TCP Segmentation Offloading - TCP-IP

This is a discussion on TCP Segmentation Offloading - TCP-IP ; Hi all, I'm working on TCP segmentation Offload and I'm using Linux -2.6.14 kernel for this. I'm getting about a 6x increase in sending by turning off TCP segmentation offload. But When I turn on the TSO my performance is ...

+ Reply to Thread
Results 1 to 5 of 5

Thread: TCP Segmentation Offloading

  1. TCP Segmentation Offloading

    Hi all,

    I'm working on TCP segmentation Offload and I'm using Linux -2.6.14
    kernel for this. I'm getting about a 6x increase in sending by turning
    off TCP segmentation offload. But When I turn on the TSO my
    performance is getting degraded. But when I switched back to Linux
    2.6.9 kernel I'm getting 16Mbps more when TSO is enabled.

    Thanks In Advance
    Anand


  2. Re: TCP Segmentation Offloading

    AnandV wrote:
    > I'm working on TCP segmentation Offload and I'm using Linux
    > -2.6.14 kernel for this. I'm getting about a 6x increase in sending
    > by turning off TCP segmentation offload. But When I turn on the TSO
    > my performance is getting degraded. But when I switched back to
    > Linux 2.6.9 kernel I'm getting 16Mbps more when TSO is enabled.


    ISTR that the TSO implementation was in non-trivial flux until maybe
    2.6.16 or so? Do you _have_ to be on 2.6.14? 2.6.18 might not be a
    bad place to be for TSO.

    rick jones
    --
    oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
    these opinions are mine, all mine; HP might not want them anyway...
    feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

  3. Re: TCP Segmentation Offloading

    On Mar 2, 11:56 pm, Rick Jones wrote:
    > AnandV wrote:
    > > I'm working on TCP segmentation Offload and I'm using Linux
    > > -2.6.14 kernel for this. I'm getting about a 6x increase in sending
    > > by turning off TCP segmentation offload. But When I turn on the TSO
    > > my performance is getting degraded. But when I switched back to
    > > Linux 2.6.9 kernel I'm getting 16Mbps more when TSO is enabled.

    >
    > ISTR that the TSO implementation was in non-trivial flux until maybe
    > 2.6.16 or so? Do you _have_ to be on 2.6.14? 2.6.18 might not be a
    > bad place to be for TSO.
    >
    > rick jones
    > --
    > oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
    > these opinions are mine, all mine; HP might not want them anyway...
    > feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...



    Thanks Rick,

    Is there any patch for 2.6.14 or is it necessary that I have to move
    to 2.6.18.

    Best Regards
    Anand


  4. Re: TCP Segmentation Offloading

    On Mar 2, 11:56 pm, Rick Jones wrote:
    > AnandV wrote:
    > > I'm working on TCP segmentation Offload and I'm using Linux
    > > -2.6.14 kernel for this. I'm getting about a 6x increase in sending
    > > by turning off TCP segmentation offload. But When I turn on the TSO
    > > my performance is getting degraded. But when I switched back to
    > > Linux 2.6.9 kernel I'm getting 16Mbps more when TSO is enabled.

    >
    > ISTR that the TSO implementation was in non-trivial flux until maybe
    > 2.6.16 or so? Do you _have_ to be on 2.6.14? 2.6.18 might not be a
    > bad place to be for TSO.
    >
    > rick jones
    > --
    > oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
    > these opinions are mine, all mine; HP might not want them anyway...
    > feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...



    Thanks Rick,

    Is there any patch for 2.6.14 or is it necessary that I have to move
    to 2.6.18.

    Best Regards
    Anand


  5. Re: TCP Segmentation Offloading

    AnandV wrote:
    > Is there any patch for 2.6.14 or is it necessary that I have to move
    > to 2.6.18.


    Honestly, I've no idea. All my work is crash and burn which means I
    tend to just follow the bouncing kernel flavor of the week rather than
    back-port patches.

    rick jones
    --
    portable adj, code that compiles under more than one compiler
    these opinions are mine, all mine; HP might not want them anyway...
    feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

+ Reply to Thread