Messages in this thread |  | | Date | Fri, 22 Sep 2000 12:25:39 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: [ANOTHER PATCH] Re: [PATCH] 2.4.0 i386 watchpoint problems |
| |
On Fri, 22 Sep 2000, Andi Kleen wrote: > > Also unless I'm missing another thing ptrace allows you to put any addresses > including kernel address into the debug registers, so you could certainly > get debug traps everywhere, making my original objection valid.
See:
if(addr < (long) &dummy->u_debugreg[4] && ((unsigned long) data) >= TASK_SIZE-3) break;
> How does this patch look ?
Horrible. It WILL NOT WORK.
It does a access_ok() that uses the debugger address space to determine if the debuggee is ok.
Also, debugreg6 is not a linear address at all. Only 0-3 are, so only they should be tested.
Which is exactly what the test that already exists does.
No, it may not be obvious, but it's all there.
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |