1. 26 11月, 2008 3 次提交
  2. 25 11月, 2008 6 次提交
    • J
      DCB: fix kconfig option · 7a6b6f51
      Jeff Kirsher 提交于
      Since the netlink option for DCB is necessary to actually be useful,
      simplified the Kconfig option.  In addition, added useful help text for the
      Kconfig option.
      Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      7a6b6f51
    • S
      netdev: add HAVE_NET_DEVICE_OPS · 47fd5b83
      Stephen Hemminger 提交于
      As a concession to vendors who have to deal with one source for different
      kernel versions, add a HAVE_NET_DEVICE_OPS so they don't end up hard
      coding ifdef against kernel version.
      Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      47fd5b83
    • I
      111cc8b9
    • I
      tcp: Try to restore large SKBs while SACK processing · 832d11c5
      Ilpo Järvinen 提交于
      During SACK processing, most of the benefits of TSO are eaten by
      the SACK blocks that one-by-one fragment SKBs to MSS sized chunks.
      Then we're in problems when cleanup work for them has to be done
      when a large cumulative ACK comes. Try to return back to pre-split
      state already while more and more SACK info gets discovered by
      combining newly discovered SACK areas with the previous skb if
      that's SACKed as well.
      
      This approach has a number of benefits:
      
      1) The processing overhead is spread more equally over the RTT
      2) Write queue has less skbs to process (affect everything
         which has to walk in the queue past the sacked areas)
      3) Write queue is consistent whole the time, so no other parts
         of TCP has to be aware of this (this was not the case with
         some other approach that was, well, quite intrusive all
         around).
      4) Clean_rtx_queue can release most of the pages using single
         put_page instead of previous PAGE_SIZE/mss+1 calls
      
      In case a hole is fully filled by the new SACK block, we attempt
      to combine the next skb too which allows construction of skbs
      that are even larger than what tso split them to and it handles
      hole per on every nth patterns that often occur during slow start
      overshoot pretty nicely. Though this to be really useful also
      a retransmission would have to get lost since cumulative ACKs
      advance one hole at a time in the most typical case.
      
      TODO: handle upwards only merging. That should be rather easy
      when segment is fully sacked but I'm leaving that as future
      work item (it won't make very large difference anyway since
      this current approach already covers quite a lot of normal
      cases).
      
      I was earlier thinking of some sophisticated way of tracking
      timestamps of the first and the last segment but later on
      realized that it won't be that necessary at all to store the
      timestamp of the last segment. The cases that can occur are
      basically either:
        1) ambiguous => no sensible measurement can be taken anyway
        2) non-ambiguous is due to reordering => having the timestamp
           of the last segment there is just skewing things more off
           than does some good since the ack got triggered by one of
           the holes (besides some substle issues that would make
           determining right hole/skb even harder problem). Anyway,
           it has nothing to do with this change then.
      
      I choose to route some abnormal looking cases with goto noop,
      some could be handled differently (eg., by stopping the
      walking at that skb but again). In general, they either
      shouldn't happen at all or are rare enough to make no difference
      in practice.
      
      In theory this change (as whole) could cause some macroscale
      regression (global) because of cache misses that are taken over
      the round-trip time but it gets very likely better because of much
      less (local) cache misses per other write queue walkers and the
      big recovery clearing cumulative ack.
      
      Worth to note that these benefits would be very easy to get also
      without TSO/GSO being on as long as the data is in pages so that
      we can merge them. Currently I won't let that happen because
      DSACK splitting at fragment that would mess up pcounts due to
      sk_can_gso in tcp_set_skb_tso_segs. Once DSACKs fragments gets
      avoided, we have some conditions that can be made less strict.
      
      TODO: I will probably have to convert the excessive pointer
      passing to struct sacktag_state... :-)
      
      My testing revealed that considerable amount of skbs couldn't
      be shifted because they were cloned (most likely still awaiting
      tx reclaim)...
      
      [The rest is considering future work instead since I got
      repeatably EFAULT to tcpdump's recvfrom when I added
      pskb_expand_head to deal with clones, so I separated that
      into another, later patch]
      
      ...To counter that, I gave up on the fifth advantage:
      
      5) When growing previous SACK block, less allocs for new skbs
         are done, basically a new alloc is needed only when new hole
         is detected and when the previous skb runs out of frags space
      
      ...which now only happens of if reclaim is fast enough to dispose
      the clone before the SACK block comes in (the window is RTT long),
      otherwise we'll have to alloc some.
      
      With clones being handled I got these numbers (will be somewhat
      worse without that), taken with fine-grained mibs:
      
                        TCPSackShifted 398
                         TCPSackMerged 877
                  TCPSackShiftFallback 320
            TCPSACKCOLLAPSEFALLBACKGSO 0
        TCPSACKCOLLAPSEFALLBACKSKBBITS 0
        TCPSACKCOLLAPSEFALLBACKSKBDATA 0
          TCPSACKCOLLAPSEFALLBACKBELOW 0
          TCPSACKCOLLAPSEFALLBACKFIRST 1
       TCPSACKCOLLAPSEFALLBACKPREVBITS 318
            TCPSACKCOLLAPSEFALLBACKMSS 1
         TCPSACKCOLLAPSEFALLBACKNOHEAD 0
          TCPSACKCOLLAPSEFALLBACKSHIFT 0
                TCPSACKCOLLAPSENOOPSEQ 0
        TCPSACKCOLLAPSENOOPSMALLPCOUNT 0
           TCPSACKCOLLAPSENOOPSMALLLEN 0
                   TCPSACKCOLLAPSEHOLE 12
      Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      832d11c5
    • I
      e1aa680f
    • E
      net: avoid a pair of dst_hold()/dst_release() in ip_append_data() · 2e77d89b
      Eric Dumazet 提交于
      We can reduce pressure on dst entry refcount that slowdown UDP transmit
      path on SMP machines. This pressure is visible on RTP servers when
      delivering content to mediagateways, especially big ones, handling
      thousand of streams. Several cpus send UDP frames to the same
      destination, hence use the same dst entry.
      
      This patch makes ip_append_data() eventually steal the refcount its
      callers had to take on the dst entry.
      
      This doesnt avoid all refcounting, but still gives speedups on SMP,
      on UDP/RAW transmit path
      Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      2e77d89b
  3. 24 11月, 2008 3 次提交
    • E
      eth: Declare an optimized compare_ether_addr_64bits() function · 1f87e235
      Eric Dumazet 提交于
      Linus mentioned we could try to perform long word operations, even
      on potentially unaligned addresses, on x86 at least. David mentioned
      the HAVE_EFFICIENT_UNALIGNED_ACCESS test to handle this on all
      arches that have efficient unailgned accesses.
      
      I tried this idea and got nice assembly on 32 bits:
      
      158:   33 82 38 01 00 00       xor    0x138(%edx),%eax
      15e:   33 8a 34 01 00 00       xor    0x134(%edx),%ecx
      164:   c1 e0 10                shl    $0x10,%eax
      167:   09 c1                   or     %eax,%ecx
      169:   74 0b                   je     176 <eth_type_trans+0x87>
      
      And very nice assembly on 64 bits of course (one xor, one shl)
      
      Nice oprofile improvement in eth_type_trans(), 0.17 % instead of 0.41 %,
      expected since we remove 8 instructions on a fast path.
      
      This patch implements a compare_ether_addr_64bits() function, that
      uses the CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS ifdef to efficiently
      perform the 6 bytes comparison on all capable arches.
      Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      1f87e235
    • E
      net: Convert TCP/DCCP listening hash tables to use RCU · c25eb3bf
      Eric Dumazet 提交于
      This is the last step to be able to perform full RCU lookups
      in __inet_lookup() : After established/timewait tables, we
      add RCU lookups to listening hash table.
      
      The only trick here is that a socket of a given type (TCP ipv4,
      TCP ipv6, ...) can now flight between two different tables
      (established and listening) during a RCU grace period, so we
      must use different 'nulls' end-of-chain values for two tables.
      
      We define a large value :
      
      #define LISTENING_NULLS_BASE (1U << 29)
      
      So that slots in listening table are guaranteed to have different
      end-of-chain values than slots in established table. A reader can
      still detect it finished its lookup in the right chain.
      Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      c25eb3bf
    • G
      dccp: Set per-connection CCIDs via socket options · b20a9c24
      Gerrit Renker 提交于
      With this patch, TX/RX CCIDs can now be changed on a per-connection
      basis, which overrides the defaults set by the global sysctl variables
      for TX/RX CCIDs.
      
      To make full use of this facility, the remaining patches of this patch
      set are needed, which track dependencies and activate negotiated
      feature values.
      Signed-off-by: NGerrit Renker <gerrit@erg.abdn.ac.uk>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      b20a9c24
  4. 22 11月, 2008 9 次提交
  5. 21 11月, 2008 9 次提交
  6. 20 11月, 2008 10 次提交
    • P
      pkt_sched: add DRR scheduler · 13d2a1d2
      Patrick McHardy 提交于
      Add classful DRR scheduler as a more flexible replacement for SFQ.
      
      The main difference to the algorithm described in "Efficient Fair Queueing
      using Deficit Round Robin" is that this implementation doesn't drop packets
      from the longest queue on overrun because its classful and limits are
      handled by each individual child qdisc.
      Signed-off-by: NPatrick McHardy <kaber@trash.net>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      13d2a1d2
    • P
      netlink: avoid memset of 0 bytes sparse warning · 0c19b0ad
      Patrick McHardy 提交于
      A netlink attribute padding of zero triggers this sparse warning:
      
      include/linux/netlink.h:245:8: warning: memset with byte count of 0
      
      Avoid the memset when the size parameter is constant and requires no padding.
      Signed-off-by: NPatrick McHardy <kaber@trash.net>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      0c19b0ad
    • P
      filter: add SKF_AD_NLATTR_NEST to look for nested attributes · d214c753
      Pablo Neira Ayuso 提交于
      SKF_AD_NLATTR allows us to find the first matching attribute in a
      stream of netlink attributes from one offset to the end of the
      netlink message. This is not suitable to look for a specific
      matching inside a set of nested attributes.
      
      For example, in ctnetlink messages, if we look for the CTA_V6_SRC
      attribute in a message that talks about an IPv4 connection,
      SKF_AD_NLATTR returns the offset of CTA_STATUS which has the same
      value of CTA_V6_SRC but outside the nest. To differenciate
      CTA_STATUS and CTA_V6_SRC, we would have to make assumptions on the
      size of the attribute and the usual offset, resulting in horrible
      BSF code.
      
      This patch adds SKF_AD_NLATTR_NEST, which is a variant of
      SKF_AD_NLATTR, that looks for an attribute inside the limits of
      a nested attributes, but not further.
      
      This patch validates that we have enough room to look for the
      nested attributes - based on a suggestion from Patrick McHardy.
      Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
      Acked-by: NPatrick McHardy <kaber@trash.net>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d214c753
    • E
      net: listening_hash get a spinlock per bucket · 5caea4ea
      Eric Dumazet 提交于
      This patch prepares RCU migration of listening_hash table for
      TCP/DCCP protocols.
      
      listening_hash table being small (32 slots per protocol), we add
      a spinlock for each slot, instead of a single rwlock for whole table.
      
      This should reduce hold time of readers, and writers concurrency.
      Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      5caea4ea
    • S
      netdev: expose ethernet address primitives · ccad637b
      Stephen Hemminger 提交于
      When ethernet devices are converted, the function pointer setup
      by eth_setup() need to be done during intialization.
      Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      ccad637b
    • S
      netdev: introduce dev_get_stats() · eeda3fd6
      Stephen Hemminger 提交于
      In order for the network device ops get_stats call to be immutable, the handling
      of the default internal network device stats block has to be changed. Add a new
      helper function which replaces the old use of internal_get_stats.
      
      Note: change return code to make it clear that the caller should not
      go changing the returned statistics.
      Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      eeda3fd6
    • S
      netdev: network device operations infrastructure · d314774c
      Stephen Hemminger 提交于
      This patch changes the network device internal API to move adminstrative
      operations out of the network device structure and into a separate structure.
      
      This patch involves some hackery to maintain compatablity between the
      new and old model, so all 300+ drivers don't have to be changed at once.
      For drivers that aren't converted yet, the netdevice_ops virt function list
      still resides in the net_device structure. For old protocols, the new
      net_device_ops are copied out to the old net_device pointers.
      
      After the transistion is completed the nag message can be changed to
      an WARN_ON, and the compatiablity code can be made configurable.
      
      Some function pointers aren't moved:
      * destructor can't be in net_device_ops because
        it may need to be referenced after the module is unloaded.
      * neighbor setup is manipulated in a couple of places that need special
        consideration
      * hard_start_xmit is in the fast path for transmit.
      Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d314774c
    • M
      cpuset: update top cpuset's mems after adding a node · f481891f
      Miao Xie 提交于
      After adding a node into the machine, top cpuset's mems isn't updated.
      
      By reviewing the code, we found that the update function
      
        cpuset_track_online_nodes()
      
      was invoked after node_states[N_ONLINE] changes.  It is wrong because
      N_ONLINE just means node has pgdat, and if node has/added memory, we use
      N_HIGH_MEMORY.  So, We should invoke the update function after
      node_states[N_HIGH_MEMORY] changes, just like its commit says.
      
      This patch fixes it.  And we use notifier of memory hotplug instead of
      direct calling of cpuset_track_online_nodes().
      Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
      Acked-by: NYasunori Goto <y-goto@jp.fujitsu.com>
      Cc: David Rientjes <rientjes@google.com>
      Cc: Paul Menage <menage@google.com
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      f481891f
    • U
      reintroduce accept4 · de11defe
      Ulrich Drepper 提交于
      Introduce a new accept4() system call.  The addition of this system call
      matches analogous changes in 2.6.27 (dup3(), evenfd2(), signalfd4(),
      inotify_init1(), epoll_create1(), pipe2()) which added new system calls
      that differed from analogous traditional system calls in adding a flags
      argument that can be used to access additional functionality.
      
      The accept4() system call is exactly the same as accept(), except that
      it adds a flags bit-mask argument.  Two flags are initially implemented.
      (Most of the new system calls in 2.6.27 also had both of these flags.)
      
      SOCK_CLOEXEC causes the close-on-exec (FD_CLOEXEC) flag to be enabled
      for the new file descriptor returned by accept4().  This is a useful
      security feature to avoid leaking information in a multithreaded
      program where one thread is doing an accept() at the same time as
      another thread is doing a fork() plus exec().  More details here:
      http://udrepper.livejournal.com/20407.html "Secure File Descriptor Handling",
      Ulrich Drepper).
      
      The other flag is SOCK_NONBLOCK, which causes the O_NONBLOCK flag
      to be enabled on the new open file description created by accept4().
      (This flag is merely a convenience, saving the use of additional calls
      fcntl(F_GETFL) and fcntl (F_SETFL) to achieve the same result.
      
      Here's a test program.  Works on x86-32.  Should work on x86-64, but
      I (mtk) don't have a system to hand to test with.
      
      It tests accept4() with each of the four possible combinations of
      SOCK_CLOEXEC and SOCK_NONBLOCK set/clear in 'flags', and verifies
      that the appropriate flags are set on the file descriptor/open file
      description returned by accept4().
      
      I tested Ulrich's patch in this thread by applying against 2.6.28-rc2,
      and it passes according to my test program.
      
      /* test_accept4.c
      
        Copyright (C) 2008, Linux Foundation, written by Michael Kerrisk
             <mtk.manpages@gmail.com>
      
        Licensed under the GNU GPLv2 or later.
      */
      #define _GNU_SOURCE
      #include <unistd.h>
      #include <sys/syscall.h>
      #include <sys/socket.h>
      #include <netinet/in.h>
      #include <stdlib.h>
      #include <fcntl.h>
      #include <stdio.h>
      #include <string.h>
      
      #define PORT_NUM 33333
      
      #define die(msg) do { perror(msg); exit(EXIT_FAILURE); } while (0)
      
      /**********************************************************************/
      
      /* The following is what we need until glibc gets a wrapper for
        accept4() */
      
      /* Flags for socket(), socketpair(), accept4() */
      #ifndef SOCK_CLOEXEC
      #define SOCK_CLOEXEC    O_CLOEXEC
      #endif
      #ifndef SOCK_NONBLOCK
      #define SOCK_NONBLOCK   O_NONBLOCK
      #endif
      
      #ifdef __x86_64__
      #define SYS_accept4 288
      #elif __i386__
      #define USE_SOCKETCALL 1
      #define SYS_ACCEPT4 18
      #else
      #error "Sorry -- don't know the syscall # on this architecture"
      #endif
      
      static int
      accept4(int fd, struct sockaddr *sockaddr, socklen_t *addrlen, int flags)
      {
         printf("Calling accept4(): flags = %x", flags);
         if (flags != 0) {
             printf(" (");
             if (flags & SOCK_CLOEXEC)
                 printf("SOCK_CLOEXEC");
             if ((flags & SOCK_CLOEXEC) && (flags & SOCK_NONBLOCK))
                 printf(" ");
             if (flags & SOCK_NONBLOCK)
                 printf("SOCK_NONBLOCK");
             printf(")");
         }
         printf("\n");
      
      #if USE_SOCKETCALL
         long args[6];
      
         args[0] = fd;
         args[1] = (long) sockaddr;
         args[2] = (long) addrlen;
         args[3] = flags;
      
         return syscall(SYS_socketcall, SYS_ACCEPT4, args);
      #else
         return syscall(SYS_accept4, fd, sockaddr, addrlen, flags);
      #endif
      }
      
      /**********************************************************************/
      
      static int
      do_test(int lfd, struct sockaddr_in *conn_addr,
             int closeonexec_flag, int nonblock_flag)
      {
         int connfd, acceptfd;
         int fdf, flf, fdf_pass, flf_pass;
         struct sockaddr_in claddr;
         socklen_t addrlen;
      
         printf("=======================================\n");
      
         connfd = socket(AF_INET, SOCK_STREAM, 0);
         if (connfd == -1)
             die("socket");
         if (connect(connfd, (struct sockaddr *) conn_addr,
                     sizeof(struct sockaddr_in)) == -1)
             die("connect");
      
         addrlen = sizeof(struct sockaddr_in);
         acceptfd = accept4(lfd, (struct sockaddr *) &claddr, &addrlen,
                            closeonexec_flag | nonblock_flag);
         if (acceptfd == -1) {
             perror("accept4()");
             close(connfd);
             return 0;
         }
      
         fdf = fcntl(acceptfd, F_GETFD);
         if (fdf == -1)
             die("fcntl:F_GETFD");
         fdf_pass = ((fdf & FD_CLOEXEC) != 0) ==
                    ((closeonexec_flag & SOCK_CLOEXEC) != 0);
         printf("Close-on-exec flag is %sset (%s); ",
                 (fdf & FD_CLOEXEC) ? "" : "not ",
                 fdf_pass ? "OK" : "failed");
      
         flf = fcntl(acceptfd, F_GETFL);
         if (flf == -1)
             die("fcntl:F_GETFD");
         flf_pass = ((flf & O_NONBLOCK) != 0) ==
                    ((nonblock_flag & SOCK_NONBLOCK) !=0);
         printf("nonblock flag is %sset (%s)\n",
                 (flf & O_NONBLOCK) ? "" : "not ",
                 flf_pass ? "OK" : "failed");
      
         close(acceptfd);
         close(connfd);
      
         printf("Test result: %s\n", (fdf_pass && flf_pass) ? "PASS" : "FAIL");
         return fdf_pass && flf_pass;
      }
      
      static int
      create_listening_socket(int port_num)
      {
         struct sockaddr_in svaddr;
         int lfd;
         int optval;
      
         memset(&svaddr, 0, sizeof(struct sockaddr_in));
         svaddr.sin_family = AF_INET;
         svaddr.sin_addr.s_addr = htonl(INADDR_ANY);
         svaddr.sin_port = htons(port_num);
      
         lfd = socket(AF_INET, SOCK_STREAM, 0);
         if (lfd == -1)
             die("socket");
      
         optval = 1;
         if (setsockopt(lfd, SOL_SOCKET, SO_REUSEADDR, &optval,
                        sizeof(optval)) == -1)
             die("setsockopt");
      
         if (bind(lfd, (struct sockaddr *) &svaddr,
                  sizeof(struct sockaddr_in)) == -1)
             die("bind");
      
         if (listen(lfd, 5) == -1)
             die("listen");
      
         return lfd;
      }
      
      int
      main(int argc, char *argv[])
      {
         struct sockaddr_in conn_addr;
         int lfd;
         int port_num;
         int passed;
      
         passed = 1;
      
         port_num = (argc > 1) ? atoi(argv[1]) : PORT_NUM;
      
         memset(&conn_addr, 0, sizeof(struct sockaddr_in));
         conn_addr.sin_family = AF_INET;
         conn_addr.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
         conn_addr.sin_port = htons(port_num);
      
         lfd = create_listening_socket(port_num);
      
         if (!do_test(lfd, &conn_addr, 0, 0))
             passed = 0;
         if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, 0))
             passed = 0;
         if (!do_test(lfd, &conn_addr, 0, SOCK_NONBLOCK))
             passed = 0;
         if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, SOCK_NONBLOCK))
             passed = 0;
      
         close(lfd);
      
         exit(passed ? EXIT_SUCCESS : EXIT_FAILURE);
      }
      
      [mtk.manpages@gmail.com: rewrote changelog, updated test program]
      Signed-off-by: NUlrich Drepper <drepper@redhat.com>
      Tested-by: NMichael Kerrisk <mtk.manpages@gmail.com>
      Acked-by: NMichael Kerrisk <mtk.manpages@gmail.com>
      Cc: <linux-api@vger.kernel.org>
      Cc: <linux-arch@vger.kernel.org>
      Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      de11defe
    • J
      include/net net/ - csum_partial - remove unnecessary casts · 07f0757a
      Joe Perches 提交于
      The first argument to csum_partial is const void *
      casts to char/u8 * are not necessary
      Signed-off-by: NJoe Perches <joe@perches.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      07f0757a