|WARNING| mem: fix fbarray placement colliding with PCI BARs

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: WARNING
https://dpdk.org/patch/171875

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-10-23

# DPDK Patch Review

## ERRORS

### 1. Variable Name in Loop Cleanup Contradicts Preceding Loop

**Line 190:**
```c
for (i = 0; i < n_msl; i++) {
```

The cleanup loop uses `i` which was already declared and used earlier in the function (line 32: `int i, hpi_idx, msl_idx, n_msl = 0, ret = -1;`). The earlier loops iterate over `hpi_idx` and `cur_type`, then `msl_idx` is used as an index into `mcfg->memsegs[]`. The cleanup loop must iterate over exactly the memseg lists that were created in the "Create the segment lists" block (lines 141-158), which uses `msl_idx` as the loop counter and index.

Using `i` in the cleanup loop is correct (it was declared at function scope), but the bound `n_msl` tracks `msl_idx` from the creation loop, so the two are consistent. However, using a different variable name than the creation loop (`msl_idx` vs `i`) obscures the correspondence and risks a future maintainer changing one without the other.

**Why it matters:** While not currently a bug (both `i` and `msl_idx` are declared and `n_msl` correctly tracks the count), the inconsistency in naming makes the cleanup logic harder to verify and introduces maintenance risk.

**Suggested fix:** Use `msl_idx` in the cleanup loop to match the creation loop:
```c
for (msl_idx = 0; msl_idx < n_msl; msl_idx++) {
    msl = &mcfg->memsegs[msl_idx];
    rte_fbarray_destroy(&msl->memseg_arr);
    memset(msl, 0, sizeof(*msl));
}
```

---

## WARNINGS

### 1. Missing Release Notes for Behavioral Change

This patch fixes a crash in secondary processes but does not include a corresponding entry in the release notes. Fixes that resolve user-visible issues (crashes, correctness bugs affecting deployed systems) should be documented in the release notes so users understand what has been repaired.

**Suggested fix:** Add an entry to `doc/guides/rel_notes/release_XX_YY.rst` under the "Fixed Issues" section describing the intermittent secondary process crash that this patch resolves.

---

## INFO

None.

---

## SUMMARY

The patch correctly addresses the VA space collision issue by reordering memseg list creation before VA reservation. The cleanup path on error now properly destroys the created fbarrays, preventing resource leaks. The use of `i` in the cleanup loop is technically correct but inconsistent with the creation loop's use of `msl_idx`, which is a minor maintenance concern worth addressing for clarity.

Review-Result: WARNING