http://llvm.org/bugs/show_bug.cgi?id=17714
Bill Wendling <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |INVALID --- Comment #3 from Bill Wendling <[email protected]> --- I don't think that this is real non-deterministic behavior. What's happening is that the memory for the loop analysis is sometimes being reused between function passes. So the map is sometimes associating a basic block with the address of a previous LoopInfo object. For instance, if you iterate over the map and try to emit the LoopInfo object, it will fail because that object is no longer in valid memory. Anyway, it sucks, but this isn't an LLVM bug. :-( -- You are receiving this mail because: You are on the CC list for the bug.
_______________________________________________ LLVMbugs mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
