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 X-Received: by 2002:a05:6214:411e:: with SMTP id kc30mr18613374qvb.94.1640687095308; Tue, 28 Dec 2021 02:24:55 -0800 (PST) X-Received: by 2002:a05:6902:1246:: with SMTP id t6mr26523567ybu.305.1640687094926; Tue, 28 Dec 2021 02:24:54 -0800 (PST) Path: eternal-september.org!reader02.eternal-september.org!news.misty.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Tue, 28 Dec 2021 02:24:54 -0800 (PST) In-Reply-To: <87pmpiubmb.fsf@nightsong.com> Injection-Info: google-groups.googlegroups.com; posting-host=80.229.20.55; posting-account=WsVe0AoAAABheGmBjlLgPWhgIw6kxcL6 NNTP-Posting-Host: 80.229.20.55 References: <87tuevtblb.fsf@nightsong.com> <19527aed-3b3a-44c2-acc8-2221dbc7a3b6n@googlegroups.com> <87pmpiubmb.fsf@nightsong.com> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: <279d8891-c296-4935-ab17-11e1ce2a40ecn@googlegroups.com> Subject: Re: is Ada used in James Webb Space Telescope software? From: John McCabe Injection-Date: Tue, 28 Dec 2021 10:24:55 +0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Xref: reader02.eternal-september.org comp.lang.ada:63294 List-Id: On Monday, 27 December 2021 at 00:37:03 UTC, Paul Rubin wrote: > John McCabe writes:=20 > > as it looked like we (Matra Marconi Space) might be forced to use the= =20 > > RTX2010 as it was one of very few space qualified processors with=20 > > hardware floating point support. In the end we used the MA31750, with= =20 > > Ada, instead. > Interesting. I hadn't heard of the MA31750 but it appears to be a 16=20 > bit processor that implements the MIL-STD-1750A instruction set(!),=20 > which I didn't know about either. Apparently it was made in the 1980s=20 > but has since been superseded by SPARC architecture cpu's.=20 There were 3 or 4 different implementations of the MIL-STD-1750A instructio= n set architecture around the time. It was an interesting one; it was fairl= y small, but had some relatively complex instructions that were really usef= ul. The MA31750 was GEC-Plessey Semiconductors' 2nd version, I believe, alt= hough if I remember correctly, this was the one that had the FPU, or maybe = it was the MMU, integrated into a single device, using silicon-on-sapphire = for rad-hardness. There were two other implementations I particularly remem= ber that were rad-hard, one by IBM, which had better claimed performance bu= t was really expensive and special order only (I think we paid =C2=A37500 o= r so for each MA31750, so you may be able to imagine what I mean by "really= expensive"), and one by another US company that went into Chapter 11 prote= ction around the time we were talking to them!=20 > I wonder if targeting GCC to the RTX2010 might have been feasible.=20 > Can I ask what Ada compiler you used for the MA31750? It looks like GCC= =20 > supported the MA31750 until version 3.1, but I don't know whether GNAT=20 > existed then. I'm almost 100% sure GNAT wasn't available for the MIL-STD-1750A; it was a = very niche market and we weren't aware of any C compilers we could've used = at the time, even if we'd wanted to.=20 The Ada compiler we used was the same as Nikolas; TLD. I was also working o= n part of ENVISAT (the Tile Control and Interface Unit - TCIU, although som= e of my colleagues were also using it on the main ASAR control system). Alt= hough Nikolas mentions Matra Marconi Space mandating TLD, that would've com= e down from Dornier who'd apparently done a deal with TLD. I don't know wha= t happened with TLD after that, but some geezer from the Irvine Compiler Co= rporation contacted me once when they were following up on some unpaid lice= nse fees related to part of the TLD compiler.=20