|SUCCESS| [v6] mempool/dpaa2: update for new mempool cache algorithm

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/164492

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-05-27

Reviewing patch: mempool/dpaa2: update for new mempool cache algorithm

git sha: (from series patch 164492)
Author: Morten Brorup <mb at smartsharesystems.com>

Subject: [PATCH v6] mempool/dpaa2: update for new mempool cache algorithm

This patch removes code that previously modified the mempool cache's
flushthresh field, which has become obsolete due to an improved mempool
cache algorithm.

> diff --git a/drivers/mempool/dpaa2/dpaa2_hw_mempool.c b/drivers/mempool/dpaa2/dpaa2_hw_mempool.c
> index 02b6741853..ee001d8ce0 100644
> --- a/drivers/mempool/dpaa2/dpaa2_hw_mempool.c
> +++ b/drivers/mempool/dpaa2/dpaa2_hw_mempool.c
> @@ -54,8 +54,6 @@ rte_hw_mbuf_create_pool(struct rte_mempool *mp)
>  	struct dpaa2_bp_info *bp_info;
>  	struct dpbp_attr dpbp_attr;
>  	uint32_t bpid;
> -	unsigned int lcore_id;
> -	struct rte_mempool_cache *cache;
>  	int ret;
>  
>  	avail_dpbp = dpaa2_alloc_dpbp_dev();

[ ... ]

> @@ -152,18 +150,6 @@ rte_hw_mbuf_create_pool(struct rte_mempool *mp)
>  	DPAA2_MEMPOOL_DEBUG("BP List created for bpid =%d", dpbp_attr.bpid);
>  
>  	h_bp_list = bp_list;
> -	/* Update per core mempool cache threshold to optimal value which is
> -	 * number of buffers that can be released to HW buffer pool in
> -	 * a single API call.
> -	 */
> -	for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++) {
> -		cache = &mp->local_cache[lcore_id];
> -		DPAA2_MEMPOOL_DEBUG("lCore %d: cache->flushthresh %d -> %d",
> -			lcore_id, cache->flushthresh,
> -			(uint32_t)(cache->size + DPAA2_MBUF_MAX_ACQ_REL));
> -		if (cache->flushthresh)
> -			cache->flushthresh = cache->size + DPAA2_MBUF_MAX_ACQ_REL;
> -	}
>  
>  	return 0;
>  err4:

The patch is straightforward and correctly removes obsolete code. However,
there's a dependency concern that should be addressed:

Does the dependent patch (patch-164427) properly handle the performance
optimization that was being done here? The removed code was adjusting
flushthresh to cache->size + DPAA2_MBUF_MAX_ACQ_REL to match the hardware
buffer pool capabilities. Is the new mempool cache algorithm aware of
DPAA2_MBUF_MAX_ACQ_REL limitations, or does it need driver-specific
hints about optimal batch sizes?

The commit message states the field is "obsolete" and modifying it "has no
effect", but can you confirm that the new algorithm achieves equivalent or
better batching behavior for this hardware without the driver needing to
provide hints?