Skip to content

J815: slow the CPUs on the hottest SoC die reading - #1908

Open
m0d16l14n1 wants to merge 3 commits into
aurora-silicon:integration/aurorafrom
m0d16l14n1:j815/t8142-j815-thermal
Open

m0d16l14n1 wants to merge 3 commits into
aurora-silicon:integration/aurorafrom
m0d16l14n1:j815/t8142-j815-thermal

Conversation

@m0d16l14n1

@m0d16l14n1 m0d16l14n1 commented Nov 14, 2026 •

Copy link
Copy Markdown
Contributor

Rebased onto integration/aurora 0abc2f3 now that #1532 (the full J815 device tree, with the J815's SMC die-key list) has merged: the branch is this PR's three commits. The earlier dt-checks failure came from the #1532 merge commit (t6030 ALS compatibles), not from these patches. The trip values are still estimates, not taken from macOS yet (below).

The J815 has no fan and no thermal zone with trips: its SoC die zone (the hottest of the 26 apple,soc-die-keys) only reports.

Change

  1. dt-bindings, hwmon (macsmc): a node with apple,discover-temperatures has no stable channel indices, so a DT zone can't name one of its channels. When such a node is a thermal sensor provider, the hottest-die reading is sensor 1000, above any channel index the hwmon core registers. A zone on it replaces the tripless macsmc_soc_die zone; without one the driver registers that zone as before.
  2. t8142-j815: one soc-die-thermal zone on sensor 1000 with the power allocator and the clusters' energy model (arm64: dts: apple: T8142 Airs: CPU topology, capacity and energy model #1537): switch-on at 60 °C, control at 100 °C, critical at 115 °C, both CPU clusters as cooling devices (E weighted 3:1, as on J700 arm64: dts: apple: t8140-j700: control the CPU zones at 105 C and 4.2 W (+8-9% sustained, under macOS limits) #1043).

Sustainable power, 9000 mW, is an estimate. The J815 ADT's clpc pkg-sustainable-power-target is 12.5 W (0xc8000, 16.16). The J700's is 5.5 W, and macOS holds the J700's CPUs at 4.1–4.3 W, about 76%. The same share of 12.5 W gives 9 W. (The CPU SPL key isn't in mW: the J700's 0x880 against its measured 4.2 W.) The 100 °C control point is below the J700's 105 °C until a macOS soak on a J815 gives its own value.

Tested on J815 (MacBook Air 15" M5), macOS 27.2 stub, 3 min soak with every core busy (2 s samples):

  • Zone: policy=power_allocator, sustainable=9000, trips 60000/100000/115000, cooling devices on both clusters.
  • Die 83 °C at start → peak 99.4 °C, settling at ~98.7–99.1 °C. P cluster throttled from 4464 to 2508 MHz (cooling state 13), E cluster kept at 2964 MHz. No critical trip.

CHECK_DTBS (apple,smc-hwmon, thermal-zones) on t8142-j815/j813/j704: nothing new. checkpatch --strict: clean per patch.

Part of the J815 bring-up: board results are summarised in discussion #135.

Assisted-by: Claude Opus 5.5

@m0d16l14n1

Copy link
Copy Markdown
Contributor Author

@DjDeveloperr out of draft, parent-first on #1532 (alias fix merged in). J815 soak: peak 99.4 °C, settles at about 99 °C with the P cluster held at 2508 MHz by the IPA governor; trips 60/100/115 °C. Please review.

Assisted-by: Claude Opus 5.5

A node with apple,discover-temperatures has no stable channel indices, so
a thermal zone cannot name one of its temperature channels. Let such a
node be a thermal sensor provider for one sensor, 1000: the hottest of
its apple,soc-die-keys readings. A zone on it replaces the driver's
tripless macsmc_soc_die zone.

Assisted-by: Claude Opus 5.5
Signed-off-by: Nick <nekreshen@yandex.ru>
A board that discovers its SMC temperatures (apple,discover-temperatures,
as the M5 Airs do) has no stable channel indices, so a device tree thermal
zone cannot name one of its temperature channels, and the hottest-die
zone is registered without trips. Such a board then has no thermal
policy of its own.

When the node is a thermal sensor provider, offer the hottest-die reading
as sensor 1000, above any channel index the hwmon core registers. A zone
on it gets the board's trips and cooling maps and replaces the tripless
macsmc_soc_die zone; without one the driver registers that zone as before.

Assisted-by: Claude Opus 5.5
Signed-off-by: Nick <nekreshen@yandex.ru>
The J815 has no fan and no thermal zone with trips: its SoC die zone
(the hottest of the 26 apple,soc-die-keys) only reports. Make the SMC
hwmon node a thermal sensor provider and give the board one zone on the
die sensor, with the power allocator and the clusters' energy model:
switch-on at 60 C, control at 100 C, critical at 115 C, and both CPU
clusters as cooling devices (the E cluster weighted 3:1, as on J700).

Sustainable power is an estimate, 9000 mW: the J815 ADT's clpc
pkg-sustainable-power-target is 12.5 W against the J700's 5.5 W, where
macOS holds the J700 CPUs at 4.1-4.3 W (about 76%). The control point is
below the J700's 105 C until a J815 soak gives its own value.

Assisted-by: Claude Opus 5.5
Signed-off-by: Nick <nekreshen@yandex.ru>
@m0d16l14n1
m0d16l14n1 force-pushed the j815/t8142-j815-thermal branch from 83a0fe3 to 8d5e6d5 Compare November 19, 2026 07:27
@m0d16l14n1 m0d16l14n1 changed the title J815: slow the CPUs on the hottest SoC die reading (on #1532) J815: slow the CPUs on the hottest SoC die reading 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.

1 participant