CVE-2024-42242

What it actually does

The SDHCI driver sets blk_queue_max_segment_size below PAGE_SIZE under certain conditions (notably 64 KiB pages), which blk_validate_limits now rejects with -EINVAL instead of silently raising the value.

Why it matters to the business

On systems with 64 KiB page sizes, MMC/SD storage devices may fail to initialize, causing loss of block I/O and potential boot or runtime storage failures; the issue is widespread in the Linux kernel but carries low urgency absent KEV listing or active exploitation.

How an attack unfolds

A local attacker with permission to trigger MMC/SDHCI device enumeration or block-queue setup causes the driver to register an invalid max_segment_size; the subsequent blk_validate_limits call fails, returning an error that aborts device initialization and produces a denial-of-service condition.

How to fix it

Upgrade the Linux kernel to a build containing commit bf78b1accef46efd9b624967cb74ae8d3c215a2b; for AlmaLinux 8 systems also ensure the chrony package is at version 4.1-1.el8.alma or later.

How to confirm the fix worked

After patching, confirm that sdhci no longer sets max_segment_size below PAGE_SIZE by inspecting kernel logs for absence of blk_validate_limits warnings and by verifying the segment size value on affected MMC devices.

Typical attack path

1. Obtain local access with privileges sufficient to access block devices or trigger driver probe
2. Cause SDHCI/MMC device initialization or queue configuration
3. Trigger blk_validate_limits validation that detects max_segment_size < PAGE_SIZE
4. Receive -EINVAL and abort device setup, resulting in loss of storage availability

How to detect it

Search kernel and dmesg logs for messages containing blk_validate_limits, max_segment_size, or EINVAL originating from MMC/SDHCI initialization paths; correlate with device probe events for SDHCI controllers.

What your security team should do

• Apply the referenced kernel commit or distribution update to all affected systems
• Reboot or reload MMC drivers to activate the corrected segment-size logic
• Monitor boot and udev logs for segment-size validation errors until patches are deployed