|SUCCESS| [v1] rib: reduce node mempool memory footprint

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/172318

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-10-30

# DPDK Patch Review

## ERRORS

None identified.

## WARNINGS

None identified.

## INFO

### 1. Performance optimization for node allocation

The patch reduces mempool memory footprint by removing unnecessary alignment and spreading for CPU-only RIB node structures. The approach is sound:

- Nodes are accessed only by CPU, not DMA
- The structures are small (40 bytes for IPv4, 56 bytes for IPv6)
- Cache-line alignment and memory-channel spreading waste space (192 bytes per object)
- The flags `RTE_MEMPOOL_F_NO_CACHE_ALIGN`, `RTE_MEMPOOL_F_NO_SPREAD`, and `RTE_MEMPOOL_F_NO_IOVA_CONTIG` are appropriate for this use case

**File: lib/rib/rte_rib.c, lines 433-437**
```c
	node_pool = rte_mempool_create(mem_name, conf->max_nodes,
		sizeof(struct rte_rib_node) + conf->ext_sz, 0, 0,
		NULL, NULL, NULL, NULL, socket_id,
		RTE_MEMPOOL_F_NO_SPREAD | RTE_MEMPOOL_F_NO_CACHE_ALIGN |
		RTE_MEMPOOL_F_NO_IOVA_CONTIG);
```

**File: lib/rib/rte_rib6.c, lines 496-500**
```c
	node_pool = rte_mempool_create(mem_name, conf->max_nodes,
		sizeof(struct rte_rib6_node) + conf->ext_sz, 0, 0,
		NULL, NULL, NULL, NULL, socket_id,
		RTE_MEMPOOL_F_NO_SPREAD | RTE_MEMPOOL_F_NO_CACHE_ALIGN |
		RTE_MEMPOOL_F_NO_IOVA_CONTIG);
```

The performance numbers (13% and 29% improvement in add/delete) and memory savings (390 MiB to 128 MiB for IPv4) support the change.

### 2. Comment accuracy

The comments accurately describe the intent and note that `RTE_MEMPOOL_F_NO_SPREAD` is implied by `RTE_MEMPOOL_F_NO_CACHE_ALIGN` but is passed explicitly for documentation. This is good practice for code clarity.

### 3. Missing release notes

This patch introduces a significant performance improvement and memory usage reduction. While it does not change public API, the behavioral change (memory footprint reduction, performance improvement) is significant enough that users would benefit from knowing about it in release notes. However, since this is an internal implementation detail and not an API change, release notes are not strictly required per the guidelines.

Review-Result: CLEAN