Re: [REGRESSION] ACPI battery charging state regression on GLK MRD / Gemini Lake platform after bb1256e0ddc7

[Rafael J. Wysocki]

Hi,

On Mon, Jun 8, 2026 at 5:36 PM goldi fiche <goldifiche@xxxxxxxxxx> wrote:
>
> Hi,
>
> by "correctly" I mean the mismatch between observed charging behavior and reported power state.
>
> On my system:
> - AC is connected
> - battery percentage increases over time
> - energy-rate remains 0 W (no charging current reported)
> - state stays "pending-charge"
>
> So the battery appears to be charging (capacity increases), but the system reports no charging current and keeps an intermediate "pending-charge" state instead of clearly reporting "charging".
>
> Additionally, I observe that between ~90% and 100% the system correctly reports "charging" in both kernel and GNOME UPower, including a non-zero energy-rate. This suggests that the issue is state-dependent and may be related to firmware reporting behavior at lower battery levels.
>
> The inconsistency is between actual charging progress and the reported power_supply state at low to mid charge levels.

So I'm wondering if you have been able to test the patch at:

https://lore.kernel.org/linux-acpi/5111702.31r3eYUQgx@rafael.j.wysocki/