about summary refs log tree commit diff homepage
path: root/lib/Core/Memory.cpp
diff options
context:
space:
mode:
authorDan Liew <daniel.liew@imperial.ac.uk>2015-12-16 18:13:11 +0000
committerDan Liew <daniel.liew@imperial.ac.uk>2015-12-18 11:22:50 +0000
commit53ff7a002a8213a5d5e778bef2a895998d9890e1 (patch)
tree026a4f1b16c1996755954f7824e0d10f8ed0ef8e /lib/Core/Memory.cpp
parent7e75fa79b2e76251c2cd417a7eae8a7620b014ae (diff)
downloadklee-53ff7a002a8213a5d5e778bef2a895998d9890e1.tar.gz
Fix memory leaks of ``Array`` objects detected by ASan.
Some of these leaks were introduced by the factory constructor for Array
objects (f049ff3bc04daead8c3bb9f06e89e71e2054c82a) but a few others have
been around for far longer.

This leak was fixed by introducing a ``ArrayCache`` object which has two
purposes

* Retains ownership of all created ``Array`` objects and destroys them when
  the ``ArrayCache`` destructor is called.
* Mimic the caching behaviour for symbolic arrays that was introduced
  by f049ff3bc04daead8c3bb9f06e89e71e2054c82a where arrays with the same
  name and size get "uniqued".

The Executor now maintains a ``arrayCache`` member that it uses and
passes by pointer to objects that need to construct ``Array`` objects (i.e.
``ObjectState``). This way when the Executor is destroyed all the
``Array`` objects get freed which seems like the right time to do this.

For Kleaver the ``ParserImpl`` has a ``TheArrayCache`` member that is
used for building ``Array`` objects. This means that the Parser must
live as long as the built expressions will be used otherwise we will
have a use after free. I'm not sure this is the right design choice.
It might be better to transfer ownership of the ``Array`` objects to
the root ``Decl`` returned by the parser.
Diffstat (limited to 'lib/Core/Memory.cpp')
-rw-r--r--lib/Core/Memory.cpp21
1 files changed, 11 insertions, 10 deletions
diff --git a/lib/Core/Memory.cpp b/lib/Core/Memory.cpp
index 12ca7104..72f0a1fb 100644
--- a/lib/Core/Memory.cpp
+++ b/lib/Core/Memory.cpp
@@ -14,6 +14,7 @@
 #include "klee/Solver.h"
 #include "klee/util/BitArray.h"
 #include "klee/Internal/Support/ErrorHandling.h"
+#include "klee/util/ArrayCache.h"
 
 #include "ObjectHolder.h"
 #include "MemoryManager.h"
@@ -110,9 +111,9 @@ ObjectState::ObjectState(const MemoryObject *mo)
     readOnly(false) {
   mo->refCount++;
   if (!UseConstantArrays) {
-    // FIXME: Leaked.
     static unsigned id = 0;
-    const Array *array = Array::CreateArray("tmp_arr" + llvm::utostr(++id), size);
+    const Array *array =
+        getArrayCache()->CreateArray("tmp_arr" + llvm::utostr(++id), size);
     updates = UpdateList(array, 0);
   }
   memset(concreteStore, 0, size);
@@ -176,6 +177,11 @@ ObjectState::~ObjectState() {
   }
 }
 
+ArrayCache *ObjectState::getArrayCache() const {
+  assert(object && "object was NULL");
+  return object->parent->getArrayCache();
+}
+
 /***/
 
 const UpdateList &ObjectState::getUpdates() const {
@@ -215,15 +221,10 @@ const UpdateList &ObjectState::getUpdates() const {
       Contents[Index->getZExtValue()] = Value;
     }
 
-    // FIXME: We should unique these, there is no good reason to create multiple
-    // ones.
-
-    // Start a new update list.
-    // FIXME: Leaked.
     static unsigned id = 0;
-    const Array *array = Array::CreateArray("const_arr" + llvm::utostr(++id), size,
-					    &Contents[0],
-					    &Contents[0] + Contents.size());
+    const Array *array = getArrayCache()->CreateArray(
+        "const_arr" + llvm::utostr(++id), size, &Contents[0],
+        &Contents[0] + Contents.size());
     updates = UpdateList(array, 0);
 
     // Apply the remaining (non-constant) writes.