CVE-2024-42477

CVSS V2 None CVSS V3 None

Description

llama.cpp provides LLM inference in C/C++. The unsafe `type` member in the `rpc_tensor` structure can cause `global-buffer-overflow`. This vulnerability may lead to memory data leakage. The vulnerability is fixed in b3561.

Overview

CVE ID
CVE-2024-42477

Assigner
GitHub_M

Vulnerability Status
PUBLISHED

Published Version
2024-08-12T15:02:40.980Z

Last Modified Date
2024-08-12T15:02:40.980Z

Weakness Enumerations

CWE	URL
CWE-125	http://cwe.mitre.org/data/definitions/125.html

References

Reference URL	Reference Tags
https://github.com/ggerganov/llama.cpp/security/advisories/GHSA-mqp6-7pv6-fqjf	x_refsource_CONFIRM
https://github.com/ggerganov/llama.cpp/commit/b72942fac998672a79a1ae3c03b340f7e629980b	x_refsource_MISC

Sources

Source Name	Source URL
NIST	https://nvd.nist.gov/vuln/detail/CVE-2024-42477
MITRE	https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2024-42477

History

Created	Old Value	New Value	Data Type	Notes
2024-08-13 13:09:55				Added to TrackCVE