netkvm_buffer_shortage: allow non-increasing packet lose

[heywji]

This patch fixes two places.

1. Hotfix the extra params for nic1
2. Change the strict decreasing requirement to allow a plateau at optimal performance. Fixes false failure when packet loss remains 0%.

ID: 4186

Signed-off-by: Wenkang Ji [email protected]

Summary by CodeRabbit

• New Features

  • Network validation now accepts stable or decreasing packet loss across test runs (plateaus, including 0%, are considered acceptable).

• Improvements

  • Test messaging updated to state packet loss may remain stable or decrease and to include actual results for troubleshooting.

• Chores

  • Increased NIC queue sizes in the network test configuration to improve reliability under buffer-shortage scenarios.

[coderabbitai]

Walkthrough

Adds RX/TX queue size parameters to the NIC config string and relaxes the test's ping-loss evaluation to accept non-increasing (stable or decreasing) packet-loss trends; associated comments and error message were updated.

Changes

Cohort / File(s)	Summary
Config: NIC queue sizes qemu/tests/cfg/netkvm_buffer_shortage.cfg	Appends ,rx_queue_size=1024,tx_queue_size=256 to the nic_extra_params_nic1 string.
Test logic: ping result evaluation qemu/tests/netkvm_buffer_shortage.py	Changes comparison from strictly decreasing (>) to non-increasing (>=) when analyzing ping loss. Updates error message and adds comments noting plateaus (including 0%) are acceptable.

Sequence Diagram(s)

sequenceDiagram
    autonumber
    participant Tester as Test Runner
    participant QEMU as QEMU VM
    participant Guest as Guest OS
    participant Analyzer as analyze_ping_results

    Tester->>QEMU: start VM with updated `nic_extra_params_nic1` (rx_queue_size, tx_queue_size)
    QEMU->>Guest: boot
    Tester->>Guest: run ping workload
    Guest-->>Tester: ping results (packet loss sequence)
    Tester->>Analyzer: analyze ping results
    alt non-increasing or decreasing
        Analyzer-->>Tester: pass (accept stable/decreasing loss)
    else increasing trend
        Analyzer-->>Tester: fail with updated error message (include results)
    end

Loading

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

I twitch my whiskers, configs align,
Queues set tidy—RX and TX shine.
Pings may plateau, steady and bright,
No sudden losses in the night.
A rabbit hops; tests pass light. 🥕✨

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title Check	✅ Passed	The title clearly summarizes the primary change in the pull request by indicating that the netkvm_buffer_shortage test now allows non-increasing packet loss rather than requiring a strict decrease, making it specific and directly related to the main behavior update.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

📜 Recent review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 587dae2 and 753683a.

📒 Files selected for processing (2)

• qemu/tests/cfg/netkvm_buffer_shortage.cfg (1 hunks)
• qemu/tests/netkvm_buffer_shortage.py (1 hunks)

🚧 Files skipped from review as they are similar to previous changes (2)

• qemu/tests/netkvm_buffer_shortage.py
• qemu/tests/cfg/netkvm_buffer_shortage.cfg

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[heywji]

Test Result: PASS

 (1/1) Host_RHEL.m10.u1.ovmf.qcow2.virtio_scsi.up.virtio_net.Guest.Win2022.x86_64.io-github-autotest-qemu.netkvm_buffer_shortage.q35: STARTED      
 (1/1) Host_RHEL.m10.u1.ovmf.qcow2.virtio_scsi.up.virtio_net.Guest.Win2022.x86_64.io-github-autotest-qemu.netkvm_buffer_shortage.q35:  PASS (1275.0
5 s)                                                                                                                                               
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0                                                                  
JOB HTML   : /root/avocado/job-results/job-2025-10-14T13.33-af09327/results.html                                                                   
JOB TIME   : 1276.06 s                                                     

[heywji]

@leidwang Hi Leidong, could you please help me review this patch?

[leidwang]

Why do we need to add these parameters to nic1 now?

[heywji]

I initially pushed the wrong code version to our new test case creation. I found it while debugging the KVMAUTOMA-4186 Jira issues.

[leidwang]

Okay, if so, you can add some notes in commit message, to clarify why add this line.

[heywji]

All done. Please help review the rest. Thanks :)