qcs6490: VP9 fix backported from mailing list

[HeyMeco]

Description

Upstream Venus driver has a broken implementation regarding VP9 playback. After reporting it to qcom this patch was proposed in the LKML.

Please squash merge this PR

Test

• Playback of Big-Buck-Bunny-1080P-VP9 on V4L2 enabled chromium by @amazingfate making sure decoder is V4L2VideoDecoder

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules

Summary by CodeRabbit

• Bug Fixes
  • Fixed VP9 decoder End of Stream (EOS) handling for IRIS2 devices running older firmware (<= v1.0.87), improving video decoding reliability on those platforms.
  • Preserved previous EOS behavior for other hardware so non-IRIS2 platforms remain unaffected.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

Restricts the VP9 EOS quirk to IRIS2 cores by replacing the previous IS_V6-based condition with an IRIS2-specific check; sends a NULL EOS address only when running on IRIS2 with firmware ≤ 1.0.87, preserving the non-zero dummy address for other cores.

Changes

Cohort / File(s)	Summary
VP9 EOS Quirk Scoping patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch	Replace generic IS_V6 guard with an IRIS2-specific condition; send NULL EOS address only for IRIS2 with firmware ≤ 1.0.87; add explanatory block comment; keep non-zero dummy address for non-IRIS2 cores.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

• Single targeted hardware-specific change in one patch file.
• Review items: conditional correctness for IRIS2 check, firmware version comparison, and comment clarity.

Suggested labels

Ready to merge

Suggested reviewers

• igorpecovnik
• paolosabatino
• joekhoobyar

Poem

🐰 A little hop for IRIS2's tune,

EOS now settled under the moon,
Firmware checked, the logic neat,
VP9 hums with a steady beat,
Hooray — decoding's snug and sweet! 🎬✨

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 accurately reflects the main change: backporting a VP9 fix from the mailing list for the qcs6490 platform, which directly corresponds to the patch modification in the changeset.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

🧪 Generate unit tests (beta)

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

---

📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Jira integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between e3cd5c5 and ff0d043.

📒 Files selected for processing (1)

• patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch (1 hunks)

🚧 Files skipped from review as they are similar to previous changes (1)

• patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 0

🧹 Nitpick comments (1)

patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch (1)

24-24: Minor typo: missing space after comma.

Line 24 has "IRIS2 (SM8250),for firmware" - missing space after the comma.

Apply this diff to fix the formatting:

-		/* Send NULL EOS addr for only IRIS2 (SM8250),for firmware <= 1.0.87.
+		/* Send NULL EOS addr for only IRIS2 (SM8250), for firmware <= 1.0.87.

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Jira integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between fd0a6d7 and e3cd5c5.

📒 Files selected for processing (1)

• patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch (1 hunks)

🧰 Additional context used

🧠 Learnings (2)

📓 Common learnings

Learnt from: SuperKali
Repo: armbian/build PR: 8608
File: lib/functions/compilation/patch/drivers_network.sh:361-363
Timestamp: 2025-09-11T06:06:52.328Z
Learning: For the RTW88 SDIO RF path detection fix patch (004-rtw88-sdio-rf-path-detection-fix.patch), SuperKali prefers using exact version matching (eq 6.1 || eq 6.16) rather than range checks, as the patch has only been tested on those specific kernel versions and may be mainstreamed soon.

📚 Learning: 2025-09-11T06:06:52.328Z

Learnt from: SuperKali
Repo: armbian/build PR: 8608
File: lib/functions/compilation/patch/drivers_network.sh:361-363
Timestamp: 2025-09-11T06:06:52.328Z
Learning: For the RTW88 SDIO RF path detection fix patch (004-rtw88-sdio-rf-path-detection-fix.patch), SuperKali prefers using exact version matching (eq 6.1 || eq 6.16) rather than range checks, as the patch has only been tested on those specific kernel versions and may be mainstreamed soon.

Applied to files:

• patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch

🔇 Additional comments (2)

patch/kernel/archive/qcs6490-6.18/0050-6.18-VP9-Fix.patch (2)

1-12: LGTM! Clear patch metadata.

The patch metadata follows standard git format, and the subject/description clearly explain the fix scope and rationale.

---

29-32: IS_IRIS2 macro is defined in kernel 6.18 Venus driver.

The change from IS_V6 to IS_IRIS2 correctly restricts the EOS quirk to IRIS2 cores only. The IS_IRIS2 macro is defined in drivers/media/platform/qcom/venus/core.h as a vpu_version check and is used throughout the Venus driver in 6.18, including in hfi_venus.c and core.c.

[leggewie]

Please squash merge this PR

rebase done, thus dropping the merge commit. I assume that's what you were looking for? Or do you want to further squash the remaining two commits into 1?

[HeyMeco]

rebase done, thus dropping the merge commit. I assume that's what you were looking for?

Nice but yeah we don't need this as two commits when merged it should be merged as one with reference to the PR

[leggewie]

thank yo for your work

[github-actions]

✅ This PR has been reviewed and approved — all set for merge!