Skip to content

K
kernel_linux

OpenHarmony / kernel_linux
上一次同步 3 年多

通知 13
Star 8
Fork 2
K
kernel_linux
切换分支/标签
  1. 07 4月, 2011 28 次提交
  3. 05 4月, 2011 5 次提交
    • B
      sfc: Implement ethtool_ops::set_phys_id instead of ethtool_ops::phys_id · c5e129ac
      由 Ben Hutchings 提交于 
      Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
      c5e129ac
    • B
      sfc: Implement generic features interface · abfe9039
      由 Ben Hutchings 提交于 
      Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
      abfe9039
    • B
      sfc: Enable all TSO features on VLANs · f82d9a67
      由 Ben Hutchings 提交于 
      The TSO code already supports IPv6 on VLAN, so enable it.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
      f82d9a67
    • T
      net: Allow no-cache copy from user on transmit · c6e1a0d1
      由 Tom Herbert 提交于 
      This patch uses __copy_from_user_nocache on transmit to bypass data
cache for a performance improvement.  skb_add_data_nocache and
skb_copy_to_page_nocache can be called by sendmsg functions to use
this feature, initial support is in tcp_sendmsg.  This functionality is
configurable per device using ethtool.

Presumably, this feature would only be useful when the driver does
not touch the data.  The feature is turned on by default if a device
indicates that it does some form of checksum offload; it is off by
default for devices that do no checksum offload or indicate no checksum
is necessary.  For the former case copy-checksum is probably done
anyway, in the latter case the device is likely loopback in which case
the no cache copy is probably not beneficial.

This patch was tested using 200 instances of netperf TCP_RR with
1400 byte request and one byte reply.  Platform is 16 core AMD x86.

No-cache copy disabled:
   672703 tps, 97.13% utilization
   50/90/99% latency:244.31 484.205 1028.41

No-cache copy enabled:
   702113 tps, 96.16% utilization,
   50/90/99% latency 238.56 467.56 956.955

Using 14000 byte request and response sizes demonstrate the
effects more dramatically:

No-cache copy disabled:
   79571 tps, 34.34 %utlization
   50/90/95% latency 1584.46 2319.59 5001.76

No-cache copy enabled:
   83856 tps, 34.81% utilization
   50/90/95% latency 2508.42 2622.62 2735.88

Note especially the effect on latency tail (95th percentile).

This seems to provide a nice performance improvement and is
consistent in the tests I ran.  Presumably, this would provide
the greatest benfits in the presence of an application workload
stressing the cache and a lot of transmit data happening.
Signed-off-by: NTom Herbert <therbert@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      c6e1a0d1
    • I
      xen: netfront: assume all hw features are available until backend connection setup · fc3e5941
      由 Ian Campbell 提交于 
      We need to assume that all features will be available when registering the
netdev otherwise they are ommitted from the initial set of
dev->wanted_features. When we connect to the backed we reduce the set as
necessary due to the call to netdev_update_features() in xennet_connect().
Signed-off-by: NIan Campbell <ian.campbell@citrix.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      fc3e5941
  5. 04 4月, 2011 2 次提交
  7. 02 4月, 2011 5 次提交