Re: Unexecutable stack

Khimenko Victor (khim@sch57.msk.ru)
Tue, 28 Dec 1999 01:55:50 +0300 (MSK)

In <19991227191253.B7283@oxygene.terra.vein.hu> Gabor Lenart (lgb@oxygene.terra.vein.hu) wrote:
GL> On Mon, Dec 27, 1999 at 08:37:59PM +0300, Khimenko Victor wrote:
>> In <38673F5C.1010C79A@katren.ru> Mike Karmyshev (mike@katren.ru) wrote:
>> > Hello.
>> > I've recently played a bit with Solar Designer's patch and it looks that
>> > it doesn't have any
>> > significant overhead. Shoudn't it be in the kernel by default(at
>> > least,SECURE_STACK)?
>>
>> Last time when this question was raised was more then year ago (if I recall
>> correctly) and Linus said that his feeling about unexecutable stack is that
>> it does not make exploits impossible but insted give you false sense of safety.
>> So answer is "no". You can add such patch by hands if you wish...

GL> Hmmm. But kernel contains features marked 'experimental'.

`Experimental', not meaningless.

GL> Like experimental things, secure Linux patch can go into kernel with some
GL> remark like experimantal. (in this case: "big warning, ...").

It's not experimental. Experimental == Linus thinks that feature is cool but
it's not yet debugged enough to be used in production environment. And it's
NOT a case.

GL> Pre-XFree 4.0 servers seem to segfault with this patch. Anyone who has got
GL> similar sympthoms ?

GL> BTW, restricted proc fs should go into kernel tree (do not care in this case
GL> if unexecutable stack goes in or not), because it's the minimum to have
GL> an ability to hide my processes from others. It's VERY simple and trivial
GL> patch, only alters file access permissions in /proc.

Once again: you should convinience Linus that it's good thing to do :-)

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