This patch implements Proportional Rate Reduction (PRR) for TCP.
PRR is an algorithm that determines TCP's sending rate in fast
recovery. PRR avoids excessive window reductions and aims for
the actual congestion window size at the end of recovery to be as
close as possible to the window determined by the congestion control
algorithm. PRR also improves accuracy of the amount of data sent
during loss recovery.
The patch implements the recommended flavor of PRR called PRR-SSRB
(Proportional rate reduction with slow start reduction bound) and
replaces the existing rate halving algorithm. PRR improves upon the
existing Linux fast recovery under a number of conditions including:
1) burst losses where the losses implicitly reduce the amount of
outstanding data (pipe) below the ssthresh value selected by the
congestion control algorithm and,
2) losses near the end of short flows where application runs out of
data to send.
As an example, with the existing rate halving implementation a single
loss event can cause a connection carrying short Web transactions to
go into the slow start mode after the recovery. This is because during
recovery Linux pulls the congestion window down to packets_in_flight+1
on every ACK. A short Web response often runs out of new data to send
and its pipe reduces to zero by the end of recovery when all its packets
are drained from the network. Subsequent HTTP responses using the same
connection will have to slow start to raise cwnd to ssthresh. PRR on
the other hand aims for the cwnd to be as close as possible to ssthresh
by the end of recovery.
A description of PRR and a discussion of its performance can be found at
the following links:
- IETF Draft:
http://tools.ietf.org/html/draft-mathis-tcpm-proportional-rate-reduction-01
- IETF Slides:
http://www.ietf.org/proceedings/80/slides/tcpm-6.pdfhttp://tools.ietf.org/agenda/81/slides/tcpm-2.pdf
- Paper to appear in Internet Measurements Conference (IMC) 2011:
Improving TCP Loss Recovery
Nandita Dukkipati, Matt Mathis, Yuchung Cheng
Signed-off-by: Nandita Dukkipati <nanditad@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The l4_rxhash flag was added to the skb structure to indicate
that the rxhash value was computed over the 4 tuple for the
packet which includes the port information in the encapsulated
transport packet. This is used by the stack to preserve the
rxhash value in __skb_rx_tunnel.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
RCU api had been completed and rcu_access_pointer() or
rcu_dereference_protected() are better than generic
rcu_dereference_raw()
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As rt_iif represents input device even for packets
coming from loopback with output route, it is not an unique
key specific to input routes. Now rt_route_iif has such role,
it was fl.iif in 2.6.38, so better to change the checks at
some places to save CPU cycles and to restore 2.6.38 semantics.
compare_keys:
- input routes: only rt_route_iif matters, rt_iif is same
- output routes: only rt_oif matters, rt_iif is not
used for matching in __ip_route_output_key
- now we are back to 2.6.38 state
ip_route_input_common:
- matching rt_route_iif implies input route
- compared to 2.6.38 we eliminated one rth->fl.oif check
because it was not needed even for 2.6.38
compare_hash_inputs:
Only the change here is not an optimization, it has
effect only for output routes. I assume I'm restoring
the original intention to ignore oif, it was using fl.iif
- now we are back to 2.6.38 state
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a gcc 4.4.3, warnings are emitted for a possibly uninitialized use
of ecn_ok.
This can happen if cookie_check_timestamp() returns due to not having
seen a timestamp. Defaulting to ecn off seems like a reasonable thing
to do in this case, so initialized ecn_ok to false.
Signed-off-by: Mike Waychison <mikew@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make sure skb dst has reference when moving to
another context. Currently, I don't see protocols that can
hit it when sending broadcasts/multicasts to loopback using
noref dsts, so it is just a precaution.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
The raw sockets can provide source address for
routing but their privileges are not considered. We
can provide non-local source address, make sure the
FLOWI_FLAG_ANYSRC flag is set if socket has privileges
for this, i.e. based on hdrincl (IP_HDRINCL) and
transparent flags.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP in some cases uses different global (raw) socket
to send RST and ACK. The transparent flag is not set there.
Currently, it is a problem for rerouting after the previous
change.
Fix it by simplifying the checks in ip_route_me_harder
and use FLOWI_FLAG_ANYSRC even for sockets. It looks safe
because the initial routing allowed this source address to
be used and now we just have to make sure the packet is rerouted.
As a side effect this also allows rerouting for normal
raw sockets that use spoofed source addresses which was not possible
even before we eliminated the ip_route_input call.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
IP_PKTOPTIONS is broken for 32-bit applications running
in COMPAT mode on 64-bit kernels.
This happens because msghdr's msg_flags field is always
set to zero. When running in COMPAT mode this should be
set to MSG_CMSG_COMPAT instead.
Signed-off-by: Tiberiu Szocs-Mihai <tszocs@ixiacom.com>
Signed-off-by: Daniel Baluta <dbaluta@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
compare_keys and ip_route_input_common rely on
rt_oif for distinguishing of input and output routes
with same keys values. But sometimes the input route has
also same hash chain (keyed by iif != 0) with the output
routes (keyed by orig_oif=0). Problem visible if running
with small number of rhash_entries.
Fix them to use rt_route_iif instead. By this way
input route can not be returned to users that request
output route.
The patch fixes the ip_rt_bug errors that were
reported in ip_local_out context, mostly for 255.255.255.255
destinations.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
Computers have become a lot faster since we compromised on the
partial MD4 hash which we use currently for performance reasons.
MD5 is a much safer choice, and is inline with both RFC1948 and
other ISS generators (OpenBSD, Solaris, etc.)
Furthermore, only having 24-bits of the sequence number be truly
unpredictable is a very serious limitation. So the periodic
regeneration and 8-bit counter have been removed. We compute and
use a full 32-bit sequence number.
For ipv6, DCCP was found to use a 32-bit truncated initial sequence
number (it needs 43-bits) and that is fixed here as well.
Reported-by: Dan Kaminsky <dan@doxpara.com>
Tested-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gergely Kalman reported crashes in check_peer_redir().
It appears commit f39925dbde (ipv4: Cache learned redirect
information in inetpeer.) added a race, leading to possible NULL ptr
dereference.
Since we can now change dst neighbour, we should make sure a reader can
safely use a neighbour.
Add RCU protection to dst neighbour, and make sure check_peer_redir()
can be called safely by different cpus in parallel.
As neighbours are already freed after one RCU grace period, this patch
should not add typical RCU penalty (cache cold effects)
Many thanks to Gergely for providing a pretty report pointing to the
bug.
Reported-by: Gergely Kalman <synapse@hippy.csoma.elte.hu>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When assigning a NULL value to an RCU protected pointer, no barrier
is needed. The rcu_assign_pointer, used to handle that but will soon
change to not handle the special case.
Convert all rcu_assign_pointer of NULL value.
//smpl
@@ expression P; @@
- rcu_assign_pointer(P, NULL)
+ RCU_INIT_POINTER(P, NULL)
// </smpl>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert array index from the loop bound to the loop index.
A simplified version of the semantic patch that fixes this problem is as
follows: (http://coccinelle.lip6.fr/)
// <smpl>
@@
expression e1,e2,ar;
@@
for(e1 = 0; e1 < e2; e1++) { <...
ar[
- e2
+ e1
]
...> }
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (32 commits)
tg3: Remove 5719 jumbo frames and TSO blocks
tg3: Break larger frags into 4k chunks for 5719
tg3: Add tx BD budgeting code
tg3: Consolidate code that calls tg3_tx_set_bd()
tg3: Add partial fragment unmapping code
tg3: Generalize tg3_skb_error_unmap()
tg3: Remove short DMA check for 1st fragment
tg3: Simplify tx bd assignments
tg3: Reintroduce tg3_tx_ring_info
ASIX: Use only 11 bits of header for data size
ASIX: Simplify condition in rx_fixup()
Fix cdc-phonet build
bonding: reduce noise during init
bonding: fix string comparison errors
net: Audit drivers to identify those needing IFF_TX_SKB_SHARING cleared
net: add IFF_SKB_TX_SHARED flag to priv_flags
net: sock_sendmsg_nosec() is static
forcedeth: fix vlans
gianfar: fix bug caused by 87c288c6e9
gro: Only reset frag0 when skb can be pulled
...
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If a device event generates gratuitous ARP messages, only primary
address is used for sending. This patch iterates through the whole
list. Tested with 2 IP addresses configuration on bonding interface.
Signed-off-by: Zoltan Kiss <schaman@sch.bme.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix improper protocol err_handler, current implementation is fully
unapplicable and may cause kernel crash due to double kfree_skb.
Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rt_tos was changed to iph->tos but it must be filtered by RT_TOS
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
icmp_route_lookup() uses the wrong flow parameters if the reverse
session route lookup isn't used.
So do not commit to the re-decoded flow until we actually make a
final decision to use a real route saved in 'rt2'.
Reported-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because the ip fragment offset field counts 8-byte chunks, ip
fragments other than the last must contain a multiple of 8 bytes of
payload. ip_ufo_append_data wasn't respecting this constraint and,
depending on the MTU and ip option sizes, could create malformed
non-final fragments.
Google-Bug-Id: 5009328
Signed-off-by: Bill Sommerfeld <wsommerfeld@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPv6 fragment identification generation is way beyond what we use for
IPv4 : It uses a single generator. Its not scalable and allows DOS
attacks.
Now inetpeer is IPv6 aware, we can use it to provide a more secure and
scalable frag ident generator (per destination, instead of system wide)
This patch :
1) defines a new secure_ipv6_id() helper
2) extends inet_getid() to provide 32bit results
3) extends ipv6_select_ident() with a new dest parameter
Reported-by: Fernando Gont <fernando@gont.com.ar>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Compiler is not smart enough to avoid double BSWAP instructions in
ntohl(inet_make_mask(plen)).
Lets cache this value in struct leaf_info, (fill a hole on 64bit arches)
With route cache disabled, this saves ~2% of cpu in udpflood bench on
x86_64 machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the future dst entries will be neigh-less. In that environment we
need to have an easy transition point for current users of
dst->neighbour outside of the packet output fast path.
Signed-off-by: David S. Miller <davem@davemloft.net>
This will get us closer to being able to do "neigh stuff"
completely independent of the underlying dst_entry for
protocols (ipv4/ipv6) that wish to do so.
We will also be able to make dst entries neigh-less.
Signed-off-by: David S. Miller <davem@davemloft.net>
It's just taking on one of two possible values, either
neigh_ops->output or dev_queue_xmit(). And this is purely depending
upon whether nud_state has NUD_CONNECTED set or not.
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that there is a one-to-one correspondance between neighbour
and hh_cache entries, we no longer need:
1) dynamic allocation
2) attachment to dst->hh
3) refcounting
Initialization of the hh_cache entry is indicated by hh_len
being non-zero, and such initialization is always done with
the neighbour's lock held as a writer.
Signed-off-by: David S. Miller <davem@davemloft.net>
Get rid of all of the useless and costly indirection
by doing the neigh hash table lookup directly inside
of the neighbour binding.
Rename from arp_bind_neighbour to rt_bind_neighbour.
Use new helpers {__,}ipv4_neigh_lookup()
In rt_bind_neighbour() get rid of useless tests which
are never true in the context this function is called,
namely dev is never NULL and the dst->neighbour is
always NULL.
Signed-off-by: David S. Miller <davem@davemloft.net>
We currently can free inetpeer entries too early :
[ 782.636674] WARNING: kmemcheck: Caught 32-bit read from uninitialized memory (f130f44c)
[ 782.636677] 1f7b13c100000000000000000000000002000000000000000000000000000000
[ 782.636686] i i i i u u u u i i i i u u u u i i i i u u u u u u u u u u u u
[ 782.636694] ^
[ 782.636696]
[ 782.636698] Pid: 4638, comm: ssh Not tainted 3.0.0-rc5+ #270 Hewlett-Packard HP Compaq 6005 Pro SFF PC/3047h
[ 782.636702] EIP: 0060:[<c13fefbb>] EFLAGS: 00010286 CPU: 0
[ 782.636707] EIP is at inet_getpeer+0x25b/0x5a0
[ 782.636709] EAX: 00000002 EBX: 00010080 ECX: f130f3c0 EDX: f0209d30
[ 782.636711] ESI: 0000bc87 EDI: 0000ea60 EBP: f0209ddc ESP: c173134c
[ 782.636712] DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[ 782.636714] CR0: 8005003b CR2: f0beca80 CR3: 30246000 CR4: 000006d0
[ 782.636716] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[ 782.636717] DR6: ffff4ff0 DR7: 00000400
[ 782.636718] [<c13fbf76>] rt_set_nexthop.clone.45+0x56/0x220
[ 782.636722] [<c13fc449>] __ip_route_output_key+0x309/0x860
[ 782.636724] [<c141dc54>] tcp_v4_connect+0x124/0x450
[ 782.636728] [<c142ce43>] inet_stream_connect+0xa3/0x270
[ 782.636731] [<c13a8da1>] sys_connect+0xa1/0xb0
[ 782.636733] [<c13a99dd>] sys_socketcall+0x25d/0x2a0
[ 782.636736] [<c149deb8>] sysenter_do_call+0x12/0x28
[ 782.636738] [<ffffffff>] 0xffffffff
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current tcp/udp/sctp global memory limits are not taking into account
hugepages allocations, and allow 50% of ram to be used by buffers of a
single protocol [ not counting space used by sockets / inodes ...]
Lets use nr_free_buffer_pages() and allow a default of 1/8 of kernel ram
per protocol, and a minimum of 128 pages.
Heavy duty machines sysadmins probably need to tweak limits anyway.
References: https://bugzilla.stlinux.com/show_bug.cgi?id=38032
Reported-by: starlight <starlight@binnacle.cx>
Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hi,
Reinhard Max also pointed out that the error should EAFNOSUPPORT according
to POSIX.
The Linux manpages have it as EINVAL, some other OSes (Minix, HPUX, perhaps BSD) use
EAFNOSUPPORT. Windows uses WSAEFAULT according to MSDN.
Other protocols error values in their af bind() methods in current mainline git as far
as a brief look shows:
EAFNOSUPPORT: atm, appletalk, l2tp, llc, phonet, rxrpc
EINVAL: ax25, bluetooth, decnet, econet, ieee802154, iucv, netlink, netrom, packet, rds, rose, unix, x25,
No check?: can/raw, ipv6/raw, irda, l2tp/l2tp_ip
Ciao, Marcus
Signed-off-by: Marcus Meissner <meissner@suse.de>
Cc: Reinhard Max <max@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Calling icmp_send() on a local message size error leads to
an incorrect update of the path mtu. So use ip_local_error()
instead to notify the socket about the error.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We might call ip_ufo_append_data() for packets that will be IPsec
transformed later. This function should be used just for real
udp packets. So we check for rt->dst.header_len which is only
nonzero on IPsec handling and call ip_ufo_append_data() just
if rt->dst.header_len is zero.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make the case labels the same indent as the switch.
git diff -w shows no difference.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
