TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy

TEAM-ADA@LISTSERV.ACM.ORG

Options: Use Forum View

Use Proportional Font
Show HTML Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Ed Falis <[log in to unmask]>
Reply To:
Date:
Wed, 16 Apr 1997 12:31:04 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (69 lines)
I was tasked to run the ACES for a bid we did last year.  We decided to bid
an Ada 95 compiler.  I encountered a number of problems.

1.  The suite infrastructure appear to assume an Ada 83 style library
management system rather than a source-based system like in at least GNAT
and ObjectAda, and probably Ada-multi from Green Hills.  This required a
great deal of manual massaging of the scripts.

2.  The harness program has several program components that are not legal
Ada 95 (having to do with use of uninitialized actual parameters  for mode
"in out" formals.  These only worked if the harness was built with checks
suppressed.

3.  The same error occurred in a fairly large number of test cases.

4.  A fair number of tests also use obsolete floating point attributes.

5.  There were a few other miscellaneous Ada 95 incompatibilities.

Altogether, it took me over 6 weeks full-time to generate the results,
which were still not complete.  And those who know me, know that I'm
reasonably sophisticated at dealing with benchmarking efforts.  (The first
time I ran the suite, on an Ada 83 system, took about 3-4 weeks, and also
involved some customization of the scripts).

My personal opinion is that the suite delivers small bang for the buck
involved in setting it up and running it, and that its architecture is
over-engineered to the characteristics of Ada 83 development systems (and
that it would need a very substantial reworking to fix it).

Hey! You asked, so I'm telling you what I think.

My opinions are my own, and not necessarily those of Aonix.

- Ed

java/lang/Class.

 ----
>From: Edward Colbert <[log in to unmask]>
To: [log in to unmask]
Date: Wednesday, April 16, 1997 2:50 AM
Subject: Are there Problems with the ACES?

Hi Phil,

I consulting on a project that's having a compiler vendor create custom Ada
95 code generator.  When we told the vendor that we plan to start testing
the compiler they just delivered with a subset of the ACES, the vendor made
a vague statement that there are problems with the ACES.   Are you (or
anyone on the Team Ada mail list) aware of any problems that we should
know?  I don't see any thing in ACES documentation that should be a
problem.

Thanks in advance.

Take Care,
Ed

----------------

Edward Colbert
President
Absolute Software Co., Inc.
4593 Orchid Dr. Los Angeles, CA   90043-3320
Phone:  (213) 293-0783
FAX:    (310) 215-FAX4
E-mail: [log in to unmask]

ATOM RSS1 RSS2