This patch implements part of RegionStoreManager. Only SetRVal() and getInitialStore() method is implemented. The store model is implemented as a map from Region* to RVal. Global and parameter pointers are assumed to point to an AnonTypedRegion initially, which resides in a new Unknown MemSpaceRegion. Is this reasonable? A new field 'PointedBy' is added to AnonTypedRegion, which represents the pointer variable that points to this region. Can we have a better name for it? Later we have to have more information to differentiate AnonTypedRegion. For example, how to differentiate heap objects created at the same program point along different paths?
regionstore.patch
Description: Binary data
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
