Few Windows errors are as disruptive as a Blue Screen of Death (BSOD). Among the most common is IRQL_NOT_LESS_OR_EQUAL (Stop code: IRQL NOT LESS OR EQUAL, bugcheck 0xA). It typically appears during normal use, at boot, or while launching drivers-heavy tasks (gaming, video conferencing, virtualization), and it indicates that something at a low level (usually a driver or hardware) tried to access memory it shouldn’t at an inappropriate interrupt request level (IRQL).
This guide goes far beyond generic advice. You’ll learn exactly what this stop code means, the most likely causes, and how to methodically fix it using real minidump analysis with WinDbg and BlueScreenView. We’ll start with safe preliminary checks, then move through a structured, step-by-step troubleshooting process, culminating in advanced diagnostics like Driver Verifier and, if necessary, an in-place repair. By the end, you’ll know how to identify the offending driver or component and prevent this BSOD from coming back.
Understanding the Error
What “IRQL_NOT_LESS_OR_EQUAL” Means (in plain language)
Windows uses Interrupt Request Levels (IRQLs) to prioritize work in the kernel. High IRQL code must not access paged memory or perform operations that can be interrupted. The IRQL_NOT_LESS_OR_EQUAL bugcheck (0x0000000A) occurs when kernel-mode code (often a driver) tries to access memory at an IRQL that isn’t allowed—for example, dereferencing an invalid pointer or touching pageable memory at a high IRQL.
In plain terms: a kernel component (usually a driver) did something unsafe with memory. It can be a coding bug, memory corruption, bad RAM, firmware issues, or a disk/driver error that led to invalid memory access.
Common Scenarios That Trigger It
- Installing or updating a device driver (GPU, network, audio, storage).
- Heavy network or disk I/O (copying large files, torrents, virtualization).
- Waking from sleep/hibernation, plugging in USB/Thunderbolt devices or docks.
- Overclocking/undervolting CPU, GPU, or enabling XMP/EXPO for RAM.
- After a Windows Update or BIOS/UEFI update.
- Interactions with third-party antivirus, VPN, or performance-tuning software.
Common Causes
Skim this list to identify potential culprits:
- Faulty or outdated device drivers (GPU, NIC, storage, chipset, USB, audio).
- RAM issues (unstable XMP/EXPO, failing DIMMs, mismatched kits).
- Disk/SSD issues (corruption, failing sectors, outdated NVMe firmware).
- BIOS/UEFI or firmware bugs (chipset, NVMe, GPU VBIOS).
- Windows updates and compatibility conflicts.
- Third-party antivirus/VPN/filter drivers (kernel hooks).
- Overclocking/undervolting or aggressive power/thermal tuning.
- Malware or rootkits (rare but possible for kernel-level errors).
- Page file disabled or misconfigured (required for minidumps and stability).
Quick reference table of symptoms and causes:
| Symptom | Likely Cause |
|---|---|
| BSOD shortly after booting to desktop | Startup driver/service conflict, antivirus/VPN filter drivers |
| BSOD during gaming/graphics workload | GPU driver, DirectX runtimes, unstable overclock, RAM |
| BSOD on resume from sleep/hibernation | Chipset/storage/USB drivers, fast startup conflicts |
| BSOD while copying files or using VMs | Storage/NVMe driver, network driver, disk issues |
| Random BSODs with “memory_corruption” in dump | Faulty RAM, unstable XMP, or a misbehaving driver corrupting memory |
Preliminary Checks
Before deep diagnostics, stabilize the system.
Boot to Safe Mode
Safe Mode loads a minimal set of drivers.
- Windows 11/10 (when you can log in):
- Start > Settings > System > Recovery > Advanced startup > Restart now.
- Troubleshoot > Advanced options > Startup Settings > Restart.
- Press 4 (or F4) for Safe Mode.
- If Windows won’t boot:
- Interrupt boot 3 times (power off during the Windows loading spinner). Windows Recovery Environment (WinRE) appears. Go Troubleshoot > Advanced options > Startup Settings > Restart > Press 4 (F4).
- Or from an elevated Command Prompt in WinRE:
- To force next boot into Safe Mode:
bcdedit /set {current} safeboot minimal - To exit Safe Mode later:
bcdedit /deletevalue {current} safeboot
- To force next boot into Safe Mode:
Back Up Important Data
If the system is unstable, back up now.
- Copy files to external storage or cloud.
- Example (robocopy, run in an elevated Command Prompt):
robocopy C:\Users\YourName\Documents D:\Backup\Documents /E /R:1 /W:1
Run Basic Health Checks
- Check disk file system
- Quick online scan (no reboot):
chkdsk C: /scan - Full fix (reboot required):
chkdsk C: /f
- System file integrity
- System File Checker:
sfc /scannow - Deployment Image Servicing and Management (repair component store):
DISM /Online /Cleanup-Image /RestoreHealth
Then run sfc /scannow again.
- SMART/drive health (PowerShell as Administrator)
- Basic:
Get-PhysicalDisk | Select FriendlyName, HealthStatus, OperationalStatus - For deeper SMART, use your SSD vendor tool (Samsung Magician, Crucial Storage Executive, Western Digital Dashboard) or a utility like CrystalDiskInfo.
If these checks find and fix issues, test for stability. If the BSOD persists, continue.
Step-by-Step Troubleshooting
Follow these steps in order. Reboot and test between steps. Create a restore point before major changes.
- Undo Recent Changes
- Roll back any recently installed drivers, Windows updates, or software (AV/VPN, tuning tools).
- Device Manager > Right-click device > Properties > Driver tab > Roll Back Driver (if available).
- Settings > Windows Update > Update history > Uninstall updates.
- Uninstall recently added apps in Settings > Apps > Installed apps.
- Update Windows and Drivers (but do it smartly)
- Settings > Windows Update > Check for updates. Install critical/security updates first.
- Check Optional updates for drivers—but prefer vendor/OEM sites for GPU, chipset, storage/NVMe.
- Update these in order:
- Chipset drivers (Intel, AMD): from your PC/motherboard OEM first, else from Intel/AMD.
- Storage/NVMe drivers: Intel RST, Samsung NVMe, AMD RAID if applicable.
- GPU drivers: NVIDIA/AMD/Intel (use Clean Install option if available).
- Network and audio drivers: from OEM.
- Avoid third‑party “driver updaters.”
- Revert Overclocking and XMP/EXPO
- Restore BIOS/UEFI to defaults (Load Optimized Defaults).
- Disable CPU/GPU overclocking and undervolting (Intel XTU, Ryzen Master, Afterburner).
- Disable XMP/EXPO to test at JEDEC speeds; if stable, later try a more conservative profile.
- Check Antivirus/VPN/Filter Drivers
- Temporarily uninstall third-party security suites, VPN clients, and endpoint agents known to install kernel drivers (e.g., filter/firewall). Use vendor cleanup tools if available.
- Use built-in Microsoft Defender during testing.
- Test Memory
- Windows Memory Diagnostic:
mdsched.exe
Choose Restart now and check for problems. - For thorough testing, use MemTest86 (bootable USB). One full pass minimum; 4 passes preferred. Any errors = faulty/unstable RAM or timings. Test each stick alone to isolate.
- Storage and Firmware
- Update SSD/NVMe firmware with vendor tools.
- For SATA disks: replace flaky cables, test on another port.
- Run vendor diagnostics (e.g., SeaTools, Samsung Magician benchmarks).
- Ensure adequate free space (at least 15–20% on system drive) and that the page file is enabled and on C:.
- BIOS/UEFI and Device Firmware
- Suspend BitLocker (if used): Control Panel > BitLocker > Suspend protection.
- Update BIOS/UEFI and device firmware (motherboard, SSDs, Thunderbolt, docks) from OEM support pages.
- After update, load defaults, then adjust only necessary settings.
- Clean Boot to Isolate Conflicts
- msconfig > Services tab > Hide all Microsoft services > Disable all.
- Task Manager > Startup > Disable nonessential items.
- Reboot; if stable, re-enable services/startups in batches to find the offender.
- System Restore (if applicable)
- If the issue started recently and Restore was enabled:
Control Panel > Recovery > Open System Restore > choose a point before the BSODs.
- Minidump Analysis (the decisive step)
You’ll now identify the faulty driver or module using minidumps—small crash files Windows writes at BSOD time.
Enable and Locate Minidumps
- Control Panel > System > Advanced system settings > Advanced tab > Startup and Recovery (Settings):
- Write debugging information: Small memory dump (256 KB).
- Dump file: %SystemRoot%\Minidump
- Ensure page file is enabled on C: and not set to “No paging file.”
- After a crash, minidumps are in:
C:\Windows\Minidump
If there are no dumps:
- Verify the page file is system-managed on C:, ensure enough free space, and disable “Automatically restart” in Startup and Recovery to see if dumps begin to appear.
Option A: BlueScreenView (easiest)
- Download BlueScreenView (NirSoft).
- Open it; it auto-loads minidumps.
- Look at “Bug Check String” (IRQL_NOT_LESS_OR_EQUAL) and “Caused By Driver.”
- Inspect the lower pane call stack. If a third-party driver (e.g., nvlddmkm.sys, rt640x64.sys, aswNetSec.sys) appears at or near the top, that’s a strong clue.
- Action: update/roll back or uninstall that specific driver/software.
Option B: WinDbg (most powerful)
- Install WinDbg (Preview) from Microsoft Store.
- Launch WinDbg (x64), File > Open dump file > choose latest .dmp from C:\Windows\Minidump.
- Set symbols (only once needed):
- In the command window:
.symfix
.sympath srvC:\Symbolshttps://msdl.microsoft.com/download/symbols
.reload
- In the command window:
- Run the analysis:
!analyze -v - Read the output:
- Bugcheck code: 0xA (IRQL_NOT_LESS_OR_EQUAL).
- Probably caused by: driver.sys (don’t blindly trust “ntoskrnl.exe” as it’s often a victim).
- “MODULE_NAME” and “IMAGE_NAME” show the suspected module.
- “STACK_TEXT” reveals which drivers were active; look for third-party modules near the fault.
- Inspect a suspect driver:
lmvm drivername
This shows version, timestamp, and company. - If the stack shows memory_corruption or pool corruption:
- Strongly suspect RAM/XMP instability or a driver overwriting memory. Validate with MemTest86 and consider Driver Verifier (below).
Sample WinDbg session:
!analyze -v
lmvm rt640x64
kv
Interpretation tips:
- If a vendor driver appears repeatedly across multiple dumps, update or remove it.
- If only ntoskrnl.exe is blamed, dig into the stack and module list; correlate with Event Viewer logs and recent changes.
- Cross-reference the timestamp: very old drivers are high risk on modern Windows builds.
- Replace or Repair the Culprit
- GPU/network/storage/audio driver stuck in stack? Clean-install the driver:
- Use the vendor’s “clean install” checkbox or DDU (Display Driver Uninstaller) for GPUs in Safe Mode, then install the latest stable driver.
- Bad or unstable RAM? Replace the faulty DIMM(s) and/or run at stock (no XMP).
- Failing SSD/HDD? Replace the drive and restore from backup or image.
- In-Place Repair (Repair Install)
If corruption persists and you can boot:
- Download the latest Windows ISO (Media Creation Tool or MS site).
- Mount the ISO, run setup.exe, choose “Keep personal files and apps.”
- This refreshes Windows system files while preserving your data and programs.
Advanced Diagnostics
Use Driver Verifier (with caution)
Driver Verifier stresses drivers and will often trigger a BSOD for the misbehaving one—this is by design.
- Create a restore point and ensure you can boot into Safe Mode.
- Open Command Prompt (Admin) and run:
verifier /standard /all - Reboot and use the system until it BSODs again. Analyze the new minidump as above. Look for a specific third-party driver.
- When done (or if you get a boot loop), disable Driver Verifier:
verifier /reset
Reboot.
Tips:
- You can target non-Microsoft and unsigned drivers only for fewer false positives:
verifier
Choose Create custom settings > Select individual settings from a full list > check standard options + additional like Force IRQL checking and Pool tracking > Select driver names from a list > exclude Microsoft drivers.
Event Viewer Correlation
- eventvwr.msc > Windows Logs > System.
- Filter for “BugCheck” (Event ID 1001) and driver/service errors around the crash time.
- Windows Error Reporting (WER) entries can also provide clues about faulting modules.
Collect System Info
- msinfo32 (File > Save) to capture system environment, driver versions, BIOS.
- Useful when seeking help from forums/pros.
When to Seek Professional Help
Consider professional repair or hardware replacement when:
- MemTest86 reports any errors (RAM must be replaced or timings corrected).
- SSD/HDD shows reallocated/pending sectors or “Caution/Bad” SMART status.
- BSODs persist after a clean Windows install with only OEM drivers.
- Physical symptoms: freezes combined with coil whine, PSU instability, overheating throttles, or GPU artifacts suggest hardware failure.
- You’re not comfortable updating BIOS/UEFI or handling internal components.
A technician can swap parts, test with known-good hardware, and perform board-level diagnostics.
Prevention Tips
- Keep drivers updated—but prioritize OEM and chipset/storage drivers. Avoid random driver packs.
- Update BIOS/UEFI and firmware when release notes mention stability fixes.
- Avoid aggressive overclocks; if you do overclock, stress test thoroughly (Prime95, MemTest86, OCCT).
- Leave the page file enabled and on the system drive to ensure minidumps are created.
- Make regular system images (e.g., Macrium Reflect, Windows Backup) and store backups offline.
- Install only one antivirus. Uninstall old security/VPN tools completely before switching vendors.
- Be cautious with USB/Thunderbolt docks and hubs: keep their firmware and drivers current.
- After big changes (major Windows Feature Update, BIOS update), run sfc /scannow and DISM as a sanity check.
Conclusion
IRQL_NOT_LESS_OR_EQUAL (0xA) BSODs are fixable. They point to low-level memory access issues—most often caused by drivers, RAM instability, or storage/firmware problems. By starting with Safe Mode and basic health checks, then moving to targeted updates and minidump analysis with BlueScreenView or WinDbg, you can identify the exact culprit and resolve the issue. If needed, Driver Verifier and an in-place repair provide deeper remediation. With good driver hygiene and regular backups, you can prevent this BSOD from returning.
Stay patient, follow the steps in order, and you’ll typically find the root cause.
FAQ
1) Can I ignore the IRQL_NOT_LESS_OR_EQUAL BSOD?
No. This is a kernel-level error that typically won’t resolve on its own. Ignoring it risks data loss and further instability. Use minidumps to identify the offending driver or component.
2) Does this BSOD mean my hardware is failing?
Not necessarily. Many cases are driver-related. However, if dumps point to memory_corruption, or if MemTest86/SMART diagnostics show errors, hardware (RAM or storage) may indeed be failing.
3) I only see ntoskrnl.exe in BlueScreenView. What now?
ntoskrnl.exe is often a victim, not the root cause. Use WinDbg for deeper analysis, check the call stack, and correlate with Event Viewer. Try Driver Verifier to force the real culprit to surface in the next dump.
4) Why are there no minidumps on my system?
Common reasons: page file disabled or not on C:, insufficient disk space, or system is hard powering off before writing the dump. Ensure Small memory dump is enabled, leave the page file on C:, keep free space, and disable automatic restart temporarily.
5) Is Driver Verifier safe?
Yes, but it intentionally stresses drivers and can cause repeated BSODs while testing. Use it only when you can boot to Safe Mode, and disable it with verifier /reset after you’ve captured a useful dump.
Command Reference (copy/paste)
- CHKDSK (online scan):
chkdsk C: /scan - CHKDSK (fix on reboot):
chkdsk C: /f - System File Checker:
sfc /scannow - DISM (repair image):
DISM /Online /Cleanup-Image /RestoreHealth - Force Safe Mode (enable/disable):
bcdedit /set {current} safeboot minimal
bcdedit /deletevalue {current} safeboot - Driver Verifier (enable/disable):
verifier /standard /all
verifier /reset - WinDbg symbol setup:
.symfix
.sympath srvC:\Symbolshttps://msdl.microsoft.com/download/symbols
.reload - WinDbg analysis:
!analyze -v
lmvm drivername
kv
By using the structured approach above—especially the minidump analysis—you’ll have the best chance of resolving the IRQL_NOT_LESS_OR_EQUAL BSOD quickly and permanently. Good luck, and don’t hesitate to reach out to a professional if hardware faults are suspected.
