Full Disclosure mailing list archives

Re: grsecurity 2.1.0 release / 5 Linux kernel advisories


From: Marcus Meissner <meissner () suse de>
Date: Wed, 19 Jan 2005 11:34:03 +0100

On Fri, Jan 07, 2005 at 01:18:53PM -0500, Brad Spengler wrote:
Let's try this again, since web archives don't like multipart 
attachments.

grsecurity 2.1.0 release / Linux Kernel advisories
--------------------------------------------------------------------

Table Of Contents:
...
4) 2.6 scsi ioctl integer overflow and information leak


4) 2.6 scsi ioctl integer overflow and information leak

In drivers/block/scsi_ioctl.c:

at sg_scsi_ioctl():
       struct request *rq;
       int err, in_len, out_len, bytes, opcode, cmdlen;
        ^ in_len, out_len are signed int
       char *buffer = NULL, sense[SCSI_SENSE_BUFFERSIZE];

       /*
        * get in an out lengths, verify they don't exceed a page worth of data
        */
       if (get_user(in_len, &sic->inlen))
        ^ in_len is user-controlled
               return -EFAULT;
       if (get_user(out_len, &sic->outlen))
        ^ out_len is user-controlled
               return -EFAULT;
       if (in_len > PAGE_SIZE || out_len > PAGE_SIZE)
        ^ signed int only has upper bound checked
               return -EINVAL;

Well spotted.

I actually spent 30 minutes unsuccessfully trying to get my kernel to
crash with this before a C guru here explained it to me.

The C compiler actually rises to the rescue and safes the kernel.

"A comparison will always be evaluated in the largest integer context
available." (or so)

PAGE_SIZE is defined as (1UL << PAGE_SHIFT) for all architectures on
2.6 kernels.

signed int has 31bits, unsigned long has either 32 or 64 bit.

So this expression will always be evaluated in unsigned long mode. 

In this case the signed int is not converted, but the binary representation
is used, -1 would be in this case 0xffffffff.

        => No problem.

Ciao, Marcus

Attachment: _bin
Description:

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.netsys.com/full-disclosure-charter.html

Current thread: