kernel/linux.git/security/selinux/include/security.h, branch v6.19.12

memfd,selinux: call security_inode_init_security_anon()

2025-10-22T23:28:27+00:00

Prior to this change, no security hooks were called at the creation of a memfd file. It means that, for SELinux as an example, it will receive the default type of the filesystem that backs the in-memory inode. In most cases, that would be tmpfs, but if MFD_HUGETLB is passed, it will be hugetlbfs. Both can be considered implementation details of memfd. It also means that it is not possible to differentiate between a file coming from memfd_create and a file coming from a standard tmpfs mount point. Additionally, no permission is validated at creation, which differs from the similar memfd_secret syscall. Call security_inode_init_security_anon during creation. This ensures that the file is setup similarly to other anonymous inodes. On SELinux, it means that the file will receive the security context of its task. The ability to limit fexecve on memfd has been of interest to avoid potential pitfalls where /proc/self/exe or similar would be executed [1][2]. Reuse the "execute_no_trans" and "entrypoint" access vectors, similarly to the file class. These access vectors may not make sense for the existing "anon_inode" class. Therefore, define and assign a new class "memfd_file" to support such access vectors. Guard these changes behind a new policy capability named "memfd_class". [1] https://crbug.com/1305267 [2] https://lore.kernel.org/lkml/20221215001205.51969-1-jeffxu@google.com/ Signed-off-by: Thiébaud Weksteen Reviewed-by: Stephen Smalley Tested-by: Stephen Smalley Acked-by: Hugh Dickins [PM: subj tweak] Signed-off-by: Paul Moore

selinux: enable per-file labeling for functionfs

2025-09-07T16:54:56+00:00

This patch adds support for genfscon per-file labeling of functionfs files as well as support for userspace to apply labels after new functionfs endpoints are created. This allows for separate labels and therefore access control on a per-endpoint basis. An example use case would be for the default endpoint EP0 used as a restricted control endpoint, and additional usb endpoints to be used by other more permissive domains. It should be noted that if there are multiple functionfs mounts on a system, genfs file labels will apply to all mounts, and therefore will not likely be as useful as the userspace relabeling portion of this patch - the addition to selinux_is_genfs_special_handling(). This patch introduces the functionfs_seclabel policycap to maintain existing functionfs genfscon behavior unless explicitly enabled. Signed-off-by: Neill Kapron Acked-by: Stephen Smalley [PM: trim changelog, apply boolean logic fixup] Signed-off-by: Paul Moore

selinux: Remove unused function selinux_policycap_netif_wildcard()

2025-08-11T15:59:11+00:00

This is unused since commit a3d3043ef24a ("selinux: get netif_wildcard policycap from policy instead of cache"). Signed-off-by: Yue Haibing Acked-by: Stephen Smalley Signed-off-by: Paul Moore

selinux: introduce neveraudit types

2025-06-19T21:23:04+00:00

Introduce neveraudit types i.e. types that should never trigger audit messages. This allows the AVC to skip all audit-related processing for such types. Note that neveraudit differs from dontaudit not only wrt being applied for all checks with a given source type but also in that it disables all auditing, not just permission denials. When a type is both a permissive type and a neveraudit type, the security server can short-circuit the security_compute_av() logic, allowing all permissions and not auditing any permissions. This change just introduces the basic support but does not yet further optimize the AVC or hook function logic when a type is both a permissive type and a dontaudit type. Suggested-by: Paul Moore Signed-off-by: Stephen Smalley Signed-off-by: Paul Moore

selinux: constify network address pointer

2025-04-11T20:29:50+00:00

The network address, either an IPv4 or IPv6 one, is not modified. Signed-off-by: Christian Göttsche Signed-off-by: Paul Moore

selinux: support wildcard network interface names

2025-03-07T20:11:10+00:00

Add support for wildcard matching of network interface names. This is useful for auto-generated interfaces, for example podman creates network interfaces for containers with the naming scheme podman0, podman1, podman2, ... To maintain backward compatibility guard this feature with a new policy capability 'netif_wildcard'. Netifcon definitions are compared against in the order given by the policy, so userspace tools should sort them in a reasonable order. Signed-off-by: Christian Göttsche Signed-off-by: Paul Moore

Merge tag 'selinux-pr-20250121' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux

2025-01-22T04:09:14+00:00

Pull selinux updates from Paul Moore: - Extended permissions supported in conditional policy The SELinux extended permissions, aka "xperms", allow security admins to target individuals ioctls, and recently netlink messages, with their SELinux policy. Adding support for conditional policies allows admins to toggle the granular xperms using SELinux booleans, helping pave the way for greater use of xperms in general purpose SELinux policies. This change bumps the maximum SELinux policy version to 34. - Fix a SCTP/SELinux error return code inconsistency Depending on the loaded SELinux policy, specifically it's EXTSOCKCLASS support, the bind(2) LSM/SELinux hook could return different error codes due to the SELinux code checking the socket's SELinux object class (which can vary depending on EXTSOCKCLASS) and not the socket's sk_protocol field. We fix this by doing the obvious, and looking at the sock->sk_protocol field instead of the object class. - Makefile fixes to properly cleanup av_permissions.h Add av_permissions.h to "targets" so that it is properly cleaned up using the kbuild infrastructure. - A number of smaller improvements by Christian Göttsche A variety of straightforward changes to reduce code duplication, reduce pointer lookups, migrate void pointers to defined types, simplify code, constify function parameters, and correct iterator types. * tag 'selinux-pr-20250121' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux: selinux: make more use of str_read() when loading the policy selinux: avoid unnecessary indirection in struct level_datum selinux: use known type instead of void pointer selinux: rename comparison functions for clarity selinux: rework match_ipv6_addrmask() selinux: constify and reconcile function parameter names selinux: avoid using types indicating user space interaction selinux: supply missing field initializers selinux: add netlink nlmsg_type audit message selinux: add support for xperms in conditional policies selinux: Fix SCTP error inconsistency in selinux_socket_bind() selinux: use native iterator types selinux: add generated av_permissions.h to targets

selinux: constify and reconcile function parameter names

2025-01-08T04:14:38+00:00

Align the parameter names between declarations and definitions, and constify read-only parameters. Signed-off-by: Christian Göttsche [PM: tweak the subject line] Signed-off-by: Paul Moore

selinux: match extended permissions to their base permissions

2025-01-05T01:58:46+00:00

In commit d1d991efaf34 ("selinux: Add netlink xperm support") a new extended permission was added ("nlmsg"). This was the second extended permission implemented in selinux ("ioctl" being the first one). Extended permissions are associated with a base permission. It was found that, in the access vector cache (avc), the extended permission did not keep track of its base permission. This is an issue for a domain that is using both extended permissions (i.e., a domain calling ioctl() on a netlink socket). In this case, the extended permissions were overlapping. Keep track of the base permission in the cache. A new field "base_perm" is added to struct extended_perms_decision to make sure that the extended permission refers to the correct policy permission. A new field "base_perms" is added to struct extended_perms to quickly decide if extended permissions apply. While it is in theory possible to retrieve the base permission from the access vector, the same base permission may not be mapped to the same bit for each class (e.g., "nlmsg" is mapped to a different bit for "netlink_route_socket" and "netlink_audit_socket"). Instead, use a constant (AVC_EXT_IOCTL or AVC_EXT_NLMSG) provided by the caller. Fixes: d1d991efaf34 ("selinux: Add netlink xperm support") Signed-off-by: Thiébaud Weksteen Signed-off-by: Paul Moore

selinux: add support for xperms in conditional policies

2024-12-13T21:35:38+00:00

Add support for extended permission rules in conditional policies. Currently the kernel accepts such rules already, but evaluating a security decision will hit a BUG() in services_compute_xperms_decision(). Thus reject extended permission rules in conditional policies for current policy versions. Add a new policy version for this feature. Signed-off-by: Christian Göttsche Acked-by: Stephen Smalley Tested-by: Stephen Smalley Signed-off-by: Paul Moore