Public bug reported: ## Summary On a Lenovo ThinkPad P16s Gen 1 AMD, resuming from s2idle intermittently deadlocks: the machine enters `s2idle` (typically on lid-close) and never resumes. The screen stays black and unresponsive, no VT is reachable, and the kernel logs nothing after the suspend-entry line. Only a forced power-off recovers. The **first** suspend/resume after boot succeeds; a **later** sleep is what hangs, so it presents as intermittent.
## System - Model: Lenovo ThinkPad P16s Gen 1 AMD - CPU/APU: AMD Ryzen (Rembrandt); iGPU Radeon 680M, amdgpu DCN 3.1, PCI 0000:04:00.0, device id 1002:1681 - BIOS: R23ET87W (ver 1.63), EC R23HT68W *(last available, not downgradable, potentially related to 1.61 change originally meant for Windows sleep issue)* - OS: Ubuntu 22.04, HWE kernel **6.8.0-134-generic** (6.8.0-134.134~22.04.1, published ~2026-07-09) - Last-known-good kernel: 6.8.0-124-generic - Desktop: Cinnamon (X11); home directory on eCryptfs - Sleep type: **s2idle only** (`/sys/power/mem_sleep` = `[s2idle]`, no `deep`) ## Symptom / impact - Enters s2idle, never resumes. Black, unresponsive display. - `Ctrl+Alt+F3` gives at most a blinking cursor (no usable VT). - No kernel panic/oops; `/sys/fs/pstore/` is empty. - `/sys/power/suspend_stats/fail` = 0 (PM core does not record a failed suspend — the hang is on the resume path, not a rejected suspend). - Requires forced power-off. ## Reproduction 1. Boot 6.8.0-134. 2. Suspend once (lid close) and resume — works. 3. Use the machine, then suspend again (lid close) later. 4. On some later cycle, resume hangs hard. Frequency is intermittent; lid-close-triggered s2idle observed in the failing case. ## Evidence (from persistent journal, boot -1, kernel 6.8.0-134) Good cycle: ``` 17:37:27 kernel: PM: suspend entry (s2idle) 17:37:28 kernel: amd_pmc AMDI0007:00: Last suspend didn't reach deepest state 17:37:28 kernel: amdgpu 0000:04:00.0: amdgpu: SMU is resumed successfully! 17:37:28 kernel: PM: suspend exit ``` Failing cycle (triggered by lid close) — note total silence after entry: ``` 17:49:27 systemd-logind[1428]: Lid closed. 17:49:30 systemd[1]: Reached target Sleep. 17:49:30 kernel: PM: suspend entry (s2idle) 17:49:30 systemd-sleep[124948]: Entering sleep state 'suspend'... <no further log entries — machine hung on resume, forced power-off> ``` - Chronic on every cycle (incl. good ones): `amd_pmc AMDI0007:00: Last suspend didn't reach deepest state` — s0i3 is not being reached. - Boot log shows PSR enabled: `[drm] PSR support 1, DC PSR ver 0, sink PSR ver 3 DPCD caps 0x7b su_y_granularity 4`. ## Ruled out - **Userspace/lock screen**: early in a *different* (recoverable) resume, a VT login prompt was reachable and PAM/fingerprint/eCryptfs errors appeared — but in the fatal case the kernel itself logs nothing after suspend entry, so the hang is in the kernel/firmware resume path, not the greeter. - **PM core rejecting suspend**: `suspend_stats/fail` = 0. ## Confounders (important) Two things changed in roughly the same window and must be disentangled: 1. Kernel upgraded from 6.8.0-124 to 6.8.0-134. 2. System BIOS updated (~two weeks ago), from 1.53 to 1.63. See 1.61 change for Windows in https://download.lenovo.com/pccbbs/mobiles/r23uj85w.html "- Fixed an issue that 10 minutes after system enters modern standby, the system will may hang up when resume." A BIOS-update-induced s2idle resume regression on Lenovo AMD ThinkPads is a known current pattern (e.g. T14 Gen 6 AMD after BIOS 1.17). Therefore the kernel s *not yet proven* to be the cause. The A/B test below is required. Source: https://github.com/fwupd/firmware-lenovo/issues/562 Looked https://bugs.launchpad.net/ubuntu/+source/linux- oem-6.17/+bug/2148538 but that is for kernel 6.17. ** Affects: linux-hwe-6.8 (Ubuntu) Importance: Undecided Status: New ** Package changed: linux-oem-6.17 (Ubuntu) => linux-hwe-6.8 (Ubuntu) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2160878 Title: s2idle resume hang on ThinkPad P16s Gen 1 AMD (kernel 6.8.0-134) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-hwe-6.8/+bug/2160878/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
