Hi Zhongxing, No we can't. I'm still seeing symbolic regions wrapped in TypedViewRegions. Reverting.
On Apr 28, 2009, at 6:49 AM, Zhongxing Xu wrote: > Author: zhongxingxu > Date: Tue Apr 28 08:49:42 2009 > New Revision: 70293 > > URL: http://llvm.org/viewvc/llvm-project?rev=70293&view=rev > Log: > Now we can remove the 'blast-through' code. > > Modified: > cfe/trunk/lib/Analysis/CFRefCount.cpp > > Modified: cfe/trunk/lib/Analysis/CFRefCount.cpp > URL: > http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Analysis/CFRefCount.cpp?rev=70293&r1=70292&r2=70293&view=diff > > = > = > = > = > = > = > = > = > ====================================================================== > --- cfe/trunk/lib/Analysis/CFRefCount.cpp (original) > +++ cfe/trunk/lib/Analysis/CFRefCount.cpp Tue Apr 28 08:49:42 2009 > @@ -1904,13 +1904,6 @@ > > const TypedRegion* R = dyn_cast<TypedRegion>(MR->getRegion()); > > - // Blast through TypedViewRegions to get the original > region type. > - while (R) { > - const TypedViewRegion* ATR = dyn_cast<TypedViewRegion>(R); > - if (!ATR) break; > - R = dyn_cast<TypedRegion>(ATR->getSuperRegion()); > - } > - > if (R) { > // Is the invalidated variable something that we were > tracking? > SymbolRef Sym = state.GetSValAsScalarOrLoc > (R).getAsLocSymbol(); > > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
