On Sun, 30 May 2010 00:46:18 -0700, Jordy Rose <jedik...@belkadan.com>
wrote:
> For looking up super regions' direct bindings, I tried commenting out
that
> entire section, but it makes the case you mentioned fail
> (no-outofbounds.c). Looking at each of the inside IFs:
> 2. Same as your example, but with x not yet defined:
> int x;
> char *p = &x;
> return p[0]; // expected-warning {{Undefined}}
Of course, I forgot that I had actually changed the condition from
isUnknown() to isUnknownOrUndefined() in my working copy. So right now the
second IF doesn't seem to do /anything/, since an unknown value is clearly
not a LazyCompoundVal.
> It seems like if there's any chance of the feature being useful (even
> UnknownVal vs UndefinedVal) this section has to stay. Between
> no-outofbounds.c and the commit message about
misc-ps-region-store-x86_64.m
> crashing; I didn't pry much deeper.
Right. Possibly only useful with that change, but more likely still needed
for edge cases.
Jordy
_______________________________________________
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
