Re: [REGRESSION] ACPI processor/cpuidle change in 6.18.52 breaks bare-metal Xen dom0 boot

[Support TRINITY]

I have now tested Linux 6.18.52 with commit:

0089ce1c056a ("ACPI: processor: Update cpuidle driver check in __acpi_processor_start()") applied on the affected Alpine Xen dom0 bare-metal host.

Result: it boots successfully.

Xen dom0 comes up correctly, /proc/xen is mounted, xenstored starts, SSH
is reachable, and xl info, xl list, etc... work as expected.

So this confirms that 0089ce1c056a fixes the regression on this affected
Alpine system.

Thank you Rafael for identifying the missing stable dependency.

Regards,
Tony


-----Message original-----
De: Sasha Levin <sashal@xxxxxxxxxx>
à: Rafael J. Wysocki (Intel) <rafael@xxxxxxxxxx>
Cc: Sasha Levin <sashal@xxxxxxxxxx>; Marek Marczykowski-Górecki <marmarek@xxxxxxxxxxxxxxxxxxxxxx>; regressions <regressions@xxxxxxxxxxxxxxx>; linux-acpi <linux-acpi@xxxxxxxxxxxxxxx>; linux-pm <linux-pm@xxxxxxxxxxxxxxx>; xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>; linux-kernel <linux-kernel@xxxxxxxxxxxxxxx>; Stable <stable@xxxxxxxxxxxxxxx>; Support TRINITY <support@xxxxxxxxxxxxxxx>
Envoyé: mardi 22 septembre 2026 04:23 CEST
Sujet : Re: [REGRESSION] ACPI processor/cpuidle change in 6.18.52 breaks bare-metal Xen dom0 boot


> I will try testing 6.18.52 with commit 0089ce1c056a ("ACPI: processor:
> Update cpuidle driver check in __acpi_processor_start()") applied,
> instead of using my lifecycle revert.

Queued for 6.18, thanks.

Please still post the result of that test once you have it.

-- 
Thanks,
Sasha