libdav1d CVEs in Older Decoder Versions

This article provides a concise overview of the Common Vulnerabilities and Exposures (CVEs) associated with older versions of libdav1d, the popular open-source AV1 video decoder. It details specific historical security flaws, including memory corruption and denial-of-service risks, and outlines how these vulnerabilities impact software relying on the library.

libdav1d is an extremely fast, cross-platform AV1 decoder developed by VideoLAN and the VideoLAN project community. Because it is written primarily in C and assembly to maximize performance, older versions of the library have historically been susceptible to memory safety issues. Software developers integrate libdav1d into web browsers, media players, and operating systems, meaning vulnerabilities within the library can have widespread downstream effects.

Several notable CVEs have been documented for older versions of libdav1d:

The primary impact of these vulnerabilities is client-side exploitation. Since libdav1d is commonly used in web browsers (such as Firefox and Chrome) and media engines to decode streaming video, an attacker could host a malicious AV1 file on a website. When a user with an outdated browser or media player attempts to play the video, the vulnerability triggers, resulting in a application crash (Denial of Service) or, in worse cases, remote code execution on the user’s system.

To mitigate these security risks, developers and system administrators must ensure that libdav1d is regularly updated to the latest stable release. The VideoLAN development team actively patches these memory-related bugs in newer versions, making upgrading the most effective defense against known exploits.