CVE-2026-99136: dm-crypt: fix a tiny race condition in crypt_dec_pending

Published Oct 28, 2026

Updated

dm-crypt: fix a tiny race condition in cryptdecpending

Affected Software

0 affected components

Event History

Oct 28, 2026

CVE Published

via MITRE·10:34 AM

Data Sourced

via MITRE·10:34 AM

Description

Data Sourced

via NVD·11:17 AM

Description

Oct 29, 2026

Data Sourced

via Microsoft·08:05 AM

DescriptionSeverityWeakness

Frequently Asked Questions

What is the practical impact of this race?

A failed read may be reported as successful even though the data was never decrypted. A failed write may also be reported as successful, which can cause callers to believe I/O completed correctly when it did not.

Under what conditions can the issue occur?

The race requires one context to read the I/O error status while another context, such as crypt_endio executing from an interrupt, sets that error and drops its reference. The stale status can then be used when the final reference is released and the bio is completed.

Is an additional memory barrier required by the fix?

No. The fix reads io->error after atomic_dec_and_test, and atomic_dec_and_test is fully ordered, so no additional barrier is needed.

CVE-2026-99136: dm-crypt: fix a tiny race condition in crypt_dec_pending

Affected Software

Event History

Frequently Asked Questions

What is the practical impact of this race?

Under what conditions can the issue occur?

Is an additional memory barrier required by the fix?

Company

Resources

Contact

Frequently Asked Questions

What is the practical impact of this race?

Under what conditions can the issue occur?

Is an additional memory barrier required by the fix?