CVE intelligence and bounded remediation
CVE-2021-41220 — TensorFlow is an open source platform for machine learning
TensorFlow is an open source platform for machine learning. In affected versions the async implementation of `CollectiveReduceV2` suffers from a memory leak and a use after free. This occurs due to the asynchronous computation and the fact that objects that have been `std::move()`d from are still accessed. The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, as this version is the only one that is also affected.
- Severity
- High
- CVSS
- 7.8 (3.1)
- Published
- 2021-11-05
- CISA KEV
- Not currently listed
- Ecosystem
- software/application
- Weaknesses
- CWE-416
Affected products
- google / tensorflow
- google / tensorflow / 2.7.0
Matched remediation archetype
Use-after-free, double free, and expired resource use
This catalog composition supplies bounded fallback guidance. Explicitly reviewed curated workflows load with the complete record below.
Check exposure
- Trace ownership, references, callbacks, asynchronous tasks, and teardown paths around the affected object or resource.
- Identify reachable inputs and timing or state transitions that can release the object while references remain.
- Confirm affected builds, allocators, feature flags, architectures, and process privileges.
Remediate safely
- Apply the maintained ownership or lifetime fix and rebuild all artifacts containing the affected native code.
- Use explicit ownership, safe reference management, cancellation and join semantics, and idempotent teardown.
- Add deterministic lifetime tests plus isolated sanitizer and concurrency coverage for shutdown and error paths.
Authoritative sources
Complete CVE record and remediation plan
The detailed catalog view below loads this exact record, its source evidence, and the full seven-phase agentic change plan.