|WARNING| app/dma_perf: skip case if worker maps to main lcore

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: WARNING
https://dpdk.org/patch/165968

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-06-23

# DPDK Patch Review

## Summary
The patch adds validation to reject DMA performance test cases when any worker is configured to run on the EAL main lcore. The implementation is straightforward but has one correctness issue and one style issue.

---

## Errors

### 1. Missing bounds check on array access
**File:** `app/test-dma-perf/main.c`  
**Line:** Loop at line 126

The loop accesses `case_cfg->dma_config[i]` without verifying that the array has been allocated or that `num_worker` is within valid bounds. If `case_cfg->dma_config` is `NULL` (e.g., due to allocation failure in config parsing) or if `num_worker` exceeds the allocated size, this causes a NULL pointer dereference or out-of-bounds access.

**Fix:** Add a NULL check before the loop:

```c
if (case_cfg->dma_config == NULL) {
	printf("Case %u: invalid configuration (dma_config is NULL).\n", case_id);
	return;
}

for (uint32_t i = 0; i < case_cfg->num_worker; i++) {
	if (case_cfg->dma_config[i].lcore_dma_map.lcore == main_lcore) {
		printf("Case %u: worker %u cannot run on the EAL main lcore (%u).\n",
		       case_id, i, main_lcore);
		return;
	}
}
```

Alternatively, if the code structure guarantees `dma_config` is always valid when `run_test()` is called, add an assertion:

```c
assert(case_cfg->dma_config != NULL || case_cfg->num_worker == 0);
```

---

## Warnings

### 1. Implicit boolean comparison
**File:** `app/test-dma-perf/main.c`  
**Line:** 127

The comparison `case_cfg->dma_config[i].lcore_dma_map.lcore == main_lcore` is used directly in the `if` statement. While this is functionally correct, DPDK coding style prefers explicit comparisons for clarity when the result is not a `bool` type.

However, in this specific case, the comparison operator `==` already returns an implicit boolean result (the comparison itself is the condition), so this is actually acceptable under DPDK style. The guidelines primarily target avoiding implicit truthiness on pointers and integers (e.g., `if (ptr)` should be `if (ptr != NULL)`), not comparisons that already produce boolean results.

**Status:** Not an issue upon closer inspection. Ignore this item.

---

## Info

### 1. Consider adding debug context
The error message could include the configured lcore ID to help users understand which worker binding is problematic:

```c
printf("Case %u: worker %u (lcore %u) cannot run on the EAL main lcore (%u).\n",
       case_id, i, case_cfg->dma_config[i].lcore_dma_map.lcore, main_lcore);
```

This is minor but improves debuggability.

---

## Final Verdict

**Must Fix:**
- Add NULL check or assertion before accessing `case_cfg->dma_config[i]` to prevent potential NULL pointer dereference

**Optional:**
- Enhance error message with configured lcore ID for better diagnostics