Based upon a report and initial patch by Peter Lieven.
tcp4_md5sig_key and tcp6_md5sig_key need to start with
the exact same members as tcp_md5sig_key. Because they
are both cast to that type by tcp_v{4,6}_md5_do_lookup().
Unfortunately tcp{4,6}_md5sig_key use a u16 for the key
length instead of a u8, which is what tcp_md5sig_key
uses. This just so happens to work by accident on
little-endian, but on big-endian it doesn't.
Instead of casting, just place tcp_md5sig_key as the first member of
the address-family specific structures, adjust the access sites, and
kill off the ugly casts.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch cleans up duplicate includes in
net/ipv6/
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
As discovered by Evegniy Polyakov, if we try to sendmsg after
a connection reset, we can do incredibly stupid things.
The core issue is that inet_sendmsg() tries to autobind the
socket, but we should never do that for TCP. Instead we should
just go straight into TCP's sendmsg() code which will do all
of the necessary state and pending socket error checks.
TCP's sendpage already directly vectors to tcp_sendpage(), so this
merely brings sendmsg() in line with that.
Signed-off-by: David S. Miller <davem@davemloft.net>
After clearing all passwords for IPv6 peers, we need to
set allocation count to zero as well as we free the storage.
Otherwise, we panic when a user trys to (re)add a password.
Discovered and fixed by MIYAJIMA Mitsuharu <miyajima.mitsuharu@anchor.jp>.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based on <draft-ietf-ipv6-deprecate-rh0-00.txt>.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current IPSEC rule resolution behavior we have does not work for a
lot of people, even though technically it's an improvement from the
-EAGAIN buisness we had before.
Right now we'll block until the key manager resolves the route. That
works for simple cases, but many folks would rather packets get
silently dropped until the key manager resolves the IPSEC rules.
We can't tell these folks to "set the socket non-blocking" because
they don't have control over the non-block setting of things like the
sockets used to resolve DNS deep inside of the resolver libraries in
libc.
With that in mind I coded up the patch below with some help from
Herbert Xu which provides packet-drop behavior during larval state
resolution, controllable via sysctl and off by default.
This lays the framework to either:
1) Make this default at some point or...
2) Move this logic into xfrm{4,6}_policy.c and implement the
ARP-like resolution queue we've all been dreaming of.
The idea would be to queue packets to the policy, then
once the larval state is resolved by the key manager we
re-resolve the route and push the packets out. The
packets would timeout if the rule didn't get resolved
in a certain amount of time.
Signed-off-by: David S. Miller <davem@davemloft.net>
When a transmitted packet is looped back directly, CHECKSUM_PARTIAL
maps to the semantics of CHECKSUM_UNNECESSARY. Therefore we should
treat it as such in the stack.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The skb transport pointer is currently used to specify the start
of the checksum region for transmit checksum offload. Unfortunately,
the same pointer is also used during receive side processing.
This creates a problem when we want to retransmit a received
packet with partial checksums since the skb transport pointer
would be overwritten.
This patch solves this problem by creating a new 16-bit csum_start
offset value to replace the skb transport header for the purpose
of checksums. This offset is calculated from skb->head so that
it does not have to change when skb->data changes.
No extra space is required since csum_offset itself fits within
a 16-bit word so we can use the other 16 bits for csum_start.
For backwards compatibility, just before we push a packet with
partial checksums off into the device driver, we set the skb
transport header to what it would have been under the old scheme.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ip_hdrlen() buddy, created to reduce the number of skb->h.th-> uses and to
avoid the longer, open coded equivalent.
Ditched a no-op in bnx2 in the process.
I wonder if we should have a BUG_ON(skb->h.th->doff < 5) in tcp_optlen()...
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now the skb->nh union has just one member, .raw, i.e. it is just like the
skb->mac union, strange, no? I'm just leaving it like that till the transport
layer is done with, when we'll rename skb->mac.raw to skb->mac_header (or
->mac_header_offset?), ditto for ->{h,nh}.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the places where we need a pointer to the network header, it is still legal
to touch skb->nh.raw directly if just adding to, subtracting from or setting it
to another layer header.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ipv6_fl_socklist from listening socket is inadvertently shared
with new socket created for connection. This leads to a variety of
interesting, but fatal, bugs. For example, removing one of the
sockets may lead to the other socket's encountering a page fault
when the now freed list is referenced.
The fix is to not share the flow label list with the new socket.
Signed-off-by: Masayuki Nakagawa <nakagawa.msy@ncos.nec.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do this even for non-blocking sockets. This avoids the silly -EAGAIN
that applications can see now, even for non-blocking sockets in some
cases (f.e. connect()).
With help from Venkat Tekkirala.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch contains the following possible cleanups:
- make the following needlessly global functions statis:
- ipv4/tcp.c: __tcp_alloc_md5sig_pool()
- ipv4/tcp_ipv4.c: tcp_v4_reqsk_md5_lookup()
- ipv4/udplite.c: udplite_rcv()
- ipv4/udplite.c: udplite_err()
- make the following needlessly global structs static:
- ipv4/tcp_ipv4.c: tcp_request_sock_ipv4_ops
- ipv4/tcp_ipv4.c: tcp_sock_ipv4_specific
- ipv6/tcp_ipv6.c: tcp_request_sock_ipv6_ops
- net/ipv{4,6}/udplite.c: remove inline's from static functions
(gcc should know best when to inline them)
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
... into anonymous union of __wsum and __u32 (csum and csum_offset resp.)
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Just some mis-placed ifdefs:
net/ipv4/tcp_minisocks.c: In function ‘tcp_twsk_destructor’:
net/ipv4/tcp_minisocks.c:364: warning: unused variable ‘twsk’
net/ipv6/tcp_ipv6.c:1846: warning: ‘tcp_sock_ipv6_specific’ defined but not used
net/ipv6/tcp_ipv6.c:1877: warning: ‘tcp_sock_ipv6_mapped_specific’ defined but not used
Signed-off-by: David S. Miller <davem@davemloft.net>
Throughout the TCP/DCCP (and tunnelling) code, it often happens that the
return code of a transmit function needs to be tested against NET_XMIT_CN
which is a value that does not indicate a strict error condition.
This patch uses a macro for these recurring situations which is consistent
with the already existing macro net_xmit_errno, saving on duplicated code.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
This removes two redundancies:
1) The test (skb->protocol == htons(ETH_P_IPV6) in tcp_v6_init_sequence()
is always true, due to
* tcp_v6_conn_request() is the only function calling this one
* tcp_v6_conn_request() redirects all skb's with ETH_P_IP protocol to
tcp_v4_conn_request() [ cf. top of tcp_v6_conn_request()]
2) The first argument, `struct sock *sk' of tcp_v{4,6}_init_sequence() is
never used.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
We currently allocate a fixed size (TCP_SYNQ_HSIZE=512) slots hash table for
each LISTEN socket, regardless of various parameters (listen backlog for
example)
On x86_64, this means order-1 allocations (might fail), even for 'small'
sockets, expecting few connections. On the contrary, a huge server wanting a
backlog of 50000 is slowed down a bit because of this fixed limit.
This patch makes the sizing of listen hash table a dynamic parameter,
depending of :
- net.core.somaxconn tunable (default is 128)
- net.ipv4.tcp_max_syn_backlog tunable (default : 256, 1024 or 128)
- backlog value given by user application (2nd parameter of listen())
For large allocations (bigger than PAGE_SIZE), we use vmalloc() instead of
kmalloc().
We still limit memory allocation with the two existing tunables (somaxconn &
tcp_max_syn_backlog). So for standard setups, this patch actually reduce RAM
usage.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A while ago Ingo patched tcp_v4_rcv on net/ipv4/tcp_ipv4.c to use
bh_lock_sock_nested and silence a lock validator warning. This fixed
it for IPv4, but recently I saw a report of the same warning on IPv6.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Function sk_filter() is called from tcp_v{4,6}_rcv() functions with arg
needlock = 0, while socket is not locked at that moment. In order to avoid
this and similar issues in the future, use rcu for sk->sk_filter field read
protection.
Signed-off-by: Dmitry Mishin <dim@openvz.org>
Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
Based on MIPL2 kernel patch.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace CHECKSUM_HW by CHECKSUM_PARTIAL (for outgoing packets, whose
checksum still needs to be completed) and CHECKSUM_COMPLETE (for
incoming packets, device supplied full checksum).
Patch originally from Herbert Xu, updated by myself for 2.6.18-rc3.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This automatically labels the TCP, Unix stream, and dccp child sockets
as well as openreqs to be at the same MLS level as the peer. This will
result in the selection of appropriately labeled IPSec Security
Associations.
This also uses the sock's sid (as opposed to the isec sid) in SELinux
enforcement of secmark in rcv_skb and postroute_last hooks.
Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This labels the flows that could utilize IPSec xfrms at the points the
flows are defined so that IPSec policy and SAs at the right label can
be used.
The following protos are currently not handled, but they should
continue to be able to use single-labeled IPSec like they currently
do.
ipmr
ip_gre
ipip
igmp
sit
sctp
ip6_tunnel (IPv6 over IPv6 tunnel device)
decnet
Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP over IPV6 would incorrectly inherit the GSO settings.
This would cause kernel to send Tcp Segmentation Offload packets for
IPV6 data to devices that can't handle it. It caused the sky2 driver
to lock http://bugzilla.kernel.org/show_bug.cgi?id=7050
and the e1000 would generate bogus packets. I can't blame the
hardware for gagging if the upper layers feed it garbage.
This was a new bug in 2.6.18 introduced with GSO support.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refer to RFC2012, tcpAttemptFails is defined as following:
tcpAttemptFails OBJECT-TYPE
SYNTAX Counter32
MAX-ACCESS read-only
STATUS current
DESCRIPTION
"The number of times TCP connections have made a direct
transition to the CLOSED state from either the SYN-SENT
state or the SYN-RCVD state, plus the number of times TCP
connections have made a direct transition to the LISTEN
state from the SYN-RCVD state."
::= { tcp 7 }
When I lookup into RFC793, I found that the state change should occured
under following condition:
1. SYN-SENT -> CLOSED
a) Received ACK,RST segment when SYN-SENT state.
2. SYN-RCVD -> CLOSED
b) Received SYN segment when SYN-RCVD state(came from LISTEN).
c) Received RST segment when SYN-RCVD state(came from SYN-SENT).
d) Received SYN segment when SYN-RCVD state(came from SYN-SENT).
3. SYN-RCVD -> LISTEN
e) Received RST segment when SYN-RCVD state(came from LISTEN).
In my test, those direct state transition can not be counted to
tcpAttemptFails.
Signed-off-by: Wei Yongjun <yjwei@nanjing-fnst.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current users of ip6_dst_lookup can be divided into two classes:
1) The caller holds no locks and is in user-context (UDP).
2) The caller does not want to lookup the dst cache at all.
The second class covers everyone except UDP because most people do
the cache lookup directly before calling ip6_dst_lookup. This patch
adds ip6_sk_dst_lookup for the first class.
Similarly ip6_dst_store users can be divded into those that need to
take the socket dst lock and those that don't. This patch adds
__ip6_dst_store for those (everyone except UDP/datagram) that don't
need an extra lock.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Certain subsystems in the stack (e.g., netfilter) can break the partial
checksum on GSO packets. Until they're fixed, this patch allows this to
work by recomputing the partial checksums through the GSO mechanism.
Once they've all been converted to update the partial checksum instead of
clearing it, this workaround can be removed.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
[IPV6]: Added GSO support for TCPv6
[NET]: Generalise TSO-specific bits from skb_setup_caps
[IPV6]: Added GSO support for TCPv6
[IPV6]: Remove redundant length check on input
[NETFILTER]: SCTP conntrack: fix crash triggered by packet without chunks
[TG3]: Update version and reldate
[TG3]: Add TSO workaround using GSO
[TG3]: Turn on hw fix for ASF problems
[TG3]: Add rx BD workaround
[TG3]: Add tg3_netif_stop() in vlan functions
[TCP]: Reset gso_segs if packet is dodgy
This patch adds GSO support for IPv6 and TCPv6. This is based on a patch
by Ananda Raju <Ananda.Raju@neterion.com>. His original description is:
This patch enables TSO over IPv6. Currently Linux network stacks
restricts TSO over IPv6 by clearing of the NETIF_F_TSO bit from
"dev->features". This patch will remove this restriction.
This patch will introduce a new flag NETIF_F_TSO6 which will be used
to check whether device supports TSO over IPv6. If device support TSO
over IPv6 then we don't clear of NETIF_F_TSO and which will make the
TCP layer to create TSO packets. Any device supporting TSO over IPv6
will set NETIF_F_TSO6 flag in "dev->features" along with NETIF_F_TSO.
In case when user disables TSO using ethtool, NETIF_F_TSO will get
cleared from "dev->features". So even if we have NETIF_F_TSO6 we don't
get TSO packets created by TCP layer.
SKB_GSO_TCPV4 renamed to SKB_GSO_TCP to make it generic GSO packet.
SKB_GSO_UDPV4 renamed to SKB_GSO_UDP as UFO is not a IPv4 feature.
UFO is supported over IPv6 also
The following table shows there is significant improvement in
throughput with normal frames and CPU usage for both normal and jumbo.
--------------------------------------------------
| | 1500 | 9600 |
| ------------------|-------------------|
| | thru CPU | thru CPU |
--------------------------------------------------
| TSO OFF | 2.00 5.5% id | 5.66 20.0% id |
--------------------------------------------------
| TSO ON | 2.63 78.0 id | 5.67 39.0% id |
--------------------------------------------------
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds GSO support for IPv6 and TCPv6.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
While debugging a TCP server hang issue, we noticed that currently there is
no way for a user to get the acceptq backlog value for a TCP listen socket.
All the standard networking utilities that display socket info like netstat,
ss and /proc/net/tcp have 2 fields called rx_queue and tx_queue. These
fields do not mean much for listening sockets. This patch uses one of these
unused fields(rx_queue) to export the accept queue len for listening sockets.
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Locks down user pages and sets up for DMA in tcp_recvmsg, then calls
dma_async_try_early_copy in tcp_v4_do_rcv
Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No code changes, just tidying up, in some cases moving EXPORT_SYMBOLs
to just after the function exported, etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch extends {get|set}sockopt compatibility layer in order to
move protocol specific parts to their place and avoid huge universal
net/compat.c file in the future.
Signed-off-by: Dmitry Mishin <dim@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Consolidating open coded sequences in tcp and dccp, v4 and v6.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implementation of packetization layer path mtu discovery for TCP, based on
the internet-draft currently found at
<http://www.ietf.org/internet-drafts/draft-ietf-pmtud-method-05.txt>.
Signed-off-by: John Heffner <jheffner@psc.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fix dst reference counting in tcp_v6_send_synack
Analysis:
Currently tcp_v6_send_synack is never called with a dst entry
so dst always comes in as NULL.
ip6_dst_lookup calls ip6_route_output which calls dst_hold
before it returns the dst entry. Neither xfrm_lookup
nor tcp_make_synack consume the dst entry so we still have
a dst_entry with a bumped refrence count at the end of
this function.
Therefore we need to call dst_release just before we return
just like tcp_v4_send_synack does.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ip6_xmit() function now assumes that its sk argument is non-NULL,
which isn't currently true when TCPv6 code is sending RST or ACK
packets. This fixes that code to use a socket of its own for sending
such packets, as TCPv4 does. (Thanks Andi for the pointer).
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move nextheader offset to the IP6CB to make it possible to pass a
packet to ip6_input_finish multiple times and have it skip already
parsed headers. As a nice side effect this gets rid of the manual
hopopts skipping in ip6_input_finish.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
As DCCP needs to be called in the same spots.
Now we have a member in inet_sock (is_icsk), set at sock creation time from
struct inet_protosw->flags (if INET_PROTOSW_ICSK is set, like for TCP and
DCCP) to see if a struct sock instance is a inet_connection_sock for places
like the ones in ip_sockglue.c (v4 and v6) where we previously were looking if
sk_type was SOCK_STREAM, that is insufficient because we now use the same code
for DCCP, that has sk_type SOCK_DCCP.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Renaming it to inet6_hash_connect, making it possible to ditch
dccp_v6_hash_connect and share the same code with TCP instead.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
So that we can share several timewait sockets related functions and
make the timewait mini sockets infrastructure closer to the request
mini sockets one.
Next changesets will take advantage of this, moving more code out of
TCP and DCCP v4 and v6 to common infrastructure.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It was already non-TCP specific, will be used by DCCPv6.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Out of tcp6_timewait_sock, that now is just an aggregation of
inet_timewait_sock and inet6_timewait_sock, using tw_ipv6_offset in struct
inet_timewait_sock, that is common to the IPv6 transport protocols that use
timewait sockets, like DCCP and TCP.
tw_ipv6_offset plays the struct inet_sock pinfo6 role, i.e. for the generic
code to find the IPv6 area in a timewait sock.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using sk->sk_protocol instead of IPPROTO_TCP.
Will be used by DCCPv6 in the next changesets.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
And move it to struct inet_connection_sock. DCCP will use it in the
upcoming changesets.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
And inet6_rsk_offset in inet_request_sock, for the same reasons as
inet_sock's pinfo6 member.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
More work is needed tho to introduce inet6_request_sock from
tcp6_request_sock, in the same layout considerations as ipv6_pinfo in
inet_sock, next changeset will do that.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Here is the patch that introduces the generic skb_checksum_complete
which also checks for hardware RX checksum faults. If that happens,
it'll call netdev_rx_csum_fault which currently prints out a stack
trace with the device name. In future it can turn off RX checksum.
I've converted every spot under net/ that does RX checksum checks to
use skb_checksum_complete or __skb_checksum_complete with the
exceptions of:
* Those places where checksums are done bit by bit. These will call
netdev_rx_csum_fault directly.
* The following have not been completely checked/converted:
ipmr
ip_vs
netfilter
dccp
This patch is based on patches and suggestions from Stephen Hemminger
and David S. Miller.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch randomizes the port selected on bind() for connections
to help with possible security attacks. It should also be faster
in most cases because there is no need for a global lock.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Arnaldo and I agreed it could be applied now, because I have other
pending patches depending on this one (Thank you Arnaldo)
(The other important patch moves skc_refcnt in a separate cache line,
so that the SMP/NUMA performance doesnt suffer from cache line ping pongs)
1) First some performance data :
--------------------------------
tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established()
The most time critical code is :
sk_for_each(sk, node, &head->chain) {
if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif))
goto hit; /* You sunk my battleship! */
}
The sk_for_each() does use prefetch() hints but only the begining of
"struct sock" is prefetched.
As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far
away from the begining of "struct sock", it has to bring into CPU
cache cold cache line. Each iteration has to use at least 2 cache
lines.
This can be problematic if some chains are very long.
2) The goal
-----------
The idea I had is to change things so that INET_MATCH() may return
FALSE in 99% of cases only using the data already in the CPU cache,
using one cache line per iteration.
3) Description of the patch
---------------------------
Adds a new 'unsigned int skc_hash' field in 'struct sock_common',
filling a 32 bits hole on 64 bits platform.
struct sock_common {
unsigned short skc_family;
volatile unsigned char skc_state;
unsigned char skc_reuse;
int skc_bound_dev_if;
struct hlist_node skc_node;
struct hlist_node skc_bind_node;
atomic_t skc_refcnt;
+ unsigned int skc_hash;
struct proto *skc_prot;
};
Store in this 32 bits field the full hash, not masked by (ehash_size -
1) Using this full hash as the first comparison done in INET_MATCH
permits us immediatly skip the element without touching a second cache
line in case of a miss.
Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to
sk_hash and tw_hash) already contains the slot number if we mask with
(ehash_size - 1)
File include/net/inet_hashtables.h
64 bits platforms :
#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash))
((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \
((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
32bits platforms:
#define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash)) && \
(inet_sk(__sk)->daddr == (__saddr)) && \
(inet_sk(__sk)->rcv_saddr == (__daddr)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
- Adds a prefetch(head->chain.first) in
__inet_lookup_established()/__tcp_v4_check_established() and
__inet6_lookup_established()/__tcp_v6_check_established() and
__dccp_v4_check_established() to bring into cache the first element of the
list, before the {read|write}_lock(&head->lock);
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Support several new socket options / ancillary data:
IPV6_RECVPKTINFO, IPV6_PKTINFO,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS,
IPV6_RECVDSTOPTS, IPV6_DSTOPTS, IPV6_RTHDRDSTOPTS,
IPV6_RECVRTHDR, IPV6_RTHDR,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS
Old semantics are preserved as IPV6_2292xxxx so that
we can maintain backward compatibility.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Of this type, mostly:
CHECK net/ipv6/netfilter.c
net/ipv6/netfilter.c:96:12: warning: symbol 'ipv6_netfilter_init' was not declared. Should it be static?
net/ipv6/netfilter.c:101:6: warning: symbol 'ipv6_netfilter_fini' was not declared. Should it be static?
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Doing this we allow tcp_diag to support IPV6 even if tcp_diag is compiled
statically and IPV6 is compiled as a module, removing the previous restriction
while not building any IPV6 code if it is not selected.
Now to work on the tcpdiag_register infrastructure and then to rename the whole
thing to inetdiag, reflecting its by then completely generic nature.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the same way as was done with the v4 counterparts, this will be moved
to inet6_hashtables.c.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This changeset basically moves tcp_sk()->{ca_ops,ca_state,etc} to inet_csk(),
minimal renaming/moving done in this changeset to ease review.
Most of it is just changes of struct tcp_sock * to struct sock * parameters.
With this we move to a state closer to two interesting goals:
1. Generalisation of net/ipv4/tcp_diag.c, becoming inet_diag.c, being used
for any INET transport protocol that has struct inet_hashinfo and are
derived from struct inet_connection_sock. Keeps the userspace API, that will
just not display DCCP sockets, while newer versions of tools can support
DCCP.
2. INET generic transport pluggable Congestion Avoidance infrastructure, using
the current TCP CA infrastructure with DCCP.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
That groups all of the tables and variables associated to the TCP timewait
schedulling/recycling/killing code, that now can be isolated from the TCP
specific code and used by other transport protocols, such as DCCP.
Next changeset will move this code to net/ipv4/inet_timewait_sock.c
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This also moved inet_iif from tcp to inet_hashtables.h, as it is
needed by the inet_lookup callers, perhaps this needs a bit of
polishing, but for now seems fine.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This creates struct inet_connection_sock, moving members out of struct
tcp_sock that are shareable with other INET connection oriented
protocols, such as DCCP, that in my private tree already uses most of
these members.
The functions that operate on these members were renamed, using a
inet_csk_ prefix while not being moved yet to a new file, so as to
ease the review of these changes.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This paves the way to generalise the rest of the sock ID lookup
routines and saves some bytes in TCPv4 TIME_WAIT sockets on distro
kernels (where IPv6 is always built as a module):
[root@qemu ~]# grep tw_sock /proc/slabinfo
tw_sock_TCPv6 0 0 128 31 1
tw_sock_TCP 0 0 96 41 1
[root@qemu ~]#
Now if a protocol wants to use the TIME_WAIT generic infrastructure it
only has to set the sk_prot->twsk_obj_size field with the size of its
inet_timewait_sock derived sock and proto_register will create
sk_prot->twsk_slab, for now its only for INET sockets, but we can
introduce timewait_sock later if some non INET transport protocolo
wants to use this stuff.
Next changesets will take advantage of this new infrastructure to
generalise even more TCP code.
[acme@toy net-2.6.14]$ grep built-in /tmp/before.size /tmp/after.size
/tmp/before.size: 188646 11764 5068 205478 322a6 net/ipv4/built-in.o
/tmp/after.size: 188144 11764 5068 204976 320b0 net/ipv4/built-in.o
[acme@toy net-2.6.14]$
Tested with both IPv4 & IPv6 (::1 (localhost) & ::ffff:172.20.0.1
(qemu host)).
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Also expose all of the tcp_hashinfo members, i.e. killing those
tcp_ehash, etc macros, this will more clearly expose already generic
functions and some that need just a bit of work to become generic, as
we'll see in the upcoming changesets.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This required moving tcp_bucket_cachep to inet_hashinfo.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This should really be in a inet_connection_sock, but I'm leaving it
for a later optimization, when some more fields common to INET
transport protocols now in tcp_sk or inet_sk will be chunked out into
inet_connection_sock, for now its better to concentrate on getting the
changes in the core merged to leave the DCCP tree with only DCCP
specific code.
Next changesets will take advantage of this move to generalise things
like tcp_bind_hash, tcp_put_port, tcp_inherit_port, making the later
receive a inet_hashinfo parameter, and even __tcp_tw_hashdance, etc in
the future, when tcp_tw_bucket gets transformed into the struct
timewait_sock hierarchy.
tcp_destroy_sock also is eligible as soon as tcp_orphan_count gets
moved to sk_prot.
A cascade of incremental changes will ultimately make the tcp_lookup
functions be fully generic.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is to break down the complexity of the series of patches,
making it very clear that this one just does:
1. renames tcp_ prefixed hashtable functions and data structures that
were already mostly generic to inet_ to share it with DCCP and
other INET transport protocols.
2. Removes not used functions (__tb_head & tb_head)
3. Removes some leftover prototypes in the headers (tcp_bucket_unlock &
tcp_v4_build_header)
Next changesets will move tcp_sk(sk)->bind_hash to inet_sock so that we can
make functions such as tcp_inherit_port, __tcp_inherit_port, tcp_v4_get_port,
__tcp_put_port, generic and get others like tcp_destroy_sock closer to generic
(tcp_orphan_count will go to sk->sk_prot to allow this).
Eventually most of these functions will be used passing the transport protocol
inet_hashinfo structure.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
From tcp_v4_rebuild_header, that already was pretty generic, I only
needed to use sk->sk_protocol instead of the hardcoded IPPROTO_TCP and
establish the requirement that INET transport layer protocols that
want to use this function map TCP_SYN_SENT to its equivalent state.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This trips up a lot of folks reading this code.
Put an unlikely() around the port-exhaustion test
for good measure.
Signed-off-by: David S. Miller <davem@davemloft.net>
Make TSO segment transmit size decisions at send time not earlier.
The basic scheme is that we try to build as large a TSO frame as
possible when pulling in the user data, but the size of the TSO frame
output to the card is determined at transmit time.
This is guided by tp->xmit_size_goal. It is always set to a multiple
of MSS and tells sendmsg/sendpage how large an SKB to try and build.
Later, tcp_write_xmit() and tcp_push_one() chop up the packet if
necessary and conditions warrant. These routines can also decide to
"defer" in order to wait for more ACKs to arrive and thus allow larger
TSO frames to be emitted.
A general observation is that TSO elongates the pipe, thus requiring a
larger congestion window and larger buffering especially at the sender
side. Therefore, it is important that applications 1) get a large
enough socket send buffer (this is accomplished by our dynamic send
buffer expansion code) 2) do large enough writes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow using setsockopt to set TCP congestion control to use on a per
socket basis.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow TCP to have multiple pluggable congestion control algorithms.
Algorithms are defined by a set of operations and can be built in
or modules. The legacy "new RENO" algorithm is used as a starting
point and fallback.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This chunks out the accept_queue and tcp_listen_opt code and moves
them to net/core/request_sock.c and include/net/request_sock.h, to
make it useful for other transport protocols, DCCP being the first one
to use it.
Next patches will rename tcp_listen_opt to accept_sock and remove the
inline tcp functions that just call a reqsk_queue_ function.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ok, this one just renames some stuff to have a better namespace and to
dissassociate it from TCP:
struct open_request -> struct request_sock
tcp_openreq_alloc -> reqsk_alloc
tcp_openreq_free -> reqsk_free
tcp_openreq_fastfree -> __reqsk_free
With this most of the infrastructure closely resembles a struct
sock methods subset.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
