Skip to content

dt-bindings: phy: describe the T8122 DPTX registers - #1940

Closed
techmore wants to merge 7 commits into
aurora-silicon:integration/aurorafrom
techmore:j773g/dptx-binding-20261114
Closed

techmore wants to merge 7 commits into
aurora-silicon:integration/aurorafrom
techmore:j773g/dptx-binding-20261114

Conversation

@techmore

Copy link
Copy Markdown
Contributor

Describe the T8122 DPTX PHY register layout used by the J773g HDMI stack, stacked on draft PR #1192.

The binding now requires the core and dptx register names. The DTS patch names and orders the existing T8122 register windows accordingly. This is a binding-only / DTS-description change; it does not change runtime code and has no hardware boot of its own.

Dependency: this series is intentionally stacked on #1192 head 700491843df09e2a8c2182215769025609895d5a, whose T8122 PHY node consumes this binding. The dependent PHY implementation commit 39d81c9ac9a1 remains Niklas Sheth's original commit with its original trailers.

Validation:

  • exact unsigned head a829ccde195d7db1d0125a6fd558dea608be98ee passed agents/m4 pr-check with zero reported warnings;
  • the resulting Image is SHA-256 195fbc7a1a9289164f1766665cc743d98fd4c82b941c475ef690a5a87da4daf2;
  • apple/t8132-j773g.dtb is SHA-256 de99bb21c4131b453bbb662aa97e9c6f9429c1e3086045dff427a9de1618f04d;
  • make ARCH=arm64 DT_SCHEMA_FILES=Documentation/devicetree/bindings/phy/apple,dptx-phy.yaml dt_binding_check passed with dtschema 2026.9.

Full-tree schema validation was not run because the tree has a known existing apple,tunable type crash. No separate J773g boot is claimed.

Assisted-by: LLM

techmore and others added 7 commits November 13, 2026 15:28
A T8132 main DCP wired to fixed HDMI uses six firmware segments and shares
the retained SID 0/SID 16 frontend contract with the external display path.
Qualify that layout from the available main DCP and its connector role.
Validate its physical framebuffer separately from inherited mappings and
preserve both frontend streams without a board-name gate.

Assisted-by: LLM checkpatch
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
Some T8132 desktops route the main DCP through a fixed DPTX-to-HDMI path
instead of an internal panel. Use the connector-type property as that role
contract, prepare the PHY and converter rails, adopt the retained firmware
session, and register it through the external-output KMS path. Accept the
retained physical framebuffer without manual bind controls or machine-name
checks.

Use a J773g M4 profile without panel timing and continue to refuse machines
without a profile. J773g does not use the J713 panel admission parameters
because the external path skips apple_pmp_j713_display_prepare() and has no
internal panel timing or panel power admission.

Keep an empty external device from replacing simpledrm, and retain an
adopted session when its initial HDMI DPTX connect fails. The external KMS
layer still caps the device at two outputs, so two adopted dcpext outputs
leave the fixed HDMI output unregistered.

Assisted-by: LLM checkpatch
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
T8122 (as used on T8132) needs its own cold activation: the T8112 core
power steps, moved enable/clear bits (0x2218, 0x2234), a different
0x2230 setting, no 0x4008/0x420c/0x4404/0x4408/0x4600 steps, and an AUX
impedance calibration whose result is copied into 0x4228. At link-rate
change it clears 0x222c bit 1 instead of 0x2220 bit 7, takes the T6020
settings for 0x20b0, the first 0x20b4 write and 0x20b8 bit 16, then
clears 0x20b4 bit 1 and skips the 0x20b8 bit 21 clear.

For T8122 only, enable runtime PM before creating the PHY so the generic
PHY core propagates runtime-PM references to the provider. The T8112,
T6020 and T6050 paths are unchanged.

Link: GravityLinux/linux@877c574
Assisted-By: OpenAI Codex (GPT-5.6 Sol)
Signed-off-by: Niklas Sheth <hi@niklassheth.com>
[Sean: extracted the T8122 hunk; rebased over T6050; runtime PM for T8122 only; calibration code from the readback]
Assisted-by: LLM checkpatch
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
The Mac mini routes its main DCP through the T8122 DPTX PHY and an HDMI
converter controlled by two SMC GPIOs. Describe that fixed path and enable
the retained DCP, mailbox, and display DART nodes. Do not attach the DCP or
mailbox to DISP_CPU: iBoot leaves that domain in auto mode and direct PMGR
writes fault on this machine.

Assisted-by: LLM checkpatch
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
Name and order the existing T8122 DPTX PHY register windows as core and
dptx to match the driver and binding.

Assisted-by: LLM
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
With only the HDMI path described, the Mac mini still has no display:

- The DPTX PHY sits on ps_dptx_phy, which the J713 PMGR policy (applied
  to every M4 board that does not state the domain) keeps from Linux.
  phy-apple-dptx and the DCP behind it wait for it until the deferred
  probe timeout (-110). State the domain for this board, as it already
  does for the ATC2 and ATC3 domains.
- The DCP's PIODMA child owns iommu-addresses entries in
  /reserved-memory, so the DCP node needs #address-cells and
  #size-cells; the display DART's admission otherwise warns that both
  are missing.
- The loader and the display DART look the PIODMA stream up by the
  disp0_piodma alias, as on T8122; add it.

Assisted-by: LLM
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
The T8122 driver maps its core and DPTX windows by name. Require those
resource names for DPTX PHY nodes.

Assisted-by: LLM
Signed-off-by: Sean Dolbec <sdolbec1@gmail.com>
@techmore

Copy link
Copy Markdown
Contributor Author

One corrected-config boot on our J773g exercised signed head bcb17f6f0be18bd61f194de2f3b70502f023b391, which has the same tree as reviewed unsigned commit a829ccde195d7db1d0125a6fd558dea608be98ee, using the PR's own apple/t8132-j773g.dtb; all CPUs came online, while Ethernet did not enumerate and only loopback was present.
The retained DCP firmware 26.6.2 stub was adopted (DCPLink READY; external DCP adopted), DPTX services enumerated, and the RTKit crash count was 0.
The receipt has no PHY-bind line naming the core and dptx resources; it does show apple-dcp 30ae00000.dcp: fixed HDMI PHY and converter rails prepared and no new PHY probe failure.
At the timed capture HDMI-A-1 was disconnected and disabled, then HPD connected it and drove 1920x1080 at 60 Hz and 120 Hz; the receipt shows the first #1940 attempt and the #1192 head alone asserting HPD within 0.1 seconds of each other, while the corrected boot asserted HPD earlier.
G16's firmware-identity rejection is expected because this stack does not contain #1718.
The first attempt used a mis-configured test kernel without CONFIG_ARM64_CTXR_EXECMEM, leaving 9 of 10 CPUs offline, so it was repeated with the correct config; no PR-attributable regressions were found.

@techmore

Copy link
Copy Markdown
Contributor Author

Closing: stacked on #1192's old head, and #1192 does not need it (its J773g node already names the DPTX windows and the driver maps by name). The two small commits (reg-names required; naming the T8122 node's windows) can be sent on #1192's current head once checked against the T6041/T6050 nodes.

@techmore techmore closed this Nov 19, 2026
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