|WARNING| [v4] examples/l3fwd: fix RFC 1812 checks

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: WARNING
https://dpdk.org/patch/172033

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-10-24

Thank you for submitting this patch. I have reviewed it against the DPDK contribution guidelines.

## ERRORS

**1. Endian conversion misplaced in comparison (lines 215 and 234)**

```c
if (link_len < rte_cpu_to_be_16(pkt->total_length))
```

The comparison mixes a host-endian `link_len` with a big-endian conversion of `total_length`.
`rte_cpu_to_be_16()` converts from CPU byte order to big endian, but the `pkt->total_length` field
is already stored in network byte order (big endian) in the packet. The comparison should convert
from network order to CPU order for arithmetic:

```c
if (link_len < rte_be_to_cpu_16(pkt->total_length))
```

This affects both the new check at line 215 and the existing check at line 207.

**2. Existing endian bug also present at line 207**

The same issue exists in the unmodified line 207:
```c
if (rte_cpu_to_be_16(pkt->total_length) < rte_ipv4_hdr_len(pkt))
```

This should be:
```c
if (rte_be_to_cpu_16(pkt->total_length) < rte_ipv4_hdr_len(pkt))
```

While this is pre-existing code, the patch changes this line by replacing `sizeof(struct rte_ipv4_hdr)`
with `rte_ipv4_hdr_len(pkt)`, so the endian bug should be fixed at the same time.

## WARNINGS

**3. Switch statement missing default case (line 173)**

The switch on `RTE_MBUF_F_RX_IP_CKSUM_MASK` processes all four known values but lacks a `default:` case.
While all four enum values are covered, adding a default case improves robustness against future enum additions
and makes defensive coding explicit:

```c
switch (ol_flags & RTE_MBUF_F_RX_IP_CKSUM_MASK) {
case RTE_MBUF_F_RX_IP_CKSUM_UNKNOWN:
	/* Not checked in H/W; check it. */
	if (rte_ipv4_cksum(pkt) != 0)
		return -2;
	break;
case RTE_MBUF_F_RX_IP_CKSUM_GOOD:
	break;
case RTE_MBUF_F_RX_IP_CKSUM_BAD:
	return -2;
case RTE_MBUF_F_RX_IP_CKSUM_NONE:
	/* Packet integrity known good, checksum field unset; don't check it. */
	break;
default:
	/* Unknown checksum status; conservatively reject packet */
	return -2;
}
```

## INFO

**4. Comment formatting style (line 184)**

The comment `/* Packet integrity known good, checksum field unset; don't check it. */`
exceeds the preferred single-line comment length. For better readability, consider multi-line format:

```c
/*
 * Packet integrity known good, checksum field unset;
 * don't check it.
 */
```

This is a minor style preference and not a guideline violation.

**5. Simplified checksum validation is an improvement**

The patch correctly removes the unnecessary zeroing and restoration of the checksum field.
`rte_ipv4_cksum()` returns zero when the checksum is correct, making the direct comparison cleaner
and eliminating the modification of packet data during validation.

Review-Result: ERROR