Re: [PATCH] ptrace RSE bug

From: Petr Tesarik <>
Date: 2007-10-18 05:48:47
Petr Tesarik wrote:
>   4. While talking about RNAT, does the RBS syncing back and forth
> handle correctly the case when part of the RNAT stored in the backing
> store belongs to the kernel registers? It must not be possible to change
> the NAT bits for kernel registers from userspace!
>      Maybe it's not an issue, because I tried to actually exploit this
> bug, and my attempts failed.

I've just verified that put_rnat() does the right thing here (even
prevents setting NaT for syscall arguments), so modifying RNAT bits is
no problem. The other things still apply.

Petr Tesarik
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to
More majordomo info at
Received on Thu Oct 18 05:46:43 2007

This archive was generated by hypermail 2.1.8 : 2007-10-18 05:46:58 EST