A sudden WHEA_UNCORRECTABLE_ERROR 0x124 stop code indicates that the Windows Hardware Error Architecture has intercepted a fatal physical subsystem failure before corrupt data could compromise the operating system kernel. Unlike software-driven Blue Screens of Death caused by null pointer dereferences or unhandled driver exceptions, bug check 0x124 is triggered directly by hardware reporting mechanisms built into modern microprocessors and chipsets. When this kernel panic occurs, the operating system halts all I/O execution immediately to protect data integrity, dumping telemetry into physical memory before rebooting. Resolving this critical condition requires structured diagnostic analysis rather than blind component swapping or generic registry cleaning.
For systems engineers and infrastructure administrators managing enterprise workstations or server environments, analyzing the raw crash artifact is the only deterministic path to root-cause isolation. Within the enterprise ecosystem, unresolved hardware halts degrade operational reliability across local services and networked environments cataloged under our Windows Errors & BSOD troubleshooting hub. By extracting the Common Platform Error Record (CPER) binary structure through Microsoft WinDbg, you can identify the exact PCIe root port, memory controller channel, or processor execution core that initiated the hardware alert.
Architectural Breakdown of WHEA_UNCORRECTABLE_ERROR 0x124
The Windows Hardware Error Architecture (WHEA) serves as the centralized abstraction layer between physical hardware error reporting mechanisms and the Windows NT kernel. When an electrical anomaly, bus parity failure, or physical component timeout occurs, hardware error sources notify the operating system via standard reporting pathways. These mechanisms include the processor Machine Check Architecture (MCA), PCI Express Advanced Error Reporting (AER), and system firmware error records defined by ACPI specifications.
When the kernel receives an uncorrected error notification, it invokes KeBugCheckEx with parameter 0x124. The operating system attaches four architecture-specific arguments to the crash stack frame, as documented in the official Microsoft Learn Bug Check 0x124 WHEA_UNCORRECTABLE_ERROR documentation. Parameter 1 determines the error source type, while Parameter 2 stores the virtual 64-bit memory address pointing directly to the internal _WHEA_ERROR_RECORD payload.
| Bug Check Parameter | Hardware Designation | Architectural Significance |
|---|---|---|
| Arg1: 0x0000000000000000 | Machine Check Exception (MCA) | Direct CPU core hardware error, L1/L2/L3 cache parity fault, or bus interconnect timeout. |
| Arg1: 0x0000000000000004 | PCI Express Error (AER) | PCIe root port, switch, NVMe controller, or discrete GPU link training/TLP packet failure. |
| Arg1: 0x0000000000000005 | NMI / Generic Error | Non-Maskable Interrupt triggered by motherboard sensor telemetry or legacy system timer. |
| Arg1: 0x0000000000000007 | Boot Error Record | Hardware state inconsistency captured during early platform firmware boot initialization. |
Configuring and Extracting Kernel Memory Dumps in WinDbg
Default Windows crash dump settings generate Small Memory Dumps (minidumps) limited to 256 KB. Minidumps frequently truncate the secondary payload offsets of large CPER records, making full device identification impossible. Production systems and high-throughput workstations must be configured to generate an active or complete kernel memory dump (MEMORY.DMP) stored under %SystemRoot%.
To verify and enforce kernel dump generation on target endpoints, execute the following administrative PowerShell command sequence to inspect and adjust the crash reporting subsystem:
# Query active crash dump configuration
Get-ItemProperty -Path "HKLM:\SYSTEM\CurrentControlSet\Control\CrashControl" | Select-Object CrashDumpEnabled, DumpFile
# Enforce Kernel Memory Dump mode (DumpType = 2) and enable overwrite
Set-ItemProperty -Path "HKLM:\SYSTEM\CurrentControlSet\Control\CrashControl" -Name "CrashDumpEnabled" -Value 2
Set-ItemProperty -Path "HKLM:\SYSTEM\CurrentControlSet\Control\CrashControl" -Name "Overwrite" -Value 1
Once the system captures a crash event following an unrecoverable freeze, launch Microsoft WinDbg (available via the Windows SDK or modern Windows Terminal package) as Administrator. Configure the public Microsoft Symbol Server path and open the dump file:
# Set symbol search path within WinDbg console
.sympath srv*https://msdl.microsoft.com/download/symbols
# Reload symbols and force module resolution
.reload /f
5 Steps to Parse CPER Records for WHEA_UNCORRECTABLE_ERROR 0x124
Diagnosing the physical failure requires querying the error record descriptor via the dedicated debugger extension. Follow this five-step sequence within WinDbg to parse the hardware diagnostic frame down to the physical silicon entity:
Step 1: Execute Automated Kernel Analysis
Run the primary automated analysis extension to determine basic parameter allocation and identify the reporting error source:
kd> !analyze -v
Inspect the output block under BUGCHECK_P1 through BUGCHECK_P4. If Arg1 displays 00000000`00000004, the subsystem has suffered a PCIe bus communication drop. If Arg1 displays 00000000`00000000, a CPU execution unit has flagged a Machine Check Exception.
Step 2: Parse the Common Platform Error Record
Locate the 64-bit pointer displayed in Arg2. Pass this address directly to the WinDbg error record extension as detailed in the Microsoft WinDbg !errrec extension reference:
kd> !errrec ffffd081`7b2e9028
WinDbg parses the raw hexadecimal structure into human-readable headers, displaying the record GUID, error severity, and associated hardware section descriptors:
===============================================================================
Common Platform Error Record (CPER) Section:
Section Type : PCIe Error (d29a5690-e076-47a3-b526-eb3ec640047b)
Severity : Fatal (1)
Length : 208
FRU Text : OEM NVMe Storage Array
-------------------------------------------------------------------------------
PCIe Root Port / Device Descriptor:
Port Type : Root Port
Version : 2.0
Command / Stat : 0x0010 / 0x0000
Device ID : VendorID 0x8086, DeviceID 0x7AB8, SubVendor 0x1043
Bus:Dev:Func : 0x00 / 0x1B / 0x04
Step 3: Decode the Bus, Device, and Function (BDF) Coordinates
The CPER output isolates the physical location of the failing endpoint via its Bus, Device, and Function registers (e.g., 0x00 / 0x1B / 0x04). Use the WinDbg PCI extension to query the Windows PCI bus driver for that hardware hardware node:
kd> !pci 100 0 1b 4
Match the resulting DeviceID against your hardware topography. When an M.2 solid-state drive encounters severe controller thermal throttling or voltage collapse under prolonged I/O loads, it suddenly disconnects from the PCIe bus. When the root port experiences a link-down event without clean link negotiation, WHEA flags an unrecoverable AER condition. This operational scenario frequently impacts high-speed storage components evaluated in our Storage & RAM hardware evaluations.
Step 4: Inspect the Advanced Error Reporting (AER) Bitmask
Examine the PCIe AER registers embedded within the !errrec output block. Key registers include the Uncorrectable Error Status register:
- Bit 4 (Data Link Protocol Error): Physical layer packet corruption across copper traces.
- Bit 5 (Surprise Down Error): An endpoint abruptly lost power or dropped off the bus during execution.
- Bit 14 (Completion Timeout): A storage controller or GPU command buffer failed to respond within specification limits.
- Bit 18 (Malformed TLP): A Transaction Layer Packet failed CRC check due to unstable signal integrity or incorrect PCIe bifurcation.
Step 5: Verify Machine Check Architecture (MCA) Registers
If Arg1 denotes a processor MCA fault (0x0), run the machine check bank inspection extension to evaluate specific core telemetry:
kd> !mca
Examine the MCi_STATUS register output. A Bank 0 or Bank 1 error points to instruction fetch or data cache parity failures, often caused by undervolting or thermal degradation on specific CPU cores analyzed in our Processor architecture benchmarks.
Deterministic Hardware Remediation Matrix
Once WinDbg isolates whether the failure originates from the central processor, system memory, or the PCI Express fabric, implement targeted hardware remediation using the engineering reference matrix below:
| Identified Subsystem | WinDbg Signature | Primary Hardware Root Cause | Remediation Protocol |
|---|---|---|---|
| PCIe NVMe Controller | Section: PCIe Error, AER Bit 5 (Surprise Down) | ASIC thermal throttling beyond TMT2 or auxiliary rail voltage drops. | Reseat M.2 drive, replace degraded thermal interface material, and disable PCIe ASPM in UEFI. |
| Processor Core (MCA) | Section: Processor Generic, Bank 0-4 Parity Fault | Unstable Vcore curve, aggressive LLC droop, or degraded silicon. | Disable CPU undervolting offsets, increase Load-Line Calibration (LLC), and flash vendor microcode. |
| System Memory (DRAM) | Section: Memory Error, Single-bit / Multi-bit ECC Fault | DRAM sub-timing training failures or memory controller signal reflection. | Disable aggressive XMP/EXPO memory profiles, run MemTest86+ across four passes, and verify DIMM socket tension. |
| Discrete GPU / Accelerator | Section: PCIe Error, AER Bit 14 (Completion Timeout) | PCIe riser cable signal degradation or 12VHPWR connector thermal creep. | Remove PCIe riser assemblies, verify direct motherboard slot mating, and audit 12V rail voltage stability. |
Frequently Asked Questions
Can a corrupted device driver trigger WHEA_UNCORRECTABLE_ERROR 0x124?
While third-party drivers do not directly generate bug check 0x124, a driver that incorrectly programs hardware registers (such as disabling link training or writing invalid ASPM low-power states to a controller) can cause the underlying silicon to trigger a physical hardware interrupt. The root cause remains a hardware response, but driver-level power management bugs can precipitate the condition.
What is the difference between a minidump and a complete kernel dump for 0x124?
A minidump records only the active call stack and loaded driver lists. It frequently omits the full CPER record memory segment referenced by Parameter 2. A kernel dump preserves physical kernel structures, enabling the !errrec debugger command to extract all PCIe AER descriptors, vendor IDs, and hardware FRU strings.
How does PCIe Active State Power Management (ASPM) affect 0x124 stability?
Aggressive ASPM configurations force high-speed PCIe devices into low-power states (L0s and L1). If an NVMe controller or network interface card exhibits slow L1 exit latency, the root port flags a completion timeout before the endpoint wakes, causing an instantaneous WHEA 0x124 kernel crash.
For homelabs, remote testing, and staging environments, we run NVMe VDS instances on Erahost. Fast provisioning, European routing, and predictable I/O.
View NVMe Plans →