[PATCH] D103440: [WIP][analyzer] Introduce range-based reasoning for addition operator

[Valeriy Savchenko via Phabricator via cfe-commits]

vsavchenko added a comment.

In D103440#2797991 <https://reviews.llvm.org/D103440#2797991>, @manas wrote:

>> I also would like to see tests where the ranges are not going all the way to either INT_MIN or INT_MAX (if we talk about int), but overflow still might happen, and cases where overflow might happen, but we still can identify the overflowing results precisely (e.g. the result is `[INT_MIN, INT_MIN + 10] and [INT_MAX - 5, INT_MAX]`)
>
> If I understood correctly, does a case like: `c, d in [INT_MAX/2 - 10, INT_MAX/2 + 10]` works? It will produce an overflowing range of `[INT_MIN, INT_MIN + 18] U [INT_MAX - 21, INT_MAX]`. I will add that to the test-set, if that is so.

Yes, exactly!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103440/new/

https://reviews.llvm.org/D103440