1. 26 Apr, 2016 2 commits
    • Alexander Duyck's avatar
      i40e/i40evf: Add support for IPIP and SIT offloads · 577389a5
      Alexander Duyck authored
      Looking over the documentation it turns out enabling IPIP and SIT offloads
      for i40e is pretty straightforward.  As such I decided to enable them with
      this patch.  In my testing I am seeing an improvement of 8 to 10 Gb/s
      for IPIP and SIT tunnels with this offload enabled.
      Signed-off-by: default avatarAlexander Duyck <aduyck@mirantis.com>
      Tested-by: default avatarAndrew Bowers <andrewx.bowers@intel.com>
      Signed-off-by: default avatarJeff Kirsher <jeffrey.t.kirsher@intel.com>
      577389a5
    • Alexander Duyck's avatar
      i40e/i40evf: Clean up feature flags · b0fe3306
      Alexander Duyck authored
      The feature flags list for i40e and i40evf is beginning to become pretty
      massive.  I plan to add another 4 or so features to these drivers and
      duplicating the flags for each and every flags list is becoming a bit
      repetitive.
      
      The primary change here is that we now build our features list around
      hw_encap_features.  After that we assign that to vlan_features,
      hw_features, and finally map that onto features.  In addition we end up
      throwing features onto hw_encap_features that end up having no effect such
      as the Rx offloads and SCTP_CRC.  However that should have no impact and
      makes things a bit easier for us as hw_encap_features is one of the less
      updated features maps available.
      
      For i40evf I went through and sanity checked a few features as well.
      Specifically RXCSUM was being set as a read-only feature which didn't make
      much sense.  I have updated things so we can clear the NETIF_F_RXCSUM flag
      since that is really a software feature and not a hardware one anyway so
      disabling it is just a matter of ignoring the result from the hardware.
      Signed-off-by: default avatarAlexander Duyck <aduyck@mirantis.com>
      Tested-by: default avatarAndrew Bowers <andrewx.bowers@intel.com>
      Signed-off-by: default avatarJeff Kirsher <jeffrey.t.kirsher@intel.com>
      b0fe3306
  2. 25 Apr, 2016 36 commits
  3. 24 Apr, 2016 2 commits
    • Eric Dumazet's avatar
      tcp-tso: do not split TSO packets at retransmit time · 10d3be56
      Eric Dumazet authored
      Linux TCP stack painfully segments all TSO/GSO packets before retransmits.
      
      This was fine back in the days when TSO/GSO were emerging, with their
      bugs, but we believe the dark age is over.
      
      Keeping big packets in write queues, but also in stack traversal
      has a lot of benefits.
       - Less memory overhead, because write queues have less skbs
       - Less cpu overhead at ACK processing.
       - Better SACK processing, as lot of studies mentioned how
         awful linux was at this ;)
       - Less cpu overhead to send the rtx packets
         (IP stack traversal, netfilter traversal, drivers...)
       - Better latencies in presence of losses.
       - Smaller spikes in fq like packet schedulers, as retransmits
         are not constrained by TCP Small Queues.
      
      1 % packet losses are common today, and at 100Gbit speeds, this
      translates to ~80,000 losses per second.
      Losses are often correlated, and we see many retransmit events
      leading to 1-MSS train of packets, at the time hosts are already
      under stress.
      Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
      Acked-by: default avatarYuchung Cheng <ycheng@google.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      10d3be56
    • Parthasarathy Bhuvaragan's avatar
      tipc: fix stale links after re-enabling bearer · 8cee83dd
      Parthasarathy Bhuvaragan authored
      Commit 42b18f60 ("tipc: refactor function tipc_link_timeout()"),
      introduced a bug which prevents sending of probe messages during
      link synchronization phase. This leads to hanging links, if the
      bearer is disabled/enabled after links are up.
      
      In this commit, we send the probe messages correctly.
      
      Fixes: 42b18f60 ("tipc: refactor function tipc_link_timeout()")
      Acked-by: default avatarJon Maloy <jon.maloy@ericsson.com>
      Signed-off-by: default avatarParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
      Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8cee83dd