On 28 July 2013 06:56, Jonathan S. Shapiro <[email protected]> wrote:
> Oh. I'm being brain-dead.
>
> NoDetect is kind of an odd case, but many of the others are subsumed by
>
> reads(%r)  -  Procedure reads no region larger than %r
> writes(%r) -  Procedure writes no region larger than %r

Talpin, Jouvelot, Tofte and the group at ENS really did a fantastic
job.  Asking what is missing from their region and effect system is a
question I've burned years on.

Which arguments escape and where to are really questions for the region system.

DynamicRegionAlloc feels a bit odd.  I think it's actually
documentation.  I do have a gut feeling that, for example, finalisers
should be total functions, and should allocate a finite amount of
memory.  But functions can satisfy those constraints and still
completely miss what that gut feeling is really trying to say.  So I
think that making a requirement like that would actually be of little
practical value.

-- 
William Leslie

Notice:
Likely much of this email is, by the nature of copyright, covered
under copyright law.  You absolutely may reproduce any part of it in
accordance with the copyright law of the nation you are reading this
in.  Any attempt to deny you those rights would be illegal without
prior contractual agreement.
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to