Maxime Jacquemin authored 1 month ago and David Bühler committed 1 month ago

As the format is not supported for now, we have removed it.
Witnesses are provided to represent the fact that long
floats are encoded as double. The doc is not yet updated.

Also renames round_to_single_precision_float into round_to_single_precision.

Floating_point contains utilitary functions over ocaml float.
Typed_float provides a typed representation of floating-point numbers, which
encodes the format of the represented number.

Maxime Jacquemin authored 6 months ago and David Bühler committed 1 month ago

Asked as a helper function to minimize modifications and ensures
backward compatibility. Will be removed when every thing relies
and the typed floats, which means probably never.

Maxime Jacquemin authored 6 months ago and David Bühler committed 1 month ago

- Renaming [round_to] as [change_format] to avoid confusion with [round]
- Renaming [of_float] as [represents ~float ~in_format] to better convey
  the goal of this module and why we hide the float type
- Adding [@@noalloc] to external functions [set_rounding_mode] and
  [get_rounding_mode] as a (rather small) optimisation

Maxime Jacquemin authored 8 months ago and David Bühler committed 1 month ago

The module now handles explicitly the floating point formats using the
type system, thus ensuring strict manipulation of float numbers.

Add a record with 2 fields : tnode and tattr

The new parameter is -eva-cache-size.

Also take a Hptmap_sig.cache_type argument instead of ~name and ~temporary.
This is consistent with all other cached functions in hptmap.

The warning messages for missing assigns and missing from now have a
category that is set to `error` by default. Moreover, the message is
expanded to note that `assigns \nothing` and `assigns \from \nothing` is
assumed instead and that the analysis is probably incorrect.

... to assigns:invalid-location