Re: [dpdk-dev] [PATCH 5/8] net/ice/base: improvements to Flow Director masking

["Yang, Qiming" <qiming.yang@intel.com>]

Hi, Qi

> -----Original Message-----
> From: Zhang, Qi Z
> Sent: Friday, September 27, 2019 12:17 PM
> To: Lu, Wenzhuo <wenzhuo.lu@intel.com>; Yang, Qiming
> <qiming.yang@intel.com>
> Cc: dev@dpdk.org; Ye, Xiaolong <xiaolong.ye@intel.com>; Zhang, Qi Z
> <qi.z.zhang@intel.com>; Nowlin, Dan <dan.nowlin@intel.com>; Stillwell Jr,
> Paul M <paul.m.stillwell.jr@intel.com>
> Subject: [PATCH 5/8] net/ice/base: improvements to Flow Director masking
> 
> Currently, 3-tuple FD matching is implemented using masking. However, this
> is using up twenty-four of the thirty-two FD masks available. This patch uses
> the swap register more efficiently to implement the 3-tuple matches, which
> saves all FD masks for other uses.
> 
> Added IPV6 versions of DSCP, TTL and Protocol fields for FD use.
> 
> Signed-off-by: Dan Nowlin <dan.nowlin@intel.com>
> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
> ---
>  drivers/net/ice/base/ice_flex_pipe.c |  71 +++++++++------------
>  drivers/net/ice/base/ice_flex_type.h |   4 +-
>  drivers/net/ice/base/ice_flow.c      | 118 ++++++++++++++++++++---------------
>  drivers/net/ice/base/ice_flow.h      |  10 ++-
>  4 files changed, 108 insertions(+), 95 deletions(-)
> 
> diff --git a/drivers/net/ice/base/ice_flex_pipe.c
> b/drivers/net/ice/base/ice_flex_pipe.c
> index 75bb87079..8f8cab86e 100644
> --- a/drivers/net/ice/base/ice_flex_pipe.c
> +++ b/drivers/net/ice/base/ice_flex_pipe.c
> @@ -1248,25 +1248,6 @@ void ice_free_seg(struct ice_hw *hw)  }
> 
[snip]

> @@ -4342,29 +4321,39 @@ ice_update_fd_swap(struct ice_hw *hw, u16
> prof_id, struct ice_fv_word *es)
>  		si -= indexes_used;
>  	}
> 
> -	/* for each set of 4 swap indexes, write the appropriate register */
> +	/* for each set of 4 swap and 4 inset indexes, write the appropriate
> +	 * register
> +	 */
>  	for (j = 0; j < hw->blk[ICE_BLK_FD].es.fvw / 4; j++) {
> -		u32 raw_entry = 0;
> +		u32 raw_swap = 0;
> +		u32 raw_in = 0;
> 
>  		for (k = 0; k < 4; k++) {
>  			u8 idx;
> 
>  			idx = (j * 4) + k;
> -			if (used[idx])
> -				raw_entry |= used[idx] << (k *
> BITS_PER_BYTE);
> +			if (used[idx] && !(mask_sel & BIT(idx))) {
> +				raw_swap |= used[idx] << (k *
> BITS_PER_BYTE); #define
> +ICE_INSET_DFLT 0x9f

Is this macro defined within this function?

> +				raw_in |= ICE_INSET_DFLT << (k *
> BITS_PER_BYTE);
> +			}
>  		}
> 
[snip]


Qiming