[PATCH] D43791: [analyzer] Suppress MallocChecker positives in destructors with atomics.

[George Karpenkov via Phabricator via cfe-commits]

george.karpenkov added inline comments.


================
Comment at: lib/StaticAnalyzer/Checkers/MallocChecker.cpp:527
   private:
+    bool shouldSuppressOnAtomicReferenceCountingPointers(
+        const Stmt *S, const LocationContext *CurrentLC);
----------------
Docstring.


================
Comment at: lib/StaticAnalyzer/Checkers/MallocChecker.cpp:2847
+
+  AtomicExpr::AtomicOp Op = AE->getOp();
+  if (Op != AtomicExpr::AO__c11_atomic_fetch_add &&
----------------
IMO would be slightly easier to read with logic reversed, e.g.

```
if (AE == dyn_cast<>(...))
   if (Op == add || Op == sub)
      for (...)
        if (...)
          return true
return false
```

But this is a preference and can be ignored.


================
Comment at: lib/StaticAnalyzer/Checkers/MallocChecker.cpp:2899
+      // to find out if a likely-false-positive suppression should kick in.
+      for (const LocationContext *LC = CurrentLC; LC; LC = LC->getParent()) {
+        if (isa<CXXDestructorDecl>(LC->getDecl())) {
----------------
I'm not sure what is going on here.
We are just traversing the stack frame, finding the first destructor in the `isReleased` branch?


Repository:
  rC Clang

https://reviews.llvm.org/D43791