CVE-2024-42479

llama.cpp allows write-what-where in rpc_server::set_tensor

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.

Affected

ggerganov · llama.cpp

CVSS v3.1 NVD

9.8 CRITICALCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

CVSS v3.1 GITHUB

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

EPSS

2.61% chance of exploitation in the next 30 days, 85th percentile

Published

2024-08-12

Updated

2024-08-13

Proof-of-concept exploits (1)

• ggerganov/llama.cpp/security/advisories/GHSA-wcr5-566p-9cwj

Security advisories

• ggerganov/llama.cpp/commit/b72942fac998672a79a1ae3c03b340f7e629980b

References

• CVE Record
• NVD entry