CVE-2026-98721
NoneCVSS
—
EPSS
0.2%
Published
Oct 18, 2026
CISA due date
—
Cloudanix analysis
A startup timing problem in certain storage hardware drivers could cause a computer to crash if an interrupt arrives at exactly the wrong moment.
- Exploitation
- Local Privilege
- Attacker needs
- Local User
- Complexity
- High
- Impact
- Denial Of Service
- Component
- OS kernel
- Class
- Race Condition
- Remediation
- Patch
- Effort
- Low
- Patch risk
- Low
What it actually does
Race condition allowing a pending interrupt to access uninitialized IRQ lock and completion objects because the handler is registered before these primitives are initialized in the Cadence NAND driver.
Why it matters to the business
Potential kernel instability, crashes during boot or driver load on systems using Cadence NAND controllers, increasing downtime risk for storage-dependent workloads; low EPSS and no KEV status indicate limited immediate operational urgency.
How an attack unfolds
An attacker triggers or schedules a pending interrupt during the brief window after IRQ registration but before lock and completion initialization, causing the handler to dereference uninitialized synchronization state and induce memory corruption or panic.
How to fix it
Initialize the IRQ lock and completion object before calling request_irq in the driver; apply the upstream kernel commit that resolves the ordering issue.
How to confirm the fix worked
Inspect the rawnand cadence driver source to confirm lock and completion initialization precede IRQ registration; stress-test driver load paths with simulated pending interrupts to verify absence of the race.
Typical attack path
- Load or probe the Cadence NAND driver (module init or device enumeration)
- Arrange for a pending interrupt to fire immediately after request_irq
- Interrupt handler executes with uninitialized lock/completion, triggering undefined behavior or crash
How to detect it
Review kernel boot logs and dmesg for oops, panic, or warning messages referencing the cadence NAND interrupt handler or uninitialized synchronization objects during driver probe.
What your security team should do
- Apply the fix to all kernels containing the Cadence NAND driver within standard patch cycles
- Include the updated driver in next kernel build or distribution update
- Monitor mainline kernel changelogs for related NAND changes
The record
In the Linux kernel, the following vulnerability has been resolved: mtd: rawnand: cadence: Initialize IRQ state before requesting IRQ The Cadence NAND interrupt handler uses both the IRQ lock and completion object. Registering the IRQ before initializing them leaves a window in which a pending interrupt can access uninitialized synchronization state. Initialize them before registering the handler.
Affected packages
| Ecosystem | Package | Introduced | Fixed |
|---|