Re: [PATCH net-next 5/9] net/sched: support per action hw stats

[Simon Horman <simon.horman@corigine.com>]

On Wed, Feb 01, 2023 at 06:10:34PM +0200, Oz Shlomo wrote:
> There are currently two mechanisms for populating hardware stats:
> 1. Using flow_offload api to query the flow's statistics.
>    The api assumes that the same stats values apply to all
>    the flow's actions.
>    This assumption breaks when action drops or jumps over following
>    actions.
> 2. Using hw_action api to query specific action stats via a driver
>    callback method. This api assures the correct action stats for
>    the offloaded action, however, it does not apply to the rest of the
>    actions in the flow's actions array.
> 
> Extend the flow_offload stats callback to indicate that a per action
> stats update is required.
> Use the existing flow_offload_action api to query the action's hw stats.
> In addition, currently the tc action stats utility only updates hw actions.
> Reuse the existing action stats cb infrastructure to query any action
> stats.
> 
> Signed-off-by: Oz Shlomo <ozsh@nvidia.com>
> ---
>  include/net/flow_offload.h |  1 +
>  include/net/pkt_cls.h      | 29 +++++++++++++++++++----------
>  net/sched/act_api.c        |  8 --------
>  net/sched/cls_flower.c     |  2 +-
>  net/sched/cls_matchall.c   |  2 +-
>  5 files changed, 22 insertions(+), 20 deletions(-)
> 
> diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h
> index d177bf5f0e1a..27decadd4f5f 100644
> --- a/include/net/flow_offload.h
> +++ b/include/net/flow_offload.h
> @@ -597,6 +597,7 @@ struct flow_cls_offload {
>  	unsigned long cookie;
>  	struct flow_rule *rule;
>  	struct flow_stats stats;
> +	bool use_act_stats;
>  	u32 classid;
>  };

Hi Oz,

It's probably not important, but I thought it is worth bringing
to your attention.

The placement of use_act_stats above  puts it on a different
cacheline (on x86_64) to stats. Which does not seem to be idea
as those fields are accessed together.

There is a 4 byte hole immediately above above cookie,
on the same cacheline as stats, which can accommodate use_act_stats.

>  
> diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
> index be21764a3b34..d4315757d1a2 100644
> --- a/include/net/pkt_cls.h
> +++ b/include/net/pkt_cls.h

...

> @@ -769,6 +777,7 @@ struct tc_cls_matchall_offload {
>  	enum tc_matchall_command command;
>  	struct flow_rule *rule;
>  	struct flow_stats stats;
> +	bool use_act_stats;
>  	unsigned long cookie;
>  };

I believe the same logic applies to this change too.

...

> diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
> index cb04739a13ce..885c95191ccf 100644
> --- a/net/sched/cls_flower.c
> +++ b/net/sched/cls_flower.c
> @@ -502,7 +502,7 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f,
>  	tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false,
>  			 rtnl_held);
>  
> -	tcf_exts_hw_stats_update(&f->exts, &cls_flower.stats);
> +	tcf_exts_hw_stats_update(&f->exts, &cls_flower.stats, cls_flower.use_act_stats);
>  }
>  
>  static void __fl_put(struct cls_fl_filter *f)
> diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c
> index b3883d3d4dbd..fa3bbd187eb9 100644
> --- a/net/sched/cls_matchall.c
> +++ b/net/sched/cls_matchall.c
> @@ -331,7 +331,7 @@ static void mall_stats_hw_filter(struct tcf_proto *tp,
>  
>  	tc_setup_cb_call(block, TC_SETUP_CLSMATCHALL, &cls_mall, false, true);
>  
> -	tcf_exts_hw_stats_update(&head->exts, &cls_mall.stats);
> +	tcf_exts_hw_stats_update(&head->exts, &cls_mall.stats, cls_mall.use_act_stats);
>  }
>  
>  static int mall_dump(struct net *net, struct tcf_proto *tp, void *fh,