This is a great question, but let me give an idea for why I think it might be impossible (or at least very hard) in general. Presumably, such an implementation would almost be a C++ interpreter, or at least a compiler for something more like Lisp or Java. It would need to keep extra data … Read more
I think the problem is that Asan and Msan both want to control the heap, and both want to reserve a large amount of memory to use as “shadow memory” which tracks the allocations and usage of the memory your program uses. They can’t both be active because they would be trying to track the … Read more
While breaking on the detection functions (as described by @Mark Plotnick and @Iwillnotexist Idonotexist) is one option, a better approach is breaking on the functions that report these issues after detection. This approach is also used for ASAN where one would break on __asan_report_error. Summary: You can stop on an ubsan report via a breakpoint … Read more
This is more a framework for adding such checks than an attempt to detect all forms of undefined behavior (which is almost certainly impossible in the “halting problem” sense). The GCC documentation lists these as the currently supported checks: -fsanitize=undefined Enable UndefinedBehaviorSanitizer, a fast undefined behavior detector. Various computations will be instrumented to detect undefined … Read more
This is a great question, but let me give an idea for why I think it might be impossible (or at least very hard) in general. Presumably, such an implementation would almost be a C++ interpreter, or at least a compiler for something more like Lisp or Java. It would need to keep extra data … Read more