CVE-2026-112073
Received Received - Intake

Out-of-Bounds Read in Tencent ncnn MatMul Component

Publication date: 2026-10-19

Last updated on: 2026-10-19

Assigner: VulDB

Description
A vulnerability was identified in Tencent ncnn up to 44b02dabdcf97499281a4692ebd65bad73dcf8b6. Affected by this vulnerability is the function matmul_transb of the file src/layer/matmul.cpp of the component MatMul. The manipulation leads to out-of-bounds read. The attack can be initiated remotely. The exploit is publicly available and might be used. The project was informed of the problem early through an issue report but has not responded yet.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-10-19
Last Modified
2026-10-19
Generated
2026-10-19
AI Q&A
2026-10-19
EPSS Evaluated
N/A
NVD
CVE-2026-112073
EUVD
EUVD-2026-104055

Affected Vendors & Products

Showing 1 associated CPE
Vendor Product Version / Range
Tencent ncnn 44b02dabdcf97499281a4692ebd65bad73dcf8b6

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-125 The product reads data past the end, or before the beginning, of the intended buffer.
CWE-119 The product performs operations on a memory buffer, but it reads from or writes to a memory location outside the buffer's intended boundary. This may result in read or write operations on unexpected memory locations that could be linked to other variables, data structures, or internal program data.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

CVE-2026-112073 is a heap-buffer-overflow vulnerability in Tencent ncnn's MatMul layer. It occurs when the width of the first input matrix exceeds the width of the second input matrix during matrix multiplication. The MatMul layer fails to validate input dimensions before processing, causing the underlying Gemm layer to read beyond allocated memory.

Detection Guidance

To detect this vulnerability, compile ncnn with AddressSanitizer (ASAN) and run inference on models with mismatched matrix dimensions (A.w > B.w). Check for heap-buffer-overflow errors in logs. Example command: ASAN_OPTIONS=detect_leaks=0 ./ncnn_test --model malicious_model.param --input input.bin

Impact Analysis

This vulnerability can cause denial of service via crashes when processing malicious model files. It may also lead to information leakage if the overflowed memory contains sensitive data. Remote code execution is not possible as the issue does not provide write primitives.

Compliance Impact

This vulnerability could potentially impact compliance with GDPR and HIPAA by enabling information leakage if the out-of-bounds read exposes sensitive data in memory. GDPR requires protection of personal data, and HIPAA mandates safeguarding protected health information. The vulnerability does not allow remote code execution but could lead to unauthorized data access during inference operations.

Mitigation Strategies

Update ncnn to the latest version if available. Validate input matrix dimensions (A.w <= B.w) before matrix multiplication. Restrict model loading to trusted sources only. Monitor for crashes or memory corruption during inference.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-112073. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart