1. 11 May, 2009 6 commits
  2. 07 May, 2009 2 commits
  3. 06 May, 2009 7 commits
  4. 05 May, 2009 3 commits
    • Jiri Pirko's avatar
      net: introduce a list of device addresses dev_addr_list (v6) · f001fde5
      Jiri Pirko authored
      v5 -> v6 (current):
      -removed so far unused static functions
      -corrected dev_addr_del_multiple to call del instead of add
      
      v4 -> v5:
      -added device address type (suggested by davem)
      -removed refcounting (better to have simplier code then safe potentially few
       bytes)
      
      v3 -> v4:
      -changed kzalloc to kmalloc in __hw_addr_add_ii()
      -ASSERT_RTNL() avoided in dev_addr_flush() and dev_addr_init()
      
      v2 -> v3:
      -removed unnecessary rcu read locking
      -moved dev_addr_flush() calling to ensure no null dereference of dev_addr
      
      v1 -> v2:
      -added forgotten ASSERT_RTNL to dev_addr_init and dev_addr_flush
      -removed unnecessary rcu_read locking in dev_addr_init
      -use compare_ether_addr_64bits instead of compare_ether_addr
      -use L1_CACHE_BYTES as size for allocating struct netdev_hw_addr
      -use call_rcu instead of rcu_synchronize
      -moved is_etherdev_addr into __KERNEL__ ifdef
      
      This patch introduces a new list in struct net_device and brings a set of
      functions to handle the work with device address list. The list is a replacement
      for the original dev_addr field and because in some situations there is need to
      carry several device addresses with the net device. To be backward compatible,
      dev_addr is made to point to the first member of the list so original drivers
      sees no difference.
      Signed-off-by: default avatarJiri Pirko <jpirko@redhat.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      f001fde5
    • Pablo Neira Ayuso's avatar
      netfilter: xt_cluster: fix use of cluster match with 32 nodes · 280f37af
      Pablo Neira Ayuso authored
      This patch fixes a problem when you use 32 nodes in the cluster
      match:
      
      % iptables -I PREROUTING -t mangle -i eth0 -m cluster \
        --cluster-total-nodes  32  --cluster-local-node  32 \
        --cluster-hash-seed 0xdeadbeef -j MARK --set-mark 0xffff
      iptables: Invalid argument. Run `dmesg' for more information.
      % dmesg | tail -1
      xt_cluster: this node mask cannot be higher than the total number of nodes
      
      The problem is related to this checking:
      
      if (info->node_mask >= (1 << info->total_nodes)) {
      	printk(KERN_ERR "xt_cluster: this node mask cannot be "
      			"higher than the total number of nodes\n");
      	return false;
      }
      
      (1 << 32) is 1. Thus, the checking fails.
      
      BTW, I said this before but I insist: I have only tested the cluster
      match with 2 nodes getting ~45% extra performance in an active-active setup.
      The maximum limit of 32 nodes is still completely arbitrary. I'd really
      appreciate if people that have more nodes in their setups let me know.
      Signed-off-by: default avatarPablo Neira Ayuso <pablo@netfilter.org>
      Signed-off-by: default avatarPatrick McHardy <kaber@trash.net>
      280f37af
    • Patrick McHardy's avatar
      netfilter: add missing linux/types.h include to xt_LED.h · a7ca7fcc
      Patrick McHardy authored
      Pointed out by Dave Miller:
      
        CHECK   include/linux/netfilter (57 files)
      /home/davem/src/GIT/net-2.6/usr/include/linux/netfilter/xt_LED.h:6: found __[us]{8,16,32,64} type without #include <linux/types.h>
      Signed-off-by: default avatarPatrick McHardy <kaber@trash.net>
      a7ca7fcc
  5. 04 May, 2009 3 commits
    • Marcel Holtmann's avatar
      Bluetooth: Fix issue with sysfs handling for connections · a67e899c
      Marcel Holtmann authored
      Due to a semantic changes in flush_workqueue() the current approach of
      synchronizing the sysfs handling for connections doesn't work anymore. The
      whole approach is actually fully broken and based on assumptions that are
      no longer valid.
      
      With the introduction of Simple Pairing support, the creation of low-level
      ACL links got changed. This change invalidates the reason why in the past
      two independent work queues have been used for adding/removing sysfs
      devices. The adding of the actual sysfs device is now postponed until the
      host controller successfully assigns an unique handle to that link. So
      the real synchronization happens inside the controller and not the host.
      
      The only left-over problem is that some internals of the sysfs device
      handling are not initialized ahead of time. This leaves potential access
      to invalid data and can cause various NULL pointer dereferences. To fix
      this a new function makes sure that all sysfs details are initialized
      when an connection attempt is made. The actual sysfs device is only
      registered when the connection has been successfully established. To
      avoid a race condition with the registration, the check if a device is
      registered has been moved into the removal work.
      
      As an extra protection two flush_work() calls are left in place to
      make sure a previous add/del work has been completed first.
      
      Based on a report by Marc Pignat <marc.pignat@hevs.ch>
      Signed-off-by: default avatarMarcel Holtmann <marcel@holtmann.org>
      Tested-by: default avatarJustin P. Mattock <justinmattock@gmail.com>
      Tested-by: default avatarRoger Quadros <ext-roger.quadros@nokia.com>
      Tested-by: default avatarMarc Pignat <marc.pignat@hevs.ch>
      a67e899c
    • Omar Laazimani's avatar
      usbnet: CDC EEM support (v5) · 9f722c09
      Omar Laazimani authored
      This introduces a CDC Ethernet Emulation Model (EEM) host side
      driver to support USB EEM devices.
      
      EEM is different from the Ethernet Control Model (ECM) currently
      supported by the "CDC Ethernet" driver.  One key difference is
      that it doesn't require of USB interface alternate settings to
      manage interface state; some maldesigned hardware can't handle
      that part of USB.  It also avoids a separate USB interface for
      control and status updates.
      
      [ dbrownell@users.sourceforge.net: fix skb leaks, add rx packet
      checks, improve fault handling, EEM conformance updates, cleanup ]
      Signed-off-by: default avatarOmar Laazimani <omar.oberthur@gmail.com>
      Signed-off-by: default avatarDavid Brownell <dbrownell@users.sourceforge.net>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      9f722c09
    • Satoru SATOH's avatar
      tcp: Fix tcp_prequeue() to get correct rto_min value · 0c266898
      Satoru SATOH authored
      tcp_prequeue() refers to the constant value (TCP_RTO_MIN) regardless of
      the actual value might be tuned. The following patches fix this and make
      tcp_prequeue get the actual value returns from tcp_rto_min().
      Signed-off-by: default avatarSatoru SATOH <satoru.satoh@gmail.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      0c266898
  6. 03 May, 2009 1 commit
  7. 01 May, 2009 2 commits
  8. 29 Apr, 2009 8 commits
  9. 28 Apr, 2009 7 commits
    • Stephen Hemminger's avatar
      netfilter: revised locking for x_tables · 942e4a2b
      Stephen Hemminger authored
      The x_tables are organized with a table structure and a per-cpu copies
      of the counters and rules. On older kernels there was a reader/writer 
      lock per table which was a performance bottleneck. In 2.6.30-rc, this
      was converted to use RCU and the counters/rules which solved the performance
      problems for do_table but made replacing rules much slower because of
      the necessary RCU grace period.
      
      This version uses a per-cpu set of spinlocks and counters to allow to
      table processing to proceed without the cache thrashing of a global
      reader lock and keeps the same performance for table updates.
      Signed-off-by: default avatarStephen Hemminger <shemminger@vyatta.com>
      Acked-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      942e4a2b
    • Marcel Holtmann's avatar
      Bluetooth: Add different pairing timeout for Legacy Pairing · 052b30b0
      Marcel Holtmann authored
      The Bluetooth stack uses a reference counting for all established ACL
      links and if no user (L2CAP connection) is present, the link will be
      terminated to save power. The problem part is the dedicated pairing
      when using Legacy Pairing (Bluetooth 2.0 and before). At that point
      no user is present and pairing attempts will be disconnected within
      10 seconds or less. In previous kernel version this was not a problem
      since the disconnect timeout wasn't triggered on incoming connections
      for the first time. However this caused issues with broken host stacks
      that kept the connections around after dedicated pairing. When the
      support for Simple Pairing got added, the link establishment procedure
      needed to be changed and now causes issues when using Legacy Pairing
      
      When using Simple Pairing it is possible to do a proper reference
      counting of ACL link users. With Legacy Pairing this is not possible
      since the specification is unclear in some areas and too many broken
      Bluetooth devices have already been deployed. So instead of trying to
      deal with all the broken devices, a special pairing timeout will be
      introduced that increases the timeout to 60 seconds when pairing is
      triggered.
      
      If a broken devices now puts the stack into an unforeseen state, the
      worst that happens is the disconnect timeout triggers after 120 seconds
      instead of 4 seconds. This allows successful pairings with legacy and
      broken devices now.
      
      Based on a report by Johan Hedberg <johan.hedberg@nokia.com>
      Signed-off-by: default avatarMarcel Holtmann <marcel@holtmann.org>
      052b30b0
    • Roger Quadros's avatar
      Bluetooth: Ensure that HCI sysfs add/del is preempt safe · f3784d83
      Roger Quadros authored
      Use a different work_struct variables for add_conn() and del_conn() and
      use single work queue instead of two for adding and deleting connections.
      
      It eliminates the following error on a preemptible kernel:
      
      [  204.358032] Unable to handle kernel NULL pointer dereference at virtual address 0000000c
      [  204.370697] pgd = c0004000
      [  204.373443] [0000000c] *pgd=00000000
      [  204.378601] Internal error: Oops: 17 [#1] PREEMPT
      [  204.383361] Modules linked in: vfat fat rfcomm sco l2cap sd_mod scsi_mod iphb pvr2d drm omaplfb ps
      [  204.438537] CPU: 0    Not tainted  (2.6.28-maemo2 #1)
      [  204.443664] PC is at klist_put+0x2c/0xb4
      [  204.447601] LR is at klist_put+0x18/0xb4
      [  204.451568] pc : [<c0270f08>]    lr : [<c0270ef4>]    psr: a0000113
      [  204.451568] sp : cf1b3f10  ip : cf1b3f10  fp : cf1b3f2c
      [  204.463104] r10: 00000000  r9 : 00000000  r8 : bf08029c
      [  204.468353] r7 : c7869200  r6 : cfbe2690  r5 : c78692c8  r4 : 00000001
      [  204.474945] r3 : 00000001  r2 : cf1b2000  r1 : 00000001  r0 : 00000000
      [  204.481506] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM Segment kernel
      [  204.488861] Control: 10c5387d  Table: 887fc018  DAC: 00000017
      [  204.494628] Process btdelconn (pid: 515, stack limit = 0xcf1b22e0)
      Signed-off-by: default avatarRoger Quadros <ext-roger.quadros@nokia.com>
      Signed-off-by: default avatarMarcel Holtmann <marcel@holtmann.org>
      f3784d83
    • Eric Dumazet's avatar
      net: netif_tx_queue_stopped too expensive · 6a321cb3
      Eric Dumazet authored
      netif_tx_queue_stopped(txq) is most of the time false.
      
      Yet its cost is very expensive on SMP.
      
      static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
      {
      	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
      }
      
      I saw this on oprofile hunting and bnx2 driver bnx2_tx_int().
      
      We probably should split "struct netdev_queue" in two parts, one
      being read mostly.
      
      __netif_tx_lock() touches _xmit_lock & xmit_lock_owner, these
      deserve a separate cache line.
      Signed-off-by: default avatarEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      6a321cb3
    • Eric Dumazet's avatar
      net: Avoid extra wakeups of threads blocked in wait_for_packet() · bf368e4e
      Eric Dumazet authored
      In 2.6.25 we added UDP mem accounting.
      
      This unfortunatly added a penalty when a frame is transmitted, since
      we have at TX completion time to call sock_wfree() to perform necessary
      memory accounting. This calls sock_def_write_space() and utimately
      scheduler if any thread is waiting on the socket.
      Thread(s) waiting for an incoming frame was scheduled, then had to sleep
      again as event was meaningless.
      
      (All threads waiting on a socket are using same sk_sleep anchor)
      
      This adds lot of extra wakeups and increases latencies, as noted
      by Christoph Lameter, and slows down softirq handler.
      
      Reference : http://marc.info/?l=linux-netdev&m=124060437012283&w=2 
      
      Fortunatly, Davide Libenzi recently added concept of keyed wakeups
      into kernel, and particularly for sockets (see commit
      37e5540b 
      epoll keyed wakeups: make sockets use keyed wakeups)
      
      Davide goal was to optimize epoll, but this new wakeup infrastructure
      can help non epoll users as well, if they care to setup an appropriate
      handler.
      
      This patch introduces new DEFINE_WAIT_FUNC() helper and uses it
      in wait_for_packet(), so that only relevant event can wakeup a thread
      blocked in this function.
      
      Trace of function calls from bnx2 TX completion bnx2_poll_work() is :
      __kfree_skb()
       skb_release_head_state()
        sock_wfree()
         sock_def_write_space()
          __wake_up_sync_key()
           __wake_up_common()
            receiver_wake_function() : Stops here since thread is waiting for an INPUT
      Reported-by: default avatarChristoph Lameter <cl@linux.com>
      Signed-off-by: default avatarEric Dumazet <dada1@cosmosbay.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      bf368e4e
    • Jesse Brandeburg's avatar
      sctp: add feature bit for SCTP offload in hardware · 8dc92f7e
      Jesse Brandeburg authored
      this is the sctp code to enable hardware crc32c offload for
      adapters that support it.
      
      Originally by: Vlad Yasevich <vladislav.yasevich@hp.com>
      
      modified by Jesse Brandeburg <jesse.brandeburg@intel.com>
      Signed-off-by: default avatarJesse Brandeburg <jesse.brandeburg@intel.com>
      Signed-off-by: default avatarVlad Yasevich <vladislav.yasevich@hp.com>
      Signed-off-by: default avatarJeff Kirsher <jeffrey.t.kirsher@intel.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8dc92f7e
    • Magnus Damm's avatar
      smsc911x: add fifo byteswap support V2 · 833cc67c
      Magnus Damm authored
      This is V2 of the smsc911x fifo byteswap patch.
      
      The smsc911x hardware supports both big and little and endian
      hardware configurations, and the linux smsc911x driver currently
      detects word order.
      
      For correct operation on big endian platforms lacking swapped
      byte lanes the following patch is needed. Only fifo data is
      swapped, register data does not require any swapping.
      Signed-off-by: default avatarMagnus Damm <damm@igel.co.jp>
      Acked-by: default avatarSteve Glendinning <steve.glendinning@smsc.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      833cc67c
  10. 27 Apr, 2009 1 commit