Commit Graph

544 Commits

Author SHA1 Message Date
Eric Dumazet 5037e9ef94 net: fix IP early demux races
David Wilder reported crashes caused by dst reuse.

<quote David>
  I am seeing a crash on a distro V4.2.3 kernel caused by a double
  release of a dst_entry.  In ipv4_dst_destroy() the call to
  list_empty() finds a poisoned next pointer, indicating the dst_entry
  has already been removed from the list and freed. The crash occurs
  18 to 24 hours into a run of a network stress exerciser.
</quote>

Thanks to his detailed report and analysis, we were able to understand
the core issue.

IP early demux can associate a dst to skb, after a lookup in TCP/UDP
sockets.

When socket cache is not properly set, we want to store into
sk->sk_dst_cache the dst for future IP early demux lookups,
by acquiring a stable refcount on the dst.

Problem is this acquisition is simply using an atomic_inc(),
which works well, unless the dst was queued for destruction from
dst_release() noticing dst refcount went to zero, if DST_NOCACHE
was set on dst.

We need to make sure current refcount is not zero before incrementing
it, or risk double free as David reported.

This patch, being a stable candidate, adds two new helpers, and use
them only from IP early demux problematic paths.

It might be possible to merge in net-next skb_dst_force() and
skb_dst_force_safe(), but I prefer having the smallest patch for stable
kernels : Maybe some skb_dst_force() callers do not expect skb->dst
can suddenly be cleared.

Can probably be backported back to linux-3.6 kernels

Reported-by: David J. Wilder <dwilder@us.ibm.com>
Tested-by: David J. Wilder <dwilder@us.ibm.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-12-14 23:52:00 -05:00
Eric Dumazet 1b8e6a01e1 tcp: md5: fix lockdep annotation
When a passive TCP is created, we eventually call tcp_md5_do_add()
with sk pointing to the child. It is not owner by the user yet (we
will add this socket into listener accept queue a bit later anyway)

But we do own the spinlock, so amend the lockdep annotation to avoid
following splat :

[ 8451.090932] net/ipv4/tcp_ipv4.c:923 suspicious rcu_dereference_protected() usage!
[ 8451.090932]
[ 8451.090932] other info that might help us debug this:
[ 8451.090932]
[ 8451.090934]
[ 8451.090934] rcu_scheduler_active = 1, debug_locks = 1
[ 8451.090936] 3 locks held by socket_sockopt_/214795:
[ 8451.090936]  #0:  (rcu_read_lock){.+.+..}, at: [<ffffffff855c6ac1>] __netif_receive_skb_core+0x151/0xe90
[ 8451.090947]  #1:  (rcu_read_lock){.+.+..}, at: [<ffffffff85618143>] ip_local_deliver_finish+0x43/0x2b0
[ 8451.090952]  #2:  (slock-AF_INET){+.-...}, at: [<ffffffff855acda5>] sk_clone_lock+0x1c5/0x500
[ 8451.090958]
[ 8451.090958] stack backtrace:
[ 8451.090960] CPU: 7 PID: 214795 Comm: socket_sockopt_

[ 8451.091215] Call Trace:
[ 8451.091216]  <IRQ>  [<ffffffff856fb29c>] dump_stack+0x55/0x76
[ 8451.091229]  [<ffffffff85123b5b>] lockdep_rcu_suspicious+0xeb/0x110
[ 8451.091235]  [<ffffffff8564544f>] tcp_md5_do_add+0x1bf/0x1e0
[ 8451.091239]  [<ffffffff85645751>] tcp_v4_syn_recv_sock+0x1f1/0x4c0
[ 8451.091242]  [<ffffffff85642b27>] ? tcp_v4_md5_hash_skb+0x167/0x190
[ 8451.091246]  [<ffffffff85647c78>] tcp_check_req+0x3c8/0x500
[ 8451.091249]  [<ffffffff856451ae>] ? tcp_v4_inbound_md5_hash+0x11e/0x190
[ 8451.091253]  [<ffffffff85647170>] tcp_v4_rcv+0x3c0/0x9f0
[ 8451.091256]  [<ffffffff85618143>] ? ip_local_deliver_finish+0x43/0x2b0
[ 8451.091260]  [<ffffffff856181b6>] ip_local_deliver_finish+0xb6/0x2b0
[ 8451.091263]  [<ffffffff85618143>] ? ip_local_deliver_finish+0x43/0x2b0
[ 8451.091267]  [<ffffffff85618d38>] ip_local_deliver+0x48/0x80
[ 8451.091270]  [<ffffffff85618510>] ip_rcv_finish+0x160/0x700
[ 8451.091273]  [<ffffffff8561900e>] ip_rcv+0x29e/0x3d0
[ 8451.091277]  [<ffffffff855c74b7>] __netif_receive_skb_core+0xb47/0xe90

Fixes: a8afca0329 ("tcp: md5: protects md5sig_info with RCU")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-18 23:02:47 -05:00
Eric Dumazet 00fd38d938 tcp: ensure proper barriers in lockless contexts
Some functions access TCP sockets without holding a lock and
might output non consistent data, depending on compiler and or
architecture.

tcp_diag_get_info(), tcp_get_info(), tcp_poll(), get_tcp4_sock() ...

Introduce sk_state_load() and sk_state_store() to fix the issues,
and more clearly document where this lack of locking is happening.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15 18:36:38 -05:00
Eric Dumazet 49a496c97d tcp: use correct req pointer in tcp_move_syn() calls
I mistakenly took wrong request sock pointer when calling tcp_move_syn()

@req_unhash is either a copy of @req, or a NULL value for
FastOpen connexions (as we do not expect to unhash the temporary
request sock from ehash table)

Fixes: 805c4bc057 ("tcp: fix req->saved_syn race")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ying Cai <ycai@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-05 15:57:51 -05:00
Eric Dumazet 805c4bc057 tcp: fix req->saved_syn race
For the reasons explained in commit ce1050089c ("tcp/dccp: fix
ireq->pktopts race"), we need to make sure we do not access
req->saved_syn unless we own the request sock.

This fixes races for listeners using TCP_SAVE_SYN option.

Fixes: e994b2f0fb ("tcp: do not lock listener to process SYN packets")
Fixes: 079096f103 ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Ying Cai <ycai@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-05 14:36:09 -05:00
Eric Dumazet 5e0724d027 tcp/dccp: fix hashdance race for passive sessions
Multiple cpus can process duplicates of incoming ACK messages
matching a SYN_RECV request socket. This is a rare event under
normal operations, but definitely can happen.

Only one must win the race, otherwise corruption would occur.

To fix this without adding new atomic ops, we use logic in
inet_ehash_nolisten() to detect the request was present in the same
ehash bucket where we try to insert the new child.

If request socket was not found, we have to undo the child creation.

This actually removes a spin_lock()/spin_unlock() pair in
reqsk_queue_unlink() for the fast path.

Fixes: e994b2f0fb ("tcp: do not lock listener to process SYN packets")
Fixes: 079096f103 ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-23 05:42:21 -07:00
Eric Dumazet dc6ef6be52 tcp: do not set queue_mapping on SYNACK
At the time of commit fff3269907 ("tcp: reflect SYN queue_mapping into
SYNACK packets") we had little ways to cope with SYN floods.

We no longer need to reflect incoming skb queue mappings, and instead
can pick a TX queue based on cpu cooking the SYNACK, with normal XPS
affinities.

Note that all SYNACK retransmits were picking TX queue 0, this no longer
is a win given that SYNACK rtx are now distributed on all cpus.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-18 22:26:02 -07:00
Eric Dumazet f03f2e154f tcp/dccp: add inet_csk_reqsk_queue_drop_and_put() helper
Let's reduce the confusion about inet_csk_reqsk_queue_drop() :
In many cases we also need to release reference on request socket,
so add a helper to do this, reducing code size and complexity.

Fixes: 4bdc3d6614 ("tcp/dccp: fix behavior of stale SYN_RECV request sockets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-16 00:52:18 -07:00
Eric Dumazet ef84d8ce5a Revert "inet: fix double request socket freeing"
This reverts commit c69736696c.

At the time of above commit, tcp_req_err() and dccp_req_err()
were dead code, as SYN_RECV request sockets were not yet in ehash table.

Real bug was fixed later in a different commit.

We need to revert to not leak a refcount on request socket.

inet_csk_reqsk_queue_drop_and_put() will be added
in following commit to make clean inet_csk_reqsk_queue_drop()
does not release the reference owned by caller.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-16 00:52:17 -07:00
Eric Dumazet 4bdc3d6614 tcp/dccp: fix behavior of stale SYN_RECV request sockets
When a TCP/DCCP listener is closed, its pending SYN_RECV request sockets
become stale, meaning 3WHS can not complete.

But current behavior is wrong :
incoming packets finding such stale sockets are dropped.

We need instead to cleanup the request socket and perform another
lookup :
- Incoming ACK will give a RST answer,
- SYN rtx might find another listener if available.
- We expedite cleanup of request sockets and old listener socket.

Fixes: 079096f103 ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-13 18:26:34 -07:00
Eric Dumazet ed53d0ab76 net: shrink struct sock and request_sock by 8 bytes
One 32bit hole is following skc_refcnt, use it.
skc_incoming_cpu can also be an union for request_sock rcv_wnd.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-12 19:28:22 -07:00
Eric Dumazet e994b2f0fb tcp: do not lock listener to process SYN packets
Everything should now be ready to finally allow SYN
packets processing without holding listener lock.

Tested:

3.5 Mpps SYNFLOOD. Plenty of cpu cycles available.

Next bottleneck is the refcount taken on listener,
that could be avoided if we remove SLAB_DESTROY_BY_RCU
strict semantic for listeners, and use regular RCU.

    13.18%  [kernel]  [k] __inet_lookup_listener
     9.61%  [kernel]  [k] tcp_conn_request
     8.16%  [kernel]  [k] sha_transform
     5.30%  [kernel]  [k] inet_reqsk_alloc
     4.22%  [kernel]  [k] sock_put
     3.74%  [kernel]  [k] tcp_make_synack
     2.88%  [kernel]  [k] ipt_do_table
     2.56%  [kernel]  [k] memcpy_erms
     2.53%  [kernel]  [k] sock_wfree
     2.40%  [kernel]  [k] tcp_v4_rcv
     2.08%  [kernel]  [k] fib_table_lookup
     1.84%  [kernel]  [k] tcp_openreq_init_rwin

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:46 -07:00
Eric Dumazet ca6fb06518 tcp: attach SYNACK messages to request sockets instead of listener
If a listen backlog is very big (to avoid syncookies), then
the listener sk->sk_wmem_alloc is the main source of false
sharing, as we need to touch it twice per SYNACK re-transmit
and TX completion.

(One SYN packet takes listener lock once, but up to 6 SYNACK
are generated)

By attaching the skb to the request socket, we remove this
source of contention.

Tested:

 listen(fd, 10485760); // single listener (no SO_REUSEPORT)
 16 RX/TX queue NIC
 Sustain a SYNFLOOD attack of ~320,000 SYN per second,
 Sending ~1,400,000 SYNACK per second.
 Perf profiles now show listener spinlock being next bottleneck.

    20.29%  [kernel]  [k] queued_spin_lock_slowpath
    10.06%  [kernel]  [k] __inet_lookup_established
     5.12%  [kernel]  [k] reqsk_timer_handler
     3.22%  [kernel]  [k] get_next_timer_interrupt
     3.00%  [kernel]  [k] tcp_make_synack
     2.77%  [kernel]  [k] ipt_do_table
     2.70%  [kernel]  [k] run_timer_softirq
     2.50%  [kernel]  [k] ip_finish_output
     2.04%  [kernel]  [k] cascade

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:43 -07:00
Eric Dumazet 079096f103 tcp/dccp: install syn_recv requests into ehash table
In this patch, we insert request sockets into TCP/DCCP
regular ehash table (where ESTABLISHED and TIMEWAIT sockets
are) instead of using the per listener hash table.

ACK packets find SYN_RECV pseudo sockets without having
to find and lock the listener.

In nominal conditions, this halves pressure on listener lock.

Note that this will allow for SO_REUSEPORT refinements,
so that we can select a listener using cpu/numa affinities instead
of the prior 'consistent hash', since only SYN packets will
apply this selection logic.

We will shrink listen_sock in the following patch to ease
code review.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ying Cai <ycai@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:41 -07:00
Eric Dumazet aa3a0c8ce6 tcp: get_openreq[46]() changes
When request sockets are no longer in a per listener hash table
but on regular TCP ehash, we need to access listener uid
through req->rsk_listener

get_openreq6() also gets a const for its request socket argument.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:40 -07:00
Eric Dumazet ba8e275a45 tcp: cleanup tcp_v[46]_inbound_md5_hash()
We'll soon have to call tcp_v[46]_inbound_md5_hash() twice.
Also add const attribute to the socket, as it might be the
unlocked listener for SYN packets.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:38 -07:00
Eric Dumazet 38cb52455c tcp: call sk_mark_napi_id() on the child, not the listener
This fixes a typo : We want to store the NAPI id on child socket.
Presumably nobody really uses busy polling, on short lived flows.

Fixes: 3d97379a67 ("tcp: move sk_mark_napi_id() at the right place")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-03 04:32:37 -07:00
Eric Dumazet 0536fcc039 tcp: prepare fastopen code for upcoming listener changes
While auditing TCP stack for upcoming 'lockless' listener changes,
I found I had to change fastopen_init_queue() to properly init the object
before publishing it.

Otherwise an other cpu could try to lock the spinlock before it gets
properly initialized.

Instead of adding appropriate barriers, just remove dynamic memory
allocations :
- Structure is 28 bytes on 64bit arches. Using additional 8 bytes
  for holding a pointer seems overkill.
- Two listeners can share same cache line and performance would suffer.

If we really want to save few bytes, we would instead dynamically allocate
whole struct request_sock_queue in the future.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:10 -07:00
Eric Dumazet f964629e33 tcp: constify tcp_v{4|6}_route_req() sock argument
These functions do not change the listener socket.
Goal is to make sure tcp_conn_request() is not messing with
listener in a racy way.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:09 -07:00
Eric Dumazet 0c27171e66 tcp/dccp: constify syn_recv_sock() method sock argument
We'll soon no longer hold listener socket lock, these
functions do not modify the socket in any way.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:09 -07:00
Eric Dumazet 72ab4a86f7 tcp: remove tcp_rcv_state_process() tcp_hdr argument
Factorize code to get tcp header from skb. It makes no sense
to duplicate code in callers.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:07 -07:00
Eric Dumazet bda07a64c0 tcp: remove unused len argument from tcp_rcv_state_process()
Once we realize tcp_rcv_synsent_state_process() does not use
its 'len' argument and we get rid of it, then it becomes clear
this argument is no longer used in tcp_rcv_state_process()

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:07 -07:00
Eric Dumazet a00e74442b tcp/dccp: constify send_synack and send_reset socket argument
None of these functions need to change the socket, make it
const.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-29 16:53:07 -07:00
Eric Dumazet 0f935dbedc tcp: constify tcp_v{4|6}_send_synack() socket argument
This documents fact that listener lock might not be held
at the time SYNACK are sent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-25 13:00:39 -07:00
Eric Dumazet b83e3deb97 tcp: md5: constify tcp_md5_do_lookup() socket argument
When TCP new listener is done, these functions will be called
without socket lock being held. Make sure they don't change
anything.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-25 13:00:38 -07:00
Eric Dumazet b40cf18ef7 tcp: constify listener socket in tcp_v[46]_init_req()
Soon, listener socket spinlock will no longer be held,
add const arguments to tcp_v[46]_init_req() to make clear these
functions can not mess socket fields.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-25 13:00:36 -07:00
Eric Dumazet d8ed625044 tcp: factorize sk_txhash init
Neal suggested to move sk_txhash init into tcp_create_openreq_child(),
called both from IPv4 and IPv6.

This opportunity was missed in commit 58d607d3e5 ("tcp: provide
skb->hash to synack packets")

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-24 14:52:30 -07:00
Eric Dumazet 58d607d3e5 tcp: provide skb->hash to synack packets
In commit b73c3d0e4f ("net: Save TX flow hash in sock and set in skbuf
on xmit"), Tom provided a l4 hash to most outgoing TCP packets.

We'd like to provide one as well for SYNACK packets, so that all packets
of a given flow share same txhash, to later enable bonding driver to
also use skb->hash to perform slave selection.

Note that a SYNACK retransmit shuffles the tx hash, as Tom did
in commit 265f94ff54 ("net: Recompute sk_txhash on negative routing
advice") for established sockets.

This has nice effect making TCP flows resilient to some kind of black
holes, even at connection establish phase.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <tom@herbertland.com>
Cc: Mahesh Bandewar <maheshb@google.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-17 21:01:04 -07:00
David S. Miller 182ad468e7 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Conflicts:
	drivers/net/ethernet/cavium/Kconfig

The cavium conflict was overlapping dependency
changes.

Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-13 16:23:11 -07:00
Eric Dumazet 3257d8b12f inet: fix possible request socket leak
In commit b357a364c5 ("inet: fix possible panic in
reqsk_queue_unlink()"), I missed fact that tcp_check_req()
can return the listener socket in one case, and that we must
release the request socket refcount or we leak it.

Tested:

 Following packetdrill test template shows the issue

0     socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+0    setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+0    bind(3, ..., ...) = 0
+0    listen(3, 1) = 0

+0    < S 0:0(0) win 2920 <mss 1460,sackOK,nop,nop>
+0    > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK>
+.002 < . 1:1(0) ack 21 win 2920
+0    > R 21:21(0)

Fixes: b357a364c5 ("inet: fix possible panic in reqsk_queue_unlink()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-10 21:17:45 -07:00
Tom Herbert 877d1f6291 net: Set sk_txhash from a random number
This patch creates sk_set_txhash and eliminates protocol specific
inet_set_txhash and ip6_set_txhash. sk_set_txhash simply sets a
random number instead of performing flow dissection. sk_set_txash
is also allowed to be called multiple times for the same socket,
we'll need this when redoing the hash for negative routing advice.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-29 22:44:04 -07:00
Eric Dumazet dbe7faa404 inet: inet_twsk_deschedule factorization
inet_twsk_deschedule() calls are followed by inet_twsk_put().

Only particular case is in inet_twsk_purge() but there is no point
to defer the inet_twsk_put() after re-enabling BH.

Lets rename inet_twsk_deschedule() to inet_twsk_deschedule_put()
and move the inet_twsk_put() inside.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-09 15:12:20 -07:00
Eric Dumazet 98da81a426 tcp: remove redundant checks II
For same reasons than in commit 12e25e1041 ("tcp: remove redundant
checks"), we can remove redundant checks done for timewait sockets.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-07 01:55:01 -07:00
Eric Dumazet 12e25e1041 tcp: remove redundant checks
tcp_v4_rcv() checks the following before calling tcp_v4_do_rcv():

if (th->doff < sizeof(struct tcphdr) / 4)
    goto bad_packet;
if (!pskb_may_pull(skb, th->doff * 4))
    goto discard_it;

So following check in tcp_v4_do_rcv() is redundant
and "goto csum_err;" is wrong anyway.

if (skb->len < tcp_hdrlen(skb) || ...)
	goto csum_err;

A second check can be removed after no_tcp_socket label for same reason.

Same tests can be removed in tcp_v6_do_rcv()

Note : short tcp frames are not properly accounted in tcpInErrs MIB,
because pskb_may_pull() failure simply drops incoming skb, we might
fix this in a separate patch.

Signed-off-by: Eric Dumazet  <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-04 01:04:40 -07:00
Marcelo Ricardo Leitner 2efd055c53 tcp: add tcpi_segs_in and tcpi_segs_out to tcp_info
This patch tracks the total number of inbound and outbound segments on a
TCP socket. One may use this number to have an idea on connection
quality when compared against the retransmissions.

RFC4898 named these : tcpEStatsPerfSegsIn and tcpEStatsPerfSegsOut

These are a 32bit field each and can be fetched both from TCP_INFO
getsockopt() if one has a handle on a TCP socket, or from inet_diag
netlink facility (iproute2/ss patch will follow)

Note that tp->segs_out was placed near tp->snd_nxt for good data
locality and minimal performance impact, while tp->segs_in was placed
near tp->bytes_received for the same reason.

Join work with Eric Dumazet.

Note that received SYN are accounted on the listener, but sent SYNACK
are not accounted.

Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-21 23:25:21 -04:00
Daniel Borkmann 492135557d tcp: add rfc3168, section 6.1.1.1. fallback
This work as a follow-up of commit f7b3bec6f5 ("net: allow setting ecn
via routing table") and adds RFC3168 section 6.1.1.1. fallback for outgoing
ECN connections. In other words, this work adds a retry with a non-ECN
setup SYN packet, as suggested from the RFC on the first timeout:

  [...] A host that receives no reply to an ECN-setup SYN within the
  normal SYN retransmission timeout interval MAY resend the SYN and
  any subsequent SYN retransmissions with CWR and ECE cleared. [...]

Schematic client-side view when assuming the server is in tcp_ecn=2 mode,
that is, Linux default since 2009 via commit 255cac91c3 ("tcp: extend
ECN sysctl to allow server-side only ECN"):

 1) Normal ECN-capable path:

    SYN ECE CWR ----->
                <----- SYN ACK ECE
            ACK ----->

 2) Path with broken middlebox, when client has fallback:

    SYN ECE CWR ----X crappy middlebox drops packet
                      (timeout, rtx)
            SYN ----->
                <----- SYN ACK
            ACK ----->

In case we would not have the fallback implemented, the middlebox drop
point would basically end up as:

    SYN ECE CWR ----X crappy middlebox drops packet
                      (timeout, rtx)
    SYN ECE CWR ----X crappy middlebox drops packet
                      (timeout, rtx)
    SYN ECE CWR ----X crappy middlebox drops packet
                      (timeout, rtx)

In any case, it's rather a smaller percentage of sites where there would
occur such additional setup latency: it was found in end of 2014 that ~56%
of IPv4 and 65% of IPv6 servers of Alexa 1 million list would negotiate
ECN (aka tcp_ecn=2 default), 0.42% of these webservers will fail to connect
when trying to negotiate with ECN (tcp_ecn=1) due to timeouts, which the
fallback would mitigate with a slight latency trade-off. Recent related
paper on this topic:

  Brian Trammell, Mirja Kühlewind, Damiano Boppart, Iain Learmonth,
  Gorry Fairhurst, and Richard Scheffenegger:
    "Enabling Internet-Wide Deployment of Explicit Congestion Notification."
    Proc. PAM 2015, New York.
  http://ecn.ethz.ch/ecn-pam15.pdf

Thus, when net.ipv4.tcp_ecn=1 is being set, the patch will perform RFC3168,
section 6.1.1.1. fallback on timeout. For users explicitly not wanting this
which can be in DC use case, we add a net.ipv4.tcp_ecn_fallback knob that
allows for disabling the fallback.

tp->ecn_flags are not being cleared in tcp_ecn_clear_syn() on output, but
rather we let tcp_ecn_rcv_synack() take that over on input path in case a
SYN ACK ECE was delayed. Thus a spurious SYN retransmission will not prevent
ECN being negotiated eventually in that case.

Reference: https://www.ietf.org/proceedings/92/slides/slides-92-iccrg-1.pdf
Reference: https://www.ietf.org/proceedings/89/slides/slides-89-tsvarea-1.pdf
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mirja Kühlewind <mirja.kuehlewind@tik.ee.ethz.ch>
Signed-off-by: Brian Trammell <trammell@tik.ee.ethz.ch>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Dave That <dave.taht@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-19 16:53:37 -04:00
Eric Dumazet cd8ae85299 tcp: provide SYN headers for passive connections
This patch allows a server application to get the TCP SYN headers for
its passive connections.  This is useful if the server is doing
fingerprinting of clients based on SYN packet contents.

Two socket options are added: TCP_SAVE_SYN and TCP_SAVED_SYN.

The first is used on a socket to enable saving the SYN headers
for child connections. This can be set before or after the listen()
call.

The latter is used to retrieve the SYN headers for passive connections,
if the parent listener has enabled TCP_SAVE_SYN.

TCP_SAVED_SYN is read once, it frees the saved SYN headers.

The data returned in TCP_SAVED_SYN are network (IPv4/IPv6) and TCP
headers.

Original patch was written by Tom Herbert, I changed it to not hold
a full skb (and associated dst and conntracking reference).

We have used such patch for about 3 years at Google.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Tested-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-05 16:02:34 -04:00
Eric Dumazet b357a364c5 inet: fix possible panic in reqsk_queue_unlink()
[ 3897.923145] BUG: unable to handle kernel NULL pointer dereference at
 0000000000000080
[ 3897.931025] IP: [<ffffffffa9f27686>] reqsk_timer_handler+0x1a6/0x243

There is a race when reqsk_timer_handler() and tcp_check_req() call
inet_csk_reqsk_queue_unlink() on the same req at the same time.

Before commit fa76ce7328 ("inet: get rid of central tcp/dccp listener
timer"), listener spinlock was held and race could not happen.

To solve this bug, we change reqsk_queue_unlink() to not assume req
must be found, and we return a status, to conditionally release a
refcount on the request sock.

This also means tcp_check_req() in non fastopen case might or not
consume req refcount, so tcp_v6_hnd_req() & tcp_v4_hnd_req() have
to properly handle this.

(Same remark for dccp_check_req() and its callers)

inet_csk_reqsk_queue_drop() is now too big to be inlined, as it is
called 4 times in tcp and 3 times in dccp.

Fixes: fa76ce7328 ("inet: get rid of central tcp/dccp listener timer")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-24 11:39:15 -04:00
Eric Dumazet 789f558cfb tcp/dccp: get rid of central timewait timer
Using a timer wheel for timewait sockets was nice ~15 years ago when
memory was expensive and machines had a single processor.

This does not scale, code is ugly and source of huge latencies
(Typically 30 ms have been seen, cpus spinning on death_lock spinlock.)

We can afford to use an extra 64 bytes per timewait sock and spread
timewait load to all cpus to have better behavior.

Tested:

On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1
on the target (lpaa24)

Before patch :

lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
419594

lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
437171

While test is running, we can observe 25 or even 33 ms latencies.

lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms
rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2

lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms
rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2

After patch :

About 90% increase of throughput :

lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
810442

lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
800992

And latencies are kept to minimal values during this load, even
if network utilization is 90% higher :

lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms
rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-13 16:40:05 -04:00
Eric Dumazet b52e69217b tcp: md5: fix a typo in tcp_v4_md5_lookup()
Lookup key for tcp_md5_do_lookup() has to be taken
from addr_sk, not sk (which can be the listener)

Fixes: fd3a154a00 ("tcp: md5: get rid of tcp_v[46]_reqsk_md5_lookup()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-09 18:11:16 -04:00
Ian Morris 00db41243e ipv4: coding style: comparison for inequality with NULL
The ipv4 code uses a mixture of coding styles. In some instances check
for non-NULL pointer is done as x != NULL and sometimes as x. x is
preferred according to checkpatch and this patch makes the code
consistent by adopting the latter form.

No changes detected by objdiff.

Signed-off-by: Ian Morris <ipm@chirality.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-03 12:11:15 -04:00
Ian Morris 51456b2914 ipv4: coding style: comparison for equality with NULL
The ipv4 code uses a mixture of coding styles. In some instances check
for NULL pointer is done as x == NULL and sometimes as !x. !x is
preferred according to checkpatch and this patch makes the code
consistent by adopting the latter form.

No changes detected by objdiff.

Signed-off-by: Ian Morris <ipm@chirality.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-03 12:11:15 -04:00
David S. Miller 9f0d34bc34 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Conflicts:
	drivers/net/usb/asix_common.c
	drivers/net/usb/sr9800.c
	drivers/net/usb/usbnet.c
	include/linux/usb/usbnet.h
	net/ipv4/tcp_ipv4.c
	net/ipv6/tcp_ipv6.c

The TCP conflicts were overlapping changes.  In 'net' we added a
READ_ONCE() to the socket cached RX route read, whilst in 'net-next'
Eric Dumazet touched the surrounding code dealing with how mini
sockets are handled.

With USB, it's a case of the same bug fix first going into net-next
and then I cherry picked it back into net.

Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-02 16:16:53 -04:00
Eric Dumazet 41d25fe092 tcp: tcp_syn_flood_action() can be static
After commit 1fb6f159fd ("tcp: add tcp_conn_request"),
tcp_syn_flood_action() is no longer used from IPv6.

We can make it static, by moving it above tcp_conn_request()

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Octavian Purdila <octavian.purdila@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-29 12:17:18 -07:00
Eric Dumazet 0144a81ccc tcp: fix ipv4 mapped request socks
ss should display ipv4 mapped request sockets like this :

tcp    SYN-RECV   0      0  ::ffff:192.168.0.1:8080   ::ffff:192.0.2.1:35261

and not like this :

tcp    SYN-RECV   0      0  192.168.0.1:8080   192.0.2.1:35261

We should init ireq->ireq_family based on listener sk_family,
not the actual protocol carried by SYN packet.

This means we can set ireq_family in inet_reqsk_alloc()

Fixes: 3f66b083a5 ("inet: introduce ireq_family")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-25 00:57:48 -04:00
Eric Dumazet fd3a154a00 tcp: md5: get rid of tcp_v[46]_reqsk_md5_lookup()
With request socks convergence, we no longer need
different lookup methods. A request socket can
use generic lookup function.

Add const qualifier to 2nd tcp_v[46]_md5_lookup() parameter.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-24 21:16:30 -04:00
Eric Dumazet 39f8e58e53 tcp: md5: remove request sock argument of calc_md5_hash()
Since request and established sockets now have same base,
there is no need to pass two pointers to tcp_v4_md5_hash_skb()
or tcp_v6_md5_hash_skb()

Also add a const qualifier to their struct tcp_md5sig_key argument.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-24 21:16:30 -04:00
Eric Dumazet ff74e23f7e tcp: md5: input path is run under rcu protected sections
It is guaranteed that both tcp_v4_rcv() and tcp_v6_rcv()
run from rcu read locked sections :

ip_local_deliver_finish() and ip6_input_finish() both
use rcu_read_lock()

Also align tcp_v6_inbound_md5_hash() on tcp_v4_inbound_md5_hash()
by returning a boolean.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-24 21:16:29 -04:00
Michal Kubeček d0c294c53a tcp: prevent fetching dst twice in early demux code
On s390x, gcc 4.8 compiles this part of tcp_v6_early_demux()

        struct dst_entry *dst = sk->sk_rx_dst;

        if (dst)
                dst = dst_check(dst, inet6_sk(sk)->rx_dst_cookie);

to code reading sk->sk_rx_dst twice, once for the test and once for
the argument of ip6_dst_check() (dst_check() is inline). This allows
ip6_dst_check() to be called with null first argument, causing a crash.

Protect sk->sk_rx_dst access by READ_ONCE() both in IPv4 and IPv6
TCP early demux code.

Fixes: 41063e9dd1 ("ipv4: Early TCP socket demux.")
Fixes: c7109986db ("ipv6: Early TCP socket demux")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-23 22:38:24 -04:00
Fan Du c69736696c inet: fix double request socket freeing
Eric Hugne reported following error :

I'm hitting this warning on latest net-next when i try to SSH into a machine
with eth0 added to a bridge (but i think the problem is older than that)

Steps to reproduce:
node2 ~ # brctl addif br0 eth0
[  223.758785] device eth0 entered promiscuous mode
node2 ~ # ip link set br0 up
[  244.503614] br0: port 1(eth0) entered forwarding state
[  244.505108] br0: port 1(eth0) entered forwarding state
node2 ~ # [  251.160159] ------------[ cut here ]------------
[  251.160831] WARNING: CPU: 0 PID: 3 at include/net/request_sock.h:102 tcp_v4_err+0x6b1/0x720()
[  251.162077] Modules linked in:
[  251.162496] CPU: 0 PID: 3 Comm: ksoftirqd/0 Not tainted 4.0.0-rc3+ #18
[  251.163334] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[  251.164078]  ffffffff81a8365c ffff880038a6ba18 ffffffff8162ace4 0000000000009898
[  251.165084]  0000000000000000 ffff880038a6ba58 ffffffff8104da85 ffff88003fa437c0
[  251.166195]  ffff88003fa437c0 ffff88003fa74e00 ffff88003fa43bb8 ffff88003fad99a0
[  251.167203] Call Trace:
[  251.167533]  [<ffffffff8162ace4>] dump_stack+0x45/0x57
[  251.168206]  [<ffffffff8104da85>] warn_slowpath_common+0x85/0xc0
[  251.169239]  [<ffffffff8104db65>] warn_slowpath_null+0x15/0x20
[  251.170271]  [<ffffffff81559d51>] tcp_v4_err+0x6b1/0x720
[  251.171408]  [<ffffffff81630d03>] ? _raw_read_lock_irq+0x3/0x10
[  251.172589]  [<ffffffff81534e20>] ? inet_del_offload+0x40/0x40
[  251.173366]  [<ffffffff81569295>] icmp_socket_deliver+0x65/0xb0
[  251.174134]  [<ffffffff815693a2>] icmp_unreach+0xc2/0x280
[  251.174820]  [<ffffffff8156a82d>] icmp_rcv+0x2bd/0x3a0
[  251.175473]  [<ffffffff81534ea2>] ip_local_deliver_finish+0x82/0x1e0
[  251.176282]  [<ffffffff815354d8>] ip_local_deliver+0x88/0x90
[  251.177004]  [<ffffffff815350f0>] ip_rcv_finish+0xf0/0x310
[  251.177693]  [<ffffffff815357bc>] ip_rcv+0x2dc/0x390
[  251.178336]  [<ffffffff814f5da3>] __netif_receive_skb_core+0x713/0xa20
[  251.179170]  [<ffffffff814f7fca>] __netif_receive_skb+0x1a/0x80
[  251.179922]  [<ffffffff814f97d4>] process_backlog+0x94/0x120
[  251.180639]  [<ffffffff814f9612>] net_rx_action+0x1e2/0x310
[  251.181356]  [<ffffffff81051267>] __do_softirq+0xa7/0x290
[  251.182046]  [<ffffffff81051469>] run_ksoftirqd+0x19/0x30
[  251.182726]  [<ffffffff8106cc23>] smpboot_thread_fn+0x153/0x1d0
[  251.183485]  [<ffffffff8106cad0>] ? SyS_setgroups+0x130/0x130
[  251.184228]  [<ffffffff8106935e>] kthread+0xee/0x110
[  251.184871]  [<ffffffff81069270>] ? kthread_create_on_node+0x1b0/0x1b0
[  251.185690]  [<ffffffff81631108>] ret_from_fork+0x58/0x90
[  251.186385]  [<ffffffff81069270>] ? kthread_create_on_node+0x1b0/0x1b0
[  251.187216] ---[ end trace c947fc7b24e42ea1 ]---
[  259.542268] br0: port 1(eth0) entered forwarding state

Remove the double calls to reqsk_put()

[edumazet] :

I got confused because reqsk_timer_handler() _has_ to call
reqsk_put(req) after calling inet_csk_reqsk_queue_drop(), as
the timer handler holds a reference on req.

Signed-off-by: Fan Du <fan.du@intel.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Erik Hugne <erik.hugne@ericsson.com>
Fixes: fa76ce7328 ("inet: get rid of central tcp/dccp listener timer")
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-23 21:40:48 -04:00