iDefense reported a vulnerability discovered by regenrecht affecting Xorg X server:

DESCRIPTION

Local exploitation of a memory corruption vulnerability in the X.Org X server, as included in various vendors' operating system distributions, could allow an attacker to execute arbitrary code with the privileges of the X server, typically root.

The vulnerable code exists within multiple functions in the XInput extension, and occurs when swapping the byte order of client requests. By claiming to be sending integer data in the opposite byte order of the server, the client can cause the server to swap the byte order of a request. The number of bytes swapped can be controlled by the client, and is not properly validated by the server. This results in the corruption of heap memory located after the client's request data.

The vulnerable code is shown below:

From Xi/chgkmap.c:SProcXChangeDeviceKeyMapping()

87 REQUESTATLEASTSIZE(xChangeDeviceKeyMappingReq);

88 p = (long )&stuff[1];

89 count = stuff->keyCodes stuff->keySymsPerKeyCode;

90 for (i = 0; i < count; i++) {

91 swapl(p, n);

92 p++;

93 }

On line 89, the count value is calculated by multiplying two values taken directly from the client request. The for loop then proceeds to swap the byte order of the request data, and eventually of other heap memory as well.

Duplicate vulnerabilities are also found in the following functions:

Xi/chgprop.c, SProcXChangeDeviceDontPropagateList()

Xi/grabdev.c, SProcXGrabDevice()

Xi/grabdevb.c, SProcXGrabDeviceButton()

Xi/grabdevk.c, SProcXGrabDeviceKey()

Xi/selectev.c, SProcXSelectExtensionEvent()

Xi/sendexev.c, SProcXSendExtensionEvent()

ANALYSIS

Exploitation allows an attacker to execute arbitrary code with root privileges. In order to exploit this vulnerability, an attacker must be able to send commands to an affected X server. This typically requires access to the console or access to the same account as a user who is on the console. One method of gaining the required access is to remotely exploit a vulnerability in, for example, a graphical web browser. This would then allow an attacker to exploit this vulnerability and elevate their privileges to root.

If an X Server is configured to listen for TCP based client connections, and a client is granted access to create sessions (via the xhosts file), then the vulnerability can be exploited remotely.

WORKAROUND

iDefense is currently unaware of any workarounds for this issue. The XInput extension is normally compiled into the X Server; as such, it's not possible to disable it from being loaded in the configuration file.

The Linux kernel before 2.6.25.10 does not properly perform tty operations, which allows local users to cause a denial of service (system crash) or possibly gain privileges via vectors involving NULL pointer dereference of function pointers in (1) hamradio/6pack.c, (2) hamradio/mkiss.c, (3) irda/irtty-sir.c, (4) pppasync.c, (5) pppsynctty.c, (6) slip.c, (7) wan/x25asy.c, and (8) wireless/strip.c in drivers/net/.

The Kerberos 4 support in KDC in MIT Kerberos 5 (krb5kdc) does not properly clear the unused portion of a buffer when generating an error message, which might allow remote attackers to obtain sensitive information, aka "Uninitialized stack values."

yast2-backup 2.14.2 through 2.16.6 on SUSE Linux and Novell Linux allows local users to gain privileges via shell metacharacters in filenames used by the backup process.

Race condition in the directory notification subsystem (dnotify) in Linux kernel 2.6.x before 2.6.24.6, and 2.6.25 before 2.6.25.1, allows local users to cause a denial of service (OOPS) and possibly gain privileges via unspecified vectors.

Description of problem: Zoltan Sogor noticed this VFS behaviour while testing UBIFS. Lookup can install a child dentry for a deleted directory. This keeps the directory dentry alive, and the inode pinned in the cache and on disk, even after all external references have gone away.

This isn't a big problem normally, since memory pressure or umount will clear out the directory dentry and its children, releasing the inode. But for UBIFS this causes problems because its orphan area can overflow.

http://lkml.org/lkml/2008/7/2/83 http://www.linux-mtd.infradead.org/doc/ubifs.html

The docoredump function in fs/exec.c in Linux kernel 2.4.x and 2.6.x up to 2.6.24-rc3, and possibly other versions, does not change the UID of a core dump file if it exists before a root process creates a core dump in the same location, which might allow local users to obtain sensitive information.