From: larry@JPL-VLSI.ARPA
Subject: Real-time Ada Use
Date: Wed, 2-Apr-86 16:20:26 EST [thread overview]
Date: Wed Apr 2 16:20:26 1986
Message-ID: <8604030800.AA09670@ucbvax.berkeley.edu> (raw)
The following was addressed to me personally, so I'm leaving off the name.
Gee, Larry, I dunno about Ada being tailored to a specific hardware
environment. Ada isn't an ideal real-time programming system. Since Ada
leaves so many things 'implementation-defined', you have to work hard
to write high-performance systems (which are exactly what ALL real-time
systems are). I haven't a great deal of actual experience yet, but I
don't see that real-time systems are that much easier to write
USING ADA-THINK than real-time systems written in C (for instance). Porting
an Ada application that needs to access queued tasks in a particular order
(priority system, with dynamic priorities) is going to be a tough act.
The one area in which Ada is weak is in its parallel processing capability.
Every other aspect of Ada has been tested in actual production languages,
though in (often only slightly) different form. Concurrency, however, is
both inherently complex and an area of computer science that is still pretty
much a frontier. Ada's minimal parallel-processing capability is probably
a good thing. In 1988 when the Ada standard goes up for change I suspect
this is where most of the changes will take place, building on the work of
all of us who are trying to use Ada in real-time systems.
Larry @ jpl-vlsi.arpa
reply other threads:[~1986-04-02 21:20 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox