Congestion window recover after loss depends upon the fact
that if we have a full MSS sized frame at the head of the
send queue, we will send it. TSO deferral can defeat the
ACK clocking necessary to exit cleanly from recovery.
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>
In tcp_clean_rtx_queue(), if the TSO packet is not even partially
acked, do not waste time calling tcp_tso_acked().
Signed-off-by: David S. Miller <davem@davemloft.net>
Everything stated there is out of data. tcp_trim_skb()
does adjust the available socket send buffer space and
skb->truesize now.
Signed-off-by: David S. Miller <davem@davemloft.net>
Only put user data purely to pages when doing TSO.
The extra page allocations cause two problems:
1) Add the overhead of the page allocations themselves.
2) Make us do small user copies when we get to the end
of the TCP socket cache page.
It is still beneficial to purely use pages for TSO,
so we will do it for that case.
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_snd_test() is run for every packet output by a single
call to tcp_write_xmit(), but this is not necessary.
For one, the congestion window space needs to only be
calculated one time, then used throughout the duration
of the loop.
This cleanup also makes experimenting with different TSO
packetization schemes much easier.
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_snd_test() does several different things, use inline
functions to express this more clearly.
1) It initializes the TSO count of SKB, if necessary.
2) It performs the Nagle test.
3) It makes sure the congestion window is adhered to.
4) It makes sure SKB fits into the send window.
This cleanup also sets things up so that things like the
available packets in the congestion window does not need
to be calculated multiple times by packet sending loops
such as tcp_write_xmit().
Signed-off-by: David S. Miller <davem@davemloft.net>
'nonagle' should be passed to the tcp_snd_test() function
as 'TCP_NAGLE_PUSH' if we are checking an SKB not at the
tail of the write_queue. This is because Nagle does not
apply to such frames since we cannot possibly tack more
data onto them.
However, while doing this __tcp_push_pending_frames() makes
all of the packets in the write_queue use this modified
'nonagle' value.
Fix the bug and simplify this function by just calling
tcp_write_xmit() directly if sk_send_head is non-NULL.
As a result, we can now make tcp_data_snd_check() just call
tcp_push_pending_frames() instead of the specialized
__tcp_data_snd_check().
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_write_xmit() uses tcp_current_mss(), but some of it's callers,
namely __tcp_push_pending_frames(), already has this value available
already.
While we're here, fix the "cur_mss" argument to be "unsigned int"
instead of plain "unsigned".
Signed-off-by: David S. Miller <davem@davemloft.net>
Put the main basic block of work at the top-level of
tabbing, and mark the TCP_CLOSE test with unlikely().
Signed-off-by: David S. Miller <davem@davemloft.net>
The tcp_cwnd_validate() function should only be invoked
if we actually send some frames, yet __tcp_push_pending_frames()
will always invoke it. tcp_write_xmit() does the call for us,
so the call here can simply be removed.
Also, tcp_write_xmit() can be marked static.
Signed-off-by: David S. Miller <davem@davemloft.net>
When we add any new packet to the TCP socket write queue,
we must call skb_header_release() on it in order for the
TSO sharing checks in the drivers to work.
Signed-off-by: David S. Miller <davem@davemloft.net>
It reimplements portions of tcp_snd_check(), so it
we move it to tcp_output.c we can consolidate it's
logic much easier in a later change.
Signed-off-by: David S. Miller <davem@davemloft.net>
This just moves the code into tcp_output.c, no code logic changes are
made by this patch.
Using this as a baseline, we can begin to untangle the mess of
comparisons for the Nagle test et al. We will also be able to reduce
all of the redundant computation that occurs when outputting data
packets.
Signed-off-by: David S. Miller <davem@davemloft.net>
On each packet output, we call tcp_dec_quickack_mode()
if the ACK flag is set. It drops tp->ack.quick until
it hits zero, at which time we deflate the ATO value.
When doing TSO, we are emitting multiple packets with
ACK set, so we should decrement tp->ack.quick that many
segments.
Note that, unlike this case, tcp_enter_cwr() should not
take the tcp_skb_pcount(skb) into consideration. That
function, one time, readjusts tp->snd_cwnd and moves
into TCP_CA_CWR state.
Signed-off-by: David S. Miller <davem@davemloft.net>
The ideal and most optimal layout for an SKB when doing
scatter-gather is to put all the headers at skb->data, and
all the user data in the page array.
This makes SKB splitting and combining extremely simple,
especially before a packet goes onto the wire the first
time.
So, when sk_stream_alloc_pskb() is given a zero size, make
sure there is no skb_tailroom(). This is achieved by applying
SKB_DATA_ALIGN() to the header length used here.
Next, make select_size() in TCP output segmentation use a
length of zero when NETIF_F_SG is true on the outgoing
interface.
Signed-off-by: David S. Miller <davem@davemloft.net>
Below a patch to preallocate memory when doing resize of trie (inflate halve)
If preallocations fails it just skips the resize of this tnode for this time.
The oops we got when killing bgpd (with full routing) is now gone.
Patrick memory patch is also used.
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
- rt_check_expire() fixes (an overflow occured if size of the hash
was >= 65536)
reminder of the bugfix:
The rt_check_expire() has a serious problem on machines with large
route caches, and a standard HZ value of 1000.
With default values, ie ip_rt_gc_interval = 60*HZ = 60000 ;
the loop count :
for (t = ip_rt_gc_interval << rt_hash_log; t >= 0;
overflows (t is a 31 bit value) as soon rt_hash_log is >= 16 (65536
slots in route cache hash table).
In this case, rt_check_expire() does nothing at all
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- rt hash table allocated using alloc_large_system_hash() function.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Locking abstraction
- Spinlocks moved out of rt hash table : Less memory (50%) used by rt
hash table. it's a win even on UP.
- Sizing of spinlocks table depends on NR_CPUS
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Inflating a node a couple of times makes it exceed the 128k kmalloc limit.
Use __get_free_pages for allocations > PAGE_SIZE, as in fib_hash.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Robert Olsson <Robert.Olsson@data.slu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
Makes IPv4 ip_rcv registration happen last in af_inet.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In 2.6.12 we started dropping the conntrack reference when a packet
leaves the IP layer. This broke connection tracking on a bridge,
because bridge-netfilter defers calling some NF_IP_* hooks to the bridge
layer for locally generated packets going out a bridge, where the
conntrack reference is no longer available. This patch keeps the
reference in this case as a temporary solution, long term we will
remove the defered hook calling. No attempt is made to drop the
reference in the bridge-code when it is no longer needed, tc actions
could already have sent the packet anywhere.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
In an smp system, it is possible for an connection timer to expire, calling
ip_vs_conn_expire while the connection table is being flushed, before
ct_write_lock_bh is acquired.
Since the list iterator loop in ip_vs_con_flush releases and re-acquires the
spinlock (even though it doesn't re-enable softirqs), it is possible for the
expiration function to modify the connection list, while it is being traversed
in ip_vs_conn_flush.
The result is that the next pointer gets set to NULL, and subsequently
dereferenced, resulting in an oops.
Signed-off-by: Neil Horman <nhorman@redhat.com>
Acked-by: JulianAnastasov
Signed-off-by: David S. Miller <davem@davemloft.net>
This should help up the insertion... but the resize is more crucial.
and complex and needs some thinking.
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
I think there is a small bug in ipconfig.c in case IPCONFIG_DHCP is set
and dhcp is used.
When a DHCPOFFER is received, ip address is kept until we get DHCPACK.
If no ack is received, ic_dynamic() returns negatively, but leaves the
offered ip address in ic_myaddr.
This makes the main loop in ip_auto_config() break and uses the maybe
incomplete configuration.
Not sure if it's the best way to do, but the following trivial patch
correct this.
Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
I followed Thomas' proposal to see every martian destination as a case
where the ipInAddrErrors counter has to be incremented. There are
two advantages by doing so: (1) The relation between the ipInReceive
counter and all the other ipInXXX counters is more accurate in the
case the RTN_UNICAST code check fails and (2) it makes the code in
ip_route_input_slow easier.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mostly missing initialization of padding fields of 1 or 2 bytes length,
two instances of uninitialized nlmsgerr->msg of 16 bytes length.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds mangling of ARP requests (in addition to replies),
since ARP caches are made from snooping both requests and replies.
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: <pageexec@freemail.hu>
$subject was fixed in 2.4 already, 2.6 needs it as well.
The impact of the bugs is a kernel stack overflow and privilege escalation
from CAP_NET_ADMIN via the IP_VS_SO_SET_STARTDAEMON/IP_VS_SO_GET_DAEMON
ioctls. People running with 'root=all caps' (i.e., most users) are not
really affected (there's nothing to escalate), but SELinux and similar
users should take it seriously if they grant CAP_NET_ADMIN to other users.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
It doesn't seem to make much sense to let an "If unsure, say N." option
default to y.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since it is tristate when we offer it as a choice, we should
definte it also as tristate when forcing it as the default.
Otherwise kconfig warns.
Signed-off-by: David S. Miller <davem@davemloft.net>
Create TCP_CONG_ADVANCED option, akin to IP_ADVANCED_ROUTER, which
when disabled will bypass all of the congestion control Kconfig
options and leave the user with a safe default.
That safe default is currently BIC-TCP with new Reno as a fallback.
Signed-off-by: David S. Miller <davem@davemloft.net>
Most users need not be concerned with a complex choice of what
FIB lookup algorithm to use. So give them the safe default of
IP_FIB_HASH if IP_ADVANCED_ROUTING is disabled.
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>
This patch implements Tom Kelly's Scalable TCP congestion control algorithm
for the modular framework.
The algorithm has some nice scaling properties, and has been used a fair bit
in research, though is known to have significant fairness issues, so it's not
really suitable for general purpose use.
Signed-off-by: John Heffner <jheffner@psc.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
H-TCP is a congestion control algorithm developed at the Hamilton Institute, by
Douglas Leith and Robert Shorten. It is extending the standard Reno algorithm
with mode switching is thus a relatively simple modification.
H-TCP is defined in a layered manner as it is still a research platform. The
basic form includes the modification of beta according to the ratio of maxRTT
to min RTT and the alpha=2*factor*(1-beta) relation, where factor is dependant
on the time since last congestion.
The other layers improve convergence by adding appropriate factors to alpha.
The following patch implements the H-TCP algorithm in it's basic form.
Signed-Off-By: Baruch Even <baruch@ev-en.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP Vegas code modified for the new TCP infrastructure.
Vegas now uses microsecond resolution timestamps for
better estimation of performance over higher speed links.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP Hybla congestion avoidance.
- "In heterogeneous networks, TCP connections that incorporate a
terrestrial or satellite radio link are greatly disadvantaged with
respect to entirely wired connections, because of their longer round
trip times (RTTs). To cope with this problem, a new TCP proposal, the
TCP Hybla, is presented and discussed in the paper[1]. It stems from an
analytical evaluation of the congestion window dynamics in the TCP
standard versions (Tahoe, Reno, NewReno), which suggests the necessary
modifications to remove the performance dependence on RTT.[...]"[1]
[1]: Carlo Caini, Rosario Firrincieli, "TCP Hybla: a TCP enhancement for
heterogeneous networks",
International Journal of Satellite Communications and Networking
Volume 22, Issue 5 , Pages 547 - 566. September 2004.
Signed-off-by: Daniele Lacamera (root at danielinux.net)net
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sally Floyd's high speed TCP congestion control.
This is useful for comparison and research.
Signed-off-by: John Heffner <jheffner@psc.edu>
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is the existing 2.6.12 Westwood code moved from tcp_input
to the new congestion framework. A lot of the inline functions
have been eliminated to try and make it clearer.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP BIC congestion control reworked to use the new congestion control
infrastructure. This version is more up to date than the BIC
code in 2.6.12; it incorporates enhancements from BICTCP 1.1,
to handle low latency links.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Enhancement to the tcp_diag interface used by the iproute2 ss command
to report the tcp congestion control being used by a socket.
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 patch creates a new kstrdup library function and changes the "local"
implementations in several places to use this function.
Most of the changes come from the sound and net subsystems. The sound part
had already been acknowledged by Takashi Iwai and the net part by David S.
Miller.
I left UML alone for now because I would need more time to read the code
carefully before making changes there.
Signed-off-by: Paulo Marques <pmarques@grupopie.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
