Re: [PATCH v2 08/10] x86-64: Emulate vsyscalls

[Andrew Lutomirski]

On Mon, May 30, 2011 at 3:46 AM, Ingo Molnar <mingo@xxxxxxx> wrote:
>
> * Andy Lutomirski <luto@xxxxxxx> wrote:
>
>> There's a fair amount of code in the vsyscall page, and who knows
>> what will happen if an exploit jumps into the middle of it.  Reduce
>> the risk by replacing most of it with short magic incantations that
>> are useless if entered in the middle.  This change can be disabled
>> by CONFIG_UNSAFE_VSYSCALLS (default y).
>
> btw., please flip the default or consider removing the option
> altogether.
>
> We want to improve security and we want safe vsyscalls the default,
> and it's no good if we make it too easy for users to keep the fire
> door open all the time! :-)

I'd advocate waiting until glibc 2.14 comes out with this change:

http://sourceware.org/git/?p=glibc.git;a=commit;h=a8509ca540427502bd955f35296ff7b727c7a8a1

I want to add a warning (ratelimited to an extremely low rate) in v3
whenever any of the vsyscalls get used telling users that their legacy
code is suffering a performance impact, but it seems like bad form to
tell people to build glibc from git to avoid a regression.

The other option is to make the change unconditional for gettimeofday
and vgetcpu but leave time alone for awhile.  There's no a priori
reason why leaving vtime around is worse than vread_tsc and
vread_hpet.


--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/