Repository navigation
J813 S2R: SEP setup, PMP shutdown and GPIO IRQ recovery - #4
Merged
m0d16l14n1 merged 5 commits intoNov 16, 2026
Merged
Conversation
Describe the J813 SEP, mailbox and power domains, publish its alias for aura's firmware/manifests handoff, and admit J813 to the T8142 profile. Leave sep_dart disabled until the SEP module enables it. Describe the ADT's SID0 DMA aperture directly with no DART offset. The profile fallback gives the DMA API a low address and adds BIT(40) only to the page-table mapping; the SEP consumed that registration without replying. The explicit high window matches the existing J813 Touch ID work (7003518) and the macOS s05 registration trace. On J813, fresh attach with xart_writes=0 provision_keybag=0 trusted_keys=0 filevault_api=0 changes from zero replies at IOVA 0xfffc0000 to all 12 endpoints ready at IOVA 0x100fff80000. The key store opens and hwrng registers. The sleep dry run passes with eight coprocessors; a real commit still resets with no aura wake marker. This is sleep bring-up, not Touch ID or suspend/resume qualification. Signed-off-by: Thomas Glanzmann <thomas@glanzmann.de>
…commit Record the SEP power word in PMU 0x8310..13 and the seven forced-on AON actual-state nibbles in 0x8314..17. The writes precede the complex power words and their 33 ms deadline. On J813 the SEP reads 0x1000270f, matching macOS's gated state, while all seven AON words read actual state 4. The real commit still resets with PMU note 0x68 and no aura wake marker. Signed-off-by: Thomas Glanzmann <thomas@glanzmann.de>
Do not apply T8132's forced state-4 audio list on T8142. All seven J813 domains are no_ps in the ADT; the captured macOS AP never accesses them across sleep/wake, and the stopped snapshot has them at state 0. Keep the test telemetry reading the same words independently of the match list so this experiment can be compared with the previous run. Hardware A/B on J813: SEP reaches 0x1000270f in both runs. The seven AON actual-state nibbles change from 0x04444444 to 0x0f00f004 (low nibble first: 4,0,0,f,0,0,f). Both real commits reset at PMU note 0x68 with no aura wake marker. This removes an unsupported inherited write sequence, but neither fixes deep sleep nor reproduces macOS's all-off audio state. The audio shutdown still needs investigation. Signed-off-by: Thomas Glanzmann <thomas@glanzmann.de>
The PMP firmware was idled and its RUN bit cleared, but its always-on power domain stayed at target f. macOS lowers the AP-owned PMP domain after the final idle and SRAM backup. PMS_SRAM then gates automatically while retaining target f. Do the same before the final memory and CPU sequence. On the combined J813 test stack this permits a real physical sleep and return, with the boot ID preserved and /sys/power/state returning 0. Device recovery and the other local preparation changes are still under investigation. Signed-off-by: Thomas Glanzmann <thomas@glanzmann.de>
After a J813 deep-sleep wake, headphone GPIO 100 can be in level-low interrupt mode while its Linux IRQ is disabled and marked masked. The generic mask_ack_irq() fallback skips irq_mask() in that state, leaving the asserted input to repeatedly trigger AP GPIO parent IRQ 347. CPU0 can stay in IRQ dispatch and starve I2C completions and resume timers. Provide irq_mask_ack so the pin is always masked in hardware before acknowledging it. A clean-module hardware repeat completed parallel device resume, with CS42L84 and AOP resume succeeding, without the diagnostic interrupt or polling used to investigate the stall. Signed-off-by: Thomas Glanzmann <thomas@glanzmann.de>
m0d16l14n1
marked this pull request as ready for review
November 16, 2026 10:57
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
J813 completes a first real deep-sleep/wake on the experimental test stack after powering off PMP and correcting GPIO interrupt handling. PMP previously stayed powered at the final commit; a headphone GPIO interrupt storm could then hold CPU0 in IRQ dispatch during resume.
This PR was merged at commit
2f16ec945349on 2026-11-16 at 10:57:59 UTC. Its changes are SEP setup/high DMA-window registration, PMP shutdown, GPIO mask-and-ack, and explicitly test-only SEP/AON measurements and audio power policy.Hardware validation of the clean GPIO fix at 10:48:23 UTC returned success with an unchanged boot ID, all ten CPUs online and a successful CPU0 task. Codec resume took 27 ms and AOP 3 ms with normal audio and parallel device resume.
The tests required the combined J813 test stack, including external-DCP support and manual shutdown of inactive SIO. This merged PR alone was not shown to resume. Wi-Fi/Bluetooth abort and repeated sleep remained unresolved at merge time.
Subsequent external-DCP retention work is on the experimental kernel branch and will be submitted as a separate follow-up. Those later commits are not part of this merged PR.