v1.18 attached (`mei-vsc-hibernate-1.18.tar.gz`). Use this instead of
1.9 and 1.16.

1.16 already survived two S4s by not hanging `OVTI02E1-N` on the post-S4
ACPI i2c client’s secondary (that is the `acpi_dev_pm_detach` 0x8 Oops).
Freeze #1 still WARNed `refcount_t underflow` because we `put_device`’d
the `acpi_get_first_physical_node` ref after `i2c_unregister_device`.
ACPI unbind already dropped that ref.

1.18: unregister the ACPI i2c client and do **not** put that lookup ref.
If that unregister ran, skip the swnode find (same device on cold boot).
Swnode-client freeze is still put-then-unregister (the
`i2c_find_device_by_fwnode` get is extra).

Verified 2026-09-07, same boot, overlay `mei-vsc-hibernate/1.18`, kernel
7.0.0-31-generic:

- S4 #1 00:40 → 00:42: `unregister ACPI i2c i2c-OVTI02E1:00` with **no** 
underflow WARN → `swnode i2c client 18-0010 companion=none`; ov02e10 bound.
- S4 #2 00:45 → 00:48: `unregister swnode i2c 18-0010 before S4` (no WARN) → 
same `companion=none`; ov02e10 bound; ISYS/PSYS/CSI up; no Oops.

Pass/fail is still ov02e10 + ivsc_csi + IPU6 isys/psys in sysfs. A USB
webcam is not a pass. First I2C `-5` then bind ~12s after resume is
expected (USB LJCA).

ov02e10 still needs `i2c_device_id` / `alias: i2c:ov02e10` (patch in the
tarball).

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2166564

Title:
  IPU6/IVSC camera dead after hibernate (S4): mei-csi recreated without
  fwnode; ov02e10 unbound (Precision 5690)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-hwe-7.0/+bug/2166564/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to