CVE-2014-7284

The net_get_random_once implementation in net/core/utils.c in the Linux kernel 3.13.x and 3.14.x before 3.14.5 on certain Intel processors does not perform the intended slow-path operation to initialize random seeds, which makes it easier for remote attackers to spoof or disrupt IP communication by leveraging the predictability of TCP sequence numbers, TCP and UDP port numbers, and IP ID values.

This kernel-linus update is based on upstream -longterm 3.14.23 and fixes the following security issues:

The kvm_iommu_map_pages function in virt/kvm/iommu.c in the Linux kernel through 3.16.1 miscalculates the number of pages during the handling of a mapping failure, which allows guest OS users to (1) cause a denial of service (host OS memory corruption) or possibly have unspecified other impact by triggering a large gfn value or (2) cause a denial of service (host OS memory consumption) by triggering a small gfn value that leads to permanently pinned pages (CVE-2014-3601).

The assoc_array_gc function in the associative-array implementation in lib/assoc_array.c in the Linux kernel before 3.16.3 does not properly implement garbage collection, which allows local users to cause a denial of service (NULL pointer dereference and system crash) or possibly have unspecified other impact via multiple "keyctl newring" operations followed by a "keyctl timeout" operation (CVE-2014-3631).

The xfs_da3_fixhashpath function in fs/xfs/xfs_da_btree.c in the xfs implementation in the Linux kernel before 3.14.2 does not properly compare btree hash values, which allows local users to cause a denial of service (filesystem corruption, and OOPS or panic) via operations on directories that have hash collisions, as demonstrated by rmdir operations (CVE-2014-7283).

The net_get_random_once implementation in net/core/utils.c in the Linux kernel 3.13.x and 3.14.x before 3.14.5 on certain Intel processors does not perform the intended slow-path operation to initialize random seeds, which makes it easier for remote attackers to spoof or disrupt IP communication by leveraging the predictability of TCP sequence numbers, TCP and UDP port numbers, and IP ID values (CVE-2014-7284)

The pivot_root implementation in fs/namespace.c in the Linux kernel through 3.17 does not properly interact with certain locations of a chroot directory, which allows local users to cause a denial of service (mount-tree loop) via . (dot) values in both arguments to the pivot_root system call (CVE-2014-7970).

The do_umount function in fs/namespace.c in the Linux kernel through 3.17 does not require the CAP_SYS_ADMIN capability for do_remount_sb calls that change the root filesystem to read-only, which allows local users to cause a denial of service (loss of writability) by making certain unshare system calls, clearing the / MNT_LOCKED flag, and making an MNT_FORCE umount system call (CVE-2014-7975).

For other fixes included in this update, read the referenced changelogs.

This kernel-tmb update is based on upstream -longterm 3.14.23 and fixes the following security issues:

The kvm_iommu_map_pages function in virt/kvm/iommu.c in the Linux kernel through 3.16.1 miscalculates the number of pages during the handling of a mapping failure, which allows guest OS users to (1) cause a denial of service (host OS memory corruption) or possibly have unspecified other impact by triggering a large gfn value or (2) cause a denial of service (host OS memory consumption) by triggering a small gfn value that leads to permanently pinned pages (CVE-2014-3601).

The assoc_array_gc function in the associative-array implementation in lib/assoc_array.c in the Linux kernel before 3.16.3 does not properly implement garbage collection, which allows local users to cause a denial of service (NULL pointer dereference and system crash) or possibly have unspecified other impact via multiple "keyctl newring" operations followed by a "keyctl timeout" operation (CVE-2014-3631).

The xfs_da3_fixhashpath function in fs/xfs/xfs_da_btree.c in the xfs implementation in the Linux kernel before 3.14.2 does not properly compare btree hash values, which allows local users to cause a denial of service (filesystem corruption, and OOPS or panic) via operations on directories that have hash collisions, as demonstrated by rmdir operations (CVE-2014-7283).

The net_get_random_once implementation in net/core/utils.c in the Linux kernel 3.13.x and 3.14.x before 3.14.5 on certain Intel processors does not perform the intended slow-path operation to initialize random seeds, which makes it easier for remote attackers to spoof or disrupt IP communication by leveraging the predictability of TCP sequence numbers, TCP and UDP port numbers, and IP ID values (CVE-2014-7284)

The pivot_root implementation in fs/namespace.c in the Linux kernel through 3.17 does not properly interact with certain locations of a chroot directory, which allows local users to cause a denial of service (mount-tree loop) via . (dot) values in both arguments to the pivot_root system call (CVE-2014-7970).

The do_umount function in fs/namespace.c in the Linux kernel through 3.17 does not require the CAP_SYS_ADMIN capability for do_remount_sb calls that change the root filesystem to read-only, which allows local users to cause a denial of service (loss of writability) by making certain unshare system calls, clearing the / MNT_LOCKED flag, and making an MNT_FORCE umount system call (CVE-2014-7975).

Other fixes: The X86_SYSFB config option has been disabled as it prevents proper KMS setup on some systems (mga#13098)

BFS cpu scheduler has been updated to 0.454.

BFQ io scheduler has been added and set as default.

For other fixes included in this update, read the referenced changelogs.

This kernel-vserver update provides an upgrade to the upstream 3.14 -longterm branch, currently based on 3.14.23 and fixes the following security issues:

The kvm_iommu_map_pages function in virt/kvm/iommu.c in the Linux kernel through 3.16.1 miscalculates the number of pages during the handling of a mapping failure, which allows guest OS users to (1) cause a denial of service (host OS memory corruption) or possibly have unspecified other impact by triggering a large gfn value or (2) cause a denial of service (host OS memory consumption) by triggering a small gfn value that leads to permanently pinned pages (CVE-2014-3601).

The assoc_array_gc function in the associative-array implementation in lib/assoc_array.c in the Linux kernel before 3.16.3 does not properly implement garbage collection, which allows local users to cause a denial of service (NULL pointer dereference and system crash) or possibly have unspecified other impact via multiple "keyctl newring" operations followed by a "keyctl timeout" operation (CVE-2014-3631).

The xfs_da3_fixhashpath function in fs/xfs/xfs_da_btree.c in the xfs implementation in the Linux kernel before 3.14.2 does not properly compare btree hash values, which allows local users to cause a denial of service (filesystem corruption, and OOPS or panic) via operations on directories that have hash collisions, as demonstrated by rmdir operations (CVE-2014-7283).

The net_get_random_once implementation in net/core/utils.c in the Linux kernel 3.13.x and 3.14.x before 3.14.5 on certain Intel processors does not perform the intended slow-path operation to initialize random seeds, which makes it easier for remote attackers to spoof or disrupt IP communication by leveraging the predictability of TCP sequence numbers, TCP and UDP port numbers, and IP ID values (CVE-2014-7284)

The pivot_root implementation in fs/namespace.c in the Linux kernel through 3.17 does not properly interact with certain locations of a chroot directory, which allows local users to cause a denial of service (mount-tree loop) via . (dot) values in both arguments to the pivot_root system call (CVE-2014-7970).

The do_umount function in fs/namespace.c in the Linux kernel through 3.17 does not require the CAP_SYS_ADMIN capability for do_remount_sb calls that change the root filesystem to read-only, which allows local users to cause a denial of service (loss of writability) by making certain unshare system calls, clearing the / MNT_LOCKED flag, and making an MNT_FORCE umount system call (CVE-2014-7975).

Other fixes: The X86_SYSFB config option has been disabled as it prevents proper KMS setup on some systems (mga#13098)

The vserver patch has been updated to vs2.3.6.13

The util-vserver userspace tools has been updated to 0.30.216-pre3062

For other fixes included in this update, read the referenced changelogs.

The net_get_random_once implementation in net/core/utils.c in the Linux kernel 3.13.x and 3.14.x before 3.14.5 on certain Intel processors does not perform the intended slow-path operation to initialize random seeds, which makes it easier for remote attackers to spoof or disrupt IP communication by leveraging the predictability of TCP sequence numbers, TCP and UDP port numbers, and IP ID values.

Sasha Levin reported a flaw in the Linux kernel's point-to-point protocol (PPP) when used with the Layer Two Tunneling Protocol (L2TP). A local user could exploit this flaw to gain administrative privileges. (CVE-2014-4943)

Salva Peiró discovered an information leak in the Linux kernel's media- device driver. A local attacker could exploit this flaw to obtain sensitive information from kernel memory. (CVE-2014-1739)

A bounds check error was discovered in the socket filter subsystem of the Linux kernel. A local user could exploit this flaw to cause a denial of service (system crash) via crafted BPF instructions. (CVE-2014-3144)

A remainder calculation error was discovered in the socket filter subsystem of the Linux kernel. A local user could exploit this flaw to cause a denial of service (system crash) via crafted BPF instructions. (CVE-2014-3145)

A flaw was discovered in the Linux kernel's handling of hugetlb entries. A local user could exploit this flaw to cause a denial service (memory corruption or system crash). (CVE-2014-3940)

Don Bailey and Ludvig Strigeus discovered an integer overflow in the Linux kernel's implementation of the LZ4 decompression algorithm, when used by code not complying with API limitations. An attacker could exploit this flaw to cause a denial of service (memory corruption) or possibly other unspecified impact. (CVE-2014-4611)

Tuomas Räsänen reported the Linux kernel on certain Intel processors does not properly initialize random seeds for network operations, causing TCP sequence numbers, TCP and UDP port numbers and IP ID values to be predictable. A remote attacker could exploit this flaw to spoof or disrupt IP communication. (CVE-2014-7284)