[m-rev.] for review: make foreign_procs a top-level item

[Julien Fischer]

Hi Zoltan,

On Fri, 4 Aug 2023, Zoltan Somogyi wrote:

> For review by anyone.
>
> This diff caused two tests to fail. The diff to make_hlds_passes.m contains
> a ZZZ comment documenting the reason for this, and outlines a solution:
> adding marker pragmas to the HLDS separately from other kinds of
> impl pragmas. Does anyone object to this solution? Does anyone object
> to doing it by making marker pragmas their own kind of pragma, separate
> from decl, impl and gen pragmas? I would like feedback on this in the next
> day or so, since I want to do that next.

I have no objection.

> That comment also outlines a longer term proposal: deprecating promise
> pure/semipure pragmas. I would also like feedback on that, but this question
> is not that urgent.

In principle, I don't have any objection to removing them.  As per the
m-dev thread Peter mentioned, I suspect that the pragmas are less error
prone than correctly placing a promise scope around the top-level goal.
Forgetting to unify the I/O state inside the promise scope is rather
easy to do.

I think a lot of what motivated the 2008 discussion has gone away due
to the addition for trace goals to the language.

The diff looks fine.

Julien.