[m-rev.] for review: Region runtime system

[Quan Phan]

Quoting Julien Fischer <juliensf at csse.unimelb.edu.au>:

> >
> > Implement the region runtime system and partly integrate it with the
> runtime
> > system of Mercury. This partly integrated runtime only supports
> "deterministic"
> > programs (deterministic programs in which no region-related actions happen
> in
> > condition of if-then-elses, at least no removals of regions happen).
> > For example, it now works with nrev, qsort, life, which are some
> > small deterministic benchmarks I used in my related paper.
> 
> Works, as in works with the HLDS->HLDS transformation that you've 
> implemented?
> 

That is if we compile those programs with the RBMM-enabled compiler (HLDS->HLDS
transformation and code generation for RBMM, and the partly integrated runtime)
the executables we get will be able to run with dynamically-allocated memory
handled by RBMM.

I am modifying other things as per your comments. Thanks.

Quan
--------------------------------------------------------------------------
mercury-reviews mailing list
Post messages to:       mercury-reviews at csse.unimelb.edu.au
Administrative Queries: owner-mercury-reviews at csse.unimelb.edu.au
Subscriptions:          mercury-reviews-request at csse.unimelb.edu.au
--------------------------------------------------------------------------