Hi Kevin,

I hope someone who actually did this finds this question, but I could think of a few reasons why this would not have been done.

Firstly, this requires special tooling, and we have to consider what happens when Installer runs.

Secondly, we cannot debug natively compiled code. This could become a big problem in the most undesirable of times: when you have a problem in production. If we want to debug in production, we’ll then have two problems instead of one.

Finally, and most importantly, the performance gains in Oracle natively compiled code is typically well apparent only when there is a lot of number crunching is going on in PL/SQL, and not much I/O. Even much of the online examples are full of matrix multiplications. But typically IFS code is full of pure SQL statements which translate to I/O and not a whole lot of computation in most cases (MRP, MS, Costing, and the likes barely being the exceptions). Another big part in IFS code would be string operations, and I would think those string manipulation APIs are well optimized for decades. So I have reservations about how the fabled performance gains would translate to IFS business logic.

In fact, this has come up time to time in the past and I can see traces of experiments in our internal documents. But it doesn’t seem like native compilation got any traction. But this still is in our backlog of ideas and might be implemented some day.

Anyway, hope you find the answer for this sooner that that. Thank you.

Best regards,
Sampath