From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.6
Path: eternal-september.org!reader02.eternal-september.org!aioe.org!RKN7TKnHC01q0gdg6EhkbQ.user.46.165.242.75.POSTED!not-for-mail
From: Simon Wright <simon@pushface.org>
Newsgroups: comp.lang.ada
Subject: Re: C time_t 2038 problem s-os_lib.ads
Date: Sat, 25 Sep 2021 12:23:53 +0100
Organization: Aioe.org NNTP Server
Message-ID: <ly4ka8x5iu.fsf@pushface.org>
References: <8936f386-3fdb-43b3-b912-317906d59631n@googlegroups.com>
	<sii2q2$fcn$1@dont-email.me>
	<4431fad9-d297-4d68-8c0f-fa771c6710f6n@googlegroups.com>
	<cd4a94de-e8b9-4194-8d80-24a009a558ban@googlegroups.com>
	<874kabm5mp.fsf@nosuchdomain.example.com>
	<3c0272f8-4117-46a4-9051-5419d1edfdc6n@googlegroups.com>
	<87ilyplh4d.fsf@nosuchdomain.example.com>
Mime-Version: 1.0
Content-Type: text/plain
Injection-Info: gioia.aioe.org; logging-data="433"; posting-host="RKN7TKnHC01q0gdg6EhkbQ.user.gioia.aioe.org"; mail-complaints-to="abuse@aioe.org";
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.2 (darwin)
X-Notice: Filtered by postfilter v. 0.9.2
Cancel-Lock: sha1:8PuOoNn9dbfBrjm9HQoBQ1HItak=
Xref: reader02.eternal-september.org comp.lang.ada:62838
List-Id: <comp.lang.ada>

Keith Thompson <Keith.S.Thompson+u@gmail.com> writes:

> There are several common data models in the C and C++ world:
>
> Name       ILP32  LP64  IL32P64
> ====       =====  ====  =======
> char       8      8     8
> short      16     16    16
> int        32     32    32
> long       32     64    32
> long long  64     64    64
> pointer    32     64    64
>
> 32-bit systems (which are becoming rarer for non-embedded systems)
> typically use ILP32, and 64-bit Linux/Unix systems typically use LP64.
> 64-bit Windows uses IL32P64 (and hardly anything else does).
>
> It's *seems* almost obvious that Ada's types
>     Character
>     Short_Integer
>     Integer
>     Long_Integer
>     Long_Long_Integer
> should correspond to the similarly named C types, but it's not required.
> (I don't know whether GNAT does so consistently or not.)

Package Standard in FSF GCC 11.2.0 on macOS (which you can see by
compiling something with -gnatS) has

   type Integer is range -(2 **31) .. +(2 **31 - 1);
   for Integer'Size use 32;

   subtype Natural  is Integer range 0 .. Integer'Last;
   subtype Positive is Integer range 1 .. Integer'Last;

   type Short_Short_Integer is range -(2 **7) .. +(2 **7 - 1);
   for Short_Short_Integer'Size use 8;

   type Short_Integer is range -(2 **15) .. +(2 **15 - 1);
   for Short_Integer'Size use 16;

   type Long_Integer is range -(2 **63) .. +(2 **63 - 1);
   for Long_Integer'Size use 64;

   type Long_Long_Integer is range -(2 **63) .. +(2 **63 - 1);
   for Long_Long_Integer'Size use 64;

   type Long_Long_Long_Integer is range -(2 **127) .. +(2 **127 - 1);
   for Long_Long_Long_Integer'Size use 128;

I didn't know about the last, which is new in FSF GCC 11/GNAT CE 2021
... I could build my Analytical Engine simulator with 40 digit wheels
(i.e. capable of 40 decimal digits) instead of 50 using
Long_Long_Long_Integer instead of GNATColl.GMP.Integer.