Re: [PATCH] x32: Mask away the x32 syscall bit in the ptrace codepath

[Andy Lutomirski]

On Wed, May 28, 2014 at 2:19 PM, H. Peter Anvin <hpa@xxxxxxxxxxxxxxx> wrote:
> On 05/28/2014 02:15 PM, Andy Lutomirski wrote:
>>>
>>>> 3. The OOPS you're fixing doesn't seem like it's fixed.  What if some
>>>> other random high bits are set?
>>>
>>> There is a range check in entry_*.S for the system call.
>>
>> I can imagine that causing a certain amount of confusion to fancy
>> seccomp users.  Oh, well.  No one that I know of has complained yet.
>>
>
> I don't know how seccomp or audit deal with out-of-range syscall
> numbers, so that might be worth taking a look at.

audit oopses, apparently.  seccomp will tell BPF about it and follow
directions, barring bugs.

However: are you sure that entry_64.S handles this?  It looks like
tracesys has higher priority than badsys.  And strace can certainly
see out-of-range syscalls.  And I can oops audit by doing:

auditctl -a exit,always -S open
./a.out

where a.out is this:

#include <stdio.h>
#include <sys/syscall.h>

int main()
{
  long i;
  for (i = 1000; ; i += 64 * 4096)
    syscall(i, 0, 0, 0, 0, 0, 0);
  return 0;
}

I would *love* to deprecate the entire syscall auditing mechanism.  Or
at least I'd love to have distros stop using it.

I'll go ask for a CVE number.  Sigh.


--Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/