Hernan Ponce de Leon

@ThomasHaas you mentioned that you discussed (before the improvement that you already implemented) with Rene some potential improvements. Do those differ of what was already implemented? I.e. can we improve...

Right now, it is less blocking than it used to be, but I would not complain if we can get the times faster. Since we already have ideas on how...

@ThomasHaas can you add to this issue the llvm file of tree-RCU so we have a concrete test showing the problem?

I still have not manage to find an example where this new alias pays off: while we have a more precise analysis, the extra optimization makes the program harder for...

Does the first bullet point mean the new analysis is always more precise than the old ones? Do we have any concrete example where we get noticeable better performance?

@ThomasHaas would it make sense to add EBR to our CI? IIRC it is supposed to be buggy wrt IMM but correct under hardware models

It seems that when I rebase the branch, I overwrote the default alias type. It should be fixed by now. BTW, as Thomas said, the new alias analysis shows to...

It seems you are right (even if I'm pretty sure I saw `field_sensitive` in the log when I called dartagnan from the console with no alias options). We should be...

I am testing this branch in some new benchmarks and I get a bunch of warnings like this ``` [04.03.2024] 00:14:56 [WARN] InclusionBasedPointerAnalysis.postProcess - empty pointer set for bv64 117:r24...

> Yes, this means the are no possible targets to read from. It could be a bug in the analysis but also a correct result. For example, this warning also...