Re: Checks off for release code - was Re: Size code Ada and C

[William Clodius <wclodius@lanl.gov>]

Robert Dewar wrote:
> <snip>
> 
> In the Ada 83 world, and in the Ada 95 world, different compilers use
> different approaches here. Why on earth did you choose an Ada 83 compiler
> that used implicit heap allocation? and once you had done so, why on earth
> did you use constructs that caused implicit heap allocation? Sounds like
> a case of the customer not being sufficiently well informed of this issue
> (which is a bit surprising to me, it is an issue that has always been a
> high profile one).
> 
> <snip>

As I understand earlier messages, his team chose a compiler precisely
because it did not use implicit heap allocations, and then the vendor
changed the implementation so that it used implicit heap allocations.
The vendor had "lost sight of the best way". Quoting the earlier
message:

> As an example I give the use of implicit heap allocation in Ada83 compilers.  
> Back in 1983 and 1984 my company compared the few compilers that existed for 
> embedded applications using a given processor.  One of the compilers used heap 
> allocation for unconstrained records; one did not.  We picked the one that did 
> not.

> It might have been the next update to the compiler in which they started using 
> heap.  When we complained, we were told that they had had many complaints that 
> people were getting Storage_Error when they had unconstrained types. ...

-- 

William B. Clodius		Phone: (505)-665-9370
Los Alamos Nat. Lab., NIS-2     FAX: (505)-667-3815
PO Box 1663, MS-C323    	Group office: (505)-667-5776
Los Alamos, NM 87545            Email: wclodius@lanl.gov