From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00,MSGID_SHORT
	autolearn=no autolearn_force=no version=3.4.5-pre1
Date: 12 Jan 93 22:12:18 GMT
From: visicom!rlk@nosc.mil  (Bob Kitzberger)
Subject: Re: Basic embedded concepts and Ada
Message-ID: <262@visicom.com>
List-Id: <comp.lang.ada>

hawkinsg@saifr00.cfsat.honeywell.com (Glenn Hawkins) writes:

> Although our compiler is 
> pretty efficient with word and byte level accesses, it pretty much chokes on 
bit 
> level accesses, so representing the different registers as records of bits 
> (appropriately rep spec'ed, of course) doesn't seem practical.

Too bad; that's the preferred approach.

>I think that to set bits in these registers it would be easiest for
>me to define the objects representing the registers as 32-bit integers, allowi
ng
>me to simply add constants (1, 2, 4, 8, etc.) representing the different bits
>I want to set. Just like I would in JOVIAL, C, etc.

Don't forget that unsigned 32-bit integers are rarely supported in Ada
on 32-bit CPUs, so it's not as easy as you might wish to set the MSB.

>Seeing if a bit has been set seems more difficult.  [...] I would like to
avoid machine code insertions if possible.  Any ideas?

Since bit-level rep specs appear to be broken in your compiler, I'd use
machine code insertions to implement ANDs, ORs, etc.  I know you want
to avoid them (for portability reasons?), but if they're documented
as a workaround you can mitigate risk.

Then, call your Ada compiler vendor and ask them to fix rep specs.

	.Bob.
----------------
Bob Kitzberger          VisiCom Laboratories, Inc.
rlk@visicom.com         10052 Mesa Ridge Court, San Diego CA 92121 USA
                        +1 619 457 2111    FAX +1 619 457 0888