Skip to content

drm/asahi: g17: keep the commit-stop raise to the T8142 - #2521

Merged
DjDeveloperr merged 1 commit into
integration/aurorafrom
djdev/g17s-no-commit-raise
Nov 20, 2026
Merged

DjDeveloperr merged 1 commit into
integration/aurorafrom
djdev/g17s-no-commit-raise

Conversation

@DjDeveloperr

@DjDeveloperr DjDeveloperr commented Nov 20, 2026 •

Copy link
Copy Markdown
Contributor

Keeps the G17 commit-stop domain raise (f5b8062, narrowed by #2469) to the T8142. One line in asahi_g17_commit_raises() plus its comment.

What it changes:

  • M5 boards (J714s, J716c), behaviour: nothing. They stop the GPU through their own M5 commit stop (asahi_g17_m5_commit_stop, picked by asahi_g17_m5_provider()). That stop never ran the raise and is byte-identical to the M5 default kernel's.
  • M5 boards, the log: with the apple,agx-g17s entry, asahi_g17_commit_finish() printed GPU domain at the commit stop: 0x0, raise to full 0 there, from fields the M5 stop never fills. That line read as a raise that never happened. It goes away.
  • The other G17S boards (K114c, K116s): they take the generic stop, where the raise was never measured, and go back to the stop their sleep was tested with.
  • T8142: unchanged; it keeps the raise on the g17 and g16 paths.

Correction to the first version of this PR: it blamed an M5 hang after a deep sleep (the ship gate's intg28, 2026-11-19 23:49) on the raise. An independent review showed the raise never ran on the J714s, so this PR does not address that hang; it is still open and being bisected in the ship lane.

It is in the M5 ship candidate (7.3.0-m5-intg28c) because it is harmless there.

🤖 Generated with Claude Code

#2469 kept the GPU domain raise at the sleep commit's stop on the
T8142 and on every apple,agx-g17s part. The G17S M5 boards (J714s,
J716c) never run it: they stop the GPU through their own M5 commit
stop, which this doesn't touch. On them the flag only made
asahi_g17_commit_finish() print "GPU domain at the commit stop: 0x0,
raise to full 0" from fields the M5 stop never fills, which reads as a
raise that didn't happen.

The other G17S boards (K114c, K116s) take the generic stop, where the
raise was never measured; their sleep was tested before it came in.
Keep the raise to the T8142, the SoC it was made for (the J815's GPU
gates itself between frames and its registers may not answer below full
power): the other G17S boards go back to the stop they were tested
with, and the M5 boards stop printing the empty line.

Signed-off-by: Diljit Singh <hey@djdev.me>
@DjDeveloperr
DjDeveloperr force-pushed the djdev/g17s-no-commit-raise branch from 3decff5 to 10db28c Compare November 20, 2026 05:23
@DjDeveloperr DjDeveloperr changed the title drm/asahi: g17: don't raise the G17S domain at the commit either drm/asahi: g17: keep the commit-stop raise to the T8142 Nov 20, 2026
@DjDeveloperr
DjDeveloperr merged commit e0ee46a into integration/aurora Nov 20, 2026
1 check 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.

1 participant