Memory disclosure via unchecked RPC tensor type
Published Aug 12, 2024 · Updated Aug 13, 2024
Out-of-bounds read in llama.cpp before b3561 allows remote attackers to disclose process memory through a crafted RPC GET_TENSOR request. The RPC server passes an attacker-controlled rpc_tensor.type through tensor construction to ggml_type_size, which indexes the global type_traits table without validating the enum range. The flaw is reachable without authentication only when the experimental RPC backend is built and its TCP service is accessible; the documented consequence is memory disclosure.
Summary
What happened
Out-of-bounds read in llama.cpp before b3561 allows remote attackers to disclose process memory through a crafted RPC GET_TENSOR request. The RPC server passes an attacker-controlled rpc_tensor.type through tensor construction to ggml_type_size, which indexes the global type_traits table without validating the enum range. The flaw is reachable without authentication only when the experimental RPC backend is built and its TCP service is accessible; the documented consequence is memory disclosure.
The record
- CVE
- CVE-2024-42477
- Published
- Aug 12, 2024
- Updated
- Aug 13, 2024
- Vendor
- Georgi Gerganov
- Product
- llama.cpp
- Classifications
- CWE-125, T1210
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Aug 12, 2024CVE publishedPublication date reported by the CVE source.
- Aug 13, 2024Record 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 <b3561
- Affected versionversion=< b3561
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
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.
Public exploit references
- Upstream RPC GET_TENSOR proof of conceptproof 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