[PATCH] D35109: [Analyzer] SValBuilder Comparison Rearrangement

[Gábor Horváth via Phabricator via cfe-commits]

xazax.hun added a comment.

In https://reviews.llvm.org/D35109#926078, @baloghadamsoftware wrote:

> So still the options are to fix it in the checker or fix it in the engine (the max/4 or the type extension solution), but leaving it unfixed is not an option. I am open to any solution, but only full solutions and no partial solutions, because they will not help us enough.


Just to be clear, what is the source of the problem here? As far as I understand, your concern is not supporting `$x + N > $x  + M`. Does this affect the mismatched iterator part of the checker? What is the effect on the other parts? Not finding some issues? Having more false positives?
If there are parts of the check that does not depend on evaluating such constraints, maybe we could proceed with upstreaming those, so the whole process is not blocked while we still figuring out what to do with the rest. What do you think?


https://reviews.llvm.org/D35109