CVE-2024-42479 | llama.cpp provides LLM inference in C/C++. The unsafe `data`… | CVSS 10.0 CRITICAL

Description

llama.cpp provides LLM inference in C/C++. The unsafe `data` pointer member in the `rpc_tensor` structure can cause arbitrary address writing. This vulnerability is fixed in b3561.

Metadata

CVE ID: CVE-2024-42479
State: PUBLISHED
Assigner: GitHub_M
Reserved: 2024-08-02 14:13 UTC
Published: 2024-08-12 15:07 UTC
Last updated: 2024-08-13 13:47 UTC
Primary CWE: CWE-123
CWE-123: Write-what-where Condition
Vendor / Product: ggerganov / llama.cpp
Sources: cve.org · NVD

Severity & Metrics

10.0 CRITICAL CVSS 3.1

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

SSVC — CISA Coordinator

Exploitation: PoC
Automatable: no
Tech. Impact: total

Affected products (1)

Vendor	Product	Platform	Versions
ggerganov	llama.cpp	—	< b3561

Weakness (CWE)

CWE	Source	Description
CWE-123	cna	CWE-123: Write-what-where Condition

CVSS scores (1)

Score	Severity	Version	Source	Vector
10.0	CRITICAL	3.1	cna	CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

References (2)

https://github.com/ggerganov/llama.cpp/security/advisories/GHSA-wcr5-566p-9cwj https://github.com/ggerganov/llama.cpp/security/advisories/GHSA-wcr5-566p-9cwj
https://github.com/ggerganov/llama.cpp/commit/b72942fac998672a79a1ae3c03b340f7e629980b https://github.com/ggerganov/llama.cpp/commit/b72942fac998672a79a1ae3c03b340f7e629980b