Freebsd

Concurrent removals of certain anonymous shared memory mappings by using the UMTX_SHM_DESTROY sub-request of UMTX_OP_SHM can lead to decreasing the reference count of the object representing the mapping too many times, causing it to be freed too early. A malicious code exercizing the UMTX_SHM_DESTROY sub-request in parallel can panic the kernel or enable further Use-After-Free attacks, potentially including code execution or Capsicum sandbox escape.Show less

The ctl_report_supported_opcodes function did not sufficiently validate a field provided by userspace, allowing an arbitrary write to a limited amount of kernel help memory. Malicious software running in a guest VM that exposes virtio_scsi can exploit the vulnerabilities to achieve code execution on the host in the bhyve userspace process, which typically runs as root. Note that bhyve runs in a Capsicum sandbox, so malicious code is constrained by the capabilities available to the bhyve process. A malicious iSCSI initiator could achieve remote code execution on the iSCSI target host.Show less

An insufficient boundary validation in the USB code could lead to an out-of-bounds write on the heap, with data controlled by the caller. A malicious, privileged software running in a guest VM can exploit the vulnerability to achieve code execution on the host in the bhyve userspace process, which typically runs as root. Note that bhyve runs in a Capsicum sandbox, so malicious code is constrained by the capabilities available to the bhyve process.Show less

A signal handler in sshd(8) may call a logging function that is not async-signal-safe. The signal handler is invoked when a client does not authenticate within the LoginGraceTime seconds (120 by default). This signal handler executes in the context of the sshd(8)'s privileged code, which is not sandboxed and runs with full root privileges. This issue is another instance of the problem in CVE-2024-6387 addressed by FreeBSD-SA-24:04.openssh. The faulty code in this case is from the integration of blacklistd in OpenSSH in FreeBSD. As a result of calling functions that are not async-signal-safe in the privileged sshd(8) context, a race condition exists that a determined attacker may be able to exploit to allow an unauthenticated remote code execution as root.Show less

A logic bug in the code which disables kernel tracing for setuid programs meant that tracing was not disabled when it should have, allowing unprivileged users to trace and inspect the behavior of setuid programs. The bug may be used by an unprivileged user to read the contents of files to which they would not otherwise have access, such as the local password database.Show less

When mounting a remote filesystem using NFS, the kernel did not sanitize remotely provided filenames for the path separator character, "/". This allows readdir(3) and related functions to return filesystem entries with names containing additional path components. The lack of validation described above gives rise to a confused deputy problem. For example, a program copying files from an NFS mount could be tricked into copying from outside the intended source directory, and/or to a location outside the intended destination directory.Show less

A security regression (CVE-2006-5051) was discovered in OpenSSH's server (sshd). There is a race condition which can lead sshd to handle some signals in an unsafe manner. An unauthenticated, remote attacker may be able to trigger it by failing to authenticate within a set time period.Show less

ping reads raw IP packets from the network to process responses in the pr_pack() function. As part of processing a response ping has to reconstruct the IP header, the ICMP header and if present a "quoted packet," which represents the packet that generated an ICMP error. The quoted packet again has an IP header and an ICMP header. The pr_pack() copies received IP and ICMP headers into stack buffers for further processing. In so doing, it fails to take into account the possible presence of IP option headers following the IP header in either the response or the quoted packet. When IP options are present, pr_pack() overflows the destination buffer by up to 40 bytes. The memory safety bugs described above can be triggered by a remote host, causing the ping program to crash. The ping process runs in a capability mode sandbox on all affected versions of FreeBSD and is thus very constrained in how it can interact with the rest of the system at the point where the bug can occur.Show less

The implementation of lib9p's handling of RWALK messages was missing a bounds check needed when unpacking the message contents. The missing check means that the receipt of a specially crafted message will cause lib9p to overwrite unrelated memory. The bug can be triggered by a malicious bhyve guest kernel to overwrite memory in the bhyve(8) process. This could potentially lead to user-mode code execution on the host, subject to bhyve's Capsicum sandbox.Show less

A particular case of memory sharing is mishandled in the virtual memory system. This is very similar to SA-21:08.vm, but with a different root cause. An unprivileged local user process can maintain a mapping of a page after it is freed, allowing that process to read private data belonging to other processes or the kernel.Show less

The jail(2) system call has not limited a visiblity of allocated TTYs (the kern.ttys sysctl). This gives rise to an information leak about processes outside the current jail. Attacker can get information about TTYs allocated on the host or in other jails. Effectively, the information printed by "pstat -t" may be leaked.Show less

`bhyveload -h <host-path>` may be used to grant loader access to the <host-path> directory tree on the host. Affected versions of bhyveload(8) do not make any attempt to restrict loader's access to <host-path>, allowing the loader to read any file the host user has access to. In the bhyveload(8) model, the host supplies a userboot.so to boot with, but the loader scripts generally come from the guest image. A maliciously crafted script could be used to exfiltrate sensitive data from the host accessible to the user running bhyhveload(8), which is often the system root.Show less

When dumping core and saving process information, proc_getargv() might return an sbuf which have a sbuf_len() of 0 or -1, which is not properly handled. An out-of-bound read can happen when user constructs a specially crafted ps_string, which in turn can cause the kernel to crash.Show less

The 802.11 beacon handling routine failed to validate the length of an IEEE 802.11s Mesh ID before copying it to a heap-allocated buffer. While a FreeBSD Wi-Fi client is in scanning mode (i.e., not associated with a SSID) a malicious beacon frame may overwrite kernel memory, leading to remote code execution.Show less

The e1000 network adapters permit a variety of modifications to an Ethernet packet when it is being transmitted. These include the insertion of IP and TCP checksums, insertion of an Ethernet VLAN header, and TCP segmentation offload ("TSO"). The e1000 device model uses an on-stack buffer to generate the modified packet header when simulating these modifications on transmitted packets. When checksum offload is requested for a transmitted packet, the e1000 device model used a guest-provided value to specify the checksum offset in the on-stack buffer. The offset was not validated for certain packet types. A misbehaving bhyve guest could overwrite memory in the bhyve process on the host, possibly leading to code execution in the host context. The bhyve process runs in a Capsicum sandbox, which (depending on the FreeBSD version and bhyve configuration) limits the impact of exploiting this issue.Show less

Handlers for *_CFG_PAGE read / write ioctls in the mpr, mps, and mpt drivers allocated a buffer of a caller-specified size, but copied to it a fixed size header. Other heap content would be overwritten if the specified size was too small. Users with access to the mpr, mps or mpt device node may overwrite heap data, potentially resulting in privilege escalation. Note that the device node is only accessible to root and members of the operator group.Show less

A user-provided integer option was passed to nmreq_copyin() without checking if it would overflow. This insufficient bounds checking could lead to kernel memory corruption. On systems configured to include netmap in their devfs_ruleset, a privileged process running in a jail can affect the host environment.Show less