The way to do this is not to use the shadow memory, but to clobber
directly the variable itself with specified byte pattern -- though
error diagnostics won't be as nice.
To use shadow memory, the scope start marker is also needed.
David
On Tue, Oct 16, 2012 at 9:12 AM, Jakub Jelinek <ja...@redhat.com> wrote:
> On Tue, Oct 16, 2012 at 09:06:22AM -0700, Xinliang David Li wrote:
>> I don't get it. Clobber marks the end of lifetime of a variable so it
>> is safe to emit code to really clobber its value -- otherwise how
>> would clobber based slot sharing work?
>
> If you at CLOBBER protect the var in the shadow mem as unaccessible, the
> thing is where you make it accessible again.
> Consider:
>
> int i;
> for (i = 0; i < 2; i++)
> {
> int tmp;
> bar ();
> baz (&tmp);
> bar ();
> }
>
> where the first bar () can't possibly access tmp, but the second bar () can.
> Now we unroll it and have:
> bar ();
> baz (&tmp);
> bar ();
> tmp ={v} {CLOBBER};
> bar ();
> baz (&tmp);
> bar ();
> tmp ={v} {CLOBBER};
> So, if you want to *mem_to_shadow (&tmp) = 0xff; at {CLOBBER} time, where
> and how you would insert *mem_to_shadow (&tmp) = 0; again? In the above
> code it would need to go before the second baz (&tmp);, the question is
> how you'd find that out...
>
> Jakub
