[PATCH] D37120: [analyzer] Fix modeling arithmetic
Alexander Shaposhnikov via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Thu Aug 24 14:44:17 PDT 2017
alexshap created this revision.
Herald added a subscriber: xazax.hun.
This diff attempts to fix modeling of arithmetic expressions
where pointers are treated as integers (i.e. via C-style / reinterpret casts).
In particular, it resolves https://bugs.llvm.org/show_bug.cgi?id=34309
Test plan: make check-all
Repository:
rL LLVM
https://reviews.llvm.org/D37120
Files:
lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
test/Analysis/ptr-arith.cpp
Index: test/Analysis/ptr-arith.cpp
===================================================================
--- test/Analysis/ptr-arith.cpp
+++ test/Analysis/ptr-arith.cpp
@@ -105,3 +105,9 @@
return 0;
return N;
}
+
+// Bug 34309
+bool ptrAsIntegerSubtractionNoCrash(long x, char *p) {
+ long y = (long)p - 1;
+ return y == x;
+}
Index: lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===================================================================
--- lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -364,6 +364,13 @@
rhs.castAs<nonloc::LocAsInteger>().getLoc(),
resultTy);
case nonloc::ConcreteIntKind: {
+ // Evaluate pointers treated as integers
+ // (for example, results of C-style casts (long)((void *)ptr))
+ // in arithmetic expressions with integers.
+ if (!BinaryOperator::isComparisonOp(op))
+ return makeSymExprValNN(
+ state, op, lhs.castAs<nonloc::LocAsInteger>(),
+ rhs.castAs<nonloc::ConcreteInt>(), resultTy);
// Transform the integer into a location and compare.
// FIXME: This only makes sense for comparisons. If we want to, say,
// add 1 to a LocAsInteger, we'd better unpack the Loc and add to it,
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D37120.112614.patch
Type: text/x-patch
Size: 1359 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20170824/99d5cbfb/attachment.bin>
More information about the cfe-commits
mailing list