On 7 August 2013 17:57, Bennie Kloosteman <[email protected]> wrote:
> re sub graphs  , when you use a borrowed pointer and pass to another task it
> transfer all the data by value , this is what its  probably referring to.
> They also have some object cloning mechanism ..
>
> How does deep copy on return break region unfication ?

Break in a good way!  It means the return region is disjoint from all
the others, which means we can de-allocate the others wholesale.

It could be considered a good thing to have, because de-allocation
decisions may not be entirely obvious (from the control flow that
remains in the SSA form).  It could be interesting for data locality,
and also for effect based optimisations, because initialisation of
some fields of the object could be deferred if the fields are not
accessed before the copy.

I'm a bit uncomfortable about the compiler inserting random copy code
into my function just because I added an if statement, but I guess it
wouldn't be the only option anyway.

-- 
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