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 10.66.221.137 with SMTP id qe9mr588383pac.4.1392300885492; Thu, 13 Feb 2014 06:14:45 -0800 (PST) X-Received: by 10.140.91.227 with SMTP id z90mr35526qgd.3.1392300885437; Thu, 13 Feb 2014 06:14:45 -0800 (PST) Path: border1.nntp.dca3.giganews.com!backlog4.nntp.dca3.giganews.com!backlog4.nntp.ams3.giganews.com!backlog4.nntp.ams.giganews.com!border2.nntp.ams.giganews.com!nntp.giganews.com!xlned.com!feeder3.xlned.com!news.glorb.com!c10no20416258igq.0!news-out.google.com!s3ni17711qas.0!nntp.google.com!f11no19389134qae.1!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Thu, 13 Feb 2014 06:14:45 -0800 (PST) In-Reply-To: <%r2Lu.9395$yo1.4292@fx26.fr7> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=189.77.226.1; posting-account=TRgI1QoAAABSsYi-ox3Pi6N-JEKKU0cu NNTP-Posting-Host: 189.77.226.1 References: <44f053fc-72a9-44f0-8a0d-c919632d57e8@googlegroups.com> <%r2Lu.9395$yo1.4292@fx26.fr7> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: Subject: Re: Tiva + Ada From: "Rego, P." Injection-Date: Thu, 13 Feb 2014 14:14:45 +0000 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Original-Bytes: 4567 Xref: number.nntp.dca.giganews.com comp.lang.ada:184822 Date: 2014-02-13T06:14:45-08:00 List-Id: On Thursday, February 13, 2014 8:58:51 AM UTC-3, Brian Drummond wrote: > TAMP is probably the best starting point. When I got Tiva I actually expected that I could use a more plug-and-code f= ull Ada builder, but TAMP looks to be a very good starting point. Better th= at the documentation is very organized, so I have the feeling that it will = not be difficult to begin with.=20 The only problem (not major) is that it is intended for ZFP only. For using= my Arduinos (which avr-elf-windows from Adacore is also ZFP) I had to impl= ement several clock/scheduling routines, and I bricked some boards mixing u= p some fuses to make debug to run, and it delayed a bit the application dev= elopment. And it would be very good if could use tasking and secondary stac= k. Fortunately Tiva has ready-to-use debug, so at least half of the problem= s I think I will not have with it. > I am currently re-visiting the MSP430-Ada project now that gcc4.9 has=20 > official support for the msp430. > https://sourceforge.net/projects/msp430ada/ > (hasn't been updated to reflect gcc4.9 yet...) Great. > But I also have interest in the ARM-Cortex (and what I learn on the MSP43= 0=20 > will largely transfer) : I have a couple of TI Launchpads (Stellaris, the= =20 > precursor to Tiva ... I'm a little behind!) and a Hercules RM42x. Some days ago I began a github project for using Raspiberry Pi (ARM11) devi= ces with Ada by mapping the BCM2835 chip, but it's still in the very beginn= ing. But RPi does not expose all the pins that I would need. But Tiva does = not have Ada ready-to-use, things could be a bit easier, right? > The latter is especially interesting as an Ada target : it's a dual- > processor ARM Cortex operating in lockstep, aimed at high reliability=20 > applications, on a devboard costing under =A315. >=20 > http://uk.farnell.com/texas-instruments/launchxl-rm42/rm42-hercules- > launchpad-dev-board/dp/2341348 >=20 > There are 2 versions : the RM42 aimed at industrial/medical applications= =20 > and an almost identical one tweaked for automotive applications > http://uk.farnell.com/texas-instruments/launchxl-tms57004/tms57004- > hercules-launchpad-dev/dp/2341346 The cost of them is really very attractive.=20 > And TI expect you to program this high integrity dual-processor platform= =20 > in ... what? That's the point. In C? For God... Ok, no problem with C. TI has a very goo= d development studio in C based on Eclipse with several good tools and inte= grated with the debug API. But I think I could extract the best of the boar= d with Ada if it were available. Maybe if GNAT GPL would include the ARM cross-compiler, it would be good to= o (even with ZFP). > Incidentally I can find several small runtimes (offering the Ravenscar=20 > profile) that may be worth adapting to the ARM (or are already targetted= =20 > to it) but so far they all seem to be pure GPL, with no GMGPL (or GPLv3 > +runtime exception) ... does anyone know otherwise? > ORK : http://www.dit.upm.es/~ork/index.html/ > AVR32 (not ARM) : https://github.com/gregerts/GNATforAVR32 > ARM-Ada : https://sourceforge.net/projects/arm-ada/ > As this last already targets the ARM it may be the best starting point. Interesting. Regards, Rego.