On Thu, 17 Oct 2024 15:19:52 GMT, fabioromano1 <d...@openjdk.org> wrote:
>> Will approve at the beginning of next week to let you add some last minute >> (not substantial) changes during the next few days. > >> In any reduction approach, you still need to prove that the reduced problem >> is equivalent to the original. This is to say that the running variables (in >> code and proof) must somehow be related to the initial values. > > Yes, although in the proof by "reducing the size of the problem" this should > be evident by induction... > Will approve at the beginning of next week to let you add some last minute > (not substantial) changes during the next few days. I thought that maybe could be useful add a comment highlighting that `(intVal * 2^powsOf2) / 10^z` and `scale - z` are invariant values, to point out the substructure of the problem better... ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/21323#discussion_r1805580561
