Skip to main content

Missing Release of Memory after Effective Lifetime

CVE-2022-23578

Medium
4.3/10

Summary

Tensorflow is an Open Source Machine Learning Framework. If a graph node is invalid, TensorFlow can leak memory in the implementation of `ImmutableExecutorState::Initialize`. Here, we set `item->kernel` to `nullptr` but it is a simple `OpKernel*` pointer so the memory that was previously allocated to it would leak. The fix will be included in TensorFlow 2.8.0rc0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.

  • LOW
  • NETWORK
  • NONE
  • UNCHANGED
  • NONE
  • LOW
  • NONE
  • LOW

CWE-401 - Missing release of memory after effective lifetime (memory leak)

'Missing release of memory after effective lifetime (memory leak)' is a weakness that occurs when software doesn't effectively release allocated memory after it is used. If not addressed, this enables attackers to launch denial of service attacks (by crashing or hanging the program) or take advantage of other unexpected behavior resulting from low memory conditions.

References

Advisory Timeline

  • Published