Heap corruption via overflowed convolutional buffer sizing
Published Aug 20, 2026 · Updated Aug 25, 2026
Integer overflow in Darknet 6.0 and earlier allows attackers to corrupt heap memory by loading a crafted convolutional .cfg file. In convolutional_layer.cpp, unchecked 32-bit products set l.nweights and l.outputs for xcalloc, while the forward pass recalculates GEMM dimensions in a different order and accesses beyond the undersized buffers. Inference or training must load the file; the demonstrated results are an out-of-bounds read, heap-metadata corruption, and a process crash, not code execution.
Summary
What happened
Integer overflow in Darknet 6.0 and earlier allows attackers to corrupt heap memory by loading a crafted convolutional .cfg file. In convolutional_layer.cpp, unchecked 32-bit products set l.nweights and l.outputs for xcalloc, while the forward pass recalculates GEMM dimensions in a different order and accesses beyond the undersized buffers. Inference or training must load the file; the demonstrated results are an out-of-bounds read, heap-metadata corruption, and a process crash, not code execution.
The record
- CVE
- CVE-2026-72852
- Published
- Aug 20, 2026
- Updated
- Aug 25, 2026
- Vendor
- Hank AI
- Product
- Darknet
- Classifications
- CWE-190, CWE-787, T1499
- Attack vector
- local
- Privileges
- unauthenticated
Timeline
How it unfolded
- Aug 20, 2026CVE publishedPublication date reported by the CVE source.
- Aug 25, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=0 <=6.0
What conditions does exploitation require?
What is affected?
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- GitHub Issue 148 reproducerproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo