[adilalperenciftci]

Fixes #4906. Corrects UDP checksum calculation when encapsulated frames include inner padding up to declared UDP length.

[Copilot]

Copilot review overview

🟢 Approval recommended

The fix is localized to UDP checksum construction and is directly covered by a regression test reproducing the reported mismatch.

Review effort: Lite
Findings: None

What changed in this PR

This PR fixes UDP checksum computation for IPv6/IPv4 underlayers when the UDP payload contains encapsulated-frame padding bytes that are present on the wire (and within the declared UDP length) but are appended during build_padding() after UDP.post_build() runs.

Changes:

• Adjust UDP checksum calculation to include encapsulated payload padding bytes up to the declared UDP length when rebuilding packets.
• Add a UTScapy regression test reproducing issue #4906 (VXLAN/IPv6 example) and asserting the recalculated checksum matches the received value.

File	Description
scapy/​layers/​inet.py	Includes payload padding (added via build_padding()) in the bytes used for UDP checksum computation when the UDP length exceeds the built header+payload length.
test/​scapy/​layers/​inet6.uts	Adds a regression test validating checksum recalculation matches the captured checksum for an encapsulated/padded UDP-over-IPv6 packet.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.07%. Comparing base (5166573) to head (07da874).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #5191      +/-   ##
==========================================
- Coverage   80.37%   80.07%   -0.31%     
==========================================
  Files         375      375              
  Lines       97683    97686       +3     
==========================================
- Hits        78516    78220     -296     
- Misses      19167    19466     +299     

Files with missing lines	Coverage Δ
scapy/layers/inet.py	72.10% <100.00%> (+0.05%)	⬆️

... and 21 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Copilot review overview

🟢 Approval recommended

The change is narrowly scoped to checksum input bytes, aligns with existing padding semantics, and includes a regression test covering the reported failure mode.

Review effort: Lite
Findings: None

[gpotter2]

@p-l- Hi, would you have the time to review this since you did #4906 ? Thanks :P

[adilalperenciftci]

The two red jobs look unrelated to this change: the Windows failure is the J1939 BAM frame-delay test, while this PR only touches inet.py and inet6.uts. Could someone rerun the failed jobs?

[adilalperenciftci]

I reviewed the two remaining red jobs against this PR's diff. The Python 3.7 job fails because gzip.compress(..., mtime=0) is unsupported there in test/regression.uts; the Windows Python 3.14 job fails the J1939 SlowTestSocket frame-delay assertion (expected 4 frames, got 1). These are known matrix issues outside the UDP checksum change in scapy/layers/inet.py and test/scapy/layers/inet6.uts. The checksum-focused coverage and all other checks pass, and @polybassa has approved the patch. Could a maintainer treat these as unrelated or rerun the two jobs?

[adilalperenciftci]

I rebased the UDP checksum change onto the current master, which removes Python 3.7 from the matrix and includes the existing Windows timing fixes. Workflow 40506575396 is waiting for maintainer approval: https://github.com/secdev/scapy/actions/runs/40506575396

Could a maintainer approve this workflow run?

[adilalperenciftci]

Thanks for the review and approval @polybassa! Whenever you have a moment, could you please approve the workflow run for the updated commit?