CVE-2024-36887

What it actually does

The e1000e PHY MDIC access routines use usleep_range which cannot be called from atomic context, leading to kernel panic during events such as cable disconnect/reconnect on vPro systems.

Why it matters to the business

Systems running affected Linux kernels with Intel e1000e hardware may experience unexpected crashes, causing downtime and potential loss of network connectivity on vPro platforms.

How an attack unfolds

An attacker or environmental trigger such as repeated cable events forces the driver into atomic context while sleeping, triggering the panic; no remote code execution path exists.

How to fix it

Update to Linux kernel versions containing commit f8a139656c95db893a543159873c57a470d7376d or later stable releases such as 6.6.31 and newer; rebuild and reboot affected systems.

How to confirm the fix worked

Confirm kernel version is at or above the listed fixed releases; test PHY access paths under atomic conditions and monitor for absence of panic messages.

Typical attack path

1. Trigger PHY access from atomic context via cable state change or similar event
2. usleep_range executes in non-sleepable context
3. Kernel panics due to sleeping function in atomic section

How to detect it

Review kernel logs for panic traces referencing e1000e MDIC or usleep_range in atomic context; correlate with cable events on affected hardware.

What your security team should do

• Prioritize patching kernel packages on all systems using e1000e drivers
• Schedule reboot windows for production vPro systems
• Monitor vendor advisories for backported fixes in distribution kernels