Affected: GNU cpio

GNU cpio contains a Path Traversal vulnerability in its tar archive extraction functionality. When extracting a tar archive in copy-in mode with the --no-absolute-filenames option, the extracted file name is normalized but the tar hard-link target is passed to the link_to_name function without equivalent sanitization before calling link function. A tar archive provided by an attacker, containing a hard-link entry whose linkname is set to an absolute path outside the extraction directory, can cause cpio to create a hard link to an existing file outside the intended extraction directory, breaking the expected guarantee of --no-absolute-filenames and allowing archive-controlled linkage to external files.

This issue has been fixed in commit e2b9cbdd3354d2b1569b7390d1bc15c1930559ad

Affected: Ubuntu:Pro:14.04:LTS: cpio, Ubuntu:Pro:16.04:LTS: cpio, Ubuntu:Pro:18.04:LTS: cpio, Ubuntu:Pro:20.04:LTS: cpio, Ubuntu:22.04:LTS: cpio, Ubuntu:24.04:LTS: cpio, Ubuntu:26.04:LTS: cpio

It was discovered that cpio incorrectly sanitized hard-link targets when extracting tar archives in copy-in mode. If a user or automated system were tricked into extracting a specially crafted tar archive, an attacker could possibly use this issue to create hard links to files outside the extraction directory, even when using the --no-absolute-filenames option. (CVE-2026-66484)

It was discovered that cpio did not properly bound the stack memory allocated for pathnames during archive extraction. If a user or automated system were tricked into extracting a specially crafted cpio archive, an attacker could possibly use this issue to cause cpio to crash, resulting in a denial of service. (CVE-2026-66485)

It was discovered that cpio did not properly escape archive member names when listing archive contents. If a user or automated system were tricked into listing a specially crafted archive, an attacker could possibly use this issue to inject misleading output or malicious terminal control sequences. (CVE-2026-66486)