public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* suggestion.. check for read/write/size/TASK_SIZE in kernel segment (early on) ?
@ 2003-01-07  6:29 Silvio Cesare
  0 siblings, 0 replies; only message in thread
From: Silvio Cesare @ 2003-01-07  6:29 UTC (permalink / raw)
  To: linux-kernel

Hi.. I'm not sure if this is valid or not (or is compliant with everything),
but it seems like a reasonable suggestion IMO.

In read/write system call, check that the read/write buffer address
resides not in the kernel segment.  ie, check buffer address and TASK_SIZE,
address + size etc..

That may seem silly.. but it would stop any later problems by drivers
which dont check this themselves, or handle it incorrectly or inconsistantly..

1) clear security integrity check (so drivers etc are less worried)
2) clearly defined behaviour (error codes etc) when this occurs (easier
   on the drivers again).

If its not possible to do this in the system call directly any chance of
putting such a check early on in the generic driver code present?
perhaps the buffer in read/write means something other than a userspace
buffer/array, to someone, but i cant think of any read/write behaviour to
counter-claim this..

perhaps a dumb suggestion.. open to feedback :)

--
Silvio

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2003-01-07  7:48 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-01-07  6:29 suggestion.. check for read/write/size/TASK_SIZE in kernel segment (early on) ? Silvio Cesare

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox