1. 22 12月, 2011 3 次提交
  2. 21 12月, 2011 2 次提交
    • G
      net: have ipconfig not wait if no dev is available · cd7816d1
      Gerlando Falauto 提交于
      previous commit 3fb72f1e
      makes IP-Config wait for carrier on at least one network device.
      
      Before waiting (predefined value 120s), check that at least one device
      was successfully brought up. Otherwise (e.g. buggy bootloader
      which does not set the MAC address) there is no point in waiting
      for carrier.
      
      Cc: Micha Nelissen <micha@neli.hopto.org>
      Cc: Holger Brunck <holger.brunck@keymile.com>
      Signed-off-by: NGerlando Falauto <gerlando.falauto@keymile.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      cd7816d1
    • T
      sctp: Do not account for sizeof(struct sk_buff) in estimated rwnd · a76c0adf
      Thomas Graf 提交于
      When checking whether a DATA chunk fits into the estimated rwnd a
      full sizeof(struct sk_buff) is added to the needed chunk size. This
      quickly exhausts the available rwnd space and leads to packets being
      sent which are much below the PMTU limit. This can lead to much worse
      performance.
      
      The reason for this behaviour was to avoid putting too much memory
      pressure on the receiver. The concept is not completely irational
      because a Linux receiver does in fact clone an skb for each DATA chunk
      delivered. However, Linux also reserves half the available socket
      buffer space for data structures therefore usage of it is already
      accounted for.
      
      When proposing to change this the last time it was noted that this
      behaviour was introduced to solve a performance issue caused by rwnd
      overusage in combination with small DATA chunks.
      
      Trying to reproduce this I found that with the sk_buff overhead removed,
      the performance would improve significantly unless socket buffer limits
      are increased.
      
      The following numbers have been gathered using a patched iperf
      supporting SCTP over a live 1 Gbit ethernet network. The -l option
      was used to limit DATA chunk sizes. The numbers listed are based on
      the average of 3 test runs each. Default values have been used for
      sk_(r|w)mem.
      
      Chunk
      Size    Unpatched     No Overhead
      -------------------------------------
         4    15.2 Kbit [!]   12.2 Mbit [!]
         8    35.8 Kbit [!]   26.0 Mbit [!]
        16    95.5 Kbit [!]   54.4 Mbit [!]
        32   106.7 Mbit      102.3 Mbit
        64   189.2 Mbit      188.3 Mbit
       128   331.2 Mbit      334.8 Mbit
       256   537.7 Mbit      536.0 Mbit
       512   766.9 Mbit      766.6 Mbit
      1024   810.1 Mbit      808.6 Mbit
      Signed-off-by: NThomas Graf <tgraf@redhat.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      a76c0adf
  3. 20 12月, 2011 8 次提交
  4. 19 12月, 2011 4 次提交
  5. 17 12月, 2011 20 次提交
  6. 16 12月, 2011 3 次提交