[Intel-wired-lan] [next PATCH S52-V2 12/15] i40e/i40evf: napi_poll must return the work done

Bowers, AndrewX andrewx.bowers at intel.com
Thu Nov 17 18:27:23 UTC 2016


> -----Original Message-----
> From: Intel-wired-lan [mailto:intel-wired-lan-bounces at lists.osuosl.org] On
> Behalf Of Bimmy Pujari
> Sent: Tuesday, November 08, 2016 1:05 PM
> To: intel-wired-lan at lists.osuosl.org
> Subject: [Intel-wired-lan] [next PATCH S52-V2 12/15] i40e/i40evf: napi_poll
> must return the work done
> 
> From: Alexander Duyck <alexander.h.duyck at intel.com>
> 
> Currently the function i40e_napi-poll() returns 0 when it clean completely
> the rx rings, but this foul budget accounting in core code.
> 
> Fix this by returning the actual work done, capped to budget - 1, since the
> core doesn't allow to return the full budget when the driver modifies the
> napi status
> 
> This is based on a similar change that was made for the ixgbe driver by Paolo
> Abeni.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck at intel.com>
> Change-ID: Ic3d93ad2fa2fc8ce3164bc461e69367da0f9173b
> ---
> Testing Hints:
>         This may slightly improve the responsiveness of the system, but the
>         big gain for us is actually when using i40e to perform busy polling
>         since the return value is used to verify if we have received any
>         packets.
> 
>         For our purposes the best test is probably just an Rx performance
>         test to make sure we didn't cause a regression.
> 
>  drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 2 +-
>  drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Tested-by: Andrew Bowers <andrewx.bowers at intel.com>




More information about the Intel-wired-lan mailing list