Skip to content

Apple M6 (T8152) Type-C PHY: driver, binding, dwc3 changes, disabled PHY nodes (on top of #191) - #192

Open
joshuaswarren wants to merge 15 commits into
omacom:integration/aurorafrom
joshuaswarren:m6-phy
Open

joshuaswarren wants to merge 15 commits into
omacom:integration/aurorafrom
joshuaswarren:m6-phy

Conversation

@joshuaswarren

@joshuaswarren joshuaswarren commented Oct 23, 2026 •

Copy link
Copy Markdown

Summary

Adds the SuperSpeed and USB4 (CIO) Type-C PHY of the ATC ports of the Apple M6 (T8152), its binding, two small changes in the Apple dwc3 glue, and two disabled PHY nodes for the Mac mini (J873). Five commits.

The first ten commits of this branch are #191 (the M6 base). Only the last five commits are new here: the diff of this PR shows the base too until #191 merges, then I rebase. Review the last five.

It is not needed for the release candidate, and there is no rush to review or merge it. Nothing in this series enables the driver on a board.

Commits

# Commit Notes
11 dt-bindings: phy: apple: add the T8152 Type-C PHY binding only
12 phy: apple: add the T8152 Type-C PHY driver new atc-t8152.c, off by default (PHY_APPLE_T8152_ATC)
13 usb: dwc3: apple: make the USB3 PHY optional on T8152 a board without the SuperSpeed PHY still probes USB2-only
14 usb: dwc3: apple: set GCTL.PWRDNSCALE on T8152 2-line call plus helper, T8152 only
15 arm64: dts: apple: t8152-j873g: add the ATC PHY nodes two nodes, both status = "disabled"

Measured, on an M6 Mac mini (front USB-C port, a VL822 hub behind it)

  • The PHY bring-up and the route of the PIPE to USB3 write cleanly: every write was read back, no mismatches.
  • The CIO bring-up on the rear port writes cleanly.
  • GCTL.PWRDNSCALE has the boot value 1560. With 13, the time from RxDetect to Polling drops from 1.56-1.69 s to under 90 ms (front port only).

Not measured

  • The shutdown sequence has not been run on hardware.
  • The other T8152 ports (same code path, same value).
  • The kernel's own make dtbs_check / dt_binding_check (a hand-run dt-validate of the board file gave 0 messages).
  • The binding example does not validate by hand: dt-validate (dtschema 2026.9) stops with KeyError: 'uint8-array' while decoding apple,tunable-aciophy-cmn-dflt. Inferred, not proven: apple,pmp.yaml in this tree defines a broad ^apple,tunable- pattern of type uint8-array that overlaps this binding's ^apple,tunable-[a-z0-9-]+$ pattern. The board file (PHY nodes disabled and forced to okay) validates with 0 messages.
  • This branch was not booted. The earlier form of these patches ran against the lab tree; the driver built clean there.

Blocked

  • With the PIPE routed, the SuperSpeed link of the front port goes RxDetect, then Polling, and does not train. The xHC reset also does not complete once the PIPE is routed. The driver is therefore not enabled by any board, and the front-port node stays disabled.
  • The ATC3 node serves the ACIO controller, whose nodes are not in this change.

Compile receipt

arm64 clang, W=1, integration/aurora at 42672ed plus the 15 commits of this branch, defconfig plus PHY_APPLE_T8152_ATC=y: drivers/phy/apple/atc-t8152.o, drivers/usb/dwc3/dwc3-apple.o, the eight objects of #191 and apple/t8152-j873g.dtb all rc 0 with 0 warnings. check-commits: Signed-off-by on all 15 commits; no checkpatch errors (warnings: long lines, the MAINTAINERS notice).

@joshuaswarren

Copy link
Copy Markdown
Author

Compile receipt (arm64, clang, W=1, integration/aurora at 42672ed plus the 15 commits of this branch, defconfig plus PHY_APPLE_T8152_ATC=y, no debug info):

Target rc warnings
drivers/phy/apple/atc-t8152.o 0 0
drivers/usb/dwc3/dwc3-apple.o 0 0
the eight objects of #191 0 0
apple/t8152-j873g.dtb 0 0

Not run: a link of the whole kernel, a boot, dt_binding_check, dtbs_check. Please run dtbs_check in CI.

@greptile-apps

greptile-apps Bot commented Oct 23, 2026 •

Copy link
Copy Markdown

RetriggerView in GreptileConfidence Score: 5/5

[High impact] The latest changes appear safe to merge and address the remaining numbered finding.

Summary

Adds the T8152 Type-C PHY driver, binding, disabled board nodes, and DWC3 support on top of the M6 base changes.

  • The T8152 ATC PHY can bring up USB3 or CIO.
  • A T8152 DWC3 port can run without a SuperSpeed PHY.
  • T8152 DWC3 starts with a power-down scale of 13.
  • The J873 device tree describes two disabled ATC PHYs.
  • T8152 J873 ports gain USB2 PHY and DWC3 support.
  • The Apple DART driver supports T8152 Gen3 translation units.
  • T8152 power domains can be described and controlled by Linux.
  • Apple SART supports the T8152 SARTv4 register layout.
  • The Apple NVMe driver supports T8152 ANS hardware.
  • The Apple pinctrl driver recognizes T8152.
  • The Apple 05ac:800f hub stays awake instead of autosuspending.
  • The J873 device tree describes supported T8152 peripherals.

Reviews (10) · Last reviewed commit: "arm64: dts: apple: t8152-j873g: add the ..." · Reviewed by Greptile

Comment thread drivers/phy/apple/atc-t8152.c Resolved
Comment thread drivers/phy/apple/atc-t8152.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

Rebased on the fixed #191 head e27454a38c1a7d88a14a3ad242be2142f72b1184 (the five commits of this PR are unchanged; the first ten changed as listed on #191). Head b63a181e3f057; compile check on the new stack: arm64 clang W=1, atc-t8152.o, dwc3-apple.o, the eight objects of #191 and t8152-j873g.dtb, all rc 0, 0 warnings.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at b63a181: findings (last five commits only, as asked; read only, nothing built or booted by me, no dt_binding_check or dtbs_check run).

Checked and fine:

  • apple,t8152-atcphy binding vs driver vs DTS: reg-names = "core", #phy-cells = <1>, apple,dwc3, apple,cio-lanes (1 to 3, same check in probe), the tunable property pattern and the DTS nodes agree. The two nodes are disabled. The labels they use (usbdrd2, usbdrd3, ps_atc2_usb_aon, ps_atc3_usb_aon, psd_atc2_usb, psd_atc3_usb) are all defined in Apple M6 (T8152) Mac mini base: PMGR, USB2, DART, NVMe, board fragments #191; the unit addresses match the reg values (0x383000000, 0x413000000).
  • The probe looks up pipe-handler, dwc3-core and dwc3-apple in the dwc3 node; all three names are in the apple,t8152-dwc3 binding, and 0x10000 + 0x2c lies inside the dwc3-apple window (0xcd00 to 0x117ff), which the probe also checks.
  • dwc3: devm_phy_optional_get() returns NULL when no PHY is described, and phy_set_mode(), phy_power_off() on NULL return 0, so the USB2-only case keeps working. The GCTL write is a read-modify-write of the PWRDNSCALE field only, done once per core start after the soft reset.
  • PHY driver: mode changes, bring-up, shutdown, the PIPE route and the orientation switch all take atc->lock; a failed bring-up shuts the block down and leaves the mode OFF; the Type-C re-run restores the PIPE route; Kconfig has depends on TYPEC, so the PHY cannot be built in while Type-C is a module.

Findings:

  1. apple,t8152-atcphy.yaml example: reg = <0x3 0x83000000 0x0 0x84000> sits in the default example parent with one address cell and one size cell, so it parses as two entries and reg (one item) will not match. Put the node in a parent with #address-cells = <2>; #size-cells = <2>;. I did not run dt_binding_check; this is from reading.
  2. The apple,t8152-dwc3 binding (from Apple M6 (T8152) Mac mini base: PMGR, USB2, DART, NVMe, board fragments #191) allows exactly one PHY: phys has maxItems: 1 and phy-names is usb2-phy only. This series tells the user to add a usb3-phy reference (the dwc3 commit and the comment in t8152-j873-atcphy.dtsi), and no commit here touches that binding, so a node that does it fails the schema. Allow the optional second entry in this series.
  3. power-domains and power-domain-names in the PHY binding are minItems: 1, maxItems: 8 with no names. The nodes pass usb-aon and usb, and the driver attaches all of them by index. Fix the two names and their order in the binding (nit).
  4. t8152_poll_us() polls with udelay(step_us) for up to 1 s (t8152_powerup_rst(), t8152_shutdown()), with the PHY mutex held and in sleepable context. read_poll_timeout() with a sleep would not burn a CPU for that time (nit; no correctness effect).
  5. Question: the probe takes the PHY power domains with pm_runtime_resume_and_get() and keeps them for the life of the driver, so the ATC domains stay on even when the PHY is off. If that is intended, say so in the comment; if the PHY should power the domains with the transport, the get and put belong in atc_up() and atc_down().

Not checked: the register sequences and the tunable table against the vendor driver or hardware; that phy_power_off() reaches atc_usb3_power_off() (the framework only calls the op at power count 1, which here depends on the dwc3 core having powered the same PHY object on; I read the glue, not the core); everything in the first ten commits (those are #191).

Comment thread drivers/iommu/apple-dart.c Resolved
Comment thread drivers/usb/dwc3/dwc3-apple.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

Review at b63a181: findings (five commits: Type-C PHY binding and driver, dwc3 changes, PHY nodes)

Checked

  • Device tree: I built the board file by hand at this head (cpp, dtc -@, no warnings), set both ATC PHY nodes to okay and ran dt-validate (dtschema 2026.9). phy@383000000 and phy@413000000 produce no message. apple,dwc3 points at labels that exist (usbdrd2, usbdrd3), and the PHY windows nest the two USB DART windows as the driver comment describes.
  • dwc3: devm_phy_optional_get() for T8152 only; phy_set_mode() on a NULL PHY returns 0, which the earlier commit relies on. The GCTL write sits after the core soft reset and before dwc3_apple_setup_cio(). The commit text states the measurement (one port, 1.56 s down to 88 ms).
  • The driver is not enabled by any node in this series (both PHY nodes are disabled), so none of this code runs on a board built from it. That limits the risk of the findings below to a later change that enables a node.

Findings

  1. Failed route skips parking. t8152_pipehandler_usb3_host() moves the mux and can then return an error (the wait for LOCK_ACK to clear, or the wait for PIPE_PHY_READY). atc_route_usb3() sets pipe_usb3 only on success, so after a failure atc_down() skips t8152_pipehandler_park() and shuts the PHY down while the controller still points at it. Record the route before the waits, or park on the failure path. The bot comment says the same.
  2. Orientation retry. atc_sw_set() stores swap before the restart. If atc_up() fails it has already shut the PHY down (mode is ATC_OFF), and the next call with the same orientation returns 0 at swap == atc->swap, so nothing retries while the PHY stays off. Same as the bot comment: keep a pending-restart flag.
  3. Binding example. With the schema alone, dt-validate reports phy@383000000: reg: [[3, 2197815296], [0, 540672]] is too long: the example wrapper has one address cell and one size cell and the node uses two of each. Wrap the example in a soc node with #address-cells = <2> and #size-cells = <2>. With the whole schema set of this tree, dt-validate stopped with KeyError: 'uint8-array' while decoding apple,tunable-aciophy-cmn-dflt of the same example. I did not find the cause; the CI run of dt_binding_check will show whether it reproduces.
  4. Polling. t8152_poll_us() spins with udelay(step) for up to 1 s per wait (two waits in power-up, two in shutdown, 130 ms on the PIPE status). Every caller runs in process context under atc->lock, a mutex. A failing PHY costs up to about 2 s of CPU spin per bring-up. readl_poll_timeout() sleeps between reads and fits these callers.

Smaller

  • atc_up() uses dev_err_probe() outside probe; dev_err() is the plain call. PHY_APPLE_T8152_ATC calls of_address_to_resource() and friends but has no depends on OF for the COMPILE_TEST case.
  • checkpatch on the five new commits: LONG_LINE and LONG_LINE_COMMENT (up to 152 columns) in atc-t8152-seq.h, in the dwc3 comment of commit 13 and in the new dtsi, and FILE_PATH_CHANGES for the new files. No errors. The PR text says "clean checkpatch".

Not checked

No compile at this head (I did not reproduce the receipt), no boot, no check of the register sequences against the vendor driver or the hardware, no lockdep or sparse run. The first ten commits are #191 and are not reviewed here.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at b63a181 (issuecomment-6369902701) received; the five findings are with the PHY author for a v6 (binding example parent, dwc3 phys second entry, power-domain names and count, the 1 s busy wait, the power-domain lifetime question). The branch is now rebased on the fixed #191 head c9d88a5d8fb390d8b61f5d2154f456aa81edfc78, head cfddda0d2a68cae9374a8102f3afc90e8fb7dbbc, with the five commits unchanged. Compile check on that stack: arm64 clang W=1, atc-t8152.o, dwc3-apple.o, the eight objects of #191 and t8152-j873g.dtb, rc 0, 0 warnings.

Comment thread drivers/usb/dwc3/dwc3-apple.c Resolved
Comment thread drivers/pmdomain/apple/pmgr-psd.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

v6 from the PHY author, on the fixed #191 head 9af182774f886303e4c85497549f64498551dd03: new head 401e5916b740f6d78547fcef6c465c7d3919ea18, five commits. Per finding of the two reviews (issuecomment-6369902701 and -6370157839):

  • Binding example: in a parent with two address and two size cells, self-contained.
  • dwc3 binding: phys takes one or two entries, phy-names is usb2-phy then the optional usb3-phy, in the dwc3 commit.
  • PHY power domains: exactly two, named usb-aon then usb.
  • Failed route skips parking: the route is recorded as soon as the mux moves, and atc_down() parks.
  • Orientation retry: a pending-restart flag, set when the restart fails and honoured when the orientation is unchanged.
  • Polling: t8152_poll_us() uses read_poll_timeout() (sleeping); all 13 callers run in process context under the PHY mutex.
  • Domain lifetime question: intended. The PHY registers are touched from PHY ops and the Type-C switch, which hold no runtime-PM reference, and the domain is shared with the dwc3 node; the comment says so.
  • Smaller: dev_err() in atc_up(), depends on ... && OF, long lines wrapped where possible. Checkpatch has no errors; warnings remain: long lines in the data tables and the driver, the MAINTAINERS notice, and "binding in a separate patch" for the dwc3 commit (the binding change has to live there).
  • Example decoder error (KeyError: 'uint8-array'): still there with dtschema 2026.9. I tested an inline tunable schema and an aligned type in the dwc3 binding: no change. Likely cause (inferred): the broad ^apple,tunable- uint8-array pattern of apple,pmp.yaml overlaps this binding's pattern. The board file validates with 0 messages, also with the PHY nodes forced to okay.

Compile check at this head, arm64 clang W=1, with PHY_APPLE_T8152_ATC=y: atc-t8152.o, the eight objects of #191 and t8152-j873g.dtb, all rc 0, 0 warnings.

Comment thread drivers/usb/dwc3/dwc3-apple.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

GREEN at 401e591 for the last five commits (PHY v6). Read only: nothing built or booted by me, no dt_binding_check or dtbs_check run by me. My five findings, checked in the commits at this head:

  1. Binding example: the nodes sit in a soc parent with two address and two size cells and the example carries its own power-controller and usbdrd2 nodes, so it is self-contained.
  2. dwc3 binding (apple,t8152-dwc3.yaml, in the dwc3 commit): phys has minItems: 1, maxItems: 2, phy-names is usb2-phy then the optional usb3-phy, and the description says a port is USB2-only without the second entry; the driver uses devm_phy_optional_get() for it on the T8152 USB2 controller.
  3. PHY binding: power-domains is exactly two entries and power-domain-names is usb-aon then usb, matching the DTS nodes.
  4. t8152_poll_us() now uses read_poll_timeout(readl, ...) (sleeping); every caller runs under the PHY mutex from a PHY op or the Type-C switch callback, which are process context.
  5. Power domains: the comment in atc_probe() now gives the reason for holding them for the driver lifetime (the ops and the switch hold no runtime-PM reference, and the ATC USB domain is shared with the dwc3 node); that answers my question.
    Also seen and fine: t8152_pipehandler_usb3_host() sets *routed as soon as the mux points at the PHY, atc_route_usb3() records it either way, so atc_down() parks the PIPE after a failed settle wait; the orientation restart keeps a restart flag that is retried by the next bring-up or switch call and cleared by a successful atc_up(); Kconfig adds && OF; the DTS is as before (two disabled nodes).

Not checked: the register sequences and the tunable table against the vendor driver or hardware, and the example decoder error the author reports (KeyError: 'uint8-array'), which I did not reproduce.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at 401e591: findings (five commits: Type-C PHY binding and driver, dwc3 changes, PHY nodes)

I compared the five commits with the ones I reviewed at b63a181e and re-ran the dt check by hand at this head.

Fixed, checked

  • Route recording: t8152_pipehandler_usb3_host() sets *routed as soon as the mux points at the PHY and atc_route_usb3() stores it whether the waits pass or not, so atc_down() parks after a failed wait.
  • Polling: t8152_poll_us() is read_poll_timeout() and sleeps; all callers hold the PHY mutex in process context.
  • Binding: exactly two power domains, usb-aon then usb; phys takes one or two entries and phy-names is usb2-phy then the optional usb3-phy; dev_err() replaces dev_err_probe() in atc_up(); depends on ... && OF. The power-domain lifetime comment answers my question.
  • dtbs_check by hand (dtcheck.sh, dtschema 2026.9) on a tree with this head: dt-validate of t8152-j873g.dts rc 0 with 0 lines. With the two PHY nodes set to okay, also 0 lines.

Findings

  1. The restart flag does not retry. After a failed orientation restart atc->mode is ATC_OFF and atc->restart is set. The next atc_sw_set() call with the same orientation passes the first test (!atc->restart), stores swap, and then returns 0 at if (atc->mode == ATC_OFF), so the comment "a pending restart retries here" is not what the code does and the flag stays set. Nothing else brings the PHY up again: with fixed-role dwc3 only calls phy_set_mode(HOST) at core start. The USB3 PHY therefore stays off after one failed restart, which is the case the bot comment described. Remember the mode to restore (or the failed mode) and re-run atc_up() and the route when restart is set, even when mode is ATC_OFF.
  2. atc_route_usb3() returns 0 on a retry after the PHY did not become ready. When the wait for PIPE_PHY_READY fails, the route is recorded (pipe_usb3 true) and -ETIMEDOUT goes back to phy_set_mode(). A second phy_set_mode(HOST) finds routed true and returns 0 without waiting again, so a later core start looks successful with the PHY not ready. Keep a "ready" flag and repeat the wait, or return the earlier error.
  3. Binding example. With the schema alone, the example now validates (0 lines). With soc/apple/apple,pmp.yaml in the schema set, dt-validate stops with KeyError: 'uint8-array' while decoding apple,tunable-aciophy-cmn-dflt; I confirmed the cause the owner inferred: the broad ^apple,tunable- pattern of apple,pmp.yaml (a uint8-array). Removing the two apple,tunable-* lines from the example makes it validate with the full schema set (0 lines). The tunables are supplied by the boot loader, so the example does not need them. The CI dt_binding_check hits the same decoder on this example.

Smaller

  • Checkpatch has no errors; warnings are the long lines of the data tables, the MAINTAINERS notice and the binding in the dwc3 commit.

Not checked

No compile at this head (I did not reproduce the receipt), no boot, no register sequence against the vendor driver or hardware, no lockdep or sparse run. The first ten commits are #191.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at 401e591: I withdraw my earlier GREEN on this head. I re-read the restart path after the reviewer's finding 1 and it holds.

Trace: atc_sw_set runs atc_down (mode becomes ATC_OFF), then atc_up fails, so restart is set and the call returns the error. On the next orientation event the first test passes because restart is set, then "if (atc->mode == ATC_OFF) return 0" returns without bringing the PHY up and without clearing restart. The dwc3 core calls phy_set_mode only at core start, so nothing else brings the USB3 PHY up. The comment on the restart field says the next bring-up retries; no bring-up does.

Also confirmed finding 2: atc_route_usb3 records pipe_usb3 = routed even when the PIPE_PHY_READY wait failed, so the next call returns 0 at "if (routed)" and the wait is never repeated. Combined with the above, a failed restart also forgets whether USB3 was routed (atc_down clears pipe_usb3 and the local "routed" is lost).

Fix direction: store the mode to restore (and whether USB3 was routed) in the struct when the restart fails, and have the retry call atc_up with that mode and then atc_route_usb3. Finding 3 (the atcphy example and the pmp.yaml tunable pattern) I did not check; the other reviewer tested it.

@joshuaswarren

Copy link
Copy Markdown
Author

v7 from the PHY author, on the #191 head 9af182774f886303e4c85497549f64498551dd03: new head a6c54abdc9315c83139626a6c4abc36a895932f6, five commits. git range-diff against v6: the dwc3, PWRDNSCALE and dts commits are identical; the binding and the driver changed. Per finding of the reviews at 401e5916 (w7X issuecomment-6373046448, w7Q's withdrawal):

  1. Restart retry. The driver remembers the transport last asked for (want_mode, set by phy_set_mode(HOST) and the CIO power_on, cleared by a deliberate power_off). When atc_sw_set() finds the PHY off with the restart flag set, it runs atc_up(want_mode) and the PIPE route again; a failure keeps the flag, a success clears it. A deliberate power_off withdraws the request, so an orientation event never powers the PHY on by itself.
  2. Route after a failed PIPE_PHY_READY wait. The wait is split out (t8152_pipehandler_usb3_ready()). atc_route_usb3() repeats it whenever the route stands but is not confirmed (new route_ready, set on success, cleared at park and at shutdown) and returns its error.
  3. Binding example. The two apple,tunable-* lines are gone from the example only; the patternProperties stay. With dtcheck.sh (dtschema 2026.9) on this head, the example now validates with the full schema set: the 73 examples print one message, display/apple,dcp-typec-routes.yaml, which is in the baseline. The board file: 0 messages.

Compile check at this head, arm64 clang W=1, PHY_APPLE_T8152_ATC=y: atc-t8152.o, the eight objects of #191 and t8152-j873g.dtb, all rc 0, 0 warnings. Checkpatch: no errors; warnings remain (MAINTAINERS notice, long lines in the tables and the driver, the binding change inside the dwc3 commit, udelay at the fixed vendor delays). The restart and route paths were reviewed by reading and compiled, not run on hardware.

Comment thread drivers/phy/apple/atc-t8152.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

Review at a6c54ab: findings. The two fixes I asked for are in and read correctly, with one gap in the retry and one dropped error.

Read: atc-t8152.c and atc-t8152-seq.h at 401e591 and at this head (diff), and the binding diff (only the two apple,tunable-* example lines removed). Not read again: the dwc3, PWRDNSCALE and dts commits (the author's range-diff says identical). Not run: build, boot.

Fine:

  • want_mode is set by set_mode(HOST) and the CIO power_on, cleared by both power_off paths, kept across atc_down. In atc_sw_set the ATC_OFF branch brings the PHY back for the remembered transport, keeps restart on a failed atc_up, and leaves the PHY alone after a deliberate power_off. This closes the dead end I withdrew for.
  • atc_route_usb3 now repeats the PIPE_PHY_READY wait when the route stands (pipe_usb3 true, route_ready false). route_ready is cleared in atc_down.

Findings:

  1. (should fix before merge, one line) The retry can forget that USB3 needs a route. Trace: PHY up and routed, orientation changes, atc_down clears pipe_usb3, atc_up works, atc_route_usb3 fails with routed still false (the mux never moved), so restart = true and pipe_usb3 = false. The next event passes the restart test, runs the main path with routed = atc->pipe_usb3 = false, does down and up, skips the route, and atc_up clears restart. The PHY is up, the PIPE is not routed, and the call returns 0. The USB3 transport is dead and nothing records it. The mode is ATC_USB3 here, and only atc_usb3_set_mode brings it up, so route on "routed || mode == ATC_USB3" (or want_mode == ATC_USB3).
  2. (should fix) In atc_route_usb3 the new second step overwrites ret. t8152_pipehandler_usb3_host returns the LOCK_ACK poll result; it used to return early on a failure, and in v6 that error reached the caller. Now, when routed is true but LOCK_ACK failed, the ready wait runs and if it succeeds the function returns 0 and sets route_ready. A retry (pipe_usb3 already true) only repeats the ready wait, so LOCK_ACK is never checked again. Keep the first error: if the host call failed with routed true, set route_ready = false and return that error without the ready wait, and on retry re-check lock ack as well, or state in a comment why a ready PHY makes lock ack irrelevant.
    GREEN would apply to a head with 1 and 2 fixed or answered. The other reviewer's findings 1 and 2 at 401e591 are resolved by this head apart from the cases above.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at a6c54ab: findings (five commits)

I compared v7 with the head I reviewed (401e5916): commits 13, 14 and 15 are identical; the binding example and the driver commit changed.

Fixed, checked

  • Binding example: the two apple,tunable-* lines are out of the example. dtcheck.sh (dtschema 2026.9) on a tree with this head: dt-validate of the board file rc 0 with 0 lines, and 72 of the 73 apple binding examples validate; the one message is display/apple,dcp-typec-routes.yaml, which is in the baseline. The apple,t8152-atcphy example now validates with the full schema set.
  • Restart retry: atc_sw_set() now brings the PHY back for want_mode when the PHY is off and a restart is pending, and keeps the flag on failure. want_mode is set by phy_set_mode(HOST) and by the CIO power_on (after their busy checks) and cleared by power_off, so an orientation event does not power the PHY on by itself. I walked the paths: failed restart then same-orientation event retries; power_off then event does not; a success clears the flag; a failed route after a successful atc_up() sets it again.
  • PIPE_PHY_READY retry: the wait is its own function and atc_route_usb3() repeats it on every call while the route stands, so a second phy_set_mode(HOST) no longer returns success without waiting.

Findings

  1. atc_route_usb3() drops the error of the lock release. t8152_pipehandler_usb3_host() now ends with return t8152_poll_us(h, T8152_PH_LOCK_ACK, ...) after it has set *routed. In atc_route_usb3() the line ret = t8152_pipehandler_usb3_ready(...) overwrites the ret of the host function whenever the mux moved (routed true). A timeout of the final LOCK_ACK wait is therefore reported as success when PIPE_PHY_READY is set, and the override lock may still be held. Before the split the function returned that error. Return it first: if (ret) { atc->route_ready = false; return ret; } after the routed test.
  2. want_mode is withdrawn by the other transport. atc_usb3_power_off() sets want_mode = ATC_OFF before it tests atc->mode != ATC_USB3, and atc_cio_power_off() does the same for CIO. A power_off of the CIO PHY while USB3 is up (or the reverse) clears the request of the transport that is running, and a later failed restart then does not retry. Clear it only when it names this PHY's transport: if (atc->want_mode == ATC_USB3) atc->want_mode = ATC_OFF; and the CIO twin.

Smaller

  • route_ready is written in two places and read nowhere; either use it (skip the repeated wait when it is set and the ready bit still reads set) or drop it.
  • Checkpatch has no errors. The driver commit still has 79 LONG_LINE_COMMENT, 6 LONG_LINE and 4 LONG_LINE_STRING warnings (most in atc-t8152.c, the probe comments); the reply says the lines were wrapped where possible.

Not checked

No compile at this head (I did not reproduce the receipt), no boot, no register sequence against the vendor driver or hardware, no lockdep or sparse run. The first ten commits are #191.

@joshuaswarren

Copy link
Copy Markdown
Author

v8 on the #191 head 9af182774f886303e4c85497549f64498551dd03: new head 3b4c3b7abab669329196bb00108fbb993189ffc1, five commits. git range-diff against v7: only the driver commit changed. Per the findings on a6c54abd (w7X issuecomment-6373881693, w7Q issuecomment-6373870751):

  1. LOCK_ACK error. atc_route_usb3() returns the error of t8152_pipehandler_usb3_host() first when the mux moved but the lock-release wait failed. pipe_usb3 stays recorded and route_ready stays false, the ready wait is not run, and a retry re-runs the host sequence, which checks the lock ack first.
  2. want_mode withdrawal. atc_usb3_power_off() clears it only when it is ATC_USB3, and atc_cio_power_off() only when it is ATC_CIO.
  3. route_ready is read. A second phy_set_mode(HOST) returns early only when the route stands and route_ready is set; otherwise the ready wait runs again. It is cleared at shutdown. (Known nit: the inner if (!pipe_usb3 || !route_ready) is always true after the early return; kept for readability.)
  4. Unrouted PIPE after a restart. The main path of atc_sw_set() routes when routed || mode == ATC_USB3, so a restart after a route whose mux never moved no longer leaves the PHY up with the PIPE on the dummy backend.

Compile check at this head, arm64 clang W=1, 16K pages, PHY_APPLE_T8152_ATC=y: atc-t8152.o, the eight objects of #191 and t8152-j873g.dtb, all rc 0, 0 warnings. dtcheck.sh (dtschema 2026.9): board file 0 messages; the 73 examples print one message, the baseline one. Checkpatch: no errors; warnings as before. These paths are checked by reading and compiled, not run on hardware.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at 3b4c3b7: one item left, then GREEN.

Read: atc-t8152.c and atc-t8152-seq.h at this head against a6c54ab (seq.h identical; the driver diff is the five hunks the author lists).

Resolved:

  • Item 1 of my last review (route forgotten after a failed restart): atc_sw_set now routes on "routed || mode == ATC_USB3". Correct, since atc_up only sets ATC_USB3 for the set_mode path.
  • Item 2 partly: the LOCK_ACK error is now returned before the ready wait and route_ready stays false.
  • want_mode is withdrawn per transport; route_ready is read, set on success, cleared in atc_down.

Left (should fix, three lines): the retry can forget a standing route. t8152_pipehandler_usb3_host sets *routed = false on entry. Its first step releases a lock that is still held (LOCK_ACK set) and returns the poll error if that fails, before touching the mux. After a first call where the mux moved but the lock release timed out, the retry (pipe_usb3 true, route_ready false) runs the host sequence again; if the entry release fails again, routed stays false and atc_route_usb3 does "atc->pipe_usb3 = routed", so pipe_usb3 becomes false while the mux still points at the PHY. atc_down then skips t8152_pipehandler_park and the PIPE stays routed to a powered-down PHY. Keep a standing route:

ret = t8152_pipehandler_usb3_host(&atc->r, &routed);
if (routed)
	atc->pipe_usb3 = true;
if (!atc->pipe_usb3)
	return ret;	/* the mux never moved */
if (ret)
	return ret;

Nit, no change needed: the inner "if (!pipe_usb3 || !route_ready)" is always true after the early return (the author says so too).

GREEN at a head with that fixed, or an answer on why the park can be skipped. Compile and dt check receipts are the author's; not reproduced. Not run: build, boot.

@joshuaswarren

Copy link
Copy Markdown
Author

Review at 3b4c3b7: findings (one left)

I compared v8 with a6c54abd: commits 11, 13, 14 and 15 are identical; only the driver commit changed.

Fixed, checked

  • LOCK_ACK error: atc_route_usb3() returns the error of t8152_pipehandler_usb3_host() when the mux moved and the lock release failed, before the ready wait.
  • want_mode: atc_usb3_power_off() clears it only when it is ATC_USB3, atc_cio_power_off() only when it is ATC_CIO.
  • route_ready is read: the early return needs pipe_usb3 && route_ready.
  • Restart: the main path of atc_sw_set() routes when the route stood or the mode is ATC_USB3.
  • The binding, dwc3 and device tree commits are identical to v7, so the dt check of v7 stands (dtcheck.sh, dtschema 2026.9: board file 0 lines, 72 of 73 examples clean, the baseline message only). I did not re-run it at this head.

Finding

  1. A retry after a failed PIPE_PHY_READY wait re-runs the whole routing sequence, not the wait. After the early return, if (!atc->pipe_usb3 || !atc->route_ready) is true whenever route_ready is false, and route_ready is false after a failed ready wait just as it is after a failed lock release. The retry then calls t8152_pipehandler_usb3_host() on a PIPE that already points at the PHY: it re-requests the lock, rewrites the override words, and clears and sets the mux clock field again (T8152_PH_MUX clock bits to 0, then 1) before it reaches the ready wait. The reply says the ready wait runs again; the code also repeats the mux sequence on a live route. Tell the two cases apart: set a route_locked (or "mux sequence complete") flag when t8152_pipehandler_usb3_host() returns 0, clear it where pipe_usb3 is cleared, and in atc_route_usb3() run the host sequence only when !pipe_usb3 || !route_locked, then the ready wait. With that, a failed lock release still re-runs the host sequence (its lock-ack check is at the top), and a failed ready wait only repeats the wait.

Smaller

  • The inner if (!pipe_usb3 || !route_ready) is always true after the early return (the reply says so); with the fix above the condition becomes meaningful.
  • Checkpatch has no errors; the long-line warnings are as before.

Not checked

No compile at this head (I did not reproduce the receipt), no boot, no register sequence against the vendor driver or hardware. The first ten commits are #191.

Comment thread drivers/pmdomain/apple/pmgr-psd.c Resolved
@joshuaswarren

Copy link
Copy Markdown
Author

GREEN at 017ef3a

Compared the five commits with the ones I cleared at 40c82439: the binding, the device tree commit and the two dwc3 commits are identical apart from hunk offsets (the base changed); the driver commit changed in atc_down() and its callers.

Checked

  • atc_down() returns an error. If the PIPE park fails it keeps pipe_usb3 set and the mode, clears route_ready, logs an error and returns, so a later down retries the park; the PHY is not shut down under a controller that still points at it. On success the flags clear and the shutdown error (if any) is returned.
  • atc_sw_set(): when atc_down() fails the PHY is still up and routed, restart is set and the error is returned; the next orientation event retries through the main path. atc_usb3_power_off() and atc_cio_power_off() return the atc_down() result after clearing only their own want_mode. The remove path logs "the PIPE did not park; the PHY is left up and routed" because nothing retries a park once the driver is gone.
  • The retry cases of atc_route_usb3() (ready wait, lock release, no route standing) are unchanged from 40c82439.
  • dtcheck.sh (dtschema 2026.9) on a tree that carries this PR's binding and device tree together with the other stacked PRs: dtc -@ 0 warning lines, dt-validate of the board file 0 lines, 73 examples with only the baseline message.

Not checked

Compile and checkpatch receipts (the author's), hardware, and the register sequence against the vendor driver. The first ten commits are #191.

@joshuaswarren

Copy link
Copy Markdown
Author

New head e62cb4df089e706bf897330f7ba57ce971735f52: rebased on the new #191 head 9fbb3568 (only the PSD power-on rollback of the base changed); the commits of this PR are unchanged. The stack objects build with rc 0 and 0 warnings (arm64 clang W=1, 16K pages).

@joshuaswarren

Copy link
Copy Markdown
Author

GREEN at e62cb4d

Read: the commits of this head against my last GREEN head d53816b, compared per commit as diffs with the hash lines removed. The five commits of this PR itself (binding, PHY driver, two dwc3 commits, ATC dts) are identical to what I cleared. The ten base commits are the #191 head 9fbb356, where I posted the delta review (PSD power-on rollback includes the failed slot, PSD power-off restores every slot, Gen3 DART drain and unlock changes). Not run: build; the author's compile and dt receipts are not reproduced. Merge still needs CI and a second non-author GREEN.

@joshuaswarren

Copy link
Copy Markdown
Author

GREEN at e62cb4d

The commits above the #191 head are identical to the ones I cleared at 017ef3a5 (compared as diffs with the hash lines removed); the head moved because the #191 base changed (one hunk in the PSD power-on rollback, reviewed in #191). The device tree and binding results at 017ef3a5 stand.

Not checked: compile receipts and hardware. Merge still needs CI and the owner's word.

@joshuaswarren

Copy link
Copy Markdown
Author

Status at head e62cb4df089e706bf897330f7ba57ce971735f52: two independent reviews GREEN at this exact head, all Greptile threads replied and resolved, arm64 clang W=1 compile of the touched objects rc 0 with 0 warnings, device-tree schema check 0 messages. Not done: a boot on hardware, CI. Needs a maintainer's CI approval and merge. Not needed for the release candidate.

@joshuaswarren

Copy link
Copy Markdown
Author

[Joshua's agent] This PR is awaiting maintainer review. Is anything else needed from us?

@joshuaswarren
joshuaswarren force-pushed the m6-phy branch 2 times, most recently from 0ed57c9 to 3043d03 Compare November 9, 2026 22:45
@joshuaswarren

Copy link
Copy Markdown
Author

[Joshua's agent] @maralcbr @scottjones @chriskearney Could one of you review this? It is stacked on #191. I rebased it onto the new #191 head 99f5a2ad67 and integration/aurora at 49a8c15. There were no conflicts. The GCTL.PWRDNSCALE commit had the same dwc3_readl and dwc3_writel link fault as #191, so it now uses readl and writel, and the host selftest mock gained a stub for that helper. The other 4 commits are unchanged. The new head is 3043d037b1. All checks pass on it, including KUnit, compile, package and boot. It is mergeable and waits only for a review.

Review #191 first, then the last five commits here. They add the T8152 Type-C PHY driver, its binding, two small dwc3 changes, and two disabled PHY nodes. Nothing enables the driver on a board. Measured on an M6 Mac mini: every PHY and CIO bring-up write read back with no mismatch, and GCTL.PWRDNSCALE set to 13 cuts the time from RxDetect to Polling on the front port from 1.56-1.69 s to under 90 ms. The SuperSpeed link still does not train, and the PR text says so. It is not needed for the release candidate.

The T8152 GPIO block has the same layout as the T8140 one. Match
"apple,t8152-pinctrl" with the T8140 hardware data.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
T8152 has a SART with a new register layout: the per-entry config,
physical address and size registers sit at 0x00, 0x60 and 0xc0. Add the
operations for it and match "apple,t8152-sart". The binding lists the
new compatible as a stand-alone entry.

Measured on an M6 Mac mini with an earlier form of this change: the
machine boots and runs its root file system from NVMe, which uses this
SART. This version is compile-checked only.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
On T8152 the queue setup registers (admin queue, IO queues, the linear
submission queue control) live at different offsets, and a few of them
accept writes but read back zero. Add a register translation for that
layout, skip the read-back checks that cannot work, and describe the
controller with a new hardware data block matched by
"apple,t8152-nvme-ans2".

The ANS is a virtual PMGR device on this SoC and has no reset line, so
the hardware data block carries a no_reset_control flag. Only T8152 skips
the reset lookup. Every other SoC still requires a reset control and
fails the probe without one, as before.

The binding gets the new compatible. On T8152 the resets, power-domains
and power-domain-names properties are not required, and the controller
has two register regions (the NVMMU registers live in the first one).

Measured on an M6 Mac mini with an earlier form of this change: the
machine boots and runs its root file system from NVMe. This version is
compile-checked only.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
On T8152 the ATC USB blocks (and the GPU) are behind a per-slot service
request in the power state director, on top of the normal PMGR power
states. Expose each slot, or a group of slots that must come up together,
as a generic power domain. Power-on writes the slot request and waits for
the acknowledge; power-off requests state 6 in reverse order.

Add the devicetree binding.

When a power-on request times out, the slot that timed out is put back to idle with the slots before it, because its acknowledge was not seen but it may still start.

When a group power-off fails, every slot is put back to its active state, including a slot whose off request timed out and may still finish stopping, because genpd keeps the domain marked on after a failed power-off and would skip power_on at the next resume.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
The fixed host ports on T8152 need only the USB2 part of the PHY. Add a
driver that applies the bootloader-supplied calibration, handles host and
device mode, and leaves the USB3 and USB4 lanes alone.

Add the devicetree binding.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
A hub with the Apple ID 05ac:800f sits behind the front ports of the M6
Mac mini. Add that ID to the hub quirk table with
HUB_QUIRK_DISABLE_AUTOSUSPEND so the kernel keeps it powered.

The match is on the vendor and product ID alone. Every hub with this ID
loses autosuspend, not only the one on this machine. The constant is
named after the ID, USB_PRODUCT_APPLE_HUB_800F, because the ID is
Apple's and the part behind it was not identified from the descriptor.

This quirk was not measured on its own: it only keeps the hub awake.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Add a peripherals fragment to the M6 Mac mini board file: the always-on
GPIO controller with the reset hog for the front-port USB hub, and the
four PSD service slots for the USB ports.

The NVMe controller, its mailbox and its address filter are not described
here. The NVMe node needs a mailbox driver change that is not in this
tree yet.

The fragment holds only addresses, interrupt numbers and wiring. The boot
loader adds the register tunables that some blocks need. The PMGR power
domains are not described yet, so nodes that need them are left out.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Add the DART of the Apple M6 (T8152), a third-generation DART whose
topology is fixed by the hardware. Its streams are described by static
parameters instead of being probed, and it takes no MMIO access until the
first use, because the block can be power gated behind the ATC USB
domains.

The binding requires a second register region (the TLB invalidate
doorbell) for the T8152 compatible only.

The change adds:
- the apple,t8152-dart compatible and its hardware description;
- static topology and "first use" initialisation of every DART;
- the Gen3 TLB invalidate and the fault handler;
- the reset path for DARTs that the boot loader leaves locked.

A DART that the boot loader leaves locked is held powered for its
lifetime. That applies to Gen3 hardware only. The other generations keep
their behaviour. The first-use initialisation re-reads the parameters and fails if the
page size differs from the one assumed at probe. It runs under a spinlock, and
every function it calls is non-sleeping (register accesses and an atomic
poll).

Measured on an M6 Mac mini with the earlier form of this change: the
display controller DART attaches locked on every boot. This version is
rebased on current integration/aurora and was compile-checked only. The
USB DART paths need the T8152 PSD power domains and were not run on
their own.

The doorbell store and the TLBI of the Gen3 invalidate are arm64 instructions: they are built only with CONFIG_ARM64, and other architectures (COMPILE_TEST) get a stub that returns -EOPNOTSUPP. A locked Gen3 DART no longer takes the range flush through TLB_CMD, which Gen3 does not accept; it uses the full-stream Gen3 callback.

Gen3 does not take commands through TLB_CMD, so the unlock after a lock-on-error exception is not sent there: the DART logs once that this is not supported. The stale-exception drain of a reset does not read 0x704 on Gen3, like the fault handler.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Add the apple,t8152-dwc3 compatible to the Apple dwc3 glue. The ATC ports
of the M6 (T8152) controller never switch roles through the Type-C
controller. The glue starts the controller as a fixed host, with or
without a hub child, or as a fixed peripheral when the node sets
dr_mode = "peripheral".

The T8152 profile is USB2 only: it needs maximum-speed = "high-speed",
takes no reset line, and its USB3 PHY is optional. The calibration lists
come from the device tree (apple,tunable-* properties that the boot loader
fills in at runtime; the binding does not require them in the board file) and are applied to the register windows in a fixed order after
the core is reset. The pipe handler is set up for the selected role.

The apple,t8103-dwc3 paths keep their behaviour. The controller stays a
syscore device while it runs, as before: no separate suspend or resume
path is added for T8152.

The earlier form of this change ran on an M6 Mac mini USB2 front port.
This version is rebased on current integration/aurora and was
compile-checked only. SuperSpeed does not train yet and is not part of
this change.

A fixed peripheral does not require the host-only calibration lists at probe, since it never applies them. The binding lists exactly four iommus entries, two stream IDs on each of two DARTs.

If the T8152 USB2 PHY fails to initialise or power on during core start, the failure goes through the same shutdown as a failed core init, so the reset and syscore flags set earlier are cleared again.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Describe the power management blocks and the USB controllers of the Mac
mini (M6, J873) next to the peripherals fragment:
- the two PMGR blocks and their power domains, in t8152-j873-pmgr.dtsi;
- for each of the four USB ports (usb0, usb1, usb2 and usb3 of the SoC):
  two DART instances, the USB2 PHY and the dwc3 controller.

The register windows, interrupts and DART tunables are the values of the
Apple device tree of the machine. The boot loader adds the apple,tunable-*
lists of the USB2 PHY and the controller at runtime.

The bindings list the apple,t8152-pmgr and apple,t8152-pmgr-pwrstate
compatibles, and the dwc3 and USB2 PHY bindings take the tunable lists as
optional, because the boot loader adds them.

The fragment is a plain description of the hardware. The ports that the
change describes run USB2 only: SuperSpeed and USB4 are not part of it.

This commit compiles to a device tree blob with dtc without warnings. It
was not booted. The bring-up it comes from enumerated the front-port USB2
hub and a USB disk at 480 Mb/s.

The four PSD service slots get their PMGR domains (the aon and the port domain of their port) as parents, so genpd powers them before the slot request.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Describe the SuperSpeed and USB4 side of an ATC port of the Apple M6
(T8152). The PHY serves one transport at a time: USB3 for the dwc3
controller of the port (phandle argument 0) or CIO for the ACIO
controller (phandle argument 1).

The register windows of the dwc3 node (the pipe handler and the
PIPE_PHY_READY word) are not resources of the PHY node, because the dwc3
node already owns them and the platform bus rejects equal or partly
overlapping ranges. The PHY finds them through the apple,dwc3 phandle.

The boot firmware supplies the calibration as apple,tunable-<name>
properties, one per tunable list of the firmware PHY node.

The power domains are the port's USB AON domain (ps_atc*_usb_aon) and its
ATC USB domain (psd_atc*_usb), in that order, as the dts nodes describe
them. The properties stay optional: the driver probes and runs without
them (the platform bus attaches a single domain, a list handles several),
so boards with the PHY tied to one already-enabled domain do not have to
describe it.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Add a driver for the SuperSpeed and USB4 (CIO) side of the ATC ports of
the Apple M6 (T8152). The USB2 side is phy-apple-t8152-usb2.

The driver brings the PHY up for one transport at a time and routes the
dwc3 PIPE to it:

 - PHY 0, USB3: phy_set_mode(HOST) powers the PHY if needed and routes
   the PIPE. There is no power_on because dwc3 would call it before its
   core reset, as with atc.c. power_off shuts the PHY down.
 - PHY 1, CIO: phy_power_on() and phy_power_off() for the Thunderbolt
   driver.

A transport change is a shutdown followed by a full bring-up. The cable
orientation comes from the Type-C switch. Probe writes no register. The
tunable lists come from the device tree, so the driver carries no
calibration values.

Status: the bring-up sequence and the PIPE route to USB3 were run on an
M6 Mac mini (front port) and the CIO bring-up on the rear port. With the
PIPE routed, the SuperSpeed link of the front port reaches RxDetect and
Polling but does not train. The shutdown sequence has not been run on
hardware. The driver is therefore not enabled by any board file in this
series.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
A T8152 port runs USB2-only unless the board describes the SuperSpeed
PHY (apple,t8152-atcphy). Request "usb3-phy" with devm_phy_optional_get()
for the T8152 compatible, so a board without that PHY still probes.
Without the PHY every call on it succeeds, as dwc3_apple_set_role()
expects.

Other compatibles keep requiring the PHY.

The apple,t8152-dwc3 binding gains the matching optional second entry:
phys carries usb2-phy and the optional usb3-phy, so the board and the
driver agree on when the SuperSpeed PHY exists.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
The boot value of GCTL.PWRDNSCALE on T8152 is 1560. The vendor driver
writes 13. With the boot value the SuperSpeed port of the front USB-C
connector waited 1.56 to 1.69 s between RxDetect and Polling. With 13
the port was already in Polling 88 ms after the PHY was routed. This was
measured on the front port only; the other T8152 ports run through the
same function and take the same value.

Write 13 after the core is initialised, before the Apple specific setup.
The write does not make the link train; it changes the delay only.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
Describe the Type-C PHYs of the ATC2 (front) and ATC3 ports of the Mac
mini (J873) with the apple,t8152-atcphy binding and include them from
the board file. Both nodes are disabled: the SuperSpeed link of the
front port is not verified yet, and the ATC3 PHY serves the ACIO
controller, whose nodes are not part of this change. The boot loader
supplies the tunable properties.

Enable a node together with a "usb3-phy" reference in the dwc3 node or a
phys reference in the ACIO node.

Signed-off-by: Joshua Warren <816217+joshuaswarren@users.noreply.github.com>
@joshuaswarren

Copy link
Copy Markdown
Author

[Joshua's agent] The #191 head moved to 6d8665c369 on integration/aurora at bed544b, so I rebased this PR onto it. There were no conflicts, and git range-diff shows all 5 commits unchanged. The new head is d478ff49ef. All checks pass on it, including KUnit, compile, package and boot. This replaces the head named in my earlier comment.

@joshuaswarren

Copy link
Copy Markdown
Author

[Joshua's agent] Status at 2026-11-12 08:13Z: head d478ff4 is 0 commits behind integration/aurora (bed544b) and the checks that apply have passed. The PR still needs one approving CODEOWNER review before it can merge. Can a CODEOWNER take a look? Unvoiced; checked by hand.

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.

1 participant