|WARNING| net/ixgbe: avoid private driver include in testpmd

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: WARNING
https://dpdk.org/patch/171381

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-10-14

# DPDK Patch Review

## ERRORS

**1. Missing error check propagation (use-after-free risk)**

Line 979 in the new code:
```c
	}
}
```

After the loop checking bypass events (lines 968-977), the code exits the function without checking whether `rc` from `rte_pmd_ixgbe_bypass_event_show()` indicated a failure. The original code had a final check at lines 988-991 that was removed. If `rte_pmd_ixgbe_bypass_event_show()` fails partway through the loop (rc != 0 && rc != -ENOTSUP), the function silently continues and may use uninitialized or stale `event_mode` values. This is a logic error in the error path.

**Suggested fix:**
```c
	}
	if (rc != 0 && rc != -ENOTSUP)
		fprintf(stderr,
			"\tFailed to get bypass configuration for port = %d\n",
			port_id);
}
```

---

## WARNINGS

**1. Inconsistent -ENOTSUP handling pattern**

Lines 703-706:
```c
	rc = rte_pmd_ixgbe_bypass_state_set(port_id, &bypass_mode);
	if (rc == -ENOTSUP)
		fprintf(stderr, "Bypass is not enabled on this port.\n");
	else if (rc != 0)
```

Lines 789-795:
```c
		if (rc == -ENOTSUP) {
			fprintf(stderr, "Bypass is not enabled on this port.\n");
			return;
		}
```

The first pattern continues after printing the error message (falls through to the `else if` check), while the second returns immediately. For `cmd_set_bypass_event_parsed()`, when `rte_pmd_ixgbe_bypass_wd_timeout_store()` returns -ENOTSUP, the function returns early, but when `rte_pmd_ixgbe_bypass_event_store()` returns -ENOTSUP (lines 802-803), it only prints and continues.

This inconsistency in error handling flow is confusing. All -ENOTSUP cases should either return immediately or fall through consistently.

**Suggested fix:** Standardize on returning after printing -ENOTSUP errors, as bypass unavailability is a precondition failure that makes further operations meaningless.

---

## INFO

**1. Header inclusion order (style note)**

Lines 5-9:
```c
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include <rte_string_fns.h>
```

Per AGENTS.md include order guidelines, system/libc headers should be grouped together, then DPDK headers. The `stdlib.h` include appears unused in the patch context (no `malloc`, `free`, `atoi`, etc. visible). If it's genuinely unused, consider removing it in a separate cleanup.

This is a minor observation about the include block; the overall order is correct.

---

Review-Result: ERROR