|SUCCESS| vhost: fix checksum offset check in Rx path

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/171667

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-10-19

# DPDK Patch Review

## ERRORS

None identified.

## WARNINGS

None identified.

## INFO

**Checksum write bounds check (Info - improvement to pre-existing code):**

The patch correctly changes the bounds check from `off + 1` to `off + sizeof(uint16_t)` for checksum writing, which is semantically clearer and matches the actual requirement (2-byte write). This is a good improvement.

However, the check itself is incomplete in the pre-existing code (not introduced by this patch): when the checksum offset falls into a chained mbuf segment, the code silently skips writing the checksum instead of handling multi-segment packets. The existing code at line 2851 (after this patch):

```c
if (rte_pktmbuf_data_len(m) >= off + sizeof(uint16_t))
    *rte_pktmbuf_mtod_offset(m, uint16_t *, off) = csum;
```

only checks the first segment length (`rte_pktmbuf_data_len(m)`), not the total packet length. If the checksum falls beyond the first segment, it is silently dropped rather than being written via `rte_pktmbuf_mtod_offset()` on the appropriate segment or using a multi-segment write helper.

This is a **pre-existing limitation** not introduced by this patch. A future enhancement could use `rte_pktmbuf_mtod_offset()` with proper segment traversal or add a check against total packet length (`rte_pktmbuf_pkt_len(m)`) with multi-segment handling.

---

The patch itself is correct: it fixes the off-by-one in the size check to properly reflect that a 16-bit checksum requires 2 bytes of space. The observation above is about surrounding code behavior that exists regardless of this fix.

Review-Result: CLEAN