Windows 11 Arm64 24H2 boot failure on heterogeneous ARM SoC when small cores are enabled

Vadim Grinshpan 20 Reputation points
2026-06-28T09:41:43.7333333+00:00

Hello Microsoft Support,

We are investigating a Windows 11 Arm64 boot issue on a CIX CP8180-based platform with a heterogeneous ARM DynamIQ CPU topology.

With Windows 11 Arm64 24H2, the system fails to boot(PHASE0_EXCEPTION) when the small cores are enabled. The current firmware workaround is to disable the small cores, after which Windows boots successfully. This appears to be a regression or compatibility issue in Windows 11 24H2, because disabling part of the CPU topology is not an acceptable long-term solution.

We would like to confirm whether Microsoft is already aware of this issue and whether a fix is expected in a future Windows 11 Arm64 servicing update, specifically 25H2 or 26H2.

Related public references:

Questions:

  1. Does Windows 11 Arm64 24H2 require PMU-related registers such as PMCEID0_EL0 / PMCR_EL0 to be uniform across all CPU cores?
  2. Is a heterogeneous ARM topology with different PMU capabilities across big and small cores expected to be supported?
  3. Is this a known Windows 11 Arm64 24H2 regression?
  4. Is a fix planned for Windows 11 Arm64 25H2, 26H2, or a cumulative update?
  5. Are there any firmware or ACPI table requirements we should verify or adjust to avoid this boot failure while keeping all cores enabled?

We can provide firmware details, ACPI tables, boot logs, and reproduction steps if needed.

Thanks,

Vadim

Windows for business | Windows Client for IT Pros | Performance | Other
0 comments
Answer accepted by question author
Jason Nguyen Tran 26,650 Reputation points Independent Advisor
2026-06-30T07:33:48.8733333+00:00

Hi Vadim Grinshpan,

Based on what you’ve described, the PHASE0_EXCEPTION error on heterogeneous ARM SoCs with Windows 11 Arm64 24H2 does appear to be linked to how the OS handles PMU register consistency across big and small cores. Windows currently expects uniform PMU capabilities (such as PMCEID0_EL0 and PMCR_EL0) across all cores, and mismatches can trigger early boot failures.

To your specific questions: heterogeneous ARM topologies are intended to be supported, but differences in PMU registers can cause regressions in certain builds. Microsoft engineering teams are aware of this class of issue, and fixes are being evaluated for upcoming servicing updates. While I can’t commit to a specific release (25H2 vs. 26H2), the expectation is that future builds will improve compatibility with mixed-core PMU reporting.

In the meantime, firmware and ACPI table validation is critical. Please ensure that ACPI tables accurately describe the CPU topology and that PMU-related registers are exposed consistently. Providing your firmware details, ACPI tables, and boot logs will help us reproduce and confirm whether this is the same regression already under investigation.

I hope the response provided some helpful insight. If you find this answer useful, please hit “accept answer” so I know it addressed your concern.

Jason.

Was this answer helpful?

2 people found this answer helpful.
0 comments

0 additional answers

Sort by: Oldest
Most helpful Newest Oldest

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.