An issue was discovered in the Linux kernel before 4.18.7. In create_qp_common in drivers/infiniband/hw/mlx5/qp.c, mlx5_ib_create_qp_resp was never initialized, resulting in a leak of stack memory to userspace.

In the Linux kernel, the following vulnerability has been resolved: icmp: change the order of rate limits ICMP messages are ratelimited : After the blamed commits, the two rate limiters are applied in this order: 1) host wide ratelimit (icmp_global_allow()) 2) Per destination ratelimit (inetpeer based) In order to avoid side-channels attacks, we need to apply the per destination check first. This patch makes the following change : 1) icmp_global_allow() checks if the host wide limit is reached. But credits are not yet consumed. This is deferred to 3) 2) The per destination limit is checked/updated. This might add a new node in inetpeer tree. 3) icmp_global_consume() consumes tokens if prior operations succeeded. This means that host wide ratelimit is still effective in keeping inetpeer tree small even under DDOS. As a bonus, I removed icmp_global.lock as the fast path can use a lock-free operation.

The copy_from_user function in the uaccess code in Linux kernel 2.6 before 2.6.19-rc1, when running on s390, does not properly clear a kernel buffer, which allows local user space programs to read portions of kernel memory by "appending to a file from a bad address," which triggers a fault that prevents the unused memory from being cleared in the kernel buffer.

The spectre_v2_select_mitigation function in arch/x86/kernel/cpu/bugs.c in the Linux kernel before 4.18.1 does not always fill RSB upon a context switch, which makes it easier for attackers to conduct userspace-userspace spectreRSB attacks.

In the Linux kernel, the following vulnerability has been resolved: gpio: prevent potential speculation leaks in gpio_device_get_desc() Userspace may trigger a speculative read of an address outside the gpio descriptor array. Users can do that by calling gpio_ioctl() with an offset out of range. Offset is copied from user and then used as an array index to get the gpio descriptor without sanitization in gpio_device_get_desc(). This change ensures that the offset is sanitized by using array_index_nospec() to mitigate any possibility of speculative information leaks. This bug was discovered and resolved using Coverity Static Analysis Security Testing (SAST) by Synopsys, Inc.

In the Linux kernel, the following vulnerability has been resolved: fuse: Initialize beyond-EOF page contents before setting uptodate fuse_notify_store(), unlike fuse_do_readpage(), does not enable page zeroing (because it can be used to change partial page contents). So fuse_notify_store() must be more careful to fully initialize page contents (including parts of the page that are beyond end-of-file) before marking the page uptodate. The current code can leave beyond-EOF page contents uninitialized, which makes these uninitialized page contents visible to userspace via mmap(). This is an information leak, but only affects systems which do not enable init-on-alloc (via CONFIG_INIT_ON_ALLOC_DEFAULT_ON=y or the corresponding kernel command line parameter).

The __block_prepare_write function in fs/buffer.c for Linux kernel 2.6.x before 2.6.13 does not properly clear buffers during certain error conditions, which allows local users to read portions of files that have been unlinked.

The fill_event_metadata function in fs/notify/fanotify/fanotify_user.c in the Linux kernel through 3.9.4 does not initialize a certain structure member, which allows local users to obtain sensitive information from kernel memory via a read operation on the fanotify descriptor.

A use-after-free vulnerability was found in iscsi_sw_tcp_session_create in drivers/scsi/iscsi_tcp.c in SCSI sub-component in the Linux Kernel. In this flaw an attacker could leak kernel internal information.

An information disclosure vulnerability exists in the ARM SIGPAGE functionality of Linux Kernel v5.4.66 and v5.4.54. The latest version (5.11-rc4) seems to still be vulnerable. A userland application can read the contents of the sigpage, which can leak kernel memory contents. An attacker can read a process’s memory at a specific offset to trigger this vulnerability. This was fixed in kernel releases: 4.14.222 4.19.177 5.4.99 5.10.17 5.11

IBM GSKit (IBM DB2 for Linux, UNIX and Windows 9.7, 10.1, 10.5, and 11.1) uses weaker than expected cryptographic algorithms that could allow an attacker to decrypt highly sensitive information. IBM X-Force ID: 139073.

A missing address check in the callers of the show_opcodes() in the Linux kernel allows an attacker to dump the kernel memory at an arbitrary kernel address into the dmesg log.

In the Linux kernel, the following vulnerability has been resolved: bonding: Fix unnecessary warnings and logs from bond_xdp_get_xmit_slave() syzbot reported a WARNING in bond_xdp_get_xmit_slave. To reproduce this[1], one bond device (bond1) has xdpdrv, which increases bpf_master_redirect_enabled_key. Another bond device (bond0) which is unsupported by XDP but its slave (veth3) has xdpgeneric that returns XDP_TX. This triggers WARN_ON_ONCE() from the xdp_master_redirect(). To reduce unnecessary warnings and improve log management, we need to delete the WARN_ON_ONCE() and add ratelimit to the netdev_err(). [1] Steps to reproduce: # Needs tx_xdp with return XDP_TX; ip l add veth0 type veth peer veth1 ip l add veth3 type veth peer veth4 ip l add bond0 type bond mode 6 # BOND_MODE_ALB, unsupported by XDP ip l add bond1 type bond # BOND_MODE_ROUNDROBIN by default ip l set veth0 master bond1 ip l set bond1 up # Increases bpf_master_redirect_enabled_key ip l set dev bond1 xdpdrv object tx_xdp.o section xdp_tx ip l set veth3 master bond0 ip l set bond0 up ip l set veth4 up # Triggers WARN_ON_ONCE() from the xdp_master_redirect() ip l set veth3 xdpgeneric object tx_xdp.o section xdp_tx

In the Linux kernel, the following vulnerability has been resolved: USB: usbtmc: prevent kernel-usb-infoleak The syzbot reported a kernel-usb-infoleak in usbtmc_write, we need to clear the structure before filling fields.

Overlayfs did not properly perform permission checking when copying up files in an overlayfs and could be exploited from within a user namespace, if, for example, unprivileged user namespaces were allowed. It was possible to have a file not readable by an unprivileged user to be copied to a mountpoint controlled by the user, like a removable device. This was introduced in kernel version 4.19 by commit d1d04ef ("ovl: stack file ops"). This was fixed in kernel version 5.8 by commits 56230d9 ("ovl: verify permissions in ovl_path_open()"), 48bd024 ("ovl: switch to mounter creds in readdir") and 05acefb ("ovl: check permission to open real file"). Additionally, commits 130fdbc ("ovl: pass correct flags for opening real directory") and 292f902 ("ovl: call secutiry hook in ovl_real_ioctl()") in kernel 5.8 might also be desired or necessary. These additional commits introduced a regression in overlay mounts within user namespaces which prevented access to files with ownership outside of the user namespace. This regression was mitigated by subsequent commit b6650da ("ovl: do not fail because of O_NOATIMEi") in kernel 5.11.

Directory traversal vulnerability in CIFS in Linux 2.6.16 and earlier allows local users to escape chroot restrictions for an SMB-mounted filesystem via "..\\" sequences, a similar vulnerability to CVE-2006-1864.

net/ipv4/af_inet.c in Linux kernel 2.4 does not clear sockaddr_in.sin_zero before returning IPv4 socket names from the (1) getsockname, (2) getpeername, and (3) accept functions, which allows local users to obtain portions of potentially sensitive memory.

The Linux kernel before 2.6.16.9 and the FreeBSD kernel, when running on AMD64 and other 7th and 8th generation AuthenticAMD processors, only save/restore the FOP, FIP, and FDP x87 registers in FXSAVE/FXRSTOR when an exception is pending, which allows one process to determine portions of the state of floating point instructions of other processes, which can be leveraged to obtain sensitive information such as cryptographic keys. NOTE: this is the documented behavior of AMD64 processors, but it is inconsistent with Intel processors in a security-relevant fashion that was not addressed by the kernels.

Copy_from_user on 64-bit versions of the Linux kernel does not implement the __uaccess_begin_nospec allowing a user to bypass the "access_ok" check and pass a kernel pointer to copy_from_user(). This would allow an attacker to leak information. We recommend upgrading beyond commit 74e19ef0ff8061ef55957c3abd71614ef0f42f47

net/ipv4/netfilter/ip_conntrack_core.c in Linux kernel 2.4 and 2.6, and possibly net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c in 2.6, does not clear sockaddr_in.sin_zero before returning IPv4 socket names from the getsockopt function with SO_ORIGINAL_DST, which allows local users to obtain portions of potentially sensitive memory.

dm-crypt in Linux kernel 2.6.15 and earlier does not clear a structure before it is freed, which leads to a memory disclosure that could allow local users to obtain sensitive information about a cryptographic key.

In the Linux kernel, the following vulnerability has been resolved: swiotlb: fix info leak with DMA_FROM_DEVICE The problem I'm addressing was discovered by the LTP test covering cve-2018-1000204. A short description of what happens follows: 1) The test case issues a command code 00 (TEST UNIT READY) via the SG_IO interface with: dxfer_len == 524288, dxdfer_dir == SG_DXFER_FROM_DEV and a corresponding dxferp. The peculiar thing about this is that TUR is not reading from the device. 2) In sg_start_req() the invocation of blk_rq_map_user() effectively bounces the user-space buffer. As if the device was to transfer into it. Since commit a45b599ad808 ("scsi: sg: allocate with __GFP_ZERO in sg_build_indirect()") we make sure this first bounce buffer is allocated with GFP_ZERO. 3) For the rest of the story we keep ignoring that we have a TUR, so the device won't touch the buffer we prepare as if the we had a DMA_FROM_DEVICE type of situation. My setup uses a virtio-scsi device and the buffer allocated by SG is mapped by the function virtqueue_add_split() which uses DMA_FROM_DEVICE for the "in" sgs (here scatter-gather and not scsi generics). This mapping involves bouncing via the swiotlb (we need swiotlb to do virtio in protected guest like s390 Secure Execution, or AMD SEV). 4) When the SCSI TUR is done, we first copy back the content of the second (that is swiotlb) bounce buffer (which most likely contains some previous IO data), to the first bounce buffer, which contains all zeros. Then we copy back the content of the first bounce buffer to the user-space buffer. 5) The test case detects that the buffer, which it zero-initialized, ain't all zeros and fails. One can argue that this is an swiotlb problem, because without swiotlb we leak all zeros, and the swiotlb should be transparent in a sense that it does not affect the outcome (if all other participants are well behaved). Copying the content of the original buffer into the swiotlb buffer is the only way I can think of to make swiotlb transparent in such scenarios. So let's do just that if in doubt, but allow the driver to tell us that the whole mapped buffer is going to be overwritten, in which case we can preserve the old behavior and avoid the performance impact of the extra bounce.

In the Linux kernel, the following vulnerability has been resolved: KEYS: trusted: dcp: fix leak of blob encryption key Trusted keys unseal the key blob on load, but keep the sealed payload in the blob field so that every subsequent read (export) will simply convert this field to hex and send it to userspace. With DCP-based trusted keys, we decrypt the blob encryption key (BEK) in the Kernel due hardware limitations and then decrypt the blob payload. BEK decryption is done in-place which means that the trusted key blob field is modified and it consequently holds the BEK in plain text. Every subsequent read of that key thus send the plain text BEK instead of the encrypted BEK to userspace. This issue only occurs when importing a trusted DCP-based key and then exporting it again. This should rarely happen as the common use cases are to either create a new trusted key and export it, or import a key blob and then just use it without exporting it again. Fix this by performing BEK decryption and encryption in a dedicated buffer. Further always wipe the plain text BEK buffer to prevent leaking the key via uninitialized memory.

In the Linux kernel, the following vulnerability has been resolved: iavf: Fix error handling in iavf_init_module() The iavf_init_module() won't destroy workqueue when pci_register_driver() failed. Call destroy_workqueue() when pci_register_driver() failed to prevent the resource leak. Similar to the handling of u132_hcd_init in commit f276e002793c ("usb: u132-hcd: fix resource leak")

drm.c in Linux kernel 2.6.10 to 2.6.13 creates a debug file in sysfs with world-readable and world-writable permissions, which allows local users to enable DRM debugging and obtain sensitive information.

In the Linux kernel, the following vulnerability has been resolved: igb: Initialize mailbox message for VF reset When a MAC address is not assigned to the VF, that portion of the message sent to the VF is not set. The memory, however, is allocated from the stack meaning that information may be leaked to the VM. Initialize the message buffer to 0 so that no information is passed to the VM in this case.

In the Linux kernel, the following vulnerability has been resolved: netfilter: nfnetlink_osf: fix possible bogus match in nf_osf_find() nf_osf_find() incorrectly returns true on mismatch, this leads to copying uninitialized memory area in nft_osf which can be used to leak stale kernel stack data to userspace.

cryptoloop on Linux kernel 2.6.x, when used on certain file systems with a block size 1024 or greater, has certain "IV computation" weaknesses that allow watermarked files to be detected without decryption.

Signedness error in the copy_from_read_buf function in n_tty.c for Linux kernel 2.6.10 and 2.6.11rc1 allows local users to read kernel memory via a negative argument.

A "potential" buffer overflow exists in the panic() function in Linux 2.4.x, although it may not be exploitable due to the functionality of panic.

IBM Security Guardium 11.2 stores user credentials in plain clear text which can be read by a local user. IBM X-Force ID: 195770.

Linux kernel 2.6.10 and 2.6.11rc1-bk6 uses different size types for offset arguments to the proc_file_read and locks_read_proc functions, which leads to a heap-based buffer overflow when a signed comparison causes negative integers to be used in a positive context.

The JFS file system code in Linux 2.4.x has an information leak in which in-memory data is written to the device for the JFS file system, which allows local users to obtain sensitive information by reading the raw device.

Mozilla Firefox 4.x through 9.0 and SeaMonkey before 2.7 on Linux and Mac OS X set weak permissions for Firefox Recovery Key.html, which might allow local users to read a Firefox Sync key via standard filesystem operations.

A numeric casting discrepancy in sdla_xfer in Linux kernel 2.6.x up to 2.6.5 and 2.4 up to 2.4.29-rc1 allows local users to read portions of kernel memory via a large len argument, which is received as an int but cast to a short, which prevents a read loop from filling a buffer.

NVIDIA GPU and Tegra hardware contain a vulnerability in the internal microcontroller which may allow a user with elevated privileges to utilize debug mechanisms with insufficient access control, which may lead to information disclosure.

The osf_getsysinfo function in arch/alpha/kernel/osf_sys.c in the Linux kernel before 2.6.39.4 on the Alpha platform does not properly restrict the data size for GSI_GET_HWRPB operations, which allows local users to obtain sensitive information from kernel memory via a crafted call.

Integer signedness error in the osf_getdomainname function in arch/alpha/kernel/osf_sys.c in the Linux kernel before 2.6.39.4 on the Alpha platform allows local users to obtain sensitive information from kernel memory via a crafted call.

fs/proc/base.c in the Linux kernel before 2.6.39.4 does not properly restrict access to /proc/#####/io files, which allows local users to obtain sensitive I/O statistics by polling a file, as demonstrated by discovering the length of another user's password.

The open_exec function in the execve functionality (exec.c) in Linux kernel 2.4.x up to 2.4.27, and 2.6.x up to 2.6.8, allows local users to read non-readable ELF binaries by using the interpreter (PT_INTERP) functionality.

The osf_partition function in fs/partitions/osf.c in the Linux kernel before 2.6.38 does not properly handle an invalid number of partitions, which might allow local users to obtain potentially sensitive information from kernel heap memory via vectors related to partition-table parsing.

Linux kernel does not properly convert 64-bit file offset pointers to 32 bits, which allows local users to access portions of kernel memory.

The e1000 driver for Linux kernel 2.4.26 and earlier does not properly initialize memory before using it, which allows local users to read portions of kernel memory. NOTE: this issue was originally incorrectly reported as a "buffer overflow" by some sources.

The do_task_stat function in fs/proc/array.c in the Linux kernel before 2.6.39-rc1 does not perform an expected uid check, which makes it easier for local users to defeat the ASLR protection mechanism by reading the start_code and end_code fields in the /proc/#####/stat file for a process executing a PIE binary.

The do_replace function in net/bridge/netfilter/ebtables.c in the Linux kernel before 2.6.39 does not ensure that a certain name field ends with a '\0' character, which allows local users to obtain potentially sensitive information from kernel stack memory by leveraging the CAP_NET_ADMIN capability to replace a table, and then reading a modprobe command line.

The XFS file system code in Linux 2.4.x has an information leak in which in-memory data is written to the device for the XFS file system, which allows local users to obtain sensitive information by reading the raw device.

In the Linux kernel, the following vulnerability has been resolved: ext4: fix mb_cache_entry's e_refcnt leak in ext4_xattr_block_cache_find() Syzbot reports a warning as follows: ============================================ WARNING: CPU: 0 PID: 5075 at fs/mbcache.c:419 mb_cache_destroy+0x224/0x290 Modules linked in: CPU: 0 PID: 5075 Comm: syz-executor199 Not tainted 6.9.0-rc6-gb947cc5bf6d7 RIP: 0010:mb_cache_destroy+0x224/0x290 fs/mbcache.c:419 Call Trace: <TASK> ext4_put_super+0x6d4/0xcd0 fs/ext4/super.c:1375 generic_shutdown_super+0x136/0x2d0 fs/super.c:641 kill_block_super+0x44/0x90 fs/super.c:1675 ext4_kill_sb+0x68/0xa0 fs/ext4/super.c:7327 [...] ============================================ This is because when finding an entry in ext4_xattr_block_cache_find(), if ext4_sb_bread() returns -ENOMEM, the ce's e_refcnt, which has already grown in the __entry_find(), won't be put away, and eventually trigger the above issue in mb_cache_destroy() due to reference count leakage. So call mb_cache_entry_put() on the -ENOMEM error branch as a quick fix.

The ib_uverbs_poll_cq function in drivers/infiniband/core/uverbs_cmd.c in the Linux kernel before 2.6.37 does not initialize a certain response buffer, which allows local users to obtain potentially sensitive information from kernel memory via vectors that cause this buffer to be only partially filled, a different vulnerability than CVE-2010-4649.

The ocfs2_prepare_page_for_write function in fs/ocfs2/aops.c in the Oracle Cluster File System 2 (OCFS2) subsystem in the Linux kernel before 2.6.39-rc1 does not properly handle holes that cross page boundaries, which allows local users to obtain potentially sensitive information from uninitialized disk locations by reading a file.

The /proc filesystem in Linux allows local users to obtain sensitive information by opening various entries in /proc/self before executing a setuid program, which causes the program to fail to change the ownership and permissions of those entries.