Skip to content

arm64: dts: apple: t8140-j700: Show the software camera indicator (slot 1) - #2651

Merged
RyanTheTide merged 1 commit into
integration/aurorafrom
neo/j700-camera-indicator-slot1
Nov 22, 2026
Merged

RyanTheTide merged 1 commit into
integration/aurorafrom
neo/j700-camera-indicator-slot1

Conversation

@RyanTheTide

@RyanTheTide RyanTheTide commented Nov 21, 2026 •

Copy link
Copy Markdown
Contributor

Problem. The MacBook Neo's camera has no indicator light. #2561 added a best-effort software indicator, which the DCP draws in a reserved swap slot, but J700 doesn't opt in yet. The earlier opt-in (#1758) used slot 2. The DCP blends slot 2 only where the primary is transparent, so under GNOME's opaque AR30 desktop the indicator never reached the panel.

Change. DT only, J700 only. apple,camera-indicator-surface = <1> on the DCP, and apple,camera-indicator = <&dcp> on the ISP. Slot 1 sits directly above the surface DRM uses (apple,iomfb-surfaces = <0>), so it isn't one userspace could use.

Testing. On a J700, using fleet76 plus the #2561 commits, with exactly these two properties in the DTB (one-shot trial boots):

  • Opaque AR30 primary through modetest, gdm stopped: the blend CRC changes while the camera streams and comes back exactly after stop, 3/3. Slot 2 on the same primary: no change.
  • GNOME: visible 2/2, cleared at stop.
  • DPMS off/on with the stream running: the display and the dot come back, 3/3.
  • Camera stops while the display is off: the indicator is out of plane state after DPMS on, 8/8. Camera starts while the display is off: it is in plane state after DPMS on, 8/8.
  • 0 swap failed, 0 present dropped, 0 oops, 0 AVC. Row details: drm/apple, isp: add a best-effort camera indicator (unbooted) #2561 (comment) and drm/apple, isp: add a best-effort camera indicator (unbooted) #2561 (comment)
  • This DTB builds, and the two properties come out as tested: surface 0x01, ISP phandle to the DCP. checkpatch --strict is clean, and the CI dt-checks (dtbs_check) pass.

Still to run.

  • A settled CRC proof for the display-off cases. After a DPMS cycle GNOME's screen never holds still, so the CRC there is inconclusive; the plane-state results above are clean.
Notes for review

Assisted-by: LLM (Claude:Opus 5.5)

The MacBook Neo's camera has no indicator light. Point the ISP at the
DCP so the kernel shows its best-effort software camera indicator while
the camera streams, and give the indicator swap slot 1.

Slot 1 sits directly above the surface DRM uses (iomfb-surfaces <0>).
Slot 2 was tried first: the DCP blends it only where the primary is
transparent, so under an opaque primary such as GNOME's AR30 desktop the
indicator never reached the panel. With slot 1 it shows over an opaque
primary and is cleared when the stream stops.

Signed-off-by: Ryan Murray <ryan@aurorasilicon.org>
@RyanTheTide

Copy link
Copy Markdown
Contributor Author

No blockers.

DT only, on the J700 board file. apple,camera-indicator-surface = <1> passes the check in drivers/gpu/drm/apple/dcp-indicator.c:63, because the J700 DCP's only DRM surface is apple,iomfb-surfaces = <0> (t8140.dtsi:1704), so slot 1 is free and above it. Both properties are documented in the dcp and t8140-isp bindings. The commit has a human Signed-off-by and no agent trailers, and the J700 rows (opaque AR30 CRC 3/3, GNOME 2/2, DPMS 3/3 and 8/8, 0 swap failures and 0 AVC) cover the change. The DPMS plane-state behaviour you describe belongs in #2561's code and is already noted there.

Assisted-by: LLM (Claude:Opus 5.5)

@DjDeveloperr DjDeveloperr left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pre-PR checklist review (§1–§6) at 6f729e6caf89: APPROVE (0 blockers).

  • Code & DT checks (arch/arm64/boot/dts/apple/t8140-j700.dts:629,703):
    • apple,camera-indicator-surface = <1> on &dcp satisfies dcp_indicator_probe() (drivers/gpu/drm/apple/dcp-indicator.c:55-68): J700 uses apple,iomfb-surfaces = <0> (t8140.dtsi:1704, mapped to bit 0 in dcp.c:7369-7374), so slot 1 is free above surface 0 and < SWAP_SURFACES, and apple_dcp_hw_t8140 (dcp.c:9727-9737) uses H17P (not H17G), so the swap_done hook is present.
    • apple,camera-indicator = <&dcp> on &isp matches apple_isp_indicator_node() (drivers/media/platform/apple/isp/isp-drv.c:833-838) and both DT bindings (apple,dcp.yaml:195, apple,t8140-isp.yaml:93).
  • §1–§3 / §6: Board-DT-only opt-in on t8140-j700.dts (the only T8140 board in tree), default kernel command line, no hacks or naming issues.
  • §4 Attribution: Commit 6f729e6caf89 carries human author/committer and human Signed-off-by: Ryan Murray <ryan@aurorasilicon.org> only, with zero AI trailers in the commit.
  • §5 Evidence & CI: dt-checks and host-tests are green on 6f729e6caf89. Hardware rows on J700 (#2561 comments 7052690642 and 7055375971) exercise these exact two DT properties: modetest opaque AR30 blend CRC 3/3, GNOME AR30 2/2, DPMS off/on during stream 3/3, camera stop/start while dark 8/8 each, 150-frame stream rc=0, 0 AVC denials, 0 failed units, 0 oops/WARN, and 0 swap failed / present dropped. (The DPMS retained-layer plane-state note belongs to #2561's driver code, not this DT opt-in.)

Assisted-by: LLM (Gemini 4 Neon)

@RyanTheTide
RyanTheTide merged commit 8fc8c10 into integration/aurora Nov 22, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants