[PATCH v5 1/1] graph: improve node layout

[Thomas Monjalon]

Is it good to go?


15/11/2024 02:55, Huichao Cai:
> The members "dispatch" and "xstat_off" of the structure "rte_node"
> can be min cache aligned to make room for future expansion and to
> make sure have better performance. Add corresponding comments.
> 
> Signed-off-by: Huichao Cai <chcchc88 at 163.com>
> ---
>  doc/guides/rel_notes/release_24_11.rst |  2 ++
>  lib/graph/rte_graph_worker_common.h    | 10 +++++++---
>  2 files changed, 9 insertions(+), 3 deletions(-)
> 
> diff --git a/doc/guides/rel_notes/release_24_11.rst b/doc/guides/rel_notes/release_24_11.rst
> index 5063badf39..32800e8cb0 100644
> --- a/doc/guides/rel_notes/release_24_11.rst
> +++ b/doc/guides/rel_notes/release_24_11.rst
> @@ -491,6 +491,8 @@ ABI Changes
>    added new structure ``rte_node_xstats`` to ``rte_node_register`` and
>    added ``xstat_off`` to ``rte_node``.
>  
> +* graph: The members ``dispatch`` and ``xstat_off`` of the structure ``rte_node`` have been
> +  marked as RTE_CACHE_LINE_MIN_SIZE bytes aligned.
>  
>  Known Issues
>  ------------
> diff --git a/lib/graph/rte_graph_worker_common.h b/lib/graph/rte_graph_worker_common.h
> index a518af2b2a..d3ec88519d 100644
> --- a/lib/graph/rte_graph_worker_common.h
> +++ b/lib/graph/rte_graph_worker_common.h
> @@ -104,16 +104,20 @@ struct __rte_cache_aligned rte_node {
>  	/** Original process function when pcap is enabled. */
>  	rte_node_process_t original_process;
>  
> +	/** Fast schedule area for mcore dispatch model. */
>  	union {
> -		/* Fast schedule area for mcore dispatch model */
> -		struct {
> +		alignas(RTE_CACHE_LINE_MIN_SIZE) struct {
>  			unsigned int lcore_id;  /**< Node running lcore. */
>  			uint64_t total_sched_objs; /**< Number of objects scheduled. */
>  			uint64_t total_sched_fail; /**< Number of scheduled failure. */
>  		} dispatch;
>  	};
> +
> +	/** Fast path area cache line 1. */
> +	alignas(RTE_CACHE_LINE_MIN_SIZE)
>  	rte_graph_off_t xstat_off; /**< Offset to xstat counters. */
> -	/* Fast path area  */
> +
> +	/** Fast path area cache line 2. */
>  	__extension__ struct __rte_cache_aligned {
>  #define RTE_NODE_CTX_SZ 16
>  		union {
>