The patch for net-2.6 is slightly different because of the NIP6_FMT to
%pI6 conversion that was made in the seq_printf.
Changelog:
==========
V2:
* Instead of breaking the ABI by suppressing the columns that have no
meaning for unresolved entries, fill them with 0 values.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
fs/nfsd/nfs4recover.c
Manually fixed above to use new creds API functions, e.g.
nfs4_save_creds().
Signed-off-by: James Morris <jmorris@namei.org>
I should have noticed this earlier... :-) The previous solution
to URG+GSO/TSO will cause SACK block tcp_fragment to do zig-zig
patterns, or even worse, a steep downward slope into packet
counting because each skb pcount would be truncated to pcount
of 2 and then the following fragments of the later portion would
restore the window again.
Basically this reverts "tcp: Do not use TSO/GSO when there is
urgent data" (33cf71cee1). It also removes some unnecessary code
from tcp_current_mss that didn't work as intented either (could
be that something was changed down the road, or it might have
been broken since the dawn of time) because it only works once
urg is already written while this bug shows up starting from
~64k before the urg point.
The retransmissions already are split to mss sized chunks, so
only new data sending paths need splitting in case they have
a segment otherwise suitable for gso/tso. The actually check
can be improved to be more narrow but since this is late -rc
already, I'll postpone thinking the more fine-grained things.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based upon a lockdep report by Alexey Dobriyan.
I checked all per_cpu_counter_xxx() usages in network tree, and I
think all call sites are BH enabled except one in
inet_csk_listen_stop().
commit dd24c00191
(net: Use a percpu_counter for orphan_count)
replaced atomic_t orphan_count to a percpu_counter.
atomic_inc()/atomic_dec() can be called from any context, while
percpu_counter_xxx() should be called from a consistent state.
For orphan_count, this context can be the BH-enabled one.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using one atomic_t per protocol, use a percpu_counter
for "orphan_count", to reduce cache line contention on
heavy duty network servers.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using one atomic_t per protocol, use a percpu_counter
for "sockets_allocated", to reduce cache line contention on
heavy duty network servers.
Note : We revert commit (248969ae31
net: af_unix can make unix_nr_socks visbile in /proc),
since it is not anymore used after sock_prot_inuse_add() addition
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass netns pointer to struct xfrm_policy_afinfo::garbage_collect()
[This needs more thoughts on what to do with dst_ops]
[Currently stub to init_net]
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass netns to xfrm_lookup()/__xfrm_lookup(). For that pass netns
to flow_cache_lookup() and resolver callback.
Take it from socket or netdevice. Stub DECnet to init_net.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To avoid unnecessary complications with passing netns around.
* set once, very early after allocating
* once set, never changes
For a while create every xfrm_state in init_net.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Impact: Optimization
Like done in inet_unhash(), we can avoid taking a chain lock if
socket is not hashed in udp_unhash()
Triggered by close(socket(AF_INET, SOCK_DGRAM, 0));
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch let nfmark to be evaluated for routing decision for OUTPUT
packet, in mangle table, when process paquet in NFQUEUE
Until now, only change (in NFQUEUE process) on fields src_addr,
dest_addr and tos could make netfilter to reevalute the routing.
From: Laurent Licour <laurent@licour.com>
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
The earlier version was just very basic one which is "playing
safe" by always clearing the hints. However, clearing of a hint
is extremely costly operation with large windows, so it must be
avoided at all cost whenever possible, there is a way with
shifting too achieve not-clearing.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
During SACK processing, most of the benefits of TSO are eaten by
the SACK blocks that one-by-one fragment SKBs to MSS sized chunks.
Then we're in problems when cleanup work for them has to be done
when a large cumulative ACK comes. Try to return back to pre-split
state already while more and more SACK info gets discovered by
combining newly discovered SACK areas with the previous skb if
that's SACKed as well.
This approach has a number of benefits:
1) The processing overhead is spread more equally over the RTT
2) Write queue has less skbs to process (affect everything
which has to walk in the queue past the sacked areas)
3) Write queue is consistent whole the time, so no other parts
of TCP has to be aware of this (this was not the case with
some other approach that was, well, quite intrusive all
around).
4) Clean_rtx_queue can release most of the pages using single
put_page instead of previous PAGE_SIZE/mss+1 calls
In case a hole is fully filled by the new SACK block, we attempt
to combine the next skb too which allows construction of skbs
that are even larger than what tso split them to and it handles
hole per on every nth patterns that often occur during slow start
overshoot pretty nicely. Though this to be really useful also
a retransmission would have to get lost since cumulative ACKs
advance one hole at a time in the most typical case.
TODO: handle upwards only merging. That should be rather easy
when segment is fully sacked but I'm leaving that as future
work item (it won't make very large difference anyway since
this current approach already covers quite a lot of normal
cases).
I was earlier thinking of some sophisticated way of tracking
timestamps of the first and the last segment but later on
realized that it won't be that necessary at all to store the
timestamp of the last segment. The cases that can occur are
basically either:
1) ambiguous => no sensible measurement can be taken anyway
2) non-ambiguous is due to reordering => having the timestamp
of the last segment there is just skewing things more off
than does some good since the ack got triggered by one of
the holes (besides some substle issues that would make
determining right hole/skb even harder problem). Anyway,
it has nothing to do with this change then.
I choose to route some abnormal looking cases with goto noop,
some could be handled differently (eg., by stopping the
walking at that skb but again). In general, they either
shouldn't happen at all or are rare enough to make no difference
in practice.
In theory this change (as whole) could cause some macroscale
regression (global) because of cache misses that are taken over
the round-trip time but it gets very likely better because of much
less (local) cache misses per other write queue walkers and the
big recovery clearing cumulative ack.
Worth to note that these benefits would be very easy to get also
without TSO/GSO being on as long as the data is in pages so that
we can merge them. Currently I won't let that happen because
DSACK splitting at fragment that would mess up pcounts due to
sk_can_gso in tcp_set_skb_tso_segs. Once DSACKs fragments gets
avoided, we have some conditions that can be made less strict.
TODO: I will probably have to convert the excessive pointer
passing to struct sacktag_state... :-)
My testing revealed that considerable amount of skbs couldn't
be shifted because they were cloned (most likely still awaiting
tx reclaim)...
[The rest is considering future work instead since I got
repeatably EFAULT to tcpdump's recvfrom when I added
pskb_expand_head to deal with clones, so I separated that
into another, later patch]
...To counter that, I gave up on the fifth advantage:
5) When growing previous SACK block, less allocs for new skbs
are done, basically a new alloc is needed only when new hole
is detected and when the previous skb runs out of frags space
...which now only happens of if reclaim is fast enough to dispose
the clone before the SACK block comes in (the window is RTT long),
otherwise we'll have to alloc some.
With clones being handled I got these numbers (will be somewhat
worse without that), taken with fine-grained mibs:
TCPSackShifted 398
TCPSackMerged 877
TCPSackShiftFallback 320
TCPSACKCOLLAPSEFALLBACKGSO 0
TCPSACKCOLLAPSEFALLBACKSKBBITS 0
TCPSACKCOLLAPSEFALLBACKSKBDATA 0
TCPSACKCOLLAPSEFALLBACKBELOW 0
TCPSACKCOLLAPSEFALLBACKFIRST 1
TCPSACKCOLLAPSEFALLBACKPREVBITS 318
TCPSACKCOLLAPSEFALLBACKMSS 1
TCPSACKCOLLAPSEFALLBACKNOHEAD 0
TCPSACKCOLLAPSEFALLBACKSHIFT 0
TCPSACKCOLLAPSENOOPSEQ 0
TCPSACKCOLLAPSENOOPSMALLPCOUNT 0
TCPSACKCOLLAPSENOOPSMALLLEN 0
TCPSACKCOLLAPSEHOLE 12
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is preparatory work for SACK combiner patch which may
have to count TCP state changes for only a part of the skb
because it will intentionally avoids splitting skb to SACKed
and not sacked parts.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sadly enough, this adds possible divide though we try to avoid
it by checking one mss as common case.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
I knew already when rewriting the sacktag that this condition
was too conservative, change it now since it prevent lot of
useless work (especially in the sack shifter decision code
that is being added by a later patch). This shouldn't change
anything really, just save some processing regardless of the
shifter.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
I always had thought that collapsing up to two at a time was
intentional decision to avoid excessive processing if 1 byte
sized skbs are to be combined for a full mtu, and consecutive
retransmissions would make the size of the retransmittee
double each round anyway, but some recent discussion made me
to understand that was not the case. Thus make collapse work
more and wait less.
It would be possible to take advantage of the shifting
machinery (added in the later patch) in the case of paged
data but that can be implemented on top of this change.
tcp_skb_is_last check is now provided by the loop.
I tested a bit (ss-after-idle-off, fill 4096x4096B xfer,
10s sleep + 4096 x 1byte writes while dropping them for
some a while with netem):
. 16774097:16775545(1448) ack 1 win 46
. 16775545:16776993(1448) ack 1 win 46
. ack 16759617 win 2399
P 16776993:16777217(224) ack 1 win 46
. ack 16762513 win 2399
. ack 16765409 win 2399
. ack 16768305 win 2399
. ack 16771201 win 2399
. ack 16774097 win 2399
. ack 16776993 win 2399
. ack 16777217 win 2399
P 16777217:16777257(40) ack 1 win 46
. ack 16777257 win 2399
P 16777257:16778705(1448) ack 1 win 46
P 16778705:16780153(1448) ack 1 win 46
FP 16780153:16781313(1160) ack 1 win 46
. ack 16778705 win 2399
. ack 16780153 win 2399
F 1:1(0) ack 16781314 win 2399
While without drop-all period I get this:
. 16773585:16775033(1448) ack 1 win 46
. ack 16764897 win 9367
. ack 16767793 win 9367
. ack 16770689 win 9367
. ack 16773585 win 9367
. 16775033:16776481(1448) ack 1 win 46
P 16776481:16777217(736) ack 1 win 46
. ack 16776481 win 9367
. ack 16777217 win 9367
P 16777217:16777218(1) ack 1 win 46
P 16777218:16777219(1) ack 1 win 46
P 16777219:16777220(1) ack 1 win 46
...
P 16777247:16777248(1) ack 1 win 46
. ack 16777218 win 9367
. ack 16777219 win 9367
...
. ack 16777233 win 9367
. ack 16777248 win 9367
P 16777248:16778696(1448) ack 1 win 46
P 16778696:16780144(1448) ack 1 win 46
FP 16780144:16781313(1169) ack 1 win 46
. ack 16780144 win 9367
F 1:1(0) ack 16781314 win 9367
The window seems to be 30-40 segments, which were successfully
combined into: P 16777217:16777257(40) ack 1 win 46
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can reduce pressure on dst entry refcount that slowdown UDP transmit
path on SMP machines. This pressure is visible on RTP servers when
delivering content to mediagateways, especially big ones, handling
thousand of streams. Several cpus send UDP frames to the same
destination, hence use the same dst entry.
This patch makes ip_push_pending_frames() steal the refcount its
callers had to take when filling inet->cork.dst.
This doesnt avoid all refcounting, but still gives speedups on SMP,
on UDP/RAW transmit path.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can reduce pressure on dst entry refcount that slowdown UDP transmit
path on SMP machines. This pressure is visible on RTP servers when
delivering content to mediagateways, especially big ones, handling
thousand of streams. Several cpus send UDP frames to the same
destination, hence use the same dst entry.
This patch makes ip_append_data() eventually steal the refcount its
callers had to take on the dst entry.
This doesnt avoid all refcounting, but still gives speedups on SMP,
on UDP/RAW transmit path
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The rule of calling sock_prot_inuse_add() is that BHs must
be disabled. Some new calls were added where this was not
true and this tiggers warnings as reported by Ilpo.
Fix this by adding explicit BH disabling around those call sites,
or moving sock_prot_inuse_add() call inside an existing BH disabled
section.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is the last step to be able to perform full RCU lookups
in __inet_lookup() : After established/timewait tables, we
add RCU lookups to listening hash table.
The only trick here is that a socket of a given type (TCP ipv4,
TCP ipv6, ...) can now flight between two different tables
(established and listening) during a RCU grace period, so we
must use different 'nulls' end-of-chain values for two tables.
We define a large value :
#define LISTENING_NULLS_BASE (1U << 29)
So that slots in listening table are guaranteed to have different
end-of-chain values than slots in established table. A reader can
still detect it finished its lookup in the right chain.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
1) Use eq_net() in inet_netns_ok() to speedup socket creation if
!CONFIG_NET_NS
2) Reorder the tests about inet_ehash_secret generation (once only)
Use the unlikely() macro when testing if inet_ehash_secret already
generated.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes http://bugzilla.kernel.org/show_bug.cgi?id=12014
Since most (if not all) implementations of TSO and even the in-kernel
software GSO do not update the urgent pointer when splitting a large
segment, it is necessary to turn off TSO/GSO for all outgoing traffic
with the URG pointer set.
Looking at tcp_current_mss (and the preceding comment) I even think
this was the original intention. However, this approach is insufficient,
because TSO/GSO is turned off only for newly created frames, not for
frames which were already pending at the arrival of a message with
MSG_OOB set. These frames were created when TSO/GSO was enabled,
so they may be large, and they will have the urgent pointer set
in tcp_transmit_skb().
With this patch, such large packets will be fragmented again before
going to the transmit routine.
As a side note, at least the following NICs are known to screw up
the urgent pointer in the TCP header when doing TSO:
Intel 82566MM (PCI ID 8086:1049)
Intel 82566DC (PCI ID 8086:104b)
Intel 82541GI (PCI ID 8086:1076)
Broadcom NetXtreme II BCM5708 (PCI ID 14e4:164c)
Signed-off-by: Petr Tesarik <ptesarik@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now TCP & DCCP use RCU lookups, we can convert ehash rwlocks to spinlocks.
/proc/net/tcp and other seq_file 'readers' can safely be converted to 'writers'.
This should speedup writers, since spin_lock()/spin_unlock()
only use one atomic operation instead of two for write_lock()/write_unlock()
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert to network device ops. Needed to change to directly call
the init routine since two sides share same ops. In the process
found by inspection a device ref count leak if register_netdevice failed.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert to new network device ops interface.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller
Robert Olsson
Rami Rosen
Eric Dumazet
Chris Friesen
Eric W. Biederman
Frans Pop
Ilpo Järvinen
Jan Engelhardt
Arnaldo Carvalho de Melo
Shan Wei
Satoru SATOH
Stephen Hemminger
Herbert Xu
Anton Blanchard
Neil Horman
Florian Westphal
Patrick McHardy
Vlad Yasevich
Linus Torvalds
Paul Moore
Ingo Molnar
Holger Eitzenberger
Vitaly Mayatskikh
John Dykstra
Jorge Boncompte [DTI2]
Scott James Remnant
Pablo Neira Ayuso
Stephen Hemminger
Eric Leblond
Roel Kluin
Denys Fedoryshchenko
Eric W. Biederman
Hantzis Fotis
Hannes Eder
Wei Yongjun
Joe Perches
Jesper Dangaard Brouer
Jesper Dangaard Brouer
Rusty Russell
Thomas Gleixner
Patrick Ohly
Stephen Hemminger
Harvey Harrison
Benjamin Zores
Dimitris Michailidis
Timo Teras
Benjamin Thery
Evgeniy Polyakov
Willy Tarreau
James Morris
Dan Williams
Lennert Buytenhek
Alexey Dobriyan
Matt Mackall
Steven Rostedt
Doug Leith
Petr Tesarik