[m-rev.] for review: start using type_spec_constrained_preds pragmas

[Julien Fischer]

On Fri, 9 Feb 2024, Zoltan Somogyi wrote:

> On 2024-02-09 21:09 +11:00 AEDT, "Zoltan Somogyi" <zoltan.somogyi at runbox.com> wrote:
>> Attached, for your post-commit review.
>
> That was just one top-of-module comment being updated;
> the attached diff is for the other.

I've just pushed a change that fixes a few minor issues.
One review comment for you below.

> diff --git a/compiler/intermod.m b/compiler/intermod.m
> index 5c995c1c5..744f50dbb 100644
> --- a/compiler/intermod.m
> +++ b/compiler/intermod.m

...

> +% In the original implementation of intermodule optimization, everything
> +% in the generated .opt files came directly from user-written source files.
> +% This is now true *only* for the first part of .opt files. Now this first
> +% part can be optionally followed by a second part that contains
> +% compiler-generated pragmas that record the results of program analyses,
> +% which figure out things such as which pieces of code cannot throw exceptions
> +% or change the trail.
> +% 
> +% This module and its subcontractors decide what should be in the first half
> +% of the .opt file for the given module, and then write it out. The second
> +% half is handled by intermod_analysis.m. That module

That module what?

The rest looks fine.

Julien.