From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM autolearn=unavailable autolearn_force=no version=3.4.4 X-Received: by 10.107.180.17 with SMTP id d17mr18104030iof.15.1467375460828; Fri, 01 Jul 2016 05:17:40 -0700 (PDT) X-Received: by 10.157.11.173 with SMTP id 42mr787606oth.10.1467375460795; Fri, 01 Jul 2016 05:17:40 -0700 (PDT) Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!mx02.eternal-september.org!feeder.eternal-september.org!news.glorb.com!jk6no8264063igb.0!news-out.google.com!o189ni13815ith.0!nntp.google.com!jk6no8264057igb.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Fri, 1 Jul 2016 05:17:40 -0700 (PDT) In-Reply-To: Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=2601:191:8202:8510:5985:2c17:9409:aa9c; posting-account=fdRd8woAAADTIlxCu9FgvDrUK4wPzvy3 NNTP-Posting-Host: 2601:191:8202:8510:5985:2c17:9409:aa9c References: <57346ac8$0$4570$426a74cc@news.free.fr> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: Subject: Re: Ada.Strings.Fixed.Count raises Storage_Error From: rieachus@comcast.net Injection-Date: Fri, 01 Jul 2016 12:17:40 +0000 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Xref: news.eternal-september.org comp.lang.ada:30991 Date: 2016-07-01T05:17:40-07:00 List-Id: On Friday, July 1, 2016 at 6:55:50 AM UTC-4, J-P. Rosen wrote: > I don't remember anything like that. Different hat. (Management) When Ada (80) was new, it was just assumed tha= t minicomputers and PCs would have a 16-bit Integer type. Then it turned o= ut that Ada compilers couldn't really fit in a machine with too limited an = address space. So those of us who worked for hardware companies had to do = the soul searching about which hardware compilers needed to run on, and whi= ch targets could be included. The choice between 16-bit and 32-bit Integer= types was possibly just a symptom, but by 1984 it was pretty much in the r= ear view mirror. No one really wanted to buy compilers where Integer was 1= 6-bits, even for the (MIL STD) 1750A. =20 > And there is no problem for > generating correct code for a loop that spans over all values of a type, > you just have to exit in the middle - without efficiency penalty. You have a different definition of efficiency than I do. On a modern proce= ssor I want to be able to execute tight loops in one clock cycle per iterat= ion. This usually requires three or more instructions that can be executed= in parallel. The efficient workaround is to have a single loop iteration = before the actual loop code. The better way is just to run the loop in a 6= 4-bit register if that is an option. ;-)