Skip to content

thunderbolt: apple: T6050: J716c USB4 router sleeps in place - #2126

Merged
DjDeveloperr merged 2 commits into
integration/aurorafrom
j716c/thunderbolt-sleep
Nov 17, 2026
Merged

DjDeveloperr merged 2 commits into
integration/aurorafrom
j716c/thunderbolt-sleep

Conversation

@erikwb

@erikwb erikwb commented Nov 17, 2026 •

Copy link
Copy Markdown
Contributor

What this does

The J716c's right-port USB4 router (acio2, as on the J714s) sleeps in place with a dock attached,
with the J714s's rules: the USB host's binding to the router, the router's CIO frames and power
roles in the suspend, the tunnel's domain after a refused sleep, the router's sleep teardown, its
own domain (ATC2_CIO) taken down at the entry and raised after the PHY at the wake, LSTX held
high through the sleep, and no retimer scan on the dock's link. The left ports are powered off for
the sleep. The J714s's diagnostics (link state and adapter dumps, domain guard logs) stay J714s
only.

Tested

  • thunderbolt: apple: T6050: J716c USB4 router sleeps in place #2126 (j716c/thunderbolt-sleep) builds on its own base (Image.gz, modules and dtbs, arm64, LLVM).
  • On the J716c run above the dock's router came back in place after the wake, with its Ethernet up.
  • All 17 branches merged on integration/aurora were built as one kernel and run on a J716c (16" M5 Max): one suspend-to-RAM with a USB4 dock attached, about 30 s asleep, woken by the RTC alarm (wake source 76). After the wake all 18 CPUs were online, the dock's router came back in place with its Ethernet up, and the SEP answered RNG reads. The sleep's log showed "secondary clusters down first: cores 6+6, cluster 1 0x1a in 353 us, cluster 2 0x1c01a in 448 us". The PMP was not running in that build.

Dependencies

Branch #2126 (j716c/thunderbolt-sleep). Based on integration/aurora; no dependencies at build time. At run time the J716c deep sleep needs #2119 (j716c/sleep-j716c-profile) and #2123 (j716c/dts-deep-sleep).

Part of the J716c deep sleep set: #2118 (j716c/spin-table-release-cpus), #2119 (j716c/sleep-j716c-profile), #2120 (j716c/sleep-wake-source-names), #2121 (j716c/sleep-sep-auto-wake), #2122 (j716c/sep-t6050-profile), #2123 (j716c/dts-deep-sleep), #2124 (j716c/drm-apple-sleep), #2125 (j716c/dart-sleep), #2126 (j716c/thunderbolt-sleep), #2127 (j716c/usb-sleep), #2128 (j716c/phy-sleep), #2129 (j716c/pcie-sleep), #2130 (j716c/aic-sleep), #2131 (j716c/asahi-sleep), #2132 (j716c/spmi-sleep), #2133 (j716c/macsmc-sleep), #2134 (j716c/nvme-sleep).

🤖 Generated with Claude Code

… on the J714s

The J716c has the J714s's T6050 USB4 router, NHI and tunnelled PCIe,
and its right port is acio2 as on the J714s. Apply the J714s's rules
for a router that sleeps in place with a dock attached to the J716c
too: the USB host's binding to the router, the router's CIO frames and
power roles in the suspend, the tunnel's domain after a refused sleep,
the retimer scan that the dock's link does not take, and the router's
sleep teardown. The left ports are powered off for the sleep. The
J714s-only diagnostics stay J714s only.

Developed with Claude Code.

Signed-off-by: Erik Bourget <erik@ebourget.net>
…old as on the J714s

The J716c's USB4 router sleeps in place the way the J714s's does, and
needs the same two rules at its power-off:

- the router's own domain (ATC2_CIO), held on by its DART, is taken
  down at the entry and raised again only after the PHY at the wake, so
  that a dock's link trains in place;
- LSTX is held high through the sleep, so that an attached dock's CIO
  wake source does not assert right after the commit.

Apply both on the J716c too.

Developed with Claude Code.

Signed-off-by: Erik Bourget <erik@ebourget.net>
@erikwb
erikwb force-pushed the j716c/thunderbolt-sleep branch from 8742e6d to 1d2639d Compare November 17, 2026 15:09
@DjDeveloperr
DjDeveloperr marked this pull request as ready for review November 17, 2026 19:13
@DjDeveloperr
DjDeveloperr merged commit 7a49ade into integration/aurora Nov 17, 2026
1 check passed
@RyanTheTide
RyanTheTide deleted the j716c/thunderbolt-sleep branch November 17, 2026 23:53
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