CVE-2024-32878

Llama.cpp is LLM inference in C/C++. There is a use of uninitialized heap variable vulnerability in gguf_init_from_file, the code will free this uninitialized variable later. In a simple POC, it will directly cause a crash. If the file is carefully constructed, it may be possible to control this uninitialized value and cause arbitrary address free problems. This may further lead to be exploited. Causes llama.cpp to crash (DoS) and may even lead to arbitrary code execution (RCE). This vulnerability has been patched in commit b2740.

Llama.cpp is LLM inference in C/C++. There is a use of uninitialized heap variable vulnerability in gguf_init_from_file, the code will free this uninitialized variable later. In a simple POC, it will directly cause a crash. If the file is carefully constructed, it may be possible to control this uninitialized value and cause arbitrary address free problems. This may further lead to be exploited. Causes llama.cpp to crash (DoS) and may even lead to arbitrary code execution (RCE). This vulnerability has been patched in commit b2740.

Llama.cpp es una inferencia LLM en C/C++. Hay un uso de vulnerabilidad de variable de montón no inicializada en gguf_init_from_file, el código liberará esta variable no inicializada más adelante. En una POC simple, provocará directamente un bloqueo. Si el archivo se construye cuidadosamente, es posible controlar este valor no inicializado y causar problemas sin direcciones arbitrarias. Esto puede llevar aún más a ser explotado. Provoca que llama.cpp falle (DoS) e incluso puede provocar la ejecución de código arbitrario (RCE). Esta vulnerabilidad ha sido parcheada en el commit b2740.