CVE-2025-68794

What it actually does

The flaw in iomap_adjust_read_range assumes block-aligned inputs and fails to account for block offsets, causing skipped-byte calculations to over-subtract, underflow length, or return a position beyond the folio when all blocks are uptodate.

Why it matters to the business

Linux systems may encounter file-system read failures, kernel instability, or data-access errors on workloads using iomap-based file systems such as erofs, increasing operational risk and potential downtime until patched.

How an attack unfolds

An attacker with local access issues read requests at non-block-aligned offsets against an affected file system; the miscalculation returns an invalid position or length that can lead to incorrect folio handling or subsequent kernel faults.

How to fix it

Upgrade the Linux kernel to a version containing commit 82b60ffbb532d919959702768dca04c3c0500ae5 that incorporates the corrected block-offset calculation.

How to confirm the fix worked

Confirm the fix commit is present in the running kernel tree and execute targeted read tests at non-aligned offsets on erofs or similar file systems to verify correct range adjustment.

Typical attack path

1. Mount an affected file system such as erofs
2. Issue read operations with non-block-aligned start positions
3. Force iomap_adjust_read_range to evaluate fully uptodate blocks
4. Trigger length underflow or out-of-bounds position return

How to detect it

Inspect kernel logs for iomap read-range or folio-position warnings; no public exploitation artifacts are known.

What your security team should do

• Prioritize kernel updates containing commit 82b60ffbb532d919959702768dca04c3c0500ae5 across all affected hosts
• Validate the patch in a staging environment before broad deployment
• Monitor kernel stability and file-system error logs after rollout