https://bugzilla.kernel.org/show_bug.cgi?id=221179

            Bug ID: 221179
           Summary: [Lenovo IdeaPad Pro 5 14AHP9] ACPI BIOS Error on every
                    resume: AE_NOT_FOUND [^^^GPP6.RTKW] aborts
                    \_SB.PEP._DSM; EC mutex contention AE_AML_NOT_OWNER
                    [ECMT] after long suspend
           Product: ACPI
           Version: 2.5
          Hardware: AMD
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P3
         Component: EC
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Created attachment 309562
  --> https://bugzilla.kernel.org/attachment.cgi?id=309562&action=edit
ACPI tables from Lenovo IdeaPad Pro 5 14AHP9 (BIOS NBCN29WW)

Hardware:
  Model:        Lenovo IdeaPad Pro 5 14AHP9 (Xiaoxin Pro 14 2024 AMD)
  BIOS:         BIOS: NBCN29WW (updated from NBCN25WW, issue persists)
  CPU:          AMD Ryzen AI (Strix Point / Hawk Point)
  Kernel:       6.18.13-200.fc43.x86_64
  OS:           Fedora Silverblue 43
  Suspend type: s2idle


== Bug 1: Unresolved symbol [^^^GPP6.RTKW] on every resume ==

Reproducible 100% of the time on every resume from s2idle.

Kernel log:

  ACPI: EC: interrupt blocked
  ACPI BIOS Error (bug): Could not resolve symbol [^^^GPP6.RTKW], AE_NOT_FOUND
(20250807/psargs-332)
  ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0.UPHK due to previous error
(AE_NOT_FOUND) (20250807/psparse-529)
  ACPI Error: Aborting method \_SB.PEP._DSM due to previous error
(AE_NOT_FOUND) (20250807/psparse-529)
  ACPI: EC: interrupt unblocked

The DSDT method \_SB.PCI0.LPC0.EC0.UPHK references ^^^GPP6.RTKW, which does not
exist in the ACPI namespace. This causes \_SB.PEP._DSM (AMD Platform Extension
Plugin DSM) to abort on every resume, potentially degrading platform idle and
power management.


== Bug 2: EC mutex [ECMT] ownership error after long suspend (~8+ hours) ==

Appears only after longer suspend durations.

Kernel log:

  ACPI Error: Thread 3534585856 cannot release Mutex [ECMT] acquired by thread
2230648832 (20250807/exmutex-378)
  ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0.ECCC due to previous error
(AE_AML_NOT_OWNER) (20250807/psparse-529)
  ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0.RPAM due to previous error
(AE_AML_NOT_OWNER) (20250807/psparse-529)
  ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0._Q96 due to previous error
(AE_AML_NOT_OWNER) (20250807/psparse-529)
  ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0._Q93 due to previous error
(AE_AML_NOT_OWNER) (20250807/psparse-529)

EC query handlers _Q93 / _Q96 (likely triggered by hardware events queued
during
sleep) attempt to release ECMT but it is held by a different AML thread,
indicating a mutex ownership imbalance in the DSDT EC query dispatch path.


== Steps to reproduce ==

Bug 1: Resume from any s2idle suspend. 100% reproducible.
Bug 2: Suspend for ~8 hours or more, then resume.


== Notes ==

ACPI dump available on request (sudo acpidump > acpidump.dat).

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

_______________________________________________
acpi-bugzilla mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla

Reply via email to