1. 28 7月, 2012 1 次提交
  2. 20 7月, 2012 1 次提交
    • Y
      net-tcp: Fast Open base · 2100c8d2
      Yuchung Cheng 提交于
      This patch impelements the common code for both the client and server.
      
      1. TCP Fast Open option processing. Since Fast Open does not have an
         option number assigned by IANA yet, it shares the experiment option
         code 254 by implementing draft-ietf-tcpm-experimental-options
         with a 16 bits magic number 0xF989. This enables global experiments
         without clashing the scarce(2) experimental options available for TCP.
      
         When the draft status becomes standard (maybe), the client should
         switch to the new option number assigned while the server supports
         both numbers for transistion.
      
      2. The new sysctl tcp_fastopen
      
      3. A place holder init function
      Signed-off-by: NYuchung Cheng <ycheng@google.com>
      Acked-by: NEric Dumazet <edumazet@google.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      2100c8d2
  3. 12 7月, 2012 1 次提交
    • E
      tcp: TCP Small Queues · 46d3ceab
      Eric Dumazet 提交于
      This introduce TSQ (TCP Small Queues)
      
      TSQ goal is to reduce number of TCP packets in xmit queues (qdisc &
      device queues), to reduce RTT and cwnd bias, part of the bufferbloat
      problem.
      
      sk->sk_wmem_alloc not allowed to grow above a given limit,
      allowing no more than ~128KB [1] per tcp socket in qdisc/dev layers at a
      given time.
      
      TSO packets are sized/capped to half the limit, so that we have two
      TSO packets in flight, allowing better bandwidth use.
      
      As a side effect, setting the limit to 40000 automatically reduces the
      standard gso max limit (65536) to 40000/2 : It can help to reduce
      latencies of high prio packets, having smaller TSO packets.
      
      This means we divert sock_wfree() to a tcp_wfree() handler, to
      queue/send following frames when skb_orphan() [2] is called for the
      already queued skbs.
      
      Results on my dev machines (tg3/ixgbe nics) are really impressive,
      using standard pfifo_fast, and with or without TSO/GSO.
      
      Without reduction of nominal bandwidth, we have reduction of buffering
      per bulk sender :
      < 1ms on Gbit (instead of 50ms with TSO)
      < 8ms on 100Mbit (instead of 132 ms)
      
      I no longer have 4 MBytes backlogged in qdisc by a single netperf
      session, and both side socket autotuning no longer use 4 Mbytes.
      
      As skb destructor cannot restart xmit itself ( as qdisc lock might be
      taken at this point ), we delegate the work to a tasklet. We use one
      tasklest per cpu for performance reasons.
      
      If tasklet finds a socket owned by the user, it sets TSQ_OWNED flag.
      This flag is tested in a new protocol method called from release_sock(),
      to eventually send new segments.
      
      [1] New /proc/sys/net/ipv4/tcp_limit_output_bytes tunable
      [2] skb_orphan() is usually called at TX completion time,
        but some drivers call it in their start_xmit() handler.
        These drivers should at least use BQL, or else a single TCP
        session can still fill the whole NIC TX ring, since TSQ will
        have no effect.
      Signed-off-by: NEric Dumazet <edumazet@google.com>
      Cc: Dave Taht <dave.taht@bufferbloat.net>
      Cc: Tom Herbert <therbert@google.com>
      Cc: Matt Mathis <mattmathis@google.com>
      Cc: Yuchung Cheng <ycheng@google.com>
      Cc: Nandita Dukkipati <nanditad@google.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      46d3ceab
  4. 11 7月, 2012 2 次提交
  5. 20 6月, 2012 1 次提交
    • D
      ipv4: Early TCP socket demux. · 41063e9d
      David S. Miller 提交于
      Input packet processing for local sockets involves two major demuxes.
      One for the route and one for the socket.
      
      But we can optimize this down to one demux for certain kinds of local
      sockets.
      
      Currently we only do this for established TCP sockets, but it could
      at least in theory be expanded to other kinds of connections.
      
      If a TCP socket is established then it's identity is fully specified.
      
      This means that whatever input route was used during the three-way
      handshake must work equally well for the rest of the connection since
      the keys will not change.
      
      Once we move to established state, we cache the receive packet's input
      route to use later.
      
      Like the existing cached route in sk->sk_dst_cache used for output
      packets, we have to check for route invalidations using dst->obsolete
      and dst->ops->check().
      
      Early demux occurs outside of a socket locked section, so when a route
      invalidation occurs we defer the fixup of sk->sk_rx_dst until we are
      actually inside of established state packet processing and thus have
      the socket locked.
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      41063e9d
  6. 10 6月, 2012 1 次提交
  7. 09 6月, 2012 1 次提交
    • D
      tcp: Get rid of inetpeer special cases. · 4670fd81
      David S. Miller 提交于
      The get_peer method TCP uses is full of special cases that make no
      sense accommodating, and it also gets in the way of doing more
      reasonable things here.
      
      First of all, if the socket doesn't have a usable cached route, there
      is no sense in trying to optimize timewait recycling.
      
      Likewise for the case where we have IP options, such as SRR enabled,
      that make the IP header destination address (and thus the destination
      address of the route key) differ from that of the connection's
      destination address.
      
      Just return a NULL peer in these cases, and thus we're also able to
      get rid of the clumsy inetpeer release logic.
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4670fd81
  8. 18 5月, 2012 1 次提交
  9. 03 5月, 2012 1 次提交
    • Y
      tcp: early retransmit · eed530b6
      Yuchung Cheng 提交于
      This patch implements RFC 5827 early retransmit (ER) for TCP.
      It reduces DUPACK threshold (dupthresh) if outstanding packets are
      less than 4 to recover losses by fast recovery instead of timeout.
      
      While the algorithm is simple, small but frequent network reordering
      makes this feature dangerous: the connection repeatedly enter
      false recovery and degrade performance. Therefore we implement
      a mitigation suggested in the appendix of the RFC that delays
      entering fast recovery by a small interval, i.e., RTT/4. Currently
      ER is conservative and is disabled for the rest of the connection
      after the first reordering event. A large scale web server
      experiment on the performance impact of ER is summarized in
      section 6 of the paper "Proportional Rate Reduction for TCP”,
      IMC 2011. http://conferences.sigcomm.org/imc/2011/docs/p155.pdf
      
      Note that Linux has a similar feature called THIN_DUPACK. The
      differences are THIN_DUPACK do not mitigate reorderings and is only
      used after slow start. Currently ER is disabled if THIN_DUPACK is
      enabled. I would be happy to merge THIN_DUPACK feature with ER if
      people think it's a good idea.
      
      ER is enabled by sysctl_tcp_early_retrans:
        0: Disables ER
      
        1: Reduce dupthresh to packets_out - 1 when outstanding packets < 4.
      
        2: (Default) reduce dupthresh like mode 1. In addition, delay
           entering fast recovery by RTT/4.
      
      Note: mode 2 is implemented in the third part of this patch series.
      Signed-off-by: NYuchung Cheng <ycheng@google.com>
      Acked-by: NNeal Cardwell <ncardwell@google.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      eed530b6
  10. 01 2月, 2012 1 次提交
    • E
      tcp: md5: rcu conversion · a915da9b
      Eric Dumazet 提交于
      In order to be able to support proper RST messages for TCP MD5 flows, we
      need to allow access to MD5 keys without locking listener socket.
      
      This conversion is a nice cleanup, and shrinks size of timewait sockets
      by 80 bytes.
      
      IPv6 code reuses generic code found in IPv4 instead of duplicating it.
      
      Control path uses GFP_KERNEL allocations instead of GFP_ATOMIC.
      Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
      Cc: Shawn Lu <shawn.lu@ericsson.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      a915da9b
  11. 12 12月, 2011 1 次提交
  12. 01 12月, 2011 1 次提交
  13. 23 11月, 2011 1 次提交
  14. 09 11月, 2011 1 次提交
  15. 27 10月, 2011 1 次提交
  16. 21 10月, 2011 1 次提交
  17. 19 10月, 2011 1 次提交
  18. 09 6月, 2011 1 次提交
    • J
      tcp: RFC2988bis + taking RTT sample from 3WHS for the passive open side · 9ad7c049
      Jerry Chu 提交于
      This patch lowers the default initRTO from 3secs to 1sec per
      RFC2988bis. It falls back to 3secs if the SYN or SYN-ACK packet
      has been retransmitted, AND the TCP timestamp option is not on.
      
      It also adds support to take RTT sample during 3WHS on the passive
      open side, just like its active open counterpart, and uses it, if
      valid, to seed the initRTO for the data transmission phase.
      
      The patch also resets ssthresh to its initial default at the
      beginning of the data transmission phase, and reduces cwnd to 1 if
      there has been MORE THAN ONE retransmission during 3WHS per RFC5681.
      Signed-off-by: NH.K. Jerry Chu <hkchu@google.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      9ad7c049
  19. 09 12月, 2010 1 次提交
  20. 02 12月, 2010 1 次提交
    • D
      timewait_sock: Create and use getpeer op. · ccb7c410
      David S. Miller 提交于
      The only thing AF-specific about remembering the timestamp
      for a time-wait TCP socket is getting the peer.
      
      Abstract that behind a new timewait_sock_ops vector.
      
      Support for real IPV6 sockets is not filled in yet, but
      curiously this makes timewait recycling start to work
      for v4-mapped ipv6 sockets.
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      ccb7c410
  21. 01 12月, 2010 1 次提交
  22. 24 9月, 2010 1 次提交
  23. 13 7月, 2010 1 次提交
  24. 30 3月, 2010 1 次提交
    • T
      include cleanup: Update gfp.h and slab.h includes to prepare for breaking... · 5a0e3ad6
      Tejun Heo 提交于
      include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
      
      percpu.h is included by sched.h and module.h and thus ends up being
      included when building most .c files.  percpu.h includes slab.h which
      in turn includes gfp.h making everything defined by the two files
      universally available and complicating inclusion dependencies.
      
      percpu.h -> slab.h dependency is about to be removed.  Prepare for
      this change by updating users of gfp and slab facilities include those
      headers directly instead of assuming availability.  As this conversion
      needs to touch large number of source files, the following script is
      used as the basis of conversion.
      
        http://userweb.kernel.org/~tj/misc/slabh-sweep.py
      
      The script does the followings.
      
      * Scan files for gfp and slab usages and update includes such that
        only the necessary includes are there.  ie. if only gfp is used,
        gfp.h, if slab is used, slab.h.
      
      * When the script inserts a new include, it looks at the include
        blocks and try to put the new include such that its order conforms
        to its surrounding.  It's put in the include block which contains
        core kernel includes, in the same order that the rest are ordered -
        alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
        doesn't seem to be any matching order.
      
      * If the script can't find a place to put a new include (mostly
        because the file doesn't have fitting include block), it prints out
        an error message indicating which .h file needs to be added to the
        file.
      
      The conversion was done in the following steps.
      
      1. The initial automatic conversion of all .c files updated slightly
         over 4000 files, deleting around 700 includes and adding ~480 gfp.h
         and ~3000 slab.h inclusions.  The script emitted errors for ~400
         files.
      
      2. Each error was manually checked.  Some didn't need the inclusion,
         some needed manual addition while adding it to implementation .h or
         embedding .c file was more appropriate for others.  This step added
         inclusions to around 150 files.
      
      3. The script was run again and the output was compared to the edits
         from #2 to make sure no file was left behind.
      
      4. Several build tests were done and a couple of problems were fixed.
         e.g. lib/decompress_*.c used malloc/free() wrappers around slab
         APIs requiring slab.h to be added manually.
      
      5. The script was run on all .h files but without automatically
         editing them as sprinkling gfp.h and slab.h inclusions around .h
         files could easily lead to inclusion dependency hell.  Most gfp.h
         inclusion directives were ignored as stuff from gfp.h was usually
         wildly available and often used in preprocessor macros.  Each
         slab.h inclusion directive was examined and added manually as
         necessary.
      
      6. percpu.h was updated not to include slab.h.
      
      7. Build test were done on the following configurations and failures
         were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
         distributed build env didn't work with gcov compiles) and a few
         more options had to be turned off depending on archs to make things
         build (like ipr on powerpc/64 which failed due to missing writeq).
      
         * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
         * powerpc and powerpc64 SMP allmodconfig
         * sparc and sparc64 SMP allmodconfig
         * ia64 SMP allmodconfig
         * s390 SMP allmodconfig
         * alpha SMP allmodconfig
         * um on x86_64 SMP allmodconfig
      
      8. percpu.h modifications were reverted so that it could be applied as
         a separate patch and serve as bisection point.
      
      Given the fact that I had only a couple of failures from tests on step
      6, I'm fairly confident about the coverage of this conversion patch.
      If there is a breakage, it's likely to be something in one of the arch
      headers which should be easily discoverable easily on most builds of
      the specific arch.
      Signed-off-by: NTejun Heo <tj@kernel.org>
      Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
      Cc: Ingo Molnar <mingo@redhat.com>
      Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
      5a0e3ad6
  25. 22 3月, 2010 1 次提交
  26. 06 3月, 2010 1 次提交
  27. 16 12月, 2009 1 次提交
    • D
      tcp: Revert per-route SACK/DSACK/TIMESTAMP changes. · bb5b7c11
      David S. Miller 提交于
      It creates a regression, triggering badness for SYN_RECV
      sockets, for example:
      
      [19148.022102] Badness at net/ipv4/inet_connection_sock.c:293
      [19148.022570] NIP: c02a0914 LR: c02a0904 CTR: 00000000
      [19148.023035] REGS: eeecbd30 TRAP: 0700   Not tainted  (2.6.32)
      [19148.023496] MSR: 00029032 <EE,ME,CE,IR,DR>  CR: 24002442  XER: 00000000
      [19148.024012] TASK = eee9a820[1756] 'privoxy' THREAD: eeeca000
      
      This is likely caused by the change in the 'estab' parameter
      passed to tcp_parse_options() when invoked by the functions
      in net/ipv4/tcp_minisocks.c
      
      But even if that is fixed, the ->conn_request() changes made in
      this patch series is fundamentally wrong.  They try to use the
      listening socket's 'dst' to probe the route settings.  The
      listening socket doesn't even have a route, and you can't
      get the right route (the child request one) until much later
      after we setup all of the state, and it must be done by hand.
      
      This stuff really isn't ready, so the best thing to do is a
      full revert.  This reverts the following commits:
      
      f55017a9
      022c3f7d
      1aba721e
      cda42ebd
      345cda2f
      dc343475
      05eaade2
      6a2a2d6bSigned-off-by: NDavid S. Miller <davem@davemloft.net>
      bb5b7c11
  28. 03 12月, 2009 3 次提交
    • W
      TCPCT part 1g: Responder Cookie => Initiator · 4957faad
      William Allen Simpson 提交于
      Parse incoming TCP_COOKIE option(s).
      
      Calculate <SYN,ACK> TCP_COOKIE option.
      
      Send optional <SYN,ACK> data.
      
      This is a significantly revised implementation of an earlier (year-old)
      patch that no longer applies cleanly, with permission of the original
      author (Adam Langley):
      
          http://thread.gmane.org/gmane.linux.network/102586
      
      Requires:
         TCPCT part 1a: add request_values parameter for sending SYNACK
         TCPCT part 1b: generate Responder Cookie secret
         TCPCT part 1c: sysctl_tcp_cookie_size, socket option TCP_COOKIE_TRANSACTIONS
         TCPCT part 1d: define TCP cookie option, extend existing struct's
         TCPCT part 1e: implement socket option TCP_COOKIE_TRANSACTIONS
         TCPCT part 1f: Initiator Cookie => Responder
      
      Signed-off-by: William.Allen.Simpson@gmail.com
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4957faad
    • W
      TCPCT part 1d: define TCP cookie option, extend existing struct's · 435cf559
      William Allen Simpson 提交于
      Data structures are carefully composed to require minimal additions.
      For example, the struct tcp_options_received cookie_plus variable fits
      between existing 16-bit and 8-bit variables, requiring no additional
      space (taking alignment into consideration).  There are no additions to
      tcp_request_sock, and only 1 pointer in tcp_sock.
      
      This is a significantly revised implementation of an earlier (year-old)
      patch that no longer applies cleanly, with permission of the original
      author (Adam Langley):
      
          http://thread.gmane.org/gmane.linux.network/102586
      
      The principle difference is using a TCP option to carry the cookie nonce,
      instead of a user configured offset in the data.  This is more flexible and
      less subject to user configuration error.  Such a cookie option has been
      suggested for many years, and is also useful without SYN data, allowing
      several related concepts to use the same extension option.
      
          "Re: SYN floods (was: does history repeat itself?)", September 9, 1996.
          http://www.merit.net/mail.archives/nanog/1996-09/msg00235.html
      
          "Re: what a new TCP header might look like", May 12, 1998.
          ftp://ftp.isi.edu/end2end/end2end-interest-1998.mail
      
      These functions will also be used in subsequent patches that implement
      additional features.
      
      Requires:
         TCPCT part 1a: add request_values parameter for sending SYNACK
         TCPCT part 1b: generate Responder Cookie secret
         TCPCT part 1c: sysctl_tcp_cookie_size, socket option TCP_COOKIE_TRANSACTIONS
      
      Signed-off-by: William.Allen.Simpson@gmail.com
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      435cf559
    • W
      TCPCT part 1a: add request_values parameter for sending SYNACK · e6b4d113
      William Allen Simpson 提交于
      Add optional function parameters associated with sending SYNACK.
      These parameters are not needed after sending SYNACK, and are not
      used for retransmission.  Avoids extending struct tcp_request_sock,
      and avoids allocating kernel memory.
      
      Also affects DCCP as it uses common struct request_sock_ops,
      but this parameter is currently reserved for future use.
      
      Signed-off-by: William.Allen.Simpson@gmail.com
      Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      e6b4d113
  29. 22 11月, 2009 1 次提交
  30. 14 11月, 2009 1 次提交
  31. 05 11月, 2009 1 次提交
  32. 29 10月, 2009 2 次提交
  33. 20 10月, 2009 2 次提交
    • J
      tcp: accept socket after TCP_DEFER_ACCEPT period · d1b99ba4
      Julian Anastasov 提交于
      Willy Tarreau and many other folks in recent years
      were concerned what happens when the TCP_DEFER_ACCEPT period
      expires for clients which sent ACK packet. They prefer clients
      that actively resend ACK on our SYN-ACK retransmissions to be
      converted from open requests to sockets and queued to the
      listener for accepting after the deferring period is finished.
      Then application server can decide to wait longer for data
      or to properly terminate the connection with FIN if read()
      returns EAGAIN which is an indication for accepting after
      the deferring period. This change still can have side effects
      for applications that expect always to see data on the accepted
      socket. Others can be prepared to work in both modes (with or
      without TCP_DEFER_ACCEPT period) and their data processing can
      ignore the read=EAGAIN notification and to allocate resources for
      clients which proved to have no data to send during the deferring
      period. OTOH, servers that use TCP_DEFER_ACCEPT=1 as flag (not
      as a timeout) to wait for data will notice clients that didn't
      send data for 3 seconds but that still resend ACKs.
      Thanks to Willy Tarreau for the initial idea and to
      Eric Dumazet for the review and testing the change.
      Signed-off-by: NJulian Anastasov <ja@ssi.bg>
      Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d1b99ba4
    • D
      Revert "tcp: fix tcp_defer_accept to consider the timeout" · a1a2ad91
      David S. Miller 提交于
      This reverts commit 6d01a026.
      
      Julian Anastasov, Willy Tarreau and Eric Dumazet have come up
      with a more correct way to deal with this.
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      a1a2ad91
  34. 13 10月, 2009 1 次提交
    • W
      tcp: fix tcp_defer_accept to consider the timeout · 6d01a026
      Willy Tarreau 提交于
      I was trying to use TCP_DEFER_ACCEPT and noticed that if the
      client does not talk, the connection is never accepted and
      remains in SYN_RECV state until the retransmits expire, where
      it finally is deleted. This is bad when some firewall such as
      netfilter sits between the client and the server because the
      firewall sees the connection in ESTABLISHED state while the
      server will finally silently drop it without sending an RST.
      
      This behaviour contradicts the man page which says it should
      wait only for some time :
      
             TCP_DEFER_ACCEPT (since Linux 2.4)
                Allows a listener to be awakened only when data arrives
                on the socket.  Takes an integer value  (seconds), this
                can  bound  the  maximum  number  of attempts TCP will
                make to complete the connection. This option should not
                be used in code intended to be portable.
      
      Also, looking at ipv4/tcp.c, a retransmit counter is correctly
      computed :
      
              case TCP_DEFER_ACCEPT:
                      icsk->icsk_accept_queue.rskq_defer_accept = 0;
                      if (val > 0) {
                              /* Translate value in seconds to number of
                               * retransmits */
                              while (icsk->icsk_accept_queue.rskq_defer_accept < 32 &&
                                     val > ((TCP_TIMEOUT_INIT / HZ) <<
                                             icsk->icsk_accept_queue.rskq_defer_accept))
                                      icsk->icsk_accept_queue.rskq_defer_accept++;
                              icsk->icsk_accept_queue.rskq_defer_accept++;
                      }
                      break;
      
      ==> rskq_defer_accept is used as a counter of retransmits.
      
      But in tcp_minisocks.c, this counter is only checked. And in
      fact, I have found no location which updates it. So I think
      that what was intended was to decrease it in tcp_minisocks
      whenever it is checked, which the trivial patch below does.
      Signed-off-by: NWilly Tarreau <w@1wt.eu>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6d01a026
  35. 16 9月, 2009 1 次提交
    • R
      tcp: fix CONFIG_TCP_MD5SIG + CONFIG_PREEMPT timer BUG() · 657e9649
      Robert Varga 提交于
      I have recently came across a preemption imbalance detected by:
      
      <4>huh, entered ffffffff80644630 with preempt_count 00000102, exited with 00000101?
      <0>------------[ cut here ]------------
      <2>kernel BUG at /usr/src/linux/kernel/timer.c:664!
      <0>invalid opcode: 0000 [1] PREEMPT SMP
      
      with ffffffff80644630 being inet_twdr_hangman().
      
      This appeared after I enabled CONFIG_TCP_MD5SIG and played with it a
      bit, so I looked at what might have caused it.
      
      One thing that struck me as strange is tcp_twsk_destructor(), as it
      calls tcp_put_md5sig_pool() -- which entails a put_cpu(), causing the
      detected imbalance. Found on 2.6.23.9, but 2.6.31 is affected as well,
      as far as I can tell.
      Signed-off-by: NRobert Varga <nite@hq.alert.sk>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      657e9649