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=0.6 required=5.0 tests=BAYES_20,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: fd705,41969c265865be87 X-Google-Attributes: gidfd705,public X-Google-Thread: 102b75,660f5ccab731a5a3 X-Google-Attributes: gid102b75,public X-Google-Thread: 103376,9cccf6ef6149fdaa X-Google-Attributes: gid103376,public X-Google-Thread: ff68d,660f5ccab731a5a3 X-Google-Attributes: gidff68d,public From: "Marin D. Condic" Subject: Re: Ada Date: 2000/01/15 Message-ID: <3880C4F5.6AE09BBA@quadruscorp.com>#1/1 X-Deja-AN: 573059602 Content-Transfer-Encoding: 7bit References: <83reu2$2soi$1@msunews.cl.msu.edu> <38615cc4.22862595@news.shuswap.net> <84dnsu$g69@nnrp1.farm.idt.net> <84drm7$ss8$1@news.rchland.ibm.com> <855lqp$t2@nnrp4.farm.idt.net> <85l4kt$e9q@nnrp1.farm.idt.net> <85nohj$n25@spool.cs.wisc.edu> <387F8105.7C714792@quadruscorp.com> <85qcne$8e3@spool.cs.wisc.edu> Organization: Quadrus Corporation X-Sender: "Marin D. Condic" (Unverified) X-Server-Date: 15 Jan 2000 19:05:54 GMT Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 Newsgroups: alt.folklore.computers,comp.arch,comp.sys.unisys,comp.lang.ada Date: 2000-01-15T19:05:54+00:00 List-Id: Andy Glew wrote: > > > BTW: I don't see how this could be perceived as a handicap to Ada. It > > provides a facility most other languages do not, yet if your application > > has no need of it, there is no real penalty. How do you see it as a > > barrier to using Ada? > > (1) It was a barrier to ports. > > (2) A long as there is absolutely no real penalty. O.K., I can see #1 to a point. Ports to small microprocessors still don't exist in any real numbers and some of this may be due to perceived difficulties in getting tasking to work there. #2 has the conditional: "It Depends" attached to it. Certainly for compilers that target a capable OS with tasking primitives should not exhibit any penalties for tasking when it is not used, since it is the OS that holds all (most) of the code. For embedded targets where you have to supply some sort of RTK, the question becomes "how good is the compiler/linker at removing dead code?" Many modern linkers are pretty good at eliminating subroutines that are never called. And of course, it still comes down to the specific implementation. A poor quality compiler for any language is going to create penalties and WRT Ada, there are more opportunities to do so than in less sophisticated language. However, there isn't anything inherent in the language design which makes it impossible to generate efficient, tight code. There are many Ada compilers which do so. The trick is to do a good evaluation of the compilers available for the application at hand and see what sort of code generation you really get. Some are overall better than others. Each does some things well and other things poorly. You always want to build some sample code similar to the intended application and see what you get with a specific compiler. This is true for any language. MDC -- ============================================================= Marin David Condic - Quadrus Corporation - 1.800.555.3393 1015-116 Atlantic Boulevard, Atlantic Beach, FL 32233 http://www.quadruscorp.com/ m c o n d i c @ q u a d r u s c o r p . c o m Visit my web site at: http://www.mcondic.com/ "Capitalism without failure is like religion without sin." -- Allan Meltzer, Economist =============================================================